Measurement of particulate matter from diesel engine exhaust using a tapered element oscillating microbalance

Abstract

Abstract

Acquisition of real-time particulate matter (PM) data is proposed by the use of a Rupprecht and Patashnick Company, Inc. Series 110 diesel particulate monitor tapered element oscillating microbalance (TEOM) mass measuring device. The objectives of using a TEOM diesel particulate analyser were to validate its collection capability and evaluate its real-time transient characteristics. Conventional PM measured using dilute diesel exhaust filtration was used as the base line for evaluating the TEOM collection capability. Real-time mass rate data were separated into positive and negative values, then integrated over the duration of the test. The integrated positive mass was divided by the integrated negative mass to create a ratio that is indicative of real PM collected versus moisture released from the filter. Tests were performed on heavy-duty diesel engines on a stationary dynamometer using the US Heavy Duty Federal Test Procedure speed and load cycle. TEOM sample tube temperatures at 35°C yielded the best TEOM-conventional PM filtration mass ratio. However, as sample path temperatures decreased, a decrease in the positive-negative mass ratio occurred. A compromise between conventional filter agreement and real-time data was made in selecting the temperature set point of 40°C as the most desirable sampling temperature. The sample flowrate was varied from one to four litres per minute. A comparison of the TEOM with a conventional filter decreased as the TEOM flow increased. The 1 l/min set point provided the best TEOM-conventional filtration ratio. The positive-negative mass ratio increased as flow decreased. The flow-rate of 3 l/min was chosen to be a compromise between the TEOM-conventional filtration mass ratio and real-time results. The best TEOM-conventional filtration ratio measured was 0.97. The best set of test results was a three-test averaged TEOM-conventional filtration ratio of 0.92 with a coefficient of variance of 0.15 per cent.