UOA - Usless Oil Analysis?

[buster]

This isn't a thread about comparing Mobil Delvac 1 to Amsoil 0w-30 (two different oils) but just an example of why oil analysis can be misleading. Ferrography really should be used to determine more information. I think it would be worth the added cost to get this done if you are doing this type of thing once a year.

Quote:
I previously reported the invaluable aid ferrography was for a series of engines that was prone to camshaft failures. Most recently a group of Mack truck engines that had faulty cams. Spectro showed absolutely nothing; we were not able to detect which trucks in the fleet had the defective cams. With ferrography we were able to pinpoint the bad cam engines immediately. You could not even see through the slide as there was so much metal! Yet reported iron levels on spectro were the same, sometimes even lower, as the non problem engines..
George Morrison, STLE CLS



Quote:
Spectorgraphic tells us NOTHING about the type of wear that is occuring. We are comparing FERROGRAPHY of engines to determine HOW THE WEAR VARIES FROM ONE OIL TO THE NEXT. This type of test IS APPLICABLE to what is being done here.


Quote:
Look at both of the spectroanlysis oil samples taken. Based on those you would not suspect that the 0w30 engine would have any problems because the wear metals are so low. The fact is the type fo wear that is occuring is severe. The oil is working but it is not preventing dangerous types of wear from occuring.


LINK - OLD TDI thread

 

[98LSC32V]

Interesting... makes me wonder if I have been wasting money doing UOA's on my cars.

 

[Gary Allan]

Quote:
Most recently a group of Mack truck engines that had faulty cams.


And this would have altered oil selection ..or been altered by oil selection?

The process that George mentioned was good for finding faulty camshafts. I assume that it's good ..period.

Not that I'm bucking the notion that it's not lacking in a few things ..but if I had a cam failure, that UOA would have showed that the oil was just fine. Which it was.

Ferrography appears to essentially be a particle inventory.

 

[Doug Hillary]

Hello,
Buster - this was covered some years ago when some Forum Members of that time were comparing lubricant performance from UOA. Then, I raised the overall relevance of UOA - and, I made the point that unless trended with an individual engine or within a family of engines and over considerable time/distance they have little relevance

UOA is an excellent tool when used to monitor a lubricant's condition. It may assist in determining a component change interval when the engine's metallurgy is known and can then be used to monitor some aspects of engine wear performance. UOA can assist as a diagnostic "instrument" with such things as coolant leaks and etc

UOA is IMHO and experience NOT a device by which lubricants of similar specification can be compared as to engine component wear rates

The overall UOA issue is quite complex but much better (and costly) facilities exist by which individual lubricants can be compared as to component wear rate performance

When used at its limits UOA is valuable in setting economic OCIs and monitoring - as a snapshot - the health of the lubricant at a particular time

One Poster on a Porsche Forum has endeavoured to establish individual lubricant/wear performance from UOA results and some engine teardowns - this is of course always subject to so many variables as to simply only become "interesting conjecture"

If only it was this easy!!!!

Regards

 

[Merkava_4]

It would be nice if each individual engine part had its own unique genetic marker for determining the origin of individual wear particles.

 

[Vilan]

Would there be a more convenient way for car owners to find this out? Maybe a magnetic drain plug would've collected enough iron to be noticeable? Or maybe a blotter test would show some iron particles in it?

 

[buster]

Quote:
RUBBING WEAR LEVELS ARE NORMAL BUT SEVERE WEAR IS SLIGHTLY ABOVE NORMAL. MOST
SEVERE WEAR IS SLIDING AND SCUFFING ALUMINUM (POSSIBLY FROM PISTON SKIRTS). DIRT


Quote:
WEAR AND CONTAMINATION LEVELS APPEAR NORMAL. ONLY TRACES OF SEVERE WEAR WERE FOUND.
RESIDUE IS LIGHT. CONTINUE NORMAL PM. RESAMPLE AT NORMAL INT


Fe was roughly the same in the two samples, but the more severe wear was occurring with the 0w-30 (first quote). Delvac 1 was reducing the more severe wear. I'm assuming you could only know this type of wear pattern by using ferrography. A $20 doesn't appear to have been doing a good job detecting the more important wear aspect of this engine.

 

[remmious]

As mentioned UOA's are a good way to monitor oil health and can be an indicator for other potential problems but are general not comprehensive enough to determine all failures. I look at the UOA's here mostly to see how well an oil maintains its integrity within certain environments (driving conditions, engine types, etc). Remember that the wear particles measured in a UOA are under 5 microns. You need to do particle counting to determine how 'dirty' or significant the amount of debris is in your engine. Your UOA can come back with low element count but you could be wearing/falling your motor because of the type of wear and size of particles.

The short version.. UOA's are just a part of monitoring failures but still a good & easy way to trend oil life and machine health.

 

[remmious]

Another thing to consider is sample variation. I see this from time to time at my place of work when I have different techs collecting oil samples. Even though we have precise procedures and the guys are trained there can be times when the sample is compromised and not known. It shows up in the UOA though and we then resample to validate our findings.

Now most of the people that are getting into UOA here are fairly new to this so I would imagine there could be a lot of sample variation. That's why I usually recommend taking a number of samples of the same set up before making a number of changes.

 

[buster]

I would love to see more of this testing done on oils that showed higher than average wear metals to see what type of wear is actually occurring.

 

[JAG]

Outstanding thread. I need to print off the TDI thread info because a quick look at it revealed that it requires a dedicated look to fully absorb the info. Thank you buster and those here who made good posts.

