dnewton3
Staff member
UOA article - what is "normal"
- Thread starter dnewton3
- Start date
Excellent article and very usefull. One question I have is that I presume most of the folks that do oil analysis on their vehicles use fairly good quality oils, or did you include results from Wallyworld dipstick special users? In other words did you see very different results from folks using strange label oils that were of the correct viscosity??
It would also be interesting to see results from vehicles using the incorrect type of oil, because Blackstones must get quite a number of 15/40 users sending in samples for engines where a 5/40 is recommended for example.
I use the best Liqui Moly synthoil, BUT once I have confirmed my service schedule is correct I am changing to some cheap conventional oil, like Castrol GTX 10/40 or Liqui Moly Diesel 5/40 to see what the UOA results look like, so the standard deviation info was very interesting and Blackstones were even good enough to do a VOA for free as they don't see many oil types only sold in Germany.
It would also be interesting to see results from vehicles using the incorrect type of oil, because Blackstones must get quite a number of 15/40 users sending in samples for engines where a 5/40 is recommended for example.
I use the best Liqui Moly synthoil, BUT once I have confirmed my service schedule is correct I am changing to some cheap conventional oil, like Castrol GTX 10/40 or Liqui Moly Diesel 5/40 to see what the UOA results look like, so the standard deviation info was very interesting and Blackstones were even good enough to do a VOA for free as they don't see many oil types only sold in Germany.
dnewton3
Staff member
The total macro-market variability takes into account everything from bulk drum oils to syns. Yes - Wally's oil is in there. As best I could, I discounted out any bizzare lube that had no rational reason to be there.
There are a couple key points to take away from the article.
1) to really know what is "best" for any rig, you MUST run micro-analysis. That takes a LOT of time and money; more than most any person (even a BITOGer) wants to put into it
2) the variability in macro-market response is such that most anyone falls into "normal" definitions
Unfortunately, some folks are going to read into that article stuff that I never said, or see bias in it. I certainly am vocal about people wasting lube (both syns and dinos). But I think it's clear as to WHY I'm like that. Both real world data and SAE studies show that lubes can go way futher than people push them.
My analysis shows that dinos are every bit as capable as syns, when the OCIs are "normal". Now, that would NOT be true if the OCIs were greatly extended (way past the "normal" limits). But since most folks never operate out that far, my data cannot ascertain the limit yet. Synthetics most certainly can outperform conventional lubes, but that break-point of delineation is SO FAR OUT THERE that 99% or more of folks just never see it.
There are a couple key points to take away from the article.
1) to really know what is "best" for any rig, you MUST run micro-analysis. That takes a LOT of time and money; more than most any person (even a BITOGer) wants to put into it
2) the variability in macro-market response is such that most anyone falls into "normal" definitions
Unfortunately, some folks are going to read into that article stuff that I never said, or see bias in it. I certainly am vocal about people wasting lube (both syns and dinos). But I think it's clear as to WHY I'm like that. Both real world data and SAE studies show that lubes can go way futher than people push them.
My analysis shows that dinos are every bit as capable as syns, when the OCIs are "normal". Now, that would NOT be true if the OCIs were greatly extended (way past the "normal" limits). But since most folks never operate out that far, my data cannot ascertain the limit yet. Synthetics most certainly can outperform conventional lubes, but that break-point of delineation is SO FAR OUT THERE that 99% or more of folks just never see it.
Last edited:
Thanks for that and it will be interesting to see what my own UOA results yield when I change to dino next year, but I am going to do one more change with the same oil to get a rough idea of my average and deviation first.
One other question about the statistics, is did you try looking only at filter statistics in terms of wear metals for a certain type of filter after averaging out the OCI's and oil types so the only variable was the filter used??
One other question about the statistics, is did you try looking only at filter statistics in terms of wear metals for a certain type of filter after averaging out the OCI's and oil types so the only variable was the filter used??
dnewton3
Staff member
As for the filters, that is part of the "normal" variation in the macro analysis. Typically, UOAs don't acknowledge which filter is in use, unless someone directly mentions it.
But I can tell you from other data I have, that filtration selection in "normal" OCI durations has very little effect on the wear. As long as the filter used meets or exceeds the OEM spec, the wear rate isn't nearly as much affected by the filter as it is the tribochemical anti-wear layer.
There is an (in)famous filter study by GM logged in the SAE catalog. If you read the abstract, you would be lead to believe that "better" filters reduce wear. And, in that sole experiement, there was indeed corelation (but not causation). However, to bring to light the filter disparity (differences in performance), GM had to HEAVILY dose the sump with contamination past what any sane person would experience, and they also completely negated the OCIs by not changing oil. That sound "normal" to you? Lastly, near the end of the study, GM acknowledged that normal UOA wear metals would NEVER get this high, and so differentiation between filters would never be realized in the real world. Why? Because three things affect wear:
1) filter
2) OCI
3) Add-pack of the oil
In their expereiment, they had to eliminate the two (OCI and add-pack) to focus on filter efficiency. But that is NOT normal in the real world.
And I would point to my article and the two examples where fitler AND lube were contrasted. In the Vulcan engine example, the syn/premium filter did NOT outperform the dino/normal filter. In my Dmax example, the syn with bypass did not outperform the dino with normal filter. Why? Because the OCIs were normal, and the severity of exposure was not drastic enough to manifest into a tangible wear difference!
Yes- filters are important. No, you typically don't see real world wear differences in "normal" OCIs by using a premium filter.
