Sterling Doc wrote:


What you describe is how our Motronic 1.3 works too, and that's part of what causes the problem I described. Well, conjectured upon, really.

The O2 sensor plays at partial throttle so the DME can accurately control mixture while the loosened AFM spring has caused a long term fuel trim (LTFT) to be a bit lean. But at WOT the O2 sensor no longer plays and the DME is running on internal maps with only coolant temp and CPS as significant inputs. This creates a scenario where a lean LTFT can go from "bit player" to significant player".

I would never had understood what was going on if I'd not had a year of track events with an F/A meter on the dash and hooked to the Traqmate. It was dicking around with the AFM spring that caused it all. Or at least, so my theory goes.

Interesting. Thanks for the input! Neal & I have logged our A/F ratios into Traqmate for the last year. I'll go back and look at that data.

It seems unplugging the O2 sensor would keep this from being a problem?

Sterling Doc wrote:

Hmm. Well, if you did anything that screwed with the mixture the DME won't learn about it if you also disconnect the O2 sensor and therefore LTFT would be unchanged. Of course that puts you on internal F/A maps which might suit your car or might not.

It would be interesting dyno a car a couple times, disconnect the O2 sensor and do another pair of runs. Unplug DME to reset LTFT, and redo the previous 4 runs. Then move the AFM a few clicks and do 2 more runs. Then connect the O2 sensor back up, go put in a couple hundred track miles, and do another pair of runs.

That would give you a number of interesting baselines, baseline internal maps, affect of AFM monkeying, and affect of DME compensating for AFM monkeying.

But Doc is saying that his car DID change mixture at WOT after AFM adjustments without an oxygen sensor.

We can definitely change the A/F mixture a lot by messing with the AFM - we've seen this with my car, Dan Pina's, Neal Agran's, David Dirks', and it's not subtle. When I was chasing a what turned out to be a bad AFM wiper track earlier this year, I got my car so rich it made my eyes water, and shot flames out the back on over-run. The question is does the O2 sensor screw these adjustments up over the long term. I can look at the Traqmate logs over the course of a several weekends when I get home, to see if it stays the same over time.

It is interesting that Dan's car was running so lean - the O2 sensor wasn't helping there, or at least didn't have enough adjustment to fix his problem, while adjusting the AFM does.

944Racer72 wrote:

LTFT impacts internal maps. THe question is was O2 sensor every working after fooling with AFM? As long as O2 sensor remained disconnected, loosening AFM spring would allow more air. But that would also lean out mixture which isn't necessarily ideal.

You could loosen AFM spring and also put in a higher pressure fuel pressure regulator. Just gotta keep O2 sensor removed so DME doesn't adapt. That's pretty damned primitive tho. Might as well have a carburator.