Extremely Weird Airmass Dip Under Cruising Steady State

[cmitchell17]

So I just put a turbo on my 2011 Sierra Denali 6.2 L9H running about 6 psi of boost. I wanted to retain the MAF but had no way to put it in line so I am speed density now, which I enabled by setting the MAF fail frequency to 1 Hz.

So I have had the turbo on for a few days now but I feel like this weird issue just started happening and I haven't changed anything that I can really think of in the tune.

So from looking at the logs I can't really tell for sure what's coming first, the dip in timing, the dip in airmass, or dip in something else, I am thinking something is messed up somewhere and causing the ECM to think airmass is going to almost 0, so it responds by looking up a different timing value and stuff. When it happends torque management advance remains at zero. It is occurring at what seems pretty steady state, and going down the highway when the TC is completely locked up it will shock the whole truck and clunk the driveline. I may not be sampling the injector pulsewidth fast enough, but I can also see one bank go to the MIN pulsewidth of about .8ms from around 6 ms, the other bank stayed the same. So It would make sense that if the PCM falsely thinks airmass randomly and instantly decreases to almost zero at steady state, then it responds by cutting fuel to stay at stoich, then I should have somewhat of a hickup and maybe slight clunk, but its almost as though not only does the engine go to zero torque, its almost like it reverses torque as well but maybe it just feels that way.

I was able to find a pretty bad one in the log, they are not consistent between one another, some are worse than others:
Capture.JPG

Does anyone have any idea what's going on here? I am wondering if maybe some unstable coefficient got solved for correcting VVE somewhere that maybe isn't showing up in HP Tuners VVE tool, but the PCM/ECM is seeing it? I had to export a smaller log file as my original one was 10 MB and it wouldn't let me attach, but there is still a few examples of the random airmass dip there. Thanks.

White Truck Turbo (LS9 Stock Timing) ECM TCM FICM.hpt
White Truck Turbo Log 4 Random Zero Airmass Dips Causing Clunk Hiccups_Smaller Export.hpl

[blindsquirrel]

That logfile is broken, all but like 3 channels got lost during the export. That happens sometimes.

Desoot is still enabled. Catalyst test is still enabled, even though post-cat O2s are disabled. Virtual flex fuel is still enabled (bad!). PE Enrichment Rate is an unbelievable 0.05, I would use 1.00 as a default and probably never need to touch it again. I'm sure there is more but...

[GHuggins]

Also under operating system apply the speed density air mode and vacuum patches.

[cmitchell17]

That logfile is broken, all but like 3 channels got lost during the export. That happens sometimes.

Desoot is still enabled. Catalyst test is still enabled, even though post-cat O2s are disabled. Virtual flex fuel is still enabled (bad!). PE Enrichment Rate is an unbelievable 0.05, I would use 1.00 as a default and probably never need to touch it again. I'm sure there is more but...

So the 0.05 enrichment rate is what my stock calibration had in it, from my logs it seems I go into PE mode and the wideband shows richer almost immediately?

I will disable desoot mode, I had never even seen this before, but I am used to GEN V, GEN III, and Duramax stuff.

I have looked all over for the Catalyst tests and the cat warmup mode tables but I can't find them, where do you see that? I see the cat lightoff tables in relation to spark but nothing else?

Also I forgot to add I was also getting bust knock activation under steady state, half throttle, low boost conditions as well so I completely disabled burst knock retard. I always thought burst knock only activated under a delta airmass and not under steady state, but it definitely was activating under steady state.

Since I couldn't upload the log file again, here is a link to the log and another similar one but that also had weird lean issues at half throttle as well.

https://1drv.ms/u/s!AkCUE89IhD0Ng-gK...ORlEw?e=6n7O7U

https://1drv.ms/u/s!AkCUE89IhD0Ng-gM...TYkbw?e=c1mmyR

[cmitchell17]

Also under operating system apply the speed density air mode and vacuum patches.

Ok thanks, I will try that. If I interpreted you from the other thread I think you said you preferred VVE? but maybe these patches have nothing to do with VVE versus a lookup VE table.

Also for these 2 patches, it dosen't say anything about having to re-write entire, but for the other patches it does. Should I write entire after I apply them? I know the documentation says you can brick if you don't re-write entire with these patches, but maybe that was for older versions of the editor and maybe the newer ones are more idiot proof.

[blindsquirrel]

Recent versions of the editor have some of the catalyst settings removed, I'm still on an old version. Here's your file with just that one setting changed that you can't get to.

White Truck Turbo (LS9 Stock Timing) ECM TCM FICM Cat test.hpt

[cmitchell17]

Recent versions of the editor have some of the catalyst settings removed, I'm still on an old version. Here's your file with just that one setting changed that you can't get to.

White Truck Turbo (LS9 Stock Timing) ECM TCM FICM Cat test.hpt

Thanks blindquirrel, ill give it a try.

I also reverted back to my VVE table from the day before when I don't remember it doing it. So far after driving it a good while so far the issues haven't happened yet.

