Throttle hang in between shifts, engine hesitant to decel (L83 in RX-8)

[shonuff53]

I've noticed different hang behavior based on VT settings. No tuner here, but I'd wonder if lowering the respective VT cells of that part of the map might help.

[Karmadillo]

I have been thru this so many times. It is the 2016 OS. It will not make power. Look at the tps opening versus pedal. I bet the tps never goes over 50 percent and no tuning can fix it. There is something odd with the os. I will help you fix it for free. PM me if you have hptuners

You are the only one who has indicated a solution. Would you please post your fix for Rev hangs when using a manual?

[Alvin]

I feel like clutch position is critical for these.


but uggh. Friends don't let friends do GEN 5 swaps... at least right now. So much more of a PITA than gen 3/4

[GHuggins]

I've noticed different hang behavior based on VT settings. No tuner here, but I'd wonder if lowering the respective VT cells of that part of the map might help.

[XDoubleDot]

Been following this one for awhile and was curious about the Virtual Torque suggestion (thanks btw). I have a 14' M7 corvette that has this hang in between shifts at higher rpm, very annoying for drivability. You can see in the log at 7:47:45 that the throttle plate is up around 30% when clutched in during a shift. The rpm falls, but not very fast.

Also pretty ignorant on the virtual torque table modifications (mine are stock). Would the suggestion be to raise the torque in the 100-200mg rows of the airmass table(s) in order for the car to "realize" is needs to shut the throttle plate more?


tune and example log attached. Car has headers and a few iterations of maf and map tuning done on it.
Throttle Plate Hanging on Shift.hpl
Edit 26 - Reduced Min Spark Table base and long term and copied 15 decreasing tap values.hpt

thanks for the help

[GHuggins]

Been following this one for awhile and was curious about the Virtual Torque suggestion (thanks btw). I have a 14' M7 corvette that has this hang in between shifts at higher rpm, very annoying for drivability. You can see in the log at 7:47:45 that the throttle plate is up around 30% when clutched in during a shift. The rpm falls, but not very fast.

Also pretty ignorant on the virtual torque table modifications (mine are stock). Would the suggestion be to raise the torque in the 100-200mg rows of the airmass table(s) in order for the car to "realize" is needs to shut the throttle plate more?


tune and example log attached. Car has headers and a few iterations of maf and map tuning done on it.
Throttle Plate Hanging on Shift.hpl
Edit 26 - Reduced Min Spark Table base and long term and copied 15 decreasing tap values.hpt

thanks for the help

Probably not that much help, but all I can say is play with it as I've seen year to year responding the opposite from one another. Bigger numbers does seem to be the way it responds better in most cases though.

[NvAzn]

For my rpm hang fix I had to adjust Torque Management > Driver Demand > Decreasing > (Immediate Normal table / Predicted Normal table) to -6,042.1 for all the the values. Then for Torque Management > Driver Demand > Increasing make 6,041.9 the value for all the tables in increasing.

[GHuggins]

For my rpm hang fix I had to adjust Torque Management > Driver Demand > Decreasing > (Immediate Normal table / Predicted Normal table) to -6,042.1 for all the the values. Then for Torque Management > Driver Demand > Increasing make 6,041.9 the value for all the tables in increasing.

I've found this to not always work and inevitably it always goes off of the torque model. I've had vehicles showing (-1300) or so desired and current torque only to still be "self" accelerating. Only way I've got them to do commanded was the torque model. Granted all of the tables work together and must be right, but the main torque controller is the torque model...

[XDoubleDot]

thanks for the info. I initially maxed the rate limits NvAzm, but I was getting incredibly inconsistent rev hang like characteristics that made shifting incredibly annoying. Setting them back to stock made part throttle driving at least seem consistent.

I messed with raising the torque 25nm in the 3800rpm+ and the 0-200mg rows. In the one log I took the throttle stayed open ~29% where previously it was only around 24% when the clutch was pressed in after a high throttle pull. that said the temperature outside was colder, so i'm not sure if that was a variable in this. I'll try subtracting some torque next time and see if that makes a difference.

Any suggestions on what magnitude of changes should I be experimenting with here? Car only has headers/exhaust.

[GHuggins]

thanks for the info. I initially maxed the rate limits NvAzm, but I was getting incredibly inconsistent rev hang like characteristics that made shifting incredibly annoying. Setting them back to stock made part throttle driving at least seem consistent.

I messed with raising the torque 25nm in the 3800rpm+ and the 0-200mg rows. In the one log I took the throttle stayed open ~29% where previously it was only around 24% when the clutch was pressed in after a high throttle pull. that said the temperature outside was colder, so i'm not sure if that was a variable in this. I'll try subtracting some torque next time and see if that makes a difference.

Any suggestions on what magnitude of changes should I be experimenting with here? Car only has headers/exhaust.

After looking at your tune, I think you may have just changed a lot of things - or someone has - that possibly didn't need changing and I know a lot of these things cause a lot of problems, so I just made some quick adjustments for you going back to a lot of 6.2 stock stuff with a slightly altered torque model. A lof of things in DI motors affect decel from spark to torque. Hopefully these will help you. Just copy everything over to yours via comparison log and give it a try. Then for some reason I recalled you saying yours was a manual car - any reason for the auto cal?

[XDoubleDot]

thanks so much. I brought my laptop to work so i'll give this a try on the way home. I'm really curious about your comment on "auto cal". I guess I just started with the tune that was on the car when I got it. It was 5 years old at the time and only had the exhaust and what looked like a quick MAF and spark table tune when I got it. I did a lot of compares between it and a '14 and '15 z51 manual tune I found on the repository and didn't see too many changes, at at least any indication that it was for an auto. would love to know how you identified that. thanks again for the help

[GHuggins]

Hmmm. I swear when I looked at it last night there was a auto transmission cal that opened up when I clicked on the trans tab - now it shows manual. I remember it was especially odd because under the auto/manual box it was selected as auto and it had the auto gear configs. Apparently it was just too late or I'm losing my mind as I just opened it back up and it's not there.... I did confirm the changes I made were still there and correct, but now I'm really confused as to the auto part....

