Office of Transportation                  EPA420-R-06-002
United Stetes    and Air Quality                    January 2006
Environmental Protection	
Agency
         Thermal Imaging
         Cross-Validation
         Program between
         U.S. EPA and
         Briggs & Stratton, Inc.


         Summary of Testing Results

-------
                                               EPA420-R-06-002
                                                  January 2006
Thermal Imaging Cross-Validation Program between
         U.S. EPA and Briggs & Stratton, Inc.

              Summary of Testing Results
                 Assessment and Standards Division
                Office of Transportation and Air Quality
                U.S. Environmental Protection Agency

-------
Executive Summary

US EPA hosted a Stakeholder Technical Workshop regarding emissions control
of small gasoline engines on October 5, 2005.  At that workshop, EPA gave a
detailed presentation regarding the technical work we have done to evaluate the
use  of  catalytic converters for  emissions control.   A  central aspect  of  this
technical work involves evaluating the impacts on safety of adding a catalytic
converter.  Several  participants gave  presentations, including  representatives
from Briggs and Stratton Corporation.1  Briggs & Stratton representatives raised
technical concerns  regarding  the data presented by EPA,  including  specific
questions regarding  the accuracy of  EPA's thermal  images  in light of results
obtained by Briggs & Stratton in their testing.  Following the Technical Workshop,
EPA and Briggs & Stratton  agreed to  conduct a joint Thermal  Imaging Cross-
Validation Program.2 The purpose of this program was clear - to determine if the
thermal imaging equipment used  by EPA and by Briggs and Stratton produced
comparable results.

The  Thermal  Imaging  Cross-Validation   Program  involved  two phases:  (1)
independent third-party validation  of EPA and Briggs & Stratton thermal  imaging
equipment calibrations; and (2) side-by-side comparison of the two organization's
equipment when used to evaluate  the same engine at the same time.

The  Thermal Imaging Cross-Validation Program was carried out from January
11-12, 2006.  EPA and  Briggs & Stratton  engineering staff jointly participated in
the Program.3

This report documents the following results from this Program:

      1) Independent third party validation of EPA's infrared thermal imaging
      equipment and Briggs & Stratton's infrared thermal imaging equipment
      demonstrated  comparable performance between the imagers, and
      demonstrated  the temperature calibrations of the imagers were within
      manufacturers specifications.

      2) Side-by-side comparisons of thermal images taken by EPA's infrared
      thermal  imaging equipment and  by Briggs & Stratton's thermal imaging
      equipment produced comparable and repeatable results when measuring
      surface temperatures from original equipment manufacturer (OEM)
      mufflers and two catalyst muffler configurations during dynamometer tests.
1 All material presented at the October 5, 2005 Technical Workshop is available on line in the
public docket for this rulemaking activity, EPA docket OAR-2004-0008.
 Appendix A to this report contains the letter exchange between EPA and Briggs & Stratton
detailing the scope of the Thermal Image Cross-Validation program.
3 Briggs & Stratton engineering  staff included Bill Latus, Research Manager; Gary Gracyalny,
Research Engineer; and Russ Eberle, Development Technician.

-------
      3) The cross-validation test results showed that the surface temperatures
      of the OEM muffler were considerably hotter than the surface temperature
      of the catalyst muffler over the entire operational range of the two engines
      tested.

      4) Peak temperatures  measured on the catalyst-muffler by both EPA and
      Briggs & Stratton staff during the cross-validation tests were  in good
      agreement with peak temperatures reported in the Technical Workshop for
      an engine of the same  design and approximately the same number of
      hours of use.

      5) Peak temperatures  measured on OEM muffler configurations of both
      engines were higher during the cross validation tests than  those reported
      at the Technical Workshop for an engine of the same design.

      While the Thermal Imaging Cross-Validation Program was not intended to
investigate  the feasibility  of  any future  emissions  standard,  the  results of the
testing also indicate the following:

      1) The application of  a catalyst to  a small gasoline engine does  not
      increase, and can actually lower, exhaust system surface  temperatures,
      both  where the  base  engine is in  compliance with the current federal
      Phase 2  emissions standards and where the  base engine exceeds the
      Phase 2 standards.
      2) The application of a catalyst to a base engine in compliance with the
      current Phase 2 standards reduces emissions to a level that was 30
      percent below the California Tier 3 NOx+HC emissions standard.
The report contains six Sections;
  I.  Background
 II.  Validation of Temperature Calibration
 III.  Engine Dynamometer Testing at U.S. EPA-NVFEL
 IV.  Comparison between Data Collected in the Thermal Imaging Cross-
     Validation program and Data presented by EPA at the October 5, 2005
     Technical Workshop
 V.  Conclusions
 VI.  Appendixes