 

[RI_RS4]

Buster

There are different types of UOA's, and methods for spectro, FTIR and other forms of analysis, so lumping them all together is a bit of a stretch. To say, or imply that the results are "Usless" (sic) is just not true. It is a measurement, which has a range of accuracy, repeatability and usefulness, just like any measurement. It just depends upon what you are measuring, as with any tool or measurement instrument. One would not use a yardstick to measure bearing dimensions, nor would one use a micrometer to measure a bolt of fabric for a pattern.

In some cases, Spectro analysis of oil will tell you important things about the performance of the oil and engine, and in some cases it will not. To say that Spectro analysis does not find a specific problem, and there for all spectro analysis is useless, is just plain silly. It would be akin to running out of gas when your gas gage is on 1/4, and then stating that gas gauges are useless.

It sounds that you have read an email from a person that you trust (Roy at Redline) saying that UOAs are not useful for comparing the wear performance of different oils, and now you are scrambling to find other like statements, to bolster your belief. The fact is, no one measurement is the be all and end all of measurements. They have to be used in conjunction with other techniques. And yes, in some cases you can correlate UOA wear measurements to engine wear and/or problems, and thereby determine the relative merits of different oils. In some cases you can not. Just as in some cases, UOAs can be used to predict failure, and in others it cannot. Chrysler engine test engineers have a very nice SAE paper on just this subject and justified the expense of real time UOAs on engines during dyno testing.

BTW, there are more additional analysis methods than just Ferrography, including particle count analysis, and dielectric constant analysis, to name a few. In newer engines with very little ferrous metal, different methods need to be used.

 

[Pablo]

Useless? NO.
A good argument for new and better analysis techniques? YES.

Great reason for bypass filtration? Priceless!

 

[buster]

RI,

I don't necessarily trust what Redline told me, nor am I looking to bolster my belief. It's something that I've always wondered based on many different responses on the accuracy of oil analysis.

Oil analysis is definitely useful. My thread heading was basically to grab attention to the subject.

I do believe that oil analysis is detecting wear at "some" level. But, what type of wear is it? Redline is what initially made me question oil analysis. It's either an unbalanced oil, or as Roy said, not wear. Mobil 1's avg Fe is higher than their own conventional oils. Does Mobil 1 offer less protection? Just doesn't add up really. IMO.

 

[buster]

Quote:
During the course of the engines life you would monitor the regular spectrographic analysis to watch for elevated wear metals. But since the regular analysis will not indicate the severity of the actual wear this is where ferrography comes into play.

Almost all conventional oils show very low wear. Are they better?

 

[BarkerMan]

I see the other end of this stuff, the teardowns. There are some surprises with you get inside, things you don't expect because the oil analysis was not understood or the noise that everyone knew was piston slap, everyone, turned out to be a rod bearing or a wrist pin. I have seen dyno runs with just about every oil that exists over the years and the only common thread I can see is that large oil sumps are the best thing you can do for a high performance engine or an engine that you want to run for a long time, not synthetic, not special filters, not additives, not thick oil, not thin oil, just a larger sump. I know a lot of you place your faith in some special oil and have all the knowledge about oil to back it up but my opinion comes from the insides of engines I see every day. We do a lot of development for what I can expedition vehicles that don't have to produce good torque and run in the worst conditions you can imagine, all with little or no service. If there was a magic bullet oil we would use it. We built 20 vehicles for a company in Africa and these vehicles have to go 20k miles between service and have to be bullet proof. One thing they all got was a dry sump oil system that holds 5 gallons of oil and a bypass filter system that holds another gallon. We have built about 60 vehicles in 10 years for them and we have a good idea about how to make them last and oil capacity is very important.

 

[Gary Allan]

Quote:
Now most of the people that are getting into UOA here are fairly new to this so I would imagine there could be a lot of sample variation. That's why I usually recommend taking a number of samples of the same set up before making a number of changes.


I've pondered this a bit here. One would sensibly assume that none of us use clean room sterile conditions ..nor surgical technique (although I'm sure, after a bit here, we try) ..yet all of the UOA samples that I see are typically void of the gross contamination and radical variances that one would expect if coincidental contamination was a factor. There has to be some.

That is, trends don't seem to be upset. Even one time users tend to fall within UA unless some aggravating condition is also present.

This leads me to believe that the resolution at the >5um eliminates most intrusions of sampling. That and (perhaps) anything that we're likely to add isn't even read.

 

[levi]

...the only common thread I can see is that large oil sumps are the best thing you can do...

Another poster with the handle of Lonnie used to maintain the same thing. Interesting.

 

[Pablo]

Originally Posted By: levi
...the only common thread I can see is that large oil sumps are the best thing you can do...

Another poster with the handle of Lonnie used to maintain the same thing. Interesting.

Totally agree with this - the more oil volume the better...within reason.

 

[Gary Allan]

Well, assuming that the oil selected is up to the task, a larger sump would surely buffer any insult. Just the fact that you're degrading less of % of the sump in any one stress event will have impact. Any temperature swings will be slower. Peaks will take longer ..and with a large enough sump, will probably be lower if your event duration is not too long. It dampens everything.

Assuming continuous operation over extended hours over various levels of stress ..a larger sump wins hands down, imo. All things being equal, the fatigue on a smaller sump will simply occur sooner. Nothing supernatural about that. If you're operating the smaller sump'd engine into continuous duty in the upper margins of its operational capabilities ...then it's going to fatigue quicker and whatever fractional/marginal wear that occurs in those extreme conditions will accelerate in accumulative impact.

The same could be said of cooling systems in the same service. I threw in a 50% thicker rad on an electrically cooled engine ..the duty cycle of the fan was cut by 60%. Temp fluctuations were flattened ...over shooting was eliminated ..etc..etc.