Torque Model editing functionality now in beta!

[Higgs Boson]

But that doesn't make any sense as my DD tables are capped at 575 WOT and I am seeing 700 on my immediate torque requests and if I go in and raise the torque on my DD tables I actually see less torque output. Example, I was looking at a tune for another truck running basically the same setup, but his tuner ramped his DD tables up to 1200 (don't ask me why...) in WOT but his immediate never requests more than 450 and so his engine output is only hitting 450.

On my truck, trying two tunes with the exact same VTT tables I make more signigicantly power with lower numbers in the DD table. It's almost as if they inversely effect actual output. Hell, maybe I should zero them out, lol! At this point making any adjustments to anything that effects the predicted torque tables (VTT) doesn't seem to have any effect on actual output. Sure, it changes your logged predicted torque, but your actual output still follows the immediate request.

You DD table is in kilowatts.

575 *1.34 is 770.5 HP as it should be a little higher than delivered since the systems of the car take power between the engine and wheels.

[Chris@HPTuners]

The torque model is important because all torque requests and torque related functions of other controllers in the vehicle rely on it being accurate. TCM and EBCM being the most obvious users and commanders of engine torque.

The Immediate and Predicted torque requests naming is just the way GM calls it. Dodge for example call it Fast path (spark) and Slow path (throttle).

[Higgs Boson]

The torque model is important because all torque requests and torque related functions of other controllers in the vehicle rely on it being accurate. TCM and EBCM being the most obvious users and commanders of engine torque.

The Immediate and Predicted torque requests naming is just the way GM calls it. Dodge for example call it Fast path (spark) and Slow path (throttle).

Is the torque model supposed to be functional at all RPMs and airmasses and MAPs? For example, there are fields that limit Desired MAP for the torque model and one that limits MAP for the torque model. One is at 150kpa and one is at 99kpa. Why wouldn't they both be 180? Is it a limiter or does the torque model stop calculating at 99 for something? I can't believe it's nothing at all but I have not been able to observe some effect.

Immediate and Predicted corrections have always been spark/fuel or air but you're saying that it's the same for the torque model? If that's the case shouldn't they always match, you wouldn't want Immediate (Spark and Fuel) requesting 250 ft lbs and Predicted (air) requesting 300 ft lbs would you? Maybe that's why you can see the throttle go to 100% when pedal is at 40%? Is the first thing the torque model modifies the throttle blade? I guess it makes more sense to open the throttle more rather than add more spark.

Also, I have seen changes to the torque model change airmass reported. WOT can show 1.44 g/cyl and an adjustment to the model changes it to 1.15.....same delivered power. So the ECM is looking up the value that was at 1.44 and now sees it at 1.10 so it thinks that's the new airmass?

Is there a way to disable the torque model or manipulate it so we can make test runs to log real/measured airmass/flow values so we can build the torque model?

If we change negative driver demand values to 0 is it also correct to change negative torque model values to 0?

It's like we are at the point where I have more questions now than I did early on as we start messing with it and see different results. Is there a "Part 2" to the E78 Tuning Guide webpage? :-)

[Harlech]

Doesn't matter what units it is in the correlation doesn't add up. The ECM switches between predicted and immediate torque commands based on different conditions from what I am seeing and there doesn't seem to be an exact way of determining which one it is running off of. It isn't like TM where it will use the lowest calculated torque of everything as the primary, your output might follow immediate for a section of a pull, switch to predicted and then finish back off following immediate, though it seems like the only time torque output follows predicted is when predicted is within a small percentage of immediate. The majority of the time it is following immediate, but I even tried putting the DD tables to 100 on WOT and interpolate them from the stock 0% pedal, eventually it doesn't give a damn what is in your DD tables and it will start building power, granted your throttle fights it, but it still makes a load of power, possibly as a result of the blower, just fighting the TB the whole way. It would seem that both your DD tables and your VTT tables have to be around the true torque curve of the engine to make the max amount of power and we still don't know what is actually generating predicted vs. immediate torque.

[Chris@HPTuners]

there is no Immediate or Predicted for the torque model, its just the torque model. They are seperate things. As explained in the E78 guide they use fast and slow methods based on the actuator speed, the throttle is a slow thing, spark is instant. It has nothing to do with modelling the engine torque at all. It is how the ECM chooses to manage the engine torque.

So long as the torque request is above the current engine torque, that actuator will do nothing. The torque request can be fast, slow or both at the same time, its up to the requesting tq mgmt type to decide.

eg. the RPM limit can use spark to control a fast revving engine to prevent overshoot, then close the throttle to hold the limit. So initially there would be some fast (immediate/spark) reduction and some slow (predicted/throttle) reduction, then once the limiter is stable there would only be slow reduction.

