2016 F150 3.5EB, 4,498m with Schaeffers 5W-30 9000

Status
Not open for further replies.
Originally Posted By: Linctex
Originally Posted By: Trav
We read all kinds of post about filter efficiency where one filter takes a beating for its higher than FU micron filtering then we see this uoa with a torn filter no less.
Makes me wonder.


In a clean engine.... 99.99% of the time, the oil filter is "just along for the ride".


I am in that camp also and have posted about old VW air cooled engines with nothing more than a screen and seemed to hobble along nicely for years. My old 1961 HD 900 sportster (iron barrel 883) had a screen and had the original engine at 50K and 20 yrs on it.
Still, I wouldn't want torn media not only because of the filtration issue but if a small piece got loose and found its way into the oil galley the results could be catastrophic in short order.
 
Originally Posted By: DuckRyder
I have a few questions for the collective.

What is the stated purpose of a standard consumer UOA?
I don't know there is a "stated" purpose, but the general concept is to track wear trends.

How consistent is the typical consumer level UOA collection methodology?
Do you mean collecting the UOA material (oil) itself? Most take it at the drain; "dead pan" style (engine off). I take a "live" sample from the drain (engine running). Some sample from a live tap port. Some sample via tube in the sump. If you question is how consistent is the individual sample taker, I'd say fairly good. But from person to person, very inconsistent. There's a difference between repeatability and reproducibility

What particle sizes can a typical UOA pick up?
Most agree that 5um is the "max" spectral analysis will see. All the way down to sub-micron.

What particle sizes would correspond to serious, major or catastrophic wear?
UOAs can predict some doom, and yet completely miss other events. There is no 100% assurance that it will catch all. But it certainly can predict the onset, if the wear trends are known and recognized. There is no one-size-fits-all answer here

How many UOA do we have from vehicles with a known torn filter?
Few. Very few. Too few.

How many of those vehicles do we also have a similar prior UOA with a known good filter?
I'd say fewer than few ... A sub-set of a sub-set.

How many UOA with similar runs with a known good and subsequent known torn filter which also include particle counts do we have?
PCs are much rarer than UOAs. Hoping to find the trifecta (evidence of torn filter, UOA and PC all from the same OCI) is akin to finding a Unicorn.

Considering the population of known torn filters, what sample size would we need to make a statement with 95% certainty using an industry standard confidence calculator? 98% certainty? It would depend upon which mathematical model you used.

If we replace the terms "known good" and "known torn" with "high efficiency" and "low efficiency" what would our answers be?
I don't think we have enough data to delineate this.




What I can say with certainty is that, after reviewing more than 12,000 UOAs from a WIDE sense of equipment, application, environments, etc, the nuance of wear trend shift due to any particular filter used is so small that it cannot be discerned in typical equipment wear data. The variation due to "normal" (real life events) use is larger than the effects of filter selection; this from macro data analysis. Micro data analysis, (the one excellent example I have posted in my article), also shows that normal variation far exceeds lube and filter choices. However, just as with lubes, the OCI duration may well play into this. It is very likely that a premium filter would last LONGER due to certain construction characteristics. Just as a syn will last longer in service, so will a premium filter. But that only matters IF one extends the use out long enough. All my data shows that up to 15k miles, both the lube and filter choices make no statistical difference in macro analysis, concerning normality.

Whereas I have very few examples of filter media compromised combined with known UOA data (this UOA would be one of the few I know about), what's evident is that the wear data didn't shift due to the torn filter exposure. But, we have ZERO idea of how soon that void developed; was it early or late in the OCI? We'll never know. So we cannot really state anything past an assumption here.

The best way to "test" this condition would be to create an artificial void by inducing an unrestricted partial bypass of fluid, simulating a void. And then do controlled study of the wear trends.
 
Oil keeps the parts seperated no more no less . If brand X keeps the parts seperated and "lasts" through the interval brand Z with twice the everything will not do a better job until the brand X limits are exceeded. Oil filters catch the most damaging particalsin the oil. We can get to the point of diminishing returns and who has worn out a properly maintained vehicle to where the engine replacement cost isn't worth it due to the rest of the vehicle being worn out? In real life the air filter is the best oil filter.
 
