Let's talk about Flashpoint (ASTM D92) and volatility

[Hohn]

Let's start with some definitions. What is a "flashpoint"? From here.

Since we're confining our discussion to engine lubricants, there is fairly strong correlation of flashpoint to volatility as the oil's composition is predominantly hydrocarbon. I think we often consider ASTM D92 flashpoint as a proxy for volatility.

Some notably high flashpoint "Walmart aisle" oils include:

• Mobil 1 Truck and SUV 5w30: 261°C (!)
• Mobil 1 Extended Performance 10w30: 251° C
• Mobil 1 Extended Performance High Mileage 5w30: 249°C

Yet when we turn to oils that are very highly regarded (backed by UOA), we see many have considerably lower flashpoints (some even with higher KV100s:

• Mobil 1 0w40 FS: 230°C
• HPL 5w40 Supercar/Euro: 221°C
• Mobil 1 0w40 Supercar: 232°C
• Pennzoil Ultra Platium 0w40: 206°C
• HPL Premium Plus PCMO 0w40: 212°C

I was curious what some synthetic and Grp III base oils might have for flashpoints. Here's a couple:

• Synnestic 12 (AN): 258°C (published NOACK: 4.5%)
• Synnestic 5 (AN): 222°C (published NOACK: 12.7%)
• Esterex NP451 (Polyol Ester): 255°C (published NOACK: 4.6%)
• Esterex A51 (Adipate Ester): 247°C (published NOACK: 7.4%)
• SpectraSyn 4 (PAO): 220°C (published NOACK <14%)
• SpectraSyn Plus 6 (PAO): 246°C (published NOACK: <6%)
• SpectraSyn 8 (PAO): 260°C (published NOACK: 4.1%)
• Nexbase 3080 (Grp III): 260°C (published NOACK: 3.5%)
• Nexbase 3043 (Grp III): 228°C (published NOACK: 14.2%)
• Yubase 4 Plus (Grp III): 230°C (published NOACK: 13%)

Staring at all the information, I can't find any real conclusive use for Flashpoint as a measure of oil performance. It seems that by itself it tells you next to nothing. Without actually plugging in the data for many different base oils (which I haven't now the patience to do), there's a correlation between KV100 and Flashpoint and NOACK-- higher KV100s means higher flashpoints and lower NOACK.

But the correlation is weaker than we might think. Synnestic 12 is a high KV100 base (~12, duh) and has impressive flashpoint and NOACK. But NEXBASE 3080 is nearly four points lower KV100 while having even lower NOACK and higher flashpoint. And it's ~8cSt neighbor in Grp IV (Spectrasyn 8) is similarly high in flashpoint while having worse NOACK.

What do I conclude from this initial survey? Nothing really, it's not very conclusive. But a couple points might be suggested:

• Published flashpoint might indicate relative volatility, but that doesn't translate into cleanliness per se. If it's a Grp 3 and has a low flashpoint (i.e. Pennzoil Platinum 0w20 at 204°C) it might have a tendency to varnish. But a Grp IV oil just a couple degrees higher in Flash point might run superbly clean. Even if it has relatively poor NOACK it can still maintain a superbly clean engine.
• What matters more than the rate at which vapors are generated (volatility) is what those vapors are (chemically). If those vapors are light aromatics from a cheap Grp 1 or 2 oil and they initiate or spready polymerization deposits, that's very bad. If those vapors are PAO or AN (or even some paraffinic Grp 3), it might not lead to deposits at all. If those vapors are something else, they might actually lead to cleaning surfaces that aren't even oil-wetted (Valvoline Restore and Protect).

It might be that dirty running oils that tend to leave deposits have poor flashpoints. But that doesn't mean that higher flashpoint oils run cleaner or that lower flashpoint is the REASON that some oils leave more deposits behind.

I guess it's progress to go arrive at "I don't know" from "I think I know what isn't so."

 

[buster]

It seems the issue with flashpoint is repeatability.

Mobil 1 Truck and SUV 5w30 has a higher flashpoint then most of their other oils.

D

Thread 'Flash point, base oil quality, performance'

Jul 8, 2018

I always thought there was generally a strong correlation between an engine oil's flash point and the quality of its base stocks, such that an oil with a higher flash point could generally (though not always) be expected to be more resilient at high temps: less shear, fewer deposits, etc. I fully expected there to be exceptions, but I thought that was more-or-less the general trend.

Now I feel like I might have been wrong about that. On the one hand, Castrol Edge Supercar HP -- their Indy Car oil -- has a flash point of 220º C; Mobil 1 FS 0W-40 is at 226º C; and most (all?) of the...

