Auto learn code

[Scot Sealander]

Bruce Plecan wrote:
 
> What I'd like is a sensor mounted in the air stream that is fast 
> reacting, and isolated from the mounting.  There are several different > versions, some of which have a more "open" sensor.

I am using the open version of GM's air temp sensor.  You can see the
thermister inside a plastic cage. I am curious about it's time delay.
Don't know if it's a problem.  If a little slow to heat up, it will 
just make it a little too rich.  I think that is better than a little 
too lean...

You can see the sensor self heat.  One morning at power on, the ECT and 
MAT read exactly the same temp.  After some period, the MAT was drifting
up and was 2 degrees warmer.  It was not a long time, but can't give it
a real accurate time estimate on how long it took.


> This moving sensors to improve performance is just to find a cooler
> place, to enrichen the mixture.

I don't want to do that.  The goal is a more accurate method of
determining the mass of the air in the cylinder.


> I'd perfer wiith the perfect sensor to have it right down by an intake
> valve.  To me knowing the air temp as it actually enters the chamber 
> is where you need to monitor it.   Then work out the IAT
> correction/mixture strenghts.

Sure, it would be nice to know that temp exactly, but I'll accept a real 
good estimation.  It would be an enormous improvement over the present
situation.  


> Maybe a 3/8 NPT to 1/4NPT  adapter of phenolic or something would
> help...

Sounds good to me.  Have to give this some thought on just where and 
how to mount the sensor.  It seems that measuring the intercooler
output temps would be the most accurate value to use during boost
conditions.  During cruise, that temp plus a factor for manifold
heating of the air ought to come close.  I am open to a better 
suggestion!
 
Thanks, Scot Sealander