Thin cleans better, allows longer OCI than thick!

[Shannow]

Originally Posted By: Gokhan
As you can see, their better PAO, or PAO plus, is directly promoted as having a higher BOQI than their standard PAO -- plus simply meaning higher BOQI -- which immediately translates into longer oil-change intervals (OCI's) or less oxidation as their first selling point. It even claims less wear, probably because of a lower coefficient of friction.


NO, just NO...that are pointing out benefits of their oil...Improved volatility and low temperature fluidity (amongst a whole lot of other advantages)...they are not advocating your number.

"Directly promoted as having a higher BOQI"...what rubbish...

I could similarly say that they have directly promoted having a higher Viscosity Index to Pour Point ratio (in kelvins)...it would be similarly nonsensical.

Now speaking of rubbish...

Originally Posted By: Gokhan
BOQI is not really intended for Group V oils, which could be anything. The ester oil you pointed out is not even for regular runs but for maintenance. Nevertheless, ester base oil has poor hydrolytic stability and that's probably what the lower BOQI is telling you.


Please, please give us some explanation as to how you come out with THAT gem ?

CCS and NOACK relate to hydrolytic stability ?

 

[ZeeOSix]

Moon phase, ocean tides, Mars retrograde during planetary alignment and Noack must have some type of correlation.

 

[JAG]

Indeed, BOQI is not just an indicator of those things, but also provides insights into the subtle notes of vanilla, cinnamon, and chocolate residing within base oils in various proportions. I’m now convinced that it is pointless to try to reason with him.

 

[Gokhan]

The BOQI idea has been patented by Chevron and I linked and explained that. It's used in many of Chevron's base-oil publications.

So, JAG and Shannow object to the patent because they know better than the Chevron base-oil researches.

 

[SR5]

Originally Posted By: Gokhan


My understanding of BOQI is that it measures the paraffinic content. The units for 1/BOQI are pressure (stress) if you realize that the units for NOACK are inverse time. Basically it tells you how much pressure (stress) you need to pull the molecules away from each other. If you have infinite BOQI, you have zero stress required to pull the molecules apart, which means you have perfectly linear paraffinic chains, sliding effortlessly by each other.


Gokhan,
I like how you are doing unit and dimensional analysis to examine your theory. I wish all my students at work put the same effort in as you do. However dimensional analysis is one of the tools used to check if a theory is dimensionally valid, it's not really a tool to develop a theory or prove a theory. It's just a first order filter for new ideas and concepts.

Consider the concept of mechanical work.
Work = Force x Distance, and is the transfer of Energy.

Force = Mass x Acceleration
The SI units for mass is the kilogram (kg) and acceleration is meters per second per second (ms^-2). So a Force is measured in kilograms meters per second per second (kgms^-2) which we call a Newton (N) for simplicity.

So the units for Work is Newton-meters, but given it's a form of energy, it's also called a Joule (J) but in reality it's a kilogram meter squared per second per second (kg m^2 s^-2).

Now the dimensionally equivalent Torque (T) is also measured in Newton-meters (Nm) but it is a twisting force and not a form of Work or Energy in it's static form.

That's the difference between "force through a distance" (Work in Joules) and "force at a distance" (Torque in Nm). They are both dimensionally equivalent, but distinct and different physical concepts.

The physical (or chemical) concept must come first, then the details of the units are sorted out, then checked with dimensional analysis. Not the reverse order.

Dimensional / unit analysis is not the correct tool for creating a physical concept.

 

[Garak]

Originally Posted By: SR5
Dimensional / unit analysis is not the correct tool for creating a physical concept.

This, for sure. If the units don't work, the concept sure won't. The units working won't ensure that the concept does.

 

[CR94]

Originally Posted By: SR5
... Now the dimensionally equivalent Torque (T) is also measured in Newton-meters (Nm) but it is a twisting force and not a form of Work or Energy in it's static form.

That's the difference between "force through a distance" (Work in Joules) and "force at a distance" (Torque in Nm). They are both dimensionally equivalent, but distinct and different physical concepts.

The physical (or chemical) concept must come first, then the details of the units are sorted out, then checked with dimensional analysis. Not the reverse order. ...
I agree with your post.

Yet, Newton-meters of torque times a (dimensionless) angle of rotation in radians IS work---assuming the appropriate vector angle relationships. A third kind of quantity can also be measured in Newton-meters. That's freight hauling accomplished, when the vertical force vector is perpendicular to the horizontal distance traveled.

 

[Shannow]

Originally Posted By: Gokhan
This "pseudoscience" is actually the work of Chevron Global Base Oils (presentation). Perhaps you should send a message to John Rosenbaum.