If you really want to prove it one way or another, you'll have to do 30 micro-analysis UOA samples of a normal fitler, and then 30 more UOA samples of micro-analysis of a premium filter, all while holding all other inputs as steady (lube selection, useage, etc) the same. That takes HUGE amounts of time, money dedication and patience; things the average anal-retentive, A.D.D., impatient BITOG oil changer simply do not possess. And before anyone gets upset over that description, realize that I, too, was once one of the synthetic-frequent-OCI junkies at one time, too. I can poke fun where I once trod.
But I can tell you from other data I have, that filtration selection in "normal" OCI durations has very little effect on the wear. As long as the filter used meets or exceeds the OEM spec, the wear rate isn't nearly as much affected by the filter as it is the tribochemical anti-wear layer.
There is an (in)famous filter study by GM logged in the SAE catalog. If you read the abstract, you would be lead to believe that "better" filters reduce wear. And, in that sole experiement, there was indeed corelation (but not causation). However, to bring to light the filter disparity (differences in performance), GM had to HEAVILY dose the sump with contamination past what any sane person would experience, and they also completely negated the OCIs by not changing oil. That sound "normal" to you? Lastly, near the end of the study, GM acknowledged that normal UOA wear metals would NEVER get this high, and so differentiation between filters would never be realized in the real world. Why? Because three things affect wear:
1) filter
2) OCI
3) Add-pack of the oil
In their expereiment, they had to eliminate the two (OCI and add-pack) to focus on filter efficiency. But that is NOT normal in the real world.
And I would point to my article and the two examples where fitler AND lube were contrasted. In the Vulcan engine example, the syn/premium filter did NOT outperform the dino/normal filter. In my Dmax example, the syn with bypass did not outperform the dino with normal filter. Why? Because the OCIs were normal, and the severity of exposure was not drastic enough to manifest into a tangible wear difference!
Yes- filters are important. No, you typically don't see real world wear differences in "normal" OCIs by using a premium filter.
If you really want to prove it one way or another, you'll have to do 30 micro-analysis UOA samples of a normal fitler, and then 30 more UOA samples of micro-analysis of a premium filter, all while holding all other inputs as steady (lube selection, useage, etc) the same. That takes HUGE amounts of time, money dedication and patience; things the average anal-retentive, A.D.D., impatient BITOG oil changer simply do not possess. And before anyone gets upset over that description, realize that I, too, was once one of the synthetic-frequent-OCI junkies at one time, too. I can poke fun where I once trod.
dnewton3
Staff member
I cannot speak for all labs, but I'm familiar with Blackstone in particular. The "universal averages" (UAs) are simple math averages. ("n" number a samples added together, and divided by "n"). No lab we would commonly use here (Blackstone, Polaris, Cat, etc) that I'm aware of does any of the statistical processing, nor filtering of abnormalities.
I have plenty of 4-cyl data; most of it cars and not motorcycles. I can assure you that the basic tennets of my article apply to most any ICE, generally.
I have plenty of 4-cyl data; most of it cars and not motorcycles. I can assure you that the basic tennets of my article apply to most any ICE, generally.
Last edited:
Your article is saying that assuming nothing goes wrong, like coolant leaks or air filter leaks, leaving the oil and filter for 15,000 miles will yield less engine wear than 5000 mile OCI's?
dnewton3
Staff member
Not less wear; lower wear rates. I look at "wear" as a total count of the ppm, whereas "wear rates" take the wear totals and divide by the exposure. 10ppm of FE in 3k miles is a much higher wear rate than that same 10ppm of FE in 10k miles. Many wear metals will stagnate; often Cr and Cu and Pb will be low regardless of the exposure if the engine is in good shape. But the Fe will very often rise with the exposure, but the rate at which it wears will drop precipitously to very low level, often an entire order of magnitude down.
Absolutely proven, in thousands upon thousands of UOAs.
Also shown to be true in the SAE article I reference (2007-01-4133).
Now, there are a lot of built in assumptions to my positional statement here. Equipment must be in good working order, maintained appropriately, etc. Also assumes that the product choices are not inappropriate for the application, etc. But the overall data from the macro data shows this to be so very clear and true. And I have over 10,000 UOAs in my database across a very broad and diverse set of applications (cars, trucks, gassers, diesels, tractors, motorcycles, gearboxes, trannies ...)
The key to understanding this topic is the fact that OEM OCIs are set very conservative as a protective manner for them; they don't want warranty claims to be paid out. Maintenance costs them nothing; they don't pay for OCIs - we do. So it is a win-win for them; they state a conservative OCI to protect their risk, and we pay for privilege of following their dogma. Extended OCIs have shown to be very safe, when well managed, even using "normal" products. But I would NEVER, EVER advise someone to do this without understanding the risks, and being very involved in the overall maintenance program (UOAs, PCs, observations, etc).
Absolutely proven, in thousands upon thousands of UOAs.
Also shown to be true in the SAE article I reference (2007-01-4133).
Now, there are a lot of built in assumptions to my positional statement here. Equipment must be in good working order, maintained appropriately, etc. Also assumes that the product choices are not inappropriate for the application, etc. But the overall data from the macro data shows this to be so very clear and true. And I have over 10,000 UOAs in my database across a very broad and diverse set of applications (cars, trucks, gassers, diesels, tractors, motorcycles, gearboxes, trannies ...)
The key to understanding this topic is the fact that OEM OCIs are set very conservative as a protective manner for them; they don't want warranty claims to be paid out. Maintenance costs them nothing; they don't pay for OCIs - we do. So it is a win-win for them; they state a conservative OCI to protect their risk, and we pay for privilege of following their dogma. Extended OCIs have shown to be very safe, when well managed, even using "normal" products. But I would NEVER, EVER advise someone to do this without understanding the risks, and being very involved in the overall maintenance program (UOAs, PCs, observations, etc).
Last edited: