2008 TBSS E67 airmass calculation glitch?

[kjlindgr]

Hey all,

I have a 2008 TBSS with a bunch of mods that seems to periodically have a hiccup while steady state cruising. It doesn't happen all the time, and it doesn't seem to happen under heavy load, but under mostly steady state conditions, it'll happen.

What I notice is that the cylinder airmass calculation swings drastically low or high for just a split second. This causes the injectors to either shut or dump rich, as well as the timing to dip very low or swing very high. But only for a split second. Again, this is mostly steady state. I don't see any strange torque management happening. Trans isn't shifting. VSS is fairly constant. TPS isn't opening or closing. It's like the computer fails to accurately calculate the airmass load for just a brief moment.

Attached are some logs, screen shots, and the tune. Has anyone run into this before?
Tune File: TUNE_0001_withTCU_flex_iat_18.hpt

In this log, you can see at time 3:50:220, the ECU calculated a very high injector pulse width, and pulled a bunch of timing, for just a split second:
cut-1.png
Log file: cut1.hpl


And in this log at time 1:30:981, the opposite happened. The engine pulled all fuel for just a split second:
cut-2.png
Log file: cut2-fuel.hpl

[hjtrbo]

This has come a bit lately. Put the MAF back in or go to the 2bar OS seems to be the cure.

Worth a search for it. Was only in the last month. There was some good discussion that applies identically to your issue.

[kjlindgr]

Thanks a lot for the response. I will try to see if I can find the post. Was it specific to the E67 ECU by chance?

I don't see a 2bar OS patch, just a patch used to allow the injector flow rate table read into boost. The MAF is in, but I'm running SD. I'm not maxing it out so I may have to go back to a MAF tune it seems.

[Alvin]

I call it the VVE error or VVE bug.

It's been around forever. There is no fix for it unfortunately other than applying a 2 bar OS if available or turning the MAF back on.

It's annoying. The good news.. most only do it once in a blue moon.

[redhardsupra]

does it always happen in a similar MAP,RPM range?

[kjlindgr]

My suspicion is that it always happens in a similar area, but it's not super consistent. It seems to happen around 3800RPM and 80kpa, but it does vary. Some runs are clean, some runs are earlier, a few runs are vary close to these values.

Next bit of help. I tried to convert this to MAF but it barely idles. I pulled my old logs (steady state, dyno pulls, etc) to rebuild my MAF g/s values, and plugged them in to get close. Like I said, I can barely get this to idle no matter what I try. One interesting thing is that in the SD tune, injector pulse width is about 0.80-0.90ms hot and it's right around 14.5:1 AFR. With the MAF tune, pulse width is 1.20ms and it's dead lean, computer is adding 30* of timing to try to maintain idle speed (but failing), and there is zero throttle response. It's almost like the firing order is off or something very wrong. No DTC's. I'm wondering if because I've doubled stoich/IVT/etc if I need to change my approach to my MAF scaling? I really don't know.

Any suggestions?


Log and Tune file with MAF enabled:
start2.hpl
TUNE_0009.hpt

[Alvin]

does it always happen in a similar MAP,RPM range?

They will often happen in the same area but do not resolve completely by moving map x RPM boundaries.

[GHuggins]

does it always happen in a similar MAP,RPM range?

It acts like a self test going on - will do it in time or "temperature" intervals right after startup (at 60f, 90f and 120f most times on the really big trouble makers all whilst idling). The same cars will do this in steady state throttle cruise scenarios and while going wot. Almost acts like a MAF test or baro update re-running even though it probably shouldn't be - goes lean then rich in a split second and the bigger the injectors the worst it is. I know the 2 or 3 bar OS's pretty much fix it. Things they change are the baro update test as far as I can think that correspond solely to the VVE tables.

Fifth gens do this too, but only if you induce it by zeroing out the prediction coefficients. They'll dump fuel in SD mode just the same primarily while cruising though.....

[TheMechanic]

Pretty sure those are the HO2s doing a self test.
Reason for right after start is cat has not started working so it can get a good AF swing for downstreams and be able to effectively diag for slow downs in switching

[kjlindgr]

Pretty sure those are the HO2s doing a self test.
Reason for right after start is cat has not started working so it can get a good AF swing for downstreams and be able to effectively diag for slow downs in switching

I have the O2's turned off, or at least the readiness set out of normal operating range. Is there any way of effectively turning off the test?

For whatever reason, I don't have the ability on this OS to do a 2/3 bar patch. Is there a different E67 OS I can use to do that?

[Alvin]

I disagree.. It looks like a bug to me. GMVE will either hit numbers not possible in the table or crash to zero.

[GHuggins]

Pretty sure those are the HO2s doing a self test.
Reason for right after start is cat has not started working so it can get a good AF swing for downstreams and be able to effectively diag for slow downs in switching

Definitely not a cat or O2 test unless deactivating it doesn't work. Plus everytime this happens it's like it ignores airflow and drops everything dyn to 0, shoots cyl air through the roof and dumps fuel as a result. It'll do it so bad on the "oversized" inj setups that they'll stall the engines on warmup at these temp intervals. Then restart will show super rich for a brief moment. That's with fuel trims being out as little as 6 to 10 percent. You can very fine tune the ivt table for these specific temp problem areas to help by getting fueling more precisely dialed in, but the problem will still be there while cruising or everywhere else. At least that's what I've found. Definitely a glitch or something going on though.