John Rosenbaum (not me) actually patented it:

Chevron US patent on predicting the base-oil quality through CCS and NOACK


OK, for the benefit of the audience...and following on from your re-interpretation of the API ILSAC documents, that is NOT the title, nor the intent of the patent

The actual title is...
METHOD FOR PREDICTING A PROPERTY OFA BASE OIL

Originally Posted By: Patent Summary
A method for predicting a property of a base oil, and a blend chart, are provided. The method for predicting a property of a base oil includes selecting two base stocks, and preparing a chart having the viscometric property under low temperature and the volatility of both base stocks and curves between them that is used to predict whether blends of two base stocks will meet requirements for a finished lubricant.



Originally Posted By: Gokhan
The main difference between his patented "base-oil-quality prediction" and my "base-oil-quality index" is that he mixes two base stocks instead of looking at only one base stock, and he measures NOACK vs. CCS when he changes the ratio of the two base stocks to generate a NOACK vs. CCS curve. In this way, "his BOQI" = constant / (CCS * NOACK)^power, where he calculates the "constant" and "power" by varying the ratio of the two base stocks and measuring the CCS and NOACK for each ratio. In my case, I simply assume that "power = 1" and I try to estimate the "constant" empirically for 0W-xx, 5W-xx, etc., as I don't have the luxury of being able to do better as he can.

In summary, he uses two base stocks at the end points of each curve and varies the ratio to generate different base oils from them and therefore the whole curve. Then, if the base oil you're comparing to the two-base-stock-base-oil mix is below the curve (higher BOQI in my calculation), it's a superior base oil that will meet or exceed the specs of the two-base-stock-base-oil mix and if it's above the curve (lower BOQI in my calculation), it's an inferior base oil that will not meet the specs of the two-base-stock-base-oil mix.


That's not the patent either. There's a lot in it of a new basestock that they have produced of superior performance, as well as his process to "predict a property" of the base oil mix.

He produces a family of curves on NOACK and CCS for various mixes of oil basestocks...that is true.

Then for a target grade (xW - 20, 30, 40, 50, 60) use the simple blend method to target the KV100 required to meet the grade.

Then looks that the NOACK target and CCS target for the new oil formulation, and checks from his pre-prepared curves whether the blend exceeds the design intent or needs "modification" via the addition of a superior basestock to do it...via the introduction of a "trim stock".

Here comes Figure 1

The example given in Example 4 is the modelling of a 10W30 of various basestocks...a range of basestocks from the Chevron slate were chosen that would give the KV100 desired and then the combined chart was consulted to see which blend of basestocks would likely provide the target CCS (2,800 at -25C), and NOACK (14%)…


Quote:
The chart in FIG.1 was referred to, and it was found that the curve between Chevron 110RLV base oil and Chevron 220R base oil fell below and close to the point representing the base oil requirements for the 10W engine oil. This gave a good prediction that the 10W engine oil requirements could be met with a blend of only these two base oils, and not requiring any trim Stock.

0074. If it was desired that the engine oil have mostly Chevron 100R and Chevron 220R base oil, then an amount of trim stock would need to be used to meet the same base oil requirements. Based on the chart one would predict that blending in of Chevron 7R trim stock would be a good choice to bring down the Noack volatility of the base oil blend to within the desired range. Based on the chart one would also predict that blending in of Chevron 5R or Chevron 4R trim stock would be a good choice to bring down the CCS VIS at -25°C. of the base oil blend to within the desired range.


Thus figure 2...showing the plane on which those blends sit, and the likely spot of the 110R, 220R, trimmed with 7R.

The method described in the patent is to predict the potential "W" rating of blends of known basestocks to see if they meet the target CCS and Volatility requirements once the KV100 is decided on...to minimise the costs of both developing a new oil (testing those in the blend), and material cost per quart...per Example 4, should the new basestock (also embodied in the patent) be used (at more expense), or a lower quality basestock pair be "trimmed" with something more expensive.

Originally Posted By: Gokhan
So, JAG and Shannow object to the patent because they know better than the Chevron base-oil researches.


Nope, not at all...it's your BOQI that I have continually and repeatedly objected to...there's a difference.

 

[Gokhan]

Shannow, no... Did you even read the abstract?

"... a chart having the viscometric property under low temperature and the volatility of both base stocks and curves between them that is used to predict whether blends of two base stocks will meet requirements for a finished lubricant."

So, the "property" = "whether blends of two base stocks will meet requirements for a finished lubricant"

If you think I'm overreading, read Claims 1(d), 1(e), and 4.

