Maggied G8 GT Cold Startup Stalling Issue

[Aaron407]

Hi all,

I've been banging my head against the wall trying to deal with a cold start issue on my supercharged 2009 G8 GT. It has a TVS 1900 kit that I installed in the spring with a 40A boost a pump, TR6 plugs, long tube headers, high flow cats, CAI, LS9 injectors, LS9 MAP sensor, 2.8" pulley (during the summer, with a 3.3" for the winter), and a wideband. I've tuned it extensively for performance and driveability, it starts/runs absolutely beautifully during the summer, but has a cold start issue in the winter that I can't seem to track down and rectify. I'm hoping somebody might be able to offer some further suggestions.

The problem:

After the car sits for some time and I try to start it on cold days (generally near or below the freezing mark), it will crank and fire up just fine, but will either stall or nearly stall after the RPM drops down after the crank. There is no issue starting in warm weather at any time, and it seems to be related specifically to the cold. If it does stay alive, it will run rough for about ten seconds before somewhat smoothing out.

I had installed the BAP about two weeks before I installed the supercharger in the spring, and left it in bypass mode at the time. I don't recall having any issues with starting it during those two weeks, so I'm thinking it's not related to the problem. After installing the supercharger, I wired the BAP with a 2 psi Hobbs switch. At no point prior to the supercharger install (when the tune was still stock) do I recall ever having this type of stalling problem, so I'm fairly confident that some part of the install/tune has created the issue.

Tuning info:

1) I started with the supplied Magnuson tune, but modified it for the LS9 injectors and MAP sensor based on information gathered here, as well as a few other tweaks that would have nothing to do with startup (e.g. transmission torque management).
2) The car is running MAF only with a proper MAF tune.
3) The fueling is dialed in by wideband and seems just fine in normal operation.
4) According to my wideband, I have no shortage of fuel when running WOT, which suggests that the fuel pump seems ok, and it started fine before the install.
5) I've tried increasing the final idle airflow minimum table for lower RPM values in park in hopes that it would stop the RPM from dropping and stalling how it does, but it didn't help.
6) I've increased the startup airflows for cold temperatures, which had no effect on the stalling (only provided stronger cranking).
7) I extended the startup underspeed/overspeed times in cold temperatures to extend the cranking mode beyond the time where it would normally stall, but it had no positive effect on the issue.
8) I've tried using a range of open loop equivalence ratio tables in thinking that it might be an issue with the crank to run mode transition for AFR. This seemed to have no effect.
9) I increased the priming time for the fuel pump from 2 to 3 seconds. No change (as expected).
10) I increased my idle speed from 850 RPM to 1000 RPM in hopes that it wouldn't drop in RPM as much after cranking and might have a better chance to stay alive. No change in the stalling issue.

Other info:

1) I've confirmed that there are no vacuum leaks.
2) Normal operation shows fuel trims very close between the two banks, and the wideband shows proper stoich levels. This suggests to me that the O2 sensors are ok.
3) Putting the stock plugs back in had no effect.
4) I installed a new battery about a week ago. It was time for one anyway, but ruled out voltage issues.
5) MAF sensor is clean, and was replaced during the summer.
6) There are no DTCs stored.
7) My current tune is attached.

Does anyone have any suggestions of what I should look for in the tune? Since the problem didn't exist prior to installing the supercharger with supporting mods and the modified tune, and fueling seems fine while driving, I'm leaning towards it not being a fuel supply issue, and probably something hidden in the tune. As well, I'm fairly sure that it started fine with the BAP for the two weeks before I installed the supercharger kit. I haven't measured pressure at the rail yet, but I will just to rule it out if there are no tune suggestions.

I'm quite willing to run a scan on startup if it would help. I logged one in the spring and sifted through it in the past, but didn't see anything strange other than the RPM drop. Unfortunately, I didn't save the scan. If anyone can suggest what to scan for that might not be usual, let me know.

Any help would be greatly appreciated. Living in Saskatchewan doesn't afford me the option of avoiding freezing temperatures in the winter, so I'm dying to get this figured out.

[truckshop]

my old tbss used to do this when it was cold out. had to change the ect multi table. had to reduce mine at the colder temps.

[Aaron407]

You're referring to the ECT multiplier table for the minimum airflow table? I remember asking someone about that, but was told that it's not something worth adjusting.

Also, I'm going to try to snag a scan whenever I can since I know it will likely be key in figuring this out. I need to rewire my wideband so it doesn't cut out when I crank then engine, though.

