A comparison of a mini-PEMS and a 1065 compliant PEMS for on-road gaseous and particulate emissions from a light duty diesel truck.

TL;DR: There could be applications for the NCEM or other mini-PEMS for applications such as identification of potential issues by regulatory agencies, manufacturer evaluation and validation of emissions under in-use conditions, and potential use in inspection and maintenance (I/M) programs, especially for heavy-duty vehicles.

Abstract: The primary goal of this study was to compare emissions measurements between a 1065 compliant PEMS, and the NTK Compact Emissions Meter (NCEM) capable of measuring NOx, PM, and solid PN. Both units were equipped on a light-duty diesel truck and tested over local, highway, and downtown driving routes. The results indicate that the NOx measurements for the NCEM were within approximately ±10% of those the 1065 compliant PEMS, which suggests that the NCEM could be used as a screening tool for NOx emissions. The NCEM showed larger differences for PM emissions on an absolute level, but this was at PM levels well below the 1 mg/mi level. The NCEM differences ranged from −2% to +26% if the comparisons are based on a percentage of the 1.0 mg/mi standard. Larger differences were also seen for PN emissions, with the NCEM measuring higher PN emissions, which can primarily be attributed to a zero current offset that we observed for the NCEM, which has been subsequently improved in the latest generation of the NCEM system. The comparisons between the 1065 compliant PEMS and the NCEM suggest that there could be applications for the NCEM or other mini-PEMS for applications such as identification of potential issues by regulatory agencies, manufacturer evaluation and validation of emissions under in-use conditions, and potential use in inspection and maintenance (I/M) programs, especially for heavy-duty vehicles.

Summary

1. Introduction

• Portable Emissions Measurement Systems (PEMS) are tools that are designed to measure vehicle/truck emissions while operating on the road.
• PEMS serve an important role in helping to better understand and characterize the differences between laboratory certification and other testing and real-world emissions.
• As in-use emissions testing has advanced, emissions data has continued to show the importance of measuring emissions in-use to fully understand the range of emissions emitted by vehicles under different operating conditions.
• Given the complexity and cost of 1065 compliant PEMS, there is a growing interest in the development ofmini-PEMS that are not targeted at compliance with 1065 specifications, but still provide reliable emissions measurements, and are easy to deploy and less expensive.

2.1. Test vehicle, engine, and fuel

• The test vehicle is a model year 2012 Chevrolet Silverado 2500HD.
• This vehicle is equipped with an engine family CGMXD06.6355 diesel engine.
• The engine is 6.6-liter, eight cylinders, turbocharged, direct injection, and common-rail engine configuration with a six-speed automatic transmission.
• It should be noted that the vehicle was filled up several times at the same retail fueling station.
• Since the properties of in-use California ultralow sulfur diesel are tightly controlled to provide comparable emissions, the use of diesel fuel from different fill ups is expected to have minimal impact on the emissions results.

2.2. Test cycles

• The vehicle was tested over a period of 2 days using three different driving routes designed to represent local, highway, and LA downtown driving conditions.
• The characteristics of these three different cycles are shown in Table 1, alongwith the details for the FTP test for comparison.
• The local route is used to simulate the local driving and has a similar driving pattern to FTP driving cycle.
• The highway route started at UCR and went to the main campus of the University of Southern California.
• The total distance of this route was 63.6 mi.

2.3. Instruments

• The AVL PM PEMS measurement system selected is AVL's 483 micro soot sensor (MSS) in conjunction with their gravimetric filter module (GFM) option.
• The NCEM uses direct measurement sensors rather than dilution sampling.
• It can be poweredby aDC12/24 V vehicle battery and draws b10 Amp to operate.
• In these cases the sensitivity to NO2 could be slightly lower than the sensitivity to NO.

TK AVL MSS AVL PM NTK AVL

• The PM/PN sensor is based on the Pegasor PPS-M technology, where particles are charged in a corona discharge, such that the totalmeasured charge is proportional to the particle surface area, as shown in Fig. 2 (Lanki et al., 2011; Ntziachristos et al., 2011; Ntziachristos et al., 2013; Rostedt et al. 2017).
• Todetermine PN, the sensor is calibrated against a TSI scanning mobility particle sizer (SMPS).
• Both the PM and PN calibrations are performed with a soot generator that provides soot particles with a unimodal distributionwith peak concentration around 75 nm.
• Simulations using a range of possible diesel particle size distributions, however, have shown that the maximum theoretical error is 23% when using surface area as a proxy for number and 39%when using surface area as a proxy for soot mass, although the actual error is expected to bemuch less than these values (Ntziachristos et al., 2012).
• A Semtech 4-inch Exhaust Flow Measurement (EFM) system was used by both systems for the measurement of the exhaust flow to provide integrated mass emissions as well as second by second data for each pollutant.

