HTHS vis spec trumps the Kinematic 100C vis spec'

Status
Not open for further replies.
Joined
Nov 18, 2005
Messages
10,146
Location
Burlington, Ontario, Canada
The kinematic way of measuring viscosity of a fluid is very simple indeed. Take a steel ball and measure the time it takes to fall under the force of gravity through a column of (in our case) oil. The longer it takes the more viscous or thicker the oil.
50 years ago when all we had was single grade mineral oils (Okay there was also Castor oil, a natural ester, but that was used primarily for racing applications), you could reliably compare the kinematic viscosities of different brands of oil and if they had the same 100C kinematic viscosity spec' they would have the same "operating viscosity" in an engine. Operating viscosity is the flow rate of the oil at the normal operating temperature of the oil in an engine. The best proxy for the operating viscosity is oil pressure as determined by an oil pressure gauge. The higher the oil pressure reading (it actually measures back pressure) the higher the viscosity of the oil.

But with the move to multi-grade oils due to the use of polymer thickeners, that all changed. These viscosity improvers do dramatically increase the kinematic viscosity of an oil but not linearly when compared to the operational viscosity in an engine. Take for example a 10W-30 oil, which is a 10wt oil with the addition of sufficient polymer thickeners to bring it's 100C viscosity up to that of a 30wt oil. The problem is, a 10W-30 with the same 100C viscosity as a straight 30wt oil will still have a lower operational viscosity (lower oil pressure) because of the temporary shear (compression) of the polymers in the bearings of an engine. Consiquently to compensate for this affect the 10W-30 must have a sufficiently higher 100C viscosity that the non polymer containing straight 30wt oil to maintain the same operating viscosity. Of course we are just talking about temporary shear and nor permanent shear which was another problem of the early multi-grade oils.

What's different about the high temperature/high stress (HTHS) at 150C viscosity spec', other than being a much more expensive measure of viscosity, is that does incorporate the effect of shear since the measure is taken under preasure. This measure of viscosity correlates more directly with the viscosity of oil flowing through the bearings of an engine.

The introduction of synthetic oils allowed for the viscosity index (VI), a measure of an oils resistance to thinning with increased temperature, to rise significantly without the addition of polymer thickeners or much less of their use.
Naturally high VI syn oils have lower kinematic viscosities compared to mineral oils to maintain a given operating viscosity.

Today, we have PCMO's and HDEO's with every combination imaginable of base oil, GP II through GP V, and with all kinds and qualities of viscosity improvers. Consiquently trying to predict what the actual operational viscosity will be from the lowly kinematic 100C viscosity spec' is near impossible. There are 10 cSt at 100C oils that have higher operational viscosities than 14 cSt oils. The answer is the very accurate HTHS vis spec. Dispite the fact that the viscosity measure is
taken at 150C it still correlates almost uncannily precisely with the operational viscosity of an engine at 100C and at operational viscosities even lower than that.

That's one reason why fuel economy correlates with the HTHS vis' of an oil and not the kinematic 100C spec':

http://www.bobistheoilguy.com/forums/ubbthreads.php?ubb=showflat&Number=1849041&page=2

The SAE have been criticised for not incorporating the more accurate HTHS vis measure into the SAE grading system. For the 20wt grade we have a minimum HTHS vis of 2.6 cP but no maximum. Same for the 30wt grade which must be greater than 2.9 cP with no maximum, same for the 40wt grade, and nothing at all for 50wts and heavier.

http://books.google.com/books?id=Fu-99Mc...p;q&f=false

In the tradition of BITOG I've personally compared almost a dozen oils so far. The bottom line is this. Oils with the same HTHS vis have the same operational viscosity regardless of their 100C k'vis spec's. Oils with the higher HTHS vis spec' are indeed heavier than oils with lower HTHS vis spec's at normal operating temp's, again regardless of their 100C k'vis spec's.

