Failed MAF = Intermittent Stumble

[Alvin]

Exactly! And you've caught far more precious thing

Attachment 140121

Thanks, Cringer.

that is it.

You can zoom in on the VVE table by changing axis's and not find a blip like that.

[Alvin]

Here's my issue, in a screenshot and attached log. Seems like the exact same symptom. Thoughts on how this would happen on a Gen III truck?

loose ground or something along those lines is usually the cause.

F bodys would do this a lot when guys wouldn't put the battery tie down back in and let the battery rock into the fender.

[GHuggins]

I am curious to know what you can find, although I am not sure there is going to be any real fix we can do with HPT.

If Verlon can find the root cause in the coding and it's not in the hard coding then you might be surprised

[Cringer]

If Verlon can find the root cause in the coding and it's not in the hard coding then you might be surprised

What I am saying is, my guess is that there is a defect in the code. No way to update the algorithm.

My hope is there may be work arounds relating to manipulating some of these values:

• Prediction Coefficients
• MAP Filter and Sampling Coefficients
• [ECM] 3034 - Spark Airmass Blend
• [ECM] 13426 - VE Initial Correction Factor
• [ECM] 12003 - Intake Manifold Volume

These are all a shot in the dark...hoping Verlon can find the smoking gun!

[L1FTD]

What I am saying is, my guess is that there is a defect in the code. No way to update the algorithm.

My hope is there may be work arounds relating to manipulating some of these values:

• Prediction Coefficients
• MAP Filter and Sampling Coefficients
• [ECM] 3034 - Spark Airmass Blend
• [ECM] 13426 - VE Initial Correction Factor
• [ECM] 12003 - Intake Manifold Volume

These are all a shot in the dark...hoping Verlon can find the smoking gun!

Just passing on some info that seems to have fixed the issue for me.

I've been getting the same 'VVE Bug' problems. 2200rpm and below, everything drops - fuel, airmass etc. Have had some cutouts at 5200rpm too, everything drops and recovers.

I'm OL SD, E67 LSA.

I've had tune iterations with, DynAir enable/disable maxxed - random/occassional stumbles, not too bad. DynAir enable/disable down at 400rpm, still SD - stumbles much worse, very regular when 2200rpm and below.

After seeing the thread in attached pic, I checked my AirCalc state when the stumbles occur. Stumbles only seem to happen after I've gone over 4000rpm. If I drive for an hour and don't go over 4000rpm - no stumbles. Once I go over 4000rpm, stumbles come in and also transients go to shit.

This coincides with going from AirCalc state - Normal, to HiSpeed, and then getting stuck in NormalTrns for the rest of the drive.

Anyway, I've since put my DynAir enable/disable back to stock, and applied the Speed Density Air Mode patch.

Now I can go over 4000rpm, and my AirCalc will go straight back to Normal after HiSpeed. No stumbles so far and fueling tracks much smoother. Transients stay the same throughout the drive too (before was big lean spikes when stuck in NormalTrns).

AirCalcModeThread.png

[Cringer]

Just passing on some info that seems to have fixed the issue for me.

I've been getting the same 'VVE Bug' problems. 2200rpm and below, everything drops - fuel, airmass etc. Have had some cutouts at 5200rpm too, everything drops and recovers.

I'm OL SD, E67 LSA.

I've had tune iterations with, DynAir enable/disable maxxed - random/occassional stumbles, not too bad. DynAir enable/disable down at 400rpm, still SD - stumbles much worse, very regular when 2200rpm and below.

After seeing the thread in attached pic, I checked my AirCalc state when the stumbles occur. Stumbles only seem to happen after I've gone over 4000rpm. If I drive for an hour and don't go over 4000rpm - no stumbles. Once I go over 4000rpm, stumbles come in and also transients go to shit.

This coincides with going from AirCalc state - Normal, to HiSpeed, and then getting stuck in NormalTrns for the rest of the drive.

Anyway, I've since put my DynAir enable/disable back to stock, and applied the Speed Density Air Mode patch.

Now I can go over 4000rpm, and my AirCalc will go straight back to Normal after HiSpeed. No stumbles so far and fueling tracks much smoother. Transients stay the same throughout the drive too (before was big lean spikes when stuck in NormalTrns).

AirCalcModeThread.png

Wow this is very informative! Thank you for posting this info. I seemed to notice a correlation after going WOT, but not specifically 4,000 rpms.

Verlon has been working hard behind the scenes here and created a custom patched OS. I have not had a chance to test it...but his solution is different than HPT's though.

[L1FTD]

Wow this is very informative! Thank you for posting this info. I seemed to notice a correlation after going WOT, but not specifically 4,000 rpms.

Verlon has been working hard behind the scenes here and created a custom patched OS. I have not had a chance to test it...but his solution is different than HPT's though.

Ok, great stuff!

I noticed in your screenshots of the stumble, you were in that same ?NormalTrns? AirCalc too.

Putting my DynAir disable back to stock did not fix the AirCalc, but the air mode patch did.

It might not be exactly 4000rpm, but somewhere close. Whatever the rpm the switch between ?Normal? and ?HiSpeed?. Once I hit ?HiSpeed?, I would never get back to ?Normal? after. Only ?NormalTrns? or ?HiSpeed? again.

