"High end" oil filters

Status
Not open for further replies.
What is the case A -vs- case B ... people run an oil and filter 10k-12k and have for years. But what's wrong with changing the oil at 5k-6k and leave the filter for two OCI if it's designed for that capacity ... I'm going to bc 0w20 in a DI bothers me enough.
Some will say you have waisted money, some will think you're radical for leaving a dirty filter on ...
Choices
 
I think judging oil filter performance by cutting open a virgin filter is akin to judging oil performance based on virgin oil analysis. It gives you a rough idea and that's about it.
 
Originally Posted by seekingbuddha
Since some of the old oil stays behind in the engine (and in the oil filter, in case someone is using the same filter for 2 oil changes), there is likelihood that the sub 20 micron particles will stay behind in the engine for long time. So, in the case of an old car (say over 100 k or over), does it make sense to run at least one oil change with the best of the breed (Amsoil) filter even though it would cost more than twice as much as a M1 or RP ? Would this take out the sub 20 micron particles that are hanging around in the engine of an old car ? Does it really matter, if one keeps up with the recommended OCI ?
Maybe the new oil would neutralize the old residue in the engine ?

There are always sub 20 micron particles in the engine..but unless the oils dispersants have been depleted (something that's not likely to occur) the majority of sub 20 micron particles will be held in suspension and come out during a D&F. What's "left" in the engine is not something I would lose much sleep over. And if I'm correct, running an approved extended drain filter, actually does a better job of capturing these smaller particles the longer it's in use. But like most things there's a tipping point, so follow the filter mfgs recommendations.
 
Originally Posted by Mad_Hatter
And if I'm correct, running an approved extended drain filter, actually does a better job of capturing these smaller particles the longer it's in use.


Hard to say about that since there is data that shows some filters become less efficient as they load up because the increased delta-p makes them slough off (shed) already captured particles. IMO, filters that are rated higher have a better chance of holding their efficiency throughout their use period because the ISO efficiency rating is an average over the loading life of the filter. Filters that rate low in the ISO efficiency test aren't very efficient, and/or they shed more captured debris towards the end of their useful life as the delta-p increases from debris loading. The ability of the media to securely hold captured debris is an important aspect of the design of the media.
 
Originally Posted by ZeeOSix
Hard to say about that since there is data that shows some filters become less efficient as they load up because the increased delta-p makes them slough off (shed) already captured particles. IMO, filters that are rated higher have a better chance of holding their efficiency throughout their use period because the ISO efficiency rating is an average over the loading life of the filter. Filters that rate low in the ISO efficiency test aren't very efficient, and/or they shed more captured debris towards the end of their useful life as the delta-p increases from debris loading. The ability of the media to securely hold captured debris is an important aspect of the design of the media.

This thread (which you participated in) references studies by GM and filter mfgs, oil filters are at their peak filtering at the beginning of the service cycle and the end.

https://www.bobistheoilguy.com/forums/ubbthreads.php/topics/3622904/2

What I've read elsewhere online is basically that there's a finite amount of +20 micron sized openings in the filter media. As those openings become caked over (and exhausted), the sub 20 micron size particles have fewer and fewer openings through which they can pass freely on through, the filter has effectively become "more" efficienct at capturing the lower micron size.

Now there is a tipping point at which the filters performance characteristics will be adversely effected. (flow goes down, pressure goes up... risking the media blowing out and/or the oil going into bypass which case nothing is filtered)
 
Last edited:
Originally Posted by Mad_Hatter
Originally Posted by ZeeOSix
Hard to say about that since there is data that shows some filters become less efficient as they load up because the increased delta-p makes them slough off (shed) already captured particles. IMO, filters that are rated higher have a better chance of holding their efficiency throughout their use period because the ISO efficiency rating is an average over the loading life of the filter. Filters that rate low in the ISO efficiency test aren't very efficient, and/or they shed more captured debris towards the end of their useful life as the delta-p increases from debris loading. The ability of the media to securely hold captured debris is an important aspect of the design of the media.

This thread (which you participated in) references studies by GM and filter mfgs, oil filters are at their peak filtering at the beginning of the service cycle and the end.

https://www.bobistheoilguy.com/forums/ubbthreads.php/topics/3622904/2


Yep, you mean like the "hockey stick" shaped efficiency curve Purolator described to me in an email which I mentioned in that thread - see the graph below, which is also from Purolator.

