Changing the Drive-By-Wire throttle map

[Cole]

Originally Posted by D_Thissen

@drewbot and @Cole do you have any problems with you a/t?

Nope, I just put @Wayno throttle stuff into my map and went merrily on my way.

I think what we have here is finding a solution for a problem that doesn't exist.

[aagun]

dyno judgment

[renfield90]

OP is offended that his sports car with an automatic can't...whoa, stop right there. We found the problem.

If you're not happy with power output in higher gears, and you think you have too much power in lower gears, then select a gear in between. I don't like to bash slushboxes since my first sporty car had one, but you are trying to apply a tourniquet when a bandaid will do.

You have a manual mode, I suggest you use it. This is why people opt for a stick, so they can select the correct gear and throttle input for any situation.

[Lantana frs]

Perception becomes reality. I have cpb and oft and my ass tells me that they vastly improved part throttle response/performance. Unfortunately the dynos dont show that.

[PetrolioBenzina]

There is a setting that is hidden in the manual that allows you to operate the gear shift manually even in the auto, but that mode requires paying attention to whats going on around you. Give that a shot before you reinvent the OFT tunes. And welcome to 2014!

[PetrolioBenzina]

One expert has checked in with an opinion, OP fires back. Waiting for the other expert as well. Perhaps even Shiv will check in.

[Wayno]

Originally Posted by PetrolioBenzina

One expert has checked in with an opinion, OP fires back. Waiting for the other expert as well. Perhaps even Shiv will check in.

Shiv's never had an automatic development car. The only tuner to advertise that they have actively tried to improve the automatic shifting is Moto East.

For all the precise tuning involved, the coupling between the engine and the 6AT is still incredibly mechanical. Up shifts are accompanied by an ign cut and down shifts by a throttle blip. That stuff is static (or undefined) in the tune. That's why it's so sensitive to things like any intake mods at all. The TCU takes 0.3s+-x to shift and the engine is tuned to take EXACTLY 0.3s+-x to change rpm also. You make the engine "more responsive" or whatever and it becomes out of sync during shifts.

Pretty much the only thing that you should be ensuring is that after the TCU shifts in auto mode, the cell one load column higher and 800 rpm or so down has more power than the one before the shift occurs. When it has less power, that's when it feels like a turd.

The trouble is all the power mods we do start adding power around say 2400 rpm for example, so when you up shift from 2800 above where the extra power starts down to 2000 rpm or so, it doesn't feel correct. At higher rpm's the problem doesn't exist because the engine/intake has a lot more momentum. Also, this problem doesn't exist on E85 because that 2000 rpm region has lots more power, and probably the high octane US and EU blends too. Yet another problem could simply be that the OFT tunes have removed too much timing in certain load columns below 2200 rpm.

I'd happily support anyone who tries to fix this problem. The trouble with koolbrz is rather than identifying and fixing problems, he always has to start from scratch with some brand new scheme and introduce 100 new issues, before ending up back to the drawing board a month later.

It might be solved as easily as adding requested torque at 1800-2200 in the exact problem load columns, I don't know. You would only want to add as little as required, so the car doesn't think it needs to down shift. Or it might be a case of adding requested torque in the cells before the up shift, so the TCU holds on to the gear for longer. Or it might be better to not be changing the throttle map at all, but the calculated torque maps as kodename has said. In that case then you need to understand exactly how those work. Either way making heavy handed blanket changes is only going to waste your time.

With OFT it's easier to replicate specific loads because you have a screen. It's harder with tactrix. Pretty much you can only try to replicate while logging, then if you hit a bad shift, stop logging and look at the last part of the log to see what's happened.

[PetrolioBenzina]

Originally Posted by Wayno

Shiv's never had an automatic development car. The only tuner to advertise that......... look at the last part of the log to see what's happened.

Thanks for the detailed answer. I've driven the auto a few times (stock) and thought it was great. OP could save himself some headache and try the flappy-things.

[Nah]

Originally Posted by Wayno

Yet another problem could simply be that the OFT tunes have removed too much timing in certain load columns below 2200 rpm.

A little timing goes a long way...

[KoolBRZ]

If you drive a Manual Transmission, you just aren't going to get it. With a MT, you can just push the pedal down further, if it doesn't have enough power, you can down shift and feather the throttle to pace the vehicle in front of you.
With an automatic, in automatic mode, to keep pace with the vehicle in front of you, you push down the pedal, then suddenly it downshifts and quickly speeds up forcing you to let off the throttle, shifting it up again.
Since I can't recode the TCU to change the shift points, I'm trying to add more torque with the throttle map so I don't need to press the pedal as far to accelerate. Changing the throttle map has definitely changed the shift points of the TCU, but for the worse. When I added a lot more requested torque at low rpms, it short-shifted, instead of taking longer to shift, like I was hoping.
Kodename47 has a good point about the other maps in the DBW folder though, so I'm going to go back to a previous, "smoothed" map, and experiment with the other tables to determine what effect they have on the shift points and the TCU.

[Kodename47]

Calculated Torque tables have no impact on a MT so I expect that they are there to communicate with the AT TCU. Maybe try and emulate your (expected) torque map. I wouldn't go changing the max values however, use whats there to model your torque output and see what happens.

[nikitopo]

You can use sprint booster which doesn't need a tune. There is a specific one for AT cars. Personally I was using one in my MT car, but throttle response was way aggressive and it was better to de-activate it in traffic. I believe it depends the driver. Some like it, others not. Personally, I prefer the OEM settings which keep the balance. Maybe you'll like a more aggressive throttle if you want to slide the rear more.

[phrosty]

Originally Posted by Wayno

The trouble with koolbrz is rather than identifying and fixing problems, he always has to start from scratch with some brand new scheme and introduce 100 new issues, before ending up back to the drawing board a month later.

Originally Posted by KoolBRZ

Changing the throttle map has definitely changed the shift points of the TCU, but for the worse. When I added a lot more requested torque at low rpms, it short-shifted, instead of taking longer to shift, like I was hoping.
Kodename47 has a good point about the other maps in the DBW folder though, so I'm going to go back to a previous, "smoothed" map, and experiment with the other tables to determine what effect they have on the shift points and the TCU.

[KoolBRZ]

Originally Posted by Kodename47

Calculated Torque tables have no impact on a MT so I expect that they are there to communicate with the AT TCU. Maybe try and emulate your (expected) torque map. I wouldn't go changing the max values however, use whats there to model your torque output and see what happens.

I made some test tunes and found out the Calculated Torque tables are it. It works just the opposite of what I'd expect however. CT tables remade to mirror the power I'm getting now cause the AT to short-shift horribly. Then I used a trick I learned from AVCS tuning. I shifted the whole table one column to the left, saved it as +10, shifted it one column to the right, saved it as -10, and tried them out one after the other. The -10 shifted noticeably better, revving a good 1000 rpms higher before shifting, and downshifting easier without all the usual stomping to get a downshift. I'll be posting a new thread, "Modding your Calculated Torque tables for better shifting, for A/T's only."