HTHS Question

[Hokiefyd]

I have enjoyed some of the recent discussion involving various 0W-20 oils. As I was cruising home from work today, I was pondering HTHS and its involvement in engine protection and fuel economy. I came up with a few questions that I can't answer myself.

1. Is the HTHS scale linear? That is, is 2.7 cp 3.8% thicker than 2.6 cp?
2. I understand that HTHS is generally the most basic metric defining the protection vs. economy factor. That is, lower HTHS values generally offer better fuel economy and higher HTHS values generally offer a thicker film at the bearing surfaces. Do I have that right?
3. If the HTHS scale is linear, does a 3.8% thicker HTHS offer a 3.8% thicker film? Or, possibly asking the same question in the other direction, does a 3.8% thinner HTHS offer a 3.8% reduction in resistance to flow (oil friction) at the bearing surfaces?

So, my questions stem from ExxonMobil's choice to formulate their premium 20 grade oil, Mobil 1 Advanced Fuel Economy 0W-20, at 2.7 cP instead of 2.6 cP.

4. Does ExxonMobil's choice here reduce Mobil 1 Advanced Fuel Economy's potential of achieving the best possible fuel economy? If one were looking for the least possible friction, should they not choose a 2.6 cP oil?
5. Is ExxonMobil kind of acknowledging that they don't feel that 2.6 cP is enough to offer "best protection", so are making the deliberate choice of reducing the fuel economy potential in favor of adding a little more "protection capacity"?

I know that none of this would be measurable at the gas pump. But this is BITOG. I'm hoping for some technical discussion on this.

 

[Gokhan]

You've got (1) and (2) right. HTHS viscosity is measured in plain physical dynamical-viscosity units (cP), which is linear, and yes, it's the biggest prediction of fuel economy and protection against wear, which is also a compromise, especially in diesel engines.

Regarding (3), the minimum oil-film thickness (MOFT) is approximately linear in the square root of the Sommerfeld number, which has the HTHS viscosity in it:

MOFT is proportional to Square root of the {Sommerfeld number S = RPM * HTHS / Load}

MOFT ~ Square root of (RPM * HTHS / Load)

Basically, the lower the RPM, lower the HTHS viscosity, and higher the load, is the lower the MOFT.

(Reference: High-Temperature, High-Shear (Hths) Oil Viscosity by James A. Spearot)

Since the diesels produce a lot of torque (load) at low RPMs, HTHS viscosity is more critical for diesel engines than for gasoline engines. Also, diesels produce a lot of abrasive soot particles, which also necessitates a thicker oil film so that the separation between the sliding metal parts is larger than the size of the soot particles, which would score them, again necessitating higher HTHS viscosity. For gasoline engines, HTHS viscosity is less of a problem and manufacturers tend to specify as low HTHS as possible.

What you should most worry about is not 2.6 cP vs 2.7 cP but "permanent oil shear." Permanent oil shear can reduce HTHS from 2.6 cP to 2.0 cP or below easily in a few hundred miles on oil and it's more a problem in conventional oil than in synthetics. This is one of the reasons people prefer synthetics over conventionals. Again, 2.7 cP vs. 2.6 cP is almost a nonissue, and I wouldn't draw any conclusion from there. The effect in fuel economy also would be tiny. In addition, HTHS is not the only factor determining the fuel economy: cold and hot kinematic viscosity, friction of the base oil (PAO has the lowest friction) and friction modifiers also make a big difference. Used oil also tends to have noticeably less fuel economy then new oil.

Regarding wear, in addition to the HTHS viscosity, the additive package, especially the antiwear additives (ZDDP, moly, and others) but the detergents as well, is also crucial.

The lowest viscosity you can choose is not only determined by the wear but also the oil consumption. If you're burning more than about 1 qt per 1,000 miles, you should definitely switch to a thicker viscosity; otherwise, it's not practical to use a thin grade -- you're wasting money adding makeup oil, damaging emissions components, and polluting the air.