A couple of popular examples;
RL 5W-20 ( 100C vis 9.1 cSt,HTHS vis 3.3 cP) is heavier than M1 5W-30 (100C vis 11.3 cSt,HTHS 3.09 cP) and PP 5W-30 (100C vis 10.3 cSt,HTHS 3.1 cP). Both M1 and PP have the same operational viscosity.
GC (100C vis 12.2 cSt,HTHS 3.5-3.6 cP) is lighter than RL 5W-30 (100C vis 10.6 cSt, HTHS 3.8 cP) which is slightly heavier than M1 0W-40 (100C 14 cSt, HTHS 3.7 cP).

For those wanting to replicate my experience and
to make the comparison accurately, in addition to an oil pressure gauge you must have a oil temp gauge and tachometer. The oil comparisons must be made at the same engine rpm and oil temperature. Again what you're measuring is oil pressure (back pressure to be more precise) which is a proxy for the operational viscosity of the oil at that temperature.

So what's good about the 100C k'vis spec'?
Well it's a very inexpensive measure of viscosity so at the very least it gives you a ball park measure.
It's also fine when comparing oils with exactly the same chemistry and VII composition although that is hard to know at times.
It's still a necessary spec' along with the 40C k'vis spec' for determining the viscosity index of an oil.
And from a practical view point it's still a VOA reference point for UOA; ie., to determining viscosity loss. You are comparing the oil against it's virginal self.

There are an awful lot of myths associated with the HTHS viscosity measure and it's significance.
Everything from it being somewhat esoteric and not really applicable to daily street use applications that don't see temp's much if ever above 100C. That it's some sort of reserve capacity of an oil and therefore the higher the number the better. To a very critical element of an oil's specification to deal with parts of an engine that get incredably hot regardless of how cool the sump bulk oil temp's may be.

The truth is that it's an accurate measure of an oil's viscosity under stress (pressure) at 150C.
It's a very useful spec' for possibly hot running engines and as such it's an engine lubrication spec' by many manufacturers.
It's a determining factor in the fuel efficiency of an oil.
But probably it's most practical use is as a comparative determinate of an oil operational viscosity.

Cheers

Peter M
 
I do understand what you've said here, but I still think that it is a VERY valid argument that increasing the use of HT/HS specs is not useful at all due to the extreme environment that reading is taken in.

Think about the daily use MOST cars see - start up, putter 3-10 miles to work at low traffic speeds in average weather, with very little revs or throttle use at all. Oil will get warm, but nowhere near 150C.

Even if by some chance the oil DID get that hot, the other part of the equation is the stress part - HT/HS measures protection against heat and stress, usually from high revs. Not a factor in MOST daily driving scenarios.
 
That was long... Given that fuel economy correlates to HTHS and does not to 100 degree viscosity I would agree it is a more useful measurement. Obviously an engine sees a difference in one and not the other, so it's simple logic.

Unfortunately, for the purposes of UOA, it's not as cheap or easy to measure.

Also, it's not really fair that a mass market oil needs to be labeled as "energy conserving" and therefore probably needs a lower HTHS than something like RedLine which does not.... for those who will read your writing as "get the highest HTHS you can at any xw-xx and it will be the best" there may be situations where that is true, but many auto manufacturers (for whatever reason) spec "energy conserving" oil.
 
Except, if you want to predict how thick an oil actually will be in an engine at normal operating temp's you must look past the 100C k'vis spec.
Take a 20wt oil as an example. Their k'vis 100C spec' can range from 6.8 cSt (Wolfshead) to almost 9 cSt, for oils with the same HTHS vis of 2.6 cP. But in reality they all have the same operational viscosities at normal operating temps.

If you want a heavier 20wt oil you must choose an oil with a HTHS vis of 2.7 cP or higher.
 
bepparb, I agree with you and if you read what I said more closely I said getting the highest HTHS number for the grade is a misunderstanding of what the HTHS vis' spec actually is.
 
Pennzoil's detailed article on their new oil says fuel economy is related to viscosity - HTHS was not factored in.

