Comparison of Air Filtration Efficiency Measurements for Pleated and Flat Sheet Filters 970671

In the design of pleated air filters, the velocity of the flow at the filter surface is many times approximated as the “effective” velocity computed by assuming that the air flow is evenly distributed over a cross-sectional area equal to that of the “unfolded” filter. In our findings this effective velocity is much lower (by a factor of 10 to 40) than the average “duct” velocity that would be computed two to four centimeters upstream of the filter plane. This effective velocity factor would depend on parameters such as type of filter media (fiber diameter, filter solidity and material) and pleat configuration (height and spacing). When used to design engine intake or cabin air filters, the effective velocity may be misleading if used in a predictive theory for filter efficiency. In reality, the upstream portion of the pleat may be exposed to a considerably higher velocity than the downstream portion of the pleat. Thus the actual effective velocity over the filter is probably between the “duct” velocity and the “unfolded” (or “face”) filter velocity. The purpose of this paper is to investigate these effects by comparing pleated to flat sheet media filtration efficiencies.

Herein are presented measurements of the efficiencies of pleated filters and flat sheet filters, made of the same material, over a wide range of air flow rates. The two sets of efficiencies are compared in order to gain insight as to how to relate measurements and predictions for the two geometries. In addition, variations in local efficiency measurements are shown for the two geometries, both for comparison of the geometries and for examining how local efficiency is related to overall efficiency. Local efficiency was determined by Laser Doppler Velocimetry (LDV) measurements [of velocity and particle density] on an approximately 2 cm by 2 cm grid spacing within 5 cm upstream and downstream of the filter plane. In order to insure well-defined test conditions, one micron polystyrene latex spheres were used as the contaminant. Results show that there can be a local efficiency variation over the filter of 10%, and that flat media efficiency versus Stokes number trends are not related to pleated media trends by the simple ratio of their flat and unfolded areas. Results are applicable to the SAE test procedures, J726 and J1669, for engine intake and cabin air filtration, respectively.