JPI^


 **9fj^*f~
    *
National Emission Standards for
Hazardous Air Pollutants (NESHAP)
Surface Coating of Metal Furniture
Summary of Public Comments and Responses
on Proposed Rule

-------
                                             EPA-453/R-03-002
                                                January 2003
        National Emission Standards for
     Hazardous Air Pollutants (NESHAP):
      Surface Coating of Metal Furniture
Summary of Public Comments and Responses on Proposed Rule
            U.S. Environmental Protection Agency
          Office of Air Quality Planning and Standards
                Emission Standards Division
            Research Triangle Park, North Carolina

-------
                                      DISCLAIMER

       This report has been reviewed by the U.S. Environmental Protection Agency, Office of Air
Quality Planning and Standards, Emission Standards Division and approved for publication. Mention
of trade names or commercial products does not constitute endorsement or recommendation for use.

-------
                                 Table of Contents
1.0 SUMMARY	1-1
       1.1 BACKGROUND	1-1
       1.2 SIGNIFICANT CHANGES SINCE PROPOSAL	1-1

2.0  SUMMARY OF PUBLIC COMMENTS AND RESPONSES 	2J.
       2.1    APPLICABILITY	2-1
             2.1.1  Overlap with Other NESHAP 	2-1
             2.2.2  Affected Source	2-10
             2.2.3  Low Usage Exemption	2-12
             2.2.4  Other Exemptions  	2-13
             2.2.5  Definitions  	2-18
       2.3    MACT FLOORS  	2-19
             2.3.1  Basis of the MACT Floors	2-19
             2.3.2  Alternatives More Stringent Than the MACT Floors	2-26
       2.4    EMISSION LIMITS	2-31
             2.4.1  Compliance Options	2-31
             2.4.2  Work Practice Standards	2-35
       2.5    COMPLIANCE DETERMINATIONS	2-36
             2.5.1  Support for the Proposal 	2-36
             2.5.2  Compliance During Periods of Startup, Shutdown, and Malfunction
                    	2-37
             2.5.3  Compliance Period	2-38
             2.5.4  Monitoring	2-39
             2.5.5  Miscellaneous Comments  	2-39
       2.6    NOTIFICATION, RECORDKEEPING, AND REPORTING	2-43
             2.6.1  Initial Notification	2-43
             2.6.2  Consolidation and Timing of Reports  	2-45
             2.6.3  Semiannual Reports	2-47
             2.6.4  Reporting of Deviations 	2-49
       2.7    MISCELLANEOUS COMMENTS 	2-50
                                        111

-------
                                     1.0 SUMMARY







1.1 BACKGROUND



       On April 24, 2002 (67 FR 20206), the EPA proposed the "National Emission Standards for



Hazardous Air Pollutants:  Surface Coating of Metal Furniture" (40 CFR part 63, subpart RRRR)



under section 112(d) of the Clean Air Act (CAA). Public comments were requested on the proposed



rule, and comment letters were received from industry representatives, industry trade groups, and



Federal and State agencies. A total of nine comment letters was received.  Table 1-1  presents a listing



of all persons who submitted written comments, their affiliation, and air docket number and item number



for their comment letter. A public hearing was not requested.



       The written comments that were submitted on the proposed rule have been summarized, and



responses to the comments are included in the following chapter. This summary of comments and



responses serves as the basis for revisions made to the rule between proposal and promulgation.







1.2 SIGNIFICANT CHANGES  SINCE PROPOSAL







       In response to comments received on the proposed standards, we made a number of changes



to the final rule.  Many of these changes are clarifications designed to make our intentions clearer.



However, some of the changes affect the requirements specified in the proposed rule.  This section



summarizes these more significant changes to the proposed rule.
                                            1-1

-------
   TABLE 1-1. LIST OF COMMENTERS ON THE PROPOSED NATIONAL EMISSION
            STANDARDS FOR SURFACE COATING OF METAL FURNITURE


 Docket A-97-40                      Commenter and Affiliation                       Date of
  Item Number                                                                    Document

IV-D-1            American Furniture Manufacturer's Association (AFMA),             6-17-02
                  Comments of American Furniture Manufacturer's Association on
                  Proposed MACT for Surface Coating of Metal Furniture

IV-D-2            Business and Institutional Furniture Manufacturer's Association        6-20-02
                  (BIFMA), Comments on the proposed National Emission Standard
                  for Hazardous Air Pollutants for Surface Coating of Metal
                  Furniture (67 Fed. Reg. 20206, April 24, 2002)

IV-D-3            Department of the Navy, Department of Defense Comments on        6-21-02
                  the NESHAP for Metal Furniture Surface Coating Operations,
                  Proposed Rule, 67 Fed. Reg. 20206 (Apr. 24, 2002)

IV-D-4            The Boeing Company, Re: National Emission Standard for             6-24-02
                  Hazardous Air Pollutants: Surface Coating of Metal Furniture -
                  Proposed Rule

IV-D-5            The Earthjustice Legal Defense Fund, Comment on Docket No. A-      6-24-02
                  97-40: Surface Coating of Metal Furniture

IV-D-6            Georgia Department of Natural Resources, Environmental             6-24-02
                  Protection Division, Air Protection Branch, Comments on Proposed
                  Rule - 40 CFR 63, Subpart RRRR

IV-D-7            The Institute of Clean Air Companies (ICAC), ICAC Comment on      6-24-02
                  Docket No. A-97-40

IV-D-8            Lozier Corp., RE: Comments to the Proposed NESHAP: Surface        6-24-02
                  Coating of Metal Furniture

IV-D-9            The National Paint and Coatings Association, RE: National             6-24-02
                  Emission Standards for Hazardous Air Pollutants:  Surface Coating
                  of Metal Furniture; Proposed Rule
                                            1-2

-------
New and Existing Source Emission Limits

       We received comments on the procedure we used to calculate both the new source and
existing source maximum achievable control technology (MACT) floors.  In response to these
comments, we reevaluated the procedure we previously used and determined that a different procedure
was more appropriate, resulting in new MACT floor values. Consequently, we revised the new and
existing source emission limits in 63.4890 of the proposed rule.  The new source emission limit was
changed from 0.094 kilogram organic HAP per liter of coating solids used to no organic HAP
emissions. The emission limit was based on several coating technologies that contain no organic HAP
that we now believe represent in the aggregate the MACT floor for new sources.  For existing sources,
the emission limit was changed from 0.12 to 0.10 kilogram organic HAP per liter coating solids used.
This value reflects the revised MACT floor calculated using the average of the best performing 12
percent of 22 sources rather than 49 sources.
       We recognized that a small number of new sources many have particular metal furniture
components that cannot be coated with any of the organic HAP-free coating technologies due to
specialized appearance or functional characteristics. For these cases, we added a procedure at
63.4890(a)(l) to allow the source to request an  alternative emission limit.  If we grant such a request,
the alternative limit would be 0.094 kilogram organic HAP per liter coating solids used, which is the
new source emission limit we originally proposed.

Military Coatings

       Metal furniture (among other coated products) used by the military in battlefield situations or
that are integral to military war-fighting equipment will be covered under a new source category.
Although the comments we received requested that such metal furniture be covered under the upcoming
miscellaneous metal parts and products surface coating NESHAP, we believe the more appropriate
approach is to group all of the products coated with the specialized military coatings into their own
source category.  In this way, we can more effectively address the concerns unique to these coatings.
                                             1-3

-------
Compliance During Periods of Startup, Shutdown, and Malfunction

       We received a comment concerning whether the provision in 63.4900(a)(2) of the proposed
rule complied with the CAA. This paragraph stated that affected sources do not have to comply with
the emission limitations during periods of startup, shutdown, and malfunction. This provision is often
found in NESHAP in which compliance with the standards is based solely on the results of a short-term
initial performance test and short-term averaging of continuous monitoring results thereafter. In
consideration of this comment, we realized that this provision is not appropriate for the surface coating
NESHAP when these short-term tests and monitoring results are only one component of a compliance
determination that determines emissions over a long period  of time, which in this case is a month. In the
final rule, we revised and simplified the general compliance requirements in 63.4900. As part of these
revisions, we removed the statement that sources must be in compliance except during periods of
startup, shutdown, and malfunction. In  its place, we stated in 63.4900(a) of the final rule that all
affected sources must be in compliance with the emission limitations in 63.4890 at all times.  We left in
place the requirement for sources using an emission capture system and add-on control device to
develop and implement a written startup, shutdown, and malfunction plan according to 63.6(e)(3).
Additionally, in order to be consistent with this change as finalized in 63.4900(a), paragraph (h) of
proposed 63.4963 (renumbered as 63.4962 in the final rule) was not included in the final rule.

Calculation for Volume Fraction of Coating Solids

       There may be limited circumstances where the volume fraction of coating solids cannot be
determined using the specified test methods or manufacturer's data.  To account for these situations, we
added a procedure at 63.4941(b)(3) to calculate this value.
Monitoring
                                             1-4

-------
       The proposed rule contained detail monitoring provision in 63.4968 (which has been



renumbered as 63.4967 in the final rule).  These provisions used both "sensitivity" and "tolerance" to



describe a monitoring device's minimum ability to discriminate between input signals.  Comments were



received stating that this was an inappropriate use of the term "tolerance." To correct this terminology



usage, we replaced both terms with the single term "accuracy" and revised the monitoring provisions of



this section accordingly.
Title V Operating Permits







       Several commenters had concerns about possible conflicts between reporting requirements



under this rule and their approved title V programs. It is important to emphasize that a permitting



authority does not have the authority to change the reporting requirements of this rule (such as type of



report, content of report, and/or frequency of submission). Reporting requirements under this rule are



applicable requirements and sources must comply with them.



       The final rule, consistent with the proposed rule, does however allow an affected source to



submit its semiannual compliance report along with, or as part of, its 6-month monitoring report



required by 40 CFR part 70 or part 71.  See  63.4920(a)(l)(iv) and (a)(2) of the final rule.  Asa



result of comments, these two sections have been modified in the final rule to clarify when monitoring



reports are required by part 70 or part 71 (every 6 months) and when a 6-month monitoring report



must cross-reference a semiannual compliance report.  Language was also added to



63.4920(a)(l)(iv) of the final rule to ensure that a semiannual compliance report is submitted within a



reasonable time (30 days) after the end of the semiannual reporting period.



       At the request of commenters, 63.4910(c)(2) and 63.4920(a)(3)(ii) of the final rule have



been revised to ensure that certifications of truth, accuracy, and completeness for the notifications of



compliance status and semiannual compliance reports under this rule are consistent with the certification



requirements under 40 CFR part 70 or part 71.  These modifications will ensure that these reports will



meet the  certification requirements for reports submitted under part 70 or part 71.




                                              1-5

-------
Compliance Under Multiple Sets of Representative Operating Conditions







       Section 63.4962 of the proposed rule contained procedures for determining compliance when a



source operates under different sets of representative operating conditions. Upon further review of this



section, we believe this option is overly complicated and would be difficult to implement in actual



practice.  Rather than including these detailed compliance procedures in the final rule, we decided to



replace them with a general statement allowing such a compliance demonstration if you believe a



workable and enforceable procedure can be maintained to demonstrate compliance under different sets



of representative operating conditions (see 63.4891(d)(2) of the final rule).
                                             1-6

-------
              2.0 SUMMARY OF PUBLIC COMMENTS AND RESPONSES







       In the comment summaries and responses contained in the following sections, any resulting



additional rule language is represented by underlining. Any rule text that has been removed is



represented in strikeout font (i.e., strikeout).







2.1    APPLICABILITY



       2.1.1   Overlap with Other NESHAP



       Comment: One commenter (IV-D-1) expressed appreciation for EPA's inclusion of



63.4881(c)(2), which states that the rule does not apply to surface coating of metal components of



wood furniture conducted in an operation that is subject to the wood furniture NESHAP (40 CFR 63,



subpart JJ). The commenter added that this exemption is crucial to companies which manufacture



wood furniture with metal components, because many of them use the same finishing lines to surface



coat metal parts associated with wood furniture. Without the exemption, these facilities would have to



comply with both the wood furniture rule and the metal furniture rule on the same finishing lines.  The



commenter strongly advocated that the exemption be retained and included in the final MACT



standard.



       Response: We have retained this exemption in the final rule.







       Comment: Two commenters (TV-D-2 and 9) expressed concern with the applicability of the



rule as described in the preamble to the proposed rule. The definition of affected source in the



preamble is very broad in that it could include some wood furniture operations in which the operation



coats both wood and metal parts.  The preamble goes on to state that "where a  manufacturer of wood



furniture also coats metal components of that wood furniture," then "the coating  of those metal



components would be subject to subpart JJ" (that is, the wood furniture NESHAP). The preamble also



states that the future surface coating NESHAPs for miscellaneous metal parts and plastic parts will



apply if such coating operations are performed in facilities that do not apply coatings to metal furniture.
                                            2-1

-------
       The commenter noted that much more complicated scenarios are possible. "Stationary
sources" within the industry operate major complexes where both metal furniture and wood furniture
operations are collocated. In fact, there are even some operations which may sometimes run wood
furniture and other times run metal furniture. Additional clarification is needed as to which standard
would apply to such coating operations. Alternatively, the commenter suggested that EPA should
consider allowing sources to declare which standard applies and be allowed to "opt-in" or "opt-out" of
any given NESHAP depending on the characteristics of the products being manufactured. One
commenter (IV-D-9) added that simply exempting metal components of wood furniture, as EPA does
under 63.4881(c)(2) of the proposed rule, does not take into account the myriad of other possible
scenarios with the overlap of the surface coating MACTs.  Sources should be given the option of
applying one set of standards to the entire facility depending on the characteristics of the products being
manufactured.
       One commenter (IV-D-8) was concerned that the definition of metal furniture in the proposed
rule, specifically the phrase "constructed . .. partially from metal" is not defined and is not clear. No
criteria for determining, specifying, or even calculating whether furniture, which contains both wood and
metal components, is classified as wood furniture or metal furniture is provided.
       Further clarification is also needed to ensure that State and EPA Regional offices uniformly
interpret the definition of metal furniture and the applicability of the metal furniture NESHAP. With
manufacturing operations in several states, the commenter was concerned that the same manufacturing
operation located in different manufacturing facilities would be subject to different regulations and
requirements.
       Response: During the development of the proposed rule, we recognized the potential for
overlap between the metal furniture surface coating rule and other surface coating rules. We addressed
this issue in the provisions for defining the source category in 63.4881 of the proposed rule. While we
believe the language of this section adequately addressed most of the potential overlap problems, we
realized that there may be other facility-specific situations that the proposed rule did not specifically
address. Accordingly, we requested comments on this issue in the preamble to the proposed rule (see
67 FR 20208).
                                              2-2

-------
       We recognized that there is not always a clear dividing line between the affected sources of the



surface coating rules. This is evident in the furniture manufacturing industry, where both metal and



wood furniture may be produced in the same facility, and many pieces of furniture contain substantial



portions of metal and wood.  For those commenters concerned with lack of clarity between the



applicability of the metal furniture rule and other surface coating rules, in particular the wood furniture



surface coating rule (40 CFR part 63, subpart JJ), we are providing clarification through the following



examples.



