LTFT Calculation in PE mode

[Perforator]

Hi all, I was wondering if anyone knew what the PCM uses to calculate LTFT in PE Mode. I'm having a problem where, when I'm in FTCell 3 and still closed loop, my LTFT are generally around 0-2.3. When I'm in FTCell 3 and I go into Open Loop/PE Mode, my LTFT's go to 7.8-8.6. I'm running really rich in PE Mode because of the addition of the LTFT and I don't want to lean out my MAF Table because I'm afraid that I'll be really lean if the LTFT's decide to go back to 0 again. I've tried richening up the MAF Table where it goes into PE Mode and I've tried richening up the MAF table right before it goes into PE Mode and it didn't make a difference with the LTFT's. Basically I need to know what to change to lower the LTFT's in PE Mode...

I'll attach a copy of the scan and my scan config.

[Russ K]

Read up on this thread for more info.

http://www.hptuners.com/forum/showth...oop+maf+tuning

Russ Kemp

[Perforator]

Thanks Russ, I've read that before but it still doesn't answer the question.

[devildog9999]

Shoot me the bin to look at. [email protected]

[z34phoenix]

you can try and lean out your base PE AF ratio... then reset the trims... but when your in PE the trims lock and the car uses the last known trim as a default.. so watch your MAF data just before you see PE. and start adjusting there

[Perforator]

but when your in PE the trims lock and the car uses the last known trim as a default..

This is incorrect. If you filter out the scan and only include FTC3 where the STFT value is greater than 0, the values range from 0 to 2.3. If it did lock in at the last value it would use those instead of the 7.8 - 8.6 that it is using. Apparently it uses something else to determine PE fueling.

[BlackGS]

I have only been successful in PE by tuning the MAF. Even if I set PE commanded to something super rich like 10.5, the MAF controls the fueling, and I still see the same wideband afr that I did before the adjustment.

[Perforator]

Yes I agree. The problem is though, when I enter PE, my LTFT's are locking in at a high number, like 7.8 or 8.6, and it's making me run rich.
I need to change it but don't know how.
It doesn't use the O2 sensor for feedback in PE mode so it must be figuring a LTFT value out some other way...

[Perforator]

I have only been successful in PE by tuning the MAF. Even if I set PE commanded to something super rich like 10.5, the MAF controls the fueling, and I still see the same wideband afr that I did before the adjustment.

On second thought, I disagree with this, lol.
If you lower your commanded A/F value through the Base Pe or PE vs Time, whatever, your actual A/F should lower even if your MAF table is incorrect unless of course you've maxed out your injectors and they cannot provide any more fuel. I would imagine you would need to be reeeely rich to do that though... I've done this on my car before when I was trying to get my PE A/F dialed in and it worked like it's suppose to.

[BlackGS]

your actual A/F should lower even if your MAF table is incorrect.

Yup, that's what I would have thought too, but I have my commanded in PE at 10.5 now, and still only get actual readings between 11 and 12 based on the MAF settings. If I adjust the MAF though, I can get it down to 10.5.

Adjustments to my MAF will throw off the LTFT's in PE, so I still think you have to start with a MAF tune.

[Perforator]

Yup, that's what I would have thought too, but I have my commanded in PE at 10.5 now, and still only get actual readings between 11 and 12 based on the MAF settings. If I adjust the MAF though, I can get it down to 10.5.

Yea that's the way your suppose to do it. Set your base PE, adjust your MAF so your wideband reads what your base PE is. Once it matches you should be able to adjust your base PE to whatever you want and your actual should follow...

[johnh]

From what I've seen its some kind of averaging function. FTC4 can easily be tuned with the upper MAF range seen when not in PE (~5500 hz I think). FTC3, the only way I know of is to get all your LTFT nice and close to zero.

Try this: Reset your trims, get a nice long scan (after the engine is warmed up). See if there are any spots where the LTFT are more than say +6%; try to tune those out.

I will say my Trims are much more stable after just 1 round of VE tuning.

[Perforator]

I noticed it too about the VE table. I did 3 runs on it and the cruise and idle LTFT's are much more stable now than they use to be.

[Perforator]

Ok, I think I can pretty much rule out the LTFT's locking in there last value when you go into PE. At least with my car. I'll attach 2 scans. There are multiple places in both them but I'll just point out one in the first scan. Starting at frame 1223, I'm in Fuel Trim Cell 3 but not in PE yet. I'm still commanding 14.7:1. The Long Term Fuel Trim is 0. At frame 1228 the commanded A/F value goes down to 12.9, the Short Term Fuel Trim goes to 0 and my Long Term Fuel Trim goes to 8.6. If it did lock in it would be 0. It doesn't and like I said it happens in several places in both scans. I've tried increasing the MAF table by 4% above, I think, 7000hz. All that did was make me richer in PE. Last night I tried increasing the VE table using the averaged LTFT values in PE, nada. Tonight I think I'm going to try and increase the IFR table by 4% in the 100kpa cell...

Edit: I was going to upload the scans in csv format but I guess you can't... Here's a link to the scans at scandepot...

http://www.scandepot.net/viewscan.php?pk=4664
http://www.scandepot.net/viewscan.php?pk=4665

[Russ K]

I would set up the Maf histogram, plotting the STFT+LTFT pid. Then with the Maf histogram open, smoothly accelerate up to ~6000 Hz (stay out of PE).
Then copy & paste special > multiply by % to the Maf table in the editor.

Most of the positive fuel trims are occurring <4000 Hz.

Russ Kemp

[BlackGS]

Very strange that your MAF change is affecting AFR in PE, but LTFT's are not moving at all. Did you make any AE modifications?

[Perforator]

Very strange that your MAF change is affecting AFR in PE, but LTFT's are not moving at all.

It makes sense to me. The MAF is used for calculating airflow in PE so if you change it, it will change the a/f ratio. Not really sure how it comes up with LTFT in PE... I have made changes to AE but seperate to this issue. I've had the issue with the LTFT's before and after I made the AE change and my PE AFR has always changed when I changed the MAF table.

[BlackGS]

It makes sense to me. The MAF is used for calculating airflow in PE so if you change it, it will change the a/f ratio. Not really sure how it comes up with LTFT in PE... I have made changes to AE but seperate to this issue. I've had the issue with the LTFT's before and after I made the AE change and my PE AFR has always changed when I changed the MAF table.

I should have made that clearer. I was surprised that your MAF changes, which should have changed LTFT's just before entering PE, had no effect. You said you modified the MAF starting at 7000hz. At 7000hz on your scan you are still commanding 14.7, so I would have thought to see some effect on LTFT's right before and into PE. I'm as stumped as you are why you have zero LTFT a second before PE, and then it jumps to 8.6!

[Perforator]

I actually started at 7125hz... I upped the IFR 100kpa cell by 4% and just looking at my Aeroforce gauges the LTFT when I got on it a little was 8.5 . I'm going to scan later but I don't think that was it either...

[JerryH]

That is a factor... you can't just do a fueling change and expect to be able to get accurate scans immediately after. The trims need to be relearned. Fastest way "was" to reset them, go for a 15-30 min drive with at least 1 key-on-key-off in a safe location. You can actually watch the PCM learning the trims visibly while watching the STFT and LTFT PIDs.

Some year cars are very fast, others are slower. 97s need about 12 min to relearn and don't need a "KOn-KOff". My car used to take about an hour... but if I reset the trims, drove for 5-10 min, did a "KOn-KOff", drive another 5-10 min I was ready to scan and collect data for the next tune.

04's are the longest to learn from what I hear. Without a FTR, a minimum of 1-2 days is required. That makes tuning a looooooong process!