Results 1 to 9 of 9

Thread: commanded air fuel vs actual

  1. #1
    Advanced Tuner
    Join Date
    Apr 2013
    Posts
    210

    commanded air fuel vs actual

    i am going to quote something i read in this forum by 5 liter eater

    "Closed loop is the PCM reading the rich/lean from the narrow band O2's and adjusting the fuel trims to account for error. Disable closed loop. You need to see the raw AFR error without the PCM correcting for it. Do this by maxing out the closed loop enable ECT VS: IAT. Also make sure the STFT Openloop is set to disable. Then clear your fuel trims via VCM Controls. Set your PE to whatever you want to command across the board (IE: 13:1 = 1.13 in all columns).

    Then, if you wish to tune the VE, disable the MAF by setting the engine diag,airflow, MAF Fail High frequency to zero and making sure the P0103 DTC is set to MIL on first error. Set up your histo to plot the AFR error against the VE table, log, paste special, multiply by %, flash, re-log. If you have a secondary VE table you will need to copy every other row from the primary to the secondary after you make changes.

    Once you're happy with VE, re-enable the MAF (max fail high = stock) and set the dynamic airflow disable very low (below idle RPM), this will ensure you are not referencing the VE. Set up a histo to plot the AFR error against the MAF table. Log, copy, paste special, multiply by % into the MAF table, flash, re-log.

    Once you're done you can re-enable closed loop to allow the PCM to make slight adjustments."


    I completely understand what he is saying, but if the PID afr error is giving the actual error to lambda, then what is the need for a wideband? is the wideband input, in this case VE-Wideband o2 histogram, needed just to verify if the afr error is actually true?

    I am somewhat confused on how the preset PID afr error determines the error percentage without the input of a wideband?


    to keep it short, I am assuming what 5LE said above is right since he is a well known member here, how is it that the PCM can determine the AFR error deviated from lambda without the use of a wideband in open loop??? or is the AFR error derived through the use of a wideband and five liter eater didnt specify above?


    I keep hearing this all around of Commanded vs actual AFR ratio..how is that setup in a histogram to begin with?


    Thanks guys

  2. #2
    Advanced Tuner
    Join Date
    Sep 2011
    Location
    Jax, Fl
    Posts
    216
    You need a wideband to get AFR (or better yet lambda) error. Plug the wideband into the cable and set up the transfer function. Log the commanded AFR in the table and make a pid of actual/commanded. If you want to create a pid that you can just copy, paste special >multiply by % then make it (actual/commanded)x100-100 and that will give you a number similar to fuel trims.
    2006 Holden Monaro I Phantom Black Metallic I Red I One of 416
    L92/L76 I Streetsweeper HTR I SLP LT I LoudMouth 2
    JHP V2 I Powdercoated Black 18s I Kyne Splitter


    2001 GTP I Stock DD

  3. #3
    Advanced Tuner
    Join Date
    Apr 2013
    Posts
    210
    Thank you that is what I wanted to know! So the afr error uses the reading from the wideband, since it is preset, what I would have to do is assign the sensor air fuel ratio to the wideband when setting it up custom? Also when tuning the ve and maf do we want to hit 14.7 all across the board? Or in high kpa occassions we adjust the table to read rich? I'm assuming I set the ve table to 14.7 all across and then take care of high loads with PE correct?

  4. #4
    Advanced Tuner
    Join Date
    Sep 2011
    Location
    Jax, Fl
    Posts
    216
    You need to do a lot more reading.

