Page 1 of 3 123 LastLast
Results 1 to 20 of 41

Thread: NAG1/A580 Part Throttle Shift Scheduling Table Tuning

  1. #1

    NAG1/A580 Part Throttle Shift Scheduling Table Tuning Questions

    Hi all,

    I have a Sprintex supercharged 2013 Jeep Wrangler with the NAG1 trans in which I'm wanting to tune the part throttle shifting on.
    From the PT tables the shifts are scheduled by the Accelerator Pedal Position vs Output Shaft Speed.
    In these tables what is the actual PID used for the Accelerator Pedal Position?

    I have data logged the following to try and determine the shifting behavior:

    Throttle & Accel PID's:
    Accelerator Pedal Position Sensor (V)
    Accelerator Position D (%)
    Commanded Throttle Actuator (%)
    Throttle Position (%)
    Trans Pedal Position (%)

    Trans PID's
    WOT Shift (Flag)
    Upshift (Flag)
    Downshift (Flag)
    Trans Current Gear
    Trans Shift Mode
    Aggression Blend(%)
    Upshift Speed Offset (RPM)
    Downshift Speed Offset (RPM)
    Trans Turbine RPM (RPM)
    Trans Output Shaft RPM (RPM)

    Upon reviewing data logs with very light throttle applications (quasi-steady state to eliminate any other transients) I cannot seem to link the commanded upshift/downshift (flag) and corresponding output shaft speed to any throttle PID. All of them seem to be way out re: 50+ RPMs . The aggression blend is 0.0% so one would assume that the normal scheduling table would be used. On the other hand, the WOT shift table is predictable and commands the shifts once the output shaft speed is reached.

    To add to the confusion, none of the 4 throttle % PIDs I've logged are the same value. Furthermore, the only PID that reaches 100% on WOT is the trans pedal position. I have confirmed that this PID triggers the WOT shift flag, however it comes no where near the values when a shift is commanded.

    What I think I know:
    The Accelerator Position D (%) is derived from the Accelerator Pedal Position Sensor (V) and maxes out at around 91%.
    The Commanded Throttle Actuator (%) follows the Throttle Position (%) but has a -10% offset... They max out at ~76% and 86% respectively (Which one is the actual throttle blade position??).
    The Trans Pedal Position (%) seems to follow the Accelerator Position D (%) however it is scaled from 0 - 100%. Least corresponding PID to the shift scheduling tables.

    Any help would be greatly appreciated.
    Kind regards,
    Dan

    Tune and corresponding datalog attached.
    Attached Files Attached Files
    Last edited by HaasExp; 01-23-2020 at 01:27 AM.

  2. #2
    I done a few more datalogs and have attached some screen captures. Again it seems something is not right with these PT shift schedule tables as the shifts are commanded at a much higher (logged) output shaft speed than what is requested in the table.

    For example, in the following log we have a light throttle, 1-2 shift.
    1-2 upShiftTrace.JPG
    The marker line is at the exact moment the upshift flag is triggered. In the log this occurs at an output shaft speed (OSS) is 658rpm and the maximum that any accel or throttle PIDs register at that time is ~22%. However, this shift should occur at an OSS speed of ~345rpm according to the normal shift table or ~885rpm according to the performance shift table.
    1-2PTshiftTables.JPG
    Is there still some blending going on between the performance and normal shift tables? The only transmission PID I could find and log the says anything about blends is the "aggression blend". it however stays at 0.0% when steady throttle is applied.

    The full datalog is attached. It also shows this odd behavior with other gears shifts.

    Im also still at a loss at to which PID is used as the "Accelerator Pedal Position" in these tables. The only PID that swings between 0-100% is the "Trans Pedal Position".
    A virtually identical question/issue about the PT shift scheduling on a 2013 JK wrangler was previously posted in this thread but very little info was gathered:
    https://forum.hptuners.com/showthrea...ift-Scheduling

    Regards Dan

    transTest.hpl
    Last edited by HaasExp; 01-23-2020 at 01:32 AM.

