Page 1 of 2 12 LastLast
Results 1 to 20 of 22

Thread: Question about Torque Inverse Calculator

  1. #1

    Question about Torque Inverse Calculator

    I've never used the Calculator built into HP Tuners and I'm currently working on a 2015 Mustang GT with 80MM turbo the car runs great up top and at idle but certain areas during part throttle i'm seeing torque errors in my log. The car originally was on a dyno and they tuned the torque tables but once we got in on the street we noticed some issues during part throttle. I took one of the Mapped points that the issue was happening in and reduced torque by 10 in load areas .20-1.60 load using the calculator then hit the calculate inverse button and my 3d inverse table has 0's in certain areas anyone else have this issue?

    example.PNG

  2. #2
    Senior Tuner CCS86's Avatar
    Join Date
    Nov 2017
    Location
    Austin
    Posts
    1,089
    I'm not a fan of the HPT inverse calculator. It uses curve fitting, when I don't think that is applicable. Just use it on a stock GT torque table and see how much it changes the inverse. Ford didn't get it wrong.

    The high load zeros are there because your row axis on the inverse table is too big and the calculator wont extrapolate. You can try making the highest axis value 547, then rescale the rest of the axis values to be a smooth blend down.

    The low load zeros just look like bugs in the calculator. It should have all the data it needs to fill those.

  3. #3
    Quote Originally Posted by CCS86 View Post
    I'm not a fan of the HPT inverse calculator. It uses curve fitting, when I don't think that is applicable. Just use it on a stock GT torque table and see how much it changes the inverse. Ford didn't get it wrong.

    The high load zeros are there because your row axis on the inverse table is too big and the calculator wont extrapolate. You can try making the highest axis value 547, then rescale the rest of the axis values to be a smooth blend down.

    The low load zeros just look like bugs in the calculator. It should have all the data it needs to fill those.

    Do you know of a better way or a different way to calculate the inverse and Torque? the table seems pretty close it was done on a dyno but some part throttle areas I get a little bit of surging I no longer have access to the dyno so I'm trying to figure out a way to adjust the torque and inverse on the road. I seen a calculator on a different post and I setup a histogram and set the car to run in only 1 mapped point and using etc tq request and rpm and used air load and copied the airload into the inverse and smoothed and then entered the data into the spreadsheet made by murfie but my TQ values seem to increase in some cells and cells right after it in a higher RPM range are lower when comparing my original tq numbers

  4. #4
    Senior Tuner CCS86's Avatar
    Join Date
    Nov 2017
    Location
    Austin
    Posts
    1,089
    Using ETC TQ request as a histogram axis to fill airload doesn't seem like a good approach to me. Those values are essentially just coming out of the driver demand table and have no connection to the produced TQ vs RPM & Load that the inverse tables are trying to establish.

    I wrote my own excel sheet to calculate TQ from inverse and inverse from TQ.

    I have tried a number of different methods to populate these tables from scratch, and so far don't feel like I've totally solved it.

  5. #5
    Advanced Tuner
    Join Date
    Sep 2018
    Posts
    213
    Agreed, I just tried it the other day and worked somewhat to a degree but wouldn't smooth out and resulted in more issues. I am wondering if disabling torque control, air switch, and putting WOT start at 1 A/D what yield better results. I'm not exactly sure but that should remove the influence of driver demand? Also set up histogram to filter out no pedal and another with only pedal so that you can remove the noise and get some data for decel.... Maybe I'll try it an post the results

  6. #6
    Senior Tuner CCS86's Avatar
    Join Date
    Nov 2017
    Location
    Austin
    Posts
    1,089
    IMO, that is a really bad idea.

    Ford didn't go to all the trouble of creating the driver demand system because it is on par or worse than a "cable".

    To get a linear torque response, the ETC opening angle is highly non-linear.

    Give this a try: revert back to stock torque / inverse tables, but rescaled at the top end for more headroom (ie load up to 1.6). Then log IPC TQ error and Engine Brake TQ. At any combination of MP, RPM and Brake TQ that you get TQ errors, go to the inverse table and multiply that cell by 0.96, and the cells bordering it by 0.98. Don't bother trying to recalculate the torque table.

    See if that helps you smooth things out.

  7. #7
    Advanced Tuner
    Join Date
    Sep 2018
    Posts
    213
    Makes sense. I didn't have time to try it anyways. HPT doesn't have IPC torque error or engine brake torque for mine and has no plans to add it when I requested them along with others. So I have to use an old SCT with livelink . Kind of sticks because I like the VCM scanner way better. I do have engine indicated torque and current engine torque though. Regardless, how exactly do you recommend setting up the histogram? Hate to bother but if you have a picture that would be helpful . I'm curious to try this, just reading some of your other posts sounds like you have a really good understanding of the system.

