Fighting Rough Idle

[cessnageek]

Hi,
I'm fighting a rough idle problem that I can't seem to get to go away. Made a post a few weeks ago that cleared up a lot of my confusion around idle fueling, but still having idle problems.

2012 5.3 LMG, DOD & VVT delete, BTR Truck Norris cam, cold air intake, E38 controller.

It idles super rough. I've attached two logs showing this, as well as the tune I am running. The roughness is visible in the RPM graph. I've got my idle fueling close, I spent a bunch of time tuning airflow final minimum to run at my desired RPM without adaptives, I've adjusted spark overspeed/underspeed, and I've adjusted PI gains on the idle air controller. Nothing has fixed it.

I would just chalk it up to it being my first cammed vehicle, but the funny part is there are stretches where it is smooth, then it gets rough again. Plus it's a mild cam. So I know I should be able to get it to run better. I don't think it's an idle adaptives problem, because I turned them off when I did idle air tuning, and it was still pretty rough even though RPMs were somewhat consistent. I've got no clue what is going on here, and advice is appreciated.

Idle_Drive.hpl
IdleLog.hpl
08122024_AfterIdleAirTuning.hpt

[abc]

Add STIT and LTIT to the PID list and log the idle again.

[cessnageek]

Not seeing those in my scanner list. I'm not connected to the truck though. Can check at lunch.

[Alvin]

That cam while small in duration (relatively) is on a 107 LSA

It's going to idle rough.


I missed the days when people chose a cam with performance in mind and not a name/hit a lick

[cessnageek]

Got busy at work, and am now trying to figure out a tick that I've started hearing under light acceleration/load. So tuning is on hold for the moment. However, I checked and don't have STIT or LTIT in my scanner PID list.

[Cringer]

Got busy at work, and am now trying to figure out a tick that I've started hearing under light acceleration/load. So tuning is on hold for the moment. However, I checked and don't have STIT or LTIT in my scanner PID list.

Those are Gen3 PID's you don't have them for Gen4 ECMs.

[cessnageek]

Just a quick update in case this is able to help anyone else. I'm starting to get a better handle on what controls what, and I went through the Gen 4 idle tuning guide again. As best I can tell, I had way more idle spark than the motor wanted, which contributed to the rough idle. After dialing in the idle airflow, as well as my idle spark, it runs quite well. That also helped fix my clunky closed throttle downshifts. I *think* my brake booster had a small leak as well, but I still got it to idle alright even with that. I replaced that last night and need to touch up the idle tune now.

Ended up with roughly 11.5 deg of timing in case that helps anyone with a similar combination. I also put shorty headers on the truck since I made the original post. Feel free to critique my idle tune, as I'm sure it's not optimal by any means. But seems to work decently well.

08232024_Idle_Final_Pre_Brake_Booster_Replacement.hpt

[edcmat-l1]

Your improvement in idle quality came from the changes you made to the over/under tables, not to the base idle timing. 11-12 degrees is too low. That puts a considerable amount of heat in your headers. You'd be better off a few degrees higher than that and keep your over/under tables as you have them.

[cessnageek]

Didn't know that. I can revisit that tonight. What would you expect idle spark to look like then? Closer to 15 deg-ish, or is that still a little on the low end? By the same token, how much is too much? Am I trying to go as high as I can and still get a decent idle?
My lack of experience definitely shows here

[edcmat-l1]

Didn't know that. I can revisit that tonight. What would you expect idle spark to look like then? Closer to 15 deg-ish, or is that still a little on the low end? By the same token, how much is too much? Am I trying to go as high as I can and still get a decent idle?
My lack of experience definitely shows here

I would defer to your stock calibration file. Just copy that table back over and leave the rest as is for now. But specifically, you'd want somewhere in the 15-18 degrees range.

[cessnageek]

Fair enough, I'll start there.

