Got a supercharger? I need a favor!
08-21-2009, 02:14 PM
#11
2nd Gear Member
Join Date: May 2008
Posts: 213
Your car's computer can learn fuel trims at idle and part throttle. So, unless you completely screw the tuning up, your car should always idle and drive at part throttle at about 14.7, even if your MAF transfer function is pretty wrong.
However, at WOT there is no correction. The engine goes into open loop and takes whatever the MAF gives it, calculates the injector pulse width, and that's that. So any incorrectness in your MAF transfer function is directly translated into your fueling.
Without access to a datalogger, unfortunately you cannot see what correction your computer is applying at low RPMs.
To start, I would modify the top end of your MAF transfer function. You'll have to do some datalogging to figure out what MAF voltage corresponds to what RPM, but for the sake of argument let's say it's 3.0 volts. Add maybe 15% to the MAF airflow at 3.0 volts and higher, and then do the run again. Hopefully, your AF ratios will stay at something more reasonable.
Once you do this a couple of times and find out what AF you have at each voltage point above 3.0 volts, you can divide the AF you got vs the AF you commanded, and then multiply this value by the value in the respective MAF airflow cell.
Another thing you can do is modify your injector high-slope. This controls the fueling when your injector passes the breakpoint, and if you're getting too lean up top you can decrease this value to give yourself more fuel.
Finally, look at your Stabilized Open-Loop Fuel Table (SOLFT). You want to make sure your commanded fuel trim is the same as what you want. If you haven't modified this table, then there is a chance that the old values are too lean, and your computer is actually doing everything right (or at least isn't as wrong as it seems it is). If you want the AF at WOT to be 11.5:1, then make sure that is the commanded AF in the upper load cells of that table.
Hope this helps! And good luck finding the problem. Without a datalooger, it's going to be difficult.
However, at WOT there is no correction. The engine goes into open loop and takes whatever the MAF gives it, calculates the injector pulse width, and that's that. So any incorrectness in your MAF transfer function is directly translated into your fueling.
Without access to a datalogger, unfortunately you cannot see what correction your computer is applying at low RPMs.
To start, I would modify the top end of your MAF transfer function. You'll have to do some datalogging to figure out what MAF voltage corresponds to what RPM, but for the sake of argument let's say it's 3.0 volts. Add maybe 15% to the MAF airflow at 3.0 volts and higher, and then do the run again. Hopefully, your AF ratios will stay at something more reasonable.
Once you do this a couple of times and find out what AF you have at each voltage point above 3.0 volts, you can divide the AF you got vs the AF you commanded, and then multiply this value by the value in the respective MAF airflow cell.
Another thing you can do is modify your injector high-slope. This controls the fueling when your injector passes the breakpoint, and if you're getting too lean up top you can decrease this value to give yourself more fuel.
Finally, look at your Stabilized Open-Loop Fuel Table (SOLFT). You want to make sure your commanded fuel trim is the same as what you want. If you haven't modified this table, then there is a chance that the old values are too lean, and your computer is actually doing everything right (or at least isn't as wrong as it seems it is). If you want the AF at WOT to be 11.5:1, then make sure that is the commanded AF in the upper load cells of that table.
Hope this helps! And good luck finding the problem. Without a datalooger, it's going to be difficult.
08-21-2009, 06:48 PM
#12
5th Gear Member
Thread Starter
Join Date: Aug 2005
Location: Whitehall, Michigan
Posts: 2,638
Your car's computer can learn fuel trims at idle and part throttle. So, unless you completely screw the tuning up, your car should always idle and drive at part throttle at about 14.7, even if your MAF transfer function is pretty wrong.
However, at WOT there is no correction. The engine goes into open loop and takes whatever the MAF gives it, calculates the injector pulse width, and that's that. So any incorrectness in your MAF transfer function is directly translated into your fueling.
Without access to a datalogger, unfortunately you cannot see what correction your computer is applying at low RPMs.
To start, I would modify the top end of your MAF transfer function. You'll have to do some datalogging to figure out what MAF voltage corresponds to what RPM, but for the sake of argument let's say it's 3.0 volts. Add maybe 15% to the MAF airflow at 3.0 volts and higher, and then do the run again. Hopefully, your AF ratios will stay at something more reasonable.
Once you do this a couple of times and find out what AF you have at each voltage point above 3.0 volts, you can divide the AF you got vs the AF you commanded, and then multiply this value by the value in the respective MAF airflow cell.
Another thing you can do is modify your injector high-slope. This controls the fueling when your injector passes the breakpoint, and if you're getting too lean up top you can decrease this value to give yourself more fuel.
Finally, look at your Stabilized Open-Loop Fuel Table (SOLFT). You want to make sure your commanded fuel trim is the same as what you want. If you haven't modified this table, then there is a chance that the old values are too lean, and your computer is actually doing everything right (or at least isn't as wrong as it seems it is). If you want the AF at WOT to be 11.5:1, then make sure that is the commanded AF in the upper load cells of that table.