Their property is exactly the base-oil quality (save a numerical "index" attached to it as I do). They are claiming that if the base oil falls above the curve, then it's of higher quality than the tested base stocks and the finished oil will meet API CJ-4 and ACEA E9. If it falls below the curve, it's of lower quality and it needs to be trimmed with base stocks of higher quality so that the finished oil can meet API CJ-4 and ACEA E9.

 

[Shannow]

Yes, I read it...yes, I read and reread the patent.

Again, you are picking words out of an 11 page document.

As I explained (did you read my post describing the patent), they are using tables that they have generated from blends of their slate of oils to "predict" whether the mix will get to their target "W" rating and volatility...that's all.

You can well say that for two identical CCS, the one with the lower volatility is probably a better base oil. You can well say that for a given KV100 and Volatility the oil with the lower CCS is probably better basestocks.

But to claim an "Index", and that you can "rank" the oils using it is neither supported by the patent, the presentation, nor any of your arguments to date.

The fact that three basestocks from the same supplier and product line, and 6 of their blend designs using those basestocks vary so WILDLY using your index draws the veracity of the "index" into question.

The suggestion that the "index" relates to hydrolytic stability is laughable.

 

[Gokhan]

SR5 and CR94, it's great to see others with physics knowledge here! These are good comments regarding dimensional analysis.

Since the units for CCS*NOACK = (pressure*time) * (1/time) = pressure, I physically interpret it as a microscopic liquid yield strength. I may be wrong, but it seems to be indicating how much force is needed to pull the liquid molecules apart from each other. Since paraffinic molecules are more uniform than naphthenic (naphthene, not naphthalene) and aromatic (such as alkylated naphthalene) molecules, the more paraffinic a base oil is, the lower the CCS*NOACK would be, if my interpretation is correct.

PAO is the most paraffinic base oil. It also achieves the smallest CCS*NOACK or the highest BOQI. GTL is a little less paraffinic and Group III is even less paraffinic, followed by Group II, and Group I. BOQI falls in the same direction. Being more paraffinic indicates a higher base-oil quality because of higher oxidation etc. stability. The downside of paraffinic oils is that they are "dry" oils that don't work well as solvents (solvency) or for attaching to surfaces (lubricity).

Esters and alkylated naphthalenes (AN's) on the other hand are not paraffinic. This would explain why their CCS*NOACK could be much higher or BOQI could be much lower. Esters have good solvency and oxidation stability but poor hydrolitic stability. AN's have good oxidative and hydrolitic stability, decent solvency, but poor CCS and viscosity index. They both have their applications in improving the overall base-oil performance, even though they usually can't be used on their own for most applications. In the Valvoline cleaning-oil patent posted here, BOQI was probably low because of the AN, which has a really poor CCS, not because of the esters. I don't have much data on esters but the BOQI for the polyolester seems to be very high, perhaps higher than PAO.

In conclusion, perhaps a better name for the base-oil-quality index would be paraffinic-content index. Yes, higher paraffinic content usually means a better base oil, but as some strongly objected, esters and AN's -- and even Group I -- have their benefits and they can actually improve the base-oil quality in ways even if they lower the paraffinic content.

Going back to the original Chevron patent, Group V -- the odd base stocks -- are certainly not accepted as directly interchangeable base oils under the API guidelines, meaning if you interchange a base oil with Group V base oils, anything can happen, which requires retesting and recertification for that reason. So, Group V is not really within the scope of the Chevron patent or BOQI, especially with unusual, less common base stocks such as AN's that are usually used in small amounts.

 

[Shannow]

It's Newton second/m^2 for Cp...it's a shear stress...It doesn't matter whether you imagine that it's somehow related to vapour pressure...it isn't.

 

[Shannow]

Originally Posted By: Gokhan
In conclusion, perhaps a better name for the base-oil-quality index would be paraffinic-content index.


Then why on earth does your index give three Spectrasyn PAOs, and the 6 blends made from it 9 different answers ...with a span of 46 through 160 ?

 

[Gokhan]

Originally Posted By: Shannow
It's Newton second/m^2 for Cp...it's a shear stress...It doesn't matter whether you imagine that it's somehow related to vapour pressure...it isn't.

Nope, dynamic viscosity = pressure * time -- not pressure (stress) -- as you know it.

NOACK has the units of 1/time, even though Garak objected to it, saying it should be dimensionless. That's because 2 hours is arbitrary -- plenty of time for gathering reliable, repeatable data. It could be 1 hour, 1 min, or even 1 second. It's the rate of evaporation.

So, time cancels and you get pressure (stress) when you multiply CCS with NOACK.