2.4. Measurement protocols

• The experimental set up for study is shown in the Fig.
• This includes the NCEM, AVL gaseous M.O.V.E. system, AVL PM system, and the AVL PN PEMS iS.
• The power system for the set up included a Yamaha gasoline generator model EF2800i, which has two 120 V AC plugs with 20A maximum current each, a CHARGEMASTER 12 V power converter to power the AVL GasM.O.V.E system, and a Xantrex sinewave inverter powered through a twin 12 V battery pack to power the EFM and the computers.
• The purpose of the 12 V batteries were to support as a backup power source, whichwas necessarywhen switching frombuilding power to the generator power, or when powering down the generator to add more fuel.
• The NCEM was controlled through the screen of the unit with the data logged to a flash drive.

3.1. NOx emissions

• The NOx emissions results for all the testing routes are shown below in Table 2.
• The NCEM did not show a consistent bias compared to the AVL M.O.V.E system, with the NCEM reading higher for some test routes and lower for others.
• The 25 and 75 percentile points are provided in Table 3,which are the points below which 25% and 75% of the measurements fall for both instruments.
• The NOx emissions can also be compared back to early comparisons between 1065 compliant PEMS and CVS reference methods conducted as part of the Measurement Allowance program (Johnson et al., 2009, 2011a).
• Following the repairs, the vehicle was tested again over the FTP cycle on a chassis dynamometer, and NOx emission levels were found to be below the 0.2 g/mi NOx emission standard.

3.2. PM emissions

• The test vehiclewas equippedwith a DPF, so the PMemissions levels were generally low.
• This suggests a possible PM physical characteristic change between in-town driving and cruise conditions that may have caused the NCEM to report differently.

3.3. PN emissions

• For the AVL PN PEMS, the PN emissions were typically 93% below the PN standard.
• The NCEM also measures particles down to ~10 nm (Amanatidis et al., 2017), as opposed to the PN PEMS that has a 23 nm size cut off, which could contribute to higher PNmeasurements for the NCEM.
• In other work, Tikkanen et al. (2013) found the PPS reported 80% higher PN than an APC for a heavy-duty engine and somewhat higher PN emissions for a passenger car, but lower PN emissions than an APC during a regeneration due to desorption from the CVS and for a Euro 4 diesel vehicle during high speed portions of the testing.

4. Conclusions

• The primary goal of this study was to compare emissions measurements between a 1065 compliant AVL M.O.V.E.S. PEMS, and a current generation mini-PEMS capable of measuring NOx, PM, and solid PN.
• The NCEM showed larger differences for PM emissions on an absolute level, but this was at PM levels well below the 1 mg/mi level.
• While the ECM data was not collected with the NCEM used in this study, the current version of the NCEM does collect ECM data that could be utilized for determining the exhaust flow rate.
• The comparisons between the 1065 compliant PEMS and the NCEM suggest that there could be applications for the NCEM or other miniPEMS in areas where larger data sets of emissions data, or where the cost of full laboratory or 1065 compliant PEMS testing is prohibitive.
• As recentfindings have suggested that it is important tomonitor vehicle emissions under a much wider range of conditions than can be duplicated in the laboratory, the NCEM could play a role in allowing for the testing of more vehicles under a broader range of conditions.

Figures

Fig. 1. NCEM NOx measurement design schematic.

Fig. 9. Q-Q plot analysis for PN emissions for 1 day of testing.

Fig. 5. Q-Q plot analysis on NOx emissions for 1 day of testing.

Table 3 Summary of 25th and 75th percentile Q-Q plot values for NOx, PM, and PN emissions.

Table 4 Summary of correlation slope and regression statistics for NOx emissions NOx (g/s).

Table 6 Summary of correlation slope and regression statistics for PM and soot emissions.

Table 1 Summary of trips statistics for different routes and cycles.

Table 5 Summary of PM emissions.

Fig. 3. Instrument setup and power supply for on-road PEMS testing.

Fig. 2. NCEM PM and PN measu

Table 8 Summary of correlation slope and regression statistics for PN emissions.

Fig. 6. Vehicle speed based comparisons for particle emissions for 1 day of testing.

Fig. 7. Q-Q plot analysis on PM em

Fig. 4. Vehicle speed based comparisons fo