HPTuners question about "delivered torque".

[Rich Z]

One of the parameters I discovered in HPTuners while monkeying around with it is a rather interesting display called "TRQ Dlvrd (ft.lb)" which I am assuming means "Torque Delivered".

Now this presents all kinds of interesting possibilities if it actually means what it says. Does the PCM actually provide a data feed or REAL and ACCURATE torque measurements coming from the engine/drivetrain? If you take a look at this graph I took while data logging my C5Z after having the STS twin turbos installed, you can obviously see SOMETHING being shown that certainly does seem to represent power delivered SOMEWHERE by SOMETHING....



So the question is, does this REALLY provide us with some realistic data indicating the power being delivered by the engine? At least realistic enough so that a person could see some TRUE changes being made from modifications done to the engine and be able to judge relative changes in the power output of the engine? Maybe I am barking up a wrong tree, but I think this would prove to be extremely useful if it provides something akin to real world data.

Of course, there are other questions.... (1) HOW is this data derived? What sorts of sensors are used to determine how much torque is produced and HOW? (2) Where is this data collected from the drivetrain? In other words is it showing torque directly at the engine (which I suspect from the figures showing in my graph) or elsewhere along the drivetrain somewhere? And (3) How can this data be used to interpolate a horsepower figure? I'm assuming HPTuners can mingle data streams to give you a calculated result for a graph..........

Of course, the implications of this are that perhaps we all have something akin to the data produced by a dyno right at our fingertips if you have the tool to read the data...

[Bob@pm-fl.com]

You are are reading is a calculated variable, by the transmission input. I have observed this reading being over 550 ft-lbs on a supercharged customers car, in reality it made about 500 ft-lbs. I have also read about this being so far off that people were wonding what the PCM was thinking, although it's relevance to tuning isn't really there.

Hope that helps!
Bob

[miami993c297]

I opened a thread recently on HP Tuner Forum, find the link below:

http://www.hptuners.com/forum/showthread.php?t=10401

The Torque numbers you are reading are Flywheel ...JFYI

Christian

[Rich Z]

Thanks! I actually did read that thread over on HPTuners, but honestly, some of it is greek to me. There is a LOT about this stuff that I am just getting my feet wet with.

I guess what I am looking for is something that HPTuners can provide as a measure of the power output that any changes you make to the engine's parameters will produce. Since there is the capability of capturing some sort of data related to torque, this seems like perhaps it would fit the bill. The actual numbers don't have to actually be dead on accurate, just consistent. In other words, all things remaining equal, the output results would remain the same. So if I made any changes, then ran it again, those changes would show a proportional change, better or worse, than what I had before the change.

Oh, one thing I must have missed in that thread was the manner in which this torque value is determined. Is there actually a strain sensor somewhere that can determine torque values delivered by the engine?

[miami993c297]

Originally Posted by Rich Z

Thanks! I actually did read that thread over on HPTuners, but honestly, some of it is greek to me. There is a LOT about this stuff that I am just getting my feet wet with.

Oh, one thing I must have missed in that thread was the manner in which this torque value is determined. Is there actually a strain sensor somewhere that can determine torque values delivered by the engine?

Hi Rich Z,

That is excately the why of my thread on HPT Forum...

The histo in this version is ABSOLUTELY INACCURATE, then if you want any valuable analyze of your data, avoid this one...

A soon as one of my engineers will have time, I will ask him to take a look to the HP Tuner software and imagine a way to extract some valuable data to define if we can read any increase in acceleration (Time vs. Revs vs. TSP etc...)


Christian

[Rich Z]

Well, I'm not sure "accuracy" is a real concern, if we are talking about something absolute. I'm looking for "relative" figures. In other words, if I make a run and the data graph shows a peak of 550 ft lbs, then I tweak something and make another run that shows 575 ft lbs as the new peak, is it TRULY showing an increase in torque? I don't need to know the figures are accurate, only that an actual RELATIVE increase in torque has resulted from the tweak.

