CL Idle Initial Value Table and PID behavior

[lagos]

I'm trying to understand the logic behind the initial value table for close loop idle and how PID is used along with this.
From my observation it seems the initial value table is used initially, however once PID finds its target value, the found value seems to be used the next time the car comes back to idle.

So if my table at 900 rpm has a value of 40, the ecu will use this value and then allow PID to find the target until it settles on the correct value (35 for example).
Then next time the car comes back down to idle, the ecu will use the learned value of 35 instead of referencing the initial value table? What happens to this learned value once the key is turned off and back on again?

[muythaibxr]

The initial value (if enabled) is used on every return to idle.

[lagos]

What about when the car returns to normal idle after AC idle up was used for the first time? It seems to me that it goes back to a learned value in that case, and if its the very first time in the day, it will go back to whatever value was last stored in there? I don't have a datalog to prove this, but it seems to be causing issues for me with the car returning to a very low idle because its initially returning to a stored value that was obtained at a different time manifold air temp.

So basically if I drive the car with the AC during the hot day, everything will be fine. But if I drive the car during the cold night, the idle will initially be very low when the AC cycle off for the very first time because it seems that its going back to the last stored value. After driving around for a bit it will relearn and everything will be fine.

Please let me know if I'm way off track her, but if I'm not than I believe that the code for this might need to be changed so that the car goes back to the initial value table when the ac cycles off for the first time.

[arran]

You're way off track. Setting up the initial value table is a ball ache, the single best advice I can provide is to avoid setting the AC idle up duty, the fan idle up duty or the dashpot to anything more than a couple of %. Use the setting for AC and fan that increases the idle RPM by a 100 or so and let the table do almost everything.

My experience is with MS3, I don't know how different it is on MS2, the FW referred to below is MS3 stream

It is also worth noting that the return to idle code is substantially changed in Firmware 1.5 which uses engine state instead. So its going to be of limited value at this stage to invest 100s of hours (like I have over the last few years) getting this to work correctly across the matrix of variables involving cold start, warm up, warmed up, fan and AC

[lagos]

You're way off track. Setting up the initial value table is a ball ache, the single best advice I can provide is to avoid setting the AC idle up duty, the fan idle up duty or the dashpot to anything more than a couple of %. Use the setting for AC and fan that increases the idle RPM by a 100 or so and let the table do almost everything.

My experience is with MS3, I don't know how different it is on MS2, the FW referred to below is MS3 stream

It is also worth noting that the return to idle code is substantially changed in Firmware 1.5 which uses engine state instead. So its going to be of limited value at this stage to invest 100s of hours (like I have over the last few years) getting this to work correctly across the matrix of variables involving cold start, warm up, warmed up, fan and AC

Thanks for the tip, and I think you're right. I think I managed to sort out my issue by decreasing AC idle up duty and also making my EGO correction less aggressive. Setting all of this up for various times of year is more of a pain than it should be. I've had my car running on ms2 for a few years and its amazing to me that I'm still messing around with stuff like this.