Shearing, dilution and low cSt questions

[OVERKILL]

True, a fuel dillution column would be nice. But at lease they are all 1st Gen Coyote engines (only MPI) which IIRC don't suffer from fuel dilution like a GDI engine.

No, it's not as bad, but MPFI engines still can dilute, my SRT HEMI's all did and so did the S62 in my M5, considerably (5% fuel in one case).

It's unfortunate that the most popular labs don't do GC fuel dilution, it would be extremely valuable, especially in discussions like these.

 

[ZeeOSix]

Without fuel being accounted for there, it's very hard to ascribe change in viscosity exclusively to shear. This is made even more opaque by the fact that these are different cars with different operating conditions. It is in no way a controlled data set and lacks any corresponding GC data.

I see you added a sentence. If you look at the data, the colored blocks in the first column (except the last big green block) are the same cars using different oils.

 

[OVERKILL]

I see you added a sentence. If you look at the data, the colored blocks in the first column (except the last big green block) are the same cars using different oils.

I suspected that. But if we do look at that last green block, there's ~10% between the two M1 5w-20 UOA's. That's a lot of variability. Is that difference fuel? Are they both diluted? With GC data we'd know.

 

[ZeeOSix]

I suspected that. But if we do look at that last green block, there's ~10% between the two M1 5w-20 UOA's. That's a lot of variability. Is that difference fuel? Are they both diluted? With GC data we'd know.

Of course there will be nuances for whatever reasons. I never claimed it was a perfect data set or controlled data collection. My main point was all engines mechanically shear oil, and the shear rate depends on the engine design, use conditions/miles on the oil and the oil being used.

But at least it's all UOAs on the same engine design and multiple UAOs on the same cars ... something I've not seen at that level collected on this chat board.

 

[OVERKILL]

Of course there will be nuances for whatever reasons. I never claimed it was a perfect data set or controlled data collection. My main point was all engines mechanically shear oil, and the shear rate depends on the engine design, use conditions/miles on the oil and the oil being used.

But at least it's all UOAs on the same engine design and multiple UAOs on the same cars ... something I've not seen at that level collected on this chat board.

Yes, the debate has never been whether some mechanical shear of VII polymers takes place, it's how much of the change in viscosity is shear vs fuel dilution. From what has been shared by formulators, it appears that fuel is the primary driver with shear playing a secondary role.

 

[ZeeOSix]

Yes, the debate has never been whether some mechanical shear of VII polymers takes place, it's how much of the change in viscosity is shear vs fuel dilution. From what has been shared by formulators, it appears that fuel is the primary driver with shear playing a secondary role.

What's the correlation between fuel dilution % and shear rate % amount of oil when the mechanical shear is zero?

Obviously engines that have high fuel dilution will probably have a higher impact on oil shear amount. But engines that don't fuel dilute much can still shear oil pretty good.

 

[OVERKILL]

What's the correlation between fuel dilution % and shear rate % amount of oil when the mechanical shear is zero?

Obviously engines that have high fuel dilution will probably have a higher impact on oil shear amount. But engines that don't fuel dilute much can still shear oil pretty good.

Problem is, we don't know which engines don't dilute much because most UOA's don't have GC fuel % results.

We also don't know if the visc has changed because of oxidative thickening either, and then thinned back down due to fuel. There are myriad combinations of fuel dilution/oxidation/shear that can only be determined if we know visc, fuel % and virgin visc. The oxidation figure can also be valuable, but can be skewed by certain base oils.

Just as an example, if we look at the MB stay in grade requirements:

Your 0w-40, which may be barely in the 40 range, has to stay in grade after 90 cycles for 229.5.

 

[ZeeOSix]

Problem is, we don't know which engines don't dilute much because most UOA's don't have GC fuel % results.

We also don't know if the visc has changed because of oxidative thickening either, and then thinned back down due to fuel. There are myriad combinations of fuel dilution/oxidation/shear that can only be determined if we know visc, fuel % and virgin visc. The oxidation figure can also be valuable, but can be skewed by certain base oils.

Of course there are many factors going on at the same time. Not many UOAs show an increase in viscosity from use from what I've seen.

But just looking at fuel dilution keeping other factors at zero, does 5% fuel dillusion equate to a 5% decrease in viscosity? Is it that linear or not? If not, what's the correlation? Is there actual data to verify the correlation? Ie, if someone took virgin oil and dilluted it with fuel by 5% would the viscosity decrease by 5% or something else?

 

[ZeeOSix]

Problem is, we don't know which engines don't dilute much because most UOA's don't have GC fuel % results.

We also don't know if the visc has changed because of oxidative thickening either, and then thinned back down due to fuel. There are myriad combinations of fuel dilution/oxidation/shear that can only be determined if we know visc, fuel % and virgin visc. The oxidation figure can also be valuable, but can be skewed by certain base oils.

Just as an example, if we look at the MB stay in grade requirements:
View attachment 75631
Your 0w-40, which may be barely in the 40 range, has to stay in grade after 90 cycles for 229.5.