A couple questions if you get the time:

1). If I understand correctly, GM calculates dynamic air based on some secret sauce blending/filtering of MAF and VVE in my case, since I'm using both. Is dynamic air still the airflow model that controls the injectors even when at idle? If so, how does my airflow final minimum table play into this? Is basically a lower limit on how low the controller tries to bring dynamic air at idle? I'm not entirely sure what that whole relationship looks like.

2). How do we define when the controller is in idle mode? Is idle any time I'm not pressing the throttle pedal, or are there transition periods? For example, if I'm doing a pull and let off the throttle suddenly, does it immediately jump into idle control, or is there some sort of ramp down from partial/WOT to idle?

Thanks!

[Cringer]

Fair enough, I'll start there.

A couple questions if you get the time:

1). If I understand correctly, GM calculates dynamic air based on some secret sauce blending/filtering of MAF and VVE in my case, since I'm using both. Is dynamic air still the airflow model that controls the injectors even when at idle? If so, how does my airflow final minimum table play into this? Is basically a lower limit on how low the controller tries to bring dynamic air at idle? I'm not entirely sure what that whole relationship looks like.

2). How do we define when the controller is in idle mode? Is idle any time I'm not pressing the throttle pedal, or are there transition periods? For example, if I'm doing a pull and let off the throttle suddenly, does it immediately jump into idle control, or is there some sort of ramp down from partial/WOT to idle?

Thanks!

1) Yes, dynamic air controls the injector at idle.

Airflow Final Min is a lower cap on the calculated airflow. Idle air is complex. I believe the ECM has an internal torque model that takes a bunch of things into account. You are telling it the base idle spark and what RPM's you want to idle at. The ECM then has to determine how much torque is needed to achieve the desired RPMs, then it has to calculate [how much air] + [this much spark] = [the desired torque] which will hit those RPMs.

In your scanner you might be able to log (HPT does not map all these for all OS's):

• Idle SS Air (which is the [how much air] part of the equation)
• Idle Proportional Air
• Idle Integral Air
• Idle Minimum Airflow
• Idle Desired Air

So...

• [Idle Desired Air] = [Idle SS Air] + [Idle Proportional Air] + [Idle Integral Air]

Then the ECM compares [Idle Desired Air] against [Idle Minimum Airflow].

• If Desired Air is less than Min Airflow, then Min Airflow is used.
• If Desired Air is greater than Min Airflow, then Desired Air is used.

Ideally, when tuned properly your Desired Air should always be greater than Min Air. Min Air should be a last ditch backup that is rarely used.

2) You can add PIDs in the scanner called:

• Idle Mode
• Idle SC mode

But I am not sure if those will really help. As far as when idle control is active, you can also look on the Idle > RPM tab and look at the Adaptive Idle sections. You will see a few rules at the top relating to min RPM, VSS, throttle delay, and RPM error for both Prop and Int controllers.


I recently made some videos that goes into more details if you are interested:

• https://youtu.be/QeS81Ig_EWE
• https://youtu.be/wovq5hi3Db0

[smokeshow]

2). How do we define when the controller is in idle mode? Is idle any time I'm not pressing the throttle pedal, or are there transition periods? For example, if I'm doing a pull and let off the throttle suddenly, does it immediately jump into idle control, or is there some sort of ramp down from partial/WOT to idle?

Throttle follower manages the handoff from driver torque control to idle speed control. After the follower has ramped out, the difference between the present dynamic airflow and steady state idle airflow is set as the new integral airflow value to finish the handoff. Once the speed falls below the adaptive idle threshold, proportional idle airflow is re-enabled along with adaptive spark. On gen 4 anyway...

[cessnageek]

Ed - stock cal file is ~18 deg in my idle area. Seems to work alright. Might pull 1 deg out. Need to get my idle air model a dialed in better first though. I think that's most of the problem right now. I think I was abusing the final airflow minimum table before as a way to not really get my VE table correct at idle. Didn't really understand that we are still using dynamic air at idle. So gotta fix that.