When the efficiency increases towards the end is when the filter is nearly completely clogged and the delta-p is increasing quickly. IMO, that's not exacrly where I'd want to run my filter to make it more efficient right before the bypass valve is ready to open and stay open. ...¬

[Linked Image]
 
Mad hatter - what you say makes sense. The way you describe is true for air filters - so one would naturally assume that it holds good for oil filters also. As the pores get clogged, perhaps they would create smaller pores, and hence capture the smaller particles. ....... unless there is something about oil that would tend to keep these very very small particles in suspension, rather than getting caught in a net. Perhaps engineers who specialize in this field ( like ZeeOSix ?) would know the test results from labs. I am not losing sleep over these details, cause this should not be a concern for an average car owner - perhaps those in racing might care. There is plenty of other aspects in a car that could go wrong, before these issues play a role on engine.
 
Thanks for that chart, ZeeOSix. It would be wonderful to see a chart like that published here for Mobile 1 Extended, Royal Purple Extended and Bosch. Would make comparison much easier.
 
Originally Posted by seekingbuddha
Mad hatter - what you say makes sense. The way you describe is true for air filters - so one would naturally assume that it holds good for oil filters also.


The pressure drop across an air filter when highly loaded is very small compared to the pressure drop across a highly loaded oil filter. Air filters might have a 1~2 psi (1 psi = 27.7 in-H2O) delta-p vs around 15 psi delta-p (typical bypass valve setting) across a nearly clogged oil filter.

Put way more than 1~2 psi across an air filter and it will probably start shedding some of its captured debris too. The large delta-p across loaded oil filters is what causes some of them (depending on media design) to shed some of the already captured particles as the graph above shows.
 
Originally Posted by ZeeOSix
Yep, you mean like the "hockey stick" shaped efficiency curve Purolator described to me in an email which I mentioned in that thread - see the graph below, which is also from Purolator.

When the efficiency increases towards the end is when the filter is nearly completely clogged and the delta-p is increasing quickly. IMO, that's not exacrly where I'd want to run my filter to make it more efficient right before the bypass valve is ready to open and stay open. ...¬

[Linked Image]



Couldn't agree more with you. I guess my point is that it technically occurs, a spike in efficiency - just before "failure".(don't think that's the best word for it but meh)....but it's one of those, just because you can doesn't mean you should situations - running the filter right up to the end of it's performance life. Like I mentioned I don't do extended drains for a number of reasons, so it's not something I lose sleep over.

Here's a question Zee.. how much of a concern is "particle population" for sub 20 micron particulates on engines doing extended drain intervals??? If a filter has a less than 50% efficiency at the critical 5~10 micron range, won't the oil be slowly accumulating (net gain) particles in that range, over the extended drain interval? Seems to me that despite the marketing of extended drain oils by the oil companies, not all vehicles are a good candidate for this. If say your car has a problem with blow by and particulates entering the system (ingress), should you be doing extended drain intervals at all???

I would think these are vehicles that should stick to 4~7.5k intervals to avoid growing the very damaging 5~10 micron particle population.

Originally Posted by seekingbuddha
I am not losing sleep over these details, cause this should not be a concern for an average car owner - perhaps those in racing might care. There is plenty of other aspects in a car that could go wrong, before these issues play a role on engine.

Ditto and agreed 👠for the most part. It could become an issue over the long term in re engine wear. Those 5~10 micron particle size accounts for something like 70% of engine wear at critical points (boundary lubrication, tight clearance) like the main bearings, rods, cam lobes, bearing journals, valve guides, piston/cylinder walls etc..
 
Last edited:
Not that I disagree with the hockey still mantra; it's plausible.

But the GM filter study from back in 98 showed odd results. They used 4 different filters of nominal values (40um, 25um, 15um, 7um). The test method was to run those filters until they reached a dP of 20psi, and then the filters were changed. What they discovered is that all filters essentially exhibited a performance result where any media used essentially ended up being about 10um in effect. All filters converged to about a point where the particles stopped were 10um and above, and anything below that was passing regardless. That's kind of interesting, but also perplexing.
40um initial ends up at 10um at dP 20 psi
25um initial ends up at 10um at dP 20 psi
15um initial ends up at 10um at dP 20 psi
7um initial ends up at 10um at dP 20psi
It's that last one that perplexes. How can a filter "open up"?

