JIM HOLSONBACK wrote in a message to ALL:

 JH> Hello, ALL.

 JH> Our 'family' car here is this older Buick sedan - a '98 LeSabre.
 JH> Engine is the GM 3.8L V6. (I've had several cars with this block
 JH> over the years, with generally good service).  IMO it gets decent
 JH> gas mileage on the recommended 87 octane. I bought gas here last
 JH> week at Costco, and it was 2.03 for 87 and 2.13 for the 91 octane.
 JH> (A bit less than 5% price premium). This is unlike a lot of the
 JH> commercial gas sellers, where the high-test is typically $0.20 or
 JH> so more than the regular.

 JH> I don't take many long trips, mostly local but not much stop and go
 JH> in heavy traffic. I was thinking if I'm averaging 20mpg now, and
 JH> could get 21.0 or better with the higher octane with my type
 JH> driving of it, I'd be doing OK.

 JH> I'm surprised I haven't seen any articles about this situation. My
 JH> question- might I be able to get more than 5% mileage increase if I
 JH> start running the high-test?

I used to run "middle grade" gas in my car,  I forget which one
by now,  but then I also had the engine tuned for it -- I'd tweaked the
distributor timing just short of pinging.  Dunno if you can do that sort of
thing with something as new as what you're talking about there,  though, 
without changing a chip.

There used to be a mailing list for that sort of thing,  I have no idea if
it's still going or not,  it was called DIY-EFI or something pretty close
to that.

There were some really knowledgable folks in there back then.

Getting the info on this stuff is the really hard part.  There were
apparently some "SAE White Papers" that covered a lot,  but you
had to _buy_ them and they weren't cheap.

---