Potential Errors and Error Propagation in Methods Used to Determine Particle Removal Efficiency

Abstract

There are many formulas that can be applied under field and laboratory conditions to determine the percent particle removal efficiency (PPRE). A lack of understanding of the limitations of each formula leads to improper application and thus the obtained results can be inconsistent and unreliable. A systematic study is required to understand the applicability of and relationship among these commonly used formulas for the purpose of identifying and possibly reducing the large variations often encountered in these applications. Towards this end, the common methods for calculating the PPRE were compared and analyzed; a controlled laboratory test was also carried out to illustrate the extent of the potential errors. The results showed that (1) when the calculated PPRE is positive, the error in this calculation represents only a fraction of the measurement errors in the sampled inlet and outlet data, but the error in the calculated PPRE will be amplified if the PPRE is negative; it is therefore important to be cautious if the negative removal efficiency is included in the PPRE calculation; (2) the PPRE calculated by the widely used event mean concentration (EMC) is more conservative (smaller) than the PPRE calculated by the mass load method because EMC does not conserve particle mass; the mass load method is a special case of EMC when EMC is at the point of conserving mass and should be used to calculate particle removal efficiency in all cases; and (3) even in systems with a short residence time, particle concentrations between the inlet and outlet may not be closely correlated.