Control of Emissions from Marine SI
   and Small SI Engines, Vessels, and
   Equipment


   Summary and Analysis of Comments
United States
Environmental Protection
Agency

-------
                        of                                 SI
                             SI
                               Equipment

              Summary       Analysis of Comments
                          Assessment and Standards Division
                          Office of Transportation and Air Quality
                          U.S. Environmental Protection Agency
EPA
                                              EPA420-R-08-016
Environmental Protection                                 _  ^  ,  onnn
Agency                                          September 2008

-------
                             TABLE OF CONTENTS




Index of Commenters




1.  General and Cross Program Issues




2.  Small SI Engines and Equipment




3.  Marine SI Engines and Vessels




4.  Evaporative Emissions

-------
INDEX OF COMMENTERS
Commenter
A. Swanson
AAR
Alcock, G.
American Honda Motor Company, Inc
American Marine Sports
American Road & Transportation Builders
Association
American Society for Testing and Materials
Arctic Cat
Attwood Corporation
Baja Marine Corporation
Bombardier Recreational Products
Briggs & Stratton
Brunswick Commercial & Government Products, Inc.
Brunswick Corporation
California Air Resources Board
Caterpillar Inc.
Challenger Powerboats, Inc
Chaparral Boats, Inc.
Cigarette Racing
Congresswoman Velazquez
Cummins Inc.
Delphi Corporation
Electro-Motive Diesel, Inc
Emission Control System Engineering, Testing and
Certification
Engine Manufacturers Association
Abbreviation

AAR

Honda

ARTBA
ASTM



BRP



California ARE







EMD
ECO
EMA
Docket ID Number
EPA-HQ-OAR-2004-0008-
-0596
-0669
-0601
-0705
-0767
-0639
-0685
-0606
-0709
-0653
-0662
-0674
-0657
-0767
-0652
-0695
-0682
-0718
-0813
-0814
-0822
-0644
-0630
-0637
-0702
-0719
-0785
-0795
-0638
-0687
-0797
-0801
-0712
-0798
-0802
-0691
-0768
-0807
-0808
-0809
-0810
-0816
-0817
-0818
                                                                 11

-------
Commenter
Enviro-Fill, Inc.
Environmental Control Corporation
Environmental Defense
Euromot
Exergy Partners Corporation
Fluoro-Seal International
Four Winns Boats L.L.C.
General Electric Transportation
General Motors
Godfrey Marine
Grady-White Boats, Inc.
Hallett Boats
Harold Haskew & Associates, Inc.
Holland, D.
Houston-Galveston Area Council
Husky Corporation
Ilmor, Inc.
IMPCO Technologies, Inc.
Inca Molded Products
Indmar Products
Industrial Truck Association
Ingenium
International Consortium for Fire Safety, Health and
the Environment
ISMA
John Deere
Johnson County Environmental Department
Keichline, R.
Kelly, A.
Kohler Company
Larson/Glastron Boats
Leggett, N.
Lowe Boats
Lund Boat Company
Maryland Department of the Environment
Massachusetts Marine Trades Association
MeadWestvaco
MECA
Abbreviation

EVCC





GE
GM





H-GAC


IMPCO
Inca
Indmar
ITA

ICFSHE
ISMA










MMTA

MECA
Docket ID Number
EPA-HQ-OAR-2004-0008-
-0684
-0806
-0608
-0654
-0648
-0649
-0766
-0627
-0646
-0650
-0625
-0679
-0786
-0787
-0645
-0677
-0713
-0640
-0595
-0633
-0803
-0658
-0614
-0692
-0812
-0700
-0667
-0800
-0616
-0711
-0671
-0767
-0619
-0561
-0643
-0703
-0626
-0603
-0612
-0660
-0655
-0722
-0634
-0723
-0724
-0668
111

-------
Commenter
Mercury Marine
Mid- American Regional Council Air Quality Forum
Motorcycle Industry Council
National Association of Clean Air Agencies
NACCO Materials Handling
National Association of State Fire Marshals
National Marine Manufacturers Association
Natural Resources Defense Council, etal
Nautigaz
New Jersey Department of Environmental Quality
New York State Department of Environmental
Conservation
Nixon, T.
North American Sleekcraft, Inc.
Northeast States for Coordinated Air Use
Management
OPW Fueling Components
Outdoor Power Equipment Institute, Inc.
Ozone Transport Commission
Paddock, B.
Parker Hannifin Corporation
Pennsylvania Department of Environmental Protection
Premier Marine, Inc.
Protectoseal
Pure Power EcoFuel LLC and Marine Propulsion
Technology, Inc.
Recreation Vehicle Industry Association
Regal Marine Industries, Inc.
Regulator Marine Inc.
Rolls Royce
S2 Yachts, Inc
Saint-Gobain, PPL
Sea Ray Boats, Inc.
Shields Marine Hose
Skeeter Products, Inc
Small Engine Repair Service
Smith, F.
Abbreviation
Mercury
MARC AQ Forum
MIC
NACAA

NASFM
NMMA
NRDC

NJDEP
NY DEC

NAS
NESCAUM

OPEI
OTC


PADEP


Pure Power
RVIA










Docket ID Number
EPA-HQ-OAR-2004-0008-
-0693
-0717
-0716
-0799
-0696
-0701
-0821
-0651
-0714
-0673
-0688
-0790
-0792
-0805
-0690
-0727
-0710
-0659
-0599
-0666
-0641
-0804
-0675
-0793
-0678
-0607
-0672
-0676
-0613
-0615
-0664
-0647
-0635
-0632
-0715
-0697
-0661
-0796
-0683
-0624
-0706
-0623
-0694
IV

-------
Commenter
South Coast Air Quality Management District
Sterling Performance
Stihl Inc.
Suzuki Motor Corporation
Trident Rubber, Inc.
Triton Boat Company
U.S. Coast Guard
Volvo Penta of the Americas
White, L.
Wisconsin Department of Natural Resources
Wood Mfg. (Ranger Boats)
Yamaha Motor Corporation USA
Yellowfin Yachts
Abbreviation
South Coast AQMD
Sterling

Suzuki



Volvo

Wisconsin DNR

Yamaha

Docket ID Number
EPA-HQ-OAR-2004-0008-
-0704
-0689
-0665
-0767
-0784
-0698
-0636
-0656
-0631
-0708
-0620
-0725
-0663
-0628
-0721
-0826
-0681

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments




1   Rulemaking Process and Cross-Program Issues

What We Proposed:

       The comments in this section generally correspond to Sections I, II, VIII, IX, XI, XII,
XIII, and XIV of the preamble to the proposed rule, where we give an overview of the
rulemaking, describe the process for public participation, and detail a range of technical
amendments that apply to programs other than those that are the focus of the proposed emission
standards.  The applicable regulatory provisions for cross-program issues are in 40 CFR parts
1027, 1065, 1068, and 1074.  In addition, we proposed technical amendments to specific
programs in 40 CFR 89, 90, 91, 94,  1033, 1039, 1042, 1045, 1048, 1051, 1054.  The Regulatory
Impact Analysis describes the calculated air quality and benefits associated with the proposed
standards in Chapters 2 and 8, respectively.

1.1  General input

1.1.1   Broad support and general observations

What Commenters Said:

       EMA commented that it has been an active participant in the development of the NPRM
for the next-phase Small  SI engine standards. Specifically, EMA has worked to help EPA staff
determine: (i) the most effective exhaust emission control technologies that could be applied to
Small SI engines; (ii) the most effective evaporative emission control technologies that could be
applied to Small SI engines and the  equipment that they power; (iii) limitations on the
applicability of such emission control technologies to Small SI engines and the equipment that
these engines power; and (iv) the optimized timeline for deploying the available emission control
technologies into the marketplace.  The net result of that collaborative process is an NPRM that
truly and properly reflects the maximum achievable emission reductions for Small SI engines
and the equipment that they power.  In that regard, EMA greatly appreciates the time and effort
that have gone into the development of the pending rulemaking — a rulemaking that has set forth
extremely challenging and dramatic, but nonetheless potentially achievable, emission reduction
targets. Indeed, the effort that has gone into this collaborative rulemaking has resulted in the
promulgation of an overall framework of technology-forcing standards and accompanying
regulations that are at the very limit of feasibility and implementability. As a consequence, that
overall framework needs to be maintained in any final rule that results from the NPRM, since
any potential increased stringency of the proposed standards or the overall regulatory program
would necessarily result in an infeasible and nonimplementable rule.

       EMA continued to comment that the NPRM properly recognizes the inherent constraints
on the transfer of advanced exhaust  emission control systems to Small SI engines, and
appropriately limits the efficiency of the required aftertreatment to achieve the proposed
standards to those levels that  can be effectively implemented taking into consideration, among
other things, noise and safety. Similarly, the NPRM properly recognizes that evaporative
emission control of Small SI engines and the equipment these engines power may involve the

                                       1-1

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
engine manufacturer, the equipment manufacturer, or the component supplier. Accordingly,
EMA supports the overall framework of the NPRM, and urges EPA to finalize a rule that
preserves that framework in order to maintain the feasibility of the pending rulemaking.

       EMA generally supports the NPRM's approach to the certification process and the
standardization of testing requirements.  Certification is critical to the creation of a level playing
field. Manufacturers must have confidence that all competitors are required to meet the same
requirements, and are subject to the same liabilities with respect to emission related product
performance and warranty. Approved alternate test procedures must be available for all
manufacturers without subsequent approval.  EMA recommends that approved alternative
procedures for emission testing, both exhaust and evaporative, and exhaust emission
deterioration factor determination be documented and posted on the EPA website for all
manufacturers to see and use.  In  addition, EMA has the following specific concerns with respect
to the NPRM's proposed certification and testing requirements that should be resolved in the
final rule,  (see section 1054.245(b)(9)).

       EMA commented on §1065.12 "Approval of alternate procedures." that approved
alternate test procedures must be  available for all manufacturers without subsequent approval.
EMA recommends that approved alternative procedures be documented and posted on the EPA
website for all manufacturers to see and use.

       Kohler Co. is committed to participating with government and regulatory agencies in the
development of responsible environmental law and regulations. Kohler believes in harmonizing,
to the greatest extent possible, the EPA Phase 3 regulation with the California Tier III regulation.
A 50-state regulation is in the best interest of the industry and the Phase 3 regulation reasonably
balances the benefit to the environment with the additional product costs of compliance. Kohler
believes the resulting compliant product will perform satisfactorily in every way, including
safety. Kohler agrees with the overall framework of the NPRM for Phase 3 regulations and has
participated in and publicly supported EPA's efforts in developing the Phase 3 regulation.
However, Kohler has some concerns with the details of the regulation. There are a few aspects of
the regulation on which Kohler feels compelled to provide  additional comment. These include:
determining engine displacement for the LSI exemption, exhaust emission calculations,
transparency of approved alternative emission test cycles, allowable maintenance for DF testing,
and record keeping/reporting.

       NMMA commented that its members are supportive of EPA's proposal to control
evaporative and exhaust emissions from marine SI engine s and fuel system components.
NMMA members are committed  to producing environmentally responsible products.  Over the
years, the recreational marine  industry has devoted significant time and resources to ensure that
they are in compliance with all applicable federal environmental and safety regulations as well as
recommended industry "best practices" and standards. NMMA also has worked closely with
EPA on the implementation of several important regulatory programs, including this current
proposal. With all of these rulemakings, NMMA has appreciated EPA's willingness to consider
additional information and data from NMMA members and work collaboratively with the
recreational marine industry to address concerns which the industry has had with the specifics in
these rulemakings.
                                           1-2

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
       Indmar commented that they support the proposed federal emission regulations for new
marine spark-ignited sterndrive/inboard engines that will substantially reduce emissions from
these engines.  There are a few areas that they would suggest alternative language or changes.
Indmar Products supports the enactment of the proposed Federal sterndrive/inboard exhaust
emissions standards with the noted changes.

       Overall, BRP commented that they are supportive of EPA's proposal to control
evaporative and exhaust emissions from marine SI engines and fuel system components and to
the revisions of the regulations impacting off-road recreational vehicles and engines. Over the
years, BRP has devoted significant time and resources to ensure compliance with all applicable
federal environmental and safety regulations as well as recommended industry "best practices"
and standards. BRP also has worked closely with EPA on the implementation of several
important regulatory programs, including this current proposal. With all of these rulemakings,
they have appreciated EPA's willingness to consider additional information and data.

       Mercury Marine supports EPA's proposal to control evaporative and  exhaust emissions
from marine SI engines and fuel  system components. They are committed to producing
environmentally responsible products. Over the years, the recreational marine industry has
devoted significant time and resources to ensure that they are in compliance with all applicable
federal environmental and safety regulations as well as recommended industry "best practices"
and standards. Mercury Marine has worked closely with EPA on the implementation of several
important regulatory programs, including this current proposal.

       Mercury Marine has a few remaining concerns regarding the technology, timing, and
implementation required by the proposal.  Catalysts and low permeation hoses are available and
can be incorporated into marine exhaust and fuel systems. However, there are market issues,
some that are out of their control, that have an impact on our abilities to meet some of the
proposed standards in the proposed timeline. Therefore, they will propose alternatives to the
items they have issues on that will provide EPA with emissions reductions but maintain their
business, customers, and employees. If the issues they have raised in these comments are
adequately addressed, then Mercury Marine fully  supports this rule.

       Yamaha is supportive of EPA's proposal with the addition of NMMA industry comments
to control evaporative and exhaust emissions from Marine SI engines and fuel systems
components. Over the past many years Yamaha has worked closely with the EPA on other
rulemakings for motorsports products on both evaporative and exhaust emission controls. With
all these rulemakings Yamaha has appreciated EPA's willingness to consider additional
information and data from Yamaha and work with the industry to address concerns and issues
relative to specifics in these rulemakings.  It is Yamaha's position that although this proposal is
very comprehensive and they are in general agreement with its intent, however there are still a
few remaining concerns outstanding that need to be addressed regarding technology availability
and implementation timing of this proposal. As they hope, these comments along with those
submitted with their approval of the NMMA will demonstrate the need for additional lead time
to design, manufacture and implement effective controls for exhaust and evaporative emissions.
                                       1-3

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
Yamaha also will offer comments on other aspects of the proposal in regards to certification,
ABT and compliance test protocols.

       Volvo Penta commented that "care for the environment" is a core corporate value of the
entire Volvo Group, including Volvo Penta. Volvo Penta supports EPA's proposal to control
exhaust emissions from SD/I engines, and has been committed to producing environmentally
responsible products.  Over the years, Volvo Penta has devoted significant time and resources to
ensure their products  comply with all applicable federal environmental and safety regulations as
well as recommended industry standards. During the past several years, Volvo Penta has worked
closely with the National Marine Manufacturers Association (NMMA) and the EPA to craft a
workable solution to emissions concerns. Throughout this process, Volvo Penta has appreciated
EPA's willingness to  consider additional information and data from the industry. These
collaborative efforts have reduced industry concerns  regarding the proposed rule. The comments
included reflect the Volvo Penta's few remaining issues of concern:

       Euromot commented that they have reviewed the proposed Phase 3  regulation within
their membership and they fully support the OPE1 comments. As established manufacturers with
a long compliance history they understand the new concept of EPA for imports (and exports).
The changes proposed by the OPEI comments are essential  for the Euromot members to be
present on the market in the future. Without these changes the regulation would be not practical
and a  dramatic burden (financial and administrative)  would be laid to the industry.

       In a public hearing, Ilmor stated that broadly  speaking, Ilmor supports the proposed EPA
Rule.

       Pleasurecraft Marine commented in a hearing that their company is  fully supportive of
emission reductions and is working diligently to solve the many complex technical  and
implementation issues associated with manufacturing a catalyst controlled engine. They would
like to thank the EPA for their insights in crafting a document that, with minor changes, will be
of great benefit to our industry, the boating community and the environment.

       Brunswick supports EPA's efforts of improving the environment through cleaner
products.

       Inca does support the need for control of evaporative emissions through low permeation
fuel tanks, diurnal emission controls, and low permeation hose. However, due to the direct
correlation their products have with boating safety they want to be sure that the requirements of
this standard do not create a low emission product that is inferior in quality to the current product
that has been successful in the marine industry for years.

       Heraeus commented that their small engine solutions provide a variety of different
catalyst solutions. They provide effective conversion of the bad stuff coming out of the engine.
They have catalysts which match the durability needed, in other words, addressing the DF factor
for the engine application. And an engine exhaust catalyst is really quite a cost-effective
approach. They support the proposed EPA Phase 3 emissions regulations for Class I and Class II
                                          1-4

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


nonroad spark-ignition engines. Heraeus is ready and believes this rule should be finalized as
soon as possible.

      MeadWestvaco Corporation supports the EPA's proposal to control evaporative
emissions from marine SI engines and fuel system components and is committed to providing
products used to control these emissions.

      NACAA strongly supports prompt EPA action to reduce emissions from these sources.
They believe this long-awaited proposal - which includes HC, NOx, and carbon monoxide (CO)
exhaust emission standards, as well as evaporative emission standards - is a critically important
step forward.

      EVCC commented that as their technology has proven reductions in HC+NOx of up to
98.9% on a two-stroke engine and 90% on a four-stroke engine (while also significantly reducing
CO), they are not only in full support of the current regulations, but they encourage our
organizations to set even more stringent standards in the near future.

      Ozone Transport Commission (OTC) and Maryland support the EPA's effort to regulate
emissions from nonroad spark-ignition engines, vessels, and equipment.  OTC has anticipated
the EPA's proposed regulation.  In a March 14, 2006 letter to the EPA Administrator several
Midwest and OTC states urged the EPA to promulgate the regulations for nonroad spark-ignition
engines. As the letter stated, this source category "has the potential to provide very significant
reductions." Additionally, the OTC adopted Resolution 06-02 on June 7, 2006 requesting that
EPA develop and implement a strong national program reflecting current technology
advancements regarding small engine emissions. The OTC is encouraged by the May 18, 2007
proposal.

      NESCAUM (Northeast States for Coordinated Air Use Management) submitted
comments on EPA's Proposed Rule for Control of Emissions from Nonroad Spark-Ignition (SI)
Engines and Equipment commending EPA and strongly supporting the goals of this rulemaking
effort. NESCAUM is an association of state air pollution control agencies in Connecticut, Maine,
Massachusetts, New Hampshire, New Jersey, New York, Rhode Island, and Vermont.

      The Mid-America Regional Council (MARC) Air Quality Forum, created in accordance
with Section 174 of the Clean Air Act to coordinate the  development and implementation of air
quality policy in the bi-state Kansas City region, wishes to express its strong support for EPA's
proposal to  set federal emissions  standards for  small spark-ignition engines.  The public health
impacts of ground-level ozone and fine particulates are well documented, and to the extent that
the proposed rule will lead to significant and measurable reductions in both pollutants nationally,
they urge EPA to implement the small engine rule as expeditiously as possible.

      The California Air Resources Board (California ARB) commented that the federal Clean
Air Act (CAA) Amendments of 1990 preempt  California from controlling emissions from new
farm and construction equipment under 175 horsepower (hp). Because of this preemption,
significant emissions from these engines are beyond California ARB's authority to regulate, and
California ARB must rely on EPA to establish  regulations.  Furthermore, it is important that EPA
                                       1-5

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
also adopt more stringent emission standards for nonpreempted engines, such as those used in
marine applications. Federally certified engines used in marine vessels can be used in California
which could impact California's ability to meet clean air goals. Adoption of the proposed
regulation outlined in the proposed rule by EPA is necessary for the protection of public health in
California and to comply with air quality standards. New stringent and cost effective standards
should be adopted for these categories in a timely manner to ensure that the cleanest engines and
equipment be introduced into the fleet at the earliest possible date.  In general, California ARB
supports the direction that EPA is taking to control emissions from nonroad spark-ignition
engines and equipment included in this notice of proposed rulemaking (NPRM).

       The South Coast Air Quality Management District (SCAQMD) staff appreciates this
opportunity to provide formal comments on the proposed regulation for Nonroad Spark- Ignition
Engines and Equipment. Although they commend the EPA's efforts in developing this proposal,
the SCAQMD staff believes that the proposed regulations must be further strengthened in order
for California and in particular, the South Coast Air Basin, to meet applicable federal fine
particulate matter (PM2.5) and 8-hour ozone air quality standards in an expeditious manner.

       The NJDEP supports the adoption of federal emission standards, which are consistent
with standards adopted by California ARB. With respect to the implementation dates, the NJDEP
encourages expeditious implementation of the federal standards and requests that the
implementation dates be advanced to align more closely with the implementation dates
established by California ARB.

       Wisconsin DNR commends EPA for its leadership in issuing this proposed rule and
seeking comments  on the merits of better emission standards. They support EPA's proposal to
set stringent emission standards for new nonroad spark-ignition engines, equipment, and vessels.
Emissions from the engines covered by this proposal are substantial that contribute to
unhealthful concentrations of PM, ozone, CO, and toxic air pollutants, which translate into
serious adverse health impacts. The proposal would significantly reduce harmful exhaust
emissions as well as evaporative emissions from these sources - which include HC, NOx, and
CO.

       New York State DEC supports EPA's proposal  to set stringent emission standards for
new nonroad spark-ignition engines, equipment, and vessels. Emissions from the engines
covered by this proposal are substantial that contribute  to unhealthful concentrations of PM,
ozone, CO, and toxic air pollutants, which translate into serious adverse health impacts. The
proposal would significantly reduce harmful exhaust emissions as well as evaporative emissions
from these sources - which include HC, NOx, and CO.

       New York State DEC continued to comment that the importance of reducing emissions
from small spark-ignition engines used in applications such as lawn and garden equipment, and
recreational marine emissions, cannot be overstated. Most of the operation of these engines
occurs during warm weather conducive to the formation of ground level ozone, and their ozone
precursor emissions are poorly controlled compared to  other classes of engines and vehicles.
Even with the proposed standards, an hour of operation of these engines will yield hydrocarbon
and oxides of nitrogen emissions comparable to driving an average (Tier 2 Bin 5) new light duty
                                          1-6

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


vehicle hundreds, in some cases thousands, of miles. Indeed, it is likely that many households
produce more ozone-forming emissions maintaining their lawn than commuting to work.
Considering the comparatively high emissions levels of these engines, and their increasing
importance to emission inventories, we believe that additional future reductions are necessary.
The Department urges EPA to continue its efforts to identify emissions control technology that
achieves emission reductions beyond what is currently proposed.

      New York State DEC continued to comment that although the standards proposed by
EPA provide significant and necessary reductions in ozone forming emissions from small land
based engines and recreational marine engines, these engines will still be relatively poorly
controlled compared to other mobile sources. These engines will still  emit ozone precursors out
of proportion to their numbers and hours of operation. The disproportionate emissions levels of
these engines compared to passenger vehicles shows that there is a need for more stringent
standards than the ones currently proposed. EPA research into nonroad spark-ignition engine
emissions control is crucial to identifying available technology which can achieve a greater
degree of emissions reductions. In particular additional work is needed to facilitate the
application of catalysts to outboard and personal watercraft engines, many of which are
automotive sized. They urge EPA to continue its efforts in this field, and build on its current
success, ultimately promulgating more stringent regulations for all of these classes of nonroad
engines.

      Pennsylvania DEP strongly supports EPA's action, but offers additional
recommendations to strengthen the final regulation. While DEP strongly supports EPA's
proposed action to reduce emissions from small land-based and marine spark-ignition engines,
equipment, and vessels, and urges EPA to promulgate the final rule as soon as practicable, DEP
understands and acknowledges that  certain aspects of the proposed rulemaking were delayed by
further study mandated by law and that Congress has also expressly precluded states from taking
any action more stringent than EPA's on small spark-ignition engines. Therefore, DEP urges
EPA to ensure that the final rule is fully implemented expeditiously to achieve the greatest
degree of emission reductions.

      The Houston-Galveston Area Council (H-GAC) Board of Directors would like to offer
general support  and comments for the Environmental Protection Agency proposed rule. They
greatly appreciate the efforts of the EPA to draft this proposed rule, which targets several of
these mobile sources for significant emission reductions that could not otherwise be achieved.

      Johnson County, Kansas commented that they are in full support of the proposed
regulations on nonroad motor engines. In an effort to lower harmful ozone causing emissions
such as those spewed from lawn equipment, they find the proposed standard to be an effective
step in improving the health of our social and biotic communities.

      Clean Air Watch commented that smog is a serious public health problem: technically
known as ozone, smog can cause asthma attacks among children and adults, send people to
hospital emergency rooms, and reduce a person's lung capacity. It has even been linked to
premature death. The evidence is quite clear that even though we have reduced air pollution
through other Clean Air Act standards, such as those for motor vehicles, we still need to make
                                       1-7

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
further progress to protect breathers. Just last Saturday, the Washington Post reported on
continuing air pollution problems in the Washington, D.C. area. And our Clean Air Watch
surveys verify that similar problems persist in many states.  Clean Air Watch continued that
EPA's independent science advisers and the agency's own scientists have concluded that existing
air quality standards for smog must be made stricter to protect kids with asthma and others. As
we clean up cars and trucks, small engines are an increasingly large part of the pollution
problem. Cleaning them up absolutely must be part of the solution. EPA's proposed standards
are a good step in the right direction. Clean Air Watch commented that it would be better for air
quality if they could take effect sooner. They hope we will resist any effort to delay or weaken
these standards.  They believe EPA's detailed studies have put to rest concerns previously raised
about safety.  So they encourage EPA to move forward and issue these  standards in final form
this year.

       Environmental Defense strongly supports the immediate issuance of a final rule to control
air pollution from spark-ignition marine and small engines in light of the serious public health
and welfare problems posed by the exhaust and evaporative emissions from these engines. They
believe a final rule that reflects, or is more protective than, the timing and level of reductions
currently required by California ARB is achievable, cost-effective and necessary to protect
human health and the environment.

       N.  Leggett offered support of the concept of requiring pollution control  technology for
spark-ignited  engines. However, some fine tuning of the proposed regulations is needed to
prevent the regulations from inhibiting progress in the invention  of new technology and the
training of technologists.

       R. Keichline is supportive of the new legislation to reduce air emissions.

       A.  Swanson commented on a recent EPA air emissions proposal (Docket ID No. EPA-
HQ-OAR-2004-0008). The commenter supported the proposal to: a) increase exhaust emission
standards for marine spark-ignition engines and small land-based nonroad engines, b) establish
new evaporative emission standards for equipment and vessels using these engines. This
proposal is reasonable, for these standards would apply only to newly manufactured products,
and it would reduce the harmful health effects of ozone and carbon monoxide from these
engines, equipment, and vessels if implemented. Thank you for the opportunity to comment, and
the commenter hopes that these reasonable regulations will be finalized and implemented in a
timely manner.

       T. Nixon commented in full support of this, and only wishes it could be implemented
sooner.  Much small-engine equipment is not needed for businesses or vital household needs,
they are luxury goods that contribute to sound and air pollution in recreational and residential
areas. The commenter believes that reducing emissions from watercraft will help make boating
more pleasant and enjoyable, and these regulations  should certainly help clear the air for people
in dense urban areas. If higher costs do reduce sales, it may further serve to increase the health
of the nation by having people use push-mowers instead of ride-on mowers, and rakes instead of
leaf-blowers.  Although, admittedly, increased use of snow shovels (vs.  snow blowers) could
                                           1-8

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


actually reduce health, as that can be back-breaking work.  Good luck getting these new
emissions regulations (or some version of them) implemented.

       B. Paddock commented that the regulatory analysis persuades the commenter that there
are substantial benefits to the human environment to be obtained by adoption of the proposed
rules on small engine emissions. The commenter urges EPA to advance the effective date and
deadline for implementation of the regulations. These improved products can be built with
existing knowledge and technologies as is evident from products sold in California. Low
emission mowers and other products should be made available throughout the U.S. at the earliest
possible date. The years of delay in the implementation of rule are unnecessary. As a consumer,
the commenter wants to buy emissions reduced products now. The cost of gasoline means the
commenter will recover the costs of fuel efficiency (and reduced emissions) in a shorter time.
Lawn mowing is becoming unaffordable. Coupled with drought and native plant landscaping,
the mower industry is likely in for slow period. New, fuel efficient, "greener" low emission
mowers offer a reason to buy sooner rather than wait for years for a better product. The same is
true for boat motors. The manufacturer who first makes rule compliant products available will
see a surge  of sales.

       Nautigaz commented in their first e-mail that it was for them very good news, it is it for
several reasons. Of course they hope to be able to work now, thanks to these new standards, but
they also hope to bring a solution which goes in the "direction of the History."  If all is not
perfect in the USA, they reassure in France it is exactly similar!  It is perhaps why our countries
are friendly since the Lafayette General.

       The Environmental Club of Colorado State University -  Pueblo commented that they are
overwhelming glad to see that  new regulations are going to be implemented restricting air
pollutants from lawn mowers and small boat engines. They are very glad that these new
regulations  are coming to pass. However, they believe that the dates which these regulations will
go into effect are too far out. They believe it is more than possible to lessen the period of time
until these regulations take effect. Please consider enacting these regulations much sooner, as our
Earth  desperately needs clean air as do we all.
                                       1-9

-------
                             Chapter 1: Rulemaking Process and Cross-Program Issues
Letters:
Commenter
NMMA
EMA
N. Leggett
Johnson County, Kansas
OTC
South Coast AQMD
Maryland
NJDEP
R. Keichline
A. Swanson
T. Nixon
B. Paddock
Wisconsin DNR
New York State DEC
Nautigaz
Houston-Galveston Area Council
NESCAUM
Environmental Defense
NACAA
Indmar
Bombardier
MeadWestvaco
Volvo Penta
Yamaha
Mercury
The Mid-America Regional Council (MARC) Air Quality Forum
California ARE
Pennsylvania DEP
Kohler
EVCC
Ilmor (hearing)
Pleasurecraft Marine (hearing)
Brunswick
Clean Air Watch (hearing)
Heraeus (hearing)
Inca Molded Products
Environmental Club of Colorado State University - Pueblo
Euromot
Document #
0688
0691
0603
0619
0678
0704
0722
0710
0561
0596
0599
0607
0663
0659
0727
0633
0641
0648
0651
0667
0674
0723/0724
0708
0721
0693
0696
0682
0676
0703
0608
0642
0642
0695
0642
0642
0700
0730
0649
                                       1-10

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


Our Response:

We appreciate all comments on the proposed rule; specific responses to the various concerns
raised by individual commenters are in the rest of this Summary and Analysis of Comments
document.

1.1.2   Legal authority

What Commenters Said:

       NACAA commented that emission control requirements for small nonroad spark-ignition
and marine spark-ignition engines and equipment should achieve the greatest reductions feasible
as soon as possible.  In Section 101(a)(3) of the Clean Air Act, Congress vests state and local
clean air agencies with "primary responsibility" for the control of air pollution. This is a
responsibility they take very seriously.  As NACAA seeks to achieve and sustain clean, healthful
air throughout the country, they must consider the full measure of emission reductions feasible
from every source of pollution as quickly as possible. With respect to nonroad spark-ignition
engines smaller than 50 horsepower, however, states and localities other than California, very
unfortunately, are preempted from adopting standards or other requirements. Therefore, it is
incumbent upon EPA to ensure that this rule achieves the greatest degree of reductions possible
as soon as possible.

       Environmental Defense commented that Congress is concerned about the air pollution
caused by nonroad mobile sources such as the spark ignition marine and small engines subject to
this rule, they enacted § 213 as part of the Clean Air Act amendments of 1990. See Pub. L. No.
101-549 § 222,  104  Stat. 2399, 2500-02 (1990) (codified at 42 U.S.C. § 7547). Section 213
instructs EPA to set emission standards for  nonroad engines that reflect the "greatest degree of
emission reduction achievable". 42 U.S.C. § 7547(a)(3).  For this reason, the D.C.  Circuit
consistently has held these provisions of the CAA to be "technology-forcing." Husquvarna AB v.
EPA, 254 F. 3d 195, 200 (D.C. Cir.  2001) (upholding EPA decision to give priority to goal of
reducing emissions, over cost, noise, energy and safety factors in setting Phase 2 emission
standards for handheld small engines);  see alsoNRDC v. EPA, 655 F.3d 318,  328  (D.C. Cir.
1981) (stating that the legislative history of the CAA demonstrates that Congress intended EPA
to "press for development and application of improved technology rather than be limited by that
which exists today.") Importantly, Section  213 also directs EPA to set emission standards that:
shall take effect at the earliest possible  date considering the lead time necessary to permit the
development and application of the requisite technology, giving appropriate consideration to the
cost of compliance within such period and energy and safety. 42 U.S.C. § 7547(b). Whether a
particular standard satisfies both statutory prongs of Section 213, therefore, involves a question
of stringency as well as timeliness. See e.g. Bluewater Network v. EPA, 370 F.3d  1, 21-22 (D.C.
Cir. 2004). Clearly,  if EPA were to set  standards reflecting the "greatest degree of emission
reduction achievable" based on technology  available in 2007, but not implement such standards
until much later, the actual emission reductions resulting from such rules would be minimal, at
best. Such standards, even though technology-forcing in 2007, would likely lag behind the
technological advances made in the interim years between the initial proposal  and  the ultimate
implementation date. Consistent with the statutory mandate to implement emission standards "at
                                       1-11

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
the earliest possible date", EPA must "provide a reasonable explanation of the specific analysis
and evidence upon which the Agency relied." Id. at 21. Anything short of this fails the arbitrary
and capricious test applied to EPA rulemaking under the CAA and APA. See Motor Vehicle
Mfrs. Assn. v. State Farm Mut, 463 U.S. 29, 52 (1983).

Letters:
Commenter
Environmental Defense
NACAA
Document #
0648
0651
Our Response:

       We understand these comments to be generally reinforcing the statutory provisions upon
which we based our proposed rule. See the preamble to the final rule and the rest of this
document for a description of specific issues related to the timing and feasibility of implementing
standards that we believe represent the greatest degree of emission reductions that are achievable
at this time.

1.1.3   Process concerns

What Commenters Said:

       Environmental Defense commented that most importantly, they believe EPA must act
expeditiously in publishing and implementing final rules for these SI engines. Indeed, they
believe EPA has already acted unlawfully and in violation of its statutory duties under the Clean
Air Act (CAA) and other laws due to the amount of time that has passed since Congress required
the Agency to reduce emissions from these engines.

       Houston-Galveston Area Council (H-GAC) commented that due consideration should be
given to the economic impacts of this rule, particularly on small businesses; however, H-GAC
believes that these concerns should not outweigh the more primary concerns of protecting the
public health and achieving Clean Air Act compliance. They therefore encourage the EPA to
finalize these emission standards before the end of this year at the most stringent levels that are
technically feasible to achieve the maximum environmental benefit. H-GAC encourages the EPA
to implement the standards as quickly as it is possible  to do so, but no later than 2009. This
would provide the maximum benefit to their region's air quality improvement efforts.

       EMD urges EPA not to do what it has done here - to include in a proposed rule items that
are unconnected, or only peripherally connected, to the subject of the rulemaking. The current
proposal  includes new proposed rules on certification fees and on preemption provisions that are
only marginally connected to the spark-ignition engines and equipment that are the main subject
of the proposal. The included items are not even listed in the title of the rule proposal. EPA's
practice here makes life difficult and increases expense for regulated parties. All manufacturers
must read carefully every EPA proposed rule,  including those seemingly unrelated to their
products, to make sure that EPA has not piggybacked provisions important to their markets on
them. EPA should make proposals such as the certification fees rule and the preemption rule the
                                          1-12

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


subjects of separate rulemakings, with their own entries in the Federal Register. There is
precedent for such an action; for example, the original rule proposal extending certification fees
to nonroad engines 17 stood alone, and was not piggybacked on another rule proposal.

       ASTM commented that the following standards referenced have been updated: D 471-98
is now D 471-06; D 323-99a is now D 323-06; E29-93a is now E29-06b. If EPa would like
copies of the updated standards, they would be happy to send them.

Letters:
Commenter
EMD
ASTM
Houston-Galveston Area Council
Environmental Defense
Document #
0687
0606
0633
0648
Our Response:

       The comments from Environmental Defense summarize the position they took in
litigation regarding the applicability of the original statutory deadline for completing this
rulemaking. The district court decision upheld EPA's position that the original deadlines no
longer applied once Congress adopted a separate requirement to publish a safety study related to
the safety implications of new emission standards before proposing such standards (in the docket
under EPA-HQ-OAR-2004-0008-0840).  Environmental Defense has appealed this decision. We
continue to believe that we have not acted unlawfully in completing this rulemaking. While we
have worked diligently to complete the rulemaking as quickly as possible, addressing a wide
range of technical issues required substantial time and interaction with many interested parties.
We believe the result of all of these efforts is a rule that is thorough and effective in achieving
our objectives.

       We share the perspective of the Houston-Galveston Area Council that the rule should
achieve emission reductions to protect public health and meet our statutory obligations while
taking into account the particular concerns for small businesses that must meet new
requirements. We believe the final rule achieves this balance by including far-reaching emission
standards in combination with a variety of provisions to address concerns related to compliance
burdens for small businesses. The timing of the final rule is somewhat behind the schedule
envisioned at the time of the proposal, but we are able to preserve the most important portions of
the implementation schedule described in the proposed rule.

       We understand the concerns raised by EMD, however we do not believe they are
sufficient to justify excluding the issues from this rulemaking. The inclusion of such broad
issues is this rulemaking is appropriate, especially under the new "plain-language" regulatory
construct, which allows the use of common procedures across multiple categories.  All changes
to these common regulations are made through public rulemakings such as this. In addition, we
made great efforts to reach out to all affected stakeholders in such rulemakings.
                                       1-13

-------
                              Chapter 1: Rulemaking Process and Cross-Program Issues
       ASTM has further updated its standards beyond the changes described in its comments
on the proposed rule.  We have included the latest ASTM standards for every reference and
appreciate ASTM's eager cooperation in supporting this effort.

1.1.4   Commenters referencing other commenters

What Commenters Said:

       Euromot reviewed the proposed Phase 3 regulation within their membership and they
fully support the OPE1 comments. As established manufacturers with a long compliance history
Euromat understands the new concept of EPA for imports (and exports). The changes proposed
by the OPEI comments are essential for the Euromot members to be present on the market in the
future.  Without these changes the regulation would be not practical and a dramatic burden
(financial and administrative) would be laid to the industry.

       Honda commented that they are a member of the Engine Manufacturers Association
(EMA), Outdoor Power Equipment Institute (OPEI), and the National Marine Manufacturers
Association (NMMA) and supports the comments submitted by each of these trade associations.

       Kohler is a member of the Engine Manufacturer Association (EMA) and the Outdoor
Power Equipment Institute (OPEI) and as such supports the written comments being submitted
by them.

       Arctic Cat is in full support of comments from MIC and ISMA and have no intention of
diminishing the importance of additional issues raised by either association by not including
specific comments here.

       BRP supports the comments submitted by the National Marine Manufacturer's
Association (NMMA), International Snowmobile Manufacturer's Association (ISMA), and the
Motorcycle Industry Council (MIC). In addition, BRP has individual comments on certain
aspects of this regulation which could not be addressed through these organizations. Detailed is a
summary of these comments along with additional information to support the NMMA, ISMA,
and MIC comments.

       Yamaha is supportive of EPA's proposal with the addition  of NMMA industry comments
to control evaporative and exhaust emissions from Marine SI engines and fuel systems
components.  As they hope, these comments along with those submitted with their approval of
NMMA will demonstrate the need for additional lead time to design, manufacture and implement
effective controls for exhaust and evaporative emissions.

       EMD has read, agrees with, and supports the comments to be submitted by the
Association of American Railroads. They urge EPA to act in accordance with those comments.

       Heraeus is very supportive and  a member of OPEI Fuel and Exhaust Clean Air Act
Committee. Heraeus is also a member of MECA with membership and workshop participation.
                                         1-14

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
       OTC and Maryland commented in finalizing this rule. OTC advises EPA to examine
closely the recommendations made by the Northeast States for Coordinated Air Use
Management (NESCAUM) and the National Association of Clean Air Agencies (NACAA). The
recommendations made by these organizations will improve upon EPA's proposal and offer
states much needed reduction in ozone forming precursors.

       The Mid-America Regional Council (MARC) Air Quality Forum, created in accordance
with Section 174 of the Clean Air Act to coordinate the development and implementation of air
quality policy in the bi-state Kansas City region, wishes to express its strong support for the EPA
proposal to set federal emissions standards for small spark-ignition engines. MARC also shares
some concerns raised by the National Association of Clean Air Agencies (NACAA) in its June 5,
2007, testimony to EPA.

       The NJDEP supports the comments submitted by the National Association of Clean Air
Agencies (NACAA) and the Northeast States for Coordinated Air Use Management
(NESCAUM) for the proposed rule.

       Pennsylvania DEP concurs in the technical recommendations made by the National
Association of Clean Air Agencies (NACAA).

       The comment from the Natural Resources Defense Council related to federal preemption
of state regulation was co-signed by representatives of Environmental Defense, Friends of the
Earth, and Coalition for Clean Air.

Letters:
Commenter
EMD
Arctic Cat
OTC
Maryland
NJDEP
NRDC
Bombardier
Yamaha
The Mid-America Regional Council (MARC) Air Quality Forum
Honda
Euromot
Pennsylvania DEP
Kohler
Heraeus (hearing)
Document #
0687
0709
0678
0722
0710
0690
0674
0721
0696
0705
0649
0676
0703
0642
                                      1-15

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
1.2   Scope

1.2.1   Handheld exhaust standards

What Commenters Said:

       MECA noted that EPA chose not to consider any change in the current Phase II exhaust
emission limits for Class III, IV, or V engines used typically on handheld equipment (e.g.,
chainsaws, string trimmers, hedge cutters). They commented that catalysts are already being
used in many (but not all) of these handheld equipment applications. However, engine
technology improvements continue to be made on these small spark-ignited engines to further
improve engine-out emission characteristics. MECA believes that it is time for EPA to assess
the need for further emission reductions from these smaller engines based on the application of
advanced engine technologies with properly engineered and cost effective exhaust emission
controls like catalyzed mufflers.

       NESCAUM also noted that EPA has declined to establish more stringent exhaust
emissions standards for handheld equipment beyond the Phase II  standards  adopted in 2000.  The
Phase II standards were affirmed by EPA in 2004, based on a technology review, with the final
standards taking effect in 2007 for all handheld engine classes. According to the technology
review, EPA determined that handheld engines would meet the exhaust emissions standards on
schedule, mostly by modifying two-stroke  designs to incorporate  stratified scavenging with lean
combustion, with or without catalytic aftertreatment. Accordingly, NESCAUM fails to see why
HC+NOx exhaust emissions standards for  Class V handheld engines should remain 44 percent
higher than the standards for smaller handheld engines. Their concern is heightened under this
proposed rulemaking because, in effect, the Class V engine category will be expanded to
incorporate all Class I engines with cylinder displacements less than 80 cc,  regardless of whether
these engines are used in handheld or nonhandheld applications. While NESCAUM does not
object to treating these smaller Class I engines in all respects as Class V engines, they urged EPA
to revisit and  strengthen the Class V exhaust emissions standards through this rulemaking.

       NESCAUM commented that at the time of EPA's technology review in 2004,
manufacturers were concentrating their Phase II development efforts on Class IV and smaller
displacement engines because these standards were to take effect two years ahead of the Class V
engine standards. The speculative concerns regarding technology transfer, safety, performance,
weight, and other factors affecting Class V engines were primarily due to the fact that
manufacturers had not begun to focus their attention on this particular engine category. EPA's
subsequent Technical Study, while confined to larger Class I and  to Class II engines, has since
established that catalysts can be effectively incorporated into larger engine designs and function
without causing some of the problems envisioned by the manufacturers. In addition,
NESCAUM noted that at least one equipment manufacturer, Stihl, already has a line of
professional grade chainsaws on the market that uses the smaller Class I (soon to be Class V)
engines, incorporating stratified scavenging technology and/or catalytic converters to meet
emissions standards. They see no basis for allowing Class V engines to certify to the most
lenient HC+NOx exhaust standards among small SI engines and therefore urge EPA to adopt
more stringent standards. At a minimum, NESCAUM commented that Class V engine standards
                                          1-16

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


should be aligned with those for smaller engines. In addition, they urged EPA to update its
technology review of exhaust emissions standards for Class III and IV engines, and as necessary,
adopt more stringent standards through subsequent rulemaking.

       NY DEC noted that EPA's proposal contains no new standards for engines in handheld
equipment, where the proximity of engine exhaust to the operator's breathing space leads to
individual toxic exposure concerns in addition to ambient air quality concerns.  They commented
that additional effort is necessary to identify technology supporting more stringent handheld
engine regulations. They urged EPA to continue its efforts in this field, and build on its current
success, ultimately promulgating more  stringent regulations for all of these classes of nonroad
engines.

Letters:
Commenter
MECA
NESCAUM
NY DEC
Document #
0668
0641
0659
Our Response:

       In response to the comments that EPA should adopt more stringent emission standards
for handheld engines, EPA believes the current Phase 2 standards are the appropriate standards
for handheld engines at the current time. In fact, the Phase 2 standards for handheld engines are
not yet fully phased-in. For Class III and IV engines, the Phase 2 standards became fully
effective, including small volume engine families, in 2007. For Class V engines, the Phase 2
standards became effective in 2007 except for small volume engine families.  However, small
volume engine families in Class V, which include about half of the Class V engine families, have
until 2010 to comply with the Phase 2 standard. Therefore, most of the handheld engine
manufacturers are still in the process of redesigning the remaining small volume engine families
to meet the Phase 2 standards.

       An analysis of the certification data for 2008 model year handheld engines shows that the
standards have resulted in widespread use of catalysts on Class III and IV engines,  as expected in
the final rule. Based on sales estimates, approximately 70% of Class III and IV engines are
using catalysts to demonstrate compliance with the Phase 2 standards.  (The remaining engines
are either 4-stroke engines which can meet the standards without a catalyst, or 2-stroke engines
without a catalyst many of which are certified through the use of ABT credits and emit at levels
higher than the Phase 2 standard.)  In addition, the 2008 certification data shows that few of the
Class V engines are using catalysts to demonstrate compliance with the Phase 2 standards,  as
expected in the final rule.  (EPA's certification data can be found on the internet at the following
address:  http://www.epa.gov/otaq/certdata.htnrfsmallsi)  As detailed in the development of the
Phase 2 standards, EPA set the Class V standard at 72 g/kW-hr in response to concerns over heat
issues related to the use of catalysts since much of the Class V equipment is used in chainsaws
where compact packaging requirements make it hard to design the engine with the increased
cooling needed with a catalyst (see 65 FR 24269, April 25, 2000). In response to the comment
that the recent testing of catalyst-equipped engines by EPA shows catalysts could be
                                       1-17

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
incorporated into larger engine designs and function "without causing some of the problems
envisioned by the manufacturers," EPA notes that the testing was performed on Class I and II
nonhandheld engines which are 4-stroke engines.  The results of that testing are not applicable to
Class V handheld engines which are 2-stroke engines and completely different types of
equipment applications offering their own issues with catalyst design and engine cooling.

       In conclusion, given that the Phase 2 standards are not fully effective, and given the
technologies being used to comply as demonstrated in the most current certification data, EPA
believes the Phase 2 emission standards for handheld engines are the appropriate standards for
handheld engines at this time. This should not be interpreted to mean that EPA will not revisit
the standards for handheld engines in the future. Indeed, under section 213(a) of the Clean Air
Act, EPA is required to "promulgate (and from time to time revise) standards" for nonroad
engines.

1.2.2   Hobby engines

What Commenters Said:

       N. Leggett (0603) commented that independent inventors and experimenters should be
encouraged because they develop inventions that are different from the technologies developed
by large corporations. There are hobby engines used for radio control models that are larger than
50 cubic centimeters  in per-cylinder displacement. Presumably these larger engines would be
allowed by the statement in the rules that: "Hobby engines are compression-ignition engines with
a per-cylinder displacement of less than 50 cubic centimeters or spark-ignition engines installed
in reduced-scale models of vehicles that are not capable of transporting a person." The
commenter's interpretation of this definition is that engines larger than 50 cubic centimeters per-
cylinder displacement are considered hobby engines if they are installed in a model vehicle.  Is
this interpretation of this statement correct? If it is not correct, the statement should be modified
to make it clear that a small engine in a model vehicle is not covered by these regulations. It is
important to keep the model scale engines as a free area for the development and use of engines.
This maintains an activity where engine designers and inventors are free to make their own
engines without mandated design features. Some of these people use desk top machine shops
such as the Sherline miniature lathe and milling machine systems (Reference 1). More people are
getting involved with their own machining using computer-controlled systems. Still others are
using full-scale metal working lathes and milling machines to build their engines.  This
developmental freedom is a contrast to the regulated engine world where people are blocked
from "tampering" with engine features. Indeed, the very process of invention and creative
technology design is  the basically playful activity of tampering with and departing from
conventional engine design. The commenter stated that we need free spaces for invention to
maintain America's position in technology and manufacturing. These engine experimenters also
develop precious industrial skills.

       N. Leggett (0612) commented that the proposed exemption for reduced scale hobby
engines is an excellent idea. Designing and building hobby  engines is an excellent way to
experiment with engine technology and to develop new engine inventions.
                                          1-18

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       L. White offered concern that this proposed regulation would make illegal the production
of hobby kits for the construction of a small engine by an individual or student unless the
manufacturers submit to and pass the requisite emission tests. The necessity to get such an
engine design or personally built kit approved under these regulations  would make hobby and
educational engine building difficult at best.  Putting emission controls on these engines
sufficient to meet these standards is inappropriate for their intended historical, educational, and
hobby recreational purpose - especially when these engines are replicas of earlier commercial
designs that did not have these controls and when the  activity is primarily for hobby,
experimental, or personal educational use. The expense of certifying such engines and kits built
for personal educational, hobby, or recreational use would put an end to this hobby and this
business. The extra weight and complexity of certified engines would also most likely make
them unsuitable for a number of hobby uses including powering RC model aircraft and small
model water craft. These engines are produced in very limited numbers, are generally of low
KW output, and see a very limited number of hours of annual use. The total fuel burn and the
total emission output are simply not worth regulating. The commenter proposes that a very
specific exclusion be incorporated in the proposed regulations such that these limited output and
very occasional use engines and kits and the manufacturers or builders of them be exempted
from regulations regarding emissions.

Letters:
Commenter
N. Leggett
N. Leggett
L. White
Document #
0603
0612
0620
Our Response:

       The regulations as proposed (and currently existing) would apply the limit of 50 cc per
cylinder for hobby engines only for compression-ignition models.  This was intended to be a
threshold below which there would likely be no commercial application other than for hobby
vehicles. Given the confusion illustrated by the comment, we believe it is important to adopt a
uniform set of definitions and requirements for all hobby engines and vehicles. We are therefore
extending the proposed definition for spark-ignition engines to apply equally to compression-
ignition engines. Thus, any engine that is installed in a reduced-scale model of a vehicle that is
not capable of transporting a person is exempt from emission standards. This change would also
mean that compression-ignition engines smaller than 50 cc per cylinder that are installed in other
applications would no longer be exempt. However, we are not aware of any such practice today.

       While we agree that there is value in allowing for innovation, development, and training
with hobby engines that outweighs the potential environmental effects, we need to draw clear
lines to prevent widespread use of an exemption to produce engines that could be used for other
purposes. In particular, we see no  need to expand the hobby-engine exemption to include larger
engines or engines that are not used in reduced-scale models of vehicles for exploring innovation
or for education. The Clean Air Act and the regulations contemplate the need for such
innovation and development with the testing exemption (§1068.210) and the manufacturer-
owned exemption (§1068.215).
                                       1-19

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
1.3  Cross-cutting issues

1.3.1   Import-specific information for certification

What Commenters Said:

       California ARB supports EPA's "Special Provisions for Compliance Assurance," and
specifically supports the provisions regarding importation data, the assurance of warranty
coverage, and bond requirements.

       Mercury commented that the proposal contains a number of new requirements in §
1045.205 for the content of the certification application. In particular, EPA is proposing to
require engine manufacturers to include additional information for imported engines in §
1045.205(z). Mercury Marine fully endorses these changes and supports EPA's efforts to
enforce these rules on imported engines.

       OPEI commented that it is difficult to keep intended ports of entry updated in the
certification applications, particularly if external shipping firms are used (like DHL etc for air
shipments).  Customs exist at all ports of entry in the US so this requirement seems unjustified.
EPA/Customs keeps a list of all US ports of entry available.  All ports of entry should be
monitored by US customs and the EPA and they should reinforce the need for proper importation
paperwork being submitted and checked to verify compliance via the EPA database. The
requirement should be deleted.
       EMA commented that the specific port that a manufacturer will use to import engines
changes from shipment to shipment.  For this reason, the requirement to identify the port(s)
where a manufacturer will import engines (see §90.107(d)(15)(i) Application for certification)
cannot be maintained in a certification document.  If required to identify a port in the application,
the manufacturer would have to submit a list of all potential ports where it may import engines.
Such a list would not provide the sought-after information and would fail to provide the intended
benefits associated with having this information.  Therefore, this section should be deleted.
       EMA commented that the NPRM does not clearly identify who is required to name an
agent for acceptance of service on behalf of a manufacturer. While it appears that such a
requirement applies only to manufacturers that do not have a U.S. presence capable of accepting
service, the final rule should clearly state that such a requirement applies only to entities without
a U.S. presence. EMA commented that this can be accomplished by moving section 1054.205(z)
so that it is included under (aa). However, if the requirement to name an agent for service
remains as a separate requirement under subsection (z), the section should be revised to read as
follows: "If you do not have a physical office in the United States with employees capable of
being served then you must name an agent for service in the United States. Service on this agent
                                          1-20

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       OPEI commented that EPA requires the location of test facilities in the US where the
manufacturer can test engines if EPA selects them for an SEA.  OPEI feels this is acceptable for
manufacturers that do not have an established presence and compliance history (minimum of 5-
years) in the US. For those manufacturers that have an established compliance history, it should
not be required that all families can be tested in the US. Handheld engines are very specialized
and larger engines are hard to test. This section should be revised to indicate options as follows:

   1) Conduct the test in the lab where the certification is conducted, or
   2) Guarantee EPA access into any of your test facilities anywhere in the world, or
   3) EPA can select a lab to have engines tested in the US (if 1 and 2 are not possible).

       NMMA commented that the proposal contains a number of new requirements in §
1045.205 for the content of the certification application. In particular, EPA is proposing to
require engine manufacturers to include additional information for imported engines in §
1045.205(z). While NMMA understands that some of this information may be helpful to track
engines and avoid problems with  foreign companies that are importing and selling engines in the
U.S. that may not be covered by valid certificates of conformity, some of these requirements are
overly burdensome. Notably, the requirement to list the location of test facilities where engines
can be tested may not be available at the time the application is submitted. Several Marine SI
engine manufacturers test their engines outside of the U.S. and may not have identified and/or
contracted with test labs in the U.S. In addition, there are not that many test facilities in the U.S.
that can perform  marine engine testing. This is particularly the case with OB engines. To address
this problem, NMMA recommends that § 1045.205(z)(3) be revised to state instead:
            Tthe location of test facilities in the United States where you can test your engines
if we select them for testing under a selective enforcement audit, as specified in 40 CFR part
1068, subpart E." NMMA is supportive of EPA's efforts to eliminate copy and noncompliant
engines from the market and this revision will ensure that the pertinent port and agent
information is provided but will not mandate that the manufacturer contracts with a test facility
prior to submitting the application.

       Suzuki commented that EPA is proposing to require companies that import engines into
the United States to identify test locations in the United States that would be used if the Agency
requires testing under a selective enforcement audit (SEA). Suzuki is concerned that it will be
overly burdensome to specify a test facility well in advance of any actual testing if the importer
does not have an existing business relationship with the US-based test facility. Suzuki
understands that  there are numerous entities certifying and importing outboard engines into the
US, and that there can be difficulties with ensuring that the importer will be held  accountable
should a noncompliance issue exist. EPA's proposal will help to address this concern; however,
Suzuki believes it to be overly broad. Suzuki recommends that EPA revise the proposal to
consider relevant factors when determining if an importer must declare a test facility in advance
of a SEA test order. Suzuki believes that relevant factors could include the length of time an
importer or distributor has been certifying engines in the US and/or the certification history of
the importer (this could include previous SEA test history, in-use test history, responsiveness to
prior Agency information requests, etc .) Alternatively, EPA could allow SEA testing to be
conducted at the  manufacturer's own testing location if the location was deemed appropriate by
EPA.
                                       1-21

-------
                                Chapter 1: Rulemaking Process and Cross-Program Issues
       Honda commented that at the time of certification of both Small SI and Marine SI
engines, it may not be possible to name a test facility in the United States that will be a viable
option for testing when at some future date EPA requests a manufacturer to perform testing.  For
some engines that are manufactured or are similar to engines manufactured in the United States,
this up-front designation may be reasonable. However, for engines requiring an outside (third
party) test facility, it would be more appropriate to choose that test facility at the time testing is
actually requested. Alternatively, a manufacturer could ensure that EPA representatives have
full and open access to existing test facilities located outside the United States.

       Manufacturers described that it would be burdensome to name a test lab, because they
would need to make extensive preparations and do round-robin testing periodically to ensure that
the named test lab would properly test engines.

Letters:
Commenter
NMMA
Mercury
Suzuki
California ARE
Honda
OPEI
EMA
NMMA
Document #
0688
0693
0698
0682
0705
0675
0691
0739
Our Response:

       It can be very useful for us to understand a manufacturer's general practice for importing
engines into the United States. Knowing which ports a manufacturer uses would help us target
certain engines  for more careful monitoring or inspection if there were a cause for concern or a
need to identify engines at the port for a selective enforcement audit. We acknowledge the
manufacturers'  concern that they should not be limited to any particular ports identified in the
application for certification, given the dynamic nature of shipping engines and equipment. To
address this concern without simply deleting the provision, we are modifying the regulatory
language to specify that importing manufacturers must identify the ports through which they
have imported certified products in the previous  12 months.  This limits the submission to
factual, historical information that is readily available.

       It is not  clear why EMA would object to naming an agent for service in the United States.
We have a need to know whom to contact if there is a need for official communication,
regardless of the location of the company.  It is especially important to require this for companies
based outside of the United States to ensure that there is a point of contact.  It is self-evident that
companies located within the United States can be contacted, but there is still a strong advantage
to knowing before  an issue arises that there is an established  contact person to handle official
communications.  This could be as simple as identifying the person submitting the application
for certification as  the agent for service. As such, we would understand that there is no burden
associated with this requirement and are adopting it as proposed for all companies.
                                           1-22

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
       Independent of the proposal requiring manufacturers to name a test lab in the United
States, the regulations state that we may specify any test lab for measuring emissions from
certified engines (§1068.401).  As long as the emission measurements conform to the specified
procedures in 40 CFR part 1065 and in the standard-setting part, the results would be considered
valid for determining whether or not the engines meet emission standards. As a result, the
provision to name a test lab in the United States does not expand the manufacturers' liability but
rather gives the manufacturer the opportunity to plan ahead of time to identify a lab where
arrangements can be made to ensure that the testing will be done properly.

       We believe it is important to preserve the proposed requirement to name a test lab in the
United States.  This would allow us to promptly pursue a selective enforcement audit for
imported engines where we find that to be necessary or appropriate. Under selective
enforcement audits, manufacturers test freshly manufactured engines, at their own expense, to
determine whether they meet applicable emission standards. It may not be practical to pursue
testing if the engines need to be shipped back to the country of origin. This is especially true in
countries where EPA agents would not necessarily  be able to freely travel or perform official
functions. Testing in the United States also allows  us to require  statements and submission of
information where U.S. laws apply, including the requirement to submit truthful information to
the government (with the  corresponding civil and criminal penalties for violations). This
requirement removes an inherent advantage for imported engines, since there will always be a
U.S. lab available for testing domestically produced engines. It also serves as a preventive
measure by forcing manufacturers to recognize that they are liable for the compliance of their
engines even after they have been sold.

       The manufacturers raised several specific objections, none of which address the
fundamental issues described above. First, it would be impractical to require manufacturers to
name a test lab "upon request" at the point of importation. Manufacturers have already stated
that they don't want to test engines in the United States, so this request would likely be met with
resistance and delay. The resistance would be greatest in cases where domestic testing is most
needed. As described above, we could in any case  pick any test  lab without the manufacturer's
direction, so we believe it is in the manufacturer's best interest to name the test lab  at
certification.

       Second, manufacturers could provide statements regarding their commitment to ensure
access to test labs located in other countries, but that is not always reliable. Manufacturers
would  offer such assurances at certification whether or not they intended to cooperate, or
whether or not local government officials would cooperate.  The burden would be on EPA to
identify engines for testing, then possibly find that the manufacturer is not willing or able to
follow through on its commitment.  This would  also put EPA in  an awkward position, which
would  likely again result in tests being run at a test  lab in the United States where no prior
arrangements had been made. This outcome would not be in the best interest of EPA or the
manufacturer.

       Third, we believe  any published measure of good compliance history is not  appropriate
for excluding a manufacturer from the responsibility to name a test lab in the United States.
                                       1-23

-------
                                Chapter 1: Rulemaking Process and Cross-Program Issues
However, we may take this into account in deciding whether to allow the manufacturer to
conduct testing in the country of origin or not. The testing in question may be what we need to
establish whether or not a manufacturer has been producing noncompliant engines. Also, with
the very large number of certifying manufacturers and emission families, it is easy to imagine
that a manufacturer could be in violation for a considerable period without being caught.
Creating this exception would inappropriately reward companies that are able to avoid detection
of violations.

       Fourth, we believe there are labs available for testing almost all kinds of engines,
including large and small engines, and all sizes of outboard engines. To the extent that
manufacturers depend on special test procedures or specialized test equipment, we would
cooperate with the  manufacturer to ensure that testing can be done properly. However, we have
modified the original proposal to create two exceptions.  Manufacturers are generally not
required to name a test lab for engines rated over 560 kW.  These engines are much more
expensive and are sold in much smaller volumes, so any effort to test these engines would
necessarily involve considerably more effort to make those arrangements. For engines above a
certain size,  there are also very few if any locations available for testing.  Also,  manufacturers of
Small SI engines may omit naming a test lab for engine families where testing depends on
custom test fixtures that are not available without making special arrangements. This allowance
is limited to  engine families representing less than five percent of a manufacturer's total U.S.-
directed production volume of Small SI engines. While we are waiving the requirement to name
test labs for these special cases, we may still require manufacturers to do selective enforcement
auditing with these engines by testing them in the United States to the extent that is possible, but
we are not requiring the companies to prepare for that by making these arrangements ahead of
time.

       We understand that manufacturers would be well served to invest some effort in
coordinating with the named test lab to ensure proper testing.  On the other hand, the fact that
manufacturers are concerned that another lab may get different results reinforces our concern
that this provision is necessary. Testing of certified engines should show that the  engines meet
emission standards for any valid test, regardless of the test location.

       Manufacturers might also name multiple test labs if they have made arrangements with
different companies that perform such testing. If manufacturers are confident that a valid test at
any facility will show that their engines comply and they have no relationship with testing
organizations in the United States, they might also indicate in the application for certification
that all test labs in the United States are acceptable for confirmatory testing.

       When we select imported engines for testing, we expect to work with the manufacturer to
make the necessary arrangements. We would generally plan to test engines in the United States.
However, in certain circumstances we may agree to allow testing in the country of origin if we
have reason  to believe that the testing will be properly performed and that we will have
unrestricted  access to the foreign test facility.
                                          1-24

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       We are therefore adopting the proposed requirement to name a test lab in the United
States, with the modifications noted above, as supported by Mercury Marine and California
ARE.

1.3.2   Date of manufacture on label

What Commenters Said:

       ISMA commented that they do not support the recommended change to § 1051.135(c)(6).
EPA has proposed to remove the flexibility of keeping records of the manufacture date in lieu of
printing the month and year of manufacture on the label.  This change results in a large burden
on the manufacturer with no benefit to the environment. Each vehicle must have its certified
engine family name  on the emission control information label. The vehicle or engine's model
year is clearly indicated on this label through the first character.  Since the EPA snowmobile
regulation is based on model year, not calendar year, placing the month and date of manufacture
on the label does not provide meaningful information. ISMA manufacturers will continue to
maintain records of build dates for their vehicles,  and respectfully requests EPA maintain the
current language in 40 CFR 1051.135(c)(6).

       MIC commented that § 1051.135(c)(6) allows omitting date of manufacture from the
label only if the date is stamped on the engine/vehicle. Stamping each vehicle or engine with the
build date is burdensome and unnecessary. The rationale for this change is that it is needed for
verifying that vehicles comply with standards based on their build date. However,  it is also
required that the label "state the exhaust emission standards or FELs to which the vehicles are
certified (in g/km or g/kW-hr)." Given this requirement, the build date is unnecessary.

       Arctic Cat submitted a comment regarding §1051.135(c)(6).  The new proposal will
require that the date of manufacture  is included on the Vehicle Emission Control Information
(VECI) label.  Arctic Cat requests an exemption from this requirement for replacement labels
(usually needed when a replacement tunnel is provided to a snowmobile customer). These
replacements number about one hundred per year to service all past production. Since this affects
used units that have  already cleared customs and are no longer in a dealership there is little value
in making the extra effort to ensure a correctly dated VECI label is supplied. Arctic Cat proposes
including the phrase "replacement label" in the field for the date (shortened if necessary to fit in
field). The difficulty of including the date on a replacement VECI label is that it is too resource-
intensive to create the label and manage the logistics to make sure this exact label is supplied to
the customer.

       Honda  also commented on the pending regulatory changes to §1051.135(c)(6). Honda
has not had sufficient time to evaluate the impacts and requirements this proposal will have on its
production efforts.  Honda requests EPA's approval to not include the production date for the
2009 model year.

       Honda  continued that the current regulations allow them the opportunity to not print the
date of manufacture on the VECI labels as long as they stamp the date on the vehicle or maintain
records and provide  them to EPA upon request. As EPA is aware, the proposed regulation does
                                       1-25

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
not include the option to maintain records and provide them to EPA upon request. Honda would
prefer to maintain records of the manufacturing dates, as they are currently doing, and provide
them to EPA upon request. Honda questioned whether there was a particular reason why this
option was no longer being provided.  With the issuance of the Final Rule being delayed so close
to the start of the 2009 model year, Honda noted that several months will be required to prepare
and implement the printing of the production date on the VECI labels. This can have a
significant impact on their early model year production.

       Suzuki commented that as a motorcycle manufacturer Suzuki is concerned about the
labeling requirement proposed in § 1051.135(c), as this  is a totally new requirement. They
supported the MIC comments on this issue, which reflected their position. Additionally, Suzuki
questioned whether they can assume that a MY 2010 effective date will be applied to whatever
revisions to part 1051 are finalized.

       MIC reiterated its concerns with EPA's proposed changes to the Recreational Vehicle
regulations at 40 CFR part 1051.135(c)(6). The current regulation allows the date of manufacture
to be omitted if a manufacturer keeps a record of the date and provides these records to EPA
upon request. This change is of concern to manufacturers because the increased burden is
significant. MIC argued that there is no benefit from requiring this of mainstream established
manufacturers. This change appears to be targeted at EPA's oversight of importation of
nonmainstream engines and not control of established manufacturers, therefore a more
appropriate approach should be found to address the problem without overly burdening
compliant manufacturers. MIC wants to reiterate that this requirement change will impose
significant problems for the following reasons (but not limited to):
    - Space allocation on engine emission labels is very limited; on many smaller engines
    emission label space is extremely limited
    - Emission labels are often not printed at the factory and therefore it is not possible to create
    a unique emission label reflecting the manufacture date
    - Even if it is possible to incorporate the date on the emission label, sufficient lead time is
    not provided in the regulation change to provide for 100% assurance of labeling accuracy
    - Many engines do not  have space for incorporating an engraved date and would require
    revised casting (extremely expensive)
    - Even if space is available on the engine for engraving the date, plants are not equipped for
    performing this engraving operation (expensive equipment and production line changes are
    required)

       In addition to the above implementation issues, MIC stated that recreational vehicle
engines are not considered complete until they are assembled into the vehicle. If the
manufacturer identifies the date of manufacture on the vehicle and the engine has a separate label
stating a different date of manufacture; unnecessary confusion (not to mention redundancy and
cost) will result. MIC believes that the current approach of requiring the manufacturer to have  a
readily available tracking method for the  assembly of partially complete engines is sufficient and
that an additional label on the engine adds no benefit.

       MIC is interested in  continuing to work with EPA  on developing a method for addressing
whatever the basis is for making this change in the regulation but does not agree that the current
                                          1-26

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


approach is acceptable. MIC requests that at the very least EPA should provide additional lead
time for the above issues to be properly addressed in the production process and that
implementation should be no earlier than model year 2012.

       In later comments, MIC emphasized that this was an important issue for them.  They
stated that they understood EPA's rationale for the requirement to identify the engine build date
on the engine or vehicle, but pointed out that the implementation date for this  requirement
prohibits the use of labels that have already been printed by the manufacturer. The MIC
proposed language below would allow for the engine build date to be printed on a supplemental
label that is affixed adjacent to the main emission control information label:
    § 1051.135(c)(6) State the date of manufacture [MONTH and YEAR]; however you may
    omit this from the label if you stamp or engrave it on the engine or vehicle, or if you provide
    this information on a supplemental label. If you use a supplemental label, it must be visible
    when viewing the primary emission control information label and comply with the
    placement, durability and legibility requirements as described in this part.

Letters:
Commenter
ISMA
MIC
Arctic Cat
Honda
Suzuki
Document #
0671
0701
0709
0736
0732
Our Response:

       Under the Clean Air Act, engine certification is based on annual production schedules (or
model years) where a manufacturer produces each engine during a production period such that it
is covered by a valid certificate of conformity.  Identifying an engine's build date establishes
clearly for each engine whether it is covered by a certificate of conformity for any given model
year. Properly associating each engine with the appropriate model year is important for
identifying applicable  emission standards, calculating emission credits (where applicable),
tracking emission-related defects, and executing a recall, among other things. We are adopting
regulatory provisions in this rule that further clarify the concept of build date, model year, and
the effective dates of certificates of conformity in 40 CFR part 1068 for all nonroad engine
categories and in the standard-setting parts. We believe each engine should be clearly
identifiable with a certain model year based on its build date. Having this information recorded
on the engine prevents a situation in which a manufacturer could manipulate records as needed to
gain a more favorable  outcome depending on the reported build date of any particular engine or
engines.  For example, we may find a collection of engines in violation and would want to
establish whether they are from the same model year or not, or whether they were built before or
after a given change in the application for certification.  Our experience has shown that it is very
difficult to contest a manufacturer's claimed build date, even when it defies any customary
business  or manufacturing process.  We also find it inappropriate generally to have to depend on
manufacturers to provide information that is necessary to determine whether that manufacturer
has committed a violation.
                                       1-27

-------
                                Chapter 1: Rulemaking Process and Cross-Program Issues
       A further practical constraint comes from engine inspections, especially at importation
where U.S. Customs and Border Patrol agents have limited time to evaluate large quantities of
very diverse products. Inspection of engines often depends on knowing an engine's build date to
establish which tier of emission standards apply. A straightforward inspection of an engine
should allow an inspector to determine the applicable standards.

       Having the build date on an engine would also provide a valuable piece of information
because the manufacturer makes a commitment in the assembly process by printing a specific
date on the engine (generally month and year).  This information is necessary for us to be able to
evaluate whether an engine was produced before or after the effective date of a certificate of
conformity. The printed build date information is unalterable, which is very effective for both
compliance assurance (or prevention of noncompliance) and enforcement.  For example,
manufacturers would be very reticent to put a false date (such as a postdate) on an engine if there
was a possibility that someone may inspect that engine shortly after the manufacturer introduces
it into commerce and where it would be directly evident that the date is in error. Likewise, if the
printed date is substantially earlier than the actual production date, it may be possible to inspect
associated records to evaluate the validity of the printed  date (production records by serial
number, build dates of equipment in which the engine is installed, invoices, bills of lading, etc.).
Having the ability to demonstrate that an engine was produced after emission standards started to
apply is essential both for our benefit to ensure compliance, and for the manufacturer's benefit to
prove compliance.

       Furthermore, where there is a compliance problem, it may be easier to demonstrate that a
false build  date is a violation than that the engine exceeds emission standards.  Requiring build
dates on labels requires that manufacturers make a statement to the government, where penalties
may apply  if the information is demonstrated to be false.

       We allow for applying the label and identifying the date of manufacture at any point in
the assembly process. Manufacturers could use pre-printed labels that are punched to identify
month  and  year, or the label printing could be brought in as part of the assembly process.
Manufacturers may also identify the date of manufacture elsewhere on the engine, such as on a
different label applied for other purposes. In any case, the manufacturer could take steps to
avoid mismatched dates on different labels if that is a priority  objective.

       We acknowledge the concern for labels on replacement components. We understand this
to be a relatively rare occurrence and agree that it would be rather impractical to include build
dates on these replacement components without a disproportionate effort.  We are including this
exception in the new language related to replacement labels in §1068.101(b)(7).

       We are adopting the proposed requirement to print build dates on the label. We
understand that this will involve a change in labeling practices for some companies. On the
other hand, manufacturers of Small SI and Marine  SI engines  have already been doing this for
several years so the feasibility of identifying engines this way is well established. This is part of
a broader effort to  adopt this requirement across engine categories.  We intend to further
standardize labeling with further specification related to the format of the build date.  For
                                          1-28

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


example, we believe it is not appropriate to identify the date with coded alphanumeric characters
intended to disguise the information from anyone who is not privy to the coded meaning.
Spelling out the full date is clearly acceptable. We would also consider acceptable certain
standard abbreviations, such as Sep 10 or 09/10 to indicate September 2010. We plan to propose
detailed specifications in a future rulemaking to describe a range of acceptable ways to identify
an engine's build date.

       We agree with MIC's concern that multiple dates on a single engine or vehicle would be
confusing  and unnecessary.  We have therefore modified the labeling requirement to say that
permanently applying the date elsewhere (not just by stamping or engraving on the engine)
would be acceptable for meeting specifications.  Given that manufacturers of recreational
vehicles are already putting  build dates on their vehicles, we believe there is no need to delay this
requirement beyond the 2010 model year.

1.3.3   General labeling provisions

What Commenters Said:

       EMA commented on regulatory language we introduced in a new §1068.45 to lay out
general  labeling requirements.  These comments included the following suggestions:
   •   Using the term "removable label" throughout part 1068 in place of the term "temporary
       label."
   •   Including an example to illustrate that a removable label for replacement engines should
       remain in place until the exemption no longer applies.
   •   Allowing hang tags to qualify as removable labels since they can be made durable
       enough to stay in place until they are removed.
   •   Referring to §1068.101  in §1068.45(e) in regard to improper removal of labels is vague
       and should be removed.

       EMA said the provisions in §1068.101(b)(7) regarding label removal should include an
allowance for removing and replacing labels that are incorrect, whether they were wrong initially
or they were rendered inaccurate by an engine modification.

       In addition, EMA commented that the regulations in several instances specify that labels
for exempted engines include information related to engine displacement and rated power. They
argued that this information should not be required.

Letters:
Commenter
EMA
EMA
Document #
0808
0810
Our Response:

       We believe it is quite appropriate to point readers to the provisions in §1068.101 that
describe provisions related to label removal.  This does not change any of the regulatory
                                       1-29

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
requirements, but we believe readers may not be aware of those provisions without a specific
reference. We agree with EMA's other suggested changes to §1068.45 and have incorporated
those into the regulations.

       We agree with EMA's suggestion to allow for removal  of incorrect labels. A single
statement can cover both of the scenarios EMA highlights; a label is incorrect anytime it is no
longer true, whether that was always the case or the engine's original, correct label is no longer
accurate. We are also including language to clarify that the allowance to remove and replace
labels does not change the fact that applying an original label that is incorrect may be a violation
of the prohibited acts.  We simply intend to allow for the manufacturer to rectify incorrect labels
and would separately consider the original action of applying a false label.

       In almost all cases, we want labels on exempted engines to identify the engine's
displacement.  This is important identifying information that helps to prevent a situation in which
an exempt label is applied to the wrong engine. Identifying the displacement helps the engine
manufacturer and anyone  inspecting the engine to know that the labeled engine is properly
covered by the exemption in question. We agree that rated power is generally  not needed.
However, in certain cases, we might approve an exemption only for certain power ratings for a
specific engine model. For those few exemptions where we might want manufacturers to
identify an engine's rated  power, we have modified the regulation to allow us to require that if it
is needed. In no cases do  we have a default requirement to identify the rated power for exempted
engines.

1.3.4  Special provisions for production-line testing

What Commenters Said:

       MIC commented on §1051.301(a)(2)  of the proposed regulations. MIC commented that
the exemption from PLT for small volume families  should not be left to the discretion of EPA
staff. The proposed language  states that the exemption "may" be provided. The language of
§1051.301(a)(2) should be revised to say "Engine families with a projected U.S.- directed
production volume below 150 units are exempt from testing under this subpart."

       ECO commented that EPA should allow small-volume engine manufacturers to utilize
the use of alternative testing methods (portable emissions analyzers) to demonstrate in-use field
testing compliance for production units.

Letters:
Commenter
MIC
ECO
Document #
0701
0712
Our Response:

       We continue to believe it is appropriate for manufacturers to be required to request a PLT
exemption for small-volume engine families in the application for certification. For example, we
                                          1-30

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


are concerned that manufacturers may attempt to gain an advantage by underestimating projected
sales. The request process should be fairly simple, since manufacturers are required to certify
each of their engine families and they must submit a sales estimate. We would grant or deny the
exemption request as part of the certification approval process.  We have added language to the
regulations in each of the applicable standard-setting parts to clarify that we would approve the
request if we  agreed that the projected sales volumes were made in good faith.

We agree with ECO's suggestion to clarify the language related to alternative methods for
production-line testing. The original language was intended to  allow for manufacturers to
develop different ways of testing production engines for proper quality assurance with respect to
emission controls.  The initial thinking was that a simpler test (such as ppm testing at multiple
modal points) on a large number of engines could be more effective at screening production
engines than a rigorous (certification-quality) test on a small number of engines. We continue to
believe there  is a good potential for this type of alternative test program. The specific suggestion
to allow the use of field-grade measurement equipment for production-line testing is an
appropriate additional alternative. We have modified the regulation for all spark-ignition
engines to allow for using field-grade measurement equipment, provided that the manufacturer
doubles the minimum sampling rate.  Much like the ppm testing described above, the somewhat
less precise or accurate test methods should provide an equivalent compliance demonstration by
expanding the sampling rate. We would expect portable analyzers to reduce the cost of testing
enough to more than offset the burden associated with testing additional engines. Note that
testing with portable analyzers that meet lab-grade specifications would not be  considered an
alternative test method and would therefore not be subject to EPA approval and would not
trigger the need to increase the  sampling rate.
1.3.5   Reporting and recordkeeping requirements

What Commenters Said:

       Kohler Co. is very concerned with the record keeping and reporting burden associated
with the proposed regulation. Table XIV-1 in the proposal lists the average burden for a Small SI
engine manufacturer at 885 hours annually. This is reported to be the total estimate for both new
and existing reporting requirements for total time required to "generate, maintain, retain, or
disclose or provide information to or for a Federal agency." This equates to less than half time
for a person working 40 hours per week. Kohler knows this estimate is grossly understated and
that they will need to add additional staff to deal with all of the record keeping, reporting,
correspondence with customers and auditing required by the proposed regulation.

       Kohler continued that it appears that in drafting the regulation, the Agency, in its drive to
ensure manufacturers sell compliant engines, has incrementally added requirements until the
total burden is excessive. They ask the Agency to take a careful look at the proposed regulation
from the prospective of those who need to comply. Kohler asked that we make it more "user
friendly" and cost effective by eliminating all unnecessary record keeping and reporting. They
said that the resources required to perform this unnecessary and burdensome recordkeeping,
                                       1-31

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
reporting and paperwork is time and effort that cannot be expended on cleaning up the engines to
provide real environmental benefits.

       Kohler commented on the following, but certainly not all inclusive, examples of sections
in the regulation where we feel reductions can and should be made:
     §91.1013 — Exemption for certified Small SI engines
     This section includes a reference to § 1045.605 which requires that small offroad engines
     (SORE) used as marine propulsion engines must have special labeling and record keeping.
     Kohler feels this is unnecessary since the engines are already labeled as compliant to the
     SORE regulation and represents an additional undue burden for manufacturers.
     §1054.130 — What installation instructions must I give to equipment manufacturers?
     Kohler feels there is a significant burden on engine manufacturer regarding evaporative
     emissions and general installation instructions.
     §1054.205 — What must I include in my application?
     Paragraph (a) requires for each engine configuration inwhich the maximum modal power is
     at or above 15kW a listing of the maximum power and the range of values for maximum
     engine power resulting from production tolerances. Kohler feels this is another unnecessary
     reporting burden.
     Paragraph (r) requires describing how engines comply with emission standards at varying
     altitudes and atmospheric pressures. Kohler suggests this will be a significant reporting
     burden that is not required today.
     §1054.610 — What is the exemption for delegated final assembly?
     Paragraph (c)(6) requires keeping records  to document how many engines are produced
     under this exemption. Also, manufacturers need to keep records to document contractual
     agreements under paragraph (c)(3) of this  section.
     Paragraph (c)(3) describes a contractual agreement with equipment manufacturers and the
     records required for this, so this section essentially is requiring the keeping of records on the
     keeping of records. In general, Kohler feels this section's description of the recordkeeping,
     labeling, and auditing appears overly complicated and confusing to the point that it will be
     commerce restricting.
     §1054.825 What reporting and recordkeeping requirements apply under this part?
     Kohler believes there are significantly more reports and recordkeeping than in Part 90; EPA
     needs to  review and make them more manageable.

       EMA commented on §1054.825 "What reporting and recordkeeping requirements apply
under this part?" Kohler stated that the report and recordkeeping requirements set forth in this
section are significantly more substantial than those currently required by 40 CFR Part 90 and
are overly burdensome. Kohler said that EPA should review the proposed requirements and
make whatever revisions are necessary in order to reduce such requirements and decrease the
substantially increased compliance burden associated with the proposed regulation.
       MIC commented that § 1051.825 states that "the following items illustrate the kind of
reporting and recordkeeping we require for vehicles regulated under this part." The title of this
subsection is "What reporting and recordkeeping requirements apply under this part." To be
consistent with the title and the presumed intent, the statement should be revised to read "The
                                          1-32

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


following items are the reporting and recordkeeping we require for vehicles regulated under this
part:"

Letters:
Commenter
Kohler
EMA
Motorcycle Industry Council
Document #
0703
0691
0701
Our Response:

       We agree that the proposed rule included several new reporting and recordkeeping
requirements. We have been careful to include only requirements that we believe are necessary
to allow us sufficient ability to oversee these programs to ensure that we can adequately
implement and enforce the regulatory requirements. The final rule includes several adjustments
to take into account the interest in reducing the compliance burden wherever possible. We take
Kohler's list of suggested opportunities for reducing the recordkeeping burden to illustrate our
interest:

§91.1013: We agree with Kohler that Small SI engines used for marine propulsion should not
trigger new labeling or recordkeeping requirements. For bigger engines we would be concerned
about creating a path for manufacturers to rely on an existing certification to avoid requirements
that apply specifically to marine engines. However, the relative stringency of standards and the
extent of sales for certified Small SI engines used for marine propulsion lead us to conclude that
a simple exemption from the marine requirements is appropriate.

§1054.130: We would expect that engine manufacturers are already providing equipment
manufacturers with installation instructions to address basic parameters such as inclusion of
intake or exhaust system components that meet performance specifications and placement of
exhaust components to ensure safe operation. The incremental effort to identify those items
necessary to ensure that engines and fuel systems are in the certified configuration after
installation in the equipment should be very small.  In fact, we would expect engine
manufacturers to do this  even if it were not required because they are liable for the emission
controls after the engines are installed. The installation instructions serve more to define the
limits of proper installation so it will be clear that it is the equipment manufacturer's fault if
engines were not installed according to the instructions.

§1054.205: Knowing whether engines are covered by one program or another is fundamental.  In
the case of Small SI engines, this hinges largely on the maximum engine power.  We believe it is
very reasonable to require manufacturers to identify the maximum engine power for engines that
are approaching the thresholds established in the regulation. We are reducing this burden for the
final rule by revising the regulation to more carefully identify those engines that are close
enough to the threshold to warrant this reporting.

§1054.205: Manufacturers are required under part 90 to comply with emission standards at
altitude, though the regulations allow manufacturers to do this with an altitude kit. However,
                                       1-33

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
since part 90 includes no reporting or recordkeeping requirement, we have no reason to believe
manufacturers are taking any steps to ensure that their engines meet emission standards at
altitudes different than at the lab used for certification testing.  We believe the new regulations
provide a minimal reporting and recordkeeping burden associated with the conditional allowance
to meet standards at high altitudes based on the use of engine modifications to install an altitude
kit.

§1054.610: We understand delegated assembly to be an optional provision that manufacturers
can exercise to help in cooperative relationships with component suppliers and equipment
manufacturers to assemble finished products.  Since catalysts are such a fundamental part of the
emission control system, several measures are needed to ensure that engines in final installations
are properly assembled such that they are in the certified configuration. The  contractual
arrangements, labeling, audits, and other measures are necessary to give us the confidence that
engines will be routinely assembled properly.  If the burdens of this oversight are too great,
manufacturers can simply default to the normal plan contemplated in  the regulation, which
involves engine manufacturers shipping only engines that are already in the certified
configuration. This is common across EPA programs today.

       With regard to the comment on §1051.825, the new text in §1051.825 (and similar
sections in other programs) is intended to help us administratively in the effort to maintain
current information collection requests with the Office of Management and Budget and to align
with the list of approved information collections in 40 CFR part 9.  While we have attempted to
provide a complete list of recordkeeping requirements in this new section, we cannot be certain
that it is absolutely comprehensive. This becomes especially true from a long-term perspective,
since we may add requirements and inadvertently omit those requirements from  §1051.825. We
would not want a manufacturer to be able to claim that a reporting or  recordkeeping requirement
that is clearly stated elsewhere in the regulation is not valid simply because it was omitted from
§1051.825. We are therefore keeping the language unchanged as illustrative  of the requirements
that apply throughout part 1051.

1.3.6  Inventor issues

What Commenters Said:

       N. Leggett commented that we need to encourage the independent inventors who design
and build their own full-scale spark-ignited engines. Experimental engines should be
automatically exempt from regulations. The inventor should not have to apply for some sort of
exemption. Rather the exemption should be automatically applicable to single engines that are
built by experimenters themselves. These individual experimenters should not be considered to
be engine manufacturers.

       N. Leggett commented further that the opportunities to experiment with engines are of
major importance to the economic and social future of the United States. Independent inventors
and experimenters should be encouraged to experiment with engines.  Their activities will lead to
the development of new engine technologies.
                                          1-34

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       N. Leggett also commented that a community of skilled hobbyists restores antique
gasoline engines to operating condition. They operate their engines on occasion including
showing them in action at shows and rallies of antique engine enthusiasts. Some of these people
install their restored engines in restored or rebuilt antique boats. These people perform a useful
function in the study and appreciation of American history. They too should be encouraged in
their efforts to bring antique technology back alive. These engines should have automatic
exemptions to the proposed emissions rules.  In addition, a related activity to the restoration of
antique engines is the building of replica antique engines from kits. Many of these kits are in the
form of rough castings that the builder then machines into an operational engine. This type of kit
building is a challenge to one's shop skills. Other kits are based on already machined parts.
Many of these engines are  small displacement hobby engines, but not all of them are small.
These kits should also be automatically exempted.

       N. Leggett commented that these exemptions to the proposed regulations are needed to
establish a free zone where experimenters and inventors are free to develop their own engine
designs and inventions. This is a contrast to the regulated engine world where people are
blocked from "tampering"  with engine features. Indeed, the very process of invention and
creative technology design is the basically playful activity of tampering with and departing from
conventional engine design. We need free spaces for invention to maintain America's position in
technology and manufacturing.

Letters:
Commenter
N. Leggett
Document #
0603
Our Response:

       We believe it is very problematic to introduce an exemption specifically for inventors'
experimental engines,  or homemade engines.  An experimental-engine exemption as
recommended in the comments would be impossible to enforce.  Without a request or approval
process, there would be no way of confirming that engines produced under such an exemption
would in fact meet even the minimal conditions described for the exemption. The Clean Air Act
and the regulations contemplate the need for experimental engines with the testing exemption
(§1068.210) and the manufacturer-owned exemption (§1068.215), as described in Section 1.2.2.
The testing exemption in particular would allow inventors to build, test, and operate their
experimental engines.  EPA's role in evaluating such exemption requests would be to confirm
that the scope of the exemption is appropriate for the company or individual requesting the
exemption and that the applicant understands the responsibilities associated with the exemption.

       The commenter's interest in an exemption for antique engines is mostly unnecessary.  To
the extent that someone restores engines that have already been placed into service or installs
such engines in restored vehicles, these engines are not subject to standards and therefore no
exemption is needed. Building replicas of antique engines is a different matter.  To the extent
that these antique engines meet our definition of "engine," they are subject to emission standards
and therefore must either be certified or qualify for an exemption. These engines may qualify in
some cases for the hobby-engine exemption. In other cases, the Clean Air Act and the
                                       1-35

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
regulations contemplate this scenario with the display exemption (§1068.220).  However, this
exemption is available only to companies that hold a valid certificate of conformity with EPA.
As described above, adding an exemption as recommended by the commenter would lead to a
situation where we would be unable to confirm that engines are being exempted appropriately.
For example, the more common form of building replica engines is for a company to build
knockoff engines, imitating an engine by disassembling it and "designing" parts for a new
assembly by carefully measuring the original parts.  This is a very significant compliance
problem. We believe there is not enough value in preserving an allowance to build new, fully
functioning, replica engines to outweigh the compliance and enforcement problems that would
result.  If someone wants to build such an engine, that would be allowable as long as the engine
does not include a crankshaft.  Such an exercise would still provide plenty of challenge for
machining and assembly; however, the value of the resulting assembly would be limited to
display purposes, without the benefit of producing usable power.

       Innovation is clearly  important to our future economic health and welfare.  Our interest is
in preserving a free zone for innovation without creating a zone where companies are free to
produce large numbers of noncompliant engines. Even under the current requirements we are
finding many thousands of engines that are being sold illegally.  Any relaxation of current
requirements would therefore need to be done very carefully to avoid making this situation
worse. We understand that some inventors may find the paperwork and approval requirements
to be burdensome, but we believe the current regulatory framework allows for innovation with a
minimum of administrative requirements.

1.4  Amendments to engine-testing provisions in 40 CFR part 1065

     We adopted extensive  changes to the test procedures in part 1065 as part of the rulemaking
to set emission standards for locomotive and marine diesel engines.  We have identified a few
additional revisions that we are including in this final rule. Some of these changes are necessary
to address issues related to Small SI or Marine SI engines. Other changes involve corrections or
clarifications of a more general nature.

What Commenters Said:

       California ARB commented that EPA is proposing to allow the use of non-dispersive
ultraviolet analyzers (NDUV) to measure NOx emission levels in addition to the currently
accepted chemiluminescent detector (CLD). California ARB generally allows the use of
alternative measuring methods if a manufacturer can demonstrate equivalency with the current
accepted method. Recently, a manufacturer requested California ARB's approval to use NDUV
to measure NOx levels for small off-road spark-ignited engines.  In response, California ARB
requested the manufacturer provide data to show equivalency between emission results from
NDUV and CLD. The manufacturer referenced a 2002 study conducted at EPA facilities wherein
testing of NDUV technology for NOx measurement was performed  on gasoline powered light
duty passenger vehicles. However, it is worthwhile to note that despite the results of the study,
EPA has not approved the use of NDUV for NOx measurement for light-duty gasoline vehicles.
The manufacturer also provided limited test data from a single small spark-ignited engine. The
data provided by the manufacturer did not indicate high correlation between the two methods
                                         1-36

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


and, instead, raised questions about the use of NDUV for official certification purposes at this
time.  California ARB requested EPA establish a test program to generate data and determine if
high correlation between NDUV and CLD measurement technologies exists before allowing its
use to measure NOx emission levels in small off-road spark-ignited engines.

       EMA commented that the requirement to obtain speed and load data at 5 Hz update rate
is not necessary for steady state testing (§1065.110(e)).  EMA recommends that for steady-state
testing the data acquisition requirement be amended to 0.5 Hz minimum.

       EMA commented that new analyzers are not configured for adjustability of FID response.
Accordingly, §1065.360 should be revised to reflect current analyzer industry practice.

       EMA commented that the complete engine mapping procedure defined in §1065.510 is
not required for Small SI engines.  The Part 1065 requirement should include clarification that
this requirement can be omitted per the standard setting part.

       Cummins commented on draft language to amend the requirements related to cycle-
validation criteria in §1065.514.  They would still have to use §1065.514(f)(3)(i) because (ii) [the
new provision based on existing requirements in part 90] does not use the statistical method. If
there was going to be a mode-by-mode validation allowance in part 1065, they wanted to
generalize it for all discrete-mode tests for consistency.  The need for mode-by-mode verification
did not seem to be unique to a specific engine size or technology. It appeared that the request
was made to allow a lower-cost alternative method. It would be cheaper to only log average,
min, max, etc. values for each mode rather than log 1 Hz data. If it is intended that this option
will go away in the future, then Cummins recommended pulling it out now and putting it in part
1054 for Small SI engines. It could then be dropped from Small SI as appropriate in the future
without confounding testing for other engine categories.

Letters:
Commenter
California ARB
EMA
Cummins
Document #
0718
0691
0795
Our Response:

       The performance specifications adopted for NDUV analyzers are intended to ensure that
measurements will properly characterize an engine's emission levels. For example, Small SI
engines may have somewhat higher levels of lubricating oil in the exhaust stream, which could
cloud the lens and other components of the instrument, leading to inaccurate results. However,
under this arrangement the NDUV analyzer would not reliably meet performance specifications
that would allow for a valid test. To avoid a situation where a manufacturer meets calibrations
and then performs testing with other engines that may cause such a problem, we are revising the
regulation to note that good engineering judgment may preclude manufacturers from using an
NDUV analyzer if sampled exhaust from test engines contains oil (or other contaminants) in
sufficiently high concentrations to interfere with proper operation.
                                       1-37

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
       We revised the provisions of §1065.110 and §1065.512 to specify 5 Hz measurements for
transient testing and 1 Hz measurements for steady-state testing.  This addresses the
manufacturers' concern without foregoing measurement accuracy.  We made these changes as
part of the rulemaking to set standards for locomotive and marine diesel engines (73 FR 37096,
June 30, 2008).

       We made extensive changes to §1065.360 as part of the rulemaking to set standards for
locomotive and marine diesel engines (73 FR 37096, June 30, 2008).  These changes were made
in collaboration with EMA members. However, it is important to note that adjustability is
critical to proper use of a FID for measuring exhaust hydrocarbons. We therefore believe it is
not appropriate to use a FID that lacks adjustability.

       Part 1065 already includes general language stating that the standard-setting part governs
when there is any  difference in the specified procedures for a particular set of engines. If we
were to reference every case where one of the standard-setting parts included additional or
differing provisions, we would forego much of the advantage of adopting a comprehensive set of
regulations that are not category specific. We are therefore not making the change to part 1065
to include a specific reference to the lack of mapping requirements for engines subject to part
1054.

       We agree with the suggestion from Cummins to move the new approach to cycle-
validation criteria to part 1054 so it applies only to Small SI engines.

1.5  Amendments to general compliance provisions in 40 CFR part 1068

1.5.1   Definition of "engine" and provisions related to partially complete engines

What Commenters Said:

       EMA commented that the NPRM proposes a number of changes to Part 1068 which
extend well beyond Small SI engines. Many of the proposed changes are technical clarifications
or corrections to existing programs, and have been previously discussed with the affected
regulated entities.  While EMA has some comments on those technical clarifications, they have
no objection to their being finalized in this rulemaking. Other changes are significantly more
substantial and raise major new issues which have not been thoroughly discussed with the
affected stakeholders. For  example, EPA appears to be taking action, in the guise of a
definitional change,  which  will  substantially change the existing requirements for all nonroad
engine service parts and engine rebuild practices.

       EMA and its members have no objection to working with EPA to better understand the
issues that EPA is trying to address and, if necessary, to develop appropriate regulatory
guidance.  However, EPA should not adopt regulatory changes that will impact aftermarket
engine service parts and engine rebuild practices and programs, without a separate rulemaking
and adequate notice and opportunity for discussion, analysis and  comment by all stakeholders.
In that regard, EMA noted that the Nonroad Tier 4 rulemaking was developed, with tremendous
success, by an extraordinary level of cooperation and outreach by and between EPA and all of


                                          1-38

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


the affected stakeholders. EMA finds it odd that EPA would now try to adopt rules that
significantly impact nonroad engines without a similarly thorough level of outreach.