Apologies in advance to the original poster for dragging this at least somewhat off the topic of the actual oil and onto the somewhat related topic of the torn filter. There are a couple of discussions along this line going on, in the general OA questions and also in the micro green filter thread, If Dave or other moderator wants to split this discussion off it might be more appropriate in one of those.

I won’t go quote by quote but I agree with his points.

One point that I’d like to elaborate on is the one that UOA can predict or detect certain wear and miss others. I believe that increased wear from a torn filter could well fall under the umbrella of wear particles that the UOA could miss. I base this hypothesis on the 5micron upper limit (note that I’ve also read 8 and 10 depending on method, and that the 5 is a hard cutoff but is a droplet vs sphere) as there could well be an increase in particles above the upper limit which could be classified as wear but not caught by the UOA.

Stated another way, I don’t believe that a UOA is the tool to measure the impact of a torn filter on the vehicle/engine.

Here is a thread that we might find of interest:

https://bobistheoilguy.com/forums/ubbthreads.php/topics/4131246/Amsoil_EA15K51_-_42,938_miles_#Post4131246

Please note that though I posted this, I only did so to host the images for DrDave, he is the one who had the evaluation done. I was somewhat disappointed that this did not go into further discussion. Unfortunately I surmise that this sort of evaluation may be more costly that can be reasonably justified, though I suppose that I could e-mail and ask.
 
Originally Posted By: DuckRyder
Apologies in advance to the original poster for dragging this at least somewhat off the topic of the actual oil and onto the somewhat related topic of the torn filter. There are a couple of discussions along this line going on, in the general OA questions and also in the micro green filter thread, If Dave or other moderator wants to split this discussion off it might be more appropriate in one of those.

I won’t go quote by quote but I agree with his points.

One point that I’d like to elaborate on is the one that UOA can predict or detect certain wear and miss others. I believe that increased wear from a torn filter could well fall under the umbrella of wear particles that the UOA could miss. I base this hypothesis on the 5micron upper limit (note that I’ve also read 8 and 10 depending on method, and that the 5 is a hard cutoff but is a droplet vs sphere) as there could well be an increase in particles above the upper limit which could be classified as wear but not caught by the UOA.

Stated another way, I don’t believe that a UOA is the tool to measure the impact of a torn filter on the vehicle/engine.

Here is a thread that we might find of interest:

https://bobistheoilguy.com/forums/ubbthreads.php/topics/4131246/Amsoil_EA15K51_-_42,938_miles_#Post4131246

Please note that though I posted this, I only did so to host the images for DrDave, he is the one who had the evaluation done. I was somewhat disappointed that this did not go into further discussion. Unfortunately I surmise that this sort of evaluation may be more costly that can be reasonably justified, though I suppose that I could e-mail and ask.



To some degree I agree with your assessment, at least in theory. Any void in the filter media will allow particles much larger than 5um to pass, and therefore would not show up in a UOA. Those particles will in turn have some significant size to them. Not all of them, but some of them.

However - I'm playing the anti-advocate here ... If large particles are now passing through that void, and getting to critical bearing areas and such, then the results of large particles abrading against other materials (15um FE against a Cu-Pb bearing, etc) will induce that attacked material to likely shed it's own particles, and very likely those would, at least in part, be small enough for a UOA to detect.

So while I agree that the filter void may not DIRECTLY impact via large particles in circulation as seen in a UOA, it is probable it will cause a cascade effect in creating other smaller ones that the UOA will see.

Simply put, a media void will pass ALL sizes of particles. But it would be bad logic to presume that larger particles passed downstream would ONLY begat other large particles. The likely result would be a ratio of both big and small stuff; X%large to Y%small particles will be created. So, it is VERY LOGICAL to believe that large particles passing the filter will generate both large and small particles upon contact, and therefore some "uptick" in wear metals in the UOA would reasonably be expected.


So if a torn filter is discovered and is associated with a decent UOA, can we not conclude that the void really didn't harm as much as some would fear?
 
