Throttle Decay

[69halfton]

Hi Guys,

First time posting, but have been cruising this forum for a while now trying to get a grasp on HPtuners software & tuning in general.

I've come up to a roadblock trying to work the bugs out of the tune on my LS3 in a 69 chevy truck. I've got the hanging throttle that many people have/had and cant seem to shake it. I've search for days, read post after post regarding this problem, and followed 5 liter eater's sticky on tuning idle, but still cant shake it.

I read on LS1 tech about a someone selling a mail order tune that has the "cure" for it... but no details on if his cure did or didn't work for the guy..

I have searched until I am blue in the face, and tried everything that people suggest with no success.

I have read that GM builds a throttle decay into their tunes for "emissions," but since this is a transplanted engine into a 1969 model year truck, there is obviously no emission requirements to be met. Further, there is no emissions equipment on the vehicle at all. I find it hard to believe that since GM programmed this into the ECM, we're stuck with it, even with the tuning software that's available to us.

I have attached my tune file, and a log file of the issue. In the log file, I ran ETC and pedal position. You can clearly see in between each shift the pedal drops strait to 19%, and that throttle position falls some, but then stops and seems to follow a decay rate from ~24% to 19%.. between shifts sometimes the throttle body doesn't even make it to 19%. I realize this is supposed to make it easier for shifts, but I'd like to be able to adjust this decay rate to speed it up some, as completing a shift smoothly takes too long due to waiting for rpms to drop.

Anyone know why this is happening, and how to disable/solve the issue?

[gmtech16450yz]

Totally fixable, don't worry I or somebody else on here will walk you through fixing it.

First step is to get a better layout file. You weren't logging pedal position and the throttle position PID's aren't the best either. Use these files and do some more logging. Post it here and I'll look at it and have a better idea on which way to get you headed.

Put these in their respective folders and open them in your scanner...


LS3 Channel gmtech.Channels.xml

LS3 Charts gmtech.Charts.xml

LS3 layout gmtech.Layout.xml

[69halfton]

Thanks for the confidence! I'm leaving in the morning for work which will keep me away from the truck for about two weeks. When I return I'll run some more logs and then post up my findings...

Thanks again!

[gmtech16450yz]

Thanks for the confidence! I'm leaving in the morning for work which will keep me away from the truck for about two weeks. When I return I'll run some more logs and then post up my findings...

Thanks again!

No problem! I am confident that you can fix it yourself with a little guidance.

I switched those layout files with some I had with better idle PIDs so if you downloaded them before this post, download the ones I just put up instead. I'll try to remember to check back here in a couple weeks.

[69halfton]

Hey There,

Finally back from work and ready to get back to some tuning..

Of the three files you posted, I can only open the one titled "layout"in the scanner.

once opened & logging, it appears to me that only some of the PIDs are actually logging data, and others remain blank.

Anyways, here is the log file that I ran using the only layout that would open. Is there something I'm missing here?

[NJ_Phil]

I'm also really interested in improving this....

[GHuggins]

I don't remember which side of the scale you have to raise, but under idle - proportional and integral airflow - you want it to take air out quicker by raising positive error areas. You can also change your spark coast down error table to help get the idle down... Throttle follower airflow max - put these back to stock so the follower tables should go back to helping too...

[NJ_Phil]

I don't remember which side of the scale you have to raise, but under idle - proportional and integral airflow - you want it to take air out quicker by raising positive error areas. You can also change your spark coast down error table to help get the idle down... Throttle follower airflow max - put these back to stock so the follower tables should go back to helping too...

I did half the entire proportional and integral airflow tables when I did idle tuning which was a big help, car idles at target +-25 RPM and is really nice for a healthy cam. Maybe I can adjust the entries above 64 RPM and lower than -64 RPM error to avoid messing up my idle band There are also "Normal" and "Coastdown" and maybe I only need to adjust the coastdown table to get the TB to match the pedal on decell.

Would love to have it decell as fast as it accells during throttle snaps. Guess first thing is get the TB to follow the pedal, then take some spark out with the coastdown table

EDIT: Looks like that would be "Idle Adaptive Spark Control -> Overspeed-> P/N" table... It's only pulling -10* max. The "Gear" table pulls -20* max

[69halfton]

Exactly..

