LS6 Speed Density and MAF tuning HELP

[BuiltLifestyl3]

I've recently started making my first revisions to my original stock tune. I've made several logs during MAF tuning (dynamic air off, steady state disabled, dfco off, cot off, open loop, low octane spark...etc) and several with speed density (MAF fail 0, steady state disabled, dfco off, cot off, open loop, low octane...etc) tuning following all of Goat Rope Garage's videos on gen 3 tuning. I have his scanner layout I bought off his website and use it to log both kinds of tuning WITH AN AEM WIDEBAND. The car has long tube headers and a catless full exhaust done by previous owner, but it has always been on a stock tune until now. Though, I've datalogged the stock tune and the ltft's always showed lean (5s to teens). Even after making numerous MAF and VE changes, the ltfts still show lean in the teens. Stfts have always looked good, but not ltfts.

The first pics are the stock VE tables and graphs. The second pair of pics are my current table and graph. I've made all adjustments incrementally with lots of interpolation in various areas during each revision. Before saving each file after making changes, I'd smooth the graph once and upload those changes onto the car and do another log.

I've also included my latest log of ltfts, stfts, MAF error, and VE AFR error. On my error histograms, the low KPA rows never seem to change from their super lean state. Even with dfco off, I can never seem to fix. I don't know what steps I'm missing or what I'm doing wrong, but the fuel trims need to change. Though the car runs fine and doesn't knock. Wideband gauge doesn't usually ever go upwards from 15.6, but it sits around stoich most the time. StockVEtable.JPGStockVEgraph.JPGFullTune VE table.JPGFullTune VE graph.JPGltft.JPGstft.JPGVE AFR Error.JPG

[blindsquirrel]

Can't check a screenshot to see if you overlooked or misunderstood some essential setting to get it into SD or MAF, must post tune file (tune files - the ones you used for each method). I believe there is even a sticky thread at the top of this forum that you sailed right past that tells you this.

[SiriusC1024]

Have you confirmed the wideband results against the narrowbands?

[NBMY2KTA]

I had the same problem. He has the wrong values for the Column and Row Axis in both your STFT and LTFT Graphs. RPM and Map should be exactly the same as whats in your primary VE Table. You will not be able to copy and paste your data until they are the same. Hope this helps.

[aaronc7]

The last picture "LTFT Airmass" looks pretty good to me. For the low kPa cells, it's probably commanding some very small injector pw value, that due to the injector deadtime characteristcs, it actually isn't firing at all or very little fuel-- so its running lean. That would normally be DFCO region of course. It's probably fine as is. Or you can increase the injector min pw tables little by little until they start to appear to be firing correctly with fuel. I wouldn't really expect this on stock injectors but who knows.

VE is a bit like, but like mentioned put them into a histogram that matches the calibration axis exactly so you can copy paste initial corrections in.

And for MAF tuning, put the corrections into MAF table axis/format so you can apply proper correction to it. Just a guess but it might need a little tweak in the airmass/hz range, but mostly looks good as already mentioned above. If it's having to add a lot of fuel at low load/airmass, high vacuum only then it could indicate a vacuum leak as well for the MAF side.

[BuiltLifestyl3]

Ah! That might help a little bit. Yeah the vertical axis for the fuel trims are in increments by 2.5 and the VE is by 5s. Good catch, though I question why Kyle did it like that. I'll update the axis and update this thread.

[BuiltLifestyl3]

The last picture "LTFT Airmass" looks pretty good to me. For the low kPa cells, it's probably commanding some very small injector pw value, that due to the injector deadtime characteristcs, it actually isn't firing at all or very little fuel-- so its running lean. That would normally be DFCO region of course. It's probably fine as is. Or you can increase the injector min pw tables little by little until they start to appear to be firing correctly with fuel. I wouldn't really expect this on stock injectors but who knows.

VE is a bit like, but like mentioned put them into a histogram that matches the calibration axis exactly so you can copy paste initial corrections in.

And for MAF tuning, put the corrections into MAF table axis/format so you can apply proper correction to it. Just a guess but it might need a little tweak in the airmass/hz range, but mostly looks good as already mentioned above. If it's having to add a lot of fuel at low load/airmass, high vacuum only then it could indicate a vacuum leak as well for the MAF side.

Hm. Interesting take. That does sound a bit strange for stock injectors. As far as MAF goes, each cell is in the +5 to -2 range and is looking pretty good. That last picture is actually the VE AFR error. Not LTFT airmass. My mouse was just hovering over the LTFT Airmass page.

Alright so I fixed the row/column axis on all pages, but it didn't change much at all. LTFTs still show very lean. I guess I'll have to make another SD log and revision asap and see if the changes actually work this time. I checked the injector pulse width on those 0.15-0.20 KPA cells and they're between 0.4 and 1.7. All the healthy cells are between 2-4. Which table do I use to adjust the pulse widths?

