3 labs test amsoil ATM 10w30 VOA

[demarpaint]

Originally Posted By: simple_gifts

Quote:

What you say is true, however you have to have some faith in the data or it's worthless.

How much faith is warranted? Very few people understand the testing methodology, the allowed variation etc. It is just a WAG that "these numbers aren't good enough"

Quote:

At this point I'd have no clue on which one to believe,

It is not about belief. Ask a few carpenters to cut something @ 5' 3 3/4" inches. It won't be 5' 3 3/4" but that doesn't matter as it will be "good enough" for what it is being used. I understand your concern. I'm not defending the variations wrt to viscosity; I'm suggesting (having worked in mfg) that "knowing" is a fleeting thing. Any measuring device needs to be 10 times more accurate than the precision to which you are trying to measure. If an oil fails into the 'light 30 weight' it might read 'high 20, light 30 even medium 30' it may make little difference if the oil reads 50 weight after 11K miles of service. If anything it may highlight the potential foolishness of choosing one oil over another based on .x variations in cst.

Your point is well taken, and I'm not trying to be confrontational please understand that. Again my beef is with the 40C and 100C numbers, % wise there's quite a difference. If in your carpenter example one carpenter was off as much % wise in his measurements there'd be a big problem, and he'd probably be looking for another job.. I work with them and I know how they measure, the difference could be in the tape measures used, or how one reads it, etc. Maybe I'm reading a little too much into this.

 

[Artem]

I always thought that labs should use the same, standard equipment for any type of analysis. Otherwise, who do you believe if one lab says you've got a problem but another shows you're fine? What if I go get an employee drug tested and one lab says he's on drugs while another shows minor signs, therefor giving him a clean bill of health?! What kind of [censored] is what? Labs need to use the same equipment otherwise these types of tests as the OP posted makes one lose all faith in the process. Because of this, I stopped looking at oil additive numbers in the lab reports a long time ago. The same oil shows different numbers from different labs. I now mostly looks at viscosity, TBN and General wear numbers for data. Even that could be flawed though so it's anyone's guess... I say we test available labs some more, find which ones give consistent numbers and stick to that. Everything else = waste of money because the results are flawed.

 

[Donald]

I would believe OAI as its a Polaris private label. Polaris is mainly involved in commercial, industrial and fleet. I think they can afford top of the line equipment. They are a much larger company than one like Blackstone who caters to auto enthusiasts.

 

[demarpaint]

Originally Posted By: Artem

I always thought that labs should use the same, standard equipment for any type of analysis. Otherwise, who do you believe if one lab says you've got a problem but another shows you're fine? What if I go get an employee drug tested and one lab says he's on drugs while another shows minor signs, therefor giving him a clean bill of health?! What kind of [censored] is what? Labs need to use the same equipment otherwise these types of tests as the OP posted makes one lose all faith in the process. Because of this, I stopped looking at oil additive numbers in the lab reports a long time ago. The same oil shows different numbers from different labs. I now mostly looks at viscosity, TBN and General wear numbers for data. Even that could be flawed though so it's anyone's guess... I say we test available labs some more, find which ones give consistent numbers and stick to that. Everything else = waste of money because the results are flawed.

Exactly- or monitoring important blood levels and having results vary by 10% or more in this case.

 

[JHZR2]

Good standard deviation for chemical process analytical systems is around 6%. So you could do a hokey standard deviation analysis and see what comes of these results. Cleanout/cleanup of systems between samples is also important.

 

[demarpaint]

Originally Posted By: JHZR2

Good standard deviation for chemical process analytical systems is around 6%. So you could do a hokey standard deviation analysis and see what comes of these results. Cleanout/cleanup of systems between samples is also important.

In the case of the 40C and 100C numbers the difference is over 10%. As I said earlier the ppm numbers really don't matter all that much [imo] but the 40C and 100C numbers are way off.

 

[wemay]

Very discouraging for the end user.

 

[BrianC]

Here's a paragraph from machinery lubrication: "POLARIS Laboratories specializes in testing and analyzing oils, fuels, coolants and water-based industrial fluids for equipment reliability. The company currently supports more than 100 private label fluid analysis programs for customers that include Chevron, ConocoPhillips, Kendall, Ingersoll Rand, Terex, MTU and Cummins Filtration. It services more than 80,000 customers in over 15 countries worldwide." Oil Analyzers INC. is one of those customers. I would think if Chevron, ConocoPhillips, etc. use them, we can feel confident that their tests are reasonably accurate considering the cost involved. Here's another article from 2013: http://www.indystar.com/story/money/2013...y-sick/3143515/ The company I work for gets its UOA kits from the local Chevron oil distributor. I think we use Delo 400 15w-40 in our fleet of Motorcoaches. These kits are Polaris kits with Chevron's name on them. Bottom line is Polaris is a big company doing over $20 million annually in sales. That wouldn't be happening if they were doing inaccurate testing.