Challenger Scatapack, Throttle delay

[bluegoat06]

My latest tune, best throttle response so far

[smokeshow]

The throttle sonic flow model is really only for diagnostics and in case of a MAP sensor failure can provide airflow estimation. The ECM is constantly calculating a manifold pressure independent of the measured reading from the sensor. If they differ by a little bit, the model is corrected. If they differ by a lot, the logic assumes the MAP sensor measurement is no longer reliable and defaults to calculated MAP. I have yet to look at any of your files...I always forget. It still sounds like buck...make sure to compare driver torque request to engine torque request below 3000rpms. This will tell you if something is shaping/damping the torque request.

[bluegoat06]

You can see the delay here in full action, i have my foot all the way to the floor, look at the expected torque, but the car doesn't do anything until later, and still, she doesn't roast the tires once the power is delivered to the wheels
throttle_delay2 by Wrangler RSM, on Flickr

[smokeshow]

Definitely looks like a buck routine. Unfortunately I can't see any of those knobs available in HPT. I'll look a little deeper when I get a minute. In the mean time when you get data again, change the raster rate in your scanner for throttle and accelerator pedal position. Looks like they are 200ms right now, too slow to get meaningful data from. Do the same for timing advance, actual spark and MAP. Speed up everything as much as you can, down to 20ms or better.

There may be some less than optimal tricks to pull here to make this issue less severe or perhaps go away entirely, but without changing the right calibrations you may lose functionality elsewhere. For example, if you fooled the torque calculation into thinking the spark advance had a much greater effect on torque output than it does, it would pull less timing when buck activates and deliver torque faster. However this would waterfall into other areas, like idle spark correction and transient load corrections. You can 'scale' them so to speak, but it would be a task no doubt.

[bluegoat06]

Smoke, i'm not sure how to change the rate in the scanner, what rate is this ?
thanks again for looking into this.

[smokeshow]

I think it's called 'measure rate' in the scanner. Right click on the variable in the PID list and look for that. You'll need data refreshed at a higher rate to pick out trends and figure out exactly which feature is managing throttle position.

[bluegoat06]

Thanks again Smokeshow, i couldn't find that pid in the scanner, maybe something that may changed in the last beta ? i don't even know what this could be ?

I'll try another example, this is a 2d gear WOT quick test, Im logging every single throttle and accelerator pedal pid i see now, trying to correlate the behavior when the pedal goes to the metal. as well as other torque PID's that are new in the beta, btw, i just found that mode#3 is the one that my ca goes when i'm switching from Normal to Sport, not mode#2 as a i thought initially.

In any case, the marker here denotes the moment the pedal is all the way down, torque request goes to 457ft/lbs, notice every other throttle or acceleration pdeal doesn't show anything related to WOT, except for one, throttle position sensor 2, that one shows 3.43V, which at least is the closest thing i can see that could indicate i'm giving gas to the floor, i don't see anything that relates to TPS 2 in the calibration though..

2ndGear_WOT_test by Wrangler RSM, on Flickr

[smokeshow]

TPS 2 is just the inverse of TPS used for rationality purposes. Don't pay attention to TPS 2, as the WOT pedal threshold will be looking at the non-inverted TPS.

Change this for raster/measure/polling rate...

[bluegoat06]

Thanks Smoke, Now i see exactly what you're referring to.
I'll change that raster rate for throttle and accelerator pedal position, timing advance, actual spark and MAP as you indicated in post#24
Can i change only this PID's to 20ms while leaving the others to 200ms ? I've never had to mess witn this before, i'm sure you can tell

[bluegoat06]

Smoke, i have some new logs now with the new raster rate, thanks a million for this.
No load vs load here indicates whether or not i'm under throttle, before giving her gas.
The biggest delay seems to happen while my foot is out of the throttle, then quickly press it, regardless of how quick, of i press 1" down, or all the way down.
If i have load in the engine, that is, i'm riding on the gas to keep speed or RPM, and then i give it more, the change is pretyt much instataneous, though power delivery at WOT seems to be just too smooth, and i gues this is the reason why the tires doesn't get overpowered from a 1st gear roll

This is a 5th gear with no load under the thottle, then i do a quick stab on the gas, as i'm trying to accelerate quicker, i'm measruing the markers as the expected torque vs actual initially peak, Green and purple lines
5thgear_noload_to_quickstab by Wrangler RSM, on Flickr

07:11:54:139
07:11:54:695
diff 556

Same thing, but now engine is loaded, pressing on the throttle, then quickly giving her more gas
5thgear_load_to_quickstab by Wrangler RSM, on Flickr

07:06:24:452
07:06:24:610
diff 158


Now 1st gear WOT
1stWOT_load_stab by Wrangler RSM, on Flickr

03:24:36:747
03:24:36:904
diff 157

[GHuggins]

Smoke - any chance you can make the changes one at a time to a stock cal using your diablo setup, then read it out with hpt and then send the cals to hpt requesting the missing tables? This is what we've been doing to the GM side to get all of the missing tables added in... It's time consuming - believe me, but will help everyone a lot...

[bkg06gto]

Mine does the exact same thing and i have fought it tooth and nail with nothing working. From a stop and a WOT jab mine will say epected 368 actual 89 then at about 2000rpms its only like 10-15 off and the car feels like it has power. Mines a RT so wonder if its something to do with that. hopefully you figure something out because it does indeed suck

[smokeshow]

Yep, still sounds like buck lol. The response is that way because during off pedal/decel situations, the engine is slowing the car down. In order to accelerate, you have to cross gear lash in several locations so a fast torque transient could feel really nasty to the average person. The other thing you're feeling in 1st gear where torque application feels too smooth could be fixed by setting all of the ESP bits to 'disabled'. So don't use spark, throttle, or fuel cut to control torque when traction is limited. You lose ESP entirely, but it won't reduce low gear torque. I'll have to look at your latest data to be sure, I will this weekend if I remember.