[truckshop]

yes. on cold start mine would hunt for idle and sometimes kill. it was commanding too much airflow. pcm would pull all the timing to get the idle to come down and then it would start idle hunting. if it was below 30 degrees out it would kill.

[Aaron407]

Thanks. Maybe my attempts to increase startup airflow actually had the opposite effect than was intended. I'm going to put it back to stock and check out the ECT table, and maybe try tweaking the startup flare control timing.

[SultanHassanMasTuning]

something that will help a bit on start up is increase your

STARTUP IDLE RPM add say 150 rpms and give it a try.

as mentioned if you cold start (airflow) is too high your adaptive spark will retard too much and cause a shutoff

edit:

i see that your going for PURE maf mode, but to help stablize idle i would have a hybrid say upto 600/500 under dynamic

[Aaron407]

Thanks. Unfortunately, it looks like I don't have a startup idle speed table to change in my ECM. I had increased my cold temperature target idle speed cells by 150 previously, though, which I presume is as close as I'll get to controlling the trouble areas.

As for going MAF only versus hybrid for stabilizing idle, I've considered that as well. However, I've been hesitant to do it since I've nailed down a tune where I can simply swap supercharger pulleys for the seasons, and I'm thinking that the pulley might affect the effective volumetric efficiency. I can't say I'm sure of that, though.

[gmtech16450yz]

Do you have a log of the startup? I see a bunch of stuff in your tune that would cause your problems. I'll try to take some screenshots of changes for you to make that should fix it, or at least get you in the right direction.


Ok try these changes in your tune.They're kinda crude obviously and since I don't have any logs to go by I have no idea what the ign timing or airflow actually look like. BTW, you can request any idle speeds you want, if the airflow and ign timing isn't right, you'll never get those requested speeds. (Sorry if that's old news, just wanted to make sure.)

Most of the changes I made were in 10% increases. They may actually be too far, but I usually like it when that happens because then not only do you know you're in the right direction, you can adjust it down easily by comparing the changes and dropping 50% of the change or whatever.






Multiply these two tables by 10. These will control how long your fast idle stays on for. A lot of times it's too short and it goes from startup airflow to running airflow and dies. This will keep a fast idle much longer when first started.

[Aaron407]

I don't have a proper scan yet, unfortunately, and probably won't be able to snag one with my wideband rewired for a little while yet. I'm intrigued by what you might have found that could cause the issue, particularly since most things related to startup are in stock form (with the exception of minimum airflow). I'm all ears.

[gmtech16450yz]

The wideband isn't really a factor here, I wouldn't worry about that for right now. Just get some logs of the complete start so we can get an idea of if it needs more air or spark.

[Aaron407]

Wow, thanks for the info! I had extended the startup times similar to how was noted, but it didn't seem to fix the issue. The problem might be buried in some of the other table changes that you have specified, though.

For whatever reason, I can't find the 'Startup Idle Speed' tables anywhere in the tabs in v2.24.260, and I can't seem to find a newer version. Thoughts?

[gmtech16450yz]

(It's on the left under "Base Idle RPM".)

Edit again... I just realized you mentioned freezing temps where you live. My changes in those tables were mostly down to 40 degrees or so. You might want to carry the changes I made further down to freezing, and make sure they're smooth, I also noticed I had a jump up in rpm between 32 and 44 degrees at one point. You want to make sure the changes are all going the right directions, if that makes sense? Like you wouldn't want the requested rpm to go from 900 at 32 degrees, then 800 at 44 degrees, then 850 at 56 degrees. All table changes should be smooth and make sense.

I'm at 2.24.888. If you log into your customer center page on the HPTuners main site you can download the latest version.

Kinda funny, I'm helping another guy through emails that is having pretty much the same issues as you are. He has a GMC Sierra with a 6.2 and a Maggie also. I took the easy way out and sent him the link to this thread so I can help two guys at once! lol. I'm not sure if he's a member here but I'm sure he'll be reading this. Say HI to Aaron, Paul! Hopefully I can get both of you guys fixed up! Like I told Paul via email, don't be afraid to experiment with these tables, you can't blow anything up and making mistakes will almost always teach you something about what the tables are doing. I also told him to make sure his throttle plate and throttle body are clean Make sure yours is too.