 

[CATERHAM]

Some good questions which indicates you already have a pretty good grasp of HTHSV.

1 - Yes the measure is linear.
2 - That's right, but to simplify it further just think of HTHSV as an accurate measure of how thick or thin an oil is at operating temp's in an operating IC engine.
3 - Yes
4 - You may be interested to know that the Canadian version of Mobil 1 Advanced Fuel Economy 0W-20 has a HTHSV of 2.6cP. I wouldn't put to much emphasis on the 2.7cP figure Mobil is currently showing. It does very from batch to batch. Mobil used to give their HTHSV figures to two decimal places. For all we know the US version is 2.66cP and the Cdn version 2.64cP so in reality after the rounding error the difference is neglible.
5 - I don't think so as explained above.
A difference of 0.1cP represents about 3-5 psi in oil pressure in an IC engine. In an engine that's spec'd for a 20wt 2.6cP oil, at normal operating temp's the oil pressure safety margin is still likely 10-15 psi above the optimum minimum OP.

The second most important viscosity measure to HTHS is an oil's viscosity index since most engine spend a great amount of time operating with oil temp's below the normal hot temperature.
The Mobil made Toyota 0W-20 has a 216 VI vs "only" 173 for AFE.
As a consiquence even at room temperature the Toyota oil will be about 30% lighter than AFE.

 

[Tom NJ]

Originally Posted By: Hokiefyd
So, my questions stem from ExxonMobil's choice to formulate their premium 20 grade oil, Mobil 1 Advanced Fuel Economy 0W-20, at 2.7 cP instead of 2.6 cP.


2.6 cP is the minimum specification. In all likelyhood they aimed a little higher than the minimum to allow for batch-to-batch variations and test reproducibility.

Tom NJ

 

[Gokhan]

Originally Posted By: CATERHAM
3 - Yes

Not linear. It's proportional to the square root of HTHS. See the formula I gave above. So, 4% higher HTHS results in 4% / 2 = 2% thicker oil film.

 

[Hokiefyd]

Thanks guys for the great discussion. I'm currently using Mobil 1 Advanced Fuel Economy 0W-20 in the Honda and Pennzoil Platinum 0w20 in the Acura. The PP has a higher viscosity index than the M1 (175 vs. 173), and the M1 has a higher HTHSV (2.7 cp vs. 2.6 cP). I'm of the general opinion that both of those differences are too negligible to even consider.

That said, the PP seems to be thinner at operating temperatures (8.39 cSt vs 8.7 cSt). It's also thinner at 40*C (42.8 cSt vs. 44.8 cSt). Comparing these two widely available off the shelf oils, the PP looks like it's generally the thinner oil. I don't start the engines in -40*C temperatures, so M1's excellent MRV valve doesn't appear to be relevant to me. If I'm looking for "thin", and the HTHSV isn't really that dramatic of a difference, it looks like PP would be the choice.

The Toyota oil is certainly also a consideration, but I'm just looking at these two off the shelf synthetics right now.

Thanks again all.

 

[Gokhan]

Don't forget that these are all new-oil viscosities. Used-oil viscosities, including HTHS, will be quite different, depending on how much the oil permanently shears. Permanent oil shear already takes place before 1,000 miles or so.

Also, don't forget to consider the Toyota Genuine Motor Oil 0W-20, which is made by ExxonMobil.

 

[Oil Changer]

Great discussion!

Hokiefyd,

The very first Toyota dealer I walked in to had three cases of the outstanding SM GTMO for $6/qt and SM Eneos for $4/qt. Might want to give it a shot in your area.

 

[CATERHAM]

Originally Posted By: Hokiefyd

That said, the PP seems to be thinner at operating temperatures (8.39 cSt vs 8.7 cSt). It's also thinner at 40*C (42.8 cSt vs. 44.8 cSt).