Hope this helps! And good luck finding the problem. Without a datalooger, it's going to be difficult.
However, at WOT there is no correction. The engine goes into open loop and takes whatever the MAF gives it, calculates the injector pulse width, and that's that. So any incorrectness in your MAF transfer function is directly translated into your fueling.
Without access to a datalogger, unfortunately you cannot see what correction your computer is applying at low RPMs.
To start, I would modify the top end of your MAF transfer function. You'll have to do some datalogging to figure out what MAF voltage corresponds to what RPM, but for the sake of argument let's say it's 3.0 volts. Add maybe 15% to the MAF airflow at 3.0 volts and higher, and then do the run again. Hopefully, your AF ratios will stay at something more reasonable.
Once you do this a couple of times and find out what AF you have at each voltage point above 3.0 volts, you can divide the AF you got vs the AF you commanded, and then multiply this value by the value in the respective MAF airflow cell.
Another thing you can do is modify your injector high-slope. This controls the fueling when your injector passes the breakpoint, and if you're getting too lean up top you can decrease this value to give yourself more fuel.
Finally, look at your Stabilized Open-Loop Fuel Table (SOLFT). You want to make sure your commanded fuel trim is the same as what you want. If you haven't modified this table, then there is a chance that the old values are too lean, and your computer is actually doing everything right (or at least isn't as wrong as it seems it is). If you want the AF at WOT to be 11.5:1, then make sure that is the commanded AF in the upper load cells of that table.
Hope this helps! And good luck finding the problem. Without a datalooger, it's going to be difficult.
WOT Air Charge Correction Factor - Default is 1, I set to 2 (maximum)
base_fuel_table - commanded lambse for 70% TPS and above was 12.8 (I think the value was .83), I filled in 12.0 at 4000 RPM, 11.6 at 5000 RPM and 11.0 at 6000 RPM.
I also set the MAF.ia to a flat 5% richer - though I don't need it and it can go back to stock.
I FINALLY GOT SOME FULL PULLS IN! I'm only hitting 6-7 psi at the high end, which seems pretty reasonable given the pulley size, how procharger's bypass valve doesn't 100% seal, and how there's a high chance I have belt slip (again, due to procharger's HORRIBLE kit design - you get almost no articulation out of the stock belt tensioner).
But since those two changes I'm getting AFRs that match what the ECU is commanding via lambse, or maybe a little richer. I'm guessing a key part of hitting that mark was the change in the WOT charge correction factor - 1.0 being atmospheric. The MAF can't really differentiate the difference between pressures.
So far it's beautiful, and thank you for the suggestions!
08-21-2009, 10:24 PM
#13
3rd Gear Member
Join Date: Aug 2008
Location: Oklahoma
Posts: 912
Cool, thanks.
I know the CBAZA strategy has multiple methods of calculating load, by just changing the value of a scalar, you can switch it between load based on TPS voltage, calculated VE, or load as a percentage of possible engine load (based off injector pulse width). So for the first, you'd get a clean 0-100%, for the second you could get anything - 10-200%+, and the last, you'd likely see 2-60% with a proper setup.
All depends I guess. I hope it doesn't exceed 100% load. I am keeping the stock MAF but probably adding that MAF.ia... we'll see. If I add the mafia and change the MAF transfer I know I'll end up over 100%. Just trying to avoid that.
I know the CBAZA strategy has multiple methods of calculating load, by just changing the value of a scalar, you can switch it between load based on TPS voltage, calculated VE, or load as a percentage of possible engine load (based off injector pulse width). So for the first, you'd get a clean 0-100%, for the second you could get anything - 10-200%+, and the last, you'd likely see 2-60% with a proper setup.
All depends I guess. I hope it doesn't exceed 100% load. I am keeping the stock MAF but probably adding that MAF.ia... we'll see. If I add the mafia and change the MAF transfer I know I'll end up over 100%. Just trying to avoid that.
All of the tuning I have done bases the spark tables on VE.
Log your MAF voltage / ad counts (whatever your software calls it) and log your A/F. Make your corrections from there.
Look at where you were at 4000rpm on your MAF Scale. Then take your desired A/F and actual A/F and do the math
For example if you were at 4000rpm and your actual A/F was 13.6 13.6/11.8 is 1.1525. So you now take the Lb/Min scale and multiply it by 1.1525. This will well tell the computer it’s getting 15% more air at that given count and it will add fuel according to the base fuel table. This should get you really close to your desired A/F. I wouldn’t worry about making large changes even 20-30% if the math calls for it.
Also this is what most supercharged Mustangs Spark tables look like as far as the X & Y values. Keep in mind that I have SVO heads with a very different camber design you will not get away with this much spark advance on a PI head.
Thread
Thread Starter
Forum
Replies
Last Post
AMAlexLazarus
AmericanMuscle.com
0
10-01-2015 09:21 AM