Cringer - Thanks for the info! I'll give those videos a watch. Unfortunately, I have none of those idle PIDs on my E38. Maybe you explain in the video, but is SS airflow just my dynamic air at idle before the PI controller corrects on top of it?

smokeshow - That makes a lot of sense. I wasn't quite sure what the throttle follower parameters were about. Thank you!

edit - just watched the first video. Never mind my SS airflow question

[Cringer]

Cringer - ... is SS airflow just my dynamic air at idle before the PI controller corrects on top of it?

No, SS (steady state) airflow is not related to Dynamic Air. SS Air is the amount of air the ECM calculates it needs to achieve the desired idle speed, then yes, P and I make corrections on top of SS Air when the idle goes high/low. Dynamic Air is still calc'd at idle in the traditional way — input from MAF and VVE. You will still need to calibrate MAF and VVE for idle conditions to produce fuel trims of 0%.

[cessnageek]

Man this thing idles rough now that I've increased idle timing back to stock. Fuel trims at idle are under 4% error. Not great, but close enough that I'm thinking something else might be causing the roughness. I've been running in VVE only for the time being as I've been trying to dial that in. I've not had a ton of success with the MAF + VVE simultaneous method. Could be user error though.

Attached a 10 min idle log and the tune if anyone has got any ideas. My VVE table is ugly because I've only been working in the idle region for the time being. Also attached the stock tune for comparison. Ignore the date in the tune file name - I need to update that. Thanks!

Cringer - so SS air + P and I is what is controlling the throttle body then? And my dynamic air is reacting to that and controlling my fueling appropriately?

Edit - Also attached for reference a log from when I had the timing pulled at idle to 11.5 deg. It's the drive from Dairy Queen log. The first half of the log is all idle while I'm in the drive though, and I really liked how it idled here. This is what I'm shooting for, but without 22% throttle and super low timing

08232024_Idle_After_Brake_Booster_Replacement_3_VE_ONLY_3.hpt
IdleLog3.hpl
08022024_StockTune.hpt
DriveFromDairyQueen.hpl

[Cringer]

Man this thing idles rough now that I've increased idle timing back to stock. Fuel trims at idle are under 4% error. Not great, but close enough that I'm thinking something else might be causing the roughness. I've been running in VVE only for the time being as I've been trying to dial that in. I've not had a ton of success with the MAF + VVE simultaneous method. Could be user error though.

Attached a 10 min idle log and the tune if anyone has got any ideas. My VVE table is ugly because I've only been working in the idle region for the time being. Also attached the stock tune for comparison. Ignore the date in the tune file name - I need to update that. Thanks!

Cringer - so SS air + P and I is what is controlling the throttle body then? And my dynamic air is reacting to that and controlling my fueling appropriately?

Edit - Also attached for reference a log from when I had the timing pulled at idle to 11.5 deg. It's the drive from Dairy Queen log. The first half of the log is all idle while I'm in the drive though, and I really liked how it idled here. This is what I'm shooting for, but without 22% throttle and super low timing

08232024_Idle_After_Brake_Booster_Replacement_3_VE_ONLY_3.hpt
IdleLog3.hpl
08022024_StockTune.hpt
DriveFromDairyQueen.hpl

Give this a shot.

08232024_Idle_After_Brake_Booster_Replacement_3_VE_ONLY_3 - cringer.hpt

[2xLS1]

That cam while small in duration (relatively) is on a 107 LSA


I missed the days when people chose a cam with performance in mind and not a name/hit a lick

Tooley was a genius when he named that cam because who would pick that cam just on the specs without the name.

[RDF1]

Tooley was a genius when he named that cam because who would pick that cam just on the specs without the name.

Smart marketing ploy.... I dont care for those cams. But all the cool kids love them cause they sound like a race car....

[cessnageek]

Thanks! Definitely runs better. Looks like you changed a bunch of stuff. Anything in particular major that helped it improve, or just a bunch of little things?

Here is a datalog of ~5 min warmup and ~8 min warm idle.