Just how stringent is ASTM D7109? From here: https://www.astm.org/Standards/D7109.htm

It might be a good way to comparison test oils in the lab, and it might be a relatively easy test to pass. But in the field it might be a whole different story, even in terms of just the plain mechanical shearing.

"5.2 This test method may be used for quality control purposes by manufacturers of polymeric lubricant additives and their customers.

5.3 This test method is not intended to predict viscosity loss in field service in different field equipment under widely varying operating conditions, which may cause lubricant viscosity to change due to thermal and oxidative changes, as well as by the mechanical shearing of polymer. However, when the field service conditions, primarily or exclusively, result in the degradation of polymer by mechanical shearing, there may be a correlation between the results from this test method and results from the field."

 

[OVERKILL]

Just how stringent is ASTM D7109? From here: https://www.astm.org/Standards/D7109.htm

It might be a good way to comparison test oils in the lab, and it might be a relatively easy test to pass. But in the field it might be a whole different story, even in terms of just the plain mechanical shearing.

"5.2 This test method may be used for quality control purposes by manufacturers of polymeric lubricant additives and their customers.

5.3 This test method is not intended to predict viscosity loss in field service in different field equipment under widely varying operating conditions, which may cause lubricant viscosity to change due to thermal and oxidative changes, as well as by the mechanical shearing of polymer. However, when the field service conditions, primarily or exclusively, result in the degradation of polymer by mechanical shearing, there may be a correlation between the results from this test method and results from the field."

Seems pretty self explanatory, it's design to produce results that should correlate with the result of shearing of VII polymers in service, but cannot account for oxidation and other factors (like fuel) weighing into viscosity loss, which is in the line immediately following the last one you highlighted:

However, when the field service conditions, primarily or exclusively, result in the degradation of polymer by mechanical shearing, there may be a correlation between the results from this test method and results from the field.

Now, how rigorous is the protocol?

It's a diesel injector, which was chosen because it will introduce significant shearing:

This test method covers the evaluation of the shear stability of polymer-containing fluids. The test method measures the viscosity loss, in mm2/s and percent, at 100 °C of polymer-containing fluids when evaluated by a diesel injector apparatus procedure that uses European diesel injector test equipment. The viscosity loss reflects polymer degradation due to shear at the nozzle. Viscosity loss is evaluated after both 30 cycles and 90 cycles of shearing.

NOTE 1: This test method evaluates the shear stability of oils after both 30 cycles and 90 cycles of shearing. For most oils, there is a correlation between results after 30 cycles and results after 90 cycles of shearing, but this is not universal.

I wouldn't expect Mercedes to use it as part of their test protocol if they didn't find it valuable or didn't produce relevant results.

 

[PimTac]

The air/fuel mIxture of GDI can be so poor, particularly during cold start

Regarding this bit of your response, what makes the air/fuel mixture of a GDI motor different from a non GDI motor?

 

[OVERKILL]

Of course there are many factors going on at the same time. Not many UOAs show an increase in viscosity from use from what I've seen.

We've had a few. AMSOIL was known for showing a bit of oxidative thickening on longer runs. This is again another parameter that the Euro certs cover:

But just looking at fuel dilution keeping other factors at zero, does 5% fuel dillusion equate to a 5% decrease in viscosity? Is it that linear or not? If not, what's the correlation? Is there actual data to verify the correlation? Ie, if someone took virgin oil and dilluted it with fuel by 5% would the viscosity decrease by 5% or something else?

You should be able to plug those figures into a visc calc and get an idea. Fuel is extremely thin, so 5% fuel in a lubricant will have a big impact. Gas is somewhere around 0.4 to 0.7cSt @ 100C, so we'll go with 0.55. I plug that into the Widman calc @ 5% and that drops my visc from 13.5cSt to 11.55cSt, which correlates pretty well with my UOA:

 

[ZeeOSix]

Seems pretty self explanatory, it's design to produce results that should correlate with the result of shearing of VII polymers in service, but cannot account for oxidation and other factors (like fuel) weighing into viscosity loss, which is in the line immediately following the last one you highlighted:

However, when the field service conditions, primarily or exclusively, result in the degradation of polymer by mechanical shearing, there may be a correlation between the results from this test method and results from the field.

They also mentioned it's not designed to predict in the field: "... as well as by the mechanical shearing of polymer."

How, how rigorous is the protocol?

It's a diesel injector, which was chosen because it will introduce significant shearing:

I wouldn't expect Mercedes to use it as part of their test protocol if they didn't find it valuable or didn't produce relevant results.

Running the oil through "90 cycles" equates to what typical OCI in terms of use in a typical engine? I don't think there is any real correlation. What is "1 cycle" defined as?

I'm sure it has some value, and as said in ASTM D7109: "5.2 This test method may be used for quality control purposes by manufacturers of polymeric lubricant additives and their customers."

 

[ZeeOSix]

You should be able to plug those figures into a visc calc and get an idea. Fuel is extremely thin, so 5% fuel in a lubricant will have a big impact. Gas is somewhere around 0.4 to 0.7cSt @ 100C, so we'll go with 0.55. I plug that into the Widman calc @ 5% and that drops my visc from 13.5cSt to 11.55cSt, which correlates pretty well with my UOA:

View attachment 75634

Yes, fuel dilution has a large impact. Can you imagine what the viscosity goes to on some of the Hondas that have 1 qt of fuel dilution added to the 4 qts in the sump?

