The effect of pleating density and dust type on performance of absolute fibrous filters

The importance of clean air to the well-being of people and for the protection of industrial equipment has highlighted the critical role of air filter performance. The objective of this thesis is to study the filter performance characteristics namely; the pressure drop and the fractional efficiency of standard full scale (592x592x400 mm) mini-pleated HEPA absolute fibrous filters. Accurate filter performance prediction plays a significant role in estimating the lifetime of filters and reducing energy and maintenance operating costs. To ensure the appropriate filter selection has been made for a specific application, filter design must be further investigated to include pleat count and its corresponding surface area. The investigations undertaken in this work were based on using standard SAE coarse and fine dust. This guaranteed that the results would be applicable globally irrespective of the geographical location or the field of application of the filtration technology. However, the performance of air filters used in gas turbine and HVAC applications tend to deviate from that predicted by laboratory results using standard air dust. This is especially true in regions known to have dust with characteristics deviating from that of standard dust, such as in Kuwait. Therefore, as part of this thesis, the Kuwaiti atmospheric dust has been characterized both chemically and physically in order to investigate the possible impact of these characteristics on the results of the filter performance. It transpired, however, that the characteristics of dust with extreme properties, represented in this case by Kuwaiti dust, fell between those of the characteristics of the two standard types of dust, namely, SAE coarse and fine dust. This finding, therefore, provides additional confidence in the generality of the results pertaining to the filter performance. The work investigated the effects of ten different flow rates ranging from 500 to 5000 m3/h with increment of 500 m3/h. The four different pleating densities used to construct the filter were 28, 30, 32 and 34 pleats per 100mm. This experimental work was conducted while keeping other parameters such as filter media class unchanged. Pleating density may play a major role in achieving the optimum pressure drop and the required efficiency expected from such a filter. Such optimization was expected to ii facilitate design alternatives supported by experimental results. A testing facility located in Limburg Germany at the EMW Filtertechnik GmbH was used for this testing. Two different particle size counters were used to cover a considerable particle size range. The results of the particle counter with size range of: 0.065 – 0.9 μm was used for the analysis since it covered the study of the Most Penetrating Particle Size (MPPS) with respect to the filter pleat density and face velocities. This experimental work involved testing ten industrial full scale HEPA filters, which were divided into three groups. The first two Groups (Group A and B), each consisted of four filters manufactured with different pleat densities of 28, 30, 32 and 34 pleat per 100 mm. The third Group C consisted of two filters; the first filter had horizontal pleat orientation while the second pleats were oriented vertically. Both filters in Group C had a pleat density of 28 pleat per 100 mm. Filters of Groups A and B were challenged with SAE coarse and fine dust, respectively. In the case of filters of Group C, only the initial pressure drop and efficiency measurements were conducted. This experimental work has highlighted the underlying reasons behind the reduction in filter permeability due to the increase of face velocity and pleat density. The reasons which led to surface area losses of filtration media are due to one or combination of the following effects: pleat crowding, deflection of the entire pleated panel, pleat distortion at the corner of the pleat and/or filtration medium compression. The experimental data for fractional efficiency were fitted using a modified Lee and Liu [1982a] model. The proposed modified model was verified to show a good agreement with the experimental results. It is evident from entire array of experiments that as the particle size increases, the efficiency decreases until the MPPS is reached. Beyond the MPPS, the efficiency increases with increase of particle size. The MPPS shifts to a smaller particle size as the face velocity increases and the pleating density and orientation did not have a pronounced effect on the MPPS. The second part of this experimental work involved the dust loading process which showed a higher efficiency and pressure drop response of SAE fine dust when compared to SAE coarse dust. Increasing the mass of dust loads and flow rates have a significant effect on the filter efficiency while the effect of varying pleating density was negligible. Throughout this study, optimal pleat count which satisfies both initial and dust loaded pressure drop and efficiency requirements may not have necessarily existed. This experimental work has also suggested that a valid comparison of the pleat densities iii should be based on the effective surface area which participates in the filtration action and not the total surface area the pleat density provides. The work in this thesis has presented novel contribution in four aspects. Firstly, the full scale nature of the experiments resulted from using full scale standard industrial size HEPA filters constructed in V-shape banks cartridge in all the tests. Secondly, a novel explanation of when the surface area losses become a dominant factor in the filter permeability reduction. Thirdly, the discovery of the fact that increasing the pleating density could be counterproductive in terms of effective filtration surface area and filter permeability. Finally, the work has proposed new design alterations for maintaining effective surface areas. All design improvements are currently under review as they might require developmental work and investigation prior to any possible future implementation.