MAF tuning - error % keeps getting worse

[ssbowtie1]

Hi all,

I did 3 logs yesterday using OL and my wideband to record MAF % error and I thought I was on the right track but on my last log, my errors just shot up and I can't seem to get/keep the error % close to zero. I'm just posting the logs, because the only changes I made to the tune was to multiply by % -error after each log but if you'd like the see the tunes that are associated with the logs I will post them as well.

[Lakegoat]

Did you go to scanner--clear trims, before you started logging after writing calibration?

[ssbowtie1]

I did, just out of habit, but does it matter since I've only been logging in open loop?

[Lakegoat]

Good question. I do it out of habit also.

[TCSS07]

Im mobile atm so I can view attachments I will later but if you still have fuel trims enabled while doing o/l tuning it will mess with the data

[ssbowtie1]

Im mobile atm so I can view attachments I will later but if you still have fuel trims enabled while doing o/l tuning it will mess with the data

I disabled fuel trims in the tune but I've attached the tune for review.

Tune 5 corresponds to the changes I made based on the error of log 5.

Tune 6 corresponds to the changes I made based on the error of log 6.

Log 7 was run after tune 6 was uploaded.

[TCSS07]

i actually like the data from how your file was during log 5 - always make sure the injector data is correct
i tune differently - i just tune in closed loop and get the fuel trims to be + or - 5% to start then switch to o/l tuning to dial it in a little closer (but i usually don't) and for maybe during acceleration cells if i get some lean stumble and then just log my wot fueling against my afr error% vs maf and adjust accordingly because maybe its different for some people but on my car my wide band sways alot (no cat) and makes it hard to get proper data because one log itll be ok for average and then the next log i could be rich or lean for an average

[ssbowtie1]

i actually like the data from how your file was during log 5 - always make sure the injector data is correct
i tune differently - i just tune in closed loop and get the fuel trims to be + or - 5% to start then switch to o/l tuning to dial it in a little closer (but i usually don't) and for maybe during acceleration cells if i get some lean stumble and then just log my wot fueling against my afr error% vs maf and adjust accordingly because maybe its different for some people but on my car my wide band sways alot (no cat) and makes it hard to get proper data because one log itll be ok for average and then the next log i could be rich or lean for an average

Thanks for the feedback. I'm actually running stock injectors. I thought open loop would be the easiest because I wouldn't have to worry about filtering out PE data. I thought log 5 was great also, and log 6 was just to confirm and get some WOT data for the higher Hz MAF cells.

I checked the WB for drift and it's reading on point. The only thing I can think of is I must have developed some kind of leak between log 6 and log 7, which is sending unmetered air into the motor. Is there anything in the log that I can look out for that could confirm this?

Needless to say, I was pretty bummed out after doing so many runs and thinking that I was about to prefect the MAF and be able to move onto the VE table.

[TCSS07]

I never filter data I just make another histogram logging what I want like make one for command afr vs maf and one for afr error vs maf and just compare against my ltft vs maf histogram and see which cells hit pe but when I tune i dial in my fuel trims with doing normal driving and some acceleration cells - im boosted so I have pe happen around 3psi - for wot data I do a pull from a low rpm like 2500 in a higher gear like 3rd gear like I would on a dyno and which cell it fell in first id consider it wot fueling from there in up

[TCSS07]

If you want to know for a vac leak pop it back in closed loop and at idle the trims should go positive

[ssbowtie1]

Well I threw in the towel for a bit but I decided to take the car out for a drive today without changing the tune. The MAF error % values look much better than log 7.

I'm pretty confused, why would trims read 8% on one run, then 1%-3% on another run, using the same tune

[gmorris]

Was you injector tip temp similar between runs? I've had that throw trims off by 4-5% between a cool morning and a hot afternoon.

[Area47]

you need to drive it a few miles before you start logging as you're running against injector temp adders/ibt/inj temp gain.

[ssbowtie1]

Thanks for the feedback. I have my injector temp gain table set to zero. Are there other tables that I should investigate?

[jmanson62]

SS, I too have run into this. I found that logging a run in the morning and then without changes log again in the afternoon, my trims will go leaner. It's my understanding that air temp has a lot to do with it. I've read on other sites that when tuning try and stay consistent with weather and pressures. Just my two cents.

[gmorris]

Thanks for the feedback. I have my injector temp gain table set to zero. Are there other tables that I should investigate?

That is really not the correct way to do this. By zeroing out that table you run the risk of running very lean at high temps. You're better off spending some time actually calibrating that table rather than zeroes. There are a couple good threads on here about it.

https://forum.hptuners.com/showthrea...ector+tip+temp

[jsllc]

Regardless of what may be on the internet. ZEROing injector temp gain table is VERY bad. Just because it can be done does not mean it should. It will result in excessively rich and lean extremes depending on conditions.

[superman07]

I thought you only zeroed it out while on the dyno in a controlled environment, then you set it back and dial in temp compensations as you identify issues when driving.

[g26er]

I thought you only zeroed it out while on the dyno in a controlled environment, then you set it back and dial in temp compensations as you identify issues when driving.

This is also what I took from it... just for the dyno and then back to stock or where ever it was before once you were done.

[jsllc]

Guess if you are looking for an excessively lean condition while placing your engine under extreme stress. That would be one way to do it. Not how I would recommend running the engine.