LSJR Calls Out YouTube AI-generated Video Referencing Him

[IndyFan]

LSJR just run out of ideas around Christmas and New Year and decided to argue with an AI video to create more content.

In all 3 point the AI video are correct:
- Engines run 0W-20 only in ghe US and 5W-30 (and higher grades) elsewhere
- 0W-20 shears faster (goes out of grade) and then its protection is not optimal (we've seen this in many UOA)
- 5W-30 provides 40% better protection because even if it shears down it still protects the engine better than 0W-20 if shears down

His claims that 0W-20 shears down twice less (actually on the graph is triple less) than 5W-30 and that 0W-20 provides the same protection (only if it stays in grade) if not better as 5W-30 is data in his imaginary world that we don't know how true it is and what were the oil samples from (engines, oil brands, OCI, mileage etc.).

#Your first one is NOT correct.
Your second one is debatable because it is very difficult to prove.
Your 3rd one is also debatable.

The current generation Pentastar specifies 0w20 all over the world. It is what it was designed to run, was tested on, and it is recommended in all markets where it is sold. I am sure there are other engines that are the same, particularly from Stellantis, but possibly from others, too.

 

[gfh77665]

LSJR just run out of ideas around Christmas and New Year and decided to argue with an AI video to create more content.

At some point, more will figure out he is Scotty 2.0.

 

[Japanese]

Going out of grade isn't the same as shear. Yes, mechanical shear of the VII can cause the oil to "go out of grade" but that's not the only cause.

Thanks.
Where can I read more about that?

 

[kschachn]

Thanks.
Where can I read more about that?

Just Google it. Only a multi-viscosity oil can experience mechanical shear since the oil molecules themselves are too small. VII polymers can have molecular weights in the millions and tend to present themselves in a linear orientation when warm. So they can be cleaved. But oil molecules are far too small to be cleaved.

When this happens the overall viscosity of the oil is reduced. However, most engines do not have a particular problem with shear and most good quality oils use sufficiently shear-resistant VII so that it is not an issue.

Here on Bitog many people claim the oil has sheared when they see a viscosity deviation on a used oil analysis. However, most of those used oil analysis are from Blackstone and they are unable to measure fuel dilution, and subsequently are incapable of properly characterizing a change in viscosity. When properly analyzed for fuel by a direct method nearly all the deviations we see here are from fuel.

Which doesn't mean it never happens. Actual mechanical shear of the VII can be mitigated by a shorter OCI, an oil with superior VII, or with the use of a monograde oil if your climate allows and if you can find one with the proper specs.

 

[rstsco]

Where can I read more about that?

As much as other's may protest, this LSJR video is on viscosity breakdown, specifically due to shear. He's in HPL's lab where HPL's plant manager has blended up some different bespoke 5W-30 oils using different viscosity modifiers, and some OTS oils, then ran them through HTHS and KRL testing.

Note: HPL's NO VII 5W-30 has the least HTHS % shear of all the blends.

 

[OVERKILL]

As much as other's may protest, this LSJR video is on viscosity breakdown, specifically due to shear. He's in HPL's lab where HPL's plant manager has blended up some different bespoke 5W-30 oils using different viscosity modifiers, and some OTS oils, then ran them through HTHS and KRL testing.

Note: HPL's NO VII 5W-30 has the least HTHS % shear of all the blends.

View attachment 318625

Yes, the KRL is a much more severe shear test than anything an oil is going to experience in service, and this isn't a used oil analysis shootout with unknown amounts of fuel in the oils. Mobil 1 Full Synthetic 0W-40 did quite well there, shearing less than Castrol Edge 5W-40, despite what many on here would have you think about it.

 

[IndyFan]

One thing LSJr. says is to do used oil analysis´s on your own vehicle to see how different oils behave in your engines with your driving styles.

Here´s what I have seen, and I typically run the oil weight recommended by the manufacturer. I have NEVER had a 0w20 or 5w20 go out of grade. Not once. I have done more used oil analysis´s on 20 weight oils than anything. None of my engines running 20 weight have been DI or turbo engines, so I don´t think they are hard on oil, but they are not just commuters, either. I tow and go off road in extreme conditions frequently with them and sometimes drive one of them pretty hard....the 2018 Jeep which has had the majority of the used oil analysis´s.

With 5w30, I saw it drop out of grade several times, mostly with a 3.5 Ecoboost in a 15 passenger Ford Transit. Of course, I would expect that. I did 10k runs and that DI/turbo engine was much harder on oil than my Jeep 3.8 and 3.6 engines where I ran the 20 weight. Never had an issue with the Ecoboost. If I still had it I would probably continue with Mobil 1 Extended Performance 5w30. I think it did fine with that oil.

