Emission Laboratory Correlation Study Between EPA and Honda Motor Company, Inc. by Richard E. Lowery April 1974 Environmental Protection Agency Office of Air Programs Office of Mobile Source Air Pollution Control Emission Control Technology Division Procedures Development Branch Ann Arbor, Michigan ------- ABSTRACT This report presents the results of a laboratory correlation study between EPA and the Honda Motor Company, Inc. It specifically compares Honda's certification test site at its Suzuka laboratory to EPA's certification test site #6. In general, the test sites compared in this study showed a high degree of equivalency. No significant differences were dis- covered in dynamometer or CVS characteristics, and vehicle emission test results correlated rather well. The only significant discre- pancy was laboratory measurement of vehicle CO emissions, which could have been the result of laboratory barometric pressure dif- ferences. Vehicle variability made an accurate assessment of this effect very difficult. ------- Emission Laboratory Correlation Study Between EPA and Honda Motor Company, Inc. Introduction: The goals of a laboratory correlation program are inherently two-fold; to identify differences in equipment, techniques, and procedures, and to relate the effect of these differences to emis- sion levels of various motor vehicles. In this program, emphasis was placed on evaluating the effects of laboratory differences on vehicle emissions rather than isolating the differences. This re- port will discuss probable differences between EPA's certification test site #6 and a test site of the Honda Motor Company, Inc. and explore the effect of these differences on the emission levels of several Honda vehicles. Purpose: It is the intent of this report to present the findings of the correlation study between EPA and Honda and to derive from those findings conclusions regarding the usage of production vehicles as correlation tools. Test Vehicles and Equipment: During this correlation study, four Honda vehicles were tested at EPA. All were front wheel drive vehicles equipped with four- cylinder engines, and all were tested in the 2000 pound inertia class. Data of these vehicles are tabulated below: V.I.D. 657LF-1009 657LF-1004 SBA-2112073 SG-1000249 Model Civic CVCC Civic CVCC Civic Civic CVCC C.I.D. 90.8 90.8 75.5 90.8 Transmission 4-speed 4-speed 4-speed 2-speed System Engine Engine Air Engine Modification Modification Injection Modification Standard Federal California Federal California Statutory Interim Interim Interim The model Civic CVCC vehicle employs Honda's stratified-charge engine. Other equipment used in this study included nine calibration gas cylinders, a Honda critical-flow orifice for CVS calibration, timing deivces to measure dynamometer characteristics, and appropriate trans- ducers to measure vehicle speed, manifold vacuum, and temperatures. ------- -2- Sequence of Correlation Testing: All vehicles tested at EPA in this program were previously tested at Honda's Suzuka laboratory in Japan. The calibration gases provided by Honda were not analyzed in Japan prior to EPA's analysis. Test Procedures: Emission Tests - All vehicles involved in this correlation study underwent the same types of emission tests. These tests .may be cate- gorized as three types: (a) 1975 certification tests, without evap- orative measurements, (b) steady-state dilute exhaust samples,' and (c) steady-state raw exhaust samples. In conducting the 1975 certi- fication tests, the