Resolved! EFI + MAC-14 maxing ataround 5k RPM

[Ben Swann]

OK I'm certain I've figured out what was going on.  First here is what I did to narrow
things down.  As of yesterday AM I was not sure even if it was an ignition dropout or
problem with fuel injection.  I went through the ignition and rechecked things -
basically all wiring is at or above factory spec, and there is virtually direct battery
connection on coil terminal 15.  I checked to make sure tachometer from UrQ was not
putting any parasitic load on the coil, since this was a difference.  The MAC-14 uses a
dedicated output from the ECU to the model 200 tach.  But there was no difference, and
engine still broke up at around 5000 RPM.

I decided it was time for compression check and to look at the plugs.  I ran the engine
to slightly warm and revved to duplicate the problem, then shut off.  I pulled plugs one
by one as I did a compression check.  Well all the plugs were fouled rich - not gassy
but fairly thick carbon.  I was not surprised and sort-of expected this.  Since they
were not wet, that supported that it was not ignition dropout.  Compression check was
138 +/- 3 PSI on all cyl.  So cleaned plugs and back to diagnosis.  

I studied the latest log and noticed that in a very short interval, the Pulsewidth would
spike just as the RPM was going from about 5K to over 5500 RPM.  The pulsewidth would be
about 7ms and jump quckly to over 14 ms.  It seems that the injectors would go from just
about right to putting way too much fuel in and drowning out the combustion.  Due to all
the cautions I'd seen and heard about tuning too lean,  I had been reluctant to lean
things out, but erred too much on the side of caution.  I had been leaning the table out
awhile back, but it was so out of range, I was not able to notice if I was helping or
hurting how the engine ran. 

I flattened out the VE table map significantly and as I did, the engine was able to
attain and maintain a higher RPM.  I was getting up to around 7000 RPM even.  I leaned
the map out some for the load range, but it is probably still too rich.  I went for  JNR
run.  Jesus!  I know it is not right yet, but basically I can't tune it much better
without an assistant and subsequently a dyno.  Even going up a long steep hill the car
simply gets going too fast to monitor events as they are happening.  It was raining and
It scared the S#|+ out of me! 

So the upshot of this is, the VE table setting in the over 5000 RPM range were way too
high and the engine became too rich to fast.  The symptoms made me look for an ignition
problem, but now I see what was going on.

I probably have other settings incorrect - For one I set up the base injector settings
to match a 42 lb injector and later found they were 37 lb.  There are more things to
adjust and I don't even have the modified MAC-14 ECU in there, so only hitting around 2
bar - if that.  Even now, this car is scary fast, and I'm able to drive it around.  I'll
be installing a Wide Band O2 sensor, and hope to get my G-Tech back from the fellow with
the slipping clutch I left in his car now in Canada.

Thanks for all the feedback and suggestions - even when I check things that aren't
related to the problem, I end up making them better in the process.

Ben


From: Ben Swann [mailto:benswann at verizon.net] 
Sent: Tuesday, October 23, 2007 9:42 AM
To: urq at audifans.com; quattro at audifans.com
Cc: 'Ben Swann'
Subject: EFI + MAC-14 maxing ataround 5k RPM
 [SNIP]