  3. #3
    Advanced Tuner
    Join Date
    Nov 2017
    Posts
    542
    I'm not familiar with the Nag1 tables.

    Is shift ID loggable? It could be that there is another table somewhere it is referencing.

    What happens if you change the table some? Do shifts follow your change at least somewhat? Or are your changes ineffective?

  4. #4
    Turn the Performance Driving Factor switch to disable.

    Or try taking it out of 4WD.

    4WD may make it reverence the Performance shift tables.

    Then just tune the Normal shift tables. Or make both the Normal and Performance part throttle tables the same.

    In your logs the shifts are basically what the Performance part throttle tables are set too.
    Last edited by Moparmatty; 01-23-2020 at 09:42 AM.
    2006 Chrysler 300C SRT-8
    JBA 3" stainless catless mid pipes
    3" Flowmaster American Thunder stainless cat-back
    Injen CAI
    Innovate dual WBO2 sensor kit
    Custom tuning by me via HPT

  5. #5
    Thanks for the replies so far guys.

    The only PID that comes close to a shift ID is the "Trans Shift Mode" which always has the value of "4WD" in the case of the Jeep wrangler. The only reference to this shift mode is in Torque Convertor -> TCC Pattern where this is set to "Normal" for the 4WD shift mode.

    I haven't played with the tables yet as I cannot predict their behavior, but I shall try disabling the "Performance Driving Factor" switch and make both the Normal and Performance part time shift tables the same to try and get some predictable PT shift characteristics as per Moparmatty's suggestion.

    I guess I should again just start with what should be the most simplest question: What is the actual PID used for the "Accelerator Pedal Position (%)" in these PT shift tables? All these tables are referenced to 0-100% Accelerator Pedal Position, even in stock guise and the only PID I've observed that swings 0-100% is the "Trans Pedal Position (%)" as previously stated. Is this the correct PID?

    Cheers,
    Dan

  6. #6
    So I think I may have made a bit of progress trying to figure out this PT shifting behaviour.
    I've disabled the "Performance Driving Factor" switch and have mimicked the values from the Normal PT shift tables into the Performance PT shift tables.

    In the latest data log I've noticed that the shifts are scheduled very close to the desired "Trans Output Shaft RPM" when the "Upshift Speed Offset" = 0rpm.
    As shown in the following capture, the 3-4 shift is scheduled at a trans output shaft speed of 952rpm at a corresponding trans pedal position of 4%. This is pretty much a match to the desired 950rpm at 0% pedal position requested in the table.
    3-4shift.JPG
    3-4shiftTables.JPG

    Thus it seems that this "Upshift Speed Offset" PID is an adder to the OSS values in the PT tables. Im led to believe that the "Upshift Speed Offset" is determined by the "Driving Factor", which is calculated through the "Term Coefficients" and "Decay Divisors".
    Can anyone confirm this or know the actual relationship between all these variables?
    Is it also possible to log this "Driving Factor"?
    DrivingFactor.JPG

    Cheers,
    Dan

  7. #7
    Scrap the idea of the trans shifting at the desired Accel Pedal/OSS when the upshift/downshift offset speed is 0.
    Shown in the following log is a 1-2 & a 2-3 shift where the OSS = 0rpm.
    2-3shiftOSS=0.JPG

    Here the 1-2 shift occurs at an OSS of 628rpm. This would coincide with an accelerator pedal between 60 - 80% from the PT shift schedule tables. However, as you can see, there are no accel pedal PIDs anywhere near this value.
    In similar fashion the 2-3 shift occurs at an OSS of 893rpm which coincides with a pedal value between 40 - 60% from the tables.
    1-2-3shiftTables.JPG

    What is actually going on here?? Im beginning to believe there must be a bug or a parameter missing in the scanner/editor. The PT shift behaviour simply cannot be explained unless I'm completely missing something here.