Federal Test Procedure was modified to eliminate unnecessary procedures. Each vehicle was preconditioned before its first test, but not before subsequent tests. No evaporative measure- ments were taken on any vehicle, but before each test, the vehicle's fuel was heated from 60°F to 84°F as specified in the evaporative section of the Federal Test Procedure. Each vehicle underwent three valid 1975 tests, the second of which was driven by a Honda technician. The Honda driver was used in an attempt to identify the effect of driver differences on vehicle emissions. The steady-state dilute samples were obtained from each vehicle at constant speeds of 50, 30, and 10 mph. The vehicle was warmed on the dynamometer and then run at each steady state speed for five minutes. A separate set of collection bags was deployed for each steady-state speed, and these bags were analyzed by EPA following normal certification procedures. All of these tests were driven by the Honda driver. The steady-state raw samples were obtained from each vehicle at speeds of 60, 50, 40, 30, 20, 10, and idle. A sample probe was placed directly in the vehicles exhaust and the emission levels were measured by Honda technicians. These tests were also driven by the Honda driver, Calibration Gas Check - The calibration gases provided by Honda were analyzed at EPA on certification train 21. Dynamometer Check - Honda technicians obtained characteristics of EPA dynamometer #6 by running a series of coast-downs and accelera- tions at the 2000 pound inertia weight. Coast-downs were run both with the vehicle on and off the dynamometer rolls. Accelerations were'run, using a 4-speed Honda Civic CVCC, from 30 to 50 mph in 4th gear and 20 to 40 mph in 3rd gear. Both accelerations were at wide open throttle. ------- -3- CVS Check - With the aid of EPA technicians, Honda engineers conducted a check of EPA's CVS unit 9C. A Honda critical-flow orifice was used to calibrate the flow rate of the CVS system. Data Sources: All data presented in this report were generated at EPA test site #6 or at Honda's Suzuka laboratory. Data relating to the dyna- mometer check were generated and analyzed by Honda engineers. Analysis of Data: Emission Tests - Two types of emission tests will be used to compare EPA and Honda test sites - 1975 certification tests and steady-state dilute sample tests. Data of these tests may be found in Appendix I. The four vehicles used in this laboratory correlation produced greatly different data comparisons, but some general trends can be seen. Using average certification test emissions of the vehicles at the two laboratories, NOx and C02 values appear to be equivalent. The maximum NOx discrepancies occurred on the last two vehicles tested, with Honda's lab reporting a 13% lower value than EPA on vehicle SG-1000249, but a 13% higher value on vehicle SBA-2112073. C02 levels were higher at Honda's facility than at EPA on three ve- hicles (maximum 6%), but vehicle 657LF-1004 showed a 10% lower C02 value at Honda. More consistent certification test emission differences were seen between laboratories in CO and HC measurements. Average levels of the four vehicles tested showed Honda's CO values to be 10-31% lower than EPA's, and Honda's HC values to be 27%, 24%, and 19% lower and 1% higher on the respective vehicles. The vehicle with only 1%'HC dif- ference between laboratories was the CVCC model designed to meet the 1975 federal statutory standards. The steady-state emission tests tended to show very