I have some theory on this. (theory only, just my guess) There must be a size of particle prominent in the test that had an affinity to move about freely, but just barely. The dP was enough to make the particles just below that size break loose, perhaps? And we don't really know what was happening when the dP was only 10psi; they didn't report it. All they did was change filters at dP 20 and then note the typical size of particle stopped via PC analysis. My take away from this is that no matter what efficiency filter you select, it's going to perform reasonably close to it's nominal value efficiency. As it ages it will get more efficient, but it might take LONGER for the more porous media to build up that cake layer. But they all converge as they load up to a similar dP. Most of our filters have a BP relief around 10psi or so (with some notable exceptions). So the GM test really does not apply here (for a lot of reasons besides dP), because we'd be in perpetual BP long before we even saw 20psi dP. Would the four filters still converge at 8 psid? Perhaps, but it might take longer? I, for one, have no intention of ever running my engines under the conditions of the GM filter study, so it's results are totally, completely, utterly worthless to me. I don't run my filters to a 20psid, and I don't run my oil to a 570k OCI. Hence, that study is not what happens in my engines.


But, as you all know, filters are an input to the wear. I care about wear rate results. What I've seen in my data streams is that filtration efficiency makes no difference in wear rates when the sumps are reasonably fresh in a modern engine.

It's completely true to say that a higher eff filter won't hurt an engine. But the real world data also shows it really does not help, as long as you're using it in a typical "normal" OFCI. The lab tests we all discuss don't really manifest into results in our garages. There's no correlation exhibited; none whatsoever. And without correlation, there can be no causation.
 
Last edited:
Originally Posted by dnewton3
Not that I disagree with the hockey still mantra; it's plausible.
..........................

7um initial ends up at 10um at dP 20psi
It's that last one that perplexes. How can a filter "open up"?

..................................

I care about wear rate results. What I've seen in my data streams is that filtration efficiency makes no difference in wear rates when the sumps are reasonably fresh in a modern engine.




A 7 micron filter will not tear open to 10 micron level, unless the media was bad quality OR the pressure was so high that it can not deliver the promised 7 micron filtration. But that is irrelevant to me if i consider your closing comment (seen above). Your closing statement has the following implications :
  • Doing shorter oil changes (thus keeping fresh sumps) is better than extended oil changes (which is great for those who don't have time to attend to cars)
  • Paying 3 times more price for a premium filter is not needed, if one sticks to shorter oil changes. Simply buy the cheaper filter and use the new one, rather than sticking to a reuse of the premium filter (for 2 or 3 OCI)
  • If the premium filter is capable of filtering better at sub 20 microns, that does not translate to longer engine life (because of shorter OCI)


When i went through the specs of Royal purple and a cheaper alternative (like Purolator or Bosch), there was a difference in their efficiency at sub 25 micron levels. The other big difference was in the filter media (as can be seen at Synthetic media in RP, Amsoil, M1 and WIX filters only ). Mobile 1 extended is actually a premium blend, not pure synthetic.


Perhaps an average car owner should not have to be so concerned about all these technicalities ? ( even in the case that he wants to run the car to 200 k miles ) ? Because the wear rates will not be substantially different when 200k (or even 300k) is reached ?
 
My personal opinion is that not all cars are good candidates for extended drain intervals. Just because you can, doesn't mean you should. If you have a monster of a contaminant producing engine, the particle population in the 5~10micron range (the most damaging size) is going to continue to build over that extended interval, to levels very damaging to the engine...[if] you're not filtering them out faster than they're created/entering the system.

If I were inclined to go beyond a 7.5k oci, i would do UOA's w/particle and insolubles counts.. at least for the first time I ran the oil beyond 7.5k.
 
Originally Posted by dnewton3
Not that I disagree with the hockey still mantra; it's plausible.

But the GM filter study from back in 98 showed odd results. They used 4 different filters of nominal values (40um, 25um, 15um, 7um). The test method was to run those filters until they reached a dP of 20psi, and then the filters were changed. What they discovered is that all filters essentially exhibited a performance result where any media used essentially ended up being about 10um in effect. All filters converged to about a point where the particles stopped were 10um and above, and anything below that was passing regardless. That's kind of interesting, but also perplexing.

