help tuning ltft's

i have a 98 grand prix gtp with a basic 3.4 setup, i am needing some help tuning. mods are cold air intake, headers/3" downpipe no cat, no egr, 3.4 pulley, casper fuel pump rewire, al 605 plugs, zzp catback. i am new to tuning and going off of http://www.grandprixforums.net/start...ers-43997.html but i cant seem to figure out how to tune my ltft's. i do everything the guide says to do. disabled pe, reset my fuel trims went on a 30min drive loaded all data and my maf/ltft histogram is 0.0 or 0.1 in all boxes. before i reset my fuel trims they were all over tho. ive read a bunch of different things and all of them say different things. some say only reset fuel trims 1x, and the guide i linked says to reset every time. why are my fuel trims 0.0/0.1? did i setup my histogram wrong? i know my fuel trims will be like alot farther off than that without resetting. my aeroforce reads +8/-8. is there a better way to tune my ltft's?

Try this.....

Make a histogram that averages the LTFT. The top X axis is MAF Hz, 1500,1625,1750,1875,2000,2125,2250,etc.........

Filter the data by eliminating Fuel Trim Cells that are 1 and lower.... [PID.6310]>1

That will eliminate idle and decel fueling from messing with the numbers

Copy the averages and transfer to excel or other equivalent doc editor. Kingsoft SPreadsheets are free.

Take the LTFT numbers and convert to a multiplier. (LTFT*0.01)+1=Multiplier

That will give you 1.01 for 1%, 0.98 for -2%, 1.1011 for 1.1%, and so on.

Once you have the multipier calculated, multiply that against the MAF table used in the PCM file from the log. The modified maf table is the corrected table. Paste that in, flash, reset trims and log to see how well it did.



If your LTFT log is chaotic looking, reduce the extreme points some so you have a reasonable curve. Try filtering data by 3+ max cell hits. That will help refine the data as well.

I would use no less then a 30 minute log. 1 hour is good, 2 hours is fantastic, anything more is just more accuracy and how far you are willing to go.




LTFT vs MAF Hz Histogram
Attachment 39347

Custom PID for LTFT conversion
FORMULA:
([PID.7]*0.01)+1

Conversion sheet
Attachment 39348

****Be sure to point the histogram average to the custom PID created. This way you can simply cut and paste multiply against the MAF table in excel. For some reason I never got the paste special Mullitply % function to work well for me.

This method I use has worked VERY well for me. Of course, you can get more detailed by doing your own fuel trim difference calculations (with fuel trims turned off), log and view the differences at key points and tune that way as well.

If your commanded AFR is 12.2 and the wideband is 12.5, then take the MAF lb/min and divide by the commanded AFR. This is the PCM commanded fuel lb/min. The multiply that against the wideband, this will give you the actual air flow. This works well but you have to be accurate with the Hz and can not floor to 125 w/o offsetting the tune process. That kind of tune is done on a case by case point basis on a log or through a complex excel doc.

Try this.....

Make a histogram that averages the LTFT. The top X axis is MAF Hz, 1500,1625,1750,1875,2000,2125,2250,etc.........

Filter the data by eliminating Fuel Trim Cells that are 1 and lower.... [PID.6310]>1


i did this and still getting 0.0 in all boxes, besides a few with -0.1. i set pe enable to 100% tps, reset the fuel trims before i started scanning, and all of my scans are 30min plus. am i doing something wrong? i am very new to this so its possible im missing something easy and havent noticed.

Sorry for the delay in reply, I'll post up a scanner file you can load. It is possible your tune is that close already.

Load this in the scanner histogram editor.
Attachment 39565

Your LTFT is stuck at 0%, not sure why, but that's a good thing on these V6 cars. Or did you turn off fuel trim learn using the scanner controls? If not, tune the STFT's. Or use LTFT+STFT if your LTFT's are working.

Just use the Maf Freq histogram and use LTFT+STFT as the plot value. No need for excel.

Russ Kemp