Monitoring Filter Efficiency over time

Status
Not open for further replies.
Joined
May 24, 2009
Messages
1,227
Location
So Cal
Was thinking about this regarding filter efficiency in the real world. As part of UOA reports are particle counts and size. Is there a study that focus on samples taken over the span of one filter use that can graph particle count. I know Zee has a interesting graph of a test that was around an hour and a half long, but was thinking it would be easy to get real world on the road particle count on one filter.
 
That graph from Purolator/Mann+Hummel simply shows that some oil filter media can "slough off" already captured debris as the filter loads up and the delta-p across the media keeps increasing. That Machine Lubrication article you linked to also mentioned that same phenomena, which I pointed out in another thread LINK . IMO, some oil filters that rate low in efficiency per ISO 4548-12 most likely are not good at holding already captured particulate as the delta-p increases. If you know how the ISO 4548-12 is conducted then you'd see what I mean as the ISO test calculated the average efficiency over the test run, so if there is some decrease in efficiency as delta-p increases, then the overall ISO efficiency takes that into account.

I guess someone could do a test where they ran an oil filter and oil out to 15K miles for instance, and did PC samples every 5K miles to see how the particle count changed with time. They would have to run the same oil the whole time so they aren't flushing the system with a fresh oil change. They could do that with a low efficiency filter (50% @ 20u or 99% @ 40u) and then again with a high efficiency filter 99% @ 20u). Thing is, they would have to run both scenarios as identical as possible to try and keep any variables out of the equation that could skew the data. This is why laboratory testing can give answers under more controlled situations.

Another interesting test might be to run a low efficiency filter (50% @ 20u) for say 8K miles and do a PC test. Then swap out just the filter for a high efficiency filter (99% @ 20u), and do not add any more new top-off oil, then run it for another 2K miles and go another PC to see how much the better filter cleaned up the oil. Similar to the Frantz PC test.
 
I like that test you suggested Zee.. As long as the filter is still on the engine most of the particulates that enters a filter will still be there and what little make up oil to take the sample should not compromise the test. (IMO) There are variants with your suggestion of changing the filter too.

The ISO 4548-12 is one test. If I was to sell filters that would be a test I would target to look good at. If I took UOA data tests for a living and looked at all the data received, that might direct my choice on air and oil filters for my car.
 
Originally Posted by Bill_W
I like that test you suggested Zee.. As long as the filter is still on the engine most of the particulates that enters a filter will still be there and what little make up oil to take the sample should not compromise the test. (IMO) There are variants with your suggestion of changing the filter too.


Ideally, you would not want to add any more new oil because it would still skew the PC. Solid testing results are not achieved by changing multiple variables.

Originally Posted by Bill_W
The ISO 4548-12 is one test. If I was to sell filters that would be a test I would target to look good at. If I took UOA data tests for a living and looked at all the data received, that might direct my choice on air and oil filters for my car.


The ISO 4548-12 has been the oil filter standard used for the last 20 years, so that should say something. If there was a better test then the ISO test it would have been superseded by some other filter efficiency test method. If a filter manufacturer doesn't give any ISO test numbers then there is no way to know the filtering performance. It could be anyone's guess.
 
Status
Not open for further replies.
Back
Top