Have fun guys! I'll check back later. Don't be afraid to ask questions on stuff you can't figure out, I'll try to answer as simply and clearly as I can. The "Throttle Follower" tables are VERY powerful with this kind of airflow stuff, and are tables that not many people really understand. They don't just effect the throttle when it's moving, they also effect it when it's at idle. So many "tuners" are good with the WOT stuff, which is really pretty easy, but it's the drivability stuff that is really important and the most misunderstood.

[gmtech16450yz]

Wow, thanks for the info! I had extended the startup times similar to how was noted, but it didn't seem to fix the issue. The problem might be buried in some of the other table changes that you have specified, though.

For whatever reason, I can't find the 'Startup Idle Speed' tables anywhere in the tabs in v2.24.260, and I can't seem to find a newer version. Thoughts?

"Idle Startup Airflow vs. ERT vs. ECT" is what dictates the actual values. The "Idle Underspeed and Overspeed Startup Airflow Time" is what determines how long those airflow values are held. If the airflow values are too low, extending the airflow time won't do a thing. Also, it's hard to know how much to raise or lower a table until you experiment. I raised the startup airflow time table by 10 TIMES. That's fine, it just means the startup airflow will be held for like 12 seconds instead of 1.2 seconds. You can see by raising that table 10%, it would have almost no effect. If you raise other tables by 10%, it could be a HUGE change, you just have to experiment or try to figure out how much change to try by reading the descriptions of the particular tables.

[Aaron407]

Thanks for all of the info, this is great. I had previously extended the times to 8 or 10 seconds as well, which acted as expected, but it still largely had the stalling issue.

I haven't yet updated VCM Suite, but the changes (minus the startup idle speed since it's not there in my current version) seem to have already made a big difference when I started it up today. I'll update and adjust the startup idle speeds, then take it from there in terms of fine tuning.

Thanks again for the help. This has plagued me for a long time, and I'm thrilled to see it being fixed up. I shouldn't have waited so long to post!

[Aaron407]

I found the tables after updating to 2.24.965. It took a bit of thought to make sure everything transitioned properly from the startup idle speed tables to the target idle speed table after the startup mode time elapsed. Unless I'm misunderstanding things, I set the startup time to 10 for both under and overspeed across the board so it should reference only the first two rows in the startup idle speed tables before transitioning to the target idle speed table. It was a bit tricky to make sure that it transitioned down smoothly without major step drops, but I found something that should work. One question that I have, though, is about what is referenced for startup. After the initial crank, is the startup airflow table used, or is the startup idle speed table used to control the throttle if still within the time specified in the startup under/overspeed time tables?

[gmtech16450yz]

Ok, so let me see how well, or how badly I can explain startup and idle control. lol.

The tables that have actual rpm values in them are ONLY desired values. You could put an idle speed of 9000rpm in there if you want, if it doesn't have the airflow and ign timing to obtain that, it's not going to. It's always been a confusing concept to grasp, you'd think "Ok, I want to raise my idle speed from 550 to 750". So you change all the idle speed tables from 550 to 750. You start the car after loading the tune and guess what? It's still idling at 550. Dammit!

So here's what the computer tries to do when it sees you're requesting 750rpm instead of 550rpm. It will try to raise the airflow (open the throttle more) and/or raise the ign timing (which raises the rpm). THIS is when those airflow and idle timing tables come into play. The airflow tables are kind of a target value, or a value that the ECM THINKS will get the idle to what is desired (750). It will try those base values first. If the idle speed is too slow still, then it looks at the adaptive idle tables to see how much more airflow it is allowed to give. If those tables don't have enough headroom to give it the airflow it needs, then it looks at timing.

(I actually don't know which the ECM looks at FIRST, I'm assuming it's airflow, then ign timing. I think the truth probably is that it looks at both at the same time essentially. Airflow will always be a slower, more coarse adjustment and ign timing will be a much faster, finer adjustment.) So let's say it's given it all the airflow it can, and the idle speed is still below the desired 750. The only other thing it can do is to raise the ign timing to try to meet the desired idle speed. Adaptive Idle Advance is a great PID to look at in your logs to see what ign timing is doing in relation to controlling idle speed. If you don't have enough airflow in your tune for the desired idle speed, the ign timing will be much higher than normal. If you have too much airflow, the ign timing will be way more retarded than normal. Again, there are a bunch of tables that tell the ECM how much, how fast, when, what kind of delay, etc that it can advance or retard the timing to control the idle. Where it gets fun is when you want a specific idle ign timing value and you have to change the airflow tables to get it!