I have not done a used oil analysis on my Vette because I´ve run 3 different oils in 3 years. (2 different weights, 5w30 and 0w40).

 

[garageman402]

As much as other's may protest, this LSJR video is on viscosity breakdown, specifically due to shear. He's in HPL's lab where HPL's plant manager has blended up some different bespoke 5W-30 oils using different viscosity modifiers, and some OTS oils, then ran them through HTHS and KRL testing.

Note: HPL's NO VII 5W-30 has the least HTHS % shear of all the blends.

View attachment 318625

Thank you @rstcso for mentioning the video from a year ago, I got this from it.

He probably put the Mobil 1 next to the Infinium Star because EOM & Shell are working with Infinium, the shear is similar.

This chart illustrates how the AI video in question is flawed. The blanket statement than a 0w-20 will shear FASTER than a 5w-30 could be comparing apples to oranges:

A 0w-20 with OCP will most definitely shear more (maybe not faster) than a 5w-30 with Lubrizol Star. Conversely, a 0w-20 with LZ Star will outperform a 5w-30 with OCP

Now separating fuel dilution from shearing is another question, but an oil with a better polymer will resist viscosity loss better from fuel dilution than one that uses, say, OCP. OCP is what the Supertech & similar oils use, per LSJr

 

[kschachn]

As much as other's may protest, this LSJR video is on viscosity breakdown, specifically due to shear. He's in HPL's lab where HPL's plant manager has blended up some different bespoke 5W-30 oils using different viscosity modifiers, and some OTS oils, then ran them through HTHS and KRL testing.

Note: HPL's NO VII 5W-30 has the least HTHS % shear of all the blends.

View attachment 318625

I think that’s well within the margin of error on the no-VII. I’ve done experiments with high impact, high shear loading on similar length hydrocarbons and there was no shear. This was determined by gas chromatography so it was a direct method of measuring chain length. There’s no way a mechanical process is going to disrupt the carbon bonds on such a short molecule.

 

[kschachn]

One thing LSJr. says is to do used oil analysis´s on your own vehicle to see how different oils behave in your engines with your driving styles.

Here´s what I have seen, and I typically run the oil weight recommended by the manufacturer. I have NEVER had a 0w20 or 5w20 go out of grade. Not once. I have done more used oil analysis´s on 20 weight oils than anything. None of my engines running 20 weight have been DI or turbo engines, so I don´t think they are hard on oil, but they are not just commuters, either. I tow and go off road in extreme conditions frequently with them and sometimes drive one of them pretty hard....the 2018 Jeep which has had the majority of the used oil analysis´s.

With 5w30, I saw it drop out of grade several times, mostly with a 3.5 Ecoboost in a 15 passenger Ford Transit. Of course, I would expect that. I did 10k runs and that DI/turbo engine was much harder on oil than my Jeep 3.8 and 3.6 engines where I ran the 20 weight. Never had an issue with the Ecoboost. If I still had it I would probably continue with Mobil 1 Extended Performance 5w30. I think it did fine with that oil.

I have not done a used oil analysis on my Vette because I´ve run 3 different oils in 3 years. (2 different weights, 5w30 and 0w40).

A good example of how key it is to determine exactly why a viscosity deviation may be occurring and why it is essential to pick a proper lab. Shear and fuel dilution have different causes and have different possible mitigation.

 

[rstsco]

I think that’s well within the margin of error on the no-VII. I’ve done experiments with high impact, high shear loading on similar length hydrocarbons and there was no shear. This was determined by gas chromatography so it was a direct method of measuring chain length. There’s no way a mechanical process is going to disrupt the carbon bonds on such a short molecule.

LSJR says even though HPL's NO VII base oil wouldn't be expected to show any shear, other additives for anti-wear, etc, will shear. Thus the small percentage.

 

[kschachn]

Now separating fuel dilution from shearing is another question, but an oil with a better polymer will resist viscosity loss better from fuel dilution than one that uses, say, OCP. OCP is what the Supertech & similar oils use, per LSJr

Yes and no. They won’t resist simple mechanical dilution by a lower viscosity fluid any better, but they will resist temporary and permanent viscosity loss through VII degradation.

 

[kschachn]

LSJR says even though HPL's NO VII base oil wouldn't be expected to show any shear, other additives for anti-wear, etc, will shear. Thus the small percentage.