-------
/.  Background
At the request of Briggs & Stratton, Inc., U.S.  Environmental Protection Agency
(EPA) participated in joint testing with  Briggs & Stratton to cross-validate infrared
(IR) thermal imaging instruments. These instruments are used by both EPA and
Briggs  & Stratton engineering staff to  acquire surface temperature  data  from
engine and exhaust system surfaces.  On December 23, a final test plan for the
cross-validation was agreed upon by EPA and Briggs & Stratton. The cross-
validation consisted of two types of testing:

    1. Validation  of  the calibration  of each instrument relative to National
       Institute of Science  and Technology (NIST) traceable,  temperature-
      controlled near-black-bodies by an independent third party laboratory.
    2. Joint testing at the EPA National Vehicle and Fuel Emissions Laboratory
      (NVFEL) engine  dynamometer test facility for measurement of surface
      temperatures of  different  exhaust system  configurations over a broad
      range of engine operating conditions.  The tested engine and exhaust
      system configurations included:
          a.  a high-hour engine (~125 hours run-time prior to testing) with OEM
             "box-style" muffler
          b.  a high-hour engine (~125 hours) with high-hour catalyst-muffler
             (~135 hours of catalyst run-time) provided by EPA
          c.  a low-hour engine with OEM "box-style" muffler
          d.  a low-hour engine with low-hour catalyst muffler provided by Briggs
             & Stratton
EPA currently uses Infrared Solutions,  Inc.  "IR Snapshot" and "IR Flexcam  T"
imagers.  Briggs & Stratton uses a FLIR Systems, Inc. imager.

The two engines tested were similar Briggs & Stratton Quantum side-valve Class
I, Phase 2 engines (EPA engines 257 and 258) used for walk-behind lawn mower
applications.   Both catalyst mufflers  were a modified version of the  Briggs &
Stratton European catalyst-muffler with doubled catalyst substrate volume.  The
OEM muffler tested on both engines was one of two basic muffler types sold  by
Briggs & Stratton with the Quantum engine for various applications.

The high-hour catalyst muffler was operated in the field  for approximately 110
hours, with the remaining hours accumulated during dynamometer testing.  The
high-hour catalyst was previously tested by EPA using a field-aged engine (EPA
engine # 6820).  The field-aged engine was disassembled approximately 1  year
ago for inspection and wear and deposit rating, and thus was no longer available
for testing.  As a result,  a new engine (engine 257) was operated  on  an  engine
dynamometer  over a ramped-modal G2 six-mode cycle for approximately 110
hours in order to provide an engine for testing with the high-hour catalyst muffler.
When combined with the hours of  testing for emissions  and development, the
total number of hours of operation on engine 257 was approximately 125.

-------
Briggs and Stratton provided a similar catalyst-muffler for testing.  This catalyst
muffler had relatively low-hours of operation (approximately 10).  EPA provided
an engine with a similar number of hours of operation for testing with this catalyst
muffler (engine 258).

//.  Validation of Temperature Calibration
 EPA and Briggs & Stratton, Inc. agreed to have the temperature calibrations of
the infrared imagers validated by Infrared Solutions,  Inc.  in Plymouth, Minnesota.
The choice of Infrared Solutions was based on
      their  maintenance  of  NIST-traceable  near-black-body   temperature
       standards in the temperature range of interest (ambient to 1000 C)
      their general familiarity with the IR imagers participating in the validation
      they were a neutral third party
      their relatively close proximity to the Briggs & Stratton and EPA facilities