Somehow lots of cars have zero problems with supposedly less than optimal HTHS ratings.
Somewhere it has been said it had to be 3.1 or better [whatever].
This has been repeated and defended by others.
More protection is better, but real life is overwhelmingly showing less is OK.
So many oils have 2.6-2.8 or so, and are great.
 
I thought this all was common knowledge. I think fifty years ago there was multi vis. What came after a ball dropping in oil goes faster or slower if oil is thinner or thicker? Please don't say "to you not much." JK, I learned that recently.
 
Last edited:
Originally Posted By: addyguy
I do understand what you've said here, but I still think that it is a VERY valid argument that increasing the use of HT/HS specs is not useful at all due to the extreme environment that reading is taken in.

Think about the daily use MOST cars see - start up, putter 3-10 miles to work at low traffic speeds in average weather, with very little revs or throttle use at all. Oil will get warm, but nowhere near 150C.

Even if by some chance the oil DID get that hot, the other part of the equation is the stress part - HT/HS measures protection against heat and stress, usually from high revs. Not a factor in MOST daily driving scenarios.


It is usefull regardless of what conditions as the HTHS is measured for areas in the engine where the oil does reach that temp. Its only in the high load areas isnt it ? My understanding is that the oil will reach those temps regardless in the bearings and piston area, cam lobes etc. Correct me if i am wrong.

I also thought this was the reason that higher hths affected fuel economy as the oil just did not thin/shear down as much as people thought it would in those sections of the engine. Somehow they must be able to measure the oil as different areas on the engine as its flowing.
 
Last edited:
That is one of the myths I mentioned.

The most viscosity sensitive areas of the engine are the bearings followed by the piston ring/cylinder area. These areas, particularly the bearings, get a rapid flow of oil. At maximum oil flow a gallon of oil is flowing through the bearings and the entire engine in about 12 seconds and if you have a high capacity oil pump as little as 5 seconds. So on a single pass, so to speak, not a great amount of heat is transferred from the bearings, pistons and cylinder walls to the oil. Therefore in a typical engine with a 4 to 6 quart sump there isn't much difference between the sump oil temps and the oil in these critical parts of the engine.

The valve train has the lowest viscosity demands in an engine, high rapid flow is what's important, particularly at start up.
 
Good job Caterham. The pressures in the HTHS test are relatively low. There was some thread where we discovered that...I doubt I'd find it without a lot of searching. Stress and strain rates are very high in the HTHS test though. Recent research has been done on effects of localized oil pressures encountered in some parts of engines and the main conclusion of one of the researchers was that effect of high pressures is just as important as effect of high shear rates. The researcher I'm speaking of is Scott Bair...see here: http://www.bobistheoilguy.com/forums/ubb...true#Post941621
 
I dunno if I can agree here.

GC has a HT/HS of 3.5cP, so does 15w-40 HD (E7)

In Audi-Castrol tests, GC 0w-30 gained 2% mpg over the 15w-40 control. Credit need be given to the reduced kinematic visc, especially during warm-up, since the HT/HS is actually the same.

Pumping losses are certainly not modeled by centi-poise at +150c.
 
Originally Posted By: Tom NJ
As for the effect of these viscosities on fuel economy, most frictional losses occur in the bearings and ring/cylinder wall interface. Both of these areas are under high shear rates, so all else being equal the HTHS viscosity should correlate better with mileage than kinematic viscosity. Of course, all things are rarely equal, so friction modifiers, polar base oils, VI Improver quantity and type, engine shear rates, and temperature will have some influence on this correlation. Furthermore, if the HTHS viscosity gets too low, friction can increase as parts move into elastohydrodynamic or boundary regimes (Stribeck curve), so the correlation of HTHS viscosity to fuel economy is only valid within a range.

So in conclusion, fuel economy generally correlates to HTHS viscosity, except when it doesn't.
grin2.gif
 
Caterham, you write:


So what's good about the 100C k'vis spec'?
Well it's a very inexpensive measure of viscosity so at the very least it gives you a ball park measure.
It's also fine when comparing oils with exactly the same chemistry and VII composition although that is hard to know at times.
It's still a necessary spec' along with the 40C k'vis spec' for determining the viscosity index of an oil.
And from a practical view point it's still a VOA reference point for UOA; ie., to determining viscosity loss. You are comparing the oil against it's virginal self.



