Closed Loop Idle won't hit target,too high

[snowrx]

My GM IAC on a MS3 is giving me too fast an idle. It seems to control the IAC OK, but will not pull the steps down to my target idle (~950 rpm) If I force the IAC in the output test to a 55 count it will idle the right speed, but left to it''s own control it idles 1-300 rpm faster with the steps changing, but to keep it at a steady fast idle( 67-80 steps) . My target rpms are correct in the target curve. If I let the engine idle 20 minutes it will get down close to the target and regulate there, but if I restart or drive the car I get a high idle again, sometimes 2000 rpm! I don't detect any vacuum leaks, and If I had one I would expect the steps to run down to compensate rather than stay higher.

What am I missing? The PID's don't seem very sensitive at this point for idle and I think I need to dial them in for the EGO control as well given the cycling of the idle speed
Anyone care to suggest nominal settings for idle or EGO PID's if I'm way off?

[sheek]

u need to add more I
try P.I.D 70. 70. 0

[snowrx]

PID = 70/70/0 brought the idle down. Now the engine dies when I put in the clutch after driving load, so I need to chase that.

[piledriver]

PID = 70/70/0 brought the idle down. Now the engine dies when I put in the clutch after driving load, so I need to chase that.

Raising the minimum value of the stepper (or PWM%) may help keep the revs up during this period, and not require .ini fiddling.
Make sure the fuel cut restores fuel soon enough if is active.

There are some programmed delays in the idle logic, 1 and 2 second minimums, range is set in the .ini.
One was IIRC changed so the min is zero, but IIRC default is still 2 seconds
The PID delay min was IIRC still 2 seconds last I looked.

I haven't checked lately when the IAC goes to its preset but I think it may be after that 2 second delay, which is too late in many cases.
Changing the min value should help, at some loss of control range.

[sheek]

PID = 70/70/0 brought the idle down. Now the engine dies when I put in the clutch after driving load, so I need to chase that.

do u have your speed sensor run through the ecu? if so under closed loop settings choose use vss to activate pid
minimum vss to around 20kphidle activation threshold tps around 3%


also a dashpot adder of around 10 may help.


also drop ur sensitivity to around 900. do these things one at a time and note if it gives positive or negative changes.


read the tooltips in the idle settings and understand what each setting does and play with them and note changes.

[snowrx]

Yes, I forgot I had zeroed my dashpot setting flailing about to reduce the 2K idle, closed loop works fine now with a bit of mechanical idle stop, 10 of dashpot and a minimum duty cycle. Thanks for the input

[Yves]

PID = 70/70/0 brought the idle down. Now the engine dies when I put in the clutch after driving load, so I need to chase that.

Raising the minimum value of the stepper (or PWM%) may help keep the revs up during this period, and not require .ini fiddling.
Make sure the fuel cut restores fuel soon enough if is active.

There are some programmed delays in the idle logic, 1 and 2 second minimums, range is set in the .ini.
One was IIRC changed so the min is zero, but IIRC default is still 2 seconds
The PID delay min was IIRC still 2 seconds last I looked.

I haven't checked lately when the IAC goes to its preset but I think it may be after that 2 second delay, which is too late in many cases.
Changing the min value should help, at some loss of control range.

You can get the engine to return to the current idle by tweaking the initial value table. I set this to CLT. If for instance your normal iac steps at idle is 30, then you set this to 25 or so in that row.

[piledriver]

PID = 70/70/0 brought the idle down. Now the engine dies when I put in the clutch after driving load, so I need to chase that.

Raising the minimum value of the stepper (or PWM%) may help keep the revs up during this period, and not require .ini fiddling.
Make sure the fuel cut restores fuel soon enough if is active.

There are some programmed delays in the idle logic, 1 and 2 second minimums, range is set in the .ini.
One was IIRC changed so the min is zero, but IIRC default is still 2 seconds
The PID delay min was IIRC still 2 seconds last I looked.

I haven't checked lately when the IAC goes to its preset but I think it may be after that 2 second delay, which is too late in many cases.
Changing the min value should help, at some loss of control range.

You can get the engine to return to the current idle by tweaking the initial value table. I set this to CLT. If for instance your normal iac steps at idle is 30, then you set this to 25 or so in that row.

Sadly that doesn't work in all cases as it doesn't kick the table in until after the CL delay, so the RPM will fall back to cranking before recovering in many cases.(I have played with the initial value table until blue in the face)
The min duty cycle works as a lower limit to keep the revs from falling too far... most of the time.
(Which can be reduced with .ini edits, which seems to be the better fix, although I have not tried the most recent firmware due to some MS2 hardware testing, should have the car back on the road with MS3 this week with any luck)

[Yves]

So far, I found my rpms to drop to slow during a decel, and when looking at my logs it had to do with the initial value in my case.
However, you also have the delay settings. I remember there is one setting that cannot be set lower than one second (not at home so I cannot look at my ts to see what setting it was). So maybe your problem is in that area. What I want to say is that if the delay is for instance one second, that the closed loop might not open the valve untill after that delay, by then the engine may have died already because the valve is too far closed.
Just thinking out loud, but maybe you can set the idle screw of your throttle blade to just below your regular idle rpm, so this acts as a minimum value to prevent the engine from stalling in that one second delay period.

[muythaibxr]

If rpm falls too low when you lift off throttle, you need to work on fuel and ignition settings, not idle speed control settings assuming it is not in PID when the problem happens. PID can't correct fuel and ignition issues. In that situation PID should not be engaged. If it is then the activation settings need tweaked or else the high RPM will make PID close the valve.