Engineering staff from EPA and  Briggs & Stratton met at the  Infrared Solutions
facility on January 11, 2006 to observe the validation testing.  Infrared Solutions
used  an  internal acceptance  test  procedure  (Procedure  # WI-05101,  see
Appendix B) to validate the calibration of the "IR Flexcam T" and FLIR imagers at
5, 100, 350 and 600 C  0.1 C.  Infrared Solutions also used a slightly different
automated acceptance test procedure to validate the "IR Snapshot" imager from
5 to 1200 C.  Results for the acceptance  test procedures are presented in
Tables 1  and 2.
Table 1: Summary of results for the validation of temperature calibrations for the "FlexCamT"
and "FLIR" imagers. Both imagers were adjusted to account for the emissivity of the temperature
targets and an ambient temperature of 25 C. Both  imagers passed the acceptance criteria of
2% of point.
                           EPA IR Flexcam T
                                Point
                             Temperature
Emissivity of
Temperature
   Target
    0.98
    0.93
    0.97
    0.93
   Target
Temperature
    (C)
     5
    100
    350
    600
                                 4.6
                                99.1
                                351.8
                                590.1
                                                 Briggs & Stratton FLIR
   Point
Temperature
                                                               Average
                                                             Temperature
    4.9
    102
    350
    602
                                                                 5.6
                                                                 101.6
                                                                 351.5
                                                                 601.6

-------
Table 2:  Summary of results for the validation  of temperature calibration for the EPA "IR
Snapshot" imager. The imager passed the acceptance criteria of  2% of point and was issued a
new calibration certificate.
Emissivity of
Temperature
Target
0.98
0.98
0.98
0.98
0.98
0.98
0.98
0.98
0.98
0.98
0.98
0.98
0.98
0.98
0.98
0.98
Target
Temperature
(C)
5
20
37
50
75
100
240
300
350
600
700
800
900
1000
1100
1200
Average
Temperature
(C)
3.51
19.57
36.44
49.19
74.18
98.99
239.8
301.85
350.88
594.65
694.06
793.47
901.97
986.42
1091.14
1192.84
///.  Engine Dynamometer Testing at U.S. EPA-NVFEL
Engineering staff from EPA and Briggs & Stratton met at the EPA-NVFEL facility
on January 12, 2006.  Testing was to be conducted on the high hour Briggs and
Stratton Quantum engine (engine 257) with the OEM muffler and the high-hour
catalyst-muffler first.  Afterwards engine 258, the  low-hour Briggs and Stratton
Quantum engine,  would be setup on the dynamometer for additional testing of
the OEM muffler and the Briggs & Stratton supplied low-hour catalyst-muffler.

Immediately prior to the start of testing, Briggs  & Stratton  engineering staff
requested installation  of K-type  thermocouples onto the surfaces of the OEM
muffler  and  high-hour catalyst-muffler  via  brass brazing  of welded bare
thermocouple wire.   Two such  thermocouples were  already  installed on  the
surface of the low-hour catalyst muffler supplied by Briggs & Stratton.  Concerns
were raised by  EPA staff that modification of the muffler and catalyst muffler in
this  manner would  change  the  thermal  conductivity  and  heat  rejection
characteristics of the exhaust components near the thermocouple and brazing.
After some discussion, EPA agreed to this request.

Prior to testing, all exhaust system surfaces were painted with a flat-black high-
temperature paint to give a consistent,  dull surface finish across the measured
surfaces.    During  collection  of IR  thermal   image  data, an emissivity  of
approximately 0.90 (dull painted finish, steel) was used.

-------
Testing for both engines was conducted using the EPA A-cycle test, which is the
test cycle  used for regulatory compliance with emissions standards for this type
of engine.  The A-cycle test is a 6-mode steady-state test cycle  that provides
engine loading covering the entire operational range of the engine.   Infrared
emissions were measured following stabilization of cylinder head temperature to
a value of approximately:

      AT/At < 1 C/minute

Note  that  this was  a considerably  more stringent  stabilization  than  the
stabilization  criteria established in 40 CFR   90.409 for  emissions testing.
Stabilization for both engines required approximately 8 minutes of operation in
mode 1  and  approximately 5 minutes of  operation in  modes 2 through 6.
Following  stabilization of cylinder head temperatures, IR thermal  images were
simultaneously acquired using the "IR  Snapshot" and "FLIR" imagers.  Exhaust
emissions were also measured using both continuous-dilute-sampling and dilute-
bag-sampling from a constant volume sampling system.

