PDA

View Full Version : TB Air Flow Strangeness


TT331FB
09-26-2005, 08:16 AM
A9S, 3" Pro Mass Air 42#, blow through.

My tune is pretty good except the motor dies occasionaly comming down to idle. I decided to check TB air flow. I had .46 VDC average MAFV at 600 RPM, ICS disconnected and spout out. I data logged normal idle and got 797 RPM, .45 MAFV average. Huh? Data log shows the idle spark is all over the place. The only thing I can figure is that spark is causing fluctuating MAFV giving an average lower than TB airflow. My plan is to data log idle again but with the spout out to remove the EEC efforts to control idle.



Any input?

Eric

93Cobra#2771
09-26-2005, 08:33 AM
Your idle spark jumping all over shows your idle is not stable - the EEC uses spark and ISC to control idle, with the quicker reacting spark getting the nodd on a surging idle.

More timing equals faster idle, less timing equals slower idle.

When coming down to an idle, is this when driving, or when sitting still? Dashpot decay may be something you need to look at...

TT331FB
09-26-2005, 10:06 AM
Just for grins, I gonna log idle with ISC connected, spout out. I think the spark is adding instability into the loop.

93Cobra#2771
09-26-2005, 10:50 AM
It can do that - even spark takes SOME time to change, and it could end up chasing itself to a nice surging, dying tune...

86GT
09-26-2005, 10:56 AM
That does seem kind of weird that the normal idle MAF is less than the base idle with ISC disconnected and spout out.

From what I have experienced, The ISC will compensate for idle if it is within some calibratable tollerance. I wish I knew the limits. If the desired RPM minus the actual RPM is out side of these limits then spark is used to bring it back into tollerance. I believe the upper limit is set by the scalar ISCLPD.

Look at function FN810. It tell the EEC what DC to send to the ISC for a given RPM error. Since your idle has been changed to 800 or so then the FN810 will need to be shifted to the right by the same amount you increase the idle. If you left it stock then the ISC never has a chance to correct until it is to late and then spark control takes over.

800- 672 = 128
Try bumping the entire curve up by 128 RPM.

I also played around with the gain control on the ISC FN860. I believe this is the same as the 'P' in a PID loop.

TT331FB
09-26-2005, 02:24 PM
That does seem kind of weird that the normal idle MAF is less than the base idle with ISC disconnected and spout out.

From what I have experienced, The ISC will compensate for idle if it is within some calibratable tollerance. I wish I knew the limits. If the desired RPM minus the actual RPM is out side of these limits then spark is used to bring it back into tollerance. I believe the upper limit is set by the scalar ISCLPD.

Look at function FN810. It tell the EEC what DC to send to the ISC for a given RPM error. Since your idle has been changed to 800 or so then the FN810 will need to be shifted to the right by the same amount you increase the idle. If you left it stock then the ISC never has a chance to correct until it is to late and then spark control takes over.

800- 672 = 128
Try bumping the entire curve up by 128 RPM.

I also played around with the gain control on the ISC FN860. I believe this is the same as the 'P' in a PID loop.

Clint, I have a A9S, none of these functions are available to me. :( What bugs me is how the MAFV can be higher at lower RPM!?

86GT
09-26-2005, 02:44 PM
I looked at the A9S strategy and it has FN810. It might help. I agree about the weirdness with the higher MAF at lower RPM. What about temp. Is it the same on both idles?

TT331FB
09-26-2005, 03:41 PM
I looked at the A9S strategy and it has FN810. It might help. I agree about the weirdness with the higher MAF at lower RPM. What about temp. Is it the same on both idles?

Clint, What is the name of this function in Tweecer Terms?

86GT
09-26-2005, 06:23 PM
'ISC adder vs RPM'

If you look down in the lower right corner it will tell you the function number. I usually refer to those because for some reason Mike calles these things something different between stratieges even though they are the same thing.

Disney Lincoln
09-27-2005, 12:51 AM
How do you raise that a certain RPM??

Mine looks like this:
X 0 0 0 2457.5 16384
Y .03125 .03125 .03125 .033447 .033447

86GT
09-27-2005, 08:38 AM
This is what the A9L or X3Z looks like:

X----0----350----450---600---16384
Y---0.5---0.5---.22988---0------0

I just add 128 to the lower 4 'X' cordinates.

TT331FB
09-27-2005, 09:15 AM
Looks like the A9S is air flow and the other is duty cycle.

86GT
09-27-2005, 09:45 AM
Even if one calibreation is in air flow and not DC, the RPM row (X) should be the same. Maybe Mike has that function all messed up on the A9S.