       EMA commented that EPA's proposed definition of an engine - a cylinder block plus one
component - is not viable and, as a real-world matter, cannot be implemented practically. EMA
understands that EPA has raised concerns about its ability to enforce standards if uncertified,
incomplete engines enter into commerce.  However, the proposed definition is not a viable
means to address the concern.  EMA noted that it is common for manufacturers to utilize global
product manufacturing processes for engine parts, including engine cylinder blocks. Therefore,
it is common practice for cylinder blocks manufactured outside  of the U.S. to be imported into
the U.S. for future assembly. In many cases, those cylinder blocks include additional parts
depending on the economics and universal nature of the features. It is not uncommon for the
same engine block assembly to be utilized in new engine manufacturing of products certified in
different product categories or for use as repair parts for older engines already in service. EMA
also commented that the proposed definition of a "partially complete  engine" further compounds
the confusion regarding the definition of an engine because it is inconceivable that an incomplete
engine has substantially more parts than an engine (as would be the case pursuant to the
proposed definitions).  EMA believes the proposed definitions clearly require significant
discussion with a broader industry group including both new engine manufacturers and
remanufacturers. They believe the proposed new definition should be removed from this
rulemaking and deferred to a subsequent process.

       NACOO commented that the new wording found in the "engine" definition of section
1068.30 is not clear at all.  It appears to be making the statements more than once with the
addition of conflicting requirements of a complete engine and incomplete  engine (i.e. the third
sentence indicates what is not included engine blocks with no attached components). Then the
next sentence states "This includes complete and partially complete as follows:" NACOO
commented that the way those sentences are written seems to be saying that a complete engine is
not an engine.  NACOO commented that EPA needs a definition of "Engine Block." They
question whether an "Engine Block" is just a block with no crank pistons cams and just an empty
cast iron block? They believe it is not clear what the term "Engine Block" means under the
proposed regulations.

       ISMA commented regarding §§ 1068.260 and 1068.262: It is their understanding from
EPA certification workshop discussions that production configuration engines destined for
installation in a certified recreational vehicle (e.g., snowmobile) are considered "partially
complete engines" under 40 CFR 1068.330. ISMA understands that these  engines do not need to
be permanently labeled with an emission control information label since they are not by
themselves a certified entity - the completed recreational vehicle is. Furthermore such engines
are covered by the temporary exemption in 1068.330 when they are imported into the U.S. The
engines are designated on the EPA 3520-21 Form using Box F (recreational spark-ignition
vehicles or engines) and Box 16 (incomplete engines). A hang tag or  other non-permanent means
of identifying the engines should be sufficient at the time of import. They  would like to be clear
that the ability to efficiently import these engines is critical to the snowmobile manufacturers.
                                       1-39

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
       MIC commented that there are several cases impacting the MIC member companies, such
as:
   •   importation of engines not subject to stand-alone engine-based certification requirements,
       that are destined for installation in certified vehicles
   •   transportation of engines not subject to stand-alone engine-based certification
       requirements, that are manufactured by a certifying vehicle manufacturer and transported
       from one of their locations to another
   •   temporary labeling requirements for these and other circumstances

MIC continued to comment that specific provisions may be needed to address engines used in
recreational vehicles and other vehicles certified by EPA.  They think the generic language in this
section of Part 1068 does not allow proper treatment of the nuances that exist between engine-
based and vehicle-based certification categories. Efficient movement of these vehicle
components is critical to the success of several MIC member companies.

       Arctic Cat commented regarding §1068.262. This  section changes the requirements for
shipping partially complete engines within the US. They are confused by this section and are still
studying its effect on their operations. For example, it is not clear to them how shipments from
Arctic Cat's ATV engine factory in St. Cloud, MN to their assembly line in Thief River Falls,
MN are affected because in this case no "secondary manufacturer" is involved. Arctic Cat
requests additional time to study this section and provide additional comments.

       IMPCO proposes that § 1068.262 be removed in its entirety and be re-written and re-
introduced at a later date. IMPCO understands EPA's intent behind the proposed language, but
it is impractical and will be near-impossible to implement. The main logistical issue  is how the
engine manufacturer will be able to identify not only the certifying manufacturer, but the engine
family name. For example, IMPCO purchases engines through distributors, not from the
manufacturer directly. IMPCO questioned how the engine manufacturer can be involved in this
process. Additionally, IMPCO noted that they can have two or more different engine families
using the same engine. In such a situation, IMPCO questioned which engine  family will be used
on the temporary label sent from the engine manufacturer.

       ECO commented that EPA has proposed to restrict the importation of base engines prior
to certification approval. In addressing this issue, it is critical that EPA incorporate flexibility for
engine manufacturers that are in the process of certifying engine families, of which the base
engine is sourced from outside the U.S. In these instances, the Manufacturer  of Records (MORs)
are often required to import the base engines and initiate the emission control system upfit
process prior to receiving final certification approval. In these instances, it is necessary for EPA
to allow MORs importation flexibility for engines that are not completely assembled.

       California ARB supports the proposed requirement for partially complete engines to
prevent manufacturers from selling partially complete engines as a strategy to circumvent
certification procedures. Under the proposed definition, the short blocks or three-quarter blocks
without fuel systems would need to be certified. However, EPA's current production line testing,
in-use testing, and warranty requirements are not designed for partially complete engines. In
addition, EPA allows the large spark-ignition manufacturers to ship partially complete engines to
                                          1-40

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


the secondary engine manufacturers without emission-related components in some
circumstances. This would make these provisions (1068.240, 1068.255, and 1068.260) difficult
to enforce.  California ARB thus recommends that EPA prepare certification guidelines on how
to certify partially complete engines under different engine categories. Partially complete engines
could be designed to use either gasoline or diesel fuel systems and subsequently they might be
installed on stationary equipment, on-road vehicles, off-road equipment, or used as replacement
engines. Preparing certification guidelines would help in circumstances when it may be difficult
for the engine manufacturers to determine the partially complete engines' final destination.

       OPEI commented that the definition of engine (in particular "partially complete engine")
in section 1068.30 is missing a critical example in paragraph (2). OPEI notes its engines are not
considered complete until final carburetor adjustments have been done to bring engine into
proper emission and performance compliance. OPEI requests to add subparagraph (2)(vi) as
follows:  "(vi) An engine that has been assembled (except for final labeling) but has not
undergone final carburetor or other tuning to bring it into compliance with manufacturer's
specifications and these standards."

       For handheld engines, OPEI does not believe that "short blocks" are "replacement
engines," and they should not be tested and labeled as such. Their short block consists of the
crankcase, cylinder and crankshaft assembly.  The block cannot run until the cooling, ignition,
intake, carburetor, fuel system and shrouding are put on. Manufacturers sell this block for
rebuilding because the cost of the block is less than the cost of the repair. OPEI requests EPA
add language that this provision does not apply to handheld engines.

       GM commented that they have considered the impacts of EPA's starting proposal.
Although the concept of what EPA intends to do is simple,  it truly creates a very difficult
position for GM, and they suspect their OEM customers as well. GM hoped that they can find a
better alternative to accomplish EPA's goals.

       GM noted that its current business model works something like this. GM has a portfolio
of engines that they offer to their marine customers.  All of the engines they offer are partial
engines (and GM is the primary engine manufacturer per EPA's definitions). These engines are
not capable of running in the as shipped configurations. GM's customers (Mercury, Volvo,
Indmar, Flagship, Kodiak, PleasureCraft and Marine Power) need to do an extensive upfit to
"dress" these engines for the final boatbuilder - these customers are the "secondary engine
manufacturers."  The value added by these customers is significant and includes intake and/or
exhaust manifolds, fuel systems, accessory drive, cooling systems, engine controllers and wire
harness, etc. In all cases, GM is not the Manufacturer of Record, the Secondary Engine
Manufacturers are.

       The engines GM offers are based  off automotive variants and may include some unique
marine hardware. GM may make multiple variants (flavors) of the same engine to meet
customer applications differences (ex: sterndrive vs. inboard). They note this does create a
proliferation of engine assemblies they  offer and their manufacturing plants / logistics operations
need to deal with this proliferation - of course, at a cost. As an example, today (2008) GM builds
5 variants of the 5.7L (350 cubic inch) marine engine.  Key differences include:
                                       1-41

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
    With vs. without intake manifold and fuel system
    Front ring gear (sterndrive) vs. Rear ring gear (inboard)
    Partial ignition system vs. no ignition system
    Mechanical throttle body vs. Electronic throttle body
Thus, GM's engineering release and engine plant have 5 part numbers (P/N's) to deal with. An
engine assembly P/N is required for each unique parts list for the final shipped product. GM
noted that because most of their marine customers use more than one type of engine from this
5.7L family, this would create the need to create a very large number of labels - an extreme
would be 5 engines x 7 customers = 35 labels.  This would also require the GM system to replace
the 5 engine assemblies with 35 engine assemblies - and that's just the 5.7L at one plant. GM
continued to comment that obviously EPA can see the initial burden.  This would significantly
affect GM's flexibility as today they can ship the same engine P/N to any of the customers who
order that particular part number.  This could be compounded more as the customers may sell
globally (GM doesn't know if the engine is for sale in US or exported).  Some engines may be
used by their customers for their service needs.

       GM requested EPA to consider an option. They referenced some of EPA's initial
verbiage: "Manufacturers may introduce into US commerce, partially complete engines as
described in this section if they have a written request for such engines from a secondary engine
manufacturer that has certified the engine and will finish the engine assembly."  GM questions
whether this could be interpreted that if a secondary engine manufacturer requests (via purchase
order) to procure an engine from a primary engine manufacturer a specific P/N - then this would
be sufficient to meet EPA's needs.  GM noted that the engine will have engine P/N and broadcast
codes, as well as information on the bill of lading where the engines come from. The secondary
engine manufacturer could possibly also include in the PO any references as to engine family
and certification intents. GM commented that EPA needs to discuss this issue with a broader
group of affected companies.

       EMA commented on draft regulatory language allowing the movement of partially
complete engines between different locations of the same parent company to affirm the principle
and request that we clarify that current business practices are not required to change substantially
and additional product identification or labeling are not required.

       EMA also suggested that the regulations allow engine manufacturers to ship partially
complete engines before a Certificate of Conformity is approved. This may also be covered by
the manufacturer-owned exemption and/or test exemption.

       In later comments, GM emphasized that they wanted to avoid labeling partially complete
engines or, if labels are needed, to be sure that GM's label is generic enough that they would not
need to identify the destination or a valid engine family name on the label.  Putting the receiving
company's name and address on the bill of lading is acceptable, but the bill of lading should not
need engine family names.  GM also objected to any provision that would put them in a position
of ensuring that secondary engine manufacturers have a valid certification or exemption that
allows them to receive shipment of partially complete engines. In addition, GM raised questions
about how the regulations would allow for shipping engines for which the secondary engine
                                          1-42

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


manufacturer has an exemption (testing, replacement, etc.), and for secondary engine
manufacturers to ship engines to each other (such as for managing excess inventory).

       The Industrial Truck Association asked for clarification regarding engines that qualify for
the manufacturer-owned exemption or the test exemption. They especially wanted to know how
to apply the provisions of §1068.262 for engines that are  exempt for other reasons.

       IMPCO noted that in certain cases, the equipment manufacturer purchases the engine and
aftertreatment. This is effectively a pass through where they charge for the fuel system and
installation, but are neither involved with the acquisition  nor do they factor the cost of the engine
or aftertreatment directly into the cost of the certified engine.  Given that these are not test
engines, they are not the certificate holder, and they will never be the certificate holder, IMPCO
question how the equipment manufacturer would acquire these engines from the engine
manufacturer. Also, the regulation should not require the manufacturer to state unconditionally
that engines will comply with applicable regulations in their final configuration.  As with §
1068.261(j), the certifying manufacturer cannot be held liable for engines that are not in their
final configuration when installed in the equipment, unless the certifying manufacturer was  in
some way negligent when it comes to specifying part numbers, installation instructions, etc.
Allow the manufacturers to include qualifying language in their statement to recognize the
equipment manufacturers' need to follow installation instructions.

       IMPCO further commented that the proposed language related to revoking the exemption
seems to allow EPA to revoke the exemption for the secondary engine manufacturer in its
entirety. Therefore, EVIPCO questioned whether EPA would have the authority to stop shipment
of all GM engines to IMPCO, even if the noncompliance only occurred with one engine family.
Given the liability placed on IMPCO for items that are outside of its control, IMPCO commented
that EPA should have to prove the noncompliance on an engine family-specific basis. They
believe this paragraph is too general and too far-reaching. Finally, IMPCO questioned what it
means to say: "if that manufacturer sells engines that are not in a certified configuration".
IMPCO noted that when engines are shipped without aftertreatment or other components, they
are all "not in a certified configuration."
                                       1-43

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
Letters:
Commenter
ISMA
IMPCO
MIC
Arctic Cat
NACOO
ECO
California ARE
OPEI
EMA
GM
EMA
GM
Industrial Truck Association
IMPCO
Document #
0671
0692
0701
0709
0714
0712
0682
0675
0691
0747
0809
0787
0800
0812
Our Response:

       At the time of the proposal, it became clear that manufacturers had somewhat varying
interpretations of regulatory provisions related to partially complete engines. To the extent that
manufacturers took the view that partially complete engines were not subject to emission
standards, we understand that our proposed language to clarify the definition of "engine" would
be a very meaningful clarification regarding the scope of EPA regulations. We believe it would
be reasonable to consider the proposed definition (or a variation of it) to be the proper
interpretation of regulatory requirements that were adopted earlier. However, we understand that
manufacturers may not have been operating with that understanding. The proposal included a
description of the concerns that led us to make this change, in particular the need to address the
prevailing practice of shipping short blocks and long blocks to secondary engine manufacturers
for certification as Marine SI engines and to address the increasing occurrence of noncompliant
imported engines.

       We believe it is both necessary and appropriate to finalize the new definitions in this
rulemaking. We received extensive input from a wide range of manufacturers during and after
the public comment period on these issues,  and have provided updated draft regulatory language
to manufacturers representing other industry sectors.  While the clarified scope of the regulation
is broader than some manufacturers have understood to be the case previously, we have been
careful to include exemption provisions to avoid unwarranted disruption of a wide range of
legitimate business practices related to assembling and distributing engines.

       We agree with the manufacturers' comments that the proposed definition of "engine"
(block plus one attached component) was too broad. In particular, the first attached components
may be dowels, pins, bushings, or plugs, none of which are fundamental to initiating the engine-
assembly process.  Discussions led us to conclude that installation of the crankshaft serves as a
clear, objective, and fundamental point in the production process that can be considered as the
time when the engine block becomes an engine (when the engine  is "born").  As a result, we will
                                          1-44

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


consider engines subject to standards in any case where an engine block has an installed
crankshaft, whether or not there are any additional components attached or assembled.  This
means that each one of these "engines" that is introduced into U.S. commerce needs to be
covered by a valid certificate of conformity or an exemption (temporary or permanent).  This
approach applies equally to handheld engines, locomotive engines, and everything in between.

       We note in the definition that there are two types of engines—complete and partially
complete engines. This is intended to allow for the situation where special exemption provisions
apply for engines that are not yet in a certified configuration. Engines needing carburetor
adjustments represent one possible example of that; however, this does not change the fact that
we find it necessary to consider these partially complete engines to be subject to standards.  See
the discussion below regarding date of manufacture for further discussion of issues related to the
flow of products and stages of assembly.

       We agree with the commenter that fully operational engines intended for installation in a
recreational vehicle should be considered partially complete engines under the regulation.  The
rulemaking changes clarify these relationships and add a variety of oversight provisions (such as
labeling, consistent with what was recommended) to prevent circumvention of the regulations.
The provisions currently in §1068.330 are being expanded and codified in §1068.262. We
believe this approach should be a minor change from current practice for companies currently
buying engines for installation in their certified  recreational vehicles. Nevertheless, we
understand that manufacturers may need time to adapt their ordering and shipping practices to
follow these new requirements.  The definition of engine and the corresponding provisions
related to partially complete engines take effect immediately once the final rule is effective. We
are therefore specifying in §1068.40 that manufacturers may have up to 12 months to comply
with new requirements. In the case of §1068.262, we are giving advance approval for waiving
the documentation and tracking requirements related to partially complete engines.

       For the particular question about certificate holders shipping a partially complete engine
from one plant to a different plant within the company, we address this in §1068.260 by
specifying  simply that manufacturers should notify us in their application for certification that
they will be shipping these partially complete engines to another of their facilities. An approved
certification represents an approval of the exemption that would allow for this  transaction.  No
labeling or additional recordkeeping requirements apply. We learned that some companies rely
on third-party companies to arrange for  inventory and transport of engines even if they are
shipping the engines between two of their own facilities. Sometimes this even involves
transferring ownership of the engines to the other company. We chose to address this by adding
a provision allowing for third-party companies to be involved in these engine shipments, as long
as the certifying manufacturer demonstrates that the engines will be transported only according
to its specifications. These provisions are intended to allow manufacturers to continue current
practices, but we would not agree that manufacturers should be able to continue their current
practices if they do not conform to these minimal requirements.  Since the certificate-holder
controls these engines at all times, there is no need for labeling or other identification beyond
what the manufacturer would do for its normal business practice.
                                       1-45

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
       We have made some revisions to the exemption provisions in §1068.262 for shipping
engines to secondary engine manufacturers that certify the engines. First, we specify that the
documentation must identify a valid engine family name for the particular engine model. The
secondary engine manufacturer would simply pass along this information when ordering the
engine.  The family name could represent a marine, industrial, or stationary application.  The
shipped engines would not all need to be built up to match the given engine family. For
example, the secondary engine manufacturer could provide a valid family name for an order of
4.3-liter engines, then build those engines to be covered by that certificate, or any other
certificate.  These engines could also be covered by an exemption (for export, for example).
Such an exemption must be approved before the original engine manufacturer may ship engines
under the provisions of §1068.262.  Second, we specify that the removable label may be
simplified to include only the shipping manufacturer's name, a statement that the engine is being
shipped to the certifying company, and a reference to the bill of lading. This allows the shipping
engine manufacturer to make a universal label that would apply for all the engines it produces
and ships under these provisions. Third, we are allowing manufacturers to apply a single label to
engines that are packaged together. For example, if 30  Small SI engines are shipped in a pallet-
mounted box, the manufacturer may label the box instead of labeling the engines individually.
Fourth, we are including provisions allowing manufacturers to ship engines to secondary engine
manufacturers while an application for certification is pending. This would allow secondary
engine manufacturers to start producing engines after sending an application for certification.
This is similar to what we allow for other manufacturers; see Section 1.5.1 for general  provisions
that apply for these early-production engines. Fifth, we have revised the requirement for
secondary engine manufacturers to make an unconditional statement of compliance. The revised
statement attests that the manufacturer has distributed engines that conformed to the regulations,
rather than attesting that the engines in the final configuration will be compliant.  This  focuses
the secondary engine manufacturer's statement on the activities it can control.  Sixth, we have
revised the regulation to specify that it is a violation "if that manufacturer sells engines that are
not in a certified configuration in violation of the regulations."  This avoids the confusion that
might arise from the provisions related to selling engines without aftertreatment devices under
the delegated-assembly provisions.

       The final rule does not include a requirement for original manufacturers to include engine
family information on the engine label or on the bill of lading.  We believe the documentation
provisions related to ordering the engines and the requirement to ship the engines directly to the
secondary engine manufacturer should be sufficient to ensure that engines reach a certified
configuration before reaching the ultimate purchaser.  However, this reduced information on the
label requires that we specify that the original manufacturer assumes some responsibility for
ensuring the validity of the information specified by the secondary engine manufacturer.
Accordingly, we specify that we may void the original engine manufacturer's exemption if the
engines are shipped to the wrong destination or if engines are not properly labeled.

       The language describing how EPA might revoke the exemption generally applies to a
secondary engine manufacturer's engine family, but we clearly should be able to void or revoke
an exemption more narrowly or more broadly if available information allows us to identify how
specific the violation is.  For example, we may void or revoke the exemption with respect to a
particular engine model or for all engine models shipped to the secondary engine manufacturer,
                                          1-46

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


depending on whether or not the violation is unique to a particular engine model. We may also
void or revoke the exemption from the original engine manufacturer with respect to a single
secondary engine manufacturer or all affected secondary engine manufacturers, depending on
whether or not the violation is unique to a particular secondary engine manufacturer.

      We believe the final regulation and the guidance contained in rulemaking documents
provides sufficient guidance to implement the new provisions. We will expect to interact
extensively with companies as they follow these requirements and will be prepared to publish
any necessary clarifications as the need arises.  In particular, our understanding is that most of
the exemption provisions related to partially complete engines are temporary, which means that
the normal requirements (production-line testing, warranty, etc.) will all apply at a later point in
the assembly  and distribution process. The provisions for handling short blocks as replacement
engines  are permanent exemptions, but this is consistent with the way we have handled
exemptions for complete replacement engines in the past.

      Note that we are adding a new §1048.601(b) to describe how the replacement-engine
provisions of § 1068.240 apply for engines subject to part 1048  in conjunction with the
secondary engine manufacturer provisions  in § 1068.262. For cases in which the secondary
engine manufacturer completes assembly of the engine, these provisions apply as written. If the
secondary engine manufacturer arranges for a third party to complete engine assembly, some
additional provisions apply. Most significantly, the ultimate purchaser must purchase (or
otherwise order) the replacement engine from the secondary engine manufacturer, and the
secondary engine manufacturer and engine assembler are both responsible if the engine is
installed in new equipment or otherwise violates the circumvention provisions of § 1068.240.

1.5.2 Definition of "date of manufacture" and issues related to "model year"

What Commenters Said:

      EMA  commented that EPA is proposing a significant regulatory change - not a
clarification - that would define an engine's date of manufacture based on when the engine is
capable  of running or on when an incomplete engine is imported to the U.S.  This significant
change is both unwarranted and misguided. EPA's proposed change will alter the way in which
manufacturers currently operate, will impose significant costs and administrative burdens, and
will not provide any emission benefit.  Indeed, EPA has not identified any "problem" that it is
attempting to  solve.  The fact of the matter is that EPA's new definition will create problems.

      EMA  continued commenting that under EPA's rules, a manufacturer on a calendar year
model year (which is the vast majority of manufacturers) cannot produce an engine for the
current model year after December 31 of that model year. However, EPA's proposed new rules
would require the manufacturer that begins production of an engine in one model year to meet
the regulatory requirements of the next model year (when the engine can actually run or when
the incomplete engine is imported). That is wrong and unfair.  It is obvious that, as a practical
matter, an engine built on December 31st could not run or would not be imported until January
1 st.  So, the December 31 st engine would either have to be built to the next model year's
standards, or it could not be sold.  It is just as obvious that such  a scenario could occur for
                                       1-47

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
engines built on December 30th, and probably for all of December, and, indeed, for engines built
even earlier in that model year. The nature of the manufacturing process, exacerbated by the
location of engine plants throughout the world and the non-integrated nature of the industry,
makes EPA's proposed new date of manufacture definition not only impractical, but likely
impossible.

       EMA commented that manufacturers cannot live with a rule that prohibits them from
assigning the date of manufacture at the point within the engine assembly process that is relevant
to their specific manufacturing workflow.  The assignment process established for a given
manufacturing facility does not (and should not) change from day to day or from the beginning
to the end of a model year.  For products that are imported as incomplete engines, the date of
manufacture likewise is determined and assigned during the manufacturing process - - a timing
and process that is independent of shipment and importation.

       EMA continued that to a large degree, EPA's proposed change seems to be a "solution in
search of a problem."  Certainly, for those model years where there is no change in standards
(i.e. most model years), the potential  for somehow "gaming" the system around defining the date
of manufacture simply does not exist. But, even for those model years in which there is a change
in standards, any potential for "gaming" can be eliminated by EPA simply requiring that
whatever process and procedure the manufacturer uses for establishing date of manufacture
remain constant for an engine family throughout all of its model years.

       Finally, EMA noted that engine manufacturers are required to control inventory of either
in-process engines, or incomplete engines being imported, to normal levels through both EPA
anti-stockpiling requirements and also normal business practices.  Accordingly, manufacturers
should be allowed to maintain their current production process for the determination of the
engine date of manufacture.

       OPEI commented that the definition for Date of Manufacture in section 1068.30 is too
obscure. Paragraph (l)(i) of the definition has a very narrow interpretation.  An engine being
able to run is different than an engine set to run properly for emission and performance.  OPEI
suggests this wording be revised as follows:  "The date on which the engine is assembled and
adjusted to the point of being able to properly run for compliance to these standards."

       OPEI also commented that EPA should clarify section 1060.201.  The certificate of
conformity will list an effective date (signature date).  The manufacturer may not introduce into
commerce before this date  but may produce equipment/engines prior to the effective date.

       California ARB recommended that procedures be adopted to prevent any stockpiling of
engines that could be used to circumvent the regulations.

       OPEI/EMA suggested that EPA clarify the allowance for equipment manufacturers to use
up inventories of previous MY engines, adding an allowance for engine manufacturers to sell
engines that they had built in the previous MY.
                                          1-48

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       IMPCO commented that § 1068.103(c) appropriately allows for production of engines
while an application for certification is pending, but the regulation should prohibit introduction
into U.S. commerce only rather than also prohibiting the selling and offering to sell such engines.

       Manufacturers also raised a variety of issues related to our proposal to adopt certain
restrictions on naming an engine's model year for importation, as described in Section 2.10.3.

Letters:
Commenter
OPEI
EMA
California ARE
IMPCO
Document #
0675
0691
0682
0812
Our Response:

       Until now, the regulations have not specified the point in the assembly process that
should serve as the basis for establishing an engine's date of manufacture.  For the large majority
of engines, this is not an issue, since total assembly time from start to finish is measured in hours
or perhaps days.  As a result, it is relatively uncommon for there to be any uncertainty regarding
an engine's date of manufacture for purposes of deciding which standards apply.  Nevertheless,
we have learned that there are widely diverging practices for establishing an engine's date of
manufacture, which means there is a different effective date of new emission standards for
different manufacturers. This is especially of interest for larger engines, which are more likely to
be assembled in multiple stages at different facilities. We believe it is important to establish a
clear requirement in this regard to avoid ambiguity and different interpretations. A consistent
approach preserves a level playing field and may prevent some manufacturers from manipulating
their build dates to circumvent the regulations.

       We expected that the proposed definition of "date of manufacture," based on  reaching a
final, running configuration, was the most straightforward and logical interpretation.  The
comments received and the ensuing discussions made clear that this interpretation was not
universally held. The diversity of views underscores the need for the regulations to establish a
clear and uniform requirement.  Once we are  able to establish such a requirement, we believe
there would be a "cost and burden" only for those companies that would otherwise be attempting
to delay complying with new emission standards. Requiring only that manufacturers continue
their normal business practice or maintain a consistent approach from year to year would not do
enough to establish uniform and enforceable requirements related to the transition to new
emission standards.

       However, we recognize the concern that manufacturers need a rather high degree of
certainty regarding applicable emission standards when they initiate assembly of an engine.  Any
number of variables in the production process could affect how long it takes to finish building an
engine. We therefore believe it is most appropriate to match up the definitions for "date of
manufacture" and "engine" by specifying that an engine's date of manufacture should be based
on the  date that the crankshaft is installed in the engine.  This provides manufacturers with the
                                       1-49

-------
                                Chapter 1: Rulemaking Process and Cross-Program Issues
control they need to determine which emission standards apply when they start to build the
engine.

       We are aware that secondary engine manufacturers may have inventory and assembly
procedures that are not tied to the actual date of crankshaft installation by the original engine
manufacturer.  We are therefore specifying for this situation that the date of manufacture is
generally the date the secondary engine manufacturer receives shipment of the partially complete
engine.  The manufacturer may alternatively specify a date of manufacture up to 30 days earlier
as long as that date is not earlier than the date the crankshaft was actually installed in the engine.
This puts the secondary engine manufacturer in a similar position relative to companies with sole
responsibility for assembling complete engines, without placing unreasonable expectations on
secondary engine manufacturers.

       Some manufacturers would be interested in naming a date of manufacture that is later
than we specify in the regulation, as  suggested in the comments. This may be for marketing
purposes, managing inventories of engine components, or for other recordkeeping or product-
development reasons.  There is no risk of manufacturers gaining an advantage of being subject to
less stringent standards by delaying the date of manufacture for an engine, so we would have no
objection to that. However, we limit the  selection of date of manufacture to a later point in the
assembly process.  Selecting a date of manufacture after the end of the assembly process for an
engine would raise concerns about the risk of manipulating emission credits for a given model
year and about ensuring that engine assembly and dates of manufacture are always within the
production period established for a given engine family, as described in the certificate of
conformity or the manufacturer's records. We see no legitimate reason to select a date of
manufacture after completing assembly for an engine.

       This approach addresses manufacturers' concerns for knowing which standards apply to
an engine, but we are concerned that manufacturers could ramp up production of engine blocks
with installed crankshafts as a method to delay compliance with new emission standards.  EPA
regulations have always included provisions describing limits on inventory and stockpiling
practices for equipment manufacturers. The regulations until now have not clearly addressed
issues related to stockpiling for engine manufacturers.  We agree with the suggestion from
commenters that anti-stockpiling provisions would be appropriate.  The Clean Air Act
contemplates the need for such provisions in §202(b)(3) where there is direction for EPA to
consider establishing a definition of  model year that prevents stockpiling. At the same time, we
received other comments related to production periods and model year, leading us to adopt a
collection of related provisions in §1068.103.

       The new text in §1068.103 includes three main provisions that are already in place for
motor vehicles and heavy-duty highway engines in §§85.2304 and 85.2305.  First, we are
clarifying that the scope of a certificate of conformity may be limited to established engine
models, production periods, or production facilities.  Any such limits would be included in the
manufacturer's application for certification or in the certificate of conformity. Second, we are
defining the limits on selecting production periods for purposes of establishing the model year.
Third, we are clarifying that engine manufacturers may start producing engines after they submit
an application for certification and before the certification is approved. This includes provisions
                                          1-50

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


to address the manufacturers' responsibility to ensure (1) that engines are not introduced into
U.S. commerce or offered for sale until the certification is approved, (2) that all engines are
assembled consistent with the certification, including any changes that may have come from the
certification review process, and (3) that manufacturers make these early-production engines
available for production-line testing or selective enforcement audits, as appropriate.

       In addition, we are adding provisions to establish limits on stockpiling for engine
manufacturers.  We are doing this by stating that manufacturers must use their normal inventory
and assembly processes for initiating assembly of their engines. We include a clarifying
expectation that we would expect normal assembly processes to involve no more than one week
to complete  engine assembly once the crankshaft is installed. We understand that assembly
processes in some cases are more complicated, and that engine manufacturers may  be unable to
complete engine assembly in some cases based on delivery of certain components.  To put some
boundaries on these exceptional situations, the regulation specifies a presumption that the engine
manufacturer has violated the stockpiling prohibition if engine  assembly is completed more than
30 days after the end  of the model year.  This presumption date  is 60 days after the end of the
model year for engines with per-cylinder displacement above 2.5 liters.  This generally
distinguishes engines that may have relatively high sales volumes (including heavy-duty
highway engines) from bigger engines that are only sold in lower sales volumes.

       Two additional provisions are intended to minimize potential burden and disruption
related to transitioning to new model years. We specify that the restrictions related to date of
manufacture and model year do not apply if there is no change  in emission standards for the
coming model year. We are also  including hardship provisions to allow manufacturers to request
approval to extend the final assembly deadline for their engines if circumstances outside their
control prevent them  from completing engine assembly in time. We would approve such a
request only if the manufacturer could not have avoided the situation and took all possible  steps
to minimize  the extent of the delay.

       Note that we are also clarifying in the standard-setting parts that the certificate is valid
starting with the indicated effective date, but that it is not valid for any production after
December 31 of the model year for which it is issued.  We are also adopting a provision to
preclude issuance of certificates after December 31 of a given model year.  This will avoid a
situation in which a manufacturer receives certification after it is no longer valid for further
production.

       Finally,  note that we are adopting a provision specifying that imported products may not
have a model year more than one year earlier than the calendar year of importation, as described
in Section 2.10.3. We proposed this in part 1054 for Small SI engines and requested comment
on including it in part 1068 for all nonroad engines. Manufacturers generally had no  objection to
expanding the scope of this provision to other categories of nonroad engines. We are therefore
adopting this provision in §1068.360.

       We understand Impco's interest in making arrangements to sell engines  once they have
submitted an application for certification for a given engine family.  However, making such
commitments to supply products before the certification is approved would put EPA in an
                                       1-51

-------
                                Chapter 1: Rulemaking Process and Cross-Program Issues
difficult position if the application included significant shortcomings. If the manufacturer would
need to do further testing, modify the engine design, or make other changes to adequately
demonstrate compliance with applicable requirements, there could be substantial delays in the
certification process. During this time, the manufacturer would likely insist on accelerating the
approval because of their premature business commitments. We believe this could interfere with
the normal review process.  Furthermore, the Clean Air Act prohibits selling or offering to sell
engines that are not yet covered by a certificate of conformity, so it is not clear how we could
create such an allowance that is consistent with the Act.

       Finally, we note that we are adding a new paragraph to §1054.601 to clarify how engine
manufacturers can sell engines after the end of the model year.  This text does not change the
prohibition in 40 CFR 1068.103(f) against engine manufacturers deviating from normal
production and inventory practices to stockpile engines with a date of manufacture before new or
changed emission standards take effect. It does add a requirement that manufacturers get our
prior approval for model years in which emission standards change if their normal practice for
producing engines includes maintaining engines in inventory for some engine families for more
than 12 months.  Manufacturers would be required to show that this is necessary and consistent
with their normal business practice.  They would also be required to include relevant inventory
and production records from the preceding eight years.

1.5.3   Retailer liability

What Commenters Said:

       OPEI shares EPA's concerns that many U.S. importers, distributors and retailers currently
do not fully appreciate that they can be found responsible and liable for selling non-compliant
small engines. Section 203(a) of the CAA prohibits any "person" "causing" the importation of
non-compliant or uncertified engines or vehicles. U.S. retailers that purchase non-compliant or
uncertified engines from an importer may be found responsible for "causing" the importation of
illegal products.  The final regulations and preamble discussion should make it clear that retailers
selling non-compliant products may be subject to the enforcement provisions set forth in
Sections  113, 204 and 205 of the Clean Air Act.

       Analogous case law holds that retailers act as de facto "importers" if they are inducing
and causing the importation. See Terry Haggerty Tire Co. v. United States, 899 F.2d 1199 at
1200 (Fed. Cir.  1990). In this case, the Court found that a tire retailer, who merely purchased
goods from a Canadian company, but did not arrange for or participate in the shipment or
importation of the goods, had caused the sale sufficient to be held as the importer.

       In California, retailers and  distributors, as well as equipment manufacturers, are
potentially "strictly liable" for offering for sale or selling non-compliant small-engines or lawn
and garden equipment.  In  assessing liability, California ARB looks at each individual case to
determine which parties were principally at fault in causing the violation. California ARB's
general enforcement policy has been to not impose penalties on innocent manufacturers or
retailers who undertook "reasonable prudent precautions" to ensure they are selling certified and
compliant products. However, if, for example, a retailer purchased products that are less than
                                          1-52

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


half of the purchase price of the normal, low-end range of a certain category, California ARB
will question whether that retailer failed to take any precautions whatsoever, and should have
known he was buying suspect, non-compliant products. The California ARB enforcement
approach has provided a much-stronger deterrent than the current EPA program vis-a-vis the
offering for sale of non-compliant products.

      In response to EPA's concerns about duplicative certifications being filed, even the
equipment manufacturers, retailers and distributors that have received substantial penalties under
California ARB settlements are not re-testing or re-certifying products with an emission label.
However, these retailers are incentivized to buy compliant products from reputable, rather than
"fly by night" companies, and to contractually require their suppliers to sell only certified and
emission-compliant products, subject to indemnifications for any violations. EPA should
develop policies that will achieve these same incentives in the national marketplace. OPEI urges
EPA to pull ahead and make effective in 2007 the regulatory clarifications that retailers and
distributors that "cause" a prohibited act are potentially liable parties.

      In addition, various OPEI member companies sent in separate letters encouraging us to
take a position consistent with the OPEI recommendations described above.

Letters:
Commenter
OPEI
Stihl
Honda
John Deere
Briggs & Stratton
Document #
0675
0767
0767
0767
0767
Our Response:

       We agree with the suggestion from OPEI to include in the regulations a clear statement
that we consider it a violation to cause someone to commit a prohibited act. The preamble to the
final rule also describes the basis and context we would consider for evaluating possible
"causation" violations.

1.5.4   Defect reporting

What Commenters Said:

       OPEI appreciates the changes for defect reporting and believes that the changes have
leveled the requirements between large and small volume manufacturers with no negative impact
on the environment. OPEI also agrees with the effective date of 2009 for Small SI.

       MIC commented that § 1068.501(a)(l)(ii) should be modified to delete the reference to
"connectors" for which no permeation standards or test procedures have been defined.
                                       1-53

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
       GE commented that EPA should not extend the tracking requirements for other nonroad
engines regarding defect investigation and reporting to locomotives and locomotive engines but
should instead retain the approach currently used under part 92. GE also restated detailed
comments made during the recently finalized locomotive rulemaking.

Letters:
Commenter
GE
MIC
OPEI
Document #
0679
0701
0675
Our Response:

       We disagree with the comment from MIC to eliminate fuel line connectors from the list
of evaporative emission components subject to defect reporting.  While we do not have separate
emission standards for connectors, defective connectors have the potential to result in emissions
just as significant as defective fuel lines.  By including them on the list we are merely requiring
that the manufacturer investigate defects when they become aware of them and report them to us.
This is not an overly burdensome requirement.

       GE's comments were addressed in the locomotive rulemaking by revising the reference
to §1068.501 in 40 CFR 1033.601.

1.5.5  Delegated assembly

What Commenters Said:

       EMA commented, and as they have discussed their comments with EPA at length, that
the existing delegated assembly provisions applicable to land-based nonroad engines are
inadequate. Because they include a provision not found in any other mobile source regulation -
the mandatory requirement to include the price of the aftertreatment with the price of the  engine,
they create a major economic penalty for manufacturers. By requiring the price of the
aftertreatment to be included in the price of the engine, importers have to pay an import duty on
the price of the aftertreatment even if the  aftertreatment is manufactured in the United States.
Similarly, for aftertreatment systems that are imported separately from the engine, the duty is
paid twice. The proposed allowance for importers to segregate the cost of the separately  shipped
components to avoid duplicate duties is not acceptable and, in any event, is not within EPA's
regulatory authority.

       EMA continued to comment that it is not fair, appropriate or necessary for EPA to
impose this requirement on  land-based nonroad engines. In fact, EPA currently provides an
option in the delegated assembly provisions applicable to heavy-duty on-highway engines.
Manufacturers either can include the price of the aftertreatment with their engine, or they can  opt
to exclude the cost of the aftertreatment and meet certain audit requirements. EPA should
finalize a similar provision for nonroad engines. In addition, nonroad engine manufacturers
                                          1-54

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


should be provided with the delegated assembly flexibility being proposed for Small SI engine
manufacturers associated with the sale of engines through distributors.

       EMA commented that both of the changes recommended above are needed and
appropriate for land-based nonroad engines and can be implemented by EPA within the scope of
the existing regulatory process.

       In response to draft regulations that address the concerns described above, EMA
suggested clearer language stating that air filters are not subject to delegated-assembly
requirements if they are not specified by part number in the manufacturer's application for
certification.

       EMA (0738) commented that we should include preamble language to clearly describe
when air-intake systems are subject to delegated-assembly requirements.  They also requested
that we delay the labeling requirements associated with delegated assembly for heavy-duty
highway engines until 2010.  In addition, they requested that we not require manufacturers to
investigate assembled engines after an initial, successful audit.

       IMPCO requested that we clarify what records would be appropriate for showing that
parts were randomly collected to prepare for production-line testing with engines that participate
in delegated assembly.  They also objected to the requirement for the certifying manufacturer to
get written confirmation that an equipment manufacturer has ordered the appropriate
aftertreatment devices and to the inclusion of air filters in the delegated-assembly provisions.
They believe these provisions are redundant with all the other requirements for documentation
and verification.

       IMPCO commented that revoking the exemption should be narrowly related to the
equipment manufacturer and engine family that were the subject of a violation. They also
objected to the regulatory provision assigning liability for in-use compliance to the  engine
manufacturer, stating that it would be irrational to expect the engine manufacturer to be
responsible for anything past delivery to the equipment manufacturer, assuming the auditing and
other applicable provisions have been followed.

       EMD provided comments arguing against applying the proposed §1068.260 to
locomotives.  They also opposed applying §1068.260(b)(4) and (b)(6) to C2 marine engines.

Letters:
Commenter
EMD
EMA
EMA
EMA
IMPCO
Document #
0687
0691
0738
0818
0812
                                       1-55

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
Our Response:

       It is important to begin by emphasizing that the delegated assembly provisions do not
represent additional requirements, but rather are voluntary provisions intended as a flexibility to
manufacturers.  With respect to liability, we expect engine manufacturers to take appropriate
"steps to ensure that all engines will be in a certified configuration when installed by the
equipment manufacturer." Manufacturers that do not believe they  can effectively ensure that
engines are in a certified configuration when installed by the equipment manufacturer should not
use these delegated assembly provisions. We believe that holding  engine manufacturers liable
for the final assembly is the best way to ensure they take appropriate steps to prevent problems.

       We do not agree that pricing engines and aftertreatment together is an unreasonable
requirement. In fact, it is not a regulatory requirement under the current program as much as a
constraint on exercising an allowance to depart from regulatory requirements.  If the pricing
provisions are more burdensome than shipping engines with aftertreatment devices, then
manufacturers could simply choose not to participate in delegated assembly.  For those
manufacturers wanting to pursue delegated assembly, we believe the pricing requirement is
important in preventing vehicle or equipment manufacturers from being in a  situation where they
would gain a financial advantage by installing engines without the proper emission controls in
place.  We have confirmed with the U.S.  Customs and Border Protection that inappropriate
payment of import duties for components that are not shipped with a given engine can be
avoided with documentation showing that the price of the engine includes a charge for
components that are not included in that particular shipment. This could most easily be
accomplished by itemizing the invoice to identify the value of the missing components relative to
the value of the rest of the engine. The regulations now include these specific instructions
regarding  invoicing with respect to import duties.

       We understand that engine manufacturers have competing interests both to maintain the
ability to arrange flexible assembly procedures and agreements, and to ensure that their engines
are introduced into commerce only after being assembled in the certified configuration.  We
share those objectives and believe the regulations serve the purpose of creating a framework for
balancing these different concerns. By applying these provisions in the regulations,
manufacturers will not find themselves in a situation where competitiveness concerns cause them
to take steps to reduce costs at the risk of producing noncompliant products.

       We agree with EMA, however, that it would be appropriate to apply the delegated-
assembly framework for heavy-duty highway engines to other nonroad engines. The main
difference between programs is the allowance for heavy-duty highway engines to rely either on
the pricing strategy described above or on audits of vehicle manufacturers, but not necessarily
both, to ensure that installed engines are in the certified configuration. While we are concerned
about the incentive for vehicle and equipment manufacturers to gain a financial advantage if
aftertreatment components are not priced together with the engine, we believe that requiring
engine manufacturers to confirm that vehicle or equipment manufacturers have ordered the
required aftertreatment components and to perform audits of vehicle or equipment manufacturers
is generally sufficient to provide the proper assurances that engines are being properly assembled
and installed. Conversely, we believe that pricing aftertreatment and engines together is a strong
                                          1-56

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


enough assurance of proper assembly and installation procedures that audits are not always
necessary as an additional oversight measure. We note that these provisions spell out a
minimum level of oversight for engine manufacturers. There may be instances, such as a new
relationship with a vehicle or equipment manufacturer or some other reason to have less
confidence in proper assembly procedures, where the engine manufacturer would want or need to
take extra steps to ensure that engines are assembled properly.

       We recognize the inconvenience of requiring engine manufacturers to obtain written
confirmation that the equipment manufacturers have ordered aftertreatment components before
shipping the engines. Thus, the final regulations only require this be done for the initial
shipment.  This will significantly limit the inconvenience, while ensuring that the equipment
manufacturers understand their obligations before they start receiving engines.

       We believe there is a strong advantage in implementing requirements uniformly across all
the engine programs, both for EPA and for manufacturers.  Aside from the pricing and auditing
requirements described above, we are making the following provisions part of the final program,
which were part of one or both of the separate programs in parts  85 and 1068:
    •   Auditing rates are generally set at four vehicle or equipment manufacturers per year, or
       enough to rotate through all the equipment manufacturers over a four-year period,
       whichever is less. A reduced rate may apply after several years of successful
       implementation of these requirements.
    •   We are continuing the approach already adopted to provide for a streamlined
       demonstration for integrated manufacturers where the auditing would effectively be an
       internal practice.

       In addition, we are including the following provisions in the unified approach to
delegated assembly that were part of the proposal for Small SI engines:
    •   Distributors may participate in delegated assembly, but only to the extent that they act as
       equipment manufacturers, adding aftertreatment devices before shipping the engines to
       vehicle or equipment manufacturers.  Allowing distributors to further delegate engine
       assembly to another set of companies raises fundamental  questions about the ability of
       engine manufacturers to adequately ensure proper final assembly of their engines.  We
       are making a temporary allowance for this for Small SI engines to accommodate the
       transitional provisions allowing equipment manufacturers to gradually work toward
       making Phase 3 products.
    •   If engine manufacturers design their air-intake systems such that they depend on specific
       parts (identifiable by part number) to achieve proper air flow through the engine, that
       raises concerns that are similar to aftertreatment devices.  In fact, we are currently
       pursuing an enforcement case where an equipment manufacturer did not follow the
       engine manufacturer's directions to use a specific air filter.  We are specifying that air
       filters identified by part number must be included in delegated assembly, though we
       require audits related to air filters only if audits are already occurring for exhaust
       systems. If manufacturers specify intake air systems by performance parameters such as
       maximum pressure drop across the air filter, the delegated-assembly provisions do not
       apply.  This is similar to the way we have treated exhaust components for systems not
       requiring exhaust aftertreatment. See §1068.260(a).
                                       1-57

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
   •   Vehicle or equipment manufacturers submitting annual affidavits must include a count of
       aftertreatment devices received to verify that there were enough of the right models of
       aftertreatment devices for the number of engines involved.
   •   Engines need to be labeled to identify their status as delegated-assembly engines, either
       with a removable label or with "Delegated Assembly" noted on the engine's permanent
       label.  This ensures that engines will not be introduced into commerce without an
       indication of their status relative to the certified configuration.
   •   Engine manufacturers must confirm that vehicle or equipment manufacturers have
       ordered aftertreatment devices corresponding to an engine order, but this confirmation is
       limited to the initial shipment of engines for a new certification and may occur up to 30
       days after the engines have been ordered.
   •   For engines subject to requirements for production-line testing  or selective enforcement
       audits, we specify that aftertreatment components must be randomly procured. We agree
       with the suggestion in the comments to broaden the allowance for randomly procuring
       components. As long as manufacturers use a method to randomly select components that
       are appropriate for the particular engine configuration, these components may come from
       any point in the normal  distribution chain.

       We agree that the labeling requirements are new for heavy-duty highway engines and are
therefore allowing until the 2010 model year for manufacturers to start meeting requirements for
these engines.

       We agree that delegated assembly provisions do not apply for components that are not
emission-related.  See §1068.260(b). However, we disagree with IMPCO's comment to exclude
air filters. Nevertheless, §1068.261(e) describes a less burdensome approach for air filters.  We
are including preamble language to further clarify the distinction between intake systems that are
specified in the application for certification by part number or by performance specification.

       We believe it is not appropriate for the regulations to specify that a single audit showing
proper assembly procedures is a sufficient basis for discontinuing future audits. We are
concerned that engine and equipment manufacturers must have an extended period of complying
with these provisions with significant communication, oversight, and verification to ensure that
engines are being assembled properly.  The regulations in all our programs (proposed and
adopted) have specified a reduced auditing schedule only after cycling through two four-year
auditing periods. We continue to believe this is appropriate for the universal program.

       With respect to production-line testing, we would expect manufacturers to need to keep
simple records describing the algorithm used to make the selection.

       We have modified parts 89 and 94 to allow manufacturers to use the delegated-assembly
provisions and other provisions related to partially complete engines for land-based nonroad
diesel engines and marine diesel engines. It is not likely that these engines will be using
aftertreatment devices to meet current standards, but there may be circumstances where this may
apply.  The new regulatory provisions also clarify exemptions and special provisions that apply,
for example, for within-company shipments between facilities and for shipping engines without
certain engine components.
                                          1-58

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
       We acknowledge that manufacturing processes and regulatory requirements for
locomotives and locomotive engines warrant special treatment with respect to shipment and
assembly of aftertreatment devices. We have therefore adopted such provisions in 40 CFR part
1033 that apply specifically to locomotives instead of the delegated assembly provisions we are
adopting in §1068.261.

1.5.6   Engine rebuilding

What Commenters Said:

       Cummins requested clarification in the rebuild requirements of Parts 89 and 1039 via Part
1068 that a rebuilt engine may be used to replace any equivalent engine regardless of model
year. (i.e. a rebuilt Tier 1 engine may replace a TPEM engine or an AB&T engine (assuming
they are identical) of a later model year. They noted that this issue has been raised within
Cummins particularly with large engines (>560kW) where there are often times fleets of vehicles
that are supported with spare engines owned by either the mine owner or by a Cummins
distributor. At the time of engine failure, or rebuild, Cummins believes it is acceptable to replace
engines on a "like-for-like" basis regardless of model year. Because of TPEM and AB&T
flexibility, the use of the term model year in 1068.120 could be interpreted to not allow this like-
for-like replacement. Therefore, Cummins requested a revision to the regulatory language for
§1068.120 as noted below. Cummins also commented on §89.1003 and §1068.240, as noted
below, and commented that those changes may not be needed given that those sections deal with
new replacement engines. However, Cummins commented that they would like EPA to consider
those changes as well.

       1068.120 : Rebuilding
       (f) If the rebuilt engine replaces another certified engine in a piece of equipment, you
       must rebuild it  to a certified configuration of the same model year or Tier level as, or a
       later  model year or Tier level than, the engine you are replacing. In circumstances
       involving a TPEM engine or an engine certified to a later Tier level using  AB&T credits,
       the engine is considered interchangeable with a previous Tier engine for the purpose of
       installation in a piece of equipment as long as the engine is identical in all material
       respects to the engine being replaced.

       89.1003(b)(7): New Replacement Engines
       (v) Where the replacement engine is intended to replace an engine that is certified to
       emission standards that are less stringent than those in effect when the replacement
       engine is built,  the replacement engine shall be identical in all material respects to a
       certified configuration of the same or later model year or Tier level as the  engine being
       replaced. In circumstances involving a TPEM engine or an engine certified to a later Tier
       level using AB&T credits, the  engine is considered interchangeable with a previous Tier
       engine for the purpose of installation in a piece of equipment as long as the engine is
       identical in all material respects to the engine being replaced..
                                       1-59

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
       1068.240(a)
       (5) You make the replacement engine in a configuration identical in all material respects
       to the engine being replaced (or that of another certified engine of the same or later model
       year). This requirement applies only if the old engine was certified to emission standards
       less stringent than those in effect when you produce the replacement engine. In
       circumstances involving a TPEM engine or an engine certified to a later Tier level using
       AB&T credits, the engine is considered interchangeable with a previous Tier engine for
       the purpose of installation in a piece of equipment as  long as the engine is identical in all
       material respects to the engine being replaced.

       California ARB  encourages EPA to address the status of remanufactured engines as being
fully subject to the requirements  for rebuilt engines in 40 CFR  1068.120, and to adopt labeling
requirements for rebuilt engines similar to those in California's Off-Road Diesel Regulation at
13 CCR 2423(1).

       OPEI commented that their units do not have hour meters and an exact hour of use is not
always possible.  OPEI  suggests  language in section  1068.120(j)(l) be reworded to indicate
approximate hours of use or in-use service time and method used to determine such estimates.

       EMA commented that the proposed provisions regarding engine rebuilding are acceptable
but incomplete because  they do not adequately address the difference between model year
requirements and emission standard requirements. In the final rule EPA should clarify that an
engine that is rebuilt may be used to replace any equivalent engine model regardless of the model
year of the equipment.

       EMA recommended that we include a definition for "rebuilding."

       OPEI believes EPA may be creating burdens  on industry segments unaware of this rule
and incapable of providing the amount of burdensome records required by this part.  OPEI
proposes EPA exempt engines/equipment subject to part 1054 from this provision.

       EMA commented that Small SI engines are significantly different than larger engines
currently regulated under §1068.120. The  listed requirements must either be excluded for Small
SI engines or modified to allow the appropriate requirements for Small SI engines.
Letters:
Commenter
Cummins
California ARB
OPEI
EMA
EMA
Document #
0719
0682
0675
0691
0808
                                          1-60

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
Our Response:

       We agree with Cummins that the regulations should more carefully address rebuild
requirements for special cases such as spare engines for maintaining a fleet, engines with Family
Emission Limits above  or below the standards, and engines produced under the Transition
Program for Equipment Manufacturers.  We have revised the regulations in §1068.120(f) to
describe how the rebuilding provisions apply under several such scenarios, consistent with the
approach recommended by Cummins.

       We agree with California ARB that remanufactured engines are generally subject to
rebuilding requirements. We note, however, that remanufactured engines might become "new
engines," for example, if they are installed in new vehicles or equipment.  In this case they would
need to be certified before being introduced into commerce like any other new engine.
Remanufactured engines that qualify as "new engines" are typically used as replacement
engines. As such the provisions for the replacement-engine exemption  in §1068.240 would
apply, including the labeling requirements, as described in Section  1.5.7.

       We agree that the rebuilding provisions should acknowledge that approximate service
hours (or miles) are adequate for engines without hour meters (or odometers) and have changed
the regulation accordingly.

       We agree with EMA that a rebuilt engine may be used to replace any equivalent engine
model regardless of the model year of the equipment and have changed the regulation
accordingly.

       The current regulations include language describing what qualifies (and does not qualify)
as rebuilding. Absent any specific recommendations, we believe it is not appropriate to more
carefully define what constitutes engine rebuilding.  Moreover, the same requirements (except
recordkeeping) generally apply for routine maintenance as for rebuilding so we believe it is also
not necessary to add this kind of clarification.

       We understand the concerns raised by OPEI for small spark-ignition engines. Engine
repairs for a single-cylinder engine might involve replacing the piston or piston rings. This
should still be done such that an engine remains in its certified configuration, but we agree that
someone might easily do this maintenance without realizing that they have triggered a regulatory
requirement. This is especially true for lawn and garden applications, but the same dynamic
applies for small outboard engines and small engines used with recreational vehicles. We are
therefore revising the regulation to waive the recordkeeping requirements for spark-ignition
engines below 225  cc. Larger engines are much more commonly used in commercial
applications where operators and repair professionals would be more  likely to have maintenance
and rebuilding practices that resemble those for diesel engines or for larger spark-ignition
engines.
                                       1-61

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
1.5.7   Replacement engines

What Commenters Said:

       EMA commented that the existing replacement engine exemption was unworkable for
partially complete engines. They recommended a separate replacement-engine exemption for
partially complete engines, including any engine assembly in which some components necessary
for engine operation were missing. They recommended that such engines could be used for
replacement purposes with no restriction other than labeling the engines to identify them as
"service use only". They also recommended that we define the terms "replacement engine" and
"destroyed."

       In response to draft regulatory language describing an allowance for a limited number of
partially complete replacement engines without requiring the usual tracking,  demonstration,
recordkeeping, etc., EMA commented that these provisions should also apply for engines
between 2.5 and 7 liters per cylinder, and that they should apply for marine diesel engines. They
also noted that the labeling requirements in §1068.265 appeared to be in conflict with the
labeling specifications in §1068.240.

       ECO responded to questions about industry practices for replacement engines. They
described Marine SI manufacturers as sending long blocks out only on an as-needed basis. There
are some marinas willing to pay the flooring charges to keep blocks immediately available, but it
sounds like this is probably the exception to the rule. Regardless, ECO noted that for the small
manufacturers that might sell 200-300 engines per year, the 0.5% allowance is only 1 to 1.5
engines. ECO commented that this is not much. Based on input from Kodiak Marine Parts,
ECO commented that 1% would probably be better, as it would give a little extra flexibility.

       IMPCO asked whether §1068.240 applies only to engines that are specifically built as
replacement engines, or to all engines that might be used as replacement engines, and noted that
they do not build replacement engines for engines that are not in production.  IMPCO noted, for
example, that if the block cracks, the engine would typically be 'replaced' by  bringing the piece
of equipment to a service facility, the service facility orders an engine (block plus crankshaft)
from the OEM or a warehouse, they remove the fuel system components, install the new block,
and install the other, used fuel system components. IMPCO also asked how the OEM is
supposed to know whether the short block is used for a currently certified engine family.
IMPCO noted that the proposed regulations state that the partially complete engine exemption is
not valid when engines are shipped to a non certificate-holder.  IMPCO questioned whether
service engines are covered? Finally, IMPCO also questioned why an exemption is needed to
sell a replacement engine from their current production.

       IMPCO also asked if an engine produced by a secondary engine manufacturer is
considered a New Replacement Engine once assembly is complete at its facility or after it is
installed in the equipment.  They stated that "the standard-setting part defines a New Nonroad
Engine as the time when it is fully assembled for the first time."
                                          1-62

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


      Some marine engine manufacturers pointed out that that boat owners sometimes want to
upgrade their engines, not because they are at the end of their life but because they want more
power, additional features, or new technologies. They suggested that manufacturers should be
able to supply noncompliant replacement engines to meet the demand for these engines.

      Kubota commented that they would like EPA to advise Kubota on the proper wording for
a replacement engine label that is to be used on a "replacement engine" that is going in place of a
failed TPEM (Flex) engine.  From the statement of the replacement engine label, Kubota cannot
legally install this label onto a TPEM engine. Per the CFR and CCR, the statement would be as
follows; "THIS ENGINE COMPLIES WITH CALIFORNIA OFF-ROAD AND U.S. EPA
NONROAD EMISSION REQUIREMENTS FOR 2004 ENGINES UNDER13 CCR 2423(j)
AND 40 CFR 89.1003(b)(7). SELLING OR INSTALLING THIS ENGINE FOR ANY
PURPOSE OTHER THAN  TO REPLACE AN OFF-ROAD ENGINE BUILT BEFORE
JANUARY 1 2008 MAY BE AVIOLATION OF CALIFORNIA AND FEDERAL LAW
SUBJECT TO  CIVIL PENALTY." Because their TPEM engines would be produced after
January 1, 2008, they have a conflict.

      Kubota suggested wording be something like: "THIS ENGINE COMPLIES WITH
CALIFORNIA OFF-ROAD AND U.S. EPA NONROAD EMISSION REQUIREMENTS
FOR 2004 ENGINES UNDER13 CCR 2423(j) AND 40 CFR 89.1003(b)(7).  SELLING OR-
INSTALLING THIS ENGINE FOR ANY PURPOSE OTHER THAN TO REPLACE AN OFF-
ROAD ENGINE BUILT BEFORE JANUARY 1 2008 OR A FLEXIBILITY PROGRAM
ENGINE MAY BE AVIOLATION OF CALIFORNIA AND FEDERAL LAW SUBJECT TO
CIVIL PENALTY."

      Cummins asked that the following sentence be added to §1068.240(a):  "In circumstances
involving a TPEM engine or an engine certified to a later Tier level using AB&T credits, the
engine is considered interchangeable with a previous Tier engine for the purpose of installation
in a piece of equipment as long as the engine is identical in all material respects to the engine
being replaced."

Letters:
Commenter
Cummins
Kubota
ECO
IMPCO
EMA
EMA
Document #
0719
0744
0802
0812
0808
0809
Our Response:

      While we do not necessarily agree with EMA's assertion that previous version of the
replacement engine exemption would be unworkable for partially complete engines, we do
believe that the approach being finalized is more appropriate. Under the revised approach, which
is intended to address EMA's concerns, manufacturers will be allowed to produce replacement
                                   1-63

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
engines (including partially complete engines) with less tracking.  See section VIII.C.5 of the
preamble and §1068.240 of the regulations for detailed descriptions of this allowance. This
allowance will be limited for each subcategory to no more than 0.5 percent of a manufacturer's
annual sales for that subcategory. We agree with EMA that this allowance should apply for
engines between 2.5 and 7  liters per cylinder.

       We disagree with the comment suggesting a high sales/production limit (1.0 percent
instead of 0.5 percent).  The comment noted that this might be especially appropriate for small
volume manufacturers.  We note that manufacturers are always allowed to produce and sell an
unlimited number of replacement engines if they track the engines and take possession of the old
engines. This would be a workable alternative for a manufacturer producing only a handful of
replacement engines each year.

       It is important to note that these provisions are intended to allow for replacement of
engines that fail prematurely where all of the following is true:

       - The engine cannot reasonably be repaired or rebuilt.
       - A different used engine (including rebuilt engines) cannot be used.
       - No new certified engine can be used.

No matter which path the engine manufacturer uses under §1068.240, the provisions may not be
used to circumvent emission  standards that apply to new engines.  Thus, boat owners are not
allowed to use replacement engines to upgrade their engines because  they want more power or
new features.

       With respect to IMPCO's comments, the new replacement engine allowance is intended
to allow a  small supply of replacement engines where they are legitimately needed,  and not
necessarily to allow the continuation of all current business practices  without any changes.
Under the  regulations being adopted, IMPCO would be able to provide replacement engines
through service facilities. However, this may require a new administrative process.  We have
added a new paragraph to 40 CFR 1048.601 to clarify how this would work for Large SI engines.
With respect to the comment about current-tier engines, a replacement engine exemption is
needed only in the case of partially complete engines that are not in their certified condition
when introduced into commerce.

       It appears that IMPCO is confusing the definitions of "new", "engine", "replacement
engine", and "date of manufacture". An engine produced by a secondary engine manufacturer is
first considered an "engine" when the crankshaft is installed in the block (which occurs before it
reaches the secondary engine manufacturer). It is also considered "new" from that point until
title is transferred to the ultimate purchaser. However, the engine is generally considered to
have a date of manufacture based on when it arrives at the secondary  engine manufacturer.
Whether or not it is considered a "replacement engine" has no bearing on the meanings described
above.

       In response to EMA's request for clarification, we  note that destroying an engine
generally means to crush, melt, or otherwise modify the engine block so that it cannot be reused
                                          1-64

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


as an engine.  Alternatively, you may destroy an engine by modifying it to destroy its original
identity and make it a new engine.

       We have revised the regulatory text to address Cummins' concern, but are not using their
recommended text.

       Finally, we have revised the labeling provisions to address concerns such as those raised
by Kubota.

1.5.8   List of emission-related components

What Commenters Said:

       MIC commented that it is proposed that "All components comprising the combustion
chamber, including the piston, piston rings, block, head, and valves" be added to the existing list
of "emissions-related components" listed in Appendix I to Part 1068. The inclusion of these
basic engine components would be inconsistent with the definition of "emissions-related
components"  contained in section III of Appendix I, which is "any other part whose only purpose
is to reduce emissions or whose failure will increase emissions without significantly degrading
engine performance." Clearly, pistons, piston rings, the block, the head and the valves do not
have emissions control as their only purpose. Because the "failure" of any of these parts will
clearly degrade  engine performance, they should not be added to the list of emissions-related
components.  Adding such components would also be inconsistent with the way on-road vehicles
are treated under the Clean Air Act. The implications of this change  are so great that there would
need to be a careful evaluation of the cost before proceeding.

       Arctic Cat commented on Appendix I to Part 1068 - Emission-Related Components
(I)(5). This change greatly expands the definition of emissions related components to include the
following: "All components comprising the combustion chamber, including the piston,  piston
rings, block, head, and valves." Arctic Cat recognizes this as a major departure from past EPA
policy and is not consistent with how other categories are treated.  It is also their understanding
that this section is not consistent with the Clean Air Act. Arctic Cat does not feel a change of this
magnitude it appropriate here given the absence  of any discussion with industry on the nature of
failures of the newly added components or their potential effect on emissions. In the recreational
category failures of the newly  added components have unique causes and effects which should
be studied before proceeding with this policy change.

       OPEI  commented that the list of emission components is too general and can lead to
broad misinterpretation as to what is covered. OPEI suggests EPA use the California ARE list.

       EMA  commented that EPA proposes to expand the emission related parts list (as defined
in Appendix I to Part 1068) to include components comprising the combustion chamber,
including the  piston, piston rings, block, head, and valves. This expansion of the emission
related parts list is not justified as those parts are critical to the basic function of the engine.

       EMA  also  commented on Appendix I to Part 1068 - Emission-Related Components.
                                       1-65

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
The emission related parts list as defined in Appendix I to Part 1068 is proposed to include
components comprising the combustion chamber, including the piston, piston rings, block, head,
and valves. This expansion of the emission related parts list is not justified as these parts are
critical to the basic function of the engine.

       Impco commented that we should add clarifying language to the list of components
related to evaporative emissions to take into account the fact that the standards do not apply to
engines fueled by natural gas or LPG.

Letters:
Commenter
MIC
Arctic Cat
OPEI
EMA
Impco
Document #
0701
0709
0675
0691
0812
Our Response:

       We agree with the commenters that it is not necessary or appropriate to include the
proposed changes to the list of emission-related components. We are not adopting the proposed
changes.

       We also agree with Impco's input regarding natural gas and LPG engines. However, we
have addressed this concern by adding a note to the introduction to this appendix to clarify that
the list of components does not make parts "emission-related components" if the equipment in
which those components are installed is not subject to evaporative emission standards.

1.5.9   Export exemption

What Commenters Said:

       Honda recommends that EPA apply the requirement for exemption labels ONLY when
there is a reciprocal agreement between the United States and the other country to accept the
EPA regulation and labeling as demonstrating full compliance with that other country's emission
regulations. Honda also requests that EPA develop and provide a continuously updated list of
these countries and sufficient lead-time to comply with these requirements. Finally, Honda
recommends that the final rule provide another section under engines for export to allow a
manufacturer in full  control of engines or products destined for another part of the world to forgo
the temporary exemption label on individual engines or product and boxes.  Small engines and
the equipment they power are not only valuable national industries but global  industries with a
positive contribution to U.S. international trade. The ability to build engines and products in the
U.S., exempt  from EPA regulation but complying with the regulations in another country, is an
important option that exists today. Honda understands that the purpose of the  exempt label is to
prevent inadvertent introduction  of these export engines or equipment using these engine into
U.S. commerce. However, they believe the proposed label language has significantly greater
                                          1-66

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


content than is necessary to convey the message that the engines are for export and exempt from
U.S. EPA regulations. Also, in many cases the engines/equipment being exported are labeled for
compliance in another country or area, such as Europe. The combination of another country's
compliance label and a label stating that the engine is "not for sale in the U.S." would be more
likely to convey some meaning to a potential label reader and it will still be useful (and less
likely to be misunderstood) when the engine is in another country. Furthermore, a temporary
exemption label is not the only means of assuring that engines or products are exported and not
introduced into U.S. commerce. If a manufacturer has full control of the engines, or equipment
with these engines, from the point of final assembly until they have left the U.S., a temporary
label would serve no purpose.

OPEI provided the following comments related to the export exemption:
    •  It is not clear why §1068.230 is required at all.  At a minimum OPEI believes
       §1068.230(a) needs revision. The term "with emission standards identical to ours"
       creates conflict with European regulation which may have identical emission standards
       but are not identical because of the lack of an ABT program, exemptions etc. OPEI
       suggests EPA means to say "with emission regulations identical to ours."
    •  Regarding §1068.230(c), OPEI believes each engine does not need to be labeled or
       tagged if the following conditions apply: 1) the shipping container is marked and 2) A
       manufacturer has procedures to insure they are not sold in the US market. 3) If an
       engine contains an emission label from another country.  Otherwise, this paragraph
       should not apply.

Caterpillar expressed a concern that they (as the certifying manufacturer) appeared to be
responsible even if another company would take an engine intended for export and sell that
illegally in the United States. Caterpillar suggested that we instead identify the act of selling an
export engine in the United States as a violation.

       Impco recommended keeping the requirement for a permanent label on export engines
and suggested requiring that the label include the corporate name and trademark.

Letters:
Commenter
Honda
OPEI
Caterpillar
Impco
Document #
0705
0675
0813
0812
Our Response:

       The Clean Air Act is quite clear in describing the terms for exempting engines for export.
Specifically, the Act calls for limiting the exemption for cases in which the other country has
emission standards that differ from those that apply in the United States.  The Act also specifies
that exempt engines must be labeled. As a result, we have a consistent set of requirements for
the various categories of nonroad engines, including Small SI engines regulated under part 90.
This seems to be working today, so we do not see that there is a need for major changes to allow
                                       1-67

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
for the free flow of commerce.  Nevertheless, we are making three adjustments to the export
exemption in §1068.230 to address the concerns raised by commenters.

       First, we are establishing a streamlined path for certifying export-only engines where the
destination country has the same emission standards. In particular, we would look for
certification or other approval from the destination country as the basis for approving the export-
only certification. Also, the requirement to pay certification fees would be waived for these
engines.  This special certification would be valid only for purposes of exporting the engines and
would not be sufficient for selling the engines in the United States.

       Second, we are clarifying what requirements apply for removable (or temporary) labels.
In contrast to Honda's concern, we do not require permanent labels today for exported engines,
nor did we propose such a requirement.  For example, we specify that manufacturers can apply a
removable label to exempted engines by labeling the container where multiple engines are
packaged together. This is especially advantageous  for Small SI engines where dozens of
engines may be packaged together. In place of the removable label, manufacturers may
alternatively apply a permanent label specified by the destination country if such requirements
apply, in which case the bill of lading would need to also state that the engines must be sent to
the named destination country to avoid a violation.

       Third, we are revising the regulation to state  that the exemption expires when the engine
leaves the country. Anyone subsequently importing such an engine would therefore be guilty of
a violation, rather than the original manufacturer.  However, if at any point in the manufacturing
or distribution process such an engine is placed into  service in the United States, the certifying
manufacturer would be held responsible for the violation. Manufacturers would do well to take
steps to ensure that anyone responsible for installing or distributing such engines understands
how important it is to avoid a situation where these engines are placed into service in the United
States.

       We agree that it is necessary to clarify what provisions apply for defining when another
country has emission standards identical to ours. We would generally understand standards to be
identical  if they specify the same numerical level of the standard, the same test procedures
(including fuel specifications), and the same approach for allowing the use of emission credits.
For example, Canada currently  specifies that  Small SI engines used there must meet U.S. EPA
standards. These engines would therefore not be eligible for an export exemption and must
therefore be certified with EPA. However, Canada also specifies that a limited number of
engines used in specialty applications may meet the  Phase 1 standards rather than the current
Phase 2 standards. These engines would be exempt  from our emission standards and
certification requirements because the applicable standards are clearly different than those that
apply to Small SI  engines produced for the United States for the given model year.  If a
manufacturer requests an exemption to export engines/equipment to a country that has standards
similar but not identical to ours, we may ask the manufacturer to specify how such standards
differ from ours. We may attempt in a future rulemaking to add language to the regulation to
clarify when we would consider standards to be identical.
                                          1-68

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       Note that engine manufacturers do not need an exemption or any special approval to ship
engines that are certified and labeled for the U.S. market if some of those engines may be
diverted for export by distributors or retailers.

       Impco incorrectly pointed out that current regulations specify a permanent label for
export engines. The regulations specifically state that the label need not be permanent. This is
appropriate since there is no benefit of having the label once the engine is outside the United
States. Moreover, manufacturers have pointed out that a permanent label would be problematic
once the engine is sold and placed into service in another country. We agree that the label
should identify the manufacturer's corporate name and trademark and have changed the
regulation accordingly.

1.5.10 Manufacturer-owned exemption

What Commenters Said:

EMA commented that adding "possession" to the list of qualifying criteria for the manufacturer-
owned exemption adds  a significant constraint on manufacturer's ability to complete all the
requirements associated with introducing new products into the marketplace.

Impco similarly objected to including the word "possession" to the list of qualifying criteria for
the manufacturer-owned exemption, noting that another company should be able to execute a
development program on their behalf without triggering the need for EPA approval under the
testing exemption.

Letters:
Commenter
EMA
Impco
Document #
0808
0812
Our Response:

       The existing manufacturer-owned engine exemption specifies that "an engine may be
exempt without a request if it is a nonconforming engine under your ownership and control and
you operate it to develop products, assess production methods, or promote your engines in the
marketplace." We proposed to add the word "possession" to this language as a clarification
because we discovered that some manufacturers mistakenly believed this exemption allowed
them to provide these engines to customers for testing. Clearly, the existing language prohibits
anyone other than the manufacturer from operating the engines. Thus, this revision is not
changing the provisions that currently apply under the manufacturer-owned engine exemption;
we are adding the word "possession" simply to eliminate any confusion.

       It is a separate issue whether we should allow manufacturers to relinquish possession  of
these engines without obtaining a test exemption. However, we continue to believe that engines
exempted as manufacturer-owned engine need to be  strictly controlled since this exemption does
                                       1-69

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
not require prior EPA approval. Anyone wishing to arrange for someone else to operate engines
may ask for EPA approval under the testing exemption.

1.5.11  Other issues

What Commenters Said:

       GE commented that the proposed §1068.420 as it would apply to an engine family when
evaluated as part of a Selective Enforcement Audit (SEA) establishes failure criteria that are less
stringent for new locomotives than the criteria established in §1033.415 for in-use locomotives
that have between 50% and 75% of useful life service. The criteria in §1068.420 and Appendix
A indicates that a failure rate of 60 % or more is required to establish that an engine  family is
noncompliant whereas §1033.415 mandates a test procedure that when extended to the maximum
allowable number of 10 test locomotives logically establishes that a failure rate of 40% or more
could be considered a noncomplying engine family based on the failure rate and EPA's judgment
and consideration of other test results such as average emissions levels, existence of any defects,
and other unspecified test factors. GE believes that it is illogical and unrealistic to expect
locomotives that have been in service for 50% to 75% of their useful life to have an emissions
compliance rate greater than the EPA has established for new locomotives and recommends that
the compliance criteria for in-use testing in §1033.415 be changed to be no more stringent than
that set for new locomotives as established in proposed §1068.420 and Appendix A.

       California ARB encourages EPA to revise the provision in 1068.250(j) that would allow
the extending of compliance deadlines for small businesses for up to three years total. At most,
California ARB believes that the relief should be granted for two years in cases of extreme
hardship. A two-year period should be sufficient to provide  any manufacturer, even a small
business entity, adequate time to achieve compliance.

       California ARB also commented that the provisions of part 1068 apply to nonroad diesel
engines, large spark-ignition engines, recreational vehicles,  small spark-ignition engines, and
marine spark-ignition engines. Although a unified program may appear to be easier for
manufacturers to follow, it may be difficult to implement due to the distribution of various
engine types in different business sectors.

       OPEI commented  on the following items:
    •  §1068.110(e) suggests owners can do diagnosis and  repair themselves and charge
       manufacturer.  This needs to be reworded to clarify  that owners should file their
       warranty claims through dealers or other authorized representatives.
    •  §1068.210 conflicts with language in §1060.215(b).  The provisions in §1068.215(b)
       should apply for development and test engines.  OPEI asks whether §1068.215(b) does
       not apply to a test exemption.  When developing new products, an important component
       of the program is field testing in real world conditions. Often units are placed with
       various test crews that use them in their normal day-to-day business activities. The units
       are monitored by the manufacturer and at the end of the test period, collected, reviewed
       and usually scrapped. While the crew is generating revenues by the use of the
       equipment, the manufacturer is not. OPEI asks how they apply this exemption for
                                          1-70

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       testing/developing new engines/equipment. OPEI also asks why §1068.315 does not
       include test exemptions.
    •  OPEI requests EPA include emails as an acceptable form of owner notification of recall
       in §1068.520(a). OPEI notes that §1068.520(a)(9) requires a self-addressed card. OPEI
       asks about using an Internet site or toll-free number instead.

       EMA commented on the following proposed items.
   •   The proposed definition of engine-based test appears to relate exclusively to exhaust
       emissions that are reported in g/kW-hr.  Instead of engine-based test, the defined term
       should be "Engine-based Exhaust Emission Test" in order to differentiate between testing
       an engine for exhaust emissions and testing associated with demonstration of compliance
       with evaporative requirements that are either already in place (such as those for LSI and
       Recreational Vehicle), or those being proposed (such as those for Small SI or marine SI
       engines).
   •   The proposed definition of an incomplete engine assembly does not make sense and must
       be removed or carefully revised in a subsequent rulemaking in light of the discussion
       above regarding the definition of an engine.
   •   The proposed definition of a secondary engine manufacturer also will need to be revised,
       as necessary, based on the ultimate resolution of the definition of an engine and/or
       incomplete engine.  As such, it should be deferred to a subsequent rulemaking.
   •   The proposed language addressing the practice of engine sector changes needs to be
       modified to make it clear that an engine that changes sectors, for example from nonroad
       to stationary, retains the engine's original date of manufacture and is subject to that
       sectors  applicable standards for that original date of manufacture.
   •   The NPRM requests comments on applying any or all of the 'special compliance
       provisions' proposed for Small SI engines in Part 1054 to all nonroad engines by
       incorporation into Part 1068. The NPRM provisions for Small  SI engines include several
       topics including: warranty coverage, bonding requirements, model year naming for
       imported engines, reporting requirements, etc. that require significant discussion with the
       various affected industries covered by Part 1068 prior to implementing these changes.
       EMA is ready to work with the agency to develop the required outreach and appropriate
       changes to Part 1068 but is concerned that the addition of this burden will delay the Final
       Regulation. (See 72 FR at 28212.)
   •   EMA supports the optional early adoption of 40 CFR Part 1068 (rather than the similar
       compliance provisions in parts 89, 90, 91 and 94) but objects to making this early-
       adoption mandatory. Many of the Part 1068 changes require significant resources to
       implement, and manufacturers may desire to implement these changes at the same time as
       the emissions standards change. (See 72 FR at 28212.)

       Arctic Cat commented that sometimes a wrecked snowmobile can be fixed by replacing
the tunnel (upon which the emissions label is affixed). The snowmobile must be wrecked in just
the right way to call for tunnel replacement, otherwise it is usually totalled. A handful of times a
year Arctic Cat has customers order tunnels and they have been asked about the emissions label.
Arctic Cat assumes a new emissions label should or must be affixed as part of the repair since the
old label would be discarded with the damaged tunnel. They cannot find any rules governing
this situation. Such rules or guidance documents would be helpful in defining our policies.
                                       1-71

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
       Impco suggested that we modify §1068.105 to clarify that equipment manufacturers are
subject to penalties if they do not follow the regulations, including emission-related installation
instructions, and to add explanatory language to make it clearer what equipment manufacturers
must do.

Letters:
Commenter
GE
California ARE
OPEI
EMA
Arctic Cat
Impco
Document #
0679
0682
0675
0691
0731
0812
Our Response:

       GE's comment on the proposed §1068.420 is no longer relevant because the recently
finalized 40 CFR 1033.601 excludes locomotives from Selective Enforcement Audits.

       We intend to evaluate hardship applications submitted under §1068.250(j) on a case-by-
case basis.  We generally agree that one or two years of hardship relief should be adequate for
almost all cases.  However, we believe it would not be appropriate to change the regulation to
rule out even the possibility of considering longer hardship relief for exceptional circumstances.

       Part 1068 indeed applies broadly. The scope of part 1068 now includes all nonroad
engine categories, though some types of engines will not be subject to part 1068 until new
standards apply sometime in the future.  We believe there are strong advantages to including
these general compliance provisions in a single place in the regulations. Keeping one set of
regulations current is straightforward, since we will  not need to update parallel regulatory
provisions when we amend the regulations periodically.  We see no particular challenge in
implementing these general  compliance provisions for multiple engine categories as a result of
this approach to adopting the regulations in one location.  There are occasions where we need to
make distinctions for certain engine types, but this is routinely handled directly in part 1068 or
with clarifying provisions in the standard-setting part.  This is certainly no greater challenge than
having the entire program written separately for each engine category.

       We have  revised §1068.110(e) to clarify that manufacturers may require that owners
submit warranty  claims only through authorized repair facilities, consistent with OPEFs
suggestion.

       The manufacturers' comments show that they have misinterpreted the provisions in
§1068.210 and §1068.215. In particular, §1068.215 specifies that the manufacturer-owned
exemption is limited to engines owned and controlled by the manufacturer such that they are  not
used in revenue-generating service. It is incorrect to assert that this does not require actual
possession of the engines by the engine manufacturer,  or that the engines may be used to
                                          1-72

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


generate revenue, as long as the revenue does not go to the engine manufacturer.  We are
therefore adopting the clarifying change to include "possession" as an explicit requirement to
qualify for the manufacturer-owned exemption. We understand that manufacturers have a
legitimate need to share experimental engines with their customers, including those that may be
used in revenue-generating service.  However, the test exemption in §1068.215 is the appropriate
path for these engines. The test exemption is available only upon request, which gives us the
opportunity to be involved and aware of such engines. Engine manufacturers that hold a
certificate of conformity (for any engine model or application) must generally submit only a
minimal set of information and can have an exemption approved to cover a two-year period.
This involves a minimal burden for the manufacturer, but keeps us informed of the status of
these engines.  We believe it would not be appropriate to allow for this practice without requiring
the engine manufacturer to identify the basic elements of their plans to introduce such engines
into U.S. commerce.  As for importation, the testing exemption is temporary, so it is described in
§1068.325 rather than §1068.315. It may turn out that engines are scrapped before the
exemption expires, but it is still the case that the exemption applies for a given time period rather
than for the life of the engine.

       We agree that the regulation should acknowledge that Internet, e-mail, and toll-free
phone numbers are legitimate alternatives to communication by traditional mail service.

       The term "engine-based test" is used only in limited cases and in each instance the
meaning of the term clearly applies only for exhaust emission testing. This also aligns with the
intuitive understanding that engines are generally tested for exhaust emissions and equipment (or
fuel-system components) are tested for evaporative emissions. We therefore believe it is not
necessary to change the proposed definition.

       The regulations no longer use the term  "incomplete engine assembly" so we have
removed this definition from §1068.30.

       We have  revised the definition of "secondary engine manufacturer" to reflect the input
received regarding partially complete engines.  As such, we see no need to defer these regulatory
provisions or definitions until a later rulemaking.

       We agree that the regulation should include clarifying language to state that changing an
engine from stationary to nonroad would not cause the engine to be subject to standards based on
the date of the conversion.  Nonroad standards would apply based on the original date of
manufacture. We have revised §1068.31 accordingly.

       We have  chosen to include the restriction related to naming model years for imported
products in part 1068, rather than including that only for Small  SI engines. We placed the
revised regulatory language in the rulemaking  docket and interacted with the Engine
Manufacturers Association and several individual manufacturers to confirm that these provisions
could be applied more broadly than just for Small  SI engines. EMA's suggestion to waive the
restriction for engines or equipment originally  produced in the United States would be difficult to
implement. It would be difficult for Customs to differentiate incoming products based on
whether or not they had an actual point of origin within the United States.  Perhaps more
                                       1-73

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
importantly, such a policy would likely be impermissible under the rules governing international
trade, since it would clearly provide preferential treatment for domestically produced items.

       We are adopting the other special compliance provisions, such as warranty assurance and
bonding, only for Small SI engines. We may propose to apply these provisions for other engine
categories in a later rulemaking.

       In the absence of any supporting comments related to accelerating the migration to the
general compliance provisions in part 1068, we are not making any broad changes to require
earlier compliance with the regulatory provisions in part 1068.  We agree, however, that there
may be good reasons for manufacturers to opt into the part 1068 provisions before they would
otherwise apply. Since these provisions represent the long-term plans for all nonroad engines,
we are including a provision allowing manufacturers  to comply with specific provisions of part
1068 early.

       We agree that the regulations should include specific provisions to address the various
responsibilities related to replacing emission control information labels due to accidents or other
need for repairs. The engine manufacturer should be  responsible for providing duplicate labels
in these cases and should take steps to ensure that the labels are applied properly.  These
duplicate labels should include all the information from the original label except for the date of
manufacture, which would be impractical to include as described in Section 1.3.2.

       We agree with Impco's recommendation to modify §1068.105 and have changed the
regulation accordingly.

1.6   Certification fees (40 CFR part 1027)

What Commenters Said:

       EMD included a comment about the certification fees for locomotive remanufacture
systems in its written submission to docket EPA-HQ-OAR-2003-0190.11 They reiterate it here,
in its more proper venue.  The certification fees rule,  included in 40 CFR Part 85, allows a
reduced certification  fee if the fee exceeds one per cent of the "aggregate projected retail sales
price of all vehicles or engines covered by that certificate." 12 That language is carried over to
the proposed Part 1027.13 In the case of emissions  remanufacture systems, EPA interprets this
provision to mean the price of the locomotives to which the kits are applied. This interpretation
means that an engine family consisting of remanufacture systems whose price fairly reflects their
contents has no chance of qualifying for a reduced fee unless sales of the systems certified under
that family are zero, because the fee is much less than one per cent of the value of even one
locomotive.

       EMD commented that this situation is unfair to system manufacturers. EMD believes that
the intent of the rule is to give manufacturers relief if their economic benefit is not commensurate
with the cost of certification. The economic benefit to the manufacturer of a remanufacture
system is the revenue to be gained by the sale of the system, not the value of the locomotive to
which it is to be applied. It appears that the rule was written with only engines and complete
                                          1-74

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


vehicles sold by a manufacturer in mind, omitting consideration of remanufacture systems,
which are a collection of parts with only a small fraction of the value of the complete engine or
vehicle. EMD requests that this rule section be rewritten to include one per cent of the aggregate
retail sales price of remanufacture systems, not the price of the locomotives to which they are
applied, as the parameter for comparison to the certification fee to qualify for a reduced fee. 14
The modification should be applied both to the applicable section of the current rule, Part 85, and
to the new Part 1027.

       EMD continued that EPA proposes to adjust certification fees only if the adjustment
would exceed $50 per engine family. EMD opposes this proposal. EMD's engine families are in
the "Other" category, which includes locomotives and compression-ignition marine engines, as
well as other smaller classes of engines. Since fees were initially assessed in 2005, the fee for
this category has declined from $826 to $802. EMD believes that it will continue to decline as
the number of certification applications in this category increases. Therefore, the effect of EPA's
proposal would be to hold the fee at an artificially high level until the adjustment reached $50,
increasing EPA's revenue from this program at the expense of manufacturers. EPA should
continue to adjust fees annually as required by the rule currently in force.

Letters:
Commenter
EMD
Document #
0687
Our Response:

       We agree with EMD's suggestion to reconsider the cost basis for reduced fees for
remanufacturing kits.  We already have provisions in place for fuel-conversion kits in which the
regulation specifies that the basis for evaluating the one-percent threshold is the value of the kit
rather than the value of the engine. This  applies anytime the particular engine is already covered
by a certificate of conformity based on the original fuel.  We believe a remanufacturing kit is
analogous to a fuel-conversion kit for purposes of certification fees. While the remanufacturing
kits may in some cases be applied to uncertified engines, in all cases the remanufacturing kits
(and not the engine that is being modified) define the scope of the certification.  We are therefore
modifying the regulation to allow for reduced fees where the assessed fee  is more than one
percent of the value of the remanufacturing kit or remanufacturing system. This applies equally
to locomotives and marine diesel engines, which are now also subject to remanufacturing
certification provisions.

       We disagree with EMD's assessment and recommendation regarding annual fee
adjustments. We believe  it is rather short-sighted to make a long-term policy decision based on
an extrapolation of the trend from the last two or three years. While there have been substantial
additional numbers of applications for certification recently, there is reason to believe that the
calculated fee will not universally trend downward.  First, some new applications  result from the
introduction of new emission control programs, such as for Large SI engines and recreational
vehicles, which are clearly one-time effects. Second, while there has been a large number new
companies certifying products from overseas, we are concerned in some cases about the ability
of these companies to fulfill their obligations for warranty, recall, and other in-use compliance
                                       1-75

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
provisions. We are taking steps in this rule to make clear that certification is more than a one-
time requirement, which we believe will serve as a disincentive for some companies that may
otherwise have thought that certification provided a simple and clear path for introducing
products into the U.S. market. We believe this trend will stabilize at some point.  Third, even if
these trends persist, it will not be long before the calculated fee exceeds the $50 threshold and we
will change the fees accordingly.  Fourth, inflation is part of the equation for calculating fees.
We would not expect the effect of increasing numbers of certificates to always be enough to
offset inflationary effects. Over any reasonable time frame, we would still expect the proposed
provision limiting annual fee changes to be revenue-neutral.

       EMD did not address the underlying reason for the proposal to limit annual fee changes.
Our proposal was focused on minimizing confusion and administrative errors.  To the extent that
fees do not change from year to year, there will be fewer mistakes when people make their
payments.  Under the current program, it is not uncommon for people to overpay or underpay by
a nominal amount.  It is time-consuming, awkward, and wasteful to spend the time required to
collect an additional $8 because an applicant was not aware that the fee had increased. Looked
at from the other side, the cost to the government of issuing an $8 refund is about $75. We
believe the public benefit of avoiding administrative errors far exceeds the benefit to EMD from
reducing certification fees by $24.

       Finally, there is a place for being cost-conscious; however, we believe the amounts in
question hardly warrant controversy. In the most extreme case, under the proposed rule we
would have a fee that is $49 less than it could be with an automatic annual readjustment. If
EMD would sell a single locomotive in the family, the certification fee would be about 0.003
percent of their revenue for that one locomotive. With actual  sales in the hundreds of units and
expectation that these cost differentials will be much closer to zero, even this estimate vastly
overstates the relative burden represented by the fee that is higher than the calculated value.

       We find that our original reasons for limiting annual fee adjustments are unchanged.  We
are therefore finalizing these provisions as proposed.

1.7   Preemption of state regulations (40 CFR part 1074)
       We have addressed comments related to preemption of state regulations in "Response to
the Petition of American Road and Transportation Builders Association to Amend Regulations
Regarding the Preemption of State Standards Regulating Emissions from Nonroad Engines,"
July 25, 2008.

1.8   Technical amendments for Large SI engines (40 CFR part 1048)

1.8.1   Fuel tank permeation

What Commenters Said:

       IMPCO commented that EPA proposed that nonmetal  fuel tanks must use a qualifying
design specified in  1060.240 § 1048.245 How do I demonstrate that my engine family complies
with evaporative emission standards?
                                          1-76

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       (e) (1) (i) ... Nonmetal fuel tanks must also use one of the qualifying designs for
controlling permeation emissions specified in 40 CFR 1060.240.

       According to §1060.1, these new LSI permeation standards are effective Model Year
2007.IMPCO requests that this be changed to MY2010. Given that it is now August 2007, it is
too late for MY2007 design to change and MY2008 designs are frozen. Additionally, because
MY2009 is an emissions carryover year, engine manufacturers  did not anticipate making any
design changes until MY2010.

       ECO commented that they think the real concern for Genie is when the new requirements
will come  into effect.

Letters:
Commenter
Impco
ECO
Document #
0692
0741
Our Response:

       We agree that allowing some lead time for implementing the design requirements related
to plastic fuel tanks is appropriate.  Delaying the implementation date until the 2010 model year
should make it possible for companies to work with fuel tank suppliers to coordinate plans for
making a smooth transition toward making compliant products.
1.8.2   Diurnal emission testing

What Commenters Said:

       IMPCO noted that the proposed regulations state that a gas cap must be tethered or self-
closing and stay sealed up to a positive pressure of 24.5 kPa. IMPCO commented that the
definition of 'sealed' is  still somewhat nebulous. IMPCO suggested that, among other allowable
designs, the following should also be considered 'sealed':
       - A fuel cap design used by an automotive OEM that has been certified under the EPA or
       California ARB  enhanced evaporative emissions standards, or
       - A fuel cap that is listed under UL 558, 'Industrial Trucks, Internal combustion Engine-
       Powered'
       - Calculated HC emissions from the gas cap are less than some percentage of the evap
       standard (possibly less than 20%) over the 24-hr diurnal test (include a standard value to
       use for the % of HC in air at a certain temp)

       Protectoseal believes their fuel caps can comply with the proposed requirements to stay
sealed. Protectoseal noted that they have defined the set point of our caps as the positive
pressure at which the  caps first exhibit bubble leakage.  Depending on the cap size and style, this
bubble tight set point  is  between 3.5 psig and 4.5 psig.  The caps are further designed to provide
venting relief as the positive tank pressure increases beyond this set point. The caps also provide
                                      1-77

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
vacuum relief (under negative pressure conditions) to allow make-up air to enter the fuel tank
and allow smooth flow of fuel to the engine.  (Protectoseal included literature sheets that provide
representative data of the flow/leakage data of the caps under pressure and vacuum conditions.)
Protectoseal noted that they have worked with Underwriters Laboratories to make sure that the
caps meet all their functional and safety related requirements (as evidenced by the UL Listing for
our new designs) while also minimizing emissions into the atmosphere when the cap assembly is
subjected to positive pressure.

Letters:
Commenterd
Impco
Protectoseal
Document #
0614
0615
Our Response:

       We understand that an unqualified requirement to maintain sealed fuel tanks can be
problematic, if only to recognize that a tank with undetectable leaks may nevertheless experience
an infinitesimal vapor loss through very small imperfections in gasket materials, especially as
materials age in normal service.  Protectoseal's experience demonstrates that the concern for
leaks is not a question of feasibility with respect to detectable leaks as measured by normal
diagnostic tests. As a result, we do believe it is not necessary to add the several options
recommended by Impco for demonstrating that a fuel cap will adequately keep a fuel tank sealed.
Rather, we are revising the regulation to require only that sealed tanks prevent measurable leaks.
This should avoid a situation where someone feels at risk of being noncompliant based on
extremely low leakage rates. Underwater "bubble tests" would be one appropriate method for
establishing whether there is a measurable leak.
1.8.3   Certification related to evaporative emission standards

What Commenters Said:

       ECO commented in order to address the inherent complications related to evaporative
certification for LSI and SD/I engine families, ECO suggests that EPA consider establishing a
component certification process for engine system / equipment components that have a bearing
on the evaporative emissions of the engine / equipment. For instance, EPA should consider
allowing equipment OEMs, or component manufacturers, to conduct their own independent
certification of evaporative system components, including fuel tanks, lines, and caps. The format
of this program could follow the format that the California ARB utilizes for Small SI engine
evaporative components.

Letters:
Commenter
ECO
Document #
0712
                                          1-78

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


Our Response:

       We adopted  evaporative emission requirements with the understanding that the large
majority of gasoline-fueled engine models would be in installed in metal fuel tanks. In this
scenario, designing systems to control permeation and diurnal emissions is very straightforward.
Now that we are implementing these requirements, it has become clear that plastic fuel tanks are
an important exception to accommodate. We have seen that it is impractical to expect engine
manufacturers to be responsible for including plastic fuel tanks in their application for
certification, since they generally are not involved in designing, shipping, or installing the fuel
tanks.  We believe it is appropriate to allow fuel tank manufacturers or equipment manufacturers
to certify fuel tanks separately. We have revised part 1060 to allow for this certification path.

1.8.4   Definition of small-volume manufacturer

What Commenters Said:

       ECO commented that 1048.801 defines a small volume manufacturer as an engine
manufacturer with U.S.-directed production volumes, subject to Part 1048, totaling no more than
2,000 engines per year. It also defines a SVM as an engine manufacturer with fewer than 200
employees. Although ECO agrees with the redefinition of SVM, the new definition is missing
the distinction that a company qualifies as a SVM by meeting either of the two criteria, not both.
To provide a precise definition of SVM, the first sentence of the definition should read "Small-
volume engine manufacturer means  either one of the following."

       California ARB's current large spark-ignition engine regulations do not have assigned
DFs for small volume  engine manufacturers. However, California ARB accepts EPA-approved
"assigned DFs" for small volume engine manufacturers during the certification process. Due to
variability of emission control technologies designed for large spark-ignition engines, California
ARB will continue to work closely with EPA to review and approve "assigned DFs" on a case-
by-case basis.
Letters:
Commenter
ECO
California ARB
Document #
0712
0682
Our Response:

       ECO's interpretation of the proposed change to the definition of small-volume engine
manufacturer is consistent with our intent. We have changed the wording of the definition to
make this distinction clearer.

       We look forward to working with California ARB further regarding assigned
deterioration factors.
                                       1-79

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
1.8.5   Additive deterioration factors

What Commenters Said:

       IMPCO strongly supports allowing manufacturers to use either Multiplicative or Additive
DFs. With more sophisticated engine technologies and ever-decreasing emissions, even the
slightest increase in emissions can tremendously affect a multiplicative DF calculation. For
example, with a low-hour value of 0.1 g/kW-hr and a 2,500-hr value of 0.3 g/kW-hr, the
extrapolated 5,000-hr multiplicative DF is 5.000. However, an additive DF would be 0.400
g/kW-hr, which is far more representative of in-use deterioration.

       ECO commented that newly added text to 40 CFR §1048.240(c)(2) states that engine
manufacturers may utilize additive DFs for engine families with low-hour emission levels below
0.3 g/kW-hr.  The mandatory use of multiplicative deterioration factors (DF) penalizes low
emission engines, as the multiplicative deterioration process creates an exponential penalty for
engines that produce extremely low 0-hour emissions. Because the allowance to use additive DFs
encourages the development of the lowest emitting engine technologies, ECO and their
stakeholder group fully support this change and encourage EPA to retain the additive DF
allowance in the final rule.

       In subsequent comments, ECO suggested changing the threshold for using an additive DF
from 0.3 g/kW-hr (for all pollutants), to one based on measured low-hour test results less than 11
percent of the applicable standard.  Alternatively, the threshold for CO could be increased to 0.48
g/kW-hr. These adjustments would take into account the higher numerical standard for CO.

