Two Labs, Same Oil, Different TBN Results...

I done the same thing, sending 2 samples to separate labs, on 2 occasions.
AGAT vs FluidLife, & AGAT vs Blackstone.

I also found that Blackstone test came back with different results than AGAT.

Fluidlife, & AGAT were quite similar, & I have stuck with AGAT just because I can drop off my own samples, & do not need to spend any thig on postage, or shipping.

I was mostly concerned about fuel dilution. Both AGAT, & Fluid Life us Gas Chromotography, where Blackstone used an interpolation method.

AGAT flagged my samples, with 3% fuel dilution, & Boron depletion & recommended oil change, while Blackstone suggested my samples were thin but no fuel dilution.

Thanks for your info. AGAT was another option I was considering. I may do this again with CAT and AGAT for fun.
 
Ok BS has re-tested the oil and replied. This was certainly a impressive level of customer service that went above and beyond what I expected.

Here is the response:

Hi XXXXX,

We re-did the TBN test and it came back slightly higher at 3.5. It's unclear what might have causes the discrepancy but you can rest assured that all three readings show plenty of active additive left in the oil. I passed this situation along to our company president, Ryan Stark, and he said there are multiple ways to test for the TBN, which are tailored to different applications. The a modified version of ASTM

D4739 method we use is most applicable to used oil, though there are other methods more applicable to unused oil. Maybe different testing methods factor into the difference. I can say that 6.2 would be an usually high TBN for oil with 200+ hours on it, according to the method we use, since many virgin oils start out around there, or slightly higher. Let me know if you have any other questions.

Thanks,

Suzi
 
  • Like
Reactions: a5m
That almost full number difference isn't surprising since D4739 uses a weaker acid and less polar solvent system that doesn't give as repeatable results.

If you were to contact CAT and ask them to rerun it, assuming their reported 6.2 TBN isn't a typo, I would expect the result to be closer.

The real thing that we should be looking at is oxidation. After all, that's ultimately what we're trying to prevent is oxidation that causes varnish and sludge. A concern for me, using D4739 could completely ignore a large concentration of weaker organic acids. It could show a decent TBN of >2.5 with "plenty of active additive left" but the oxidation could be steadily increasing. A better picture of the oil's remaining service life would be to look at D2896 and oxidation value. I don't agree with them saying D4739 is better for used oil.
 
Last edited:
RDY4WAR, thanks for your insights.

On the report from cat it has data for oxidation. States 14 UFM Oxidation. No idea what a "good" number is for oxidation.
 
I don't agree with them saying D4739 is better for used oil.
It's not just Blackstone saying this. From astm.org's page on D4739:
5.3.2 When the base number of in-service or at-term oil is required, this test method is preferred because in many cases, especially for internal combustion engine oils, weakly basic degradation products are possible. Test Method D2896 will titrate these, thus giving a false value of essential basicity.
This test method may or may not titrate these weak acids.

5.3.3 When the loss of base number value, as the oils proceed in service, is the consideration, this test method is to be preferred and all values including the unused oil shall be determined by this test method. Base numbers obtained by this test method shall not be related to base numbers obtained by another test method such as Test Method D2896.
 
  • Like
Reactions: a5m
Another lab that still uses D2896 for UOAs is Wearcheck. Here is what they wrote to me some time back:

We use ASTM D2896, which I guess is the older method. However, it is more accurate than the newer one (D4739) which does give lower numbers. On average, D4739 is 90% of what you would get with D2896.

Not sure that I'd agree with the "more accurate" part of their statement. They do include oxidation test as well, which Blackstone does not.
 
It's not just Blackstone saying this. From astm.org's page on D4739:

Yes, that's what they say. However, I still don't agree with it. I would rather look at D2896 with oxidation data. D4739 uses a weaker acid and less polar solvent system. This only catches hard base and allows more room for error, as evidenced by the nearly full number difference in the results above.


RDY4WAR, thanks for your insights.

On the report from cat it has data for oxidation. States 14 UFM Oxidation. No idea what a "good" number is for oxidation.

It depends on the oxidation of the oil when it started. Oils that aren't loaded with ester will typically start around 6-12. They're usually done once they exceed ~30. Oils with a lot of ester like Amsoil SS, HPL PCMO and HDMO, and Red Line HP series can show 40-110 in virgin form.
 
JWC86, OK, so now you have two Blackstone Labs TBN readings, 2.6 and 3.5, both of which are a long way from CAT's 6.2 reading. As a chemist who has run many analyses, I would have to recommend that you ask CAT to rerun the TBN before you make any conclusions as to which lab is "correct" with respect to the TBN.

I do agree with Blackstone's observation that a TBN of 6.2 is more in line with the TBN of a virgin oil.
 
Just want to add that I did do a bit of searching to see what TBN method the CAT S.O.S service uses, and, assuming CAT uses the same service across the board for their S.O.S program, it would indeed appear to confirm what @RDY4WAR suspected and that is that they use the D2896 method:
Screen Shot 2022-07-17 at 12.30.10 PM.jpg
 
I've sent identical samples to BS and Polaris labs several time over the yrs and there has never been much difference in the #s. But their comments differ greatly. Polaris is much conservative.
 
Not to kick a dying horse, but regardless of what RDY4WAR stated, and yes, I agree that different titration methods can give different results, but, in this case we are talking about several orders of magnitude of difference between the two testing laboratories.

Going from a pH of ~3.0 (average of the 2 Blackstone values) to a pH of 6.2 is not like going from (and I am making a loose analogy here) 30 mph to 62 mph, it's more along the line of going from 30 mph to 6,000,000 mph.

Just as an FYI, pH is expressed as the negative log(10) of the actual concentration.

Hence, there is a LARGE difference between the results for TBN.
 
Back
Top