Uncertainty of laboratory and portable solid particle number systems for regulatory measurements of vehicle emissions

Abstract

In the European Union’s emissions regulations, limits for solid particles >23 nm are applicable for the type-approval and in use compliance of vehicles. Consequently, particle number (PN) systems are used very often for both research and development of engines and vehicles, both in the laboratory and on the road. The technical specifications of the laboratory and portable on-board systems are not the same resulting in different measurement uncertainties. Furthermore, particles, in contrast to gases, can be lost in the transfer lines making comparisons at different sampling locations difficult. Moreover, the size dependent counting efficiency of the systems can result in high discrepancies when the measured particle sizes are close to the decreasing steep part of the curves. The different sampling locations (tailpipe or dilution tunnel) and thermal pretreatments of the aerosol further enhance the differences. The studies on the measurement uncertainty are scarce, especially for the PN systems measuring from 10 nm that will be introduced in the future regulations. This study quantified the uncertainty sources of the PN systems: (i) due to the technical requirements and the calibrations, (ii) due to the unknown particle sizes during measurement, (iii) due to particle losses from the vehicle to the PN systems at the tailpipe or the dilution tunnel, (iv) other parameters needed for the calculation of the emissions, non-related to the PN systems, e.g. flow and distance. The expanded uncertainty of the 23 nm laboratory systems sampling from the dilution tunnel was estimated to be 32%, with 18% originating from the calibration procedures, while of those sampling from the tailpipe 34%. For the 23 nm portable systems measuring on-road the uncertainty was 39%. The values were 2–8% higher for the 10 nm systems.

Highlights

• •

  Measurement uncertainty of laboratory and portable particle number systems.

• •

  The differences between systems are due to the technical specifications, the unknown particle size and the particle losses.

• •

  The expanded uncertainty, depending on the application, was 32%–42%.