Results 1 to 5 of 5

Thread: What kind of math is this?

  1. #1
    Senior Tuner Cringer's Avatar
    Join Date
    Aug 2021
    Location
    Somewhere smoothing your VVE table
    Posts
    1,111

    What kind of math is this?

    I know there are labeling issues in HPT, but there seems to be some real inconsistency going with the Idle Adapt Prop and Int airflow tables perhaps from GM. Unless this is some kind of hack for the PI controller to make it stable, then IDK what to think.

    Code:
    [ECM] 12196 - Idle Proportional Airflow vs. RPM Error: Proportional idle airflow adjustment. When Error RPM is negative (RPM high) then values are subtracted from idle airflow, otherwise values are added.
    [ECM] 12195 - Idle Integral Airflow vs. RPM Error: Integral idle airflow adjustment. When Error RPM is negative (RPM high) then values are subtracted from idle airflow, otherwise values are added.
    When we look at these tables we see...
    The Int table has positive numbers in the negative RPM range as expected to make the subtraction operation work logically (to reduce air).
    The Prop table has negative values in the negative RPM range. Subtracting a negative would be the same as adding a positive and thus increasing the airflow which would increase the airflow and make the high RPM even worse.

    basic math.jpg

    <<< A standard approach will give you standard results >>>

    My Free Tuning Software:

    VVE Assistant [update for v1.5]
    MAF Assistant
    EOIT Assistant

  2. #2
    Senior Tuner
    Join Date
    Jul 2020
    Location
    VIC Australia
    Posts
    1,515
    Prop is a gain, whilst int is an adder in my old text books for pid control

  3. #3
    Senior Tuner Cringer's Avatar
    Join Date
    Aug 2021
    Location
    Somewhere smoothing your VVE table
    Posts
    1,111
    Quote Originally Posted by hjtrbo View Post
    Prop is a gain, whilst int is an adder in my old text books for pid control
    There are many different styles, algorithms and implementations of PID controllers though. Hard to say what is going on inside the ECM. I wish I had the source code!

    <<< A standard approach will give you standard results >>>

    My Free Tuning Software:

    VVE Assistant [update for v1.5]
    MAF Assistant
    EOIT Assistant

  4. #4
    Senior Tuner
    Join Date
    Aug 2010
    Location
    Detroit, MI
    Posts
    1,101
    Proportional is indexed by the RPM error; integral is indexed and multiplied by the RPM error, as well as the loop timestep. Writing it out shows that it doesn't matter where the sign goes, so long as it is correct.

  5. #5
    Senior Tuner Cringer's Avatar
    Join Date
    Aug 2021
    Location
    Somewhere smoothing your VVE table
    Posts
    1,111
    Quote Originally Posted by smokeshow View Post
    Proportional is indexed by the RPM error; integral is indexed and multiplied by the RPM error, as well as the loop timestep. Writing it out shows that it doesn't matter where the sign goes, so long as it is correct.
    OK so the negative Int value multiplied by the negative RPM produces a positive value...which then matches the Prop sign value. Interesting.
    Last edited by Cringer; 06-15-2024 at 03:16 PM.

    <<< A standard approach will give you standard results >>>

    My Free Tuning Software:

    VVE Assistant [update for v1.5]
    MAF Assistant
    EOIT Assistant