Viscosity Study

[OldCowboy]

Recently, I read a post that alleged that at any temperature above -10*, there was no appreciable difference between a 0W oil and a 10W oil. Of course, there are no readily available data for viscosity between 40*C (104*F) and -25*C to -35*C, so validating that representation would require some actual measurements.

First, I designed a high-tech, ultra-sophisticated viscosity measuring device pictured here

:

This consists of a small dishwashing detergent bottle with the bottom cut out. I marked two lines on it; the top one to start the stopwatch and the bottom to stop the stopwatch. It is inserted into the neck of a washed Wish Bone Italian Dressing bottle.

For each oil, I took three time measurements at 75*F, 37*F and 0*F respectively. Then I averaged the times and compared them to the times for water at 75*F. Here is the data:

Code:


Relative

Temp 1st 2nd 3rd Avg Viscosity



75*F

Water 28.78 28.73 29.68 29.06 1.00



10W40 59.64 59.24 60.42 59.77 2.06

0W40 50.49 54.05 51.49 52.01 1.79

5W20 40.00 40.60 39.23 39.94 1.37



37*F

10W40 170.37 166.16 168.79 168.44 5.80

0W40 121.74 119.16 120.68 120.53 4.15

5W20 82.07 80.63 81.72 81.47 2.80



0*F

10W40 581.66 591.75 587.23 586.88 20.20

0W40 395.21 402.36 407.41 401.66 13.82

5W20 281.33 278.16 285.39 281.63 9.69




It's pretty obvious that at 75*F, there is not a huge difference between a 10W40 and a 0W40 oil. Of course, the 5W20 is less viscous than either of the 40 weight oils.

At 37*F, the 10W40 is 28% more viscous than the 0W40 oil. The 10W40 oil is 52% more viscous than the 5W20 oil and the 0W40 oil is 33% more viscous than the 5W20 oil.

At 0*F, the 10W40 is 32% more viscous than the 0W40 oil. The 10W40 oil is 52% more viscous than the 5W20 oil and the 0W40 oil is 30% more viscous than the 5W20 oil.

While I agree that at 75*F there is no appreciable difference between 10W and 0W oil, it is clear that once the temp drops to 40*F there starts to be an increasing viscosity difference between 10W and 0W oil.

 

[BobFout]

If you plug those oils into widman's calculator does it agree?

 

[friendly_jacek]

Well, it all depends on various oils. I can tell you that GC 0W30 is thicker in most temps than typical 10W30.

 

[tinmanSC]

Wow! nifty data! What oils were these? all M1?

 

[crinkles]

great! thanks for your effort, but why 5w-20 and not 5w-40 to stay consistent with the 10w and 0w-40 oils?

 

[chubbs1]

OldCowboy, I would have to say you have brought a lot of thought and enthusiasm to BITOG. I appreciate your posts.

 

[A_Harman]

That's pretty cool! You've picked up on something that has been bothering me about oil viscosity reporting in the API universe: what the heck is viscosity between the 40C KV number, and the subzero DV number? Most winter-weather starts are done between the temperatures of 0F and 40F, so it would be good to know an oil's viscosity in that range. I have tried using the Widman viscosity calculator to predict viscosity down to the CCS temperatures, and it doesn't even come close to the numbers listed on product data sheets. The oil guru of the company where I work told me once that viscosity calculators can only be relied on down to about 68F.

Since you are measuring the time it takes for oil to flow through an orifice under the influence of variable hydrostatic pressure, that sounds to me like a kinematic viscosity measurement since the density of the oil will effect the pressure and consequently the time. If I recall correctly, it is similar to a Saybolt viscosity measurement, which also reports time for an oil to flow in a standardized apparatus. I don't think that the viscosity is linearly proportional to time, though. With a little analysis, I think we could boil your data down to an actual viscosity number.

 

[chad8]

nice work. doesn't this also prove that people using 0w oils are fooling themselves by thinking they flow better that 5w oil .
oil probably reaches 75 degrees within a minute of start up.