       ECO also suggested in the later comments that the regulation should state that a given
engine might use an additive DF for one pollutant and a multiplicative DF for another  pollutant.

       California ARB supports using an additive DF if the emission levels are below 0.3 g/kW-
hr. The use of an additive DF for engines with very low emission levels can accommodate the
mathematical effects during the durability calculation.

Letters:
Commenter
Impco
ECO
California ARB
ECO
Document #
0692
0712
0682
0798
Our Response:

       We agree that a slight increase in the accommodation for additive DFs for CO are
appropriate.  Since the mathematical and technological effects that lead to additive DFs are
driven by the magnitude of the emission levels, not the compliance margin, we believe the best
approach is to adopt a fixed threshold of 0.5 g/kW-hr for CO.  This approach involves a
consistent level of precision relative to the threshold for HC+NOx emissions.
                                          1-80

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
       The regulatory language in §1048.240 clearly provides for a separate decision regarding
additive and multiplicative deterioration factors for HC+NOx and CO. This should make clear
that the decision for an additive deterioration factor for one standard does not require an additive
deterioration factor for the other standard.

1.8.6   Field-testing demonstration for constant-speed engines

What Commenters Said:

       Intertek/Carnot recommended an approach to demonstrating for certification that
constant-speed engines meet the field-testing standards.  We requested comment on this because
there is no longer a transient test requirement for constant-speed engines. The recommended
testing involves ten minutes of engine operation consisting of steady operation for 20 seconds
under several different load conditions.  The test would be run like a ramped-modal cycle, except
that the transitions between modes would last three seconds for increasing engine load in 20-
percent increments and they would last five seconds for increasing load in 40-percent
increments.  The cycle does not include operation at full  load, since spark-ignition engines
generally don't operate for an extended time at full load. See the figure below for the detailed
sequence of engine loads.

Letters:
Commenter
Intertek Carnot
Document #
0740
Our Response:

       We believe the recommended cycle provides a good tool for evaluating whether constant-
speed engines will adequately control in-use emissions under engine operation not included in
the D2 cycle used for steady-state testing. This includes changes in engine load that would be
common for typical in-use operation (increasing pump output, adding devices powered by a
generator, etc.). These changes would not always occur with this frequency, but we believe this
schedule of engine operation is realistic for many in-use scenarios. If manufacturers have
information available to show that a different approach is more appropriate for their engines, we
would consider approving alternative demonstrations.  This might involve testing with the
original constant-speed transient test, or some other schedule of engine operation to better reflect
a relatively worst-case scenario of engine operation.
                                       1-81

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
                          Constant Speed Brgine Ramped Modal Cycle
  b"
  1L
"100 ^

	 90-
7H-,
fin ,
Conditioning
m~,

WMHN™ Ramp Modal Cycle

I ™^™™120-9ec Rail ing Load Avg
i rr
;. 1 i
1 "
L ^J, L,
• t^n
1 jLp ~d
; Ar--'-lL J_




120secrdlingavgincludesemissions/load I



i


1 ~ ^r—
, . 1 ___
rj


: I ^ T
| ;| <3 Sec Ramos. 20 sec jl
''. i 	
00 0 100


J

„ — ,t


H-<
' TT 'pj

T^t
j ~^*~ ™^» ', I i
H


<5 Sec Rarrps, 2(
u^-----K>«.-H---
~~l r" i
- ^i— 4 4— 4- - | - -
<3 Sec Ramos. 20 sec , :
; :5 Sec Rarrps, 20
ill 	 .L.J.i sec Hold . L ., (
200 300 400 500 «
                                      Duration, [sec]
1.8.7   Other issues

       We included several technical amendments for part 1048, which led to a variety of
comments on relatively minor issues.  This section describes these comments and our responses
to them. We additionally address comments related to applying an engine's build date on the
emission control information label in Section 1.3.

What Commenters Said:

       IMPCO commented that EPA proposed to remove the requirement for constant-speed
engines to meet the transient emissions standards in 1048.101(a).  IMPCO strongly supports this,
as the transient test cycle or constant-speed engines is not at all characteristic of constant-speed
engine operation. The steady-state test cycles are far more representative and appropriate.

       IMPCO also commented that EPA proposed to remove (b)(l) through (b)(l 1) of the
AECD detailed description in the application for certification.  EVIPCO strongly supports this. It
is important to document all AECDs to ensure that the proper regulatory guidelines are being
followed. However, the additional time required to detail every aspect presently identified in the
regulations creates an unnecessary burden on the manufacturer while providing little to no
incremental benefit to EPA.

       IMPCO also commented regarding § 1048.120(c), 1048.240(b), 1048.605(c)(2),
1048.610(d)(2), 1068.101(b)(2), 1068.120(a), 1068.501(a). Several areas throughout the
proposed regulations refer to an increase in exhaust emissions of any pollutant. Pollutant should
be defined only as those regulated under Parts 63, 1048, et al.
                                          1-82

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       ECO commented on page 28148 of the Federal Register notice, section E(l)
Deterioration Factors proposes an allowance for small volume engine families to utilize assigned
deterioration factors. Additionally, on page 28213 of the proposal, EPA requests input on the use
of assigned DFs for small volume engine families and requests comment on the appropriate
production threshold for this allowance. ECO and their stakeholders feel this provision is
necessary to allow flexibility for small volume SSI engine families. ECO also agrees with EPA,
that the allowance to use assigned DFs for small volume engine families should also be
incorporated into the large spark-ignition (LSI) engine rules contained in 40 CFR 1048. As it
stands, there are numerous companies that do not meet the definition of small volume
manufacturer (ref. 40 CFR 1048.801), but that produce one or more families of engines with
production quantities of only a few hundred per year. This small production quantity does not
allow MORs the ability to recoup costs associated with a typical durability operating cycle
(2,500 hours), and often results in MORs discontinuing the certification and production of
marginal product lines. ECO recommends that EPA consider the use of assigned deterioration
factors for all LSI engine families with  annual production quantities less than 300 units.

       California ARB commented on the following sections:
   -   22. Diurnal Temperature Cycle (40 CFR part  86.133-96): 40 CFR part 86 133-96
       requires a diurnal temperature cycle not common in California or much of the southwest.
       The temperature profile used by California ARB  to represent California conditions is 65-
       105°F. California ARB recommends EPA modify the rule to require the more restrictive
       profile, which in addition to harmonizing, would ensure the expected emissions
       reductions are achieved even in the warmer parts of the country.
   -   23. Diurnal Standard (40 CFR part 1048.105): 40 CFR part 1048.105 requires tanks to
       meet the diurnal standard of 0.2 grams per gallon-day (g/gal-day). California ARB agrees
       that 0.2 g/gal-day is currently achievable. California ARB has test data that shows 0.1
       g/gal-day may also be achievable. California ARB suggests that EPA propose a future
       date that includes the more stringent 0.1 g/gal-day standard.
   -   24. Pressure Standard (40 CFR part 1048.245): 40 CFR part 1048 245 requires a
       standing pressure test of 3.5 pounds per square inch absolute (PSIA). California ARB test
       data shows that on occasion tanks in California and presumably other warmer states can
       reach above 4.0 PSIA. California ARB recommends increasing the test standard to 5.0
       PSIA.
   -   25. Production Line Testing Exemption (40 CFR part 1048.301(a)(2)): EPA's
       proposal exempts large spark-ignition engine families with a projected U.S.-directed
       production volume below 150 units from routine production line testing.  However,
       production line testing is an important tool to ensure that manufacturers are meeting the
       requirements.  Therefore,  EPA should reconsider  this exemption. California ARB plans
       to retain its  current requirement for small volume manufacturers to test one percent of
       their California production.
   -   26. Production Line Testing Procedures (40 CFR part 1048.305(a)): EPA's proposal
       requires manufacturers to use either the steady-state or transient testing procedures to
       show that the production-line engines meet the exhaust emission standards. However, it is
       still not quite clear which test cycle should be used to generate DFs for production line
       testing. EPA should clarify that it will not allow manufacturers to use a DF generated by
       a steady-state test cycle to apply to a production line test using a transient test cycle.
                                       1-83

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
       27. Exhaust Emission Standards for Large Spark-Ignition Engines <1 liter (L) (40
       CFR part 1054.105(a) Table 1): California ARE recommends that EPA modify its
       proposed HC+NOx phase-in schedule for large spark-ignition engines < 1 L to harmonize
       with the California small off-road engine exhaust emission standards for class II engines,
       8 g/kW-hr at 2008 model year. This would provide for significant emission reductions
       from this category. Furthermore, a harmonized program would help reduce the problem
       of higher emission 49-state large spark-ignition engines traveling into the California fleet.

       EMA commented that total engine displacement should be rounded to the nearest whole
cubic centimeter.  Accordingly, §1048.615(a)(l) should be revised to read as follows: "The
engine must have a total displacement at or below 1000  cc after rounding to the nearest whole
cc."

       Caterpillar suggested that we make the following changes to the regulations related to
natural gas  engines:

   •   Modify 1048.620 to point towards 40CFR60 subsection JJJJ as the optional approach for
       transportable SI engines. This would allow either factory certification or site compliance
       testing, account for the varying fuels common to these applications, and still assuring
       emissions compliance.

   •   Remove the 250kw cutpoint and include all engines down to 25 hp, as that is where part
       90 requirements are in effect for all small SI engines. This will then harmonize with the
       regulatory strategy in 40CFR60 subsection JJJJ.

   •   As 1048 already regulates gasoline and rich-burn LPG, this exemption need not apply to
       these fuels. However, all other fuel types should be included in the exemption. This
       harmonizes with the approach taken by 40CFR60 subsection JJJJ as well.
Letters:
Commenter
Impco
ECO
California ARE
EMA
Caterpillar
Document #
0692
0712
0682
0691
0814
Our Response:

       We agree that the constant-speed transient cycle should be omitted from the regulation, as
described in the proposed rule.  We may pursue a more appropriate constant-speed transient duty
cycle in a future rulemaking.

       We agree that simplified AECD descriptions in the application for certification are
appropriate, as described in the proposed rule. Manufacturers have the incentive to be thorough
in describing their AECDs in the application for certification to avoid a situation where AECDS
                                          1-84

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


are detected outside of the certification process and questions are raised regarding whether the
AECD qualifies as a defeat device.

       We agree that all the references to pollutants in part 1068 should specifically refer to
"and regulated pollutant." We have also changed the regulation at §1048.120 to refer to "any
regulated pollutant." The reference in §1048.240 is clearly written in the context of pollutants
that are subject to emission standards, so this change is unnecessary.  The provisions in
§1048.605 and §1048.610 are based on the expectation that the engine is installed in a nonroad
piece of equipment without any changes that would affect fuel or air intake, combustion, or
aftertreatment. It is therefore not clear that the regulation should specifically limit this provision
to regulated pollutants. While we would not normally monitor compliance with respect to
nonregulated pollutants, we would be interested in understanding why such an increase might
occur.

       As noted in the preamble, we are adopting provisions for assigned DFs for Marine SI and
Small SI engines for small-volume engine manufacturers and for small-volume engine families.
This is intended to address the concern that the costs of generating DFs  can be quite large
relative to revenues for niche products, even for very large companies.  This same dynamic
applies for Large  SI engines, though to a larger degree. With a 5,000-hour useful  life, the cost of
generating DFs for an engine family are much greater than for Marine SI or Small SI engines.
We agree with ECO's recommendation to adopt a provision allowing for assigned DFs for
engine families with annual U.S.-directed production volumes at or below 300 units.

       We adopted a temperature profile for diurnal emission testing that is consistent with the
approach we take for light-duty vehicles. California's approach for light-duty vehicles is to
specify the higher temperatures but specify a test fuel with lower volatility. EPA and California
ARB have concluded that these two sets of test parameters yield very similar results.  It is
unlikely that changing the temperature and test fuel would lead to any design changes for
improved control  of emissions. We therefore believe it is appropriate to keep our existing
regulation intact.

       We adopted the diurnal emission standard expecting that most manufacturers would opt
for the certification alternative to keep fuel tanks sealed up to 3.5 psi. This effectively achieves
complete control for all but the large majority of summer days.  For example, a fuel tank filled
halfway with 9 RVP fuel would reach a pressure of 3.5 psi if ambient temperatures ranged from
72 to 96°F.  We specify evaporative testing using these conditions to represent nearly a worst-
case condition.  The 3.5 psi specification is also consistent with the industry standard under
UL558, so manufacturers have considerable experience in supplying products that comply with
this pressure requirement. It is not clear that there would be a significant environmental benefit
with higher-pressure fuel caps, or that such caps would be readily available for the full range of
equipment that would need them.  For manufacturers electing to test their systems to demonstrate
compliance with the diurnal emission standard, it is not clear that a more stringent standard is
appropriate. We are in the second year of implementing the diurnal emission standards and will
be learning from this experience. Canister capacities and purge  systems, the most likely
alternative design solution, have been used for many years with  light-duty vehicles, but there is
little information available to show that a more stringent standard is appropriate. We would be
                                       1-85

-------
                                Chapter 1: Rulemaking Process and Cross-Program Issues
interested in reviewing California ARB's information when it is available. However, it is not
clear that the more stringent standard would lead to a meaningfully improved emission controls
with in-use engines and equipment.

       Production-line testing for Large SI engines relies on CumSum statistical calculations.
To be able to make a pass/fail decision for an engine family, two test results are required to
initiate calculations related to sample sizes and overall compliance. California ARB intends to
limit testing to 1 percent of production, but for production volumes below 150, a one-percent rate
in this case translates to a single engine. This was the primary basis for EPA's decision to waive
production-line testing requirements for these small families. California ARB may continue to
require testing for these engine families, but it is not clear single-engine testing can be reconciled
with the need to use CumSum calculations to reach a conclusion.

       Emission measurement during transient engine operation is very important for testing an
engine's ability to control emissions over a wide range of in-use operation. We believe
manufacturers should be able to choose whether to establish a single DF (for a pollutant) based
on a comparison of emission measurements using transient operation only, or to generate
separate DFs for transient and steady-state testing. We agree that it is not appropriate to allow
manufacturers to use a DF from steady-state testing to characterize the aging effect for transient
emission measurements. Selecting the type of engine operation for service accumulation is very
different. We believe  a variety of approaches can be used to properly age an engine.  The most
important parameters to consider are engine load (torque) and exhaust temperatures.  It is not
apparent that service accumulation based on transient engine operation is an important factor
aside from considerations of average engine load and exhaust temperatures.

       Manufacturers  of engines at or below 1000 cc must meet the emission standards that
apply for Class II Small SI engines to be exempt from the more stringent Large  SI emission
standards. We are  including an update to reference the new standards for Small SI engines in
part 1054. This provision requires manufacturers to meet the currently applicable phase of
standards, so no further regulatory change will be necessary if or when we adopt an additional
phase of standards  for  Class II engines.

       We agree that §1048.615 should continue to apply for engines at or below 1000 cc  (not
1000.0 cc), consistent  with the original regulation.

       The Clean Air  Act requires that nonroad engines be certified before they are introduced
into U.S. commerce. We are therefore unable to modify §1048.620 to make certification
optional,  as suggested  by Caterpillar.  This requirement does not apply to stationary engines.  We
may consider in a future rulemaking to specify the emission standards in part 60, subpart JJJJ, as
being sufficient for certifying engines under §1048.620, but we would need to go through the
notice-and-comment process for such an initiative.  We established a threshold of 250 kW  for
these engines to avoid competitive effects where automotive-based natural gas engines would be
potentially serving the same markets as the diesel-derived natural gas engines.
                                          1-86

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


1.9  Technical amendments for recreational vehicles (40 CFR part 1051)

1.9.1   Maintenance

What Commenters Said:

       MIC commented that 1051.125(d) defines the components specified in 40CFRPart 1068,
Appendix I as being items of noncritical emission-related maintenance, but some of these
components are also listed as critical emission-related components specified in § 1051.801,
which causes mismatch. MIC recommends that the language be changed as follows:
"...any  other emission-related maintenance on the components other than critical
emission-related components." While spark plugs are noncritical emission related components,
the maintenance and inspection of them can be critical emission related maintenance.
Manufacturers would like to have the ability to inspect and clean them during service
accumulation on the emission-data vehicles. Federal Register Vol. 67, No. 217, pg. 68321 C.2
indicates that the EPA may allow changing spark plugs even though they are aware that spark
plugs may affect emissions.

Letters:
Commenter
MIC
Document #
0701
Our Response:

       We have revised §1051.125 to clarify the relationship between critical and noncritical
emission-related maintenance.
1.9.2   Test procedures

What Commenters Said:

       ISMA understands that raw gas sampling is permanently allowed for snowmobiles under
the provisions of 40 CFR Part 1065. However, they note that calculation procedures similar to
the fuel flow method in 40 CFR Part 90.419(c ) and 40 CFR 91.419(c ) for converting the raw
gas sampling measurements into g/kW-hr emission levels are not explicitly provided in Part
1065. The fuel flow method is commonly used by the snowmobile manufacturers under the
interim provision of 1051.145(e)(l), which allow use of the raw sampling  procedures from parts
90 or 91 through the 2009 model year. 40 CFR  1065.601(c)(l) is intended to implicitly allow a
functionally equivalent fuel-flow calculation method, based on reference to the ISO 8178
standards. The indirect allowance of 40 CFR 1065.601(c)(l) and the fact that the ISO 8178
standards are not freely available in the public domain have caused considerable uncertainty for
the snowmobile manufacturers regarding the available raw-gas sampling options beyond model
year 2009.  The manufacturers note that the direct final rule for ATV's published in the Federal
Register on April 26, 2007 extends the use of the raw gas sampling methods in 40 CFR part 90
or part 91 through the 2014 model year for ATV's. Since all of the ISMA members are ATV
                                      1-87

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
manufacturers familiar with these established methods, and since ATV manufacturers are
allowed to continue using these methods for ATV's, ISMA proposes that EPA also extend the
use of the raw sampling procedures in parts 90 or 91 through the 2014 model year for
snowmobiles. This would provide the Agency with the time to codify more explicitly in Part
1065 the fuel  flow method allowance that we understand is embodied in ISO 8178. It would also
allow the ISMA member companies to use just one method for all of their raw gas sampling
during the remaining time period that engine-based testing is permitted for ATV's.

Letters:
Commenter
ISMA
Document #
0671
Our Response:

       We agree that an allowance for continued use of raw sampling procedures under part 90
or 91 is appropriate.  As an interim measure, we are revising the regulation to allow this for
demonstrating compliance with the Phase 1 and Phase 2 standards.  We intend to revisit this
question as part of the effort to revise the Phase 3 standards. We will at that time decide when
manufacturers would appropriately be subject to testing requirements under part 1065.

1.9.3   Determining maximum engine power and displacement

What Commenters Said:

       MIC commented that §1051.140 is a new section describing how to determine an
engine's maximum power and displacement. This section references the 40 CFR part 1065
mapping procedures to determine maximum power. Those procedures require test equipment that
can measure engine power during transient conditions as engine speed is changing. Compliance
with the proposed new requirement involves significant equipment costs that do not contribute to
greater emissions control. MIC therefore recommends that SAE J1349 be allowed as an
alternative method for measuring maximum engine power.  In calculating displacement, the
proposed language requires "using enough significant figures allow determination of the
displacement to the nearest 0.1 cc" while also stating that "An engine configuration's
displacement is the intended swept volume of the engine rounded to the nearest 0.5 cubic
centimeter". The  standards applicable to ATVs are defined based whether the engine size is less
than 225 cc or equal to or greater than 225 cc. It would therefore be more consistent to specify
that the swept volume should be rounded to the nearest cc, rather than the nearest 0.1 cc. There is
also an error in the example calculation incorporated in the proposed section. The correct
displacement for the example values used is 176.7 cc, not 176.5.

Letters:
Commenter
MIC
Document #
0701

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


Our Response:

       Manufacturers depend on the proper value of maximum engine power to calculate
emission credits for averaging, banking, and trading. Manufacturers also need to determine
maximum engine power to establish whether an offroad utility vehicle is above or below 30 kW.
It is important to use an objective criterion to establish maximum engine power to ensure that
manufacturers don't use available discretion (as under SAE J1349) to manipulate credit
calculations or change the standards that apply for their products.  It is also important to ensure
that different programs use the same  metric to establish whether standards apply to avoid overlap
or gaps between programs. Maximum engine power is fundamentally a design value and does
not necessarily require testing. The reference to the mapping procedures clarifies which design
value is appropriate. We also specify that the nominal value for maximum engine power must
fall within the range of values from production engines.  This testing is not required, but it allows
us to verify that the declared value is appropriate, especially for preventing manufacturers from
gaining an advantage by declaring a value for maximum engine power that does not represent
production engines.

       We agree that the regulation specifies displacement values (such as 225 cc) only to the
nearest cubic  centimeter. We have therefore revised the regulation to require manufacturers to
report displacement values to the nearest centimeter. However, to ensure that these reported
values  are accurate, we are keeping the requirement to use methods that allow for determining
the  displacement to the nearest 0.1 cc before rounding.

       The comment pointing out an error in the calculation underscores the need for an
example to illustrate the regulatory provision. The example  calculation is correct, though the
question is moot because of the change in precision described above.

1.9.4   Deterioration factors

What Commenters Said:

       MIC commented that 1051.243(b)(6) says "You may use other testing methods to
determine deterioration factors, consistent with good engineering judgment, as long as we
approve those methods in advance." MIC's concern with this language is that resource
constraints at EPA  sometimes preclude detailed consideration of alternative testing methods even
when they have been  developed using "good engineering judgment." To address this practical
concern, this subsection needs to be revised to read, "You may use other testing methods to
determine deterioration factors, consistent with good engineering judgment, unless we provide
an engineering analysis within 30 days  demonstrating that the proposed method is not
acceptable."

Letters:
Commenter
MIC
Document #
0701
                                       1-89

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
Our Response:

       We understand that there may be good reasons to modify plans for generating DFs for a
given engine family.  At the same time, evaluating such changes often involves more than a
simple assessment. We are especially sensitive to the need for review of such requests based on
many recent experiences with companies trying to exploit every possible angle to reduce the
burden associated with certification, with no apparent regard to the environmental impact of
these decisions. While an expedited response would clearly be of interest for manufacturers in
many cases, but we believe it would be a good policy—or even in the manufacturer's best long-
term interest—in setting a deadline for EPA's response. For example, if we would discover after
30 days that there is a problem with the manufacturers plan, there would need for further testing,
either for extended service accumulation for a given emission-data vehicleor for restarting the
durability testing altogether.  If the EPA approval step is unacceptable, the only viable option is
to remove the provision for alternative DF  demonstrations.

1.9.5  Other issues

What Commenters Said:

       ISMA commented recognizing the  litigation that impacted on Phase 3 of the US EPA
Rule for snowmobiles, they believe the published standard for Phase 3 in 1051.103 may be
appropriately designated as "reserved" status.

       ISMA commented that they do not  support the proposed change in 1051.205(t).
Production volumes are based on market analysis and behavior.  These volumes can shift
significantly from year to year from circumstances outside of a manufacturer's control. It is
inappropriate for EPA to require a "justification" for a change in estimated production volumes
from year to year.

       ISMA does not support the recommended change in 1051.250 (a). Clearly this should not
apply to corporate averaging engines which are already required to file two separate reports
throughout the year. The snowmobile manufacturers provide actual production number on a
quarterly basis in their PLT reports. In addition, manufacturers are required to provide final
production volumes in their corporate  average report. ISMA recommends EPA exempt
manufacturers not using small-volume compliance provisions from this requirement.

       ISMA does not support the recommended change to 1051.310 (b)(3). ISMA agrees with
EPA it is appropriate  for carry-over engine families to combine the last PLT test result from the
previous model year with the first PLT test result from the current model year to determine the
number of PLT tests required. However, EPA has proposed to add two sentences to 1051.310
(b)(3) which serve no purpose except to increase the test burden on a manufacturer, "Use the last
test result from the previous model year only for this first calculation. For all subsequent
calculations, use only results from the current model year. ISMA requests the last two sentences
from this paragraph be deleted.
                                          1-90

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       MIC and ISMA commented that it does not appear that the proposed deletion of the
reference to "vehicle" testing from section 1051.315 is appropriate because some engine families
are certified based on vehicle testing.

       ISMA question why in 1051.350 the limit for keeping files has been switched from 1 year
to 8 years. Also, additional definition is needed regarding the specific records being referenced.
       MIC commented that proposed changes to 1051.350 increase the requirement for
retaining paper records from 1 year to 8 years. If the retention period is going to be extended
such a large amount then the requirement for "paper" records should be replaced with "paper or
electronic" records.

       MIC commented that the explanation for proposed changes to §1051.701 includes the
statement that "We are also clarifying that a single family may not generate emission credits for
one pollutant while using emission credits for another pollutant, which is common to all our
emission control programs." However, the proposed requirement is not common (e.g., this issue
was specifically addressed for snowmobiles during the recent rulemaking and EPA agreed that
credits for one pollutant could be used by a family generating credits for another pollutant. This
proposed change needs to be deleted.

       ISMA commented that in 1051.801 EPA proposes to amend 40 C.F.R. § 1051.801 to
redefine the term "designated compliance officer," in such a way as to treat snowmobiles
differently than all other types of vehicles regulated under 40 C.F.R. part 1051 (i.e., off-highway
motorcycles, ATVs, and certain utility vehicles and other types of motorcycles).  Specifically,
snowmobile issues would be handled by the Manager of the Heavy-Duty and Nonroad Engine
Group in Washington, D.C., while all other part 1051-regulated vehicle issues would be handled
by the Manager, Light-Duty Engine Group in Ann Arbor, Michigan. ISMA opposes this
amendment and notes that EPA nowhere explains this proposal in the preamble to the proposed
rule, and at the very least must do so in the preamble to the final rule. ISMA's members are
entitled to expect EPA to minimize and not proliferate administrative compliance costs, and to
that end should face a compliance approach that is designed to partner with manufacturers to
pragmatically solve day-to-day issues arising from applying the part 1051 regulations. The
Section 1051.801 proposal is not pragmatic.  ISMA can see no rational basis on which EPA
could cleave off compliance activities for snowmobiles from compliance activities for other
types of vehicles regulated under part 1051. Thus, ISMA does not know how EPA could justify
requiring companies that already face significant administrative compliance costs to duplicate
their efforts across disparate parts of the Agency. This is particularly apparent to ISMA because
each of ISMA's members manufacture both snowmobiles and ATVs. In general, it makes little
sense to have different compliance officers interpreting and applying part 1051. Moreover, such
a questionable organizational change fragments, and renders inefficient EPA's own operations,
and it imposes added administrative compliance costs on manufacturers of products that
arbitrarily must deal with one set of officials for one set of vehicles, and another for another set
of vehicles, even though both types of vehicles are regulated by precisely the same part 1051
regulations. It makes sense to position the compliance officer for all vehicles regulated under
part 1051 at the location that currently serves the greatest number of vehicles.
                                       1-91

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
       MIC commented that having a unique Designated Compliance Officer for snowmobiles
would require companies that manufacture other types of vehicles to deal with two different
offices on very similar, if not identical issues. For efficiency and consistency, it would be
preferable to have snowmobile issues handled by the Ann Arbor office.

       MIC commented regarding 1051.730. The intent of this subsection would be clarified if
the term "sales weighted" is replaced with "U.S.-directed production weighted" instead of simply
"production weighted."

       Robin America requested the following change to the Part 1051. If an engine
manufacturer has general purpose engines approved (exhaust & evap) to either Parts 90 or 1054
they should be able to use these engines in ATV's without having to meet the 1051 exhaust
specifications. This will enable engine manufactures to sell general purpose standard spec engine
to ATV manufactures without any additional testing on the part of either manufacturer. This
change should be stated such that they can use previously approved engine from now until 2015
when the mandatory chassis testing is required.

       Sierra Research commented that they heard that EPA was going to state 2010 model year
at least for the labeling requirements. Also, they know in the Heavy-Duty world the use of older
testing methodologies are allowed for some time while part 1065 requirements are made to lab
facilities; will the same be allowed for the Recreational Vehicle category under 1051.140? Will
older test data be useable as carry-over once the 1065  testing requirements become applicable?

Letters:
Commenter
ISMA
MIC
Arctic Cat
Robin America
Sierra Research
Document #
0671
0701
0709
0743
0742
Our Response:

       There are two issues related to the Phase 3 standards for snowmobiles.  First, the court
decision stated that we do not have legal authority under the Clean Air Act to set NOx standards
for snowmobiles and required that we remove the NOx component from the Phase 3 standards.
We have addressed this in a separate rulemaking (72 FR 35946, June 25, 2008).  Second, the
court required that we provide further clarification and justification for the Phase 3 standards we
set. We intend to address this second question in a separate rulemaking well before the Phase 3
standards are scheduled to take effect. We may  or may not conclude that the Phase 3 standard
needs to change. We therefore believe it is not necessary or appropriate to remove the Phase 3
standard from the regulation at this time.

       The current requirement for manufacturers to estimated projected production volumes
does not allow us to require realistic estimates. There are provisions in the regulations that
depend on realistic projections, so we believe it is necessary for us to specify that these
                                          1-92

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


projections be based on proper assumptions. Records of previous production volumes and an
assessment of current market conditions allow companies to plan for ongoing production rates.
We would not expect to ask manufacturers to justify projected volumes for every family, but we
might ask for more information if the sales projections depart substantially from those of earlier
model years, especially where the projection would seem to provide an advantage for the
company under the regulations.  At the same time, we understand the recreational vehicles are
particularly prone to periods of irregular demand, with the resulting variation in year-to-year
sales figures for their products. We believe it is appropriate to keep the proposed requirement to
use realistic values for projecting sales volumes for the upcoming model year.

       We have changed the reporting requirements in §1051.250 to align with the reporting that
manufacturers already do for production-line testing and credit reports. Specifically, we are
requiring that manufacturers report their production volumes to the extent that those figures are
not already included in these other reports. This is merely intended to complete the reporting
requirement for cases where there are vehicles produced after PLT reports are complete (or if
there is no production-line testing required for the family) or in cases where the manufacturer is
not participating in the emission-credit program.

       The changed language in §1051.310 (b)(3) will not increase manufacturers' test burden.
The use of the test result from the previous model year is necessary in the first quarter of the new
year is necessary to avoid an automatic second test in the first quarter.  In the second quarter, the
result from the first quarter can be combined with the new result from the second quarter to make
the required CumSum calculations. There is therefore no longer any need to consider the result
from the previous model year in the calculations for the current model year. The new language
is intended merely to clarify the existing requirements, so we expect no additional test burden.
We are adopting the new language as proposed.

       We agree that the term "vehicle" should be preserved in §1051.315. This change was
inadvertent.

       We proposed to increase the time frame for keeping records to eight years to allow us to
review the validity of the production-line testing throughout the time that these vehicles will be
operating.  This is consistent with the recordkeeping we require in our other emission control
programs.  We agree that keeping electronic records is sufficient, as long as the manufacturer can
provide a printed copy of these records upon request. Since these records will generally stored
electronically, we expect there will be little if any additional effort to keep the records for a
longer time. The records that need to be kept are specified in considerable detail in §1051.345.

       We agree that the current regulations allow manufacturers to generate emission credits
for one pollutant while using emission credits for another pollutant with a given engine family.
Disallowing this practice would effectively increase the stringency of the emission standards.
We are therefore removing this provision from the final rule. We may revisit this issue for any
more stringent standards that we set in future rulemakings.

       The role of the Designated Compliance Officer in the regulations is simply intended to
provide a point of contact for submitting information and requesting approvals.  The regulation is
                                       1-93

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
not intended to dictate Agency decisions related to work flow or decision-making relative to
internal organizational structures. We will be deciding how to manage and implement the
certification process for recreational vehicles independent of the description and address noted in
the definition of "Designated Compliance Officer" in §1051.801. Nevertheless, to reflect the
fact that two of the three types of recreational vehicles are handled out of the Ann Arbor office,
we are revising the regulation to specify that as the default location for the Designated
Compliance Officer.

       We agree that §1051.730 should use the defined term "U.S.-directed production volume".

       We understand Robin America's interest in using certified Small SI engines for all-terrain
vehicles (ATVs).  We believe that conventional ATVs (straddle seat, handlebars, etc.) should be
certified under part 1051 since those products are generally well established, high-volume
products.  It would not be appropriate to accommodate the interest in simplifying the
certification process for these products. In contrast, the ATV  definition also captures other
rough-terrain vehicles that in many cases are low-volume niche products. We believe these
vehicles can many times benefit appropriately from using certified Small SI engines based on the
fact that these products are not as well established as high-volume recreational products. We are
therefore adopting this allowance for these ATVs through the 2014 model year. Starting in
2015, manufacturers must certify their vehicles based on a chassis test, after which the ATV
emission standards will be inherently different than the Small  SI standard, as acknowledged by
Robin America in its comment.

       The regulations in part 1065 specifically state that manufacturers may delay complying
with amended requirements for up to twelve months after those changes become effective.
Many of the changes we are making to the regulations do not impose new requirements, but
rather add flexibility or clarify existing requirements.  The provision to delay complying with
amended requirements allows for a more gradual transition when a regulatory change indeed
imposes a new or different requirement.  We are adding a similar provision in §1068.40 that
would allow for delayed compliance with technical amendments in part 1068 or in the standard-
setting part. Regarding the date of manufacture on engine labels, we are specifically stating that
the new requirement does not apply until the 2010 model year.

1.10  Technical amendments for heavy-duty highway engines (40 CFR parts 85 and 86)

What Commenters Said:

       EMA recommended that we delay the requirement to use the procedures in part 1065
until July 2010. To prevent this from impacting the stringency of the standards, they
recommended specifically excluding certain provisions from the delay. They recommened that
the regulations specify the following  provisions from part 1065:
       (1) Generate a map of your engine according to 40 CFR 1065.510(b)(5)(ii) and generate
       test cycles according to 40 CFR 1065.610.  Validate your cycle according to 40 CFR
       1065.514.
       (2) Follow the provisions of 40 CFR 1065.342 to verify the performance of any sample
       dryers in your system.  Correct your measurements according to 40  CFR 1065.659,
                                          1-94

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       except use the value of Kw in §1342-90(1) as the value of (l-xH2Oexh) in Equation
       1065.659-1.
       (3) Verify your NO2-to-NO converter according to 40 CFR 1065.378.
       (4) For diesel engine testing, correct NOx emissions for intake-air humidity according to
       40 CFR 1065.670.
       (5) You must comply with the provisions related to analyzer range and drift in 40 CFR
       1065.550. If drift correction is required, correct your measurements according to 40 CFR
       1065.672, but use the emission calculations specified in this subpart N rather than those
       specified  in 40 CFR 1065.650.
       (6) You must comply with 40 CFR 1065.125, 1065.127, and 1065.130, except for
       references to 40 CFR 1065.530(a)(l)(i), 1065.640, and 1065.655.
       (7) Follow the provisions of 40 CFR 1065.370 to verify the performance of your CLD
       analyzer with respect to CO2 and H2O quench. You are not required to follow 40 CFR
       1065.145(d)(2), 1065.248, or 1065.750, which are referenced in 40 CFR 1065.370.

Letters:
Commenter
EMA
Document #
0768
Our Response:

       We agree with the manufacturers' comment and have revised the regulation accordingly.


1.11 Technical amendments for stationary engines (40 CFR part 60)

What Commenters Said:

       EMA commented that Small SI engines generally are considered mobile due to their
small size and relatively light weight. However, there are engines that meet the definition of
Small SI yet, in fact, are utilized in stationary product applications. The EPA New Source
Performance Standard (NSPS) for spark-ignition engines correctly requires such Small SI
engines to meet the same emission requirements as their mobile counterparts. There are many
cases where  a Small SI engine that is certified, produced, and labeled as a mobile engine will
actually be utilized in a stationary application.  This industry practice must be maintained, and
should not be adversely affected by the imposition of unnecessary labeling requirements that
provide no environmental benefit. The final regulation should clarify that engines that are labeled
as compliant with nonroad standards may be utilized in stationary sources without alteration, or
additional labeling requirements.

       EMA also stated that Small SI engine manufacturers do not have the ability to determine
if an engine family may contain models that are subsequently utilized in a stationary application.
As the  proposed requirements for stationary and nonroad engines are intended to be identical,
this differentiation is not significant. Accordingly, §90.107(d)(13) should be deleted.
                                       1-95

-------
                                Chapter 1: Rulemaking Process and Cross-Program Issues
       Exergy commented that it is their understanding the proposed rule does not apply to
engines for stationary power generation and/or shaft driven equipment using natural gas and/or
propane. If EPA's proposed rule does apply, is there an exemption for stoichiometric and/or lean
burn 19 kW continuous duty or long duty-cycle (greater than 1,000 hours per year) engines for
stationary power generation and/or shaft driven equipment using natural gas and/or propane?

       EMA suggested that we add clarifying language to §1068.3 l(c) to specify that the 12-
month limit does not apply for fixed engines (i.e., neither portable nor transportable).

       Cummins raised question about how to apply the definition of "nonroad engine" for
engines that are installed in a fixed location. Aside from clarifying whether the requirement for
such an engine to operate for at least  12 months to be considered stationary, they suggested
adding regulatory language that would:
          •   Allow certain engines to be considered 'stationary' even if not at a given site for
              more that 12 months.  Aside from the seasonal engine provision, this should
              include engines in natural gas production that are connected to the fixed fuel
              supply but that might need to move early due to lack of production. (Note that
              these engines are typically mounted in a frame/skid with compressing equipment,
              radiator, etc, and typically sit on a prepared dirt  site.)
          •   NOT allow other engines connected to a fixed fuel supply for less than 12 months
              to be considered 'stationary'. An example recently discussed would be an SI
              genset mounted in a trailer on wheels.  Such a unit ought to be able to connect to a
              fixed, natural gas fuel  source for, say, three months without being considered
              stationary.
       Because of the need to have engines in different applications, attached to a fixed fuel
supply for a short period of time, to be treated differently, perhaps this would be better addressed
by 'intent'.  In the first example above, the intent would be for the unit to  operate more than 12
months in the given location.  In the second example, the intent would be to operate less than 12
months.  The first example ought to be a stationary situation; the second example ought to be
nonroad.  Cummins did not recommend specific language to accomplish this, but suggested the
following adjustment to paragraph (2)(iii) of the nonroad engine definition:
       "An 1C engine is not a nonroad engine if it meets any of the following criteria: ... (iii)
The engine otherwise included in paragraph (l)(iii) of this definition remains, or will remain, or
is intended to remain at a location for more than 12 consecutive months..." [emphasis added]

       In response to draft language for §1068.3 l(c) to clarify  the status  of stationary engines
that are neither portable nor transportable, Caterpillar suggested revised wording to make clear
that the residence-time restrictions are the subject of the sentence, not the definition of a nonroad
engine. The initial language was:
       "Note that the  definition of 'nonroad engine' in §1068.30 generally does not apply the residence-time
restriction to engines that are neither portable nor transportable in actual use."
       The suggested revision was:
       "For engines that are neither portable nor transportable in actual use, residence time restrictions generally
do not apply to the definition of a non-road engine in 1068.30."
                                           1-96

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
Letters:
Commenter
EMA
Impco
Exergy
Cummins
Caterpillar
EMA
Document #
0691
0692
0627
0785
0822
0808
Our Response:

       The regulation setting standards for stationary SI engines below 19 kW recognizes the
points raised by EMA about the interchangeability of nonroad and stationary engines in this size
range.  The regulation specifies a single label for engines below 19 kW whether they will be used
in nonroad applications, stationary applications, or both. It is important to note, however, that
state and local governments are not preempted from setting standards for stationary engines,
even if the engines are already  certified to meet EPA's nonroad standards.

       While the requirements for stationary and nonroad engines are identical, we may approve
certain procedures separately for one or the other of these applications. We would expect only
that the manufacturer would make a good-faith indication of how its engines might be used.
There would be no violation if that expectation turns out not to be true  in a way that could not
have been predicted ahead of time.

       It is not clear how Exergy could have thought that the proposed standards in part 1054 do
not apply to stationary engines, since we stated clearly in the preamble and in the regulations that
the standards apply equally to stationary and nonroad engines below 19 kW. There is no
exemption available for any of the engine types or applications noted by Exergy.

       The definition  of "nonroad engine" applies residence-time requirements to portable and
transportable engines to ensure that they are considered stationary only if they remain in one
location for an established duration. Fixed engines (lacking the features that would make them
portable or transportable) are inherently stationary, so the residence-time requirements do not
apply to them. We would expect such engines to remain in one location for longer than the times
we specify for portable engines, but it would not be a violation to move an engine before the
specified period was complete. We agree that it is appropriate to  add the clarifying language in
§1068.31 to state that "for engines that are neither portable nor transportable in actual use, the
residence-time restrictions in the definition of "nonroad engine" generally do not apply."
Cummins' suggestion  to rely on intent to determine whether an engine is stationary or not is
unworkable. This would effectively make it impossible to hold someone responsible for moving
an engine more frequently (or sooner) than is allowed under the regulation.

       We have no objection to Caterpillar's suggested adjustment to the language in
§1068.30(c) and have  modified the regulation accordingly.
                                       1-97

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
1.12  Technical amendments for nonroad diesel engines (40 CFR parts 89, 92, 94,1033,
      1039, and 1042)

What Commenters Said:

General Electric suggested that we modify the regulation in part 1033 to allow varying dilution
ratios for different test modes.

Letters:
Commenter
General Electric
Document #
0786
Our Response:

       We agree that it would be appropriate to address very small PM sample rates either by
extending the sampling time or by adjusting dilution ratios, consistent with good engineering
judgment.  The original requirement in §1068.515 specified only the extended sampling time, but
we believe the varying dilution ratios can be equally effective in making an accurate
measurement.
1.13  Benefit calculations for ozone mortality

What Commenters Said:

       Environmental Defense commented regarding EPA's Omission of Ozone Health Benefits
is Arbitrary and Capricious and Contrary to Law. The harmful effects of ozone on human health
and the environment are well documented.  Indeed, the body of science linking ozone with
premature mortality is one of the most significant developments in the last decade. The current
proposed rule, if implemented, will greatly reduce emissions of ozone-precursors thereby
achieving significant ozone-related health benefits.  In light of the robust  health benefits of this
proposal Environmental Defense objects to EPA's failure to quantify any ozone benefits,
including ozone-mortality. In its initial draft impact analysis, EPA estimated that by 2030 the
proposed spark-ignition engine standards would result in the reduction of 631,000 tons of
volatile organic hydrocarbon and 98,200 tons of oxides of nitrogen emissions. EPA projected
that these reductions will likely correspond to significant reductions in the formation of ground-
level ozone and would prevent between 60 and 360 ozone-related premature deaths, 800
hospitalizations, and almost 50,000 work days lost. EPA estimated the total benefits of this
proposed rule range between $3.9 billion and $6.1 billion annually. However, EPA ultimately
deleted all references to the above-listed quantified ozone benefits in response to pointed
comments by the White House Office of Management and Budget. In the final Draft Regulatory
Impact Analysis, EPA admits that it "typically quantifies ozone-related health impacts in its
regulatory impact analyses when possible" and that "[I]n the analysis of past  air quality
regulations, ozone-related benefits have  included morbidity endpoints and welfare effects such as
damage to commercial crops." Nevertheless, EPA states that it is deviating from its "typical"
practice due to a lack of conclusive scientific information as to how to quantify the benefits of
                                          1-98

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


ozone-related mortality. Furthermore, in order not to "provide an incomplete picture of all of the
benefits associated with reductions in emissions of ozone precursors," EPA declined to quantify
other health and welfare benefits.

       Environmental Defense strongly disagrees with the assumption that there is insufficient
information to include a valuation of mortality benefit. They also strongly object to EPA's
failure to quantify any ozone benefits, such as crop damage, lost work days or hospitalizations.
Environmental Defense believes that EPA's failure to consider the ozone mortality and
nonmortality benefits associated with the proposed rules is arbitrary, capricious and contrary to
law. See State Farm, 462  U.S. 29, 43 (finding agency failure to consider "an important aspect of
the problem"  arbitrary and capricious). Indeed, consideration of the full benefits of the emission
reductions at stake only underscores the imperative of the Agency adhering to its statutory
mandate under section 213 of the CAA to immediately take final action adopting emission
standards that reflect the "greatest degree of emission reduction achievable" and "take effect at
the earliest possible date."

Letters:
Commenter
Environmental Defense
Document #
0648
Our Response:

        Though omitted in the proposal for this rulemaking, EPA agrees that there is sufficient
evidence to include a valuation of the mortality benefit.  We therefore quantify and monetize the
ozone-related health impacts associated with the final rule, including both mortality and non-
mortality impacts.  This reflects EPA's most current understanding of the science surrounding
ozone impacts on human health and welfare, consistent with the recent ozone criteria document,
the analysis of the final ozone NAAQS, and the recently published report (April, 2008) by the
National Research Council titled, "Estimating Mortality Risk Reduction and Economic Benefits
from Controlling Ozone Air Pollution."
1.14 Air quality analysis

What Commenters Said:
       NACAA commented that state and local clean air agencies across the country are facing
the enormous challenge of developing strategies to achieve and maintain the health-based
National Ambient Air Quality Standards (NAAQS) for ozone and fine particulate matter (PM^.s).
Air quality in approximately 120 areas of the nation currently violates the 8-hour ozone standard,
the PM2.5 standard or both, exposing more than 150 million people to unhealthful levels of air
pollution. Clearly,  considerable efforts by EPA, states and localities will be needed to reduce the
widespread health  and environmental impacts associated with emissions from contributing
sources. In addition, EPA has already taken action to tighten the PM2.5 NAAQS and recently
proposed similar action on the ozone standard, which will increase the challenges facing states
                                       1-99

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
and localities. Further, many areas of the country are plagued by unacceptably high levels of
toxic air pollution.

       NACAA continued to comment that emissions from the nonroad spark-ignition engines
covered by this proposal are substantial. Use of lawn and garden equipment totals more than 3
billion hours a year. A gasoline-powered push mower currently emits as much hourly pollution
as 11 cars, a riding mower as much as 34 cars. Recreational watercraft can emit as much hourly
as 348 cars. The resulting emissions contribute to unhealthful concentrations of PM2.5, ozone,
CO and toxic air pollutants, which translate into serious adverse health impacts, including
premature death, heart disease, aggravated asthma and other respiratory conditions, as well as a
host of environmental harms, such as visibility impairment and acid rain.

       NACAA commented, as EPA appropriately acknowledges, absent action to reduce
emissions, by 2020 these engines will contribute more than one quarter (1,352,000 tons) of
mobile source volatile organic compound (VOC) emissions, nearly a third (16,374,000 tons) of
mobile source CO, 16 percent (39,000 tons) of mobile source PM2 5 and 4 percent (202,000 tons)
of mobile source NOx. However, the agency's proposal, by 2030, will reduce annual emissions
from affected sources by an estimated 630,000 tons of VOCs, 2.7 million tons of CO, 98,000
tons of NOx and 6,300 tons of direct PM2.5. Among the quantifiable benefits that would, in turn,
occur from these reductions is the prevention, annually, of an estimated 450 PM-related
premature deaths,  500 hospitalizations and 52,000 lost work days.  The total annual benefits  in
2030 are  estimated at $3.4 billion versus $240 million in annual costs.

       NJ DEP commented that advancing the federal implementation dates would provide more
timely air quality and health benefits to the residents  of New Jersey. Ozone continues to be one
of the most pervasive air quality problems in New Jersey.  The 2002 New Jersey Emissions
Inventory indicates that approximately 14% of the volatile organic compounds (VOCs) are from
spark-ignited nonroad engines and equipment rated at 25 horsepower or less.  On June 15, 2007,
New Jersey submitted its proposed State Implementation Plan for the attainment of the 8-hour
ozone National Ambient Air Quality Standard (NAAQS). It was an enormous challenge to
develop strategies to reduce ozone in order to meet the 8-hour ozone NAAQS. When the USEPA
revises the 8-hour ozone standard to be more protective of human health, New Jersey will face
greater challenges to develop strategies to reduce ozone. Aligning the federal implementation
dates more closely to the California ARB implementation dates will assist New Jersey in meeting
these challenges.

       South Coast AQMD commented on the South Coast Air Basin Air Quality Setting. As
EPA is aware, the South Coast Air Basin (Basin) is designated nonattainment for the federal
annual PM2.5 and 8-hour ozone ambient air quality standards and must attain these standards by
2015 and 2024, respectively. To achieve these deadlines, attainment must be demonstrated in the
preceding years, e.g. 2014 for PM2.5. Compared  to the nation's other nonattainment areas, the
South Coast Air Basin has the highest population-weighted ozone exposure of any area,
representing 24 percent of the nation's 8-hour ozone  exposure as well as its highest ozone design
value. Almost 90 percent of the nation's total population-weighted exposure to fine parti culates
occurs in California. In addition, 52 percent of the nation's total exposure to fine parti culates
occurs in the South Coast Air Basin alone.
                                         1-100

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
       SCAQMD continued that these pollutant exposures result in severe public health impacts
in the South Coast Air Basin. Numerous studies - conducted locally, nationally and
internationally - confirm that ozone and particulate pollution have a direct impact on respiratory
health, increasing asthma attacks, bronchitis, emphysema, chronic obstructive pulmonary
disease, lung cancer and premature death. For example, studies in Southern California have
found a significant risk of irreversible decline in lung function among children growing up in
areas with relatively high particulate pollution. In addition, California ARB has estimated that
particulate pollution in the Basin causes 5,400 premature deaths, 2,400 hospital admissions,
140,000 asthma and respiratory symptoms, 980,000 lost workdays, and 5 million restricted
activity days for minors, every year.

       SCAQMD commented that the region is moving ahead with efforts to attain the federal 8-
hour ozone and annual PM2 5 standards with the recent adoption of the 2007 Air Quality
Management Plan (AQMP) for the South Coast Air Basin. The attainment challenges are
significant given that stationary sources are now generally controlled to over 90 percent, and
about 80 percent of the emissions in the Basin are caused by mobile sources. The attainment
demonstrations provided in the 2007 AQMP show that the Basin cannot timely attain federal air
quality standards without significant emission reductions from all sources, including nonroad
spark-ignited engines. By 2023, pleasure craft will be the third highest VOC  emission source
category in the Basin (about 35 tpd) surpassing passenger cars, petroleum marketing, and
architectural coating categories. To address this need, as part of the proposed state strategy for
the California State Implementation Plan, California ARB is proposing new standards for
outboard/personal watercraft to be implemented in 2013 that will reduce exhaust emissions of
NOx and VOC by close to 70%.  The 2007 AQMP also includes an additional control strategy
that calls for accelerated turnover of outboard and personal watercraft engines to engines
meeting the most stringent existing California standards as well as more stringent standards
adopted in the 2014 timeframe for inboard and stern drive marine engines in  order to achieve
reductions. Achieving the maximum amount of emission reductions from nonroad spark-ignited
engines is critical to the South Coast Air Basin for meeting the federal 8-hr ozone and annual
PM2.5 standards.

       NY DEC commented that the ozone forming emissions of the engines targeted by this
proposal are significant, and occur primarily in the summer ozone season. As EPA notes in
Section XII-A of the Preamble, recreational marine and small land spark-ignition engines
account for over a quarter of the national mobile source VOC inventory. The contribution of
these emissions to ground level ozone formation is even greater because most of the use of these
engines occurs during the summer ozone season, when most water-borne recreation, lawn and
landscape maintenance, and outdoor construction takes  place. Marine recreation, and the
associated emissions are also concentrated geographically in areas with suitable waterways. Thus
the fraction of the national inventory comprised of emissions from these engines likely
understates their impact on ozone levels and National Ambient Air Quality Standard violations.
There is clearly a need for the standards proposed by EPA.

       Pennsylvania DEP commented that small land-based, spark-ignition engines and
equipment and marine spark-ignition engines and vessels contribute significantly to the
                                      1-101

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
precursors of ground level ozone. At the existing emissions rates, these engines and equipment
are expected to contribute as much as 10 to 15 percent of all the volatile organic emissions in
2009 in the 37 counties in Pennsylvania that are currently designated by EPA as eight-hour
ozone nonattainment areas. As Pennsylvania and other states face the challenges in attaining and
maintaining the existing and anticipated more protective ozone National Ambient Air Quality
Standards and the fine particulate standards, EPA must move forward expeditiously with its full
complement of controls on new mobile source engines. The projected emission reductions and
health benefits anticipated by 2030 under this proposal are significant (631,000 tons of volatile
organic hydrocarbon emissions, 98,200 tons of Nitrogen Oxides emissions, and 6,300 tons of
direct particulate matter PM2.5 emissions, and 2.69 million tens of carbon monoxide emissions)
and should, therefore, be achieved expeditiously.

       Houston-Galveston Area Council  commented that the Houston-Galveston-Brazoria
(HGB) region is currently classified as a non-attainment area for the 8-hour ozone National
Ambient Air Quality Standard (NAAQS). It should be noted that in a letter dated June 15, 2007,
Texas Governor Rick Perry requested that the HGB region be reclassified as  severe
nonattainment for the 8-hour ozone NAAQS. Additionally, this region is very close to exceeding
the PM2 5 standard and has elevated levels of air toxics in localized areas. These air quality issues
result in negative economic impacts,  ecosystem damage, negative health effects, and a reduction
in the quality of life in the HGB area. The unique industrial characteristics of the region
combined with being one of the largest urbanized population centers in the nation present a
particularly difficult challenge in terms of improving air quality.  Though efforts to improve air
quality have yielded some progress over the past decade, it is clear that much more needs to be
done. The active support and participation of the federal government has been vital to air quality
improvement efforts; however more than half of the ozone-forming pollution in this region is
generated by mobile sources. Regulation of the emissions from mobile sources is outside the
authority of individual states such as  Texas.

       Environmental Defense commented that it is well documented that, despite their size,
spark-ignition ("SI") marine and small engines contribute significantly to the formation of ozone
and other harmful air pollutants. EPA estimates that absent the implementation of these
proposed rules, the emissions from spark-ignition marine and small engines will account for 27%
of volatile organic hydrocarbon compounds (1,352,000 tons), 31% of carbon monoxide
(16,374,000 tons), 4% of oxides of nitrogen (202,000 tons) and 16% of particulate matter
(39,000 tons) from the mobile source sector by 2020.

       Environmental Defense continued to comment that spark-ignition marine and small
engines contribute to unhealthy air pollution concentrations of ozone, carbon monoxide (CO)
and PM in numerous areas nationwide. The air pollutants emitted by these engines are associated
with a host of adverse public health effects including acute respiratory problems, asthma,
aggravation of cardiovascular conditions and decreased lung function. Acute exposure to CO can
cause death and non-fatal poisoning.  Gaseous vapors that escape from the fuel lines and tanks
during gas refueling and accidental spills  cause and contribute to carcinogenic and non-
carcinogenic health problems. Exposure to ozone and particulate matter can cause premature
death. The immediate final promulgation  and implementation of EPA's proposal will help to
ensure cleaner air and improved health for millions of Americans.
                                         1-102

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
Environmental Defense commented on the following four pollutants:
       A. Ozone
       The SI small and marine engines subject to this proposal consist primarily of lawn and
       garden equipment and recreational boats utilized mostly during the hot summer months.
       As a result, the emissions from these engines play a particularly significant role in the
       formation of ground-level ozone. Ozone is formed by the combination of HC and NOx in
       the presence of heat and light.  According to EPA, spark-ignition marine and small
       engines not only produce about one fourth the amount of smog forming HC as all of the
       cars on the road today but their emissions are concentrated during conditions especially
       conducive to ozone formation. California officials report that, on a gallon for gallon
       basis, these engines discharge  93 times more smog forming emissions than model year
       2006 cars.

       Approximately 157 million people are exposed to levels of ozone or "smog" that exceed
       the current national health-based standard. Ozone causes acute respiratory problems,
       asthma, reduced lung function and increased hospital admissions. Children and the
       elderly are most at risk. Recently, a federal advisory panel recommended EPA tighten the
       existing 8-hour ozone health standard due to mounting evidence that it fails to protect
       adequately human health.  Scientific studies  from the United States and Europe link short
       term increases in ozone levels  to increased rates of death from respiratory and
       cardiovascular disease. Day-to-day increases in ozone concentrations during the summer
       have been linked to an increase in premature death. Final promulgation and
       implementation of EPA's proposal would aid significantly in preventing ozone-related
       illnesses,  work absences, and deaths.

       Recent scientific  information also demonstrates the harmful effects of ozone on plants
       and ecosystems. According to  the EPA, ozone impairs crops, native vegetation, and
       ecosystems "more than any other air pollutant." Indeed, in examining forest productivity
       and ecosystem diversity, ozone may be the pollutant with the "greatest potential for
       regional-scale forest impacts." Exposure to ozone weakens plants, making them more
       susceptible to disease, insects and climatic changes. Changes in the biodiversity of plants
       and trees  can affect entire ecosystems given the central role vegetation and forests play in
       providing food and habitat for many species offish, birds and mammals.

       B. Particulate matter
       Spark-ignition marine and small engines are also significant producers of particulate
       matter. PM is a mixture of soot, smoke, and tiny particles due to direct PM as well as PM
       formed in the atmosphere from precursors such as sulfur dioxide, nitrogen oxides,
       ammonia and other pollutants. Scientific studies have shown a statistically  significant
       relationship between short term exposure to PM from mobile source emissions and
       mortality. Results from a recent study on the contribution from mobile source emissions
       of PM in  14 U.S. cities indicates that mobile sources, such as the small and marine
       nonroad engines affected by this proposal, have a greater effect on the toxicity of ambient
       air than other sources.
                                      1-103

-------
                         Chapter 1: Rulemaking Process and Cross-Program Issues
Approximately 88 million people across the country are exposed to unhealthy levels of
PM. Another 27 million are likely to live in areas with unhealthy levels of PM if steps are
not taken to reduce PM emissions. Exposure to particulate matter can cause acute
respiratory symptoms, decreased lung function and increased hospital and emergency
room visits. Exposure to PM  has also been linked to death from cardiopulmonary
disease, premature death, lung cancer  and infant mortality. People with heart or lung
disease, the elderly, and children are most at risk. The proposed standards, if
implemented, would prevent 460 premature deaths, 52,000 days of missed work, 500
hospital admissions, and 310,000 restricted-activity days. They would also greatly  assist
states and local governments in attaining or maintaining air pollution concentrations
below the health-based NAAQS.

Particulate matter also causes a host of adverse environmental effects. PM impairs
visibility, both by contributing to local and regional haze. The brown clouds that hang
over many urban areas and haze surrounding our national parks and wilderness areas are
caused by particulate matter. Reducing the PM emitted from small and marine engines
will improve human welfare by helping to reduce these forms of visibility impairment.
The SO2 and NOx that can transform  in the atmosphere to PM also causes atmospheric
deposition and acid rain. Acid rain is primarily responsible for elevated levels of acid in
the many fresh-water bodies that dot the U.S. upper- Midwest and Northeast. High
acidity in lakes and streams alters the  chemical composition of the waters, leading  to
changes in vegetation, species loss and contamination. Atmospheric deposition occurs
when SO2 and NOx deposit into streams, lakes and forest beds. The deposition alters
water quality and vegetation and can lead to toxic algae and plankton blooms which can
threaten human health and welfare.  Immediate implementation of these standards will go
a long way in improving human health and welfare and protecting our streams, lakes,
forests and their inhabitants.

C. Carbon monoxide
In 1994 EPA determined that the lawn and garden equipment subject to these proposed
rules contribute significantly to unhealthy CO concentrations. EPA currently is proposing
to make a similar determination with respect to CO emissions from SI marine boats. EPA
estimates that approximately 15 million people live in areas with unhealthy levels of CO.

Like exposure to ozone and particulate matter, exposure to CO causes a number of
serious health effects. CO reduces the  delivery of oxygen to the body's organs and is
associated with impairment of visual perception, work capacity, manual dexterity,
learning ability and performance of complex tasks. The health threat posed by CO  is
particularly acute for individuals suffering from cardiovascular disease. CO emissions
also contribute to the formation of ground-level ozone. Exposure to acute levels of CO
can result in fatal and non-fatal CO  poisoning. Between 1984 and 2004 there were 113
reported deaths and 458 non-fatal poisonings caused by exposure to CO. Recreational
boaters, inhabitants of house boats,  and people swimming around docks are primarily at
risk of accidental fatal and non-fatal CO poisoning. A number of federal agencies have
issued health advisories to warn recreational boaters of the serious threats posed by
exposure to CO. Immediate finalization of the proposed rules will reduce the number of
                                   1-104

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       CO-related deaths and illnesses and help to ensure that people living, working, or
       recreating around SI marine boats and vessels can do so safely.

       D. Toxic Air Pollutants
       Gaseous air toxics, such as benzene, 1, 3 butadiene, formaldehyde, acetaldehyde and
       napthalene, comprise another major category of air pollutants emitted by these small
       engines. Exposure to the vaporous air toxics emitted from these engines causes
       carcinogenic and non-carcinogenic health effects. According to EPA's 1999 National-
       Scale Air Toxics Assessment, all of the air toxics emitted by these small engines, with the
       sole exception of acetaldehyde, comprise a significant portion of the total inhalation
       cancer risk from mobile sources. Air toxics also cause a number of other serious
       noncancer health problems involving the neurological, cardiovascular, liver, kidney,
       respiratory, immune and reproductive systems.

       One of the air toxics emitted by these small engines, benzene, poses a particularly serious
       threat to human health. Benzene is one of the most significant contributors to cancer risk
       of all air toxics in the ambient air. Assuming continuous exposure to 1999 levels of all
       outdoor air toxics, the nationwide lifetime population cancer risk was 42 per million.
       According to EPA, benzene was responsible for 24% of this cancer risk, and was
       responsible for 42% of the total inhalation cancer risk from mobile source air toxics.
       EPA's proposal to require more stringent measures to control the toxic evaporative
       emissions from these engines is essential in reducing the cancer, and non-carcinogenic,
       inhalation risk from mobile sources.

Letters:
Commenter
South Coast AQMD
NJDEP
NY DEC
Houston-Galveston Area Council
Environmental Defense
NACAA
Pennsylvania DEP
Document #
0704
0710
0659
0633
0648
0651
0676
Our Response:

       We agree that emissions from small SI and marine SI engines are significant and often
occur during the ozone season.  The final rule estimates that these engines emit over 2 million
tons of volatile organic compounds (VOCs) and almost 170,000 tons of NOx annually, and
contribute to adverse health and welfare effects associated with ozone, PM, NOx, VOCs
including toxic compounds, and carbon monoxide (CO).  Without this rule, emissions from
Small SI and Marine SI engines, equipment and vessels would continue to grow and would
become a larger percentage of total mobile source emissions.  By 2030 this final rule will reduce
VOCs by 604,000 tons and NOx by 132,200 tons annually.
                                      1-105

-------
                               Chapter 1: Rulemaking Process and Cross-Program Issues
       This rule will help states to reduce air toxics and meet the health and welfare based
National Ambient Air Quality Standards (NAAQS) for ozone, PM and CO. As of March 12,
2008 there are approximately 140 million people living in 72 areas designated as nonattainment
with the 1997 8-hour ozone NAAQS. In addition, approximately 88 million people live in areas
that are designated as nonattainment for the 1997 PM2.5 NAAQS and 850,000 people live in
areas that are designated as nonattainment for the CO NAAQS. Both the ozone and PM2 5
NAAQS have been amended in the last few years and are now more stringent. States with
nonattainment areas are required to take action to bring those areas into compliance in the future.
We expect many of the ozone and PM2.5 nonattainment areas will need to adopt additional
emissions reduction programs to attain and maintain the NAAQS.  The emission standards being
finalized in this action will become effective between 2009 and 2013 and will be useful to states
in both attaining and maintaining the NAAQS.  For discussion on the timing of the standards, see
Sections 3.2.2, 3.3.2 and 3.4.2

       According to air quality modeling performed in conjunction with this rule, the emissions
reductions will result in nationwide improvements in ambient ozone concentrations as well  as
decreases in PM2 5 concentrations. By 2030 these reductions will annually prevent 230 PM-
related premature  deaths (based on the ACS cohort study), between 77 and 350 ozone-related
premature deaths (assuming a causal relationship between ozone and mortality), 1,700 hospital
admissions and emergency room visits, 23,000 work days lost, and approximately 590,000 minor
restricted-activity  days.

1.15 Other issues

What Commenters Said:

       OPEI noted that substantial evaporative benefits and exhaust emissions benefits will be
achieved through the investment of OPEI members to produce Phase 3 compliant products.
However, OPEI shares the concerns of state organizations and environmental groups, like the
American Lung Associations and Clean Air Watch, that these air quality benefits and related
investments could be undermined if EPA approves any new waiver for fuels containing greater
that 10% ethanol.  OPEI commented that EPA must fulfill its statutory obligations under section
21 l(f)(4) of the Clean Air Act to carefully review and respond to any waiver for mid-level
ethanol fuel blends. The 250 million Americans that own and operate over 400 million
lawnmowers, chainsaws, boats, motorcycles, snowmobiles and automobiles are relying on EPA
to make sure that neither their products, nor the environment, are damaged through the approval
of fuels containing greater that 10% ethanol. OPEI also submitted a technical report to EPA to
highlight the expected adverse impacts of mid-level ethanol blends (see docket item EPA-HQ-
OAR-2004-0008-0746).

Letters:
Commenter
OPEI
Document #
0675
                                         1-106

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


Our Response:

       Although there has been interest expressed by some areas for a mid-level ethanol fuel
blend (i.e., containing more than the current 10 percent ethanol blend), EPA has not received an
application for a waiver request at this time.  Should EPA receive such a request, EPA would
fulfill its statutory requirements under the Clean Air Act in responding to the waiver request. It
can be noted that the recent Energy Independence and Security Act of 2007, revised section
21 l(f)(4) of the Clean Air Act.  While the basic criteria for analyzing a waiver request stayed the
same, the revised language states that EPA must also analyze the impact on nonroad engines and
vehicles. In addition, EPA "must take final action to grant or deny an application submitted
under this paragraph, after public notice and comment, within 270 days of the receipt of such an
application."  Prior to the recent change, section 21 l(f)(4) said the waiver would be treated as
being granted if EPA did not act within 180 days.
                                      1-107

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


2   Exhaust Emission Standards and Related Requirements for Small SI Engines

What We Proposed:

       The comments in this section generally correspond to Sections V and VII of the preamble
to the proposed rule, where we describe the proposed emission standards and certification
procedures associated with exhaust emissions from Small SI engines.  The applicable regulatory
provisions for these proposed requirements are in 40 CFR parts 90 and 1054.  The Regulatory
Impact Analysis describes the feasibility of these standards, special provisions that apply to small
businesses, and alternative standards under consideration in Chapters 4, 10, and 11, respectively.

   See Chapter 1 of this document for a discussion of issues that apply more broadly than only
for Small SI engines.  See Chapter 4 of this document for a discussion of issues related to
evaporative emissions.

       2.1   Scope and applicability

   2.1.1   Definition of handheld

What Commenters Said:

       OPEI commented that EPA appears to have two definitions of handheld indicated in
1054.101(c) and 1054.801. In order to prevent the unintended reclassification of these products
in 1054.801, OPEI  commented that EPA should keep the newly proposed weight limits intact but
make a revision to the definition of a handheld engine in 1054.801 by adding paragraph (6) Is
used in a portable hand-supported jackhammer/rammer, compactor (vibratory or other) or other
similar product. As an alternative, EPA could add a statement to the definition in 1054.801
indicating all engines/product less than 80cc are automatically handheld regardless of weight etc.
OPEI also commented that paragraph (4) of the definition should be revised to eliminate "one-
person" since many augers using handheld engines can be operated by two-persons.

       EMA commented that the NPRM properly categorizes equipment utilizing engines less
than or equal to 80cc total displacement as "handheld." The NPRM also correctly categorizes
equipment utilizing engines with larger than 80cc total displacement, but also meeting additional
requirements, as eligible for categorization as handheld.  Such engines should be allowed to
continue to meet handheld exhaust standards, and should be considered handheld
engines/equipment for purposes of the new evaporative standard requirements. In addition,
equipment that EPA has historically approved as meeting the definition of "handheld," such  as
compactors/rammers, should be allowed to continue to be categorized as handheld and should be
specifically included in  the regulation in order to ensure that all industry and agency personnel
are aware of the appropriate determining factors.

       EMA submitted comments on EPA's  proposal to modify the handheld definition by
increasing each of the specified weight limits by 1 kilogram (72 FR at 28141). EMA agrees  that
an adjustment is required. However, they commented that the proposed adjustment is
insufficient for the  conversion of prior emission control engines to either catalyzed two-cycle
                                       2-1

-------
                                                            Chapter 2: Small SI Engines
engines or four-cycle engines, as required to achieve exhaust emission standards. EMA
recommended that the handheld definition be adjusted by increasing each of the specified weight
limits by 2 kilograms.

       The California Air Resources Board (CARB) believes categorizing handheld equipment
is best done by engine size such as the 80cc limit set by California. This gives the engine
manufacturers an emissions design target at the beginning of the process.  To the extent EPA
believes it necessary to maintain the handheld category above 80cc, CARB supports the change
in weight limits for handheld equipment.  CARB commented that the increase of one kilogram,
representing the approximate additional weight related to switching to a four-stroke engine, is an
appropriate adjustment.

       Honda submitted comments in agreement with EPA's proposal that would allow engines
less than 80cc to comply with both handheld exhaust and evaporative emission  standards.
Honda commented that the language on evaporative emissions should be clarified to include
these engines. Additionally, Honda commented that engines above 80cc could  then use the
equipment-based handheld definitions to quality for the handheld category. Honda
recommended that EPA specifically add earth rammers to the handheld category rather than
continuing to rely on the Phase 2 guidance that they qualify as  handheld products.  Finally, with
the direct inclusion of 0 to 80cc engines in the handheld emission category, Honda believes the
proposal's definition for handheld equipment should be given careful reconsideration.  This may
be particularly true for products with weight limits of 14 or 20kg.  Honda questioned whether it
is necessary for generators and pumps less than 14kg (15 kg proposed) to be considered
handheld by definition.

Letters:
Commenter
OPEI
EMA
CARB
Honda
Document #
0675
0691
0682
0705
Our Response:

       EPA does not believe the regulations contain two different definitions of handheld. The
EPA regulations define "handheld" in §1054.801 by specifying the criteria that are to be used to
determine if an equipment application is handheld and therefore subject to the various handheld
requirements of Part 1054. Section 1054.101 describes which exhaust standards apply to the
different types of engines. Paragraph (a) notes that all handheld engines (i.e., those that meet the
definition in §1054.801) must meet the handheld exhaust standards. In addition, paragraph (c)
notes that all engines at or below 80cc will be subject to the handheld engine standards,
regardless of the type of application the engine is ultimately placed in. The provision in
paragraph (c) does not mean the engine is a handheld engine. It only means that the engine is
subject to the handheld exhaust standards. Therefore, EPA believes both of the  regulatory
provisions noted above are necessary and have been retained in the final rule.
                                          2-2

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       In response to the comment on the 80cc outpoint, EPA cannot use 80cc as the only
criteria for whether an engine is subject to the handheld exhaust standards because there are
many products that use engines above  80cc which qualify under the criteria contained in the
handheld definition of §1054.801. CARB uses the 80cc cutpoint in its regulations.  However,
CARS's regulations do not apply to those products above 80cc considered handheld under
EPA's  definition due to the construction and farm equipment pre-emption provisions of section
209(e)  of the Clean Air Act. Therefore, EPA is retaining a definition of "handheld" in the final
rule.

       It should be noted that the proposal based the cutpoint for the applicability of the
handheld provisions on engines "less than" 80cc.  As noted in the proposal, this change was
intended to harmonize with the displacement-based requirements for CARB. During
development of the final rule, it came to EPA's attention that the CARB cutpoint is based on
engines "at or below" 80cc.  EPA has modified the final rule regulations to include this approach
so that  the EPA and CARB requirements are the same.

       With regard to the comments on hand-supported jackhammers, rammers and compactors,
EPA agrees that the definition of handheld should include a specific reference to such
applications. In response to requests from equipment manufacturers in the past, EPA has
approved the manufacturers' request to consider such applications as handheld based on the
criteria spelled out in the handheld definition on multi-position use. Therefore, EPA believes it
makes sense to include the hand-supported jackhammers, rammers, and compactor applications
specifically in the handheld definition.

       With regard to the comment on augers, EPA is removing the "one-person" term from the
auger description in the handheld definition.  EPA acknowledges that some augers can be
operated by two people, but still have other attributes that would lead to the equipment being
considered a handheld application, including the dry weight of the equipment. Therefore, EPA
believes the "one-person" terminology is not needed with respect to augers.

       In response to the comments on whether a special provision for pumps and generators is
needed given the requirement that all engines at or below 80cc can meet the handheld standards,
EPA investigated the current certification information to see how many engines above 80cc are
used exclusively in pumps and generator applications that would fall under the 15 kilogram
weight limit (engine and equipment combined) included in the proposed definition. While EPA
was able to identify a few engine model used in such applications, sales of such engines were
extremely low. EPA sees no technical reason why such applications would need to use engines
certified to the handheld standards and is therefore removing the pump and generator language
from the handheld definition in §1054.801 of the regulations

       In response to the comments on the proposed weight limits in the handheld definition,
EPA looked at similar equipment applications in which the engine is similarly sized, but
powered by either a 4-stroke engine or a 2-stroke engine.  Based on an analysis of similarly
designed string trimmers, the dry weight of a 4-stroke trimmer with a 25 cc engine was
advertised at 13 pounds, whereas the dry weight of two different 2-stroke trimmers with similar
sized engines (24.5cc and 28cc) was advertised at 9.5 and 10.6 pounds. Therefore the 2-stroke
                                       2-3

-------
                                                            Chapter 2: Small SI Engines
trimmers were 3.5 and 2.4 pounds (1.6 and 1.1 kilograms) lighter than the 4-stroke trimmer.
Based on this comparison, EPA agrees that it is reasonable to raise the weight limits in the
handheld definition by 2 kilograms instead of the proposed 1 kilogram increase to account for
the increased weight of switching to a 4-stroke engine. Therefore, EPA is adopting a 16
kilogram weight limit in the handheld definition for most equipment with a 22 kilogram weight
limit for augers.

       Finally, with regard to Honda's comments that the language on evaporative emissions
should be clarified to include all engines at or below 80cc under the handheld evaporative
requirements, EPA agrees in principle.  For the purposes of the exhaust emission standards,
engines at or below 80cc are subject to the handheld exhaust standards.  Under the new
regulations, equipment manufacturers are allowed to use engines at or below 80cc in either
handheld or nonhandheld equipment. Because the applicability of the evaporative emission
standards is based on the type equipment, an engine at or below 80cc used in a nonhandheld
piece of equipment (that is subject to the handheld exhaust standards) would be subject to the
nonhandheld equipment evaporative standards. EPA believes this could be difficult, especially
with regard to running loss requirements that apply to nonhandheld equipment but not handheld
equipment. Therefore, the final regulations require nonhandheld equipment to comply with the
nonhandheld evaporative emission standards unless it is using an engine at or below 80cc.  In
that case, the equipment manufacturer would need to demonstrate  compliance with both the fuel
line and fuel tank requirements in 2012.  The running loss requirement would not apply to
nonhandheld equipment using engines at or below 80cc.  (It can be noted that EPA is adopting a
similar provision for nonhandheld engines which are used in handheld equipment.  In such a  case
the equipment would be subject to the handheld evaporative emission standards which do not
require control of running losses. The fuel line and fuel tank requirements would apply and take
effect in 2012.)

   2.1.2  Small SI vs. Large SI

What Commenters Said:

       EMA noted that the current differentiation between Small SI and Large SI nonroad
engines is principally determined based on the power of the engine (e.g., less than or equal to 19
kW).  In addition, engine manufacturers have the discretion to categorize engines that have
power greater than 19 kW, but less than or equal to 30 kW, with total engine displacement less
than or equal to 1,000 cc, as Small  SI engines. The current differentiation should not be
changed.

       EMA noted that the NPRM introduces restrictions regarding total engine displacement
through the addition of one significant figure to the displacement determination, and provides a
clarification  stating that all engines produced must be included in the displacement
determination. Such clarification requires that all production tolerances be included in the
determination of maximum production displacement. EMA commented that the regulatory
requirements should be clarified in order to avoid confusion regarding the product category
applicability, and the final rule should include the proposed clarification that all engines,
including tolerances, must be within a category.  However, the proposed additional significant
                                          2-4

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


figure for representation of engine displacement is not necessary. In fact, the addition of such
significant figure may result in unintended consequences associated with engine designs that are
currently classified as Small SI.  Accordingly, EMA commented that EPA should not require that
engine displacement be calculated to an additional significant figure.

       EMA noted that §90.116(a) requires the total engine displacement to be rounded to the
nearest whole cubic centimeter, but paragraph (g) requires the total displacement to be rounded
to the nearest 0.1 cc. EMA suggested that §90.116(g) be revised to reflect the nearest whole
cubic centimeter as required by §90.116(a).

       EMA commented on §1054.615(b) "What is the exemption for engines certified to
standards for Large  SI engines?"  EMA noted paragraph (b) refers to paragraph (f) of the same
section, however the section does not include a paragraph (f). EMA believes the correct
reference should be  to paragraph (d).

       Kohler noted that EPA is proposing to modify the  criteria used to determine the
displacement for the large SI one liter exemption. Kohler commented that it is opposed to this
change. Kohler provided comments on specific sections in the proposed regulations. They noted
that §90.116(g) has been added which limits the displacement of each engine produced to 1000.0
cc after rounding to  the nearest 0.1 cc.  This is a change to the previous requirement of
calculating displacement using nominal engine values and rounded to the nearest whole cubic
centimeter. This changes the rules established in §1048.615(a)(l) and 90.1 16(a) after the
regulations have been implemented.  Kohler requested that the previous wording be retained. If
it is not, any engine  families certified to the current Part 90 wording should be grandfathered and
this change should not take effect until the Phase 3 regulation is implemented in 2011. Kohler
noted that §1054.140(d) limits the displacement of each engine produced to 1000.0 after
rounding to the nearest 0.1 cc. This is a change to the previous requirement of calculating
displacement using nominal engine values and rounded to the nearest whole cubic centimeter.
Kohler doesn't believe this change is justified and requested the current wording in Part 90.116
be retained.

Letters:
Commenter
EMA
Kohler
Document #
0691
0703
Our Response:

       We agree with the commenters' position that the differentiation between Small SI and
Large SI engines should not be changed, and specifically that the 1000 cc threshold should not
be changed to 1000.0. If we had done this originally, manufacturers could have easily planned
for that and taken steps to ensure that nominal engine dimensions and production tolerances were
adequately controlled to stay below the threshold. Since we did not adopt the more precise
threshold, manufacturers have in good faith designed their engines consistent with the
regulations as published.  We do not believe there is a sufficient environmental benefit
corresponding to the more precisely defined threshold to justify the costs associated with
modifying engine designs in this way.


                                       2-5

-------
                                                            Chapter 2: Small SI Engines
       The concern related to ensuring that every engine is below the threshold comes from the
realization that we might have had a difficult time establishing that there was a violation if the
manufacturer had declared a nominal value that was below the threshold, even though production
variability could arguably lead to substantially higher displacement values. In the context of
highway motorcycles that are subject to different standards if they are over 50 cc, we have seen
examples of wide variations in displacement values above 50 cc where the manufacturer claimed
to be in compliance with regulatory requirements.  Kohler has pointed out that their particular
situation involves production variability that would be problematic if the threshold were 1000.0
cc, but not if the threshold were 1000 cc.  We are modifying the regulatory language to  specify
that the declared displacement value must be within the range of actual values for production
engines, taking into account normal production variability.  This approach is similar to what we
specify for declaring maximum engine power in §1054.140. This should allow us to
meaningfully implement and enforce the 1000-cc threshold without changing the meaning of the
current regulations for those who are already complying in good faith.

       We have modified the regulation language to more clearly state that engines voluntarily
certified to the exhaust emission standards for Large  SI engines  in part 1048 are also subject to
evaporative emission standards under part 1048. Since Large SI evaporative requirements fall to
the engine manufacturer, there should not be  a situation where an equipment manufacturer
becomes subject to EPA standards because of the engine manufacturer's choice to certify to
more stringent exhaust emission standards. In fact, equipment manufacturers may in the end
meet evaporative requirements for Small SI engines (especially for running loss control) even
though they don't need to. This would not be a violation.  We believe this regulatory
arrangement represents the clearest and most natural  division of responsibilities among  the
affected companies.
   2.1.3  Maximum engine power and displacement

What Commenters Said:

       EMA believes the NPRM introduces significant additional complexities with respect to
the determination of maximum power. EMA commented that the final rule should clarify that
the power reported by the engine manufacturer may appropriately be determined utilizing the
engine manufacturer's good engineering judgment and the appropriate industry standard for
power measurement.

       EMA commented on §1054.140 "What is my engine's maximum engine power and
displacement?" EMA believes the proposed language is both excessive and incomplete.
Specifically, they commented that the requirement to map engines pursuant to 40 CFR Part 1065
is not appropriate for small air cooled Small SI engines, and the maximum engine power does
not specify a rating procedure. The proposed requirement to include all engines in the
displacement determination, as well as including the additional significant figure to the
displacement reporting is not appropriate. EMA believes this section should include only those
requirements that are significant to the determination of whether an engine family should be
                                          2-6

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


classified as a Small or Large SI engine. Accordingly, EMA commented that this section should
be revised to read as follows:
       (a) An engine configuration's maximum power is the power level assigned by the engine
       manufacturer as determined using an industry standard power measurement procedure.
       Engine families where the maximum modal power of the emission-data engine is greater
       than 15 kW at the test speed designated require manufacturers to include the brake power
       for engines in the certification application for the family as prescribed by 1054.205.
       (b) An engine configuration's displacement is the intended swept volume of all the
       engine's cylinders.  The swept volume of the engine is the maximum product of the cross
       section area of the cylinder bore, the stroke length, and the number of cylinders including
       all tolerances. Determine the final value by rounding the final result to the nearest 1 cc.
       (c) Deleted in its entirety.
       (e) Deleted in its entirety.

       EMA noted that §1054.1(a)(l) states that the requirements of Part 1054 apply to engines
with "maximum engine power at or below 19kW." EMA commented that it is not clear what
type of power level is being described.  As the definitions set forth in §1054.801 include a
definition of "brake power," EMA commented that §1054.1(a)(l) should be revised to read as
follows: "maximum engine brake power at or below 19kW."

       OPEI noted that EPA explains in 1054.205(a) that this section only applies if the engine
is 15 kW or greater. OPEI commented that the language of §1054.140 should be modified to
explain it is not applicable to engines less than 15 kW or less than 0.95 liters.

       Kohler noted that §1054.801 defines "Displacement" to have the meaning given in
1054.140, which is changed from the current provisions in §90.116. Kohler commented that the
current meaning in  §90.116 should be retained. Kohler also objected to the extensive reporting
and recordkeeping requirements in the proposed rule. One of the items Kohler noted was the
requirement to report maximum engine  power in the application for certification.

Letters:
Commenter
EMA
OPEI
Kohler
Document #
0691
0675
0703
Our Response:

       Appropriately defining terms to establish an engine's displacement and maximum engine
power are important for ensuring that the regulations specify objectively and consistently which
emission regulations apply. Every engine should be unambiguously subject to a single set of
emission regulations—there should be no overlaps or gaps.  For maximum engine power, the
regulations need to differentiate Small SI engines from Large SI engines (and in some cases from
recreational vehicle engines). For displacement, the regulations need to assign each engine to a
Class for determining which standards apply under part 1054.  EMA accurately summarizes our
                                       2-7

-------
                                                            Chapter 2: Small SI Engines
objective by stating that we should "include only those requirements that are significant to the
determination of whether an engine family should be classified as a Small or Large SI engine."

       Part 90 in particular does not include such procedures and specifications for establishing
clear and objective determinations of power and displacement. We have chosen to adopt the new
approach in part 1054 in combination with the Phase 3 standards rather than introducing these
regulatory provisions as amendment to the Phase 2 program in part 1054.  Such a change could
cause unintended consequences by forcing an engine to be subject to a different set of standards
even though we are intending to leave the current standards intact.

       To accomplish this, the regulations must use consistent and objective parameters for
making these determinations.  It would not be appropriate to rely on a manufacturer's judgment
in establishing maximum engine power, because it would be impossible to ensure a proper
delineation between Small  SI and Large SI engines. Without an objective specification or
procedure, manufacturers would be free to manipulate the declared value to  choose the less
stringent standards. Both maximum engine power and displacement can be  measured using
standard procedures and specifications, so we believe the regulation should rely on these
procedures and specifications to determine those values.

       While engine mapping is not required to test Small SI engines, we believe it is entirely
appropriate to do engine mapping for engines where there is a need to demonstrate that the
engine's power falls within the specifications for regulation as a Small SI engine.  Mapping
procedures are specified in part 1065.  This measurement can be readily made when an engine is
mounted on an engine dynamometer.  The specified mapping procedure and the instructions for
determining maximum engine power constitute a complete rating procedure for these engines.
This may be different than  the manufacturers' current practice, but it is a rating procedure
nonetheless.

       We specify that the power and displacement values determined under §1054.140 fall
within the range of actual values from production engines. Any departure from this would
clearly be inappropriate, since the production engines clearly would not be appropriately
represented by those values determined during the certification process. We describe in Section
2.1.2 how we specify displacement limits relative to the 1000 cc threshold for Large  SI engines.

       The regulations use consistent and appropriate terminology to characterize maximum
engine power.  Brake power is a separately defined term to clarify which accessory loads are
properly counted toward any measured power value. The regulations in §1054.140 simply
specify that maximum engine power is the maximum value of an engine's measured brake power
over an engine map.

       It would not be appropriate to limit the applicability of §1054.140. This section
establishes definitions and  specifications that dictate how the regulations apply. These
definitions apply universally, but by themselves they require no action.  Other regulatory
provisions, such as the requirements to report maximum engine power in §1054.205, determine
whether action is required to make a demonstration.
                                          2-8

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       We proposed to require manufacturers to report their maximum engine power for engines
with a measured power at or above 15 kW under the specified emission test procedure. This was
intended to require this reporting only as needed to ensure that engines were not exceeding 19
kW based on the proposed approach to defining maximum engine power. We believe we can
more carefully craft this provision, given the 30 kW threshold that applies for engines with total
displacement at or below 1000 cc. As a result, we are modifying the regulation to require
reporting of maximum engine power only where the maximum power for testing is at or above
25 kW for engines with total displacement at  or below 1000 cc, and above  15 kW for larger
engines.


   2.1.4  General concerns

What Commenters Said:

       J. Snell would like to urge EPA to leave small engines exempt from emission controls.
The commenter believes it would raise the price of items like lawn mowers, pressure washers
and go karts which do not have a large enough impact on the environment to justify this increase
especially in rural areas. The commenter stated EPA could at least consider a horsepower or a
cubic centimeter limit. The commenter also believes that 4 cycle motorcycles should also be
exempt for they are a tiny percentage of the machines in the world that put out emissions.

Letters:
Commenter
J. Snell
Document #
0623
Our Response:

       The Clean Air Act directs us to set emission standards for nonroad engines, including all
Small SI engines, such that we achieve that greatest degree of emission control possible after
considering lead time, costs, and other factors. We have made an extensive effort to set
standards that are achievable with costs that are commensurate with the air quality benefit
associated with the reduced emissions. We are not changing the emission standards that apply
for highway motorcycles.

       2.2   Standards and lead time

   2.2.1  NHH standards-level

What Commenters Said:

       EMA noted that they were an active participant in the development of the NPRM for the
next-phase Small SI engine standards.  EMA commented that the net result of that collaborative
process is an NPRM that truly and properly reflects the maximum achievable emission
reductions for Small SI engines and the equipment that they power. EMA commented that the
rulemaking has set forth extremely challenging and dramatic, but nonetheless potentially
                                       2-9

-------
                                                            Chapter 2: Small SI Engines
achievable, emission reduction targets. Indeed, EMA believes the effort that has gone into this
collaborative rulemaking has resulted in the promulgation of an overall framework of
technology-forcing standards and accompanying regulations that are at the very limit of
feasibility and implementability.  As a consequence, EMA commented that the overall
framework needs to be maintained in the final rule, since any potential increased stringency of
the proposed standards or the overall regulatory program would necessarily result in an
infeasible and non-implementable rule.

       EMA commented that exhaust emission control technologies for ground supported Small
SI engines are similar to, but cannot be derived from, other nonroad engine applications or on-
highway applications.  Ground supported Small SI engines and the equipment that they power
operate under significantly different environmental and cost considerations.  Such considerations
pose major obstacles to any wholesale transfer of advanced exhaust emission control systems
and necessarily prevent the fuel and exhaust control technologies used in on-highway (or even
nonroad large spark-ignition) from being applicable to these products.

       MECA commented that it supports EPA harmonizing HC+NOx exhaust emission
standards for Class I and Class II engines used chiefly on nonhandheld equipment with the
CARB standards that were adopted in 2003 and began their implementation in 2007.  MECA
also concurs with the EPA staff analysis and conclusion that the proposed Phase 3 HC+NOx
exhaust emission standards for Class I and Class II engines are technologically feasible and that
catalyst technology can be fully optimized as part of a complete engine/emission control/exhaust
system to help achieve these proposed limits.

       MECA noted that both EPA and CARB test programs have shown that catalysts can be
applied to Class I and Class II engines without increasing safety risks associated with exhaust
component surface temperatures. Integration of catalyst into small engine mufflers utilizes
uncomplicated manufacturing techniques that should allow for the design and validation of
compliant engines within the lead-time provided by the EPA regulations.  The 30 years of
catalyst experience in general and the over 10 years of experience with applying catalysts to
smaller SI on-highway and nonroad engines provide an experience base that has enabled catalyst
technology to continue to be improved.  This small engine experience has provided an increased
understanding of how to optimize the engine/catalyst/exhaust system to work effectively, and
will facilitate application of catalyst technology on Class I and Class II engines to help meet the
proposed standards.

       MECA commented that issues raised by small off-road engine and equipment
manufacturers, such as heat management, packaging, poisoning, and durability, are
straightforward engineering challenges that are well understood and can be readily addressed.
They noted that these types of issues have been raised virtually every time the use of catalyst
technology has been proposed for use on a spark-ignition engine, be it an automobile, heavy
truck, off-road engine over 25 hp (such as a forklift), a motorcycle or moped, or a small engine
used on handheld or non-handheld equipment.  In each case, all of these issues were successfully
addressed for each application through sound engineering principles and design strategies.
MECA believes the situation is no different in the  case of Class I and Class II  nonroad engines.
                                          2-10

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       CARB supported the HC+NOx exhaust emission standard levels proposed by EPA.

       NESCAUM supported EPA's effort to harmonize the federal emissions standards with
those standards already adopted in California. In many respects, the proposed federal standards
are identical to or analogous with California standards.  This approach will make it easier for the
engine and equipment manufacturers to provide 50-state products to the U.S. market.

       Environmental Defense supported EPA's proposal to set more stringent HC and NOx
standards for Class I and II nonhandheld small spark-ignition engines and a new CO standard for
use in marine generator applications. These Phase III standards can be  met by the use of
catalysts, improved engine or fuel delivery systems, or the addition of electronic controls or fuel-
injection systems. According to EPA, several engine families selling nationwide currently
produce engines that meet the proposed Phase III standards.  Therefore, "a number of families
either will not need to do anything or will require only modest reductions" in order to comport
with the new federal standards.  The proposed standards are consistent with those previously
adopted by CARB. Once in place, EPA's proposal should achieve emissions reductions of
approximately 35% below the current federal levels.

       The National Association of Clean Air Agencies (NAC AA) commented that it supports
the federal adoption of exhaust emission standards for small spark-ignition engines consistent
with those adopted by the California Air Resources Board.  Based on the EPA's March 2006
safely study and the Regulatory Impact Analysis for this proposal, as well as public statements
by engine makers, it is evident that additional, more stringent emission  standards are feasible for
small spark-ignition engines, especially commercial equipment, which operates hundreds, if not
thousands of hours a year.  Therefore, NACAA recommended that EPA consider adding another
tier of more rigorous standards for Class I and Class II engines.

       The Pennsylvania Department of Environmental Protection (DEP) commented that it
supports EPA's adoption of a regulation for small spark-ignition engines and equipment that is
consistent with regulations adopted by the CARB.  Consistent with EPA's findings set forth in
the March 2006 Safety Study and the Regulatory Impact Analysis for the proposed rulemaking,
the Pennsylvania DEP recommended that EPA add a tier of more stringent standards for Class I
and Class II engines.

       The Mid-America Regional Council (MARC) Air Quality Forum commented that
because EPA and small engine manufacturers have both asserted that Class I and II engines can
be feasibly designed to meet emissions standards more rigorous than those in the proposed rule,
EPA should consider  incorporating an additional tier of more stringent  standards.

       The Wisconsin Department of Natural Resources (DNR) commented that EPA should
consider adding another tier of more rigorous emission  standards for Class I and Class II spark-
ignition engines as more  stringent emission standards are feasible for these engines, especially
commercial equipment, which operate hundreds, if not thousand of hours a year.

       The New York Department of Environmental Conservation (DEC) noted that EPA is
proposing standards similar to existing California standards for small spark-ignition engines.
                                      2-11

-------
                                                            Chapter 2: Small SI Engines
The proposed standards, to be implemented in 2011 and 2012, will result in a 35% reduction in
combined hydrocarbon and NOx emissions for engines in non-handheld equipment.  The New
York DEC noted that it is well aware of industry opposition to these proposed regulations, and
supports expedient adoption of the standards as proposed.  They also believe that further
emissions reductions are needed, and will be feasible in the future.

       Environmental Control Corporation (EVCC), a developer of catalytic mufflers for small
spark-ignition engines (both two-stroke and four-stroke) commented that their technology has
proven HC+NOx reductions of up to 98.9% on a two-stroke engine and 90% on a four-stroke
engine while also significantly reducing CO. They noted that they are in full support of the
current regulations, but they encouraged EPA to set even more stringent standards in the near
future.

       EVCC provided information on their catalytic mufflers. First, EVCC noted that its
patented catalytic muffler is linearly designed and can be modified to fit virtually any spark-
ignition engine. EVCC has successfully completed emissions testing for a variety of engine
applications (both two-stroke and four-stroke), including lawn mowers, snowmobiles, out-board
motors, water-pumps, and more. The company just recently completed a durability test on a 163
cc four-stroke engine (in compliance with EPA and CARB certification parameters), and is
currently in the process of completing durability testing for a two-stroke application. Second,
EVCC noted that its catalytic mufflers are both cost-effective and compact in size. The unique
airflow design of these mufflers allows them to achieve unprecedented emission reductions while
using minimal materials and space.  In its most recent 163cc four-stroke test (lawn-mower
application), EVCC  noted that its catalytic muffler was smaller than that of the OEM and did not
require external air, baffles, perforated pipes or sound chambers.

       EVCC noted that it is very concerned that EPA plans to continue regulating the emissions
of small non-road engines by grouping HC and NOx together as one value.  EVCC commented
that manufacturers of four-stroke spark-ignition engines (i.e. lawn and garden equipment) are
able meet both current and future emission regulations for CO and HC+NOx by engine
modifications and minor carburetor calibrations (by running the engine on a lean fuel mixture,
for example). While EVCC believes these engine modifications and minor carburetor
calibrations will reduce CO and HC, in most instances there will be a concomitant increase in
NOx. This is highly undesirable, as increases in NOx will  have a drastic impact on both human
health and the environment.

       EVCC noted that the emission regulations in which HC and NOx are combined together
for a total emission certification value permits an increase in NOx levels as a trade-off to
reducing HC. This loop-hole is completely unnecessary in the small engine sector, as three-way
catalytic converters are fully capable of reducing all three emission  values simultaneously.  In
addition, NOx is one of the most harmful by-products of fossil-fuel  combustion, and
manufacturers should by no-means be permitted to increase NOx levels needlessly.  All three of
the emission values in question are individually regulated in the automotive sector, and it is now
time to carry this practice to the small engine industry.
                                         2-12

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       Frank Smith, a retired Chemistry Professor from the Memorial University of
Newfoundland, commented that in changing the emission regulations applicable to small spark-
ignition engines, the requirements should be set such that the technology exists to achieve the
levels selected. In that respect, the ability of catalytic mufflers developed by Environmental
Control Corporation (EVCC) should be seriously considered.  These devices have been tested
both at Environment Canada's test facility and at Carnot Emission Services establishment.  In
January 2007, at the latter facility, results over an extended test of 125 hours, achieved for a four-
stroke engine (Honda GX-160), included NOx emissions of less than 0.5 g/kW-hr, HC emissions
of less than 5 g/kW-hr, and CO emissions of less than 300 g/kW-hr of operation, without
reduction of performance.

       In the same tests the combined HC+NOx emissions were also less than 5 g/kW-hr.
Consequently, for all pollutants considered the emissions were well below those of current
CARB regulations and also those proposed by EPA. In June 2006, tests of EVCC's catalytic
muffler fitted to a two-stroke 185 cc Class I nonhandheld engine at Environment Canada's test
facility gave similarly  low emissions of all three pollutants. A  six-mode test (in accordance with
US EPA and CARB regulations) using 20 LPM of air injection in the catalyst  was conducted by
officials at Environment Canada. Of particular interest is the 99 % reduction achieved in
hydrocarbon emissions from around 250 g/kW-hr with the original muffler to  2.5 g/kW-hr with
the catalytic muffler.  This data strongly suggests that control of all three pollutants separately is
feasible at levels below those currently proposed.

