Created December 2009 from tests performed from October to December 2009.
Last updated 30 December 2009
<meta name="google-site-verification" content="Ae5dm5TbRZSehAVzYZBkmxXGkCwI5MsxczCGIhPb16I" />

There are a number of auto enthusiast websites about oil filters...

Usually these go no further than cutting the filters open and making a visual examination of their construction. There is some of that here, too, but the focus of this site is to compare the filter medias - the part that actually does the filtering. 

Filters are for filtering

Overall construction quality of the oil filter you use is very important, but I wanted to see how one filter compares to another in actual filtering and flow in a simple bench test.

The test

The media elements were cut out of the oil filters and submerged in a pan of deliberately contaminated motor oil. The contaminated oil was freshly and thoroughly stirred immediately before the elements were placed in to it. This "dirty" oil was allowed to flow by gravity from the outside to fill the center tube - the way that oil flows through it in actual use. Several filters were tested at once, all in the same pan, to keep things even. The filtrate that flowed through filter element was stirred in the center tube of the element, and a sample was drawn from about the middle of the element.
The motor oil used was clean, new oil and the contaminates in the oil were varying proportions of baking flour and talcum powder.
The key was to find a contaminate that would neither be completely filtered out, nor completely pass through so that there was something in the filtrate to compare.
I found that baking flour, talcum powder, or a mixture of the two, worked well.

Efforts were madi n in each series to test different brand filfters of popular part numbers intended for the same application, or at least of similar size. There is the possibility that different models from at least some filter makers use different media and would yield different results.

The test tubes that were used for the samples was 3/16" O.D. clear vinyl flexible tubing, 12 inches long. The filtrate samples were drawn into the tubing with a medicine syringe until they were completely full, and the ends of the tubes were stoppered.
After at least 48 hours, the test tubes were examined and compared for the relative amount of grit that had settled out of the oil. The elements were tested multiple times, both new and dry, and oily and dirty. Photos represent the best representative results.

Notes were also taken as to the relative time it took for oil to fill the filter center tubes, and adjustments were made to account for variations in center tube volumes to get an approximate comparison of which filter elements passed oil faster and easier.

What this test is, and is NOT:

This is not a test to give percent efficiency at a certain partical size. This test is not an official SAE or ISO test that filter manufacturers use to rate their oil filters.
This is just a simple bench test to give a visual comparison of how easily contaminants passed though the media elements with the oil, and approximately how easily cold oil passed through as well. Every effort was made to keep things even and fair within the limitations of the testing procedure. I had no favorites going into the testing. I believe the results are quite reasonably valid and seem to correlate well with the comparative filter efficiency ratings given by the manufacturers who publish these ratings. As mentioned above, different part numbers or models from the same brand and name may yield different results due to the use of different medias.

How to read the results

Since these filtration "runs" were done over a four month period and it was impractical to mix exact masses and proportions of contamination products and oil for every test, the amount of sediment in the tubes varies a great deal from test picture to test picture. This is a side result of starting with the idea of testing only a few popular brands all at one time for my own satisfaction. Thanks to interest, and encouragement  from my friends at the forums, these tests continued for quite a while.

Because of this, the important thing to note is that in any given run (picture), the viewer must observe which of the filter medias of the set tested at the same time "wins" so to speak.
I other words, if "Brand A" appears to filter better than "Brands B and C" and in another test "Brand D" is run against "Brand A" (again) and appears to outperform it, then "D" did the best so far.
Still with me here?

And finally, the disclaimer

Although I think it is important to have some guide for the filters that do not publish efficiency data, there are a lot of variables in my simple tests, including one-off defects or media leaks, and things I haven't thought of that may skew the results. I am not an engineer or an expert on filtration, and as mentioned above, this is a simple bench test. Yes, I use this to compare filters for my own selection and use, but it is not intended to recommend any filter over any other for your engine or to replace filter manufacturers' testing and data. Manufacturer data can often be obtained by contacting their technical departments. Furthermore, the general information in this website is true to the best of my knowlege at the time of writing. Some of it is what I consider common or background knowlege that I have read or discovered at some point, but cannot necessarily prove or disprove. And, of course, there is always the possibility of typos or other errors in this website.