Anyone unhook the MAF to SD tune first?

[BlackSS]

I've been reading a few tutorials on the net mostly about 3800 V6s and they also have a MAF sensor. It says to unhook the MAF to make the car fail into the VE tables to tune the Long Term Fuel Trims. Then you get a acurate tune when you hook the MAF back up and tune the Maf Calibration Freq. Has anyone tried doing this way of tuning? Also the IAT1 sensor is in the MAF sensor I think so what do we have to do there?

[TheWastedYears]

I know of one person that's doing it so far. I will once I have a better understanding of what measurement is being used on the VE tables.

[ptparker]

I ran all weekend in SD tune... Don't use the 06 Cobalts VE MAP. Use the 05 Cobalt or the 05 redline. It is pretty interesting.

Don't unhook your MAF, just go to the system/diagnostic menu on VCM editor, and set the MAF Fail frequency to 100 from 10000. And disable the SES light for P0103. Don't disable the whole DTC, just the light, or drive around with the SES light on.

The fuel cell values are not the standard 0-16, and 17,18,19. They are weird. There is 12-17, 20, 33-38 cell values (add the fuel cell as a PID). It seems to be more RPM based, and less map based (some of the cells cover the entire map range) other are very narrow. Kind of counter intitutive to what I would expect.

Overall the car drives fine. It is a little more choppy when starting/changing gears, there is a noticable delay between when the throttle opens (i have a manual boost gauge, so I can see it on there) and when the car gets its fuel.

It does seem to cruise richer and for a lot of fun, turn off DFCO, it will pop and gurgle more then an SRT.

Last week, before I switched over to SD, I have my MAF calibrated to be perfect (-1 to +3 LTFT all the way across). Overall I think that a good MAF tune, is going to be smoother, but if I were to ever switch to a turbo (or add one on), I would definatly give a hard look at going streight SD.

[BlackSS]

Ok, today I've been playing around all day with the VE settings. I set the Maf fail freq down and it seems to be running rough like you said. I've been copying and pasting(using add function since the numbers are all in the negatives) the LTFT % histogram. I've done it about 8 times, driving between 15 to 30 minutes a pop. It doesn't seem to be doing anything for the LTFT. I've made sure when I copy and paste everything works out correctly and it does. I am just wondering do I keep doing this over and over untill it's right? My LTFT always seems to go back to -14.1 eventually no matter what. In the histogram at an idle each time shows between -12 and -18. Keep in mind I've done this over and over so I figured it would be getting close by now. Am I just wasting my time or do I keep doing it? I'm not using the original file, I keep using the one from the last flash just to clarify. Thanks.

[ptparker]

Originally Posted by BlackSS

Ok, today I've been playing around all day with the VE settings. I set the Maf fail freq down and it seems to be running rough like you said. I've been copying and pasting(using add function since the numbers are all in the negatives) the LTFT % histogram. I've done it about 8 times, driving between 15 to 30 minutes a pop. It doesn't seem to be doing anything for the LTFT. I've made sure when I copy and paste everything works out correctly and it does. I am just wondering do I keep doing this over and over untill it's right? My LTFT always seems to go back to -14.1 eventually no matter what. In the histogram at an idle each time shows between -12 and -18. Keep in mind I've done this over and over so I figured it would be getting close by now. Am I just wasting my time or do I keep doing it? I'm not using the original file, I keep using the one from the last flash just to clarify. Thanks.

You should be SUBTRACT from the VE tables to compinsate for a negative FT value. Clear your VE back to stock.

Couple of things I don't think you are doing that you should.
1) create a custom PID that is the LTFT + STFT value. I called mine L/STFT.This will give you a more accurate reading then just one or the other.

2) Create a custom Histogram that monitors the L/STFT value instead of just the Long or just the short.

3) On your custom histogram, don't copy any cells with less then 40 (some would say 400, but our cars learn fast) hits. This is not enough data for calculating.

