Coiled Membrane Filter (CMF) vs. Straight Membrane Filter (SMF) Comparison Test
Because of the intrinsic difference in flow resistance, the CMF will always have a higher pressure drop compared to the SMF under identical flow conditions. This pressure drop also serves as a driving pressure to push the processed water to penetrate the membrane to generate permeate. Since the side-by-side comparison test between the CMF and SMF was performed using an identical type of pump (positive displacement pump), the results could give the impression that the CMF had a higher output (flux rate) compared to SMF simply because the driving pressure was higher. Another concern was that the presented charts for the comparison tests (such as the one shown below) did not start at the same performance level. This gave the impression that the test was biased to the CMF because for testing with brand new modules (both the CMF and SMF have the same number and the same type of fibers and the only difference is the configuration), they should display identical performance at the start of the test. These two concerns are discussed in this blog.
To resolve the issue regarding initial data points for the comparison test chart, a test with a brand-new 19HF09 CMF and a brand-new 19HF09 SMF using distilled water as feed initially and surface water later was performed side-by-side to investigate the effect of the boundary layer. The test was started with the distilled water on the first day (so the boundary layer is not playing a role). The distilled feed water was replaced by surface water (acquired from Budd Lake) in the second day. The test results are summarized in the table below.
Based on the distilled water test, the CMF flux is 3.6% higher than the SMF flux (233 gfd vs. 225 gfd). However, after normalizing the flux against the feed pressure, which is called the permeability coefficient, the CMF is 7.1% lower than that of the SMF (11.7 gfd/psi vs. 12.6 gfd/psi). Based on feed comprised of the surface water, the CMF flux is 57.6% higher than the SMF flux (208 gfd vs. 132 gfd). After normalizing the flux against the feed pressure, the permeability coefficient of the CMF is 64.7% higher than that of the SMF (8.4 gfd/psi vs. 5.1 gfd/psi).
A similar test, i.e. started with distilled water as feed and then switched to surface water, was performed in a different occasion, confirming repeatability. The same dramatic change in performance improvement was observed. For testing under distilled water, the normalized flux (permeability coefficient) from the CMF is lower than that for the SMF by 7 to 10%. This is most likely caused by the non-uniform pressure distribution in the CMF. Since the pressure distribution for a SMF is uniform, it is efficient for the permeate to penetrate the tube. While in the CMF, there is more flow fighting to come out from one side of the tube wall (outer side) than the other side. However, this effect is secondary and insignificant as soon as the boundary layer starts to play a role (and the permeate flow is much lower). The reversal of the Normalized % Improvement seen with the change in feed quality also proved that the Dean Flow advantage took effect as soon as the boundary layer started to build up.
The chart below is a revision of the last chart in terms of filter media permeability. The figure not only resolves the CMF "head start" concern but also emphasizes the advantage in permeability of the CMF, which is greatly related to the boundary layer improvement due to Dean Flow. Based on the flux (last chart) of the overall test period, the performance improvement was 94.2%; based on the permeability of the overall test period (chart below), the performance improvement was 170.7%. Based on the flux from the last wash period, the performance improvement was 114.9%; based on the permeability of the last wash period, the performance improvement was 177.3%.