Constant overshoot when VE tuning

[G-Body]

I`m kind of a newb when it comes to tuning as all I have done before was my 0411 pcm`ed 6.0l el camino and I remember taking forever to get the ve table close enough for me to accept it.

Now I`m trying to tune another 6.0l TR220 cam, longtubes, controlled by a 5.3l 6106 blue/green pcm. I started by copying over a stock 6.0l ve table and tuning from there. But no matter what I do it just seems to make the ve table worse. When I started out I was way rich at idle like 40% according to my log and fairly rich for most of the part throttle cruise areas of the table. I copied the whole table, then did the paste special multiply by %. I went from being 40% rich (by the exhaust smell I don`t doubt it was) to 60% lean. I backed out quickly went back to the stock table and did a paste special multiply by % - half. Again it was lean, but only 35% at idle. I have tried to get the VE adjusted multiple times before on this car and I cannot seem to get it within +/- 10%. Either it massively under or over corrects.

I also noticed that I cannot put any sort of filter string on my histogram, every time I do it does not report any data. My VE - WB AFR error hisogram reports no data, while VE - WB AFR error 1 hit 1 has no filter but reports just fine. I tuned the MAF and the car runs pretty well in open loop MAF, but I really want to be able to re-enable the VE.

[G-Body]

The rest of the scans

[G-Body]

Ok, I figured out why the filter in the WB AFR did not work, I copied it off a histogram I downloaded and it had a > sign when it should have been <. I still can`t get the VE table to be consistent though. I started out this morning trying to just get the VE table dialed in at idle. I noticed last night that the wideband would hit approx 14.2 if I used the scanner to command a 13.2 AFR so I added 7% to the ve table in that area. The afr then went 6-7% rich. I thought I was getting closer and multipled the 400 and 800 rpm ve columns by 0.97 to cut 3% of the fuel. When I reflashed and started logging I was now 40% lean. I reset the fuel trims after each flash, so ??

[G-Body]

On my last scan I did notice that I always seem to be in fuel trim cell 22 could that be part of the issue? what would cause me to be stuck there?

[mowton]

The .cfg file and scans you posted don't show any means for determining fueling errors? I don't see enough PIDs in you chart either? How did you perfrom the above logging/updates? Are you just looking at the wideband display?

Will be in cell 22 because you are in Open loop and failed MAF...

Double check the attachments so I can see if the problem is on my side.....

Ed M

[G-Body]

I am logging my wideband through the fuel tank pressure sensor. All of the data comes up and populates the hisotgram ok. The problem comes in when I copy it and paste special multiply it into the ve histogram. Sometimes it barely changes the mixture, other times a 3% cut in fuel takes me from being 7% rich to 40 some % lean. I attached the screenshots of my PID`s and histogram

[mowton]

I am logging my wideband through the fuel tank pressure sensor. All of the data comes up and populates the hisotgram ok. The problem comes in when I copy it and paste special multiply it into the ve histogram. Sometimes it barely changes the mixture, other times a 3% cut in fuel takes me from being 7% rich to 40 some % lean. I attached the screenshots of my PID`s and histogram

I had to replicate your WB error PID per the above, in the .cfg you supplied, not sure why but got inserted now. First question, How was the interface/results validated for accuracy and correct results?

Ed M

[Nastruck]

Are you truly suppose to copy paste special multiply by % into the ve tables? My truck wont hardly run if I ever do that. If I paste add % it seems fine but is that wrong?

[mowton]

Has to be multiply %.........

Ed M

[G-Body]

I`m running an innovate LC-1 wideband 0-5v 7.35-22.39AFR. I did the calibration on the sensor per the innovate instructions, but I guess that I can`t say I have ever calibrated it to anything else. As of yesterday I was wondering if the wideband was giving me bad data. I was so pissed I cleared everything out, started over and tried to get the low load / part throttle VE tuned in using VE vs STFT. The result was the same, I generally overshoot the correction. After thinking about it some more, I think the wideband and the wideband afr error PID are working correctly as I was able to use them to tune the MAF in. I have the MAF table tuned within 5% using the wideband.

