Page 1 of 7 12345 ... LastLast
Results 1 to 20 of 135

Thread: How do YOU smooth AFR with the MAF table...

  1. #1
    Advanced Tuner
    Join Date
    Jul 2009
    Posts
    509

    How do YOU smooth AFR with the MAF table...

    I'm trying to get a better technique when it comes to smoothing my AFR to get within .2 of my commanded AFR. I know I'm new to this so I'm sure there is a better method to getting things in line quicker.

    What I've been doing is logging a third gear run from ~2000 RPMs to ~red line.
    I'll then look at my AFR from that log and I'll note any lean or rich points and their respective MAF Hz. I'll then go into the MAF table and increase or reduce the Hz for those specific cell. I have got my AFR pretty close but it's taken me over a week and at least 2 to 3 logged runs per day. I don't have a dyno and I have a full time job. Any tips that anyone could suggest that would make this process more efficient would be appreciated. I want to try the smoothing but I'm not sure if it I should do small portions or large portions.
    2016 GMC Sierra 1500 6.2L

  2. #2
    Tuner
    Join Date
    Dec 2008
    Location
    san diego
    Posts
    71
    Disclaimer: I am no expert and maybe one can chime in and let me know if I am full of it......

    Here is what I am doing for fine tuning the AFR.

    The MAF is not perfect and at certain air flows it may not read just right Vs the required amount of fuel. Most everyone I have talked to wants a nice smooth MAF curve, me to until I saw that I was going lean at WOT at ~9375 frequency. To fix this I need a bump in the MAF curve, now it is no longer smooth! So, first tune the MAF to get it as close as possible, which I believe you have done. Then record the MAF freq Vs AFR. By doing this it will tell you what the AFR is per MAF frequency cell. Repeat several times to make sure the numbers are repeatable and then go into the MAF table and adjust the numbers by the percent it is off.........one cell at a time. The bigger the # the more fuel. This I believe will get you really dialed in but the MAF curve will be lumpy. Maybe hit smooth once? More than this it will start washing out the change. This will be something you may have to play with before you get it just right.

    What do you think experts?
    *************************
    Scott in San Diego with a L-67 Super Stalker.
    Mods in the next month or so, yeah its a moving target!
    Gen 5 blower, N* throttle body, LQ4 MAF,3.4 pulley for starters.

  3. #3
    Senior Tuner DSteck's Avatar
    Join Date
    Mar 2008
    Location
    St. Louis, MO
    Posts
    4,859
    Set up a custom PID that is (Actual AFR - Commanded AFR) / Commanded AFR * 100. I can't remember if you actually need the * 100 or not, but this will give you AFR error. Log this against MAF Hz, and apply it just like you would with fuel trims. I had my AFR dialed in after 3 highway pulls. It probably won't end up as a perfect 3rd order polynomial fit... but this is the real world, and not a theoretical one.

    DSX Tuning - Authorized HP Tuners Dealer
    http://www.dsxtuning.com
    http://www.facebook.com/dsx.tuning
    Just say no to bull s***.
    IF YOU WANT HELP, POST A FILE!

  4. #4
    Senior Tuner
    Join Date
    Jun 2005
    Location
    Southern FL
    Posts
    2,044
    One of the problems with letting the scanner build you an error% is that you'll usually end up with an erratic/etchy MAF curve. This is why I do everything myself based on the log.
    Formerly known as RWTD

    Toys: '22 Tesla Model S Plaid / '20 Chevy Duramax / ?20 Sea-Doo RXT-X (2)

  5. #5
    Senior Tuner DSteck's Avatar
    Join Date
    Mar 2008
    Location
    St. Louis, MO
    Posts
    4,859
    Quote Originally Posted by RWTD View Post
    One of the problems with letting the scanner build you an error% is that you'll usually end up with an erratic/etchy MAF curve. This is why I do everything myself based on the log.
    I use a scanner-built error to get me in the ballpark, but I go back and fine tune it manually. My car shot a straight 12.5 AFR across the board, but my MAF curve doesn't look like a perfect mathematical model.

    DSX Tuning - Authorized HP Tuners Dealer
    http://www.dsxtuning.com
    http://www.facebook.com/dsx.tuning
    Just say no to bull s***.
    IF YOU WANT HELP, POST A FILE!