If he says so. But I would ask what his error bars are for this test. Given his imprecise data analysis in the past I’m not completely buying that explanation.

 

[SammyChevelleTypeS3]

Just Google it. Only a multi-viscosity oil can experience mechanical shear since the oil molecules themselves are too small. VII polymers can have molecular weights in the millions and tend to present themselves in a linear orientation when warm. So they can be cleaved. But oil molecules are far too small to be cleaved.

When this happens the overall viscosity of the oil is reduced. However, most engines do not have a particular problem with shear and most good quality oils use sufficiently shear-resistant VII so that it is not an issue.

Here on Bitog many people claim the oil has sheared when they see a viscosity deviation on a used oil analysis. However, most of those used oil analysis are from Blackstone and they are unable to measure fuel dilution, and subsequently are incapable of properly characterizing a change in viscosity. When properly analyzed for fuel by a direct method nearly all the deviations we see here are from fuel.

Which doesn't mean it never happens. Actual mechanical shear of the VII can be mitigated by a shorter OCI, an oil with superior VII, or with the use of a monograde oil if your climate allows and if you can find one with the proper specs.

Thanks for spelling that out so well. I did not know about Blackstone until word was passed here about the guess work (it could be called) they use to determine the fuel % listed in their reports.
I and other folks had been trusting Blackstone Labs without realizing their weakness + methods.
I have used a place since learning about Blackstone called Polaris labs for my last sample which I hardly every do these days. May run one for the fun of it off the new SUV we got back in October. The engine is a Non-Turbo GDI. So far I like no dropping or raising dip stick level and especially zero fuel smells.....

 

[rstsco]

If he says so. But I would ask what his error bars are for this test. Given his imprecise data analysis in the past I’m not completely buying that explanation.

All I can say is the testing was done at HPL's lab using their personnel and equipment. In theory, all he had to do was copy down the numbers. Then again, I've used SD several times and had to ask why some values were 0 instead of an actual value in a field with an expected value above 0.

 

[RazorsEdge]

Artificial Ignorance will replace the masses

Jobs will be replaced by AI

True

 

[kschachn]

All I can say is the testing was done at HPL's lab using their personnel and equipment. In theory, all he had to do was copy down the numbers. Then again, I've used SD several times and had to ask why some values were 0 instead of an actual value in a field with an expected value above 0.

Every test has accuracy, as well as repeatability and reproducibility. It’s the fault of the person doing the test if they don’t know or don’t report this. It can make a huge difference in the results, sometimes a test can have large differences in results but statistically they’re all the same.

A perfect example of this is the ASTM test for carbon black conductivity. A very inaccurate test with huge error bars. It was virtually impossible to get anything that analyzed out above random.

We see this junk on here all the time. People reporting values for tests when there is no indication whether or not the values you’re looking at are significant. Then they go on to make conclusions based on that data. There is an egregious website that does exactly this to rank motor oils. Another example is claiming one oil is better than another when the spectrographic analysis shows 3 ppm less iron.

As an aside, I will tell you that viscosity measurements definitely have a tolerance. Without analysis there is no way I’d guess a 1% test value is particularly significant. I’m still calling noise on that one based upon my laboratory experience with shearing and impact force tests on hydrocarbons.

 

[Patman]

Jobs will be replaced by AI

True

It’s true in my case. After 31 years as a Photoshop retoucher (and almost 14 years at my last company), they eliminated my job because AI could do it faster. Nowhere near as good, but faster. So I decided it was time to retire instead of staying in that dying career. Now I have a part time job that I absolutely love and it can never be replaced by AI. And even if someone handed me 50 million dollars I would still be out there giving tours in Niagara Falls

 

[RazorsEdge]

It’s true in my case. After 31 years as a Photoshop retoucher (and almost 14 years at my last company), they eliminated my job because AI could do it faster. Nowhere near as good, but faster. So I decided it was time to retire instead of staying in that dying career. Now I have a part time job that I absolutely love and it can never be replaced by AI. And even if someone handed me 50 million dollars I would still be out there giving tours in Niagara Falls

It started last year. People in the customer service are going to be hit hard this year.

 

[SammyChevelleTypeS3]

Makes one wonder is there anything the USA is doing or should be to attempt to make certain we have enough workers trained to do the many new types of jobs AI is going to produce? For instance..... if they use it in manufacturing, then they will need a host of newly trained mechanics and even operators to keep those places going. Another should be more and more IT specialists jobs to keep all the AI controlled things moving along.....