    I don't use the preset one I just make a lambda error pid using the formula I posted before. And as long as you are doing AFR error and have a decent WB it doesn't matter too much if you are commanding stoich or in PE it will give you a percent error to correct off of ie 14.5/14.7 = -1.36% error 12.3/12.5 = -1.6% error. Commanded AFR is a pid you can add to the table so don't just put 14.7 in your pid use commanded. I do cruising and as much non PE as I can before doing a few WOT runs to get it all dialed in.
    2006 Holden Monaro I Phantom Black Metallic I Red I One of 416
    L92/L76 I Streetsweeper HTR I SLP LT I LoudMouth 2
    JHP V2 I Powdercoated Black 18s I Kyne Splitter


    2001 GTP I Stock DD

  5. #5
    Advanced Tuner
    Join Date
    Apr 2013
    Posts
    210
    Thank you for replying

    I am going to try and keep this as simple as I can. What I am confused about is the following: I am familiar with stft ltft tuning for bolt on vehicles, this is the first time I use a wideband with hptuners, but have tune other japanese vehicles with other systems using a wideband. I am familiar with the concept of lambda.

    My problem is that the afr error will give me the percentage to change just like the ltft method but this is more accurate and is also in open loop. How would I go about tuning either the ve table or maf table without interference from pe enrichment? Do I disable pe enrichment and try to hit as many cells as possible to stoich? You mentioned that even if I am in PE I can still modify the ve table accordingly, is it that easy? For example I am in 80 kpa with a tps value of 60percent, at 1500 rpm, the commanded is 12.8 afr and my error is 3 percent. Just an assumption, so I modify the cell to 3 percent increase so the commanded is the same as actual. But what if I part throttle on that same cell..wouldn't the commanded change back to stoich since I am not in PE and revert the ve table back to a lower value because I originally change the ve table to what the measure was.

    Sorry if this is confusing.

  6. #6
    Advanced Tuner
    Join Date
    Sep 2011
    Location
    Jax, Fl
    Posts
    216
    You are correcting the airflow error, the VE is the airflow model. In PE and normal cruising at the same axis the engine is really flowing the same amount of air just adding different amounts of fuel. You are correcting the airflow model so it works whether you are using more fuel or not. Make sense?

    A simple example is say you have 100 units of air flowing at the load and rpm cell you said. If not in PE it takes this number and divides by 14.7 and adds 6.8 units of fuel, then in PE at same load and rpm it takes that 100 divides it by 12.8 (for example) and adds 7.8 units of fuel. The airflow is the same and any error at that load should be the same in closed loop or PE, the number will still be 100. If there is a +3% error you add 3% to the 100 making it 103. Now that's really simple and doesn't take into account temp and everything but that should be the same too. Also remember the VE is steady state so hitting the load/rpm cells needs to be done slowly not just revving the engine.
    Last edited by Kyne; 04-22-2013 at 05:36 PM.
    2006 Holden Monaro I Phantom Black Metallic I Red I One of 416
    L92/L76 I Streetsweeper HTR I SLP LT I LoudMouth 2
    JHP V2 I Powdercoated Black 18s I Kyne Splitter


    2001 GTP I Stock DD

  7. #7
    Advanced Tuner
    Join Date
    Apr 2013
    Posts
    210
    I wish I could invite you for a beer buddy I completely understand the pe and ve now! Since the PE is a multiplier only, as long as we have the ve in the correct airflow, the pe will be close to 0 error! No matter if I am in steady state or wot the airflow reading is the same but depending on my pe settings, the afr will change accordingly to the ve . If I have a crappy ve map that is running rich I will run even richer when I hit pe because the ve is not corrected in airflow! Thank you, the same applies to maf? Wow that afr error histogram makes it TOO easy huh?

  8. #8
    Advanced Tuner
    Join Date
    Sep 2011
    Location
    Jax, Fl
    Posts
    216
    As long as your injector data is correct in theory it should be fine. Lambda makes it easier and way cleaner.
    2006 Holden Monaro I Phantom Black Metallic I Red I One of 416
    L92/L76 I Streetsweeper HTR I SLP LT I LoudMouth 2
    JHP V2 I Powdercoated Black 18s I Kyne Splitter


    2001 GTP I Stock DD

  9. #9
    Advanced Tuner
    Join Date
    Apr 2013
    Posts
    210
    Right this is considering the fuel has been correctly adjusted to begin with. Is there a repository of fuel injector data or is that supplied by the mfg? Thanks a lot bud