small differ- ences in emission levels of the two laboratories. C02 levels at Honda's laboratory were consistently lower than at EPA's, especially at the 10 mph speed (4-25% lower). Differences in other emittants were not con- sistent as speeds and vehicles varied. Calibration Gases - Because the calibration gas cylinders were analyzed at EPA prior to analysis in Honda's laboratory, no data com- parison is available at this time. Results of EPA's analysis are tabulated in Appendix II. Dynamometer Checks - The dynamometer coast-down data acquired by Honda was used to derive a comparison with their dynamometer in ------- -4- Japan. (See Appendix III) EPA's dynamometer #6 was characterized as absorbing greater horsepower over the entire speed range, with a maxi- mum difference of approximately 0.9 horsepower at 40 mph. Discussion: Because of the design of this correlation study, an accurate assess- ment of the effects of test site differences is not easy to make. No data are available which would compare the effects of analysis system differences, and the use of four vehicles increases the variability which an analysis must consider. Some differences between test sites can be isolated, but their effects can only be approximated, at best. Dynamometer tests revealed that EPA's dynamometer #6 exerted a slightly greater load than Honda's test dynamometer, but that added loading did not seem to affect C02 or NOx levels significantly. Data acquired by Honda technicians in their CVS check are not available for this report, but their analysis indi- cated no significant differences would be experienced because of varia- tions in the CVS systems. Another possible difference, that of driver characteristics, was not experienced in a comparison of tests driven by EPA and Honda drivers. One difference which is very obvious is the test site barometric pressure, but the effect of this difference could vary between vehicles, making its isolation quite difficult. Using the vehicle emission tests as comparators, Honda's labora- tory produced significantly lower CO and slightly lower HC levels than EPA's laboratory. Twice during testing, vehicle problems caused un- usually high HC values. The first test on vehicle SBA-2112073 pro- duced high HC because an engine vacuum line was leaking. The problem was found by Honda technicians and corrected. Vehicle SG-1000249 made use of a throttle control valve which was sensitive to barometric pressure, and that valve was adjusted prior to the final test on that vehicle. Reasons for the trend toward lower HC and CO at Honda's laboratory are not clear, but the effect of barometric pressure may be the main cause of the discrepancy. As has been the case in previous EPA correlation programs, vehicle variability poses the major problem in judging test site equivalency. If one looks at the most repeatable vehicles used in this study, CO discrepancies seem to be the only problem. Vehicle #657LF-10.04 was most repeatable on HC, CO and NOx, and Honda's discrepancies with EPA's average values on that vehicle were only 1%, 10%, and 3.4% respectively. The most repeatable C02 vehicle, #657LF-1009, showed only a 2% differ- ence in average C02 levels. Other vehicles showed larger deviations, but they also exhibited less operational consistency. Unfortunately, one cannot rely on vehicles to isolate test site differences, but, in this study, vehicles were used in an attempt to reach that end. ------- -5- Concluslons: Results of this correlation study indicate