    Any help would be greatly appreciated.
    Dan
    Attached Files Attached Files

  8. #8
    have you made any more progress on this? I have a 16 jeep wrangler that's doing the same thing. it has an sps supercharger on it and im struggling with pt shifting from 3-4 it will rev up to 3k and will stay there until I release the pedal. when I'm commanding it to shift at way lower rpms. I made a calculator on excel to translate oss to engine rpm and to mph and it still does whatever it wants. its running the egs53 controller on a a580 transmission. customer is from out of town and came here from another shop that didn't do too much of a job tuning it and I'm stuck cleaning up the trans tune and engine tune.
    2017 mustang gt

  9. #9
    Hi Tadias48,

    Glad to hear I'm not alone. There are also a few other threads on this forum of others having issues with the NAG1 part throttle shift tuning on their Jeep Wranglers. While most upshifts occur at reasonable engine speeds (but still dont follow the pt shift tables) the 3-4 shift is is by far the worst. If around 25% throttle is applied, mine will hold 3rd seemingly forever and will not command an upshift until the pedal is released. This was the main reason I purchased HP tuners in the first place.

    Its almost like the PT tables are not even being used and some other table(s)/parameter(s) that haven't been mapped out are determining the shift.

    Im not sure if this is only related to the NAG1 in the Jeep wrangler platform. I was hoping that some of the Dodge tuners could chime in here to see if this issue extends to other vehicles.

    Unless I'm missing something really simple, it seems like we cannot predictably tune the NAG1 part throttle shifting behavior in these Wranglers through HP tuners.

    What is the best point of contact through HP tuners to try and get this issue resolved?

  10. #10
    Potential Tuner
    Join Date
    Oct 2019
    Posts
    2
    Same situation here (2013 10A JKR, Edelbrock SC, 35's) and I'm running Hemituna's tune. Probably about 95% of it, anyway. And I was having the same stupid crazy shifting issues. Mine would even not want to downshift into 1st gear at stops. And the 3rd gear hang was pure insanity. After tweeking / downloading numerous times, without any progress, I finally tried something.

    I did a throttle re-learn under the VCM Scanner. Open Scanner, connect, go to Special Functions, and it's in the same screen as the "Reset Adaptives". It tells you exactly what to do and when. If it's quiet, you'll hear the throttle body clicking and whirring. I did this (and I reset the Adaptives while I was there, I do this every time I do a download), and then things shifted as they should. The husband's JKR (same setup) is a 6-spd, so he doesn't have all the fun and games with the trans!

    Will this fix yours? Don't know. But by God it helped mine. And straightened out the idle some too. Will it stay that way? Don't know that either, as these things tend to re-learn and may do stupid crap again. I drive mine about once a week, so I'm not constantly fiddling with it. It's worth a try, especially to save your sanity.

    I'm still wondering about a few things, but it appears that I'm not the only one who has these strange issues!

  11. #11
    well damn it never crossed my mind to do a throttle relearn. all I did was lower the shift points by 25% I know its not the proper way to do things but more of a bandaid due to the time constraints. the jeep shifted better and not as harsh going into gear either. customer was satisfied. I will be seeing him back to do a watermeth kit in the next few months.
    2017 mustang gt

  12. #12
    Potential Tuner
    Join Date
    Oct 2019
    Posts
    2
    Hemituna's tune is a great tune, I am very happy with it overall, I just had to do some tweeks due to the 35's I'm running. But the more I tweeked, the worse everything seemed to get... kind of like it was never really overwriting anything. Just a weird experience overall, I can't really find a word that fits. I tried doing extreme cuts and such, and nothing seemed to change it!

    The fact that doing JUST the trans download takes over 2 1/2 minutes, compared to the ECU that takes under 30 sec is a pain, only to find out very quickly that what you just spent the last little while changing didn't do a damned thing, or made it worse.

    But, so far so good with mine. Whenever I do any adjustments now, I just do the Adaptives reset and a Throttle Relearn at the same time. That way I know I'm starting at 'zero', so to speak. I'm just trying to get my torque converter lock / unlock dialed in a tad better, and I'll be happy.

    I'm no expert by any means, but if I find something that worked for me, I'll pass it along in case it works for some other poor soul who is on their last nerve. Even a blind hog will find a truffle every now and then.