• d00df00d
• Replies: 35
• Forum: Passenger Car Motor Oil (PCMO) - Gasoline Vehicles

L

Post in thread 'Flash Point and NOACK'

Feb 24, 2018

Originally Posted By: Triple_Se7en
We haven't had a NOACK Check thread in a while here. I wonder what conventionals and which synthetics have the best NOACK numbers these days also.


I wonder the same thing.

Pennzoil Ultra Platinum GTL oils used to be really low NOACK, now they are up around where everybody else is.

• Linctex

 

[BMWTurboDzl]

D92 (Cleveland Open Cup) has poor repeatability so it's a rather useless measure.

 

[Brian123]

D92 (Cleveland Open Cup) has poor repeatability so it's a rather useless measure.

So, is the Closed Cup Test Better? I am guessing the Closed Test is D93?

 

[wemay]

So, is the Closed Cup Test Better? I am guessing the Closed Test is D93?

ASTM D93​

ASTM D93 is generally considered a better test than D92 for volatility because it uses a closed cup method that prevents vapors from escaping, resulting in a flash point that is typically a few degrees lower than that obtained from the open cup method (D92)2. This makes D93 more suitable for petroleum products with lower flash points3. Additionally, D93 is preferred for testing lower flash points, while D92 is better for high-viscosity oils3.

S

Thread 'Flashpoint Testing Method ASTM D-92 vs. ASTM D-93'

Nov 3, 2006

Does anyone know what the difference is in the test methods between the two?

I recently checked the Dino Pennzoil 5w30 flashpoint as 216 deg. F (ASTM D-93) and Pennzoil Platinum 5w30 flashpoint as 196 deg. F (ASTM D-92).

At first I thought, why would the Dino Pennzoil have a that much of a higher flashpoint than their "best" Platinum line. Then I realized the test methods were different. So it led me to the question of what's the difference in the two methods?

• silverbullet
• Replies: 3
• Forum: Passenger Car Motor Oil (PCMO) - Gasoline Vehicles

 

[twX]

If it's a Grp 3 and has a low flashpoint (i.e. Pennzoil Platinum 0w20 at 204°C) it might have a tendency to varnish. But a Grp IV oil just a couple degrees higher in Flash point might run superbly clean. Even if it has relatively poor NOACK it can still maintain a superbly clean engine.

That Pennzoil Platinum 0w20 flashpoint is based on D93, not D92. The D92 flashpoint would be at least 20°C higher. For 4 cST base stocks, PAO has a much lower flashpoint than GTL or Group III, around 200°C for PAO vs 230°C for GTL and 240°C for Group III. In the thicker grades, they're more similar.

The Pennzoil Ultra Platinum 0W-40 flashpoint is also based on D93.

I don't think there's any causal relationship between deposits and volatility. Oils that use thinner base stocks do tend to produce more piston deposits, so there is a correlation, but I believe that this is mostly due to the fact oils with thinner base stocks tend to use more viscosity modifier.

 

[kschachn]

I no longer have access to all the ASTM tests, but back in the day when I ran flashpoint tests both the closed and the open-cup tests had poor repeatability. It’s just a nature of the test. These days with automated testers it’s likely better but considering the real propose for the test, it is quite close enough.

Open cup and closed cup measure different properties. It’s not as if one is better, worse or more representative of some translatable “quality” indicator.

 

[Hohn]

It sure seems like D93 would be better and more repeatable. It's just that almost nobody is using it and publishing it, so we only have what we have to go on.

 

[kschachn]

It sure seems like D93 would be better and more repeatable. It's just that almost nobody is using it and publishing it, so we only have what we have to go on.

It’s not better really. It’s a different test for a different property. Both values are used.

 

[RDY4WAR]

So long as the flash point is >205°C (>400°F), I don't give it much thought. The only time I give it much attention is with racing ATF since the transbrake increases fluid temp at a rate of +30-50°F per second.

Flash point can correlate with volatility but not always. HPL PCMO 10W-20 has a FP of 230°C, lower than the M1 offerings first mentioned, but the Noack is just 3.8%.

 

[buster]

Weird claim by boutique oil company Driven. Driven also said they were first to the market with an LSPI oil which is technically 4 years behind Mobil 1.

Mobil 1 Truck and SUV 5w30 has a very high flashpoint of 261. Possibly from AN?

https://www.drivenracingoil.com/blo...DkmM0OmcQZoFqk5fywOmL0egSHsS_sS574jlH1b8yF_GQ


"Another reason why flash point matters for high-performance oils is evaporation losses. Oils with high flash points are unlikely to vaporize in hot temperatures. Excessive evaporation can leave engine components under-lubricated, resulting in increased friction, heat generation, and potential engine damage."