a high degree of equivalency between Honda's test site and EPA's site #6. The only major discrepancy observed was Honda's tendency to produce lower CO emission levels than EPA. The large difference in barometric pressure between the two facilities may be a major contributor to that dis- crepancy, but the variability between test vehicles and the small number of tests on each vehicle make an accurate estimation of this barometric effect impossible. Summary and Recommendations: Results of the laboratory correlation study between EPA and Honda Motor Company, Inc. have been presented and discussed. In general, there was a very good correlation between the test sites of the faci- lities involved. In future correlation studies, programs should be designed to use vehicle emission tests as a final check of site equivalency. Prior to any vehicle tests, both EPA and the manufacturer should identify dif- ferences relating to CVS, dynamometers, and analyzers. In this program, there was no confidence that the laboratories' equipment was equivalent before vehicles were tested. Unless this confidence is established before vehicles are tested, emissions tests will not be helpful in isolating the effects of test site differences. ------- APPENDICES ------- APPENDIX I Emission Test Data Section Vehicle IA 657LF-1009 IB 657LF-1004 1C SBA-2112073 ID SG-1000249 ------- Section IA Vehicle 657LF-1009 Federal Statutory Standard ------- HONDA CORRELATION DATA fLA 41 Vehicle Civic 657LF-1009 Standard Statutory Test Lab HC CO NO, CO 2 Fuel Consumption (mpg) Barometric Pressure (in. Hg) K-Factor Cold Transient (g) EPA 3.04 21.83 5.82 1267.91 24.3 SuzuKa 2.73 16.39 6.46 1306 Cold Stabilized (g) PPA 0.92 11.27 4.16 1508.49 22.7 Suzuka 0.48 9.62 4.51 1516 Hot Transient (g) EPA 1.59 12.32 5.76 1164.75 26.8 Suzuka 1.56 8.64 6.48 1203 Total (g/mile) EPA 0.42 3.69 1.33 362.35 24.4 29.04 0.8167 Suzuka 0.34 2.88 1.46 368 30.19 ------- Honda Correlation Data (LA4) Cold Transient (g) Cold Stabilized (g) Hot Transient (g) Total . (g/mile) Vehicle Civic CVCC 657LF-1009 Test Lab. EPA Test No. 1 2 3 1 2 3 1 2 3 1 2 3 HC 2.64 3.55 2.93 0.91 1.02 0.82 1 .48 1.77 1.51 0.385 0.475 0.392 CO 18.94 26.82 19.72 11.30 13.68 8.83 10.15 15.28 1 1 . 54 3.36 . 4.52 3.19 NOX 5.63 "5.62 6.20 4.08 3.. 98 4.42 5.79 5.54 5.94 1.31 1.27 1.40 C02 1266.37 1253.11 1284.26 1512.27 1 488 . 24 1524.85 1173.20 1162.55 1158.50 363.42 358.63 364.99 Fuel Consumption (mpg) 24.4 24.4 24.0 22.6 ' 22.9 22.5 26 .Y 26.7 27,0 24.4 24.5 24.3 Barometric Pressure (in. Hg) / / / 29.19 28.83 29 . 09 K-Factor / / / 0.8114 0.8151 0.8237 ------- HnnHa Correlation Data (LA4) Vehicle civic cvcc 657LF-1009 Test Lab. Suzuka Test No. HC CO NOv C02 Fuel Consumption (mpg) Barometric Pressure (in. Hg) K-Factor Cold Transient •• (g) 2.65 16.98 6.85 1324 2.52 15.65 6.46 1305 3.03 16.55 6.08 1289 Cold Stabilized (9) 0.44 9.08 4.60 1513 0.51 9.97 4.54 1501 0.50 9.80 4.40 1521 Hot Transient (g) 1.08 7.73 6.29 1187 1.80 8.65 6.49 1205 1.79 9.55 6.65 1218 Total (g/mile) 0.29 2.77 1.48 368 0.35 0.37 2.88 2.98 1.47 1.44 367 369 30.36 30.12 30.09 ------- HONDA CORRELATION DATA ("CRUISING") Vehicle Civic 657-1009 Standard Statutory 50 mph (g/mile) 30 mph (g/mile) 10 mph (g/mile) Test Lab EPA Suzuka EPA Suzuka EPA Suzuka HC 0.01 0.007 0.01 0.004 0.10 0.06 CO 0.50 0.60 0.56 0.66. 2.65 2.76 NOv 1.34 1.44 0.60 0.56 0.57 0.48 CO? 256.99 265.6 257.06 269.2 538.50 520.8 Fuel Consumption fmpgl 34.4 34.4 16.3 Barometric Pressure fin.Hgl 29.12 29.12 . 