 

[OVERKILL]

They also mentioned it's not designed to predict in the field: "... as well as by the mechanical shearing of polymer."

Clearly, you and I are interpreting that quote differently. Let's go over this again:

This test method is not intended to predict viscosity loss in field service in different field equipment under widely varying operating conditions,

Emphasis on the last bit, because:

which may cause lubricant viscosity to change due to thermal and oxidative changes, as well as by the mechanical shearing of polymer.

So, multiple factors in play as a result of wildly varying operating conditions leads to non-predictive results and the inability to isolate the effects of shear. This brings us to the point I quoted:

However, when the field service conditions, primarily or exclusively, result in the degradation of polymer by mechanical shearing, there may be a correlation between the results from this test method and results from the field.

Pretty straight forward. When the field conditions are NOT wildly varying and primarily or exclusively result in shear, a correlation may exist.

Running the oil through "90 cycles" equates to what typical OCI in terms of use in a typical engine? I don't think there is any real correlation.

I mean, that's the entire purpose of the test, to shear VII polymers by forcing the lubricant through an injector nozzle to promote viscosity loss. If there was no correlation in service, there would be absolutely no purpose in using the test. Clearly, Mercedes sees value in it.

I don't think it matters whether you think there is a correlation or not, the OEM, who knows more about this than both of us, sees value in it. I don't think it's going to be valuable for us to turn this into a pissing match on that point, but that does seem to be where you are wanting to push this and you know I'm always game

Otherwise, I can maybe send an inquiry to Mercedes and ask? That seems like a more reasonable approach, lol.

 

[OVERKILL]

Yes, fuel dilution has a large impact. Can you imagine what the viscosity goes to on some of the Hondas that have 1 qt of fuel dilution added to the 4 qts in the sump?

Yup, exactly, it would be absolutely nuts how much visc loss takes place. Are you and I both thinking about those small Honda DI engines that weren't heating up and diluting the oil to the point where the Chinese government took action? It's gotta be seriously bad for that to occur.

 

[OVERKILL]

Another data point, Shell uses ASTM 7109 on their diesel oils apparently. The 10w-30, which clearly would have less VII, shows significantly less shear than the 10w-40:

 

[OVERKILL]

Yes, fuel dilution has a large impact. Can you imagine what the viscosity goes to on some of the Hondas that have 1 qt of fuel dilution added to the 4 qts in the sump?

With respect to my UOA, the majority of my visc loss was from fuel (5%) and the remainder would be the result of some mechanical shear. The S62 is a DOHC dual VANOS V8, somewhat similar to the Coyote in many respects, with a massive double-roller chain, so I thought it an interesting case to present here because, as I noted earlier, most people don't get GC fuel % on their UOA's.

 

[ZeeOSix]

Clearly, you and I are interpreting that quote differently. Let's go over this again:

This test method is not intended to predict viscosity loss in field service in different field equipment under widely varying operating conditions,

Emphasis on the last bit, because:

which may cause lubricant viscosity to change due to thermal and oxidative changes, as well as by the mechanical shearing of polymer.

So, multiple factors in play as a result of wildly varying operating conditions leads to non-predictive results and the inability to isolate the effects of shear. This brings us to the point I quoted:

However, when the field service conditions, primarily or exclusively, result in the degradation of polymer by mechanical shearing, there may be a correlation between the results from this test method and results from the field.

Pretty straight forward. When the field conditions are NOT wildly varying and primarily or exclusively result in shear, a correlation may exist.


I mean, that's the entire purpose of the test, to shear VII polymers by forcing the lubricant through an injector nozzle to promote viscosity loss. If there was no correlation in service, there would be absolutely no purpose in using the test. Clearly, Mercedes sees value in it.

I don't think it matters whether you think there is a correlation or not, the OEM, who knows more about this than both of us, sees value in it. I don't think it's going to be valuable for us to turn this into a pissing match on that point, but that does seem to be where you are wanting to push this and you know I'm always game

Otherwise, I can maybe send an inquiry to Mercedes and ask? That seems like a more reasonable approach, lol.

I'm not trying to make it into a "pissing match" ... you know I'm game for those sometimes too. It's just that I don't think it's anything that can be correlated directly to field use ... buy I agree it has value in comparing different oils under the same shearing conditions. How rigorous the test is, or how well the test represents actual shearing in an engine is unknown. And how well do non-Euro specified oils do in this test? Is there a similar test done under API or ILSAC?

 

[ZeeOSix]

With respect to my UOA, the majority of my visc loss was from fuel (5%) and the remainder would be the result of some mechanical shear. The S62 is a DOHC dual VANOS V8, somewhat similar to the Coyote in many respects, with a massive double-roller chain, so I thought it an interesting case to present here because, as I noted earlier, most people don't get GC fuel % on their UOA's.

Every Blackstone UOA I've seen posted in the UOA forum has a "Fuel %" measurement. A lot of them come in <0.5%, yet the oil has still sheared down some but are typically still "within grade".