Torque Table Inverse Calculator

[txtailtorcher]

Hello everyone, just thought I'd share for those of you who don't have it but here's my calculator for calculating the Inverse tables in the TQ management. I will prolly revise it so you can just Ctl+C and then Ctl+v back into the tables but for now you have to C then transpose into the table to get the values. If you're really lazy like me and dont mind repetitive work then you can Copy the values from the TQ table and paste then Copy again and transpose/paste then copy the answers and paste values only and then copy and transpose and then paste into the table. I will also see if I can get it calculate TQ values based off of the inverse tables.

[murfie]

Do you have the 3.7 beta?

You can click edit and use the built in torque inverse calculator.

[txtailtorcher]

I don't

[aksubiedubie]

If you compare the inverse calculator output of the various excel files posted on the forum to the HPTuners calculated output, it differs slightly for some reason to better match what ford put in. If you add 0.001 ft lb and calculate inverse on a stock file, it is also different, especially at low RPM high load sections. Kinda makes you wonder. Murfie? Lol
- Isaac

[veeefour]

The same with effective area calc - not sure what equations they use but they are different than mine

[02reaper]

If you compare the inverse calculator output of the various excel files posted on the forum to the HPTuners calculated output, it differs slightly for some reason to better match what ford put in. If you add 0.001 ft lb and calculate inverse on a stock file, it is also different, especially at low RPM high load sections. Kinda makes you wonder. Murfie? Lol
- Isaac

I noticed this also while using the hptuners calculator. Even If you convert all stock values using the editor calculator, it deviates somewhat from what the original calculation is. Any insight on this hptuners?

[murfie]

If you compare the inverse calculator output of the various excel files posted on the forum to the HPTuners calculated output, it differs slightly for some reason to better match what ford put in. If you add 0.001 ft lb and calculate inverse on a stock file, it is also different, especially at low RPM high load sections. Kinda makes you wonder. Murfie? Lol
- Isaac

When you plot it out, there's a strong logarithmic regression between torque OR load AND RPM.

ln(x) lets us plug in growth and get the time it would take. The regression goes Y= M*ln(X)+b.

Y being load or TQ
X being RPM

e^x is the inverse if ln(x). e^x can give you the amount of growth given x amount of time.

I'm still playing with different numbers trying to put together a spreadsheet. The difference you see in the all the beta calculators values are negligible. It's better than the spreadsheets I've seen. All you really need to do with it, is make sure your axis values are well spread over the resolution of the torque values you are entering.

[Steven@HPTuners]

The Torque Inverse Calculator uses an actual model of engine torque to smooth the values you see.
If you run it on stock values, you essentially smooth it out in regards to an "ideal" model of torque. Most other calculators I saw just do linear fits, so that's why you'll see the difference.

Real life never follows this ideal quite perfectly so even if you take a Ford calibrated table, it'd act to "smooth" the table, but such is life when using models. It will still calculate inverse correctly (it inverts the model directly), but yes, it may not match exactly what is already there.
The reason we did it this way is because we're thinking of use case: You'll be putting numbers in based on good estimations of actual (most of us don't have engine dynos to do this with), and the calculator acts to try to take your ballpark estimates and make them fit a really good model for engine torque.

If you use an actual engine dyno to calibrate it, a la Ford, its going to naturally "smooth" out your values, but you'll find the model still fits quite well. The R-squared if you're familiar with statistics is over 0.999 when we apply this equation to most Ford tables, with the worst fit spot usually being very high load low RPM.



The throttle equation likewise uses a throttle body model, similar to one found in Modeling and Control of Engines and Drivelines.
As far as we're aware from what little we learn about Ford calibration from the inside, they have a tool that can convert flow bench data to that table using similar equations.
Again R-squared is really good, worst fits at the lowest throttle angles (where throttle leakage is highest).

The tools are designed to help get you models that are smooth and match actual physics involved while fitting closest to the data you measure, but once again, its a tool. You'll have to give it good, well measured data, to get a fit that works.
Even after use the "inverse" tool, you may have reality set in and have to correct some areas by hand that don't match models the best, but you'd have this same issue with other "fits" as well.

We will continue to improve these tools based on your guys' feedback when actually using them.

[veeefour]

I can confirm that Effective Area calc works very good - some would just change the predicted throttle only but the effective area is an
INVERSE(more less) for that. Every time predicted throttle gets tuned the effective area "inverse" calculations has to follow - I see this mistake very often.

I was doing this manually before - now life's easier

[txtailtorcher]

does anyone have a method to tuning Effective area? i've been messing with mine and i cant really grasp how it works.

[Bill@HPTuners]

does anyone have a method to tuning Effective area? i've been messing with mine and i cant really grasp how it works.

You should be able to log
ETC Throttle Angle Error
ETC Effective Throttle Area
ETC Vacuum

in order to dial in the effective area tables.

[txtailtorcher]

Thank you Bill!

[Ptman22]

Why does my figures not equal the same figures in the torque inverse calculator?
My numbers are higher than the calculators

[txtailtorcher]

Why does my figures not equal the same figures in the torque inverse calculator?
My numbers are higher than the calculators

which calculator? mine or HPT's? from what i understand HPT's works on a "best fit" calculation vs mine which is "direct fit" if you will.

[18f1505.0roush]

Why does my figures not equal the same figures in the torque inverse calculator?
My numbers are higher than the calculators

Asking also?

[18f1505.0roush]

YouTube video? trying to grasp this thanks