"Oils with a high flash point resist oxidation more effectively. "

 

[Tom NJ]

If the flashpoint results are accurate there should be a rough correlation between flash point and volatility within a chemical family.

Several reasons why the correlation is lacking:

- Motor oils are mixtures containing different families of chemistry. Polar components for example have a higher flash point for their molecular weight and/or viscosity due to intermolecular attractions, e.g., 4 cSt PAO = 425°F, 4 cSt Diester = 440°F, and 4 cSt POE = 480°F.

- The flash point test has poor reproducibility, compounded by batch-to-batch variations.

- Highly volatile ingredients that affect Noack may not show in the flash point since the flame is not passed until 50°F before the expected flash point, by which time the most volatile components may have partially boiled off.

- The flash point reported on Technical Data Sheets may not be representative of the actual product. Some companies correctly report the average of actual plant batches, some report their higher/est results, some report their lower/est results - depends on who is preparing the data sheet and their intent. Some TDS preparers are non-technical and report whatever they are told by whomever they ask, some are HS&E folks who prefer conservative results, and some are marketing folks who prefer a more shall-we-say liberal interpretation for comparison to competitors. I have seen substantial differences in flash points on TDSs among manufacturers of the very same pure ester. We just don't know who came up with the number, or how, or why.

Personally I disregard reported flash points. They can be useful as a quality control test and if accurate can give some clues on an oil's composition, but don't really tell you anything meaningful about a motor oil's performance.

 

[buster]

Incredibly misleading blog post by Driven. I know they make some fantastic oils but the claims I've seen them make are not good. Looking at flash point to help choose an oil is incredibly stupid. They also talk about their oils having low Noack when their 5w30 DI oil is 11.5%. Unreal....

 

[ltslimjim]

Incredibly misleading blog post by Driven. I know they make some fantastic oils but the claims I've seen them make are not good. Looking at flash point to help choose an oil is incredibly stupid. They also talk about their oils having low Noack when their 5w30 DI oil is 11.5%. Unreal....

Classic marketing driven blog. “Trust what we say, but don’t look at what we actually do when compared to what was said.”

 

[ltslimjim]

If the flashpoint results are accurate there should be a rough correlation between flash point and volatility within a chemical family.

Several reasons why the correlation is lacking:

- Motor oils are mixtures containing different families of chemistry. Polar components for example have a higher flash point for their molecular weight and/or viscosity due to intermolecular attractions, e.g., 4 cSt PAO = 425°F, 4 cSt Diester = 440°F, and 4 cSt POE = 480°F.

- The flash point test has poor reproducibility, compounded by batch-to-batch variations.

- Highly volatile ingredients that affect Noack may not show in the flash point since the flame is not passed until 50°F before the expected flash point, by which time the most volatile components may have partially boiled off.

- The flash point reported on Technical Data Sheets may not be representative of the actual product. Some companies correctly report the average of actual plant batches, some report their higher/est results, some report their lower/est results - depends on who is preparing the data sheet and their intent. Some TDS preparers are non-technical and report whatever they are told by whomever they ask, some are HS&E folks who prefer conservative results, and some are marketing folks who prefer a more shall-we-say liberal interpretation for comparison to competitors. I have seen substantial differences in flash points on TDSs among manufacturers of the very same pure ester. We just don't know who came up with the number, or how, or why.

Personally I disregard reported flash points. They can be useful as a quality control test and if accurate can give some clues on an oil's composition, but don't really tell you anything meaningful about a motor oil's performance.

Very well explained and reasoned on. It’s more important the oils overall performance than trying to grab one data point, put it in a vacuum, and say oil x is better than oil y *if* only comparing the one value. Well, characteristics for some more favorable ingredients may look worse on a single test but in practice in the real world, the fully formulated product is very well rounded. So, chasing a concern that is not actually well correlated to actual performance is sort of the catch. Related to this, isn’t there a newer testing standard being developed to replace Noack, for example? Iirc, the test is a lower temp for longer duration to simulate more accurately the actual engine conditions rather than when subjected to the higher temp for shorter duration…

 

[Tom NJ]

Related to this, isn’t there a newer testing standard being developed to replace Noack, for example? Iirc, the test is a lower temp for longer duration to simulate more accurately the actual engine conditions rather than when subjected to the higher temp for shorter duration…

Yes, but it has not hit the API specs yet.