  13. #13
    Cheers for the tips teamrm. I must say I have tried the throttle relearn procedure a few times (in an attempt to remove to initial dead pedal feel on these) but I can say that hasn't done anything in terms of trying to get predictable part throttle shifts. It still seems to have a mind of its own when it comes to these shifts. I now know of 5 users on this forum that are experiencing this same issue with the NAG1 in the Jeep Wrangler, so it must be an issue with HP Tuners on this platform. We all can't be doing the same thing wrong when it comes to tuning the part-throttle shifts on these trans.

    I've submitted a ticket with HP tuners and referenced this thread. So hopefully I will get a reply and a solution.

  14. #14
    Tuner in Training
    Join Date
    Sep 2018
    Location
    GA USA
    Posts
    34
    It's not just about the PT schedule, other things need to be adjusted as well.

    I do a lot and I mean a lot of Wranglers almost all are FI. Everything will fall into line if you get the JK out of your head and think of a high horsepower FCA with a NAG1 and how you get them to shift.

  15. #15
    Hi Mag0842, thanks for chiming in.

    I guess what I'm trying to figure out here is how to actually command the shift as requested in the PT schedule tables under light throttle conditions (convertor locked, negligible convertor slip).
    So if I wanted to command a 2-3 upshift in my PT table for an OSS of 600rpm and 20% percent throttle how do I actually do that?

    Currently I have made all my PT tables (Normal/Performance/Winter) the exact same, in terms of accelerator pedal position breakpoints and OSS values, and I have also set performance driving factor switch to disabled.
    But the shift actually gets commanded at an OSS of ~900rpm. Its wayyyy off the what I actually want, a factor of 50%! This occurs even at low acceleration rates (gentle daily driving) where the convertor slip is negligible.

    Any clues as to what other things need adjustment to make my actual PT shifts match my requested PT shift schedules in the tables?

    Cheers,
    Dan

  16. #16
    Still persisting with this issue. I've been in communication with HP tuners about it but so far nothing constructive to report.

    While browsing through numerous threads on here I did come across an early thread by bluegoat06 regarding WA580 and the shift point strategy.
    https://forum.hptuners.com/showthrea...t406143http://
    What I found of interest here were the Pedal % breakpoints and table ID's on his screen shot from the stock TCM tune off his 2013 JK wrangler (same year and model as mine).

    bluegoat06 factoryShiftPoints.JPG

    However, upon opening up the actual tune (where this screenshot presumably originated from that thread) with v4.4.4 the pedal breakpoints and the table ID's do not match....

    bluegoat06 factoryShiftPoints HP v 4.4.4.JPG

    As a comparative example: table parameters screenshot of the original for the 1-2 normal upshift table Vs the 1-2 normal upshift table parameters as read from the tune using VCM editor v4.4.4

    Table ID = 51650________________________________________Table ID = 45993
    Pedal (%) Breakpoints = 0, 6, 20, 31, 42, 50, 96______________Pedal (%) Breakpoints = 0, 20, 42, 62, 81, 92, 100
    Shift Speeds = 300, 360, 480, 645, 1020, 1305, 1605__________Shift Speeds = 300, 360, 480, 645, 1020, 1305, 1605

    The post is dated 11-28-2015 so I dont know what version of VCM editor was used to do the read but why the discrepancy??
    Shouldn't the pedal breakpoints and table ID's still match regardless of what version of VCM editor is used?
    Scaling issue? Different table addresses?

    If I were to ignore the actual pedal breakpoints in my PT tables and overlay them with the much lower breakpoints as per this screenshot, then my PT shifts would be very close to what I'm actually commanding. Currently it seems that the values of the pedal breakpoints in my PT tables (as read by v4.4.4) are far higher than what they actually are in reality (datalogs) eg at an actual pedal of 20% my shifts occur at shift speeds closer to that of whats under the 40-60% pedal breakpoints in the PT tables...

    I've attached bluegoat06's tune from the mentioned thread. Does anyone have an earlier version of HP tuners (circa 2015) to open it and check the PT breakpoints and table ID's for my sanity please?

