Oil "volatility"/"evaporation": The REAL story

Oil does NOT "evaporate" or "boil off" in normal use. Its rate of
evaporation in regular daily use is minimal to non-existent.

Who says so? ExxonMobil for one, and AMSoil for another.

I emailed both companies aking them whether the "evaporation" thing was
true or not. I cited the AMSoil page found here:
<http://www.bestsynthetic.com/volatility.shtml>

ExxonMobil sent an emailed reply. When I then asked for more info, a
tech rep phoned me, and we had a ten-minute conversation.
AMSoil gave me three emailed responses from three individuals (one was a
dealer, the other two were AMSoil employees).

Guess what? Both Exxon and AMSoil say the SAME thing: Oil does NOT
"evaporate" or "boil off" or exhibit any sort of noticeable volumetric
shrinkage due to lighter fractions disappearing over time.


The myth of "evaporation" seems to originate from both the chart on the
Best Synthetic page referenced above, and the basic specs for the
related ASTM D-5800 lab test for motor oils (the test required by the
API SM standard, and which specifies no more than 15% volumetric loss of
engine oil after 1-hr at about 500F.)

The AMSoil reps told me that volumetric loss in D-5800 has no meaning in
the real-world, for the very simple reason that oil in the real-world
never gets much hotter than 180-200F (AMSoil's numbers!).


The ExxonMobil rep had even /more/ information for me. I have condensed
it below:

1) ASTM D-5800 is a lab test ONLY. It does measure evaporation as a
proxy for actual evaporation in a real-world engine, but the numeric
results can in NO way be taken to mean that any given engine will
experience oil evaporation to any specific degree or in any specific
amount.

2) In the context of the test, a result of 15% or less is a trivial
number, translating to a trivial level of evaporation in the real-world.
My source specifically likened the difference between 9% in the test and
15% in the test to the difference between a car getting 18.23 mpg versus
18.20 mpg. My understanding is that, in order to see actual evaporation
that would be detectable by the naked eye, you'd need to be comparing an
oil that scored 9% in the test with an oil that scored, oh, 50%. And
such a 50% oil could never be sold with the API starburst anyway. API
mandates 15% or less /specifically/ because it's below the threshold
where evaporation would affect measurable oil levels in the real world.


3) Engine-oil levels are affected by very many factors, such as: basic
engine design, operating load, operating temperature, and internal
contamination or wear. This makes

it nearly impossible for anybody with a real-world engine to be able to
determine how much of the inevitable decline in oil level is due to
evaporation, and how much is due to

other factors. That is exactly why the API specifies that oils should
meet the D-5800 test as part of its SM classification.

4) ASTM test D-5800 can be conducted in minorly different ways, each of
them possibly resulting in a different number. My source made it very
clear that he had no idea how

AMSoil's lab conducted their tests, so he has no idea how they came up
with the numbers they did. But the point made was that it is possible to
slightly affect the numbers

with small variations in test methodology, so the numbers presented
can't necessarily be taken at face-value.

 

you're asking them loaded questions then dude. as we discussed before,
oil does not get hot enough to boil and i'm sure they confirmed that for
you. but just like water evaporates from a glass if left standing, even
though it's not boiling, components of the oil will evaporate at
operating temperatures. that's just fact.

eh? so why do we bother to test then?

that can't be right.

18.23 / 18.20 gives a delta of 0.16%.

18.20 x 9% = 1.64 mpg difference, or 19,8mpg.

otoh, losing 9% of the 4.5l in the oil pan of your integra is 0.4l.
that's roughly halfway between the two dipstick marks - both non-trivial
and very much detectable "with the naked eye".

you need to sort your numbers out - none of the above computes.

not believing amsoil's numbers is healthy. denying the facts of
physical chemistry is something else.