<<I expect this may start another lengthy debate (not my intention), and I
realize that this is just one guy's opinion, not the result of a controlled
study.  However the opinion comes from a guy with a lot of experience.  Also
for what it's worth, my local A&P prefers straight weight.

Tom Gourley>>

Okay, since you started it...My information comes from a long time ago and
was relative to automotive engine design, but I think it still holds true.

There are a number of attributes that can be used to compare oils, two of
which are viscosity and lubricity.  Viscosity is measured by essentially
dripping oil through an orifice by gravity, or some other equivalent way.
The point is it is measured when the oil is under no "stress".  So-called
"straight-weight" oils are measured at one temperature, hot.  Viscosity
index is a measure of how well the oil keeps that viscosity as it gets
colder and none keep it very well.  At 0 to -20 the viscosity may be 1,000
times what it is at 200F.  Viscosity index improvers are additives that
react with the base oil molecules to create bonds between molecules at high
temperature, albeit weak ones.  Therefore, with this additive it is possible
to take a SAE 20 base stock and make it test like a SAE 50 oil hot.  To
accommodate this effect, SAE created a cold test spec called "Winter", hence
the "W" after the first number of 20W-50 oil.  I behaves like a normal 20
oil cold and an SAE 50 oil hot.  That doesn't mean that the viscosity is
lower cold than hot, but it just doesn't get as high as it would have if it
were a 50 oil.  Incidentally, I don't know where the "Weight" nomenclature
came from, but I assume that the vernacular evolved to call more viscous
oils "heavy" and less viscous oils "light."  I don't know where the "W"
comes from in the Shell label, but I think it means "Detergent."

The rub (literally) is that the bonds created by the additive are weak and
when the oil is sheared in the bearings at high speed the bonds tend to
break, leaving part of the additive with one molecule and part with the
other.  Therefore, a multi-viscosity oil will gradually revert to the
viscosity of the base stock; in our case SAE 20 and therefore it will have a
finite life.  Also, the lab measurement doesn't correlate exactly to the
actual engine condition and as a consequence the engine will be running with
oil that is closer to an SAE 20 than a 50.  Viscosity is what holds the main
and rod bearings apart while running and the higher the viscosity the
greater the clearance and therefore more room for dirt to go through the
bearings.

Another characteristic is lubricity, measured by pressing a rotating steel
ball against three others and gradually increasing the pressure until the
lubricating film is breached and galling occurs.  Pure oil is very good in
this test, the higher viscosity the better.  Adding any additive to the oil
simply reduces the amount of oil in the mixture, making the performance in
this test poorer.  They fix this by adding an Extreme Pressure (EP) additive
such as Molybdenum disulphide ("Moly").  I think it is these additives which
can interfere with the break-in of the engine.  Lubricity is what protects
the engine parts when the metals contact, such as between the cam and lifter
and rings when they are at the top of the cylinder.

Bottom line?  Viscosity determines how easily the engine will crank cold and
how much viscous drag there is when running.  Lower viscosity means less
power goes into the oil, leaving more power at the crankshaft.  The 20W-50
semi-synthetic oils have been shown to give maybe 1% or more power and fuel
economy.  At high rpm a low viscosity oil probably provides enough bearing
clearance anyway.  Synthetic oil has an inherently high viscosity index, so
by mixing some with a base stock a 20W-50 can be made with no, or at least
less additive.  Therefore, this oil doesn't degrade to a different viscosity
with use and can be run longer between oil changes, neglecting the effects
of contamination.  Unfortunately, this oil has less lubricity, so it has to
have an additive to improve that attribute.

So why use a "pure" oil ("straight weight" is an imprecise term)?  Higher
cold viscosity will keep the oil on the camshaft longer between running, and
that's about the only reason I can think of.  Not related, engines can "look
better" if they have been run with a detergent oil, keeping deposits off the
metal.

Gary Casey