    Regards,
    Dan

    JK_4_7_trans1_bluegoat06.hpt
    Last edited by HaasExp; 03-01-2020 at 04:45 PM.

  17. #17
    So I managed to open up bluegoat06's tune in question with and earlier version of VCM Editor (VCM Editor 3.0.8).
    When I do this I can replicate all the pedal % breakpoints in his original screenshot.
    bluegoat06_VCM Editor 3.0.8.JPG

    Given the fact that the pedal % breakpoints and the table ID's are different when opening up the exact same tune between VCM Editor versions 3.0.8 and 4.4.4, it surely points to a bug somewhere Either the actual tables are different (given the different ID's) or the breakpoints are being scaled incorrectly.

    Unfortunately, the early version of VCM editor does not seem to be compatible with the MPVI2 and I could not read my current tune with v3.0.8 (nor open them) to compare pedal % breakpoints. Just like bluegoat06's tune I suspect they would be much lower than what v4.4.4 reads. The lower breakpoints as shown in PT shift tables in v3.0.8 would be much more realistic and closer to what the scanner has been reporting.

    Where do I go from here??

    Cheers,
    Dan
    Last edited by HaasExp; 03-04-2020 at 05:04 PM.

  18. #18
    Quote Originally Posted by HaasExp View Post
    With special thanks to spoolboy I managed to open up bluegoat06's tune in question with and earlier version of VCM Editor (VCM Editor 3.0.8).
    When I do this I can replicate all the pedal % breakpoints in his original screenshot.
    bluegoat06_VCM Editor 3.0.8.JPG

    Given the fact that the pedal % breakpoints and the table ID's are different when opening up the exact same tune between VCM Editor versions 3.0.8 and 4.4.4, it surely points to a bug somewhere Either the actual tables are different (given the different ID's) or the breakpoints are being scaled incorrectly.

    Unfortunately, the early version of VCM editor does not seem to be compatible with the MPVI2 and I could not read my current tune with v3.0.8 (nor open them) to compare pedal % breakpoints. Just like bluegoat06's tune I suspect they would be much lower than what v4.4.4 reads. The lower breakpoints as shown in PT shift tables in v3.0.8 would be much more realistic and closer to what the scanner has been reporting.

    Where do I go from here??

    Cheers,
    Dan
    Do yourself a favour and download the latest BETA software. Whenever you're having an issue with HP Tuners software always give the newest BETA software a try. They fix their mistakes all the time without anyone knowing. I didn't see the issue because I was always looking at your logs and tunes in 4.5.???? software.

    When I open the last tune you posted in latest BETA software (4.5.1409) shift tables look just like the last table pictures you posted. I also compared the breakpoints of a stock tune and they are the same. I also compared both in 4.4.4 software and they are not the same as 4.5.1409 which to me means HP Tuners had something messed up.

    As an FYI for you. You can change the breakpoints to suit your liking. I spent a lot of time doing this on my SRT8 300 a couple of years ago. Made the transmission shift at the exact throttle position and output shaft speed I wanted so I didn't have to interpolate anything. I looked at the throttle positions I routinely hit around town for my driving habits and set up the break points for where my foot just naturally wanted to sit. I then looked at ground speed I wanted each shift to happen at and then correlated that the the exact output shaft speed. I put the output values under the appropriate breakpoint. As an example, I left the first breakpoint at 0%, the 2nd break point at 20%, the 3rd one at 28% and the 4th one at 35%. I set them that way because I never drove under 20% throttle, so why have any breakpoints lower than that? Taylor them to your driving style. I made all the break points in the Part Throttle upshift and downshift tables the same. Other than the first breakpoint being set to 0%, I see no reason to have any breakpoints lower than where the vehicle is driven.

    BUT in your case..... If the breakpoints weren't mapped right, the numbers you changed them too might not have made any sense. Now that it appears they have changed something. Download the latest BETA software and read out the tune currently in the Jeep and try modifying and flashing that file.

