Calibration and accuracy of a particle number measurement system

The particle number method was introduced in the Euro 5/6 light-duty regulation and will also be introduced in the Euro VI heavy-duty regulation. The legislation requires yearly calibration or validation of the particle number systems with monodisperse aerosol for the determination of the particle number concentration reduction factor (PCRF); however, there are some open issues, like the material to be used. In this paper, we describe the new particle number system from AVL (APC) and its calibration procedure which gives repeatability of its PCRFs after 1 year of better than ±6% (two standard deviations of the differences). We show that the most important effect on the calibration results can come from the generation method (material and thermal pre-treatment). The use of neutralizers downstream and upstream of the DMA was found to have a small effect on the results. We also suggest faster and easier procedures for more frequent checks of the proper operation of the systems: (i) the check with polydisperse aerosol with a mean around 50 nm and (ii) the dilution factor check with gases. A third fast way is to compare the systems with a reference system, as two systems measuring at the same sampling position (full dilution tunnel) have <5% differences. Calibrated systems of different manufacturers have <10% differences, while non-(well)-calibrated systems might have >10% differences. Although light-duty legislation requires measurement from the full dilution tunnel, measurement from the partial flow systems (for heavy-duty engines) or directly from the tailpipe is also possible (at least with APC). Finally, the relation between particle number and particulate mass for light-duty and heavy-duty engines of different technologies is shown