Motronic full throttle AFRs

[Sagaris]

In other threads I have seen it mentioned that some of the Spec E30 racers reset their ECU to erase long-term fuel trim for more favorable AFRs. I have some things in place in preparation for megasquirt including a wideband AFR gauge.

Part throttle AFRs bounce between 14.3 and 15.0. I am getting a full throttle enrichment that brings the AFR down to 13.6 (WOT around 3500-4000 rpm) but the AFR climbs up to 15.0 as it drives towards redline.

I remember that it would hold 13.6-13.8 last fall after an ECU reset and am wondering if this is what the SpecE30 guys are talking about. Does anyone have any full throttle AFRs to compare with? I will be checking the fuel pressure to see if that is a factor as well.

M20B25 engine, M50 injectors, MarkD tune

[jlevie]

My Spec E30 is set up so that the DME is completely powered off each time the ignition is switched off to clear any learned adaptations. I've tweaked the AFM clock spring tension to get an AFR of 13.5-13.8 up to 4500rpm (when the DME quits using AFM data). Above 4500rpm the DME is only using IAT, ECT, and RPM data and there is no control over that in a Spec E30 since we can't use an adjustable FPR or a chip, so I'm stuck with selecting a DME that has the best AFR above 4500rpm. Even though the DME ignores O2 sensor data at WOT, there is enough idle and part throttle time in a race environment for the DME to learn to pull fuel in an attempt to get back to stoic operation. I don't want that to happen, so I reset the DME on every shutdown.

What I have seen on the dyno supports the theory that learned adaptation applies across all modes of operation. So if you twiddle with sensors or the like to get a more favorable AFM, the DME will undo your work in time. That should not be the case for chip tuning since it changes fuel maps and target AFR's.

Since you don't have an M20 and a Motronic 1.3 DME, this may not be fully applicable. You can certainly find out if learned adaptions are at play. Drive the car for 1000mi or so, take it to the dyno and get 2-3 baseline runs. Then disconnect the DME for 30sec and make another pull.

[kamotors]

^ dang, that's some good info. I remember i made a thread a few years ago saying that my car is much faster when I reset the battery before doing a 0-60 test and nobody knew why; its a good 0.2 seconds quicker with a reset ECU than it was after it learned and got sluggish and I thought it was just because of my mark d chip or something.