Throttle Cracker Tuning Question {B4309}

[joecar]

Sorry guys, I've had a lot going on lately... I've been following this thread, but not able to view files while out of town.

statesman, good job

[pinstripebob]

Hey Joe, Statesman's been helping out a lot! Stay safe traveling!

I went for a drive home and got another log. I didn't have too many bucking or idleing issues. I did get my other laptop going, so I should be able to log more frequently since it's dedicated to the garage. I did not have my headlights on during the log, except near the end when noted below. Below are my notes on the log, with ECT hitting 80C at frame 1818:

Bucking while crusing:
Frames 2708-2756 (minor bucking)
Frames 4643-4802 (4th gear, very repeatable part of my drive)
Frames 10319-10405 (more undulating bucking)
Frames 10529-10552
Frames 11667-11694

Bucking while accelerating:
Frames 4512-4529
Frames 6867-6880

Bucking while decelerating:
Frames 6201-6219
Frames 6721-6759 (minor bucking)
Frames 8583-8613
Frames 11891-11913

Throttle blip with headlights off: Frames 11993-12022

Throttle blip with headlights on: Frames 12055-12087

The engine speed seems to dip lower and take longer to recover with the headlights on. The log shows the recovery period when speed starts to increase being less sharp and taking a few extra frames, but it seems fairly noticeable in the driver's seat.

Let me know what you think. I haven't noticed any more or less bucking, although it does seem to do a better job returning to idle. I haven't been able to get a good idle surge in traffic.

Log:
GTO February 19 IBC to home.efi

[statesman]

statesman, good job

Thanks joecar

[statesman]

pinstripebob, Ive been away from my computer. I've got a lot to catch up on... will try to get to your log in the next day or two.

[pinstripebob]

Not a problem! I actually tweaked B4515 yesterday and did some driving, that seemed to help reduce idle hunting/surging. I increased it by 50%. Most values in it are lower than my stock tune.

I saw this thread: https://forum.efilive.com/showthread...d-Idle-Airflow

I've worked with PID controllers in the past. To oversimplify things, P is essentially how quickly you correct your error, I is sort of like a STFT where it gives you a little "memory" of your error, and D helps damp overshoot. The thread above concluded that B4515 was the same as the P value, so increasing it should help the ECU correct idle RPM errors more quickly. I'm a little tempted to zero out the D value table (B4512) just to see how the engine reacts, but that experiment can be saved for another day.

Below is the tune with the modified B4515 table I used. From your Mod8 there aren't any other changes. I've also included my stock tune for reference.

Mod8 with B4515 change:
Attachment 22682

Stock:
Attachment 22681

[statesman]

I actually tweaked B4515 yesterday and did some driving, that seemed to help reduce idle hunting/surging. I increased it by 50%. Most values in it are lower than my stock tune.

You might be masking another problem... but if it works for you, then we'll just go with it.

I've worked with PID controllers in the past.

Then you'll know that a high P value can 'appear' to help in some situations, but will cause greater overshoot in other situations. You won't know for sure until you've driven in high ambient temps whether this was a good idea or not.

It's good that you included your latest tune file... we need to be working with the same tune file while I'm making adjustments.

Log the same like you've been doing.

[pinstripebob]

Yeah you make a few really good points. I think I was blinded by optimism.

I flashed your tune, got a log, and overall it seems like things improved. There are still a few trouble issues, but they're fewer and less severe. I logged with the headlights off, but did the throttle blip when I got home with and without headlights on. I didn't have any areas of surging in traffic, although you can see some similar symptoms with the throttle blip. Typing this out, this may be in part to my B4515 change with an excessive P value causing unwanted overshoot.

In the log, the ECT hit 80C at frame 2229. Below are the highlights:

Bucking while crusing:
Frames 2983-3005
Frames 3486-3576
Frames 3750-3865
Frames 4531-4572 (low amplitude)
Frames 5050-5109
Frames 5433-5684 (similar to frames 4643-4802 of post 52 above)
Frames 5906-6530 (pretty long portion of the log)

Bucking while decelerating:
Frames 3043-3058
Frames 4158-4181
Frames 4944-5016
Frames 6536-6569
Frames 7581-7616
Frames 8843-8900
Frames 10458-10492
Frames 10723-10763
Frames 13499-13526

Surging while accelerating:
Frames 5354-5362

Throttle blip without headlights:
Frames 14944-14971
Frames 14972-15019

Throttle blip with headlights:
Frames 15021-15057
Frames 15064-15109

Let me know what your thoughts are. Again I really appreciate the help!

Log:
Attachment 22685

[statesman]

You've posted your old tune file. Log file names end in dot efi.

[pinstripebob]

I had a horrible thought that I posted the wrong file on my drive this morning!

Here ya go, sorry about that!

GTO February 27 IBC to home.efi

[statesman]

Typing this out, this may be in part to my B4515 change with an excessive P value causing unwanted overshoot.

Yep, it looks like you've created a bit of overshoot. I'm putting that table back to what it was. Maybe you could hold off making any more random changes until after I've finished helping you.

Your fueling is almost where I want it now. When you do the throttle blips, try to make them the same for headlights off and on. In your last log, the second blip went to about double the TP of the first. I know that it's hard to make two throttle blips exactly the same, but I do need them reasonably close so I can make a good comparison.

Log the same like you've been doing.