UOA's and oil brands

[dnewton3]

Here's a very interesting read from Cummins; it's dated early 2004, before the new CJ-4 rated oils were out, but still I think a lot of the info is very interesting, and probably very valid still.

Nearly all of this is a good read, but the reason I'm posting it was to draw your attention to a particular topic. I believe that while swithing brands of oil is not a bad thing, it does skew UOA analysis, perhaps heavily.

For those who try Rotella, then Delo, then Amsoil, then Delvac, then Schaeffer's ... you might not be able to draw any direct comparisons regarding oil brand performance capabilities. Specifically look at the link I'm posting; read near the bottom, Appendix A, sub-section 1, paragraph 7. It states that abnormally high lead or copper readings from an otherwise solid performing engine may very likely be due to the changes of chemistry between oil brands reacting with the engine metals, and not due to some detrimental damage.

In other words - pick a brand you like; any brand name good quality oil will do. Stick with it. Then do continued UOA's using the same brand/viscosity. If you live in a harsh environment where condensation and temp changes are extreme, then maybe choose a wide ranging synthetic. If not, run a dino 15w-40 of your favorite brand and stick with it.

If you're posting UOA's here after rotating oils at every OCI, you're likely not giving a true indication of the oil protection or engine wear patterns.

Again, I don't think you're hurting your engine by switching all the time, but the data you're posting is misleading if you want to claim brand "x" didn't protect as well as brand "y", but better than brand "z".

My interpretation of this link, regarding UOA's, is that if you want to change oils, do so under a controlled long-term experiment. Run maybe 35,000 miles with 5k samples, then try it again with a different brand for another 35k miles. If you drive long-haul, you might have to run 200k on each (sample at 10k)to establish a pattern because you're not seeing all the seasonal changes with 35K. Notice, I didn't say you had to OCI at 5K, just sample at 5k intervals . Extended OCI's are a great money saving tool. But again, whatever you do for one oil, do exactly the same for the other oil.

Switching oil brands every 5k or 10K miles isn't telling anyone anything of value. You're not hurting your vehicle, but you're not helping establish a statistical trend either. You'll not have a clear indication of how well (or not) your engine/oil combo is doing if you switch all the time.

Here's the link. http://www.cummins.dk/fileadmin/dokumenter/Pdf_filer/Cummins_Litteratur/Olie_3810340-04.htm

Read it all please before you comment. As usual, I'm always up for a good debate! GA - where are you?

 

[buster]

Quote:

---

This often results in dramatically increased levels, often ten times, of copper or lead in used oil. Increased levels from this source is not reason for excessive concern. These components will become passive after a few oil changes with different oil. Wear metal levels will then slowly decline back into the normal range for the engine.

---

I think it makes sense and definitely applies to some brands when switching. Thanks for the link!

 

[D-Roc]

Very true, but it diminishes the interest of trying new brands.

 

[dnewton3]

Oh - I whole heartedly agree that it takes all the fun out of swapping brands.

By the grace of media programming, we've become an impatient society that expects instant results; unfortunately, this (apparently) just aint' the case with swapping motor oils.

My point was that when you do this frequent oil swapping, whatever "truth" you were looking for in performance of oil protection (by brand) and engine wear rates (for your particular engine) is likely skewed, so why do a UOA in the first place if you're constantly swapping oil?

Heck - we can swap oil all we want - we won't hurt anything. But we're not getting the "truth" due to our immediate need for gratification. In fact, I think it's fair to summize, according to the Cummins bullentin, that there is no sense in doing a UOA after you swap brands, because it takes a few flushes with that same oil to get back to "normal".

I've got a great idea! I'll start a UOA business my self. I'll have a disclaimer that says "swapping oil brands more often than 50K miles will result in a generic boilerplate UOAs being generated; all will be identical." And yet, I'll still get some business!

 

[D-Roc]

At least when swapping brands and doing uoa's you still get a better picture of wear trends than if not doing an uoa at all.