    Side note, funny how you say not to adjust the torque table but only the inverse. Before the build this truck had a Edge tune on it. I read it and read the stock file and noticed only the inverse was changed. As always I appreciate your help

  8. #8
    Advanced Tuner
    Join Date
    Sep 2018
    Posts
    213
    One other thing. Do I need to rescale for a N/A engine? I normally only see loads up to 1.05 at WOT. Running WOT is runs really well. Down low is where I have issues. Stock values make it shake bad at idle and when letting of the brake in drive.

  9. #9
    Senior Tuner CCS86's Avatar
    Join Date
    Nov 2017
    Location
    Austin
    Posts
    1,089
    No need to rescale for an NA car.

    I don't actually use histogram data for this type of work. Too much junk data gets averaged in. Filtering can help, but won't eliminate everything.

    For working on the inverse table, I have a histogram that approximates the PCM table (row axis BK TQ, column axis rpm). It doesn't matter really what you populate the table with. I usually just use it to graphically see which cell needs attention, when pointing at the time on the leading edge of an IPC TQ error spike. too bad HPT won't get that added for you. They have been pretty responsive to adding parameters for me.

  10. #10
    Advanced Tuner
    Join Date
    Sep 2018
    Posts
    213
    Quote Originally Posted by CCS86 View Post
    No need to rescale for an NA car.

    I don't actually use histogram data for this type of work. Too much junk data gets averaged in. Filtering can help, but won't eliminate everything.

    For working on the inverse table, I have a histogram that approximates the PCM table (row axis BK TQ, column axis rpm). It doesn't matter really what you populate the table with. I usually just use it to graphically see which cell needs attention, when pointing at the time on the leading edge of an IPC TQ error spike. too bad HPT won't get that added for you. They have been pretty responsive to adding parameters for me.
    I can't really complain about them not adding it. They've always been really responsive to me as well and added about 15 parameters on the tune file so I'm thankful for that. The used SCT only cost me $40.00 on Ebay so it's not too big of a deal I just hate unplugging the SCT & HPT over and over LOL. Also, like I stated earlier I like the VCM scanner a lot more. To me it's easier to set up, looks cleaner, and just overall faster. That's my opinion though. The two HPT torque PIDS I have are "Engine indicated torque reference" and "Delivered engine torque". Maybe delivered engine torque is the same as Engine Brake Torque??

    I'm still trying to wrap my head around all of this but I'm determined to figure it out LOL.

    Okay I took a few screen shots. One being my IPC torque error based on the torque table not the inverse table. I wasn't sure which torque PID to use. So the cells with high errors I should multiple the inverse table like you mentioned above?

    Thanks again!
    scanner picture.PNGMCM torque table picture.PNGtorque error picture.PNGlog torque.PNG

  11. #11
    Advanced Tuner
    Join Date
    Apr 2018
    Posts
    605
    Quote Originally Posted by momotunes View Post
    I can't really complain about them not adding it.
    What kind of car do you drive? What's the reason for them not being able to add it - a technical difficulty?? I purchased HPT for just to use on my car pretty much so if it was missing something important like IPC errors and I was having bad issues, i'd be a little upset lol. Delivered engine torque? Sounds like EBT to me...
    Knock Retard is the reduction or prevention of knock by lowering ignition timing:

    (+) Adding Knock Retard = Reducing Timing. PCM is seeing knock.
    (--) Lowering Knock Retard = Increasing Timing. PCM isn't seeing knock.
    __________________________________________________ ________

    2014 Mustang GT Premium. VMP Gen2R Supercharged with an FTI 3000rpm Converter. JLT, BMR, Steeda, Viking, etc.
    Don't fix it if it ain't broken | Maximum effort gets maximum results

  12. #12
    Advanced Tuner
    Join Date
    Sep 2018
    Posts
    213
    Quote Originally Posted by blackbolt22 View Post
    What kind of car do you drive? What's the reason for them not being able to add it - a technical difficulty?? I purchased HPT for just to use on my car pretty much so if it was missing something important like IPC errors and I was having bad issues, i'd be a little upset lol. Delivered engine torque? Sounds like EBT to me...
    Weird.... My post deleted when I tried to edit it. I'll type it again lol.

    It's a 2004 F150 5.4 3V with thumper cams, manly Pistons, IMRC delete, long tubes, etc. To be fair, when I requested it I asked requested it along with other PIDs such as Effective throttle area, inferred MAP, throttle body vacuum, and such. Their was response was they didn't have plans to add those at this time. They did add a bunch of parameters to the file and added trans throttle A/D for me though. I didn't put up a fight though as I already owned the SCT. I did just check and Delivered engine torque matches my flywheel engine torque on the SCT. No EBT on SCT though either for mine. Engine indicated torque on HPT matches SCT torque requested. Axle torque delivered transmission output torque and requested are on SCT but not HPT. Maybe I'll try flipping the switch on torque model back to "normal" now and try tuning wheel torque too. I currently have it on "engine torque" only due to not seeing anything in the VCM scanner to try and tune wheel torque.