Does anyone think that the VVE table maybe getting "corrupted" or maybe the internal solver didn't converge good enough and maybe the editor might show a valid VVE table visually, but in reality when the ECM/PCM goes to use those coefficients, maybe it goes unstable or something? Is this plausible? I remember on my Camaro GEN V one time I edited my VVE and re-generated the coefficients using the HP Tuners VVE editor and I didn't check it over and flashed the cal and the car ran like crap and I couldn't figure it out until I realized what had happened.

It's almost like trying to do slight tweaks to the VVE table then resolving the coefficients is bad maybe.

[07GTS]

raise your dynamic airflow too u still have 4000 and 3900 set those to 8100 and 8000, also thresh counts and test rpm can be zeroed in diag tab

[cmitchell17]

So the airmass dip issue came back today even when I reverted back to my old VVE table. Seems like it's almost randomly consistent and the issue comes in waves.

[cmitchell17]

I thought the dynamic airflow rpm was to tell the PCM when to stop using the VE table and rely solely on the MAF? So if I raised these to the highest rpm then I would effectively disable SD mode and I have no MAF so what would the PCM do for airmass estimation?

[GHuggins]

Yes, it's to tell it when to start using MAF at 100%, so by raising it the ecm wont be looking for it. Honestly when I do SD only tunes I do some things differently. I start by unpinning one pin out of the MAF connector, which you stated the MAF wasn't there so not sure if you have it plugged in somehow or not - then I zero out the entire MAF tabless - then I set the P0102 codes to fail on cranking and then set the P0101 codes to never test and the map/maf correlation codes to never test. You might want to do that and see where your at. I don't think you have a dip causing weird issues in your VVE. You can always run a 2bar custom os if in doubt. Your VVE looks wrong for the setup just looking at it. I'm pretty sure you have a test of some sort going on with the maf that's most likely causing your problem where something isn't possibly failed 100%.

AND since your now running boost, something else I've seen but rarely run into is the torque model being off. Usually this is OS to OS basis. You'll drive on such light throttle that it'll compute that your in decel causing the throttle to close and timing to pull, but this should go away the second you re-apply throttle if that's the case. You should also be able to see this by logging throttle states or something to that nature.

[cmitchell17]

Yes, it's to tell it when to start using MAF at 100%, so by raising it the ecm wont be looking for it. Honestly when I do SD only tunes I do some things differently. I start by unpinning one pin out of the MAF connector, which you stated the MAF wasn't there so not sure if you have it plugged in somehow or not - then I zero out the entire MAF tabless - then I set the P0102 codes to fail on cranking and then set the P0101 codes to never test and the map/maf correlation codes to never test. You might want to do that and see where your at. I don't think you have a dip causing weird issues in your VVE. You can always run a 2bar custom os if in doubt. Your VVE looks wrong for the setup just looking at it. I'm pretty sure you have a test of some sort going on with the maf that's most likely causing your problem where something isn't possibly failed 100%.

AND since your now running boost, something else I've seen but rarely run into is the torque model being off. Usually this is OS to OS basis. You'll drive on such light throttle that it'll compute that your in decel causing the throttle to close and timing to pull, but this should go away the second you re-apply throttle if that's the case. You should also be able to see this by logging throttle states or something to that nature.

Yes I agree that trying to delete the MAF has something to do with this whole thing. So I maxxed out the dynamic filtered airmass, and also disabled all the MAF tests and P0101 stuff. Then I tried to start it and it wouldn't start. This not starting system has been present from the beginning ever when I removed the maf and put the turbo on, but then went away at some point maybe after I went back to close loop I can't remember.

So does it have something to do with me not reporting the DTC's? I've never found a clear answer, and there seems to be conflicting reports, about are you supposed to disable the DTC's and the tests or just the DTC's or what? That's why I didn't really mess with the tests and just set the MAF high and low DTC's to no error reported.

So I flashed it back with the stock settings for the MAF tests but left the dynamic filtered airmass maxxed out. Then I tried to drive it and it started fine but once I would get above half throttle and into boost it would go extremely lean and start to cut out.

[blindsquirrel]

MAF DTCs must be set for it to fail over into speed density. If you set them to No Error they will never set and it'll never go into SD and if there is no MAF installed that results in a no-start.

[cmitchell17]

MAF DTCs must be set for it to fail over into speed density. If you set them to No Error they will never set and it'll never go into SD and if there is no MAF installed that results in a no-start.

So I meant to say I have only the MAF circuit range performance set to no error reported, P0101. The high and low, P0102 and P0103, are both set to MIL on first error. I also set P1101 to no error reported. I thought that since we are using the frequency range to "disable/fail" the MAF then we would only want the high and low codes set to MIL on first error, or at least changed from MIL on second error to MIL on first error, I would assume so the first time the key is turned it triggers it?

But I did turn off the "SES Enable" check boxes for P0102 and P0103. I have heard this can mess things up if you turn off this checkbox and you are supposedly supposed to leave it on? I have heard if you want to disable something you turn it to no error reported but leave the SES enable on, but really who knows with this kind of stuff, since we have to trial and error millions of combinations of things, or dig through millions of lines of code.

