Iron in M1 UOA's?

Status
Not open for further replies.
"I've calculated the variations as proposed in the article and they are less than the variations that we measure."

I guess this explains as well as anything else about how credible your conclusions are, as the article doesn't mention anything about dealing with variation due to sample size, samples being representative, etc.

The article just proposes how to deal with some variables affecting metal concentrations in oil, and chooses to ignore others. A metric is proposed, but it specifically mentions that 'statistically significant' samples are supposed to be used, and leaves the definition of significant to be addressed elsewhere.

In summary there is nothing in the article that you presented to address the problem of 'stastistical significance', but you seem to think that it does and as a result seem to have missed the point completely. The article shows that there is so much variation engine to engine that one cannot distinguish the groups of engines from each other, instead one would need to consider each engine on it's own, as is stated. This is similar to what has been suggested in the 'article of the month' posted for the site.


http://www.obitet.gazi.edu.tr/makale/makale/internalcombustionengines/047.pdf

4. Evaluation parameters
The compensated wear rate obtained represents the wear behaviour of an engine in a defined period of time, but the consequent question is “What level of wear rate is normal or abnormal for the engine studied?”. Answering this question will be the final target of a condition monitoring system for predictive maintenance based on oil analysis. At this point, it is important to take into account the important factors mentioned before: engine age, metallurgy, type of service, environmental conditions, etc.
A comparative parameter is proposed to answer this question. The parameter is defined as:

Z = EwrE¯ wr / EMwr, (14)

The reference wear rate represents the normal wear rate values for a determined sample population (from an engine or engine model). These are defined as statistically significant from statistical studies with a large population of samples. These reference values are not static and they evolve with the number of samples analysed and take into account certain restrictions [10].

The expression of Z takes into account not only the deviation from the engine reference wear rate, but also by using the engine model reference wear rate, compares the current situation to a larger population behaviour such as all the engines of the same model. The value is also transformed into a non-dimensional magnitude.

With sufficient historical data, limits for Z parameter can be set by the user (fleet owner, oil analysis laboratories, etc) to reflect maintenance goals. Hence, maximum benefit can be gained by using oil analysis as a predictive tool for condition monitoring.
 
1st,

so, if I understand you correctly, you have no issue with the measurement methodology and accuracy thereof of what the authors term the engine reference wear rate. Where you have issues with is determination of the engine model reference wear rate, and the statistical significance of it's determination.

Is that correct?
 
Last edited:
As kind of a summary, statistics basically allows one to make better bets, but one needs to be careful about the assumptions used. Significance is only important when one needs to be more confident about the bet that one is making, due to high costs, bad failure modes, life, limb, etc.

If there is little downside to a bet then one can dispense with 'signifance' and just do what looks right with the data at hand, like most people seem to do with their UOAs. This might be engineering, maybe not, but as mentioned previously using a $20 UOA doesn't mean that one is 'doing science and thus critiques are unwarranted'. Using a UOA doesn't mean that somehow Fe in a UOA is automatically correlated to wear, and that any number of conclusions can be made with one, or a hundred.

It's a hobby, no need to treat a UOA like it's the Holy Hand Grenade of Anticoh, to be used on the heathen unbelievers;

"First shalt thou take out the Holy Pin, then shalt thou count to three, no more, no less. Three shall be the number thou shalt count, and the number of the counting shall be three. Four shalt thou not count, neither count thou two, excepting that thou then proceed to three. Five is right out. Once the number three, being the third number, be reached, then lobbest thou thy Holy Hand Grenade of Antioch towards thy foe, who being naughty in my sight, shall snuff it."
 
1st,

I understand statistics reasonably well as an electrical engineer, but I'm much better versed in stochastic processes for signal processing.

What I have a problem with is your immediate discount of "a $20 UOA" ... In my case for the study I reference, the UOA's were not $20, as they were done through Terry's ISO certified lab, which has correlation to ASTM reference standards. However, in either case (Dyson, Blackstone or other) we can assign error bounds for the actual metrology being performed. I believe both you and I can agree with that. What remains is our understanding of what the usefulness of the measurements are.