[aaronc7]

Min injector pw
Default Injector pw
Short Pulse adder
Injector offset vs Battery voltage vs kPa Vac

no shit min pw I believe is min pw + injector offset (+ maybe short pw, but this probably isn't hugely significant anyways). 1.7 sounds about right quickly looking at a stock 04 C5 file. 1.25 + ~0.45 (depending on voltage and vacuum).

I'm not sure if that's it, but I have seen that exact thing happen when I was tuning my car with Deka 80 injectors. I just increased min pw table little by little until I found the actual min pw for the injectors and it didnt run incredibly lean (and jerky). Worst scenario (min pw) ended up being in 1st or second gear around 2k rpm... think like slow parking lot speeds.

So overall am I correct that AFR error is small (good) in open loop, both VE and MAF? And the issue is when you turn closed loop back on, that trims are way off? Maybe the stock O2s are suspect? Or wideband is suspect but coincidentally off enough to show good results? I do most of my low load fuel tuning closed loop, but my wideband/open loop has always matched closed loop/narrowband trims pretty well.

[BuiltLifestyl3]

Min injector pw
Default Injector pw
Short Pulse adder
Injector offset vs Battery voltage vs kPa Vac

no shit min pw I believe is min pw + injector offset (+ maybe short pw, but this probably isn't hugely significant anyways). 1.7 sounds about right quickly looking at a stock 04 C5 file. 1.25 + ~0.45 (depending on voltage and vacuum).

I'm not sure if that's it, but I have seen that exact thing happen when I was tuning my car with Deka 80 injectors. I just increased min pw table little by little until I found the actual min pw for the injectors and it didnt run incredibly lean (and jerky). Worst scenario (min pw) ended up being in 1st or second gear around 2k rpm... think like slow parking lot speeds.

So overall am I correct that AFR error is small (good) in open loop, both VE and MAF? And the issue is when you turn closed loop back on, that trims are way off? Maybe the stock O2s are suspect? Or wideband is suspect but coincidentally off enough to show good results? I do most of my low load fuel tuning closed loop, but my wideband/open loop has always matched closed loop/narrowband trims pretty well.

23-10-13 19-53-13.hpl Here is the latest log. Be the judge. 02s were replaced a few months ago.

[aaronc7]

Well, to me it looks like narrowbands and wideband are on the same page. Wideband bounces right around target AFR/EQ at steady state conditions.

It's odd at idle how trims are very minimal, then everywhere else they are very high. Is it possible that there's a more logical explanation for the big difference... i.e. when you thought you were tuning open loop MAF, it was still in SD mode for example? Not trying to insult your knowledge at all, I get goofy results sometimes and can't figure it out and turns out it was something simple/obvious I overlooked or thought I changed but didn't, or whatever. May be an unexpected but more simple explanation.

Did you tweak the MAF table at all during this? Is it on a stock intake or what intake setup? Does the MAF curve look goofy at all in any spots or other likely things that are off/weird?

As mentioned in first reply I think at this point we'd all need to take a look over the tune to help much more-- might be able to spot something out of place to give direction or more insight.

[BuiltLifestyl3]

Here's the latest tune revisions for MAF, SD, and Closed Loop/Normal. The MAF table was tweaked in correlation to the MAF error log. Changes were inputed with multiplying % by half. That goes for each kind of tune I've been revising. Stock intake, but with a cone filter. The MAF curve doesn't look wonky at all.ViciousMAFtuning7.hptViciousSDTuningCorrection10.hptViciousFullTune1(afterMAF7).hpt

[aaronc7]

I took a quick look and unfortunately I got nothing. All of the changes look sane to me and really in the grand scheme of things the MAF and VE table just had minor tweaks.... which makes sense given the few mods but otherwise very stock car. I'm kind of stumped right now and don't have any good ideas to explain the apparent strange behavior of the fuel trims.

After I typed this I opened the log one more time (I figured out the first attachment is actually a hpl log file)... it says fuel system is OL, but closed loop fuel trims are clearly active. Maybe just a weird fluke, but only thing that has caught my eye so far.

[BuiltLifestyl3]

So I experimented. I decided to add 10% to the entire MAF table and did a log with the new changes in closed loop/normal. For the first time, I started seeing green in the ltfts. It was from stand still to about the 50KPA and 2500RPM range on the log. Then everything beyond that turned somewhat red, so I'm suspecting the MAF table is the solution. I then did a MAF only log with the 10% fuel increase and started seeing more green in the MAF error histogram. Still had some red towards higher frequencies. So I subtracted a minor amount from the very green cells, then added about 5% to the red-ish cells. I then did another MAF only log with those changes and now most of the cells are green and everything is + or - 3. So I'm leaving it for now. Now I will do another SD only log and see if I can make a few better tweaks. If that pans out, then the closed loop/normal file should be showing green trims. Fingers crossed!

[BuiltLifestyl3]

Alright. Had enough driving for the night. Made a couple SD revisions and a few logs. This is the best I can do right now. SDTuning10.14.23.JPG