Yes I agree PP will be lighter but it is simply because of the lower HTHSV and marginally higher VI. And while the KV100 spec' is also correspondingly lower it can't be relied upon to compare actual operational viscosities of different oils. In fact it's best to ignore the KV100 spec's. I explained why in the following thread; HTHSV trumps KV100:
https://bobistheoilguy.com/forums/ubbthreads.php?ubb=showflat&Number=2276634&page=1

 

[Hokiefyd]

Originally Posted By: Oil Changer
Great discussion!

Hokiefyd,

The very first Toyota dealer I walked in to had three cases of the outstanding SM GTMO for $6/qt and SM Eneos for $4/qt. Might want to give it a shot in your area.


My local Toyota dealer sells the genuine Denso oil filters for reasonable prices, and they used to have the SM 0W-20 for something like $6.50/qt, so it sounds close to what you were getting. I didn't know the SN version was less expensive. I'll check it out.

 

[mechtech2]

I don't know what the '0 zero' HTHS rating is. Is it what water is, that has some viscosity? Or is is an absolute zero?

This is the key, because we cant say 20 deg F is twice as cold as 10 deg F. We have to know the meaning of 'zero'.

 

[Gokhan]

HTHS viscosities measured at 150 C and a shear rate of 1,000,000 (1/second)

SAE xW-50: HTHS >= 3.7 cP
SAE 15w40 and 20W-40: HTHS >= 3.7 cP
SAE 0W-40, 5W-40, and 10W-40: HTHS >= 3.5 cP
SAE xW-30: HTHS >= 2.9 cP
SAE xW-20: HTHS >= 2.6 cP
Proposed SAE xW-15: HTHS >= 2.3 cP
Proposed SAE xW-10: HTHS >= 2.0 cP
Proposed SAE 0W-5: HTHS >= 1.7 cP

Water: HTHS = 0.28 cP at 100 C (vapor at higher temperatures)
Water also doesn't have good lubrication properties and is corrosive.

Note that these are SAE minimum limits and German Castrol 0w30 for example has HTHS >= 3.5 cP and so do all ACEA A3/Bx and API CJ-4 oils, regardless of SAE viscosity. (Of course, 15w40 and thicker grades has a minimum of 3.7 cP.)

Reference (Yubase article)

Regarding absolute zero viscosity, the only such fluids that exist in nature are superfluids, which are a unique quantum state of matter. For example Helium 3 or Helium 4 near the absolute zero temperature exhibit the phenomenon of superfluidity. Superfluids have various interesting properties, such as the ability to flow from capillaries spontaneously.

 

[mechtech2]

OK then. 0 Zero in HTHS engine oil terms is not water's viscosity, but more of an absolute zero viscosity.
A total vacuum would be a zero HTHS, for instance.
[I don't think those who set the parameters had superfluids or total vacuum engine operation in mind}

 

[Gokhan]

Originally Posted By: mechtech2
OK then. 0 Zero in HTHS engine oil terms is not water's viscosity, but more of an absolute zero viscosity.
A total vacuum would be a zero HTHS, for instance.
[I don't think those who set the parameters had superfluids or total vacuum engine operation in mind}

Yes, water has nonzero viscosity, and so do all normal fluids.

Not sure what you mean by "0 Zero in HTHS engine-oil terms" but viscosity (dynamic as in HTHS or kinematic) is not just an engine-oil term but a more general physical quantity used for all fluids. Also, as you know, of course, 0W in SAE 0W-x grades is not a measure of viscosity but only some cold-temperature viscosity-grade specification. NLGI also uses 0, 00, 000, etc. for the thinner grease grades.

The only fluids in nature that have zero viscosity are superfluids. In fact, the only known superfluid is liquid helium near the absolute zero temperature, less than 2 degrees kelvin above absolute zero for the abundant He-4 isotope of helium. It becomes a normal fluid at higher temperatures and boils around 4 degrees kelvin above absolute zero. Less abundant He-3 isotope also exhibits superfluidity but at even much lower temperatures and due to a different physical mechanism.

