DEVELOPMENT OF A CHASSIS BASED INSPECTION AND MAINTENANCE

DEVELOPMENT OF A CHASSIS BASED INSPECTION AND MAINTENANCE PROGRAM FOR HEAVY-DUTY DIESEL POWERED VEHICLES; INCLUDING, SIMULTANEOUS EMISSION MEASUREMENTS WITH A SEMTECH-D PORTABLE EMISSIONS ANALYZER California Environmental Protection Agency Air Resources Board The Statements and Conclusions presented are not necessarily those of the California Air Resources Board. The mention of commercial products reported herein is not to be construed as an actual or implied endorsement of such products.

Overview California Air Resources Board is investigating the feasibility of a widescale emissions Inspection and Maintenance (I/M) program for heavyduty diesel engines and also a field screening test for NOx. This is one of several strategies being considered in order to meet SIP Measure M 17 commitments. The objective is a 10 ton per day decrease in NOx emissions from HDDE. The concept of the HDDE I/M program is to utilize existing water-brake chassis dynamometers in Repair Facilities, coupled with portable emissions measuring equipment. A short test cycle would be conducted on commercial HDD vehicles to screen for high emitters. In order to determine the feasibility of this concept, CARB has conducted a pilot program at their Stockton facility that has been instrumented with laboratory-grade emissions equipment. A SEMTECH-D portable analyzer from Sensors, Inc. was operated in parallel for a portion of the testing in order to determine relative accuracy and feasibility of using a portable analyzer for HDD I/M testing.

HDD I/M Program and Field Screening Concept In order to implement a wide-scale HDD I/M program, existing repairgrade dynamometers could be utilized. There is already an extensive network of repair-grade, water-brake dynamometers throughout the state. For emissions measurement, portable analyzers could be used. In addition, an exhaust flowmeter is required to compute mass emissions. In this study, an air turbine was used to measure the intake airflow of the engines. For field screening, the portable emissions analyzer could be used along with a portable roadside dynamometer. En-route trucks would be detached from their trailers and operated under a prescribed test cycle. Vehicles that exceed established NOx limits on the I/M test or the field screening test would be required to undergo repairs. The test cycle evaluated in this study is a standard “power curve”.

Evaluation Method for Portable Analyzer In order to evaluate the portable analyzer, the coefficient of variation was determined on a g/whp-hr basis at a 95% confidence interval. This involved the following steps: 1. Obtain simultaneous gas concentration measurements with CARB and SEMTECH analyzers. There were 27 simultaneous tests used in this analysis that were pre-screened for quality assurance. 2. Time align each gas concentration with the intake airflow measurement and dynamometer power. 3. Compute real-time mass emissions (g/sec) for each system using the time aligned variables. Integrate the mass over the test cycle, and divide by the work to get g/whp-hr for each test on each analyzer. 4. Perform regression analysis on the data set for each gas. Determine the standard error of the regression line, and multiply by 2 to obtain the 95% confidence interval measurement uncertainty. 5. Divide the 95% CI Measurement Uncertainty by the applicable emissions standard.

Evaluation Criteria for Portable Analyzer A SEMTECH-D portable emissions analyzer from Sensors, Inc. measured emissions in parallel for a portion of the testing. This phase of testing only evaluated the gas concentration measurements, although ECM data was also collected when available. The evaluation criteria used was based the Engine Manufacturers Association (EMA) Calibrations and Standards Task Force acceptance criteria for lab-to-lab variation in emissions testing. These acceptance criteria allows a coefficient of variation (COV) of 10% based on a 95% confidence interval.

ARB’s Stockton Laboratory Truck ready for testing. ARB staff performing power curve test. Laboratory grade emissions analyzers. Clean lab ready for next truck.

Sensors’ SEMTECH-D Portable Emissions Analyzer

Power-curve Test Cycle This graph illustrates the test performed on each truck brought to the lab in order to evaluate the effectiveness of a NOx screening test. After warming up the truck, the driver puts the truck in direct gear and gives it full throttle. A load is slowly applied, using the dynamometer until the engine lugs down to past its peak power. We call this test cycle a power curve. This cycle was chosen to permit NOx emissions testing, under high engine load conditions. Analysis Region 100 hp line Ramp Lugdown

SEMTECH-D Correlation to CARB Laboratory Analyzer CO 2 Mass Emissions

SEMTECH-D Correlation to CARB Laboratory Analyzer NOx Mass Emissions

SEMTECH-D Correlation to CARB Laboratory Analyzer CO Mass Emissions Mean flow weighted CO Concentration < 100 ppm for all tests. Range is insufficient for accurate slope or r^2 determination. Regression line was forced through zero to estimate slope. Standard error and measurement uncertainty and COV determination are valid for this range. COV is consistent with published data and exceeds EMA acceptance criteria.

SEMTECH-D Correlation to CARB Laboratory Analyzer CO Concentrations per Test (Flow Weighted) Mean flow weighted CO Concentration < 100 ppm for all Power Curve Tests. Analyzer accuracy specification is +/- 50 ppm. Analyzer performed within specifications, and exceeds accuracy required to meet COV acceptance criteria on mass basis.

SEMTECH-D Correlation to CARB Laboratory Analyzer THC Mass Emissions

SEMTECH-D Correlation to CARB Laboratory Analyzer THC Concentrations per Test (Flow Weighted) Mean flow weighted THC Concentration < 100 ppm for all Power Curve Tests. Measurements agreed within 10 ppm for all but 1 test, which showed 14 ppm difference. Analyzer accuracy exceeds accuracy required to meet COV acceptance criteria on mass basis.

SEMTECH-D Correlation to CARB Laboratory Analyzer O 2 Mass Emissions

Proof of Concept SEMTECH-D Correlation Results Summary CO 2 NOx Concentration Range 6. 1 – 7. 8% 481 – 1385 25 – 93 ppm 17 – 103 ppm. C Mass Emissions Range g/whp-hr 520 – 700 4. 4 – 11. 3 0. 1 – 0. 5 0. 04 – 0. 3 Emissions Standard g/Bhp-hr NA 4– 6 15. 5 1. 3 95% CI Measurement Uncertainty, g/whp-hr 12. 8 0. 23 0. 12 0. 036 Coefficient of Variance, % of standard 2. 1% 4. 6% 0. 7% 2. 8% @ 610 g/whp-hr @ 5 g/whp-hr @ 1. 3 g/whp-hr (mean flow weighted) CO THC

Proof of Concept Evaluation of SEMTECH-D The 95% CI measurement uncertainty of the SEMTECH-D was consistent with published data. The coefficient of variance was less than 5% for all gases, which is lower than acceptance criteria established by EMA. There was a systematic error of 7% observed in the NOx correlation and 4% for CO 2, as reflected by the slope of the regression lines. While this does not constitute measurement uncertainty, it is inconsistent with other studies and will be investigated. Common sources of systematic errors are gas bottle accuracy, differences in sample conditioning, sample dilution etc. Overall, this analysis demonstrates that the use of portable emissions analyzers in a wide-scale I/M program is feasible.

Implementation Considerations This study has also provided valuable information regarding practical implementation requirements if a portable analyzer is to be used in an I/M program. These include: • Automated software for analyzer audits, calibrations and record keeping. Automated readiness checks. • Integration of dynamometer and SEMTECH data and automated data processing. • Automated test launch based on external triggers, such as dyno speed. • Automated QA monitoring during tests. • Time alignment verification.