RomRaider

ok found it but something very strange is going on.... the log attached shows the lean out but just before it at the rpm/load/mrp/tps the final base was 14.7-1 - this log shows mid to high 15's low 16's is randomly does this in my troubled rpm range. I logged boost error/ battery voltage/ iat/ coolant temp against it and none of which had any effect on the lean out or final base lean out...

This is obviously my problem, as it leans out and then gets a huge correction factor and then the next time target is 14.7 and it goes nuts trying to get things right making my CL choppy as hell...

My closed loop is close to stock configuration - what in the bloody hell is causing my final base target to go to 16.5-1?????

this could be interesting..

-Jerod

btw, the first part of the log is for reference, there are a few spikes (lean outs) in the final base but the end shows the real problem..

Not sure what you are referring to - where it goes lean with final fueling is right after letting off the throttle.

Wrong log then - it was during steady driving when it happened..


-Jerod

We'll have to get back to this in a few weeks. 3rd gear had other things to do today, like shear all the teeth off All that at 15mph in 3rd putting in traffic haha


6mt will be here in a week or so and we will resume trying to figure out why I'm leaning out...

merchgod,

Here is my issue, graphed. It shows that the CL target stays steady, however the final fueling base drops way rich. Ideas? This issue is causing my AFR to spike lean on throttle change, cause knock, and drop my IAM (sometimes). I would really like to get it resolved..

http://pasichnyk.net/wrx/tip-in-issues- ... target.png (woudn't let me upload a decent resolution image).

Thanks,
pwrx

That is strange. That is one grey area that I haven't completely worked out where there is some cl target compensations that are not applied to final fueling, while some are, and apparently, from your log, there are enrichment factors which aren't accounted for in the cl target. Not sure of the reason behind that. I would have thought that, in closed loop, the calculated fuel would factor in any changes to the CL target from 1.0 lambda, allowing short-term correction to correct to the target as necessary. That doesn't always seem to be the case and, in fact, there is, what would seem to be, unnecessary use of short-term correction to correct to a target which is already known.

The only way I could narrow it down would be to send you some custom logger defs.

If you have some custom defs or could throw some together to help isolate my issue, it would be very greatly appreciated. I've read a bunch of other threads where it seems people probalby are having the same issues, and think it is just a problem with their tip-in. I think if this extra correction can be isolated and addressed, then it will save a lot of people from pulling out their hair.

Let me know if you come up with a custom def that may help.

Thanks,
pwrx

I'm game but its going to be a week or so - Don't know how far its going to make it until something cool happens - like complete failure of the trans

merchgod,

Any eta on when you could have some updated def's to help isolate this fueling issue?

I know your busy, but thought I'd check.

Thanks,
pwrx

maybe after the next def release

I think I found the problem on my car... And you should definately check yours as well - the Vent from the PCV valve to the block had a crack running on the bottom side and was not vissible until removed yesterday..

This would make sense as I go lean about the time I go into boost...


-jerod

Jerod,

It looks like you have incorrect latency\capacity or MAF calibrations. This is evident in CL mode. Less latency and higher capacity digits may fill the bill. After that transition fueling should not require those huge CL corrections.

From the other hand OL fueling looks fine. How may it come with improper MAF\latency\capacity calibrations?

Is it possible that you have (different\dirty) injectors with different flow capacities or latencies?

Jerod,
Were you seeing issues only when in boost? I'm seeing my issues when doing stationary revs too. Did your Closed Loop and Final Fueling targets look just like mine? You sure fixing the leak resolved your issue?

Thanks,
pwrx

Brand new injectors with 100% known latency values... .01ms either way and I start getting weird fbkc when I don't now..

It could me maf related, but when I scale my maf sensor to compensate the ecu just takes away fueling on long term.. Almost like its not consistent lean out... Say I go to 17-1 at 2600 rpms and 2.10v on the maf sensor - So I add around 14% maf scaling to compensate (close to that anyway) my long term fuel trims in that learning area will almost max out negative... Then I'm back to square 1... But then 5 min late when I hate the same values at the same RPM I'm at 12-1 afrs.... Its almost like I have an intermittent leak. Its always right at or real close or just past -0- MRP... That leads me to believe that its related to a leak...

I haven't had a chance to test it out as my car is on jack stands waiting on DS1 to come up with a 6speed for me (They are good at pissing me off btw) Once that is done I will have some results.. I bought ALL new Silicon hoses for all of my engine ventilation... All new fittings and everything to ENSURE its airtight..

We shall see...

-Jerod

merchgod,

I know your still working on the next def release, so I didn't want to bother you, but I'm still plagued by this fueling target issue. Any idea of things to look into to help determine the root cause until you get some time for some custom def's?

I was always under the impression that I had some physical issues as I always had IAM issues back as far as I remember, and I believe I had them since i got the car, just based on the "seat of the pants" dyno...

Thanks man,
pwrx