Engine Dynamometer Test Results
Exhaust emissions from the tested engine configurations  are  summarized in
Table 3.   The NOx+HC emissions of engine 257 at approximately 125  hours
were 1.6 g/kW-hr above the Phase 2 emission standard to which the engine was
certified.  In this case, the engine appeared to be running an excessively rich air-
to-fuel ratio.  EPA has  previously accumulated engine hours via field operation
(field aging) for four engines in this engine family, and via engine operation on a
dynamometer  (dynamometer aging)  for three  engines in  this engine  family.
Following  aging to near 125 hours,  EPA testing showed only one engine out of
eiqht  (EPA  engine  6820) achieved  NOx+HC  emissions below the  Phase 2
emissions standards.   Depending  on the  engine, this  appears to be due to
excessively  lean operation, excessive rich operation (as with engine 257), or
excessive oil consumption.   The overly rich air-to-fuel ratio reduced  NOx+HC
catalyst  efficiency to approximately 22% from its previous  performance of 35-
40% using a similar engine that met the Phase 2  standards at high hours (EPA
engine 6820).  The excessively rich air-to-fuel ratio reduced catalyst efficiency
both by limiting available  oxygen for  HC oxidation and by overwhelming the
catalyst with excess HC.

Catalyst performance with engine 258 was much closer to previous low and high
hour emission  results on a percent emissions reduction basis (~37% reduction in
NOx+HC).  Engine-out emissions of engine 258 were consistent with low-hour
emissions  results from engines  previously  tested  by EPA  using  Federal
Certification  Fuel.   NOx+HC emissions for  engine 258 were approximately 6%
higher than the  low-hour  emissions  test  results  submitted  by the engine
manufacturer for certification of this engine family.

-------
Table 3: Summary of A-cycle emissions results. A total of 4 tests were conducted (no repeats)
 NOx (g/kW-hr)
 HC (g/kW-hr)
 NMHC (g/kW-hr)
 CO (g/kW-hr)
 CO2 (g/kW-hr)

 NOx+HC (g/kW-hr)
 NOx+NMHC (g/kW-hr)
 BSFC (g/kW-hr)
Engine 257
OEM Muffler
High-hours
2.10
15.63
13.44
577
1217
Engine 257
EPA Catalyst
Muffler
High-hours
1.41
12.44
10.37
491
1188
Engine 258
OEM Muffler
Low-hours
2.48
8.34
6.89
485
1243
Engine 258
Catalyst
Muffler
Low-hours
1.18
5.88
4.58
380
1235
17.73
15.54
 593
13.84
11.78
 546
10.82
 9.37
 554
7.07
5.76
505
Infrared thermal images for all of the tested configurations and for each of the six
modes of the A-cycle test are shown in Figures 1-12.   Peak  temperatures
measured from the thermal images are compared in  Figures 13 and 14.  The
effect of brazing the thermocouple onto the OEM muffler can be clearly seen in
the large gradient of reduced  temperature in the upper right corner of the OEM
muffler, particularly during A-cycle modes 1-3.  The large  temperature gradient
increased the difficulty  in obtaining  a point temperature  near the  estimated
position of the thermocouple junction.  Within the software for both imagers, a set
of "cross-hairs" can be positioned to allow measurement of point temperatures,
such as that of the thermocouple junction.  Moving the "cross-hairs" even a few
pixels in the presence of such  a large temperature gradient would result in a fairly
large  change in temperature for the measured  point.   The Briggs and Stratton
engineers considered the point temperatures  measured  by  both  of  the  IR
imagers to be in good general agreement with the temperature data collected
from the K-type thermocouples.

The thermal images from the "FLIR" imager used by Briggs and Stratton and the
"IR Snapshot" imager  used by EPA gave  peak temperature measurements that
were  within the 2% tolerance for the instruments for every tested mode and
condition with the sole exception of mode  1 with the low-hour engine (#258) and
catalyst-muffler (that  is,  EPA  and  Briggs  &  Stratton  peak temperature
measurements  "overlapped"  when  including  the 2%  tolerances,  with  the
exception of the one mode).   In  that case, the difference appears to have been
due to a slightly different camera angle used with the "IR Snapshot" imager that
brought more  of the surface  near  the exhaust outlet into the  IR image of the
catalyst muffler surface.