  6. #6
    Advanced Tuner
    Join Date
    Jul 2009
    Posts
    509
    I'm going through the log and analyzing each frame to see what the AFR is vs. the Hz. If I see a lean or rich spike I'll note it. Is it uncommon for the Hz to bounce around a little? What does it mean when the Hz reads a hair lean in one spot but later in the run (a few frames further) it will read slightly different but it's close to the same Hz level?
    2016 GMC Sierra 1500 6.2L

  7. #7
    Advanced Tuner
    Join Date
    Feb 2006
    Location
    Rancho Santa Margarita, CA
    Posts
    905
    Same issue here, an i do it like dsteck has indicated. the variances i'm seen from weather like i said per you other post, is nothing but this, i don't know if it's related to a jagged out MAF reading, or WB-HPT ground slope settings, but it's annoying as crap, i'm so anal that i won't stop touching and re-touching the MAF until the error is 0% everywhere , but then the afr readings are not exactly spot on ins some areas, at the same hz, it could lean, and later on, same hz goes rich.

    Like for instance here, the histogram says 0, but the afr on that spot is 12.19 from a 12.41 commanded.



    That was yesterday, today i'm testing new timing, 1* or 2* in some areas, untouched MAF tables, and the fuel didn't change that much, although i'd love to see more steady values, i think it could be related to MAF readings, notice how the MAF HZ readings are not that steady, or maybe grounding slop issues ???



    In contrast, check this afr readings from the last GTO i tuned, heads/cam,full exhaust yada yada, (last 2 squares, this was a launch fack up ), of course don't even pay attention to the histogram because is a 1/4 mile run, but notice 3rd and 4th gear at the very end of the graph and how steady the AFR looks , the guy is using a NGK WB, i think hooked up straight to the battery with an inline fuse, i don't know the details on how he grounded the WB, but i'll love to be able to read an AFR like that from my car

    Last edited by bluegoat06; 09-02-2009 at 01:17 PM.

  8. #8
    Senior Tuner
    Join Date
    Jun 2005
    Location
    Southern FL
    Posts
    2,044
    Again, why I do everything myself based on the logged data.
    Formerly known as RWTD

    Toys: '22 Tesla Model S Plaid / '20 Chevy Duramax / ?20 Sea-Doo RXT-X (2)

  9. #9
    Advanced Tuner
    Join Date
    Feb 2006
    Location
    Rancho Santa Margarita, CA
    Posts
    905
    RWTD, how you deal with the situation where at the same HZ, the afr is lean in one point,lets say 5800rpm, and then rich at 6400rpm, same hz cell ? that's what drives me nuts everytime, and everytime using the logs, not even taking the histogram into account, i'll end up over correcting, i started a brand new MAF hi table yesterday from my stock table, and took me around 5-6 runs to correct it, using NB's up to 7800hz, and WB from there up.

    This is how my MAF Hi table looks today
    Last edited by bluegoat06; 09-02-2009 at 01:34 PM.

  10. #10
    Advanced Tuner
    Join Date
    Jul 2009
    Posts
    509
    Your's looks smoother than mine. Here is mine. Is it more important to be smooth in the chart or "in line" on the wideband log?

    2016 GMC Sierra 1500 6.2L

  11. #11
    Advanced Tuner
    Join Date
    Feb 2006
    Location
    Rancho Santa Margarita, CA
    Posts
    905
    I always thought it was really important to have a smooth MAF table, same with VE, so i learned to hand smooth, i'll look at the Histo first to get me close, then i'll go by the whatever the logs says, looking at afr vs commanded, i don't even use the afr error PID, just a quick look between commaded and actual afr, and how much i need to add/remove from the maf, looking at the MAF error and actual AFR reading from the log, to get to my commanded afr, same as if you were the using the AFR error PID, is just that i didn't know how to use it before, and got used to my method, plus i can hand smooth now looking at the histo and logs with my eyes closed. never used the polynomial smoothing function in the editor.

    For those portions on the logs where at same HZ's, the afr changes from lean to rich or viceversa, i use the count function instead of the average, to see how many hits i had in that cell. then using the chart, i'll know, if it went lean or rich, whichever afr value have the most hits toward lean or rich wins, ex, if i had two hits at around 12.4's afr and only one at 12.16 for the same hz, i'll leave that cell alone since my commanded is 12.5, if it was the opposite, two 12.16's and one 12.43, then i'll remove close to the full % of maf error % to get closer to 12.5 from 12.16, if i use the entire afr error% PID value to adjust, i'll overshoot the MAF table up and down. it's like doing a paste half%, but sometimes i go less or more than half %, and the reason i prefer to do this manually when fine tuning those little spots.
    Last edited by bluegoat06; 09-02-2009 at 02:40 PM.

  12. #12
    Senior Tuner DSteck's Avatar
    Join Date
    Mar 2008
    Location
    St. Louis, MO
    Posts
    4,859
    Keep in mind that when you see a cell in the histogram for 9000Hz or something, that it's actually including data taken from 8900 to 9100 or so.

    DSX Tuning - Authorized HP Tuners Dealer
    http://www.dsxtuning.com
    http://www.facebook.com/dsx.tuning
    Just say no to bull s***.
    IF YOU WANT HELP, POST A FILE!

