Cognitive oil testing - database of friction tests

Status
Not open for further replies.
Joined
Jun 12, 2018
Messages
5
Location
Hungary
Hello Ladies and Gentlemen,

I am new guy to this forum, and I would like to start as contributing to the masses. I have found a guy in Eastern Europe who tests motor oils on friction and couple of other tests. His youtube channel is called Piotr Tester where he has all the friction tests on video with preparation and the test itself along with the measurements and final conclusions. His methods are far from an academic level but hey, samples show what's going on, and he is generally stable with his testing process. He tests on room temp, 100 degree celsius (212F?) and under high load.

I am not aiming at the person itself, but the fact that he has a neat database now with all the friction tests which is available in his video descriptions (actually 2 tables) are pretty neat. It let's everyone see how well the different brands and makes hold up against friction and in my point of view, that is one of (if not the most) important factors of oil.

Last, but not least he tests US and EU products as well. (I believe all of them are bought in the EU, so that might be a grey zone.)

The youtube channel's name is Piotr Tester.

I would be also happy to see/hear about any other opinions, faults in testing, or some similar databases somewhere.

Wish you a nice day!
 
That's not testing friction, that's testing wear. It's a great way of telling which oil is good in that particular wear testing apparatus, but there's no correlation to wear (or friction or any other parameter) in an engine.
 
Originally Posted By: weasley
That's not testing friction, that's testing wear. It's a great way of telling which oil is good in that particular wear testing apparatus, but there's no correlation to wear (or friction or any other parameter) in an engine.
Glad you explained that friction is not really the same thing as wear. Friction is just the resistance to relative motion, and wear can be zero (100% hydrodynamic conditions) with some friction.
When two parts go metal-to-metal, that is more friction with some wear too. Non-engineers use the word "friction" strictly to mean when 2 objects are touching, and we engineers use it in the broader sense.

This whole thread subject sounds like 540Rat's blog. https://540ratblog.wordpress.com/2013/06/20/motor-oil-wear-test-ranking/
which is "OK" in some sense, but there is just no replacing actual engine testing where the ACTUAL tribological conditions are there, not something from a non-engine device.

.... That said, engineers do use devices for rolling, rolling-sliding, and at different temperatures and shear rates, such as
in this video (skip to the 4:38 point):

https://www.youtube.com/watch?v=pUr7-Ug1Uqc

or a short video at: https://www.bizjournals.com/houston/video/1zbXNobToPHgjzDdsMyc76mqwx3_dKXz

They do all the "rig" testing to screen motor oil formulations, and they actively try to correlate those results to SN, GF-5, dexos1, etc. real engine tests.
 
Those so called friction testing contraptions have been around for decades. If you run into one, you can be sure a shyster is attached to it.
banana2.gif
banana2.gif
banana2.gif
 
I know that these are the "one arm bandit" tests, but since he has a bunch of oils tested, could we take anything from it? Could we link these results to some property of the oil, like the amount of some sort of aditive or base oil quality?

For example, he test Amsoil 0W40 and 5W40 with different results.
 
Originally Posted by LeoStrop
I know that these are the "one arm bandit" tests, but since he has a bunch of oils tested, could we take anything from it? Could we link these results to some property of the oil, like the amount of some sort of aditive or base oil quality?

For example, he test Amsoil 0W40 and 5W40 with different results.

No not a thing. There is a large error band associated with that type of testing and there is no statistical analysis of the results being performed. No conclusions whatsoever can be determined from that "data".

It's actually worse than that but this is enough to kill it by itself.
 
Originally Posted by GaryPoe
I wouldnt consider it a one arm bandit test. They are testing their own products for results to make their own product better.

Then I'd never buy one of their products since they have no idea what they are doing.
 
Status
Not open for further replies.
Back
Top