Although the measurements  compared  well within their respective accuracy
specifications, there was a consistent trend of peak temperatures from  the "FLIR"
imager  being approximately  12  to  20  C higher  than  peak  temperatures
measured using the "IR Snapshot" imager. This was probably due to  the higher
image resolution available from the "FLIR"  imager.
                                                                       8

-------
For all of the tested configurations and all of the operating conditions, the peak
temperatures of the OEM muffler were significantly  higher  than  those of  the
tested catalyst-mufflers (see  Figures 15  and 16).   The heat-affected, high
temperature area  adjacent to the areas of peak  temperature also covered a
larger surface area of the OEM muffler relative to that of the catalyst-muffler.
This is particularly apparent for A-cycle modes 1-4 (see Figures 1-4 and 7-10).
This was most likely due to differences in how heat is rejected from the surfaces
of the two exhaust systems. The catalyst-muffler was installed in an area of the
engine where much of its  surface area,  particularly surface area adjacent to  the
catalyst substrate,  was swept with cooling air from  the engine cooling fan as  the
cooling air exited from the  cooling fins on the engine's cylinder.  The OEM muffler
was located largely out of the direct path of airflow exiting the engine.   The
catalyst muffler also had a larger surface area to reject heat across, and routed
the exhaust in  a  manner  that lengthened the path of flow and provided more
internal surface area to aid heat rejection.

-------
           B&S Data
        EPA Data
                                   OEM
                                  Muffler
                     -400


                     -300


                     - 200
  Peak Temperature: 533.2 C
Peak Temperature: 511.1 C
                                  Catalyst
                                  Muffler
   Peak Temperature: 444.0 C                Peak Temperature: 470.6 C
Figure 1:  Comparison of surface temperature measurements by Briggs and Stratton (left) and
EPA (right) of engine 258 operated during A-cycle mode 1 (100% load, WOT) with the small, box-
style OEM muffler (top) and a European catalyst-muffler with doubled substrate volume (bottom)
at low hours (<15 hours).
          B&S Data
        EPA Data
                                  OEM
                                 Muffler
                     -400


                     -300


                     -200


                     -100
  Peak Temperature: 466.7C
Peak Temperature: 446.6C
                                 Catalyst
                                  Muffler
   Peak Temperature: 399.5C                 Peak Temperature: 377.8C
Figure 2:  Comparison of surface temperature measurements by Briggs and Stratton (left) and
EPA (right) of engine 258 operated during A-cycle mode 2 (75% load) with the OEM muffler (top)
and catalyst-muffler (bottom) at low hours.
                                                                                  10

-------
           B&S Data
        EPA Data

                                   OEM
                                  Muffler
                           I
    Peak Temperature: 431  C
Peak Temperature: 412.5C
                                  Catalyst
                                   Muffler
    Peak Temperature: 367.5 C                Peak Temperature: 350.8C
Figure 3:  Comparison of surface temperature measurements by Briggs and Stratton (left) and
EPA (right) of engine 258 operated during A-cycle mode 3 (50% load) with the OEM muffler (top)
and catalyst-muffler (bottom) at low hours.
         B&S Data
      EPA Data
                                 OEM
                                Muffler
 Peak Temperature: 408.7C
                                Catalyst
                                 Muffler
  Peak Temperature: 349.8C                   Peak Temperature: 335.1 C
Figure 4:  Comparison of surface temperature measurements by Briggs and Stratton (left) and
EPA (right) of engine 258 operated during A-cycle mode 4 (25% load) with the OEM muffler (top)
and catalyst-muffler (bottom) at low hours.
Peak Temperature: 389
                                                                                 11

-------
           B&S Data
          EPA Data
                                   OEM
                                  Muffler
                       300


                       200
   Peak Temperature: 416.9C
 Peak Temperature: 396.9C
                                  Catalyst   p,
                                   Muffler
    Peak Temperature: 350.4C                 Peak Temperature: 332.4C
Figure 5:  Comparison of surface temperature measurements by Briggs and Stratton (left) and
EPA (right) of engine 258 operated during A-cycle mode 5 (10% load) with the OEM muffler (top)
and catalyst-muffler (bottom) at low hours.
         B&S Data
        EPA Data
                                 OEM
                                Muffler
 Peak Temperature: 422.5C
Peak Temperature: 401.1 C
                     IT 600.0
                                Catalyst
                                 Muffler
  Peak Temperature: 344.4C                 Peak Temperature: 331.6C
Figure 6:  Comparison of surface temperature measurements by Briggs and Stratton (left) and
EPA (right) of engine 258 operated during A-cycle mode 6 (high-speed-idle) with the OEM muffler
(top) and catalyst-muffler (bottom) at low hours.
                                                                                  12