  13. #13
    Advanced Tuner
    Join Date
    Sep 2018
    Posts
    213
    Quote Originally Posted by momotunes View Post
    Weird.... My post deleted when I tried to edit it. I'll type it again lol.

    It's a 2004 F150 5.4 3V with thumper cams, manly Pistons, IMRC delete, long tubes, etc. To be fair, when I requested it I asked requested it along with other PIDs such as Effective throttle area, inferred MAP, throttle body vacuum, and such. Their was response was they didn't have plans to add those at this time. They did add a bunch of parameters to the file and added trans throttle A/D for me though. I didn't put up a fight though as I already owned the SCT. I did just check and Delivered engine torque matches my flywheel engine torque on the SCT. No EBT on SCT though either for mine. Engine indicated torque on HPT matches SCT torque requested. Axle torque delivered transmission output torque and requested are on SCT but not HPT. Maybe I'll try flipping the switch on torque model back to "normal" now and try tuning wheel torque too. I currently have it on "engine torque" only due to not seeing anything in the VCM scanner to try and tune wheel torque.

    I need a sanity check lol.

    If driver demand is greater than engine torque what happens?

    If driver demand is lower than engine torque what happens?

    Thanks

  14. #14
    Senior Tuner CCS86's Avatar
    Join Date
    Nov 2017
    Location
    Austin
    Posts
    1,089
    Nobody seems to know definitively.

    Generally, if DD is greater nothing happens. if DD is lower you might get IPC TQ errors.

  15. #15
    Advanced Tuner
    Join Date
    Sep 2018
    Posts
    213
    Well that's reassuring 😭. Okay so if torque tables = 100 and driver demand input is 100 than all is well. Torque tables =100 and driver demand input is 60 you'll get IPC errors which would reduce throttle blade opening?

  16. #16
    Senior Tuner CCS86's Avatar
    Join Date
    Nov 2017
    Location
    Austin
    Posts
    1,089
    Quote Originally Posted by momotunes View Post
    Well that's reassuring ��. Okay so if torque tables = 100 and driver demand input is 100 than all is well. Torque tables =100 and driver demand input is 60 you'll get IPC errors which would reduce throttle blade opening?


    It's not that simple. Nobody here has put forth a complete explanation that holds water. There are a lot of theories, but I have tune/log combos that break all the theories I have heard.

  17. #17
    Advanced Tuner
    Join Date
    Jul 2017
    Posts
    529
    I've been playing with my dd, torque, & tq inv all evening, I go round & round, I think I have it figured out, then one more tweak, & it goes all to hell. There's gotta be a better way!

  18. #18
    Advanced Tuner
    Join Date
    Apr 2018
    Posts
    605
    Quote Originally Posted by CCS86 View Post
    Nobody seems to know definitively.

    Generally, if DD is greater nothing happens. if DD is lower you might get IPC TQ errors.
    I'm sure that's a fail-safe. If it's putting out less power than requested-it's a slow poke, no biggie :P If the car is putting out more power than requested it might be accelerating out of control....!! lol.
    Knock Retard is the reduction or prevention of knock by lowering ignition timing:

    (+) Adding Knock Retard = Reducing Timing. PCM is seeing knock.
    (--) Lowering Knock Retard = Increasing Timing. PCM isn't seeing knock.
    __________________________________________________ ________

    2014 Mustang GT Premium. VMP Gen2R Supercharged with an FTI 3000rpm Converter. JLT, BMR, Steeda, Viking, etc.
    Don't fix it if it ain't broken | Maximum effort gets maximum results

  19. #19
    Advanced Tuner
    Join Date
    Sep 2018
    Posts
    213
    Quote Originally Posted by MRRPMBRP View Post
    I've been playing with my dd, torque, & tq inv all evening, I go round & round, I think I have it figured out, then one more tweak, & it goes all to hell. There's gotta be a better way!
    Same here. I really wish I knew the workings and formula.

  20. #20
    Advanced Tuner
    Join Date
    Apr 2015
    Posts
    212
    lol. Same boat.

    Im tuning a Boss 302 with a Kenne Bell 3.6 and mono blade 168 on it. I finally got the surge/oscillation mostly worked out and other than a few minor things I hope to iron out this season, it drives pretty decent for the most part. The TQ error however is off the charts in some areas. However while its having a fit about tq error it still drives fine.

    I tried 99 different methods to tune the TB airflow table and the Tq tables based on the 99 different methods ive seen on these forums. some worked sorta and others didnt. Id get to a point where it would start getting worse using the same methods, or TQ error would reduce but it would still drive like garbage.

    I gave up and started winging it. I literally just kept moving the tq tables up or down and adjusting DD (mostly to soften the transition to boost when the bypass would close because its kinda harsh) until the car drove smooth and would accelerate smooth with a steady throttle. I did reference a few other blower tunes to get a general idea of where i should be, but in large part the tq tables i use in that car are just ones I made up for now and it drives pretty great.