[4wheelinls1]

This is glitch I have seen on many occasions and have logged multiple tickets with HP over, you cannot run the VVE without the MAF, you need to use the speed density OS upgrade. One issue is the ecu has not got great computing power and when calculating VVE it seems to have this odd fault, wait until it happens at 6500 rpm on the dyno at WOT, it has to me. I was unable to link it to an OS, I have examples of it in multiple OS. The feedback I got from Chris P was that it is a GM error that won't be fixed. If you can retain the MAF do otherwise do the SD OS. i have seen this in the early pre 2009 E38 even with the SD OS, you would swear it was de-soot but once again I have been told the early E38 does not have de-soot.
Set the MAF to fail like the image, you want to set the counters and RPM to 1 so it fails straight away otherwise after a code clear of flash, even a stall it will look for the absent MAF and not start.


Capture.JPG

[cmitchell17]

This is glitch I have seen on many occasions and have logged multiple tickets with HP over, you cannot run the VVE without the MAF, you need to use the speed density OS upgrade. One issue is the ecu has not got great computing power and when calculating VVE it seems to have this odd fault, wait until it happens at 6500 rpm on the dyno at WOT, it has to me. I was unable to link it to an OS, I have examples of it in multiple OS. The feedback I got from Chris P was that it is a GM error that won't be fixed. If you can retain the MAF do otherwise do the SD OS. i have seen this in the early pre 2009 E38 even with the SD OS, you would swear it was de-soot but once again I have been told the early E38 does not have de-soot.
Set the MAF to fail like the image, you want to set the counters and RPM to 1 so it fails straight away otherwise after a code clear of flash, even a stall it will look for the absent MAF and not start.


Capture.JPG

Thank you very much. So much stuff was not making logical sense, inconsistent errors, working but not working, working sometimes, changing this and it working, then not working other times. I've tried so many combinations of things. But from what you are saying makes perfect sense now.

I haven't been able to test all permutations and combinations but I think I can make this work (minus the random airmass dip making the driveline clunk) if you leave the Dynamic Airflow High RPM disable at stock along with the MAF test thresholds and only the MAF high and low DTCs set to first error.

However, I bet there would still be all kinds of weird stuff.

So I was trying to retain the MAF I just can't figure out how I am going to physically put it in the charge pipe.

But one more thing, are you saying I need to run a custom SD OS? or I need to apply this patch:
Capture.JPG
I haven't found a clear answer to this but I believe most people had customized OS they were full flashing over theirs in order to make a SD only tune, but maybe not.

Also can I apply the patch with just a regular flash or do I need to full flash it after enabling the patch or patches?

Thank you so much for replying as I would have been spending months on this trying ever stupid combination trying to make things work as I hate giving up on stuff like this haha. I will enter a ticket for this, although I already have but I was blaming it on the VVE coefficients being badly calculated, but I will bring it up to them again so maybe we can get something to happen on it.

[cmitchell17]

Also another thing I was thinking, what if I just plugged the MAF back in but let it just sit in free air? I guess the PCM would still see its resistance value.

Maybe what is complicating matters too is that I have the IAT sensor broken out, which maybe is confusing the PCM even more.

[GHuggins]

I made the necessary changes to this one for you to try. The VE may be slightly off. I changed your map boundaries to get smoother table transitions. Doesn't appear you have a sd os available. Wouldn't think the maf hanging would make any difference as it's still failed.

[4wheelinls1]

Thank you very much. So much stuff was not making logical sense, inconsistent errors, working but not working, working sometimes, changing this and it working, then not working other times. I've tried so many combinations of things. But from what you are saying makes perfect sense now.

I haven't been able to test all permutations and combinations but I think I can make this work (minus the random airmass dip making the driveline clunk) if you leave the Dynamic Airflow High RPM disable at stock along with the MAF test thresholds and only the MAF high and low DTCs set to first error.

However, I bet there would still be all kinds of weird stuff.

So I was trying to retain the MAF I just can't figure out how I am going to physically put it in the charge pipe.

But one more thing, are you saying I need to run a custom SD OS? or I need to apply this patch:
Capture.JPG
I haven't found a clear answer to this but I believe most people had customized OS they were full flashing over theirs in order to make a SD only tune, but maybe not.

Also can I apply the patch with just a regular flash or do I need to full flash it after enabling the patch or patches?

Thank you so much for replying as I would have been spending months on this trying ever stupid combination trying to make things work as I hate giving up on stuff like this haha. I will enter a ticket for this, although I already have but I was blaming it on the VVE coefficients being badly calculated, but I will bring it up to them again so maybe we can get something to happen on it.

Oh, I just saw there is no OS upgrade option, that sucks. Log the VE airflow channels, if they drop to 0 its an ecu error.

[GHuggins]

I didn't do it in that one, but you need to apply the tcs patch too then do a write entire. You know it's funny, I've never had this issue with SD, but I always have this issue with turbo's and MAFs. This is interesting...