-------
           B&S Data
        EPA Data
                                   OEM
                                  Muffler
   Peak Temperature: 538.0C
 Peak Temperature: 522.0C
                                  Catalyst
                                  Muffler
   Peak Temperature: 419.9C                  Peak Temperature: 401.6C
Figure 7:  Comparison of surface temperature measurements by Briggs and Stratton (left) and
EPA (right) of engine 257 operated during A-cycle mode 1 (100% load, WOT) with the small, box-
style OEM muffler (top) and a European catalyst-muffler with doubled substrate volume (bottom)
at high hours (>120 hours).
           B&S Data
        EPA Data
                                   OEM
                                  Muffler
                     -400


                     -300 '


                     -200


                     -100

                     25.0
   Peak Temperature: 453.4C
Peak Temperature: 440.7C
                                  Catalyst
                                  Muffler
   Peak Temperature: 365.1 C                 Peak Temperature: 351.5C
Figure 8:  Comparison of surface temperature measurements by Briggs and Stratton (left) and
EPA (right) of engine 257 operated during A-cycle mode 2 (75% load) with the OEM muffler (top)
and catalyst-muffler (bottom) at high hours.
                                                                                  13

-------
           B&S Data
        EPA Data
                                  OEM
                                  Muffler
   Peak Temperature: 415.5C
Peak Temperature: 402.8C
                                 Catalyst
                                  Muffler
   Peak Temperature: 330.3C                 Peak Temperature: 317.5C
Figure 9:  Comparison of surface temperature measurements by Briggs and Stratton (left) and
EPA (right) of engine 257 operated during A-cycle mode 3 (50% load) with the OEM muffler (top)
and catalyst-muffler (bottom) at high  hours.
           B&S Data
        EPA Data
                                  OEM
                                  Muffler
   Peak Temperature: 394.3C
                                 Catalyst
                                  Muffler
   Peak Temperature: 318.8C                 Peak Temperature: 307.7C
Figure 10:  Comparison of surface temperature measurements by Briggs and Stratton (left) and
EPA (right) of engine 257 operated during A-cycle mode 4 (25% load) with the OEM muffler (top)
and catalyst-muffler (bottom) at high hours.
Peak Temperature: 381
                                                                                 14

-------
           B&S Data
        EPA Data
                                  OEM
                                  Muffler
   Peak Temperature: 396.8C
                                 Catalyst
                                  Muffler
   Peak Temperature: 327.9C                 Peak Temperature: 316.6C
Figure 11:  Comparison of surface temperature measurements by Briggs and Stratton (left) and
EPA (right) of engine 257 operated during A-cycle mode 5 (10% load) with the OEM muffler (top)
and catalyst-muffler (bottom) at high hours.
Peak Temperature: 381
           B&S Data
        EPA Data
                                  OEM
                                  Muffler
   Peak Temperature: 404.6C
Peak Temperature: 390.6C
                                 Catalyst
                                  Muffler
   Peak Temperature: 318.0C                 Peak Temperature: 306.7C
Figure 12:  Comparison of surface temperature measurements by Briggs and Stratton (left) and
EPA (right) of engine 257 operated during A-cycle mode 6 (high-speed-idle)  with the OEM muffler
(top) and catalyst-muffler (bottom) at high hours.
                                                                                 15

-------

0*400
o
1-
D
3 300
i/>
s
s
200
mn

I
I
i

i

i

i I J
I
i  ;

+ B&SFLIR
| B&S Fl IP
EPAISIR
EPAISIR
IR Imaaer, OEM Muffler
IR Imager, Catalvst-Muffl
Imaaer, OEM Muffler
Imaaer Catalyst-Muffler


       1            23456
                              A-Cycle Test Mode #

Figure  13:   Comparison  of  peak  surface  temperatures  from the  IR thermal images for
measurements of engine 258 operated during A-cycle mode 6 (high-speed-idle) with the OEM
muffler (top) and catalyst-muffler (bottom) at low hours. The error bars represent 2% of absolute
temperature measurement accuracy.  Note that the peak surface temperature of the OEM muffler
is significantly hotter than the catalyst muffler for all of the tested conditions.
    600
    500
  O 400
  i/>
  8
    300
    200
    100
+  B&S FLIR IR Imaaer, OEM Muffler

|  B&S FLIR IR Imaaer, Catalyst-Muffler
    EPA IS IR Imaaer, OEM Muffler

    EPA IS IR Imaaer Catalyst-Muffler
       123456
                              A-Cycle Test Mode #
Figure  14:   Comparison  of  peak  surface  temperatures  from the  IR  thermal images for
measurements of engine 257 operated during A-cycle mode 6 (high-speed-idle) with the OEM
muffler  (top)  and catalyst-muffler (bottom)  at high hours.  The  error bars represent  2%  of
absolute temperature  measurement accuracy.  Note that the  peak surface temperature of the
OEM muffler is significantly hotter than the catalyst muffler for all of the tested conditions.
                                                                                     16