Similar for trans shifts, normal upshift would generally only use fast reduction (spark). But for a downshift blip, you can't increase the engine RPM with spark, so it must request a torque increase from the engine using the slow (throttle).

The torque model is important becuase when the trans shifts it is making all these acceleration, slip and pressure calculations based on how much torque the ECM is telling it the engine is making. If the model is too low, the shifts will flare becuase the TCM won't request enough tq reduction and calculate all the pressures too low. If the torque model is too high, the TCM thinks you are putting down some power and bumps everything up for a firmer, faster shift which is also wrong.

The torque model needs to be in the ballpark, like 5-10%. I know GM TCM guys want 5% accuracy for the trans to behave nicely. <== This is the reason we added the virtual torque feature in the first place

[Chris@HPTuners]

Doesn't matter what units it is in the correlation doesn't add up. The ECM switches between predicted and immediate torque commands based on different conditions from what I am seeing and there doesn't seem to be an exact way of determining which one it is running off of. It isn't like TM where it will use the lowest calculated torque of everything as the primary, your output might follow immediate for a section of a pull, switch to predicted and then finish back off following immediate, though it seems like the only time torque output follows predicted is when predicted is within a small percentage of immediate. The majority of the time it is following immediate, but I even tried putting the DD tables to 100 on WOT and interpolate them from the stock 0% pedal, eventually it doesn't give a damn what is in your DD tables and it will start building power, granted your throttle fights it, but it still makes a load of power, possibly as a result of the blower, just fighting the TB the whole way. It would seem that both your DD tables and your VTT tables have to be around the true torque curve of the engine to make the max amount of power and we still don't know what is actually generating predicted vs. immediate torque.

The PIDs for immediate tq cmd and predicted tq cmd are the actuator torque requests. There are also PIDs that tell you the current source for these commands so you can see exactly what system is commanding them, it's all there for both engine tq and axle tq

[Higgs Boson]

there is no Immediate or Predicted for the torque model, its just the torque model. They are seperate things. As explained in the E78 guide they use fast and slow methods based on the actuator speed, the throttle is a slow thing, spark is instant. It has nothing to do with modelling the engine torque at all. It is how the ECM chooses to manage the engine torque.

So long as the torque request is above the current engine torque, that actuator will do nothing. The torque request can be fast, slow or both at the same time, its up to the requesting tq mgmt type to decide.

eg. the RPM limit can use spark to control a fast revving engine to prevent overshoot, then close the throttle to hold the limit. So initially there would be some fast (immediate/spark) reduction and some slow (predicted/throttle) reduction, then once the limiter is stable there would only be slow reduction.

Similar for trans shifts, normal upshift would generally only use fast reduction (spark). But for a downshift blip, you can't increase the engine RPM with spark, so it must request a torque increase from the engine using the slow (throttle).

The torque model is important becuase when the trans shifts it is making all these acceleration, slip and pressure calculations based on how much torque the ECM is telling it the engine is making. If the model is too low, the shifts will flare becuase the TCM won't request enough tq reduction and calculate all the pressures too low. If the torque model is too high, the TCM thinks you are putting down some power and bumps everything up for a firmer, faster shift which is also wrong.

The torque model needs to be in the ballpark, like 5-10%. I know GM TCM guys want 5% accuracy for the trans to behave nicely. <== This is the reason we added the virtual torque feature in the first place

Ok good, it's more along what I thought, I may have misunderstood one of your previous posts.

How does the torque model affect a manual trans car with no TCM? As long as you keep the throttle open without it running away then you're good?

While we have your attention, can you comment on the effects of the fields I was asking about, the Max Desired and Max for the Torque Model kpas? Can you also comment on what the Max Torque Timing table does on these cars? I have seen everything from transient fueling to max limit on spark to nothing at all.....I think I have seen them affect reported torque numbers as well. Any input on these two items?

[Harlech]

there is no Immediate or Predicted for the torque model, its just the torque model. They are seperate things. As explained in the E78 guide they use fast and slow methods based on the actuator speed, the throttle is a slow thing, spark is instant. It has nothing to do with modelling the engine torque at all. It is how the ECM chooses to manage the engine torque.

So long as the torque request is above the current engine torque, that actuator will do nothing. The torque request can be fast, slow or both at the same time, its up to the requesting tq mgmt type to decide.

eg. the RPM limit can use spark to control a fast revving engine to prevent overshoot, then close the throttle to hold the limit. So initially there would be some fast (immediate/spark) reduction and some slow (predicted/throttle) reduction, then once the limiter is stable there would only be slow reduction.