As for Fe being correlated to wear, I would qualify this only slightly, it is correlated to loss of material in the engine. For an engine that has an aluminum block, aluminum cylinder walls, and a high percentage of other aluminum components, there are few sources of Fe other than bearing material. In this particular engine, the Audi RS4, if Fe is not correlated to wear, then what pray tell is your theory about what it is correlated to, since it did not appear from vacuum fluctuations? Feel free to propose a better theory.
 
".... In my case for the study I reference, the UOA's were not $20, as they were done through Terry's ISO certified lab, which has correlation to ASTM reference standards. "

It wouldn't matter if NIST had performed the measurements, it's the perception that 'science' is being performed because a UOA was used, thus all conclusions using UOA data are scientific. You then provide a link to an article on using UOAs to estimate wear rates as proof that 'science is being done', using UOAs in an attempt to determine 'what oils produce less wear', but the article by itself in fact shows the oposite, that the there is so much engine to engine variation in UOAs that one can't do much more than attempt to develop better methods for trending an engine.
 
"...For an engine that has an aluminum block, aluminum cylinder walls, and a high percentage of other aluminum components, there are few sources of Fe other than bearing material. In this particular engine, the Audi RS4, if Fe is not correlated to wear, then what pray tell is your theory about what it is correlated to, since it did not appear from vacuum fluctuations? Feel free to propose a better theory."

The engine is linerless, the block seems to use a high silicon content for structural and thermal stability as well as wear resistance. If rings are wearing at high rates one would expect to see some high rates of other metals too, like chrome. It has large cast iron bearing blocks / caps, something commonly seen in higher load engines with alloy blocks, but it doesn't appear to be the bearing material as the cast iron parts are not stated as such. The DOHCs have roller followers. The main source of Fe appears like it could be the four timing chains, and my experience with chains is that the critical wear points are the hardened rollers to the hardened pins, and not the softer link plates.

Fe in a UOA isn't conclusive that it's due to wear over sampling period, as some have observed higher Fe levels with no engine operation. Low Fe levels in a UOA have sometimes come engines with catastrophic failures where the particles were largely being filtered out. Fe levels can also increase when an oil breaks down varnishes / skudges left by other oils, so again we have higher Fe levels not due to increased wear. There is always wear of some sort going on, but without a teardown or other indicators like other metals. particle counts, etc., if may be relatively benign, or localized, changing with load, etc., so again I don't automatically by into 'lower is better'.

In fact this thread, 'why does the evil Mobil 666 have higher levels of Fe', still has the problem of addressing why so many makers of high performance vehicles continue using M1 as a factory fill in spite of higher Fe levels, makers who have tjhe benefit of doing teardowns when qualifying oils. M1 is apaprently common among some types of racers too, people who also have the benefit of doing teardowns to see how different oils perform.
 
"Where you have issues with is determination of the engine model reference wear rate, and the statistical significance of it's determination."

I have issues with using engine UOA results to do more than determine if a specific engine is having an issue. Even large sample sizes won't address problems of biased sampling, non representative sampling, and the inability to block individual sources of variation. In a basic 'experiment' one would need to select a sample of vehicles and use a sample of oils in each vehicle over a period of time. Even then one is still left with trying to correlate some observed difference with some apparent difference in wear.
 
For a moment 1sttruck, invert his UOA results. Suppose RLI was the factory fill and recommended fluid for use.

If he then switched to M1 0w-40 and saw spikes in Fe to the degree that he does (beyond some sensible standard deviation from "noise"), what would be your sensible conclusion??
 
Looks like bearing shells are being dispensed with. I'd expect higher 'wear in' on the thrust bearing(s) than on the bearing journals. Since the area is a lot less than what would be on a cylinder liner it's still not clear wha twould be the larges tsource if Fe, the timing chains or the crank and main bearings. I'll guess the timing chains.

http://www.sae.org/automag/global_viewpoints/02.htm

Like BMW, Audi also uses vermicular graphite cast-iron for the cylinder block. Instead of having individual main bearing shells, Audi opted for what it describes as "an extra-strong main bearing frame," which incorporates the five bearing shells, and is supported at each side and bolted from beneath at several points.
 
"Suppose RLI was the factory fill and recommended fluid for use.

If he then switched to M1 0w-40 and saw spikes in Fe to the degree that he does (beyond some sensible standard deviation from "noise"), what would be your sensible conclusion?? "

No different than stated.