29.12 K-Factor 0.8697 0.8697 0.8697 ------- Section IB Vehicle 657LF-1004 California Interim Standard ------- HONDA CORRELATION DATA TLA 41 Vehicle Civic 657LF-1004 Standard Interim California Cold Transient Cg) Test Lab, EPA Suzuka HC 4.93 5.18 CO 31.28 28.43 NO, 6.24 5.80 CO 2 1172.92 1068 Fuel Consumption (mpg) 25.7 barometric Pressure (in. Hg) K-Factor Cold Stabilized (8) PPA 2.60 20.56 4.78 1346.11 25.0 Suzuka 2.66 19.38 .4.65 1200 Hot Transient (g) EPA 2.10 16.69 6.22 1056.73 29.2 Suzuka 2.03 13.04 6.15 972 Total .. (g/mile) EPA 0.789 5.80 1.47 327.04 26.5 Suzuka 0.80 5.20 1.42 295 29.01 0.8057 30.15 ------- Honda Correlation Data Cold Transient (g) . Cold Stabilized (g) Hot Transient (g) Total (g/mile) Vehicle Civic CVCC 657LF-1004 Test Lab. EPA Test No. 1 2 3 1 2 3 1 2. 3 1 2 3 HC 4.81 4.83 5.14 2.47 2.77 2.55 1.93 2.09 2.28 0.752 0.806 0.808 CO 31.42 33.16 29.27 21.49 20.55 19.65 15.87 18.08 16.12 5.87 . 6.02 5.52 MOX 5.96 6.43 6.34 4.88 4.54 4.91 6.04 6.10 6.52 1.45 1.44 1.51 C02 1179.62 1159.31 1179.83 1373.26 1308.88 1356.20 1076.56 1029.49 1064.13 332.55 319.22 329.34 Fuel Consumption (mpg) 25.6 26.0 25.6 24.5 25.7 24.9 28.8 29.9 29.0 26.1 27.0 26.4 Barometric Pressure (in. Hg) / / / 28.94 29.19 28.89 K-Factor / / / 0.8128 0.8013 0.8031 ------- Honda Correlation Data (LA4) Vehicle Civic CVCC 657LF-1004 Test Lab. Suzuka Test Mo. HC CO NO, C02 Fuel Consumption (mpg) Barometric Pressure (in. Hg) K-Factor Cold Transient (g) 5.37 29.42 5.68 1055 4.64 25.98 5.90 1077 5.54 29.88 5.82 1071 Cold Stabilized (g) 2.64 18.26 4.66 1201 2.72 18.68 4.48 1198 2.63 21.20 4.80 1202 Hot Transient (g) 1.97 13.01 5.88 981 2.05 12.89 6.07 964 2.06 13.23 6.51 970 Total (g/mile) 0.81 5.11 1.39 295 0.78 0.82 4.96 5.54 1.39 1.47 295 295 30.31 30.06 30.08 ------- HONDA CORRELATION DATA ("CRUISING! Vehicle Civic 657LF-1004 Standard Interim California 50 mph (g/mile) 30 mph (g/mile) 10 mph (g/mile) Test lah EPA Suzuka EPA Suzuka EPA Suzuka HC 0.08 0.05 O.TO 0.08 0.36 0.58 CO 2.39 2.05 2.36 2.02 5.86 5.34 NOV 1.61 1.58 0.60 0.40 0.46 0.36 C02 215.86 196.6 220.41 191.6 486.28 364.8 Fuel Consumption Cmpg") 40.4 39.5 17.9 Barometric Pressure fin.Hgl 28.94 28.94 28.94 K-Factor 0.8128 0.8128 0.8128 ------- Section 1C Vehicle SBA-2112073 Federal Interim Standard ------- HONDA CORRELATION DATA CLA 41 Vehicle Civic SBA-2112073 Standard Federal Interim Test Lab, HC CO NO, CO 2 Fuel Consumption (mpg) Barometric Pressure (in. Hg) K-Factor Cold Transient (g) EPA 5.77 54.42 6.48 1100.47 26.5 Suzuka 4.55 44.57 7.26 1216 Cold Stabilized (g) FPA 7.56 47.22 3.73 1332.59 24.2 Suzuka 5.43 36.82 4.23 1382 Hot Transient (g) EPA 5.92 35.59 7.15 1001.45 29.6 Suziika 4.91 25.02 8.11 1084 Total (g/mile) EPA 1.79 12.1 1.41 316.88 26.5 Suzuka 1.36 9.37 1.60 335 28.83 0.8852 30.04 ------- Honda Correlation Data Q-A41 Cold Transient (g) Cold Stabilized (g) Hot Transient (g) Total (g/mile) Vehicle Civic SBA-21 12073 Test Lab. EPA Test No. 1 2 3 1 2 3 1 2 3 1 2 3 HC 6.94 4.78 5.58 8.99 6.04 7.66 7.52 4.87 5.36 2.17 1.45 1.75 . CO 54.53 59.00 49.72 47.90 47.46 46.30 38.21 38.23 30.32 12.4 12.6 11.3 NOX 6.52 6.31 6.62 3.88 3.55 3.76 6.97 6.95 7.53 1.42 1.36 1.45 C02 1108.73 1104.88 1087.80 1352.69 1316.11 1328.98 1018.54 996.95 988.87 321.33 314.60 314.72 Fuel Consumption (mpg) 26.2 26.3 26.9 23.8 24.6 24.3 28.9 29.7 30.2 26.1 26.6 26.9 Barometric Pressure (in. Hg) / / / 28.82 28.48 29.20 K-Factor / / / 0,9121 0.8926 0,8508 ------- Honda Correlation Data (LA4) Vehicle Civic SBA-2112073 Test Lab. Suzuka Test No. HC CO NO, C02 Fuel Consumption (mpg) Barometric Pressure (in. Hg) K-Factor Cold Transient (9) Cold Stabilized (g) Hot Transient (g) Total (g/mile) 4.34 5.55 3.77 5.28 5.48 5.54 4.43 5.18 5.12 1.29 1.44 1.35 41.35 43,86 48.50. 