We pretend they don’t exist.
They don’t even do gas chromatography. They try to estimate it off of flash point.
HPL SAE 40 PCMO
- Thread starter Hohn
- Start date
They don’t even do gas chromatography. They try to estimate it off of flash point.
Patman
Staff member
Their refusal to use the proper method for determining fuel % still baffles me. They must be the most profitable oil analysis company in the world but yet they don’t spend the money needed to have state of the art equipment? Don’t they care about accuracy? They must be reading BITOG and hearing our complaints for all of these years. Give customers what they want
It’s somewhat less than a refusal. There is an ASTM method for using flash point to estimate fuel, but the method notes that it is a screening test. It likely has a low accuracy. Nor do we know if Blackstone is even using that particular ASTM test or performing it properly. It would be awesome if these analysis companies would publish the stated accuracy numbers with each test.
Gas chromatographs are expensive, as are properly trained technicians who properly calibrate and operate the machines. Sometimes I wonder if all these analysis companies actually run the ASTM tests exactly as published, end-to-end. As someone who ran many standardized tests in the past that takes a lot of time and careful care. If you’re being paid for each analysis and wish to have competitive pricing it’s a lot of time, materials and labor to do that.
Looks like they use ASTM D92 for the flashpoint and infer fuel dilution from that.
https://www.blackstone-labs.com/standard-oil-testing/
https://www.blackstone-labs.com/about-oil-analysis/
Well, I manage to get some time on the phone with an analyst at Polaris. First, let me say how refreshing it is that, in just a couple minutes, you can get a human being on the phone who is VERY knowledgeable about analysis. Instantly and obviously knew PDSC, GC, and all the relevant technical aspects of analysis. I assume that the person I spoke with is someone who actually runs samples or has in the past.
The analyst told me they did a partial re-run and 2/3 oxidation re-runs confirmed the elevated oxidation, likewise for the fuel dilution. I asked them to re-run KV100 just to be sure because it seems incompatible with the other data points.
But now I'm convinced that I sent them a contaminated sample because I had cleaned the sample pump tube by blowing it out with non-chlorinated brake cleaner and then compressed air. Reading that CRC can, you can see it has Acetone, N-heptane and all kinds of other hydrocarbons that would 100% show up as "fuel" and as oxidation products. The procedure of blowing out the tubing with compressed air probably not only failed to remove heavier solvents, but likely added traces of water and who knows what else.
I think it's the only plausible explanation as to how I ended up with nearly 4% dilution with no proportional change in KV100. It also explains the high oxidation values and nitration values that are simply not realistic.
Thoughts?
I'm going to take another sample here at 2k and use some virgin tubing. We'll see what that sample shows. I'm guessing it shows fuel at closer to 2.5%, oxidation back in the 20s, and nitration barely moved at all.
The analyst told me they did a partial re-run and 2/3 oxidation re-runs confirmed the elevated oxidation, likewise for the fuel dilution. I asked them to re-run KV100 just to be sure because it seems incompatible with the other data points.
But now I'm convinced that I sent them a contaminated sample because I had cleaned the sample pump tube by blowing it out with non-chlorinated brake cleaner and then compressed air. Reading that CRC can, you can see it has Acetone, N-heptane and all kinds of other hydrocarbons that would 100% show up as "fuel" and as oxidation products. The procedure of blowing out the tubing with compressed air probably not only failed to remove heavier solvents, but likely added traces of water and who knows what else.
I think it's the only plausible explanation as to how I ended up with nearly 4% dilution with no proportional change in KV100. It also explains the high oxidation values and nitration values that are simply not realistic.
Thoughts?
I'm going to take another sample here at 2k and use some virgin tubing. We'll see what that sample shows. I'm guessing it shows fuel at closer to 2.5%, oxidation back in the 20s, and nitration barely moved at all.
Last edited:
Those compounds in the brake cleaner would not show up as the same peaks on a GC analysis. And even if they did, to get to 4% in your sample would mean a hecka lot in there.
Well, Polaris did a partial re-run and the dilution is now showing 3%.
I'm thinking this was a contaminated sample and that the 2000 mile sample with virgin tubing will show considerably different oxidation and dilution values.
I'm thinking this was a contaminated sample and that the 2000 mile sample with virgin tubing will show considerably different oxidation and dilution values.
Well, we have a useful update on the SAE 40 monograde experiement. As of today, we're about 2600 miles into this OCI and I just got back my 2000 mile UOA.
First, the 2000 mile UOA in context:
Interesting dilution reduction. I attribute this to the much warmer weather we've had since the April sample was pulled. Yawn-inducing nothingburgers on Iron, Aluminum, Oxidation, Nitration, and even KV100. We're still on the 1ppm/1k iron trend if we can even see it within sample noise. Likewise for Aluminum-- the delta is too low to reliably measure.
The additive package is showing how brute tough it is and yawning at my duty cycle.
The dilution data spiking so much early strongly suggests that even at 10k OCI we're still likely in to be < 4.5% dilution. The collected data set above, when analyzed, produces a Tau value (time constant) of roughly 2000 miles.
Which means that even this small dataset strongly suggests my dilution trend will follow what many people have observed here: the dilution levels off pretty rapidly and the difference between 5k and 10k miles is very small.
The statistical model here suggests a 5k OCI dilution value of 4% and a 10k value of 4.3%.
The only part of the data that I'm still puzzled by is the oxidation value. After 4400 miles on the HPL 15w-40, the oxidation was nearly unmoved from virgin with a value of 25.