Letters:
Commenter
EMA
MECA
CARB
NESCAUM
Environmental Defense
NACAA
Pennsylvania DEP
MARC AQ Forum
Wisconsin DNR
NY DEC
EVCC
EVCC
F. Smith
Document #
0691
0668
0682
0641
0648
0651
0676
0696
0663
0659
0608
0654
0694
Our Response:

        Section 213(a)(3) of the Clean Air Act specifies the criteria EPA must use in establishing
new emission standards. Under the statute, EPA is directed to set emission standards that
achieve the greatest degree of emission reduction achievable through the application of
technology which EPA determines will be available for the engines or vehicles to which such
standards apply, giving appropriate consideration to the cost of applying such technology within
the period of time available to manufacturers and to noise, energy, and safety factors associated
                                       2-13

-------
                                                           Chapter 2: Small SI Engines
with the application of such technology.  In addition, and specific to this rulemaking only, under
section 205 of PL 109-54, EPA, in coordination with other appropriate federal agencies, was
required to complete and publish a technical study analyzing the potential safety issues
associated with the proposed standards, including the risk of fire and burn to consumers in use.
The technical study was to be completed and published before the publication of the notice of
proposed rulemaking.  Given these criteria and requirements and our assessment of the
comments, EPA continues to believe that the proposed Phase 3 standards are the appropriate
standards for nonhandheld engines for the years in which they were proposed for
implementation.  (See Section 2.2.2 for further discussion of the comments on lead time for the
Phase 3 nonhandheld engine standards.)

       The Phase 3 standards for nonhandheld engines are technology forcing and are expected
to result in the use of modified calibrations, engine improvements, catalysts, and fuel injection to
achieve the required emission reductions. The mix of technologies will vary depending on the
engine design. As detailed in Chapter 4 of the Final RIA, EPA developed several aftertreatment
and fuel-injection systems to demonstrate that the Phase 3 standards could be met.  In addition,
EPA assessed the impacts of the new standards on cost  as detailed in Chapter 6 of the Final RIA.
Finally, EPA expended considerable effort in analyzing the potential safety impacts of engines
designed to meet the proposed Phase 3 standards to comply with the requirements of section 205
of PL 109-54.  ("EPA Technical Study on the Safety of Emission Controls for Nonroad Spark-
Ignition Engines  <  50 Horsepower," EPA420-R-06-006, March 2006, docket item  EPA-HQ-
OAR-2004-0008-0333.)  Taking all of this information  into consideration, EPA believes the
Phase 3 standards meet the criteria specified in the Clean Air Act for the time frame in which the
standards are to be  implemented.

       EPA received several comments that we should set more stringent standards than those
being adopted today, but we do not concur. All of these commenters except one (as noted
below) provided no supporting analysis or data on any of the relevant statutory factors in  support
of their request. One commenter did submit emission data showing very low emission  levels for
a Class I engine.  In fact, EPA itself generated emissions data for the proposal showing low
levels as well.  For example, in Class I, EPA tested a number of engines that had HC+NOx
emission levels as low as 3.9 g/kW-hr at  low hours and were projected to be as low as 5.7 g/kW-
hr HC+NOx at the end of their regulatory useful life. Likewise, for Class II engines, EPA tested
engines with HC+NOx emission levels as low as 1.8 g/kW-hr at low hours and projected to be as
low as 2.3 g/kW-hr at the end of their regulatory useful  life. (All of the emissions data  generated
by EPA and summarized here, is presented in Chapter 4 of the Final RIA.) However, as noted
above, the requirements for establishing new emission standards are dependent on  more than
demonstrating certain emission levels. So while EPA had its own emission data showing lower
emission levels are achievable, EPA determined for the proposal that under section 213(a)(3) of
the CAA, the Class I standard of 10.0 g/kW-hr HC+NOx and the Class II standard of 8.0 g/kW-
hr HC+NOx were the appropriate emission standards. Section V.G. of the proposed preamble
and Chapter 4 of the Draft RIA for the proposed rule laid out EPA's assessment of the proposed
standards in the context of all of the CAA criteria. As noted above, none of the commenters
asking for more stringent standards addressed these other factors.
                                         2-14

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       In addition, in Chapter 11 of the draft RIA for the proposed rule, for both Class I and
Class II engines, EPA considered the appropriateness of more stringent emission standards (i.e.,
8.0 g/kW-hr HC+NOx for Class I engines and 4.0 g/kW-hr HC+NOx for Class II engines) based
on the CAA criteria. EPA noted that while more stringent standards may be feasible, EPA
concluded that more leadtime would be required for such standards. This was based on the fact
that more stringent standards would require more fundamental and significant changes in engine
design in both Classes I and II. For Class I engines, we projected that manufacturers would
likely need to convert their side valve engine designs (which represent two-thirds of sales
currently) to overhead valve designs along with using a more efficient catalyst and addressing
emissions deterioration.  For Class II engines, we projected that manufacturers would need to
convert all engines to fuel injection, upgrade their residential  engine designs to improve
emissions deterioration characteristics (e.g., those with a 250  hour useful life), and use a more
efficient catalyst.  Such redesigns would involve significantly more development work for all
manufacturers (and likely more cost) compared to the changes projected for the Phase 3
standards adopted in this rule.  This could not happen as soon as the 2011 and 2012 timeframe
being adopted for the Phase 3 standards, and would result in a delay in achieving air quality
benefits.

       In further response to comments that EPA should have promulgated more stringent
standards than we proposed, it is important to note that setting more stringent standards for either
Class I or Class II or both, would require a more robust analytical record. Those suggesting that
more stringent standards should be established now (either in a single or two phases) did not
provide input on factors such as cost, lead time, and the other CAA criteria for public
consideration. EPA could have pursued such further analysis  at this time, but it would likely
have required an additional notice/comment step which would further delay this action. The
states with air quality problems would benefit more from the  earlier reductions due to the
standards being adopted in this final rule rather than waiting for further reductions.  Therefore,
EPA concluded that the proposed Phase 3 standards (which we are adopting with today's rule)
are the appropriate standards under the CAA.

       With regard to the comment on having separate HC and NOx standards instead of a
combined HC+NOx standard, EPA is retaining the standards  based on a combined HC+NOx
level as is the case with the current Phase 2 standards.  EPA believes a combined standard offers
flexibility to manufacturers in designing technology to comply with the standards, especially
since not all engine designs respond identically to the same control techniques in catalyst design
where it is generally easier to reduce HC  emissions compared to NOx emissions. While it is true
that mathematically a combined standard could result in a decrease in overall HC+NOx
emissions with a rise in NOx (or HC), EPA does not expect that would happen to any significant
degree as manufacturers redesign engines to comply with the Phase 3 standards. This is
especially true if a manufacturer uses a catalyst to comply with the Phase 3 standards because a
catalyst would be expected to reduce both HC and NOx, although not at the same rate.  This also
would likely be true for those engines that might rely on engine modifications to comply with the
Phase 3 standards.  The latter conclusion  is based on a comparison  of three 2008 model year
Class II engine families that have certification levels below the Phase 3 standards compared to
similarly sized Phase 2 engines from the same manufacturer.  For this comparison, all of the
engines were OHV engines and were certified without catalysts.  In all three cases, the overall
                                       2-15

-------
                                                            Chapter 2: Small SI Engines
HC+NOx emissions for the 2008 model year engines were lower by up to 27 percent, and the
individual HC levels and NOx levels were also lower. While most of the decrease in HC+NOx
emissions was from decreases in HC emissions, the NOx emissions decreased in all cases as
well.  Therefore, while a combined HC+NOx standard has the potential to lead to higher levels
of one pollutant relative to that pollutants level under the Phase 2 requirements, EPA believes
that the Phase 3 HC+NOx standards should also result in lower HC emissions and lower NOx
emissions for the fleet.

      Additionally, it should be noted that the Phase 3 standards being adopted by EPA are the
same as those adopted by CARB, which are based on the combined HC+NOx level.  Having the
same requirements as CARB helps manufacturers by allowing them to certify the same designs
with both agencies, which might not be possible if EPA were to adopt separate HC and NOx
standards. In addition, it is important to note that the further control of HC and NOx emissions
from these engines is being driven by the  need to reduce ambient ozone concentrations.  Both HC
and NOx contribute to the formation of tropospheric ozone, so a slight mix in the relative
reductions among engine designs does not deter achieving the ozone air quality improvement
goal which is a key basis for this action.

   2.2.2   NHH standards-lead time

What Commenters Said:

      OPEI  commented that under Section 213 of the Clean Air Act, EPA must make sure that
adequate lead-time is provided to allow all equipment manufacturers, as well as their separate
engine and exhaust system suppliers, to develop and test the new materials, technologies, and
safeguards, including low-permeation fuel tanks and catalyzed-exhaust systems, and to ensure
operational risks are mitigated under all the expected operating conditions including off-nominal
conditions.  As long as they  are provided with adequate lead-time and the other related
flexibilities, OPEI noted that its members will be able to design their products to utilize  catalyzed
exhaust systems that would be required to meet the proposed EPA Phase 3 standards. OPEI
provided a number of reasons why  any acceleration of the proposed Phase 3 effective dates or
dilution of the proposed lead-time flexibilities would undermine and  potentially jeopardize the
manufacturers' ability to build and test products to ensure they would not have any incremental
risks.

      First, the inclusion of aftertreatment systems into an equipment manufacturer's exhaust
system will require a much broader set of changes than just packaging the catalyst into an
existing muffler, as implied by the EPA in the proposed Preamble discussion. Second, non-
integrated equipment manufacturers must work closely with a variety of suppliers to design and
install all the different components into the final product. Third, it will take an extraordinary
amount of time and effort to develop a single piece of equipment with an effective and safe
catalyzed muffler that has been thoroughly evaluated - under both nominal  and off-nominal
conditions.  Fourth, given the volumes and diversity of these Class II exhaust systems, and
limited resources, OEMs are concerned that there will be several bottlenecks (with all their
suppliers, independent test labs, and certification officials) that will further delay the production
and certification process.  Fifth, most of the non-integrated equipment manufacturers expect they
                                         2-16

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


will be forced to offer only limited CARB, Tier Ill-compliant products for the California market.
The national market will require the industry to address many of the most challenging muffler
applications and configuration that will not be offered in California.

       MECA commented that EPA's proposal to implement the exhaust emission standards for
Class I and Class II engines in the 2011-2012 timeframe provides more than adequate lead time
for engine and equipment manufacturers.  MECA urged EPA not to push out these
implementation dates beyond the proposed 2011 and 2012 dates. MECA believes that an even
faster implementation schedule for these exhaust emission standards is feasible given the
implementation schedule adopted by California.

       CARB commented that they believe the timing of the new standards should be
implemented sooner.  Small spark-ignition engine manufacturers have already been preparing to
meet the California standards which are the same as EPA's proposed standards.  The
manufacturers already have the technological ability to meet the standards.  Under EPA's current
proposal, manufacturers of Class II engines will have three years from the time California
standards have gone into effect and Class I engine manufacturers have five years. It should not
take these manufacturers three to five additional years to meet these standards nationwide,
particularly since EPA also allows for credit generation which gives manufacturers additional
flexibility.  CARB suggested that an alternative would be for EPA to give the manufacturers that
do not currently sell their products in California extra time to meet the standards.

       CARB recommended that EPA modify its proposed HC+NOx phase-in schedule for large
spark-ignition engines < 1 L to harmonize with the California small off-road engine exhaust
emission standards for Class II engines, 8  g/kW-hr at the 2008 model year.  This would provide
for significant emission reductions from this category. Furthermore, a harmonized  program
would help reduce the problem of higher emission 49-state large spark-ignition engines traveling
into the California fleet.

       NACAA questioned the need for the substantial additional lead time that EPA has
proposed beyond the implementation dates enacted by California - five years (until 2012) for
Class I engines and three years (until 2011) for Class II engines. They believe an accelerated
federal schedule is technically feasible and recommended that EPA give consideration to more
rapid implementation.

       Pennsylvania DEP commented that they are concerned about the need for the substantial
additional lead-time of three to five years proposed by EPA and strongly suggests more rapid
implementation to afford greater protection of human health and the environment.

       The MARC AQ Forum noted that the proposal sets implementation deadlines 2012 for
Class I engines and 2011 for Class II engines.  They urged EPA to accelerate its implementation
timeline.

       NESCAUM commented that they oppose the protracted timelines for compliance with
the standards, proposed for manufacturers of small land-based SI engines and equipment. The
analogous California exhaust emissions standards are fully phased-in between 2005 and 2008.  In
                                      2-17

-------
                                                            Chapter 2: Small SI Engines
contrast, the proposed phase-in period for the proposed federal standards does not even begin
until 2010 and, with special provisions afforded to small to medium volume manufacturers, full
compliance is delayed until as late as 2014.  NESCAUM does not believe there are valid reasons
for delaying the incorporation of Phase 3 engines into various types of equipment nationally
when manufacturers will already be supplying the California market with lower-emitting Phase
III engines and equipment years earlier. This approach for protracted delays is inconsistent with
the approach taken in the same rulemaking for SD/I marine engines where EPA chose to closely
track effective dates for the California standards:  "EPA is proposing that the Federal SD/I
standards take effect for the 2009 model year, one year after the same standards apply in
California. We believe a requirement to extend the California standards nationwide after a one-
year delay allows manufacturers adequate time to incorporate catalysts across the product lines
as they are doing in California. Once the technology is developed for use in California, it would
be available for use nationwide soon thereafter." NESCAUM requested that the  exhaust
emission standards for land-based small SI engines be fully implemented, beginning with the
2009 model year, consistent with the proposed compliance dates for SD/I engine standards.

       Wisconsin DNR commented that EPA should accelerate the implementation dates of the
exhaust emission standards for Class I and Class II small spark-ignition engines consistent with
those adopted by CARB.

       NJ DEP noted that the CARB standards for exhaust emissions are fully phased-in
between 2005 and 2008, whereas the proposed phase-in dates for the corresponding federal
standards do not begin until 2010. Of most  concern, NJ DEP highlighted the  special provisions
for small and medium manufacturers which may delay full compliance until 2014. In light of the
fact that manufacturers will already  be providing cleaner engines and equipment to California
and that technology issues will not be a factor, these cleaner engines and equipment should be
required to be made  available sooner nationwide.

       Environmental Defense commented  that they object to the much delayed  implementation
dates for these important standards.  EPA's  proposed engine  exhaust limits for nonhandheld
Class I and Class II engines do not go into effect until model year 2012 and 2011 respectively,
while California's comparable standards take effect in 2008 and 2009. In justifying the proposed
near-term implementation dates for  SD/I and OB/PWC standards, EPA relies on  the fact that
many manufacturers currently design and sell cleaner engines capable of achieving the proposed
standards.  Environmental Defense agrees with EPA that the availability of cleaner technology
weighs in favor of near-term implementation dates since the cost and burden to manufacturers in
meeting a more stringent standard is low in  this instance. For this reason, they fail to understand
why EPA has reached such a different conclusion in setting the implementation dates for the
small SI engine exhaust standards. Technological advances in the SI small market, just like
those in the SI marine sector, have resulted in the wide-spread availability of cleaner engines
capable of achieving greater emissions reductions. In addition, EPA's proposal provides small
SI engine manufacturers with substantial flexibility by allowing them to choose from a number
of aftertreatment technologies in order meet the new standards.  The breadth of available
technologies capable of reducing small engine emissions to the proposed Phase 3 levels weighs
in favor of shorter implementation dates, not longer.  EPA's failure to explain adequately its
                                          2-18

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


basis for delaying the implementation dates by some 4-5 years is arbitrary and capricious and
contrary to law.

       Mr. Dan Holland commented that he opposes delaying immediate and full
implementation of the proposed rules. He believes that requiring water craft to comply with the
new standards in 2009 but not require "land-based" small engines to comply until 2011 is
arbitrary and capricious. There is no need — or excuse — to wait until 2011 to implement the new
regulations with respect to all new small engines. He believes that proven technology is
commercially available now that can make all new small engines compliant with the more
stringent, proposed regulations that the EPA is authorized to promulgate. Delaying
implementation of the new standards with respect to new land-based small engines until 2011
can only be interpreted  as "political" bias in favor of the Senator from Missouri that has long
opposed emissions  regulation and emissions reduction  on the specious grounds that the addition
of catalytic converters etc. cause small engines to run "hot" and/or cause external fires, both of
which studies by the EPA and others have disproved.  Substantial  emissions reductions can
readily and easily be achieved by adding existing, proven,  inexpensive technologies to new
engines, and this wait-until-2011 "free pass" for land-based engines is  simply unacceptable, and
it is legally indefensible in light of Congress's mandate in section 428(b) of the 2004
Consolidated Appropriations Act and existing Section 213(a)(3) of the Clean Air Act which
contemplates an immediate business response during a 12-month business-design cycle, not a
business cycle "four years from now."

Letters:
Commenter
OPEI
MECA
CARS
NACAA
Pennsylvania DEP
MARC AQ Forum
NESCAUM
Wisconsin DNR
NJDEP
Environmental Defense
D. Holland
Document #
0675
0668
0682
0651
0676
0696
0641
0663
0710
0648
0595
Our Response:

       EPA continues to believe that the proposed Phase 3 standards are the appropriate
standards for nonhandheld engines for the years in which they were proposed. (See Section 2.2.1
for further discussion of the comments on the level of the standard for the Phase 3 nonhandheld
engine standards.) As noted above, EPA believes the new Phase 3  standards for nonhandheld
engines are technology forcing and are expected to result in the use of technologies including
engine improvements, catalysts, and fuel injection to achieve the required emission reductions.
Engine manufacturers will need substantial time to redesign all of their engine families to
comply with the new standards.
                                       2-19

-------
                                                            Chapter 2: Small SI Engines
       A look at the current certification data for the 2008 model year provides useful
information to gauge the level of effort required by engine manufacturers to comply with the
new standards. (EPA's certification data can be found on the internet at the following site:
http://www.epa.gov/otaq/certdata.htm#smallsi) There are a total of 87 manufacturers with
nonhandheld engine families certified with EPA. For the following discussion, we have focused
on the 15 manufacturers that historically have been selling in the small engine market. (The
remaining 62 manufacturers, primarily from China, are recent participants in the small engine
market and generally have only 1 to 5 engine families certified with EPA with relatively low
sales volumes.)  For these 15 manufacturers of nonhandheld engines, there are currently 66
engine families certified in Class I and 121 engine families certified in Class II. (These numbers
exclude engines used exclusively in snowblowers which do not have to comply with the
HC+NOx standards). While some of these engine families have emission levels below the Phase
3 standards, manufacturers will need to redesign the bulk of the designs to meet the Phase 3
standards. For these 15 manufacturers, EPA estimates that 53 of the Class I engines and 83 of
the Class II engines will have to be redesigned to meet the Phase 3 standards. ("Analysis of
2008 Small SI Nonhandheld Engine Certification Data," EPA memo from Phil Carlson to EPA
Docket OAR-2005-0008, August 28, 2008, docket item EPA-HQ-OAR-2004-0008-	.)

       For the six manufacturers with the highest numbers of nonhandheld engine families (i.e.,
Briggs and Stratton, Fuji Heavy Industries, Honda Motor Company, Kawasaki Heavy Industries,
Kohler Company, and Tecumseh Products Company), EPA estimates that they will need to
redesign over 19 engines families, on average, to comply with the new Phase 3 standards.  (The
range in the number of engine families needing to be redesigned for these manufacturers is from
12 to 35 engine families.) Given that we are finalizing the Phase 3 standards in late-2008,
manufacturers will have only 2 years before the Class II engine standards take effect and 3 years
before the Class I engine standards take effect.  As described below, we believe that engine
redesign will require a significant level of effort for engine manufacturers.  Given the level effort
needed and the number  of engine families needing to be redesigned, EPA does not believe it
would be possible to reduce the lead time for the new standards.

       The Phase 3 emission standards for Class I engines are expected to result in engine
improvements and the use of catalysts. Catalysts have been implemented on few of these
engines to date and therefore the expected widespread use will require significant technology
development and investment from engine manufacturers. In addition to the catalyst brick
formulation, other technology requirements include muffler design for desired pollutant
conversion (which they will want to optimize for minimum precious metal loading to reduce
costs),  consideration of regulatory useful life emission requirements, addressing cooling
requirements related to muffler skin temperature and exhaust temperature, and testing of the
engines in real-world applications.  While EPA believes the technological challenges can be met
by manufacturers, each  of these steps will take considerable resources and time to address for
each of their engine families. As noted in Chapter 6 of the Final RIA (as well as Chapter 6 of the
Draft RIA for the proposal), EPA estimates that engine modifications will take 4 months of
design work and 6 months of development work for each engine design. In addition, EPA
estimates that applying catalysts will take 2 months of design work and 5 months of development
work for each engine design.
                                         2-20

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
       Likewise, the Phase 3 emission standards for Class II engines are expected to result in
both engine redesign and the application of catalysts on many engines. For those Class II
engines using catalysts, engine manufacturers will need to address the same issues noted above
for Class I engines.  In addition, they will need to communicate closely with their Class II engine
users (i.e., equipment/vehicle manufacturers) since most Class II engines are sold without an
exhaust system. Due to the wide number of exhaust systems used on these engines, equipment
manufacturers will either have to modify the existing equipment design to utilize a manufacturer
provided muffler,  or they will have to develop their own muffler using the engine manufacturer's
provided catalyst brick specifications and then do the certification of that engine. Although EPA
believes these issues can be addressed, all of these efforts will take time.  As noted in Chapter 6
of the Final RIA (as well as Chapter 6 of the Draft RIA for the proposal), EPA estimates that
engine modifications will take 4 months of design work and 6 months of development work for
each engine design.  In addition, EPA estimates that applying catalysts will take 2 months of
design work and 5 months  of development work for each engine design.

       Finally, under the Phase 3 program, EPA is requiring the certification of engines using
new test procedures under part 1065 by the 2013 model year.  These new procedures require
engine manufacturers to implement changes to their current test setup in order to incorporate new
test cell operation procedures and new emissions calculations.  If a manufacturer is going to
spend the resources to certify a new engine, they will likely want to do it only once so as to use
the carryover data option in certification for a number of years. Therefore, it is likely
manufacturers will want to certify in 2011 or 2012 with the new procedures.  The effort it will
take to convert manufacturer's facilities depends on the age of the manufacturer's current testing
equipment and will add to the time and effort required to comply with the new Phase 3 standards.

       Given the number of engine families that need to be redesigned, the types of
technological issues that will need to be addressed for each engine family, and the new test
procedure requirements to which manufacturers will need to convert, EPA believes the 2012
requirement for Class I and 2011 requirement for Class II are the appropriate leadtime for the
new standards.

       With regard to the comments that EPA should move up the implementation dates because
California's Tier 3 standards are already in effect, an analysis of the 2008 model year
certification data from CARB for the six engine manufacturers with the highest number of
nonhandheld engine families (as noted above) provides some useful information. While
CARS's  Tier 3 standard for Class I engines took effect in 2007, only 9 out of 29 engine families
are certified by these manufacturers at or below the 10.0 g/kW-hr HC+NOx standard. For Class
II engines, where the Tier 3 standard takes effect in 2008, only 19 out of 60 engine families are
certified by these manufacturers at or below the 8.0 g/kW-hr HC+NOx standard. While these
manufacturers have  redesigned some of their engines to meet CARB's Tier 3 standards, they are
using emission credits to certify the remaining engines. Therefore, even though CARB's Tier 3
standards are already in effect, manufacturers have a significant amount of work to finish
certifying their engines for California.  We continue to believe the Phase 3 implementation dates
of 2012 for Class I and 2011 for Class II provide the appropriate leadtime for manufacturers to
redesign their engines to comply with EPA's Phase 3 standards.
                                       2-21

-------
                                                             Chapter 2: Small SI Engines
   2.2.3  CO standard for marine generators

What Commenters Said:

       MECA supported EPA's proposal to establish a Phase 3 CO standard of 5 g/kW-hr for
marine generators. They noted that existing commercial applications of catalyst-equipped
marine generators provide strong evidence that EPA's proposed low CO standard for marine
generators is technically feasible.

       EMA commented that dedicated marine generator engines that are permanently installed
into vessels (such that they can take advantage of features such as water cooling, vessel DC
electrical systems, electronic closed loop feedback fuel control  systems, and three way catalyst
aftertreatment systems) may be able to comply with the proposed CO emission standard (5
g/kW-hr). However, many auxiliary marine engines are either not dedicated to the vessel or are
not integrated in a manner consistent with the technology that would be required in order to
achieve the proposed CO emission level. Accordingly, EMA commented that the final
regulation must clarify that the proposed CO standard is  only applicable to the fully-integrated
marine generator engines described in the NPRM.

Letters:
Commenter
MECA
EMA
Document #
0668
0691
Our Response:

       We agree with EMA that marine auxiliary engines that are not generators should not be
subject to the more stringent CO standard. This was reflected in the proposed rule. We do not
believe it is appropriate to specify some degree of integration for marine generators before the
more stringent standards apply. This information is generally not available to engine
manufacturers at the point of certification and it would be difficult to specify an objective
measure that would make this enforceable.  The final regulation is unchanged, requiring all
marine generators to meet the 5 g/kW-hr CO standard.
   2.2.4  Useful life

What Commenters Said:

       EMA noted that there are a wide variety of usage patterns for the engines and equipment
governed by the proposed regulation.  EMA commented that the proposed maximum time span
of 5 years for the emission durability period is acceptable provided that the final rule clearly
states that the durability period is the lesser of either hours or years.

       OPEI commented that the last line of the useful life definition, "If an engine has no hour
meter, the specified number of hours does not limit the period during which an engine is required
                                          2-22

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


to comply with emission standards..." should be deleted.  OPEI supported a 5 year time limit on
useful life. This means the specified number of hours or 5 years (whichever comes first).

       CARB believes that it is appropriate to limit the useful life period to five years or the
specified number of operating hours, whichever comes first.  Limiting the useful life period
would be favorable to both industry and regulatory agencies. It would allow manufacturers to
limit warranty coverage by a time period rather than operating hours which can be difficult to
determine. For regulatory agencies, it provides more flexibility in limiting the length of time
credits may be used.

Letters:
Commenter
OPEI
CARB
EMA
Document #
0675
0682
0691
Our Response:

       All the commenters supported the provision that would define the useful life period as
five years or a specified number of hours of engine operation, whichever occurs first. We are
therefore adopting this provision, including a clear statement that either one of these two age
indicators would be sufficient to establish the end of the useful life.

       The last sentence in the definition of useful life clarifies how to apply the definition with
respect to hours of operation if the engine has no hour meter. Leaving out this specification
would leave this ambiguous and would require that we make a judgment in guidance to the
industry. We believe it is therefore fitting to include this clarification in the regulation.
Moreover, we believe the proposed provision establishes a very reasonable approach, such that
the hours-based limit on useful life is meaningful only if the extent of operation can be
established without the missing hour meter. For example, if an engine is certified based on a
useful life of 250 hours, we would intend to be able to do in-use testing on such engines that
have been in service in consumer use in a riding lawnmower throughout the five-year period
representing the useful life, unless it  is clear that the engine has operated for more than 250 hours
(if, for example,  the lawnmower has  been used in commercial service long enough to
demonstrate that it has operated longer than 250 hours.

       See Section 2.4.2 for additional discussion related to selection of useful life values for
certification.
   2.2.5  Crankcase controls

What Commenters Said:

       EMA commented on §1054.115(a)(l)(i) "What other requirements apply?" EMA noted
that the section states that engines must be manufactured in a way to allow crankcase emissions
                                       2-23

-------
                                                            Chapter 2: Small SI Engines
to be routed into the emission measurement sampling system.  EMA commented that it is
impractical for an engine manufacturer to meet this requirement given the diversity of exhaust
emission measurement equipment. This requirement should be revised to replace the testing
requirement with an engineering judgment/test requirement as described in §90.109.
       EMA commented on §1054.115(a)(l)(ii) "What other requirements apply?" EMA
commented that because the exhaust emission measurements utilized to determine the
deterioration factor must include the crankcase emissions pursuant to §1054.115(a)(l), this
section's requirement to include deterioration in crankcase emissions in the determination of
deterioration factors is illogical. However, if the requirement in §1054.115(a)(l)(i) is revised to
incorporate the language from §90.109 (as suggested above), then this section would not require
any additional revisions.

Letters:
Commenter
EMA
Document #
0691
Our Response:

       It is unclear on what basis manufacturers should be using good engineering judgment
regarding an engine's ability to meet emission standards considering vented crankcase emissions
if those emissions cannot be measured. We believe emission-measurement systems should be
capable of measuring crankcase emissions where a manufacturer would want to make a separate
measurement of crankcase emissions for adding to the conventional emission results.  This is
especially true in the case of dilute testing. However, regardless of the method used to measure
emissions, we allow for a test setup in which the engine is modified such that the crankcase
emissions are vented into the exhaust before sampling.  This should be readily achievable for any
system that can make a valid exhaust emission measurement with Small SI engines.

       We find it entirely logical to consider measured  changes in crankcase emissions over an
engine's  service life in the determination of deterioration factors.  The effect of changing
crankcase emissions could be quantified separately (if crankcase emissions are measured
separately) or the manufacturer could use a single deterioration factor that combines the
crankcase and conventional exhaust emissions at all points.
   2.2.6  Safety

What Commenters Said:

       OPEI commented that the ability of manufacturers to produce and accurately evaluate the
potential hazards of any new technology, including catalyzed mufflers, depends on EPA
providing adequate lead-time and all the related proposed flexibilities. There will be substantial
development work and costs associated with the development and installation of heat-shielding,
and other safeguards to ensure that catalyzed exhaust systems (at the efficiency levels discussed
in the proposal) do not pose any increased risks or hazards.  As EPA's administrative record
demonstrates, any more stringent exhaust standards, or more accelerated effective dates (than
                                          2-24

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


those EPA has proposed) would not meet that statutory requirements in Section 213 of the Clean
Air Act, including feasibility, safety, lead-time, and costs.

       OPEI continued that the industry is becoming much better informed on how to build and
evaluate catalyzed products primarily as a result of the continued research and development
work internally conducted by manufacturers. Manufacturers are also becoming better informed
on the exhaust gas temperatures and the muffler surface temperatures (where grass clippings and
other debris could potentially ignite) through the comprehensive study that was released this
spring by the National Institute of Standards and Technology (NIST).  Manufacturers will
become more knowledgeable through the current related study on the heat-related challenges of
catalyzed mufflers that has been conducted by the  SP Technical Research Institute of Sweden for
the International Consortium for Fire, Safety, Health and the Environment (ICFSHE).  The OPEI
Education and Research Foundation funded both the NIST and SP studies in order to promote
our understanding of the heat-related challenges with both catalyzed and non-catalyzed mufflers
so that manufacturers can build even safer products that respond to these challenges. From
OPEFs perspective, the NIST and SP studies are solely intended to inform manufacturers as they
develop new ANSI standards for "Mitigation Of Heat-Related Hazards From Mufflers On All
Ground-Supported Equipment". For example, the SP study will ultimately help manufacturers
develop procedures (to mimic in their laboratories) the most challenging and complex  off-
nominal conditions (such as single spark-plug misfire or malfunction on a twin cylinder engine).
The EPA and SP studies (as well as manufacturers' current  experience in California) also
generally confirm the enormous challenges and lead-time needed to design, build and internally
evaluate all their diverse catalyzed equipment to ensure these products will be durable, emission-
compliant and minimize the risks under the complex, nominal and off-nominal operating
conditions.

       OPEI commented that while there is still much work that remains to be done, OPEI
members are working with all their different suppliers to develop catalyzed products and to  draft
and finalize the new, helpful ANSI standards. The smaller OEMs (with the least internal staff
and resources) will benefit the most from the information supplied in the ANSI process. ANSI
standards development is a voluntary consensus-based process. The actual time to develop a
standard varies based upon the iterative notice and comment procedures. The ANSI committee
is currently on track to develop the final ANSI standards before the Phase 3-exhaust standards
are proposed to become effective in 2011.

       OPEI stated that EPA's proposed exhaust standards  combined with the proposed
effective dates should allow time for the entire industry to build catalyzed products that do not
increase any risks.  The proposed effective dates should allow time for the new ANSI standards
to be finalized and issued to the public before the Phase 3 exhaust standards begin to apply.

       MECA concurred with the conclusions reached by EPA staff that the application of
catalysts to nonroad equipment or marine generators with either Class I or Class II spark-ignited
engines can be accomplished using available engineering exhaust system design principles in a
manner that does not increase the safety risk relative to today's uncontrolled equipment. In
particular, the EPA safety study on non-handheld equipment equipped with catalyzed mufflers
represents the most thorough safety study completed to date on this class of spark-ignited
                                      2-25

-------
                                                            Chapter 2: Small SI Engines
engines.  The results of this EPA study showed that properly designed catalyzed mufflers pose
no incremental increase in safety risk (and in many cases even lower muffler surface
temperatures) relative to currently available non-handheld equipment sold without catalysts.
Catalysts have also been used voluntarily on lawn mowers in certain European markets since the
late 1990s and on a range of handheld equipment with no significant, reported safety issues,
providing additional support that catalysts can be integrated into the mufflers of Class I and
Class II engines in a safe manner.

       During testimony at the public hearing, Mr. Richter of Heraeus noted that they have
supplied catalysts for European "green" products in response to certain European states that have
requirements for catalyst-based systems on some of their walk behind equipment.  Mr. Richter
noted that nearly a million walk behind mowers have been produced with a catalyst in Europe in
response to these requirements. When asked if he was aware of any problems, performance
issues or anything related to use of the catalysts, Mr. Richer responded that there were none
whatsoever.

Letters:
Commenter
OPEI
MECA
Heraeus (hearing)
Document #
0675
0668
0642
Our Response:

       Section 213 of the Clean Air Act directs us to consider the potential impacts on safety,
noise, and energy when establishing the feasibility of emission standards for nonroad engines.
Furthermore, section 205 of EPA's 2006 Appropriations Act requires us to assess potential safety
issues, including the risk of fire and burn to consumers in use, associated with the new emission
standards for nonroad spark-ignition engines below 50 horsepower.  We expect that the new
exhaust and evaporative emission standards will have no adverse affect on safety.

       In response to industry comment that proposed exhaust standards combined with the
proposed effective dates should allow time for the entire industry to build catalyzed products that
do not increase any risks, we are finalizing the proposed standards in the years in which they
were proposed.  Given that we are finalizing the Phase 3 standards in mid-2008, manufacturers
will have a little over two years to redesign their Class II engines and a little over three years to
redesign their Class I engines.

       The safety analysis performed by EPA for Class I and II engines and SP Technical
Research Institute for Class II engines both indicate that the addition of catalyst technology can
be safely implemented with the proper design strategies. Both of these studies are available in
the docket. ("EPA Technical Study on the Safety of Emission Controls for Nonroad Spark-
Ignition Engines < 50 Horsepower," EPA420-R-06-006, March 2006, docket item EPA-HQ-
OAR-2004-0008-0333.) ("Scientific Evaluation of the Risk Associated with Heightened
Environmental Requirements on Outdoor Power Equipment - Phase II," SP Technical Research
Institute of Sweden, docket item EPA-HQ-OAR-2004-0008-711.1.) In addition, a detailed
                                          2-26

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


analysis of both studies is included in a Memo to Docket EPA-HQ-OAR-2004-0008 titled
"Nonhandheld SI Exhaust System Safety Analysis."

       The scope of our safety study included Class I and Class II engine systems that are used
in residential walk-behind and ride-on lawn mower applications, respectively. We conducted the
technical study of the incremental risk on several fronts. First, working with CPSC, we
evaluated their reports and databases and other outside sources to identify in-use situations that
create fire or burn risk for consumers. From this information, we identified ten scenarios for
evaluation covering a comprehensive variety of in-use conditions or circumstances that could
lead to an increased risk of fire or burn.  Second, we conducted extensive laboratory and field
testing of both current technology (Phase 2) and prototype catalyst-equipped advanced-
technology engines and equipment (Phase 3) to assess the emission control performance and
thermal characteristics of the engines and equipment.  Third, we conducted a design and process
Failure Mode and Effects Analyses (FMEA) comparing current Phase 2 and Phase 3 compliant
engines and equipment to evaluate incremental changes in risk probability as a way of evaluating
the incremental risk of upgrading Phase 2 engines to meet Phase 3 emission standards.

       Our technical work and subsequent analysis of all the data and information strongly
indicate that effective catalyst-based standards can be implemented without an incremental
increase in the risk  of fire or burn to the consumer either during or after using the equipment.

    2.2.7   Altitude

What Commenters Said:

       OPEI recognized that altitude provisions for handheld engines are controlled by
§1054.145(c)(4). This paragraph specifies that handheld engines must meet applicable emission
requirements up to  an altitude of 1100 feet (96 kPa). Kit information should be supplied in the
operator's manual and application for certification.  Handheld engines are small, compact and
also cannot bear the cost of automatic altitude compensation systems. Such engines also run
under high thermal  and mechanical load, which make them sensitive to increased air-fuel ratio
that would follow having to comply with the emissions standard also at high altitude settings. In
general, A/F ratio changes as a function of the square  root of the air density/fuel density.  This
may vary based on  unique engine characteristics. Depending on a manufacturer's compliance
margin and production line auditing values, OPEI believes the 1100-foot (96 kPa) requirement
should be acceptable.

       OPEI supported the proposed requirement that altitude kits must be available for non-
handheld products sold in geographic locations with higher altitudes. They believe the
prescribed ambient pressure limitation for determination of compliance is acceptable and more
appropriate than the referenced altitude. The manufacturer's ability to demonstrate compliance
with the prescribed exhaust emission standard levels at atmospheric pressures lower than 94.0
kPa should utilize a combination of historical  information, engineering analysis, and good
engineering judgment in the determination of altitude kit information to be included in the engine
family certification application and owner's manual. OPEI commented that the regulatory
requirements should be minimized in order to continue to allow the various manufacturer
processes that have and will continue to provide this service to their respective customers.


                                       2-27

-------
                                                             Chapter 2: Small SI Engines
       EMA supported the proposed requirement that altitude kits must be available for products
sold in geographic locations with higher altitudes. The prescribed ambient pressure limitation
for determination of compliance is acceptable, and a more appropriate metric than using actual
altitude. However, EMA commented that the manufacturer should be allowed to demonstrate
compliance with the prescribed exhaust emission standard levels at atmospheric pressures lower
than 94.0 kPa utilizing altitude kit information included in the engine family certification
application and owner's manual. Altitude kit design should be determined using a combination
of historical information, engineering analysis, and good engineering judgment. EPA should
minimize the regulatory requirements in order to continue to allow the various manufacturer
processes that have and will continue to provide this service to their respective customers.

       EMA noted that engine manufacturers can (and do) provide the necessary parts and
training for modification of products that are used in high-altitude conditions.  Customers that
operate equipment in high-altitude areas are well aware of the need for these modifications.  In
order for these provisions to be viable, EPA must allow a means for the manufacturer to provide
these altitude kits to consumers (dealer network, distribution system, etc.). However, EMA
believes that the owner's manual information should be limited to altitude effects that owners
will understand. Specifically, the owner's manual should include information that would inform
the consumer that engines operated at altitudes greater than the manufacturer prescribed
minimum may require the engine/equipment to be modified in order to ensure proper operation.
The owner's manual also should instruct the consumer to contact  the engine manufacturer for
further information. In addition, the information provided to the ultimate customer must identify
the range of altitude the product is  expected to operate in, the applicable engine specifics
required to determine the appropriate altitude kit, and where the customer can either obtain the
required kit or have the kit installed.

       EMA commented on §1054.115(c) "What other requirements apply?"  EMA commented
that the reference to 40 CFR Part 1065.520  should clarify that the specified barometric pressure
range of 94.0 to 103.325 kPa is an  exception rather than an additional requirement.  Further,
EMA commented that the meaning of the reference to a "standard configuration" is unclear.
Accordingly, the language should be revised to read as follows: "Engines must meet the
applicable emission standards for valid tests conducted under the  ambient conditions specified in
40 CFR Part 1065.520 except using a barometric pressure range from 94.0 to 103.325 kPa. This
requirement is applicable to nonhandheld engines distributed to all areas that do not exceed 2000
feet in elevation above sea level. See §1054.145(c) for handheld engine provisions.  For higher
altitude distribution, and resulting lower barometric pressures, carburetor modifications by the
use of altitude  kits is acceptable provided that these kits are specified in the certification
application and information is provided to the customer that identifies the altitude kit
requirements."
       EMA commented on §1054.501(b)(3) "How do I run a valid emission test?"  EMA noted
this section directs the manufacturer to perform testing under the ambient conditions specified in
40 CFR Part 1065.520, however, the ambient pressure range specified in  §1065.520 is a range
from 80.0-103.325 kPa and the pressure range specified in §1054.115(c) is 94.0-103.325 kPa.
Therefore, EMA commented that §1054.501(b)(3) should be revised to reflect the pressure range
                                          2-28

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


applicable to Small SI engines pursuant to §1054.115(c) and the reference to §1065.520 should
be deleted.
       EMA commented on §1054.205(r) "What must I include in my application?" EMA
commented that this section should be revised to clarify what  information must be submitted in
the certification application, and what information must be made available to consumers.  Engine
manufacturers routinely utilize engineering analysis to determine the altitude kit requirements
which are correlated to engine function performance at different altitudes. However, the engine
manufacturer does not have the ability to directly confirm emission compliance.  EMA
commented that this section should be revised so that it is clear that while the information
provided to the engine owner must be accurate, it also should  be easy to understand and not
overly technical.  Specifically, the information required to be provided to the engine owner
should either enable the engine owner to determine whether or not an altitude kit is appropriate
and necessary  for their operating location or provide contact information for a resource that can
assist with such determination.

Letters:
Commenter
OPEI
EMA
Document #
0675
0691
Our Response:

       The altitude-related requirements appropriately specify that emission standards apply for
nonhandheld engines throughout the range of atmospheric pressures identified in §1065.520,
specifically from 80 to 103.325 kPa. We are adopting special testing and compliance provisions
related to altitude.  We are requiring that nonhandheld engines meet emission standards without
an altitude kit, but will allow, in certain cases, testing at  barometric pressures below 94.0 kPa
(which is roughly equivalent to an elevation of 2,000 feet above sea level) using an altitude kit.
(An altitude kit may be as simple as a single replacement part for the carburetor that allows a
greater volumetric flow of air into the carburetor to make the engine operate as it would at low
altitudes.) Such kits were allowed under part 90 and we are keeping the provisions that already
apply in part 90 related to descriptions of these altitude kits in the application for certification.
This includes a description of how engines comply with  emission standards at varying
atmospheric pressures, a description of the altitude kits, and the associated part numbers.
We agree that §1054.501 should reference the pressure-related provisions in §1054.115, but do
not agree that the reference to §1065.520 should be deleted.

       OPEFs comments generally supported the proposed standards and related requirements
for complying with the regulations based on operation at high altitude. The requirement for
nonhandheld engines to meet standards up to 94.0 kPa without an altitude kit and for
manufacturers to specify the need for altitude kits to continue to comply with emission standards
at lower pressures (or higher altitudes) fits with the recommendations spelled out in the
comments.  Also, EMA's description of an approach to including altitude-related information in
the owner's manual is an excellent summary of what we would hope to see. The regulations are
                                       2-29

-------
                                                            Chapter 2: Small SI Engines
somewhat less descriptive than EMA describes, but we would have no objection if manufacturers
include the additional information suggested in the comment.

       We believe the proposed provisions requiring manufacturers to describe altitude-related
information in the application for certification are clear. The regulations in §1054.205 specify
simply that manufacturers must describe their plan for making information and parts available
such that they would reasonably expect kits to be widely used in high-altitude areas. The
example noted includes a very basic expectation that owners should have ready access to
information describing when an altitude kit is needed and how to obtain this service. The
detailed description included in EMA's comments would be a satisfactory approach to meet
these requirements.

       One thing engine manufacturers could consider adding in their communication to owners
would be geographic-based information.  For example, we identify in the regulation those
counties with median altitudes greater than 4000 feet above sea level. Owner's manuals or
websites could include specific information to identify those areas as needing altitude kits for
proper engine operation, if applicable.

       2.3  Averaging, banking, and trading

   2.3.1   Use of Phase 2 credits (and early Phase 3 credits)

What Commenters Said:

       EMA noted that the implementation of Phase 3 exhaust emission standards clearly will
play an important role in the continued improvement of the environment.  Early introduction of
clean technology that would further benefit the environment should be encouraged, and
manufacturers should be afforded meaningful incentives for the early introduction of these
cleaner Phase 3 products. It is imperative that EPA recognize that the ability of manufacturers to
comply with the Phase 3 program is tied to their ability to use existing Phase 2 credits and the
creation of transitional and enduring Phase 3 credits from the early introduction of Phase 3
product.  Engine manufacturers that have either provided a historical benefit or are eager to
provide additional environmental benefit through either early compliance or the introduction of
over achieving nonhandheld engines should be encouraged to do so. EMA recommended a
number of changes to the NPRM in order to ensure the success of this important program.  OPEI
also recommended the same changes in their comments.

       First, EMA and OPEI commented that the requirement that both Phase 3 transition credits
and Phase 3 enduring credits must be used prior to using Phase 2 credits should be revised.
Engines that are introduced early and produce Phase 3  enduring credits are providing a
substantial environmental benefit that should be encouraged.  As proposed, there is no incentive
for manufacturers to certify engines to FEL levels below the Phase 3 standard level because EPA
would require the resulting enduring credits to be used prior to Phase 2 credits. Said another
way, one of the costs to the manufacturer in investing in engines certified below the Phase 3
level is the likely loss of Phase 2 credits.  That is unfair, and makes no sense.  As a practical
matter, certifying below the Phase 3 level prior to expiration of Phase 2 credits would have no
benefit to the manufacturer as currently proposed. Accordingly, in order to promote and


                                          2-30

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


encourage the early introduction of Phase 3 and over-achieving nonhandheld products into the
marketplace as soon as possible, engine manufacturers must be allowed to preserve the Phase 3
enduring credits that they have the capability of generating (see §1054.740(c)).

       Second, EMA and OPEI commented that at the present time, the proposed AB&T
program creates substantial concern and potential exposure to engine manufacturers because of
their inability, despite their intentions, to be able to plan for such unforeseen factors as weather
and customer demand. When that uncertainty is coupled with the proposed combined limitations
on the use of Phase 2 and Phase 3 credits, manufacturers' ability to ensure compliance is
jeopardized. In order to prevent such unintended consequences, manufacturers should be
allowed to utilize Phase 2 credit banks discounted by 20% per year or, in the alternative, if no
Phase 2 credits exist, to carry a negative credit balance for up to two model years.  The option for
a manufacturer to maintain a negative credit balance would be at EPA's discretion, based on the
manufacturer's ability to provide information demonstrating that any negative credit balances
would be eliminated no later than the 2016 model year.  In addition, EMA and OPEI envision
that EPA would not allow a negative credit balance situation based on planned engine family
PEL and volume projections, but only on unexpected volume adjustments occurring within an
averaging set or the need to make an unanticipated upward adjustment to FEL due to an
insufficient compliance margin, as determined from production line testing (see §1054.740).

       Finally, EMA and OPEI commented that engine manufacturers which have provided a
benefit to the environment through the early introduction of Phase 3 credit generating
nonhandheld engines should not be penalized regarding the use of those credits for the continued
certification of their handheld engine families.  Handheld engine families will continue to
comply with the same  exhaust emission standards under the proposed Phase 3  standards as the
current Phase 2 engine families. A requirement for these carry-over handheld engine families to
utilize Phase 3 nonhandheld generated credits is inappropriate and should not be included in the
final rule. OPEI commented that these specific handheld families should be allowed to continue
to use Phase 2 credits under the provisions EPA has outlined.

       CARB commented that in general it supports restrictions on credit generation and use to
ensure that emissions benefits represented by the credits are accounted for properly. CARB
commented that it would like to strongly discourage the concept of carrying an emission credit
deficit. Manufacturers have sufficient time to plan for the change to new engines and  they have
the option to make the changes earlier than the deadline. If however, U.S. EPA chooses to allow
the use of credit deficits, CARB would strongly encourage a stiff penalty to be added to the
deficit as well as a  set time limit as to the length of time the deficit may be carried.

       CARB also commented that EPA should allow only Phase 3 nonhandheld engine credits
to be used under the handheld engine credit provisions after 2013.  They believe that prohibiting
the use of Phase 2 nonhandheld engine credits for demonstrating compliance with the  Phase 3
nonhandheld engine standards after 2013 is reasonable.
                                       2-31

-------
                                                            Chapter 2: Small SI Engines
Letters:
Commenter
EMA
OPEI
CARS
Document #
0691
0675
0682
Our Response:

       EPA believes ABT programs are an important element in setting emission standards that
are appropriate under Clean Air Act section 213(a) with regard to technological feasibility, lead
time, and cost, given the variety of engines covered by the small SI standards. Depending on
their design, ABT programs can create an incentive for the early introduction of new technology,
allowing certain engine families to act as trailblazers for new technology.  This can help provide
valuable information to manufacturers on the technology before they apply the technology
throughout their product line.  Early introduction of such engines can also secure earlier emission
benefits. Thus, EPA believes it is beneficial to design ABT programs to encourage use of the
ABT program, especially provisions that encourage the early introduction of new technologies.

       EPA agrees that the proposed provisions requiring manufacturers to use their enduring
Phase 3 credits before being allowed to use their Phase 2 credits allowance does not encourage
the early introduction of Phase 3 engines during the Phase 2 timeframe. In order to encourage
the introduction of Phase 3 compliant engines prior to implementation of the Phase 3 standards,
EPA is eliminating the proposed provision that would require a manufacturer to use their
enduring Phase 3 credits before using their Phase 2 credit allowance. Therefore, under the Phase
3 ABT program, engine manufacturers will be required to use their Phase 3 transitional credits
first. If their Phase 3 transitional credit pool is not sufficient, the manufacturer will be able to
use their Phase 2 credit allowance second. Should that still not be sufficient, then the
manufacturer will be allowed to use their Phase 3 enduring credits last of all to demonstrate
compliance.

       With regard to the comments on credit deficits, EPA does not believe such  provisions are
necessary for the Phase 3 program. Given the amount of lead time before the new  standards are
scheduled to take effect and the provisions allowing use of limited Phase 2 credits, EPA believes
manufacturers should be able to monitor their production levels and establish conservative PEL
values (especially during the introduction of new engine families) to avoid situations where a
deficit situation occurs. While EPA understands that manufacturers could find themselves in a
situation where sales volume adjustments occur within an averaging set or an PEL  needs to
adjusted upward, EPA believes that such changes should be relatively small and within the
manufacturers control to a great degree.  Manufacturers participating in the ABT program would
need to take these potential outcomes into consideration when making plans for complying with
the Phase 3 standards through  the use of the ABT program.

       Finally, in response to the comments on requiring manufacturers to use only Phase 3
nonhandheld credits for handheld engines, EPA is not adopting such a requirement in the final
rule. Under the Phase 3 program, in which we are not changing the Phase 2 exhaust standards
for handheld engines, manufacturers  of handheld engines will be allowed to use credits from
                                          2-32

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


Phase 2 handheld engines for their Phase 3 handheld engines without any restriction. Therefore,
EPA believes it can allow manufacturers to use Phase 2 credits from nonhandheld engines to
offset their high-emitting handheld engines under the constraints specified in the rule. (As noted
in Section 2.3.2, EPA is adding an annual sales limit of 30,000 handheld engines for which a
manufacturer can use nonhandheld engine credits.)


   2.3.2  Averaging sets and other restrictions

What Commenters Said:

       EMA commented that the restrictions regarding cross-class trading of credits is
appropriate during the introductory period when unrestricted trading could effect standard
implementation, use of Phase 2 credits, and other factors.  However, those restrictions should be
removed beginning with the 2013 model year.  Therefore, EPA should clarify that there are no
restrictions for nonhandheld credit trading beginning with the 2013 model year (see
§1054.740(d)).

       Honda commented that EPA should allow averaging, banking and trading (ABT) of
credits across all engine categories, including handheld and nonhandheld engines.  The proposed
rule uses a stepped form of dividing engines and their respective emission levels into three
categories, 0 to 80cc, 80 to 225cc, and above 225cc, and allows averaging and banking only
within these separate categories. The inability to average and bank credits inherently applies
more significant technical and economic challenges for engines of smaller displacements, within
their class, to comply with the specific standard. The ability to supply engines in all
displacement and horsepower categories is enhanced by the ability to "smooth" through
averaging, exhaust standard steps in a way that resembles the feasible horsepower/displacement
curve function.  Honda understands that the proposed rule limits the use of existing Phase 2
credits in order to "pull-ahead" the effective implementation of the regulatory standards.
However, Honda believes that treatment of Phase 3 credits, including early Phase 3 credits,
should be independent of the Phase 2 credits and commented that EPA should consider allowing
averaging, banking and trading across Phase 3 categories,  in the same manner allowed in Phase
2.

       Honda commented that EPA should clarify in the final  rule when and if an engine less
than 80cc would be categorized as nonhandheld for ABT purposes if EPA does not allow Phase
3 cross class averaging. Clarification or added guidance in the final rule would be useful where
an engine less than 80cc is used in a nonhandheld product would qualify as nonhandheld for
purposes of ABT, such as an engine used in a ground-supported mini-tiller.

Letters:
Commenter
EMA
Honda
Document #
0691
0705
                                       2-33

-------
                                                            Chapter 2: Small SI Engines
Our Response:

       With regard to the comment on cross-class trading of nonhandheld credits within the
nonhandheld engine classes, EPA proposed to allow such trading starting in model year 2013.
EPA has added language to specifically state that this is allowed in §1054.740 of the regulations.

       EPA believes the proposed restrictions on credit exchanges between handheld and
nonhandheld engines in the Phase 3 ABT program should be retained in the final rule (with the
limited exception as noted below). While EPA is adopting more stringent exhaust standards for
nonhandheld engines in this rule, EPA is not revising the Phase 2 exhaust standards for handheld
engines. While most manufacturers tend to be in either the handheld engine market or the
nonhandheld engine market, there are a few manufacturers that have a mix of engines falling into
both categories.  Under the Phase 2 program, where EPA allowed unrestricted averaging across
both handheld and nonhandheld engine categories, some manufacturers were able to use credits
from one category to delay introduction of cleaner technology engines in the other category.
This gave these manufacturers a potential advantage in the market compared to other engine
manufacturers that implemented the new technologies and did not have the ability to average
with engines in the other category. EPA does not believe an ABT program should encourage
such situations in the market. For this reason, EPA is retaining the averaging set restrictions for
the Phase 3 rule which prevents averaging of emissions between handheld and nonhandheld
engines (except as noted below).

       EPA is adopting the proposed provisions which allow manufacturers to use nonhandheld
engine credits for handheld engines if the engine family was certified in 2008 based on carryover
emissions data and the PEL does not increase above the level selected for the 2007 model year.
Based on current certification data, only a small number of engine manufacturers would be
impacted by these provisions and the number of handheld engines potentially affected is very
small, with overall sales being less than 1 percent of handheld engine sales.  However, because
of concerns that manufacturers could increase their sales of such high PEL handheld engines,
EPA is adopting  one additional constraint. Under the final regulations, manufacturers may use
nonhandheld credits for up to 30,000 handheld engines per year. EPA believes that the
constraints  being adopted regarding the use of nonhandheld engine credits for handheld engines
should ensure that the sales of these handheld engines remain at their currently low levels.

       In regard  to the comments on which engines certified to the handheld engine standards
can generate nonhandheld engine credits, EPA proposed to allow manufacturers to generate
nonhandheld ABT credits from engines below 80cc for those engines a manufacturer has
determined are used in nonhandheld applications.  EPA is retaining that provision in the final
rule. Therefore,  a manufacturer can generate nonhandheld engine credits from engines at or
below 80cc that are subject to the handheld engine standards if the manufacturer determines they
are used in  nonhandheld applications (i.e., applications that do not meet the handheld definition
in §1054.801 of the regulations). Because the engines are subject to the handheld engine
standards, the credits would be generated against the applicable handheld engine standard.
These nonhandheld credits could be used within the Class I and Class II engine classes to
demonstrate compliance with the Phase 3 exhaust standards, subject to applicable restrictions.
Given the restriction on mixing credits between handheld and nonhandheld engines, credits
                                          2-34

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


generated by engines at or below 80cc used in handheld applications could only be used for
handheld engines.


   2.3.3  FEL caps

What Commenters Said:

       EMA supported the proposal that FEL caps should be established at the Phase 2 standard
levels.  EMA commented that the engine families previously considered Class I-B under the
Phase 2 regulation and that are set to become Class I engines under the proposal must be allowed
to utilize the prior Class I-B standard level at the FEL cap as those engines were not subject to
the Phase 2 Class I standard levels.

Letters:
Commenter
EMA
Document #
0691
Our Response:

       EPA agrees with the comment that FEL caps should be established at the Phase 2
standard levels. The final rule includes such a requirement in §1054.103(b) by stating that a
manufacturer may not specify a family emission limit that exceeds the applicable Phase 2
standards as specified in 40 CFR 90.103 and summarized in Appendix I of Part 1054.
   2.3.4  Credit life

What Commenters Said:

       Both EMA and OPEI commented that they oppose the proposition that any engine-
exhaust or evaporative credits generated by a manufacturer should have an arbitrary life period.
Emission credits are either generated through the voluntary early implementation of new
emission control technology or introduction of products that are cleaner than required by the
applicable emission standard.  Such credits are generated at a cost to the manufacturer, and are
granted in exchange for the manufacturer's independent decision to produce products that
provide additional benefits to the environment. These credits are important assets that should not
be arbitrarily lost due to time or actions not under the manufacturer's control.

       CARB commented that it strongly urges EPA to limit the credit life of exhaust credits
earned to five years. They commented that emission credits should not outlast the equipment
which allowed the manufacturer to attain the credits. CARB also commented that a five year
limit on the credit lifetime would also be consistent with the proposed useful life requirements
under which the engine manufacturers would be required to warrant the engine for five years.
                                       2-35

-------
                                                            Chapter 2: Small SI Engines
       Briggs & Stratton commented that it opposes any limitation on the life of ABT credits.
Engine manufacturers should not be punished for not using the credits in an arbitrary time frame.
Briggs and  Stratton also  commented that the proposed ABT provisions almost completely
eliminate the Phase 2 emission credits that have been generated by small engine manufacturers.
Engine manufacturers in good faith generated Phase 2 credits under the current regulations. The
proposal by EPA is a significant loss to the engine manufacturers.

Letters:
Commenter
OPEI
EMA
CARS
Briggs and Stratton
Document #
0675
0691
0682
0657
Our Response:

       We are retaining the unlimited lifetime for Phase 3 ABT credits, as proposed.  While
EPA is retaining the unlimited lifetime, EPA notes that manufacturers should not assume that
Phase 3 credits will be available without any restrictions on their use if, and when, EPA should
consider a new round of emission standards in the future.  In setting new emission standards,
section 213(a) requires of the CAA requires EPA to set emission standards that achieve the
greatest degree of emission reduction achievable through the application of technology which
EPA determines will be available for the engines or vehicles to which such standards apply,
giving appropriate consideration to the cost of applying such technology within the period of
time available to manufacturers and to noise, energy, and safety factors associated with the
application of such technology.  If manufacturers have a large pool of ABT credits available to
them, EPA must consider ways to ensure that those credits do not result in an unnecessary delay
of the standards. This can be done in a variety of ways, and has been done in the Phase 3 final
rule by allowing only limited numbers of Phase  2 credits to be used for a limited period of time
during the transition to the  new Phase 3 standards.

       EPA does not believe a limit on the life of Phase 3 credits is needed at this time for the
ABT program adopted with today's program. Phase 3 credits will be generated at a cost to
manufacturers and thus they will have a value to the manufacturers. EPA believes provisions
which limit a manufacturer's ability to use credits during the Phase 3 timeframe, such as a limit
on credit life, will reduce the incentive for manufacturers to invest in the development and
introduction of new technology. However, as mentioned above, manufacturers should not
assume that an unlimited life for Phase 3 credits means those credits will be available without
any restrictions on their use if, and when, EPA should consider a new round of emission
standards in the future.  As part of any future rulemaking, EPA would expect to consider ways to
ensure that the Phase3 credits existing at that time would not result in  an unnecessary  delay of
any future standards.

       With regard to the comment on the loss of Phase 2 credits, EPA does not believe
manufacturers have a right to use those credits indefinitely. In fact, EPA would like to point out
that such a scenario was clearly a possibility  and was noted in the Summary and Analysis of
                                          2-36

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


Comments document for the Phase 2 nonhandheld rule.  ("Summary and Analysis of Comments,
Phase 2 Emission Standards for New Nonroad Spark-Ignition Nonhandheld Engines At or Below
19 kW," March 3, 1999, docket item EPA-HQ-OAR-2004-0008-	.) In response to
comments on the unlimited lifetime of Phase 2 ABT credits, EPA stated that while it was
adopting an unlimited credit lifetime for Phase 2 ABT credits at the time, EPA did not wish to
limit its ability to address possible unforeseen conditions that arise as a result of the program in
future rulemakings.  EPA further stated that it would be able to reconsider the appropriate life of
Phase 2 ABT credits in connection with any post-Phase 2 rulemaking.

   2.3.5  Other ABT  Issues

What Commenters Said:

       EMA submitted a number of comments on specific regulatory sections. EMA
commented on §1054.105(b) "What exhaust emissions standards must my non-handheld engines
meet?" EMA noted that while the proposed AB&T program is restricted to HC+NOx emissions,
the NPRM does not expressly state that CO emission standards cannot use AB&T (as previously
included in Part 90.201). Because the proposal includes a compliance requirement with the
significantly lower CO standard for marine generator engines, EMA commented that the final
rule should clarify that AB&T is not applicable to CO emissions.
       EMA commented on §1054.715(b) "How do I bank emission credits?"  They commented
that reserve credits cannot be traded. Therefore, EMA recommends that the reference to
"trading" should be deleted from this section. (Also included in Section 4.4.5)
       EMA commented on §1054.725(b)(2) "What must I include in my application for
certification?"  They commented that engine families that generate or use credits at the time of
certification  should not be required to designate their credit destination or origin within the
averaging  set. (Also included in Section 4.4.5)
       EMA commented on §1054.730(f)(3) "What ABT reports must I send to EPA?" EMA
commented that if an error mistakenly increases a manufacturer's balance of emission credits,
correction of the errors and recalculation of the balance of emission credits should be undertaken
at the manufacturer's discretion. They manufacturer should not be required to  correct the errors
and recalculate the balance of emission credits as currently proposed. (Also included in  Section
4.4.5)
       EMA commented on §1054.735(d) "What records must I keep?" EMA commented that
the requirement to keep additional records for each engine or piece of equipment including the
engine identification number, build date and assembly plant is excessive and beyond the current
requirements of 40 CFR Part 90.209. They commented that these additional record keeping
requirements either should be deleted or replaced with engine manufacturer records associated
with products produced. (Also included in Section 4.4.5)
       EMA commented on §1054.735(e) "What records must I keep?" EMA commented that
this section appears to be arbitrary and capricious.  EPA should not be allowed to require
manufacturers to keep additional unspecified records or demand additional information not
required by the rule without a proper purpose or for cause. EPA should be required to support
any imposition of additional record keeping requirements or demand for additional information
                                      2-37

-------
                                                            Chapter 2: Small SI Engines
with specific and appropriate reasons. Further, such decisions should not be made unilaterally by
EPA, and the manufacturer must have the ability to question any such request and, if necessary,
request a formal hearing process. (Also included in Section 4.4.5)
      EMA commented on §1054.625(j)(2) "What requirements apply under the Transition
Program for Equipment Manufacturers?" EMA commented that the requirement to multiply
credits generated by an engine family by 0.9 must be limited to engine families that are actually
utilized by equipment manufacturers under the §1054.625 flexibility provisions. In  addition,
EMA commented that engine manufacturers should have the option to track the correct number
of engines that utilize the §1054.625 flexibility provisions and adjust ABT credit calculations
based on the actual number of engines.

      OPEI commented that the "flex" provisions will not be implemented if excessive
regulatory burdens discourage engine manufacturers from participating in that program. OPEI
believes an unintended possible outcome of the proposed ABT credit adjustment program is the
creation of a disincentive for engine manufacturers to participate in the flexibility program.  Such
a credit adjustment requirement would be unfair (in terms of lost, banked, credits) and would
also be overly burdensome to administer. To OPEFs knowledge, such ABT credit adjustments
are  not part of EPA's other similar equipment flexibility programs. If such an adjustment
program is required, OPEI recommended that the credit adjustment provisions for Delegated
Assembly be clearly defined as applicable only to those discrete engine families that utilize the
Delegated Assembly provisions and are also participating in the flexibility program  pursuant to
Section  1054.625(c)(2). There are many circumstances where no Delegated Assembly engines
will be utilized for equipment manufacturer flexibility programs and discounting of credits
should not occur under any scenario. If a credit adjustment program is required, engine
manufacturers must be given the option to participate in the flexibility program pursuant to
§1054.625(c)(2).

Letters:
Commenter
EMA
OPEI
Document #
0691
0675
Our Response:

       With regard to the comment on CO, the ABT program for small SI engines does not
cover CO emissions.  EPA agrees that language stating that CO is not part of the ABT program
should be included in the Part 1054 regulations. EPA has revised §1054.103 and §1054.105 to
include such language.

       EPA disagrees with the comment on §1054.715(b) suggesting that reserved credits cannot
be traded. The existing Phase 2 ABT regulations in 90.206(b), allow manufacturers to trade
current model year credits. Current model year credits are "reserved" credits by definition,
because manufacturers do not submit their end-of-year and final reports until after the model
year is finished. Therefore, EPA believes it is appropriate to include similar language in the
Phase 3 ABT regulations stating that reserved credits can be traded.  Therefore, §1054.720 of the
regulations states in paragraph (b) that a manufacturer may trade reserved emission credits.
                                          2-38

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


Also, it should be noted that the language of §1054.715, paragraph (b) which was noted by the
commenter in their comments has been changed to fix an inconsistency in the regulations. The
proposed language stated that a manufacturer's credit projections submitted at the time of
certification were considered "reserved credits."  This is not the case as "reserved credits" are
defined in §1054.701 as credits that have been generated, but not yet verified by EPA. In order
to generate credits, the manufacturer must actually build engines, not just project that they will
build engines.  However, the revised language of 1054.715(b) also allows reserved credits to be
traded.

       EPA agrees with the comment on §1054.725(b)(2) suggesting that manufacturers not
have to designate the credit destination or origin for each of its engine families and has removed
that requirement from the regulations. However, for engine families that are projecting to use
emission credits (i.e., the engine family has a negative credit balance), EPA believes that a
manufacturer should provide information on where they will obtain credits for that engine
family. Therefore, at the time of applying for certification, a manufacturer will be required to
submit certain information to EPA.  For the Phase 3 ABT program, the regulations require
manufacturers to provide the PEL for the engine family, and detailed credit calculations for the
engine family,  as well as where they will obtain credits for their credit using families (i.e., from
banked credits, from averaging with other  current engine families certified with FELs below the
standard, or from trading with other engine manufacturers). EPA does not believe it is necessary
to require manufacturers to provide any further information, including a detailed accounting of
where they plan to use their credits or if the credits they plan to use are actual or reserved, as
proposed.

       EPA disagrees with the comment on §1054.730(f)(3) suggesting that a manufacturer be
allowed to fix errors in the credit reports at its discretion. If errors are discovered at any time
showing that a manufacturer has earned too many credits, then EPA believes a manufacturer
should be required to correct the error. The ABT program  is meant to ensure that the average
emission level  of all participating engines meet the applicable standard.  An error in the ABT
reporting that results in more credits being generated than should be generated could result in the
average emission level being above the emission standard even though the original credit
calculations did not show such a result.  Requiring manufacturers to fix such errors would allow
EPA to then address any resulting noncompliance.

       In response to the comment on §1054.735(d) that the information a manufacturer is
required to keep for ABT is excessive, EPA is making some changes to the regulations. EPA
believes the changes will still allow us to have access to important information if needed,
especially if a noncompliance situation arises. Under §1054.730 of the final regulations,
paragraph (b) requires manufacturers to report a variety of information for engines participating
in the ABT program including family designation, PEL, useful life, and the production volumes
for each participating family.  Under §1054.735, manufacturers will be required to keep a copy
of the reports submitted to EPA under §1054.730 along with a record of the identification
number for each engine produced.  If there are multiple FELs in an engine family, the
manufacturer will need to keep records of the identification number associated with each FEL.
Manufacturers may identify these numbers as a range. Manufacturers will not be required to list
                                       2-39

-------
                                                            Chapter 2: Small SI Engines
the build date for each engine produced, nor will they be required to keep information on the
assembly plant as originally proposed.

       With regard to the comment on §1054.735(e) on the provisions requiring manufacturers
to keep or allowing EPA to request additional unspecified records or relevant information, EPA
believes such a requirement is allowable and necessary.  Section 208 of the Clean Air Act (which
applies to nonroad engines under section 213(d) of the Clean Air Act) describes the information
collection requirements for manufacturers. Under those requirements, manufacturers must
provide information EPA may reasonably require to determine whether manufacturers have
acted in compliance with regulations.  While EPA has listed the specific information a
manufacturer must keep for the Phase 3 ABT program in  §1054.735, it is possible in the future
that we may identify other information that would be needed to deal with a specific situation.
The provisions in paragraph (e) of §1054.735 would allow us to request such information.  Of
course, EPA would only request such additional information if it were in accordance with the
law,  such as provided for in section 208 of Clean Air Act. In addition, EPA would only expect
manufacturers to keep and provide such information after we have put such a request into effect,
either through a rulemaking change or guidance to manufacturers.  In response to the comment,
EPA has revised the language of §1054.735 paragraph (e) to reflect that EPA will request
information if it is in accordance with the law.  Finally, in response to the comment on
requesting a hearing if a manufacturer believes EPA's request is inappropriate, the proposed
regulations allow a manufacturer to request a hearing from EPA under §1054.745, paragraphs (c)
and (d). EPA has retained those requirements in the final regulations.

       In regard to the comments on §1054.625(j)(2) on adjusting ABT credits for credit-
generating engine families that are available under the delegated-assembly provisions, EPA
continues to believe it is appropriate to adjust such credits. Rather than imposing a disincentive
from participating in the transition program for equipment manufacturers (TPEM) program, the
credit adjustment merely accounts for the fact that equipment manufacturers may in many cases
legally install a non-catalyzed muffler on an engine that is part of a family whose certification
depends on the use of a catalyst. It is true the EPA has not adopted this adjustment for other
engine categories, but this is because most other engine categories do not have a TPEM program
and none of them allow engine manufacturers to produce  these engines without specifically
identifying them as exempt TPEM engines. EPA wishes  to clarify that the adjustment applies
only to engine families that are available under the delegated assembly provision and are also
participating in the TPEM program. As noted in the proposal, the proposed credit adjustment
factor of 0.9 is intended to represent the maximum estimated usage of the TPEM program across
the broad range of equipment manufacturers. However, EPA understands engine manufacturers'
concerns that the adjustment may not reflect the actual number of engines that are downgraded
for use in the TPEM program.  Therefore, for the final rule, EPA is retaining the 0.9 adjustment
factor.  In addition, EPA is including an option that will allow engine manufacturers to track the
final configuration of the engines to determine the actual  number of engines that were
downgraded for the TPEM program.  A manufacturer would need to track sales for all of the
equipment manufacturers purchasing the given engine family. The engine manufacturer could
use the resulting number of engines that were not downgraded in its calculation of ABT credits
for that specific engine family.
                                         2-40

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       2.4   Certification

   2.4.1  Changing the FEL mid-year (and printing FELs on labels)

What Commenters Said:

       OPEI and EMA disagreed with EPA's negative assessment of FEL changes made during
the model year.  Currently, engine manufacturers may set their initial FEL levels to a level that
they are confident will pass production audits and will not result in compliance concerns
(provided compliance with the standard requirements is achieved). If, based on the actual
production audit results and subsequent Cum-Sum analysis, the manufacturer determines that the
FEL could have been lower from the beginning of the model year (or implementation of a
running change), OPEI commented that the manufacturer should be allowed to claim credit for
the environmental benefit actually provided.  The ability to correct the FEL level also provides
the manufacturer with a limited ability to recoup credits under the EPA program that are
otherwise available to manufacturers under CARS's PLT credit program.

       OPEI agreed that lowering the FEL should have a limit. A manufacturer that has already
submitted production line test data for a family should not be allowed to retroactively (even at a
point in the 4th quarter that would apply back to the first quarter) lower the FEL later in the
model year to a level that would result in a CUM-SUM failure from earlier tests in the model
year. This in effect sets a cap on the FEL change.

       OPEI and EMA commented that the proposed rule's requirement to include the FEL
numerical reference on the engine emission label would prevent the manufacturer from being
able to accurately represent (in the CARB PLT credit case) or retroactively change FEL levels.
They commented that the final regulation should provide the engine manufacturer the ability to
make retroactive FEL adjustments. Also, in order to allow such adjustments, OPEI and EMA
believe it is essential that EPA drop the proposed FEL labeling requirement.

       OPEI noted that CARB does not require either exhaust or evaporative family emission
levels (FELs) to be placed on the emission label. EPA's proposal to add individual evaporative
and exhaust family emission levels (FELs) on the label would be inconsistent with CARB, would
further confuse consumers, and would be totally impractical for manufacturers. For example,
consumers could end up unintentionally buying a product with more horsepower that emitted
greater mass emissions because the consumer did  not realize the FELs are normalized to a single
kilowatt. Such FEL labeling will also facilitate additional local purchase restrictions and use
bans in direct violation of Section 209(e) of the Clean Air Act and the related legal precedent on
federal pre-emption (as discussed below in Section XIV).  For all these reasons, OPEI
commented that EPA should drop completely its proposed exhaust and evaporative FEL labeling
requirement. (Comment also included in 4.6.1 and 4.6.3)

       EMA commented on §1054.701(e) "General provisions."  EMA believes the requirement
that an FEL can only be adjusted applicable to future production is not appropriate and should be
deleted. For example, if  a manufacturer determines, based on PLT test results, that the margin
                                      2-41

-------
                                                            Chapter 2: Small SI Engines
for compliance is inadequate and that credits exist either from the current model year or prior
model year banked credits, it should be allowed to increase the FEL for the entire model year.

       With regard to FEL changes, CARB commented that it agrees with EPA that any
revisions to the FEL should only apply to engines produced after the FEL  change. They agree
that it would be difficult to test the engines before the FEL change because verifying the engine
emissions from previously produced engines would be difficult.

       Briggs and Stratton noted that EPA asked for comments on including the FEL on the
emission labels. Briggs & Stratton disagrees with this proposal. The addition of the FEL on
emission labels provides no benefit to consumers, EPA, or the environment. However, this
proposal does impose a significant burden on both the engine manufacturer and the OEM. They
noted that supplemental labels are required for many applications where the emission label is
obscured in the final product.  If the FEL must be printed on the emissions label a new emission
label is required whenever the FEL is changed. This creates more costs and labels to manage for
the engine manufacturer and the equipment manufacturer with no commensurate benefit to the
environment. Briggs and Stratton commented that EPA should delete the requirement for the
FEL to be printed on the label.

Letters:
Commenter
OPEI
CARB
Briggs and Stratton
EMA
Document #
0675
0682
0657
0691
Our Response:

       We maintain two principles that contradict the manufacturers' comments regarding FEL
changes and printing FELs on emission labels.  First, we believe that each engine a manufacturer
produces should be associated with a family emission limit at the point of production.  This is
important for ensuring proper accountability and enforceability. If manufacturers are able to
assign FELs after production with the only restriction being related to compliance with statistical
calculations for production-line testing, there is a great concern that it would be very difficult to
confirm that FELs were assigned appropriately.  Similarly, if accountants change FELs
retroactively, it would be very difficult to test engines after they have been placed into service
and establish whether it meets emission standards or not.  There would seem to be no clear way
of knowing which FEL applied to which engine.

       Second, the intent of production-line testing and the underlying statistical calculations
depend on the engine having a specific and permanent applicable standard (with or without an
FEL). The statistical calculations are based on a given number of engines passing or failing the
applicable emission standard out of a bigger population representing the complete emission
family.  Repeating the CumSum calculations after the end of the year has the effect of simulating
the  engine family as if the tested engines were the complete population. Aside from the bad
math, we believe manufacturers should set their FELs with the understanding that they are  liable
                                          2-42

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


for the test results as they are generated. Waiting until the end of the year to set the real and final
FEL sets up an incentive for the manufacturer to use a high FEL through the year, then simply
reduce the FEL at the end of the year as much as the statistics allow. This puts the manufacturer
in a position of having almost no liability from production-line testing. In contrast, we believe
manufacturers should set the FEL for a family only as low as they can based on the
understanding that tested emissions must comply with the named FEL. If manufacturers learn
early in the model year that their FEL is higher than it needs to be, they may decrease the FEL as
much as can be justified based on prior testing and use that lower FEL for most of the model
year to generate larger quantities of credits (or use smaller quantities).

      We proposed to require manufacturers to print FELs on emission labels. This is common
across our programs and is intended to help us clearly establish the applicable emission standard
for each engine. In discussions after the close of the comment period, manufacturers agreed with
us that it would be as effective for the manufacturers instead to keep records to correlate  engine
build dates with changing FELs.  For example, if a manufacturer would change the FEL for an
engine family for production engines starting June 14 of a given year, it would keep a record of
engine identification numbers that would allow them to identify the applicable standard for each
engine.  If manufacturers choose to identify their build dates by month  and year (without the
specific date), the presumed build dates would default to least favorable dates for the
manufacturer.  In the case of an FEL increase on June 14, this means the manufacturer would
apply the new FEL starting with engines produced on June 1;  conversely, a decreased FEL
would apply starting with engines produced on June 30.  This flexible approach would allow the
manufacturer to forego some emission credits for the advantage of being less careful with
tracking engine serial numbers with build dates.  This approach for assigning dates for
calculating emission credits may be slightly different than the timing associated with the revised
certificate that we would issue for the engine family; however, we believe this should not be a
problem.

      Along with the requirement to keep records of engine build dates with FEL changes, we
are adopting a requirement for the manufacturer to report this  information in the ABT reports
submitted after the end of the model. These engine identification numbers may be submitted as a
range of values to streamline the report as much as possible.
   2.4.2  Useful life implementation (and labeling)

What Commenters Said:

       CARB commented that it agrees with U.S. EPA that a numerical value is the best way to
describe the useful life of equipment. If other terminology is used, CARB suggests that both the
descriptive terminology and the numerical value should be used.  If only one can be used, then
CARB suggests that the numerical value representing the useful life be retained.

       NESCAUM commented that it supports EPA's proposal to require engines and
equipment be labeled in a manner that will help the user better understand the intended useful
life of the equipment. They believe using descriptors such as Residential, Premium Residential,
                                      2-43

-------
                                                            Chapter 2: Small SI Engines
Commercial, and Heavy Commercial will be helpful in this regard, provided that there is a
means to match the descriptor against a specified useful life period in terms of operating hours or
years.

       OPEI commented that it is concerned that a consumer may become confused if only the
hours are listed on the label.  A 125-hr handheld product is considered "Premium Residential"
according to EPA while a 125-hour Class I engine would be classified as "Residential". To
avoid this confusion OPEI requested the EPA allow for the use of hours, or the use of descriptive
terms (Light Use, Medium Use, Heavy Use), or the use of both hours and descriptive terms on
the label. OPEI agrees that Class I and Class II engines and their applications require different
terms of usage.

       OPEI requested that a handheld product manufacturer would have the choice of using any
of these three options to describe the Useful Life period.  OPEI supports the use of the terms
"Light Use", "Medium Use", and "Heavy Use" to characterize the three useful life categories
applicable to handheld engines instead of the terms EPA has proposed. OPEI believes their
proposed terms best meet EPA's objective of accurately describing the intended use to the
purchaser. It is possible that  a commercial operator may buy  a product with a lower useful life
but the usage pattern of the product would fit the description of the useful life. The integrated
nature of handheld products allows a more transparent understanding of the durability of the
product through marketing and other means.

       OPEI noted that §1054.107(a)(4) deals with keeping information available to  support
Useful life selection. In addition, OPEI noted that page 88 of the Preamble states EPA intention
to review Useful-life selection if not highest value. By default, if a manufacturer certifies to
highest value, they are showing through cert testing that the engine meets the useful life period.
OPEI commented that EPA should add language in §1054.107(a)(4) to confirm that EPA will
approve the manufacturer's useful life selection without further demonstration if the
manufacturer selects the highest available useful life value and submits data showing that the
engine lasted that long as part of the durability demonstration for certification.

       OPEI notes that §1054.107(a)(4)(i) life time surveys are a point of interest for most
manufacturers; "If a manufacturer has data to support an engine/product has the majority of its
family sales sold to a market  (for example homeowner use) then the manufacturer may certify
the product/engine to an appropriate useful life provided the data supports that majority of the
product built and sold does not exceed the actual usage time." OPEI commented that if a
manufacturer has a family where 70% of sales can be proven  to be to homeowners and you can
prove that a large majority, for example greater than 75%, of those homeowners will  never use
the product more than 125 hours before scrapping it, the manufacturer should be allowed to
certify it to 125 hours even if the engine can be demonstrated to last longer.

       EMA  commented that it is critical that EPA recognize an  engine's useful life period, as
determined by the engine manufacturer prior to certification and production, does not dictate the
ultimate equipment manufacturer  or ultimate consumer's usage of the engine.   There  are a
significant number of engines produced in  this product category that will never be  used for the
prescribed emission durability period regardless of the years  of use.  There are also a very small
                                          2-44

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


number  of engines  in this category  that  will accumulate hours at a much faster  rate,  and
depending  on the equipment design,  may  be replaced prior to the expiration of the emission
durability period. In many cases this usage  disparity exists within a single given engine family.

       EMA commented that EPA's criteria for acceptance of useful life must be  expressly
established and documented in order to assure consistent treatment and a level playing field. A
manufacturer's decision to  select the longest useful life  period  should require the same
justification as other useful  life period  selection. The criteria used to approve emission durability
periods must clearly be identified in the regulatory text or preamble language.   Manufacturers
must receive guidance from EPA regarding what types of records EPA expects to review in the
event it  asks a manufacturer to substantiate  its selection of an engine's durability period.  In
addition, EMA commented that the final rule  should expressly acknowledge that industry survey
information regarding product categories  usage  patterns,  such as previous OPEI surveys, is
acceptable documentation of a manufacturer's useful life selection.

       EMA commented on the proposed statement required by  §1054.135(c)(4) "How must I
label and identify the engines I produce?" EMA objects to the statement and commented that the
manufacturer should have the option to include language associated with the emission durability
period in the compliance statement. Accordingly, there should be a reference to
§1054.135(c)(12) in this section.
Letters:
Commenter
OPEI
NESCAUM
CARS
EMA
Document #
0675
0641
0682
0691
Our Response:

       This is the only program in which we allow manufacturers the discretion to select an
engine family's useful life.  We believe this is a necessary accommodation for the reality that
similar engines can be designed and used for widely varying purposes, users, and applications.
Making this selection is nevertheless fundamental to defining the stringency of the standards that
apply to the engine family so we strongly believe we should set up clear, objective, and practical
guidelines for choosing an appropriate useful life in each case. We should also have a role of
monitoring compliance with these guidelines and intervening in cases where a manufacturer is
misusing the available discretion to assign an inappropriately short useful life. We have
observed several cases under the Phase 2 program where manufacturers select the shortest
available useful life for an engine family where the engines are clearly designed and marketed as
long-life products for commercial applications. In contrast, some manufacturers have chosen a
mix of useful-life values that seems to appropriately match the varying design parameters and
intended usage patterns. Our intent is to create a program in which we can ensure that all
companies are together taking this approach of responsibly pairing useful life with the expected
in-use operating life.
                                       2-45

-------
                                                             Chapter 2: Small SI Engines
       As described in the proposed rule, we believe emission labels need to clearly state the
manufacturer's selected value for the useful-life period, in hours.  Under the regulation,
manufacturers are directed to select the useful life that most closely correlates with the
equipment's expected lifetime of service. We believe this decision is important not only for
emission controls, certification, and compliance but also for consumers.  If a manufacturer puts
in additional engineering and product features such that an engine can operate twice as long as a
competitor's engine while meeting emission standards throughout the longer useful-life period,
that should be clearly identifiable to the consumer as a superior product. The current approach
of identifying the useful life with a single-letter code does not communicate useful-life
information clearly enough.  Similarly, we believe that descriptive terms may be helpful in
communicating useful-life information, but they cannot replace the objective value of identifying
useful life with a universal and clearly understood metric. Including the engine operating hours
to identify the useful life is the best way to achieve this.

       We believe it may also be helpful to add descriptive terms to further characterize an
engine's useful life. We will therefore allow manufacturers the option of using prescribed
wording in addition to identifying the hour value for the useful life. We are adopting the terms
described in the proposal for nonhandheld engines.  We believe these terms are well matched to
the range of uses for nonhandheld applications. We have no objection to the wording suggested
for handheld engines.  Using different terms may be helpful to avoid any confusion that may
result from attaching the same descriptive terms to different useful-life values for handheld and
nonhandheld engines.
Application
Handheld
Nonhandheld Class I
Nonhandheld Class II
Useful Life (hours)
50
125
300
125
250
500
>500
250
500
1000
>1000
Descriptive Terms
Light use
Medium use
Heavy use
Residential
Extended life residential
(or general purpose)
Commercial
Heavy commercial
Residential
Extended life residential
(or general purpose)
Commercial
Heavy commercial
       We agree with EMA that the criteria for establishing an engine family's useful life should
be clearly defined and evenly applied for certifying engines. We also agree that the selected
value should not prevent equipment manufacturers from installing engines according to the their
own judgments about which engine is best suited to their particular equipment models, and that
owners should not be restricted in how (or how long) they use their engines or equipment.
Information about how equipment manufacturers and owners are selecting, installing, and using
                                          2-46

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


engines may factor into the engine manufacturer's decision regarding useful life, but once the
engine family has a useful life, that value should not be limiting for equipment manufacturers or
owners.

       The regulation text we proposed and are adopting in §1054.107 is changed very little
from the current regulation in §90.105, which was adopted in 1999. As described above, many
engine manufacturers have been taking a responsible approach in exercising their discretion to
select useful-life values.  We therefore believe that the proposed regulation, with a few minor
modifications, suitably defines the process for defining the terms and criteria for selecting useful
life. Fundamentally, the regulations require that manufacturers select the useful life value that
best represents the expected median in-use life of the equipment in which the engine will be
installed, including specification of a variety of information types for supporting the selection.
We do not  expect a dramatic change from current practice for those manufacturers that are
already making a good-faith effort to make proper selections. Making the effort to document the
basis for making these  selections,  which is already required under §90.105,  and subjecting those
decisions to EPA review will ensure that all manufacturers receive equal treatment under the
regulations. This will be a  substantial  improvement over the Phase 2 program where
manufacturers may find themselves at a competitive disadvantage by making responsible useful-
life selections.

       While we believe the regulation is sufficiently clear in establishing the meaning of useful
life and the process for making selections, we agree that further guidance will be helpful in
taking the next step of making concrete decisions about which useful-life value most
appropriately represents a particular engine family. This will give manufacturers assurance that
useful-life  selections will be made consistently across the industry, and will further help to
ensure an orderly process for certification.

       We agree with OPEI that engine manufacturers selecting the longest nominal value
would not need to do any more than submit certification data showing that an engine
representing the engine family operated long enough to appropriately establish deterioration
factors. If there were any reason for a manufacturer to select a useful life that is too long (such
as artificially generating credits from an engine family with a low family emission limit), we
would see that under the Phase 2 program where we have not asked manufacturers to justify their
selections.  We have observed no  such abuse under the Phase 2 program, so we have no reason to
believe that would occur in the future.  With no potential to require manufacturers to select a
longer useful life, we therefore believe it is unnecessary for manufacturers to provide any
additional information to justify their selection of the longest available nominal value for the
useful life.

       The regulation allows manufacturers to rely on product-specific surveys to establish the
median life span of equipment in the field. It would not be appropriate to rely on broader
surveys that characterize usage patterns or lifetimes for aggregated products, since that would
provide no information that would demonstrate any greater reliability or durability that may
apply for any particular engine family. It would not be appropriate to use industry averages to
justify lifetime estimates for individual models.  On the other hand, if a manufacturer has two
engines with similar designs and technical features (such as one- and two-cylinder versions of an
                                        2-47

-------
                                                             Chapter 2: Small SI Engines
otherwise common engine), it may be possible to draw conclusions about useful life for both
engine families from a single survey. We would expect such a survey to avoid sampling criteria
or other statistical methods that would distort the results.

       Conducting a survey is most straightforward when an engine is installed in a small
number of equipment models and is generally placed into service such that usage characteristics
are relatively uniform.  This situation is common for the highest-volume handheld and
nonhandheld products. A more challenging situation occurs when engines are installed in a wide
range of equipment models and users have widely varying usage patterns.  In these cases, we
would expect manufacturers to make sound judgments in selecting dominant equipment models,
applications, and usage characteristics to determine a useful-life value that best represents the
median lifetime of the range of equipment in which the engines are installed. This may reflect a
combination of commercial and residential use.  Surveys could also take into account the
possibility that individual owners may choose to retire a piece of equipment before it has reached
the point of no longer being able  to run (for example, by upgrading to a new model with
additional features).  The manufacturers are generally selecting the useful life from three nominal
values, so the goal of any survey  is limited to establishing the proper useful life only to that level
of precision.  We would not expect manufacturers to estimate the median lifetime of in-use
equipment to the nearest hour to be able to select the useful life for an engine family for
certification.

       In discussions following the close of the comment period, some manufacturers expressed
concern that gathering information from the field to determine appropriate useful-life values for
each engine family would be very costly and time-consuming.  We will be learning together how
detailed that information needs to be and to what extent the information can be shared across
engine families. In the meantime, we would also encourage nonhandheld engine manufacturers
to consider the alternative specified in the regulations allowing for useful-life determinations
based on engineering evaluation.  Toward that end, we have made an effort to correlate engine
design features with useful-life values. We considered including these design features directly in
the regulation, but chose to continue with the broader approach consistent with the current
specifications in §90.105.  To the extent that nonhandheld engine manufacturers are unable to
easily gather information to establish median equipment lifetimes corresponding to their engine
families, we would consider the engine design features in the following table to be an adequate
basis for establishing the useful life for a given engine family.  Manufacturers using the values as
indicated in the table would need to provide no additional information.  We are aware that
pressurized lubrication and cast-iron cylinder liners can take different forms (or have different
degrees of quality and durability), but we would consider any form of these technologies to
correspond to the indicated useful-life values, since they are clearly intended (and expected) to
provide substantial improvements in engine operating life. We may revise this approach to
correlating  engine design features to useful-life values based on testing or other information that
allows us to more carefully establish median lifetimes for specific designs.
                                          2-48

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
       Design Features for Nonhandheld Engines and Corresponding Useful Life Values
Design Features
Pressurized lubrication and more than one cylinder

Pressurized lubrication or more than one cylinder
Engine displacement at or above 225 cc or a cylinder liner
Any other engine design
Useful Life
1000 hours
500 hours
250 hours
125 hours
       Finally, manufacturers may choose to do testing instead of relying on survey information.
In this case, we would envision the manufacturer assembling five pieces of equipment that best
represent the engine family. Testing could also be done with engines on a dynamometer.  These
engines could be exercised until the point of failure under normal operating conditions with
proper maintenance throughout. The point of failure for the third failing engine would determine
the median lifetime for the engine family. The appropriate useful-life selection would be the
nominal value that is at least as high as the measured median lifetime. Manufacturers would
need to use good judgment in making a determination regarding the point of failure, including
consideration of the cost and ease of repair in the case of component failure and including
consideration of equipment performance in the case of reduced power output (from lost
compression, for example).  It would not be appropriate to consider a piece of equipment to be at
the end of its lifetime if a typical consumer with access to a reliable mechanic would have it
repaired or would otherwise continue using it. We would not accept the idea that a typical
consumer would as a matter of course dispose of equipment where an evaluation of the cost of
maintenance would justify continued use of the equipment instead of purchasing a new unit.
   2.4.3  Other labeling issues

What Commenters Said:

       EMA commented that EPA must recognize the fact that in order to fit on products that
typically are small, engine/equipment emission labels for Small SI engines are very small by
necessity.  Given the small size of the emission label, EPA should reconsider the labeling
requirements incorporated in the final rule. EPA should only require the most relevant
information to be on the label. Currently, the NPRM requires that the engine/equipment
emissions label include all of the following information: a numerical designation of the emission
durability period; Family Emission Limit (FEL); rated or intermediate speed; identification of the
emission control system; adjustment/tune-up information; altitude kit requirements; fuel and
lubricant requirements; and winter use identification. The inclusion of all of this information is
not only impossible due to the size of the label, but unnecessary.  Most of the information
required to be included on the label by the NPRM is information that is included in the
certification application and more appropriately included in the owner's manual. In addition,
emission labels are easy to counterfeit and the presumption that the inclusion of such additional
information will prevent or dissuade counterfeiting is not valid.
                                       2-49

-------
                                                            Chapter 2: Small SI Engines
       EMA believes the information provided on an engine/equipment emission label must be
easily read and understood.  If EPA requires too much information on such a small label, the
label will become cluttered, the size of the print will be extremely small, and the label will be
difficult to read.  Adding to the content of the label will not ensure compliance. In fact, adding
additional content to the label will thwart EPA's labeling goals because it will prevent the easy
identification of (i) the engine/equipment manufacturer; (ii) compliance applicability; and (iii)
date of manufacture. EMA commented that in order to ensure a label that is effective, and easy
to read and understand, the emission labeling requirements should be limited to the inclusion of
the following important information:

       a.  Manufacturers corporate name or trademark
       b.  Engine family name (exhaust)/ evaporative code (evap)
       c.  Date of manufacture (month and year) unless it is stamped or engraved elsewhere on
       the engine/equipment
       d.  The following statements of compliance (where applicable, the word "engine" would
       be replaced with the word "equipment"):
           i.     Exhaust - "This engine complies with U.S.  EPA Exh. Stds."
           ii.    Evap. - "This engine complies with U.S. EPA Evap. Stds."
           iii.    Exhaust & Evap. - "This engine complies with U.S. EPA EXH/EVP STDS."

       EMA commented that the proposed requirement to include the FEL on the emission label
is not acceptable because it precludes the manufacturer from (i) accurately representing the FEL
(in the CARB PLT Credit situation); and (ii) making necessary retroactive changes to an FEL
level. In addition, this requirement will impose an undue burden on both engine and equipment
manufacturers because it will require the manufacturer to create new labels (and dispose of old
label inventory) every time an FEL is changed, and to maintain both original and supplemental
labels. The addition of the FEL to the engine label does not add information that is valuable to
the equipment manufacturer, the ultimate purchaser, or EPA, and creates additional unjustified
burden on the manufacturer. For these reasons, and in light of the limited available space on the
engine label, manufacturers should not be required to include  the FEL on the engine label.

       EMA commented that due to the limited  size of the engine label, information that is more
appropriately included in the owner's manual should not be required to be included on the
emission label. They commented that EPA should require the following information to be
included in the owner's manual instead of on the emission label: (i) identification of the emission
control system; (ii) adjustment / tune-up information; (iii) altitude kit requirements; and (iv) fuel
and lubricant requirements.

       EMA is opposed to the new emission label requirements for winter exclusive engines.
Winter exclusive engines are uniquely configured to run in cold climates (e.g., they do not
typically have air cleaners, and often have winter calibrations  and hot air ducting), and would not
run well or last long in other types of applications. Winter exclusive engines already are
adequately identified and discernable from non-winter exclusive engines by the engine family
name, and the engine manufacturer's scheme for encoding this information into their family
naming convention. Accordingly, EMA commented that there is no need to include this
                                          2-50

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


information on the label.  Such a requirement would unnecessarily take up space on an already
crowded label.

       EMA also opposed the additional requirement to identify rated or intermediate speed
application restrictions on the emission label. This information does not add any value to the
label content and should therefore be eliminated.

       If EPA determines that it is necessary to identify delegated assembly engines on the
emission compliance label, EMA commented that EPA should only do so with the use of an
identifying mark on the permanent label, such as "DA" as an approved abbreviation for
"delegated assembly."

       Wherever possible, EMA commented that EPA should strive to minimize differences
between EPA's and CARB's labeling requirements. The NPRM requires the following emission
label heading: "EMISSION CONTROL INFORMATION"; while CARS requires either
"IMPORTANT ENGINE INFORMATION" or "IMPORTANT EMISSION INFORMATION".
There is no valid reason for EPA and CARB to have different emission label heading
requirements. As such, EPA and CARB should align on this issue. In the past, EPA has
accepted CARB label headings as an approved alternative. EMA urged EPA to include the
CARB heading as an option in the final  rule in order to avoid confusion and the need for
additional approvals to achieve this critical alignment.

       EMA believes the emission label is the appropriate location for identification of the
manufacturer responsible for compliance and related emission warranty requirements.  However,
the NPRM appears to preclude those engine manufacturers that certify a complete engine (e.g.,
both exhaust and evaporative requirements) from using an integrated emission compliance label.
EMA commented that engine manufacturers certifying a complete engine should be allowed to
label products using a single emission compliance label.  If the equipment manufacturer is the
party responsible for introducing the complete evaporative control system into commerce,  the
equipment manufacturer should be allowed to provide the emission compliance label.