-------
Muffler
O

b
E ,r
s
>.
S i
re
u
O nr
rom muffler t
J N
1 C
emperature f
i it
1 U
Change in 1
I o





1
Catalyst
hotter

*B&S FLIR IR
 EPA IS IR



23456
A-Cycle Mode #


s hotter
Figure 15:  Change in peak surface temperature between the OEM muffler and  the catalyst
muffler for engine 258 at low hours. The OEM muffler was approximately 40 to 90 C hotter for
the tested conditions over the A-cycle. The "FLIR" imager used by Briggs and Stratton generally
showed a larger temperature increase for the OEM muffler relative to the catalyst-muffler than the
"IR Snapshot" imager used by EPA.

   Muffler is hotter
-muffler (AT, C)
a [j
V)
ire from muffler to catal
itj M
01 01
"re
ige in Tempe
ij
Ol
re
O
-195
| .











I
*B&S FLIR IR
EPA IS IR
* 




2345
A-Cycle Mode






    Catalyst is hotter

Figure 16:  Change in peak surface temperature between the OEM muffler and  the catalyst
muffler for engine 257 at high hours. The OEM muffler was approximately 70 to 120  C hotter for
the tested conditions over the A-cycle.
                                                                                    17

-------
IV.  Comparison between Data Collected in the Thermal Imaging
Cross-Validation program  and Data  presented  by EPA  at the
October 5, 2005 Technical Workshop

At the October  5, 2005 Technical  Workshop, EPA's presentation included
emissions and thermal imaging data on a Briggs & Stratton side-valve Quantum
engine, EPA test engine 6820.  A copy of this presentation material is included in
Appendix C of this report.

      Emissions  Result  Comparison:  October 5  Technical Workshop and
January Thermal Imaging Cross-Validation Program

EPA test  engine  257  was included  in this Thermal  Imaging Cross-validation
program in order to accommodate Briggs & Stratton's request to  bring a their
own  catalyst-designed muffler, which was  designed  for a  Briggs &  Stratton
Quantum engine, which engine 257 is.  Engine 6820, also a Quantum engine,
was not available for repeat testing during the Thermal Imaging Cross-Validation
Program  as it was completely disassembled for inspection,  wear and deposit
rating and comparison to other high-hour engines at the request of Briggs &
Stratton early in 2005.

The emissions performance of  engines 6820 and engine 257, without a catalyst,
are significantly  different, as are  other engine performance characteristics.  At
high-hours, engine 6820's  emissions  (without catalyst) were  15.5  g/kw-hr
NOx+HC,  while engine 257's  emissions  (without catalyst)  were 17.7 g/kw-hr
NOx+HC.4 The EPA Phase 2  NOx+HC standard for this class of engine is 16.1
g/kW-hr.  Therefore, engine 6820 met the existing Phase 2 standard while engine
257 does  not.  Engine 6820's  engine-out emissions performance at high hours
was approximately 4 percent below the Phase 2 standard, while engine 257 is
approximately  10  percent above the standard.    Engine   257  exhibited a
significantly richer air to fuel ratio,  as indicated by its higher CO emissions.

The  emissions   performance  of  engine 6820 with  a  catalyst resulted  in
approximately a  39 percent reduction in NOx+HC emissions, while the emissions
performance of engine  257 with a catalyst resulted in approximately a 22 percent
reduction in NOx+HC emissions.

Engine 6820, when equipped with a prototype catalyst, achieved the California
Tier  3 standard of 10.0 g/kw-hr  NOx+HC standard.   The EPA  developed
prototype catalyst used on both engines was designed specifically for an engine
which was capable of achieving emissions at or below  the  existing Phase 2
standard.  The  catalysts design  (e.g., size, location  in the  muffler, etc.) was
selected to reduce NOx+HC emissions from  an engine capable of achieving the
  See Appendix C, page 33.


                                                                     18

-------
Phase 2 standard and operating in a specific engine-out air-fuel ratio range as a
function of engine load.   The emissions  performance of engine 6820  with a
catalyst resulted in approximately  a 39 percent reduction in NOx+HC emissions,
achieving a level that complies with the California Tier 3 standard.