       Example 1. Coating operations at facilities currently subject to the wood furniture rule (40



CFR part 63, subpart JJ) would continue to be subject to that rule. This would be the case even if the



items coated contained metal components, as long as the items meet the definition of "wood furniture"



or "wood furniture component" in 63.801 (a).



       Example 2. Coating operations at facilities that coat metal furniture (as defined in



63.4881(a)(2) of the final rule) constructed either entirely or partially from metal (but not qualifying as



wood furniture components under subpart JJ) would be subject to the metal furniture rule.



       Example 3. Facilities that coat only metal furniture components such as knobs, hinges, and



screws (that is, components that are of a more generic nature and could have broader uses in products



other than metal furniture), and  provide these components exclusively to metal furniture manufacturing



facilities, would be subject to the metal furniture rule.



       Example 4. The applicability of the surface coating rules when the item coated is composed of



both metal and wood components in approximately equivalent percentages will depend primarily on the



functionality of the entire unit. A common  example of such an item is a commercial shelving unit



constructed of a metal base and wood backing. For reasons related to structural stability and



appearance, the functionality  of this particular shelving unit depends more on the metal components than



the wood components.  The surface coating of this shelving unit would be regulated under the metal



furniture rule. Thus, the surface coating of all components of this shelving unit would be regulated under



the metal furniture rule, regardless of whether they are made of metal or wood,  as long as the facility is



a metal furniture manufacturing  facility. This would be true even if the metal furniture manufacturing



facility dedicated a coating line  exclusively to the coating of the wood components.  However, if the




                                                20
                                               -3

-------
wood components were coated at a facility that could not be classified as either a wood furniture or a
metal furniture manufacturing facility, then the coating of the wood components would not be covered
under either the wood furniture or metal furniture rule.
       Example 5.  Coating operations such as those presented in Example 5 may not involve items
that can be readily classified according to functionality. For these situations, the applicability
determination would be made on a case-by-case basis taking into account factors other than
functionality. These factors may include as many relevant factors other than functionality as necessary,
such as the primary North American Industrial Classification System (NAICS) code (or Standard
Industrial Classification (SIC) code) for the facility, amount of surface area coated for each type of
substrate, and how the coating operations have been classified for other surface coating rules (such as
NSPS and State rules).
       The examples we have provided here are necessarily simplistic in nature compared to many of
the situations encountered in the metal furniture manufacturing industry and are intended only to provide
guidance. Even so, the examples demonstrate the complex applicability issues related to this rule and
why precisely defining applicability among the surface coating rules has proved to be a challenge.
While we realize that many of the situations encountered in the metal furniture manufacturing industry
can be far more complicated than presented here, discussion of these more complex situations is
beyond the scope of this document.  For these reasons, we intend to provide additional  guidance
documents in the future that will specifically address some of the more complex applicability issues. In
order to address the specific concerns raised by the metal furniture industry, we are planning to involve
all interested stakeholders in the development of these guidance documents.
       We understand that it could be beneficial to consolidate regulatory requirements at facilities
where coating operations belonging to different source categories (such as metal furniture,
miscellaneous metal parts, and plastic parts) are collocated. Consolidation may reduce the amount of
records, reports, or compliance calculations that the facility would have to maintain.  Some commenters
suggested that the final rule include a compliance option that would allow this consolidation of different
regulatory requirements within a facility. Section 112(d)(2) of the CAA states that all major sources
within a regulated source category must meet the maximum degree of emission reduction that we
                                              2-4

-------
determine to be achievable. We do not believe that the commenters' recommendation of allowing a



facility to choose which coating operation to opt into, or out of, a particular NESHAP would ensure



that the MACT level of control was met for all HAP emission points within each source category.



Therefore, to comply with these CAA requirements, we have not included the compliance option



suggested by the commenters in the final rule.



       As an alternative to the commenters' recommendation, we explored the possibility of including



a compliance option that would allow a facility with collocated surface coating operations (such as



metal furniture, miscellaneous metal parts, and plastic parts) to comply with the requirements of the



most stringent applicable NESHAP. Upon consideration, we found that implementing this option posed



several complex issues that we could  not resolve, and therefore we are not including this compliance



option in the final rule.



       To illustrate the most stringent compliance option, consider a metal furniture surface coating



facility that also has collocated operations that are affected sources under the wood furniture surface



coating NESHAP and the plastic parts and products surface coating NESHAP. The test to determine



the most stringent rule would require you to determine the facility-wide emission rate by individually



applying each applicable surface coating rule to all surface coating operations at the facility and



choosing the rule that results in the lowest emission rate. In this example, you would calculate the



facility-wide emission rate by applying the metal furniture emission limits to all surface coating



operations at the facility. You would  then repeat the calculation for the wood furniture rule emission



limits, then repeat it again for the plastic parts and products emission limits. The rule that results in the



lowest facility-wide emission rate is the one you must comply with under the most stringent rule option.



       However, when considering how to implement this procedure, we encountered difficulties with



a number of practical considerations.   One area of concern is the differing scope of the affected source



between the individual surface coating rules.  Metal furniture surface coating facilities clean the metal



substrate prior to coating application.   Therefore, we included these cleaning operations in the metal



furniture rule affected source. Wood furniture surface coating operations, however, do not have an



analogous cleaning operation. The emission limits in the wood furniture rule (see 63.802) cover



cleaning of application equipment and spray booths, but not the substrate being coated. Thus, when




                                              2-5

-------
performing the test for determining the most stringent rule and applying the wood furniture rule to metal



furniture surface coating operations, the issue arises as to how HAP emissions from metal substrate



cleaning operations should be counted. One way would be to disregard them completely since the



wood furniture rule does not limit their HAP emissions.  Another way would be to count all HAP



emissions in the facility total since the operations are uncontrolled.



        Another difficulty that arose in our analysis of the most stringent rule compliance option is how



we would treat exemptions that are specific to one of the surface coating rules applicable to a facility.



For example, the wood furniture rule allows an exemption from the emission limits for an "incidental



wood furniture manufacturer" that uses less than 100 gallons per month of finishing materials or



adhesives (see 63.800(a)).  The possibility exists that a metal furniture surface coating facility could



have collocated wood furniture surface coating operations that meet the definition of incidental wood



furniture manufacturer. Such a facility could argue that they should be allowed to determine which of



the two rules (metal furniture or wood furniture surface coating) is more stringent.  A similar issue arises



when a facility has only a small amount of their production (for example, less than 5 percent of their



coating usage) covered under one of the multiple surface coating rules applicable to the facility. We



have not yet determined in such cases whether it is appropriate to allow the lesser coating operations to



dictate which rule should apply to the facility.



        We also have had difficulty determining the appropriate basis on which to evaluate which rule is



more strict.  By "basis," we mean the level of annual production or coating usage for products in each



source category that you would use to determine the total annual emission rate under each applicable



rule. Options we have considered include an average of the five years immediately prior to making the



determination or a prediction over some set period of time in the future of what level of production the



facility will achieve.  We also considered whether the determination should be revised periodically,



perhaps when the facility's title V permit is renewed.



        Given the number and complexity of the issues involved with this compliance option, we



concluded that we could not adequately resolve all  of them and still promulgate  the final rule in a timely



manner.  Therefore, we have decided not to include the most stringent rule compliance option in the



final rule at this time. As we continue to develop other surface coating rules, we will attempt to resolve




                                               2-6

-------
the issues presented here, as well as others we have identified.  In the future, if we find this compliance
option to be in conformance with the requirements of the CAA and that it is workable and enforceable
in practice, we will consider amending the final rule. Until such time, you are required to comply with
each surface coating rule that is applicable to your facility.

       Comment: Three commenters (IV-D-2, 6, and 9) were concerned that under the proposed
Miscellaneous Organic Chemical Manufacturing (MOCM) NESHAP and the Miscellaneous Coating
Manufacturing (MCM) NESHAP (see 67 FR 16154), it appears that certain ancillary operations at
metal furniture surface coating facilities could be regulated by the MOCM/MCM rather than by the
metal furniture NESHAP.  In particular, as proposed, the MOCM/MCM could be interpreted as
covering distribution of coatings inside of metal furniture surface coating facilities including the piping
and pumps associated with the distribution of the coatings from the paint mix rooms to the paint booths,
as well as mixing of coating raw materials to make a coating for use in-house by the metal furniture
surface coating facility.  The commenters recommended clarification that material distribution systems
are not within the  affected source of the metal furniture NESHAP.
       Response: The proposed MOCM/MCM rule to which the commenters refer would regulate
coating manufacturing operations and would require controls on the following emission sources in these
operations: storage tanks, process (mixing) vessels, equipment components, wastewater treatment and
conveyance systems, transfer operations, and ancillary sources such as heat exchange systems. Thus, if
an operation is determined to be an affected source under the MOCM NESHAP or the MCM
NESHAP, it would have to comply with the applicable requirements under those rules.  However, the
requirements of the future MOCM NESHAP and the MCM NESHAP will not apply to operations
within the affected source of the metal furniture surface coating NESHAP.
       In 63.4882 of the proposed rule, the affected source is described, in part, as "all storage
containers and mixing vessels" and "all manual and automated equipment and containers used for
conveying coatings, thinners, and cleaning materials." It was our intent when developing this regulatory
language that it be general enough that piping and pumps associated with the distribution of coatings (as
well as thinners and cleaning materials) would be included in the affected source. However, based on
                                             2-7

-------
these comments, the proposed rule may have been too general and could lead to confusion. We



modified 63.4882(b)(3) in the final rule as follows to clarify our intent:



               (3)  All manual and automated equipment and containers and all pumps and piping



       within the affected source used for conveying coatings, thinners, and cleaning materials;



       In addition, we realized that our use of the term "mixing" in 63.4882(b)(2) of the proposed



rule was not explicit enough for a source or permitting authority to determine whether the MCM



NESHAP would apply to certain metal furniture operations. We believe that some metal furniture



facilities may have operations that could be considered coating manufacturing. While not explicitly



stated in the proposed rule, it was our intent to include such operations in the metal furniture NESHAP



affected source, so long as the coating was manufactured solely for use within the metal furniture



surface coating facility in which it was made.  To clarify that such mixing operations, when they occur at



a metal furniture surface coating facility and for the sole use of that facility, are within the affected source



of the metal furniture surface coating NESHAP, we added the following definition to 63.4981 of the



final rule to work in conjunction with the affected source definition in 63.4882(b) of the final rule:



              Mix, mixed, or mixing means 1) to blend an as-purchased material to assure



       uniformity of its ingredients or 2}  to combine two or more ingredients to produce a coating.



       thinner, or cleaning material solely for the use by the facility in which it is mixed for



       manufacturing or refurbishing metal furniture.







       Comment: One commenter (IV-D-3) stated that military coatings are applied to tactical



vehicles, support equipment, and other battlefield support material such as temporary base camps,



deployable medical systems with metal racks, refrigeration and heating units and tactical shelters used to



house communications and electronics equipment with metal racks, tables, desks and air



conditioner/heater units.  Such military equipment would potentially be subject to the metal furniture,



large appliance, miscellaneous metal parts, wood furniture, and plastic parts surface coating rules.



Chemical agent resistant coating (CARC) systems and other military-unique coating systems are used



almost exclusively in these scenarios. Thus, one singular coating system might potentially have to
                                              2-8

-------
demonstrate compliance with a number of different NESHAPs, adding a tremendous administrative



burden.



       The commenter believed it would be difficult and costly to segregate the amount of coating used



on metal furniture, appliances, and miscellaneous metal parts especially since the components are often



painted at the same time using the same spray guns containing the same battlefield coating. The



coatings used are combat oriented with specific military performance requirements.  These battlefield



coatings must perform many functions not required of metal furniture coatings typically found in



commerce, including the following: resistance to chemical and/or biological agents, ability to be



decontaminated without degrading the performance of the coating system, low observability  and/or



reductions in signature/detection properties, provide electromagnetic field (EMF) shielding, and provide



field service protection of munitions for a period of 20 years.



       The commenter suggested adding language to 63.4881(c) of the proposed rule stating that



surface coating of metal furniture used by the military in a battlefield environment are to be regulated by



the miscellaneous metal parts and products NESHAP.



       Response: As stated by the commenter, this issue is not unique to the metal furniture surface



coating surface category. The specialized military coatings are used on products within at least the



metal furniture, large appliances, and miscellaneous metal parts surface coating source categories.



Rather than forcing all of these products into one of the existing source categories (such as



miscellaneous metal parts and products as suggested by the commenter), we believe the more



appropriate approach is to create a new source category for equipment used by the military in



battlefield situations or that are integral to military war-fighting equipment. In this way, we can more



effectively address the concerns unique to these coatings. We note, however, that surface coating



operations for the metal furniture addressed by this commenter will be subject to the final metal furniture



surface coating rule until we propose a rule for the new source category.







       2.2.2  Affected Source



       Comment: Two commenters (IV-D-2 and 9) noted that the proposed rule "essentially states



that companies are subject to the Metal Furniture NESHAP if they:  1) apply surface coatings to metal




                                             2-9

-------
furniture; 2) where metal furniture includes furniture or components of furniture constructed either



entirely or partially from metal; and 3) the affected source is a major source of HAP or is located at a



major source of HAP. The commenters requested specific language excluding non-major and/or area



sources.



        Response: We disagree that additional language is needed to exclude area sources. Paragraph



63.4881(b) of the proposed rule clearly states that an affected source must be a major source or



located at a major source in order for it to be subject to the rule. The change requested by these



commenters could result in the interpretation that any  affected source that by itself is an area source,



would not be subject to the rule.  This interpretation is not correct when such an area source is located



at a major source.  In order to avoid this possible misinterpretation, we have not made the change



suggested by these commenters in the final rule.







        Comment: Two commenters (IV-D-2 and 9) noted that the preamble discusses an affected



source as "the collection of all operations associated with the surface coating of metal furniture or



components of metal furniture that are performed at a contiguous area under common control." It is



unclear why EPA has not defined the affected source in a manner similar to other standards by using



terms such as facilities, major source, stationary source, or other terms familiar and well defined by



rules and policy memoranda.



        Response: As we explained in the preamble to the proposed rule, our primary goal in selecting



the affected source was to ensure that MACT is applied to the HAP-emitting operations or activities



within the source category. We can define the affected source in terms of a single stationary source



(emission point), group of stationary sources, or part of a stationary source.  We chose a broadly



defined group of stationary sources primarily to give you increased flexibility in determining how to



comply with the rule. Defining the affected source in this way provides you with the opportunity and



incentive to utilize control strategies that are more cost effective than if separate standards were



established for each emission point. Thus, you can choose where to most effectively implement



controls for your particular facility.
                                             2-10

-------
       Comment: Two commenters (IV-D-2 and 9) were concerned that 63.4881(c)(l) of the



proposed rule says that the rule does not apply to surface coating operations "conducted at a source



that uses only coatings, thinners, and cleaning materials that contain no organic HAP." First, the



commenters believed this is an example of where the rule is unclear with respect to the term "source."