       EMA commented on § 1054.135(c)(5) stating that the requirement to include engine
displacement on the label adds no value to either the customer or EPA and should be deleted
from the labeling requirements.

       EMA commented on § 1054.135(g) stating that the proposed language would preclude
engine manufacturers that certify a complete engine to both the exhaust and evaporative
requirements defined in 40 CFR Part 1060 from using a viable integrated label. EMA
commented that this section should be revised to read as follows: "Manufacturers that certify
compliance to both the exhaust and evaporative requirements of 40 CFR Part 1054 and 40 CFR
Part 1060 may meet the labeling requirements using a single label that provides all of the
required information from both parts."
       EMA commented on §1054.136 "How must I permanently label the equipment I
produce?" EMA commented that this section is redundant and should be deleted.
       Honda requested that EPA reconsider the entire proposed requirements for engine
labeling. Honda's evaluation of the proposed label and contents that would be required for many
                                      2-51

-------
                                                           Chapter 2: Small SI Engines
engines indicated that the label size and content would be significantly increased due to
declarative statements and other label information, with little or no added value. Honda believes
that a simplified label with the certifying organization identification or logo, engine
manufacturer identification (corporate name or trademark) and a single alpha-numeric designator
could fully signify engine regulatory compliance.
       Honda commented that the information on a certification label has extremely limited
value to anyone in the supply chain other than a U.S.  Customs or EPA Inspector attempting to
match the manufacturer and engine with a specific Certificate of Conformity and that engine's
certification information. Furthermore, they believe the emission label information is of limited
value to the engine purchaser, both individual and corporate, because they rely in their purchase
decisions on business-relevant information such as model and engine type or catalog number
which are typically stamped in the engine block or on another label. Fundamentally, the
emission label does not receive any level of attention by or provide any usefulness to buyers or
users, regardless of unit production volume.

       Honda recognized EPA's concern for counterfeiting  of labels, but they do not believe that
EPA's proposal will prevent counterfeiting. Honda also recognized EPA's desire to provide
distinction between uncertified and certified products. Nevertheless, they believe there is a much
better and effective approach to addressing these two needs than merely expanding the
information on a label. Specifically, Honda suggested that EPA work with industry to establish
revisions to the certification application that would provide data for an EPA database that could
be electronically accessed by those with a need to know (U.S. Custom's inspectors) and
correlated with information that is part of the engine itself, e.g., stamped engine identification
information or an engine identification number on the label.  Perhaps the month and year of
engine manufacture would be a necessary supplement if the  manufacturer does not maintain a
readily available database of serial number and corresponding date of manufacture. However,
the manufacturer name on the label may also be redundant since it is typically on the engine
itself and also coded into the engine family name.

       OPEI noted that EPA's regulatory language states the label must contain month and year
of manufacture with no allowance for variation (see 1060.13 5(b)(2) and 1054.13 5(c)(6)). OPEI
commented that the minimum requirement should be month and year. Production time intervals
less than a month should also be allowed, for example, week or day.  OPEI also requested that
the date of manufacture be allowed in a code on the label. (For example A06 means January
2006, B06 means February 2006.) OPEI stated that EPA currently allows for coded date of
manufacture and should reflect this in the regulatory language.

       OPEI commented that EPA has set precedence in the past for allowing for the deletion of
the specific model year on the label and replacing it with a term like "this product complies with
EPA Phase 2 standards" or "this product complies with EPA standards for 2002 and later."
Since Class III, IV, and V handheld products have no exhaust changes, OPEI requested that EPA
add language to §1054.135(c)(12) that will allow the use of standard language that will not need
pre-approval for EPA such as: "THIS ENGINE COMPLIES WITH U.S. EPA PHASE 3
REGULATIONS FOR ..." or "THIS ENGINE COMPLIES WITH U.S. EPA EXHAUST
REGULATIONS FOR 2010 AND LATER MODEL YEAR"
                                         2-52

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
Letters:
Commenter
EMA
Honda
OPEI
Document #
0691
0705
0675
Our Response:

       All the information that we proposed to require on emission labels relates fundamentally
to compliance with emission standards.  This information is useful in varying degrees to
consumers, equipment manufacturers, and EPA and Customs inspectors. We also note that
manufacturers have been successful so far in creating and applying labels with all the
information we require under the current regulations, without creating confusion or otherwise
thwarting our labeling goals. Nevertheless, we agree with the suggestion in the comments that
we  should pursue alternative means to make some of this information available.  In large part,
our interest in narrowing the required label content is based on the disproportionate amount of
time it takes to handle requests for variations from the regulatory specifications.  As a result, we
have gone through the effort to reduce the required label content to the minimum needed for the
label to ensure compliance, given our (and the manufacturers') current and projected abilities to
manage the additional information. By reducing the label content in this way, we believe we
have also reached a point at which we can disallow any variations from the specified label
content for these few pieces of information.  This will significantly streamline the preparation,
review, and approval of emission labels.

       We generally agree with EMA's assessment regarding the essential elements  of the
emission label.  The manufacturer's corporate name and the applicable family identification must
be included. Manufacturers may add their trademark, but this is not required.  The date of
manufacture must be included, unless it is stamped or engraved elsewhere on the engine. A
modified compliance statement must be included, as described below. We believe the label
should include two additional items. First, as described in Section 2.4.2, the label must identify
the  engine's useful  life. Second, if the family includes engines with differing displacement
values, the displacement of each engine should be identified on the label.  This would be the only
way to readily determine which standards apply to each engine since  the displacement
information embedded in the engine family name would not necessarily apply. If manufacturers
want to avoid separately identifying displacement information on the label in this situation, they
could simply certify the engines in different engine families.

       Further reducing content to include only a code for looking up all the information may be
possible in the future, but we believe the EMA comments represent a more realistic middle
ground for the foreseeable future.  A label with nothing but a  code for looking up relevant
information would prevent the label from having any value without being able to access the
database. We believe there will be times when owners, equipment manufacturers, and EPA and
Customs inspectors should be able to identify the basic engine and compliance information by a
simple visual inspection of the engine.
                                       2-53

-------
                                                             Chapter 2: Small SI Engines
       The label items included in the proposal but not in the final rule can be appropriately
included in the owner's manual. This includes the identification of the emission controls system,
tune-up specifications, information related to operation at high altitude, fuel and lubricant
specifications, limitations on engine use at rated or intermediate speeds (if applicable), and
limitations on engine use in wintertime equipment (if applicable).

       While the owner's manual is useful for identifying these additional items, this is of little
value to EPA or Customs inspectors or even to owners if they don't have or don't use the
owner's manual, as is commonly the case.  To address this concern, we may pursue system
improvements that would allow us to readily access the database that includes this information.
In this  scenario, an inspector with  a laptop or handheld device with Internet access would be able
to use the  engine family identification number to quickly look up all the highlighted information
that is relevant for a given engine. This would allow us to create a very accessible virtual label
without being constrained by space limitations.

       As described in Section 2.4.1, we agree that emission labels do not need to include the
applicable family emission limit. This is based on the alternative requirement to track changing
family emission limits by date of manufacture and serial number rather than the reasons
identified  by the commenters.

       See Section 1.3.2 for a discussion of build dates on labels,  compared with engine
manufacturers keeping records with this information. We agree that build dates should be based
on identifying the month and year at a  minimum.  We don't believe it is appropriate to use coded
information to identify the build date.  This is especially important given the discretion we are
allowing to create family codes for compliance with evaporative emission standards, as
described  in Section 4.6. Identifying the full month and year would be preferred (e.g., February
2009).  We would also find standard abbreviations acceptable, such as Feb 09 or 02/09.  We
intend to pursue regulatory amendments to clarify the format of build dates on engines or
emission labels, with the goal of adopting uniform specifications across all our programs.

       Fundamental to certifying engines under the Clean Air Act is the idea that the certificate
is valid for a given model year. Manufacturers must recertify all their engines for every new
model year.  In some cases a manufacturer may produce certified engines in a given year and not
renew certification for the following year. This is a case where the model year information
would be necessary to identify the compliance status  of the engines properly produced under a
valid certificate and to avoid improperly labeling for the engines produced when there was no
valid certificate.  We are also adding regulatory language to ensure that manufacturers properly
align their build dates and overall production periods with the dates defining the model year for
the particular engine family, as identified by the effective dates for the certificate. See Section
1.5.2 for further discussion of issues related to build dates and model years.

       The information related to  wintertime use and rated-speed/intermediate-speed operation
is mostly intended for equipment manufacturers to ensure that engines are installed in equipment
consistent with any applicable limits on the engines' certification.  We believe these items should
be included in the owner's manual for  completeness.  We also separately require that engine
manufacturers make clear in their installation instructions that equipment manufacturers install
                                          2-54

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


engines such that they remain in a certified configuration if there are any limits on the range of
applications covered by the certificate.

       We agree that an abbreviation for "Delegated Assembly" may be necessary. However we
believe the abbreviation should be no shorter than "DEL ASSY".  Such an abbreviation will
allow for continued recognition of the terms  for an informed reader/inspector, without resorting
to a two-letter code that could ultimately be overlooked or misunderstood. This is especially
important given the discretion we are allowing to create family codes for compliance with
evaporative emission standards, as described in Section 4.

       As described above, our primary motivation to reduce the label content as much as
possible is to  standardize labels and avoid requests for alternative wording and formatting.
Accordingly,  we do not believe it is necessary or appropriate to create a path for alternative
labeling for the label heading. Our understanding is that California has agreed to allow
manufacturers to meet their requirements with EPA's label heading, so this should not be an
issue under the Phase 3 program.

       We believe the proposed language in § 1054.135(g) clearly and explicitly allowed
integrated manufacturers to use a single label for meeting requirements for compliance with both
exhaust and evaporative standards.  We have nevertheless modified the wording to align with the
language suggested in the comment.

       We agree that the proposed §1054.136 does not add new requirements and is not
necessary for highlighting other requirements that apply for equipment manufacturers.  We have
removed this  section for the final rule.

   2.4.4  Maintenance

What Commenters Said:

       OPEI  noted that the maintenance provisions for handheld engines are outlined according
to §1054.145(c)(3). This paragraph allows the continued use of maintenance provisions outlined
in EPA Phase 2 for certification and deterioration factor (DF) engines. OPEI also noted this
provision has no expiration date.  OPEI further noted that the maintenance provisions outlined in
1054.125 do not apply to handheld engines.  OPEI requested that EPA add language to 1054.125
indicating this section does not apply to handheld engines.

       EMA  and OPEI commented that EPA should allow the following critical emission-
related maintenance practices during the determination of deterioration factors based on the
maintenance schedule provided to users: air filter, spark plug, valve lash adjustment, and two-
cycle exhaust port carbon removal.  These practices are well understood in the market place and
have been utilized for many years in order to ensure that engines perform their intended function
for their expected lifetime. EMA and OPEI also commented that EPA should explicitly
acknowledge  that the following maintenance practices are critical emission-related maintenance
that cannot be conducted during the determination of deterioration  factors: internal combustion
                                       2-55

-------
                                                             Chapter 2: Small SI Engines
chamber deposit removal, valve or valve seat reconditioning (including lapping, grinding, or
cutting), and replacement of exhaust aftertreatment components.

       EMA commented that air filter maintenance generally is prescribed by the engine
manufacturer for all customers.  Such maintenance instructions typically include provisions that
address adverse environmental  conditions that may require more frequent maintenance.
Depending on the air filter design, such maintenance could include cleaning or replacement.
Engine deterioration factor determination must be allowed to utilize the maintenance as
prescribed to the customers operating in a clean environment typically utilized for engine aging.

       EMA commented that if there is a concern that manufacturer defined maintenance
intervals are too close to prescribed emission testing points, the final rule should require both
pre- and post-maintenance emission tests on a case-by-case basis. For example, CARB requires
pre- and post-maintenance testing if the emission test point is within 10 hours of the prescribed
maintenance.

       EMA commented on §1054.125(a) "What maintenance instructions must I give to
buyers?" EMA commented that the requirement to demonstrate that scheduled maintenance is
reasonably likely to be performed is impractical. Small SI engine maintenance is typically done
by either the owner or an independent dealer. For the individual home owner, maintenance
intervals are typically dictated by seasonal time and use patterns.  However, the same engine
utilized by  a semi-commercial owner/operator may be  serviced routinely on a use basis. Typical
maintenance not covered by defect warranty that involves cleaning (such as air filters) or
adjustment (such as valve clearance) do not generate any documentation available to the engine
manufacturer.  Generically available items (such as spark plugs) are impractical for engine
manufacturers to document due to the sheer number of suppliers and retail outlets selling such
merchandise. EMA believes a limited and explicit list of acceptable emission-related
maintenance must be identified in the final rule, along  with a provision that allows engine
manufacturers to demonstrate why additional critical emission related maintenance  not specified
in the rule should be allowed. EMA commented that allowable critical emission-related
maintenance during service accumulation and emissions durability determination should include
air filter cleaning/changes, valve lash adjustment and spark plug changes.  The frequency of this
maintenance must be consistent with the engine operator's manual.  EMA  commented that
internal engine maintenance, such as decarboning of the engine combustion chamber, re-seating
of the valves, or other maintenance should explicitly be included in §1054.125(a)(2).
       EMA commented that the parts identified in §1054.125(d) "What maintenance
instructions must I give to buyers?" must be revised in order to agree with the proposed revisions
to §1054.125(a). Further, the second sentence should be revised to read as follows: "Noncritical
emission-related maintenance generally includes re-seating valves, removing combustion
chamber deposits, or any other maintenance related to emission-related parts as specified in 40
CFRPart 1068, Appendix I."
                                          2-56

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       EMA commented on §1054.125(e) "What maintenance instructions must I give to
buyers?" EMA commented that based on the proposed language in §1054.125(a), valve lash
should be removed from the list of potential non-emission related maintenance.
       EMA commented on the definition of critical emission-related components set forth in
§1054.801. EMA commented that the proposed definition does not include air filters or spark
plugs.  They noted that such parts are included in the definition of critical emission related parts
elsewhere in the proposal.

       Kohler commented that it has concerns with the current maintenance allowed during DF
testing in Part 90 and the wording in the proposed regulation §1054.125(a). Kohler commented
that normal maintenance should be allowed to be required in the owner's manual without
performing surveys etc. Any maintenance such as changing sparkplugs, air filters, and oil are all
normal and accepted by industry and should not require any special survey or demonstration on
the part of the manufacturer to be  allowed to include them as a requirement in the manual.

       Kohler noted that §1054.125(d) states that you cannot change an air filter or sparkplug
during service accumulation. Kohler commented that this statement needs to be changed to "you
cannot change an air filter or sparkplug during service accumulation for DF testing at intervals
different than that specified in the owners manual."

       CARB noted that EPA proposed to allow emission-related maintenance  during DF testing
if "60 to 80 percent of in-use engines get the specified maintenance at the recommended
interval."  As noted in the preamble, the small spark-ignition engines are predominantly operated
by homeowners and experience widely varying service practices. To ensure that the DFs do
actually represent in-use engines, it is crucial that maintenance that is not likely to be performed
in-use is not allowed for test engines.  To strike a balance, CARB recommended alignment with
other maintenance-related provisions that were adopted recently for the on-road heavy-duty
category requiring an 80 percent survey and other provisions.

Letters:
Commenter
OPEI
CARB
Kohler
EMA
Document #
0675
0682
0703
0691
Our Response:

       We agree that the maintenance provisions of §1054.125 do not yet apply to handheld
engines (as specified in the proposed §1054.145(c)(3)).  This is necessary because we are not
changing the stringency of the exhaust emission standards for handheld engines. Changing the
allowable maintenance during service accumulation for certification could affect emissions in a
way that would effectively change the emission standards for those engines. We expect to apply
the provisions of §1054.125 to handheld engines without modification when we adopt the next
                                       2-57

-------
                                                            Chapter 2: Small SI Engines
phase of standards for those engines. We therefore agree that it is appropriate to add a note to
§1054.125 to clarify that the maintenance provisions of that section do not apply to Phase 3
handheld engines.

      We also agree that the regulations should clearly disallow removal of combustion
chamber deposits, reconditioning of valves and valve seats, or replacing aftertreatment
components. These would rarely be performed as normal maintenance practices by owners so
they should also not occur during service accumulation for certification.  We note too that we
have no reason to believe that carbon removal from exhaust ports on two-stroke engines can be
considered normal maintenance, so we do not believe that would be an appropriate maintenance
step during service accumulation. We expect that all two-stroke engines certified to Phase 3
standards will be handheld engines.  As described above, the provisions of §1054.125 do not
apply to handheld engines, so we will revisit the question of maintenance for these engines when
we adopt the next phase of standards.  We do not expect to allow carbon removal from exhaust
ports during  service accumulation unless there is clear evidence demonstrating that this
maintenance is typical for in-use engines.

      It is clear that some owners clean or replace air filters and spark plugs on the schedule
prescribed in the owners manual. This would be the  case for fastidious homeowners wanting to
make their equipment last as long as possible or commercial owners interested in reducing the
costs associated with repairing or replacing  aged equipment. We remain unconvinced that in-use
maintenance related to air filters and spark plugs is so prevalent that manufacturers should
perform these maintenance steps during  service accumulation. There are surely many owners
who, perhaps in spite of best intentions, fail to invest the time, effort,  and expense of preventive
maintenance. There is clearly some tendency to treat Small SI engines and equipment as
disposable items, running with minimal maintenance until a problem  surfaces, then evaluating
whether to make a repair or just replace the  equipment. Especially with low price-point
consumer products, repair costs (and even some preventive maintenance costs) would be high
enough that many owners would minimize maintenance and repairs and opt instead to purchase a
replacement  model after a few years. Even  for commercial operations, Small SI equipment
many times would represent a small part of a much larger operation.  As such, companies
operating these engines would in many cases not make it a priority to coordinate a regular
schedule of preventive maintenance. For both homeowners and commercial users, we believe
the likelihood of taking preventive maintenance steps on the prescribed schedule falls
dramatically after the first year of service (or for second owners).  Performing a survey to
establish current maintenance practices would be very helpful, but we understand the constraints
on getting this information described in the  comments.

      Limiting maintenance during service accumulation for certification to align with the
prevailing in-use practice is important to avoid a situation where manufacturers are able to
achieve the necessary level of emission control in the laboratory while in-use engines are
emitting at higher levels because these same maintenance steps are not being done.  To the extent
that maintenance might not be performed in the field, manufacturers should have the incentive to
design their engines such that they do not depend on  this maintenance to comply with emission
standards. For example,  as described in  the proposal, we are concerned that air filters may
become  coated with oil mist on the downstream side.  Intake systems can be designed to prevent
                                          2-58

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


this by carefully designing the pressure dynamics of the intake system and the venting of
crankcase gases (and oil mist) into the intake system to prevent the entrained oil from reaching
the back side of the air filter.  In contrast, if we allow routine air filter changes as prescribed in
the owners manual, there is no need to improve these designs, even though the problem would
occur with  any in-use engines that do not get the scheduled filter changes.

       Having said that, we also note that our testing to establish the feasibility of the proposed
standards involved a rigorous effort to perform maintenance as prescribed in the respective
owners manuals, which generally involved air filter maintenance every 25 hours and spark plug
changes every 100 hours. Based on this experience, we don't believe we should entirely
disallow these maintenance steps for certification for demonstrating compliance with the Phase 3
standards.  We therefore believe it would be appropriate to allow manufacturers to clean or
change air filters and spark plugs, as long as manufacturers perform emission measurements
before and  after these maintenance steps. It would be best to perform testing after each
maintenance step; however, we would find it acceptable if manufacturers tested engines before
and after maintenance after every other air filter change. Manufacturers would use the average
of these two results for calculating deterioration factors. However, every measured test point
would need to be under the emission standard to be considered in compliance. This approach
allows for continued performance of these maintenance steps, consistent with our feasibility
testing, but properly identifies the effect on emissions.

       Most Class I engines are certified with a useful  life of 125 hours. Since manufacturers do
durability testing halfway through the useful life, this would be a normal point of replacing the
air filter. If manufacturers specify filter replacements every 25 hours, this would involve only a
small adjustment to fit with the planned testing. If manufacturers specify filter replacements
every 25 hours, they would need measure emissions before and after changing the air filter after
the second  filter change at 50 hours, or they could opt for a 30-hour filter change interval and
simply test at the scheduled midpoint for service accumulation.

       Laboratories where  service accumulation occurs generally have very little dust or
airborne debris that is common in the in-use environment.  We believe it is well within reach for
manufacturers to design their engines for extended operation without needing cleaning or
replacement of air filters. We believe this approach properly balances the manufacturers'
interest of performing maintenance during certification with our interest of documenting the
emission effects of this maintenance and maintaining the incentive for manufacturers to design
their engines to be dependent on maintenance as little as possible.

       Some Class I engines and all Class II engines are certified with a useful life of 250 hours
or longer. Testing these engines at the midpoint of their service accumulation involves a
correspondingly longer period. At the extreme, a 1000-hour useful life would involve testing
after 500 hours  of operation. To avoid additional test points, manufacturers would need to
design their engines to meet standards without cleaning or changing air filters for 250 hours or
spark plugs for  500 hours. While this involves a greater challenge, we think it is even more
achievable  for these engines where the reduced price sensitivity does not impose such a
challenging constraint in properly designing and manufacturing these engines. We believe these
longer useful-life  engines should be capable of operating on a controlled test fuel in a controlled
                                       2-59

-------
                                                            Chapter 2: Small SI Engines
environment for 250 hours without servicing air filters and for 500 hours without replacing spark
plugs.  However, as described above, we believe it is appropriate to allow for more frequent
service as long as the manufacturer performs emission tests before and after the maintenance to
document the effect on emissions.

       In pursuing more stringent emission standards in the future, we intend to more carefully
demonstrate the feasibility of achieving effective emission control over the full useful life with
maintenance intervals that more appropriately reflect any reduced level of service that may be
typical of the in-use experience for Small SI engines.  We would then be able to set more careful
limits on the maintenance that manufacturers may perform during service accumulation such that
certified engines will not depend on maintenance that may not be occurring with in-use engines.

       We have modified §1054.125 and §1054.801 to include air filters and spark plugs as
critical emission-related maintenance.

       Consistent with the proposal and all our other programs, we believe that adjusting valve
lash is not emission-related maintenance. Including valve-lash adjustments in §1054.125(e)
allows manufacturers to perform this maintenance during service accumulation at the least
frequent interval specified in the owners manual. This approach addresses the manufacturers'
interest in performing this maintenance on their recommended schedule.

       The first four paragraphs of §90.118 were adopted as part of the initial phase  of
standards, in which there was no service accumulation beyond engine stabilization. When part
90 was modified for the Phase 2 standards, there were no changes in the regulation to add
specific requirements or prohibitions related to maintenance during the service accumulation
period between stabilization and the end of the useful life. As such, we have concluded that only
oil and filter changes may be done before stabilization is complete, and manufacturers may
follow the scheduled maintenance specified in the owner's manual for the rest of the  service-
accumulation period.
   2.4.5  Deterioration factors/bench aging

What Commenters Said:

       OPEI and EMA commented that EPA should allow for the future development and use of
an aftertreatment bench aging procedure.  However, due to the complexity of such development,
the limitations and appropriateness of any procedure must adequately be assessed.

       OPEI commented that if a manufacturer can show that due to field-testing, the bench DF
cycle is too aggressive, EPA may approve an alternative test cycle based on data the
manufacturer provides.

       OPEI noted per CARB requirements that the calculation of a DF must involve at least
three test points (zero/midpoint and end of test). If a maintenance interval is scheduled at a test
point, the emission test should be run both before and after the maintenance. The emission test
                                          2-60

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


results should then be averaged for the value to be used in the calculation.  OPEI commented that
EPA should specify the same requirement in §1054.245 to avoid confusion.

      CARB commented that operating and testing the complete engine is necessary to get
accurate deterioration factors (DF). They noted that some manufacturers are using bench-aging
of components, including catalysts, to identify the worst-case scenario amongst
models/components. Subsequently, DFs are developed on the worst-case model/configuration
using full service accumulation on a dynamometer or in-use. CARB commented that bench-
aging of components and other alternative procedures should be allowed only  if manufacturers
provide adequate correlation data between their aging procedure and normal service
accumulation. Regarding assigned DFs, CARB commented that these should  be limited to just
California small-volume manufacturers (less than 500 total units per year). Other manufacturers
are required by California regulations to develop their own DFs so EPA's use  of those same DFs
would not impose any burden on manufacturers.

      ECO noted that EPA proposed an allowance for small volume engine families to utilize
assigned deterioration factors and requested input on the use of assigned DFs for small volume
engine families. ECO commented that this provision is necessary to allow flexibility for small
volume engine families.

Letters:
Commenter
OPEI
CARB
EMA
ECO
Document #
0675
0682
0691
0712
Our Response:

       We understand that a bench aging procedure has the potential to provide effective
deterioration factors at a substantially lower cost compared with aging engines with complete
systems on an engine dynamometer. As noted in the proposal and reiterated in the comments
however, we would want to be very sure that a specific bench aging procedure would adequately
represent aging from complete in-use engines. A fundamental factor in evaluating the
appropriateness of any bench-aging procedure is the extent to which it simulates representative
exhaust gas composition and other in-use operating parameters.  Any bench-aging procedure
would therefore need to take into account a wide range of variables to provide an adequate
simulation.

       We agree that the regulation should be changed to require testing at the midpoint of
service accumulation.  This provides additional information and aligns  with the requirements
already in place in California.  See Section 2.4.4 for a discussion of issues related to maintenance
during service accumulation.

       We continue to believe it is appropriate to include a provision for assigned deterioration
factors for small-volume engine families, even if the certifying company is not a small business.
                                      2-61

-------
                                                            Chapter 2: Small SI Engines
There may be several cases where the manufacturer produces only engines for equipment that is
preempted from California regulations, or that is not sold in California at all. We agree that it is
not helpful to allow for assigned deterioration factors where the engine manufacturer will have to
develop its own deterioration factor for the same engine family in California. However, we
would not want to disallow the use of assigned deterioration factors for those small-volume
engine families where the manufacturer does not need to do service accumulation to establish a
deterioration factor for California.

   2.4.6  Warranty

What Commenters Said:

      EMA commented that engine manufacturers must have the ability to shorten emission
warranty periods for engines that accumulate hours at a high rate such that they exceed 50% of
their specified emission durability period prior to the expiration of the prescribed emission
warranty period.

      EMA commented that the proposed emission related warranty parts list requirements
duplicate the information provided in 40 CFR Part 1068, Appendix I. They recommended that
the emission related parts list be inclusive of the emission related components identified in the
certification application, which also references  40 CFR Part 1068, Appendix I.

      EMA commented that for engines certified using aftertreatment or intake systems
supplied by the equipment manufacturer under the delegated assembly provisions defined in
§1054.610, that the warranty requirements be transferred to the equipment manufacturer. Engine
manufacturers should be required to maintain a cross reference such that any customer request
for warranty associated with a component provided by the equipment manufacturer would be
referred to the appropriate equipment manufacturer.

      EMA commented on §1054.120 "What emission-related warranty requirements apply to
me?" EMA commented that the regulations should be revised in order to clarify to whom the
section applies. Specifically, they recommended that the section be revised to read as follows:
"The requirements of this section apply to the manufacturer that certifies compliance with the
exhaust  emission requirements of this part.  See 40 CFR Part 1060.120 for evaporative emission
warranty requirements."

      EMA noted that under §90.1103(a), the warranty period should begin on the date of sale
to the ultimate purchaser. Accordingly, EMA commented that the sixth sentence of
§1054.120(b) "What emission-related warranty requirements apply to me?" should be revised to
read as follows: "The warranty period begins on the date of sale to the ultimate purchaser."
They also commented that this section should provide an option for decreased warranty period in
order to provide a differentiation between consumer and commercial usage of non-handheld
products similar to what is provided for handheld equipment in §1054.120(b)(2).  Finally, EMA
commented that EPA should add the following language as §1054.120(b)(4): "Any end user that
purchases a Consumer Product and uses it Commercially will have a shorter warranty period."
                                          2-62

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       Honda commented that nonhandheld engines in commercial equipment should also
specifically have the same option granted for handheld seasonal equipment, to limit the warranty
time period based on the product's use even without an engine or equipment hour meter. This
suggestion is made in reference to §1054.107. Honda noted that it is in §1054.120(b)(2) where
EPA has made an allowance for the seasonal use of handheld equipment.  The regulations state,
"We may establish a shorter warranty period for handheld engines subject to severe service in
seasonal equipment if we determine that these engines are likely to operate for a number of hours
greater than the applicable useful life within 24 months. You must request this shorter warranty
period in your application for certification or in an earlier submission."  EPA has in this section
recognized that commercial equipment is very likely to be used  for many more hours in less
calendar time than would be expected for homeowner operated equipment.

Letters:
Commenter
Honda
EMA
Document #
0705
0691
Our Response:

       We believe §1054.120 appropriately defines the engine components that are subject to
the emission-related warranty. The commenter's suggestion for the warranty to cover only those
parts listed in the application for certification would allow manufacturers to avoid warranty
coverage for a given component simply by leaving it out of the description in the application.
The broader language included in §1054.120 is necessary to  ensure that components will be
covered even if manufacturers develop an emission control technology with components that
would not be  covered by the specific list given in Part 1068,  Appendix I.

       The certificate holder always bears the primary responsibility for ensuring that engines
have proper warranty coverage.  Certifying engine manufacturers may choose to cooperate with
equipment manufacturers in the interface with owners, but we would hold the certificate holder
responsible for compliance with warranty obligations.  We could also pursue recourse against
equipment manufacturers, importers, or retailers for having caused the violation if we are able to
establish that any of those parties did not take basic steps to ensure that there was an effective
plan for meeting warranty requirements.

       The provision for shorter warranty periods for handheld engines used seasonally in severe
service can work because the companies making the engines also install the engines in their own
equipment. They can therefore understand the range of expected operation in the field for their
certified engines. (We note, however, that no handheld manufacturer has requested this shorter
warranty for engines used in seasonal equipment.) This is generally not the case for
nonhandheld  engines. Even those manufacturers that also make equipment will sell many loose
engines from  the same engine family to other equipment manufacturers. It is therefore difficult
to conceive of an engine manufacturer being able to adequately demonstrate the seasonal or
severe-duty nature of the expected in-field operation. While this may occur for some engine
installations, there could be many other installations where equipment manufacturers and/or
owners simply want a more reliable engine for operation that is neither seasonal nor severe-duty.
                                       2-63

-------
                                                             Chapter 2: Small SI Engines
We note too the increasing likelihood that commercial engines will have electronic controls and
fuel injection.  While these will be simple systems, they will include the ability to clock engine
operating hours.  Since we allow for a shorter warranty period based on engine operating hours,
it would be unnecessary for manufacturers to have any special approval for a shorter warranty
period based on seasonal and severe-duty operation.

       We don't believe it is appropriate to specify that a shorter warranty period applies for
commercial use of products that are intended for consumer applications. Many types of
equipment are not clearly differentiated between consumer and commercial applications.
Similarly, a person's use of any given piece of equipment is many  times not easily
distinguishable between consumer and commercial applications. The suggested language could
therefore not be clearly applied to adjust warranty periods for these products. As described
above, we believe the best long-term approach is to anticipate that  many or most engines in
commercial service will have hour meters that will indicate an end to the warranty period based
on the engine's operating hours rather than counting months on the calendar.

       We agree that §1054.120 should more carefully state that the section applies to
manufacturers that certify with respect to exhaust  emissions, with part 1060 covering warranty
obligations with respect to evaporative emissions.  We also agree that the warranty period should
start at the point of sale rather than the date the engine is placed into service, consistent with the
prevailing practice for standard warranties on consumer products.  This avoids a situation where
owners could make unverifiable claims that they first placed the engine into service several
months  after making the purchase.
   2.4.7  Naming labs and ports for imported products

       See Section 1.3.1 for an analysis of the comments related to the requirements for
importing manufacturers to identify the ports where they import products and to name a
laboratory in the United States for testing their engines.
   2.4.8  Engine family criteria

What Commenters Said:

       OPEI commented on §1054.230(b) recommending that EPA should include in the list
that families with displacements within 15% can be grouped together. This has been proven
reliable and acceptable for EPA Phases 1 and 2 as well as CARB Tier I/II/III.

       EMA noted that engine manufacturers producing multi-fuel engines recognize that they
must evaluate the different fuel influences in order to determine the worst case configuration
associated with the compliance demonstration for any engine family.  EMA commented that it is
important that engine manufacturers be allowed to utilize their best engineering judgment in
order to determine which fuel  and resulting engine configuration represents the worst case
configuration for a given family and, therefore, be used for the certification data development
                                          2-64

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


process. For example, an engine family may include both propane and natural gas fuel options
for an engine model, but the engine manufacturer should be allowed to determine the worst case
configuration for certification testing using their best engineering judgment.

       EMA commented on §1054.230(b) "How do I select emission families?" EMA noted
that pursuant to 40 CFR Part 90.116(d)(5), engines of different displacements that are within
15% of the largest displacement may be included within the same engine family. While this
flexibility is implicit in the proposed rule, EMA requested that EPA include a statement in the
preamble clearly stating that the intent of the language is to allow  engine models of varying
displacements (such as specified in §90.116) to be combined into one family at the
manufacturer's option.
       EMA commented on §1054.230(f) "How do I select emission families?" EMA
commented that because Part 1054 does not identify the requirements associated with obtaining
an evaporative certificate of compliance, it is not appropriate for this  section to discuss
evaporative component selection. Because all of the evaporative requirements refer the engine
manufacturer to Part 1060, EMA commented that it is both redundant and confusing to include
evaporative requirements within the requirements controlling the exhaust certification process.
       Honda commented that the engine family determination criteria in the final rule should
state that engines with a 15% displacement difference (percentage based on largest engine) may
be in the same engine family if they have similar emission characteristics. Honda also
commented that if a manufacturer can demonstrate that engines with a larger displacement
difference also have similar emission characteristics, the manufacturer should also be able to get
approval for inclusion in the same family.

Letters:
Commenter
Honda
OPEI
EMA
Document #
0705
0675
0691
Our Response:

       Under part 90 we have approved the combination of engines within a single family if the
range in displacement is within 15 percent of the largest engine's displacement. The proposed
regulatory language for part 1054 specifies that engines must have the same "approximate bore
diameter of cylinders."  We have adopted this language broadly across most of our programs.
We are adopting the proposed regulatory language without modification. We believe this is the
best approach, giving a clear guideline but allowing enough discretion to be able to respond to
any particular situations that may arise.  We will continue to approve combined engine families
based on the 15-percent displacement threshold. This maintains a harmonized policy with
California and is generally consistent with the way we have implemented other EPA programs.
We may also decide in special circumstances that a different threshold should apply.
                                       2-65

-------
                                                             Chapter 2: Small SI Engines
       We agree that dual-fuel engines represent a special case for differentiating engine
families. Clearly one engine that can run on multiple fuels must be in a single engine family.
The approach EMA describes in which the engine manufacturer chooses the worst-case fuel for
certification testing is appropriate. We note, however, that an engine that fails to meet the
applicable emission standards when operating on any of the specified fuels is noncompliant. We
have revised the regulatory language to clarify that fuel type differentiates engine families,
except in the case of dual-fuel engines. We have also added a clarification to  §1054.235 to say
that we may require manufacturers to submit data using the fuel not yet included in testing, and
that such a test would be treated as if it were a  second engine rather than a replacement for the
original data.

       We agree that §1054.230 should reference part 1060 to clarify how to define emission
families with respect to evaporative emissions, rather than including that information directly.
The final regulations have been changed accordingly.

   2.4.9  Other certification issues

What Commenters Said:

       OPEI noted that EPA is asking manufacturers to report CO2 in §1054.205(p). OPEI
questioned why EPA was asking for the information and commented that if EPA wants CC>2
reported, then manufacturers should be provided with requirements on how it should be reported
(units,  calculation etc).

       OPEI commented that under paragraph 1054.640(c), if the manufacturer is responsible to
EPA, then paragraphs (a) and (b) are an unnecessary burden and should be deleted.

       EMA commented that EPA should clarify where in the certification application the
additional information required by §90.107(d)(l 1)-(15) should be included.

       EMA commented on §1054.130 "What installation instructions must I give to equipment
manufacturers?" EMA commented that installation instructions for equipment that is not subject
to the provisions of the delegated assembly requirements in §1054.610 should be limited to
features consistent with the requirement to assure that the engine is in its certified configuration.
EMA noted that these instructions are generally not explicit instructions, but rather a process
used by engine manufacturers to approve the use of their engine in any equipment according to
the engine manufacturer's requirements.  For example, exhaust back pressure or intake air
temperature rise may be specified to  assure emission compliance and also  expected performance.
Accordingly, EMA commented that this section should be substantially revised to read as
follows:
       "(a) If you sell an engine for someone else to install in a piece of equipment, make
       available the information required to  ensure that as installed the engine will be in its
       certified configuration.
       (b) If the engine does not include provisions to control evaporative emissions advise the
       equipment manufacturer to refer to 40 CFR Part 1060 for applicable requirements.
                                          2-66

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       (c) Provide information to the equipment manufacturer that if installation precludes
       visibility of the engine's emission control label that a duplicate label must be added to the
       equipment in a visible location."

       EMA commented on §1054.205(o)(l) "What must I include in my application?" EMA
commented that the reference to THC or THCE should be expanded to include NMHC as
required by §1054.103(c)(2).

       EMA commented on §1054.250(a) "What records must I keep and what reports must I
send to EPA?"  EMA commented that the requirement to submit volume reports within 30 days
is inconsistent with current EPA requirements, is not adequately defined, and is inappropriate.
EMA noted that existing reporting requirements provide manufacturers 45 calendar days for
reporting.  Accordingly, EMA commented that the reporting requirement should be revised to 45
calendar days.
       EMA commented on §1054.250(b)(4) "What records must I keep and what reports must I
send to EPA?"  EMA commented that it is not practical to require manufacturers to maintain
production volume records for each engine family by assembly plant. In many cases, there are
multiple steps in the assembly process that may be completed at different assembly plants
thereby making this information either meaningless or impractical to determine. EMA
recommended that this record retention requirement should be revised to require records
regarding the total production volume for each engine family.
       Kohler commented that a consistent test cycle between engine manufacturers is critical to
maintaining a level playing field.  This applies to both dynamometer emissions testing as well as
DF hour accumulation.  Kohler requested that in the Phase 3 regulation, EPA take action to
maintain a level playing field for all manufacturers by assigning alphanumeric designators to all
approved alternative test cycles and posting these to the EPA website.  This would include
alternate procedures for dynamometer testing as well as the approved cycles (speed/load/time)
for DF hour accumulation. Kohler had the following recommendations for specific language
modifications.
      §1054.501 (c) Alternate test procedures — EPA allows engine manufactures to request
      approval for the use of an alternate  test cycle  if they cannot run the test cycle specified in
      this part. If an engine manufacturer requests and receives approval these MUST be given
      an alphanumeric designation and posted on the EPA website and be available for anyone
      to use.

       Kohler noted that §90.104 (h)(2)(ii) currently states that engine manufacturers should . . .
       "Conduct such emission testing again following aging the engine. The aging procedure
       should be designed to allow the manufacturer to appropriately predict the in-use emission
       deterioration expected over the useful life of the engine, taking into account the type of
       wear and other deterioration mechanisms expected under typical consumer use which
       could affect emissions performance.  If more than one engine is tested, average the results
       and round to the same number of decimal places contained in the applicable standard,
       expressed to one additional significant figure".

Kohler noted that there is no specific aging procedure defined. Many manufacturers today,
including Kohler, use repetitive cycles of the 6-mode certification test cycle for aging the engine.
                                       2-67

-------
                                                             Chapter 2: Small SI Engines
However, there is no public information available that states what procedure is being used by
individual engine manufacturers. To maintain a level playing field Kohler requested that the
following wording be used in the regulation:
       "Conduct such emission testing again following aging of the engine. The aging procedure
       must accumulate service (age the engine) in a way that represents how you expect the
       engine to operate in use and be approved by EPA. EPA's approval will assign an
       identification code for the cycle to be utilized in the manufacturer's certification
       application(s) for all applicable engine families. Approved test cycles will be listed with
       their respective identification code on the Small Spark Ignition Certification website and
       available for any applicable engine family.  If more than one engine is tested, average the
       results and round to the same number of decimal places contained in the applicable
       standard, expressed to one additional significant figure."

Letters:
Commenter
OPEI
EMA
Kohler
Document #
0675
0691
0703
Our Response:

       We require manufacturers to submit emission results for CC>2 only where those
measurements are needed to determine emission levels of regulated pollutants. If this
information is routinely gathered as part of emission testing, there is a minimal reporting burden
for manufacturers.  We want to be able to access this information to help us assess the reported
results for the regulated pollutants.  We have revised the regulation to  clarify that these results
should be reported as brake-specific values (g/kW-hr).

       We believe that the branding provisions of §1054.640 include basic information
necessary for ensuring that equipment manufacturers will fulfill their warranty obligations. We
agree that we can omit the requirement for engine manufacturers to describe the specific
arrangements in their application for certification, but we believe it is necessary for the engine
manufacturer to formalize the arrangements in the  form of a contractual obligation, and it is quite
appropriate to inform us of all the equipment manufacturers with whom this relationship exists.

       The references to THC and  THCE are simply given as examples, so there is no need to
include NMHC as another example. However, it is not incorrect, so we have modified the
regulation accordingly.

       We agree with the commenter that installation instructions should be focused on ensuring
that engines are in their certified configuration after installation in equipment.  Our proposed
regulation included several specific details, such as referencing altitude specifications where
appropriate, clarifying information  related to evaporative emission controls, describing limits on
installations (such as being certified only for use in rated-speed applications),  and adding a note
that duplicate labels may be necessary.  The proposed provisions are well established in many of
our other programs and they include only as much  as we believe is necessary to achieve the
                                          2-68

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


commenters stated goal of ensuring that engines are in their certified configuration in the final
installation.

       We agree with manufacturers that submitting production reports within 45 days after the
end of the model year is sufficient.

       If it becomes clear that defective engines are limited to production processes or other
practices at a particular production facility, both we and the manufacturer would want to
understand how isolated the problem is.  This would apply both for applying a remedy and
assessing penalties, if appropriate. We believe manufacturers should be keeping these records as
a matter of course for business reasons, so we expect there is no additional burden to keep this
information.  If production is divided into multiple steps across multiple facilities, manufacturers
should still be able to identify the number of engines that were processed at each facility.

       EMA's concerns about certifying fuel lines under §90.127 are moot because we are
revising these requirements to apply to component manufacturers.

       We agree that EPA's process for approving certification and testing procedures should be
transparent.  However, we believe the best approach for accomplishing this is administratively
rather than by regulation.  We may develop a process consistent with Kohler's suggested
approach,  but we need to maintain the flexibility to develop and modify those processes based on
our continuing experiences rather than limiting ourselves to a specific approach in the regulation.
We look forward to working with manufacturers over time to continue to improve our processes
for evaluating such requests and communicating the results of this decision-making.

       2.5   Test procedures

    2.5.1  NHH duty cycle/governor
Comment
Response
EMA commented that the NPRM's requirement that
engines operate utilizing the engine's installed governor
for the idle mode is not appropriate for many engines. A
significant percentage of engines in the Small SI
category do not utilize the engine governor to control
speed at idle. Such engines utilize a fixed throttle
position, generally determined by an adjustment screw.
For engines that do not utilize the governor to control
idle speed, the test condition should represent the
expected in-use idle speed control condition rather than
the governor.	
We agree that the regulations should reflect the situation
in which no engines in the family have governors that
control idle speed. The definitions and testing
provisions in part 1054 and part 1065 specify that
engines without governors controlling idle speed should
be set at the idle speed declared by the manufacturer.
§ 1054.235(c)(4). EMA commented that it is impractical
to recalibrate an emission test engine within normal
production tolerances as described in this section.
§1054.235(a) requires the test engine to be selected
based on the identified criteria and to be "tested as they
will be produced". Artificial modification via
recalibration is an overly broad requirement that should
not be granted to EPA to use in its discretion. EMA
commented that this provision should be deleted.	
The proposed provision is limited to items that are not
considered adjustable parameters. As noted in the
definition of the term, this might include adjustments
that are not emission-related or that manufacturers ask
us to exclude. To the extent that production tolerances
allow for varying engine settings, these items should be
subject to calibration settings such that the testing
configuration represents the full range for in-use
engines. This provision is already in place under	
                                          2-69

-------
                                                                        Chapter 2: Small SI Engines
§1054.501(b)(l).  The reference to engines without
throttle control is confusing and inappropriate. As
properly defined elsewhere in the proposed rule, the
engines in this category are generally considered
constant speed engines for emission testing purposes.
Engines in this category can have a wide variety of
controls including: (i) no user control of speed, (ii) user
control of the maximum speed, and (iii) load sensitive
automatic idle speed. EMA commented that this section
should be revised to eliminate the first portion of the
second sentence that it reads as follows: "See 40 CFR
Part 1065.10 for instructions for using alternate
procedures if utilizing the procedure specified in
1054.505 would result in emissions that do not represent
in-use emissions."
We agree that the provisions in question are best
addressed elsewhere. We have removed the text in
questionfrom §1054.501(b)(l).  See §1054.650 for
provisions related to certifying engines without
governors or with variable-speed governors.
§1054.501(b)(3).  EMA commented that the proposal
disallowed correcting emissions for the effects of
humidity. EMA commented that this restriction is not
consistent with EPA's current requirements as set forth
in §90.419. Many laboratories do not have EPA's
ability to run at a controlled humidity. Accordingly,
EMA commented that the humidity correction factor for
NOx emissions calculated per §1065.670 should be
required for a valid emission test.	
We agree that the humidity correction in §1065.670 is
appropriate for Small SI engines.  We have revised the
regulations accordingly.
§1054.501(d). EMA commented that engine
manufacturers must be allowed to use good engineering
judgment in order to determine engine changes
associated with the prescribed emission test temperature.
The ambient emission test conditions are not
representative of in-use conditions for winter exclusive
products, but ambient test conditions cannot be achieved
in the test environment that equate to in-use conditions.
For example, winter exclusive engines cannot operate in
the prescribed emission test conditions without removal
or modification of air intake heating systems such that
intake air temperature during the emission test is
representative of intake air temperature when the engine
is operated in-use.  EMA recommended that the
following be added: Engines may be modified for
emission testing such that intake temperatures are
analogous to in-use conditions.	
We agree that manufacturers should be allowed to
remove intake air heaters when testing wintertime
engines at temperatures between 20 and 30°C and have
modified the regulations accordingly.  We have also
added a provision allowing manufacturers to test
wintertime engines at reduced ambient temperatures by
referencing the existing specifications for snowmobiles
in §1051.505. In addition, we are adding language to
§1054.501 to say that non-wintertime engines should be
tested in a way that properly simulates in-use intake air
temperatures. We want to avoid a situation where
manufacturers cool the intake air after it has warmed up
from exposure to engine heating.  This is clearly not
appropriate since that type of cooling does not occur
during in-use operation.
§ 1054.505(a)(l).  EMA commented that the reference to
40 CFR Part 1065.514 must clarify that these engines
are considered constant speed engines pursuant to
§1054.505(b) and therefore only torque statistics are
required.	
We agree that the reference to § 1065.514 should be
limited to torque-related measurements.  There are
certain modes where manufacturers must control speed
within certain bounds, but these are specified separately
in §1054.505.	
§1054.505(c)(2). EMA commented that if EPA does not
accept our proposed revisions to §1054.801 (see
comments in Section 2.5.3 below), this section must be
revised in order to delete the phrase "maximum test
torque" and replace it with the following language "full-
load torque value from §1054.505(d)(2)."
We agree that §1054.505(c)(2) should reference
§1054.505(d)(2) for the appropriate torque value, rather
than relying on maximum test torque as defined in
§1065.1001.
                                                  2-70

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
Letters:
Commenter
EMA
Document #
0691
   2.5.2  HH duty cycle/governor

What Commenters Said:

       OPEI commented that Max power, as defined in 1054.801 and in 1065.601, are in conflict.  OPEI
commented that EPA needs to make clear that the power from 1054 applies for handheld products during
the cert test.

Letters:
Commenter
OPEI
Document #
0675
Our Response:

       The regulations specify that maximum engine power is used only for determining
whether engines are subject to part 1054 requirements or not. Engines are tested based on the
procedures specified in part 1054, subpart F (including Appendix II), which clarify the load
settings for full-load operation during the emission test. There are no power definitions or
specifications in §1065.601, but the other places where there are power specifications in part
1065 (such as engine mapping in §1065.510) do not apply for handheld engines.
   2.5.3  Maximum test speed

What Commenters Said:

       OPEI agreed with the reasoning EPA presented for an improved basis in selecting an
appropriate wide-open throttle speed for emission testing. OPEI believes though that EPA may
be unnecessarily complicating the regulation with its current language. Handheld products such
as chain saws, trimmers, brushcutters, edger and hedge clippers (not inclusive) run at wide-open
throttle speeds in the field that may vary depending on application and load. For example a
chain saw may be used for debranching, bucking or felling. All which may have slightly
different loads and resulting speeds.  It can be said that these product types will always be
operating around the max power point but this could vary by several hundred rpm.

       Product that always operates at a fixed speed due to the load of the transmission device
(like a  power blower, pump or generator) should always be tested at the actual operating speed in
field conditions to get the best real world emission test results. OPEI suggested that the
definition for rated speed at wide-open-throttle for handheld products be revised as follows:  1)
Products that always operate at a fixed speed due to the natural load placed on the engine (such
as power blowers and pumps) should be tested at the real world operating speed, in the
                                       2-71

-------
                                                            Chapter 2: Small SI Engines
configuration intended for use by the manufacturer, (+/- 350 rpm). 2) For all other handheld
products (like clippers, trimmers chain saws, edgers etc), the emission test at wide-open throttle
should be performed at the point of peak engine power (+/- 350 rpm).

      EMA commented on two of the definitions set forth in §1054.801:  1) Maximum test
speed: The reference should not be to 40 CFR Part 1065.1001, but rather to 40 CFR Part
1054.505.  2) Maximum test torque:  The reference should not be 40 CFR Part 1065.1001, but
rather to 40 CFR Part 1054.505.

Letters:
Commenter
OPEI
EMA
Document #
0675
0691
Our Response:

       We believe ungoverned handheld engines should be tested at speeds representing the
most likely in-use operating speed. We agree with OPEFs suggested approach for identifying
the nominal test speed for engines based on whether or not they will be operating in a fixed-
speed application.

       We use the term "maximum test speed" in part 1054 only to describe how to test
governed handheld engines over the two-mode handheld duty cycle. We believe this broadly
applicable method from part 1065 is appropriate for testing these engines and are therefore
retaining the definition as proposed.

       We have revised the regulation to avoid using the term "maximum test torque," since the
meaning of this term from part 1065 does not apply to Small SI engines.

    2.5.4  Test fuel

What Commenters Said:

       OPEI, EMA and Briggs and Stratton commented that California Phase 2 Certification
fuel should be allowed with EPA approved adjustment factors for Phase 2 nonhandheld engines
as currently practiced for Phase 2 engines.  Specifically, OPEI commented that the statement in
§1054.501(b)(2) should be revised to reflect: "use commercially available fuel representative of
the fuel that in-use engines would use in the same environmental conditions as the test is
conducted. Use of CARB Phase 2 fuel is considered acceptable." EMA commented on
§1054.501(b)(2) "How do I run a valid emission test?" EPA should explicitly state that
California Phase 2 Certification fuel may be used with EPA approved adjustment factors as
currently practiced for Phase 2 engines. Briggs and Stratton added that the proposal does not
allow for the use of oxygenated fuel, which would include California Phase 2 Certification fuel.
Alternative test fuels should continue to be allowed. Requiring a change would impose a burden
on many engine manufacturers with no benefit.
                                         2-72

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       OPEFs and EMA's engine manufacturers proposed that appropriate correction factors be
developed if the EPA certification test fuel is changed to an oxygenated fuel. Currently EPA
allows an option for engine manufacturers to use California certification test fuel which is an
oxygenated fuel for exhaust emission testing with an appropriate adjustment factor for the
emission results reported. OPEFs engine manufacturers recommended that a standard
adjustment factor for 50 state correlation (CARB/EPA) be included in the final regulation. EMA
added that the emission adjustment factors should be defined either in the regulatory text or in
guidance such that all manufacturers can utilize the same approved adjustments.
       With regard to handheld engines, OPEI also supported alignment of EPA certification
fuel with the California certification fuel because the type of fuel may directly influence the
results of any testing and the ability of manufacturers to confirm that technologies evaluated are
in fact compliant with the proposed regulations. However, the means to achieve this alignment
needs to be flexible. OPEI proposed allowing the use of CARB certification fuel for handheld
engine exhaust emission testing without the need for correlation factors.  OPEI requested that
different approaches to the solution should be considered for 2-stroke and 4-stroke engines since
the emissions are different.

       OPEI commented that EPA should harmonize the fuel for exhaust and evaporative
emission testing because the same fuel can represent real world conditions in the field. EPA
should accept a ten percent ethanol blend fuel as the standard test fuel for engines without
changing the limits. If a manufacturer certifies with the ten percent ethanol blend fuel, OPEI
commented that EPA should use the  same fuel for any SEA or in-use testing conducted.

Letters:
Commenter
OPEI
Briggs and Stratton
EMA
Document #
0675
0657
0691
Our Response:

       We are requiring Phase 3 exhaust emission testing with a standard test fuel consistent
with the existing requirements under 40 CFR part 90 (see 40 CFR part 1065, subpart H).  The
existing regulatory specifications allow for no oxygenates in the test fuel. Because CARB
specifies a test fuel which contains the oxygenate MTBE (but also allows for the use of EPA's
test fuel), we understand that some engine manufacturers will have emissions data from engines
which meet EPA's Phase 3 standards based on testing to meet California's Tier 3  Small Off-
Road Engine requirements for 2007 and later model years.  In some cases this test data will be
based on California's oxygenated test fuel, although manufacturers have the option to certify
using a test fuel such as that specified by EPA in 40 CFR Part 90. To allow for a quicker
transition to  the new EPA standards, we will allow for use of this pre-existing exhaust emission
test data (based on California's oxygenated test fuel) for EPA certification purposes through the
2012 model  year (see §1054.145(k)). Manufacturers could also use the CARB test fuel for their
PFT testing, if they based their certification on that fuel. The use of the CARB data would be
subject to the provisions for carryover data for demonstrating compliance with the standards in
effect.  (The carryover provisions for Phase 3 are specified in §1054.235(d) of the regulations.)
                                       2-73

-------
                                                             Chapter 2: Small SI Engines
While we will allow use of this CARB data for certification through the 2012 model year, we
will use our test fuel without oxygenates for all confirmatory testing we perform for exhaust
emissions. We are limiting the timeframe for such a provision because we ultimately want the
exhaust emission test results to be on the EPA specified test fuel.

       After the 2012 model year, a manufacturer wanting to use the CARB test fuel for
certification purposes could request use of the  CARB test fuel under the provisions of 40 CFR
1065.701(b) which apply for alternate fuel specifications. As part of that request, the
manufacturer is required to show that use of the alternate test fuel will not affect the ability to
demonstrate compliance with all applicable emission standards.  We would expect this showing
should be straightforward for handheld engines, where we are not changing the exhaust emission
standards for Phase 3 and where many manufacturers are already using CARB test fuel and
should already be taking any emissions difference into account when certifying their engines.
For nonhandheld engines, where we are changing the exhaust standards for Phase 3, we would
expect to see emissions data showing the impact of the alternate fuel on emissions (compared to
EPA's standard test fuel) as part of the manufacturer's request to use an alternate fuel under 40
CFR 1065.701(b).  While we may allow use of alternate test fuels such as the CARB test fuel
after the 2012 model year, we will use our test fuel without oxygenates for all confirmatory
testing we perform for exhaust emissions.  Furthermore, because of the differences in engine
technologies, we do not believe it is appropriate for us to establish an "adjustment factor" for
CARB certification fuel or any other potential  alternate fuel. Each manufacturer would need to
determine the emissions impact of the alternate fuel for its specific engine designs.

       In the proposal we noted our concerns about testing with oxygenated fuels since this
could affect an engine's air-fuel ratio, which in turn could affect the engine's combustion and
emission characteristics. Because of the relatively recent dramatic increase in the use of ethanol
(another oxygenate) in the broad motor gasoline pool, we have reexamined our position (as
discussed below) and are adopting provisions that will allow manufacturers to use a 10 percent
ethanol blend for certification testing for exhaust emissions from nonhandheld engines, as an
alternative to the standard test fuel.  This option to use a 10 percent ethanol blend will begin with
the implementation date of the Phase 3 exhaust standards. The option would apply to
production-line testing as well if the manufacturer based their certification on the 10 percent
ethanol blend.  We are also committing to using a 10 percent ethanol blend for all confirmatory
testing we perform for exhaust emissions under the provisions described below.

       Ethanol has been blended into in-use gasoline for many years,  and until as recently as
2005, was used in less than one-third of the national gasoline pool. However, ethanol use has
been increasing in recent years and, under provisions of the Energy Independence and Security
Act of 2007, ethanol will be required in significantly greater quantities.  We project that
potentially 80 percent of the national gasoline pool will contain ethanol by 2010, making ethanol
blends up to 10 percent the de facto in-use fuel. As ethanol blends become the primary in-use
fuel, we believe it makes sense for manufacturers to optimize their engine designs with regard to
emissions, performance, and durability on such a fuel. We also believe manufacturers need to
know that any confirmatory testing we do on their engines will be performed on the same fuel
the manufacturer used for certification since the fuel can impact the ability to demonstrate
compliance with the emission standards.
                                          2-74

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       Limited data of nonhandheld engine emissions tested on 10 percent ethanol blends
suggests the HC emissions will decrease and NOx emissions will increase compared to
emissions from the same engine operated on current certification fuel without oxygenates.
Depending on the relative HC and NOx levels of the engines, these offsetting effects can result
in small increases or decreases in total HC+NOx emission levels.  Because the impact on
HC+NOx emissions can vary slightly from engine family to engine family, we do not want
manufacturers varying their certification fuel from one family to another to gain advantage with
regard to emissions certification.

       Therefore, if a manufacturer wishes to use a  10 percent ethanol blend for certification, we
are adopting provisions that require manufacturers to use the 10 percent ethanol blend for all of
their Phase 3 nonhandheld engines for a given engine class within three years of the Phase 3
standard taking effect (i.e., by the 2014 model year for Class I engines and by the 2013 model
year for Class II engines).  During the transition period, we will perform any confirmatory
testing on the 10 percent ethanol blend if that is the fuel used by the manufacturer for
certification.  At the end of the transition period, we will perform any confirmatory testing on the
10 percent ethanol blend if that is the fuel used by the manufacturer for certification, but only if
the manufacturer has certified all of their nonhandheld engines in that engine class on the 10
percent ethanol blend. If the manufacturer has not certified all of its engines in a given engine
class on the 10 percent ethanol blend, we could decide to test the engine on our current test fuel
without oxygenates.

       For handheld engines, where we do not have sufficient data on  the impact of ethanol
blends on emissions,  we are adopting a slightly different approach. Manufacturers will have the
option to use a 10 percent ethanol blend for certification beginning with the 2010 model year.
The option to use  a 10 percent ethanol blend would apply to PLT testing as well if the
manufacturer based their certification on the 10 percent ethanol blend.  While we will allow use
of a 10 percent ethanol blend for certification, we expect to use our test fuel without oxygenates
for all confirmatory testing for exhaust emissions. Therefore, an engine manufacturer will want
to consider the impacts of ethanol on emissions in evaluating the compliance margin for the
standard, or in setting the PEL for the engine family if it is participating in the ABT program.
We could decide at our own discretion to do exhaust emissions testing using a 10 percent ethanol
blend if the manufacturer certified on that fuel. It can be noted that both EPA and CARB are
currently running  test programs to look at the emission impacts of a 10 percent ethanol blend on
a range of small SI engines, including handheld engines. Based on the results of that test
program, we may  want to consider changes to the provisions allowing the use of a 10 percent
ethanol blend for certification and PLT testing for handheld engines. If the results of the
handheld engine testing show that emissions are comparable on both fuels, we would expect to
revise the provisions  for handheld engines and adopt similar requirements to those adopted for
nonhandheld engines as noted above.

       The test fuel specifications for the 10 percent ethanol blend are based on using the current
gasoline test fuel and adding fuel-grade  ethanol until the blended fuel contains 10 percent ethanol
by volume. It should be noted that this is the first time EPA regulations specify the use of an
ethanol test fuel for exhaust emissions testing for certification purposes. It is likely that EPA
will consider similar test fuel changes in the future for other vehicle and engine categories
including those addressed in this final rule. As part of those deliberations, it is possible that EPA
                                       2-75

-------
                                                            Chapter 2: Small SI Engines
could decide that the test fuel specifications for the ethanol blend should be different than those
adopted in this rule. Should that occur, EPA would need to consider whether changes to the test
fuel specifications adopted in this rule for the 10 percent ethanol blend are appropriate for
nonhandheld engine testing.

   2.5.5  1065 Issues for Small SI Engines

       Comments were received from industry and industry organizations on several issues
relating to the application of Part 1065 to Small SI engines.  This section describes issues that are
specific to Small SI engines.  See Section 1.4 for more general issues raised by commenters
related to engine testing under Part 1065.

What Commenters Said:

       In the proposal and in its administrative record, EPA has not clearly identified, much less
evaluated, all the potential impacts that could occur by replacing the Part 90 exhaust emission
test procedures for  small-spark-ignited engines with the generic Part 1065 exhaust emission test
procedures. The 1065 test procedures were developed for very different types of much larger
and more sophisticated engines.  OPEI stated that its members are committed to working with
EPA to fill the critical data gaps. However, there is no way EPA or industry could generate all
the needed information in the next few months, complete this evaluation, and make all the
needed improvements - before the final Phase 3 rulemaking must be issued. For these reasons
(which are discussed in more detail below), EPA should allow (on a permanent basis) small
engine exhaust testing at facilities that use equipment and procedures that are compliant with the
existing Part 90 equipment, procedures and calculations.

    1.  Equation Calculations: Kohler, B&S and OPEI stated that Part 1065 should not be
       implemented until correlation between Part 90 and Part 1065 (subsection G) calculations
       have been adequately demonstrated (documented,  correlates and validates). Before
       eliminating or making any changes to the well-established Part 90 test procedures, EPA
       should first conduct comparative testing, and identify and analyze all the impacts of
       shifting to Part 1065.  Industry has developed a database of information with part 90 over
       the past 10 years.  The proposed changes to calculate emissions on Part 1065 are
       confusing and jumbled. Simple spreadsheet calculation methods are now impossible and
       a program dedicated to an iterative solution is required.  There is no data in the record to
       show the proposed test method would yield the same results. If the correlations are not
       completed, the manufacturers of Small SI engines  must be allowed to continue to use the
       40 CFR Part 90 calculation methods.

       Specifically, OPEI stated that virtually all small SI engines operate richer than
       stoichiometry and the majority of testing is conducted using raw gas sampling methods.
       Consequently, the applicability of the equations for raw gas emission measurement for
       engines with air/fuel ratios richer than stoichiometric are critical. Part  1065  prescribes
       equations associated with the conversion of the raw analyzer measurements to the mass
       equivalent.  Much more measured data (like H2O, N2O, aldehydes) would be needed to
       prove the equivalence of the 1065 calculation on a theoretical (mass balance of O, C, N,


                                         2-76

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       H) and practical bases. Differences in exhaust sampling systems may affect these
       chemical reactions/equilibrium and thereby the exhaust gas composition and measured
       values. It is a well known effect that hydrocarbons are converted to CO at high
       temperatures (post catalyst). Probes, sampling location, temperatures and flow rates all
       have a potential effect to change the measured values within a raw gas measurement.

   2.  Test fuels: In addition, the test fuel specifications in Part 1065 are different than existing
       Part 90 fuel specifications which could also result in skewed, different reported
       emissions. Part 1065 does not include any specification that addresses the 2-stroke oil-
       grade and mixing ratio. (See 40 CFR S. 90.308). (OPEI)

       Test methods: Before eliminating Part 90 as an option for test equipment, EPA would
       first need to resolve numerous outstanding issues, make needed modifications, and
       document that Part 1065 requirements can be practically implemented with small
       engines.

   3.  Fuel flow meter issues: a) the accuracy prescribed by §1065.205 may be impossible to
       meet for small engine test cells. Current known instruments will nominally meet the 2%
       accuracy and 1% repeatability values specified  in Part 90, but may not meet the
       percentage of maximum or the percentage of point requirements specified in Table 1 of
       1065.205.  b) Linearity verification for fuel flow rates <1% (under §1065.307) are not
       feasible for small engines with low fuel flow rates,  c) Lastly, a concern expressed was
       that the wide open fuel flows of today's Part  1065 engines may reach as much as 50
       liter/hour whereas Small SI engines frequently do not exceed 0.5 liters/hour.  (Industyr
       representatives later indicated that these points were meant to raise the issue that the
       tolerances in these sections are not yet known to be achievable.)

   4.  The requirement to control torque as needed to  meet 40 CFR Part 1065.514 cycle-
       validation criteria may not be feasible for test modes with very low target set points.
       Currently, 40 CFR Part 90.410 includes a provision for Phase II testing that reads as
       follows:  "hold the specified load within the larger range provided by ±0.27 Nm (±0.2 Ib-
       ft), or ±ten (10) percent of point" for current Phase 2 engine testing.  EPA must include a
       similar provision applicable to the testing of engines with modes where torque set points
       result in impractical cycle validation, as prescribed by §1065.514. (EMA) The comments
       also stated a concern that the torque transducers called  for in Part 1065 today would
       measure up to hundreds of joules of torque whereas the transducer used for engines
       specified in Part 90 measure in the range of tenths of joules. The characteristic of these
       engines requires a transducer to handle high torque spikes yet, measure smaller torque
       ranges once the engine stabilizes.  (OPEI)

   5.  Part 1065 analyzers which are designed for larger engine emission measurements might
       not be practical or suitable for long-term emission measurements on small SI  engines - at
       least without substantial modifications. For example, the HC hang-up  specifications (2
       ppm) in  S. 1065.520 are impractical for the much higher HC emission concentrations
       measured on rich burn gasoline sampling raw gas concentrations.
                                       2-77

-------
                                                         Chapter 2: Small SI Engines
6.  The ambient conditions defined by 40 CFR Part 1065.520 are different than the
   conditions prescribed by §1054.115(c).  Specifically, the ambient pressure range
   specified in §1065.520 is a range from 80.0-103.325 kPa; and the pressure range
   specified in §1054.115(c) is 94.0-103.325 kPa. The Part 1065 ambient condition
   requirements should be clarified in order to provide that the general requirements
   prescribed in Part 1065 are pre-empted by the standard setting Part. (EMA)

7.  The NPRM does not allow exhaust emission test results to be adjusted in order to account
   for the effect of ambient humidity. However, NOx emissions test data is currently
   corrected for humidity pursuant to 40 CFR Part 90.419. Because many laboratories do
   not have the ability to run at controlled humidity (as EPA can), such corrections are often
   significant.  The final regulation must allow the correction of emission test results for
   humidity as currently prescribed by § 90.419 (and utilized by CARB). (EMA)

OTHER:

8.  Based on the limited information that EPA has provided, it is difficult for OPEI to
   comment on all the ramifications of this proposed change.  However, it appears that the
   Part 1065  test procedures could cause small engine manufacturers to spend hundreds of
   thousands of dollars on at least new calibrations and software with  no environmental
   benefits. The cost estimate of equipment upgrades will be as much as $500,000 per test
   cell with no real benefit to emissions. (OPEI)

9.  EPA has also not identified how a shift to the Part 1065 test procedures would impact
   small engine manufacturers in terms of replacing or modifying their existing Part 90-
   compliant test equipment and related software and calibrations.

10. Manufacturers noted that some manufacturers control engines  at idle by  setting the
   dynamometer to control engine speed and use operator demand to control torque (usually
   zero, but not always), while other manufacturers take the opposite approach.

11. Given numerous uncertainties, the application of Part 1065 could result in more stringent
   exhaust standards.  At least for handheld manufacturers, the Part 1065 test procedures
   could also unintentionally result in more stringent exhaust standards. To avoid these
   unintended consequences, EPA should allow small engine manufacturers to continue to
   rely on the Part 90 test procedures, which could simply be referenced in the new Part
   1065. (OPEI) In its Phase 3 proposal, EPA repeatedly indicated it would not change the
   stringency of the Phase 2, exhaust standards for Handheld (HH) products.  The Phase II
   exhaust standards for HH engines are exclusively based on data generated from Part 90
   test equipment.  EPA's proposal and  supporting administrative record do not evaluate
   whether, or to what extent, the application of the Part 1065 requirements and calculations
   could generate higher reported emissions from small engines (compared  to Part 90) -
   unintentionally resulting in more stringent standards than are supported by EPA's record.

12. The requirement to submit a written report explaining reasons  for invalidating any test
   and the need for EPA to authorize retesting is overly broad and requires clarification.
                                      2-78

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       There is no need for EPA to authorize common causes for clearly invalid tests, such as
       invalid pre- or post- span measurements, etc.  The requirement to submit the test result
       from an invalid test is acceptable provided EPA recognizes that in some cases the reason
       that the test is invalid will result in erroneous results that should not be used for any
       purpose. (EMA)

    13. OPEI stated some members have recently purchased Part 1065-compliant analyzers and
       equipment in other industry segments based on their reliance on EPA's proposal that they
       would be able (as an option) to start certifying products with this equipment immediately.
       OPEI supports EPA's proposed approach to allow (as an option) certification testing
       using Part  1065-compliant equipment.

    14. TIMING:  At this stage in the rulemaking, there is inadequate time and resources for
       EPA and the affected stakeholders (including test equipment suppliers) to gather the
       needed information and then to develop proposed tailored solutions and regulatory
       modification to address all the unresolved issues.  There is no way EPA could issue a
       final regulation in June 2008 that would address all these problems with the needed
       modifications for small engines.

    15.  Part 1065 Test Procedures would Create Discriminatory Trade Barriers

       The U.S. EPA proposed test equipment changes would contradict the agreed-to goals of
       standards-harmonization  and, in certain circumstances, could create  a barrier to trade.
       CARB  and the EU emissions regulations for small  engines  are based on the Part 90
       procedures. As indicated in CARB comments into EPA's  Phase III rulemaking, CARB
       may not accept  certification data from  Part  1065 test equipment for  small engines.
       CARB remains concerned that Part 1065 equipment will not generate the same results as
       Part 90 test  equipment for small engines.   In  a call  on January  23, 2008, CARB's
       certification office confirmed that CARB will not  genetically accept certification test data
       on small engines based on Part 1065 test results  because of the absence of any existing
       database that genetically documents  the  equivalency  of Part 1065 and Part 90 test
       procedures as applied to small engines.   Consequently, CARB will require individual
       manufacturers to  demonstrate complete and identical test equivalency on their proposed
       test equipment through a comprehensive CARB-approved test plan.  To our knowledge,
       no small engine manufacturer has made  such a  demonstration to CARB's satisfaction.
       We expect other jurisdictions, including the EU, to adopt a similar position and refuse to
       accept  Part  1065  test results  for small engine certification  in  the  absence  of  a
       comprehensive demonstration of equivalency.  Such demonstrations will be very difficult
       to  prove.  Even  after expending substantial resources to try and  establish such an
       equivalency,  it is uncertain whether individual manufacturers' test equipment will meet
       CARB's and the EU's requirements.
       Other major  countries, including China,  are  developing regulations that are primarily
       based on the EU regulations and the related Part 90 testing  procedures. EPA's proposed,
       unilateral changes to these test procedures would force at least European and Asian small
       engine manufacturers (that need to produce uniform products for the global market) to
       spend millions of dollars to purchase, install  and calibrate separate analyzers,  software,
                                      2-79

-------
                                                      Chapter 2: Small SI Engines
and instrumentation (and invest in additional personnel) to re-test dozens of different,
emission-compliant, engine families that are exported to the U.S. market.
A disproportionate cost burden would be born by smaller and  mid-size European and
Asian manufacturers that typically manufacture and  certify  small volume or "niche"
products for the U.S. market.  In fact, the higher per-unit costs of U.S. EPA certification-
testing could bar  small  volume  European  and Asian  producers  from being able to
compete and sell niche products in the U.S. market. For  example, smaller manufacturers
with only a few test  cells would likely incur at least $300,000 in additional costs in
modifying their test cells in order to test and certify (with EPA) their US products under
the Part 1065  procedures.   Assuming a 5-year amortization, this would  result (on
average) in  $73,000 in additional  testing  costs per year.  Many of the affected niche
product lines produced by European and Asian manufacturers consist of only 1,000 units
in U.S.  sales each year.  For such products, European and Asian manufacturers would
incur additional, amortized testing costs in the range of $73 per  unit for niche lawn and
garden product lines that typically sell for less than $300 per unit. Thus, the U.S. EPA's
proposed "Part 1065" test procedures could  create  discriminatory trade  barriers  that
unfairly discriminate  against the niche products and low-volume manufacturers  and
would require them to invest in expensive and redundant emission test equipment.
Euromot also stated that the changes in the test procedures as proposed by introducing
§1065 would generate a misalignment with present equipment and worldwide
harmonized procedures and would not generate additional value to the US customers.
Euromot therefore asks EPA to stay with the current part 90 test procedures.

On March 19, 2008 Euromot sent a followup letter to their comments on the NPRM and
stated that in the final Phase 3 rule, EPA should 1) continue to apply  harmonized Part 90
Test procedures to small spark-ignited engines; and 2) Commit to initiate a process to
develop Global Technical Regulation (GTR) with the coordinated participation of the EU
and other international stakeholders (including Euromot) to develop new test procedures
that are specifically tailored to the unique challenges of small spark ignition engines.

Stihl further emphasized their interest in cooperating with EPA in an effort to develop a
Global Technical Regulation.  They noted that Euromot's goal is a very precise test
method that  does not give flexibility towards "creative" test results, rather than leaving
specifications that are too general to serve as a practical instruction for companies that
may be inclined to cut corners in their testing efforts.
                                   2-80

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
Letters:
Commenter
OPEI
OPEI
Euromot
Euromot
Kohler
Briggs and Stratton
EMA
Stihl
Document #
0675
0752
0649
0766
0703
0657
0691
0784
Our Responses:

Before addressing the specific comments, it is important to note that since the time of the NPRM,
we have made changes to part 1065 in a separate rulemaking that also set new emission
standards for locomotives and marine diesel engines (73 FR 37096, June 30, 2008). As
described below, several of these changes address comments that we received on this rule.

1. Equation Calculations - correlation needed, else allow manufacturers to use Part 90.

Industry commented that correlation between Part 90 and Part 1065 calculations has not
adequately been demonstrated for these classes of engines. EMA, B&S, OPEI, Kohler stated
part 1065 not be implemented until it is shown 1065 and 90 are equivalent for raw gas
measurements.

In response to comments, we have conducted a thorough comparison of the part 90 and part 1065
calculations.1 Our initial analysis show  small but significant differences between the two
methods. Some of the initial differences are being eliminated through changes to the part 1065
equations as described below. Others that are the result of errors in the part 90 calculations are
not being eliminated. Calculations from the modified part 1065 equations and from the part 90
equations for handheld engines are presented below. The  table below shows the differences
between the Part 90 and Part 1065 data sets with Part 1065 calculations yielding lower emission
results for HC and overall HC+NOx. Existing data can be recalculated or adjusted to be
comparable to part 1065 results.
1 In January /February of 2008, EPA shared correlated data with the nonhandheld and handheld industries based on
industry submitted data.
                                       2-81

-------
                                                             Chapter 2: Small SI Engines
Percent difference between emissions calculated using part 90 versus part 1065 equations
(Negative values indicate that values calculated according to 1065 are lower than those calculated
according to 90.)

HC
NOx
CO
HH
Raw*
-2.51%
0.29%
0.23%
HH
CVS
-2.78%
-0.13%
0.28%
NHH
Test 1
-2.19%
0.41%
0.22%
NHH
Test 2**
-2.28%
0.32%
0.21%
* Errors were found in the currently published Part 90 calculations and are being corrected with this final
rulemaking.
** An error was found in the industry Kh calculation related to nonhandheld engines for Test #2; it was
corrected and comparison to 1065 was based on the corrected numbers.

       As part of the exercise to compare calculated emission results, we determined that it was
necessary to account for free hydrogen in the exhaust as part of the carbon balance.  This is
particularly important for engines that run rich of stoichiometry because of the greater
concentration of hydrogen formation with such engines. For engines that have already been
subject to testing under Part 1065 using the old equations, calculating based on a zero
concentration of hydrogen in the exhaust is a reasonable simplifying assumption. We have
modified the Part 1065 equations in this rule to reflect the need to account for hydrogen in the
exhaust for engines that run rich of stoichiometry. The hydrogen values can be calculated and
need not be measured by an analyzer.

   2.  Specifications for 2-stroke oil grade and mixing ratio will be considered.  Specifications
       for test fuels will be evaluated and considered.

   In response to comments, EPA is adding 2-stroke oil grade and mixing ratio specification to
part 1065 in subpart H.  The new language is being taken from §90.308(a)(l), which states that
the fuel/oil mixture ratio must be that which is recommended by the manufacturer for the 2-
stroke engines.

   3.  Fuel flow meter issues

       Regarding measurement of fuel flow rates, the equipment for measuring fuel flow rates
so precisely may not currently be in use by all in industry.  Currently part 90 states that test
points are to come within, 2% at non-idle and 5% at idle of the reading. According to the 1065
requirements for the calibration of the fuel flow meter, the engine manufacturers must test 10
points over the range of fuel measurement expected during the entire test.  The verification tests
then apply to this linear line and calibrations of the system are to be done.  For a handheld engine
the certification test uses only two test points during its test and may use 0-61b/hr for example.
The two modes are at WOT and idle and therefore the in between points are never used.  For a
nonhandheld engine that used 0-3 Ib/hr, a reading must be taken every 0.33 Ib/hr and there are 6
modes in the certification test.  Industry does not yet know if fuel flow measurement equipment
is available to read to this degree and does not see the  need for this precision for neither of these
test procedures are transient.
                                          2-82

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
       EPA would like to ensure linearity of the fuel flow meter within the range of testing.
This cannot be a simple two point verification, but needs to include enough points to insure
linearity between the maximum and minimum fuel flow rates.  This check only needs to be
performed yearly; therefore EPA would like the small handheld engine manufacturers to perform
the fuel flow meter linearity verification check as currently described in 1065.307.

       One possible solution is to use gravimetric technology as long as the linearity
specifications in 1065 are met.

                   Table 1:  Comparison of Requirements for Fuel Flows
Fuel Flows
Fuel flow meter
specs
Linear
Verification
Calibration and
verify
Frequency of
Calibration
Part 90
90.328: Measurement equipment
accuracy /calibration frequency table.
Table 2: Permissible deviation from
reading: fuel consumption: +1-2% at
non-idle
+1-5% idle
None . . .

Calibrate monthly or within one month
prior to the certification test.
Part 1065
(Recommendation) :
Table 1 of 1065.205
5 sec rise and fall time
IHz
Accuracy: 2% of pt/1.5% of max
Max repeatability: l%pt/.75%max
Noise .5%
Linear verification 1065.307
- 10 measurement points covering range of
test meas.
- least squares linear regression and the
linearity criteria specified in Table 1 of this
section.
1065.320
Measurement systems that require linearity
verifications ...
Torque and Fuel Flow:
Within 370 days before testing
   4.  Torque related issues

       As industry works with 1065 over the coming years, we expect to work with industry to
understand how to properly measure torque. In particular, we believe it is possible to use
equipment meeting the torque requirements for testing Small SI engines, even at the low torque
levels that are typical for these engines. We have modified the cycle-validation criteria for
torque as described in Section 2.5.7 to more carefully reflect the level of precision that is
appropriate for Small SI engines.

       EPA would like to ensure linearity of the torque meter within the range of testing. This
cannot be a simple two point verification, but needs to include enough points to insure linearity
between the maximum and minimum torque values.  This check only needs to be performed
yearly; therefore EPA would like the small handheld engine manufacturers to perform the torque
meter linearity verification check as currently described in §1065.307.
                                       2-83

-------
                                                            Chapter 2: Small SI Engines
                Table 2: Comparison of Requirements for Torque Transducers
Torque
Transducers
Calibration
procedures
Calibration
accuracy
Calibration
frequency
Part 90
90.305 Dynamometer
specifications and calibration
accuracy
... a minimum of three
calibration weights for each
range used is required. The
weights must be equally spaced
and traceable to within .5% of
NIST weights, (foreign
countries., use wtd to local gov
stds)
90.306 Dynamometer torque cell
calibration
Gives details on procedure
90.306 dynamometer torque cell
calibration
Meas torque must be within 2%
of calculated torque
90.328 Measurement equipment
accuracy/calibration frequency
table.
Torque: monthly or within one
month prior to the certification
test
Part 1065
§1065.307 Linearity verification.
(2) Engine torque. Use a series of calibration weights
and a calibration lever arm to simulate engine torque.
You may instead use the engine or dynamometer
itself to generate a nominal torque that is measured
by a reference load cell or proving ring in series with
the torque-measurement system. In this case use the
reference load cell measurement as the reference
value. Refer to § 1065.3 10 for a torque-calibration
procedure similar to the linearity verification in this
section.
1065.310: ... Apply at least six calibration-weight
combinations for each applicable
torque-measuring range, spacing the weight
quantities about equally over the range.
Table 1 of §1065.307-Measurement systems that
require linearity verifications..
specifications given for linearity
Table 1 of §1065.307-
Linearity for system (fuel flow rate and engine
torque) - every 370 days
1065.310 Torque calibration.
Calibrate all torque-measurement systems including
dynamometer torque measurement transducers and
systems upon initial installation and
after major maintenance. Use good engineering
judgment to repeat the calibration.
5.... The HC hang-up specifications (2 ppm) in S. 1065.520 are impractical for the much higher
HC emission concentrations measured on rich burn gasoline sampling raw gas concentrations.

       This language was changed in the locomotive/marine 2008 final rulemaking to address
this concern.

6.  The Part 1065 ambient condition requirements should be clarified in order to provide that the
general requirements prescribed in Part 1065 are pre-empted by the standard setting Part.
(EMA)

       Section 1065.5 states "The testing specifications in the standard-setting part may differ
from the specifications in this part.  In cases where it is not possible to comply with both the
standard-setting part and this part, you must comply with the specifications in the standard
setting part. The standard-setting part may also allow you to deviate from the procedures of this
part for other reasons." Thus the regulations are already clear in this respect.
                                          2-84

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


7. EPA needs to allow for correction of ambient humidity 90.419 (also used by CARB). Industry
does not have humidity controlled test cells.

       Parts  1045 and 1054 are being revised to allow the correction of NOx for humidity.
Equation 1065.670-1 calculates the correction.  The 2006 version of Part 90 also refers to the
same equation so there is no change in this calculation.

8. Based on the limited information that EPA has provided,  it is difficult for OPEI to comment on
all the ramifications of this proposed change. However, it appears that the Part 1065 test
procedures could cause small engine manufacturers to spend hundreds of thousands of dollars
on at least new calibrations and software with no environmental benefits.  The cost estimate of
equipment upgrades will be as much as $500,000 per test cell with no real benefit to emissions.
(OPEI)

       We disagree with the estimated cost and the supposition that upgrading to part 1065 will
have no benefits. As described in the RIA, we believe that a typical manufacturer will need to
spend much less than this to upgrade its test facilities to be part 1065 compliant. To the extent
that any manufacturer needs to spend more, it will be because they are currently using outdated
equipment that is not sufficiently accurate, precise, and/or reliable to demonstrate compliance
with EPA standards.  Clearly having more accurate and repeatable measurements is beneficial.

9. EPA has not identified how a shift to the Part 1065 test procedures would impact small
engine manufacturers in terms of replacing or modifying their existing Part 90-compliant test
equipment and related software and calibrations.

       Industry does not have to certify with part 1065 procedures until 2013 and therefore time
is available to meet with EPA on specific questions related to part 1065 compliance. With
proper planning, industry can plan out any changes over time.

10. Testing at idle
       We agree that manufacturers  should be able to choose whether to use the dynamometer or
operator demand to control speed and torque during idle operation.  We understand that in some
cases, once the engine is stable and the dynamometer controls are functioning, engines may be
tested in a configuration such that the engine operates at the specified speed or torque level
without adjustment.
11. Part 1065 could result in more stringent exhaust standards

       The nonhandheld industry provided EPA with emission test data from two engine tests
using raw emission measurement.  The handheld industry provided EPA with one raw emission
dataset and one CVS emission dataset. EPA verified industry's Part 90 calculations and then
used the data to calculate results using 1065 calculations. In each case, the numbers correlated
between Part 1065 and Part 90 within -2.3% of HC on nonhandheld test data and -2.78% HC on
the handheld raw test data. In each case the 1065 calculations yielded lower numbers for the
                                       2-85

-------
                                                            Chapter 2: Small SI Engines
pollutant of THC due to the use of Part 1065's molecular weight of hydrocarbon default number
rather than calculating it as was done in Part 90. The slight percentage decrease in effect yields
slightly more lenient exhaust standards for the four datasets used in this analysis.  Changes of up
to 0.41% for NOx and 0.21 to 0.28% for CO are slightly more stringent standards for these
pollutants in these examples.

The changes of <0.41% for NOx and <0.28% for CO on handheld and nonhandheld engines are
very small changes. The NOx levels are very small on hh engines (test data showed 0.4%-0.8%
of HC+NOx is NOx) and the emission standard for HC+NOx is either 50 g/kWhr or 72 g/kWhr,
for Classes III, IV and Class V respectively. Therefore there is only a very slight increase in
stringency in the standard.   The handheld industry  test results showed CO between 400  and 500
g/kWhr on the two sets of data for handheld engines.  A 0.28% difference would mean an
addition of approximately 1.12-1.4 g/kWhr. For nonhandheld engines, the overall change in
HC+NOx was -1.7% and -1.6% for the two tests with the decrease in HC and the increase in
NOx combined. For CO, test data shows the nonhandheld engines at 374 and 390 g/kWhr and
0.28% of these values are 1.05 and 1.09 respectively. This again is only a very slight increase.

12. The requirement to submit a written report explaining reasons for invalidating any test and
the need for EPA to authorize retesting is overly broad and requires clarification.  There is no
need for EPA to authorize common causes for clearly invalid tests, such as invalid pre- or post-
span measurements, etc. The requirement to submit the test result from an invalid test is
acceptable provided EPA recognizes that in some cases the reason that the test is invalid will
result in erroneous results that should not be used for any purpose. (EMA)

       We agree that preapproval to retest after an  invalid test is not necessary in most cases.
However, it is necessary for invalid test results to be reported along with an explanation  of why a
test was invalidated.  The revised regulations are also clear that we reserve the right to require
preapproval of using retest results in PLT calculations should we determine that a manufacturer
is inappropriately invalidating tests.

13. OPE1 supports EPA allowing immediate certification with part 1065 compliant equipment.

       Part 1065 allows for the early use of these procedures consistent with good engineering
judgment.

14. Inadequate time and resources for EPA and the affected stakeholders to gather needed
information and present tailored solutions and regulatory modification to address all unresolved
issues.
       The industry is not required to use part 1065 until 2013, which will allow over 4  years
from the time of the final rule to modify the procedures.   Also, the regulations in part 1065
include numerous provisions to provide manufacturers to use equivalent procedures. Thus we do
not anticipate any problems associated with the timing of this requirement.


75. Part 1065 Test Procedures would Create Discriminatory Trade Barriers
                                          2-86

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       In regards to the concern that CARB will not accept 1065 based certifications: CARB has
been informed of Part 1065 along the way and will be adopting 1065 test procedures for small
spark ignition engines.
       Regarding the European Union adoption of 1065 procedures for small SI engines: it is
likely that industry will begin proceedings for a GTR for small SI engines which will address
1065 requirements versus "may" options. The European community has already adopted 1065
for diesel engines.
Euromot stated that: "The changes in the test procedures as proposed by introducing §1065
would generate a misalignment with present equipment and worldwide harmonized procedures
and would not generate additional value to the  US customers. We therefore ask EPA to stay with
the current §90 test procedures. "
       EPA is moving to 1065 test procedures  and is planning to take steps to work with the
California Air Resources Board. EPA has also begun talks with industry representatives to
discuss any issues related to testing  small SI engines per part 1065. Part 90 is somewhat vague
in several areas relating to emission testing procedure.  Part 1065 provides guidance in those
areas.

Euromot requested that EPA commit to initiate a process to develop Global Technical
Regulation (GTR) with the coordinated participation of the EU and other international
stakeholders (including Euromot) to develop new test procedures that are specifically tailored to
the unique challenges of small spark-ignited engines.
       EPA will continue to interact with manufacturers on issues that arise in complying with
part 1065 as they work toward making any necessary changes to comply with the new test
procedures starting with the 2013 model year. Given that the test procedures in part 1065 have
been demonstrated to be complete and consistent with the existing procedures in part 90, EPA
believes it is not  necessary to initiate a Global Technical Regulation at this point. However, if
there is an international or other forum for exploring testing issues for Small SI engines, EPA
would expect to participate in that effort.

   2.5.6  Running loss simulation during exhaust test

What Commenters Said:

       OPEI commented that the influence of running loss vapor on exhaust emissions is
insignificant and should not be associated with  the exhaust emission testing requirement. It is
impractical for an engine manufacturer to exhaust emission test engines that are installed in a
wide variety of equipment that include a wide variety of fuel tank sizes and running loss vapor
generation characteristics. (EMA) The proposal's requirement that running loss controls be
included when conducting exhaust emission tests is not practical. Emission tests are conducted in
engine emission  dynamometer test cells that include fuel delivery systems and do not generally
include engine mounted fuel delivery systems. A given engine family may be utilized with a
large variety of fuel  tank configurations, some of which will be supplied by the engine
manufacturer and some of which will be supplied by OEM customers. In addition, the inclusion
of the running loss control system may significantly compromise the ability to comply with the
                                       2-87

-------
                                                            Chapter 2: Small SI Engines
requirements for running the exhaust emission test - i.e., measurement of fuel flow for raw gas
testing.  Accordingly, this requirement should be eliminated.

       EMA commented on §1054.501(b)(6) "How do I run a valid emission test?" This
requirement is not practical for emission testing. Exhaust emission tests are conducted in engine
emission dynamometer test cells that include fuel delivery systems and do not typically include
engine mounted fuel delivery systems. A given engine family may be utilized with a large
variety of fuel tank configurations, some of which will be supplied by the engine manufacturer
and some of which will be supplied by OEM customers. Accordingly, this section should only
include the first sentence,  and the remainder of the section should be deleted.

Letters:
Commenter
OPEI
EMA
Document #
0675
0691
Our Response:

       As described in Chapter 5 of the Regulatory Impact Analysis, we measured in-use fuel
temperatures and corresponding emission rates to quantify running loss emissions from various
types of equipment.  In some cases, measured temperature increases and emission rates were
small.  However, we noted that some equipment had fuel tanks positioned closer to the engine or
other heat sources such that they experienced significant fuel heating. One case involved a 24°C
temperature rise, which corresponded to a 69.3 g/hour emission rate. Total fuel consumption for
a Class I engine might be about 300 g/hour (220 g/kW-hr with a 3 kW engine operating at 50
percent load). In this case, the engine would be ingesting 25 percent more fuel than it was
designed for.  The engine's emission controls would clearly not be able to compensate for this
unmetered vapor load. Class II engines have higher fuel flow rates, but a similar assessment
shows that a 12 kW engine would be ingesting about 6 percent more fuel than it was designed
for. Even this smaller deviation would likely cause an engine without feedback controls to
exceed emission standards.

       Measuring emissions from an engine for which the onboard fuel tank supplies the fuel,
including any running loss vapors routed to the intake, is not difficult with dilute-sampling
equipment. We understand that this is much more difficult with raw sampling, and that
individual labs may have some safety-related or other restrictions that make it impractical to do
this testing. As a result, we are keeping the specification to include ingestion of actual or
simulated  running-loss vapors in the engine's intake during exhaust testing, but we are adding an
allowance for manufacturers to make an engineering evaluation to show that actual vapor loads
from in-use engines will  not cause the engine to exceed the emission standard (or FEL if
applicable).  This would  preserve the motivation for engine and equipment manufacturers to
minimize the heat load on fuel tanks and to account for any remaining effect in establishing their
compliance margins.  We would expect any EPA measurement of exhaust emissions to include
running-loss vapor loads (representative of in-use operation) as much as possible.  For engine-
mounted tanks, we would expect to simply duplicate (or retain) this configuration for laboratory

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


operation. For remote-mounted tanks, we may measure fuel temperatures from in-use equipment
to properly simulate the running-loss effects.

       In cases where the engine manufacturer also designs the configuration and placement of
the fuel tank, this exercise should be straightforward. We understand, however, that equipment
designs may include a wide variety of configurations that are not all within the engine
manufacturer's  control. In these cases, we would expect engine manufacturers to do
development testing with their engines to be able to understand the sensitivity and limits of their
engines' compliance with exhaust emission standards as a function of running-loss vapor loads.
For loose engine sales to equipment manufacturers that control fuel tank designs, we would
expect engine manufacturers to specify in their installation instructions some appropriate limits
on the extent of tank heating to prevent the engine from exceeding applicable emission
standards. For example, engine manufacturers could directly specify a maximum vapor load (in
grams per hour) for continuous operation in the final installation. The vapor load for a given
operating condition would vary depending on the  size of the tank. Engine manufacturers could
therefore alternatively specify a table of values for maximum fuel-tank temperature rise for fuel
tanks with a range of capacities. The specifications in these installation instructions would form
the basis of the  engine manufacturer's simulation  or analysis to demonstrate that the engine will
meet emission standards in the final installation. Engine manufacturers may need to select a
higher Family Emission Limit to include a sufficient compliance margin to take running-loss
effects into account.

   2.5.7  Cycle-validation criteria

What Commenters Said:

       EMA commented on §1065.514(f) Cycle-validation criteria. While statistical cycle
validation makes sense for transient test methods, it is an unnecessary encumbrance to steady-
state testing. The current method is to track minimum and maximum speeds and torques
observed during the sampling period; test mode acceptance requires the extreme deviations from
the desired set point to be less than a percentage of set point. If statistical steady state mode
validation is required, the test control system will  need to be revised, at considerable cost to the
manufacturer with no resulting environmental  benefit. Accordingly, statistical cycle validation
should not be required for Small SI engines.

       EMA commented that the requirement to control torque as needed to meet §1065.514 is
not feasible for  test modes with very low target set points.  Currently, pursuant to §90.410 the
torque control requirement for Phase 2 engine  testing is "hold the specified load within the larger
range provided  by ±0.27 Nm (±0.2 Ib-ft), or ±ten (10) percent of point". EMA commented that
this section must include a similar provision for testing  of engines where the required torque set
points cannot use the cycle validation criteria required by §1065.514.

       EMA commented further on cycle-validation criteria: While the prescribed cycle-
validation criteria and statistical cycle validation is viable for transient test methods, it is an
unnecessary encumbrance to steady state testing.  Currently, engine manufacturers track the
minimum and maximum speeds and torques observed during the sampling period, and the test
                                       2-89

-------
                                                             Chapter 2: Small SI Engines
mode acceptance criteria require the extreme deviations from the desired set point to be less than
a percentage of set point. No environmental benefit is achieved from the addition of statistical
cycle validation criteria for steady-state testing, and it raises serious concerns for engine
manufacturers.  Accordingly, EMA recommends that these requirements be waived for Small SI
engines.

       In response to a draft version of regulatory text suggesting updated cycle-validation
criteria for nonhandheld engines, EMA suggested that the language should be in part 1054, not in
part 1065, since it should apply specifically for Small SI engines.  EMA further suggested that
any change to what is required under §90.410 today would raise significant concerns.

Letters:
Commenter
EMA
EMA
Document #
0807
0691
Our Response:

       Specifying nominal test speeds and loads with no cycle-validation criteria is meaningless.
Without some definition of acceptable deviation from the reference values, it would be
impossible to invalidate a test no matter what speeds and loads the engine actually experienced.
Manufacturers are testing with cycle-validation criteria today.  These specifications provide a
useful starting point for setting appropriate specifications.

       The current requirements in part 90 specify that torque for Modes 1 through 3 stay within
5 percent of point. Torque for Modes 4 and 5 must stay within 0.27 Nm or within 10 percent of
point, whichever is larger. This allows for very sloppy testing, especially for small engines. For
an 85 cc engine with peak torque of 4.0 NM, the nominal torque setting for Modes 4 and 5 would
be 1.0 and 0.4 Nm, respectively.  Specifying these points as 0.4+0.27 Nm and 1.0+0.27 Nm
means that any torque value between 0.13 and 1.27 Nm would be  a valid test point except for the
narrow range of 0.67 to 0.73 Nm. This effectively allows the manufacturer to pick the most
favorable torque values for certification.  Seen another way, this could be utilized for EPA
testing as similar to not-to-exceed zones for specifying any test point around the nominal value.
This is clearly not the  intent of the specified parameters.