Last edited:
I saw magnatec dexos 1 gen 2 on the shelf at wally world, I bet it'll do just as good as these more expensive oils. $17.88 last I checked.
 
Last edited:
Originally Posted By: dnewton3

So if a torn filter is discovered and is associated with a decent UOA, can we not conclude that the void really didn't harm as much as some would fear?


Questionable in my opinion. It might very well be so, but I don’t feel like we have enough information to make that leap.

As to the other points...

It would seem logical that decreased filter efficiency would have an impact on UOA results particularly between filters of radically different efficiencies, for example a Toyota service filter (Amsoil test @~55%) a Honda Service filter (A01 similar construction to Toyota, A02 ~65% per Fram/Amsoil) and a Fram Ultra or similar (~99%) yet these seem to make no difference in UOA results (BTW Jay says the Ultra is ~80% @5micron) so with a torn filter haven’t we just gotten a reduction in efficiency? How much? Does it tear all at once, a little at a time, right out of the box or later?

Unfortunately it would be pretty close to impossible to answer the torn filter question as to efficiency. It would be interesting if someone were willing to get PC’s on a vehicle that uses a likely tear filter routinely, but even then luck of the draw might never deal that person a torn filter.
 
I have been using Motorcraft filters exclusively since 2002.
I have never had a UOA with any kind of abnormality in that whole time.
 
Originally Posted By: DuckRyder
Originally Posted By: dnewton3

So if a torn filter is discovered and is associated with a decent UOA, can we not conclude that the void really didn't harm as much as some would fear?


Questionable in my opinion. It might very well be so, but I don’t feel like we have enough information to make that leap.

As to the other points...

It would seem logical that decreased filter efficiency would have an impact on UOA results particularly between filters of radically different efficiencies, for example a Toyota service filter (Amsoil test @~55%) a Honda Service filter (A01 similar construction to Toyota, A02 ~65% per Fram/Amsoil) and a Fram Ultra or similar (~99%) yet these seem to make no difference in UOA results (BTW Jay says the Ultra is ~80% @5micron) so with a torn filter haven’t we just gotten a reduction in efficiency? How much? Does it tear all at once, a little at a time, right out of the box or later?

Unfortunately it would be pretty close to impossible to answer the torn filter question as to efficiency. It would be interesting if someone were willing to get PC’s on a vehicle that uses a likely tear filter routinely, but even then luck of the draw might never deal that person a torn filter.



The SAE "bus study" (city buses with DD 2-stroke engines) pitted tighter and looser filters against each other and tracked the wear trends. They discovered a VERY CLEAR correlation between the PC data and the UOA wear metals. More particles (from the use of looser filter media) equated to more metals in the UOA; particularly Fe.

But does it really matter if the filter is "loose" (40um nominal) or has a void (a small hold in it)? At that size, regardless if the hole is intentional (loose pores) or unintentional (a void), the result is that heavy particulate re-circulation = heavy Fe wear evidence in the UOA.

And so, if a UOA does NOT show the wear trends affected, then it's a very logical to conclude that the filter made little difference in the wear, regardless of how the particulate passes (loose pores or a void).

You cannot have it both ways; can't have your cake and eat it too. You either believe that there is correlation between particulate load (size and quantity), which results in more wear, or you don't. If you do believe it is so, they you also have to believe that the study proved beyond doubt that the wear metals will track with particulates in circulation.

No change in wear rates? No harm done, despite the filter void.



Simply put this way ...
Because Fe wear data tracks with particulate load in circulation, then the absence of Fe wear spikes means there was no substantial damage from the filter void in this OCI and UOA we see in this thread.
 
Last edited:
Of course I’m going to eat my cake, what else would I do with it?
grin.gif


I’ll try to find the “bus study” and read it as well as the points and counterpoints (they have been arguing about that thing in the filter forum for as long as I can remember). However, I’m not sure that 2 stroke diesel bus = 4 stroke petrol light duty pickup truck. I don’t know anything about DD either, we didn’t get to talk to those guys in school or the GMTC.