Source is an undefined term. The rule should only use defined terms such as "major stationary source"



or "affected source." Second, it should be clarified that the determination that a substance "contain no



organic HAP" is determined in accordance with the provisions of 63.4941(a)(l)(i) of the proposed



rule.



       Response: We agree that the use of term "source" in 63.4881(c)(l) of the proposed rule may



be vague.  In the final rule, we have replaced "a source" with "an affected source." We also added that



whether the material contains no organic HAP is determined by the procedures specified in



63.4941 (a).  This will clear up any ambiguity over whether a material containing only trace amounts of



organic HAP would qualify as containing "no organic HAP."







       Comment: One commenter (IV-D-2) generally agreed with the list of the types of sources for



which the proposed rule does not apply as presented in 63.4881(c)(5) of the proposed rule. The



commenter proposed two changes be considered. First, 63.4881(c)(5) states: "Surface  coating of



metal furniture conducted for the purpose of repairing or maintaining metal furniture used by a facility



and not for commerce is not subject to this subpart unless organic HAP emissions from the surface



coating itself are as high as the rates specified in paragraph (b) of this section." The phrase "unless



organic HAP emissions from the surface coating itself are as high as the rates specified in paragraph (b)



of this section" should be stricken.  That phrase could be interpreted to mean that any facility that was a



major source of HAPs (as defined under 63.4881 of the proposed rule) from surface coating



operations would be subject to the proposed rule if they ever coated any metal furniture. Also, this



provision uses the term "facility." This provision should apply to both affected sources and other



entities.



       Response: We disagree that the phrase specified by the commenter should be removed. Our



intent was to exclude small metal furniture surface coating operations that are incidental to other




                                             2-11

-------
manufacturing operations. However, if the repair and maintenance of metal furniture rises to the level of



being a major source itself, then it must be regulated under the final rule. To do otherwise would violate



the CAA and allow a major source to go unregulated. Also, the commenter's interpretation of the



wording of this section is correct-if you are a major source and you coat metal furniture, then you are



subject to the rule.  You may then make use of the exemptions to the rule if they apply to your



operations.



       We agree that the use of the term  "facility" does not correctly convey our intent.  We have



amended (c)(5) by substituting "facility" with "major source."  You must look at all the operations at a



major source to determine which ones fit  the description given in 63.4881(c)(5) of the proposed rule,



not just those within the affected source.







       2.2.3  Low Usage Exemption



       Comment:  Two commenters (IV-D-3 and 4) suggested that the rule should include a low use



exemption. One commenter (IV-D-3) noted that such an exemption is consistent with other



NESHAPs and many State rules upon which the NESHAPs are based. A low use exemption would



prevent insignificant operations that occur at major HAP sources from being subject to the regulation.



The commenters believed all surface coating rules should contain a low use exemption to prevent the



rule from applying to insignificant metal furniture surface coating operations that do not qualify for one of



the exemptions in the proposed rule. Such an exemption is appropriate for any major HAP source



facility that performs insignificant amounts of metal furniture surface coating. It would also be



inappropriate to regulate a private facility that is a major HAP source due to other operations, but



performs an insignificant amount of metal  furniture surface coating.



       One commenter (IV-D-4) noted that like numerous other industries with incidental



manufacturing of tools and support equipment, they fabricate a small quantity of metal furniture solely



for their own use and not for commerce.  EPA should exclude from the requirements of the proposed



rule incidental metal furniture manufacturers who are primarily engaged in the manufacture of products



other than metal furniture and who utilize  no more than 100 gallons per month or 1,200 gallons per year



of coatings to manufacture such metal furniture.




                                            2-12

-------
       Response: The proposed rule was not intended to apply to incidental metal furniture surface
coating operations when those operations were for the purpose of repairing or maintaining metal
furniture used by a major source (see 63.4881(c)(5)).  We specifically did not provide an exemption
for the manufacture of metal furniture. If we allowed such an exemption, the possibility would exist for
a large facility to manufacture most if not all of its own metal furniture free from the compliance
requirements of this rule. This could result in a negative economic impact on the metal furniture
industry. While the commenters suggested alternatives for a low usage exemption, no supporting data
were included with these comments to demonstrate that the limits would assure that no sources would
escape regulation that should be subject to the rule. We also considered establishing a low usage
exemption when developing the proposed rule (such as the exemption in subpart JJ for incidental wood
furniture manufacturers using no more than 100 gallons of finishing material  or adhesives), but we did
not have sufficient data to establish a limit.  For these reasons, we have decided not to include a usage
limit in the final rule.

       2.2.4  Other Exemptions
       Comment: One  commenter (IV-D-5) said that although section 112 of the CAA does not
allow EPA the discretion to exclude any sources which qualify as major sources, under the proposed
rule, janitorial, building, and facility maintenance operations and coating applications using handheld
nonrefillable aerosol containers are excluded from compliance even if they are emission points at a
major source. The commenter believed this exemption contravenes the CAA, as well as EPA's own
regulations.
       The proposed rule also exempts those sources that conduct surface coating of small items such
as hinges under the pretext that they  serve a wider use than the metal furniture industry. The proposed
rule also does not govern sources that repair and maintain metal furniture for non-commerce purposes.
If the  sources qualify as major sources and the emissions are the result of surface coating of metal
furniture, then it was the commenter's opinion that the EPA has no authority to exclude such sources
from compliance.  Therefore, both of these exemptions violate the CAA and EPA's own regulations.
       The proposed rule does not regulate wood furniture makers that also coat metal
                                             2-13

-------
components of that wood furniture purportedly because such sources are regulated under



subpart JJ for wood furniture manufacturing.  EPA itself has defined metal furniture to include



"components of furniture constructed either entirely or partially from metal."  Therefore, it was the



commenter's opinion that this exclusion also violates the CAA.



       Response: We disagree that the exemptions addressed by this commenter violate the CAA or



our own regulations.  Janitorial, building, and facility maintenance operations are not part of the metal



furniture surface coating source category, even though they occur at the same facility at which metal



furniture surface coating occurs. During our review of this issue, however, we found that defining



"facility maintenance operations" would be helpful to distinguish these operations from operations within



the source category.  Thus, we added the following definition to 63.4891:



               Facility maintenance operations means the routine repair or renovation (including the



       surface coating^ of the tools, equipment, machinery, and structures that comprise the



       infrastructure of the affected facility and that are necessary for the facility to function in its



       intended capacity.



       We also disagree that exempting nonrefillable handheld aerosol containers was a violation of the



CAA.  Under section 112(d)(l) of the CAA, we have the authority to "distinguish among classes,



types, and sizes of sources within a category or subcategory in establishing . . .  standards."  We



considered nonrefillable handheld aerosol containers to be  a different type of source (as compared to



typical high capacity metal furniture surface coating operations such as  spraying and dipping), because



the coating applied by this type of source must meet specific requirements in order to be sprayable from



an aerosol can. We found no practical  controls applicable  to this type of source and chose to exempt it



from the affected source to reduce the recordkeeping burden on the industry.



       We do not believe this will create a loophole in the rule to allow sources to use large quantities



of nonrefillable handheld aerosol containers to apply coatings. First, the metal furniture industry is highly



driven by  the appearance of the coating on the finished product.  The high quality appearance



demanded by customers cannot be achieved through the use of coatings applied from a handheld



aerosol container.  Second, many of the coatings used by the industry are custom made to each



manufacturer's specifications. Aerosol coatings are limited  in availability, so the wide range of colors




                                              2-14

-------
and performance characteristics (such as wear resistance and flexibility) currently offered by furniture

manufacturers could not be maintained. Third, aerosol containers apply coatings at a very slow rate

compared to typical spray guns used in the industry.  This slow application rate would slow production

rates far below current levels. Finally, on a unit basis, aerosol containers of coating are more expensive

than purchasing coatings in bulk, as nearly all furniture manufacturers do.  Thus, it would be cost

prohibitive for a furniture manufacturer to use aerosol coatings for more than just touch-up operations.

       We have said that this rule does not apply to the coating of small items such as hinges and

screws that have a wider use beyond metal furniture. Because these small items could be used by a

number of different source categories (such as metal furniture, wood furniture, and large appliances), it

would be impossible to determine which rule should apply when these parts are the only metal

components coated at a major source facility. However, if coatings are applied to these small metal

parts within a metal furniture facility, both the proposed and final rules specify that the coating operation

would be included in the affected source and thus covered under the metal furniture rule.

       We again disagree that it was improper for us to exclude surface coating operations for repair

and maintenance of metal furniture not meant for commerce.  The source category is described as

including "any facility engaged in the surface coating and manufacture of metal furniture parts or

products."1  We interpret this to mean that the parts and products are manufactured for commercial

purposes.  Therefore, operations such as repainting of metal furniture items at a hobby  shop located on

a military base or metal chairs used within a manufacturing facility for use by that facility's employees are

not within the intended source  category, so long as these repair and maintenance operations are not

major sources.


       Comment: One commenter (IV-D-5) believed that the exclusion of synthetic minor sources

(those sources which would be considered major sources without the use of enforceable permit

conditions) excludes significant sources of HAP from the proposed rule. Under National Mining
       1 Documentation for Developing the Initial Source Category List, Final Report. U.S. EPA,
Office of Air Quality Planning and Standards.  Research Triangle Park, NC. EPA-450/3-91-030. July
1992.  p. A-31.

                                             2-15

-------
Association v. United States Environmental Protection Agency. 59 F.3d 1351, 1363 (D.C. Cir. 1995),



controls may only be taken into consideration in determining a source's potential to emit (and thus,



whether it is a major source), if they are "demonstrably effective." The proposed rule, however, allows



sources to obtain synthetic minor status regardless of whether their limitations are "demonstrably



effective."  Thus, it provides no assurance that all major sources will be covered, as required by the



CAA.



       Response: The proposed rule allows a source to obtain synthetic minor status any time before



the compliance date of the final rule (three years after promulgation of the final rule). In order to



become a synthetic minor source, a facility would have to obtain a State operating permit containing



operating or emission limits that maintain the synthetic minor status. We disagree that such operating



permits cannot be classified as "demonstrably effective." All State operating permits require some level



of monitoring to assure the permitting authority that the limits are being adhered to, and reporting results



of the monitoring is also typically required.  Consequently, the facility does demonstrate that real



emission reductions are being achieved.



       Section 112 of the CAA defines different criteria for developing rules for major sources and



area sources. The proposed rule, as well as the final rule, applies only to major sources.  Once the



synthetic minor status is achieved, as long as it is before the  compliance date, we have no authority to



regulate the source under the final rule because it applies only to major sources. In the future, we may



regulate metal furniture surface coating area sources under a separate rulemaking.







       Comment: One commenter (IV-D-5) stated that the proposed rule excludes all inorganic  HAP



emitted by sources in the surface coating of metal furniture.  The commenter believed this exemption is



in violation of the CAA which requires EPA to set standards for all HAP emitted by the source



category. Although EPA argues that the use of coatings containing inorganic HAP is not currently



widespread enough to merit emissions standards, the commenter did not believe EPA has such



discretion.  The commenter pointed out the holding in National Lime v. EPA 233 F.3d 625, 633-34



(D.C. Cir. 2000), which stated that EPA has a clear statutory duty to regulate all major sources of



HAP emissions.




                                             2-16

-------
       Response:  In the preamble to the proposed regulation, we pointed out the very small usage of

coatings in the metal furniture surface coating industry that contain inorganic HAP (two coatings out a

total of approximately 680 coatings reported in industry questionnaire responses). Throughout the data

gathering efforts (including site visits, industry questionnaires, and literature searches) for this

rulemaking, we also found that coating application operations were predominately equipped with either

dry filters or waterwash systems2 to reduce the amount of overspray3 emitted to the atmosphere. Such

control systems are common in many other coating industries as well, and are installed and maintained

for a number of reasons. This type of control reduces the amount of coating droplets emitted to the

atmosphere. Because the inorganic HAP are contained in these droplets, these control systems reduce

the amount of inorganic HAP emissions. These controls have been in general use for many years.  We

know of no reason why the industry would remove these controls after the final rule is promulgated.

       If we did develop a MACT floor for inorganic HAP, we expect that it would be the very dry

filters and waterwash systems currently in place. Given the combination of very low usage of coatings

containing inorganic HAP in the metal furniture surface coating industry as indicated above and the

current (and expected continued) use of controls to reduce overspray emissions, we believe including

such controls in the rule would not result in any additional emission reduction and only add to the

regulatory burden on the industry and the permitting authorities. For these reasons, we have not added

emission limits for inorganic HAP in the final rule.


       2.2.5   Definitions
       2 See "National Emission Standards for Hazardous Air Pollutants (NESHAP) for Source
Category: Metal Furniture Surface Coating-Background Information for Proposed Standards.  EPA-
453/R-01-010. U.S. Environmental Protection Agency. Research Triangle Park, North Carolina.
October 2001.  Section 4.2.

       3 Overspray is the droplets of coating that do not adhere to the substrate being coated or the
surfaces of the spray booth and are carried in the exhaust stream of the spray booth.  Inorganic HAPs,
because of their lack of volatility, are contained in these droplets.

                                             2-17

-------
       Comment: One commenter (IV-D-2) noted that 63.4881(a)(l) defines "surface coating" as
the "application of coatings to a substrate using, for example, spray guns or dip tanks."  The commenter
thinks this definition is too broad for the rule.  Since many stationary sources impacted by this rule may
coat numerous substrates, this definition should restrict the definition of surface coating to those coatings
applied to metal furniture or components of metal furniture. Also, the method of applying the coating is
not necessary in the definition of surface coating. The commenter suggested revising the definition of
surface coating to mean "the application of coatings to a substrate which is a component of metal
furniture."
       Response: Paragraph 63.4881(a) defines the source category to which the proposed rule
applies as the "surface coating of metal furniture." Paragraphs 63.4881(a)(l) and (2) then define what
we mean by "surface coating" and "metal furniture." When these three paragraphs are read together as
intended, not individually as the commenter has done, we believe there is no ambiguity,  and the
definitions are not overly broad. Also, the application methods are provided only as an example and
do not limit the definitions in any way, and were provided simply as guidance to the reader. We believe
the definitions are appropriate as written and made no changes in the final rule as  a result of this
comment.

       Comment: One commenter (IV-D-2) noted that 63.4881(a)(2) defines  metal furniture as
"furniture or components of furniture constructed either entirely or partially from metal." The
commenter thought this definition was too broad for the rule.  It would imply that any coating on any
component constructed with any metal would be subject to this rule.  Under this definition, even a piece
of wood furniture, which contained metal fasteners or glides, would constitute metal furniture.  This sets
up a conflict between the definitions contained in  the proposed rule and the provisions of the proposed
rule whereby EPA attempts to avoid overlap of the multiple NESHAP surface coating rules. The
commenter suggests changing the term "partially" to "predominantly" or "primarily."
       Response: As we discussed in previous comment responses concerning  overlap between the
proposed rule and other surface coating rules, we  concluded that it was unworkable to define the
affected source in terms of the amount of metal in the item being coated. We necessarily would have
                                             2-18

-------
had to define "predominantly" or "primarily" if we had chosen to use those terms in the definition. We



deliberately did not include such limiting terms in the definition for the reasons discussed previously.