       We believe an  alternative approach is better for targeting the nominal torque values for
very small engines and will not increase the burden for running a valid test.  Specifically, we
believe there should be separate parameters to address a tolerance for the range of measured
values and a limit for the mean value over the sampling period.  Setting a tolerance specification
of+2 percent of point  or +0.27 Nm, whichever is greater is consistent with the part 90
specifications, allowing for an achievable range of values for high-power and low-power modes.
An additional specification to keep the mean torque value within of+1 percent of point or +0.12
Nm, whichever is greater,  ensures that the manufacturer targets a true nominal value even if
there is substantial fluctuation in torque values during the sampling period.  This prevents a
manufacturer from intentionally biasing torque values low or high to take advantage of the wide
tolerances that are necessary to accommodate the very low power  levels. Mean values are
                                          2-90

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


inherently much more stable than instantaneous values, so achieving the narrower range of
values for the calculated mean torque should also be very achievable with current engines and
with current test equipment.

        2.6   Production-line testing
                     Comment
                     Response
EMA and OPEI commented that emission tests are often
invalidated because one of the requirements specified in
the test procedures was not met. Such requirements
could be anything from a span check, a test condition
parameter out of range, or any number of criteria
required to conduct a valid test.  Accordingly, they
commented manufacturers should not be required to
explain the reasons and report the emissions results from
all tests that have been invalidated.  Reporting of all test
results would only be appropriate if EPA specifically
recognizes that data obtained from an invalid test can not
be utilized to determine  compliance. By default, the
new test should become the official test results. In EPA
Phase 1 and 2 as well as CARD, this degree of
authorization is not required.
We agree that manufacturers should not need to get EPA
approval before invalidating and repeating a test when a
problem arises. However, we continue to be concerned
that allowing manufacturers to omit reports of
invalidated tests could result in manufacturers finding a
way to invalidate a test based on the observation that the
engine has failed or will fail to meet emission standards.
We believe we can best address these competing
concerns by requiring manufacturers to document their
invalidated tests,  including the reason for invalidating
and any emission results that were recorded. The
manufacturer could include an explanation describing
why (or to what extent) the reported emission results
from the invalidated test do not reflect the engine's
actual performance. We believe the proposed regulatory
text in §1054.305(g) properly balances these concerns.
OPEI commented that EPA Phase 2 and CARB all
require that production line test reports be filed within
45 days of the end of the quarter instead of 30 days, as
proposed in § 1054.3145. EPA is now requesting a
different time period. OPEI requested keeping
harmonization with CARB (report due 45 calendar days
within end of test period).	
We agree that quarterly reports for production-line
testing should be due 45 days after the end of each
quarter, consistent with our approach under part 90.
EMA commented on §1054.301(f)  "When must I test
my production-line engines?" EMA commented that the
reference to 40 CFR 1068.27 is redundant and should be
deleted.
We agree that the reference to §1068.27 is not necessary
and have removed it from the regulation.
EMA commented on §1054.301(b) "When must I test
my production-line engines?" EMA noted that the
referenced section (§1054.32fs5) does not exist. EMA
believes that the correct reference is to §1054.325.
In regard to the reference to §1054.32fs5, we appreciate
the comment and have revised the reference as
recommended.
EMA commented on §1054.305(d) "How must I prepare
and test my production-line engines?"  EMA commented
that the requirement to adjust parameters must be clearly
limited to adjustable parameters as defined in
§ 1054.115(b).  In addition, EMA commented that
adjustment of the idle speed outside of the adjustable
range as defined in §1054.305(d)(l) is not appropriate.
Manufacturers  determine idle speed ranges and
tolerances.  Adjustments outside of the manufacturers
recommended tolerance are not appropriate.
The proposed provision related to adjusting idle speed
was derived from the current regulations at §90.508
where we describe adjustments needed to operate an
engine until it has reached stabilized emission levels.
The original specification may have been related to the
technology used for engines in that time frame.  In any
case, we are not aware of any need for making idle
adjustments as described in the proposal. This includes
a review of the testing we performed to establish the
feasibility  of the Phase 3  emission standards. We have
therefore removed this provision from the final rule.
ECO commented that EPA should allow small volume
engine manufacturers to utilize the use of alternative
testing methods (portable emissions analyzers) to
demonstrate in-use field testing compliance for
production units.	
We agree that the regulations should allow for simpler
measurement methods for production-line testing, as
described in Section 1.3.4.
                                             2-91

-------
                                                           Chapter 2: Small SI Engines
Letters:
Commenter
OPEI
EMA
ECO
Document #
0675
0691
0712
       2.7   Equipment-manufacturer flexibilities

   2.7.1  Duration and extent of allowances

What Commenters Said:

       OPEI commented that without the proposed equipment transition flexibility, the EPA
Phase 3 program cannot be implemented without causing substantial and unnecessary injury to
equipment manufacturers and to the market. Non-integrated OEMs producing outdoor power
equipment must be able to stagger their complex and iterative-development and product
evaluation process, for their most challenging catalyzed muffler configurations and for their
difficult fuel tank technologies (such as roto-molding).  OPEI commented that it supports the
allowance of 30% of one year's production for large OEMs. OPEI also supports the proposed 4-
year transition period of the 2011 through 2014 model years.

       OPEI commented that it strongly objects to EPA's suggestion in the preamble that the
proposed hardship relief measures in the Phase 3 regulation could somehow moot the
independent need for the equipment-transitional flexibility program. Both proposed elements are
critical to the industry and to the effective implementation of the final program.

       In response to EPA's request for comments on whether the transition program for
equipment manufacturers somehow moots the need for the proposed Delegated Assembly
Program (or visa versa), OPEI commented that both programs are necessary and the two
programs serve separate,  distinct purposes.  OPEI noted that EPA has failed to evaluate or
quantify (in its administrative record or in its SBREFA process) the substantial economic
damages that would result without either the proposed equipment flexibility or the Delegated
Assembly provisions. OPEI commented that EPA's other off-road emission regulations
explicitly recognize the separate and independent need for Delegated Assembly, equipment-
transition flexibilities, and hardship relief. Consequently, it would be arbitrary for EPA to
abruptly eliminate any  one of these flexibilities for small engines. (Also included in Section
2.8.1)

       EMA supported the proposed delegated assembly and equipment manufacturer flexibility
provisions included in the NPRM. EMA commented that EPA must incorporate both of these
programs into the final rule.  If EPA were to adopt only the delegated assembly program and not
the equipment manufacturer flexibility program (or visa versa), EMA commented that the
functionality of the adopted program would be significantly impaired by the absence of the other
program. (Also included in Section 2.8.1) The inclusion of aftertreatment systems into an
equipment manufacturers' exhaust system requires a much broader set of changes than just
                                         2-92

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


packaging the catalyst into an existing muffler. In many cases, the introduction of an exhaust
system that includes a catalyst will require a complete redesign of the exhaust system and/or the
equipment in order to provide the necessary space and heat management of the exhaust system.
This required redesign of the exhaust system/equipment is an enormous burden to the
manufacturer and will create a significant strain on its resources.  Accordingly, EMA believes
that the flexibility provisions are absolutely necessary and must be incorporated into the final
rule in order to ensure that manufacturers have the ability and time required to complete the
necessary redesign.

       CARB commented that it believes the proposed transition program for equipment
manufacturers is unnecessary since most equipment manufacturers are working together with
their engine manufacturers to meet California's Tier 3 standards.  The equipment manufacturers
are already working together to address concerns regarding lead-time, coordination, and other
aspects involved in meeting the standards. CARB also commented that the proposed eligibility
requirements for the TPEM  program (i.e., only those manufacturers that have primary
responsibility for designing  and manufacturing equipment and whose manufacturing procedures
include installing engines in the equipment are eligible) make it difficult to determine and
enforce which manufacturers would actually qualify for the program.

Letters:
Commenter
OPEI
EMA
CARB
Document #
0675
0691
0682
Our Response:

       In response to the comments that a TPEM program is not needed because of CARB's
Tier 3 program, EPA continues to believe a TPEM program is necessary for the manufacturers of
Class II equipment.  While CARB's Tier 3 standards took effect in the model year 2008, the
major engine manufacturers appear to be using ABT credits to certify to the standards since most
of the engines certified with CARB in the Class II category have FELs above the Tier 3 standard
of 8.0 g/kW-hr HC+NOx level. While it is not clear how long it will be before manufacturers
redesign their Class II engines for California, it is likely that the engine manufacturers will not
have a full set of engines redesigned until 2011 or later when EPA's Phase 3 standards take
effect.  Because equipment manufacturers may need to make changes to some equipment designs
to accommodate the redesigned Class II engines, EPA believes a TPEM program will help to
ensure the transition to the Phase 3 standards goes smoothly for equipment manufacturers.  EPA
believes the basic framework of the TPEM program which allows  manufacturers to use Phase 2
engines over a four year period on up to 30% of their average Class II sales is appropriate and we
are  finalizing those levels in the final rule, as proposed.

       EPA agrees with the comments that the TPEM program and hardship provisions are both
needed for the Phase 3 program. The hardship provisions are intended to help manufacturers that
are  facing economic hardship as a result of not being able to comply with the new standards.
The criteria for qualifying for hardship are set at a relatively high level,  which would likely be
                                      2-93

-------
                                                           Chapter 2: Small SI Engines
difficult for a manufacturer to demonstrate if they were having difficulty redesigning only a few
of their equipment models. The TPEM program allows an equipment manufacturer to deal with
the models which are difficult to redesign without having to demonstrate that the company would
experience hardship without the relief. Therefore, EPA agrees that both the TPEM program and
the hardship provisions are needed and is retaining both of them for the Phase 3 program.

      EPA agrees with the comments that the TPEM program and the delegated assembly
provisions are both needed for the Phase 3 program. The delegated assembly provisions allow
manufacturers to independently source their exhaust systems based on the catalyst specifications
determined by the engine manufacturer.  However, the delegated assembly provisions will not
ensure that an equipment manufacturer will be able to redesign all of their equipment models in
time to accommodate a Phase 3 engine. Therefore, EPA agrees that both the TPEM program and
the delegated assembly provisions are needed and is retaining them for the Phase 3 program.

      EPA disagrees with the comment that the criteria used to qualify manufacturers for the
TPEM program makes it difficult to determine who is eligible.  The purpose of the eligibility
criteria is to  ensure that only those companies that truly manufacture  equipment can participate
in the program.  We do not want companies that only import complete equipment or companies
that only install  engines into a pre-existing equipment chassis to be eligible for the program.
EPA has made this  clear in the regulations. If there is any question regarding a manufacturer's
qualifications, EPA can request information from the manufacturer to determine if they actually
are designing/manufacturing equipment and installing engines under  the authority granted in
Section 208  of the Clean Air Act.  Section 208 (which applies to nonroad engines under section
213(d) of the Clean Air Act) describes the information collection requirements for manufacturers
and states that manufacturers must provide information that EPA may reasonably require to
determine whether  manufacturers have acted in compliance with regulations. For this reason,
EPA is retaining the eligibility criteria for equipment manufacturers in the final regulations.
   2.7.2  Additional allowances for mid-sized companies

What Commenters Said:

       OPEI commented that it supports the proposed mechanism to allow small and mid-sized
OEMs to request up to a 70% production-based allowance (on a case-by-case basis).

       NESCAUM commented that they oppose the various provisions for small and medium
volume manufacturers of engines and equipment that extend the use of Phase 2 compliant land-
based SI engines for several years beyond the initial introduction of Phase 3 engines. However,
they would not oppose a program whereby small businesses may apply individually to EPA for
limited temporary relief from specific requirements due to economic hardship or other
circumstances beyond their control.
                                         2-94

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
Letters:
Commenter
OPEI
NESCAUM
Document #
0675
0641
Our Response:

       EPA is retaining the special TPEM provisions for small- and medium-sized companies in
the final rule.  EPA believes that such companies face a bigger challenge with regard to
equipment redesign than large companies because the small- and medium-sized companies tend
to have fewer resources (i.e., both staff and money) available to work on equipment redesign.
Therefore, EPA believes it is appropriate to offer more flexibility to these companies under the
TPEM program. For small-sized manufacturers (defined in the regulations as producing no more
than 5,000 pieces of nonhandheld equipment per year), the extra flexibility is automatic and
allows them to exempt a cumulative 200% over four years. For medium-sized companies (i.e.,
those producing between 5,000 and 50,000 units with Class II engines), the manufacturer can
request up to an additional 70% allowances over the four years, but must provide a variety of
information to EPA to justify its request.

       In regard to the comment on relying on hardship requests instead of additional TPEM
allowances for small- and medium-sized businesses, EPA does not believe that making a
hardship provision the primary means for obtaining additional allowances would be workable for
manufacturers or EPA.  As noted above, smaller companies have limited resources to allocate to
equipment redesign. Even though they may be small, many of these companies have a wide
range of equipment offerings. EPA would rather see these businesses working on the equipment
redesigns than pulling together information to request additional allowances from EPA. Plus, it
potentially would place additional significant burden on EPA to review hardship applications,
since there are over three hundred eligible small- and medium-sized equipment manufacturers.
Therefore, EPA is retaining the TPEM provisions for small- and medium-sized businesses as
proposed.

    2.7.3   Reporting and recordkeeping

What Commenters Said:

       OPEI commented that it supports the proposed EPA notification, recordkeeping and
ongoing annual reporting requirements for equipment manufacturers - these proposed provisions
should be more than adequate to protect the integrity of the program.  Any additional
requirements would be overly burdensome.

       OPEI commented that it supports the proposed provision that would allow engine
manufacturers to simply keep records showing their TPEM engines met the Phase 2 standards -
rather than re-certifying those TPEM engines for the current model year.

       EMA commented that it agrees it is not appropriate or necessary to certify Phase 2
compliant engines used in the equipment flexibility program. EMA commented that engine
                                      2-95

-------
                                                            Chapter 2: Small SI Engines
manufacturers must be allowed to measure emissions prior to the catalyst, or with and without
catalyst, during deterioration factor determination and certification emission testing being
conducted for Phase 3 certification in order to generate the required data (i.e., data showing that
the engine family, without aftertreatment, will comply with the Phase 2 standard) with the
minimum amount of extra time and expense. Specifically, EMA commented that the provisions
in §1054.625(j)(2) must allow the test data required by sub-paragraph (i) to be measured prior to
the catalyst as part of the testing requirements for certification to Part 1054  Standards.

       EMA submitted two further comments on the regulatory language for the transition
program for equipment manufacturers.  First, EMA commented on §1054.625(g)(l)(iv) "What
requirements apply under the Transition Program for Equipment Manufacturers?" EMA
commented that equipment manufacturers will not be able to provide the name and address of
the company that produces the engines that it will be using for the equipment exempted under
this section prior to June 30, 2010. They commented that this requirement is impractical and
should be deleted.  Second, EMA commented on §1054.625(g)(2) "What requirements apply
under the Transition Program for Equipment Manufacturers?" EMA commented that all
manufacturers using the program should be required to comply with the reporting requirements
set forth in this section for each year of the program, or until  the manufacturer's ability to use the
program has expired.

Letters:
Commenter
OPEI
EMA
Document #
0675
0691
Our Response:

       EPA is retaining the reporting and recordkeeping requirements for the TPEM program
with some minor modifications, as noted below, in response to the comments summarized above.
First, EPA is revising the provisions of §1054.625(j)(2)(i) to allow manufacturers to measure
emissions prior to the catalyst to show that an engine would meet the Phase 2 standards. EPA
believes it is appropriate to give manufacturers the option of measuring emissions either before
the catalyst or with a non-catalyzed version of the exhaust system to show that an engine would
meet the Phase 2 standards.  Second, EPA is revising the provisions of §1054.625(g)i)(iv) to
require equipment manufacturers to  list the names of the manufacture^s) whose engines they
expect to use under the TPEM program.  Because the information being requested is due before
the TPEM program begins, EPA agrees that it would be difficult for an equipment manufacturer
to know which manufacturer's engines it would be using for the following four years of the
TPEM program.  However, equipment manufacturers should have an idea of which
manufacturers' engines it expects to use, and such information would be useful to EPA in
monitoring the use of the TPEM program.

       In response to the comment on §1054.625(g)(2), we are not making any changes to the
regulations. EPA believes the referenced language already requires equipment manufacturers to
report their use of the TPEM program  to EPA for each year they participate in the program.
                                         2-96

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


   2.7.4  Labeling

What Commenters Said:

       OPEI commented that to promptly respond to fluctuating demand, equipment
manufacturers need the flexibility to designate their inventoried engines as either a Phase 3
TPEM engine or as a Phase 3 "Delegated Assembly" engine. The final TPEM labeling scheme
must provide the needed flexibility to the OEM to designate his Phase 3 engines - after he has
ordered and received his engine families  In this regard, OPEI supported EPA's proposed
labeling provisions for the equipment manufacturers, under which his label would simply state
the equipment manufacturer's name and clarify that this is a TPEM engine. The engine
manufacturer's original emission label will appropriately provide all the engine-emission
information. A full content, equipment manufacturer, emission compliance label would be
confusing to customers and agency personnel - regarding the certification, emission warranty,
and other information typically provided by the engine manufacturer as specified in the engine
labeling requirements. In the cases where the engine originates as a Phase 3  compliant product
utilizing the Delegated Assembly provisions, the equipment manufacturer re-labeling of the
engine must not interfere with the ability of the ultimate consumer or the agency to accurately
identify the important information included on the  engine label.

       EMA commented that the content included on the engine/equipment manufacturer label
should be sufficient to convey the fact that an engine is designated as a TPEM engine. EPA
should not require the standard engine and/or equipment manufacturer emission compliance
label to be placed on a TPEM engine.  In the situation where a TPEM engine originates as a
Phase 3 compliant engine under the delegated assembly program, the equipment manufacturer
must be required to re-label the engine in a manner that will not interfere with the original engine
label. EMA also commented on §1054.625(j)(2) "What requirements apply under the Transition
Program for Equipment Manufacturers?" EMA commented that the reference in this section to
the labeling requirement set forth in §1054.610(c)(7) is not appropriate and should be deleted.

Letters:
Commenter
OPEI
EMA
Document #
0675
0691
Our Response:

       EPA is retaining the labeling provisions for TPEM equipment as proposed with only
minor changes as described below. As supported by OPEI, the regulations allow a manufacturer
to designate engines purchased under the delegated assembly program as either a TPEM engine
(with a separate TPEM label applied by the equipment manufacturer) or a fully compliant Phase
3 engine (with the appropriate aftertreatment installed by the equipment manufacturer).  EPA
notes that the TPEM label that must be placed on the equipment does require additional
information than noted in OPEI's comment supporting the proposed labeling provisions. In
addition to  stating the name of the equipment manufacturer and noting that the engine is a TPEM
                                      2-97

-------
                                                           Chapter 2: Small SI Engines
engine, the label must also contain contact information for the equipment manufacturer and the
year in which the equipment is produced.

      EPA disagrees with EMA's comment that a standard engine/equipment label should not
be required on a TPEM engine. EPA believes it is important to require a full engine label on the
engine as well as an additional equipment manufacturer label to identify TPEM equipment.  The
information on the labels allows EPA and others to identify important information about the
engine and equipment that could be needed to verify compliance with the TPEM program. In
response to EMA's comment on §1054.625(j)(2), EPA disagrees that the language is not
appropriate.  The citation is only a reference to the labeling requirements engine manufacturers
must comply with for their delegated-assembly engines (which may end up as  TPEM engines)
and does not add any additional requirements. Therefore, EPA believes it is appropriate to
include such a reference in the TPEM program regulations. It should be noted that in revising
the regulations for the final rule, EPA has moved the labeling requirements for engines
participating in the delegated assembly program to §1068.261, and therefore the language of
§1054.625(j)(2) has been revised to reference the new section.
   2.7.5  Additional provisions for imported products

What Commenters Said:

       OPEI commented that it supports the proposed special provisions, including bonding, for
foreign equipment manufacturers and importers of equipment made outside of the U.S. using
TPEM engines.

Letters:
Commenter
OPEI
Document #
0675
Our Response:

       EPA is adopting the provisions for foreign equipment manufacturers and importers of
equipment made outside of the U.S using TPEM engines as proposed.
   2.7.6  Relationship to tank permeation requirements

What Commenters Said:

       OPEI commented that it supports the proposal to also allow equipment manufacturers to
use non-compliant rotational-molded, "flex" fuel tanks on any of their equipment with TPEM
engines. OPEI objects to EPA's proposed limit on "flex" fuel tanks requiring the OEM to first
use up available banked credits or allowances from his early compliance with the fuel tank
permeation requirements.  This restriction takes away from the incentive for manufacturers to
introduce compliant tanks early or to produce tanks with FELs below the standard.  In addition,
                                         2-98

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


they believe this restriction will overly-complicate the administration of the program with no
benefits.

       OPEI submitted an additional comment after the close of the comment period regarding
rotational-molded fuel tanks.  They supported a delay in the permeation requirements for
rotational molded fuel tanks instead of the proposed linkage to the TPEM program.

Letters:
Commenter
OPEI
OPEI
Document #
0675
0793
Our Response:

       EPA is revising the allowance for equipment manufacturers to use non-compliant
rotational-molded fuel tanks on their equipment with TPEM engines for the final rule. EPA
continues to believe that equipment manufacturers may face challenges in transitioning all of
their rotational-molded fuel tanks to meet the new permeation standards in the timeframe for the
new standards. However, based on discussions with manufacturers, we have been convinced that
there is not necessarily a direct link between the potential TPEM engines/equipment and the use
of rotational-molded tanks on those engines/equipment.  We are therefore allowing equipment
manufacturers to use noncompliant rotational-molded fuel tanks for two additional years on
limited numbers of 2011 and 2012 model year equipment using Class II engines, regardless of
whether the equipment is part of the TPEM program. Equipment manufacturers may use
noncompliant rotational-molded fuel tanks if the production volume of the fuel tank design used
in Class II equipment models is collectively no more than 5,000 units in the 2011  model year. In
the  2012 model year,  equipment manufacturers may use noncompliant rotational-molded fuel
tanks if the production volume of the fuel tank design used in Class II equipment models is
collectively no more than 5,000 units in the 2012 model year, but the total number of exempted
rotational-molded fuel tanks across the manufacturer's Class II equipment is limited to 10,000
units. If production volumes are greater than 5,000 for  a given fuel tank design, all of those
tanks must comply  with emission standards. Tank designs  would be considered identical if they
are  produced under a  single part number to conform to a single design or blueprint.  In addition,
tanks would be considered identical if they differ only with respect to production variability,
post-production changes (such as different fittings or grommets), supplier, color, or other
extraneous design variables.

       2.8  Delegated assembly

    2.8.1  Need for delegated assembly

What Commenters Said:

       OPEI noted that EPA has proposed a permanent Delegated  Assembly program (see
§1054.610) specifically designed for small spark-ignition engines.  The purpose of this program
is to create a very protective compliance program that would allow non-integrated engine
                                      2-99

-------
                                                                     Chapter 2: Small SI Engines
manufacturers to distribute their certified engines without the required emission-related parts
(i.e., catalyst) that are listed on and required for EPA certification. OPEI elaborated as follows:

        First, in the initial stage of developing the after-treatment system, the engine manufacturer must determine
        the catalyst-specific parameters including substrate size, precious metal loadings, and engine performance-
        related specifications to allow the catalyst to be packaged or canned so that it can be installed. The catalyst
        selection and packaging designs must provide the intended exhaust emission conversion and also manage
        exhaust and cooling air flow to ensure all safety concerns are addressed, and to manage sound and tonal
        qualities.

        Second, the engine manufacturer identifies the "standard" muffler packages that meet the criteria identified
        above. In many cases there is only one "standard" muffler configuration designed and developed by the
        engine manufacturer. Engine manufacturers cannot specify the diversity of mufflers required to fit into
        specific equipment or to install packaged catalysts into such customized mufflers. Such customization is a
        very time-consuming and resource-intensive exercise that adds significantly to the complexity of the engine
        manufacturer's product as supplied to the equipment manufacturer.  It is highly unlikely that engine
        manufacturers would be able to fundamentally change their business structure to supply customized
        mufflers in the future. Such a change in their business models would be tantamount to asking a major
        international supplier of lumber, which sells large volumes of stock lumber to less than one hundred
        wholesalers, to only sell customized, small volume cabinets or furniture to thousands of individual
        customers.

        Third, for most Class II products, the catalyst prescribed by the engine manufacturer must be packaged into
        the muffler system prescribed by the equipment manufacturer because there is insufficient space to allow
        separate catalyst and muffler systems. Consequently, for the vast majority of Class  II engines, engine and
        equipment manufacturers must depend on their independent muffler suppliers, who exclusively have the
        capacity and expertise to install catalysts properly into their customized mufflers. For most Class II
        products, the independent muffler supplier is the only party who  can practically install catalysts into the
        mufflers for the vast majority of Class II engines and customize these products for the various equipment
        designs.

        Fourth, without Delegated Assembly, the engine manufacturer would have to include specific catalyzed
        mufflers in the box with his shipped engines. The OEM would not be able to use many of these purchased
        and shipped mufflers because they would not fit into his final, complete applications. This problem results
        from the fact that an equipment manufacturer cannot wait for specific orders from his downstream retailers
        before he orders his engines. The OEM must typically purchase and receive large volumes of the same
        engine family, which must be used in many different models. Each model will likely have different or
        unique muffler configurations. Unpredictable market demand will drive the OEMs  ultimate production of
        specific equipment models  and therefore the volumes of the different engine-muffler combinations the
        OEM will ultimately build. When the engine is shipped to the OEM, neither the engine manufacturer, nor
        the equipment manufacturer may know which exact equipment models and/or which muffler configurations
        wil be used with each specific engine family.

        OPEI concluded their argument by noting that EPA must finalize a practical Delegated
Assembly program which allows direct shipments of catalysts from a catalyst supplier to a
muffler supplier, who will be accountable and responsible for proper catalyst canning and will
install the required catalyst in each application,  prior to shipping the specialized catalyzed
muffler to the OEM.

        OPEI commented that  as part of this preamble discussion, EPA incorrectly suggests that
many muffler geometries are fairly uniform and that it should be possible to produce more
standardized, "stock" mufflers that could be supplied directly through the engine manufacturer
(see 72 Fed. Reg. at 28152).  OPEI believes EPA grossly over-simplified the problems and
                                               2-100

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


challenges associated with applying and packaging exhaust/muffler systems onto a wide variety
of equipment.  While many mufflers have a somewhat common cylindrical geometry, the
muffler mounting brackets, internal sound bafflers, and the header and tailpipe geometries vary
greatly. One supplier of mufflers and exhaust components, released just under 100 unique
components last year alone to address the variations in the muffler and exhaust configurations
required to fulfill the OEM packaging requirements for these small engine applications.  OPEI
estimates that there are over 1000 different muffler configurations for these engines in the
national marketplace today.

       Traditionally, the muffler design is a compromise between exhaust back pressure, as
prescribed by the engine manufacturer for emission compliance, and sound attenuation.  The
various applications these mufflers are installed in have unique trade-offs. OPEI commented that
EPA needs to consider the unique requirements of the small  engine market that serves
applications such as: portable power, utility vehicles, golf carts, construction equipment, light
towers, agricultural, etc.  Other equipment may also have different requirements  which can
dramatically affect muffler designs.  The application specific trade-offs often lead to unique
internal design features of the muffler for each engine, in addition to external differences.

       Of course, the principal challenge of muffler design is thermal management issues, which
are safety-related and are a further constraint to application designs. The addition of a catalyst
will create additional complexities and challenges, especially when dealing with  off-nominal
conditions such as engine misfire situations (See Sections III and IV above). The thermal
management issues lead to a variety of different insulation and heat shield scenarios, which
result in unique muffler configurations specific to the product design. Current experience with
the development of catalyzed muffler systems that meet California Tier III regulations has
confirmed that these product-designs and complex heat and emission-related challenges will
demand customized mufflers that are supplied by a third party.

       For all these reasons,  OPEI commented that it does not believe that there  will or can be a
shift in the market place towards standardized "stock" muffler designs.  OEMs will continue to
depend on customized mufflers to facilitate their product designs as required to service their
diverse markets.  Consequently, OPEI commented that the Delegated Assembly Program is
absolutely crucial to satisfy the market needs of the small SI applications to obtain catalyzed
mufflers from their muffler suppliers.

       OPEI noted that EPA has requested comment on whether the transition program for
equipment manufacturers somehow meets the need for the proposed Delegated Assembly
Program (see 72 FR 28152).  Conversely, EPA requested comment on whether manufacturers
will need the equipment-transition program (described above in Section III) if they can
independently source their exhaust systems  based on the Delegated Assembly Program (see 72
FR 28154).  OPEI responded that the answer to both of these questions is  no. OPEI believes
both programs are necessary. In fact, the two programs serve separate,  distinct purposes.
Finally, OPEI commented that EPA has  failed to evaluate or quantify (in its administrative
record or in its SBREFA process) the substantial economic damages that would result without
either the proposed equipment flexibility or the Delegated Assembly provisions (proposed in
§1054.610).  EPA's other off-road emission regulations explicitly recognize the separate and
                                      2-101

-------
                                                            Chapter 2: Small SI Engines
independent need for Delegated Assembly, equipment-transition flexibilities, and hardship relief.
Consequently, OPEI commented that it would be arbitrary for EPA to abruptly eliminate any one
of these flexibilities for small engines.

       EMA supported the proposed delegated assembly and equipment manufacturer flexibility
provisions included in the NPRM.  EMA commented that EPA must incorporate both of these
programs into the final rule. If EPA were to adopt the delegated assembly program only and not
the equipment manufacturer flexibility program (or visa versa), the functionality of the adopted
program would be significantly impaired by the absence of the other program.

       Honda noted that in the market today for small engine products, multiple businesses
cooperate to produce parts for numerous small engine powered machines that are the design,
production, and marketing responsibility of a multitude of independent equipment
manufacturers. Each manufacturer in the process, including the final equipment manufacturer,
may design and manufacture, design and outsource the manufacture of, or simply purchase an
existing part.  This  distinction is important to identify who should take responsibility for the
actual performance of the part.

       Today, for larger sales volumes, engine manufacturers cooperate closely with equipment
manufacturers to meet the equipment manufacturers' product needs.  At a smaller manufacturing
volume there is less direct contact but the process of matching the engine to the equipment and
the documentation of this process are still in place. The key to the appropriate  use of a specific
engine in a specific application is based on a basic engineering evaluation of the engine matching
document and of the general instructions for engine use.  This concept works well when the
equipment manufacturer uses the engine as it was built by the engine manufacturer. However, in
actuality, a wide variety of equipment is used on an even  wider variety of tasks and the
equipment manufacturer must be able to tailor the engine to fit both the equipment and the task.
For example, a trencher, earth rammer, concrete equipment, and a lawn mower work in similar
dust and debris environments, but the packaging of a single engine model in the least vulnerable
location to make  a workable machine can result in very different requirements.  The proposal's
preamble includes a comparison of a handful of mowing equipment and concludes that it would
be possible to package a single engine design in all machines. Honda believes  this fails to look
beyond the most popular use of small SI engines and does not recognize the significant diversity
of small engine powered types of equipment.

       Honda commented that the need for an equipment manufacturer to have flexibility in the
final assembly of engine intake and exhaust components is critical for both large and small
volume equipment manufacturers.  An engine manufacturer cannot economically stock or supply
in a timely manner, the  array of components required by the diversity of the market. Only the
equipment manufacturer that functions as an independent business, striving to create or improve
a machine's design and target  a value price, is in a position to create the best product for its
customers.  Similarly, the engine manufacturer is in the best position to provide engine matching
tools and instructions that ensure the final engine assembly will be in compliance with applicable
regulations and will match the required certification information. In many cases the engine
manufacturer and the equipment manufacturer will work jointly, or with a third party, to ensure
that the design is compliant with the regulations.
                                         2-102

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
       Honda suggested two alternatives to the proposed means of ensuring that the final
assembly of the equipment will comply with the regulations and answer the concern about
enforcement of the regulation.

       1) Honda noted that in some cases the engine and equipment manufacturer (and
potentially a third party supplier or testing facility) work together to  fulfill the equipment
manufacturer's product design targets. During that process, they generate enough information
and working instructions that the equipment manufacturer could then submit an abbreviated
certification form and be identified to the EPA as the manufacturer with responsibility for the
specific configuration of that engine model used in the equipment manufacturer's particular final
product. EPA could then assume its rightful role and use the Selective Enforcement Audit
mechanism or confirmatory testing to verify that the product meets the requirements of the
regulation.  EPA would also know in advance which manufacturer is taking responsibility for
what part of the final assembly.

       2) Honda suggested that EPA supplement the definitions in 1068.101(b) and the text in
1054.20 so that it is clear that failure to follow the engine manufacturer's instructions for
delegated assembly is tampering and/or falls into the category of a defeat device. This option
could be applied when the equipment manufacturer does not interact directly with the engine
manufacturer and an abbreviated certificate is not submitted. The equipment manufacturer in
this case would need to follow the engine matching and  installation instructions using its own
data or engineering evaluation. This type of situation could also be treated in a manner similar to
1060.101(f) in the evaporative emissions section where the equipment manufacturer is "deemed
to be certified."  EPA will then have the necessary enforcement authority when they perform a
field or factory audit of equipment.  A thorough examination of the steps involved in engine
distribution, product design and manufacturing, and the  role of third party suppliers should make
it possible to retain EPA authority to ensure emission  compliance without disrupting the ability
of both engine and equipment manufacturers to deliver innovative and value priced product to
the consumer.

       CARB commented that EPA's overall  approach  to the proposed delegated final assembly
is reasonable  and corresponds to CARB's current certification procedures.  CARB  believes that
EPA should require engine manufacturers to be held responsible for  ensuring that the catalysts
are installed on the engines that are accumulating credits.  Since the  engine manufacturer is
receiving the  benefit of accumulating credits for these particular engines, they should make sure
that the catalysts are being installed.  Any instance of an engine found without a complete
emission control system, as certified, should be treated as noncompliant, with all possible
penalties. Allowing any exceptions would send an inappropriate message to the manufacturers.
                                      2-103

-------
                                                            Chapter 2: Small SI Engines
Letters:
Commenter
OPEI
EMA
CARS
Honda
Document #
0675
0691
0682
0705
Our Response:

       An important point to clarify before evaluating the commenters' arguments in support of
delegated assembly is that under any conceivable regulatory scenario we would require mufflers
that are shipped directly from the muffler manufacturer to the equipment manufacturer to be
specifically included in the engine manufacturer's application for certification. While the
commenters describe or imply a need for equipment manufacturers to have an unrestrained
ability to work out design parameters with  muffler manufacturers, we have neither proposed nor
considered such an outcome.  OPEFs description betrays an understanding that equipment
manufacturers should have an unrestrained ability to change muffler designs parameters even if
that affects emission levels. We acknowledge that muffler design involves compromises
between back pressure, thermal management, and sound attenuation.  A muffler design that is
not specifically part  of the engine manufacturer's application for certification will inevitably
involve design parameters that favor non-emission factors over factors important for controlling
emissions. The result would be a noncompliant engine and a situation where the certifying
manufacturer will disclaim any responsibility for the performance of its own engine.  This is
clearly unacceptable. The certifying engine manufacturer is responsible for ensuring compliance
and therefore needs to be in control of design variable that could affect whether engines meet
emission standards or not.

       We agree with the commenters that muffler manufacturers play an important role in
incorporating an engine manufacturer's specified catalyst into a muffler that appropriately
controls air flow for maintaining catalyst performance, managing external surface temperatures,
and provides proper sound attenuation. However, this fact alone does not demonstrate that
equipment manufacturers need to be able to get customized mufflers for every  equipment model.
There are many examples of current engine and equipment models in which mufflers and
catalysts flow from component manufacturer to engine manufacturer to equipment manufacturer,
with varying degrees of involvement by equipment manufacturers in the design parameters of the
exhaust components. Regardless of the extent to which engine and equipment  manufacturers
would work  out arrangements for delegated assembly, every engine manufacturer will need to
certify their engines using some number of stock mufflers. In the transition to  new emission
standards, engine and equipment manufacturers will work out the degree to which multiple
muffler configurations will be needed to meet the design needs for the range of equipment
models that will be affected.  The Transition Program for Equipment Manufacturers provides
four years of a more flexible transition to allow for these negotiations and adjustments.

       OPEI's analogy to a lumber supplier needing to start selling custom cabinets  exaggerates
the business dynamic in question. The comparison does not acknowledge that  engine
manufacturers are already selling the new product (engines with mufflers) in many cases, that
                                         2-104

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


engine manufacturers are liable for the performance of the finished product, or that equipment
manufacturers have to design their equipment around a given muffler design, whether or not the
muffler is manufactured to their specifications. Our objective is not to create a better analogy,
but the weakness of this comparison highlights our concern that the commenter is unable to
provide a straightforward rational assessment of the situation.

       The fundamental gap in the argument presented is the interplay between equipment and
engine manufacturers in coming up with final design specifications.  We believe that purchasing
agents and design engineers working for the equipment manufacturers will have an important
role in moving successfully into a new era in which engine manufacturers have a legal
responsibility to ensure that exhaust systems are properly designed and assembled for
compliance with exhaust emission standards.  Purchasing agents for equipment manufacturers
buying large volumes of engines can have a very significant influence over the engine
manufacturer's design parameters. As a result, we would expect these dominant equipment
manufacturers to effectively dictate muffler designs to ensure that available stock mufflers meet
their needs, considering physical dimensions, thermal management, and sound attenuation.
Engine manufacturers would want to make a reasonable number of muffler configurations
available, so we would envision this process playing  out such that several stock mufflers would
be available.

       Even under the broadest conceivable approach to delegated assembly, equipment
manufacturers will be unable to get customized mufflers for their small-volume products. Since
engine manufacturers need to agree to add each muffler configuration to their application for
certification and enter into a contract with equipment manufacturers creating customized
mufflers, there will be many cases where this option isn't viable or cost effective.  Engine
manufacturers may decide that a custom design presented to them by  an equipment manufacturer
is unacceptable, or they may be unable to provide the resources to make this determination.
They may be unwilling to trust the equipment manufacturer to properly procure parts for and
assemble the final products such that every engine is in its certified configuration  before delivery
to the end user.  These potential complications were given credence by one manufacturer who
communicated to us that their plan is to participate in delegated  assembly using a  custom muffler
from an equipment manufacturer only if the equipment manufacturer  performs a complete round
of testing, including service accumulation over the engine's full useful life, to show that the new
muffler design complies with the underlying certificate. This is more than we require currently,
but it illustrates a prudent approach by engine manufacturers to protect themselves from the
liability of delegating important compliance responsibilities to other companies.

       Design engineers working for equipment manufacturers  also have an important role to
play in this process. While OPEI suggests that equipment manufacturers will need to
discontinue production of equipment models if they can't procure customized mufflers, we
believe this ignores the equipment manufacturers' ability to adjust the designs of their equipment
to accommodate a specific muffler configuration supplied  to them by an engine manufacturer.
As noted in the comments, there are many examples of custom muffler designs that are tailored
to a specific type of equipment. If that custom muffler was no longer available, design  engineers
for the equipment manufacturer could, for example, adjust mounting brackets, accommodate a
different muffler orientation, or otherwise make the muffler fit to allow the equipment to
                                      2-105

-------
                                                            Chapter 2: Small SI Engines
satisfactorily perform its function. While there would be a significant effort and expense to
modify equipment designs just for a muffler change, the incremental effort of accommodating a
muffler change as part of a broader equipment redesign is much smaller. Our understanding is
that equipment models are typically redesigned every five to eight years. (The cost estimates in
Chapter 6 of the Final Regulatory Impact Analysis take into account the cost of modifying
equipment as part of an overall redesign.)

       Customizing mufflers for sound attenuation and thermal management should be easily
managed by the engine manufacturer. We believe engine manufacturers will be strongly
motivated to meet market demands by working with muffler manufacturers to create a menu of
stock mufflers that provide varying degrees of sound attenuation.  The engine manufacturer
would be well  positioned to efficiently design for sound attenuation by integrating that effort into
an overall design program to develop a catalyst and exhaust configuration that meets emission
standards. Similarly, thermal management of exhaust surfaces is fundamentally related to engine
operation. Engine manufacturers are best positioned to design mufflers (in cooperation with
muffler manufacturers) such that all possible engine operating modes are considered when
properly designing a muffler to avoid any risks associated with high surface temperatures.

       We would also caution against the tendency to overstate the extent of change in muffler
designs resulting from our proposal.  Our testing to support the feasibility of the new emission
standards showed that an existing muffler could be modified to incorporate a catalyst primarily
by rearranging the internal flow paths, without significantly changing the muffler's outer
dimensions.  We also showed that this could be done without significantly increasing external
surface temperatures. This is not to say that engine and muffler manufacturers won't develop
mufflers that have notable differences from current designs, rather that we are not expecting
dramatic changes in these designs. As a result, we believe the design challenge for equipment
manufacturers will mostly involve the transition from customized to stock mufflers.  As noted
above, this will involve little or no change for high-volume products, because equipment
manufacturers will in effect dictate that their custom design becomes one of the standard
configurations from the engine manufacturer.  For the remaining equipment models, we are
confident that equipment manufacturers will be able to make the changes needed to
accommodate a stock muffler, such as rearranging mounting brackets, repositioning mufflers, or
otherwise to make the mufflers fit into the overall equipment design.

       We also believe that the commenters grossly overstate the current need for customized
mufflers. We stand by our observations in the preamble of our proposed rule regarding the
standardization of mufflers in current products.  The products we observed with relatively
uniform muffler configurations represented a wide range of models, brands, and applications.
Moreover, the  general observation was that the nature of mufflers and exhaust systems is that
they need space to safely and effectively route hot exhaust gases away from the engine and into
the atmosphere. We suspect that the large number of muffler configurations produced today is
mostly related to proper mounting, orientation, and plumbing to fit the muffler into the
equipment. Redesigning most equipment for a standard muffler configuration should involve
only modest changes to shift the position of the muffler or to change the cage or shielding or
frame that currently  houses the muffler.  That is not to say that there aren't examples of mufflers
that are more carefully tailored to specific equipment models, rather that we believe this practice
                                         2-106

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


is much less common than suggested by the commenters. We understand that these changes will
take time and effort, but we believe that they are well within reach for equipment manufacturers,
especially as part of an overall equipment redesign. The time available before the standards take
effect and the flexibility provisions built into the final rule should allow equipment
manufacturers to work with engine manufacturers for an orderly transition in their equipment
designs to the extent that is needed.

       Two examples from observations made at the 2006 Louisville Expo for lawn and garden
equipment highlight the difference in opinion as to the extent of the redesign that will be
necessitated by this rulemaking.  First, we observed a four-wheel drive utility vehicle in which
the compartment to house the engine and the whole exhaust system were internal to the body of
the vehicle, located primarily behind and under the driver's seat.  This would seem to be a prime
example of a need for a specialized muffler confined in a limited space, since every amount of
space devoted to the engine and exhaust system was space made unavailable for passengers and
payload. Despite this trade-off of utility and comfort, the vehicle design included a relatively
large cavity for the muffler and other exhaust components.  Clearly this amount of space was
needed to provide adequate clearance from the exhaust surfaces to avoid exposing other parts of
the vehicle to such high temperatures. We suspect that an effort to change to a different muffler,
even one with a very different shape, would not be impossible.

       In a second example, an equipment manufacturer complained vehemently that their
engine supplier insisted on supplying the muffler with the engine, leaving them with the
extraordinary burden of fitting the stock muffler into their equipment.  The representative
claimed to have a worst-case equipment model on display- a riding lawn mower with several
premium features, including a plastic collection bin mounted behind the mower and over top of
the exhaust system.  In this case the muffler was mounted in a cage for preventing accidental
contact with hot exhaust surfaces, again with rather generous spacing around the muffler. It was
apparent that changing this equipment model would require significant time to address design
concerns such as fit, weight distribution, exposure to radiant heat, etc.  We believe, however, that
these design challenges could all be addressed even by a company with very limited engineering
resources. Having several years to plan and execute these changes seemed to be a very
reasonable expectation, even in this worst-case configuration. These observations support our
conclusion that equipment manufacturers  will be able to respond to changing muffler designs in
the context of the Phase 3 standards, especially if they have a transition period that will allow
them to factor in the necessary changes in advance. Furthermore, the fact that there is already an
example of an engine manufacturer telling its customers that only stock mufflers are available
demonstrates that this can be a business decision negotiated between companies rather than one
that is inherently and necessarily subject to the control of equipment manufacturers.

       It is important to note the comparison with nonroad diesel engines, as we are expecting
those engines to include new aftertreatment devices to meet Tier 4 standards. These
aftertreatment devices will be new, relatively large components added to exhaust systems (not
incorporated into existing mufflers) that equipment manufacturers will need to accommodate.
We will allow the equipment manufacturers the flexibility of using limited numbers of previous-
tier engines for several years (much like the Transition Program for Equipment Manufacturers
described above).  There is an expectation for the long term that equipment manufacturers will
                                      2-107

-------
                                                            Chapter 2: Small SI Engines
be able to use stock aftertreatment devices from the engine manufacturers as part of their
equipment design.  The expected design effort for Small SI equipment pales in comparison to the
efforts expected from the nonroad diesel equipment manufacturers.  We also note that the
program for nonroad diesel engines is the only other one in which we have adopted both
delegated assembly and a Transition Program for Equipment Manufacturers. These are not
universal and inherent aspects of our compliance programs, as suggested by OPEL

       Aside from the question of who designs catalyst and muffler configurations, we
acknowledge that there are business reasons to prefer shipping mufflers directly from the muffler
manufacturer to the equipment manufacturer.  This was the original purpose of the delegated-
assembly provisions we adopted in §1068.260. OPEI's comments appropriately describe the
situation for an equipment manufacturer in that situation, needing to manage large numbers of
equipment models, each with multiple engine and muffler configurations. The dynamics of
managing inventories to produce all of these equipment models causes us great concern that
every assembled unit is built properly in its certified configuration.  This is the basis of the
extensive protective measures we believe are necessary to ensure that engines are properly
assembled.

       We  also acknowledge that, with proper constraints and controls, engine manufacturers
can work with equipment manufacturers that they trust to install properly designed catalyzed
mufflers. Some of these mufflers may have been designed by the equipment manufacturer
together with the muffler manufacturer and coordinated with the engine manufacturer,  such that
final engine assemblies will meet the required standards. As a result, we believe it is appropriate
for the final rule to include a carefully constructed delegated-assembly program for Small SI
engines in addition to the Transition Program for Equipment Manufacturers that will allow
manufacturers a flexible transition period to incorporate any engine or muffler design changes
resulting from compliance with the Phase 3 standards. This transition period will allow time for
market forces to work toward a sensible degree of accommodation between engine and
equipment manufacturers as they find the best way of dividing design and assembly
responsibilities such that they preserve the engine manufacturers' ultimate control over design
and compliance responsibility and at the same time recognize the equipment manufacturers'  need
to make equipment that functions within the limitations in muffler design and  specifications
required due to certification.  There is a continuing need for delegated assembly after this
transition period, but we believe this is more of a business decision regarding the most efficient
method of designing and shipping product than an inherent necessity for equipment
manufacturers to be able to produce equipment with certified engines that can be used in a
multitude of applications. Accordingly, we are adopting delegated-assembly provisions for
Small SI engines that include greater initial flexibility, after which a narrower set of provisions
apply, as described in Section 2.8.2.

       We  believe the proposed regulations already  reflect Honda's suggested approaches for
dividing responsibilities among engine and equipment manufacturers.  The idea that  equipment
manufacturers rely on an abbreviated certification for designs that fall outside of the  engine
manufacturer's certified configurations was proposed in §1054.612. This gives the equipment
manufacturer the ability to recertify an engine family without generating a new deterioration
factor or conducting production-line tests. Also, the current regulations in §1068.105 clearly
                                         2-108

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


state that equipment manufacturers violate the tampering prohibition if they fail to follow an
engine manufacturer's installation instructions. However, it would not be appropriate for the
engine manufacturer's installation instructions to simply specify broad design parameters that
equipment manufacturers would then follow, with some unspecified testing or engineering
evaluation to support a conclusion that the resulting engine design is covered by a certificate.
The "deemed certified" approach proposed in §1060.101 is limited to requirements that are so
straightforward that they can be established by simple observation. Evaluating compliance with
exhaust emission standards is far from straightforward. We therefore believe that approach of
considering engines certified through an informal demonstration does not meet the requirement
under the Clean Air Act for engines to be certified based on a demonstration that measured
emission levels are within prescribed limits.  We also believe manufacturers would be unwise to
delegate this level of responsibility to another company, since they would be held liable for any
noncompliance resulting from any designs that fall short of meeting emission standards.


   2.8.2  Specific provisions for delegated assembly

What Commenters Said:

General:
       OPEI noted that EPA requested comment on the need for the specific provisions of the
Delegated Assembly for small engines with catalyzed mufflers in comparison to other non-road
engine/equipment categories as defined by current regulations (72 Fed. Reg. at 28149-28152).
OPEI commented that the proposed small SI  provisions are essential in order to respond to the
following unique constraints of the small spark ignition engine and equipment industry:  (1) the
cost sensitive nature of the products produced; (2) the retail distribution system employed; and
(3) the diversity of products.  OPEI commented that the current generic Delegated Assembly
Program fails to respond to each of these unique factors and would create totally impractical
burdens (see §85.1713 and §1068.260).  In turn, this would have a dramatic, adverse impact on
both large and  small outdoor power equipment manufacturers resulting in the elimination of
many equipment models.

       EMA commented that the Small  SI engine and equipment industries have specific needs
regarding delegated assembly that have been appropriately balanced with the regulatory
requirements as specified in §1054.610 of the proposal along with the other changes
recommended by EMA in this section. EMA commented that the final rule should not integrate
these requirements with the general provisions prescribed in 40 CFRPart 1068, but should rather
retain their independence in Part 1054.

Written confirmation:
       OPEI commented that the regulations need to allow a small engine manufacturer to
obtain written confirmation (within 30 days after shipping engines) that his OEM  customer has
ordered the appropriate catalysts as part  of the initial shipment -approval process for delegated
engines.
                                      2-109

-------
                                                            Chapter 2: Small SI Engines
       EMA commented on §1054.610(c)(9) "What is the exemption for delegated final
assembly?"  EMA commented that the proposed requirement is not viable. An engine
manufacturer cannot "have written confirmation ... for an initial shipment of engines..." and
also ".  . . receive the written confirmation within 30 days of shipment." Accordingly, EMA
commented that the language should be revised to read as follows: "You must advise the
equipment manufacturer that (i) written confirmation that the appropriate aftertreatment has been
ordered is required within 30 calendar days of the initial engine shipment for a given model year;
and (ii) if written confirmation is not received future engine shipments will not be allowed.  The
equipment manufacturer can meet the written confirmation requirement by notification to the
engine manufacturer that engines will be used under the equipment manufacturer flexibility
program defined in 40 CFR Part 1054.625."

Audits:
       OPEI commented that engine manufacturers' audits of their respective OEM production
practices (and confirmation that products meet the certified configuration) can be effectively
accomplished in many different ways.  The regulatory requirements should not constrain the
options an engine manufacturer may utilize. OPEI commented that the final program should
allow the small engine manufacturers to conduct audits of either the OEM's production process
or his final assembled products (pursuant to EPA's proposal).

       EMA commented that there are many different ways an engine manufacturer can
effectively audit an OEM's production practices and confirm that products meet the certified
configuration. The regulatory requirements should not place undue restraint on the engine
manufacturer's ability to use the many viable options available.  In order to accommodate the
wide variety of engine manufacturer/OEM business relationships, the auditing requirements must
be flexible. Each engine manufacturer has a variety of OEM customers ranging from the very
sophisticated large business (where engine orders/deliveries are coordinated with equipment
build schedules for just in time production) to small companies that may only place a single
order each year. EMA commented that requiring certification documentation of all the various
options an engine manufacturer may utilize is burdensome and ineffective. Certification
documentation should be limited to an acknowledgement from the engine manufacturer of the
need for the required audits and its intent to utilize the delegated assembly provisions.

       EMA commented on §1054.610(c)(10) "What is the exemption for delegated final
assembly?"  The requirement to select individual equipment manufacturers equally among the
volume quartiles is overly prescriptive with no added benefit to the environment. EMA
commented that this section should be revised in order to provide for selection  of equipment
manufacturers from each quartile as much as possible. This will allow  engine manufacturers to
select equipment manufacturers for auditing based on their confidence in the equipment
manufacturers processes.

Point of final assembly:
       OPEI and EMA commented that the "point of final assembly" (when the exemption no
long applies) will vary depending on the equipment manufacturer production process. Engines
that are scheduled to be utilized in one equipment model may be pre-assembled with the
expected exhaust system.  Due to production-demand changes, these engines may be returned to
                                         2-110

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


inventory with the intent of later assembly only to be reconfigured at a later date for installation
into different equipment.  This may result in the exhaust system being replaced.  For purposes of
determining whether an exemption has expired, OPEI commented that the "point of final
assembly" should be defined as the point at which the final equipment is totally complete and
ready to be introduced into commerce.

Labeling:
       OPEI commented that if there are provisions required for the designation of Delegated
Assembly engines on the emission compliance label, OPEI supports the use of an identifying
mark on the permanent label, such as "DA" as an approved abbreviation for "Delegated
Assembly."
       CARB  recommended a change to the proposed labeling requirements.  EPA is proposing
a partially completed label (temporary label) be placed by the engine manufacturer which would
subsequently be replaced by a final permanent label by the equipment manufacturer upon
completion of delegated  assembly. Since the engine manufacturer is ultimately responsible for
the final assembly and product as the holder of the Executive Order (in California) and
Certificate of Conformity (federally), CARB recommended requiring the following:
     Option 1:  The engine manufacturer applies a partial permanent label, and following
delegated assembly, the equipment manufacturer adds a supplemental permanent label (placed
just below the original label) completing the labeling requirement.  This procedure is similar to
the approach used for rebuilt/replacement off-road compression-ignition engines.
     Option 2:  The engine manufacturer applies a complete permanent label and ships the
incomplete engine to the equipment manufacturer who subsequently completes the delegated
assembly.  This option would have an added requirement that the engine manufacturer must
demonstrate, as part of the certification process, that there are quality control procedures in place
to ensure that the final assembly occurs correctly.

Production-line testing:
       OPEI and EMA commented that engine manufacturers should be allowed flexibility
regarding the equipment manufacturer supplied exhaust systems required for PLT testing,
including the ability to inventory randomly selected samples for future PLT testing requirements.

       EMA commented on §1054.610(c)(12) "What is the exemption for delegated final
assembly?" EMA commented that this section should be revised in order to clarify that engine
manufacturers  may inventory equipment manufacturer supplied exhaust systems for production
line testing, provided that such systems are randomly selected components that are representative
of equipment manufacturer production.

Class I engines:
       OPEI and EMA commented that Class I engines are generally sold complete with the
engine manufacturer supplied exhaust system. However, there  are a limited number of specialty
products where this is not possible. Some Class I products have all of the same equipment
manufacturer/customer demands that are necessary to provide a Delegated Assembly option for
the larger Class II engines. OPEI and EMA commented that these Class I products should not be
precluded  from this required flexibility based only on their respective class.
                                      2-111

-------
                                                            Chapter 2: Small SI Engines
Liability:
       EMA commented that engine manufacturers utilizing the delegated assembly provisions
and meeting all specified requirements (e.g., provide the equipment manufacturer with all
information necessary to complete the engine assembly to it's certified configuration, and
conduct the required audits) must be assured that the equipment manufacturer is responsible for
delivering compliant product into commerce.

       Honda noted that the NPRM takes the position that the engine manufacturer, as the
engine certifying party, becomes responsible for the actions of the equipment manufacturer, an
independent business.  The NPRM essentially appoints the engine manufacturer to be the
"Selective Enforcement Authority"  and to perform audits of the equipment manufacturer as
though the engine manufacturer were a government agency with authority to enter a business and
inspect records. Honda does not believe EPA intends to relinquish its independent enforcement
authority nor does Honda believe it is reasonable to ask the equipment manufacturer to submit,
by contract or otherwise, to inspection by competing engine manufacturers, all of whom sell to
his competitors, and in some instances produce the same type of product within their own
company or a wholly  owned subsidiary.  Honda also noted that the NPRM states that the
equipment manufacturer must follow the engine manufacturer's instruction or the equipment is
not covered by the certificate of conformity and not legal to introduce into commerce.

Air filters:
       EMA commented on §1054.610(e) "What is the exemption for delegated final
assembly?"  EMA commented that manufacturers must have the ability to certify engines
without identifying a specific part number for the air filter.  This ability must either be
specifically incorporated into the regulatory language, or included in a clarifying regulatory
support document.  Current Certification Guidance and submission templates require inclusion of
air filter part numbers as a condition of certification. However,  this section would allow engine
manufacturers to provide a definitive parameter, such as intake restriction range, to define the
certified configuration.  Therefore, equipment manufacturer installed intake systems meeting the
engine manufacturer prescribed parameter would not be subject to these provisions.

References:
       EMA commented on §1054.610(g)(2) "What is the exemption for delegated final
assembly?"  EMA commented that §1054.610(g)(2) includes an incorrect reference to paragraph
(g)(2). This reference should be corrected to refer to paragraph  (g)(l).

Within-company shipments:
       EMA commented on §1054.610(m) "What is the exemption for  delegated final
assembly?"  EMA noted that as set forth in §1054.610(d), engine manufacturers that install
engines into equipment are not required to request an exemption or take any other extraordinary
steps in order to do so.  Likewise, engine manufacturers should be allowed to complete
production of engines at different facilities without being required to request an exemption.
Accordingly, EMA commented that this section should be deleted.
                                         2-112

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


Evaporative systems:
       EMA commented on §1054.610 "What is the exemption for delegated final assembly?"
EMA commented that this section should be revised to add a provision similar to §1054.610(c)
that would apply to the situation where an engine manufacturer certifies compliance to the
evaporative standards and delegates final assembly of the evaporative system to the equipment
manufacturer.  Such a provision is of particular importance to small equipment manufacturers
that cannot use fully integrated engines and do not have the resources to design and certify
pursuant to the 40 CFR Part 1060 requirements.

Letters:
Commenter
OPEI
CARS
EMA
Honda
Document #
0675
0682
0691
0705
Our Response:

General:
       As noted in the proposal and in the comments, we have already adopted delegated-
assembly provisions for heavy-duty highway engines in part 85 and for nonroad engines in part
1068 in addition to what we proposed for Small SI engines in part 1054. We have made a
comprehensive review of these various regulations to create a hybrid program that allows us to
take what we believe is a robust approach that uniformly and broadly addresses the concerns
related to the cooperative efforts of engine, equipment,  and component manufacturers in
designing and assembling certified systems. The combined approach, which incorporates
elements of each of the three programs, is written in a new §1068.261.  There is also an
abbreviated version of §1068.260 remaining to describe a framework and general provisions
related to the arrangements between engine and equipment manufacturers in taking engines
through the assembly process to reach a certified configuration.  Section 1 .xx describes the
approach we took to creating this unified program. The rest of this section describes how the
final program differs from the proposal and responds to the specific concerns related to Small SI
engines and equipment raised in the comments.

       There are three principal differences between the proposed and final regulations for Small
SI engines and equipment.  First, we are allowing distributors to participate in delegated
assembly, but distributors would need to act as equipment manufacturers, adding catalyzed
mufflers where appropriate for shipment to equipment manufacturers.  We proposed to allow
distributors to act as agents on behalf of engine manufacturers to further delegate assembly to
equipment manufacturers. We are allowing this only for the first four years of the Phase 3
standards (2011 through 2014 model years). While a more flexible approach is needed for the
transition to new standards, as described above, we believe this is not appropriate for the long
term because of concerns about the ability of engine manufacturers to ensure that engines will be
assembled in the certified configuration. As described in the comments, assembling engines
involves a very significant effort to differentiate different models and manage engines and
components coming from multiple suppliers. We believe that there is too much risk of
                                      2-113

-------
                                                            Chapter 2: Small SI Engines
miscommunication or misbehavior where a distributor is acting on behalf of the engine
manufacturer to do design work, arrange for shipment, and manage audits and other oversight
steps to ensure that potentially large numbers of very small equipment manufacturers properly
assemble engines. Given the complexity and diversity of these arrangements, we expect that the
question would be how extensive the noncompliance is, not whether there would be
noncompliant engines. Such problems would be difficult to find and, if we do discover a
problem, it would be difficult to hold any particular company accountable, given the distribution
of responsibility among the several companies. Nevertheless, we expect to learn a lot from the
experience of implementing these provisions.  If we see that manufacturers can observe the
regulatory requirements in a way that alleviates the concerns described here, we would be open
to a regulatory amendment to continue the provisions related to distributors that we are adopting
on an interim basis.

       We understand that some companies will be too  small to get an engine manufacturer to
agree to participate in delegated assembly for some or all of their equipment models. In these
cases, we believe distributors will in many cases be able to provide design support for the
equipment manufacturer. Small equipment manufacturers could benefit from a distributors
ability to participate in delegated assembly, but only to the extent that the distributor can
coordinate muffler designs with the muffler manufacturer, the engine manufacturer, and the
equipment manufacturer. Distributors often serve an important role in helping small equipment
manufacturers with system integration to properly install engines and to maximize the
performance of the equipment to match the engine's design parameters and specifications.
Allowing distributors to participate in delegated assembly would be a natural fit with this role.
We also recognize that some equipment manufacturers would have such small volumes or
distinct equipment parameters t that they would not benefit from this  limited role of distributors
in delegated assembly. As a result, these companies would need to redesign their equipment as
needed to be able to use one of the stock muffler configurations available from the engine
manufacturer or distributor.  As noted above, we believe this is achievable by the time the
transition provisions expire in 2015.

       Second, the final rule requires that audits minimally involve inspection of assembly
procedures and production records, investigation of assembled engines, and confirmation that the
number of aftertreatment devices shipped were sufficient for the  number of engines produced.
The proposal specified that an audit could include any one of these three things.  As described
above, we are concerned that insufficient oversight would lead to a situation where equipment
manufacturers assemble engines such that they are not in their certified configuration, either as a
simple mistake or to take advantage of the discretion allowed to get away with changes that
reduce costs or change design parameters for some performance  advantage.  We believe the three
activities noted are basic steps that should be part of any audit. Moreover, we specifically
identify these as minimum steps for performing an effective audit.  If we learn over time that
these steps are insufficient, either for specific manufacturers or the industry as a whole, we may
require additional auditing steps to ensure that engines are properly assembled.

       A current enforcement case highlights the need for active oversight with delegated
assembly. An engine manufacturer has been relying on  installation instructions to ensure that
equipment manufacturers install the proper air filter, which is identified specifically by part
                                         2-114

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


number in the application for certification.  It turns out that an equipment manufacturer was
found to be substituting a different air filter for some perceived advantage, either for cost or
performance, which caused the engines to be sold in an uncertified configuration. The engine
manufacturer had taken steps to make the information available to equipment manufacturers, but
this was clearly not enough to ensure that final assemblies involved only engines in a certified
configuration. In anticipation of engines using catalyzed mufflers, we see the incentive for
departing from an engine manufacturer's installation instructions only increasing. We therefore
believe that delegated assembly can be successfully done only with an active program to oversee
and document compliance with installation instructions.

       Third, we specify a different schedule for the number of audits that engine manufacturers
must perform after the first four years.  In fact, the change involves a smaller number of audits,
based on our expectation that a smaller number of equipment  manufacturers will be participating
in delegated assembly after the transition to the Phase  3 standards is complete.

       The following paragraphs respond to the individual concerns expressed in the comments.

       Written confirmation:  OPEI's suggestion is  consistent with our proposal.  The final
rule preserves this provision, not only for small businesses but for all companies.

       We believe this requirement is quite clear and viable, reflecting the need for confirmation
with the business realities of ordering and shipping engines. In particular, we believe  it is not
sufficient for equipment manufacturers to satisfy the requirement for written confirmation simply
by notifying the engine manufacturer that they are aware of the regulatory requirements.  This
requirement is in the context of a scenario in which the equipment manufacturer is separately
procuring and paying for aftertreatment devices. There is a substantial risk for engine
manufacturers to send out noncompliant products without this assurance, so we believe engine
manufacturers would want to treat the regulatory requirement as a minimum for ensuring that
their engines do not reach ultimate purchasers in an noncompliant configuration.

       Audits: In a situation where delegated assembly does  not require engine manufacturers to
include the price of aftertreatment with the  price of the engine, we are concerned that there is too
great an incentive for equipment manufacturers to deviate from the specified installation
instructions, either to reduce costs or to gain some perceived performance advantage.  As
described above, we believe  an effective  audit that minimally  includes the three elements
specified under the current regulations in §1068.260 is essential for maintaining proper oversight
of the assembly process. The application for certification should include enough information to
make clear that the certifying engine manufacturer will properly fulfill its auditing
responsibilities.

       We believe it is appropriate for engine manufacturers to follow an auditing plan that
involves reasonably objective directions for selecting equipment manufacturers.  Adding "as
much as possible" to this direction would make it meaningless. As noted by EMA, certain
equipment manufacturers will have earned  more or less confidence based on their relationship
with the engine manufacturer and their past performance.  Allowing engine manufacturers more
discretion in this regard would only allow them to delay auditing equipment manufacturers for
                                       2-115

-------
                                                            Chapter 2: Small SI Engines
which there is less confidence that everything is in order.  If engine manufacturers are
particularly concerned about any one equipment manufacturer, they should be sure to audit that
company independent of the specific regulatory requirements, or simply terminate the
arrangement for that company.

       Point of final assembly: We specified in the proposal that the exemption expires at the
point of final assembly because there is a need to avoid a situation in which a delegated-
assembly engine is introduced into commerce in an uncertified configuration where we would
not want to consider that a violation. The exemption therefore covers a shipment from the
engine manufacturer to the equipment manufacturer (or from one of the equipment
manufacturer's facilities to another). There is no need for an exemption for other internal
processes after the equipment reaches the point of final assembly, because its engine needs to be
in a certified configuration the next time it is introduced into commerce. There is no violation
for an engine that is placed into inventory at the end of the assembly line and then pulled back
for trading out exhaust components to be in a different certified configuration. Note however,
that if an EPA inspection of an equipment manufacturer's inventory of completed products turns
up engines that are not in a certified configuration, we would take steps to address the
nonconformity, as allowed under the regulations.

       Label: We agree that abbreviating "Delegated Assembly" may be appropriate, so we are
revising the regulation to allow labels with "DEL ASSY"  where space prevents the full
designation. Especially with the approach we are taking for labeling with respect to evaporative
emission families, further abbreviating the term would only be confusing or inappropriate.

       The proposed rule included labeling requirements consistent with CARB's second
recommended option.  We are adopting similar labeling requirements for the final rule, including
the option of either applying a temporary label or identifying "delegated assembly" on the
permanent label.  This ensures that the engine will be properly identified at every point in the
assembly (and shipping) process. We believe equipment manufacturers  should not be
responsible for labeling engines where they are simply assembling the exhaust system.

       Production-line testing: We agree that manufacturers should be able to maintain an
inventory of randomly selected components for testing. We have revised the regulations
accordingly.

       Class I engines: We agree that engine manufacturers may need to use the delegated-
assembly provisions for Class I engines, though this should be far less common than for Class II
engines. We are therefore preserving this provision in the final rule.

       Liability: The regulations appropriately state that engine manufacturers are liable for the
in-use compliance of every certified engine.  The delegated-assembly provisions are an option
that engine manufacturers may exercise based on their business interests and their relationships
with equipment manufacturers. Choosing to use these provisions does not change the
fundamental responsibility associated with certifying  engines, to ensure that engines comply with
the regulations throughout the useful life.  In addition, the regulations also make clear that
                                         2-116

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


equipment manufacturers are in violation if they introduce equipment into commerce without
following the engine manufacturer's installation instructions.

       If an equipment manufacturer has been found to be in violation, we specify that we may
require the engine manufacturer to discontinue the use of delegated assembly for that
manufacturer (revoking the exemption). We would generally not hold engine manufacturers
responsible for noncompliant engines where the equipment manufacturer is fully responsible for
the noncompliance.  However, we would hold engine manufacturers in violation if they
intentionally submitted false or incomplete information (voiding the exemption).

       Honda correctly notes that we are not  surrendering our enforcement authority with
respect to delegated assembly.  However, we  have described our basis for being concerned that
engine manufacturers do more than simply send incomplete engines with installation
instructions, trusting equipment manufacturers to properly complete engine assembly subject to
EPA's enforcement of applicable requirements. Delegated assembly is fully optional, so any
engine or equipment manufacturers not wanting to be subject to the required oversight functions,
or not wanting to be in a situation where confidential business information would be
compromised, may choose not to participate in delegated assembly. Engine manufacturers could
also take the middle ground, participating in delegated assembly but including the price of
aftertreatment in the price of the engine.  In this case, the regulations specify a significantly
lighter oversight burden.  Since the engine manufacturer is choosing to participate in delegated
assembly, it is unclear why there would be any thought that they should take steps to ensure that
engines are assembled properly. Third-party  auditors could do on-site visits if there is a
sensitivity regarding access to a competitor's  facilities or records.  Moreover, we specifically
state in the regulation that information  submitted between companies under these regulatory
provisions is considered to have been equivalent to a submission to EPA. The prohibitions in
§1068.101 and the corresponding civil  and criminal  penalties apply for any false information that
a company submits to another company.

       Air filters: The regulations include the clarifying language requested by EMA in which
we specify that air filters are subject to the delegated-assembly requirements only if the
manufacturer's certification depends on identifying the air filter by part number. In contrast, if
the manufacturer certifies an engine based on specified intake restrictions, the delegated-
assembly provisions do not apply. In this scenario, the engine manufacturer would still be
responsible for the in-use compliance of any engines in the engine family that were assembled
following the applicable installation instructions.

       References: We have revised the regulations such that this reference is  obsolete.

       Within-company shipments:  The final regulations include the streamlined provisions
for engine manufacturers that also manufacture equipment and install their own engines.

       Deleting the provisions related to completing production at different facilities would
disallow this practice entirely. We need to be aware of this practice and to be able to set
conditions or require specific steps to ensure that the exemption is  not abused. We therefore
need to base this exemption on EPA approval; however, we specify that the manufacturer must
                                      2-117

-------
                                                            Chapter 2: Small SI Engines
simply describe this practice in the application for certification. Approving the certification is
considered approval of the exemption.  We are therefore retaining this provision as proposed.

       Evaporative systems: Engine manufacturers must comply with evaporative emission
standards to the extent they assemble fuel-system components. They are not responsible for
further assembly of the fuel system by equipment manufacturers so there is no need for an
exemption or other provisions analogous to delegated assembly.
       2.9   Equipment manufacturer recertification

What Commenters Said:

       OPEI supported EPA's proposal to allow the re-certifying equipment manufacturers to:
1) only conduct low-hour emission testing on the "green" modified exhaust system; and 2) rely
on and apply the engine manufacturer's previously established deterioration factors. OPEI
commented that EPA has also appropriately proposed not to apply PLT testing to the re-
certifying OEM as this would overly-complicate this process without any benefits since the
engine would already be subject to PLT. OPEI commented that this re-certification provision
should be permanent and not expire.  OEMs will still require muffler certifications on a long-
term basis to produce certain critical equipment  models.

       Regarding equipment manufacturer recertification, CARB believes that such a provision
would conflict with anti-tampering regulations.  CARB commented that an alternative would be
the equipment manufacturer working with the engine manufacturer (holder of the executive
order) to include his/her variation as a running change and re-testing for a new worst-case
model/configuration. However, if EPA does adopt the provision to allow equipment
manufacturer recertification, CARB commented that EPA should require production line testing
and impose an expiration date for the program.

Letters:
Commenter
OPEI
CARB
Document #
0675
0682
Our Response:

       We agree that there may be a continuing need for equipment manufacturers to rely on the
streamlined certification proposed in §1054.612 where they rely on a catalyst from an already-
certified engine family. The streamlined certification would allow the equipment manufacturer
to assemble that catalyst in a custom muffler configuration. We believe this situation calls for a
reduced certification burden, especially for developing deterioration factors.

       We also believe that there will be a reduced need for this as time passes.  As described
above, the four-year transition program should allow time for engine and equipment
manufacturers to work out arrangements for designing and producing mufflers in compliant
                                         2-118

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


configurations.  As a result, we believe it is appropriate to limit the provisions for streamlined
certification starting in 2015. In discussion with manufacturers, there was general agreement
that an appropriate threshold would be annual sales of 5,000 units, which is already established
as the threshold for defining small-volume engine families.

       There is no violation of the tampering prohibition because the engine would never be
introduced into U.S. commerce in an uncertified configuration.

       We agree that changes coming in response to an equipment manufacturer's needs could
be factored in as a running change for the certifying engine manufacturer (with new testing as
needed).  This would require no new regulatory provisions; however, the proposed approach
addresses the situation where the engine manufacturer does not want to be responsible for the
changes called for by the equipment manufacturer.

       We will  monitor the use of this provision over time, both for its frequency of use and the
degree of compliance.  We may choose to discontinue the streamlined recertification provisions
in the future, but we believe there is enough chance that equipment manufacturers will depend on
it that it can  be appropriately applied beyond 2014 for small-volume emission families as
described above.

       2.10  Compliance provisions

   2.10.1 Warranty assurance

What Commenters Said:

       OPEI and EMA noted that the proposal implements a change in the requirements for
manufacturers to provide emission warranty service including provisions that deal with people
living more than 100 miles from an authorized service center starting in 2009 model year.  OPEI
and EMA understand the agency's concern that customers must have access to sources of
emission warranty but they do not support the prescriptive solution associated with authorized
service centers within 100 miles of every customer. It will be virtually impossible for engine or
equipment manufacturers to identify where the ultimate purchaser of a piece of equipment may
use the equipment and therefore impossible to properly identify for the agency that the
requirement has been met. The relief purported to be provided regarding sparsely populated
areas is also not viable. If any provision is required beyond the need for at least one distributor
within the United States, OPEI and EMA recommended that the  servicing dealer requirement be
linked to population centers with a 2000 U.S. Census population in excess of 100,000 people.
(See §90.1103 Emission warranty, warranty period and §1054.120(f)(3) and (4) What emission-
related warranty requirements apply to me?)

       CARB supported EPA's "Special Provisions for Compliance Assurance," and
specifically supported the provisions regarding the assurance of warranty coverage.
                                      2-119

-------
                                                             Chapter 2: Small SI Engines
Letters:
Commenter
OPEI
CARS
EMA
Document #
0675
0682
0691
Our Response:

       Certifying manufacturers must not only sell a product that meets emission standards, but
also meet obligations over a defined period of service.  The most obvious requirements related to
in-use engines are warranty and recall. We are aware that many low-cost engines are sold by
foreign manufacturers with little or no presence in the U.S. market for honoring warranty claims.
This is a violation of the regulations, subject to substantial penalties.  We believe it is very
important to take the preventive step at certification to have companies describe their plan for
meeting warranty obligations than to wait until there is a violation. The proposed approach was
an attempt to reasonably balance a consumer's need to be able to access an authorized service
center with the manufacturer's burden to maximize coverage with their repair facilities.

       We believe it is clearly necessary to require more than a single parts distributor in the
United States to expect a manufacturer to be able to provide effective warranty coverage for
consumers. We agree with the approach recommended by the manufacturers to say that they can
demonstrate adequate warranty coverage by placing authorized service centers in all U.S.
population  centers with a census count of 100,000 or more. Table 2-1 identifies 251 areas from
the 2000 census that qualify, listed alphabetically by state. We have modified the regulations to
allow for this demonstration.

       We are also aware that some companies may not sell engines throughout the United
States, in which case they would not be expected to maintain authorized service centers in all the
identified population centers. We are keeping a modified version of the proposed requirement
as an alternative to the commenters' suggestion to rely on the list of population centers.  This
would allow manufacturers to choose from a variety of methods for demonstrating an ability to
respond to warranty claims.

        We are adopting two main changes to the proposed approach related to warranty
demonstrations. First, we are clarifying that the distance from consumers is based only on the
contiguous United States.  This allows us to avoid an expectation that manufacturers maintain
multiple service centers across Alaska or in every U.S. territory.  Second, we are revising the
provisions related to sparsely populated areas. While the proposal allowed for up to 10 percent
of sales to be to owners living more than 100 miles from an authorized service center, we agree
that this would be difficult for manufacturers to implement. We are instead specifying that the
100-mile limit does not apply in states with any high-altitude areas (see 40  CFR part 1068,
Appendix III). Identifying states with high-altitude areas aligns quite closely with low
population  density.

       To the extent that the 100-mile approach or the population centers doesn't fit well
nationwide for a given manufacturer, we would also allow for a combined approach in which the
                                          2-120

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


manufacturer would rely on one method for certain states and another method for other states.
However, we would require each state to have at least one authorized service center unless the
manufacturer is able to meet the 100-mile specification without having an authorized service
center in a given state.

       Also, we proposed to apply these requirements in the 2009 model year, but we believe
the timing of the final rule dictates that we allow an additional year for manufacturers to meet
these new requirements.  We have therefore modified the regulations to require manufacturers to
comply starting with the 2010 model year.
                                      2-121

-------
Birmingham, AL
Huntsville, AL
Mobile, AL
Montgomery, AL
Anchorage, AK
Chandler, AZ
Gilbert, AZ
Glendale, AZ
Mesa, AZ
Peoria, AZ
Phoenix, AZ
Scottsdale, AZ
Tempe, AZ
Tucson, AZ
Little Rock, AR
Anaheim, CA
Antioch, CA
Bakersfield, CA
Berkeley, CA
Burbank, CA
Chula Vista, CA
Concord, CA
Corona, CA
Costa Mesa, CA
Daly City, CA
Downey, CA
El Monte, CA
Elk Grove, CA
Escondido, CA
Fairfield, CA
Fontana, CA
Fremont, CA
Fresno, CA
Fullerton, CA
Garden Grove, CA
Glendale, CA
Hayward, CA
Huntington Beach, CA
Inglewood, CA
Irvine, CA
Lancaster, CA
Long Beach, CA
Los Angeles, CA
Modesto, CA
Moreno Valley, CA
Norwalk, CA
Oakland, CA
Oceanside, CA
Ontario, CA
Orange, CA
Oxnard, CA
Palmdale, CA
Pasadena, CA
Pomona, CA
Rancho Cucamonga, CA
Richmond, CA
Riverside, CA
Roseville, CA
Sacramento, CA
Salinas, CA
San Bernardino, CA
San Buenaventura (Ventura), CA
San Diego, CA
U.S. Population Centers
 San Francisco, CA
 San Jose, CA
 Santa Ana, CA
 Santa Clara, CA
 Santa Clarita, CA
 Santa Rosa, CA
 Simi Valley, CA
 Stockton, CA
 Sunnyvale, CA
 Thousand Oaks, CA
 Torrance, CA
 Vallejo, CA
 Visalia, CA
 West Covina, CA
 Arvada, CO
 Aurora, CO
 Colorado Springs, CO
 Denver, CO
 Fort Collins, CO
 Lakewood, CO
 Pueblo, CO
 Thornton, CO
 Westminster, CO
 Bridgeport, CT
 Hartford, CT
 New Haven, CT
 Stamford, CT
 Waterbury, CT
 Washington, DC
 Cape Coral, FL
 Clearwater, FL
 Coral Springs, FL
 Fort Lauderdale, FL
 Gainesville, FL
 Hialeah, FL
 Hollywood, FL
 Jacksonville, FL
 Miami, FL
 Miami Gardens, FL
 Miramar, FL
 Orlando, FL
 Pembroke Pines, FL
 Port St. Lucie, FL
 St. Petersburg, FL
 Tallahassee, FL
 Tampa, FL
 Athens-Clarke County, GA
 Atlanta, GA
 Augusta-Richmond County,
 Columbus, GA
 Savannah, GA
 Honolulu, HI
 Boise City, ID
 Aurora, IL
 Chicago, IL
 Joliet, IL
 Naperville, IL
 Peoria, IL
 Rockford, IL
 Springfield, IL
 Evansville, IN
 Fort Wayne, IN
 Indianapolis, IN
 Table 2-1
 over 100,000 - U.S. Census, 2000*
       South Bend, IN
       Cedar Rapids, IA
       Des Moines, IA
       Kansas City, KS
       Olathe, KS
       Overland Park, KS
       Topeka, KS
       Wichita, KS
       Lexington-Fayette, KY
       Louisville-Jefferson County, KY
       Baton Rouge, LA
       Lafayette, LA
       New Orleans, LA
       Shreveport, LA
       Baltimore, MD
       Boston, MA
       Cambridge, MA
       Lowell, MA
       Springfield, MA
       Worcester, MA
       Ann Arbor, MI
       Detroit, MI
       Flint, MI
       Grand Rapids, MI
       Lansing, MI
       Sterling Heights, MI
       Warren, MI
       Minneapolis, MN
       St. Paul, MN
       Jackson, MS
       Independence, MO
       Kansas City, MO
       Springfield, MO
       St. Louis, MO
       Lincoln, NE
       Omaha, NE
       Henderson, NV
       Las Vegas, NV
       North Las Vegas, NV
       Reno, NV
       Manchester, NH
       Elizabeth, NJ
       Jersey City, NJ
       Newark, NJ
       Paterson, NJ
       Albuquerque, NM
       Buffalo, NY
       New York, NY
GA    Rochester, NY
       Syracuse, NY
       Yonkers, NY
       Cary, NC
       Charlotte, NC
       Durham, NC
       Fayetteville, NC
       Greensboro, NC
       Raleigh, NC
       Winston-Salem, NC
       Akron, OH
       Cincinnati, OH
       Cleveland, OH
       Columbus, OH
       Dayton, OH
Toledo, OH
Norman, OK
Oklahoma City, OK
Tulsa, OK
Eugene, OR
Portland, OR
Salem, OR
Allentown, PA
Erie, PA
Philadelphia, PA
Pittsburgh, PA
Providence, RI
Charleston, SC
Columbia, SC
Sioux Falls, SD
Chattanooga, TN
Clarksville, TN
Knoxville, TN
Memphis, TN
Nashville-Davidson, TN
Abilene, TX
Amarillo, TX
Arlington, TX
Austin, TX
Beaumont, TX
Brownsville, TX
Carrollton, TX
Corpus Christi, TX
Dallas, TX
El Paso, TX
Fort Worth, TX
Garland, TX
Grand Prairie, TX
Houston, TX
Irving, TX
Laredo, TX
Lubbock,  TX
McAllen,  TX
Mesquite, TX
Pasadena, TX
Piano, TX
San Antonio, TX
Waco, TX
Wichita Falls, TX
Salt Lake  City, UT
West Valley City, UT
Alexandria, VA
Arlington CDP
Chesapeake, VA
Hampton, VA
Newport News, VA
Norfolk, VA
Richmond, VA
Virginia Beach, VA
Bellevue,  WA
Seattle, WA
Spokane, WA
Tacoma, WA
Vancouver, WA
Green Bay, WI
Madison, WI
Milwaukee, WI
       *Source: U.S. Census Bureau (see http://www.demographia.com/db-usmuni2004.htm)

                                                         2-122

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


   2.10.2 Bonding

What Commenters Said:

       OPEI noted that its members are facing an enormous threat from manufacturers of non-
compliant engines - particularly as the costs increase to produce even cleaner, EPA-compliant
products. The current EPA framework is not designed with the safeguards needed to address the
imminent threat from "bad actors" with no U.S. assets. Certain off-shore manufacturers have
become very sophisticated in relying on "shell importers" in order to avoid any meaningful
enforcement exposure.

       OPEI commented that the final Phase 3 small engine regulations should require a foreign
manufacturer (that has no U.S. assets) to post a bond to cover a portion of his engines in case
they do not comply with the EPA emission standards. These bonding requirements are really the
only meaningful mechanism EPA has to take action against a "bad foreign actor" who  sells non-
compliant engines through a "shell  importer" and disappears if his non-compliant products are
discovered.  OPEI therefore supported EPA's proposed bonding requirements for foreign
manufacturers and importers with no U.S. assets to create an even and effective compliance and
enforcement program. OPEI urged EPA to pull ahead and make effective in 2007 the bonding
requirements.

       OPEI commented that it does not believe it would be necessary or appropriate to impose
such bonds on established manufacturers that have adequate U.S. assets to cover  non-compliance
events. Even with EPA's new proposed, bonding requirements, manufacturers with substantial
U.S. assets will  still have dramatically greater compliance exposure (and incur greater  costs) than
a foreign manufacturer which just submits a bond.

       OPEI commented that there should not be any other exemptions from the bonding
requirements given the difficulty in defining an objective and practical criterion for preventing
enforcement abuses.  OPEI is skeptical that EPA can develop clear and objective regulatory
language that would establish an exemption to the bonds for manufacturers that have a
demonstrated long-term record of no violations. Moreover, OPEI is concerned that many
manufacturers have previously certified engines, but not shipped any products into the U.S.
market. Thus, the fact there has not been a known prior violation does not really indicate that
such a manufacturer is a "responsible" company.  OPEI also does not believe EPA will be able to
establish clear and objective standards to exempt from the bonding requirements  either
manufacturers or importers who had been certified to voluntary industry standards for production
quality (such standards do not currently exist) or who performed voluntary in-use testing.
Deliberate "bad actors" intent on circumventing the regulations will be willing to also fabricate
their compliance with production quality standards or voluntary in-use testing.

       Euromot commented that, as importers, they accept the bonding requirements (or
equivalent U.S.  assets) and concept of a stronger market surveillance option within the proposed
regulation.
                                      2-123

-------
                                                            Chapter 2: Small SI Engines
       CARB commented that it supports U.S. EPA's "Special Provisions for Compliance
Assurance," and specifically supports the provisions regarding importation data, the assurance of
warranty coverage, and bond requirements.  The posting of bonds to cover compliance or
enforcement-related obligations for importers who have not yet proven financial stability is
crucial. Without the bonds, consumers may not be able to obtain needed warranty coverage.
Also, if the imported engines are found not to meet the standards then enforcement actions can
be made using the bond funds.  Once a company gets into good financial standing, determined
by EPA, then the company can be refunded the bond funds. Overall, CARB agreed with EPA
that a bond requirement is necessary. However,  CARB asked that in the proposed regulatory
language EPA not preclude California from adopting a similar program should CARB deem it
appropriate in the future for California certified engines.

       Briggs & Stratton commented that it supports the bonding requirements in the NPRM.  It
is imperative for the small engine industry that all manufacturers are accountable for meeting the
emission regulations, not just those located in the U.S. who are therefore susceptible to EPA
enforcement actions. Companies with U.S. assets sufficient to cover enforcement actions should
not be required to post import bonds, but companies without such U.S. assets should post bonds
to ensure uniform enforcement for all manufacturers.

       EMA  supported the bonding requirements set forth in the NPRM. Such requirements are
an important step to creating a level playing field among all competitors. Engine manufacturers
that do not have sufficient assets in the United States to avoid the bonding requirement also are
unlikely to have adequate resources in the U.S. to audit equipment manufacturer use of the
delegated assembly provisions. Manufacturers that have significant physical assets in the United
States can easily be identified, and EPA can take appropriate legal action as required when/if
there is a compliance concern. EPA does not have access to manufacturers without assets in the
United States, making it difficult, if not impossible, to take enforcement action against such
entities. EMA commented that the proposed bonding provision  correctly requires all  parties
responsible for compliance with the Phase 3 regulations to have  assets in the U.S. (whether
physical assets, or a posted bond) that may be attached in connection with an enforcement action.
If the proposed bonding provisions are not adopted, EMA commented that it is imperative that
EPA adopt another means to ensure that it has the ability to take enforcement action against
manufacturers that do not have assets located in the United States.  In addition, the enforcement
provisions associated with Part 1054 and Part 1060 apply to any party that introduces product
into commerce in the United States and EPA should exercise its  authority accordingly.

       EMA  commented on §1054.690 "What are the bond requirements for importing certified
engines and equipment?"  EMA commented that the last part of the last sentence in paragraph (a)
does not make sense as drafted.  Accordingly, EMA suggested that the sentence should be
revised to read as follows (new language is in italics): "For example, it would be a sufficient
demonstration if you show that you have manufactured or imported engines for the U.S. market
for a significant period of time without failing a test conducted by EPA officials or being found
to be substantially not in compliance with EPA regulations"

       The National Association of State Fire Marshals commented that their preliminary review
suggests that the Chinese are capable of making  a significant impact on the United States market.
                                         2-124

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


They noted that they have seen this before, with Chinese manufactured All-Terrain Vehicles
(ATVs) capturing over 30% of the U.S. market in just a few years.  ATV's which they have
tested, and provided to CPSC, failed to meet the applicable American National Standard and they
recommended that they be recalled.

       The National Association of State Fire Marshals commented that EPA's proposed Phase 3
Certification and compliance provisions are well suited for the legacy engine and equipment
manufacturers that have an established track record for meeting EPA's Phase 2 requirements.
They noted that EPA recognized the concerns with imported products, and their plans are
noteworthy. However, new entrants from China can be expected to defy these provisions. Their
experience enforcing CPSC regulations has shown that Chinese manufacturers and importers are
willing to falsify conformance with CPSC regulations and to "port shop" until entry into the
United States is achieved.

Letters:
Commenter
OPEI
CARS
Euromot
Briggs and Stratton
EMA
National Association of State Fire Marshals
Document #
0675
0682
0649
0657
0691
0673
Our Response:

       We agree with the comments noting the need for bonding provisions to ensure that
companies without substantial U.S. assets should be subject to bonding requirements to ensure
enforcement with their obligations associated with certifying engines.  Bond payments would
allow EPA to compel companies to take actions or pay penalties where there might otherwise be
no way of enforcing regulatory requirements. The bond payment would not apply for companies
with substantial physical assets in the United States, since they are inherently subject to EPA's
enforcement of regulatory requirements because we have access to the company's personnel and
facilities to compel compliance or payment of penalties.

       We also note that bonds are generally not paid in a lump sum and then refunded after
some period.  Rather, companies pay a premium to a bond agent who then opens a policy or
account with a face value equal to the amount of the bond obligation, much like an insurance
policy. Any EPA judgments against the company would generally be paid by the bond agent out
of the account. As a result, the expense for maintaining a bond is simply the regular premium for
maintaining a valid bond.

       We are not including in the regulation any provision that would preclude California from
adopting its own requirements for bond payments.  However, any bond requirements in
California would need to conform with any prevailing legal authority related to international
trade.
                                     2-125

-------
                                                            Chapter 2: Small SI Engines
       We believe that basing bond payments on adherence to industry standards such as ISO
14000 would not be effective in assigning bond responsibilities where that would be necessary or
appropriate,  as described in the comments. We also believe it would not be appropriate to
simply waive bond requirements based on some measure of compliance history, though we are
prepared to set a lower threshold as an asset test as described below.

       In discussions following the end of the comment period, manufacturers made three
recommendations regarding the implementation of a bond requirement. First, they pointed out
that there should be a minimum bond value rather than relying only on the published per-engine
bond values. This would prevent small-volume importers from being responsible for
maintaining  a bond whose value is too small to provide any reasonable assurance of compliance
or any practical ability to cover possible financial judgments if the company or its products are
found to be in violation. We believe it would be necessary to require a bond value of $250,000 if
the calculated value based on a per-engine calculation is less than that. This would ensure that
the bond would cover a violation involving eight engines (or eight days where penalties are
calculated per day). We believe  any smaller bond value would be insufficient to achieve the
objectives described above.

       Second, manufacturers suggested $10 million of physical assets as a threshold value for
determining whether a company  has enough of a presence in the United States to avoid a bond
payment. This would include any property to which the company possesses a clear title. The
value of any given property should be based on a commercial appraisal. A mortgage or other
debt obligation associated with the property would not affect the value with respect to
determining whether bond requirements apply. We believe a $10 million threshold is high
enough to avoid a situation where foreign manufacturers can make a token property investment
to avoid bond payments, without imposing bond obligations on companies that have sufficient
assets for demonstrating an ability to meet compliance and enforcement obligations. However,
we believe smaller amounts would be appropriate for secondary engine manufacturers, where the
capital  investment for a given level of engine production may be much smaller as a result of the
business practice of buying engines that are already nearly complete.  We therefore believe $6
million is an appropriate threshold for secondary engine manufacturers. Also, we are aware that
there is a reduced need for bond payments where  companies have a consistent record of meeting
their certification and compliance obligations. As such, we believe a reduced threshold of $3
million in U.S.-based assets is appropriate for companies that have certified for the previous ten
years without being found in noncompliance.

       Third, manufacturers pointed out that the bond payment should not be a condition of
certification. We  agree that manufacturers should not be required to post a bond before they
certify their engines. However, we believe it is necessary for companies to describe in the
application for certification why  they should be exempt from the bonding requirements, if
applicable.  This would allow us to take any appropriate steps to verify claimed assets before
importation, rather than trying to correct a problem after a violation occurs. If bond payments
are required  for a given manufacturer, the bond would need to be in place for any 2010 model
year engines introduced into U.S. commerce on or after January 1, 2010.
                                         2-126

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


   2.10.3 Restricted model year

What Commenters Said:

       Regarding restrictions related to naming model years, CARB commented that it believes
it is reasonable to require that model year engines and equipment may be at most one year earlier
than the calendar year of the importation during the change of the emission standards.
Whichever requirements EPA chooses to adopt, CARB recommended that procedures be
adopted to prevent any stockpiling of engines that could be used to circumvent the regulations.

       EMA commented that the proposal's requirement that imported engines be identified by
either model year of importation or one model year earlier is a viable and appropriate approach
to preventing the stockpiling of older engines/equipment. However, given the seasonal nature of
lawn and garden products, there are limited situations where the one year limitation could be too
restrictive. Accordingly, EMA commented that EPA should give itself the authority to extend
the time frame in special circumstances. (See  §90.616 and §1054.695(b).)

       In later comments, EMA suggested that we allow an additional year for products that
were produced in the United States, exported, and subsequently are imported again into the
United States.

Letters:
Commenter
CARB
EMA
EMA
Document #
0682
0691
0808
Our Response:

       We are adopting the proposed provisions, as supported by the comments. We are not
adopting the suggested allowance for approval under unique circumstances to allow a longer
time frame to import products from earlier model years. We believe any such provision would
invite any number of requests, each with unique circumstances.  It is difficult to imagine a test
that would allow us to establish a threshold that would appropriately differentiate legitimate
requests from those that could or should have been avoided. In contrast, we believe the one-year
allowance provides a generous amount of time to complete production for filling orders and
shipping products to the United States.

       While the allowance is for a one-year difference between calendar year and model year, it
is important to clarify that 12 months is the minimum time interval that would apply.  This would
be the case, for example, for an engine produced in December 2009 with new emission standards
applying for the 2010 model year. Manufacturers would then have twelve months for shipment
to  an equipment manufacturer for installation and importation into the United States (or
importation of the loose engine). Especially with the awareness that new emission  standards
have taken effect, we believe this presents a reasonable deadline for manufacturers  to complete
their production and shipping to get products into the United States.  If manufacturers end their
                                      2-127

-------
                                                             Chapter 2: Small SI Engines
model year before December of a given year, that would provide an additional margin for
importing products by the end of the following calendar year.  For products manufactured before
the end of the model year, there would also be a correspondingly longer time until the
importation deadline would apply.

      We are aware that seasonal products may pose unique challenges. However, we are not
adopting a requirement that products reach the ultimate purchaser by the end of the calendar year
following the named model year. Rather, these products must simply be imported before the
deadline applies. We would expect manufacturers, distributors, or dealers to maintain their
normal inventories of unsold products at their facilities within the United States without regard to
the importation deadline described here.

      We believe it is also not appropriate to modify the regulation to accommodate products
that are exported and are later imported. We believe this represents a rather unusual scenario,
since it would be limited to products that are certified and labeled for current EPA standards
even though they are  exported. The engines or equipment would then need to be unused for
more than a year before being sent back to the United States. Adopting such an exception would
likely also be contrary to policy requirements related to international trade,  since it would apply
preferential treatment to domestically produced engines.

       See Section 1.5.2 for a discussion of issues related to stockpiling engines  and equipment.

   2.10.4 Adding or changing governors

What Commenters Said:

      EMA noted that the majority of all nonhandheld engines in this category have speed
control governors, including engines used in small utility vehicles and go-carts.  Because such
engines  have a high potential for over speed (operation at a speed higher than the intended design
of the engine), such governor systems are critical to the safety and structural integrity of these
engines.  Parties that modify engines to replace or eliminate the use of an engine  manufacturers
speed control governor should be considered the manufacturer and should be held responsible for
all aspects of the resulting product, including emissions compliance. In cases where an engine
modification is an engine manufacturer approved configuration, the engine  manufacturer must
include this configuration in its determination of a worst case emission configuration for
certification.  Accordingly, EMA commented that no additional compliance determination
should be required.

      EMA commented on §1054.650 "What special provisions apply for adding or changing
governors?" EMA noted that this section states that the special provisions in the  section apply
for engines that will not have constant-speed governors when installed in equipment. However,
there is no definition of what constitutes a "constant-speed governor." Accordingly, EMA
commented that EPA must provide such a definition in order to provide manufacturers with the
ability to determine when the special provisions apply.
                                         2-128

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
Letters:
Commenter
EMA
Document #
0691
Our Response:

       We disagree with the comment suggesting that manufacturers simply include the range
of governor strategies in an engine family by testing the worst-case configuration. The duty
cycles we specify address only constant-speed engine operation. This is typical of generators,
lawn mowers, and most other types of equipment. However, there are certain applications for
which there is a governor to prevent overspeed but otherwise allows for operation at a wide
range of engine speeds. It is not possible for manufacturers to consider the in-use operation of
these variable-speed engines as part of its certification demonstration because we provide no
standardized procedure for quantifying the emissions effect of this different operation.  We have
adopted a requirement in §1065.10(c)(l) to address this kind of mismatch between an engine's
in-use configuration (or operation) and that reflected in the certification test; this requires
manufacturers to notify us of the mismatch and allows us to work out an alternate testing
regimen to reconcile the discrepancy.  We believe it is better to address this scenario directly in
the  regulations rather than attempting  to resolve it over time under the provisions of
§1065.10(c)(l). We could adopt a unique duty cycle for variable-speed engines. However, we
believe these engines make up a very  small portion of overall sales of Small  SI engines and that
it is therefore more appropriate at this time to require manufacturers to make an engineering
demonstration that emission controls continue to work effectively at different engine speeds.  We
may pursue a different duty cycle in a future rulemaking if we find that this approach is not an
effective  way of addressing the concern.