    As another check. I looked at both my the stock and modified TCM tunes for my SRT8 300 in 4.4.4 and 4.5.1409 and they were both the same. So that tells me they don't/didn't have all the calibrations labelled/mapped wrong.
    Last edited by Moparmatty; 03-01-2020 at 08:09 PM.
    2006 Chrysler 300C SRT-8
    JBA 3" stainless catless mid pipes
    3" Flowmaster American Thunder stainless cat-back
    Injen CAI
    Innovate dual WBO2 sensor kit
    Custom tuning by me via HPT

  19. #19
    Advanced Tuner
    Join Date
    Nov 2017
    Posts
    542
    Here is a trick I just figured out for dialing in shift points - but it only works if you have the ability to shift manually.

    Start logging current gear, APP and OSS in scanner.

    Drive around shifting manually the way you want your trans to shift. Up and downshifts. Then go back and enter the OSS speeds you logged into the appropriate shift tables.
    Last edited by spoolboy; 03-01-2020 at 09:01 PM.

  20. #20
    Appreciate the input to this rather fustrating issue guys!

    Quote Originally Posted by Moparmatty View Post
    Do yourself a favour and download the latest BETA software. Whenever you're having an issue with HP Tuners software always give the newest BETA software a try. They fix their mistakes all the time without anyone knowing. I didn't see the issue because I was always looking at your logs and tunes in 4.5.???? software.

    When I open the last tune you posted in latest BETA software (4.5.1409) shift tables look just like the last table pictures you posted. I also compared the breakpoints of a stock tune and they are the same. I also compared both in 4.4.4 software and they are not the same as 4.5.1409 which to me means HP Tuners had something messed up.
    I believe I've just experienced this first hand. Only just downloaded the BETA software on Friday which gave the same breakpoints and table ID's as 4.4.4 (no difference). Low and behold I just checked the version at it was 4.5.1403. Go back to the HP tuners site and download the BETA suite again (Monday) and I got the 4.5.1409! Sneaky!

    Can confirm that when using BETA 4.5.1409 the table IDs and pedal breakpoints from bluegoat06's reference tune are the same as what 3.0.8 reads. I can also confirm that the tunes from my Wrangler that I read with 4.4.4 have different table IDs and breakpoints when I open them up in 4.5.1409. These table IDs now match that of 3.0.8. I think we might be onto a winner!

    I will do a read with BETA 4.5.1409 and see if everything is all aligned with what the scanner logs to confirm.


    Quote Originally Posted by Moparmatty View Post
    As an FYI for you. You can change the breakpoints to suit your liking. I spent a lot of time doing this on my SRT8 300 a couple of years ago. Made the transmission shift at the exact throttle position and output shaft speed I wanted so I didn't have to interpolate anything. I looked at the throttle positions I routinely hit around town for my driving habits and set up the break points for where my foot just naturally wanted to sit. I then looked at ground speed I wanted each shift to happen at and then correlated that the the exact output shaft speed. I put the output values under the appropriate breakpoint. As an example, I left the first breakpoint at 0%, the 2nd break point at 20%, the 3rd one at 28% and the 4th one at 35%. I set them that way because I never drove under 20% throttle, so why have any breakpoints lower than that? Taylor them to your driving style. I made all the break points in the Part Throttle upshift and downshift tables the same. Other than the first breakpoint being set to 0%, I see no reason to have any breakpoints lower than where the vehicle is driven.
    Very valid points! I have been implementing the same method as spoolboy suggested so I can go by the feel of the vehicle and just set the actual breakpoints and OSS from the scanner to what feels right.

    However, this issue was just making it impossible.
    Quote Originally Posted by Moparmatty View Post
    BUT in your case..... If the breakpoints weren't mapped right, the numbers you changed them too might not have made any sense.
    The Pedal% OSS relationship in these tables never made sense!

    Sidenote-> Moparmatty: Have you ever noticed neighboring breakpoints having the exact same value in a stock read? I've found a few down shift tables where the pedal % has been the same as the neighboring one although the OSS are vastly different. Odd!
    sameBreakpoints.JPG


    Hopefully everything will be sorted now and I can actually tune these shift points correctly (and have them actually match within reason in the scanner).

    Thanks again,
    Dan