I was under the impression that the 100 and 40 C kinematic viscosities were good measures of how the fluid flows at those temperatures. The 100 C performance should indicate how well the oil gets into the nooks and cranies at high temperatures. We want a low number because that means it flows more freely at that temperature.

So in my case, since I race the car frequently, I should be selecting oils with good figures in that 100 C spec.
 
Last edited:
Here's one point you've brought up a few times that I just can't rationalize.

You've stated that you can take any number of lubes and merely take their HTHS to determine their REAL operational viscosity. If that were the case, then VII would be a 100% waste of product. You could merely run without them ..if as you allege they have no effect on operational viscosity.

Where's the missing link in the logic for me?
 
Thanks JAG.

Yes, that was an excellent posting and I see you made many of the points I have made. It would be so much clearer if we simply did away with the kinematic method of measuring viscosity in an IC engine. But as you said "it's such an easy test to do" and cheap, that it won't likely be replaced any time soon.
 
river_rat, thanks.

dparm, thanks for your thoughtful comments.
The 40C kinematic viscosity spec is at least an indication of how viscose an oil is at that temperature and inconjunction with the 100C spec we can determine an oils VI. With that we can estimate using various viscosity calculators the kinematic viscosity of an oil down to at least the freezing point (0C).
So it's a useful tool, just not an especially accurate one particularly if you are using it to compare viscosities of oils of markedly different chemistries.

The kinematic 100C spec' taken by itself can be over looked if you know the more accurate HTHS vis for the oil. For a given
HTHS vis spec the kinematic viscosity of various oils could range up to 4 cSt. Using a 30-40 wt oil that could be the difference between 10 cSt and 14 cSt at 100C. But I would argue that the difference in flow under the force of gravity alone at that temperature is negligible. That same 10 to 14 cSt oil will have a kinematic viscosity at 0C of about 500 cSt and 0C is not especially cold.
 
Originally Posted By: Gary Allan
Here's one point you've brought up a few times that I just can't rationalize.

You've stated that you can take any number of lubes and merely take their HTHS to determine their REAL operational viscosity. If that were the case, then VII would be a 100% waste of product. You could merely run without them ..if as you allege they have no effect on operational viscosity.

Where's the missing link in the logic for me?


Gary, thanks for your input.

I've never said VII's have no effect on operational viscosity, quite to the contrary.
Using the simply 10W-30 vs straight 30wt mineral oil example I mentioned. The 10W-30 being basically a 10wt oil with a load of VII polymers in it will of course be lighter than the straight 30wt at cold temperatures. But at 100C if the 10w-30 has just the same kinematic viscosity as the straight 30wt your operational viscosity in an engine will be lower due to the compressable nature of the polymers under stress. To compensate you will have to add more polymers thereby raising the kinematic viscosity of the 10W-30 beyond that of the straight 30wt oil. How much more? Enought to equalize their HTHS viscosities.

Adding VII's to an oil increases both the 100C k'vis spec and the HTHS vis of the oil, they just increase the 100C k'vis spec at a disproportionately higher rate to the HTHS vis.
 
Okay. Now we're getting somewhere in flatland. Paul Harvey's "the rest of the story".

I'll blurt out one of these "how did you get there from here" utterances from time to time. You'll often find yourself jumping over a few steps in the logic train in your presentation. There's also instances where one may know what one wants to say, but not necessarily have the ability to express it well for others (there are vast differences in "others")
grin2.gif


btw- I don't think they've made a 10w-30 that way for a very long time, but it was one way to get a 10w-30.
 
Yes, I understand that VI, kV @ 40, and kV @ 100C are all related with a formula and you pretty easily estimate a missing value.
 
Status
Not open for further replies.
Back
Top