       We agree that we need clarifying language to make clear what the regulation means by
referring  to constant-speed governors  and have revised the language accordingly.

       We also agree with EMA's suggestion to disallow removal or modification of installed
governors without recertifying the engine. We have revised the language in  §1054.650 to reflect
this change.

    2.10.5 Competition exemption

What Commenters Said:

       OPEI commented that it agreed with EPA's reasoning and logic for determining what a
"competition" engine is and how to apply for exemptions for their sale and use.

       Briggs and Stratton noted that  in the current small engine regulations (40 CFR Part 90) an
engine "Used solely for competition"  is defined as ". .  . exhibiting features that are not easily
removed  and that would render its use other than in competition unsafe, impractical, or highly
unlikely" (40 CFR Part 90.3).  In the Phase 3 proposal EPA is taking a different approach as
described in the preamble on page 28140 in the Federal Register. The engines must meet all four
of the listed criteria to be considered exempt based on use solely for competition.  In order for
                                      2-129

-------
                                                            Chapter 2: Small SI Engines
new engines to be exempt per §1054.620 an engine manufacturer would have to annually apply
for the exemption and provide the information as required by EPA.

       Briggs and Stratton raised the following specific issues with regard to the proposal:

     1. Manufacturers that make engines specifically designed for competition have made
        investments to develop a product to comply with the criteria under existing regulations.
        The proposed regulations in §1054.620 create additional certification requirements,
        business limitations, and recordkeeping burdens in addition to the investment already
        made to comply with the current regulations. Briggs and Stratton suggested that the
        regulations allow engine makers to either meet the new criteria in §1054.620(c) or the
        existing criteria, which is: "Used solely for competition means exhibiting features that
        are not easily removed and that would render its use other than in competition unsafe,
        impractical, or highly unlikely".

     2. The criteria in §1054.620(c) are written assuming  that only professional racing teams
        use small engines for competition. However, amateurs competing in sanctioned events
        do much of the competitive racing using small engines.  Therefore, Briggs and Stratton
        commented that the limitation for sale to the general public in §1054.620(c)(l) is  not
        practical and this requirement should be deleted.

     3. The requirements to "document the ultimate purchaser" and "any equipment
        manufacturers requests for an exempted engine" in §1054.620(g) are not practical. As
        discussed above, amateurs that purchase engines through dealers serving the racing
        market perform much of the racing in sanctioned events. Dealers do not necessarily
        build the equipment but supply the parts used by amateur racers and engine/equipment
        builders that serve the racing market. Briggs and Stratton commented that
        §1054.620(g) should be modified to read: "If we request it, you must provide any
        information we need to determine whether the engines are used solely for competition.
        This would include  any documentation regarding the number of engines and a list of
        the engine manufacturers' customers for these engines.  Keep these records  for five
        years."

     4. Section 1068.235 allows engines to be modified for competition after they are placed
        into service, to  be modified without request, and no record keeping of these engines is
        required by the original engine manufacturer. Briggs and Stratton commented that
        §1068.235 should clarify that this exemption should not be used to circumvent the
        requirements of 1054.620.

Letters:
Commenter
OPEI
Document #
0675
                                         2-130

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


Our Response:

       The existing definition under part 90 is very broad.  Under the current program,
manufacturers would need to show only that an engine or equipment has features that make
noncompetition use impractical or unsafe. We believe this allows far too much discretion for
manufacturers to claim product as being limited to competition purposes. There is also not any
process under part 90 for EPA to review these determinations. We proposed a set of qualifying
criteria and limitations and a corresponding process to approve a manufacturer's use of this
exemption. We believe these changes are needed to prevent exempted or excluded competition
engines from being used for noncompetition purposes. The proposed provisions are very similar
to those proposed or adopted in our other programs.

       The proposed criteria to qualify for the competition engines are explicitly not limited to
professional racing teams.  We proposed to allow for sales to "professional competition teams,
professional competitors, or other qualified competitors." We also proposed an approval process
in which we could approve a competition exemption for manufacturers who could provide clear
and convincing evidence that an engine would be used solely for competition even if not all the
proposed criteria would apply.  With respect to displaying competition models for sale to the
general public, we believe it is important to avoid a situation where "unqualified competitors"
are led to believe that they can purchase competition engines.  It is therefore appropriate to keep
the proposed limitation to prevent the "display for sale" of competition models.  Allowing
manufacturers to offer competition models for sale to the general public would prevent EPA and
manufacturers from ensuring that purchasers will limit their use of these engines to sanctioned
racing events.  Manufacturers or dealers may display competition models to promote
noncompetition models where it is clear that the competition models are not for sale to the
general public. Qualified competitors should  not be dependent on a manufacturer's marketing to
the general public to be able to find the engines and parts they need. We have modified the
regulation to clarify that competition engines may be displayed at a public dealership, though
they may not be displayed as a sales item.

       We agree with the suggestion to clarify that the allowance to modify certified engines to
be used solely for competition should not be used to circumvent the requirements that apply
under §1054.620 or similar provisions in other standard-setting parts. We have modified the
language in §1068.235 to reflect this change.

   2.10.6 Alternate fuels

What Commenters Said:

       OPEI and EMA supported EPA's proposal that parties converting engines from a
certified configuration to a non-certified configuration (i.e., from gasoline to propane) be
required to certify the final product.  OPEI and EMA commented that such parties should also be
required to either remove or cover the original certified engine manufacturer emission
compliance label with their own emission compliance label. As prescribed by the regulation, the
party that certifies the final product should assume all responsibility for emission warranty, either
directly or by contract with the original engine manufacturer.
                                      2-131

-------
                                                            Chapter 2: Small SI Engines
       EMA commented on §90.1003(b)(3)(i) and said that the language does not make sense
and must be clarified.  Significant components removed in the conversion process, such as
carburetors, are not reinstalled but replaced in the conversion process. EMA also commented on
§90.1003(b)(3) (ii) and believes the reference to §1054.635 is incorrect.  EMA commented that
the correct reference should be to §1054.645.
Letters:
Commenter
OPEI
EMA
Document #
0675
0691
Our Response:

       The regulatory language in §90.1003 already refers to replaced components, as suggested
in the comment. However, we believe the wording should be revised to address this confusion.
We have therefore revised the language in §90.1003 accordingly. The revised language also
includes a corrected reference to §1054.645.
   2.10.7 Hardship exemption

What Commenters Said:

       OPEI strongly objected to EPA's suggestion in the preamble that the proposed hardship
relief measures in the Phase 3 regulation could somehow moot the independent need for the
equipment-transitional flexibility program described above.  Both proposed elements are critical
to the industry and to the effective implementation of the final program. Moreover, there is
substantial risk and uncertainty that EPA would not grant hardship relief requested by an
individual OEM, at least until it is too late.  By the time a manufacturer is in such duress that he
can demonstrate and obtain hardship relief, it will typically be too late for him to make the
needed production changes to avoid substantial economic injury.

       In its other regulatory programs, EPA has never indicated that the hardship relief was
linked to, or somehow mooted the need for, the much broader, existing transitional flexibility
programs for equipment manufacturers. This is because the hardship relief provisions are limited
to extraordinary circumstances and  require substantial administrative time and effort to obtain.
For example, both the diesel engine regulations and the general provisions applicable to diesel
engines, large spark-ignited engines (LSI), snowmobiles and off-road motorcycles include an
independent hardship relief variance request for non-integrated  equipment manufacturers.  See
§89.102(f); and §1068.255. For example, the Tier III and Tier IV diesel regulations allow for an
additional 70% allowance for OEMS that demonstrate hardship relief.  See §89.102(f) and
§1039.625(m).
                                         2-132

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
Letters:
Commenter
OPEI
Document #
0675
Our Response:

       EPA agrees that the TPEM program and hardship provisions are both needed for the
Phase 3 program.  The hardship provisions are intended to help manufacturers that are facing
economic hardship as a result of not being able to comply with the new standards.  The criteria
for qualifying for hardship are set at a relatively high level, which would likely be difficult for a
manufacturer to demonstrate if they were having difficulty redesigning only a few of their
equipment models. The TPEM program allows an equipment manufacturer to deal with the
models which are difficult to redesign without having to demonstrate that the company would
experience hardship without the relief.  Therefore, EPA agrees that both the TPEM program and
the hardship provisions are needed and is retaining both of them for the Phase 3 program.

   2.10.8 Stockpiling provisions for  engine manufacturers

What Commenters Said:

       EMA shared that there is a general understanding that the inventory allowances described
in §90.1003(b)(4) apply equally to engine manufacturers  and equipment manufacturers. They
also pointed out that it is not uncommon in the Small SI engine business for OEM's to order
engines based on sales projections and then return engines or cancel  orders after the engines are
built if market conditions  change.

       In response to draft language that would clarify the extent to  which we would
accommodate extended inventories for engine manufacturers, EMA  commented that this
approach seemed acceptable, with a remaining concern that the provision should not focus on
small engine families.  Engine families can consist of a wide variety of engine
models/configurations.  A high-volume family may include all the various models a
manufacturer produces of vertical-shaft single-cylinder engines with a given displacement. The
various models or customer-specific features may be as significant as a different crankshaft or as
minor as a different styling element.  Just because the family is high-volume doesn't mean that
engines with a specific customer feature will not be stranded due to unforeseen changes in the
market. Changing engines once they are manufactured and placed into inventory range from
moderately expensive (trading out external parts) to ridiculously expensive (exchanging
crankshafts). EMA suggested the regulatory language should state: "We will generally allow
maintaining extended inventories  only for unforeseen changes in market demand."

Letters:
Commenter
EMA
Document #
0817
                                      2-133

-------
                                                            Chapter 2: Small SI Engines
Our Response:

       The issue raised by the commenters is being addressed by adding §1068.103(f) to
explicitly prohibit stockpiling engines when new emission standards take effect and adding
§1054.601(b) to explain how §1068.103(f) will apply for Small SI engines. The provisions of
the new §1068.103(f) clarify that what is prohibited is for engine manufacturers to deviate from
normal production and inventory practices to stockpile engines with a date of manufacture before
new or changed emission standards take effect.  This recognizes that typical production practices
for most engine manufacturers involve engines remaining in the manufacturer's inventory for
some time. For most engines (especially for larger engines), since it is not economical to
maintain a significant number of engines in inventory long after the end of a model year, this
inventory time would typically be no more than a few weeks.

       However, Small SI engines can be kept in inventory for much longer times, especially for
small volume engine models. Manufacturers noted other possible cases for such extended
inventories.  In response to these concerns, we are adding §1054.601(b) which describes how
§1068.103(f) will apply for Small SI engines. This provision does not preclude manufacturers
from keeping engines in inventory for long times.  However, in recognition that normal Small SI
practices can include keeping some engines in inventory for a very long time, §1054.601(b) will
require that manufacturers obtain our approval to keep any engines in inventory for longer than
12 months. Such manufacturers would be required to show that keeping such extended
inventories is consistent with its normal business practice.  In addition, given the lead time
provided when we adopt new standards,  we are requiring the manufacturer to demonstrate that
the extended inventory (beyond  12 months) is also necessary and could not have been avoided
through prudent planning.  Consider the following examples:

       Example #1 -the manufacturer normally keeps certain small volume engines in inventory
      for up to three years.  In this case, the manufacturer would need to plan its production run
       of  such engines so that it reasonably expected to not keep any of the engines in inventory
       for more than 12 months after the new standards took effect.

       Example #2 - the manufacturer normally keeps engines in inventory for up to six months.
       In this case, the manufacturer could keep the engines in inventory for up to six months
       after the new standards took effect without seeking EPA approval.

       Example #3 - the manufacturer normally keeps engines in inventory for up to ten months,
       but receives a return of a large number of engines (unforeseen but consistent with its
       normal business practice) so that it will not use up its inventory for an additional four
       months.  In this case, the manufacturer could keep the engines in inventory for up to 12
       months after the new standards took effect without seeking EPA approval. Engines
       remaining in inventory after 12 months could be scrapped, sold as replacement engines,
       exported, or covered under another applicable exemption. Alternatively, the
       manufacturer could ask to be allowed to sell the engines under its original certificate
       beyond the 12 month period.
                                         2-134

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       It is worth noting that this 12 month limit is consistent with the provisions of §1068.360
which prohibit the importation of new engines and new equipment in any calendar year that is
more than one year after the named model year.

   2.10.9 Other issues under part 1068

What Commenters Said:

       OPEI generally supported the proposed application of the Part 1068 Compliance
Provisions to small engines. OPEI generally supported EPA's efforts to modify Part 1068 to
accommodate the application of evaporative standards which create different compliance
obligations - depending on whether the OEM certifies or merely installs a previously certified
evaporative component.  (OPEI commented that they would like to work with EPA to further
simplify/clarify this program so the component suppliers and OEMs can more readily understand
their obligations and liabilities.)

       In response to EPA's request for comment on applying these proposed requirements for
engine rebuilding and maintenance to the engines and vehicles subject to this rulemaking, OPEI
commented that it believes EPA may be creating burdens on industry segments unaware of this
rule and incapable of providing the amount of burdensome records required by this part.  OPEI
commented that EPA should exempt engines/equipment subject to part 1054 from this provision.

Letters:
Commenter
OPEI
Document #
0675
Our Response:

       We agree that the provisions in part 1068 can and should apply broadly to engine
categories, including Small SI engines.

       We also agree that the recordkeeping provisions related to rebuilding should apply
differently for handheld and Class I engines. Commercial rebuilding for these engines is quite
rare. We are concerned that applying the recordkeeping requirements for these engines will not
be very meaningful for EPA's oversight, and rebuilders could in many instances be unaware that
their service has reached a point that would qualify as a rebuild and that recordkeeping
requirements would therefore apply. These engines also generally have very simple systems for
controlling emissions, so there is less of a need to carefully document part numbers for replaced
components and other related records.  We are therefore modifying the regulations to waive the
recordkeeping requirement for engines with displacement below 225 cc. Note, however, that the
underlying requirement to rebuild engines to the original certified configuration continues to
apply.  This requirement is  simply an elaboration of the general prohibition against tampering
with certified engines. Even small businesses rebuilding small numbers of small engines should
not be exempt from the tampering prohibition.
                                      2-135

-------
                                                            Chapter 2: Small SI Engines
       In contrast, Class II engines (at or above 225 cc) are substantially more expensive and are
much more likely to be used in commercial applications where commercial rebuilding can be
expected to extend the engine's operating life. We believe that commercial entities rebuilding
these engines can be expected to maintain a standard business practice involving more careful
documentation of their work.

       As described in Section 1.5.5, we believe this distinction for rebuilding engines below
225 cc should apply equally to all spark-ignition engines (including recreational vehicles and
outboard marine engines).

       2.11  Small business issues

What Commenters Said:

       Although ECO believes that small volume engine manufacturers require flexibility to
remain competitive in the market, ECO commented that it does not agree a complete pass on
PLT testing is the correct approach. Instead,  ECO encouraged EPA to develop an approach that
maintains the integrity of the certification compliance process, while providing small volume
manufacturers the flexibility needed to remain competitive. As a minimum requirement, ECO
commented that at least one engine per family, per year, be tested to demonstrate ongoing
compliance of production engines. As a second alternative, ECO suggested that EPA allow
small volume engine manufacturers to utilize the use of alternative testing methods (portable
emissions analyzers) to demonstrate in-use field testing compliance for production units.

Letters:
Commenter
ECO
Document #
0712
Our Response:

       As part of the process of developing provisions for small businesses during the proposal,
EPA identified 10 small businesses that are also small SI engine manufacturers. Based on
estimated sales from the certification records, there companies represent less than 0.5% of small
engine sales.  The cost of performing testing for a PLT program are significant, especially for
small companies that typically do not have their own emissions facility and must test at an
independent lab. Even if we were to allow use of a portable system, the cost of such systems are
still fairly expensive for the limited testing they would be used for.  Due to the cost of running a
PLT program and limited emission impact such a program could potentially have, we continue to
believe that small volume engine manufacturers should be exempt from PLT testing.

       2.12  Other issues

   2.12.1  In-use testing

What Commenters Said:
                                         2-136

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       NACAA noted that data available in the EPA docket indicates in-use compliance failures
by various models of lawn and garden equipment.  This has been a continuing concern of
NACAA and is heightened by the fact that EPA did not propose a mandatory in-use testing
program for these engines. NACAA urged EPA to consider the addition of such an in-use
testing program, consistent with the requirements for outboard and personal watercraft engines,
to ensure in-use performance at the levels envisioned by the regulation.

       The Pennsylvania DEP noted that EPA has not proposed an in-use testing program for
small spark-ignition engines despite the fact that in other recent proposals, EPA has treated in-
use compliance as an important part of EPA's program for ensuring performance throughout the
useful life.  The Pennsylvania DEP commented that EPA should consider an in-use compliance
program.

       The MARC AQ Forum commented that the rule should establish a testing program to
ensure that small engine emissions controls do not fail prematurely.

       NESCAUM commented that it is essential that the engines affected by this rulemaking
meet the applicable standards for the entire useful life of the equipment into which they are
installed. Consequently, they believe the proposed requirements for verifying durability of
emissions controls are  inadequate, principally because there are no requirements for in-use
emissions testing. Consistent with the durability requirements pertaining to OB/PWC engines,
NESCAUM urged EPA to incorporate similar requirements for manufacturers of small SI
engines and equipment, including a robust in-use testing program.

       The Wisconsin DNR commented that EPA should consider the addition of a mandatory
testing program for various models of lawn and garden equipment, to ensure in-use performance
at the levels envisioned by the regulation.

       OPEI noted that handheld  engines are very difficult to test. OPEI requested that EPA
provide more detail in  §1054.401  of the regulations. For example, OPEI asked whether EPA
will use the same test method and fuel for an in-use test as for certification. In addition, they
asked if EPA will use the same fixtures the manufacturer used.  OPEI suggested that language be
added stating that EPA would test at the manufacturer's facility or request such fixtures from the
handheld engine manufacturer.

       EMA commented on §1054.401 of the regulations. They believe this section should
clarify that EPA will use the same test method and fuel as used for certification by the engine
manufacturer.  Accordingly, EMA commented that this section should be revised to read as
follows: "We may perform in-use testing of any engine or equipment subject to the standards of
this part using the test procedures  and test fuels utilized by the manufacturer during the
certification process."
                                      2-137

-------
                                                            Chapter 2: Small SI Engines
Letters:
Commenter
NACAA
Pennsylvania DEP
MARC AQ Forum
NESCAUM
Wisconsin DNR
OPEI
EMA
Document #
0651
0676
0696
0641
0663
0675
0691
Our Response:

       In response to the comments recommending an in-use test program for Small SI engines,
EPA is not adopting such a program in the final rule.  EPA did not propose an in-use test
program as part of the proposal and therefore it is difficult for us to adopt such a requirement
without a chance for people to comment on the specifics of an in-use testing program.  Given the
large numbers of engine designs currently certified, and the wide range of applications into
which those engines are placed, designing a testing program to gauge the performance of in-use
engines and equipment would not be an easy task. Plus, there could be significant costs
associated for manufacturers in running such a program depending on how the program is
designed.  EPA believes an in-use program for Small  SI engines is something that should be
given full consideration as part of a future rulemaking.

       While an in-use test program could be a useful tool to determine whether in-use
engines/equipment are complying with the standards, it is not the only way. In addition to
certification testing, EPA requires manufacturers to perform production line testing to
demonstrate that engines coming off the production line are emitting at the expected levels.
Furthermore,  EPA has the authority to perform selective enforcement audit (SEA) testing where
engines coming off the  production line are tested with EPA in attendance for the testing.  Finally,
EPA recently initiated its own on-going confirmatory test program that is expected to test a wide
range of small engines in the coming years (not necessarily including engines that have already
been placed into service). While none of these programs on their own can ensure engines will
meet the standard in use, each will help to encourage manufacturers to produce well-designed
engines that continue to meet the emission standards throughout their lifetime.

       In regard to the  comments that EPA should provide more details on how it would
perform its own in-use testing, EPA has made one change to §1054.401 of the regulations.  The
regulation notes that EPA will consult with the manufacturer as needed to be able to perform a
valid emission test. To the extent that engines can't be tested without unique fixtures or other
approved "special test procedures" (see §1065.10(c)(2)), we would generally duplicate the
methods used by the manufacturer for certification testing.  This could involve testing at the
manufacturer's facility  or at any test facility we designate.  This intent to duplicate the
manufacturers' procedures does not apply for approved "alternate test procedures" for in-use
testing (see §1065.10(c)(7)).  Alternate test procedures are  approved by EPA because they are
expected to result in emission levels similar to what would result from the standard test
                                         2-138

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


procedure.  Therefore, although we may choose to do so, EPA sees no reason to commit to using
an "alternate test procedure" for testing in-use engines.
       With regard to test fuel used for in-use testing, EPA has made a change to the regulatory
provisions. As described in Section 2.5.4, we are finalizing provisions that will allow
manufacturers to use a 10 percent ethanol blend for certification testing for exhaust emissions
from nonhandheld engines, as an alternative to the standard test fuel (Indolene). This option to
use a 10 percent ethanol blend for certification of nonhandheld engines will begin with the
implementation date of the Phase 3 exhaust standards and would apply to production-line testing
as well. We are also committing to using a 10 percent ethanol blend for all confirmatory testing
we perform for nonhandheld engines certified on the ethanol blend, under conditions specified in
Section 2.5.4. Our commitment to test on an ethanol blend for those nonhandheld engines
certified on an ethanol blend has been noted in §1054.501 of the regulations.

       For handheld engines, we are not committing to using the same fuel as the manufacturer
used for certification testing. EPA would expect to use Indolene for all in-use testing of
handheld engines, although we could decide, at our own discretion, to do exhaust emissions
testing using the certification fuel  used by the manufacturer.

       With regard to the fixtures used for testing handheld engines, EPA has not made any
changes to the regulations.  For any in-use testing, EPA would expect to contact manufacturers
to ensure that we are testing engines in a manner that is appropriate for operating the equipment
on an engine dynamometer. While this may include requesting a fixture from the engine
manufacturer, EPA does not believe this will always be necessary and will not commit to doing
so at this time.
   2.12.2 Voluntary green labeling program

What Commenters Said:

       NESC AUM commented that they support the concept of a "green labeling" program, as a
means to make consumers aware of which engines exhibit especially clean emissions
performance as consumers make their equipment choices. In the Phase 2 rulemakings for
handheld and nonhandheld SI engines, EPA committed to "pursue the development of a
voluntary green labeling program for small SI engines as a non-regulatory program."
NESCAUM noted that more than eight years have now elapsed since EPA made this
commitment and as yet, there is no such program.  NESCAUM urged EPA, through this
rulemaking, to renew its commitment to work with stakeholders to develop a green labeling
program.

Letters:
Commenter
NESCAUM
Document #
0641
                                      2-139

-------
                                                            Chapter 2: Small SI Engines
Our Response:

       EPA is not prepared to commit to developing a voluntary green labeling program for the
Phase 3 standards at this time. In several previous rulemakings, EPA has adopted provisions
allowing manufacturers to certify to "Blue Sky" standards in the nonroad diesel, marine diesel,
and large SI categories. These Blue Sky standards are more stringent than the regularly
applicable standards and allow manufacturers to note such compliance on the engine label.
While we have had such standards in place since 1998, no manufacturer has yet certified an
engine under these standards.  Therefore, while we could consider a voluntary labeling program,
we are not convinced that manufacturers are interested in participating in such a program. While
EPA could pursue a voluntary program in the future, we are not committing to developing a
program for the Phase 3 standards in this rule.

   2.12.3 Miscellaneous Issues

What Commenters Said:

       EMA commented that §1054.15(b) "Do any other regulation parts apply to me?" states
that Part 1065 describes procedures and equipment specification for testing engines. However,
Part 1065 only provides this information regarding exhaust emission testing, not evaporative
emission testing. Accordingly, EMA commented that this section should be revised to read as
follows: "Part 1065 of this chapter describes procedures and equipment specifications for
exhaust emission testing engines. Subpart F of this . . ."

       EMA noted that §1054.101(b) states that HC and NOx exhaust emissions are optional for
wintertime engines.  However, §1054.101(d) states that two-stroke snowthrower engines may
meet exhaust emissions standards that apply to handheld engines with the same  engine
displacement. In order to avoid any confusion between the requirements set forth in
§1054.101(b) and (d), EMA commented that §1054.101(d)  should be revised  to read as follows:
"Notwithstanding the provisions of subpart (b) of this part, two-stroke . .  ."

       EMA commented on §1054.205(a) "What must I include in my application?" EMA
commented that it is not clear what information is required for engine families where the
certification test engine has a maximum modal power in excess  of 15  kW. Accordingly, this
section should be revised to read as follows: "For each engine family  in which the maximum
modal  power of the emission-data engine is at or above 15kW, provide the nominal brake power
for engines included in the engine family as described in 40 CFR Part 1054.140."
       EMA commented on §1054.235(e) "What exhaust emission testing must I perform for
my application for certification of conformity?" EMA noted that pursuant to this section, EPA
may require a second engine to be tested. However, the section fails to define how the "official"
results of such testing will be determined.  EMA recommended that EPA's current practice -
which is to use the average of the results obtained - be included in the final rule.
                                         2-140

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       G. Alcock commented that there is a very important overriding consideration regarding
all leaf blowers. The particulates sent into the air in the process of'blowing' far outweigh the
combustion output as regards overall pollution. Leaf blowers should be banned entirely. Leaf
vacuums are far more efficient and could even be restricted to electrical power sources. All
homes have external electrical outlets. To limit the financial business loss of leaf blower
manufacturers (which should not be the criteria by  which laws are considered) would be the
conversion of leaf blowers to vacuums. Innovation would allow designs for these modifications
and a much cleaner environment would result.  In Arizona's attempts at legislation, the ban was
immediately thrown out because it would hurt the manufacturer of leaf blowers. This is the tail
wagging the dog. They said use of leaf blowers would be limited to high pollution days. Every
time a leaf blower is used the local area becomes a high pollution day.

Letters:
Commenter
EMA
G. Alcock
Document #
0691
0601
Our Response:

       We agree that §1054.15(b) should be changed to focus on exhaust emissions. The
regulations have been changed accordingly.

       We agree that §1054.101(b) and (d) from the proposal need to be reconciled. We have
combined these into a single paragraph and added the clarification that the handheld HC+NOx
standards apply to the two-stroke snowthrower engines if they are certified to the handheld
standards.

       We believe the proposed requirement to identify maximum engine power for engines
with maximum modal power over 15 kW is exceptionally clear.  Maximum engine power is a
defined term (see §1054.140), as noted by EMA's comments on that subject. Maximum engine
power is the parameter used to determine whether engines are subject to the requirements of part
1054, so any other information would not be suitable for identifying the engine family in
§1054.205(a). Note that we are revising the regulation to require reporting of maximum engine
power for engines with displacement at or below 1000 cc only if maximum modal power is at or
above 25 kW.

       We disagree with EMA's suggestion that we should specify in §1054.235(e) that the
results from a second engine tested by the manufacturer should be averaged with the results from
the first engine to determine the official result for the engine family. The regulations at
§1054.240(a) specifically state that all engines tested for certification need to comply with
emission standards.  Allowing the averaging approach would allow manufacturers to have a test
engine with emissions above the standard that is offset by an engine from the same family that
has lower emissions. This is clearly incompatible with the principle that the test engine needs to
represent the worst-case configuration and  that every engine produced under the engine family
must meet emission standards. This is consistent with the current regulations at §90.104(a),
which also require that all test engines meet applicable standards.
                                      2-141

-------
                                                            Chapter 2: Small SI Engines
       We understand that there are certain quality-of-life concerns regarding the use of Small
SI engines. We encourage the responsible use of leaf blowers and other types of equipment that
may be operated in neighborhoods or in other areas where people may be sensitive to the use of
such equipment. However, the Clean Air Act directs us to set emission standards for these
products without giving us the authority to limit the use or sale of these products.
                                         2-142

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments



3   Exhaust Emission Standards and Related Requirements for
    Marine SI Engines

What We Proposed:

       The comments in this section generally correspond to Sections III, IV, and VII of the
preamble to the proposed rule, where we describe the proposed emission standards and
certification procedures associated with exhaust emissions from Marine SI engines.  The
applicable regulatory provisions for these proposed requirements are in 40 CFR part 1045.  The
Regulatory Impact Analysis describes the feasibility of these standards, special provisions that
apply to small businesses, and alternative standards under consideration in Chapters 4, 10, and
11, respectively.  There are also several technical amendments to the regulatory provisions in 40
CFR part 91.

   See Chapter 1 of this document for a discussion of issues that apply more broadly than only
for Marine SI engines. See Chapter 4 of this document for a discussion if issues related to
evaporative emissions.

  3.1   Scope and applicability

3.1.1   Differentiating Small SI and Marine SI engines

       See Section 3.12.3 for a discussion of issues related to installation of certified Small SI
engines in marine vessels.

3.1.2   OB/PWC and SD/I definitions

What Commenters Said:

       NMMA and Brunswick commented that they have no objections to creating a single term
that would include both sterndrive and inboard engines in a single category of engines and that
also clarifies that hovercraft and air boats are specifically included in this engine category.

       BRP and Yamaha commented that they use PWC engines to propel their jet boat products
(also called "sport boats"), which would be classified as sterndrive/inboard under the new
regulations. BRP commented that both EPA and CARB currently categorize Sport Boats with
outboards  and personal watercraft. Currently, BRP certifies its Sport Boat models in the same
engine families as PWC models for both EPA and CARB. BRP and Yamaha commented that
including jet boat engines in the SD/I category creates a new more stringent set of emission
standards for these engines. Both manufacturers commented that this is only appropriate if jet
boats are given sufficient lead time to comply with the standards and the corporate average
provision is expanded to allow CO averaging.
                                      3-1

-------
                                                          Chapter 3: Marine SI Engines
Letters:
Commenter
NMMA
Bombardier
Yamaha
Mercury
Document #
0688
0674
0721
0693
Our Response:

       We are finalizing the definition of sterndrive/inboard engines as proposed. We believe
classifying engines used in hovercraft, air boats, and jet boats as SD/I engines is appropriate
because it will subject the engines in these applications to the same emission standards as other
boats with similar size, power, and usage characteristics. As described in Section 3.2.3, we are
providing flexibility in meeting the new emission standards for jet boat engines because they are
currently designed to use engines derived from OB/PWC applications  and because of their
relatively low sales volumes.  We believe that this flexibility, coupled  with the additional lead
time, addresses the comments raised by BRP and Yamaha regarding lead time and CO
averaging.

3.1.3  Maximum engine power and displacement

What Commenters Said:

       NMMA and Mercury Marine commented on § 1045.140 at which EPA defines
"maximum engine power" as the "maximum brakepower point on the  nominal power curve for
the engine configuration." 72 Fed. Reg. at 28,268.  Section § 1045.140(b) states that "[t]he
nominal power curve of an engine configuration is the relationship between maximum available
engine brake power and engine speed for an engine, using the mapping procedures of 40 CFR
part 1065, based on the manufacturer's design and production specifications for the engine." Id.
The reference to the mapping procedures in Part 1065 is inappropriate. Under EPA's current
regulations for OB and PWC engines, manufacturers use SAE J1228 to determine maximum
power, and the California regulations also require the use of SAE J1228. For the EU,
manufacturers use ISO 8665, which is equivalent to the SAE standard. EPA's proposal to require
the procedures in Part 1065 would be inconsistent with these existing requirements and,
importantly, would require significant additional testing over and above what is required for
compliance with the California and EU requirements. This considerable cost burden on
manufacturers is unjustified given there is no environmental benefit. NMMA recommends that
EPA replace the reference to Part 1065 with SAE J1228 and ISO 8665. This would ensure
consistency among the different regulatory schemes and reduce unnecessary compliance costs.

       Indmar has concern over the procedure for establishing the nominal power curve and the
resulting rated speed and rated power. California and the European Union use SAE J1228 or ISO
8665 (same except for English vs. metric). Section 1045.140(b) references 40 CFR part 1065 and
should reference SAE J1228.  They think EPA should remain common with CARB and
eliminate the possibility of duplicate testing for EPA at a slightly different power level.
                                          5-2

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       Bombardier commented that in 40 CFR 1045.140, EPA is proposing to redefine how
maximum engine power is determined on marine spark-ignited engines by changing the current
engine mapping procedures from SAE J1228 to 40 CFR 1065. Currently, the marine industry
follows the procedures of SAE J1228 for EPA and CARB and ISO 8665 (functionally equivalent
to SAE J1228) for the European Union (EU). By changing the mapping procedures for marine
spark-ignited engines, EPA is forcing manufacturers to run a different test procedure for EPA
than done for CARB and the European Union. This would impose a significant additional test
burden on a manufacturer. BRP recommends EPA replace the reference to 40 CFR part 1065
with SAE J1228 and ISO 8665 to maintain harmonization with CARB and the EU.

       Yamaha commented that EPA has elected to establish a test protocol that is without merit
and will add increased cost to certification, possible additional costs for dyno
replacements/updates and will not harmonize with what currently both the CARB and EU utilize
which is SAE 1228 or the ISO equivalent for this purpose. Yamaha requests that EPA adopt the
NMMA language of continued use of SAE J1228 for this purpose to harmonize on an
International basis.

Letters:
Commenter
NMMA
Indmar
Bombardier
Yamaha
Mercury
Document #
0688
0667
0674
0721
0693
Our Response:

       The regulations rely on the value for maximum engine power to establish which
standards apply and to calculate emission credits. For example, the regulations include emission
standards that differ for power ratings at 4.3, 30, 40, 250, 373, and 485 kW.  It is important to
have an objective method for establishing an engine's power rating for determining which
standards apply and for calculating emission credits. The current regulations and the published
SAE and ISO procedures direct the manufacturer to declare a value for rated power without any
clear direction to establish that value based on an engine's power map or other operating
characteristics.

       It is true that manufacturers would need to run an engine map for each engine, but we
expect that this is already common practice to establish the engine's power characteristics and
determine the recommended prop range.  Therefore, we disagree that the definition of maximum
engine power in 40CFR 1045.140 will increase testing costs.

      Note that maximum engine power is not related to testing engines. The relevant
parameter for testing is maximum test speed.  Manufacturers raised similar concerns about our
approach for establishing maximum test speed, which we describe in Section 3.9.1.
                                      3-3

-------
                                                          Chapter 3: Marine SI Engines
3.1.4   Fuel additives for reducing emissions

What Commenters Said:

       Pure Power commented regarding their EcoFuelTM Mach 3 Gasoline & Diesel Additive.
They claim that independent test results reported by AIDS, Inc, Ontario, CA, (recognized by the
EPA and CARB for automotive emission and fuel consumption) from both gasoline and diesel
powered cars and trucks showed "across the board" reductions as high as: NOx (44%), HC
(16.3%), CO (4.5%), opacity smoke (30.4%), particulates (18.3%), in addition to a 14%
reduction in fuel consumption.

       Pure Power also commented that their ThrustorTM & Schultz NozzleTM Marine
Propulsion System reduces fuel consumption, while increasing overall vessel performance. The
ThrustorTM is designed to mount on the anti-cavitation plate and skeg for all outboard and stern
drive boats. The Schultz NozzleTM mounts to the vessel hull.  Conservative projected fuel
savings between 10% and 20% depending on vessels size and speed.

Letters:
Commenter
Pure Power
Document #
0664
Our Response:

       Our regulations are intended to be fuel neutral and would not preclude the use of these
fuels or additives. However, anyone wishing to obtain a certificate of conformity that relies on
the use of a fuel that is not widely available or that relies on any particular additive would be
required to demonstrate that the engines would consistently operate with such fuels or additives
during in-use operation.  Moreover 40 CFR 1068.101(b)(l) prohibits using the incorrect fuel if it
renders the emission control inoperative.
3.1.5   Natural gas and LPG engines

What Commenters Said:

       Rolls Royce submitted comment asking what legislation EPA will apply to our [natural]
gas engines if they are to be used in marine application. Has EPA had a chance to check this?

       Nautigaz shared commercial information related to their system for converting gasoline-
fueled marine engines for operation on LPG. They pointed out the energy-security advantages of
LPG based on the extensive domestic production of LPG fuels within the United States.  They
also maintained that engines operating on LPG will always be less polluting than diesel-fueled or
gasoline-fueled engines. Nautigaz also pointed out various technical parameters of interest in
designing marine systems, such as corrosion control, the advantages of fuel-level indicators and
anti-deflagration devices.
                                           5-4

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
Letters:
Commenter
Rolls Royce
Nautigaz
Document #
0715
0727
Our Response:

       Oceangoing vessels that transport natural gas as a commodity product are increasingly
using the stored (and vented) natural gas to fuel the ship's propulsion engines. The comment
from Rolls Royce helped us realize that these engines would likely be subject to our Marine SI
standards under the wording of the proposed regulations. These engines might be 20,000 or
30,000 kW, so the certification and testing protocol we have developed for Marine SI
technologies would clearly not apply for these larger engines. To address this, we have revised
the regulations to specify that natural gas engines above 250 kW would need to meet the
standards that apply for marine compression-ignition engines. All automotive-type engines
using natural gas today are less than 250 kW so this threshold should properly differentiate
engines installed in conventional sterndrive  and inboard vessels from the diesel-derived natural
gas engines used in workboats and other commercial vessels. This is consistent with the recently
adopted provision of 40 CFR 1042. l(e).
       The emission standards in this rule are fuel neutral. Manufacturers may certify engines
using LPG, gasoline, or other fuels. It may be possible for LPG-fueled engines to reach lower
emission levels than gasoline-fueled engines, but our observation across the various engine
categories is that catalyst-equipped engines have comparable emission levels whether they are
fueled by gasoline, LPG, or natural gas.  Diesel-fueled engines are subject to a totally different
set of emission standards and other regulatory requirements.

  3.2   SD/I standards and lead time

3.2.1   SD/I standards-level

What Commenters Said:

       NMMA, Mercury Marine, Indmar, MECA, NACAA, Pennsylvania  DEP, New York
DEC, NESCAUM, and Environmental Defense support the HC+NOx standards of 5.0 g/kWh
and CO limit of 75 g/kWh proposed by EPA for the SD/I engines.

       NMMA stated that EPA is proposing a 5 g/kW-hr standard for HC+NOx and a 75 g/kW-
hr standard for CO for SD/I engines starting in model year 2009. 72 Fed. Reg.  at 28,263
(proposed § 1045.105). While NMMA fully supports the level of the proposed emission limits
for HC+NOx and CO in § 1045.105(a), the 2009 model year implementation date is not feasible
for the recreational marine industry.

       Mercury fully supports the exhaust standards for SD/I Engines provided that the
implementation dates are adjusted to provide necessary lead time.
                                       3-5

-------
                                                           Chapter 3: Marine SI Engines
       Sea Ray wants to take this opportunity to express its concerns about the robustness of
catalyst systems in the salt water environment. Since testing was never completed, CA will serve
as a validation and feedback opportunity to all of us. The industry needs the additional time to
understand what the problems might be should they arise.

       Indmar commented that they were also actively involved in the test program to prove the
technical feasibility of catalytic converters on SD/I engines for their useful life (480 hours) in
both fresh and salt water. They supplied two boats as well as technical support to Southwest
Research Institute to conduct the test program. They support the proposed federal emission
regulations for new marine spark-ignited sterndrive/inboard engines that will substantially reduce
emissions from these engines.

       NMMA submitted comments regarding the Southwest  Research Institute (SwRI)
Saltwater Test Program. Even though the SwRI tests never proved catalyst feasibility in salt
water, their members believe that, at this stage of catalyst development, there is little or no
additional data to be obtained by completing the tests.  The designs being tested at SwRI are not
designs that any of the engine companies are considering pursuing.  Whether or not they could
survive 480 hours is of no value. Their members have their own compliance plans that include
designs that appear to withstand saltwater operation, although  they will not know for sure until it
gets into the hands of customers. Therefore, NMMA agreed that EPA and CARB should
discontinue the SwRI saltwater test program.

       NMMA continued that in the context of EPA's recently proposed rules for exhaust
controls for marine engines, there is a continuing concern regarding catalyst and sensor
durability, especially in salt water, and  in engine technologies  not included in the SwRI test
programs, for example, personal watercraft engines installed in jetboats. The manufacturers of
those items also have been unable to provide any help to the engine manufacturers in this regard
as they have no experience in the salt water environment. NMMA stated that it is critical to both
the marine industry and the hundreds of thousands of American jobs that are created by this
industry, that EPA delay implementation of any nationwide  catalyst-based rule until the
manufacturers have studied the effect of the catalyst through a complete warranty cycle (three
years) and the manufacturers gain the necessary field experience in California. In any waiver
decision regarding catalysts for SD/I, they commented that EPA must make clear that it is not
predetermining the outcome of the ongoing rulemaking, and that if durability problems should
arise in actual use in California,  that EPA will work with CARB and engine manufacturers  to
adjust any rules applicable to these engines.

       NACAA commented that with respect to marine spark-ignition engines and vessels,
NACAA supports EPA's proposal to set CO standards for all sectors.  We also support the
agency's proposal to establish the first-ever federal standards for vessels powered by sterndrive
or inboard engines.

       Pennsylvania DEP supports EPA's proposed standards and implementation schedule for
marine spark-ignition engines and vessels.

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       New York DEC stated that EPA proposes to adopt standards similar to California's,
resulting in a 70% reduction in combined hydrocarbon and oxides of nitrogen (NOx) emissions.
The Department supports the proposed emission standards, including not-to-exceed (NTE)
standards and the requirement for closed crankcases.

       NESCAUM supports EPA's effort to harmonize the federal emissions standards with
those standards already adopted in California. In many respects, the proposed federal standards
are identical to or analogous with California standards. This approach will make it easier for the
engine and equipment manufacturers to provide 50-state products to the U.S. market.

       Environmental Defense supports EPA's proposal to establish HC+NOx exhaust emission
standard of 5g/kW-hr for sterndrive and inboard marine engines (SD/I engines). These standards
are identical to those adopted by CARB.  The proposed exhaust emissions standards represent
significant reductions of 70% in HC and NOx and 50% in CO emissions.  EPA predicts engine
manufacturers will meet these standards by incorporating catalysts into the water-cooled exhaust
systems used for these engines. Environmental Defense applauds the Agency for taking the
initiative to set a carbon monoxide exhaust emission standard for SD/I engines for the first time.
The addition of a CO standard should not impose any additional costs on engine manufactures
since the same catalyst technology used to achieve the HC and NOx standards will ensure that
the new CO standard is met as well.

       MECA stated that the technology to reduce emissions from spark-ignited inboard and
sterndrive marine engines will be based on automotive-type three-way catalyst with closed-loop
control technology. This technology has been used on well over 300,000,000 automobiles with
outstanding results and the same technologies can be adapted to marine inboard and sterndrive
engines. Here again results from EPA and ARB sponsored test programs detailed in the EPA
Draft Regulatory Impact Analysis confirm that three-way catalysts (TWCs) can be effectively
integrated into marine inboard and sterndrive engines, and TWCs have the necessary mechanical
integrity and catalytic durability to perform with high emission conversion efficiencies
throughout the entire 480-hour useful life emissions requirement for these marine engines,
regardless of operation in fresh or salt water environments. Important results from this
demonstration program included the design and integration of exhaust manifolds with TWCs that
provided relatively low exhaust manifold surface temperatures (through the use of a water-
jacketed exhaust system) and minimized the potential for water ingestion  into the region of the
manifolds that contained the TWCs. Both ceramic- and metallic-based substrates were used to
display a range of three-way catalyst formulations as a part of this durability test program, all
with good results. Thus, a variety of TWC technology options used successfully in automotive
applications were shown to be effective in these marine engine applications. The early
commercial introduction of a catalyst-equipped marine inboard engine is further proof that
catalyst can be used to achieve EPA's proposed HC+NOx and CO standards for this category of
Marine SI engines.

       SCAQMD staff believes that more stringent standards for this category are also
appropriate, technically feasible, and absolutely critical for the South Coast Air Basin to meet its
PM 2.5 and 8 hour ozone standards. Engines in this category are closely related to automobile
engines which have achieved much lower emission levels using advanced emission control
                                       3-7

-------
                                                           Chapter 3: Marine SI Engines
systems for more than 20 years.  Successful transfer of this technology to land based nonroad
engines (which are also similar to automobile engines) has lead the California Air Resources
Board and the U.S. EPA to adopt exhaust standards that will require new engines to meet exhaust
levels two times, and by 2010, five times more stringent than those levels proposed in this rule.

       See 3.2.3 for comments specifically related to jet boat engines. See 3.4 for comments
specifically related to high-performance engines.

Letters:
Commenter
Sea Ray
South Coast AQMD
NY DEC
NESCAUM
Environmental Defense
NACAA
Indmar
Mercury
MECA
Pennsylvania DEP
Document #
0683
0704
0659
0641
0648
0651
0667
0721
0668
0676
Our Response:

       As supported by the majority of commenters, we are adopting the proposed exhaust
emissions standards for SD/I engines of 5 g/kWh HC+NOx and 75 g/kWh CO. The final
HC+NOx standards are similar to the California ARB emissions standards for HC+NOx that
began in 2008.  We believe the type of catalyst used to achieve the HC+NOx standard will also
be effective in reducing CO emissions enough to meet the new standard, therefore no additional
technology will be needed to control CO emissions.

       We believe the final federal exhaust emission standards for SD/I engines represent the
greatest degree  of emission reduction feasible in this time frame. Over the past few years,
developmental programs have demonstrated the capabilities of achieving significant reductions
in exhaust emissions from SD/I engines. Chapter 4 of the Final RIA presents data from several
of the SD/I engines with catalysts that were tested as part of the development of the standards
had HC+NOx emission rates lower that 5 g/kW-hr,  even with consideration of expected in-use
emissions deterioration associated with catalyst aging. The goal of the testing was to
demonstrate catalysts that will work within the packaging constraints associated with water
jacketing the exhaust and fitting the engines into engine compartments on boats. California ARB
has acted on this information to set an HC+NOx emission standard of 5 g/kW-hr for SD/I
engines, starting in 2008. At this time, three engine manufacturers have certified SD/I engines to
these standards. In addition, Chapter 4  of the Final RIA presents data from these engines as
detailed data on several developmental SD/I engines with catalysts packaged within water-
cooled exhaust manifolds. Four of the  developmental engines in our test program were operated
with catalysts in vessels for 480 hours.  The remaining developmental engines were tested with
catalysts that had been subjected to a rapid-aging cycle in the laboratory. As stated in their

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


comments, Indmar has demonstrated the durability of catalysts over their full useful life of SD/I
engines, both in fresh and salt water. Data from these catalyst-equipped engines also support the
level of the standards.  We also performed testing on SD/I engines equipped with both catalysts
and EGR.  Although this testing showed emission results in the 2-3 g/kW-hr range, we expect
that similar reductions could be achieved more simply through the use of larger catalysts or
catalysts with higher precious metal loading.

       Past experience, in other engine categories, indicates that most manufacturers will strive
to achieve emission reductions well below the final standards to give them certainty that they
will pass the standards in-use, especially as catalysts on SD/I engines are a new technology.
Therefore, we believe the emission standards for SD/I engines represent the greatest degree of
emission reductions achievable taking into consideration the potential variability in in-use
performance and in test data mean and do not believe it would be appropriate at this time to set a
lower standard for these engines.

3.2.2   SD/I standards-lead time

What Commenters Said:

       NMMA and Mercury Marine stated that while SD/I engine manufacturers have started
the necessary research and development to produce engines and emission control systems to
comply with the 2008 CARB standards, the California market represents only a small portion of
the national marine engine market.  As a result, they argued that some manufacturers will limit
the engine models offered in California because there is not sufficient lead time to reconfigure
their entire product line. They commented that the implementation date for the federal emissions
standards must take into account the challenge of designing catalyst-based systems for all
engines across the entire SD/I engine market. Mercury Marine adds that due to the issue of a
major change in base engines supplied by GM (see below) a 2009 implementation date would
force Mercury to apply for hardship relief as soon as the rule is finalized. This is not the way
they would like to start off a new rule.

       NMMA and Mercury Marine stated that EPA is proposing a 5 g/kW-hr standard for
HC+NOx and a 75 g/kW- hr standard for CO for SD/I engines starting in model year 2009. 72
Fed. Reg. at 28,263 (proposed § 1045.105). While NMMA fully supports the level of the
proposed emission limits for HC+NOx and  CO in § 1045.105(a), the 2009 model year
implementation date is not feasible for the recreational marine industry. NMMA has worked
cooperatively with EPA over the past several years to share data and information on the status  of
the development of catalyst technology that can be used effectively and safely in both fresh and
saltwater environments. While the technology is commercially available, the ability of
manufacturers to develop catalysts and reconfigure all of their engines to accommodate catalyst-
equipped exhaust systems by model year 2009 is not realistic for several important reasons that
are specific to how the marine engine industry is structured.

       NMMA urges EPA to adopt in the final rule the third option for implementation
discussed—full compliance with the emission limits in model year 2010 for all SD/I engines
except for the replacement engines for the 4.3L and the 8.1L and personal watercraft engines
installed in jet boats, which should have until model year 2011 to comply.  Mercury Marine is


                                       3-9

-------
                                                           Chapter 3: Marine SI Engines
limiting model availability in CA for 2008 - 2009 and needs until 2010 to have all of the
horsepower levels covered for a National Rule. Mercury Marine has supplied a confidential list
to EPA of the engines and power ratings that will not be available as catalyst engines for 2008 -
2009 and will not be sold in California.

       NMMA and Mercury Marine commented as EPA notes in the preamble, a large number
of SD/I engines are based on automotive engine blocks produced by General Motors (GM). 72
Fed. Reg. at 28,115. EPA also correctly points out that GM plans to discontinue production of
the 4.3L and 8.1L engine blocks in 2009 and instead plans to offer a 4.1L engine block and a
6.0L supercharged engine as replacements. There are significant market and compliance
implications associated with GM's product plans, which the NMMA-suggested compliance
schedule would address.  From a cost perspective (which EPA correctly identifies in the
preamble), the small number of remaining years of sales of the 4.3L and 8.1L fail to justify the
considerable costs associated with developing catalyst-based exhaust systems for these engines.
From a compliance timing perspective, manufacturers that marinize the replacement engines will
only be able to start designing catalyst systems sometime late this year when it is expected that
manufacturers will see the first prototypes of the replacement engines. The development cycle
for marinizing the base engine is over two years for some companies. Thus, a model year 2009
implementation date does not allow enough lead time for the industry to marinize the
replacement engines and develop exhaust control systems.

       Mercury Marine added that CARB has already provided relief on these engines for 2008
and 2009. The development cycle for marinizing the base engine is three years. Production base
engines from GM are not scheduled to be available until 2010, and that assumes they maintain
their current schedule. They commented that they have already been advised that the GM
timeline has slipped a few months.  Furthermore, having to develop these new engines as catalyst
marine engines is taking  resources away from being able to develop catalyst versions of the
engines listed above that will not be available in California. Mercury Marine commented that the
workload is more than can be accomplished to launch all of these new and modified engines on a
national level before 2010 - 2011.

       NMMA also commented that the other option for implementation that EPA suggests is to
allow an additional year for the engines not using catalysts in California in 2008, namely the
4.3L and the 8.1L.  NMMA stated that the model  year 2009 is not practical and that an additional
year for these engines until model year 2010 is appropriate  and justified. California's marine
engine standards will require catalysts on engines (other than the two engines noted above)
starting in model year 2008. In light of this short lead time and the number of different products
offered, NMMA argued that marine engine manufacturers will not have the ability to fully
catalyze their entire line of engines for California in that time frame. Also important to consider
is that the California market constitutes only a small percentage of the marine engine market
(unlike the case with motor vehicles, which is larger than the percent of the overall population).
Thus, marine engine manufacturers will in some cases limit the engines available in California to
those that can be readily  catalyzed and will continue to sell a mix of catalyzed and noncatalyzed
engines in the other 49 states in 2008 and 2009. NMMA stated that, by model year 2010, engine
manufacturers should have the necessary lead time required by Clean Air Act § 213 to resolve
                                          3-10

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


most, if not all, of the technological challenges involved with catalyzing their entire product
lines, with the exception of the replacement engines for the 4.3L and the 8.1L.

       NMMA and Mercury Marine stated that the phase-in approach suggested by EPA in the
preamble is not a workable option for this industry. With thousands of boat builders dispersed
across the U.S., marine engine manufacturers do not have ultimate control of the type of engines
purchased and installed on boats. This is particularly the case where the engine manufacturer is
still manufacturing engines that are not catalyzed. Boat builders determine which engines are
purchased and can choose either catalyzed or non-catalyzed versions of the engines since boat
builders are not subject to emissions standards. For these reasons, they concluded that a
compliance deadline in model year 2010 for the majority of SD/I engines, with full
implementation in model year 2011, makes sense in the context of this particular industry.

       NMMA summarized comments from its members of Four Winns Boats, LLC,
Chaparral/Rodalo Boats, Massachusetts Marine Trade Association, Regal Marine Industry,
Challenger Power Boats, Godfrey, Lowe Boats, Brunswick Corporation, North American
Sleekcraft, S2Yachts, Sea Ray, Hallett, Cigarette Racing, Premier Marine Inc., and
Larson/Glastron Boats. Two manufacturers urge EPA to adopt the third option for
implementation:  full compliance with the emission limits in 2010 for all the 4.3L & 8.1L and
their replacements will have until 2011. Three manufacturers state engine supplier (Mercury
Marine/GM) needs until 2010 to have all hp levels covered for national rule-if 2009, some
models may be available for one year before phase-out.  Four manufacturers commented the
engines will need to be installed with onboard diagnostic emission notification systems - need
time to engineer approach once receive engines from engine  supplier. One commenter stated, as
a small independent builder, the technology and the products to support this are clearly not
available today and is pleading that the EPA will push this back until 2011 to allow for proper
testing and implementation. Twelve commenters stated that California imposed requirement
should be a testing ground until the system can be validated for a national  release with a few
years into the program. One commenter stated that the proposed implementation is not feasible
due to changes being made in the availability of GM based engines which would result in some
one year offering of motors and recommend gradual phase-in with full compliance by 2012.

       VolvoPenta supports full industry compliance with standards in 2010, except for 4.1 and
supercharged 6.0 which need until 2011. This option keeps a level playing field for all small
business partners and allows more time for California catalyzed units to acquire hours of actual
operation in consumer hands. The option will provide adequate time for Volvo Penta to develop
full model lines demanded by customers and ensures compliance to the rule. Volvo Penta needs
additional time to conclude its own saltwater testing and to monitor the durability of California
compliant engines.  Volvo Penta stated that, if U.S. EPA proceeds with a rule requiring full
industry compliance on January 1, 2009, with the standards for SD/I engines, then their
comments are intended to serve as Volvo Penta's application for a hardship exemption.

       Sea Ray advocated EPA to adopt, in the final rule, the third option for implementation
discussed—full compliance with the emissions limits in 2010 for all SD/I engines except for the
4.3L and 8.1L, and their replacements, which should have until 2011 to comply. Their engine
supplier, Mercury Marine, is limiting model availability in CA for 2008 - 2009  and needs until
                                       3-11

-------
                                                           Chapter 3: Marine SI Engines
2010 to have all of the Hp levels covered for a National Rule. If the 2009 date is implemented,
Sea Ray commented that some of these models may only be available for a single year before
being phased out. Moreover, since these engines will need to be installed with on board
diagnostic emission notification systems, they will need time to engineer their approach in
coordination with the products they receive from Mercury. Sea Ray also expressed its concerns
about the robustness of catalyst systems in the salt water environment. Because testing was
never completed, California will serve as a validation and feedback opportunity to all of the
industry. The industry needs the additional time to understand what the problems might be
should they arise.

       Mercury stated in a public hearing that if they had to meet the standard in 2009, as soon
as the rule is signed, they would have to apply for relief under the hardship provision. Between
the issues with GM that they have discussed, and the fact that they are not selling some models
in CA, which they can not reconcile in their product line until 2010, Mercury stated that they
need one of the options that is in the preamble. That option is compliance with the standard for
the engine families that are not changing in 2010 and an extension to 2011 for the engine
families being replaced by GM, keeping in mind that Mercury Marine will not get even
prototype level hardware for GM's new engines until late this year or next year.  According to
Mercury, the development cycle for converting these auto base engines to marine engines is 30-
36 months. This also allows for the possibility of GM missing the launch date of the new models
and the industry not having to come back to EPA for hardship relief.  Mercury will also gain
some field experience with the catalyst engines in the California market, as catalyst feasibility
testing at SwRI was terminated, with industry approval, without ever demonstrating catalyst
durability in a saltwater environment. Because that independent testing, funded by EPA and
CARB, was never completed,  Mercury commented that it is reasonable to allow for the
California market to be that testing grounds for 2 years. Lastly, when asked about phase-in
programs, Mercury stated that one of the issues that they have is that they do not control the
product mix in the field, instead the boat builder does. However, Mercury does like the
provisions in the proposed rule for banking early credits which is an incentive to get catalysts
into the market early.

       California ARE strongly encourages U.S. EPA to adopt a 2008 start date of the 5 g/kW-
hr HC+NOx standard for sterndrive/inboard engines to avoid putting  California dealerships at a
competitive disadvantage with out-of-State dealerships that would still be able to sell boats
without catalyst-equipped engines at a significantly lower purchase price (less the cost of catalyst
and associated hardware). Tooling will already exist for the catalyzed engines as a result of
California's requirements, and an extra year to implement the same standards is unnecessary
considering that the sole manufacturer already producing catalyst equipped engines for the
California rule is doing so nationwide. Not only would nationally harmonized implementation
eliminate the disparity in compliance costs between California and federal engines, it also makes
sense from an economic perspective since the economies of scale (quantity discount) involved  in
producing a harmonized engine model nationwide rather than multiple state-based models would
reduce the price of compliance both to the manufacturers and to the consumer. As EPA notes in
the preamble, a sterndrive/inboard engine manufacturer that qualifies as a small business already
offers a catalyst-equipped engine nationwide. Thus, the implementation delay and  small volume
manufacturer provisions proposed by U.S. EPA may be unnecessary, and would result in a delay
                                          3-12

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


in public health benefits. If U.S. EPA still believes it necessary to provide industry with some
sort of compliance cushion, ARB suggested restructuring the federal program such that 2008
models could be treated leniently in-use initially, providing industry with a greater learning
opportunity for fine tuning their catalyst system designs, rather than a delay in implementing the
5 g/kW-hrHC+NOx standard.

      New Jersey DEP commented that several CARB standards for exhaust emissions are
fully phased-in between 2005 and 2008, whereas the proposed phase-in dates for the
corresponding federal standards do not begin until 2010. Of most concern, the special provisions
for small and medium manufacturers may delay full compliance until 2014. In light of the fact
that manufacturers will already be providing cleaner engines and equipment to California and
that technology issues will not be a factor, New Jersey DEP stated that these cleaner engines and
equipment should be required to be made available sooner nationwide.

      MECA believes that the 2009 model year implementation date provides industry with
adequate time to meet these standards.

      NACAA noted that sterndrive and inboard engines with catalysts are already in
production and engine manufacturers are already tooled to produce catalyzed engines for
California for 2008.  Therefore, although they believe the proposed federal implementation
schedule - beginning in 2009 - is appropriate and should not be delayed, they recommend  that
EPA require that once a certified engine is available in California it be sold nationwide.

      Pennsylvania DEP supports EPA's proposed standards and implementation schedule for
marine spark-ignition engines and vessels.

      NESCAUM supports EPA's current proposal, that the SD/I catalyst-based exhaust
emissions standards take effect in 2009, one year following implementation in California. They
agree with EPA's position that  once the catalyst-based technology is introduced across product
lines in California, it should be readily available nationwide soon thereafter. They see no need
for EPA to implement the alternative approach of extending the compliance date to 2010. At the
same time, as it appears that General Motors is discontinuing supplying the 4.3 and 8.1 liter
engine blocks in 2009, they would not object to allowing additional time, as suggested, for the
orderly transition to the 4.1 and 6.0 liter blocks. Their understanding is that the engines based on
the 4.3 and 8.1 liter blocks represent a relatively small portion  of the new marine engine market,
compared to other more widely-used blocks. Presumably, the new 4.1 and 6.0 liter blocks  will
not claim a large share of the market, at least in their introductory years. Therefore, concluded
commented that the overall emissions impact should be minimal if additional transition time is
provided. They would support this approach (allowing additional time for engine blocks
representing a small fraction of the market) over the alterative  approach of allowing all engine
families to certify to a more lenient transitional standard over the 2009-10 timeframe.

      New York DEC urges EPA to require implementation of the standards in 2009 for all
engines.

      Wisconsin DNR commented that EPA shall require CARB certified sterndrive or inboard
engines available in California from 2008 to be sold nationwide.
                                       3-13

-------
                                                          Chapter 3: Marine SI Engines
       Environmental Defense commented that EPA has proposed an implementation date for
the SD/I exhaust emission standards of model year 2009. California's HC and NOx standards
take effect in model year 2008. Accordingly, the national standards will go into effect one year
after the identical standards in CA. EPA observes that a "one year delay [in implementing the
national rules] allows manufacturers adequate time to incorporate catalysts across their product
lines as they are doing in California." Environmental Defense strongly opposes any delay beyond
this proposal. Indeed, one engine manufacturer is already selling engines equipped with catalysts
nationwide.  Reducing the HC, NOx and CO emissions from these small recreational boats will
assist many states and local governments in achieving or maintaining healthy levels of ozone,
PM and CO  and will help to ensure better air quality for many Americans.

Letters:
Commenter
NMMA
Sea Ray
NJDEP
Wisconsin DNR
NESCAUM
Environmental Defense
NACAA
Volvo Penta
Mercury
MECA
NY DEC
Pennsylvania DEP
Mercury (hearing)
Hallett
Sea Ray
S2Yachts
North American Sleekcraft
Brunswick Corporation
Lowe Boats
Godfrey
Challenger Power Boats
Regal Marine Industry
Massachusetts Marine Trade Association
Chaparral/Rodalo Boats
Four Winns Boats, LLC
Cigarette Racing
Premier Marine Inc
Larson/Glastron Boats
Document #
0688
0683
0710
0663
0641
0648
0651
0708
0693
0668
0659
0676
0642
0713
0683
0697
0666
0695
0660
0645
0644
0635
0634
0630
0650
0637
0613
0626
                                          3-14

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


Our Response:

       Our SD/I standards start to take effect with the 2010 model year, two years after the same
standards apply in California. We believe a requirement to extend the California standards
nationwide after a two-year delay allows manufacturers adequate time to incorporate catalysts as
they are doing in California across all of their product lines. Once the technology is developed
for use in California, it will be available for use nationwide soon thereafter. In fact, one
company currently certified to the California standards is already offering catalyst-equipped SD/I
engines nationwide.

       To address the challenge related to the transition away from the current 4.3 and 8.1 liter
GM engines, we are adopting in the final rule a direct approval for a hardship exemption
allowing manufacturers to produce these engines for one additional year without certifying them
(see §1045.145). Starting in the 2011 model year, we would expect manufacturers to have
worked things out such that they could certify their full product lineup to the applicable
standards.

3.2.3   Issues related to jet boats

What Commenters Said:

       NMMA commented that the proposed definition includes jet boats in the SD/I category.
72 Fed. Reg. 28,290. NMMA supports the inclusion of jet boats in the definition of SD/I engine
with the condition that manufacturers of jet boats would receive until 2011 to comply with the
more stringent SD/I emissions standards. Jet boats utilize the same engine technology as personal
watercraft engines and have been regulated under the EPA standards applicable to personal
watercraft and outboards. This technology is very different from SD/I engines, which rely on
automotive-based engines. Additional lead time for compliance, therefore, is necessary to allow
engine manufacturers sufficient time to redesign and develop engines—which typically takes
three years for known technologies—that will comply with the new, more stringent SD/I
emissions limits. It is inappropriate to subject jet boats to the same implementation lead-time as
the SD/I engines considering that those manufacturers have been in product development for the
2008 implementation of the CARB standards over the last few years. It is also critical that, as
proposed, jet boats be allowed to average credits, both HC+NOx and CO, generated by other
personal watercraft and outboards to provide flexibility and ensure that jet boats will be able to
meet the SD/I emission standards. NMMA is supportive  of the proposed approach discussed in
the preamble and in the proposed regulatory text in § 1045.701(d) provided CO averaging is
included.

       NMMA also stated that related to the inclusion of jet boats in the SD/I category is the
treatment of the useful life for these engines. For PWC engines used in jet boats, NMMA
supports a 5 year, 350 hour useful life. This is consistent with the proposed useful life for
outboard and personal watercraft engines in the rule discussed below and is appropriate for jet
boats given that the engines are identical. To force dual compliance levels for identical engines
leads to confusion and increases the certification burden imposed on the engine manufacturer.
NMMA also recommends that the useful life for jet boat engines be reviewed by EPA three years
                                       3-15

-------
                                                           Chapter 3: Marine SI Engines
after the recommended model year 2011 compliance date and adjusted as experience is gained in
the field with the unproven after treatment technology.

       BRP stated in a hearing and submitted written comments that they cannot support EPA's
proposed catalyst based emission standards for stern drive and inboard engines as it presently
applies to water-jet sport boats. This product category has been regulated under the standards
applied to outboard and personal watercraft engines on the basis that water-jet sport boats utilize
the same engine technology as personal watercraft engines. It is understood that the EPA now
desires to regulate this boat category, which has exclusively utilized automotive-based engines.
These sterndrive and inboard engine manufacturers have effectively been developing a catalyst
solution in preparation for the CARB regulation since approximately 2004. It is therefore very
inappropriate to subject the water-jet sport boats to the same proposed lead-time given the
difference in basic engine technology and prior catalyst development time.

       Furthermore BRP commented there are numerous patents held by a competitive water-jet
sport boat manufacturer which represent clear and significant design constraints to BRP in order
to avoid patent infringement. There are effectively 30 related patents which have applicability to
water-jet sport boats, 13 which have specific catalyst application constraints. These constraints
include catalyst positioning, layout, cooling and sensor placement issues. The fundamental
nature of these challenges results in the need for greater development lead-time.

       BRP development and application lead-time for an established engine technology is
approximately three years. The patent issues they have briefly explained represent complex
design challenges and it is therefore not possible at this point to project the amount of additional
development time required to meet the proposed catalyst application to water-jet sport boats.

       BRP also stated that however, the lead-time challenge can be justly addressed by
providing water-jet sport boat manufacturers which utilize an outboard personal watercraft
engine technology the following allowances:
       1.  An implementation lead-time of model year 2011, and
       2.  BRP is supportive of the proposed corporate averaging provisions in 40 CFR
          1045.701 (d) which allow "Sterndrive / Inboard engines certified under 1045.660 for
          jet boats may be use HC + NOx exhaust credits generated from outboard and personal
          watercraft engines, as long as the credit-using engine is the same model as an engine
          model from an outboard or personal watercraft family." For the corporate averaging
          provision of 40 CFR 1045.701 (d) to be meaningful to a manufacturer, CO  averaging
          is essential for achieving compliance, (also listed in 3.6.2)

       Yamaha stated in a hearing, after considerable discussions with their engineers to reach  a
feasibility consensus they request that if the EPA were to agree on a MY start date of 2011
(which again for Yamaha is April of 2010) for compliance at the Inboard level of 5 g/kW-hr, this
would afford additional time to an already taxed staff, to design and build a "ground up engine"
required to meet the target levels presented in the proposal. As the EPA may be aware, PWC
engines utilized in Jet boats is a very small quantity, and short runs of catalyst based engines
would be cost prohibitive.  Therefore, this short additional lead time will have the positive effect
of bringing into a lower compliance level, a greater amount of PWC engine families to help
                                          3-16

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


offset production costs. Again by allowing this lead time, the EPA will have championed even
further emission reductions over the broad engine spectrum.

       In Yamaha's view, a MY 2011 compliance date (although aggressive) was agreed upon
internally as this appears to parallel what the EPA has considered acceptable lead time for the
SD/1 members affected by this rule as a result of new engine block design and feasibility issues.
If the industry is in fact subject to classify PWC engines used in Jet Boats as a different engine
then we need the additional lead time.

       Yamaha submitted written comments stating the proposed definition includes Jet boats
into the SD/I Category. Yamaha utilizes PWC engines (complete exact units) to propel their Jet
boat product. The number of units used currently does not quite come to 8% of their entire PWC
engine production,  Yamaha will most likely by defacto have to produce en mass a compliant 5gr
engine that will carry the day for both PWC and Jet Boat compliance. By allowing additional
requested lead time this will in actuality further reduce emissions over a larger engine group.
Even though this is an engine rule and should remain so, Yamaha will  support this
reclassification if the following flexibilities are granted or represented within the rule.
       a.  Enough lead time to develop, test and produce the necessary engine block, exhaust
          and catalyst systems required to achieve a 5gr level of HC+Nox emission and 75gr
          CO levels as proposed for SD/I engines. Yamaha estimates this to be achievable in
          M/Y 2011 which is reflective of the proposed lead time flexibility being requested by
          the SD/I engine suppliers due to the engineering challenges of a ground-up new
          engine block w/ catalyst being produce as a replacement to current available units.
       b.  Yamaha commented that language to exclude 75gr. CO requirement on PWC engines
          utilizing banked HC+NOx credits for Jet Boat reclassification compliance (between
          M/Y2009 and 2011) should be included in the rule as no banking of CO credits
          existed on previous Tier 1 requirements and currently are above the 75gr limit set for
          a catalyzed SD/I automotive based engine.

       Yamaha requests that the useful life period for Jet Boat engines be the same as current
PWC useful life of 5 years or 350 hours as these engines share same design and product use and,
upon completion of a successful EPA technical review in 2014 raise the useful life period to that
of SD/I. This would allow for proper long term durability testing of catalysts systems that would
need to be in place. This request is not a large departure than the EPA seeking comment on the
proposed reduced useful life structure of the High Performance SD/I engines. Yamaha strongly
requests for PWC engines used in Jet Boats be granted similar flexibility and remain at the same
useful life period as our PWC engines.

       Mercury Marine stated that EPA is proposing to define "sterndrive/inboard engine" in §
1045.801 as "a spark ignition engine that is used to propel a vessel, but is not an outboard engine
or a personal watercraft engine. This includes engines on propeller-driven vessels, jet boats, air
boats, and hovercraft." 72 Fed. Reg. at 28,290 (emphasis added). Mercury Marine has no
objections to creating a single term that would include both sterndrive and inboard engines in a
single category of engines and that also clarifies that hovercraft and air boats are specifically
included in this engine category. However, it is also critical that, as proposed, jet boats be
allowed to utilize credits, both HC+NOx and CO, generated by outboards to provide flexibility
                                       3-17

-------
                                                           Chapter 3: Marine SI Engines
and ensure that jet boats will be able to meet the SD/I standards, (also listed in 3.6.2) In addition,
Mercury recommends that the new SD/I standards for jet boats become effective in 2011.

       Yamaha asked whether the requirement to have smaller sales of jet boat engines than the
analogous outboard or personal watercraft engines needed to be in place for every model year.
The concern related to a scenario in which the outboard and personal watercraft versions of an
engine would be discontinued while the jet boat engines would continue in production for
another year.

Letters:
Commenter
NMMA
Bombardier
Yamaha
Mercury
Yamaha (hearing)
Bombardier (hearing)
Document #
0688
0674
0721
0693
0642
0642
Our Response:

       We are providing some flexibility in meeting new emission standards for jet boat engines
because they are currently designed to use engines derived from OB/PWC applications and
because of their relatively low sales volumes. We will finalize the proposal to allow
manufacturers to use emission credits generated from outboard and personal watercraft engines
to demonstrate that their jet boat engines meet the new HC+NOx standards for SD/I engines.
We are also adding the flexibility of CO emission averaging that was not previously included in
the NPRM. This is necessary to fulfill the intent of the proposed flexibility.

       Manufacturers of jet boat engines subject to SD/I standards and using credits from
outboard or personal watercraft engines must certify these jet boat engines to an PEL that meets
or exceeds the standards for outboard and personal watercraft engines.  We are providing
manufacturers a one year delay to meet the FEL requirement which now becomes effective in
2011.

       Jet boat engines are now by definition sterndrive/inboard engines, so the default useful
life period is 10 years or 480, whichever comes first.  However, we understand that jet boat
engines that are common to personal watercraft or outboard engine models depend on the
preexisting certification demonstration.  As such, we believe it is appropriate to allow for a 350-
hour useful life so that the original certification can continue to be valid without additional
durability demonstration for the jet boat engines. This shorter useful life does not apply for jet
boat engines that are certified independently.  Note that, under 1045(3)(2), any SD/I engine
manufacturer may request that we approve a shorter useful life on a case-by-case basis.

       We understand that there are valid business reasons to discontinue engine models in
stages for certain applications. We believe the regulations should address Yamaha's concern,
especially because their plan involves a long-term strategy to design their jet boat engines to
                                          3-18

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


comply with the SD/I standards without relying on emission credits. We have revised this
provision such that it no longer requires a demonstration of lesser sales of jet boat engines for
every model year. This would allow us to respond to a special situation such as that described by
Yamaha and acknowledge that their situation meets our intent. We would expect such a
demonstration rarely to be based on sales information from more than two consecutive model
years.

  3.3  OB/PWC standards and lead time

3.3.1   OB/PWC standards-level and form of standard

What Commenters Said:

       NMMA and Mercury Marine commented for outboard (OB) and personal watercraft
(PWC) engines in § 1045.103, EPA is proposing a HC+NOx standard of 28 - 0.3 x P g/kW-hr for
engines <= 40 kW. For engines > 40 kW, EPA is proposing 16 g/kW-hr for HC+NOx. 72 Fed.
Reg. at 28,262. EPA explains in the preamble that the HC+NOx standards are  similar in
stringency to the  2008 model year California limits but use a "simplified form" as opposed to the
one used by the CARB regulations. 72 Fed. Reg. at 28,130. While NMMA appreciates efforts to
simplify a regulatory requirement, the best approach for emissions standards for the PWC and
OB engine categories is to harmonize any new federal standards exactly with those in place in
California. To establish a separate formula for developing the federal number,  even if it is similar
in stringency, creates additional complexity for the marine industry with no environmental
benefit.

       NMMA and Mercury Marine continued that with respect to CO emission limits, EPA is
proposing in § 1045.103 for engines = 40 kW, a standard of 500 - 5.0 x P g/kW- hr, and for
engines > 40 kW, a standard of 300 g/kW-hr. The proposal also would allow manufacturers to
average, bank and trade emission credits and would require a family emission  limit (FEL) for
engines > 40 kW at a maximum of 450 g/kW-hr. The maximum value for the FEL for all other
engines would be a formula of 650 - 5.0 x P. 72 Fed. Reg. at 28,263 (proposed § 1045.103(b)).
These proposed CO levels are technologically achievable and assure that PWC and OB engines
will be able to still meet the CARB 2008 HC+NOx emission standards. From a safety
perspective, these levels are also appropriate. USCG boating safety statistics for deaths from  CO
poisoning  clearly indicate that PWC and OB engines have no history of CO poisoning. A more
stringent standard would impose a significant cost burden on these manufacturers with no health
or welfare benefits as evidenced by the science and accident statistics associated with CO
poisoning. Thus,  NMMA supports these standards in the proposal and agrees that if EPA is to set
a limit for CO, these levels are appropriate for these two engine segments.

       Mercury Marine commented that EPA has requested comment on catalyst level emissions
on OB/PWC. Mercury Marine is the only OB company that has to meet the CA SD/I catalyst
level emissions standard for 2008. Since they have been developing catalyst systems  for SD/I,
Mercury states that they are in the best position to comment on this. For SD/I engines, where
weight and packaging are much less of an issue, the cost to develop catalyst engines is in the
vicinity of $3M per engine family. The engineering challenges to deal with water intrusion,
condensation, exhaust gas temperatures, etc. have  been enormous.


                                      3-19

-------
                                                          Chapter 3: Marine SI Engines
       Mercury Marine continued that for Outboards, these challenges, and the associated costs,
are more extreme. Due to the tight packaging, under cowl thermal management, and closer
proximity to water, catalysts on outboards will be a larger, and more expensive, undertaking, and
technical feasibility is not a given. It will entail a complete redesign of every outboard engine
and, if technically feasible, will cost in the range of $8M - 15M per engine family, just in
engineering costs, and take 4-5 years after a rule is finalized to launch the first models. Any
such rule can not be finalized until technical feasibility is established. Therefore, it is
inappropriate to consider catalyst level emissions on outboards at this time.

       Honda commented on the Outboard and Personal Watercraft (PWC) Exhaust Standard
Proposal. They suggest that EPA use the  CARB's equation when setting the exhaust emission
standard for outboard engines. Emission standards in the proposal's Section 1045.103 are
described in the preamble as "simplified". Honda believes that they are not "simplified", but
simply different with no real reason or environmental benefit. The proposed standard diverges
from the original EPA standards curve and the California standard creating, not a simplified
uniform standard for the United States, but rather two separate standards.

       Bombardier commented that EPA explains these standards are of similar stringency to the
CARB 2008 standards (3-Star), but are in a simplified form. However, creating a new standard
different from the California standard complicates certification. BRP urges harmonizing the
proposed HC +NOx exhaust emissions standards for PWC and Outboards with the CARB 3-Star
emission standards.

       Suzuki appreciates EPA's attempt to simplify the certification process wherever possible;
however EPA's direction with the proposed HC+NOx standards creates a situation where some
outboard engines currently certified to comply to CARB 2008 standards will require calibration
and design changes to comply with the slightly different levels proposed by EPA while still
maintaining reasonable compliance margins . This will be an expensive and resource-intensive
effort which will not be of benefit to the environment. It is also important to note that the effort
required to calculate the appropriate emission standard for a given engine family is not materially
different between EPA's proposal and  the CARB 2008 HC+NOx requirements. Considering that
the rational for the proposed new Federal-specific HC+NOx standards is to simplify the
certification process, and the actual effect will be to increase certification cost and effort, it is not
reasonable for EPA to proceed with their proposed federal-specific HC+NOx standards.  Suzuki
requests EPA reconsideration of their proposal to create new Federal-specific HC+NOx emission
standards, and requests EPA adopt a requirement that strictly harmonizes with the CARB 2008
HC+NOx standards.

       Suzuki continued that EPA has proposed all-new CO standards of 500 - 5 .0 x P g/kW-hr
for engines <40kW, and a standard of  300 g/kW-hr for engines of >40 kW. Additionally, EPA
has proposed to limit maximum emissions of CO to levels of 150 g/kW above the applicable
standard. Suzuki believes EPA's proposal represents levels that are technically achievable given
reasonable lead-time and will allow for continued compliance with CARB 2008 HC+NOx
standards without major design changes. However, EPA's proposed CO standards will require
design changes and development for some Suzuki outboard engine families.  Assuming that EPA
                                         3-20

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


harmonizes with the CARB 2008 HC+NOx standards and a reasonable amount of lead-time is
provided, Suzuki can support EPA's new CO standards as currently proposed.

       Tohatsu stated in a hearing that contents of the new regulations should be the same as the
current CARB standards to avoid having to comply with two different sets of standards and
testing methods. They would like to see  one national standard rather than a national standard and
then also other standards set by different local areas.

       California ARE supports adoption of a federal 16 grams per kilowatt-hour (g/kW-hr)
hydrocarbon plus oxides of nitrogen (HC+NOx) standard for outboard/personal watercraft
engines greater than 40 kilowatt (kW). This standard is technologically feasible and is similar in
stringency to the existing 2008 California standards for the majority of the category.

       California ARB also supports U.S. EPA's proposed carbon monoxide (CO) standards for
outboard/personal watercraft and sterndrive/inboard engines. These standards are technologically
feasible.  ARB staff will likely propose the adoption of identical CO standards when it next
revises California's regulations for recreational marine engines.

      NESCAUM supports EPA's effort to harmonize the federal emissions standards with
those standards already adopted in California. In many respects, the proposed federal standards
are identical to or analogous with California standards. This approach will make it easier for the
engine and equipment manufacturers to provide 50-state products to the U.S. market.

      Environmental Defense supports EPA's proposal to establish more stringent HC and
NOx emission limits for outboard and personal watercraft (O/PW). The proposed  standards, if
implemented, would achieve more than a 60% reduction in HC and NOx emissions over existing
standards. These standards are consistent with those previously adopted by CARB.
Manufacturers will be able to achieve these emissions reductions by replacing older carbureted
two-stoke engines with more advanced, direct injection two-stroke or four-stroke engines. This
transition should be relatively easy and inexpensive for manufacturers as the market trend has
been moving toward the retirement of carbureted two-stoke engines in favor of cleaner two and
four-stroke engines. Environmental Defense is also pleased that EPA's proposal includes a CO
limit for OB/PWC engines.  Achieving the proposed CO standard is readily achievable as the
same two and four-stroke engines required to meet the HC and NOx standards will achieve the
CO standard.

      NY DEC stated that EPA proposes to adopt standards generally similar to  existing
California standards, yielding a 60% reduction in combined hydrocarbon and NOx emissions
compared to current federal standards. The Department supports these proposed standards.
NY DEC also stated that additional work is needed to facilitate the application of catalysts to
outboard and personal watercraft engines, many of which are automotive sized.

      NACAA commented with respect to personal watercraft and outboard engines, they
support the proposed standards for implementation in 2009. They  note that EPA anticipates
manufacturers will meet these standards with readily available technology - improved fueling
systems and other in-cylinder controls - and, therefore, question why the agency did not assess
                                       3-21

-------
                                                           Chapter 3: Marine SI Engines
the feasibility of catalysts for these engines, for the purpose of pursuing future, more rigorous
catalyst-based standards. They recommend that the agency conduct such an analysis and proceed
with additional standards accordingly.

       Pennsylvania DEP supports EPA's proposed standards and implementation schedule for
marine spark-ignition engines and vessels.  Since the standards proposed for personal watercraft
and outboard engines appear to be easily achieved by manufacturers, DEP urges EPA to assess
the feasibility of additional technology for the future as quickly as possible.

       MARC AQ Forum stated that EPA should investigate the feasibility of using catalysts to
reduce emissions from personal watercraft and outboard engines. If such technology proves
workable, EPA should move expeditiously to set more stringent emissions standards for these
engines.

       SCAQMD staff believes that more stringent catalyst based standards are appropriate for
this category. The California Air Resources Board staff in developing the outboard/personal
watercraft standards in 1998 identified catalyst based technology as one of the possible
technologies to meet the proposed standards.  Their analysis showed that challenges in bringing
catalyst technology to marine  engines existed, but concluded that they were not insurmountable.
Consistent with this conclusion, the California Air Resources Board proposed state strategy
measure will require new outboard and personal watercraft engines to meet a 5.0 g/kW-hr by
2013 (approximately three (3) times lower than the U.S. EPA  currently proposed standard). This
level of control is expected to  be reached using catalyst based  technology. Review of the Draft
Regulatory Impact Analysis document also shows that currently one personal watercraft
manufacturer has certified engines equipped with an oxidation catalyst, demonstrating that
catalyst based  technology is feasible. Therefore, they believe that a more stringent catalyst based
standard beginning in the 2013 timeframe is appropriate and they strongly urge EPA to consider
adopting this additional standard (i.e., 3 to 5 g/kW-hr) for the outboard/ personal watercraft
category as a second phase of catalyst based standards.

       Wisconsin DNR requested EPA to assess the feasibility of more stringent catalyst-based
emission standards for personal watercraft and outboard engines.
                                          3-22

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
Letters:
Commenter
NMMA
Honda
South Coast AQMD
Wisconsin DNR
NY DEC
Environmental Defense
NACAA
Bombardier
Mercury
MARC AQ Forum
Suzuki
CARS
Pennsylvania DEP
Tohatsu (hearing)
Yellowfin
Document #
0688
0705
0704
0663
0659
0648
0651
0674
0693
0696
0698
0682
0676
0642
0681
Our Response:

       Section 213(a)(3) of the Clean Air Act specifies the criteria EPA needs to consider in
revising existing emission standards. Revised emission standards are to achieve the greatest
degree of emission reduction technologically achievable taking into consideration the cost of
technology in the lead time available to manufacturers, as well as noise, energy and safety
factors. Given these criteria, EPA continues to believe that the proposed OB/PWC standards are
the appropriate standards for these engines for the years in which they were proposed.  These
standards can be met through the expanded reliance on four-stroke engines and two-stroke
direct-injection engines.

       Based on industry input, we understand that our proposed simplification of the form of
the HC+NOx standard would cause undue complexity for industry.  Therefore, we will be
finalizing a HC+NOx standard that utilizes a functional relationship to set the emission standard
for each engine family depending on the power rating, common with the CARB 2008 emission
standards. The final HC+NOx standard is roughly equivalent to the proposed standard, in terms
of stringency, and will achieve more than a 60 percent reduction from the existing 2006
standards.

       We will finalize the proposed CO emission standards for OB/PWC engines. These
standards will result in meaningful CO reductions from many engines and prevent CO from
increasing from engines that  already use technologies with lower CO emissions.  The new
emission standards are largely based on certification data from cleaner-burning Marine SI
engines, such as four-stroke engines and two-stroke direct-injection engines.

       We believe the catalyst technology that will be required to meet emission standards
substantially more stringent than we are adopting has not been adequately demonstrated for
outboard or personal watercraft engines.  Outboard engines are designed with lower units that are
                                       3-23

-------
                                                          Chapter 3: Marine SI Engines
designed to be as thin as possible to improve the ability to turn the engine on the back of the boat
and to reduce drag on the lowest part of the unit. This raises concerns about the placement and
packaging of catalysts in the exhaust stream. As such, we believe the new standards for
HC+NOx and CO emissions are the most stringent possible in this rulemaking. While there is
good potential for eventual application of catalyst technology to outboard and personal
watercraft engines, we believe the technology is not adequately demonstrated to determine
whether or when such technology would be  available. More time to gain experience with
catalysts on sterndrive and inboard engines and a substantial engineering effort to apply that
learning to outboard and personal watercraft engines may allow us to pursue more stringent
standards in a future rulemaking.

3.3.2   OB/PWC standards-lead time

What Commenters Said:

       NMMA and Mercury Marine commented that EPA's proposal has model year 2009 as
the implementation date for the proposed HC+NOx and CO standards, including the FEL caps,
for PWC and OB engines. 72 Fed. Reg. 28,262 (proposed § 1045.103). One calendar year  lead
time to comply with the federal emissions standards and the FEL caps is simply not workable for
these engine segments because of the nationwide scope of the standards. Although these
manufacturers will have some  families that will meet the model year 2008  compliance date in
California, a national rule  (with fleet-averaging and FEL caps) in  model year 2009 would
disallow the sale of older,  carbureted 2-stroke engines and would  force these companies to re-
engineer their entire product line. In turn, this would have a major impact on existing signed
supply agreements with small boat builders which will lead to product shortages and disrupted
business plans. The implementation of a national rule is a considerable undertaking that cannot
be achieved in one year. Assuming that the rule is signed by the end of 2007, manufacturers will
not see a rule published until early 2008. This means that some manufacturers could be starting
production of model year 2009 PWC and OB engines at the same time a final rule is published.
Even if a final rule is signed and published by the end of 2007, there is less than a one-year lead
time for manufacturers.  NMMA requests that EPA extend the implementation date for PWC and
OB engines until model year 2010 and delay the imposition of FEL caps for PWCs until model
year 2011. This results in the industry being able to meet the standards (with fleet averaging) in
model year 2010, and gives industry an additional year to re-engineer the remaining PWC  engine
families that might be subject to FEL caps. Individual NMMA members will provide additional
support in their separate comments as to why the additional delay of the FEL cap for PWCs is
warranted.

       NMMA and Mercury Marine continued to state that a two-year period for
implementation is well supported by several EPA rules promulgated pursuant to its authority in
CAA § 213. For example,  for Recreational Vehicles, EPA provided a four year lead time and
allowed for a phase-in. See Control of Emissions from Nonroad Large Spark-Ignition Engines,
and Recreational Engines  (Marine and Land-Based), Final Rule, 67 Fed. Reg. 68,242 (Nov. 8,
2002). In addition, for the  first marine engine standards for PWC and OB engines, EPA provided
industry with a two year lead time from the time of promulgation  of the standards until the first
implementation date for the emissions  standards. See Control of Air Pollution; Final Rule  for
                                         3-24

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


New Gasoline Spark Ignition Marine Engines; Exemptions for New Nonroad Compression
Ignition Engines at or Above 37 Kilowatts and New Nonroad Spark Ignition Engines at or Below
19 Kilowatts, Final Rule, 61 Fed. Reg. 52,087 (Oct. 4, 1996). As an alternative, EPA can draft
the implementation date as two years from the date of publication in the Federal Register to
ensure that a two-year lead time is preserved.

      NMMA suggested that another approach that also would assist manufacturers in
transitioning to a national standard is a phase-in of 50 percent in model year 2009 and the
remaining 50 percent in model year 2010 and a delay of the FEL cap until model year 2011 for
PWCs. Unlike the SD/I engine category, which is a very different market with unique
distribution and sales arrangements,  a phase- in approach for implementation is well-suited for
the PWC/OB market. This would allow manufacturers to phase out carbureted 2 stroke engines
and provide additional time for redesign and development of engines that can comply with the
standards. EPA used a similar 50-50 percent phase-in for Phase I of the standards for
Snowmobiles in the Recreational Vehicle Rule. See 40 C.F.R. §1051.103.

      Mercury Marine also requests that EPA phase-in the OB standard between 2009 and
2010. Their recommendation is to allow 10% of the manufacturer's carryover product line to be
excluded from the FEL caps in 2009. These units would still be required to utilize credits to meet
the standard.  They believe that there is no need for any exclusions, or modifications, in credit
use or calculations. Starting in 2009, all Outboards would switch over to the new credit
calculations in the new rule.  Carryover credits from the current rule would still be useable for 3
years. Further, in order to not have to recertify most of the product line in just a few months, the
final rule should allow carryover certifications, conducted under the requirements of the current
rule, to be used until recertification of the product is required for other reasons.

      Yellowfin commented that they are a low volume builder of high end offshore center
console  outboard boats. They commented that it is imperative for them to have ample supply of
a variety of engines (2-stroke and 4-stroke). EPA's  proposal that outboards meet the CARB
2008 standards nationally in 2009 would impact their business severely. They recommended
that the  CARB 2008 standards should be implemented nationally in 2010.

      Honda requests that the effective date for compliance be extended to 2010 and not be
2009. They stated that EPA has proposed a 2009 implementation date for outboard exhaust
emissions and outboard fuel lines. Honda will have certified and begun production of engines for
the 2009 MY before this regulation is projected to be finalized.  It will be quite difficult to certify
and produce product with this negative lead time.  However, they do believe that it will be
possible to exhaust certify engines beginning in the 2010 model year.

      Bombardier commented that provided EPA adopts the current 3-Star California exhaust
emission standard  for PWC and Outboards, BRP can fully comply with this standard in MY2009
if EPA allows carry-over data to be used.

      Suzuki stated that although their full outboard engine product line is currently certified to
the 2008 CARB HC+NOx standards, implementation of EPA's  new proposed CO standards will
require design changes to some Suzuki models to ensure that emissions of HC+NOx and CO are
                                      3-25

-------
                                                          Chapter 3: Marine SI Engines
attained with sufficient compliance margin. Considering that production of 2008 models has
already begun, EPA's proposed 2009 model year effective date will provide less than one year of
lead-time which is insufficient for engine families that require changes from their 2008 model
year configuration. As discussed above and assuming harmonization with CARB 2008
HC+NOx levels, compliance with the new CO standards proposed by EPA will require
development effort for some Suzuki models.  Therefore, Suzuki requests EPA adopt a
70%/100% phase-in of the new HC+NOx and CO standards for the 2009 and 2010 model years
to allow for a reasonable development process.

      Yamaha stated in a hearing that in the preamble, EPA has proposed to implement a start
date of Outboard and PWC exhaust emission levels in MY 2009. To Yamaha, Model Year 2009
would mean compliance as of April, 2008 production which may come and go without signage
of this very rule.  Due to the protracted direction and ever dynamic time frames experienced with
this NPRM, their Engineering and product planning staff are respectfully requesting that in order
for Yamaha to re-evaluate their current model line-up, readjust the mapping and fuel calculation
of current 4 stroke technology required to achieve a lower emission level across our product line
and to be allowed to utilize our emission credits earned in 2006, 2007 and 2008 Tier I, an
additional  1 model year lead time will be needed.

      Yamaha continued to comment that this would be MY 2010 which for Yamaha would be
production starting April of 2009.  At this point all elements of the emission levels including the
FEL cap within this proposal would go into effect.  This in essence would disallow the sale of
all carbureted 2 stroke engines from this point on, achieving one of EPA's objective goals.

      Yamaha commented that the implementation dates outlined in the proposal reflect a
Compliance Date of MAT (model year) 2009. For Yamaha this would mean compliance for their
line up of 25 engine  families by April of 2008 production start period. That is if the rule is even
signed on a time frame prior to this date. Yamaha recognizes the EPA is aware of compliant
engines in California under the California ARB mandates but that quantity and models sold in
CA is very small compared to a 50 state basis (10 families vs.  25 nationally). Yamaha has over
200 different model variations to supply the marine industry with appropriate designs and use
characteristics for the boating public. It is their position that the M/Y 2009 is unreasonable and
unobtainable for Yamaha based on many factors. Their current facility is working at and beyond
peak output to supply a world market. To affect new mapping and fuel calibrations to any
already taxed system will not be achievable in the proposed time period.

      Yamaha continued to comment that another area affected by the proposed dates is long
term supply contracts to many boat builders in the US that in their long term planning pre-existed
EPA action and could not foresee a start date of this proposal.  Due to its dynamic nature,  to
incorporate the necessary changes in boat design (flotation and transom strength) and sales
structure is impossible for 2009. If Yamaha cannot continue to supply these engines currently
being used by the builders a product shortage will occur causing business disruption to very
small business owners and many parties face potential litigation for breech of contract.

      Yamaha also stated that US protectorates and isolated attainment states (example Hawaii,
Puerto Rico) fall under EPA reach but are supplied by our factory in Japan with product. These
                                         3-26

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


regions due to their remote locations have not embraced (both for technology infrastructure and
cost reasons) new technology in regards to outboard product but, Yamaha USA must list these
units in our product certification process and numbers.

      Yamaha commented that as the EPA can realize, they have a monumental task ahead of
them to achieve compliance even if allowed 3 years lead time. Yamaha also realizes that they
can bring forward certain engines that can meet the new emissions levels as demonstrated by
their compliance in CA. With this knowledge Yamaha is requesting that if the EPA cannot see
the need to hold off starting exhaust emission compliance until M/Y 2010 then they propose a
phase in amount of 50% of compliant engines (based on total engine sales) in 2009 with 50%
exemption with no FEL or NTE caps and achieve 100% compliance in Model Year 2010 with all
caps in place. This extra year is consistent with lead time flexibilities allowed in other EPA
rulemakings.

      NMMA member companies such as Ranger Boats, Triton, Premier Marine Inc.,
S2Yachts, Lund Boat Co, Brunswick Corporation, Brunswick Commercial and Government
Products, Inc., Lowe Boats, Godfrey, Challenger Power Boats, Cigarette Racing, Massachusetts
Marine Trade Association submitted comments to the proposed rule. Fourteen equipment
manufacturers support 2010 (or later) for outboards due to the fact that outboard manufacturers
were planning their new OB engine designs for 2010 and moving implementation to 2009 would
result in some engine designs not being available for about one year. In order to remain
competitive and assure a smooth transition, they need to have engine designs available. Some
companies have international business and reputations that are needed to maintain for success.
Some companies work on smaller margins and need all engine designs to be available.  One
manufacturer stated their desire for a gradual phase-in with full compliance by 2012.

      Tohatsu stated in a hearing that it is quite a tough job for a small manufacturer like
Tohatsu who has total employees of less than 500 people to re-develop and set calibration fuel,
ignition timing, etc. and also comply with evaporation requirements. And naturally these
changes will also require a new batch of deterioration testing at 350 hours for all models.

      Sea Ray commented that in the rule, it is proposed that OB engines be compliant to the
CARB 2008 emission standards by 2009. It is understood that OB manufacturers have been
preparing for this changeover but with a 2010 target date in mind. Although it appears that most
engines will comply by 2009, having this extra transitional year will be beneficial to all
concerned. If the implementation date is accelerated to 2009, there may actually be some
outboard model engines that will no longer be able  to be sold in the United States. The industry
currently faces enough issues regarding sales of boats in these use categories.

      NACAA commented that with respect to personal watercraft and outboard engines, they
support the proposed  standards for implementation  in 2009.