I keep feeling like I must be forgetting to disable something when I try to tune VE. Or that I am not filtering my results enough. I noticed that since I started filtering with [SENS.121] >14.6 it seems to be slightly better. I have DFCO disabled, but when I was watching as I logged yesterday I still had the feeling that my readings were being skewed when I was moving the throttle. Maybe I need to filter out any results within x milliseconds of a change in TPS? But that still would not make sense as to why I can`t get the afr error at idle nailed down.

[WS6FirebirdTA00]

I would adjust the range for the WB. That is too large. You want to be able to have better resolution, especially if you aren't logging directly through HPT. Have you used log works and also confirmed what you see in HPT is what the WB sees? I can't open it here, but how long are your logs and what type of driving is it?

[mowton]

The voltage levels are bouncing way too much. I am struggling trying to grasp the accuracy of your WB input. Here is what I recommend......

1) Make sure that that all the grounds (WB and PCM low for the fuel pressure sensor input) are tied together and located to a good common ground point close to the PCM
2) When you do the LC-1 Free Air calibrate, you should see 20.9 AFR. Hook up your scanner and log to verify you are seeing 20.9 AFR as you WB reading through HPTuners to start and it is stable
3) If you can, put the tune back in closed loop, and monitor the WB output to verify it is reading on or very near your commanded AFR and it is somewhat stable.
4) Logging VE (and MAF for that matter) requires very stable driving practice whereas you collect data at a specific point and when it stabilizes you move to the next. Easy/gentle pedal press and slowly work to fill as much of the table as possible in each session. You don't have to do it all, pick a section, collect steady state data, copy/paste mult %, clear trims, do it again.......otherwise known on the forum as "wash, condition, rinse, repeat" :-)
5) In the LC-1 Advanced Analog Out setting table, make sure it is set to 1/6, not instant.

Make sure you clear the trims via the VCM before each logging session ......

Also you filters aren't right. Don't filter on the commanded AFR, Filter on the following......TPS%>20%, ECT>80, MAP>45kPa

Ed M

[MikeOD]

I`m running an innovate LC-1 wideband 0-5v 7.35-22.39AFR.

You've gotta change that scale...input voltage fluctuations are going to result in MASSIVE implied AFR swings, when the AFR really isn't moving that much. Make the range something more like 10:1-16.5:1 (Lambda 0.68 - 1.12) and work from there.

Also, use "Paste Special - Multiply By % - Half"...if you don't use the half, you'll more often than not find it overshooting when the error is large.

[G-Body]

It might be my wideband. The grounds and wiring are all good, but I can`t seem to plug into the LC-1 with a laptop. I updated my laptop since the last time I plugged in to it and my new laptop does not have a serial port. I cannot get the LC-1 to work with a usb-serial converter. I tried to redo the free air calibration but it reads 22.33 during and after the calibration (via HPTuners since I can`t get the logworks to hook up). Looks like I`m going to have to find a computer with a serial port and see what is going on with my lc-1. Thanks for the help.

[mowton]

It might be my wideband. The grounds and wiring are all good, but I can`t seem to plug into the LC-1 with a laptop. I updated my laptop since the last time I plugged in to it and my new laptop does not have a serial port. I cannot get the LC-1 to work with a usb-serial converter. I tried to redo the free air calibration but it reads 22.33 during and after the calibration (via HPTuners since I can`t get the logworks to hook up). Looks like I`m going to have to find a computer with a serial port and see what is going on with my lc-1. Thanks for the help.

Will stand by....

[MikeOD]

HPTuners isn't going to show you when it's reading free air, it's going to show you the leanest number your output range can show you when the wideband is reading free air.

[mowton]

HPTuners isn't going to show you when it's reading free air, it's going to show you the leanest number your output range can show you when the wideband is reading free air.

Yeah and that is 22.XX......