Tribologist discusses 0W8 motor oil

It's been pointed out in other threads that his "cold oil flow race" test doesn't always correlate to the CCS and MRV specs of the oils he's comparing. So those cold flow tests may not actually correlate to how those oils would actually behave in a cold start-up and pumpability in a real engine. These days, a lot of people buy into the hype based on only the visual aspects they see instead of the actual technical aspects.
Also known as a county fair demonstration. Many useless things get promoted this way.
 
But the reality is, in a 2012 F150. Yeah, uh… use whatever the manufacturer recommended, in whatever brand you like. Etc.
To add ... we know that what's recommended by the manufacture in this country is highly influenced by CAFE. Many threads over the years showing the same engines used in other parts of the world where the OM's show a whole range of recommended oil viscosity, many much higher than those recommended in the USA vehicles using that same engine.
 
Curious if the bearings in these engines specified for 0W-16 or lower viscosity are wider/larger etc. to carry the same loads over a larger area thereby enabling thinner oils to be used with no increase in wear or increasing odds of failure?
 
Curious if the bearings in these engines specified for 0W-16 or lower viscosity are wider/larger etc. to carry the same loads over a larger area thereby enabling thinner oils to be used with no increase in wear or increasing odds of failure?
I can't speak to all brands, but BMW has widened their journals on all engines specifying 3.0 HTHS or lower compared to former designs and those B series engines have IROX coatings on the bearings.
 
I can't speak to all brands, but BMW has widened their journals on all engines specifying 3.0 HTHS or lower compared to former designs and those B series engines have IROX coatings on the bearings.
That was my suspicion simply due to the fact the load has to be dispersed. The other part I am pondering is do wider bearings equal less friction, the same, or indeed more? Theoretically and typically, more surface area equals more friction at the advantage of less load per square "whatever".

Much of this "race to lower viscosity" is chasing MPG so it does beg the question...
 
That was my suspicion simply due to the fact the load has to be dispersed. The other part I am pondering is do wider bearings equal less friction, the same, or indeed more? Theoretically and typically, more surface area equals more friction at the advantage of less load per square "whatever".

Much of this "race to lower viscosity" is chasing MPG so it does beg the question...
One would have to assume there is still a net benefit. I'm sure they already know via simulation before they build a single prototype. It's hard to say if the wider journals are strictly due to the thinner oils - the engines also make more power than their predecessors so I'm sure part of it is simply that.
 
One would have to assume there is still a net benefit. I'm sure they already know via simulation before they build a single prototype. It's hard to say if the wider journals are strictly due to the thinner oils - the engines also make more power than their predecessors so I'm sure part of it is simply that.
It would depend on how far you go back to define "predecessor" and there are many powerful engines that use or have used the same or similar bearing sizes for years, but your point is taken.

I lean more to wider bearings are needed to sustain the load when the MOFT is approaching zero. Amazing technology at work to be sure...
 
To add ... we know that what's recommended by the manufacture in this country is highly influenced by CAFE. Many threads over the years showing the same engines used in other parts of the world where the OM's show a whole range of recommended oil viscosity, many much higher than those recommended in the USA vehicles using that same engine.

Keep in mind, some of that may be local availability and also local fuel quality. Not just cafe standards.

Just because here in the western world, you can go get any oil weight you want from a store / dealership / etc. doesn’t mean you can anywhere else.

I do a significant amount of packaging and handling for a major Japanese OEM. Like 10-12 truck loads a week. It all goes overseas. The vast majority of what we package and ship isn’t found at their dealerships in the U.S.


I get what you’re saying with Cafe standards. And sure. That can be a portion of it. But non-western markets can’t afford the oils we can. They are also more accustomed to the heavier oils, which are found more commonly due to heavy usage of motorbikes and such.

So you have to play nicely with local cultures and what they can afford.

As well, the local fuel will be different. And the local warranty market will be as well.
 
Even if you believe that his wear test method provides any valuable information, those results only differed by a whopping 5% I know it looks larger, more like a 2x difference between the best and worst, because of how close he zoomed in on the y-axis. A tried and true tactic of deceptive data presentation.

Results close enough that I’d be shocked if his test setup could even reliably measure differences that small. Something he could easily prove but never has. But I get it, content needs to be churned out and why bother doing even the smallest amount of experimental rigor when it’s just so much easier to amplify potentially meaningless differences by hard ranking results as 1st 2nd and 3rd place.
PF's lubricity test setup is clearly far more rudimentary compared to the tech that LSJr has at his disposal and I do take the results with a grain of salt. That said, it does provide a set of data points that I view as valid. The results are consistent with more advanced lubricity tests. There is a clear correlation between the amount of anti-wear additive package in oil and wear, as demonstrated by LSJr's GTX Classic vs GTX wear test I quoted in an earlier post in this thread where the difference in wear was far more dramatic. PF's arguably less accurate test reached similar conclusions. Toyota's oil had the lowest addpack of the tested oils so the result is not surprising.

Also, I think you underestimate the wear scar size difference. Wear happens in 3 dimensions so even a tiny difference on one axis makes a big (potentially cubed) difference in the volume of metal lost through wear. A 10% difference on one axis likely means a 33% increase in wear (1.1^3).

I wish someone with a professional grade lubricity tester would thoroughly and repeatedly wear-test all oils and publish results. Until then, these tests are the best measure we have. Certainly more trustworthy than marketing claims printed on the oil bottles.
 