Engine 257 had (1) higher engine  out-emissions than engine 6820, in fact above
the Phase 2 emission standard, and (2) a richer air-fuel ratio, as indicated by the
higher CO emissions.  When the catalyst used on engine 6820  was put on
engine 257,  the result was a 22 percent reduction in emissions, and a level that
did not achieve the California Phase 3 standard.  The catalyst achieved a lower
percent reduction on engine 257 because the catalyst was designed to operate
on an engine that had lower engine-out emissions, and a leaner air-fuel ratio with
a  higher oxygen content (conditions  typically expected from an  engine that
complies with the current Phase 2  emissions standards).5

The testing of engine 6820 indicates that the combination of a properly designed
catalyst and an engine capable of complying with the existing Phase 2 engine
standard can result in a design which meets the California Tier 3 standard (i.e.,
10 g/kw-hr).  The testing  of engine 257 indicates that using a  catalyst designed
for a  compliant Phase 2 engine  on an engine that does not comply with  the
current Phase 2 standards does not necessarily result in a design which meets
the California Tier 3 standard.
      Thermal Image Result Comparison:  October 5 Technical Workshop and
January Thermal Imaging Cross-Validation Program

The  results of the Thermal Imaging Cross-Validation Program indicate that for
the  Briggs  &  Stratton  Quantum  engine,  EPA  test  engine  257,  surface
temperatures for the catalyst-equipped engine were lower for all test operating
modes than the base engine without a catalyst.  This lower surface temperature
was  true for  both peak temperature  and average surface temperature.   See
Figures 7-12, 14 and 16.  These results are consistent with the results shown by
EPA at the October 5 Technical Workshop for engine 6820.  See pages 37 - 41
of Appendix C.   For both engine 257 and 6820, the testing presented at the
October 5 Technical Workshop and the testing preformed in the Cross-Validation
Program  indicate that  adding a catalyst for  emissions  performance does not
increase but can lower surface temperatures for this engine model. The catalyst
lowered temperatures more for engine 257 than for engine 6820.
  Note that even if the catalyst had achieved a 39% reduction on engine 257, as it did on engine
6820, engine 257 would still not achieve the California Phase 3 standards because its engine out
emissions were significantly higher than allowed under EPA's Phase 2 standards.
                                                                       19

-------
V.  Conclusions
Engineering staff from both EPA and Briggs & Stratton met on January 13, 2006
and agreed that:

   1.  Both the EPA and Briggs & Stratton were successfully validated versus
      MIST traceable standards.
   2.  Comparable  peak  temperatures  and  temperature  gradients  across
      exhaust system  surfaces  were acquired  by  both EPA and  Briggs  &
      Stratton using the procedures outlined in this report.

There remained differences of opinion between  EPA staff and Briggs & Stratton
staff regarding thermal images acquired by  EPA in  December 2004-January
2005 using a high-hour engine (EPA engine  #  6820) with the  same  high-hour
catalyst muffler used within  the cross-validation with  engine  257.   Briggs  &
Stratton staff was concerned about differences in the heat affected areas on both
the catalyst-muffler and the OEM muffler relative to those measured during the
cross-validation.  EPA staff agreed that the  heat affected areas  do appear to be
different; however the peak temperatures originally measured for engine 6820
with  the catalyst  muffler at  high  hours  were  generally  comparable to those
measured with engine 257 during the cross-validation. The bigger differences
were with the peak temperatures measured using the OEM muffler.  The peak
temperatures  measured with  both  engines 257  and 258 showed that the OEM
muffler configuration  was considerably hotter than comparable measurements
when engine 6820 was  equipped with a similar OEM muffler. The end result is a
even larger difference  in temperature between the OEM muffler and catalyst
muffler configurations, with the OEM muffler showing up as considerably hotter in
the more recent infrared thermal images taken by both Briggs and Stratton  and
by EPA.

The differences seen between the  IR images from one year ago and those from
the cross-validation testing are primarily due to the difference in the performance
of the  base-engines.    As  discussed in this  report, the EPA engine 6820's
emission performance  at high-hours was much lower than  the  more recently
tested  engine  257.   Engine  differences  which could result in these different
emissions performance  may include the difference in air-to-fuel ratio and possibly
differences in spark timing.
                                                                      20

-------