Originally Posted By: Izb
After these UOAs (
http://www.bobistheoilguy.com/forums/ubbthreads.php?ubb=showflat&Number=2061820&page=3 ) I don't trust in Pure One's 99.9%.
P1's UOA had the worst particicle counts.
You cannot judge anything on one contamination analysis and you are drawing way too many conclusions from too little data.
1) The "particle counts" (technically they are not that.. read on) listed above are the pore blockage type... comparable to other pore blockage tests so don't anyone compare the number there to any other types. Pore blockage is the only type of contamination analysis done by Blackstone. It's generally the least accurate type as well because it uses a 10 or 15 um screen and calculates an ISO code via restriction and computer algorithms, rather than actually optically counting the particles as do the more expensive and accurate types. Pore blockage is useful in a broad sense to check on oil cleanliness but it isn't accurate enough to tell much about filter efficiency beyond "adequate" or inadequate." Lzb, since you are so good at find and reading the old posts, find some of mine from a coupla years back where 2010FX4 and I collaborated to compare the results of pore blockage vs a true optical test. Pore blockage is most accurate in the range of the screen size used, 10 or 15 um in thais case. From my experience at BStone, they most often use the 15 um screen.
2) Notice how the numbers are all over the place. The RP filter is significantly better at 10K than at 5K while the M1 is the same at both mileages, for example. This is very typical and is a combination of lab error, sampling error and differences in contamination rates. Notice how much dirtier the oil was in the early tests than it was later. The engine was still new at the start and still having some break in, so the contamination rates were higher. Note how much better the Amsoil did in the later test than the early (there is about a 50% difference in actual particles between one ISO code and another). Unless a vehicle was operated in a climate and dust controlled lab, there will be times when the contamination inputs are higher so depending on the exact moment when the sample was taken, the oil could be dirtier one day than another. Variables, variables, variables!
3) Finally, notice the oil analysis. The tested wear metals didn't change much, did they, no matter what filter was used? That should make a light bulb go off in a thinking person's mind.
Really, the only way to compare oil filters is using the exact same test regimen on each filter, the ISO 4548-12 for example. All the filters BTAnchors tested were largely in the same rated efficiency ballpark under ISO 4548-12. What we see in his great comparison is that all of them did an admirable job of keeping the oil clean and the difference are mostly uncontrollable variables. The test would have been MUCH more useful if he had done optical testing. I wonder if he has continued with his trending? The longer he did it, the more useful it would be but I imagine by now that he's settled on a particular filter.