Yet within 1k miles on the monograde, oxidation has spiked from a baseline of 22 to 39. I had surmised that this value may have representing potential contamination from solvent cleaning my sample tube. So I was careful to use brand new virgin tube to collect the 2k sample. And oxidation is only +1 higher than 1k, but it's still higher and still +18 in only 2k miles.
I have no theory that aligns with this oxidation trend. I think that that the slow progression from 1k suggest there was perhaps some kind of one time oxidation event that the oil had to manage early on but it hasn't progressed. But what? Like I said, the oxidation is quite puzzling.
Anyway, the 2k sample suggest I'm probably good to go to 10k on this HPL oil. A 5k sample will seal the deal as to whether we push to 10k or dump at 5k.
As an aside, the tank MPG for 2600+ miles on the monograde is 30.1. It has held a solid 1mpg advantage over both the Restore and Protect 5w-30 and the HPL 15w-40. Despite the higher viscosity, this is indeed a remarkably low-traction oil as I had hypothesized many pages ago in this thread.
First, the 2000 mile UOA in context:
Interesting dilution reduction. I attribute this to the much warmer weather we've had since the April sample was pulled. Yawn-inducing nothingburgers on Iron, Aluminum, Oxidation, Nitration, and even KV100. We're still on the 1ppm/1k iron trend if we can even see it within sample noise. Likewise for Aluminum-- the delta is too low to reliably measure.
The additive package is showing how brute tough it is and yawning at my duty cycle.
The dilution data spiking so much early strongly suggests that even at 10k OCI we're still likely in to be < 4.5% dilution. The collected data set above, when analyzed, produces a Tau value (time constant) of roughly 2000 miles.
Which means that even this small dataset strongly suggests my dilution trend will follow what many people have observed here: the dilution levels off pretty rapidly and the difference between 5k and 10k miles is very small.
The statistical model here suggests a 5k OCI dilution value of 4% and a 10k value of 4.3%.
The only part of the data that I'm still puzzled by is the oxidation value. After 4400 miles on the HPL 15w-40, the oxidation was nearly unmoved from virgin with a value of 25.
Yet within 1k miles on the monograde, oxidation has spiked from a baseline of 22 to 39. I had surmised that this value may have representing potential contamination from solvent cleaning my sample tube. So I was careful to use brand new virgin tube to collect the 2k sample. And oxidation is only +1 higher than 1k, but it's still higher and still +18 in only 2k miles.
I have no theory that aligns with this oxidation trend. I think that that the slow progression from 1k suggest there was perhaps some kind of one time oxidation event that the oil had to manage early on but it hasn't progressed. But what? Like I said, the oxidation is quite puzzling.
Anyway, the 2k sample suggest I'm probably good to go to 10k on this HPL oil. A 5k sample will seal the deal as to whether we push to 10k or dump at 5k.
As an aside, the tank MPG for 2600+ miles on the monograde is 30.1. It has held a solid 1mpg advantage over both the Restore and Protect 5w-30 and the HPL 15w-40. Despite the higher viscosity, this is indeed a remarkably low-traction oil as I had hypothesized many pages ago in this thread.
Ok, I think we have an explanation for why the 1k and 2k mile samples might be a bit screwed up in oxidation. Or rather, two potential explanations.
The first is that the baseline sample for virgin oil that I sent to Polaris was not used as a product baseline because they flagged the trace aluminum result. So perhaps that’s relevant to later samples? I’m not sure why it wouldn’t affect the 8 miles sample also, then…
The second, more likely explanation: higher ester concentrations can apparently have stronger initial reactions when fresh. It doesn’t just raise the oxidation baseline, but also the magnitude of any early spectral drift with the first few heat cycles.
Which is why the relevant oxidation fact is not the hangup 1k mile point with teh massive jump to 39. Rather, the relevant point is that it’s only +1 in the next 1000 miles.
@High Performance Lubricants seeing has now there’s not a lot of content here on the monogrades here, could you perhaps comment on what is reasonable for oxidation trajectory on the PCEO monogrades? Is this just the result of slightly different kinds or amounts of ester? I really appreciate any insight you can provide.
Everything I can see says this oil is absolutely going to make it to 10k with lots of breathing room even with a good bit of dilution. But I want to be sure I’m not ignoring an oxidation signal hiding in plain sight.
The first is that the baseline sample for virgin oil that I sent to Polaris was not used as a product baseline because they flagged the trace aluminum result. So perhaps that’s relevant to later samples? I’m not sure why it wouldn’t affect the 8 miles sample also, then…
The second, more likely explanation: higher ester concentrations can apparently have stronger initial reactions when fresh. It doesn’t just raise the oxidation baseline, but also the magnitude of any early spectral drift with the first few heat cycles.
Which is why the relevant oxidation fact is not the hangup 1k mile point with teh massive jump to 39. Rather, the relevant point is that it’s only +1 in the next 1000 miles.
@High Performance Lubricants seeing has now there’s not a lot of content here on the monogrades here, could you perhaps comment on what is reasonable for oxidation trajectory on the PCEO monogrades? Is this just the result of slightly different kinds or amounts of ester? I really appreciate any insight you can provide.
Everything I can see says this oil is absolutely going to make it to 10k with lots of breathing room even with a good bit of dilution. But I want to be sure I’m not ignoring an oxidation signal hiding in plain sight.