4) Also on your custom histogram, put a filter on that only looks at cells with a TPS < 80%. When you go WOT (or near it), the car goes into PE mode, which disables all your trims anyways. Try to avoid going WOT when doing VE/MAF tuning anyways. It just isn't helpful.

5) Disable DFCO. Again, the FT's are disabled in DFCO. Just set the ECT enable temp to 250 degrees (just to keep it from being enabled). When test driving, try to keep part throttle all the time. I use cruise control on hilly roads. It hits lots of cells. Make sure you use all gears to get all rpm ranges. Turn back on DFCO when you are done tweeking it, as it saves a ton of gas.

6) If you end up with a L/STFT value of -14 for a given cell, SUBTRACT 7% (half of 14) to that cell on the ve table. BABY STEPS is the name of the game. If it is a postitve trim value, ADD from the VE value by half that percent. Remember a negative FT value is better then a positive. A range of zero to -5 is better then 0 to +5. If it is negative the car is taking fuel away. That is better then it having to add fuel, although when all sensors (mainly your O2, MAP, and MAF) are working right, it shouldn't matter.

===============================================
== NUMBER 7 VERY IMPORTANT! ======================
===============================================
AS SOON AS YOU CHANGE ANY VE VALUES, GO TO VCM SCANNER DIAGNOSTICS AND RESET FUEL TRIMS WITH THE CAR IN THE ON POSITION BUT NOT YET STARTED. (I think that is why your values never changed)

[ptparker]

^ might have been the longest post I have ever done (besides copying and pasting crap).

[BlackSS]

Alright I will try that. I was doing the adding part right I guess. Cuz I was copying a - number and adding to the ve TABLE. Example: 10+ -5 = 5. How do I set up a histo with LTFT and STFT and how do I set up the filter for the TPS% and can I set up a filter that will only show cell counts over 40? Thanks for all your help man!

[ptparker]

Originally Posted by BlackSS

Alright I will try that. I was doing the adding part right I guess. Cuz I was copying a - number and adding to the ve TABLE. Example: 10+ -5 = 5. How do I set up a histo with LTFT and STFT and how do I set up the filter for the TPS% and can I set up a filter that will only show cell counts over 40? Thanks for all your help man!

http://www.hptuners.com/help/demos/C...stom%20PID.swf

This will tell you how to make the custom PID and and how to put it in the histogram. THe only difference is use the PID values instead of the SENS values.

Also when you create your custom histo and your PID, make sure you add the new PID to the list of monitored sensors before your start recording.

Paste [PID.17]<80 in the plot filtering. PID 17 is the throttle position. Also change the cell hits required to 40. Maybe don't do the cell hits required part until you want to actually modify the VE. It makes real time monitoring a bitch.

Also don't add the number. Fuel trims are a percentage, so to compinsate for a -8, multiply the value by .96 (100% - (8%/2)). For a +12, multiply by 1.06 (100% + (12%/2)).

[WopOnTour]

Originally Posted by ptparker

http://www.hptuners.com/help/demos/C...stom%20PID.swf

This will tell you how to make the custom PID and and how to put it in the histogram. THe only difference is use the PID values instead of the SENS values.

Also when you create your custom histo and your PID, make sure you add the new PID to the list of monitored sensors before your start recording.

Paste [PID.17]<80 in the plot filtering. PID 17 is the throttle position. Also change the cell hits required to 40. Maybe don't do the cell hits required part until you want to actually modify the VE. It makes real time monitoring a bitch.

Also don't add the number. Fuel trims are a percentage, so to compinsate for a -8, multiply the value by .96 (100% - (8%/2)). For a +12, multiply by 1.06 (100% + (12%/2)).

Nice stuff, but I would either only use ONLY LTFT or perhaps alter/reduce the weighting of the STFT counts as the change in fuel control for each "step" in STFT is generally considered to be something like 1/10 of the authority in actual IPW as compared to a single step in LTFT.
Did that make sense??
STFT= x "baby steps" (actual amount varies) to equal a single LTFT step
Wop

[TheWastedYears]

My head hurts.