  13. #13
    Advanced Tuner
    Join Date
    Jul 2009
    Posts
    509
    I went in and smoothed my MAF curve. I'm going try this and see if it helps smooth my AFR.
    2016 GMC Sierra 1500 6.2L

  14. #14
    Senior Tuner
    Join Date
    Jun 2005
    Location
    Southern FL
    Posts
    2,044
    Quote Originally Posted by bluegoat06 View Post
    RWTD, how you deal with the situation where at the same HZ, the afr is lean in one point,lets say 5800rpm, and then rich at 6400rpm, same hz cell ?
    Well, first off, what CAI do you have on your goat? Most of the aftermarket CAIs on the goats cause HUGE turbulence through the MAF (especially that new POS Vararam that's out for it!). I'd love to see a log of your WOT MAF data.

    Also, what injectors are you running, and if you are running aftermarket, do you have the proper calibration data in for them?
    Formerly known as RWTD

    Toys: '22 Tesla Model S Plaid / '20 Chevy Duramax / ?20 Sea-Doo RXT-X (2)

  15. #15
    Advanced Tuner
    Join Date
    Jul 2009
    Posts
    509
    Quote Originally Posted by RWTD View Post
    Well, first off, what CAI do you have on your goat? Most of the aftermarket CAIs on the goats cause HUGE turbulence through the MAF (especially that new POS Vararam that's out for it!). I'd love to see a log of your WOT MAF data.

    Also, what injectors are you running, and if you are running aftermarket, do you have the proper calibration data in for them?
    Here is my WOT log.
    2016 GMC Sierra 1500 6.2L

  16. #16
    Tuner
    Join Date
    Apr 2009
    Location
    Georgia
    Posts
    148
    Very good read in here! I'm in the process of tuning my procharged goat and I have my VE table done. Now I'm working on the Maf cal and it's kicking my butt! lol I'm scared to do a WOT pull because I don't know how far my maf cal is off.

  17. #17
    Senior Tuner
    Join Date
    Jun 2005
    Location
    Southern FL
    Posts
    2,044
    Quote Originally Posted by 99LS1SS View Post
    Here is my WOT log.
    WOW! That is some HORRIBLE MAF turbulence going on there, and exactly what I saw on the Goat with the Vararam setup. They have a crappy filter setup on that CAI. The way they designed it didn't allow for any smoothing of the airflow going into the MAF. It may seem to work nice, due to the placement of the filter away from heat, but it's ultimately junk until they fix the MAF turbulence problem. You'll NEVER get a smooth AFR with that CAI!

    Sorry to be so blunt, bud. I think you should point other owners of the G8 and Goat Vararam setup to my comments here, or those considering purchasing it, and hopefully they will realize what's so wrong with this CAI. Vararam needs to step up to the plate and rectify their design for their customers.

    So you can understand what I'm referring to, convert your file to csv and then open with Excel or OpenOffice, then graph out your MAF Hz in a "line" style mode at WOT only. You'll see what I'm talking about.
    Last edited by RWTD; 09-02-2009 at 10:02 PM.
    Formerly known as RWTD

    Toys: '22 Tesla Model S Plaid / '20 Chevy Duramax / ?20 Sea-Doo RXT-X (2)

  18. #18
    Advanced Tuner
    Join Date
    Jul 2009
    Posts
    509
    Thanks for the input. Now I'm going to go cry because I was the first GXP with a Vararam.....j/k....about the crying part. Now I'm going to have to evaluate my next move.
    2016 GMC Sierra 1500 6.2L

  19. #19
    Advanced Tuner
    Join Date
    Feb 2006
    Location
    Rancho Santa Margarita, CA
    Posts
    905
    Quote Originally Posted by RWTD View Post
    Well, first off, what CAI do you have on your goat? Most of the aftermarket CAIs on the goats cause HUGE turbulence through the MAF (especially that new POS Vararam that's out for it!). I'd love to see a log of your WOT MAF data.

    Also, what injectors are you running, and if you are running aftermarket, do you have the proper calibration data in for them?

    I have the LPE CAI, stock injectors, and stock injector calibration in the tune as well.

    Here is a 3rd gear pull on the freeway this morning.

    Please let me know if you need the config file or the hpt file.

    Thanks again as always.
    Last edited by bluegoat06; 09-03-2009 at 12:48 AM.

  20. #20
    Advanced Tuner
    Join Date
    Jul 2009
    Posts
    509
    Given the fact that there is turbulence, do you think it "hurts" my car or just makes it impossible to achieve a smooth AFR line? Or, are they both related?
    2016 GMC Sierra 1500 6.2L