I guess you can talk about the viscosity of vacuum if you really want to but since vacuum means absence of matter, it's not really a fluid, and it would be hard to define and/or devise a method to measure its viscosity. It would also strongly depend on the vacuum pressure of course.

Unlike a superfluid, the viscosity of vacuum would also never be exactly zero. If you consider the ideal case of an exact vacuum, then you can't talk about viscosity because there is no fluid (matter). On the other hand, superfluid helium has absolute zero viscosity because of a subtle quantum mechanism. It can do a lot of interesting things. For example if you put superfluid helium in a cup, it creeps up along the inside wall and then down the outside wall until the cup is completely empty. See this picture of superfluid helium in a cup:

So, if you could refrigerate your garage to near absolute-zero temperature and put superfluid helium in your engine, thanks to having absolute zero viscosity, it would creep up your dipstick tube and you would find it all on the floor when you came back in the evening.

 

[KCJeep]

Great thread!

I can't understand all of it but HTHS intrigues me as a habitual blender. Lately I've been focused more on HTHS in my blends, but I don't really know how to calculate it. If you say it's linear will a simple average using accurate percentages work?

My Jeep is spec'd for 10w30, most PCMO of which are around 3.0 to 3.1 HTHS. It runs fantastic though on 5w30 of 2.9 to 3.0 HTHS. I am however afraid to use 5W30 in the summer due to shear concerns. I've recently found several 5w30 High Mileage oils with HTHS in the 3.2 to 3.3 range, so I'm thinking there is absolutely no reason not to run those in the warm months. They have higher VI and a higher HTHS than many 10w30's.

Am I correct in my thinking here? If so why on earth would I run a 10w30 ever?

 

[Gokhan]

Originally Posted By: KCJeep
Great thread!

I can't understand all of it but HTHS intrigues me as a habitual blender. Lately I've been focused more on HTHS in my blends, but I don't really know how to calculate it. If you say it's linear will a simple average using accurate percentages work?

My Jeep is spec'd for 10w30, most PCMO of which are around 3.0 to 3.1 HTHS. It runs fantastic though on 5w30 of 2.9 to 3.0 HTHS. I am however afraid to use 5W30 in the summer due to shear concerns. I've recently found several 5w30 High Mileage oils with HTHS in the 3.2 to 3.3 range, so I'm thinking there is absolutely no reason not to run those in the warm months. They have higher VI and a higher HTHS than many 10w30's.

Am I correct in my thinking here? If so why on earth would I run a 10w30 ever?

I wouldn't worry much about permanent oil shear with a synthetic 5w30. A conventional 5w30 shears quite a bit a synthetic 5w30 not so much. Therefore, I think a synthetic 5w30 would always be an excellent replacement for a conventional 10w30, which some conventional-oil users prefer over conventional 5w30 because of shear stability.

 

[lemonade]

Originally Posted By: Gokhan
Originally Posted By: mechtech2
OK then. 0 Zero in HTHS engine oil terms is not water's viscosity, but more of an absolute zero viscosity.
A total vacuum would be a zero HTHS, for instance.
[I don't think those who set the parameters had superfluids or total vacuum engine operation in mind}

Yes, water has nonzero viscosity, and so do all normal fluids.

Not sure what you mean by "0 Zero in HTHS engine-oil terms" but viscosity (dynamic as in HTHS or kinematic) is not just an engine-oil term but a more general physical quantity used for all fluids. Also, as you know, of course, 0W in SAE 0W-x grades is not a measure of viscosity but only some cold-temperature viscosity-grade specification. NLGI also uses 0, 00, 000, etc. for the thinner grease grades.

The only fluids in nature that have zero viscosity are superfluids. In fact, the only known superfluid is liquid helium near the absolute zero temperature, less than 2 degrees kelvin above absolute zero for the abundant He-4 isotope of helium. It becomes a normal fluid at higher temperatures and boils around 4 degrees kelvin above absolute zero. Less abundant He-3 isotope also exhibits superfluidity but at even much lower temperatures and due to a different physical mechanism.

