Desired torque implications on tuning process

Tuning discussion for all first generation Mazda RX-8 (2004-2008)
Post Reply
zenvirzan
Posts: 7
Joined: May 6th, 2022, 3:52 am

Desired torque implications on tuning process

Post by zenvirzan »

Hi all, I'm currently trying to understand how the Desired Torque table works.

Can anyone confirm if this is how the calculations are applied? Looking at the table axes, this is the only flow I can think of that makes sense.

IRL APP -> Throttle Duty table = "Requested" Throttle Position (degrees)
"Requested" Throttle Position -> Desired Engine Torque table = Virtual "torque" [and this is an abstract value, not calculated using any sort of closed loop calculation etc, simply used as an index in the next lookup?]
Virtual "torque" -> Relative Throttle Plate Position table = Actual Throttle Position

I'm assuming changing the Desired Engine Torque table has no impact on fuelling or ignition on its own, but is just used as an index for other tables?

How does the Calculated Engine Torque table affect things? Is it a cruise-control thing, considering there is no APP or TPS in the axis?

Assuming this is the case, assuming I had the time to spare, to do a full tune to the best of my ability, would it make the most sense to (ignoring ignition timing tuning for the minute):

1. disable closed loop (zero the Closed Loop Max RPM value)
2. modify the Throttle Duty table to ease-in to linear (similar to stock)
3. Flatten the Desired Engine Torque table values to be linear (from 0 to x, x can be whatever)
4. Modify the Relative Throttle Plate position table axis to be from 0 to x as set in the Desired Engine Torque table and set the values to be linear from 0 to 84
5. Set a flat AFR in the AFR table (eg. 13.8 or 12.5) for tuning purposes
6. Tune using the VE table to make the actual AFR match the Target AFR as close as possible
EDIT: as Walle has pointed out, it may be more effective to utilise the trims + the actual AFR, rather than use the possibly erroneous "target AFR" value. There is also no requirement to do step #5, however I would personally just to make the math easy.

After that process is over, the AFRs can be tweaked to go for maximum power as usual.

Steps 2 and 3 should effectively disable the Desired Engine Torque table if my understanding is correct to reduce the variables at play, making the TPS correlate to the APP as specified in the Throttle Duty table?
Last edited by zenvirzan on September 18th, 2022, 2:24 pm, edited 1 time in total.
User avatar
wcs
Posts: 55
Joined: May 12th, 2021, 12:46 am

Re: Desired torque implications on tuning process

Post by wcs »

Good questions.
Wish I could add some input.
But I'm interested in your post.

I don't know if there is anything I could do or try here to maybe help validate or test ideas?
Walle
Posts: 34
Joined: August 29th, 2021, 7:27 pm

Re: Desired torque implications on tuning process

Post by Walle »

step 5 and 6 can be done irrelevant. VE table is very rough in OEM config. viewtopic.php?f=25&t=4316&p=16161#p16161 read my post/answer in this topic. :beertoast
zenvirzan
Posts: 7
Joined: May 6th, 2022, 3:52 am

Re: Desired torque implications on tuning process

Post by zenvirzan »

Thanks Walle, I'll keep that in mind. Comforting to know that even stock it wasn't perfect.

Still curious if anyone has tried effectively removing the Desired Torque table as mentioned in the OP--I have starting issues currently and it really didn't like it when I tried it, but I also didn't try very hard (fairly sure it was due to my throttle plate not being open far enough at idle in the relative throttle position table). I'm more worried about if it'll do any weird ignition timing stuff as a byproduct.
Walle
Posts: 34
Joined: August 29th, 2021, 7:27 pm

Re: Desired torque implications on tuning process

Post by Walle »

but why would you mess with torque tables ? imho they are also used when ESC , ESP and TC are engaged. could make for a very unpredictable car.
zenvirzan
Posts: 7
Joined: May 6th, 2022, 3:52 am

Re: Desired torque implications on tuning process

Post by zenvirzan »

After some further experimentation, it looks like it works as expected! I'm super happy with the outcome, I think this may change the way we tune RX8s from now on! I know I'd never go back :bouncinggrins

See my new thread with the example tune for more information!
Post Reply