The 'standard deviation' is one of the items under question, much less unbiased sampling, variation over time, etc.,

there is still the possibility of the new oil cleaning up a mess left by the previous oil,

perhops the new oil is exhibiting normal levels of wear while the previous oil was producing hign impact failures not detected by a UOA,

etc.,

and even if there higher levels of wear one needs to determine where it is and under what conditions.

I'll trust oil choices based upon long trials with tear downs over a UOA without the same.
 
Can I take that to mean that you would continue to use an oil that kept producing inordinately high levels of Fe (over many OCI's ..since you don't think UOA is telling you anything, even on the 3rd order level, on wear) ..while other oils would not ..as long as the UOA showed that it was suitable for continued use?
 
Originally Posted By: Gary Allan
Can I take that to mean that you would continue to use an oil that kept producing inordinately high levels of Fe (over many OCI's ..since you don't think UOA is telling you anything, even on the 3rd order level, on wear) ..while other oils would not ..as long as the UOA showed that it was suitable for continued use?





I believe Doug did..... And still had an immaculate tear-down.

Whilst BuickGN had fantastic UOA's and no bearings.
 
I imagine that Doug did on his OTR applications. With any of those, you have two spec'd oils (from OEM approved lubricants). So it's either dino or synth.

The wear is what the wear is
21.gif


Whilst BuickGN?? Surely possible. I guess if he does see Pb ..he can just ignore it and hammer away, right?


..but the question was would 1sttruck (or you for that matter) continue to use an oil that produced substantially higher numbers when others did not? Long enough to eliminate all the picked nits that he came up with for their possible origins??

In RS-Audiguy's situation, just invert the usage and show that he switched to an oil that had obvious aggravating properties to it and let me hear anyone say "Good choice! Steady as she goes! Higher indicators surely shows that it's the better selection!!"

..and keep in mind that no one is going to tear down and engine and measure it. This puts the debate in an unreachable box for all but a very few people. Whether directly related or on some tertiary or inferential level, if I see more lint in the lint trap ..it came from somewhere. Even if I know what piece of clothing is shedding the cotton ..but the more that is there, something has less of it from wherever it came from.
 
Originally Posted By: Gary Allan
I imagine that Doug did on his OTR applications. With any of those, you have two spec'd oils (from OEM approved lubricants). So it's either dino or synth.


I believe Doug did testing for Castrol, Shell and Mobil using OTR trucks, yes. He was able to run D1 out to 44,000Km IIRC and DID perform tear-down testing. This was in concert with frequent UOA's to determine remaining oil life and contaminant level. But from my understanding, his experience in this field is not limited to OTR trucks, I believe this is just where he accrued the most mileage.

Quote:
The wear is what the wear is
21.gif



Not according to Redline apparently.....?

They cite (as has been quoted on here) chemical leaching as a source of Fe in UOA's.

Quote:
Whilst BuickGN?? Surely possible. I guess if he does see Pb ..he can just ignore it and hammer away, right?


I believe the issue was that he did NOT see Pb. He didn't see anything. The UOA's showed "good to go". IIRC, the particles being shed were too large to be picked-up by the UOA......


Quote:
..but the question was would 1sttruck (or you for that matter) continue to use an oil that produced substantially higher numbers when others did not? Long enough to eliminate all the picked nits that he came up with for their possible origins??

In RS-Audiguy's situation, just invert the usage and show that he switched to an oil that had obvious aggravating properties to it and let me hear anyone say "Good choice! Steady as she goes! Higher indicators surely shows that it's the better selection!!"


I don't know. I started to get into the big 'ol UOA thing and then just decided not to continue. I don't need the anxiety. I have enough experience and am familiar enough with my own junk that I'm just going to continue on doing what I'm doing; it has helped me accrue substantial mileage and left me with fantastic oil pressure, clean engines and no discernible wear. I figure if the Expedition wears like the Mustang, it's going to outlive me.

Quote:
..and keep in mind that no one is going to tear down and engine and measure it.


Doug did
wink.gif
I've had both of my 302's apart, but no measurements were made. BuickGN tears his apart almost annually
grin2.gif


Quote:
This puts the debate in an unreachable box for all but a very few people. Whether directly related or on some tertiary or inferential level, if I see more lint in the lint trap ..it came from somewhere. Even if I know what piece of clothing is shedding the cotton ..but the more that is there, something has less of it from wherever it came from.