However, even if I were to believe that “Fe wear data tracks with particulate load in circulation” when presented with a UOA with negligible change in FE (it does seem to have gone up the change before this one) and a torn filter I have to explain that somehow:

Some possible answers are:

A torn filter does not increase particles in circulation (not sure I buy that)
The filter tore very shortly before the FCI/OCI (that is possible - it also doesn’t mean that a torn filter is nothing to be concerned about)
The method of measurement is flawed (also possible)

or the base assumption that UOA Fe data tracks with particles is flawed...

I don’t know what the answer is, in fact that is my entire point...

I’m pretty sure we are never going to agree on this, but the longer I’m here and the more I see, the more I believe that unless your vehicle dumps all of the coolant in the oil or you see a BIG change in something that has a longstanding trend - UOA’s (at least at this level of sophistication) are pretty near entertainment only value.

So I guess maybe that is my point too.
 
The study certainly showed that Fe wear data correlated with particulate loading; hard to deny that.

And actually, I agree with you in part.

The study was done on old, dirty, sooty city buses with DD 2-strokes. Hence, there was plenty of particulate preset to remove from the lube. And so, the various filter choices (ranging from very loose to fairly tight) had the ability to show a disparity of particulate removal because those engine had a LOT of particulate in the lube. When there's lot's of stuff floating around, there's a lot to remove. And if it's not removed, then it's still present to do damage. Hence, high soot and particulate = more Fe wear data in a UOAs. The finer the filters, then he less particulate, and less wear.

But, IMO this actually tells us not one, but two things about today's filters and UOAs.
1) because our modern, fuel injected engines (both gas and diesel) run so much cleaner; because they run clean, there's not much to "clean up" by the filter
2) UOAs wear data certainly does track with particulate load

Hence, when there's very little particulate present in the first place, there's not going to be much difference in wear between minor filter differences, or even a void in the filter itself. Essentially, you won't see much of tell-tale Fe if there's not a lot of particulate to cause the damage from the get-go.

Remind me and I'll link it tomorrow if I get a chance.


UOAs are most certainly the low-cost, easily attained tool to track wear. Teardowns are an option; a very expensive, time consuming option and induce measurement errors upon reassembly.
 
Last edited:
Regarding the OP, Is it my imagination or are the copper, iron, and silicon numbers VERY high? Blackstone Shows no concern for that in their comments. I have this same motor, and certainly hope I get better results. Also, another member cast doubts on Blackstone's fuel dilution testing capabilities. Any links to such a discussion on that, or comments, would be appreciated...I've already got the sample bottled up, could send it elsewhere.
 
Last edited:
Good report. I like that it stayed in grade, and on regular octane gas nonetheless.
 
I use fram ultra filters for two reasons:

1. Higher probability they won't fail (fail defined as any number scenarios where the filter design failed to stand up to the application).
2. I can run them for two or even three OCI if I get lazy enough.

The filtering efficiency has little to do with my decision.
 
I have had a Fram Ultra fail, the leaf spring failed. The Fram element did not rattle when I put the filter on.
 
Originally Posted by Impatient
Regarding the OP, Is it my imagination or are the copper, iron, and silicon numbers VERY high? Blackstone Shows no concern for that in their comments. I have this same motor, and certainly hope I get better results. Also, another member cast doubts on Blackstone's fuel dilution testing capabilities. Any links to such a discussion on that, or comments, would be appreciated...I've already got the sample bottled up, could send it elsewhere.


My wear metals do appear to be higher than most EB UOAs posted here, but I'm still near the universal averages. I do have a fair amount of idling on my samples, so I think that's part of the problem. Comparing mileage between trucks and samples is just not an accurate apples-to-apples analysis...but it's what we got. Plus, a few of us have to be higher than average, otherwise the average would change.

Regarding the fuel testing capabilities of Blackstone, I honestly cannot give my own educated, unbiased opinion. Many here feel that their methods are not accurate. My level of Nerding here on BITOG has not taken me to the point of caring enough to educate myself on the various testing methods and their inherit limitations. Since this is an Internet forum, I think there are a few folks that understand how it works and then a lot of bobble heads just repeating what they read a few threads ago. I for one just use the flashpoint as an estimate.
 
Status
Not open for further replies.
Back
Top