 

[Shannow]

Originally Posted By: Gokhan
Originally Posted By: Shannow
It's Newton second/m^2 for Cp...it's a shear stress...It doesn't matter whether you imagine that it's somehow related to vapour pressure...it isn't.

Nope, it's pressure * time -- not pressure (stress).


I was explaining that viscosity is shear stress...not vapour pressure...I defined "for Cp"

Originally Posted By: Gokhan
NOACK has the units of 1/time, even though Garak objected to it. That's because 2 hours is arbitrary. It could be 1 hour, 1 min, or even 1 second. It's the rate of evaporation.

So, time cancels and you get pressure (stress).


I agree with Garak...you don't know the evaporation curve inside that 2 hour nominal period, so you can't really use it as a rate. If one oil does it's thing in the first hour, versus another taking an hour and a half...your index gets broken.

But nonetheless, the CCS is related to shear stress...not some relative of vapour pressure in spite of the fact that they share pascals as their unit.

 

[SR5]

Originally Posted By: CR94
Originally Posted By: SR5
... Now the dimensionally equivalent Torque (T) is also measured in Newton-meters (Nm) but it is a twisting force and not a form of Work or Energy in it's static form.

That's the difference between "force through a distance" (Work in Joules) and "force at a distance" (Torque in Nm). They are both dimensionally equivalent, but distinct and different physical concepts.

The physical (or chemical) concept must come first, then the details of the units are sorted out, then checked with dimensional analysis. Not the reverse order. ...
I agree with your post.

Yet, Newton-meters of torque times a (dimensionless) angle of rotation in radians IS work---assuming the appropriate vector angle relationships. A third kind of quantity can also be measured in Newton-meters. That's freight hauling accomplished, when the vertical force vector is perpendicular to the horizontal distance traveled.


Yes I agree with you. But I did make a point of saying Static torque, so I didn't need to introduce the concept of rotational kinetic energy and angular motion.

But for completeness, and keeping it brief.

Angular velocity (w) is the change of angle ( *)with respect to (wrt) time (t). w = d* / dt
Angular acceleration (@) is the change of angular velocity wrt time. @ = dw / dt
Torque (T) is a perpendicular force (Fp) at the end of a length (L). T = Fp x L
The rotational moment of inertia (I) for a mass (m) at a distance (r) is given by I = m r^2
Then it can be shown the T = I x @ which is the rotational equivalent of F = m x a
Rotational Work (Wr) = T x * which is the rotational equivalent of W = F x d
Rotational Kinetic Energy (Kr) = 1/2 x I x w^2 which is the rotation equivalent of linear kinetic energy KE = 1/2 x m x v^2

But I was trying not to go there, especially on a dumb phone, with a small screen and big thumbs.

 

[Gokhan]

Originally Posted By: Shannow
I agree with Garak...you don't know the evaporation curve inside that 2 hour nominal period, so you can't really use it as a rate. If one oil does it's thing in the first hour, versus another taking an hour and a half...your index gets broken.

So, you're saying that it's the average rate of evaporation within the two-hour window. That still doesn't make it dimensionless. It's still a rate as a physical quantity. In other words, NOACK = 9% actually means 0.09/(2 hr), not just 0.09. The unit for the physical quantity is 1/time. Otherwise, it would make no sense at all as a physical quantity because it relates to oil consumption, which is expressed as oil consumed per mile or oil consumed per hour. You never say my car consumes 0.5 quarts of oil -- you need to specify the miles or time.

 

[Shannow]

I'll accept it...but as there's a curve, it's another spanner in your works as a ranking tool for base oil quality.

For instance, if an oil is made using a "dumbbell" process, of very light, and very heavy, you will have a different curve to something made using two much closer basestocks.

Regardless, viscosity is still based on shear stress...

 

[bbhero]

You all are working this out quite well and in a good manner...

In the mean time while I await a finalized consensus... I'm going to eat some donuts

 

[SR5]

Evaporate Loss (Noack) is the weight loss % after 1 hour at 250 C. That's according to my ACEA sequences that I'm reading right now.

To me that makes the units of Noack kg per second or something similar like grams per hour. But Mass per Time in some form.

Remember velocity (or speed) is change in distance (or position) in a period of time, so it's units are meters per second or miles per hour, some sort of distance per time.

Just because the first part is a change, and just because that change is expressed as a percent, that doesn't mean the unit disappears. If I increase my speed by 10%, I'm still traveling in miles per hour. If I diet and drop my weight by 10%, it's not a dimensionless quantity, I've still lost kg. You can measure things by an absolute change or a relative change, but the relative change is a normalization of sorts, for comparison purposes, not a reduction in the original dimensions or units.

Noack is a change in mass over time, it's SI units should kg/s or similar.