[XDoubleDot]

ahh, yeah I haven't spent much time at all in there as I haven't bought the credits to reflash that module. I did probably deselect the requirement to have the clutch depressed to start the car, but to my knowledge it's not in effect on the car because I've never flashed that part on.

[XDoubleDot]

It was a bit unhappy lol. still hangs the throttle on some shifts and noticeably changed the throttle response, way more torque at lower pedal % which I think is from the driver demand changes, which stock file were you working off btw? seems a bit different than what I found. The gmve CL was off about 20-30% so I think the airflow changes in those areas might be hurting things.

Edit 26_1 log.hpl

[GHuggins]

I used a stock 17 SS camaro's file with a manual trans and then used one of my own torque models and some minor DD changes. I knew idle timing would be higher or at least it should have been however everything else should have been better. In the above log it doesn't even look like dfco is kicking in. Try disabling rate limits and going from there. I know there's no way your VE was right as it was all over the place and the VE on these makes some big differences.

This one has the rate limits and stock tm in it. Also has what should be a better VVE.

[XDoubleDot]

man these things make me wish for a throttle linkage sometimes haha. still seems to be doing it's own thing. I wasn't sure if you meant for me to max out the rate limits in there or not, I left them as is as I saw the request rate was disabled (had read that was just for neutral though?)

Edit 26_2 log.hpl

I've read people stating that the rate you push the clutch in etc has something to do with how fast/far the throttle plate closes. I've seen nothing to back that up, or a way to effect it. Right now it seems to be staying open in the mid to upper 20%'s (idle is ~5%) which is a problem because I typically let the rev's drop a little with the clutch engaged before grabbing the next gear, partially out of habit for the syncros.

has anyone else got a log where they can see the throttle plate % during high throttle/WOT shifts?

as for the vve, yeah what came with the car and what I had slightly tweaked sure doesn't match what i've seen in the gen 3/4 threads which all seem smooth. I'd read that it might look strange due to the way the zones were selected. That said i'm none the wiser.

[GHuggins]

Honestly that log looks better, but for what I now believe you are after - well I'm not sure with the current tables that it will be achievable without working around the problem. The problem being a "timer" setting if you will for changing from axle torque to cfco torque reference. We can manipulate it by changing the torque model and your DD "axle torque reference table", but inevitably the problem is the "built in" timer setting... This one has the DD more negative in the affected areas. You can even play with idle and timing settings more to get that "exhaust brake" feel you might be after. Hopefully someone on here knows how to make it transition quicker.

[XDoubleDot]

Weather and work isn't lining up with me getting a chance to drive the car and test this latest one out. The Idle -> Torque -> "External Load" table seems really promising. seems like making the values in the higher rpm areas more negative (like you did) might be what I'm looking for. The fact that it's oil temp based might explain some of the inconsistencies i've experienced.

regarding setting the cam timing back to the stock values in the cruise areas. I had tweaked those (in the cruise area) from stock to deal with what felt like a deadspot. As these are below the rpm and airmass that i'm seeing the rev hang, do you see anything issue resetting these back to how i had them?

thanks again for the help and I'll report back what I find

[GHuggins]

OK, so from everything I've seen and tested there are certain things you can do to make them more aggressive on deceleration.

Idle - Torque - external load - making them more negative into the rpm zones you want it to decel - yes this is actually a big contributor
Idle - Torque - Torque reserve - Lowering these - BUT you can only lower so much and may have their own negatives
Idle - raising the immediate and proportional idle controls - may cause their own problems and why I don't bother with

Driver Demand - making more negative in the kph cells with 0 pedal - did this in yours

Torque model - "usually" raising the lowest or most negative row - don't do much though and you'll have to tweak just before idle so you don't have stalling issues - "my torque model" above shows this, but you need to change to your idle areas as that one is for a cammed motor

Throttle follower under torque model - you can raise, but honestly if everything else is right you shouldn't have to.

[XDoubleDot]

So I tested out 26_3 and again the revs hung high. additionally the driveability was noticeably worse, I suspect this is due to the vve being further off than my existing tune (basing this on the gmve closed loop adjustment table, lots of average corrections in the 20s). question on this at the end

I went back to the original "edit 26" tune I had. I was messing around with the ARM and noticed the upshifts felt a lot better as in the revs dropped faster and the transmission took the next gear with a lot less effort, I could hear the car make that "pop" that reminds me of a fuel cut on high throttle shifts. The problem is I can't for the life of me find anything in the log that backs this up haha. half way thru the "log with ARM" i turn it on. on the high rpm shifts I'm not seeing anything different with requested torque, "fuel cut" timing, or spark that aligns with what I noticed driving. Next logs I'll have fuel injector pulse width to get a better idea on when fuel cut is actually happening, right now i'm relying on the O2 sensor with all the delay.

Edit 26 log with ARM.hpl

Edit 26_3 log.hpl

if you're wondering why I don't just drive with the ARM activated, it's because the downshift rev matching is horrendous. for 1-3 seconds after i've let the clutch out on a downshift, the car pulls forward as if my foot is on the gas, super annoying in most senarios. My assumption is that I need a better vve tune, which leads me to my last question....

why does my vve look so different? My understanding was that the gen5 stuff would look choppy due to the breakpoints. below is what I have from the repository which is very similar to mine, aside from some slight changes I made to mine in the lower rpm range from the "gmve parameter".

2015 z51 m7 repository virtual VVE.JPG