40um initial ends up at 10um at dP 20 psi
25um initial ends up at 10um at dP 20 psi
15um initial ends up at 10um at dP 20 psi
7um initial ends up at 10um at dP 20psi
It's that last one that perplexes. How can a filter "open up"?

I have some theory on this. (theory only, just my guess) There must be a size of particle prominent in the test that had an affinity to move about freely, but just barely. The dP was enough to make the particles just below that size break loose, perhaps? And we don't really know what was happening when the dP was only 10psi; they didn't report it. All they did was change filters at dP 20 and then note the typical size of particle stopped via PC analysis. My take away from this is that no matter what efficiency filter you select, it's going to perform reasonably close to it's nominal value efficiency. As it ages it will get more efficient, but it might take LONGER for the more porous media to build up that cake layer. But they all converge as they load up to a similar dP. Most of our filters have a BP relief around 10psi or so (with some notable exceptions). So the GM test really does not apply here (for a lot of reasons besides dP), because we'd be in perpetual BP long before we even saw 20psi dP. Would the four filters still converge at 8 psid? Perhaps, but it might take longer? I, for one, have no intention of ever running my engines under the conditions of the GM filter study, so it's results are totally, completely, utterly worthless to me. I don't run my filters to a 20psid, and I don't run my oil to a 570k OCI. Hence, that study is not what happens in my engines.


My take is since they pushed the delta-p to 20 psi they were probably well into the nearly totally clogged realm of the filters. If these filters were described as "nominal xx microns" of 40, 25, 15 and 7, then I'm assuming that's 50% efficiency ("nominal") at those micron levels, and that was when the filters were new I would assume. If so, then the Purolator example graph above would be around 12 microns nominal when it was in new starting condition.

If you look at the end of the Purolator graph where the delta-p is increasing due to clogging, and the different particle size efficiencies are drastically inceasing, it looks like that filter ended up being around a 7 micron nominal filter (I'm extrapolating the particulate size lines). So it doesn't seem like a big surprise that if any filter is pushed to nearly 100% clogged state that it's all of a sudden going to become very efficient. I think that's basically what happened in the GM filter testing where they pushed the delta-p to 20 psi. If they would have collected data like in the Purolator graph as the filter loaded up and the delta-p continued to increase from the initial new condition baseline, it would be much easier to see the correlation between filter loading and it's effect on efficiency for different particle sizes, and how captured debris is shed as delta-p increases which will lower the efficiency.

I agree with you dnewton, I'd never run an oil filter way past it's rated use from the manufacturer. If I was doing long OCIs or running a filter for 2 or 3 OCIs, the total mileage put on the filter would still have to be no more than what the filter was rated for.
 
seekingbuddha and Mad_Hatter ... I pretty much agree with you guys, as I've said in the past that a higher efficiency oil filter is more important on longer OCIs in order to capture as much debris in the 5u to 20u range. Short OCIs dump the sump of particulate, so filter efficiency becomes less important as the OCIs become shorter. An extreme case example is if someone had no oil filter at all and changed oil every 500 miles. The particulate in the oil would be drained instead of captured by the oil filter. On the other end of the spectrum is commercial trucks (semis, etc) that run large sump capacity and use a 2u bypass filtering system so they can run the oil for huge amounts of mileage (25~35K miles).
 
On the other hand, someone could make the following argument regarding Extendend Range oil filters like M1 or Royal purple:

These filters are designed for 15k miles of usage, and hence they will reach a peak efficiency in filtering sub 25 micron particles, as they load up on in-solubles over time. So, one could run the first oil to half the range (7.5 k miles) and then do an oil change, keeping the filters in place without removal. During the second run, these filters would perform even better in the sub 25 micron range because they are already pre loaded with particles from the first run. So, the pores would have gotten smaller and be able to trap particles faster this time around. Thus, the filter can be run for their rated 15k miles - perhaps even a third time as someone here suggested before, with peak efficiency. Only an UOA would confirm this theory, but sounds good on paper.
 
Last edited:
Originally Posted by 01_celica_gt
call me crazy but that was some really bad info and comparison but decent pictures.

In my opinion, your diagnosis and opinion (not facts) of this so called "filter comparison" is god awful...

DITTO. Totally bad info. Wow!
 
Status
Not open for further replies.
Back
Top