So the short answer to your question is "All of the above". lol. It uses all of those tables to control the idle speed. The tables themselves determine how long or how much each control does, in addition to a few dozen other variable settings. If you have an idle that drops too fast but doesn't raise fast enough to compensate for P/S loads or whatever, there are tables to control that. You can control so freakin much in these ECM's there's pretty much no issues that can't be fixed if you know where to tweak the values. Just be thankful you're not working on a Bosch E69 ECM where there ARE NO "Idle speed" tables like this. It's all in "Desired Air Load" or "Desired Torque" or "Optimum Spark".

Hope that helped and made a little bit of sense.

[Aaron407]

That actually made a lot of sense. It first uses the startup airflow values. If it doesn't reach the target RPM value, it varies the airflow within the limits of the adaptive idle airflow values (which I haven't found yet, but haven't looked terribly hard). If it hits the limits in the table and still hasn't reached the desired idle speed, it starts playing with timing via the adaptive idle timing tables to try to further control the idle speed.

I do have a question about the throttle follower changes. As noted in my original post, the problem was that the RPM would fall off a cliff after cranking and result in a stall, so I was looking to slow down the RPM drop. In looking elsewhere, people who have had the RPM drop down too quickly after startup were told to reduce the throttle follower table values to slow down the "descent", while I increased it as suggested. Am I looking at something incorrectly to think that sounds conflicting? I'm more than eager to learn, but information that seemingly contradicts stops me dead in my tracks, so hopefully you can put my mind at ease.

Thanks again for taking so much time to help me with this, I'm learning a lot.

EDIT: Maybe people were being told to take away from a decay table rather than a torque table? It was rather ambiguous in the other threads that I read.

[gmtech16450yz]

That actually made a lot of sense. It first uses the startup airflow values. If it doesn't reach the target RPM value, it varies the airflow within the limits of the adaptive idle airflow values (which I haven't found yet, but haven't looked terribly hard). If it hits the limits in the table and still hasn't reached the desired idle speed, it starts playing with timing via the adaptive idle timing tables to try to further control the idle speed.

I do have a question about the throttle follower changes. As noted in my original post, the problem was that the RPM would fall off a cliff after cranking and result in a stall, so I was looking to slow down the RPM drop. In looking elsewhere, people who have had the RPM drop down too quickly after startup were told to reduce the throttle follower table values to slow down the "descent", while I increased it as suggested. Am I looking at something incorrectly to think that sounds conflicting? I'm more than eager to learn, but information that seemingly contradicts stops me dead in my tracks, so hopefully you can put my mind at ease.

Thanks again for taking so much time to help me with this, I'm learning a lot.

EDIT: Maybe people were being told to take away from a decay table rather than a torque table? It was rather ambiguous in the other threads that I read.

YAAYYYY! You're getting it even if you think you aren't!

You're theory is exactly right, or at least what you put in the edit was. When you raise the values in the Throttle Follower Torque table, it basically makes the throttle plate open more. Reducing the numbers into the negative keeps the throttle plate closely following the pedal position. You can imagine why it would need a table like this. If you're at WOT and you quickly take your foot off the pedal, the plate would slam shut immediately. The Step Up and Step Down tables are the speed at which the torque table values are changed. In THOSE tables, the lower the numbers, the slower the reaction or rate of change. So yes, like you thought or had heard, if you want the throttle to slowly close, lower the numbers. But again, if the values in the torque table aren't right, slowing down the rate won't do a thing. Kinda like making a car that's sitting still have twice the power. It's still sitting still so the power increase does nothing.


edit, BTW, don't worry about the throttle follower step up and step down tables. Those are more for fine tuning the throttle follower. Raising the values in the main torque table will make the big changes. Like I've said, if you want to see how the table works, try putting big changes in like -100 or so in the rpm cells 2000 and 2500. Go drive it and you'll see the throttle response will be held back in that rpm range. In your situation, raise the values below 1000rpm a ton and get it to the point where the idle will actually hang at 1000rpm. Again, don't be afraid to experiment. That's when I make my biggest discoveries, or mistakes!

[Aaron407]

Wow, you have been ridiculously helpful. I think I'm understanding it well enough for now. It just needs to sink in a bit more and I'll have to experiment to dial it in. Thanks again!

[gmtech16450yz]

No problem! Glad to help. Hopefully the other guy (Paul) is reading all of this and learning how to fix his.