 

[dnewton3]

Agreed, to a large extent. Even if the Cu and Pb go up, there are other wear metals to look at that may not be effected by the frequent brand swapping. Plus all the "chemistry" to compare, such as TBN contributors. And then there is the insoluble/soot info we get as well.

I'm not saying a UOA is worthless, but according to the manufacturer of your engine, swapping brands frequently isn't going to give a true representation of wear rates.

One could argue that if you constantly swapped oil at each OCI, all other things being equal, you might get some "trend" of wear by the shear nature of consistent inconsistency. Kind of like the old saying - "The only constant is change."

 

[wjdell]

If you switch brands and get good results on the first UOA, then you can be reasonably assured that the oil is working good for you. However if you switch brands and the first UOA is poor or mediocre, that does not necessarily mean that oil is bad for your motor. Best to run three intervals to get a better representation.

 

[Triple_Se7en]

Quote:

---

If you switch brands and get good results on the first UOA, then you can be reasonably assured that the oil is working good for you. However if you switch brands and the first UOA is poor or mediocre, that does not necessarily mean that oil is bad for your motor. Best to run three intervals to get a better representation.

---

Getting good results on the first UOA could also indicate the previous oil was more potent - thus not using as much with the new brand. Sometimes, small doses of increased consumption with a new brand is a good thing initially.... then slowly reverses itself back to very little or no consumption.

So what you think is good - may not be good long-term.
So what you think is bad -- may not be bad for long.

 

[Towel_Rail]

Very interesting. I was under the impression that oils meeting a certain API certification were totally compatible and would not react with each other. Perhaps the requirement is just that they don't react *catastrophically*...

- Scott

 

[allan5oh]

you mean they dont go boom?

 

[Tim]

And this is why I have never done a UOA on the first OCI after a switch. I always wait until the second.

 

[Flywest]

To my mind thats a lotta of [censored] frankly.

IF - UOA's could be relied upon to be accurate - then maybe the article would have some validity..

How many people here send identical samples to more than one lab to cross check what they are being told?

I am probably one of the few and top be truthful you'd think they had tested oil from two different vehicles and two different brands because the results were chalk and cheese.

To then turn around and suggest some chemical created from mixing oils is responsible for elevated wear metals is frankly [censored].

Its more likely the elevated wear metals were because the lab technician was smoking weed that day or sampled someone else's oil or creatively made up the results as a guess because he couldn't be bothered actually doing the tests.

As someone who graphs all their results and has changed oil brands as well - and does this for 3 vehicles at a time - with multiple labs used, I've come to the conclusion UOA is a crock of [censored] and not worth the money it costs because non of the results on the same oil from different labs are the same.

I've had labs mix up the samples between my vehicles for example, and then blame the postman for maybe dropping the sample kits - the lids coming off and he put them back in the bottle with the wrong paperwork etc.

Quote simply the lab testing we pay for - if it were medical related - you'd be getting a leg amputation for an ear infection because the results aren't trustworthy or reliable.

Don't believe me?

Start sending two or three or more identical samples to different lab and start graphing the results alongside each other, and you'll find out what I did!

UOA sample testing is a ripoff and a fraud because the results are not reliable or trustworthy. Its something the Oil Cos provide because they have too, but they pay peanuts and have monkeys doing the tests!

If you rely on UOA for deciding your engine maintenance or upgrade schedules - you could be spending $$$ to fix problems that don't exist.

UOA's from companys like CAT who want to sell you a new machine are some of the WORST - wouldn't surprise me f the new machinery sales dudes deliberately sabbotage the results just to convince you you need a new machine.

My 2C!

Cheers

 

[bismic1]

I have only run duplicates on two sets of oil change samples. I used Blackstone and MRT Labs. One sample was on Motorcraft 15W40 and the last one was on Schaeffer 9000 5W40. There were definately differences. However, I do not think the differences were enough to cause me to draw different conclusions. The results are below (assuming I can paste - this is my first try). I look forward to what ya'll think.

