closed loop iac control

[Yves]

I finally could get closed loop idle control working. This is on an SBC V8 with ITB's with an agressive cam a lot of overlap. Basically the engine has a lopey idle as is.

The reason for going to closed loop control was that my tps opening at idle varied from cold to hot, making idle rpm usually to high for my liking. My target was 1000 rpm but at the 2.6-2.7% hot I was usually at around 1150-1200 rpm.

Now with closed loop I seem to still have it oscillating albeit not a lot, from 980-1050 rpm.

So my question is if it is possible to get a more or less stable idle or is this a futile exercise with this type of cam ?

My second question is about cam timing : I now run around 18.5-19° of idle timing. Would reducing the timing help with idle stability ?

[ANTI VENOM]

980 to 1050 seems pretty stable to me for a cam with a large amount of overlap. I just started to tune the closed loop idle, if I find anything that might help, I'll let you know. As for the timing, I'm not sure. I'm at 18 degrees but my cam is for a turbo and has a low amount of overlap. Good luck.

[nathaninwa]

I like using idle control with idle timing correction. Both work together and can create a pretty clean idle

You can even set it up so a non cammed engine rumps, rumps, rumps like one that does for those cruise in nights

[Yves]

Thanks for the replies.

I have currently 18° at 1000 rpm and 19.4 at 1150 rpm.

What kind of curve do you have for idle advance

[nathaninwa]

For a smooth idle I will flatline about 5* in the 100 rpm difference range and setup about 3* near the 30/40 mark. I'm sure each engine is different tho, just feel it out. For that rump sound I'll go as far as 15* at the 50rpm difference mark!

That reminds me, need to ask Phil for the green live ball on that screen

Edit, using timing control lets you limit idle valve closing steps so you won't come to a stall on a decel. I like to find where it likes to be and limit it there and kinda let timing take over to get the rest of the correction inline

[Yves]

I have only used the closed loop while in the garage. Not sure how she'll do when decelerating.
Isn't the dashpot adder there for that ?

[nathaninwa]

Depending on the firmware your using the dash pot added helps but was kinda complicated to tune so it's been deleted on the latest firmware iirc. Setup the initial value table using air temp for lookup and you can set a value just above what's needed to catch the idle prior to closed loop taking control

[Yves]

Help me understand what causes a possible stall in that case. It's not clear to me.

When I was doing testing with closed loop the valve shut almost completly at 2 steps. So what should I put in the table ? 5 steps ?

[nathaninwa]

Can you post your msq and a quick lot of it stalljng

[Yves]

Like I said, I only tested it in the garage.

[nathaninwa]

Ok, I see my tune has dash lot adder but I don't use it

Couple things. Start your engine cold and see what the steps are for a cold start. Let's say this is 150 steps, and fully warm idling the steps are say 30. In the closed loop idle settings set the valve max step to like 160 and the minimum steps to about 25. This will limit the valves movement and hold the valve at 25 steps so it's at least ready to idle

Also, I don't like the use last value for idle setting. If your idling 30 steps the Ecu will use a value of 30 for closed loop to start with. I like using a mat based table so I can it about 5 percent, what ever steps that is for you, so the Ecu will use a value of 35 then enter closed loop where it drops idle down. Tuning that value you can darn near nail your target each time

Maybe use the basic mode for sensitivity

I use a delay of 0 for pid


I also like to use some timing idle correction to help with closed loop. If you set it up for 0* at 0 rpm then at -50 use a value of 5* and at 50 use a value of -5*. This won't get you into trouble at all, but will help stabalize idle

If your idle steps is way above 30, then when fully warm open the blade slowly and watch it come down to about 30. This will give you max valve motion

[Yves]

OK, will try that approach today

[Yves]

Did some testing and so far the results are good. I need to let the engine cool overnight and retry as my starting wasn't what it was before (attempting to stall)

I used bigger values for PID and it seemed to do fine in the sense that it was idling between 1030 and 1070 where my set point is 1050 right now.

[Yves]

Drove it a little to get gas. Seems to run ok, but the rpm seems to want to stick at higher rpm (1100-1200 ish) for a little too long. Tried to play with the initial value table a little but it was not a lot of help.

Last thing I tried was a warm startup. Seemed to want to hang in high rpm for a pretty long time, although the engine was fully warm at the time. The tp changed from 2.0 to 2.4% during that time because of the heat soaking on the intake manifold and linkage (ITB)
(sorry no log of that. Something went wrong and there was no Y-axis or something according to mlv...)