Also, did not matter that I had DynAir disable maxed at one point. AirCalc would still change to ?HiSpeed? (i.e. MAF only mode). Interestingly, putting DynAir disable down at 400rpm, I would be in ?HiSpeed? all the time, and that made things worse.

Anyway, will keep following

[verlon]

That's awful, but yes, there is another bug in the AirCalc Mode state machine. It can get stuck in some state if DynAir Enable/Disable thresholds pair is lower than fuel Fast/Slow thresholds pair.

[SiriusC1024]

Where are the Fast/Slow fuel settings?

Is this it?
fuel fast_slow.png

[L1FTD]

That's awful, but yes, there is another bug in the AirCalc Mode state machine. It can get stuck in some state if DynAir Enable/Disable thresholds pair is lower than fuel Fast/Slow thresholds pair. Being stuck in this state has terrible consequences - engine speed variable, used for all DynAir calculations (inc. VE), freezes at the moment of stepping into this state.
HPT patch does not solves this bug.

Interesting!

Are the fuel fast/slow thresholds accessible in HPT?

I assume this would be irrelevant if the DynAir thresholds are left stock.

Occasionally at high rpm (usually around 5200) I would get a cut. Fuel drops to zero, airmass zero, MAT zero. Recovers almost instantly, but obviously very concerning.

Thinking back, these high rpm cuts only started happening when I unnecessarily maxed the DynAir thresholds. MAF is de-pinned, so I had no need to change the DynAir.

Since the AirCalc patch and DynAir back to stock, I am still yet to have a low rpm stumble or high rpm cut. Hoping I had two separate problems, with two separate fixes applied. Will see if it lasts.

Moral of the story for me - only change what is necessary, leave stock where possible Too many hidden tables

[Cringer]

Where are the Fast/Slow fuel settings?

Is this it?
fuel fast_slow.png

Here is the explanation on those fields:
https://forum.hptuners.com/showthrea...l=1#post719020

[verlon]

Where are the Fast/Slow fuel settings?

Is this it?
fuel fast_slow.png

No, that's closed loop timing.

Fuel timing is different and not defined in HPT as Chris stated.

[Alvin]

Awesome. It would be nice to have a fix. There are several platforms that do not have the 2 bar OS available. A fix is definitely needed.

[SiriusC1024]

No, that's closed loop timing.

Fuel timing is different and not defined in HPT as Chris stated.

I take it you can see it, though. Are you able to define for us a no-go range of DynAir filtered MAF enables to prevent this phenomenon?

[SiriusC1024]

Here is the explanation on those fields:
https://forum.hptuners.com/showthrea...l=1#post719020

Good info thanks for the link.

[GHuggins]

Verlon's already got a patch he's testing to my knowledge. Great to know he found it and it can be fixed.

Verlon, will these be available via xdf's? Or is a complete code patch rework required? Can we kill the baro update too?

As always, Thank You Verlon!!!

[verlon]

I take it you can see it, though. Are you able to define for us a no-go range of DynAir filtered MAF enables to prevent this phenomenon?

On L94 DynAir Enable/Disable matches fuel thresholds (3900/4000).

Verlon's already got a patch he's testing to my knowledge. Great to know he found it and it can be fixed.

VE Patch is tested. I'm preparing detailed description of the VE bug and what exactly patch does.
AirCalc Mode patch will take some time.

Verlon, will these be available via xdf's? Or is a complete code patch rework required? Can we kill the baro update too?

HPT does not allow modifying OS via xdf and they are right, it is an easy way to brick ECM.

[Cringer]

Verlon's already got a patch he's testing to my knowledge. Great to know he found it and it can be fixed.

Verlon, will these be available via xdf's? Or is a complete code patch rework required? Can we kill the baro update too?

As always, Thank You Verlon!!!

Yes, I did verify Verlon's patch worked today — it prevents the fueling spike. However, the "AirCalc Mode" getting stuck in "Normal Trns" pointed out by L1FTD is still an issue. As to what impact this actually has, I do not know. I think Verlon may be working to address that to, but that is his call.

The other crazy part is that applying this patch solved some moderate idle surging that was occurring after installing a 103mm throttlebody. So major kudos to Verlon on his work on this!

[GHuggins]

On L94 DynAir Enable/Disable matches fuel thresholds (3900/4000).



VE Patch is tested. I'm preparing detailed description of the VE bug and what exactly patch does.
AirCalc Mode patch will take some time.



HPT does not allow modifying OS via xdf and they are right, it is an easy way to brick ECM.

Maybe I should have worded differently instead of grouping everything together. Will the rpm settings be available via xdf and do they just need to be the same as the dyn settings? Added the last part but it just came up as another possible question

[L1FTD]

Here are some screenshots of the NormalTrns issue.

First three are what was happening for a long time. First part of drive before going over 3800-4000rpm = Normal, then to HiSpeed, then stuck in NormalTrns.

Second three are after the HPT SD air mode patch. Normal, then HiSpeed, then back to Normal - no NormalTrns again.

Before Air Patch - Normal.pngBefore Air Patch - HiSpeed.pngBefore Air Patch - NormalTrns.png

After Air Patch - Normal.pngAfter Air Patch - HiSpeed.pngAfter Air Patch - Normal Again.png