I feel like playing with spark at this point is only masking the issue... what I'm really trying to accomplish is having the throttle body match the pedal position.

[NJ_Phil]

Exactly..

I feel like playing with spark at this point is only masking the issue... what I'm really trying to accomplish is having the throttle body match the pedal position.

I believe that's what Ghuggins suggestion about removing airflow from the table will do. From what I understand Idle airflow is the commanded airflow the ECM is told to run at during Idle. It will adjust the TB to reach that target airflow.
If the motor is at idle, then it only used the base running airflow....If the motor is decelerating TO idle, it uses the error airflows in the table to "get to" the base airflow. More error in the table when RPM is far from the target should command less airflow and not hang the TB

[gmtech16450yz]

It's like almost midnight but I have a tidbit that might help.

You have to get the "base" airflow and throttle follower settings as close as possible. You can make the error tables "stronger", but you're only giving it more reach to try to fix a bigger error. Reduce the errors instead of giving the error controllers more range. Make sense?

I put some layout files in this other thread. I STRONGLY suggest using them when you're trying to figure out idle tuning. It will make things easier to see.

https://www.hptuners.com/forum/showt...w-when-rolling

[NJ_Phil]

One final post for me since this is halfton's thread and I feel like I hijacked it, but we have the same problem. For both of us, it's not really about idle tuning, it's about the TB not following the pedal when decelerating.
throttle_decay.png


My thought was adjusting the sections of the proportional and integral airflow tables where RPM error was high, e.g > 256, more error would be introduced into the system and close the TB by commanding a lower airflow. Since table values for lower RPM errors were unchanged, idle characteristics would remain unchanged once it got closer to the target

[gmtech16450yz]

One final post for me since this is halfton's thread and I feel like I hijacked it, but we have the same problem. For both of us, it's not really about idle tuning, it's about the TB not following the pedal when decelerating.
throttle_decay.png


My thought was adjusting the sections of the proportional and integral airflow tables where RPM error was high, e.g > 256, more error would be introduced into the system and close the TB by commanding a lower airflow. Since table values for lower RPM errors were unchanged, idle characteristics would remain unchanged once it got closer to the target

I bunch "idle tuning" into that also. So scanning the things I'm talking about and seeing them in the scanner in a different manner will most definitely relate to pedal/throttle mapping throughout the rpm and throttle ranges, including decel. Like I said, increasing the error control isn't the best way to fix this stuff. Get the pedal/throttle mapping close enough that the error control can handle the errors properly.

Running out the door right now, I'll try to elaborate more but the first step is setting up the scanner.

[GHuggins]

I bunch "idle tuning" into that also. So scanning the things I'm talking about and seeing them in the scanner in a different manner will most definitely relate to pedal/throttle mapping throughout the rpm and throttle ranges, including decel. Like I said, increasing the error control isn't the best way to fix this stuff. Get the pedal/throttle mapping close enough that the error control can handle the errors properly.

Running out the door right now, I'll try to elaborate more but the first step is setting up the scanner.

Your referring to making the driver demand decel portions of the tables more negative correct? The throttle follower tables in the above posted tune are already really overly negative - way more than he should have needed - he just has the throttle follower tables out of the equation - I believe - with his max airflow settings... I've noticed you can raise the max air settings higher to make the throttle close quicker, but can cause other issues.

The errors and main settings all work together - you typically can't just adjust one table alone and get exactly what you want...

[69halfton]

I had already tried playing with the error tables trying to get it to react faster. The best result I had was by zero'ing out the Throttle Follower Airflow Max. When I Adjusted for RPM error I did get the throttle to close almost completely, but only under a long coast after the "decay timer" seemed to be up. I could not get the throttle to close promptly with the pedal.

gmtech - I managed to get your layout/chart/channel files all to work together (the first batch you reposted titled "LS3".

The log file is attached.

It seems from this that there isn't such a pronounced delay between pedal and throttle body, but the throttle position does stay open about 2% higher that at idle during a shift.