Yep, engines specifying 0W-16 and 0W-8 have special design features to run that low viscosity - journal bearings design, special materials and coating, etc.
Would it preclude them from running higher viscosity oils?

Edit: disregard, already answered in the thread.
 
Last edited:
I wish someone with a professional grade lubricity tester would thoroughly and repeatedly wear-test all oils and publish results. Until then, these tests are the best measure we have. Certainly more trustworthy than marketing claims printed on the oil bottles.
You forgot 540 RAT data. 😄
 
PF's lubricity test setup is clearly far more rudimentary compared to the tech that LSJr has at his disposal and I do take the results with a grain of salt. That said, it does provide a set of data points that I view as valid. The results are consistent with more advanced lubricity tests. There is a clear correlation between the amount of anti-wear additive package in oil and wear, as demonstrated by LSJr's GTX Classic vs GTX wear test I quoted in an earlier post in this thread where the difference in wear was far more dramatic. PF's arguably less accurate test reached similar conclusions. Toyota's oil had the lowest addpack of the tested oils so the result is not surprising.

Also, I think you underestimate the wear scar size difference. Wear happens in 3 dimensions so even a tiny difference on one axis makes a big (potentially cubed) difference in the volume of metal lost through wear. A 10% difference on one axis likely means a 33% increase in wear (1.1^3).

I wish someone with a professional grade lubricity tester would thoroughly and repeatedly wear-test all oils and publish results. Until then, these tests are the best measure we have. Certainly more trustworthy than marketing claims printed on the oil bottles.
You truly have no idea what you are talking about. You are fooled just like the rest of their audience.

Out of every falsehood you have posted the worst one is your belief that they are presenting a valid data set.
 
PF's lubricity test setup is clearly far more rudimentary compared to the tech that LSJr has at his disposal and I do take the results with a grain of salt. That said, it does provide a set of data points that I view as valid. The results are consistent with more advanced lubricity tests. There is a clear correlation between the amount of anti-wear additive package in oil and wear, as demonstrated by LSJr's GTX Classic vs GTX wear test I quoted in an earlier post in this thread where the difference in wear was far more dramatic. PF's arguably less accurate test reached similar conclusions. Toyota's oil had the lowest addpack of the tested oils so the result is not surprising.

Also, I think you underestimate the wear scar size difference. Wear happens in 3 dimensions so even a tiny difference on one axis makes a big (potentially cubed) difference in the volume of metal lost through wear. A 10% difference on one axis likely means a 33% increase in wear (1.1^3).

I wish someone with a professional grade lubricity tester would thoroughly and repeatedly wear-test all oils and publish results. Until then, these tests are the best measure we have. Certainly more trustworthy than marketing claims printed on the oil bottles.
The issue with the lubricity test is it's meant for gear oil testing....not motor oil testing. As much as you may want to believe this is telling you how well these oils perform in your engine...it doesn't. I enjoy PFs videos but this isn't a valid way to compare oils in the context of your engine. It is however a dataset as you point out with reasonable variable control etc. just not one that can be used to help with what is the higher performing motor oil.

1000017981.jpg
 
The issue with the lubricity test is it's meant for gear oil testing....not motor oil testing. As much as you may want to believe this is telling you how well these oils perform in your engine...it doesn't. I enjoy PFs videos but this isn't a valid way to compare oils in the context of your engine. It is however a dataset as you point out with reasonable variable control etc. just not one that can be used to help with what is the higher performing motor oil.

View attachment 212094
That PF dude will never be recruited by the SwRI …
He should stick with: glue boards with A, B, C, … add weight to failure … Rat540 just needs to go away like all rats …
 
Last edited:
That PF dude will never be recruited by the SwRI …
He should stick with: glue boards with A, B, C, … add weight to failure … Rat just needs to go away like all rats …
I like them both because they make BITOGers sit in corner and cry 🤣
 
1st prize for the most pathetic post here. No arguments, only personal insults. Instant ignore.
Well let's unpack it then.

First off, any test that is intended to provide information about a material property has to be relevant to the system it's being used in. If I make a taste test of motor oil, that's not relevant to ICE operation. Here you have the same situation. This test is not relevant to engine operation as already noted by others in this thread. That's the first fatal flaw.

If you suspend disbelief that the test isn't relevant then the second fatal flaw is the use of non-standardized test equipment and no published test methodology. Why is this important? Well besides making any tests comparable, it also allows the use of pre-calculated statistical analysis of the results. That's the real key to a standard test. If you develop a test outside of the standard then it is now incumbent on the operator to analyze the methodology and results. These goofballs do none of that.

This is critical since it will show if your method is even a valid test and will give parameters such as minimum sample size as well as variability, repeatability and reproducibility. They have none of this. The actual ASTM test for gear oils has huge error bars for the results, mostly due to operator variability during the test. Large variability equates to also needing a somewhat larger sample size but again, these county fair hawkers generally perform the test once. But that's only another fatal flaw.

Getting back to your assertion that it is valid data, this is indeed the worst part of those tests. No results are presented properly showing the uncertainty in the results. This is about the large error bars. If you take the time to take their worthless results and use standard uncertainty then the real result is that all the oils in the test yield indistinguishable results. In other words they all test the same.

But even to get here that requires you to again suspend belief in at least three other aspects of the test. The whole thing is a complete dumpster fire and the last thing it has is any data validity. It has zero. None.
 
Back
Top