       We did, however, realize that the definition used in the proposed rule could lead to apparent



coverage of the same coating operations under both the metal furniture and wood furniture rules. For



this reason, we included the exemptions in 63.4881 to clarify which of those operations would be



subject to the proposed rule. We also provided guidance in previous comment responses (see Section



2.1.1 of this document) for determining applicability for several common situations.







2.3    MACT FLOORS



       2.3.1   Basis of the MACT Floors



       Comment: One commenter (IV-D-5) noted that the MACT floors, as specified in the CAA,



must reflect the actual performance of the relevant best controlled sources.  For new sources, this must



be the single best performing source.  The MACT floor for existing sources must be based on the



average of the best performing 12 percent of sources.  Rather than setting emission standards based on



actual emissions, EPA proposed standards based solely on the FLAP content of the materials used by a



source. As EPA recognizes, however, some sources use capture technologies to control their



emissions. Thus, their actual performance is affected by that technology as well as the FLAP content in



the materials.  Accordingly, the commenter believed that the MACT floors do not reflect the best



sources' actual performance as required by the CAA.



       Even assuming that EPA can base floors solely on the FLAP contents in the materials used, the



commenter believed the EPA has done so unlawfully.  The CAA expressly requires EPA to base



MACT floors on "the best performing twelve percent of the existing sources (for which the



Administrator has emissions information)." Although EPA has emissions information for only 22 of the



49 major sources it identified in the category, the MACT floors were based on the FLAP content for the



best six rather than the best three (that is,  12 percent of 49 sources rather than 12 percent of 22



sources).



       Response: We did not base the MACT floor on the FLAP content of the materials used as the



commenter stated. Rather, we calculated an emission rate for each source normalized by the amount of




                                            2-19

-------
coating solids used. For this calculation, a basic assumption we used was that all of the volatile
components of the coatings, thinners, and cleaning materials used would evaporate and be emitted to
the atmosphere. Except where add-on control devices are used, this is typically the case.  In those
instances where a facility in our database reported the use of an add-on control  device, we took into
account the capture and destruction efficiency of the control device when calculating the normalized
facility emission rate for the purpose of determining the MACT floors.  Thus, our analysis was in fact
based on the actual performance of the sources as required by the CAA.
       As discussed in the preamble to the proposed rule, there were a total of 49 facilities in the
MACT floor database.  Based on the information provided by these facilities in  response to industry
questionnaires, only 22 of these facilities provided all of the data required to calculate overall emissions
in terms of kilograms organic HAP emitted per liter of coating solids used. We had nearly complete
data on all of the other 27 facilities.  From all of these data, we estimated that the subset of 22 facilities
was an adequate representation of the 49 facilities. We believe that had we been able to determine the
emission rates of the other 27 facilities, the values would have fallen within the range represented by the
22 facilities for which we had complete data.  We believed that the existing source MACT floor would
not have changed to any considerable degree if we had complete data for all 49  facilities.  However,
we agree with the commenter that the CAA language is explicit in that we must use 12 percent of the
facilities for which we have emissions information (see section 112(d)(3)(A) of the CAA), which is 22
facilities. The existing source MACT floor would then be based on the best performing three facilities
(12 percent of 22).  This revised existing source MACT floor is  0.10 kg organic HAP per liter coating
solids used (0.83 pounds per gallon). We changed 63.4890(c) of the final rule to reflect the revised
existing source MACT floor.

       Comment:  Two commenters (IV-D-2 and 9) noted that in the Background Information
Document (BID) accompanying the proposed rule, both major sources and synthetic minor sources
were included in the calculation of the MACT floors.  The commenters objected to the incorporation of
synthetic minor sources in the MACT floor calculation because,  by definition, the MACT standard is to
apply to "major sources" of HAP.
                                            2-20

-------
       Response:  We disagree that the MACT floors must be based solely on major sources of HAP
emissions.  Section 112(d)(l) of the CAA directs us to promulgate regulations for categories of major
and area sources of HAP emissions.  Then, section 112(d)(2) mandates that these standards "shall
require the maximum degree of reduction in emissions . . . achievable for new or existing sources."
Section 112(d)(3) specifies how we are to determine the maximum degree of emission reduction and
describes it as "not less stringent than the emission control that is achieved in practice by the best
controlled similar source" for new sources, and for existing sources describes it as "the average
emission limitation achieved by the best performing 12 percent of the existing sources .  . ." Even though
Congress saw fit to distinguish between major and area sources in many other places in section 112 of
the CAA, they did not specifically require that the floor be based on major sources alone.  Throughout
section 112(d), Congress simply used the term "source." We  interpret this to mean that Congress left it
to our discretion to determine the most appropriate sources on which to base the MACT floors.
Accordingly, for the proposed rule we used both major sources and synthetic minor sources as the
basis of the MACT floors. We believe our interpretation of section 112(d) of the CAA is correct, and
no changes were made to the final rule as a result of these comments.

       Comment:  One commenter (IV-D-5) stated that the CAA requires "the maximum degree of
reduction in emissions of hazardous air pollutants" that can be  achieved through substitution of materials;
changes in processes; enclosures; collection, capture, and treatment; and design, equipment, work
practice, or operational standards.  The commenter believed that the EPA did not required the
maximum degree of emission reduction when it developed the existing source MACT floor.
       Despite the availability of coating technologies that may emit no organic HAP, such as powder,
UV, and autophoretic coatings, the EPA did not require their use. Similarly, although the record shows
that add-on capture and control technologies are available and currently in use, the EPA did not base
the existing source MACT floor on the use of this technology.  Basing the existing source MACT floor
on these coating technologies and add-on capture and control technologies would have reduced
emissions more than that achieved through the proposed rule.  For example, the EPA published a
report which shows that a capture rate of 75 percent would reduce emissions for the small, medium,
                                            2-21

-------
and large model plants to below 0.10 kg organic HAP/L coating solids used. Given that the EPA has
not provided any reason to believe that the use of add-on technology is not achievable technically or
economically, the EPA's refusal to set standards based on the use of such technology contravenes the
mandate in section 112(d)(2) of the CAA for the maximum degree of emission reduction.
       Furthermore, EPA did not require the use of low organic HAP content or organic HAP-free
liquid coatings even though this would further reduce emissions.  The record shows that some coatings
currently in use contain no organic HAP and thus would have no organic HAP emissions. The EPA's
sole reason for not incorporating such technology is that it does not consider them to be "different
emission control technology" than what is already being used by the six best performing sources.  First,
it is unclear that the two coatings are in fact the same.  Second, even if they were, the only relevant
consideration under section 112(d)(2) is whether standards based on zero organic content would
reduce emissions below the levels required by EPA's proposal and whether they would be achievable.
It is plain that they would reduce emissions further, and nothing in the record suggests that they are, in
any regard, not achievable.
       Response: From the information provided by the commenter, it was unclear whether the
comments were directed to only the existing source MACT floor or to the new source MACT floor as
well.  We decided that, since the arguments presented could apply to both MACT floors, we would
consider both in our response.
       The existing source MACT floor was based on the average emission rate of the best performing
sources, as directed by section 112(d)(3) of the CAA (see the response to the first comment in Section
2.3.1 of this document for further discussion on the sources we used to develop the  existing source
MACT floor).  We believe this was the proper procedure to use to determine the existing source
MACT floor. The argument made by this commenter that we should have considered other emission
control technologies when developing the existing source MACT floor is more appropriately addressed
in Section 2.3.2 of this document (Alternatives More Stringent Than the MACT Floors).
       In the preamble to the proposed rule, we discussed our findings that the use of organic
HAP-free coating technologies was not widespread enough to cover the range of activities found in the
entire metal furniture industry.  While developing our response to these comments, we began to further
                                            2-22

-------
consider how the state-of-the-art for new sources has changed since our initial data gathering efforts.



We have always recognized that there are certain coating technologies that may emit no organic HAP



(according to 63.4941 (a) of the final rule) such as powder coatings and liquid coatings that contain no



organic HAP.  The industry questionnaires that we conducted in 1998 (representing 1997 data)



showed that of the 49 facilities that were eventually used for the MACT floor database, six used



powder coatings exclusively. These powder coating facilities produced metal furniture items such as



office chairs, dental chairs, commercial and residential lighting fixtures, and indoor and outdoor lighting



fixtures. Of these six powder coating facilities, three had coating solids usages within the range



represented by the facilities we used to determine the MACT floor (in other words, had these three



facilities used conventional liquid coatings instead of powder coatings, we would expect them to be



major sources of HAP emissions and they would have been included in the determination of the MACT



floor).  We believe these data demonstrate the industry's current ability to exclusively use powder



coatings in many situations.



        Based on the 1998 questionnaire responses of the 22 facilities that provided complete data,



information was provided for 188 individual liquid coatings.  Eight of these coatings were reported as



containing no organic HAP. In addition, another 48 individual coatings were reported as containing less



than 1 percent by mass of organic HAP  (typically as a small component of a solvent blend such as



aromatic naphtha).  We believe that this high percentage of non-HAP (or essentially non-HAP)



coatings used by these 22 facilities indicates the coating suppliers' ability to produce and market non-



HAP coatings, and demonstrates that they are currently in use by the industry.



        Over the past five years since we sent questionnaires to the metal furniture manufacturing



industry, non-HAP coating technologies have undergone continual development.  The availability of



powder coatings in a wide range of colors has continually increased, as has the ability to produce



various  surface finishes and control film thickness. Coating manufacturers have also made significant



strides in formulating non-HAP coatings, driven in large part by the requirements of surface coating



NESHAPs for a wide variety of industries. In addition, we are aware of other coating technologies,



such as  electrocoating, that have the potential to emit no HAP.  Although we are not currently aware of
                                             2-23

-------
these coating technologies being used in the metal furniture industry, we believe they can be used in
certain circumstances and represent viable alternatives for new sources.
       We believe the continual development of these non-HAP coating technologies over the past
several years has allowed them to gain wider acceptance and use within the metal furniture surface
coating industry such that we now believe they represent in the aggregate the MACT floor for new
sources. Considering that new sources have much greater latitude than existing sources to design
manufacturing operations and the metal furniture items themselves to accommodate these coating
technologies, new sources can more readily take advantage of these coating technologies.  Accordingly,
we have revised the new source MACT floor to be no emissions of organic HAP from metal furniture
surface coating operations.  The emission limit for new sources in 63.4890 of the final rule reflects this
new MACT floor determination.
       We also recognize that there may be specialized appearance or functional characteristics can be
produced only with coating technologies employing organic HAP, even for new sources. To
accommodate these situations, we added a provision in the final rule that allows a new affected source
to demonstrate on a case-by-case basis that organic HAP-free coating technologies cannot be used.  If
we approve such a request, then the source would be required to meet an emission limit of 0.094 kg
organic HAP/L (0.78 Ib/gal) coating solids used, which is the new source emission limit originally
proposed for new sources.  The following language was added at 63.4890(b) of the final rule:
              (b) Alternative emission limit. You may request approval from the Administrator to use
       an alternative new source emission limit for specific metal furniture components or type  of
       components for which you believe the emission limit in paragraph (a) of this section cannot be
       achieved.
              (1) Any request to use an alternative emission limit under paragraph (b) of this  section
       must contain specific information demonstrating why no organic HAP-free coating technology
       can be used on the metal furniture components.  The request must be based on objective
       criteria related to the performance or appearance requirements of the finished coating, which
       may include but is not limited to the criteria listed in paragraphs (b)(l)(i) through (viii) of this

                                            2-24

-------
               (i) Low dried film thickness requirements (e.g.. less than 0.0254 millimeters (0.001



       inch)).



               (ii) Flexibility requirements for parts subject to repeated bending.



               (iip Chemical resistance to withstand chemical exposure in environments such as



       laboratories.



               (iv) Resistance to the effects of exposure to ultraviolet light.



               (v) Adhesion characteristics related to the condition of the substrate.



               (vi) High gloss requirements.



               (vii)  Custom colors such as matching the color of a corporate logo.



               (viii) Non-uniform surface finishes such as an antique appearance that requires visible



       cracking of the dried film.



               (2) If the request to use an alternative emission limit under paragraph (b} of this section



       is approved, the new source must meet an emission  limit of 0.094 kilogram organic HAP per



       liter (kg/U (0.78 pounds per gallon flb/gal)) coating solids used for only those components



       subject to the approval. All other metal furniture surface coating operations at the new source



       must meet the emission limit specified in paragraph (a^ of this section. Until approval to use the



       alternative emission limit has been granted by the Administrator under this paragraph, you must



       meet the emission limit specified in paragraph (a) of this section and all other applicable



       requirements in this subpart.







       2.3.2   Alternatives More Stringent Than the MACT Floors



       Comment: One commenter (IV-D-2) supported EPA's analysis of alternatives more stringent



than the MACT floor. The commenter thought that the floors represent stringent standards for both



new and existing sources.



       Response: We appreciate the commenter's support on this issue.  Based on other comments,



however, we revised the MACT floor for both new and existing sources as discussed in Section 2.3.1



of this document.
                                             2-25

-------
       Comment: One commenter (IV-D-7) pointed out that similar emission limits for affected new



sources have been in place for over two decades under the metal furniture surface coating NSPS (40



CFR part 60, subpart EE).  While the proposed rule may be more far reaching in the definition of



affected source, the NSPS requirements are potentially more stringent in that they more



comprehensively address VOC emissions rather than the subset of HAP emissions.  Because the



NSPS standard has been in effect for over two decades, it seems logical that the MACT standards for



both existing and new sources should be more stringent. Clearly, adopting standards for existing



sources that are less stringent than 20-year-old NSPS requirements is a problem. For perspective, by



the time the MACT standards are fully in effect later this decade, the NSPS requirements will have



been in effect for nearly 30 years. The commenter believed this is adequate justification for EPA to



reexamine the rationale that supported the rejection of alternatives more stringent than the MACT floor.



       Response: The CAA gives us authority to regulate only HAP, not VOC.  Thus, without further



justification that VOC could be used as a surrogate for HAP, we have no  authority to regulate VOC



directly through the proposed rule.  Therefore, we disagree with the assertion that because the NSPS



may regulate a broader group of pollutants, we should further consider alternatives more stringent than



the MACT floors.



       We believe it is improper to determine which of these two rules is  more stringent simply by



looking at the numerical emission limit. First, the affected source for subpart EE is each individual



coating operation. The affected source for the proposed rule is the collection of all coating operations;



all storage and mixing vessels; all equipment and containers and all pumps  and piping within the affected



source used for conveying coating, thinning, and cleaning materials; and all storage containers and



equipment and containers for conveying waste materials. Because of this more encompassing affected



source in the proposed rule, many more emission points are included, as well as more materials from



which the emissions emanate (coatings, thinners, cleaning materials, and waste as opposed to just



coatings and thinners).



       Second, while the emission limits in both rules are normalized by the volume of coating solids,



the proposed rule counts all coating solids used, while subpart EE counts all coating solids applied.



Consequently, subpart EE requires determination of transfer efficiency and subsequent calculation of the




                                             2-26

-------
amount of solids actually applied to the substrate.  In contrast, the proposed rule does not take transfer



efficiency into account. Instead, the proposed rule counts all coating solids that enter the coating



application process, regardless of their ultimate fate.



       Because of these basic and substantial differences in these two rules, a direct comparison of



their stringency is not a straightforward exercise. Accordingly, based on this comment, we do not



believe further consideration of alternatives more stringent than the MACT floor is warranted.







       Comment: One commenter (IV-D-7) noted that the purpose of the NESHAP control program



is to set national emission standards for major sources in industries that emit HAP.  The CAA clearly



identifies the MACT floor as the minimum baseline requirement under the NESHAP program. The



EPA is encouraged to then consider alternatives more stringent than the MACT floor. In fact, the CAA



requires the NESHAP to reflect the maximum degree of reduction in HAP emissions that is achievable.



Although the proposal identifies at least three alternatives that should be fully considered in establishing



the existing source MACT floor, the proposal failed to conform to the requirements of the CAA.



       The EPA  dismissed individually three alternatives: powder coatings, lower organic HAP liquid



coatings, and emission capture and control technologies.  In the case of powder coatings and lower



organic HAP content liquid coatings, the EPA reasoned that these options cannot be used on all



product types in this industry,  and  did not warrant further consideration. The capture systems and



control devices alternative was rejected solely on cost.



       The proposal states that, conservatively, a capture and control system could cost approximately



$1 million.  However, the lack of experience in applying these controls to this specific industry, rather



than lack of feasibility, and the case-by-case nature of applications,  make this cost prediction unreliable.



The commenter disagreed with the conclusion that the additional emission  reduction would not justify



the additional cost. The emission reductions are substantial  and the actual cost of control for  this



alternative is inconclusive at best.



       Table 6-3  in the BID to the proposed rule identifies the substantial emission reductions that



would be achieved by establishing more stringent emission rates or levels that can be achieved by
                                             2-27

-------
capture and control technology.  The commenter believed the conclusion to dismiss all of the more



stringent alternatives based on emission limits was inconsistent with the stated information.



       The commenter believed that all three options appear to be technically feasible. Emissions



capture and control devices have broad applicability throughout the surface coating of metal furniture



industry, and the coating alternatives, while not universally applicable, provide a strong and effective



alternative for various segments of the industry. In fact, the reductions achievable by these alternatives



should collectively form the basis for the MACT standard.  As such, given the wide range of



alternatives and their applicability, EPA should establish the MACT standard based on what is



achievable and allow industry to determine which alternative is the best fit.



       Another commenter (IV-D-5) believed that the use of alternatives more stringent than the



MACT floors was dismissed because the EPA claimed it did not have the necessary information on the



benefits which would be achieved by reducing emissions below the floor. According to the commenter,



the only benefit that EPA may consider under section 112(d)(2) is emission reduction. Congress has



mandated the maximum degree of reduction achievable, regardless of what EPA believes the benefits of



such a standard to be. Thus, EPA's doubts about the benefit of a greater degree of reduction are



statutorily irrelevant.



       Response: We agree that there are specific circumstances in which the three more stringent



alternatives are in use by existing sources and discussed this in the preamble to the proposed rule (see



67 FR 20215-6).  In that discussion, we also examined the limitations of powder coatings, lower



organic HAP content liquid coatings,  and organic HAP-free liquid coatings and concluded that all of the



coatings currently applied by existing metal furniture surface coating facilities cannot be replaced by any



combination of these coatings. Although new sources have increased flexibility over existing sources to



make these coating technologies more viable (see our discussion in Section 2.3.1 of this document) we



continue to believe that existing sources cannot make use of these coating technologies in all situations.



Therefore, existing facilities would either have to stop producing those products that  could not be



coated using one of the two alternatives (powder coatings and lower organic HAP content liquid



coatings), or produce an inferior product using one of the two alternatives. Section 112(d)(2) of the



CAA directs us to promulgate standards that require the maximum amount of emission reduction




                                             2-28

-------
achievable by all sources in the source category.  Since these alternatives cannot replace all current
coatings in use by existing sources, the potential emission reduction they represent is not achievable and
a standard based on the two alternatives would thus violate the CAA.  We also believe that stopping
production of certain products or producing inferior products would create a severe economic burden
on the industry. Because  section 112(d)(2) directs us to consider the cost of achieving this additional
emission reduction, we believe the economic burden on the industry would be unreasonable in
comparison to the emission reduction achieved. Again, we would be in violation of the CAA if we used
these alternatives as the basis for a standard more stringent than the MACT floor for existing sources.
       Furthermore, the format of the standard in the proposed rule (mass of organic HAP per volume
of coating solids used) encourages the use of these lower organic HAP technologies. We believe
existing sources will increasingly use such technologies because they beneficially affect the compliance
demonstration calculations.  We also clarify that contrary to the reasoning presented by one of the
commenters, the MACT floor cannot be based on just a coating; the MACT floor must be based on
the entire source (section  112(d)(2) and (3) of the CAA).
       We did identify add-on capture and control technology as a technically feasible alternative more
stringent than the MACT  floors. One commenter questioned the basis for our cost estimate of this
option as presented in the  preamble. We note here that the preamble contained only a brief summary
of a thorough cost analysis using accepted EPA methodology for estimating the cost of both a total
enclosure and a thermal oxidizer. This analysis is documented in the project docket (Docket No.
A-97-40). Through this analysis, we determined that the cost of additional emission reduction beyond
that achieved by the emission limits in the proposed rule, on a nationwide basis, of installing add-on
capture and control systems on existing sources would be $138,000 per megagram of additional HAP
emission reduction ($125,000 per ton). We believe these costs are unreasonable and any standard
based on them would violate the directive of section 112(d)(2) of the CAA to consider the cost of
achieving the additional emission reduction.
       One commenter also objected to our statement in the preamble that we did not consider
above-the-floor options because we did not know the benefits of implementing  such options.  When we
used the term "benefits," we meant the affect  on environmental and health risks associated with HAP
                                             2-29

-------
emissions from the metal furniture surface coating industry.  We did not reject the above-the-floor



options solely because we lacked this benefits information. We simply stated that, given the excessive



cost of additional organic HAP emission reduction that would be imposed by above-the-floor options,



we had no information on the benefits of further reducing organic HAP emissions that would compel us



to choose an above-the-floor option anyway.



       For all of these reasons, we believe our rejection of the alternatives more stringent than the



MACT floors for existing sources was correct and appropriate.  In Section 2.3.1 of this document we



discussed how we revised the new source MACT floor determination. Because the revised MACT



floor is zero organic HAP emissions, there was no need to discuss comments concerning alternatives



more stringent than the floor as directed to new sources.







2.4    EMISSION LIMITS



       2.4.1   Compliance Options



       Comment: One commenter (IV-D-2) requested that 63.4890 clearly state that the emission



limitations are to be determined on a monthly average basis.



       Response: In 63.4890, we specify that the emission limitations are to be met for each



compliance period. We disagree that this section should specify the length of the compliance period.



We detailed the length of the compliance period as one month in 63.4942(a), 63.4952(a), and



63.4963(a) (renumbered as 63.4962(a) in the final rule)  of the proposed rule, which present the



compliance determination procedures.  We believe it is clearer to have the compliance period specified



along with the compliance demonstration procedure.







       Comment: One commenter (IV-D-2) noted that  63.4891 of the proposed rule addresses



which materials must be included in  emission calculations  for purposes of demonstrating compliance.



For purposes of clarity and to avoid  any potential overlap problems between various surface coating



NESHAPs, this provision should clearly state that the materials to be accounted for are those "coatings,



thinners, and cleaning materials used for the manufacture of metal furniture."
                                            2-30

-------
       Response:  Section 63.4891 of the proposed rule directs you to "include all coatings, thinners,
and cleaning materials used in the affected source" in your compliance demonstration. Because the
affected source necessarily includes only those operations used for surface coating of metal furniture
(see 63.4882(b) of the proposed rule), only the coatings, thinners, and cleaning materials used for the
manufacture of metal furniture would be included. Thus, the additional language requested by the
commenter would be redundant, and we did not make the change in the final rule.

       Comment:  One commenter (IV-D-2) noted that 63.4891 (a) of the proposed rule describes
the compliant material option as requiring that the organic HAP content of each coating used in the
coating operation be less than or equal to the applicable emission limit in 63.4890, and that each
thinner and each cleaning material used contains no organic HAP.  This provision appears to require
double counting of thinners and cleaners.  First, they must individually not contain any HAPs, and
second, they must be included in the calculation of overall organic HAP emissions.  Although the
commenter understood that the floor was based on HAP content of coatings including thinners  and
cleaners, they did not recall any basis for requiring each thinner and each cleaning material to contain no
organic HAP.  In particular, this provision does not appear to be supported by the data reflecting
existing sources.  The commenter recommended striking the provision that each thinner and cleaning
material contain no organic HAP and instead merely include the HAP content of those materials in with
other coatings used across the affected source.
       Response:  The commenter misunderstood the purpose of this compliance option. This option
is intended for a specific subset of facilities that are not using add-on control devices to comply with the
emission limits and all the coatings they use individually meet the emission limits in 63.4890 of the
proposed rule.  Additionally, since the compliance calculations include thinners and cleaning materials,
this specific subset of facilities also must use thinners and cleaning materials that all contain no organic
HAP.  When these restrictions are met, then we have realized that the compliance demonstration
burden can be  significantly reduced. As an incentive to those facilities that choose to meet the emission
limits through these pollution prevention measures, we have included this less burdensome compliance
demonstration in the proposed rule. The changes requested by the commenter are already included in
                                             2-31

-------
the proposed rule under the heading "Compliance Requirements for the Emission Rate Without Add-



On Controls Option" (see 63.4950 - 63.4952), which contain the emissions averaging provisions



that are necessary when the above restrictions are not met (for example, when a thinner containing



xylene is added to a coating).  Therefore, no changes were made to the final rule in response to this



comment.







       Comment:  One commenter (IV-D-2) said that there is no requirement in 63.4892(a) of the



proposed rule to meet the operating limits in the rule if the affected source chooses the compliant



material option. However, the term "affected source" is not contained in this portion of the rule.



Instead, only the term "coating operation(s)" is used. The commenter suggested clarifying this provision



by adding the term "affected source".



       This commenter also recommended that the term "affected source" should be included in



63.4893(a) to clarify applicability of the provision.



       Response:  Both of these paragraphs are correct as written in the proposed rule. You may use



different compliance options on different coating lines if you so choose (as long as the proper



recordkeeping is performed). Therefore, if you choose the compliant material option on one line, but



not on another, only the coating line for which the compliant material  option is used would not be



required to meet the operating limits and work practice standards. The other coating line would have to



meet the operating limits and work practices.  Thus, the correct terminology is coating line, not affected



source.







       Comment:  One commenter (IV-D-5)  thought that the initial compliance requirements for



existing sources are inadequate to satisfy the requirements of the CAA. The proposed rule would allow



sources to exclude organic HAP present in the products as purchased if they are less than 0.1 percent



by mass for OSHA listed carcinogens and less than 1.0 percent by mass for other compounds. The



CAA explicitly considers emissions "in the aggregate" for determining whether a source  can be



characterized as a major source.  There is no justification for filtering out emissions to determine



compliance when reduction of the emissions taken as a whole is the very purpose of the CAA.




                                             2-32

-------
Accordingly, the reporting requirements allow sources to claim compliance through reducing emissions
reporting requirements, despite the fact that when these emissions are considered in the aggregate the
source may in fact be emitting more than is allowed under the standards.
       Response:  When we developed the MACT floor for the proposed rule, we used information
provided by metal furniture surface coating facilities to determine the organic HAP content of the
coatings, thinners, and cleaning materials these facilities used. Most of the information provided to us
was obtained from material safety data sheets (MSDS) compiled by the manufacturers of the materials.
These MSDS are routinely used to comply with the Occupation Health and Safety Administration
(OSHA) hazard communication requirements of 29 CFR 1910.1200.  These requirements state at 29
CFR1910.1200(d)(5)(ii):
       If a mixture has not been tested as a whole to determine whether the mixture is a health hazard,
       the mixture shall be assumed to present the same health hazards as do the components which
       comprise one percent (by weight or volume) or greater of the mixture, except that the mixture
       shall be assumed to present a carcinogenic hazard if it contains a component in concentrations
       of 0.1 percent or greater which is considered to be a carcinogen under paragraph (d)(4) of this
       section.
       Thus, MSDS generally report carcinogens present at a concentration of 0.1 percent or greater
and other components at 1.0 percent or greater. Since we developed the MACT floor using, in large
part, information from MSDS, the compliance procedure in the proposed rule must necessarily be
consistent with the MACT floor determination. Therefore, the proposed rule excluded carcinogens
present at less than 0.1 percent and other HAP present at less than 1.0  percent.
       Furthermore, when we developed the proposed rule, we looked for opportunities to reduce the
compliance burden on both the regulated industry and the permitting authorities. One such opportunity
was to allow sources to use manufacturer's data (often this is in the form of material safety data sheets
(MSDS)) to obtain relevant information on coatings, thinners, and cleaning materials to perform the
compliance calculations. We allowed this in the proposed rule with the provision that results from EPA
test methods would take precedence if there was  a difference between the manufacturer's data and the
test method results in a compliance action.  When we discussed this option with the industry, they
                                             2-33

-------
pointed out that MSDSs often do not report ingredients present in concentrations less than the OSHA



cutoffs of 0.1 percent by mass for carcinogens and 1.0 by mass for other compounds. It would be



possible, then, that the organic HAP content of coating as listed on the manufacturer's data could be



slightly less than that obtained by a Method 311 test.



       We did not want to create a situation where a source in good faith was using coatings that they



thought had organic HAP contents just below the emission limits as listed on the manufacturer's MSDS,



then have those coatings tested by the permitting authority using Method 311 and find a slightly higher



organic HAP content due to HAP not listed on the MSDS because of its low concentration.  We did



not believe this to be a fair situation when the source was  acting in good faith to comply with the rule.



Therefore, since we expect the majority of the industry to use manufacturer's data for compliance



purposes and typically this will be in the form of MSDSs, we included in the proposed rule the cutoffs



of 0.1/1.0 percent by mass for Method 311  testing.







       2.4.2   Work Practice Standards



       Comment: One commenter (IV-D-2) thought that 63.4893(b)(l) of the proposed rule



regarding work practice standards should be clarified so as to require that organic-HAP-containing



materials be stored in "normally" closed containers. The  current language requires that the materials be



stored in "closed containers" which would be impossible  since the material would have to have some



means of entering the container.



       Response: While we believe the intent of the language in the proposed rule is clear and doubt



there would be any misunderstanding in practice, we have revised this paragraph in the final rule as



follows to assure clarity:



       (1) All organic-HAP-containing coatings, thinners, cleaning materials, and waste materials must



       be stored in closed containers. You must ensure that these containers are kept closed at all



       times except when depositing or removing these materials from the container.







       Comment: One commenter (IV-D-2) stated that 63.4893(b)(2) of the proposed rule requires



spills of organic HAP materials be minimized. The commenter agreed with this provision. This is not




                                             2-34

-------
only environmentally sound, but economically sound as well. However, the commenter requested that

this provision clarify that spilled organic HAP material would not be included in compliance calculations.

Such spills would instead constitute an unusual emission which may require reporting under other CAA

provisions, but should constitute an upset or "malfunction" under which the emission limits would not

apply.

       Response:  The proposed rule already allows you to take into account the organic HAP content

contained in waste material shipped offsite for disposal (see, for example, Equation 1 of 63.4951).  If

the virgin material or waste is not disposed offsite, then we assume it is being used within the affected

source and must then be accounted for in the compliance equations.


       Comment:  One commenter (IV-D-2) stated that 63.4893(c) of the proposed rule provides

that the "U.S. Environmental Protection Agency" may grant permission to use alternative work practice

standards. We agree with and support the concept but, for purposes of consistency and clarification,

the term "Administrator" should be used in place of the name of the Agency.

       Response:  We agree with the commenter and have made changes as appropriate in the final

rule.


2.5    COMPLIANCE DETERMINATIONS

       2.5.1  Support for the Proposal

       Comment:  Two commenters (IV-D-2 and 9) supported the EPA's use of manufacturer's data

as a source of information needed for compliance determinations.  Both commenters also believed that

the Occupational Safety and Health Administration (OSHA) cut-offs used in material safety data

sheets4 (MSDS) are acceptable cut-offs for purposes of evaluating organic HAP content.  The

commenters strongly supported using MSDS in this fashion.  One commenter (IV-D-2) also supported

the proposed methodology for mass balance calculations.
       4 In many cases, MSDS report ingredients only if they exceed the following criteria:  OSHA-
defmed carcinogens as specified in 29 CFR 1910.1200(d)(4) present at 0.1 percent by mass or more;
all other compounds present at 1.0 percent by mass or more.

                                            2-35

-------
       Response:  The final rule retains the use of manufacturer's data for compliance determinations,



the OSHA cut-offs for evaluating organic HAP content, and the mass balance calculations.  However,



the facility retains responsibility should the coating be determined to be non-compliant using the



prescribed test methods.







       Comment:  One commenter (IV-D-2) supported the proposed monthly compliance period for



sources demonstrating compliance based on materials used in the coating operation.



       Response:  The monthly compliance period has been retained in the final rule.







       2.5.2  Compliance During Periods of Startup, Shutdown, and Malfunction



       Comment:  One commenter (IV-D-5) took issue with the provision in the proposed rule that



states compliance with the startup, shutdown, and malfunction plan is considered compliance with the



rule. The commenter believed that the CAA is clear that sources must comply with emission standards



continuously, and only limited exception to that rule is allowed for unavoidable deviations during startup,



shutdown, and malfunctions under a technology-based emission standard.  Allowing sources to comply



to avoid enforcement actions merely by demonstrating that they were in compliance with their own



startup, shutdown, and malfunction plans necessarily allows them to operate at less than continuous



compliance even if their deviation was not unavoidable.  Accordingly, the commenter believed the



proposed startup, shutdown, and malfunction provision does not comply with the CAA.



       Response:  In 63.4900(a)(2) of the proposed rule, we stated that "[a]ny coating operation(s)



for which you use the emission rate with add-on controls option, as  specified in 63.4891(c), must be



in compliance with the applicable emission limit in 63.4890 at all times except during periods of



startup, shutdown, and malfunction."  This provision is often found in NESHAP  in which compliance



with the standards is based solely on the results of a short-term initial performance test and short-term



averaging of continuous monitoring results thereafter.  In consideration of this comment, we realized that



this provision is not appropriate for the surface coating NESHAP when these short-term tests and



monitoring results are only one component of a compliance determination that determines emissions



over a long period of time, which in this case is a month. For the metal furniture surface coating




                                             2-36

-------
NESHAP, the source owner or operator will use the performance test and continuous monitoring



results in combination with data on coatings and other materials used over a month's period of time.



These components will be combined to calculate a monthly organic HAP emission rate.  Since there



may be many startups and shutdowns of a coating operation over the course of a month as part of



normal operation, it is not appropriate to exempt such periods from compliance with the standards. We



believe that a month-long compliance period will accommodate potential short-term higher emission



rates that might occur due to startup, shutdown, or malfunction and that the proposed exemption is not



necessary or appropriate. Therefore, it is not included in the final rule.  Additionally, in order to be



consistent with this change as finalized in 63.4900(a), paragraph (h) of 63.4963 of the proposed rule



(renumbered as 63.4962 in the final rule) was not included in the final rule.







       2.5.3  Compliance Period



       Comment: One commenter (IV-D-6) noted that 63.4968(a)(l) of the proposed rule uses the



term "successive" to mean periods of time that do not overlap, and the end of one period is the



beginning of the next time period.  The next paragraph (63.4968(a)(2)) uses the term "successive" to



define the 3-hour compliance period. This definition rules out the use of a rolling 3-hour compliance



period such that in an 8-hour workday, only 2 compliance determinations may be made. One could



conclude that the 2-hour period at the end of the 8-hour workday is not assessable for compliance



purposes due to only 2 hours being available in the third compliance period in the workday. A facility



could run the process (and control equipment) only in 2-hour blocks that would effectively prevent a



compliance determination from ever being made.  The rule could state that a 3-hour "block" average



would include all data during any part of that period, the almost universal interpretation given in this type



of situation by EPA and other permitting authorities.  However, the commenter recommended deleting



the word "successive" and instead require that "any 3-hour period" could be used.  The latter means a



rolling 3-hour period based on hourly averages computed for each hour based on any available data for



that hour.



       Response: We agree with the commenter that the word "successive" should not have been



used to describe the 3-hour periods in 63.4968(a)(2) of the proposed rule (renumbered as




                                            2-37

-------
63.4967(a)(2) in the final rule). As evidenced in Table 1 to the proposed rule (67 FR 20244), we



meant "any 3-hour period of normal operation" (that is, a rolling 3-hour period). In the final rule, we



have followed the commenter's suggestion and deleted the word "successive" in 63.4967(a)(2). The



paragraph now reads as follows:



       (2) You must determine the average of all recorded readings for each successive 3-hour period
       2.5.4  Monitoring



       Comment:  One commenter (IV-D-5) believed that the CAA requires the EPA to set standards



for the metal furniture surface coating source category based on actual emissions, not just the organic



HAP content of the liquid surface coating materials used.  The monitoring provisions for such standards



must necessarily require emissions monitoring, not just monitoring of HAP content.



       Response:  As we discussed in our response to a related comment in Section 2.3.1 of this



document, one of our basic assumptions in the development of the proposed rule was that all of the



volatile components of the coatings, thinners, and cleaning materials used by a source will evaporate



and be emitted to the atmosphere. Under this assumption, the organic HAP content of the materials is



used to derive the emission rate (kg organic HAP emitted per liter of coating solids used). The volume



of coating solids used is a good measure of the surface area coated or production rate.  Thus,



monitoring of the organic HAP content of the materials used and the volume coating solids used



provides a means of monitoring emissions based on production.







       2.5.5  Miscellaneous Comments



       Comment:  One commenter (IV-D-6) suggested that 63.4967(c)(l) of the proposed rule



(renumbered as 63.4966(c)(l) in the final rule) indicates that only one desorbing gas mass flow



measurement and one carbon bed temperature measurement is taken during the performance test,



either before or after the performance test. The paragraph following indicates that the minimum and



maximum values, respectively, are used as operating limits.  The commenter believed that the proposed



rule was meant to require monitoring and recording of desorbing gas flow and carbon bed temperature




                                            2-38

-------
either before or after each performance test run so that sufficient data can be obtained to establish the
operating limits. This change allows the determination of a minimum or maximum value (as specified in
63.4967(c)(2)) to be used in future compliance determinations.
       Response: The proposed rule is correct as written.  The desorbing gas mass flow is measured
continuously throughout the regeneration cycle, then one measurement of the total mass flow is
recorded. This value then becomes the minimum desorbing gas mass flow for subsequent regeneration
cycles. The proposed rule does not, contrary to the commenter's interpretation, say that only one
measurement is to be taken.
       A single measurement of the carbon bed temperature is recorded after completion of the
regeneration cycle. This recorded temperature value is the maximum achieved during the regeneration
cycle. In all likelihood,  the temperature will be measured continuously because generally the
temperature measuring device is permanently mounted in the carbon bed.  The carbon bed temperature
for subsequent regeneration cycles must be at least as high as the maximum temperature achieved
during the performance test. We believe these two paragraphs are correct as written, and we made no
changes to the final rule in response to this comment.

       Comment: One commenter (IV-D-9) expressed concern over the use of kg organic HAP/liter
of coating solids used as the metric for the emission limit in the proposed rule. The EPA has indicated
that eight out  of the eleven new surface coatings MACT standards that are currently under development
will use this metric. The EPA also indicated that the use of this metric is based on an "equity" issue, and
that it makes the comparison of one technology or formulation to another easier. The commenter was
concerned about this "equity" argument and stated that "if EPA is truly concerned about 'equity' why
do they, in determining the MACT floor, go through such a convoluted process to calculate facility
HAP emission rates by using, in many cases, an arbitrary 'default'  density for conversion between units
of mass and volume."
       The commenter believed the use of a mass-to-mass metric would have eliminated the need to
use an arbitrarily chosen default value to determine the volume coating solids and the use of this
                                            2-39

-------
convoluted process to determine the emission rates.  The commenter suggested that the EPA should



use the mass-to-mass metric for the emission rate.



       The commenter also questioned the use of the two ASTM test methods for the determination of



volume coating solids (ASTM D-2697 and D-6093).  The commenter believed that virtually all of the



data on volume coating solids is based on theoretical formulation values. The two test methods



referenced in the proposed rule are not routinely run (if at all) by manufacturers of metal furniture



coatings. In the preamble, the EPA makes no comment on the viability of these two test methods,



which the commenter has questioned as unrealistic and unreliable for compliance enforcement



purposes.



       Response: We selected the units of "mass of organic HAP per volume of coating solids used"



to normalize the assessment of organic HAP emissions across all affected sources. These units relate



directly to production rates assuming that average dry film coating thicknesses are fairly constant across



all product types. We believe that the use  of mass of coating solids in the denominator of the standard



would penalize operations using lower density pigmented coatings (that is, a lower denominator in the



emission calculation would lead to a higher apparent emissions value), while providing an advantage to



users of higher density coatings.  Therefore, an emission limit based on volume of coating solids was



deemed to be more equitable.



       We did not use an arbitrary default density in the calculation of volume solids. We requested



volume solids data from the industry through questionnaires sent out to a representative selection of the



metal furniture surface coating facilities. Many of the questionnaire recipients provided these data.  For



those that did not provide volume coating solids data, it was calculated from a knowledge of the volume



of volatiles  (assuming that the sum of the volatiles and nonvolatiles volumes is equal to the volume of the



coating).  When the volume of volatiles was not given, it was calculated using the actual average volatile



density as calculated from the densities of the individual volatile components (available from standard



reference sources) and the mass fraction of the volatile components from the data provided in the



questionnaire response (these data were typically from an MSDS).  The data we obtained from MSDS



were often  stated as a range.  When this was the case, we used the mean of the given range in our



calculations. We believe these calculation  methodologies are straightforward engineering calculations




                                             2-40

-------
that are readily apparent from the data we had available.  The commenter's characterization of them as
a "convoluted process" is unwarranted and greatly exaggerated.
       The format of mass of pollutant per volume of coating solids was also used in the new source
performance standards (NSPS) for metal furniture surface coating (see 40 CFR 60, subpart EE) and
several other surface coating NSPS.  A format of mass of pollutant per volume of coating less water
and exempt solvents has been used in many surface coating Control Technique Guidelines (CTG).  For
purposes of consistency with these other requirements, it is most advantageous for source owners and
operators subject to different Federal rules to use similar formats in determining compliance with these
rules. For the most part,  sources in the surface coating categories are accustomed to the volume
coating solids format in EPA regulatory materials, and they have not argued for a change to mass of
coating solids.  For example, our RACT/BACT/LAER Clearinghouse (RBLC) provides a compilation
of determinations made for the level of control considered to be RACT, BACT, and LAER.  We
searched this database for all BACT determinations (those determinations related to new source review
permits) for the period of 1988 to 2002 under the surface coating industry category. For VOC control,
there were 110 determinations that listed the primary units of the permit limits as mass VOC per volume
coating solids.  As is apparent from these data, the surface coating industry has long been familiar with
these units and has broadly incorporated them into their operating permits.
       Consistency in reporting compliance with the various standards to which a source is subject will
also  simplify the enforcement of these NESFLAP. We agree that consistency in format is also helpful to
sources subject to more than one NESFLAP.  Most of the proposed surface coating NESFLAP and
those under development use the volume coating solids format.  For these reasons, we did not change
the format of the emission limits in the final rule.
       Concerning the use of ASTM D-2697 and D-6093 for determining the volume coating solids,
the commenter provided  no data in support of any other alternatives.  In the absence of any data
showing the relationship of volume coating solids measured with the ASTM methods to theoretically
determined values, we have concluded that either means of volume coating solids quantification is
acceptable. Based on this comment, we did realize that there may be situations for some coatings
                                            2-41

-------
where either the ASTM test methods or manufacturer's data cannot be used to determine the volume


fraction coating solids. We added 63.4941(b)(3) to the final rule for this situation:


       (3) Calculation of volume fraction of coating solids. V.. If the volume fraction of coating solids


       cannot be determined using the options in paragraphs (b}(\} and (b}(2} of this section, you must


       determine it using Equation 1 of this section:




                                        TYlvolatiles
                                 V,= \-	                                     Eqn. 1
                                          JJavg



       Where:


       Vs_    =     Volume fraction of coating solids, liters coating solids per liter coating.


                     Total volatile matter content of the coating, including FLAP, volatile organic


                     compounds  (VOC1 water, and exempt compounds, determined according to


                     Method 24 in appendix A of 40 CFR part 60.  grams volatile matter per liter





                     Average density of volatile matter in the coating, grams volatile matter per liter


                     volatile matter, determined from test results using ASTM Method D1475-98.


                     information from the supplier or manufacturer  of the material, or reference


                     sources providing density or specific gravity data for pure materials.  If there is


                     disagreement between ASTM Method D1475-98 test results and other


                     information  sources, the test results will take precedence.