But as per Redline, what if it is just leaching dye from the cotton?

There are a lot of questions. I consider Doug an expert on this topic. If he says a UOA is not good at measuring engine wear, I believe him. He's been doing it longer than I've been alive, has been published and written papers that are regarded as authoritative on the subject.
 
"..but the question was would 1sttruck (or you for that matter) continue to use an oil that produced substantially higher numbers when others did not? Long enough to eliminate all the picked nits that he came up with for their possible origins??"

I have already answered that; I would use an oil recommended by a number of makers of higher performence vehicles, an oil that appears to have been selected with the benefit of tear downs, over an oil that has not and only appears to have the benefit of lower metals in a UOA. Close the loop, do the tear downs, demonstrate that some group of oils has lower wear, lower deposits, over the loads likely to be encountered.

That's just the UOA - wear correlation (we haven't mentioned deposits / sludge / varnish yet), as there is still the problem of demonstrating it in a significant manner.
 
This has been posted before, it's from the Mobil 1 site and is a list of vehicles where M1 is a factory fill. I don't see a list of approved use by make, which will be longer.

Why are these makers using an inferior oil, one that doesn't have lower wear metals ? Why don't these makers just do a UOA and pick a better oil ?


"In addition to being in all Porsche engines, Mobil 1 is original equipment (it is installed at the factory) in:

Acura RDX
Aston Martin DB9 and DBS
All Bentley Vehicles
Bristol Fighter and Bristol Fighter S
All Cadillac Vehicles
Chevrolet Cobalt SS S/C Coupe
Chevrolet Corvette C6 and Z06
Chevrolet TrailBlazer SS
Chrysler 300C SRT-8
Dodge Charger SRT-8 and Viper
All Holden HSV
Jeep Grand Cherokee SRT-8
Mercedes-Benz AMG Vehicles
Mercedes SLR McLaren
Mitsubishi Evolution and Lancer Evolution FQ400
Nissan GT-R
Opel GT
Pontiac Solstice GXP
Saab 9-3 TTId
Saturn Sky Red Line
Vauxhall VXR8"
 
Originally Posted By: Gary Allan
Whether directly related or on some tertiary or inferential level, if I see more lint in the lint trap ..it came from somewhere. Even if I know what piece of clothing is shedding the cotton ..but the more that is there, something has less of it from wherever it came from.


Hi Gary,

I like the lint trap analogy! However, what really matters is the total volume of lint generated, not just what is in the lint trap. If some of the total generated lint passes through the lint trap into the vent pipe, and some remains in the dryer, then the volume of lint in the trap alone may not directly correlate to the total volume of lint generated.

Likewise with iron in an engine. To gauge wear one must measure the total quantity of iron generated. If some of this iron is in the filter, some is tied up in sludge & varnish, and some remains dissolved or suspended in the oil, then just measuring the iron in the oil (UOA) may not correlate to total iron generated.

In fact, IF M1 dissolves or suspends some iron containing sludge or varnish, or IF it generates fewer large iron particles for the filter to trap, then less total wear can actually give higher iron in the oil as seen in the UOA.

Since we cannot measure the iron in the filter or in the sludge & varnish, we must take UOA iron with a grain of salt. I wouldn't disregard it altogether, just recognize that judging an oil's wear properties by UOA metals alone may be misleading.

Tom NJ
 
My belief is that the iron is coming from the camshaft lobes and lifters and is directly related to the performance on the Seq IVA tests.
 
Well said Tom.

It's just too complex for the average consumer to be able to determine the "best" oil based on inexpensive oil analysis testing. It's very limited. Useful, but limited. The engine sequence tests and OEM tests companies put their products through are very stringent. Some of them exceed anything a consumer could replicate.

There are too many variables and other factors at play. To single out an element or specification is the first mistake people make. It's the complete package that counts. Oils have to do many things.

The M1/Fe issue has been discussed since 2003. 6 years later M1 continues to get more endorsements and high praise. Engines are not failing. Same can be said with RL. I'm sure these companies know their products and capabilities.
 
Originally Posted By: chevrofreak
My belief is that the iron is coming from the camshaft lobes and lifters and is directly related to the performance on the Seq IVA tests.


Perhaps this is also directly related to the increased noise many people hear when people switch to Mobil 1?
 
Status
Not open for further replies.
Back
Top