 

[shanneba]

Originally Posted By: chad8
nice work. doesn't this also prove that people using 0w oils are fooling themselves by thinking they flow better that 5w oil .
oil probably reaches 75 degrees within a minute of start up.


Starting my BMW with a 49F oil temp, it takes about 4 minutes for the oil temp to reach 75F.

I don't have any recorded data from colder temps, that is about as cold as my garage gets

 

[OldCowboy]

Originally Posted By: crinkles
great! thanks for your effort, but why 5w-20 and not 5w-40 to stay consistent with the 10w and 0w-40 oils?


Well, the 5W20 was more for my interest. Plus I had it on hand.

I really wasn't trying to do any comparison between 20 weight and 40 weight oils. I was really interested in a thickening comparison between 10W and 0W oils.

 

[OldCowboy]

Originally Posted By: tinmanSC
Wow! nifty data! What oils were these? all M1?


The two 40 weight oils were M1. The 5W20 was PP.

 

[OldCowboy]

Originally Posted By: BobFout
If you plug those oils into widman's calculator does it agree?


I'm not familiar with widman's calculator. Please provide a link.

 

[A_Harman]

http://www.widman.biz/English/Calculators/Operational.html

 

[OldCowboy]

Originally Posted By: chad8
nice work. doesn't this also prove that people using 0w oils are fooling themselves by thinking they flow better that 5w oil .


I don't think it says anything about a 0w to 5W comparison, since the 5W oil was a 20 weight and the 0W was a 40 weight. If you look at Amsoil data, all three of their 30 weight oils have a viscosity of 10.5 at 100*C. At 40*C (104*F) they rank 0W
Of course, different oils have different viscosity indexes, so down to about 70*F, there probably isn't a huge difference between 0W and 5W. But as the temp continues to drop, the spread will be wider.

 

[BobFout]

Originally Posted By: OldCowboy
Originally Posted By: BobFout
If you plug those oils into widman's calculator does it agree?


I'm not familiar with widman's calculator. Please provide a link.


http://www.widman.biz/English/Calculators/Graph.html

 

[OldCowboy]

Originally Posted By: BobFout
If you plug those oils into widman's calculator does it agree?


Widman's Calculator shows more spread than I saw:

Code:


0*F 37*F 75*F

10W40 3927 759 221

0W40 1956 466 156

5W20 1503 316 100


At 0*F (-17*C) it predicted 0W40 would be 50% as viscous as 10W40 and 5W20 would be 38% as viscous as 10w40. My results were 68% and 48%.

At 37*F (3*C) it predicted 0W40 would be 61% as viscous as 10W40 and 5W20 would be 41% as viscous as 10W40. My results were 72% and 48%.

At 75*F (23*C) it predicted 0W40 would be 70% as viscous as 10W40 and 5W20 would be 45% as viscous as 10W40. My results were 87% and 67%.

Because my measuring was done at 75*F, I could explain the differences between my results and the Widman Calculator results by the oils warming during the measuring process. However, my 75*F measurements were done with everything at 75*F. So, I really don't know how to explain the differences.

 

[bustednutz]

Man, I wish I had that kind of extra time on my hands.

 

[Beagle2000]

Originally Posted By: bustednutz
Man, I wish I had that kind of extra time on my hands.


+1

Truthfully, I find it completely pointless.

 

[jstutz]

Originally Posted By: Beagle2000
Originally Posted By: bustednutz
Man, I wish I had that kind of extra time on my hands.


+1

Truthfully, I find it completely pointless.


Posts like this, is what is pointless. It added nothing to the tread at all except an insult. I really hate seeing stuff like this.

 

[river_rat]

Originally Posted By: bustednutz
Man, I wish I had that kind of extra time on my hands.

Should have thought of that before you had kids.

I think its a cool test anyway. Thanks, OldCowboy you're right there is not much easy to decipher info on where the XW starts to show a difference.