Thanks,

[gmtech16450yz]

Ok I'm here and I have some time. Let's see if we can get you fixed up, or at least get you going in the right direction.

First off, your channels don't have the PID's needed. Did you go to the channels list and click on "recent channel configs" and then "open channel config"? Do that and open the file I sent you that says it's the channel one. You're online now aren't you?

BTW, you can't have a log open when you go to load different configs or layouts.

[gmtech16450yz]

The next thing I noticed in your log is your idle timing is super low. That's going effect everything going forward, so I'd fix that FIRST. Get your idle timing to at least 20 or so, it's now below 10.

I'm actually going to have you change a bunch of stuff at once. I know it's not the best way to do it, but there are tables that you should get back on track first.
You actually don't have a whole bunch of tables changed so this shouldn't be that bad. First off, in the "Idle" tab, put the follower coast dec steps back to stock. Also put the base running airflow and throttle follower tables back to stock. In startup idle rpm, if the P/N table is how you want it, copy that table to the in gear table so they're the same.

Another BTW for all of you M6 guys. When you're trying to get back to basics or trying to figure issues out, I'd suggest you make sure all your gear dependent tables are the same for all gears and neutral. Since M6's don't have an actual gear position sensor, the ECM has to figure it out by looking at rpm, clutch position and vss. It's not always right. So what happens is you'll have something like the throttle follower values very different between say 2nd and 3 gear. As you downshift, the ECM is trying to figure out which gear you're in and it may use the 3rd gear values into neutral or 2nd gear, causing weird issues. Eliminate that possible situation by making the gear tables all the same.

[gmtech16450yz]

On the idle ignition timing... I said it's low. That's partly because you have too much idle airflow. Your throttle body is stock correct?

I just looked at your ign timing- in your "main spark advance", put that low octane table back to bone stock. Now take your high octane table, copy the whole thing and paste it into your "Idle spark advance" base and coastdown tables.

[69halfton]

Yes, Throttle Body, along with the rest of the engine internals are all stock. Only modifications are Headers, CAI, and no emissions equipment.

I have made all the changes you suggested. I will close & reopen the scanner and open each of the three LS3 files you had sent, and go do another log.

Should I be changing the idle airflow now to compliment the added timing?

[gmtech16450yz]

This is why I'm trying to get you guys to use these layout and channel configs...


Look at this spot in this log. The pedal goes to zero at right around 3300rpm. Now look down at the bottom of the scanner and you'll see a channel showing "Min- lb/h". That's the "Idle Minimum Airflow" PID. It's showing that the minimum at that moment is set to 87.66 lb/h. Remember that number.





Now look in the tune for this particular log. Look at the "Idle Airflow Minimum" table, in the 3300rpm area. It's showing around 86 lb/h. Now look at the multiplier tables, at that temp and baro, they happen to be both close to 1, so there isn't much modifying that minimum airflow value in the min idle airflow table. (There are a couple other modifiers, but don't worry about those right now.) Guess what? THEY MATCH!!!!! Holy cr@p doesn't that make figuring out how to make your pedal/throttle relationship easier?




That's the min airflow part of the picture. Now there's also the throttle follower that will play into this. Think of the throttle follower like it's a whip or a noodle. (lol, not sure that's an analogy that's gonna make sense but here goes!) The throttle follower is EXACTLY THAT, it "follows" the throttle movement. By changing these values you can get the throttle to "follow" slowly or quickly to and from it's desired angles. Kind of like swinging a wet noodle vs. a dry noodle. Very little throttle follower interaction would be like a dry noodle, it will track throttle movements up and down quickly. Putting in a lot of throttle follower authority will make it like a wet noodle, it will react to desired angles slowly. The table is in torque values, so you're basically telling the ecm how much torque you want it to "take away" or "keep ahead of". (Again, not sure I'm using the right words here.) So Jason, you had the right idea putting -100 in your throttle follower table. That's the right direction. The problem is your min airflow is too high, so it's only going to go down to that value quicker, which won't fix your problem.

So yes, you need to drop the min airflow table values way down. Go extreme, it will be easier to see what you're doing that way. And get that scanner showing those pids so you can see what's going on!