(Scott) wrote:

On Tue, 31 Jul 2007 08:23:18 -0400, in rec.aviation.homebuilt, "Peter Dohm"
wrote:

You got a little too ambitious with the clipping and the part you left in
was from a previous contributor. I only agreed that the errors in the fuel
computer would probably be consistent--electronics usually are--and then
recommended a proceedure to conduct coast down testing.

Sorry about the piggyback post, but I was replying to HarryK, as the
(edited) attribution shows. I responded via your post because his was
already expired, and I was not ambitious enough to download it again.

I agree with you about the coastdown test. I disagree that a faulty fuel
computer will necessarily be reliable in any useful way. In my GP I have an
example of a computer that not only produces incorrect results, but produces
results that are wildly inconsistent when compared to results from the usual
method of measuring fuel mileage. Perhaps others have more reliable errors,
but you can't know without testing them first.

(To be fair, my GP's computer is probably very accurate based on the data it
gets. The problem is most likely with the analog data sources for fuel flow
rate and/or fuel level.)

I think Peter's point is still valid - while the mileage computers
might not be entirely accurate, the inaccuracies should be relatively
consistent. That is, if you're doing a test where all other variables
are the same, and the computer tells you you're getting an extra 1mpg,
you can't be sure that the delta is precisely 1mpg, but you can bet
that there IS a difference, and that it's probably not TOO far from
that value.

For example, my Jeep Grand Cherokee has a computer, and I drive a lot
on fhe Arizona roads that are absolutely flat and have very little
traffic. I can get the cruise control set, let everything stabilize,
and see a 1mpg difference from clicking the overdrive on and off
(which is about what I'd expect the actual delta to be). Can I say
with any certainty that the difference is very close to 1mpg? Nope.
But can I surmise there IS an advantage to running with the overdrive
engaged, based on the computer? Absolutely.

Mark "now if I could only get it to go up 10mpg..." Hickey