I guess you can talk about the viscosity of vacuum if you really want to but since vacuum means absence of matter, it's not really a fluid, and it would be hard to define and/or devise a method to measure its viscosity. It would also strongly depend on the vacuum pressure of course.

Unlike a superfluid, the viscosity of vacuum would also never be exactly zero. If you consider the ideal case of an exact vacuum, then you can't talk about viscosity because there is no fluid (matter). On the other hand, superfluid helium has absolute zero viscosity because of a subtle quantum mechanism. It can do a lot of interesting things. For example if you put superfluid helium in a cup, it creeps up along the inside wall and then down the outside wall until the cup is completely empty. See this picture of superfluid helium in a cup:

So, if you could refrigerate your garage to near absolute-zero temperature and put superfluid helium in your engine, thanks to having absolute zero viscosity, it would creep up your dipstick tube and you would find it all on the floor when you came back in the evening.

Remind me of quantum class, He, with 2 proton, 2 neutron and 2 electron, He-4 is a boson, hence subject to Bose-einstein statistics, and boson are not subject to Pauli exclusion principle, so there's no limit to the number of bosons that may be in the same quantum state, so we have superfluid. oh my god, can't believe i still remember these.

Oh, about He-4 flow up the dipstick thing, this is called "creeping flim", i think... don't remember exactly.
I have something to add about liquid He though, if one tries to measure the viscosity by, for example, measuring the drag on a metal plate as it is passed over the surface of the liquid, the result is about the same as the one we would expect from a normal liquid, even at temperatuire below the lambda point. Inother word, there appears to be a contradiction here. Some scientist later proposed a solution stating that liquid He below lambda point is part superfluid and part normal, with increasing portion of superfluid approaching 0Kelvin once it drop below lambda point.
Way off topic.... sorry

 

[CATERHAM]

Originally Posted By: KCJeep
I can't understand all of it but HTHS intrigues me as a habitual blender. Lately I've been focused more on HTHS in my blends, but I don't really know how to calculate it. If you say it's linear will a simple average using accurate percentages work?

Yes averaging is pretty close but it's better to simply use a viscosity calculator and just substitute the HTHSV spec' for the KV100.

 

[KCJeep]

Originally Posted By: Gokhan
Originally Posted By: KCJeep
Great thread!

I can't understand all of it but HTHS intrigues me as a habitual blender. Lately I've been focused more on HTHS in my blends, but I don't really know how to calculate it. If you say it's linear will a simple average using accurate percentages work?

My Jeep is spec'd for 10w30, most PCMO of which are around 3.0 to 3.1 HTHS. It runs fantastic though on 5w30 of 2.9 to 3.0 HTHS. I am however afraid to use 5W30 in the summer due to shear concerns. I've recently found several 5w30 High Mileage oils with HTHS in the 3.2 to 3.3 range, so I'm thinking there is absolutely no reason not to run those in the warm months. They have higher VI and a higher HTHS than many 10w30's.

Am I correct in my thinking here? If so why on earth would I run a 10w30 ever?

I wouldn't worry much about permanent oil shear with a synthetic 5w30. A conventional 5w30 shears quite a bit a synthetic 5w30 not so much. Therefore, I think a synthetic 5w30 would always be an excellent replacement for a conventional 10w30, which some conventional-oil users prefer over conventional 5w30 because of shear stability.


Thank you, I don't use synthetics but a couple of the High Mileage oils I mentioned are synthetic blends, I'd think those okay then since I don't do long OCI's due to warranty?

 

[KCJeep]

Originally Posted By: CATERHAM
Originally Posted By: KCJeep
I can't understand all of it but HTHS intrigues me as a habitual blender. Lately I've been focused more on HTHS in my blends, but I don't really know how to calculate it. If you say it's linear will a simple average using accurate percentages work?

Yes averaging is pretty close but it's better to simply use a viscosity calculator and just substitute the HTHSV spec' for the KV100.


Thanks will do it that way then.