Results 1 to 15 of 15

Thread: lean after reflash

  1. #1

    lean after reflash

    i was getting some logs working on this tune and after i stopped, reflashed and started back up it was really lean. i thought maybe my fuel pressure was jacked so drove to shop and checked it but it was actually high at almost 70. so i adjusted that and drove it again and the error was better but then after stopping and looking at log started it up and lean again.

    thinking might be heat soak on the air temp sensor. at one point it was reading IAT around 160 i noticed sitting there so i got on highway and the average error did come down and IAT dropped to around 80s. it was like 70 ambient

    here are some logs showing the swings. is this a heatsoak issue or...?

    one from monday night everything seemed normal. then first today seemed normal. second showing lean and last was trip i got on highway to get some air flowing

    subaradosd33.hpt
    23-05-16 08-00-57monday evening.hpl
    23-05-17 18-33-33first wed.hpl
    23-05-17 18-52-09second wed.hpl
    23-05-17 21-09-08.hpl
    8873.jpg
    Last edited by PAQ; 05-17-2023 at 10:25 PM.

  2. #2
    Senior Tuner
    Join Date
    Mar 2022
    Location
    Indiana
    Posts
    2,903
    Turn fuel trims back on like you've got some sense. Might as well have a carburetor on the thing.

    Consider fabricating an airbox. The only time I've seen IAT's that high and sustained is during a dyno session.

    Work on the VE table.

  3. #3
    Haha ok... so higher rpm and throttle open loop for higher areas on the table and then closed loop and fuel trims enabled for part throttle cruise and idle areas?

    I just got my beefier diff in last night.

    I will see what I can do about an air box

  4. #4
    Senior Tuner
    Join Date
    Mar 2022
    Location
    Indiana
    Posts
    2,903
    No, have closed loop enabled like factory. Let the software decide when to command OL by going into PE in high load areas.

    As far as types of CL there's STFT and LTFT. Some people disable LTFT. The only rationale I've heard is that LTFT's carry over into PE. To me LTFT's correcting PE enrichment, and all the rest, is a good thing.
    Last edited by SiriusC1024; 05-19-2023 at 02:28 PM.

  5. #5
    Ohhh... ok so you tune in closed loop all the time even for wot?

    So then you have two histograms setup one being for fuel trim correction and the other for eq error?

  6. #6
    Senior Tuner
    Join Date
    Mar 2022
    Location
    Indiana
    Posts
    2,903
    You tune with feedback from the O2 sensors which means Closed Loop. PE is open loop in the way it functions, and it depends on the accuracy of the VE table.

  7. #7
    Right...but you leave closed loop enabled even when tuning the wot PE but you use the wideband data for that?

  8. #8
    Senior Tuner
    Join Date
    Mar 2022
    Location
    Indiana
    Posts
    2,903
    Yes leave closed loop enabled. The transition to open loop during PE is automatic. You'd use logged telemetry from the wideband to tune this way.

  9. #9
    Tuning Addict blindsquirrel's Avatar
    Join Date
    Apr 2017
    Location
    Meridian MS
    Posts
    7,562
    Use a filter in your LTFT+STFT graph to exclude anything where EQ isn't = 1.

    Similarly, you can also make a filter in your wideband graph to only include data when EQ is greater than 1. Filters are awesome, but clunky to describe because of the way the PIDs change from controller to controller. My E40 filters wouldn't work on your Gen 3. You just have to build them yourself using the 'New Variable' button, then adding <, >, =, avg,...

  10. #10
    This is amazing stuff! Thank you guys soo much! I didn't really like the no fuel trims and was switching back and forth trying to achieve this.

  11. #11
    Tuning Addict blindsquirrel's Avatar
    Join Date
    Apr 2017
    Location
    Meridian MS
    Posts
    7,562
    You can also filter to exclude idle (either by when idle routines are active if there's a flag for that, or by only including stuff over a certain minimum throttle pedal or TPS), and also transients by not including stuff a certain time before and after a throttle change greater than a certain amount... and removing anything when EVAP purge is active, and and and...

    If it's a channel that can be logged, you can make a filter against it.

    Make a copy of your current graph(s), then put your experimental filters into the new copy. That way you can switch easily between the graph with no filters and the ones with them.. See which filters do what.

  12. #12
    i set up filters and worked on VE table some getting it closer i think and somewhat smoother.

    was playing with MAF last night before laptop died. i believe i have the right things for maf only.

    what you think about this tune progress?
    also screenshots of my filters

    23-05-29 20-23-58.hpl
    Screenshot (5).pngScreenshot (4).pngScreenshot (6).pngScreenshot (7).pngsubaradoclmaf4.hpt
    Attached Files Attached Files
    Last edited by PAQ; 05-30-2023 at 10:03 AM.

  13. #13
    Tuning Addict blindsquirrel's Avatar
    Join Date
    Apr 2017
    Location
    Meridian MS
    Posts
    7,562
    That tune file is the opposite of 'MAF only'. If that's the one used during MAF tuning you'd be taking VE error and applying to the MAF table, and that wouldn't turn out so great.

  14. #14
    not the one that says clsd45+maf...i accidentally added that one... i was using the other one i posted -subaradoclmaf4.. or is that one wrong too?

  15. #15
    also..do those filters look right..they appear to be working right but i am an idiot soo...