2.6    NOTIFICATION,  RECORDKEEPING, AND REPORTING


       2.6.1  Initial Notification


       Comment: Two commenters (IV-D-2 and 9) requested a change in the implementation of the


proposed rule as  related to section 112(j) of the CAA. In accordance with section 112(j) and the


provisions of 40 CFR 63.53, one commenter has already submitted a Part 1 notification for the affected


source by May 15, 2002. Both commenters requested that the final rule reflect that sources that
                                            2-42

-------
submitted such notifications be exempt from the initial notification requirements of the General



Provisions (40 CFR 63, subpart A) as the notices are redundant.



       Response:  We disagree with the commenter. The requirements of both 40 CFR 63.9 and



63.53 must be met. It is also important to clarify that the requirements in 40 CFR 63.53 address



applications for case-by-case MACT determinations, not notifications, as the commenters stated.



More specifically, the provisions in 40 CFR 63.53 address the application requirements for a source



when the relevant MACT standard has not been promulgated by the date required by the CAA and the



provisions in 40 CFR 63.9 address the notification requirements for a source when it becomes subject



to the relevant standard. Lastly, the content and timing of the initial notifications required by 40 CFR



63.9 differ from the content and timing of the applications required by 40 CFR 63.53.







       Comment:  One commenter (IV-D-2) was concerned about the statement in the preamble to



the proposed rule at page 20212 that companies would be required to "send a notification of planned



construction or reconstruction of a source that would be subject to the rule and apply for approval to



construct or reconstruct."  The commenter believed this statement should be clarified to reflect that such



notice and approval would merely involve the current pre-construction/installation permit program



present in each and every State and that there is no other notification or approval process required for



construction or reconstruction.



       Response:  We disagree with this commenter. The requirements to which the commenter refers



and which are mentioned in the quoted preamble statement can be found in 40 CFR 63.9 which



addresses notification requirements and 63.5(d) which addresses applications for approval of



construction or reconstruction. These requirements from the General Provisions apply to affected



sources under part 63 and both sets of requirements apply to affected sources under this rule as



described in Table 2 to the final rule.



       It is important to note, however, that 40 CFR 63.9(a)(3) does provide that if a State requires a



notice that contains all of the information required in a notification under 40 CFR 63.9, then the owner



or operator may send the Administrator a copy of the notice sent to the State to satisfy the requirements
                                            2-43

-------
of 40 CFR 63.9.  Copies of such notifications would need to be submitted at the times specified in 40



CFR63.9.







       2.6.2  Consolidation and Timing of Reports



       Comment:  One commenter (IV-D-2) was concerned with the requirement of 63.4891 of the



proposed rule that an affected source that switches compliance options must document the switch and



report it in the next  semiannual compliance report. The commenter believed that documenting the



switch is adequate and that reporting it in the next semiannual compliance report is unnecessary and



could be confusing and difficult given that most sources will be submitting semiannual compliance



reports on forms provided by the State. These forms are generally required to be used for purposes of



semiannual reporting under programs implementing part 70 or part 71 permits, and the forms vary



widely from State to State. Most State forms do not contain sections whereby the source is to report



some affirmative action which does not result in a deviation. To avoid this conflict between new



NESHAP provisions and existing title V programs, the commenter suggested deleting this and any other



affirmative statements in the semiannual compliance report as required by the proposed rule.



       Response:  We disagree with the change suggested by the commenter.  In order for the



regulatory authority responsible for compliance to evaluate the semiannual compliance report required



by 63.4920(a), it is imperative that the compliance methods be reported. Many metal furniture



surface coating facilities are very  complex with many coating application and cleaning operations, and



the final rule allows several different methods of compliance for each of these operations. In order for



the regulatory authority to understand how such a facility demonstrated compliance, the compliance



options must be known. Moreover, there is no conflict with this requirement and a title V permitting



program, as a permitting authority must know what compliance options a source is operating under in



order to ascertain whether the source is in compliance with its applicable requirements.







       Comment:  One commenter (IV-D-2) suggested that the rule clearly state that the title V



permitting authority has the authority to set the reporting requirements so as to be consistent with title V
                                            2-44

-------
reporting obligations.  Specifically, the rule should reflect the ability of the permitting authority to



consolidate reporting dates applicable to the facility under the title V permit.



       Two commenters (IV-D-2 and 9) believed that, for purposes of reporting requirements,



particularly reports of startup, shutdown, and malfunction events under 63.4920(c), the rule should be



clarified regarding when reports should be submitted to the State permitting authority versus when



reports must be submitted to the EPA. The commenters suggested the reporting of startup, shutdown,



and malfunction events be reported consistent with the requirements of the applicable title V permit



conditions for "prompt" reporting of deviations. The commenters believed this would make the



reporting timeframes consistent with the title V permit and clarify to whom the reports must be  sent.



       Response:  In terms of consolidating reports, the final rule, consistent with the proposed rule,



allows for an affected source to submit its semiannual  compliance report along with, or as part  of, its 6-



month monitoring report required by 40 CFR part 70 or part 71.  See 63.4920(a)(l)(iv) and (a)(2) of



the final rule.  As a result of the above comments, these two paragraphs have been modified to  clarify



when monitoring reports are required by part 70 or part 71 (that is, every 6 months) and when a 6-



month monitoring report must cross-reference a semiannual compliance report. Moreover, language



was added to 63.4920(a)(l)(iv) of the final rule to ensure that a semiannual compliance report is



submitted within a reasonable time (30 days) after the  end of the semiannual reporting period.



       We disagree with the comment that the rule needs to be clarified as to whether reports,



particularly reports of startup, shutdown, and malfunction events discussed in 63.4920(c) of the final



rule, need to be submitted to a State permitting authority or to EPA.  Consistent with 40 CFR  63.12



(which is listed in Table 2 to the final rule) and 63.4980 of the final rule, whether EPA or a State,



local, or tribal agency should receive reports required under this rule is determined by the delegation



status of the rule. As discussed in 63.4980(a) of the final rule, a source should contact its EPA



Regional Office to find out if implementation and enforcement of the final rule has been delegated to its



State, local, or tribal agency.



       We also disagree with the comment that the rule should be revised so that the reporting of



startup, shutdown, and malfunction events is consistent with the requirements of the applicable title V



permit conditions for "prompt" reporting of deviations. As stated in 40 CFR 70.6(a)(3)(iii)(B), a




                                            2-45

-------
permitting authority is required to define "prompt" in relation to the degree and type of deviation likely
to occur and the applicable requirements.  Therefore, as required by this provision, applicable
requirements, including those found in 63.4920(c) of the final rule and 40 CFR 63.10(d)(5), must be
taken into account by a permitting authority when it defines "prompt." Therefore, it is the responsibility
of the part 70 permitting authority to determine whether the reporting requirements found in
63.4920(c) of the final rule and 40 CFR 63.10(d)(5) are sufficient to meet the permitting authority's
requirements for the prompt reporting of deviations. A permitting authority may decide for a particular
source or source category, or as a general matter, to impose more stringent reporting requirements
(such as type of report, content of report, and/or frequency of submission) than those specified in an
applicable requirement. However, the requirements in 63.4920(c) of the final rule and 40 CFR
63.10(d)(5) are applicable requirements and must be met by  a source which is subject to this rule
regardless of whether they are contained in a title V permit or not.
       It is important to emphasize that a permitting authority does not have the authority to change the
reporting requirements of this rule (such as type of report, content of report, and/or frequency of
submission).  Reporting requirements under this rule are applicable requirements  and sources must
comply with them.

       2.6.3  Semiannual Reports
       Comment: One commenter (IV-D-02) noted that since semiannual reports will be, for the most
part, submitted under a title V permit, the certification requirement for the metal furniture NESHAP
should clearly be the same as the certification requirement for the title V program.
       This same commenter believed that the affirmative reporting requirement in 63.4920(a)(4) of
the proposed rule in which a source is required to report that there were no deviations is unnecessary
and will be problematic for those sources which will be filing these semiannual reports in accordance
with the requirements of their title V permits due to incompatibility with existing State reporting formats.
Moreover, since there is already a requirement under the proposed rule and under title V for sources to
promptly report deviations, there does not appear to be any need to make an affirmative statement that
there were no deviations. If no deviations are reported, then  it is implicit that there were no deviations.
                                            2-46

-------
At a minimum, the commenter suggested clarifying 63.4920(a)(4) to read "the semiannual compliance



report must include a statement that there were no deviations from these emission limitations."



       Section 63.4942 of the proposed rule contains provisions describing how the affected source



must demonstrate compliance with emission limitations.  Section 63.4942(c) requires that the



semiannual report identify coating operations for which the compliant material option was selected.  The



provision goes on to require an affirmative statement of compliance with the emission limitation in



63.4890 and that no thinner or cleaning material contained  organic HAP. The commenter believed



this provision is in conflict with existing State programs and title V semiannual reporting obligations.



The requirement to make an affirmative statement regarding compliance should be deleted.



Alternatively, the commenter provided language indicating that compliance with the State program for



title V reporting constitutes compliance with this reporting obligation.



       Response: We agree with the first comment and 63.4920(a)(3)(ii) of the final rule has been



revised as follows:



               (ii)  Statement by a responsible official with that official's name, title, and signature,



       certifying the truth, accuracy,  and completeness of the content of the  report.  Such certifications



       must also comply with the requirements of 40 CFR 70.5(d^ or 71.5(d).







       We made a corresponding change in 63.4910(c)(2) of the final rule:



               (2)  Statement by a responsible official with that official's name, title, and signature,



       certifying the truth, accuracy,  and completeness of the content of the  report.  Such certifications



       must also comply with the requirements of 40 CFR 70.5(d^ or 71.5(d).







       The added sentence will  ensure that any certification submitted by a source under this rule will



also be consistent with the requirements of part 70 or part 71. Additionally, "of the content" was



deleted from the first sentence as a responsible official needs to certify that the entire submittal is



complete, not just the content of the report.



       We disagree with the commenter that the affirmative statements regarding the absence of



certain deviations required by 63.4920(a)(4) and 63.4942(c) should be deleted due to conflicts with




                                             2-47

-------
existing title V programs. As 6-month monitoring reports are not required by part 70 or part 71 to



contain such affirmative statements, there is no duplication in requiring such statements under this rule.



Such affirmative statements allow a permitting authority to quickly ascertain whether a source has



experienced certain deviations which in turn allows for the more efficient allocation of resources. We



have, however, added language to 63.4920(a)(4) of the final rule to clarify what is required in these



affirmative statements.







       2.6.4  Reporting of Deviations



       Comment:  Two commenters (IV-D-2 and 9) noted that  63.4920(a)(5)(iv) of the proposed



rule requires that the semiannual report contain a "statement of the cause of each deviation." There is a



similar requirement in 63.4920(a)(6)(iii) and in various other areas of the proposed rule.  States have



specific provisions in their title V programs which require and define "prompt" reporting. Moreover,



most State programs have provisions regarding not only the timing of reports of deviations, but also the



contents of such reports. In  some cases, the initial reports of deviations must be submitted so quickly



that the actual cause of the deviation may not be known with certainty.  Rather than set up a conflict



between existing State requirements and the proposed rule, the commenters recommended modifying



the requirement of reporting the cause of the deviation to reporting the "suspected" cause. Not only



would this eliminate conflicts between the proposed rule and various State reporting requirements,  but it



is also consistent with title V in that the stationary source has an obligation to supplement reports with



material additional information.



       Section 63.4963(b) of the proposed rule (renumbered as  63.4962(b) in the final rule) requires



deviation reports to be submitted in accordance with 63.4910(c)(6) and 63.4920(a)(7).  Again, for



purposes of consistency with existing title V programs, the commenters suggested amending that



requirement to provide that, for sources with title V permits, deviation reports must comply with the



reporting provisions of that permit. For sources which do not yet have a title V permit, deviation



reports should be submitted as required by the proposed rule. Section 63.4963 (f) of the proposed rule



(renumbered as 63.4962(f)  in the final rule) contains similar provisions for affected sources choosing



the add-on controls option.  The commenters suggested that the reporting requirements of State




                                            2-48

-------
programs and title V are controlling and that compliance with those obligations is sufficient even if they
only require deviation reporting and do not require affirmative statements of compliance.
       Response: We do not believe that changes to the rule are needed to require sources to report
the "suspected cause" for each deviation versus the "cause" for each deviation.  Reporting based on the
best information that is available at the time is all that is required whether the rule refers to the "cause"or
the "suspected cause" for a deviation. Moreover, certifications of truth, accuracy, and completeness
under 40 CFR 70.5(d) and 71.5(d) are to be based on information and belief formed after reasonable
inquiry. Therefore, there is no conflict between this rule and State title V programs.  We agree with the
commenter, however, that new information regarding the cause of a deviation must be reported to a
permitting authority when it becomes known.
       In terms of the submittal of deviation reports, the final rule establishes the reporting
requirements that must be met by subject affected sources.  These provisions are not superseded by the
requirements of a title V program. As a result, no changes have been made to the rule as a result of this
comment. Moreover, as mentioned earlier in this response to comments document, a permitting
authority does not have the authority to change the reporting requirements of this rule.  Reporting
requirements under this rule are applicable requirements and sources must comply with them.

2.7    MISCELLANEOUS COMMENTS

       Comment: Two commenters (IV-D-2 and 9) believed that some metal furniture manufacturers
may need to install add-on control equipment to comply with the emission limitations set forth in the
proposed rule. Based on the regulatory trigger for the Prevention of Significant Deterioration (PSD)
rules, the installation of the control equipment could be deemed to trigger New Source Review (NSR).
       Under the PSD program, if the source undergoes "any physical change in or change in the
method of operation of a major stationary source that would result in a significant net emissions increase
of any pollutant subject to regulation under the Act," then that could be considered a "major
modification" under 40 CFR 52.21(b)(2)(i). If the add-on control in this case would constitute a
physical change or change in the method of operation, then the company would have to evaluate
                                            2-49

-------
whether or not there was "a significant net emissions increase of any pollutant" subject to regulation
under the PSD program, such as VOC.

       The commenters stated that this situation, combined with EPA's interpretation of the CAA to
allow evaluating "significant net emissions increase" by comparing future potential emissions (that is,
permitted levels) to actual emissions, unnecessarily discourages the use of add-on control equipment.
Therefore, the final rule should specifically exempt add-on control equipment, which is installed for
purposes of complying with the final rule, from NSR and PSD requirements.
       Response: We agree that some sources may choose to install add-on control devices to
achieve compliance with the  emission limits.  Some of these add-on control devices, such as an
oxidizer,  can generate emissions of pollutants that may trigger the need for preconstruction permits
under the nonattainment new source review (NSR) or prevention of significant deterioration (PSD)
program  (referred to as "major NSR").
       In 1992, we adopted  an explicit pollution control project (PCP) exclusion for electric utility
steam generating units (see 57 FR 32314). In a July 1, 1994 guidance memorandum,5 we provided

guidance to permitting authorities on the approvability of PCP exclusions for source categories other
than electric utilities.  In that  guidance, we indicated that add-on controls and fuel switches to less
polluting fuels may qualify for an exclusion from major NSR as a PCP. To be eligible to be excluded
from otherwise applicable major NSR requirements, a PCP must, on balance, be "environmentally
beneficial," and the permitting authority must ensure that the project will not cause or contribute to a
violation of the NAAQS or PSD increment, or adversely affect visibility or other air quality related
values (AQRV) in a Class I area.  The permitting authority must additionally ensure that offsetting
reductions are secured in the  case of a project that would result in a significant increase in a
nonattainment pollutant. The permitting authority can make these determinations outside of the major
NSR process.
       5 Memorandum from Seitz, John, U.S. EPA, July 1, 1994. Pollution Control Projects and
New Source Review (NSR) Applicability. Available at www.epa.gov/ttn/nsr/poly^gui.html.