Similar for trans shifts, normal upshift would generally only use fast reduction (spark). But for a downshift blip, you can't increase the engine RPM with spark, so it must request a torque increase from the engine using the slow (throttle).

The torque model is important becuase when the trans shifts it is making all these acceleration, slip and pressure calculations based on how much torque the ECM is telling it the engine is making. If the model is too low, the shifts will flare becuase the TCM won't request enough tq reduction and calculate all the pressures too low. If the torque model is too high, the TCM thinks you are putting down some power and bumps everything up for a firmer, faster shift which is also wrong.

The torque model needs to be in the ballpark, like 5-10%. I know GM TCM guys want 5% accuracy for the trans to behave nicely. <== This is the reason we added the virtual torque feature in the first place

Okay, that makes a ton more sense! I wasn't able to extrapolate that from the E78 guide at all.

Thanks, Chris!

[veee8]

there is no Immediate or Predicted for the torque model, its just the torque model. They are seperate things. As explained in the E78 guide they use fast and slow methods based on the actuator speed, the throttle is a slow thing, spark is instant. It has nothing to do with modelling the engine torque at all. It is how the ECM chooses to manage the engine torque.

So long as the torque request is above the current engine torque, that actuator will do nothing. The torque request can be fast, slow or both at the same time, its up to the requesting tq mgmt type to decide.

eg. the RPM limit can use spark to control a fast revving engine to prevent overshoot, then close the throttle to hold the limit. So initially there would be some fast (immediate/spark) reduction and some slow (predicted/throttle) reduction, then once the limiter is stable there would only be slow reduction.

Similar for trans shifts, normal upshift would generally only use fast reduction (spark). But for a downshift blip, you can't increase the engine RPM with spark, so it must request a torque increase from the engine using the slow (throttle).

The torque model is important becuase when the trans shifts it is making all these acceleration, slip and pressure calculations based on how much torque the ECM is telling it the engine is making. If the model is too low, the shifts will flare becuase the TCM won't request enough tq reduction and calculate all the pressures too low. If the torque model is too high, the TCM thinks you are putting down some power and bumps everything up for a firmer, faster shift which is also wrong.

The torque model needs to be in the ballpark, like 5-10%. I know GM TCM guys want 5% accuracy for the trans to behave nicely. <== This is the reason we added the virtual torque feature in the first place

There must be a hard limit with the TCM on the 6L and 8L transmissions. They will max the torque around 885ft lbs. in the scanner with the engine reported torque being much higher.

[GrannySShifting]

Where do you begin with modeling the virtual torque stuff accurately? Have a 6L80 car that been rebuilt different clutches and this thing is all over the place, this being very accurate will prob help with all the other variables that are introduced with different clutch volumes etc

[viper139]

You can edit the RPM axis by opening up a coefficient table and clicking the column axis label. The coefficient tables X axis is what controls what RPM values Virtual Torque uses.

Changing the Y axis units might come later.. it's more involved.

I have tried to change the RPM axis by doing this with no luck.
Is this still available on later versions of the VCM virtual Torque editor?

[GHuggins]

Go to engine - torque management - engine - airmass - change axis here just like it has been... Axis automatically update in the editor...

[viper139]

Thank you Sir that worked.

Curious, in this particular tune the table stops at 6400 rpm.
The rev limiter is set at 6600 rpm.

Does the PCM interpolate for the coefficients beyond 6400 rpm?
If so, then there is no reason for me to change this table, even if I bump the rev limiter up to 6700 rpm it should be good.

Your thoughts please.
Thanks

[GHuggins]

As far as I know, it uses the last values in the table, but can't confirm for sure - easy enough to log and see if the numbers stay the same to redline though...

[viper139]

As far as I know, it uses the last values in the table, but can't confirm for sure - easy enough to log and see if the numbers stay the same to redline though...

Thank you Sir.

[Higgs Boson]

While we have your attention, can you comment on the effects of the fields I was asking about, the Max Desired and Max for the Torque Model kpas? Can you also comment on what the Max Torque Timing table does on these cars? I have seen everything from transient fueling to max limit on spark to nothing at all.....I think I have seen them affect reported torque numbers as well. Any input on these two items?

any word on these two years of experience later?

[Simplyjesse]

How about another 2 years later? Great thread by the way guys. E78's are ridiculous to find reliable info on.

[TriPinTaZ]

I think some of the strategies with tuning the torque model have evolved. The important thing to consider is the VVE plays a role with Virtual Torque, especially during transient conditions.