GHuggins...I don't have a diablo tuner, but that sounds like an interesting way to do it. Where is this happening at, and are they taking requests? Lol...there are some knobs on my Colorado that I wish I could turn. I know they're there, but HPT doesn't have them in the definitions.

[bluegoat06]

Thanks for the input Smoke, i've been doing more testing today before i had the time to read you post, i still have the ESP portion enable as i tought if i wanted to disbled it, i'll just use the bottom, but i can definetely try to disabled everything in the ESP and see what changes.

This is my latest tune, i'll go ahead and disable ESP next and see what's up

[bluegoat06]

K, i just disabled the ESP and went out for a spin, zero difference whatsoever, since it didn't make any difference, i just re-enabled my ESP stuff. i'd rather have it at the press of a buttom.

You metioned a fast torque transient perhaps being at play, and i see the scanner can now log a "Fast indicator fast path" and "Torque identifier slow path", however i don't see anything in the editor that will allow to change this. at least nothing with that name. so perhaps HPT is working on it already ?

I've also asked in challenger foums, there is a lot of pople having the tip-in trottle delay issue, some people say they can roast the tires in 1st gear punch, some more say they can't, just like me, one said he used to go sideways and everything, but then Uconnect did an upgrade, and then he lost Sport mode, and supposedly after they fixed that, he's saying the car does't do it anymore, so finding commonalities to this problem might an issue.

I've changed the torque model tables and used one form a 2016 Stock Challenger scat i found here, made no difference whatsoever
Also adjusted the ESP min torque from 15 to 0, then to -15. from feeling, i'll leave it set at 0, so far from everything else i've tried, it's the only thing that i feel has made a difference this morning, but as the car got hotter, the behavior started changing.

Map peaking out at 50% throttle, the hotter the car got, the higher the TPS was moving to make MAP peak, if it makes sense
For example, left the car idling for 3-4 min after a cold start this morning, then took her for a spin

MAP_good by Wrangler RSM, on Flickr

stopped to save the log, drover another min or so, then tried to repeat the same thottle stab
map_bad by Wrangler RSM, on Flickr

Don't know if this behavior is normal with the map or not, or if i'm staring at the issue, without knowing why it happening or how to fix it , but it's the only thing diffent that i can see after changing crap, and trying a bunch of other changes and suggestions.

[smokeshow]

When I said fast torque transient, I meant quick/brief. Things like fast indicated fast path mean something different in Dodge-land. Good thinking outside the box with the MAP/TPS correlation, however I wouldn't put too much stock in that without looking at other variables like RPM and cam position. Also, what is your MAP raster rate set to? It looks far too blocky in the data to be believable. You'll also notice your pressure ratio and aircharge number spikes up before MAP does, which isn't realistic. If you're running out of CAN bandwidth, get rid of some of the non-essentials...for example active exhaust position and any other redundant PIDs that aren't followed by (SAE).

To be honest I don't think the knobs to fix this are available for turning in HPT. In the end you may end up having to sort of bandaid-fix it.

[bluegoat06]

When I said fast torque transient, I meant quick/brief. Things like fast indicated fast path mean something different in Dodge-land. Good thinking outside the box with the MAP/TPS correlation, however I wouldn't put too much stock in that without looking at other variables like RPM and cam position. Also, what is your MAP raster rate set to? It looks far too blocky in the data to be believable. You'll also notice your pressure ratio and aircharge number spikes up before MAP does, which isn't realistic. If you're running out of CAN bandwidth, get rid of some of the non-essentials...for example active exhaust position and any other redundant PIDs that aren't followed by (SAE).

To be honest I don't think the knobs to fix this are available for turning in HPT. In the end you may end up having to sort of bandaid-fix it.

I only set the Intake Manifold Absolute pressure set to 20ms, i think that ill be the problem, i should have used manifold absolute pressure instead, i'll remove some of those PID's and give it try one more time.

[GHuggins]

Smoke, you request tables to be added through support... It's better if you have a tuner to change them with and then request with each change, so they can see where it's at, but you can request via names of the tables too - just may take a while to do it that way unless you can get one of the Dodge HPT engineers involved...

[bluegoat06]

Smoke, something in the raster rate seem to mess with how the data is displayed ? in any case, here are some logs with less PID's and the change to the raster rate in the manifold absolute pressure pid,
And then 2 more with the default channel rates, hopefully there will be something you can see either way.

[bluegoat06]

Finally made progress on the WOT power delivery, and i'm able to feel the tires moving out when i punch her from 1st gear, don't know why it didn't occur to me before to use the power request % mode tables above 1V, i was modifying those values only up to 1V, thinking the delay from no throttle to throttle was also the cause of the relatively smooth power delivery, in any case, i changed the power request mode 2 and 3, which are the ones my car use for Sport mode, making the tables a little more aggresive above 1V, also did a small modification on the knock sensors, i have 10% above all the KS tables, then did a comparo with the 2016 Challenger, and matched the higher values on the 2016 stock Challenger scat with mine, mostly looking to decreae my KR a little, as i'm running 2* less timing compared to the stock, and still i see KR in the 3600rpm a lot of times in 1st gear, which should be BS, also someting else i think helped was the change in ESP min TM from 15 to 0.

I'm attaching a couple of logs, one in Sport and one with ESP off, sport will bark them good now in 1st gear, so i don't neccesasrily have to have ESP off.