Buick Regal Falls on its Face

The issue is when the driver goes WOT and it goes into what feels like limp mode.
You can see the problem in the log but the short of it is that the car just hits a wall at the top of first gear, shifts to second, and then does not accelerate for a few seconds. In the log you can actually see the throttle close.

I logged torque requests and just can't figure out what's happening.

Attached is the cfg, tune file (stock and modded), and 2 logs
All taken with 2.24

Any help is hugely appreciated.
Attachment 56407
Attachment 56408
Attachment 56409
Attachment 56410
Attachment 56411

This is what it looks like to me.

You are hitting the rev limiter(log shows 6400rpm and tune is set to 6300) and your switchover from fuel/spark to ETC is set to a delay of 0 seconds. In Fuel>DFCO/cutoff look for a table named "switchover delay." You can either raise your rev limiter or increase the values in the switchover delay so it uses spark and fuel to limit revving. I haven't tuned an LDK/LHU so I can't recommend which is better, but the Cobalt valvetrains are good to 7200rpm stock FWIW

Bandit, Thanks for the response. I think you're looking too early into the log, I'm okay with the throttle closing when the limiter is hit.
At 1:36 sec into "tune3 brick wall2.hpl" the problem occurs. The RPM never goes above 5240 rpm brfore shutting the throttle.

Attachment 56436

Took another look at the problem today, was able to make it go away with the attached file, but the driver reported the issue came back a few hours later.

Logged with a different cfg too.
Attachment 56482
Attachment 56483
Attachment 56484

try setting all revlimiter values back to stock, I suppose the accel thresh vs acccel is causing it.

btw, also lots of other things in this tune make not much sense for me

What else doesn't make sense to you?
If you do a comparison with the stock file in my first post you can see all the changes I made...

i did, DD way too high, torque tables bumped up too much, lots of stuff which isn't necessary and maybe freaks it out.

There is good reason to remove the torque limiters, by setting them high. We don't want them getting in the way of the car making boost. Max boost is controlled by the Max "Boost Max Limit" table.
I appreciate the attempt to help but you might want to read up on the torque model
https://www.hptuners.com/help/vcm_ed...vanced_e78.htm

Something is commanding the engine to deliver LESS torque which the the PCM removes torque by lowering desired boost, and closing the throttle.
Your claim that the torque tables are too high is illogical. I suppose if one set the torque tables so low that the vehicle never made over stock boost that might fix the issue, but then you might as well flash it back to stock.

This throttle closing issue is somewhat documented
https://forum.efilive.com/archive/in...p/t-23580.html

Additionally by setting the driver demand high, we are making sure the torque request doesn't get too low. If the requested torque is less than the delivered limiting occurs.

"tune_7" was a significant improvement over "tune_3" the only issue now is that the PCM seems to "learn" and begins shutting the throttle after so many launches/ 1st gear rolls.

Quote:

---

Originally Posted by VodeAn

There is good reason to remove the torque limiters, by setting them high. We don't want them getting in the way of the car making boost. Max boost is controlled by the Max "Boost Max Limit" table.
I appreciate the attempt to help but you might want to read up on the torque model
https://www.hptuners.com/help/vcm_ed...vanced_e78.htm

Something is commanding the engine to deliver LESS torque which the the PCM removes torque by lowering desired boost, and closing the throttle.
Your claim that the torque tables are too high is illogical. I suppose if one set the torque tables so low that the vehicle never made over stock boost that might fix the issue, but then you might as well flash it back to stock.

This throttle closing issue is somewhat documented
https://forum.efilive.com/archive/in...p/t-23580.html

Additionally by setting the driver demand high, we are making sure the torque request doesn't get too low. If the requested torque is less than the delivered limiting occurs.

---

I suppose it s an AT car, when you bump up the Trq tables that much, the TCU possibly gets wrong engine torque readings, and tends to make strange things or requests less torque.
I did hundreds of these cars and never had any issue like this.

Yes it is an automatic, and we don't have access to the TCM so maybe the problem could be there.


I flashed a version with lower torque tables and it didn't have the issue for a few days and all of a sudden it came back.

Attachment 56594

Here is my 2 cents based on tuning some of the other newer Direct Injected Turbo ECUs from GM.

I find that when PEAK TORQUE table and BOOST MAX LIMIT produce a condition that exceeds the allowed Airmass in the TURBOCHARGER KNOCK MAX AIRMASS table, the actual vs demanded TPS takes a nose dive and the car becomes unresponsive.

I might be slightly wrong because it's been about a month since I last had one of these types of ECUs in front of me. Let me go back and double check that it wasn't the other way around (where TURBOCHARGER KNOCK MAX AIRMASS > BOOST MAX LIMIT wasn't the cause and that I have my logic backwards, etc.)

And here is where I was talking about:

Screenshot 1: This is right before the ECU tells you to totally hold your horses. You will notice that actual Cylinder Airmass is still below the Knock safety threshold, and the ECU is still commanding 100% ETC position.

Attachment 56797

Screenshot 2: You will see that a split second later your actual Airmass exceed the maximum allowed airmass for knock safety, and the ECU therefor pulls the commanded ETC position % relative to what the foot is telling the car to do. The car will then continue to protect istelf from allowing this condition to occur and you will typically feel that the car has NO DESIRE to gain RPM or allow engine load when this occurs until a brief moment later.

Attachment 56798


Also, just as a matter of opinion. You run that engine pig rich in that log. 11.17 AFR on a DI, whew!

Hope this helps you buddy!

Thanks a ton! I think that'll fix it, if you look at "tune8" I actually accidentally left the entire airmass table stock which would explain why it still did it but less frequently since I'm commanding less boost in "tune8"
In the other files where I command more boost I raised the table, but like you noted I didn't raise them enough.

Also yes the AFR is rich (stock targets), I didn't want to start leaning the vehicle out until we had all the issues sorted out.

Thanks again

No problem. Glad I can help. Sometimes when you stare at the same car for too long your brain starts to cross wires.

not a real difference from what i said, that the trq tables are bumped up too much

Quote:

---

Originally Posted by dertobi

not a real difference from what i said, that the trq tables are bumped up too much

---

Other than being specific and providing examples based off of the logs and tune file he provided to make it more clear and relevant.

dertobi let's break this down and look at your suggestions...

I suppose the accel thresh vs acccel is causing it. <- wrong

DD way too high <- wrong, was not causing any issue or negative impact (some of the low throttle torque request is from a ATS and the higher throttle torque requests are set high so they do not become a limit. With most torque based controllers max engine torque is commanded when wot is detected at the pedal, but we set the demand high just to make sure it doesn't become a limiting factor, again you might want to read up on the torque model)

the TCU possibly gets wrong engine torque readings, and tends to make strange things or requests less torque <- wrong, the TCM uses delivered engine torque for its calculations. If there is a first gear torque limit in the TCM causing the problem, then it would be directly related to the actual engine torque produced, by keeping the torque limits low in torque management you could prevent the engine from making enough torque to hit the supposed 1st gear torque limit but that totally defeats the purpose of trying to make more power. Setting the limits high does not send the TCM an incorrect torque reading because setting torque limits high does not equal a high delivered engine torque signal. I thought this was pretty clear when I explained it a few posts up.

Again, the solution here wasn't to neuter the car up with low max torque tables but rather to raise the knock airmass table so the car could be allowed to produce the all power that it can for a given desired boost level.


Never once did you mention the knock airmass table, and you want to take credit for the fix?? lol k

That's nice you wanted to help but this is a science, ideas are recognized and valid when they:
a. produce repeatable, testable, results
b. include data and logical reasoning behind them