      Pennsylvania DEP supports EPA's proposed standards and implementation schedule for
marine spark-ignition engines and vessels.
                                      3-27

-------
                                                          Chapter 3: Marine SI Engines
       CARB recommends that U.S. EPA revise the implementation date for this standard to
begin in 2008 rather than in 2009 as proposed. Although slight, the potential exists for unfair
competition between California dealerships and out-of-State outboard/personal watercraft
dealerships that would be permitted to sell higher emitting, but less expensive, outboard/personal
watercraft engines in 2008. ARB believes that sufficient flexibility already exists in federal
regulations (e.g., 40 CFR 1068.240, 245, or 250) to address the compliance concerns mentioned
in the preamble for manufacturers,  if any, that do not sell outboard/personal watercraft engines in
California and which because of this, may need more time to comply with the proposed standard.

       NJ DEP commented that specifically, several CARB standards for exhaust emissions are
fully phased-in between 2005 and 2008, whereas the proposed phase-in dates for the
corresponding federal standards do not begin until 2010. Of most concern, the special provisions
for small and medium manufacturers may delay full compliance until 2014. In light of the fact
that manufacturers will already be providing cleaner engines and equipment to California and
that technology issues will not be a factor, these cleaner engines and equipment should be
required to be made available sooner nationwide.

       Environmental  Defense commented that EPA is proposing to implement the O/PW
standards in model year 2009. California's comparable HC and NOx emissions standards take
effect in model year 2008. While Environmental Defense agrees with EPA that it is feasible to
implement these standards nationally one year after CARB's take effect, they see no reason why
the standards cannot be implemented in 2008. As EPA notes in its explanation for this near-term
implementation date, many manufacturers are already  selling lower emission engines  that meet
the CARB HC and NOx standards nationwide.  These  manufacturers will not need to  do
anything in order to comply with the proposed federal  O/PW standards. Therefore, they urge
EPA to better explain its reason for the 2009, as opposed to 2008, implementation date.
                                         3-28

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
Letters:
Commenter
NMMA
Sea Ray
Honda
NJDEP
Environmental Defense
NACAA
Bombardier
Yamaha
Mercury
Suzuki
CARS
Pennsylvania DEP
Yamaha (hearing)
Tohatsu
Ranger Boats
Triton
Premier Marine Inc.,
S2Yachts
Lund Boat Co
Brunswick Corporation
Brunswick Commercial and Government Products, Inc.
Lowe Boats
Godfrey
Challenger Power Boats
Cigarette Racing
Massachusetts Marine Trade Association
Document #
0688
0683
0705
0710
0648
0651
0674
0721
0693
0698
0682
0676
0642
0642
0628
0656
0613
0697
0655
0695
0652
0660
0645
0644
0637
0634
Our Response:

       We have considered the many comments we received supporting our proposed OB/PWC
timing or arguing for different timing.  Several air quality agencies and environmental
organizations argued that earlier implementation of technologies is feasible. Many manufacturers
commented that they will require an additional year to make their entire lineups compliant with
the national rule.

       We have considered the time required by the industry to complete the necessary design,
development, and validation activities for their product lines, and have concluded that 2010 is the
appropriate date for the new emission standards of OB/PWC engines. The option suggested by
commenters for a 50/50% phase-in for 2009 and 2010 was not a feasible option because the rule
will not be signed until after the 2009 model year begins. Essentially this phase-in would have
allowed them to sell carbureted two-stroke engines for an additional year beyond the proposed
implementation dates.  The majority of the remaining engines can meet the new standard either
directly, or through credit exchanges. By delaying the implementation date to 2010,
manufacturers still have the additional year of lead time requested, beyond the proposed
                                       3-29

-------
                                                          Chapter 3: Marine SI Engines
implementation date, to phase-out carbureted two-stroke engines.  The final rule gives two years
beyond the implementation date of the California standards of similar stringency.  In addition to
phasing-out carbureted two-stroke engines, manufacturers may need additional time to refine
emissions calibrations for engines not currently sold in California. The additional time will give
manufacturers time to address any models that may not meet the upcoming California standards
or are not sold in California.  This also accommodates the lead time concerns with the timing of
this final rule as expressed by the commenters.

       The new exhaust emission standards represent the greatest degree of emission control
achievable in the effected time frame. While manufacturers can meet the standards with their
full product line in 2010, requiring full compliance with a nationwide program earlier, such as in
the same year that California introduces new emission standards, will pose an unreasonable
requirement for manufacturers to develop entire product lines compliant with the new standards
with little to no lead time. Allowing two years beyond California's requirements is necessary to
allow manufacturers to certify their full product line to the new standards including the
additional CO requirement, not only those products they will make available in California.

  3.4   High-performance engines

3.4.1   Standards and relationship to ABT

What Commenters Said:

       NMMA commented that for CO, EPA is proposing a 350 g/kW-hr standard for high-
performance engines. NMMA supports this level and agrees that the technological challenges
faced by high performance engines require a CO standard at that level. Individual NMMA
members will provide further comments and test data supporting the CO level proposed by EPA
in the rule. With regard to HC+NOx, EPA requests comments on  the need for and level of
alternative emissions standards for high-performance SD/I engines. While EPA proposes two
possible alternatives, NMMA members believe that the most appropriate approach for the high-
performance engine segment is a modification of the second suggested alternative, which is a 15-
22 g/kW- hr standard for the high-performance segment, and to disallow credits. 72 Fed. Reg. at
28,117. NMMA recommends instead that EPA adopt a non-catalyst based standard with a cap
set at 20 g/kW-hr for engines with rated output of 373 kW-484 kW in 2010 with a further
reduction to 16 g/kW-hr in model year 2011. NMMA will also support a cap of 25 g/kW-hr for
engines with rated output of 485 kW and above in model year 2010 with a further reduction to 22
g/kW-hr in model year 2018. These recommended levels are conditioned on the option of using
the modified test cycle described below. Consistent with EPA's second alternative, NMMA also
recommends that no averaging, banking or trading of credits be allowed for either HC+NOx or
CO.  Most high-performance engine manufacturers  do not have products below the 373 kW
rating with which to average. In addition,  these manufacturers cannot rely on credits being
available on the open market from their competitors. By removing the option for averaging,
banking and trading, EPA will ensure a level playing field among  all manufacturers of high-
performance engines.

       As EPA finalizes the standards for high-performance engines, NMMA encourages the
Agency to work with C ARE to ensure that the standards for high-performance engines are


                                         3-30

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


harmonized to the greatest extent possible. CARB staff included in the "ARB Staff Report:
Initial Statement of Reasons for Rulemaking, September 30, 2005" that it was the staffs
intention to return to the Board prior to the scheduled 2009 implementation date of the standards
for "High Power Engines (greater than 373 kW)" and that they are awaiting the promulgation of
the federal regulation. Thus, harmonization is clearly a priority for the CARB and NMMA urges
EPA to work cooperatively with CARB to ensure consistency among the two regulatory
schemes.

      Timor supports fixed standards for all high-performance engine manufacturers. Two-
tiered:  373-484 kW and >485 kW. Ilmor supports a rule with no ABT for high-performance
sector.  Ilmor supports harmonized standard for high-performance engines  (>373 kw) for EPA
and ARB.  In a hearing, Ilmor commented that they estimate that 80% of the High-Performance
engines are produced by 5 or possibly 6 manufacturers. (Mercury, Ilmor, Teague,  Sterling,
Flagship, Chief). An additional 10-20 very small businesses, produce as little as 15-25 engines
per year each.

      NMMA members (North American Sleekcraft, Inc., Brunswick Corporation) commented
that catalytic converters not practical for low niche market.  They produce boats that use engines
over SOOhp.  EPA should put a cap on the current emission limitations for high performance for
level playing field for those who make such boats.  They believe EPA realizes catalytic
converters are not feasible on high performance engines.

      Brunswick makes boats that use engines over SOOHP. They commented that the only
logical choice for USEPA is to put a cap on the current emission limitations for high-
performance engines in order to create a level playing field for those few manufacturers that
make high performance engines. Brunswick believes USEPA realizes that catalytic converters
are not feasible on high performance engines.

      Mercury Marine commented that EPA's proposal recognizes the unique aspects of high-
performance engines and will provide the necessary flexibility as long as several additional
revisions are implemented.  Mercury Marine is supportive of the flexibility provided for high-
performance SD/I engines in the proposed rule. Mercury Racing produces High-Performance
Engines as a stand alone division, and competes with several  small businesses in this market.
This is a uniquely American Industry, employing several thousand people between the engine
manufacturers, boat builders and dealers. It is imperative that the same standards  apply to all
manufacturers. That said, several of these measures require additional revision in several
respects to ensure that the standards both achieve the reductions that EPA intends as well as
remain workable for the high-performance segment.

      For CO, EPA is proposing a 350 g/kW-hr standard for high-performance engines.
Mercury Marine supports this level and agrees that the technological challenges faced by high-
performance engines require a CO standard at that level.  Mercury Marine has supplied
confidential test data that supports this standard.

      EPA suggested that a possible way to reduce emissions from High Performance Engines
was to add an air pump. Mercury Marine commented that first, the size of an air pump that
                                      3-31

-------
                                                          Chapter 3: Marine SI Engines
would result in any meaningful reductions in emissions would be very large, and require
considerable power to drive it.  Further, no such pump currently exists.  Mercury Racing tested
air pumps some years ago, and was unable to get them to survive for more than 90 minutes of
operation.

       With regard to HC+NOx, EPA requests comments on the need for and level of alternative
emissions standards for high-performance SD/I engines.  While EPA proposes two possible
alternatives, Mercury Marine believes that the most appropriate approach for the high-
performance engine segment is a modification of the second suggested alternative, which is a 15-
22 g/kW-hr standard for the high-performance segment and disallow credits.  72 Fed. Reg. at
28,117. Mercury Marine recommends instead that EPA adopt a non-catalyst based standard with
a cap set at 20 g/kW-hr for engines with rated output of 373 kW-484 kW in 2010 with a further
reduction to 16 g/kW-hr in 2011.  Mercury Marine will also support a cap of 25 g/kW-hr for
engines with rated output of 485 kW and above in 2010 with a further reduction to 22 g/kW-hr in
2018.   These recommended levels are based on EPA offering the option of using the modified
test cycle described below.

       Mercury commented that these standards will provide meaningful reductions in emissions
from High Performance Engines.  Mercury Racing has tested several existing engines.  The
current engines in the under 485 kW category have shown HC + NOx values in the range of 11 -
18 g/kW-hr. They believe that there are engines, built by smaller companies, utilizing
carburetors that are considerably higher on emissions. Every engine company has access to fuel
injection technology,  and they believe that a standard that forces the use of better, available,
technology is appropriate. By 2011 they are recommending a cap of 16 g/kW-hr.

       For the category of over 485 kW, Mercury Racing currently has engines that have shown
emissions totals of over 34 g/kW-hr. As with the lower category, they believe that there are
carburetor equipped engines being produced by other manufacturers that are considerably higher
than this.  As previously stated, every engine company has access to fuel injection technology,
and Mercury believes that a standard that forces the use of better, available, technology is
appropriate.  Mercury Racing has been able to calibrate their large engines down to
approximately 21 g/kW-hr HC + NOx.  Given those results, they endorse a standard set at 25
g/kW-hr HC + NOx for this category in 2010, with a long term reduction to 22 g/kW-hr for
2018.

       Sterling Performance is a small business engaged in the building of high performance
marine engines and has been in this business for over 20 years. They are involved with racing
and pleasure boat engines of the highest performance and durability. The high performance
inboard marine sector consists of a very low volume of engines that we  estimate the total U.S.
annual sales of all builders combined to be less than 1500. These engines are generally used by
other small businesses to power the watercraft they sell.  Sterling Performance supports the
proposal of the removal of the option for allowing the averaging, banking or trading of credits
for either HC+NOx or CO. Since they do not produce engines below a rated output of 485 kW,
they have  nothing with  which to average. Sterling asks that only a "level playing field" be
considered for all manufacturers of high performance engines. They support a cap of 25 /kW-hr
for engines with a rated output of 485 kW and above in model year 2010 with a further reduction
                                          3-32

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


to 22 g/kW in model year 2018. In regards to certification testing, the E4 duty cycle overstates
the idle fraction and an alternate duty cycle that allows for a nominal load factor of 15% in mode
5 would be more appropriate. With the current ARB standard in place in California, Sterling
Performance as a small business is essentially "out of business"  in that state. It is of course their
hope that the USEPA adopts a standard which will enable them to continue to build engines and
further more that it may be harmonized with ARB so that once again the California market is
open to them as a small business manufacturer of high performance engines.

       California ARB recommends that U.S. EPA remain committed to the 5 g/kW-hr
HC+NOx standard and 2009 start-date for high performance sterndrive/inboard engines to align
with existing California requirements, or to at least pursue an approach that yields equivalent
emission benefits. ARB recognizes the challenges faced by small volume manufacturers of high
performance engines to comply with the 5 g/kW-hr HC+NOx standard;  however, they have
equity concerns over giving a more lenient standard to the  segment of industry with the product
most able to absorb the costs of compliance. Still, ARB recognizes the benefits of national
harmonization and is open to reasonable alternatives that would  preserve the emission reductions
of the existing spark-ignition marine regulations in California. ARB staff will carefully review
the final U.S. EPA decision in this matter and proceed accordingly in determining whether or not
a change is warranted for California's high performance engine requirements.

       NY DEC commented that  high performance engines available to the general public (i.e.,
not solely for competition) should be held to the same standards as all other sterndrive and
inboard engines.

Letters:
Commenter
NMMA
NY DEC
Ilmor
Sterling
Mercury
CARS
North American Sleekcraft, Inc.
Brunswick Corporation
Document #
0688
0659
0658
0665
0693
0682
0666
0695
Our Response:

       We considered all the comments and are finalizing non-catalyst based standards for high-
performance engines.  The final rulemaking sets the HC+NOx emissions standards in 2010 at 20
g/kWh for engines with output less than 485kW and 25 g/kWh for engines with output over 485
kW. In 2011 and later model years, the HC+NOx emission standards drop to 16.0 g/kW-hr for
engines at or below 485 kW and 22.0 g/kW-hr for bigger engines. The final standard maintains
the proposed 350 g/kWh CO standard that is effective in 2010.  Since the standards being
adopted for SD/I high-performance engines are less stringent than originally proposed, we are
not including the SD/I high-performance engines in the ABT program. Manufacturers are
                                      3-33

-------
                                                           Chapter 3: Marine SI Engines
required to meet the emission standards for SD/I high-performance engines without using
emission credits.

       We respect NY DEC's desire to obtain greater emission reductions in the high
performance engine segment; however, we have determined that the SD/I emission standards are
not a feasible option for the high performance engines. Catalytic converters, which are required
to meet the new SD/I emission standards, are not a viable technology in high performance
engines.  These engines produce very high exhaust flow rates and temperatures that make
catalysts  incapable of sustained and effective operation over extended engine operation. We are
therefore implementing the most stringent standards achievable through calibration development
and the expanded use of electronic fuel injection in high performance engines.

       ARB has recently relaxed its exhaust emission standards for SD/I high performance
marine engines to be reflective of emission levels that can be attained without the use of
catalysts.  These emission standards are similar to those finalized today in this rule. To
compensate for the associated shortfall in emission reductions, compared to the original
standards, ARB is requiring that high-performance vessels use evaporative emission control
systems including carbon canisters and low permeation tanks and hoses.  Similarly, we are
finalizing evaporative emission standards for all SI marine vessels subject to this rule.

3.4.2   Lead time

What Commenters Said:

       NMMA commented that the SD/I marine engine manufacturing industry will need lead
time to comply with the emissions standards in the proposal. They continued to comment that
this is especially true for the high-performance engine segment which will have to develop the
technology to ensure compliance with the emissions standards without the use of averaging. The
fact that engine manufacturers must comply with the high-performance California emission
standards in 2009 does not assure compliance with a model year 2009 implementation date for
national emission standards. As EPA states in the preamble, California represents only a small
portion of the market and manufacturers will need to develop control technology for their entire
product line. This cannot happen overnight and certainly manufacturers cannot begin the process
of developing the control technology until the levels of the standards are finalized. Therefore,
NMMA supports a model year 2010 implementation date for large businesses.

       With regard to small businesses that are in the high-performance segment, NMMA
supports the additional compliance time proposed for these manufacturers. They believe that
2011 is appropriate for high-performance small businesses and provides the requisite time for the
control technology to be developed and tested. In addition, given that NMMA is recommending
no averaging for this segment, the additional years for compliance will be critical for this
segment.

       Mercury Marine also commented that the SD/I marine engine manufacturing industry
will need lead time to comply with the emissions standards in the proposal.  This is especially
true for the high-performance engine segment which will have to develop the technology to
ensure compliance with the emissions standards without the use of averaging.  The fact that


                                          3-34

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


engine manufacturers must comply with the California emission standards in 2009 does not
assure compliance with a model year 2009 implementation date for the federal emission
standards. As EPA states in the preamble, California represents only a small portion of the
market and Mercury Racing may not offer all products in California in 2009.  Therefore, the
national rule implementation needs to be at least 2010 to provide  sufficient time to develop lower
emissions versions of these engines.

       With regard to small businesses that are in the high-performance segment, Mercury
Marine will support additional time for compliance, but that additional time should be 2011.
Allowing all of Mercury Racing's competitors to not comply until 2013 is creating an unfair
advantage to these companies which have access to the same technologies and capabilities as
Mercury Racing.

       NMMA members (North American Sleekcraft, Inc., Lowe Boats,  and Cigarette Racing)
supported 2010-2011 implementation for catalysts to evaluate and design the lower emission
engines into boats while ensuring performance and safety and soundness in economy.

Letters:
Commenter
NMMA
Mercury Marine
North American Sleekcraft, Inc
Brunswick Corporation
Baja Marine Corporation
Cigarette Racing
Lowe Boats
Document #
0688
0693
0666
0695
0726
0637
0660
Our Response:

       Given the timing of the final rule, we agree with NMMA's suggestion to delay
implementation until 2010 for large businesses. This will allow sufficient lead time to complete
the design and certification effort associated with meeting the new emission standards. We
however, will maintain the 2013 implementation date for small businesses.  Small businesses do
not currently have access to the testing equipment necessary to perform emission testing and
subsequent emissions calibration. This additional lead time will allow them sufficient time to
perform this testing and emissions calibration work. In addition, it will provide them sufficient
time to upgrade their carbureted engines to electronic fuel injection.  Given the high fuel rates of
high performance engines, custom fuel injection systems will need to be developed for many  of
these engines.
                                      3-35

-------
                                                                      Chapter 3: Marine SI Engines
3.4.3   Special provisions for high-performance engines
Summary of Comments
Response
NTE Testing: NMMA and Mercury Marine support the
proposal to not to apply NTE requirements to the high
performance engine segment. They state that many of
the manufacturers in this segment are small businesses
and the additional testing will cause significant testing
burden and costs.
We will finalize these provisions as proposed.
Therefore, we will not apply NTE requirements to the
high performance segment.
Certification Testing: NMMA, Mercury Marine, and
Ilmor support an alternative E4 test cycle for the high
performance engine segment.  They propose to increase
the idle load at the Mode 5 point in the E4 test cycle
from 0% to 15% load. The proposal is based on data
from high performance boat builders and owners. The
data reflected that high performance vessels spend
significantly less time at idle (15%) than the E4 test
weighting of 40%.  In addition, the data showed that
typically 8% of operating time is at idle with no load and
18% of the time is at idle in gear, which is represented
by the 15% load proposed at idle.	
We will adopt the optional alternate E4 test schedule for
the high performance engine segment which allows 15%
load at the Mode 5 idle point based on the data supplied
by industry.  We believe this is sufficient relief for the
high performance engine segment based on the data
provided by industry.
Portable analyzers:  NMMA suggested that portable
analyzers do not provide any meaningful relief in testing
burden. They stated that this equipment was not
developed for the high-performance segment and that
discrepancies between portable analyzers and a full test
lab would create problems.

In addition, NMMA expressed confusion that EPA
would refer to portable analyzers for in-use testing of
high performance engines given that in-use testing
requirements were not proposed for SD/I engines.
We have used currently available portable analyzers to
perform valid and accurate measurement of emissions
from high performance marine engines. It is true that
portable analyzers will in some cases have somewhat
greater variability than conventional laboratory
equipment. Manufacturers may choose to take this
greater variability into account as part of the decision
whether or not to use portable analyzers for certification.
If compliance margins are not big enough or where
engine manufacturers otherwise do not want to deal with
this greater availability, they may instead opt for the
more expensive testing with conventional laboratory
equipment. We note, however, that portable analyzers in
some cases meet laboratory specifications, in which case
no greater variability would be expected.

The final preamble clarifies that EPA is adopting a
provision that allows for SD/I high performance engine
testing to be performed with different equipment than is
specified for the laboratory with less restrictive
specifications and tolerances.  The less restrictive
specifications are typical of the specifications required
for in-use testing.	
Warranty and Useful Life:  NMMA and Mercury Marine
support the high performance warranty and useful life
limits proposed in the NPRM. The proposal limits
warranty and useful life to three years or 150 hours for
engines with 373-484 kW output and one year or 50
hours for engines with >485 kW output. They also state
that the warranty and useful life limits proposed by EPA
are consistent with CARB's limits and it makes sense
from a policy and technical perspective to harmonize the
We will finalize the proposed high performance
warranty and useful life provisions, which are
harmonized with California ARB's provisions.
                                                  3-36

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
requirements.
Broad Definition of Engine Families'. Ilmor supports the
proposed broad definition of engine families in the
NPRM. This proposal allows high performance engine
manufacturers to group all high performance engines
into a single engine family based on good engineering
judgment.

We will finalize the proposed provisions for grouping all
high performance engines into a single engine family
based on good engineering judgment.
Commenter
NMMA
Mercury Marine
Ilmor
Document #
0688
0693
0658
  3.5   Cross-category issues related to emission standards

3.5.1   NTE limits (NTE Testing Burden and Need)

What Commenters Said:

       Several commenters stated that they do not support the not-to-exceed (NTE) standards
proposed in the regulation. Suzuki does not believe that NTE standards are necessary for the
outboard engine product category in general.  Honda suggested that EPA reconsider the NTE
proposal of this marine engine regulation and not adopt NTE for marine engines. Honda also
commented that the NTE section of this marine regulation should address the basic issue of
defeat devices and not attempt to create a new undocumented test cycle with infinite test points.
NMMA and Mercury suggested that the ABT program ensures that the emissions from a
manufacturer's fleet meet the standards, therefore NTE is not necessary.

       NMMA, Mercury, and Suzuki commented that the test burden associated with NTE
standards is considerable.  NMMA and Mercury also claim that the costs associated with the
NTE tests are not adequately represented in the draft RIA. The commenters claimed that the
practical effect of this requirement is that marine engine manufacturers will have to run hundreds
more tests in the  development process for engines.  Such a resource intensive requirement is a
considerable burden for this industry with little to no benefit to the environment.

       Yamaha commented that EPA originally explained that NTE was a component of
certification only but now wants to utilize it as a form of Selective Enforcement Audit protocol
causing undue and unsubstantiated burden on the engine maker.

       As an alternative to NTE, Honda suggests that EPA consider the acquisition of data from
actual boat use (SD/I, outboard, and PWC with the full variety of engine technology that is
available to power these vessels) that represents the nominal and off-nominal operating
conditions. The data can be used to define a test procedure that is not infinitely burdensome and
can be applied to all marine engine technologies.
                                      3-37

-------
                                                          Chapter 3: Marine SI Engines
       Honda does not support adoption of an NTE provision for marine engines. They were not
in support of NTE provisions for ATVs in an earlier EPA rulemaking and the fundamental
principles behind their opposition then apply here for this marine engine proposal. As an
alternative, they would suggest that EPA consider the acquisition of data from actual boat use
(SD/I, outboard, and PWC with the full variety of engine technology that is available to power
these vessels) that represents the nominal and what is claimed to be off nominal operating
conditions. From this data it may be determined that the extremes of operating conditions can be
better defined. The data can also be used to define  a test procedure that is not infinitely
burdensome and can be applied to all marine engine technologies. On-the-water test procedures
are also a section in the proposal where EPA is attempting to create a compliance limit when
there is no test data, no test procedure, no hardware input and output parameters, and no basis to
assume that there is some actual, reliably measurable, data that could be generated and compared
with a dynamometer test.  This is the basis for Honda's suggestion that the NTE section of this
marine regulation should address the basic issue of defeat devices and not attempt to create a
new undocumented test cycle.

       Honda does not understand how the NTE sections apply specifically to outboards and
PWCs.  They assume that EPA may have intended that some of these sections apply only to SDI
vessels. Outboards and PWC do not necessarily include any sensors or controls in a basic 4-
stroke carbureted engine so including them in this requirement, especially torque value
broadcasting, would be a complete change in their configuration clearly not anticipated in either
the regulatory implementation date nor in the cost analysis associated with the emission
reductions. Without engine management,  a simple  air / fuel map of the engine in the operating
range would be sufficient to demonstrate that the engine will provide proper emission
performance and not introduce any form of "defeat device".  The basic purpose of NTE is to
prevent the use of a defeat device that would impair emissions performance under normal
operating condition or, under particular conditions, change the engine performance for some
other benefit while adversely affecting emissions. EPA seems to have clearly stepped beyond
this purpose and is in effect creating a new engine test cycle with infinite test points. Creating a
new test cycle and setting standards for that cycle without real world data demonstrating that it is
representative of boats in operation and is technically achievable by the boats / engines being
regulated are clearly a violation of the basic technical principles upon which EPA has always
developed test cycles.

       ARB commended U.S. EPA for its leadership role in developing and adopting NTE
standards and test procedures for sterndrive/inboard engines. ARB believes the standards will
allow sterndrive/inboard engine performance to be evaluated in-use under real-world operation.
ARB staff recognizes the value to industry of harmonized requirements and will carefully review
U.S. EPA's NTE program when determining what NTE standards are appropriate for
California's own program.
                                          3-38

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
Letters:
Commenter
NMMA
Bombardier
Yamaha
Mercury
Suzuki
CARS
Honda
Document #
0688
0674
0721
0693
0698
0682
0705
Our Response:

       We disagree with commenters' position that NTE is unnecessary.  NTE is a critical part
of a comprehensive emissions program that is intended to ensure that emission controls function
with relative consistency across the full range of expected operating conditions.  Without NTE,
we would not be able to ensure the emissions benefits expected from the regulation are realized
in-use.

       Commenters stated that the ABT program ensures that emissions from a manufacturer's
fleet meet the standards and therefore NTE is not required.  However, the commenter did not
explain their perceived relationship between ABT and NTE.  These are two very different
programs. ABT refers to emission credit exchanges between different engines. NTE is a set of
standard test procedures intended to ensure that emission control is achieved in-use.

       We disagree with Honda's comment that NTE should solely address the use of defeat
devices. No single test procedure or test cycle can cover all real-world applications,  operations,
or conditions. Yet to ensure that emission standards are providing the intended benefits in use,
we must have a reasonable expectation that emissions under real-world conditions reflect those
measured  on the test procedure.  The defeat device prohibition is designed to ensure that
emission controls are employed during real-world operation and as a result emission  reductions
are  achieved in the real world, not just under laboratory testing conditions. However, the defeat
device prohibition is not a quantified standard and does not have an associated test procedure, so
it does not have the clear objectivity and ready enforceability of a numerical standard and test
procedure. We believe using the traditional approach, i.e., using only a standardized laboratory
test procedure and test cycle, makes it difficult to ensure that engines will operate with the same
level of control in use as in the laboratory and therefore makes it difficult to enforce a defeat
device prohibition.  Thus, we believe there are significant advantages to establishing NTE
standards. In addition, the final NTE test procedure is flexible,  so it can represent the majority of
in-use engine operation and ambient conditions. The NTE approach thus takes all the benefits of
a numerical standard and test procedure and expands it to cover a broad range of conditions.
With the NTE approach, in-use testing and compliance become much easier because emissions
may be sampled during normal boating. In sum, by establishing an objective measurement, our
NTE approach makes  enforcement of defeat device provisions easier, provides more certainty to
EPA and the industry, and is crafted to be flexible and represent most in-use engine operation
and ambient conditions.
                                       3-39

-------
                                                          Chapter 3: Marine SI Engines
       We disagree with industry's comments that the test burden associated with NTE is
considerable with either current or future engines. Data supplied by manufacturers show that
emissions from existing low emission engines in many areas of the NTE zone are generally
below the limit today.  We believe the technology used to meet the standards over the five-mode
duty cycle will meet the caps that apply across the NTE zone. We therefore do not expect the
final NTE standards to cause manufacturers to need additional technology.  We believe the NTE
standard will not result in a large amount of additional testing, because these engines should be
designed to perform as well in use as they do over the five-mode test.  However, our cost
analysis in the Final RIA accounts for some additional testing, especially in the early years, to
provide manufacturers with assurance that their engines will meet the NTE requirements and
therefore meet applicable standards in-use.

       The test burden also will not be as great as industry assumed from the proposal because
of the lead time and carry-over provisions permitted in the final regulation. Manufacturers have
at least two years to develop efficient NTE test methods that focus on areas of high emissions
before NTE is required. We also added a small business provision that allows an additional year
of lead time. We exempted the high performance engine segment from NTE testing altogether
because we have very limited information on their detailed emission characteristics and we are
concerned about  extent of testing that would be required by the large number of affected engine
manufacturers that are small businesses.  We also considered testing burden by allowing
manufacturers to carry-over certification on engines certified prior to 2010 until 2012 for OB
engines and 2013 for PWC and SD/I. Like emissions certification, the manufacturers will be
able to carry-over NTE certification until the engine design changes significantly.

       We also disagree that the NTE testing burden is not accounted for properly in the RIA.
In the RIA Chapter 6.3.5, we recognized that manufacturers may need to adjust engine
calibrations to meet the proposed standard and collect further data to demonstrate compliance
with the proposed not-to-exceed zone. We therefore allow on average two months of R&D for
each engine family as part of the certification process. Considering two engineers and three
technicians and the corresponding testing costs for the two-month period, we estimate a total cost
of $130,000 per engine family. Unless engine designs were significantly changed,
manufacturers could recertify engine  families each year using carryover of this original test data.
Commenters did  not provide detailed information on their cost estimates for NTE testing.

       Honda commented that actual in-use boat data should be used to create the NTE zone.
We developed this zone based on the range of conditions that these engines typically see in use.
Manufacturers collected data on several engines installed on vessels and operated under light and
heavy load. Chapter 4 of the Final RIA presents this data and describes the development of the
boundaries and conditions associated with the NTE zone. Although significant in-use engine
operation occurs  at low speeds, we are excluding operation below 40 percent of maximum test
speed because brake-specific emissions increase dramatically as power approaches zero. An
NTE limit for low-speed or low-power operation will be very hard for manufacturers and EPA to
implement in a meaningful way.

       We value CARB's support for our NTE testing and we agree with them on the value of
harmonized requirements for NTE test protocol and standards.
                                         3-40

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


3.5.2   Lead time for NTE standards

       NMMA and Mercury Marine commented that there are certain NTE implementation
issues that EPA's proposal fails to consider and accommodate in the proposed requirements for
Marine SI engine manufacturers in § 1045.205. Specifically, the requirement in § 1045.205(p)
that the application for certification contain a statement that all the engines in the engine family
comply with the NTE limits and the requirement to include any relevant testing, engineering
analysis, or other information to support the statement is particularly troublesome. 72 Fed. Reg.
at 28,270. While this  requirement may not be a problem for new engine families, for engine
families that are carried over, EPA must delay the NTE requirements in the certification
application. Otherwise, manufacturers would have the impossible task of having to retest all of
their engine families,  including those that existed prior to the applicability of the NTE standards.
To address the carryover situation,  NMMA and Mercury recommend that EPA include in §
1045.205(p) language that would specify that test data for carryover engines compliant with the
standards can be carried over through model year 2014 and that certification is valid until the
engines must be recertified for other reasons. Section 1045.205(p) should be revised to state:

            (p) For new engine families, state that all the engines in the engine family comply
       with the not-to-exceed emission standards we specify in subpart B  of this part for all
       normal operation and use when tested as specified in § 1045.515. Describe any relevant
       testing, engineering analysis, or other information in sufficient detail to support your
       statement. Through model year 2014, any prior model year engine  certified under the
       Tier I standards in Part 91 may carry over test data and is not subject to NTE as long as
       the engine meets the applicable standards in this subpart.

       This additional language will ensure that manufacturers will be able to transition to the
new standards without having to retest all of their prior engine families that are already
compliant with the  standards.

       Mercury commented that they have a suggestion on the NTE Zone Implementation that
may make it easier to come to an agreement and implement. Whatever approach is put in the
rule, for 2010, 2011, 2012 manufacturers would test to it and report the results with new
certifications. They would make a good faith effort to comply with it, but there would be no
penalty for noncompliance. Then, in 2012, EPA and industry would do a tech. review and see
what worked and what didn't, modify it as needed, and future new certifications would need to
meet it. This is similar to the concept that CARB is using on catalyst monitoring, where for the
first two years, industry has to do catalyst monitoring and store fault codes, but they do not have
to activate the warning horn/MIL light.

       Provided EPA adopts the current 3-Star California exhaust emission standard for PWC
and Outboards,  BRP can fully comply with this standard in MY2009 if EPA allows carry-over
data to be used. It is not possible for BRP to re-test their PWC or Outboard engines for
compliance with the proposed Not  To Exceed (NTE) Zone requirements or proposed  change to
the maximum test speed in time for MY2009 certification. As a result, BRP is supportive of the
NMMA comment to exempt carry-over engine families from the NTE and maximum  test speed
                                      3-41

-------
                                                           Chapter 3: Marine SI Engines
provisions in this regulation through MY2013. Please refer to the carry-over certification
discussion below.

       BRP supports NMMA proposal to have carry-over engine families from the existing
marine regulation and early-certified engine families meeting the exhaust emission standards of
the proposed regulation be exempt from the proposed NTE test requirements and maximum test
speed definition change through MY2013. It is necessary for BRP to phase in engine families to
the new testing requirements over the next few model years. It is infeasible to re-test every
engine family within the next couple years to verify compliance with the NTE proposal. In
addition, allowing carry-over data to be exempt from the NTE and maximum test speed
provisions will create an incentive for BRP and other manufacturers to certify their engine
families to the new emissions standards in an earlier model year.

       Yamaha supports NMMA comments that all HC+NOx compliant engine families under
Tier 1 not be subject to NTE testing until that family undergoes a major change or resubmitted as
new model until M/Y 2014. This will help offset the time and costs associated with an NTE test.

Letters:
Commenter
Bombardier
Yamaha
Mercury
NMMA
Mercury
Mercury
Document #
0674
0721
0693
0688
0693
0716
Our Response:

       Manufacturers commented that certification to the NTE standards requires additional
testing for engine models that are already certified to the new emission standards for California.
In addition, they expressed concern that they may need to recalibrate existing engine models to
meet the NTE standards. Manufacturers commented that this would not be possible by the date
of the duty cycle standard.  For engines already certified in California, manufacturers carry over
preexisting certification test data from year to year. Manufacturers commented that additional
time would be necessary to retest, and potentially recalibrate, these engines for certification to
the NTE standards.  To address these issues regarding lead time needed to retest these engines,
we are not applying the NTE standards for 2010-2012 model year engines that are certified using
preexisting data (i.e., carryover engine families).  For new engine models, manufacturers
indicated that they will be able to perform the NTE testing  and duty-cycle testing as part of their
efforts to certify to the new standards. Therefore the primary implementation date of 2010
applies to these engines.  Beginning in the 2013 model year, all OB/PWC and conventional SD/I
engines must be certified to meet the NTE standards.

       We believe that the NTE requirements are technologically feasible in the time frame
adopted in this rule. These NTE limits are supported by data in the RIA and have been further
                                          3-42

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


confirmed by confidential data submitted by individual manufacturers. Therefore, we do not
believe that a tech review is warranted.


3.5.3   NTE zones, subzones, and test specifications

What Commenters Said:

       NMMA, Mercury, and Bombadier commented that EPA's proposed NTE requirements
do not reflect how marine engines are certified and designed and do not accommodate the
majority of engine designs. They stated that exhaust emissions vary by engine technology across
the 5- mode weighted average test cycle used to determine Marine SI emission certification
levels. In addition, the commenters stated that the emission levels  at each of the five test points
can vary significantly from the declared FEL. The commenters believe that EPA's NTE proposal
forces an area around each point of the certification duty cycle to meet the engine family's FEL
times a multiplier regardless of the certification data for that point. Suzuki commented that their
full line of outboard engines comply with the stringent CARB 2008 HC+NOx levels but EPA's
proposed NTE test requirements and emissions standards under any of the available NTE
subzone sets will be too severe for several existing engine families to attain without costly and
time consuming redesign.

       NMMA, Pleasurecraft Marine, Indmar, Mercury Marine, Bombardier, Volvo Penta, and
Suzuki support using the second alternative discussed in the NPRM preamble, which is  a
weighted average approach to the NTE limit rather than an individual NTE limit for each
subzone. Under this approach, an emission measurement would be made anywhere within each
of the subzones plus idle. The measured emissions would then be combined using the weighting
factors for the E4 modal test.  The commenters believe that the proposed alternative NTE Zone
will ensure a common test methodology to test all different types of marine engines.

       NMMA has provided EPA with a full description of a NTE zone shape that they believe
makes sense for all engine categories and addresses the open loop phase of catalyst operation
during the marine duty cycle.  The proposed shape of the  subzones was supported by industry.
NMMA's proposed a dividing line for Subzone 1 at 85% engine speed and 80% engine  torque to
accommodate all Marine SI technologies, including open-loop fueling for catalyst protection in
the SD/I engines.  They proposed that Subzones 2 and 3 are defined by the ICOMIA 5-mode
cycle, but the wide open throttle point was defined by the 85% speed and 80% torque boundary
of Subzone 1.  NMMA proposed the lower boundary for Subzone 2 at 68% of rated test speed
and Subzone 3 at 51% of rated test speed.  Subzone 4 is defined as the remaining areas of the
NTE zone.

       Bombardier commented that EPA's NTE proposal forces an area around each point of the
certification duty cycle to meet the engine family's FEL times a multiplier regardless of the
certification data for that point.  They also stated that despite the three sets of multipliers
available, this is not a proposal BRP can comply with without substantial lead time. BRP desired
to meet with EPA and other industry members to reach a consensus on the NTE requirements.
                                      3-43

-------
                                                          Chapter 3: Marine SI Engines
       Mercury stated that the key to ensuring that the NTE limits will be workable for all
engine categories is to have a multiplier that will allow for the "worst case" engines. Otherwise,
they believe EPA would need to develop subgroups to accommodate every engine category.
Mercury Marine believes that a multiplier of 2.0 with the weighted zone approach is required to
make this concept work.

       Assuming the proposed weighted-average test method is adopted, Suzuki believes an
appropriate NTE multiplier for 4-stroke outboard motors is 1.6 times the certification PEL for
HC+NOx and CO emissions. Suzuki believes this proposed multiplier will accomplish EPA's
stated objectives for NTE, and will not penalize small 4-stroke  outboard engines that that are not
equipped with fuel injection.

       Yamaha's PLT testing indicates that the multipliers outlined in the proposal are too
stringent when applying to PLT tests of various engine technologies and fuel delivery methods
with little or no break-in time beyond what is allowed for current PLT preparation. Yamaha
recommends NTE multipliers of 1.5 times the FEL (un-weighted).

       Manufacturers have commented that do not have enough information to fully evaluate the
feasibility of the NTE zone for future SD/I engines. Manufacturers have expressed concern that
the new line of supercharged GM will result in engines with higher exhaust temperatures than
current designs. The  commenters suggest that higher exhaust temperatures may require open
loop fuel operation at lower speeds and loads,  including  some operation in subzone 2.

       Several manufacturers submitted data for our analysis and development of multipliers.
The data can be found in the RIA Chapter 4.

Letters:
Commenter
NMMA
Indmar
Bombardier
Mercury
Suzuki
Pleasurecraft Marine (hearing)
Document #
0688
0667
0674
0693
0698
0642
Our Response:

       We have re-worked the NTE test protocol with industry to develop a new approach.  The
proposal discussed several approaches to the NTE testing protocol.  We requested comment from
industry on several alternatives. Industry commenters provided input to the advantages and
shortcomings of these approaches.  Manufacturers specifically stated that there are many
different engine technologies and suggested high multipliers that could be met by existing
engines. We continued to work with the manufacturers since they submitted their written
comments to address these important issues.
                                         3-44

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       The OB/PWC NTE multipliers are slightly revised from the proposed procedure to better
reflect the emissions performance of four stroke engines. We are raising the HC+NOx limit in
Subzones 1, 2, and 3 from 1.2 to 1.4. In the event where OB/PWC engines are fitted with
catalysts, manufacturers would use the NTE requirements for catalyzed engines that were
originally designed for SD/I engines (with catalysts).  This is appropriate because the emissions
characteristics for engines equipped catalysts, in the NTE zone, are driven primarily by the
catalyst efficiency rather than the engine calibration. This is especially true at high speed/power
operation when the engine may need to run rich as a catalyst protection strategy. During this rich
operation, the catalyst would not effectively reduce HC or CO.  Detailed data is included  in the
RIA Chapter 4.

       The two-stroke OB/PWC engines have apparent high engine operation variability, as
stated in the proposal.  Therefore, we singled out the two-stroke engines based on industry
recommendation.  We are adopting a single weighted limit of 1.5 times FEL for the entire zone.

       Four-stroke SD/I engines are unique from the OB/PWC engines because they are
expected to use a catalyst to meet the new standards. We are adopting changes to the subzone
shapes for SD/I in the final rule based on industry comments. First, we are modifying the shape
of the NTE zone to reflect the emissions performance differences between open loop and closed
loop fuel operations. We are combining subzones 2, 3, and 4 into a single subzone to reflect the
common closed loop engine operation in these areas.  Second, we are increasing the subzone 1
area to address the points that require open loop fuel operation to maintain safe exhaust
temperatures based on data from industry. We believe that the finalized subzone 1 area is
properly defined for catalyst-equipped engines based on current engine blocks. In addition,
initial data from General Motors indicates that the finalized subzone 1 may also be appropriate
for 6.0L supercharged engines. However, this is not certain. As engine manufacturers begin
their development of the new catalyst-equipped, supercharged, SD/I engines, more information
will become available on the exhaust temperature characteristics of these engines. If it becomes
apparent that these engines cannot be designed to meet the NTE requirements, then we would
consider revisiting the NTE subzones and limits to address this issue.

3.5.4  Altitude

What Commenters Said:

       ARB strongly encourages U.S. EPA to withdraw its proposal for exempting all
recreational marine engines from compliance with emission standards at altitudes greater than
2000 feet above sea level (< 94 kPa) as described in Section IV.D.(4) of the preamble. Although
the preamble justifies this limitation because of a presumed majority of boating activity at sea
level or low altitude, many lakes in California popular to boaters reside significantly above 2000
feet. Examples include Lake  Tahoe at 6225 feet above sea level, Lake June at 7612 feet above
sea level, and Big Bear Lake at 6743 feet above sea level. Furthermore, the proposed altitude
limitation would effectively exempt recreational marine engines from having to comply with
emission standards in-use for all of New Mexico, Wyoming, Utah, and Colorado, which reside
entirely at or above 2000 feet above sea level. Additionally, fourteen U.S. states in total have a
mean elevation at or above 2000 feet above sea level. While ARB understands that requiring
                                       3-45

-------
                                                           Chapter 3: Marine SI Engines
manufacturers to perform certification testing at high altitudes may be inconvenient, they
maintain that manufacturers must remain liable for complying with emission standards in-use, as
feasible, at all elevations where significant boating activity occurs. As a compromise, ARB
recommends allowing manufacturers to certify engines using test data generated at or around
2000 feet above sea level, but to provide an engineering evaluation stating that the engine will
still comply  with the applicable emission standards up to 8000 feet above sea level. Requests for
exemptions from the 8000 feet above  sea level threshold could be considered on a case-by-case
basis.

Letters:
Commenter
CARS
Document #
0682
Our Response:

       We acknowledge that there are lakes at elevations greater than 2000 feet above sea level.
While this boating activity is less prominent than that occurring at lower altitudes, we agree that
the regulations should not automatically exempt marine engines based on operation above 2000
feet of altitude.  For electronically controlled engines with feedback controls, designing engines
that can compensate for altitude effects is straightforward. The bigger challenge is for open-loop
engines where there is much less opportunity to incorporate design parameters that would
compensate for altitude effects.

       In discussions with engine manufacturers after the proposal, there was general agreement
that the approach we proposed for nonhandheld Small  SI engines would be appropriate to extend
to Marine SI engines. We  are therefore adopting those same requirements for Marine SI engines
in the final  rule. In summary, this would include the following provisions:
    •   Engines must comply with emission standards  in the standard configuration at all
       atmospheric pressures above 94 kPa, which generally corresponds to an altitude of 2000
       feet above sea level.
    •   Engines must comply with emission standards  at atmospheric pressures above 80 kPa,
       which generally corresponds to an altitude of about 6400 feet above sea level. This may
       involve an altitude kit, which would be described in the application for certification with
       supporting information (engineering analysis and/or test data). This atmospheric pressure
       is the lowest value for performing a valid test under 40 CFR part 1065.
    •   Manufacturers must describe their plan for making information and parts available to
       reasonably expect that altitude kits would be widely used in high-altitude areas if the
       engine depends on such a kit for complying at high altitudes.

       See the discussion of altitude-related comments for Small SI engines in Section 2.2.7 for
additional information.

3.5.5   Methane measurement

What Commenters Said:
                                          3-46

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       CARB commented that although not in alignment with California's existing regulations
for outboard/personal watercraft and sterndrive/inboard engines, the use of a total hydrocarbon
(THC) criterion for determining compliance with the HC+NOx standards is not opposed by ARB
since a numerically equivalent THC standard would be more stringent than basing compliance
on only the reactive component of hydrocarbon emissions. California's existing recreational
marine standards are based solely on non-methane hydrocarbon because methane is not an ozone
precursor. However, methane is a greenhouse gas with climate changing potential; therefore,
inclusion in the HC+NOx standard could be beneficial if methane emissions are always
decreased in proportion to non-methane components regardless of the emissions control
technology employed. As an alternative to the present proposal, U.S. EPA might consider the
adoption of a separate standard for methane to ensure more meaningful emission reduction
levels.

Letters:
Commenter
CARB
Document #
0682
Our Response:

       Whether one considers ease of measurement, climate change, or matching the form of the
standard with the available emission control technologies, the conclusion is that a total
hydrocarbon standard is a sound basis for setting emission standards for Marine SI engines.  We
agree with the observation that methane emissions will decrease as a result of setting a THC
standard. We are adopting emission standards in the form of total hydrocarbons, as proposed.

  3.6   Averaging, banking, and trading

What Commenters Said:

       NMMA and Mercury Marine supported the inclusion of an Averaging, Banking and
Trading Program for OB/PWC engines and also for SD/I engines.

       CARB encouraged EPA to rescind provisions for emission credit banking and trading for
all recreational marine engines or to at least depreciate the value of banked credits over time.
They expressed concern that it may be possible for manufacturers to certify their engines to
emission levels that are considerably lower than required, even within proposed family emission
limit (FEL) caps, which could delay the introduction of more stringent emission standards in the
future for some manufacturers (if enough credits have been banked). CARB noted that the EPA
makes a similar argument for disallowing the banking of CO credits from outboard/personal
watercraft engines, and stated that the argument is applicable to the other regulated pollutants as
well as sterndrive/inboard engines.
                                      3-47

-------
                                                          Chapter 3: Marine SI Engines
Letters:
Commenter
NMMA
Mercury
CARS
Document #
0688
0693
0682
Our Response:

       EPA is retaining the ABT programs for Marine SI engines in the final rule.  There will be
one ABT program for OB/PWC engines and a separate ABT program for SD/I engines at or
below 373 kW.  The ABT program for OB/PWC engines will include averaging, banking and
trading provisions for the HC+NOx standard and averaging provisions only for the CO standard.
The ABT program for SD/I engines at or below 373 kW will include averaging, banking and
trading provisions for both the HC+NOx standard and the CO standard. (As described in Section
3.4.1, EPA is finalizing a set of emission standards for high performance SD/I engines that do
not include ABT provisions.) EPA believes ABT programs are an important element in setting
emission standards that are appropriate under Clean Air Act section 213(a) with regard to
technological feasibility, lead time, and cost, given the variety of engines covered by the Marine
SI standards. Depending on their design, ABT programs can create an incentive for the early
introduction of new technology, allowing certain engine families to act as trailblazers for new
technology.  This can help provide valuable information to manufacturers on the technology
before they apply the technology throughout their product line.

       EPA believes the banking and trading provisions are important parts of the ABT program
for the HC+NOx and CO standards for SD/I engines at or below 373 kW and the HC+NOx
standard for OB/PWC engines and we are retaining them for final rule. (As noted in the
proposal, EPA does not believe banking and trading provisions are appropriate for the CO
standards being applied to OB/PWC engines given the level of the CO standard.) Banking
provisions, including early banking provisions (discussed below in Section 3.6.4), create an
incentive for manufacturers to go beyond the requirements set by EPA and also create an
incentive for early introduction of new technology. EPA believes this behavior should be
encouraged because early introduction can also secure earlier emission benefits. With regard to
trading, EPA believes that trading can help manufacturers that, for whatever reason, are
struggling with meeting the standards. Trading has happened very infrequently under EPA's
ABT programs, most likely due to cost and competitiveness issues.  However, it could prove
very useful to a company that is having short-term difficulty with complying with the standards,
where other means of addressing the problem do not exist.
3.6.1   Credit life

What Commenters Said:

       NMMA and Mercury Marine both supported the proposal to use an unlimited credit life
for credits used  in the ABT Programs for both OB/PWC engines and SD/I engines. In the event
that EPA determines it is necessary to limit the credit life, NMMA and Mercury Marine
                                         3-48

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


commented that EPA should apply the alternative approach suggested in the preamble, which
would be to limit the credit life to the regulatory useful life of the engine. This would mean that
the credits generated by a particular engine would be available while that particular engine is in
the fleet. This would avoid concerns voiced by EPA in the preamble about credits being used
years after the engine that generated the credits is no longer in the fleet. Moreover, NMMA and
Mercury Marine noted that the ability to continue to carry over credits generated in the existing
ABT program for OB/PWC engines into the new ABT program rewards manufacturers that have
produced engines cleaner than the standards.

       CARB commented that it would support the limitation of credits based on the useful life
of the engine as proposed. Further, CARB recommended that previously banked credits not be
applicable for use on models after a change in standards has occurred.

Letters:
Commenter
NMMA
Mercury
CARB
Document #
0688
0693
0682
Our Response:

       We are retaining the unlimited lifetime for ABT credits under the Marine SI ABT
program, as proposed. While EPA is retaining the unlimited lifetime, EPA notes that
manufacturers should not assume that these credits will be available without any restrictions on
their use if, and when, EPA should consider a new round of emission standards in the future. In
revising emission standards, section 213(a)(3) of the CAA requires EPA to set standards which
achieve the greatest degree of emission reduction that is technologically achievable, taking into
consideration such items such as cost, safety and lead time. If manufacturers have a large pool
of ABT credits available to them, EPA must consider ways to ensure that those credits do not
result in an unnecessary delay of the standards.  This can be done in a variety of ways, and has
been done in other ABT programs by allowing only limited numbers of existing credits to be
used for a limited period of time during the transition to the new standards.

       EPA does not believe a limit on the life of the credits is needed for the Marine SI ABT
program adopted with today's program.  Credits are generated at a cost to manufacturers and
thus they have a value to the manufacturers. Provisions which limit a manufacturer's ability to
use credits, such  as a limit on credit life, will reduce the incentive for manufacturers to invest in
the  development and introduction of new technology. As mentioned above, manufacturers
should not assume that an unlimited life means the credits will be available without any
restrictions on their use if, and when, EPA should consider  a new round of emission standards in
the  future. EPA would  expect to consider ways  to ensure that existing credits would not result in
an unnecessary delay of any future standards.

3.6.2   Averaging sets  and other restrictions

What Commenters Said:
                                       3-49

-------
                                                          Chapter 3: Marine SI Engines
       NMMA commented that the ability of engine manufacturers to use credits
interchangeably between OB and PWC engines is important in ensuring compliance with the
standards.

       NMMA and Mercury Marine commented that it is critical that jet boats be allowed to
average credits, both HC+NOx and CO, with OB/PWC engines to provide flexibility and to
ensure that jet boats will be able to meet the SD/I emission standards. NMMA noted its support
of the proposed approach discussed in the preamble and in the proposed regulatory text in
§1045.701(d), provided CO averaging was included.

       Bombardier commented that it supported the proposed corporate averaging provisions in
§1045.701(d) which allows SD/I engines certified under §1045.660 for jet boats to use HC+NOx
exhaust credits generated from OB/PWC engines, as long as the credit-using engine is the same
model as an engine model from an OB/PWC family. However, for the corporate averaging
provision of §1045.701 (d) to be meaningful to a manufacturer, Bombardier commented that CO
averaging is also essential for achieving compliance. In addition, Bombardier premised their
comments on the feasibility of having  their jet boat models comply with the SD/I standards
beginning with MY2011 (see Section  3.2.3, above) on the basis that §1045.701(d) is expanded
to allow CO averaging.

       NMMA and Mercury Marine recommended that EPA remove the restriction regarding
the ability of an engine to earn credits  for one pollutant when using credits to comply with the
emissions standard for another pollutant for both OB/PWC engines and SD/I engines.  They
commented that this restriction does not serve any useful  purpose. From an emission reduction
perspective, EPA will still see the pollution reduction across a manufacturer's fleet even with the
restriction lifted. NMMA and Mercury marine noted that EPA's rationale for this restriction is
that it has been imposed in other programs and is therefore justified for the marine engine
category.  They do not believe this is a sound basis for  such a restriction. NMMA commented
that U.S. Coast Guard (USCG)  data demonstrates that an  averaging approach to controlling
emissions results in emission reductions. Thus, NMMA believes a restriction is unnecessary
from an environmental perspective.

       From a technical perspective, NMMA and Mercury Marine commented that this
proposed restriction unduly penalizes certain engines in manufacturers' fleets. For example, for
OB/PWC engines, some direct injection two-stroke engines have very low CO emissions but
have higher HC+NOx emissions.  Mercury Marine noted  that many DI 2-Stroke Engines are
borderline on meeting the standard for HC+NOx, but have extremely low CO emissions, usually
under 100 g/kW-hr.) Thus, these engines would have to use HC+NOx credits for compliance but
would be ineligible to generate  CO credits.  NMAA commented that the inability to earn CO
credits for these engines will  have a significant impact  on certain manufacturer's product plans
developed to assure compliance with the standards.

       With regard to the proposed restriction for SD/I engines, Mercury Marine commented
that GM will be launching a new base engine in 2010 (6.0 L S/C) that may be negatively
impacted by this approach. The supercharged engine may need to run rich of stoichiometric at
                                          3-50

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


Mode 2 and may be high on CO emissions at that point. Mercury Marine note they are forced to
use GM base engines as they are the only ones that fit within the tight packaging requirements of
the boat builders.

       As noted earlier at the beginning of Section 3.6, CARB commented that EPA should
rescind the provisions for credit banking and trading. Should EPA decide to keep the banking
and trading provisions for marine engines, CARB encouraged EPA to depreciate the value of
banked credits over time.  CARB is concerned that it may be possible for manufacturers to
certify engines to emission levels that are considerably  lower than required, even within
proposed family emission limit (FEL) caps, which could delay the introduction of more stringent
emission standards in the future for some manufacturers (if enough credits have been banked).
They noted that EPA made a similar argument for disallowing the banking of CO credits from
OB/PWC engines, and CARB believes the argument is applicable to the other regulated
pollutants as well as SD/I engines.

       CARB also recommended that cross class trading not be allowed. Finally, CARB
recommended that deficits not be carried over to future years without significant penalties.

Letters:
Commenter
NMMA
Mercury
CARB
Bombardier (hearing)
Bombardier
Document #
0688
0693
0682
0642
0674
Our Response:

       With regard to the averaging sets for the Marine SI ABT program, EPA is adopting two
averaging sets. OB/PWC engines will be in one averaging set. SD/I engines at or below 373 kW
will be in another averaging set.  (As discussed in Section 3.4.1, the final regulations for high-
performance SD/I engines do not include ABT.) There will be no mixing of credits between the
two sets of engines, except under certain conditions for jet boat engines.  Jet boat engines are
subject to the SD/I engine standards. Manufacturers will be able to use credits generated from
OB/PWC engines to demonstrate that their jet boat engines meet the HC+NOx and CO standards
for SD/I engines. Engine manufacturers can only use this provision if the majority of units sold
in the United States from those related engine  families are  sold for use as OB/PWC engines.
Finally, the manufacturer must certify these jet boat engines to an FEL at or below the applicable
emission standards for a similarly-powered OB/PWC engine.  While the preamble to the
proposal noted manufacturers could use this special provision for jet boat engines for
demonstrating compliance with both the HC+NOx standard and the CO standard, the proposed
regulations failed to include a reference for CO. The reference to the  CO standard has been
included in the regulations for the final rule.

       With regard to restriction regarding the ability of an engine to earn credits for one
pollutant when using credits to comply with the emissions  standard for another pollutant, EPA is
                                      3-51

-------
                                                          Chapter 3: Marine SI Engines
dropping that provision for the final rule.  While EPA proposed such a restriction, it was
modeled on similar requirements in other ABT programs where there was concern that a
manufacturer could use technologies to reduce one pollutant while increasing another pollutant.
In such cases, EPA did not want to allow manufacturers to both generate credits for one pollutant
while using credits for another pollutant.  In order to comply with the standards applicable to
OB/PWC engines and SD/I engines at or below 373 kW, the types of technologies manufacturers
are expected to use technologies such as direct-injection 2-stroke engines or 4-stroke engines for
OB/PWC engines and catalysts along with engine improvements for SD/I engines.  All of these
technologies should result in reductions in both HC+NOx emissions and CO emissions
compared to current designs.  While the technologies are expected to reduce both HC+NOx
emissions and CO emissions, there could be situations where these technologies are capable of
meeting one of the emission standards but not the other. EPA does not want to preclude such
engines  from being able to certify using the provisions of the ABT program and is therefore
dropping the proposed restriction from the final rule.

      With regard to comments on discounting of emission credits, we are not adopting such
provisions for the ABT program. Discounting emission credits is similar to limiting the lifetime
of credits. Both provisions lower the value of a credit to a manufacturer. As noted earlier in the
discussion on credit lifetime, EPA believes that emission credits are generated at a cost to
manufacturers and thus they have a value to the manufacturers. Provisions which limit a
manufacturer's ability to use credits, such as a "significant" discount, will reduce the incentive
for manufacturers to invest in the development and introduction of new technology, which is a
key goal of an ABT program.

      In response to the comments on credits deficits, it can be noted that EPA did not propose
to allow credits deficits under the Marine SI ABT program. EPA is not including any deficit
provision in the final regulations for the Marine SI ABT program.
3.6.3   FEL caps

What Commenters Said:

       Mercury Marine commented that the FEL cap for jet boat engines should be the same as
the FEL cap for OB/PWC engines because jet boat engines are derived from these products.

       Bombardier noted that it is supportive of the effort to develop alternative fuels to reduce
petroleum-based fuels consumption. Bombardier commented that EPA has proposed maximum
FEL caps for marine engines which may impede a manufacturer's effort to provide alternative
fueled marine engines. Bombardier requested that engines using fuels other than gasoline,
alcohol and natural gas be exempt from the HC+NOx maximum FEL proposed in  40 CFR
1045.103 (b)(l).  Because these engine families would still be subject to the proposed corporate
averaging requirements, any increase in HC+NOx emissions would be off-set by further
HC+NOx reductions of other engine families.  Bombardier reasoned that this change would be
an emission neutral (or beneficial) change to the regulation, and would help support a
manufacturer's efforts to develop alternatively fueled marine engines.
                                         3-52

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
Letters:
Commenter
Mercury
Bombardier
Document #
0693
0674
Our Response:

       As proposed, we are classifying jet boat engines as part of the SD/I engine category,
subject to the SD/I standards. However, because many jet boats, today, use OB/PWC engines,
we are providing additional regulatory flexibility in which limited jet boat engines may be
certified using OB/PWC emission credits. To be eligible for this flexibility, the jet boat engines
must meet the OB/PWC standards. We believe that this PEL cap is necessary to limit the degree
to which manufacturers may take advantage of emission credits to produce engines that are
emitting at higher levels than competitive SD/I engines.

       The purpose of the PEL cap is to prevent the sale of very high-emitting engines. As
discussed in Chapter 4 of the RIA, engine manufacturers already certify the majority of their
engines using FELs well below the new FEL cap. This cap can be met through the use of simple
four-stroke engines or direct-injection two-stroke engines. Bombardier did not comment on
what alternative fuel they were referring to or why engines operating on this fuel could not meet
the HC+NOx cap. In addition, Bombardier did not present a rationale why high-emitting engines
using this fuel would be more acceptable than other high-emitting engines.  Therefore, we are
retaining the HC+NOx FEL cap for all OB/PWC Marine SI engines.

3.6.4   Early credits for SD/I engines

What Commenters Said:

       NMMA and Mercury Marine supported the Early Credit Program because it encourages
SD/I manufacturers to expedite the introduction of catalyst-equipped engines nationwide earlier
than what would be required in the regulation, which results in an environmental benefit.
Mercury Marine  noted that it plans to offer only catalyst-equipped versions of its Towed Sports
(Water Ski Boats) engines in 2009, as this market is sensitive to CO emissions. NMMA and
Mercury Marine  also recommended that EPA allow manufacturers to earn early credits for
engines that meet either the HC+NOx standard or the CO standard.

       NMMA and Mercury Marine commented that another important change that would need
to be made to any Early Credit Program is to ensure that the timing for the program coincides
with any adjustment to the implementation date for the standards. (In comments summarized in
Section 3.2, NMMA and Mercury Marine commented that the 2009 model year implementation
date for the SD/I exhaust standards was not realistic for the marine industry. NMMA and
Mercury Marine  recommended a 2010 compliance date for most of the SD/I engines, with a
2011 implementation date for the GM replacement engines. NMMA also recommended a 2011
implementation date for PWC engines installed in jet boats.)  In order for an Early Credit
                                      3-53

-------
                                                          Chapter 3: Marine SI Engines
Program to be useful, NMMA and Mercury Marine commented that EPA would need to adjust
the period to reflect any changes made to the implementation date.

       CARB recommended against the adoption of early introduction multipliers for the
generation of credits from SD/I engines.

       Although CARB opposes the banking and trading of emission credits, CARB commented
that the prohibition in 1045.145(b)(6) against the early banking of emission credits for SD/I
engines sold in California before 2009 should be amended or rescinded altogether if EPA decides
to implement the ABT program as proposed.  CARB understands that EPA does not want to
allow credits to be generated from engines that are already required to meet cleaner emission
standards in California. However, the blanket prohibition creates a disincentive for
manufacturers to sell cleaner engines in California beyond what is already required.
Furthermore,  CARB noted that it does not allow credit banking or trading for spark-ignition
recreational marine engines  sold in California. Therefore, any credits earned from the early
introduction of cleaner engines in California would not be subject to double-counting under
EPA's ABT program.

Letters:
Commenter
NMMA
Mercury
CARB
Document #
0688
0693
0682
Our Response:

       With regard to the early credit provisions for SD/I engines, EPA is revising the program
as a result of changes to the implementation dates for SD/I engines at or below 373 kW and
changes to the emission standards for high-performance SD/I engines.  As described in Section
3.2.2, EPA is delaying implementation of the new standards for SD/I engines at or below 373
kW until 2010 for most engines.  This is a one year delay from the proposal and is in response to
comments from manufacturers saying that additional lead time is needed to comply with the new
standards. Because EPA has agreed that additional lead time is needed to comply with the new
standards, we are revising the early credits provisions to allow manufacturers to earn early
credits prior to 2010.  However, given that manufacturers believe additional lead time is needed
to comply, EPA does not believe that manufacturers should be able to earn bonus credits for
certifying earlier than the 2010 timeframe.  Therefore, EPA will allow manufacturers to earn
early credits for SD/I engines below 373 kW that are certified before the applicable date in 2010
or 2011.  However, manufacturers will not be eligible to earn bonus credits on those engines.

       It should be noted that EPA is retaining a delayed implementation data of 2011 for small-
volume engine manufacturers to comply with the new standards for SD/I engines at or below 373
kW, as proposed.  Therefore, EPA is retaining the early credit provisions for small-volume
engine manufacturers that certify earlier than 2011 to the new standards for SD/I engines at or
below 373 kW, including the bonus factors that apply to the credit calculations.  EPA believes it
is appropriate to keep the bonus factors for small-volume engine manufacturers to encourage the
                                          3-54

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


early introduction of new technologies from those manufacturers.  Early credits, alone, may not
be enough incentives for small businesses to certify early to the standards because they may run
the risk of losing market share, during these early years, to lower cost product from competitors
who choose not to certify early. Bonus credits help provide an additional incentive for the early
introduction of low emission engines.

       EPA is retaining the requirement that engines must comply with both the HC+NOx
standard and CO standard to qualify for early credits.  The main purpose of the  early credit
program is to encourage the early introduction of engines complying with the new standards.
EPA does not believe it is appropriate to provide credits for engines that comply with only one of
the new standards, because that engine would not be a fully compliant product.  In most cases,
this should not be an issue because the anticipated emission-control technology for these engines
may be used to meet both the new HC+NOx and CO standards.

       As described earlier in Section 3.4.1, EPA is finalizing a set of emission standards for
high performance SD/I engines that do not include ABT provisions. As a result, the early credits
provisions for high-performance SD/I engines have been deleted from the final  regulations.

       In response to the comment on credits for engines sold in California, EPA is retaining the
prohibition to generate credits from such engines, as proposed.  SD/I engines sold in California
are subject to exhaust emission standards adopted by CARB.  EPA's new exhaust standards will
not apply to SD/I engines sold in California.  Therefore, it is not appropriate to allow
manufacturer to earn credits for engines subject to CARB standards, even if California does not
allow credits from those engines to be banked.

  3.7  Other requirements

3.7.1   Diagnostics

What Commenters Said:

       NMMA and  Mercury Marine commented that the proposed rule includes a requirement in
§ 1045.110 that SD/I engines be equipped with an onboard diagnostic (OBD) system that will
diagnose malfunctions of the emission control system.  As proposed, § 1045.110(b) requires the
OBD  system to have a malfunction-indicator light (MIL) that must be readily visible. 72 Fed.
Reg. at 28,265. The proposed regulatory text also states that the manufacturer "may use sound in
addition to the light  signal." Id. (emphasis added). NMMA and Mercury do not oppose the
requirement for an OBD system on SD/I engines to the extent that the requirement is not overly
complex and is consistent with the California requirements. On the automotive  side, OBD
systems that meet California requirements are deemed to comply with the federal requirements.
The OBD requirements in Part 86 provide "For light-duty vehicles, light-duty trucks, and heavy-
duty vehicles weighting 14,000 pounds GVWR or less, demonstration of compliance with
California OBD II requirements (Title 13 California Code 1968.2 (13 CCR 1968.2)), as modified
pursuant to CARB Mail-Out MSCD #02-11 (internet posting  date October  7, 2002), shall satisfy
the requirements of this section, except that compliance with  13 CCR 1968.2(e)(4.2.2)(C),
pertaining to 0.02 inch evaporative leak detection, and 13 CCR 1968.2(d)(1.4),  pertaining to
tampering protection, are not required to satisfy the requirements of this section."  40 C.F.R. §
                                       3-55

-------
                                                         Chapter 3: Marine SI Engines
86.1806-05(j) (emphasis added).  This "deemed to comply" provision has reduced the
certification burden for the automotive industry and a similar approach is appropriate for the
recreational marine industry.

       NMMA and Mercury continued to comment while proposed § 1045.110(a)(3) seems to
include the "deemed to comply" concept discussed above by allowing for a diagnostic system
approved by CARB for use with SD/I engines to "fully satisfy the requirements of [§
1045.110]," the requirement in that section for the MIL is inconsistent with the CARB
regulations. In the CARB regulations, the OBD system must have "the capability to activate an
audio or visual alert device located on the marine vessel to inform vessel occupants in the event
of a malfunction .  . . ." See CAL. CODE REGS. tit. 13, § 2444.2(b)(4) (2007) (emphasis added).
EPA's requirement of a MIL and possibly sound, if desired, is inconsistent with the CARB
requirements and will impose an additional burden on manufacturers choosing the option of
developing systems to meet both the California and future federal requirements. They
recommend that § 1045.110(b) be revised as follows:
      (b) Use either a malfunction-indicator light (MIL) or sound. If a MIL is used, the MIL
      must be readily visible to the operator; it may be any color except red. When the MIL goes
      on, it must display "Check Engine," "Service Engine Soon," or a similar message that
      they approve. Instead of a MIL you may use sound. You may also use both a MIL and
      sound. In addition to the light signal. The MIL must go on or a sound must be made under
      each of these circumstances:  72 Fed. Reg. at 28,265.
Given that CARB's OBD requirements for SD/I engines commence in model year 2008, it is
critical that EPA harmonize the federal OBD requirements with those that are already in place in
California.  Subsections 1045.110(g)(l) and (2) also require revision. As currently proposed,
these two subsections incorporate by reference two separate ISO standards: "ISO 9141-2 Road
vehicles—Diagnostic systems—Part 2: CARB requirements for interchange of digital
information, February 1994;" and "ISO  14230-4 Road Vehicles—Diagnostic systems—Keyword
Protocol 2000—Part 4: Requirements for emission-related systems, June 2000." 72 Fed. Reg.  at
28,265. These standards are inappropriate for marine engines and should be replaced with a
reference to an industry agreed to protocol developed by the American Boat and Yacht Council
(ABYC).

       Indmar commented that the final item Indmar Products believes needs clarification is the
OBD-M system.  CARB allows for a MIL or a sound device. They believe this option is
necessary to stay common with CARB. This may appear to be a minor detail but would have
significant cost and logistics impact if we have to develop and sell different OBD-M systems for
EPA and CARB.

       Volvo Penta opposes the use of ISO 9141-2 Road Vehicles and ISO 14230-4  Road
Vehicles (1045.1 lOg) for formats codes and connections. Volvo Penta has worked extensively
with CARB, SAE and the other SD/I manufacturers to draft the new marine version of SAE J-
1939 for marine onboard diagnostic purposes. Therefore, Volvo Penta supports and encourages
the EPA to harmonize the OBD requirements with CARB. Two different systems of format
codes  and connections to provide one set of data is prohibitively expensive, overly burdensome
and confusing to Volvo Penta and marine technicians in the field.
                                         3-56

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       Yamaha commented that current PWC engines for Federal or California compliance do
not require the addition of OBD currently or the near future. To add OBD both physically and
electronically for a small percentage of engines will be challenging, time consuming and very
costly due to small production quantities. These units are used in Yamaha produced Jet Boats
exclusively.  As this is a vertically integrated product, Yamaha requests exemption relief from
unnecessary OBD requirements until a stand alone Jet Boat (SD/I) engine is produced and
certified as a 5gr engine. Yamaha anticipates that this can be achieved by M/Y 2011.

       Mercury Marine and NMMA commented that EPA states in § 1045.2, who  is responsible
for compliance, that [t]he requirements and prohibitions of this part apply to manufacturers of
engines and  fuel-system components as described in § 1045.1. The requirements of this part are
generally addressed to manufacturers subject to this part's requirements. The term 'you'
generally means the certifying manufacturer. For provisions related to exhaust emissions this
generally means the engine manufacturer .... For provisions related to certification with respect
to evaporative emissions, this generally means the manufacturer of fuel-system components.
Vessel manufacturers must meet applicable requirements as described in § 1045.20. The
difficulty with this provision is that the recreational marine industry is not vertically integrated.
This means that the SD/I engine manufacturer will supply the engine, the OBD system,
connectors and installation instructions to the boat builder but will have no further role in
assuring compliance with the regulatory requirements. While § 1045.20 addresses the obligations
of the boat builder, engine  manufacturers cannot guarantee that these requirements  will be
followed.  In particular, proposed § 1045.20(d) requires boat builders to "follow all emission-
related installation instructions from the certifying manufacturers as described [in the rule]. If
you do not follow the installation instructions, we may consider your vessel to be not covered by
the certificates of conformity. Introduction of such vessels into U.S. commerce violates 40 CFR
1068.101." 72 Fed. Reg. at 28,262 (proposed § 1045.20(d)). While § 1045.20 makes it explicit
that boat builders must comply with the regulatory requirements, neither § 1045.2 nor § 1045.20
provide a "safe harbor" for an engine manufacturer in the situations where the engine
manufacturer complies with the regulations but the boat builder does not.

       To remedy this situation, Mercury Marine and NMMA recommend that EPA include in
the final rule additional language in § 1045.2 that would hold an engine manufacturer harmless
in the event that a boat builder fails to follow the requirements of the rule. This provision should
state that as long as the engine manufacturer applies the emission control label, the  OBD system,
connectors, and emission-related installation instructions, the manufacturer is deemed to be in
compliance with the requirements of the rule. This additional language will avoid any future
confusion as to the compliance obligations of the engine manufacturer.

       NESCAUM commented that they support requiring engine diagnostics to ensure
maintenance of stoichiometric control of air-to-fuel ratios.
                                       3-57

-------
                                                           Chapter 3: Marine SI Engines
Letters:
Commenter
NMMA
NESCAUM
Indmar
Volvo Penta
Yamaha
Mercury
Document #
0688
0641
0667
0708
0721
0693
Our Response:

       The final diagnostic requirement focuses solely on maintaining stoichiometric control of
air-fuel ratios. This kind of design detects problems such as broken oxygen sensors, leaking
exhaust pipes, fuel deposits, and other things that require maintenance to keep the engine at the
proper air-fuel ratio.  California ARB has adopted diagnostic requirements for SD/I engines that
involve a more extensive system for monitoring catalyst performance and other parameters.  We
will accept a California-approved system as meeting EPA requirements.  The final regulations
direct manufacturers to follow standard practices defined in documents adopted recently by the
Society of Automotive Engineers in SAE J193 9-5. We agree with commenters that the
malfunction indicator may be either a visual or audible cue and have made the corresponding
change in our final rule.

       Jet boat engines that are certified using the emission-credit provisions of §1045.660 will
not need a catalyst to meet emission standards.  Because the proposed diagnostic requirements
are geared toward closed-loop and catalyst-equipped engines, we agree that engines without
these features should not need a diagnostic system. We have revised the regulation to apply the
diagnostic requirement only to engines with catalysts.  Jet-boat engines equipped with catalysts
should be able to meet the proposed diagnostic requirements like any other SD/I engine.

       As noted in the comment, the regulations clearly state that vessel manufacturers are in
violation if they fail to properly install diagnostic systems or otherwise do not follow the
certifying engine manufacturer's emission-related maintenance instructions. We believe the
regulations do not need to go beyond this to create a safe harbor for engine manufacturers.
Where an investigation establishes that the engine manufacturer has properly designed and
produced an engine and communicated installation instructions to a vessel manufacturer, we
would generally  expect to enforce against the engine installer. On the other hand, there may be
cases where the engine manufacturer has not properly  designed or produced its engines or has
not properly communicated installation instructions to vessel manufacturers (either by oversight
or collusion). In these cases, we would not want to create an immunity for the engine
manufacturer where we can in fact establish that the fault for misbuilt or otherwise noncompliant
engines rests with the engine manufacturer.

3.7.2  Torque broadcasting

What Commenters Said:
                                          3-58

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       Mercury Marine, NMMA and BRP commented that also related to the option for in-field
testing is the requirement in proposed §1045.115(b), Torque Broadcasting, for electronically
controlled engines to "broadcast" their speed and output shaft torque. 72 Fed. Reg. at 28,265.
This section requires that engines "broadcast" engine parameters so that they can be read by a
remote device or "broadcast" directly to controller area networks. The rationale provided in the
proposed provision is that the information is necessary for testing in the field. Id. The term
"broadcast" is also used in § 1045.205(s)  in the provision related to the information required in
the certification application. This term is not defined in the proposed regulations and it is unclear
what this term means. They can only assume that "broadcast" is supposed to mean the
transmission of a signal of some kind. EPA does not specify how far the signal must be sent,
what form is acceptable, or what the design specifications are for the "receivers" for such
broadcasts. Since the equipment does not exist, and there is no currently understood
methodology to determine torque, given the nature of propeller cavitation and slip, Mercury
Marine, NMMA, and BRP request that EPA delete this provision.

       Volvo Penta opposes the need to broadcast engine torque. The proposal for
manufacturers to broadcast engine torque is new, and has not been the subject of any discussion
between EPA, NMMA and its member manufacturers. Volvo Penta has no experience with
engine torque broadcast methods. Engine torque broadcast methodology is an emerging field
without commonly accepted standards. Volvo Penta will require considerable time, resources
and testing to create a robust and reliable  method. If engine torque broadcast requirements are
implemented through rulemaking, Volvo  Penta will seek an exemption or postponement of
implementation of the rule until after 2011.

       Honda commented that they do not understand how these sections apply specifically to
outboards and PWCs. They assume that EPA may have intended that some of these sections
apply only to SDI vessels. Outboards and PWC do not necessarily include any sensors or
controls in a basic 4-stroke carbureted engine so including them in this requirement, especially
torque value broadcasting, would be a complete change in their configuration clearly not
anticipated in either the regulatory implementation date nor in the cost analysis associated with
the emission reductions. Without engine management, a simple air / fuel map of the engine in the
operating range would be  sufficient to demonstrate that the engine will provide proper emission
performance and not introduce any form of "defeat device".

Letters:
Commenter
NMMA
Honda
Bombardier
Volvo Penta
Mercury
Document #
0688
0705
0674
0708
0693
Our Response:

       As noted by Volvo, broadcasting for engines is an emerging field. For highway and
nonroad diesel engines, we adopted requirements for engines to broadcast torque and speed
                                       3-59

-------
                                                            Chapter 3: Marine SI Engines
values several years ago.  We also adopted this requirement for Large SI engines in 2002. These
systems are in the early stages of deployment, but there is a growing body of experience in this
technology.  Broadcasting simply involves electronic monitoring of engine parameters such that
the engine's electronic control unit can record values as needed to determine engine speed and
torque at any given point in time.  This is useful for performing field tests with portable
analyzers. Speed measurements are straightforward. Since torque cannot be easily measured
directly, manufacturers would need to do enough testing in the laboratory to establish
relationships between torque and other measurable parameters such as throttle position and
manifold absolute pressure. Once those relationships are established, the electronic control unit
can be programmed with a look-up table to convert measured values to torque readings in real
time.

       While we believe it is not difficult to incorporate broadcasting, we are aware that some
development time is required to establish the look-up tables for converting engine operating
parameters to torque values.  We are also aware that the value of broadcasting for performing
field tests with portable analyzers becomes prominent only after the point at which Not-to-
Exceed standards have started to  apply. We are therefore revising the regulation to require
broadcasting starting with the 2013 model year.

       We believe it is not necessary to establish protocols for codes or other details for
broadcasting. Manufacturers should be able to establish their own protocols for their engines.
This development will be in tandem with the manufacturers' effort to create diagnostic systems.
In both cases there is a need for the electronic control unit to store values that can be retrieved by
plugging in a laptop computer or some other type of reader. We expect the broadcast protocols
to be based on those for the associated diagnostic systems. We are clarifying in the regulation
that broadcasting needs to be  done in a manner that allows for emission testing.  For example, we
believe it is not necessary to specify a frequency for broadcasting engine parameters, since
testing can't be performed if the broadcasting is not frequent enough to perform a valid test
under the procedures specified in part 1065.

       We specifically object to Mercury's reference to propeller cavitation and slip as an
obstacle to proper torque broadcasting.  Engine torque is  determined by the load that is applied to
(and the rotational force that is transmitted through) the crankshaft. Any vessel-based variables
such as vessel speed, vessel direction (upstream or downstream), vessel load, or propeller
cavitation or slip would not affect the internal engine relationships between output torque and the
relevant parameters such  as throttle position and manifold absolute pressure.

       We are limiting the broadcast requirements to electronically controlled engines. We
agree that carbureted engines cannot be modified to comply with broadcasting requirements
without fundamental modifications that go beyond the intended effect of setting new emission
standards. However, we believe delaying the broadcast requirement until 2013 allows sufficient
time for manufacturers to incorporate this upgrade for electronically controlled outboard and
personal watercraft engines.  As for  SD/I engines, broadcasting will allow for greater flexibility
in performing emission tests in the future.
                                          3-60

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


3.7.3   Crankcase emission controls

What Commenters Said:

       NESCAUM commented that they support EPA's proposal to require positive crankcase
ventilation controls on SD/I engines.

Letters:
Commenter
NESCAUM
Document #
0641
Our Response:

       We are adopting the crankcase requirements as proposed.

  3.8   Certification
       The following sections describe various issues related to the certification process that are
specific to Marine SI engines. A few additional certification issues of more general interest are
described in Section 1.3.

3.8.1   Maintenance

What Commenters Said:

       Volvo Penta opposes the proposal that prohibits manufacturers from scheduling critical
emission related maintenance during useful life. Testing to date shows that there may be need to
replace O2 sensors before the useful life period of the engine is reached. The O2 sensor
manufacturer has made recommendations as to the type of O2 sensors to be used, but stated that
marine applications are different and harsher than other applications where these sensors have
been used successfully. On-going sensor durability testing has revealed significant numbers of
O2 sensors out of specification before the engine's useful life, as defined by the proposed rule.
Moreover, our O2 sensor manufacturer has informed Volvo Penta that there is currently nothing
available that will work any better in this application.

Letters:
Commenter
Volvo Penta
Document #
0708
Our Response:

       There is no reason that oxygen sensors should fail before 480 hours of engine operation
during service accumulation in the laboratory. We understand that in-use operating conditions
may be so harsh that oxygen sensors will in some cases not survive through the useful life, but
we believe that  diagnostic systems are best suited to addressing this concern.  A properly
functioning diagnostic system would readily detect a failed oxygen sensor;  the malfunction
indicator would alert the operator. Since a failed oxygen sensor would lead to a loss in available
                                       3-61

-------
                                                          Chapter 3: Marine SI Engines
power or increased fuel consumption or both, we believe owners would generally respond to the
malfunction indicator by replacing the defective component.

3.8.2   Carryover data

What Commenters Said:

       BRP supports NMMA proposal to have carry-over engine families from the existing
marine regulation and early-certified engine families meeting the exhaust emission standards of
the proposed regulation be exempt from the proposed NTE test requirements and maximum test
speed definition change through MY2013. It is necessary for BRP to phase in engine families to
the new testing requirements over the next few model years. It is infeasible to re-test every
engine family within the next couple years to verify compliance with the NTE proposal. In
addition, allowing carry-over data to be exempt from the NTE and maximum test speed
provisions will create an incentive for BRP and other manufacturers to certify their engine
families to the new emissions standards in an earlier model year.

Letters:
Commenter
Bombardier
Document #
0674
Our Response:

       BRP's comments generally affirmed the rule as proposed. We have included these
provisions in the final rule.

3.8.3   Warranty

What Commenters Said:

       NMMA and Mercury Marine commented that EPA notes in the preamble that the Agency
is proposing updated warranty periods for the new standards. 72 Fed. Reg. 28,132. The new
proposed emission-related warranty periods for PWC and OB engines in § 1045.120 are shorter
in terms of hours but longer in terms of calendar years (or months). 72 Fed. Reg. at 28,132. For
OB engines, EPA is proposing 5 years or 175 hours of operation, whichever comes first. 72
Fed. Reg. at 28,132. For PWC engines, EPA proposes 30 months or 175 hours, whichever comes
first. The new warranty provision also requires that an emission related warranty cannot be any
shorter than  any published warranty offered without charge for an engine or component. 72 Fed.
Reg. 28,266 (proposed § 1045.120(b)). NMMA does not oppose the updated warranty periods
for these engines nor does NMMA object to the requirement for the warranty period to track with
any free, published warranty; however, § 1045.120(b) should be revised to clarify that "any
published warranty" only applies to the particular engine and not the entire engine family. In
addition, NMMA commented that EP A also needs to clarify that "any published warranty" does
not include service contracts. Service contracts are those contracts that manufacturers offer for
maintaining and repairing the engine beyond the warranty period. NMMA commented that while
most service contracts require a fee, in some cases manufacturers may,  as a promotion, offer
                                         3-62

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


complimentary service contracts for a limited period of time to encourage the purchase of a new
product or to clear inventory. A service contract, however, is not a warranty and should not be
construed as such.

       To make the language clear, NMMA recommends that EPA revise §  1045.120(b)
as follows:
              (b) Warranty period. Your emission-related warranty must be valid during the
       periods specified in this paragraph (b). You may offer an emission related warranty more
       generous than we require. The emission-related warranty for the engine may not be
       shorter than any published warranty you offer without charge for the engine
                 to the           not the              Similarly, the emission-related
       warranty for any component may not be shorter than any published warranty you offer
       without charge for that component. A               is not a         If an engine has
       no hour meter, we base the warranty periods in this paragraph (b) only on the engine's
       age (in years). The warranty period begins when the engine is placed into service.  These
       changes will help clarify that only the engine and not the engine family is affected by any
       published warranty and that service contracts are not to be confused with warranties.

Letters:
Commenter
NMMA
Mercury
Document #
0688
0693
Our Response:

       We agree that extended warranties offered at no additional charge should be limited to
those components or engines that are the subject of the extended warranty. We have revised the
regulation to emphasize that the extended warranty does not apply more broadly than for the
particular engines that are the subject of the extended warranty. We believe it is not helpful to
introduce a distinction between no-cost service contracts and warranties because that would
likely become a loophole that allows manufacturers to avoid warranty requirements. In
particular, if a manufacturer offers a no-cost service contract, that represents an expectation that
the engine will operate consistently over a certain period.  We do not understand or accept the
logical construct that would say the engine manufacturer should pay for defects that are not
emission-related, but that they are not responsible for defects that are emission-related.  We are
therefore adopting these warranty requirements as proposed.

3.8.4  Family criteria

What Commenters Said:

       NMMA and Mercury commented that for SD/I certification purposes, EPA is proposing
in § 1045.230(b) to require manufacturers to group engines in the same family if they are the
same in all the following respects: combustion cycle and fuel; cooling system (e.g., raw water,
separate circuit cooling); method of air aspiration; number, location, volume and composition of
catalytic converters; the number arrangement, and approximate bore diameter of cylinders;
                                       3-63

-------
                                                            Chapter 3: Marine SI Engines
method of control for engine operation; numerical level of the emission standards that apply to
the engine. 72 Fed. Reg. 28,271. While this list is very similar to what is currently required for
outboard and personal watercraft in § 91.115, the SD/I engine segment has unique characteristics
and requires a more flexible approach that will prevent the creation of a large number of engine
families and reduce the certification and administrative burdens placed on these manufacturers
(e.g., double certification tests, durability tests, recordkeeping, etc.). To that end, in the final rule,
NMMA and Mercury comment that EPA should revise §1045.230(b) to reduce the number of
characteristics that must be identical for purposes of determining engine families.

       In particular, NMMA and Mercury stated that the requirements for identical cooling
systems and bore diameter should be removed because these are overly restrictive in practical
effect and will not have an impact on exhaust emissions from SD/I engines. Exhaust emissions
do not vary for thermostatically controlled engines regardless of whether the engine is cooled
with raw or fresh water.  Also of significance is that CARB does not require manufacturers to use
the cooling system as a criterion for distinguishing among engine families. NMMA and Mercury
commented as for the bore diameter, there are situations where similar engines of varying
displacements should be included in the same engine family. For example, GM's 5.0L and 5.7L
engines vary only in displacement and share common induction systems, number and
arrangement of cylinders, cylinder heads, and external marinization components, including
exhaust equipped catalysts. These engines have been classified historically in one engine family
and have the same emissions characteristics. For these reasons, NMMA and Mercury believe that
EPA must delete these criteria for SD/I engines from § 1045.230(b).

       NMMA and Mercury Marine commented for PWC/OB, EPA proposes requirements for
dividing product lines into engine families in § 1045.230. As discussed in  the comments related
to the requirements for SD/I engines, the list of characteristics contained in proposed §
1045.230(b) is similar to what is in § 91.115; however, there are several requirements, e.g., the
bore diameter and cooling systems, that will require SD/I manufacturers to establish too many
engine families as noted above and are not a meaningful criteria for PWC  or OB engines either.
There are also several differences between § 1045.230(b) and § 91.115 with regard to the
inclusion of the numerical level of the emissions standards and method of control for engine
operation in the characteristics that must be identical. 72 Fed. Reg. 28,271 (proposed §
1045.230(b)(6) and (7)). In light of these differences and the fact that OB and PWC engine
manufacturers have been following § 91.115 for over a decade, NMMA and Mercury Marine
recommend that EPA substitute portions of § 91.115(c) and (d) for the corresponding language
in § 1045.230 with the changes recommended for  SD/I. The following redline is provided by
NMMA and Mercury to show how this provision should be revised.
     § 1045.230 How do I select engine families?
     a.  For purposes of certification, divide your product  line into families of engines  that are expected to have
        similar emission characteristics throughout the useful life as described in this section. Your engine family
        is limited to a single model year.
     b.  To be       in the                           be       in all of the
        respects:
       (1) The          cycle;
       (2) The cylinder           (inline, vcc.          so forth);
       (3) The
       (4) The                 converters, location, volume,    composition:
       (5) The
                                           3-64

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments
                   in the                 if they arc (he     in all the
       (1)  The          cycle    fuel,
       (2)  The                                vs.
       (3)        of air                               vs.
       (4)  The
       (5)  The                                           of
       (6)                 for                               (i.e..          or
       (?)  The         level                         apply to the
     c.  At the                                 in all the        listed in         (b)
            be                                     if the                                 be
                to                                This            is          the            of
                   as:
       (1)  The bore
       (2)  The
       (3)  The                                            valve.     valve.      valve.    so
           forth);
       (4)  The               valve or         as
       (5)  The   system;
       (6)  The       system;
       (?)  The       of air

     You             a                  is                      (b)
            if you     the                            arc              the      life.

     d.   You may group engines that are not identical with respect to the things listed in paragraph (b) of this
        section in the same engine family, as follows:
       (1)  In unusual circumstances, you may group such engines in the same engine family if you show that
           their emission  characteristics during the useful life will be similar.
       (2)  If you are a small-volume engine manufacturer, you may group all your high-performance engines into
           a single engine family.
       (3)  The provisions of this paragraph (cd) do not exempt any engines from meeting all the emission
           standards and requirements in subpart B of this part.

       NMMA commented that these recommended revisions harmonize the existing
requirements in § 91.115 with the newly proposed §  1045.230. This redline also reflects the
recommendations discussed above related to SD/I engine families.

       Honda commented regarding the Engine Family Determination for Outboard Engines and
PWCs.  Honda suggests that the criteria for engine family selection outlined in Section
1045.230(7) of the proposal be deleted from the final rule.  Section 1045.230 of the proposal
makes "the numerical level of the emission  standard" a family determination criteria. The
numerical standard level would mean that each  engine horsepower would be a separate family.
This is unlike 40 CFR Part 91 where two engine models (75 & 90 hp for example) are created
from one engine (same  displacement / block and head) and are in the same family. This change
would be completely  contrary to the intended purpose of the family concept (similar engine with
similar emission characteristics). Perhaps this was incorrectly carried over from another
regulation where different classes with different displacement categories meet numerically
different standards.

       Indmar commented in §1045.230 (b) 2 the cooling system (raw-water vs. separate-circuit
cooling) could be a family discriminator. This would double the number of engine families for
                                         3-65

-------
                                                          Chapter 3: Marine SI Engines
them with no value added. Indmar offers most of their engines with raw-water or fresh-water
cooling. The control temperature for both these applications is 165 degrees Fahrenheit. The
exhaust manifolds are heated for both fresh and raw water systems so the exhaust gas feed
stream to the catalyst is not impacted differently with either system.  The emissions of the engine
will not change with either cooling system. Also of significance is that CARB does not require
manufacturers to use the cooling system as a criterion for distinguishing among engine families.

       Suzuki commented that EPA has proposed to revise the requirements for how to group
products into common engine families to include a new provision of "approximate  bore
diameter"  as a requirement for engine family grouping. Suzuki is concerned that this new
provision will inappropriately require the  creation  of additional engines families that otherwise
could be grouped together if the existing engine family grouping criteria specified in §91 .115
were employed. They are also concerned that the judgment criteria could be confusing to
implement from a certification-planning viewpoint.  Suzuki requests that EPA reconsider the
need to include this revision in the regulation. Should EPA decide to proceed with the proposed
revision, Suzuki requests that the regulatory language be revised to allow the Agency to have
discretion to approve the grouping of engines of dissimilar bore diameters if a manufacturer can
show that the proposed grouping is in agreement with good engineering practices.

       Volvo Penta opposes the family aspects (families) as outlined in the NPRM. Volvo Penta
commented that the proposed NPRM aspects will create too many engine families requiring a
multiplicity of certification testing and documentation with no resulting emissions reduction. The
proposal, therefore, is unnecessarily burdensome. Volvo Penta is a custom marinizer of General
Motors (GM) produced engine blocks. Traditionally, Volvo Penta arranged engine  families for
emissions classification by GM's engine block types and fuel  intake systems. As the engines are
catalyzed, the fuel intake systems become the same, thereby eliminating fuel intake type as a
family discriminator. Volvo Penta's current engine families for emissions purposes are:
       •3.014 Carbureted
       • 4.3 V6 Carbureted
       • 4.3 V6 EFI
       • 5.0 V8 Carbureted
       • 5.0 V8 EFI
       • 5.7 V8 EFI (All models EFI)
       • 8.1 V8 EFI (All models EFI)
Beginning with California in 2008 Volvo  Penta will identify the following engine families:
       • 3.014 (EFI + Cat)
       • 4.3 V6 (Carbureted)
       • 4.3 V6 (EFI)
       • 5.0 & 5.7 V8 (all models EFI + Cat)
       • 8.1 (all models EFI + Cat)
They anticipate that by 2011 the Volvo Penta engine families  will include:
       • 3.014 (EFI + Cat)
       • 4.1 V6 (all models EFI + Cat)
       • 5.0 & 5.7 V8 (all models EFI + Cat)
       • 6.0 SC V8 (all models EFI + Cat)
                                          3-66

-------
Nonroad Spark-Ignition Engines—Summary and Analysis of Comments


       Volvo Penta will continue to offer multiple horsepower and cooling system options
within each family as they do today. The least compliant (i.e., "dirtiest") engine within each
family is used for California and EU compliance certification purposes. That process ensures that
all engines within a particular family (however defined) meet the emissions criteria required.
Multiple families add expense without benefit.

       Pleasurecraft Marine in a hearing commented that §1045.230 outlines the criteria for
defining engine families. There are two areas that Pleasurecraft Marine commented need
reconsideration. Those areas are:
        •  Line Item 2, the cooling system (§1045.230(b)(2)
        •  Line Item 5, the number, location, volume, and approximate bore diameter of the
           cylinders (§1045.230 (b)(5)).

       Pleasurecraft Marine commented regarding Line Item 2, segregating engine families by
their cooling system accomplishes nothing more than doubling the number of engine families.
Emissions will not vary, for thermostatically controlled engines regardless of whether the engine
is raw or fresh water-cooled, therefore, the cooling system should not be a factor in determining
engine families.

       Pleasurecraft Marine commented regarding Line Item 5, there are circumstances where
similar engines of different displacements should be included in a common engine family. An
example would be the General Motors 5.0 and 5.7 liter engines. These engines vary only in
displacement and share common induction systems,  number and arrangement of cylinders,
cylinder heads, and external marinization components including exhaust equipped with catalyst.
Historically General Motors, who designed these engines, has classified them as one family. If
the larger displacement 5.7L will meet emissions standards it is safe to say that the 5.OL will do
so as well. Classifying these engines as one family, as they should be, will save small businesses,
such as theirs, tens of thousands of dollars in unnecessary certification cost.

Letters:
Commenter
NMMA
NMMA
Honda
Indmar
Volvo Penta
Mercury
Pleasurecraft Marine (hearing)
Document #
0688
0688
0705
0667
0708
0693
0642
Our Response:

       We agree that engine families should not be differentiated based on the cooling system.
The current regulations in part 91 include this specification, but it seems that the relative
uniformity of designs for outboard and personal watercraft engines has prevented this from being
an issue. We are revising the regulations to exclude the cooling system from §1045.230 for all
Marine SI engines.
                                       3-67

-------
                                                           Chapter 3: Marine SI Engines
       The intended effect of including the applicable emission standard to differentiate engine
families was two-fold. First, this would prevent SD/I engines from being included in the same
engine family with OB/PWC engines.  Second, this would prevent engines certified to different
Family Emission Limits from being in the same engine family.  Selecting different Family
Emission Limits for engines that are subject to identical standards inherently implies that the
engines will not have similar emission characteristics throughout the useful life, which is the
fundamental purpose of establishing engine families, as expressed in §1045.230(a). Contrary to
the concern raised by Honda, the regulatory language does not prevent a manufacturer from
including different power ratings in the same engine family. As specified in  §1045.103, the
applicable emission standard for an OB/PWC engine family is based on the maximum engine
power for the engine family as described in §1045.140. Section §1045.140 acknowledges that an
engine family may have multiple power ratings within the family by specifying that the
maximum engine power fo