Note:
Blackstone on the 1st and 3rd analysis; MRT on 2nd and 4th.

Motorcraft 15W40 on the 1st and 2nd and Schaeffer 9000 5W40 on the 3rd and 4th.


Miles on Truck 14723 14723 19811 19811
Miles on Oil 2901 2901 5088 5088
"National
Avg." Oil change # 5th 5th 6th 6th
3. Aluminum. 3 4 3 4
1. Chromium. 1 2 1 2
22. Iron. 21 31 23 32
3. Copper. 2 4 3 6
3. Lead. 1 0 2 2
1. Tin. 1 0 1 0
31. Molybdenum. 2 0 195 282
0. Nickel. 0 1 0 1
0. Manganese. 0 na 0 na
0. Silver. 0 0 0 0
0. Titanium. 0 0 0 0
4. Potassium. 7 0 2 0
33. Boron. 2 2 6 4
11. Silicon. 13 17 13 15
3. Sodium. 3 8 6 15
3103. Calcuim. 2282 1869 3141 2676
83. Magnesium. 7 11 7 11
1120. Phosphorous. 950 1219 1155 1233
1288. Zinc. 1074 1207 1390 1550
2. Barium. 1 3 0 0
Vanadium. 0 0
Antimony. 35
65 to 76SUS vis 210. 68.3 68 72.1 69.5
11.6-14.8cSt at 100 C. 13.51 12.9
>410 Flashpoint. 400 390 415 400
0 Antifreeze. 0 0 0 0


(The formatting changed after I posted - I will try to post a better table)

 

[dnewton3]

Sending oil samples to different labs is like comparing apples to oranges. I agree that you'll likely get different results; in fact, I would expect nothing less. Different labs have different people, machines, etc.

I work in quality engineering for my company. I do statistical analysis quite often. I can tell you that ANY process is only as good as the inputs. As they say: Garbage in, Garbage out.

By quality definition, there is a difference between "repeatability" and "reproducability". To truly understand how good a UOA lab is in it's processes, you would have to follow this doctrine. Collect a LARGE sample of oil from a LARGE sump system, so that the samples total volume represented only 10% of the total sump volume, yet able to be divided into a minimum of 30 samples. Then send the 30 samples for analysis, to the same lab. And make sure the lab uses the same person for those tests, on the same machine, on the same day, etc. ONLY THEN can you determine how good the lab is in terms of statistical capability.

Now, since most of us don't have the money or time to do as such, we can best limit the chance for errors by limiting or reducing variables. Choose a lab you like and use it all the time. That eleminates one variable. Then, collect the samples the same way every time. And so on ...

With that in mind, and since I'm the OP, I'd point you to the fact that the link was to a Cummins statement concerning oil selections; a very detailed statement in fact.

I don't own, nor do I have much experience operating Cummins products. But, I have emense respect for them and their products. I would presume, given their quality certifications and experience, that they do a reasonable job of quality/quantity ascertation when producing a doccument such as this. This statement is most likely a summary of years of research and collaboration. I, however, do take the whole thing with a grain of salt. There are fabulous and terrible workers at any company.

This paper is what it is. I posted it because I believe there is validity to the basic concept of UOA's, but that there is also inherent limitations regarding UOA's.

The point of my orignal post was that if you change oil brands, there is likelyhood that you're skewing your UOA's until the oil chemistry and engine metals come to some cooperative understanding.

 

[Jim Allen]

dnewton3: Good analysis.

Flywest: You are obviously bitter about some bad experiences there in Australia. Justifiably so, it sounds like. But your conclusion there, in whatever part of Australia in which you live, with whatever sloppy companies you dealt with, do not necessarily translate to me here in Ohio, USA. Your comments do serve to keep me vigilant.