And yes, I would think a measurement that would show the amount of time to reach a certain speed via HPTuners would be just as informative as seeing what your quarter mile times were at the track. Perhaps setting beginning and ending points for mph and determining the elapsed time between those two points would be helpful. If it takes you 4.2 seconds to go from 35 to 90 before a tweak, but only 3.5 seconds afterwards, that would be pretty reasonable evidence that your tweak had a positive change in performance.

[miami993c297]

Originally Posted by Rich Z

Well, I'm not sure "accuracy" is a real concern, if we are talking about something absolute. I'm looking for "relative" figures. In other words, if I make a run and the data graph shows a peak of 550 ft lbs, then I tweak something and make another run that shows 575 ft lbs as the new peak, is it TRULY showing an increase in torque? I don't need to know the figures are accurate, only that an actual RELATIVE increase in torque has resulted from the tweak.

And yes, I would think a measurement that would show the amount of time to reach a certain speed via HPTuners would be just as informative as seeing what your quarter mile times were at the track. Perhaps setting beginning and ending points for mph and determining the elapsed time between those two points would be helpful. If it takes you 4.2 seconds to go from 35 to 90 before a tweak, but only 3.5 seconds afterwards, that would be pretty reasonable evidence that your tweak had a positive change in performance.

We are on the same page.

It is exactelly where I pushed the moderators on HP Tuner Forum.

No accuracy in absolute, but that's not the matter.

No possibility to make any interpretation of the histogram, the pids related entered in the calculation beeing the wrong one.

I did as an exemple on a hot day (over 80°) an addition of two degrees of timing, acceleration time was worse but torque histo was showing an increase, because in the equation added timing is a positive value, then I would not take care of this histo at all.

Just my own experience

Christian

[Rich Z]

So this sounds like it is a calculation done in the program based on several inputs rather than a single data stream from a sensor. If that is the case, then it is fixable within HPTuners by fixing the code that does the calculation.

Has HPTuners addressed this "bug"?

Doesn't the PCM detect and control torque delivered to the transmission in automatics? I thought that "torque management" was one of the params you can control when tuning the PCM? That seems to imply to me that the PCM can determine torque itself from somewhere. After all, it has to know when to reduce torque during shifting operations, which means it should be able to determine if the torque is high enough to need being limited.

If not detected directly, it most certainly should be some derived value based on multiple inputs.

[miami993c297]

Hi Rich Z,

You are almost there.

In the post # 23 I guess in my thread on HPT Forum I will again quote Ken (HPT Software designer):

"Here is what Ken said about the calculation:

They convert engine airflow to torque using AFR and Spark as modifiers.

Sounds like its a pretty useless, uninvolved calculation lol"

That's why it is useless...
Christian

[dvdcta]

Originally Posted by Rich Z

One of the parameters I discovered in HPTuners while monkeying around with it is a rather interesting display called "TRQ Dlvrd (ft.lb)" which I am assuming means "Torque Delivered".

Now this presents all kinds of interesting possibilities if it actually means what it says. Does the PCM actually provide a data feed or REAL and ACCURATE torque measurements coming from the engine/drivetrain? If you take a look at this graph I took while data logging my C5Z after having the STS twin turbos installed, you can obviously see SOMETHING being shown that certainly does seem to represent power delivered SOMEWHERE by SOMETHING....



So the question is, does this REALLY provide us with some realistic data indicating the power being delivered by the engine? At least realistic enough so that a person could see some TRUE changes being made from modifications done to the engine and be able to judge relative changes in the power output of the engine? Maybe I am barking up a wrong tree, but I think this would prove to be extremely useful if it provides something akin to real world data.

Of course, there are other questions.... (1) HOW is this data derived? What sorts of sensors are used to determine how much torque is produced and HOW? (2) Where is this data collected from the drivetrain? In other words is it showing torque directly at the engine (which I suspect from the figures showing in my graph) or elsewhere along the drivetrain somewhere? And (3) How can this data be used to interpolate a horsepower figure? I'm assuming HPTuners can mingle data streams to give you a calculated result for a graph..........

Of course, the implications of this are that perhaps we all have something akin to the data produced by a dyno right at our fingertips if you have the tool to read the data...

by the seat of pants