32.00 38.62 39.85 23.39 24.47 27.19 8.42 9.52 10.16 7.85 7.46 6.46 4.56 4.10 4.02 8.49 8.09 7.75 1.71 1.59 1.50 1209 1184 1156 1400 1385 1361 1074 1081 1098 338 335 331 30.19 30.07 29.86 ------- HONDA CORRELATION DATA CCRUISING1 Vehicle Civic SBA-2112073 Standard Federal Interim 50 mph (g/mile) 30 mph (g/mile) 10 mph (g/mile) Test Lab EPA Suzuka EPA Suzuka EPA Suzuka HC 0.47 0.24 0.21 0.16 0.54 0.48 CO 4.98 2.92 4.71 4.00 13.18 11.16 NO* 2.24 1.56 0.48 0.44 0.42 0.48 CO? 198.37 223 243.91 230 520.37 474 Fuel Consumption (mpgl 42.7 35.2 16.3 Barometric Pressure rin.H£l 28.80 28.80 28.80 K-Factor 0.9123 0.9123 0.9123 ------- Section ID Vehicle SG-1000249 California Interim Standard ------- HONDA CORRELATION DATA CLA 41 Vehicle Civic SG-1000249 Standard Interim California Cold Transient (g) Test Lab, EPA Suzuka HC 4.23 3.23 CO 40.39 29.01 NO, 8.69 7.83 CO 2 1278.20 1378 Fuel Consumption Qnpg) 23.5 Barometric Pressure (in. Hg) K-Factor Cold Stabilized (g) PPA 2.01 19.64 6.41 1392.40 24.3 Suzuka 1.36 14.06 5.54 1426 Hot Transient (g) EPA 1.93 17.57 8.42 1123.87 27.5 Suzuka 1.50 10.30 7.22 1230 Total (g/mile) EPA 0.66 6.29 1.99 344.35 25.0 Suzuka 0.48 4.32 .1.74 363 28.84 0.8901 30.08 ------- Honda Correlation Data (LA4) Cold Transient (g) Cold Stabilized (g) Hot Transient (g) Total (g/mile) Vehicle Civic S6-1000249 Test Lab. EPA Test No. 1 2 3 1 2 3 1 2 3 1 2 3 HC 4.93 4.12 3.65 2.08 2.06 1.89 1.89 1.98 1.92 0.70 0.661 0.61 . CO 44.47 39.14 37.55. 21.31 19.50 18.11 18.00 19.92 14.79 6.8 6.36 5.7 NOX 8.55 8.19 9.33 6.58 5.87 6.78 8.71 7.80 8.75 2.03 1.85 2.10 COe 1244.54 1254.53 1335.54 1398.29 1361.74 1417.17 1137.99 1119.38 1114.25 344.28 338.56 350.21 Fuel Consumption (mpg) 23.9 24.0 22.7 24.1 24.8 23.9 27.2 27.5 27.9 25.0 25.3 24.6 Barometric Pressure (in. Hg) / / / 28.83 28.50 29.19 K-Factor '/ / / 0.9295 0.8924 0.8485 ------- Honda Correlation Data (LA4) Vehicle Civic SG-1QQ0249 Test Lab. Suzuka . Test No. HC CO NO, C02 Fuel Consumption (mpg) Barometric Pressure (in. Hg) K-Factor Cold Transient (g) 3.26 25.33 7.69 1378 3.53 31.19 7.53 1381 2.91 30.52. 8.26 1374 Cold Stabilized (9) 1.30 13.86 5.56 1428 1.38 14.14 5.44 1434 1.39 14.17 5.63 1416 Hot Transient (g) 1.37 9.76 7.02 1216 1.55 10.77 7.17 1253 1.59 10.37 7.46 1221 Total (g/mile) 0.47 4.04 1.72 362 0.51 0.47 4.49 4.43 1.70 1.79 366 361 30.22 30.14 29.87 ------- HONDA CORRELATION DATA CCRUISING1 Vehicle civic SG-1000249 Standard Interim-California 50 mph (g/mile) 30 mph (g/mile) 10 mph (g/mile) Test Lab EPA Suzuka EPA Suzuka EPA Suzuka HC 0.01 0.01 0.02 0.04 0.57 0.54 CO 0.57 0.62 1.31 0.86 3.09 3.36 NO* 1.63 1.37 0.68 0.56 0.47 0.54 i C02 242.16 259 210.94 204 328.11 2.64 Fuel Consumption fmpgl 36.5 41.6 26.5 Barometric Pressure rin.Hgl 28.80 28.80 28.80 K-Factor 0.9255 0.9255 0.9255 ------- APPENDIX II Calibration Gas Data ------- Honda Calibration Gases Gas C3H8 in air (ppm) CaHs in N2 (ppm) CO in N? (ppm) C02 in N2 (%) NO in N2 (ppm) NOX in N2 (ppm) NO in N2 (ppm) NOx in N2 (ppm) EPA Analysis - Tra Cylinder No. C-4552 C-4508 C-4599 C-4546 C-6"74 ' C-690 C-1260 C-192 C-1288 in 21 Scott Laboratory 7.31 22.46 27.52 44.15 70.2 222 0.93 24.1 24.4 46.0 46.5 EPA 7.2 22.6 27.7 43.9 68.2 215.4 0.919 23.0 23.3 43.2 44.2 Honda ------- APPENDIX III Dynamometer Data ------- ------- *4 HP a VEHICLE ------- |