                                             2-50

-------
       The 1994 guidance did not supercede existing NSR requirements, including approved State



NSR programs, nor void or create an exclusion from any applicable minor source preconstruction



review requirements in an approved State implementation plan (SIP). Any minor NSR permitting



requirements in a SIP would continue to apply, regardless of any exclusion from major NSR that might



be approved for a source under the PCP  exclusion policy.



       We believe the current guidance  on the PCP exclusion adequately provides for the possible



exemption from major NSR for PCP resulting from the final rule. Permitting authorities should follow



that guidance to the extent allowed under the applicable SIP in order to determine whether the



installation of an add-on control device in a given circumstance qualifies as a PCP. Projects that qualify



for the exclusion would be covered under minor source regulations in the applicable SIP, and permitting



authorities would be expected to provide adequate safeguards against NAAQS and increment



violations and adverse impacts on AQRV in Class I areas.  Only in those areas where potential adverse



impacts cannot be resolved through the minor NSR programs or other mechanisms would major NSR



apply.







       Comment:  Two commenters (IV-D-2 and 9) said that according to recent presentations made



by the EPA, there appears to be some discussion regarding using risk analysis up-front in some of the



MACT standards. This may be done through section 112(d)(4) of the CAA or possibly through the



use of delegation to some States with mature risk-based programs. If such provisions are provided for



in other MACT rules, the commenter requested that they be considered for the proposed rule as well.



       Response:  Section 112(d) of the CAA, under which MACT standards are developed, directs



us to base MACT on what is technologically achievable considering cost, energy, and non-air quality



health and environmental impacts. Air quality health impacts are not included in this list of



considerations for determining the technology-based MACT level of control. In some cases, we



evaluate the health risk impact of levels of control more stringent than the MACT floor to help



determine if requiring such levels is warranted. At this time, we are not considering implementing a risk-



based analysis  for alternatives to the technology-based standards in the proposed rule.
                                            2-51

-------
       Comment: One commenter (IV-D-2) noted that the preamble at page 20209 listed the



emission limit for new sources as 0.094 kilograms organic HAP per liter of coating solids used.



However, the limit for existing sources is listed in the preamble as 0.12 kilograms organic HAP/liter



used.  It should be clarified that the existing source limit should be described as 0.12 kilograms organic



HAP/liter of coating solids used. A similar correction should be made to the Ib/gal unit contained in



the parentheses following the metric units.  The correction should state:  1.0 Ib/gal of coating solids



used.



       Response: We thank the commenter for their attention to detail.  While there was some



ambiguity in the text of the preamble, the correct units were used in text  of the proposed rule (see



63.4890 of the proposed rule).







       Comment: Two commenters (IV-D-2  and 9) believed that the EPA has greatly underestimated



the costs of implementing this standard as proposed. For example, the statement on page 20218 of the



preamble to the proposed rule that there will be no incremental costs associated with the use of lower



organic HAP  content coatings and thinners is incorrect. Moreover, EPA states that there will be zero



cost associated with implementing this standard.  This too is inaccurate. This standard represents a



stringent new level of control for new and existing sources.  The cost of complying with these emission



limitations and other requirements will be commensurate with the stringency of the standard, and is



clearly not zero.



       Response: The issue of the cost of coatings had been raised during the development of the



proposed rule. On several occasions we asked the metal furniture surface coating industry, including



the two commenters, for the relative cost difference between typical liquid coatings currently used by



the industry and lower organic HAP content alternatives. It was our intent to take this cost differential



into account in the cost analysis.  Despite our repeated attempts, no cost data were forthcoming. Even



now, these commenters provided no basis for their statement that a higher incremental cost would be



associated with the use of lower organic HAP content coatings. Based on this, we concluded that



generally there was no difference in cost between these two types  of coatings and reflected this in our



cost estimates.




                                             2-52

-------
       We disagree with the commenters' statement that we indicated there would be no cost



associated with implementing the standard. At the same reference to the preamble to the proposed rule



that the commenters cited, we summarized the cost we estimated that individual facilities would likely



incur, as well as an extrapolation of these values to nationwide levels.  These cost estimates took into



account the incremental cost of changing to non-HAP cleaning materials as well as the monitoring,



recordkeeping, and reporting costs. As indicated in the preamble to the proposed rule, these costs



were obviously not zero as indicated by the commenters.







       Comment: One commenter (IV-D-02) suggested that the definitions in 63.4981 of the



proposed rule should be clarified by specifying that the term "Administrator" means the Administrator of



the EPA, as represented by the appropriate Regional Office, or a State or local agency which has been



delegated authority of the NESHAP program (or at least this NESHAP rule).  Alternatively, the EPA



could revise the rule so  as to clarify that all reports must be consistent with the requirements of title V,



which would clarify both timing of the reports and to whom the reports would be sent.



       Response: The meaning of "Administrator" is given in 63.2 of the General Provisions.  The



definitions in the General Provisions were referenced in the proposed rule (see 63.4981).







       Comment: One commenter (IV-D-6) noted that in the preamble to the proposed rule on page



20211, the operating limits for each capture system that is not a permanent total enclosure is described



as including both the volumetric flow rate and pressure drop. Paragraph 63.4967(e) of the proposed



rule (renumbered as 63.4966(e) in the final rule) states that the operating limit is either the gas



volumetric flow rate or  duct static pressure.  The language in Item 6 of Table 1 to the final rule



reinforces this comment.  The commenter requested that the preamble be corrected.



       Response: We appreciate the commenter's attention to detail. The preamble to the proposed



rule incorrectly states that both the volumetric flow rate and the pressure drop must be established as



operating limits for these capture systems. However, the proposed rule correctly states that either



parameter may be used, and this has been retained in the final rule.
                                             2-53

-------
       Comment:  One commenter (IV-D-6) noted that item 6 of Table 1 to the final rule incorrectly



identifies in the "For the following device" column "emission capture system that is a PTE according to



63.4965(a)." A review of the requirement at 63.4967(e) of the proposed rule indicates that the



device is actually an "emission capture system that is not a PTE according to 63.4965(a)."  Table 1



should be corrected.



       Response:  We appreciate the commenter's attention to detail.  The commenter is correct that



the word "not" was inadvertently omitted from this entry. We made the correction in the final rule.



(Note that 63.4965 and 63.4967 were renumbered as 63.4964 and 63.4966, respectively, in the



final rule).







       Comment:  One commenter (IV-D-6) noted that paragraph 63.4930(c)(2) of the proposed rule



uses the abbreviation He to represent the calculated values of the organic HAP content for each coating



and 63.4930(c)(3)(iii) uses the abbreviation HC to represent the value of the mass of HAP emissions



reduction by the emission capture system. The commenter believes this could be confusing (see



Equation 1 of 63.4961 and the nomenclature following, where both a capital "C" and a lower case "c"



are used as subscripts).  A different abbreviation needs to be used for one of these terms.  The



commenter also suggested that in 63.4962(c)(l), the abbreviation, HHAP, should be written H^. and



in 63.4962(c)(2)(i), the abbreviation HC should be written E,..



       Response:  We agree with this commenter that H, was inappropriately used as a variable



representing two different values. In the final rule, we changed H, in Equation 1 of 63.4961 of the



final rule to HR, and changed the references to this variable to HR elsewhere in the final rule. In response



to the commenter's other suggestions, we thoroughly reviewed the proposed rule for typographical



errors and made corrections as appropriate.







       Comment:  One commenter (IV-D-6) noted that the emission  standards for the proposed rule



in 63.4890 are in metric units with the English equivalent in parentheses.  However, the compliance



calculation equations in 63.4941, 63.4951, and 63.4961 only specify metric units. This incongruence



could be interpreted by the affected facility as a requirement to keep all records in only metric units.




                                             2-54

-------
This potential problem may be alleviated by adding a paragraph (d) to 63.4931 allowing records to be



maintained in either English or metric units, as long as the facility is consistent in their recordkeeping



approach.



       Response:  Calculations and recordkeeping may be performed in either metric or English units



so long as the appropriate emission limits in metric units are met. No rule change is needed.







       Comment:  One commenter (IV-D-6) mentioned that 63.4968(a)(5) and (6) of the



proposed rule use the term "associated repairs" but do not define if the association is with repairs for



the capture and control system, or the process equipment.



       Response:  We agree that the language of the proposed rule does not clearly state our intent. In



the final rule, paragraph (a)(5) of this section (which has been renumbered as 63.4967), has been



revised as follows,  and a similar change was made to paragraph (a)(6):







       . . . except during monitoring malfunctions, associated repairs to correct the monitor



       malfunctions, and required quality assurance or control activities  . . .







       Comment:  One commenter (IV-D-6) stated that 63.4968(c)(3)(ii), 63.4968(d)(2),



63.4968(e)(l), and 63.4968(f)(2)(iv) of the proposed rule use the term "sensitivity" to describe the



monitoring device's minimum ability to discriminate between two input signals. Paragraph



63.4968(g)(2)(iii) uses the term "tolerance" to define the same instrument characteristic.  The term



"tolerance" is not appropriate unless it is used to refer to the maximum or minimum monitored value that



will not cause damage to the monitoring device.



       Response:  We agree with the commenter that the terminology was not used appropriately. We



did not mean to differentiate between the two terms. Rather than use two different terms for these



measurements, we  have decided to replace both terms with the single term "accuracy" and include both



a numerical and percentage component. This section (which  has been renumbered as 63.4967 in the



final rule) has been modified to accommodate this change.
                                             2-55

-------
       Comment:  One commenter (IV-D-6) noted that 63.4968(g)(2)(iii) of the proposed rule



implies that there are two different standards, one for mechanical devices (gauge) and one for electronic



devices (transducers).  Previous references to monitoring device sensitivity in 63.4968(c)(3)(ii),



63.4968(d)(2), 63.4968(e)(l), and 63.4968(f)(2)(iv) use a numeric sensitivity and a sensitivity in terms



of a percentage of the recorded parameter.  It is unclear if the EPA is proposing a different "sensitivity"



standard for mechanical and electronic devices.



       Response:  While we recognize differences between and among monitoring approaches, we did



not mean to propose separate standards in this instance. Upon consideration of our approach, we have



decided to replace the term "tolerance" with the term "accuracy" (see the previous comment response)



and include both a numerical and percentage component. This section (which has been renumbered as



63.4967 in the final rule) has been modified to accommodate this change.







       Comment:  Two commenters (IV-D-2 and 9) noted that the preamble to the proposed rule



requested "comments on how monitoring,  recordkeeping, and reporting requirements can be



consolidated for sources that are subject to more than one rule." The commenter believed this poses a



particularly tricky problem because the same stationary  source could be subject to the wood furniture



NESHAP, the metal furniture NSPS,  and individual State permit emission limits (such as RACT,



BACT, and LAER limits), all of which have different units of compliance and different sources which



must be included.  The most efficient way to consolidate monitoring, recordkeeping, and reporting



requirements is to eliminate the need to comply with multiple standards.



       Response:  As we discussed in Section 2.2.1 of this document, we considered allowing you to



comply with only the most stringent surface coating rule applicable to an affected source.  However,



due to practical considerations related to implementing such a compliance option, we have not included



the  option in the final rule at this time. You will be required to comply with all requirements of each rule



applicable to your facility.







       Comment:  One commenter (IV-D-2) noted that 63.4920(a)(5)(i) of the proposed rule



requires that, for purposes of reporting deviations on semiannual compliance reports, the identification




                                             2-56

-------
of "each thinner and cleaning material used that contained organic HAP" is required.  Since the



commenter believed that the requirement that all thinners and cleaners contain no organic HAP should



be stricken, they believe that this corresponding reporting requirement be deleted as well.



       Response: As we responded to a related comment in Section 2.4.1 of this document, this



commenter misunderstood the use of the compliant material option.  Therefore, this reporting



requirement is necessary and has been retained in the final rule.
                                             2-57

-------
                                      TECHNICAL REPORT DATA
                                  (Please read Instructions on reverse before completing)
 1. REPORT NO.
   EPA-453/R-03-002
                                                                      3. RECIPIENT'S ACCESSION NO.
 4. TITLE AND SUBTITLE
 National Emission Standards for Hazardous Air Pollutants (NESHAP):
 Surface Coating of Metal Furniture - Summary of Public Comments and
 Responses on Proposed Rule
                  5. REPORT DATE
                   January 2003
                  6. PERFORMING ORGANIZATION CODE
 7. AUTHOR(S)
                                                                      8. PERFORMING ORGANIZATION REPORT NO.
 9. PERFORMING ORGANIZATION NAME AND ADDRESS
 U.S. Environmental Protection Agency
 Office of Air Quality Planning and Standards
 Coatings and Consumer Products Group (C539-03)
 Research Triangle Park, NC  27711
                                                                      10. PROGRAM ELEMENT NO.
                  11. CONTRACT/GRANT NO.
                                                                      68D01055
 12. SPONSORING AGENCY NAME AND ADDRESS

 Office of Air Quality Planning and Standards
 U.S. Environmental Protection Agency
 Research Triangle Park, NC 27711
                                                                      13. TYPE OF REPORT AND PERIOD COVERED
                  14. SPONSORING AGENCY CODE
                  EPA/200/04
 15. SUPPLEMENTARY NOTES
 16. ABSTRACT
 This document contains a summary of public comments and responses on the proposed NESHAP for Surface
 Coating of Metal Furniture (40 CFR 63, subpart RRRR), proposed on April 24, 2002 (67 FR 20206). This
 document also provides the EPA's response to each comment, and outlines the changes made to the regulation in
 response to public comments.
 17.
                                       KEY WORDS AND DOCUMENT ANALYSIS
                    DESCRIPTORS
                                                    b. IDENTIFIERS/OPEN ENDED TERMS
                                                                                          c. COSATI Field/Group
 Air Pollution
 Hazardous Air Pollutants
 Metal Furniture Surface Coating
Hazardous Air Pollutants
 18. DISTRIBUTION STATEMENT
   Release Unlimited
                                                    19. SECURITY CLASS (Report)
                                                      Unclassified
                                      21. NO. OF PAGES
                                           68
                                                    20. SECURITY CLASS (Page)
                                                      Unclassified
                                                                                          22. PRICE
EPA Form 2220-1 (Rev. 4-77)   PREVIOUS EDITION IS OBSOLETE

-------
United States                                   Office of Air Quality Flaming and Standards                               Publication No. EPA-45X/R-02-XXX
Environmental Protection                        Air Quality Strategies and Standards Division                              October 2002
Agency                                        Research Triangle Park, NC

-------