Getting a handle on the IAC

[Jim Blackwood]

I've not found anything yet that obviously and clearly would allow me to control the gain of the IAC motor. This is an issue because... well feedback controlled systems have a nasty tendency to go into uncontrolled oscillation if the gain gets too high and it turns out (unsurprisingly) that this system is no different.

Now a lot of different factors affect the end result of the signal going to the IAC motor in a lot of different ways but in the end, it has gain controls somewhere. I need to know what they are. I have an idle oscillation and the IAC is leading the parade.

Any suggestions?

Jim

[In-Tech]

I won't have time to look through my notes for you till probably Friday. If you get the pintle further away from the seat(close the TB more) it will help. Also negative timing on flare. On speed density stuff I use a IAC driver tool I have to set/lock the iac at a rough setting, turn off idle timing correction, then set the fuel, and once as smooth as I can get, turn back on all the GM controls :)

[Jim Blackwood]

Thanks Carl, I'll try to show you what it's doing, at this point I'm not sure it's the IAC. I ran some tests with the IAC unplugged, and then of course with the opening covered since that apparently lets it run wide open. Anyway the surge is still there once I back the throttle stop out enough, though it seemed less severe. Unfortunately I copied over that scan but I don't think it would have been much use anyway as the tank was running dry. I can duplicate it if it would help.

Also, it appears I didn't save the scan that clearly showed the IAC leading the surge with an almost identical sine wave pattern. Which is too bad but today's scans show a different result.

So, attached below is the current tune, and two scans one right after the other. (I started a new scan because I questioned the IAC trace). I changed the extension from hpl to txt because hpl is not an approved file type for upload here. So in these scans, there is one area, in the "a" file from zero to 24 seconds where the IAC could be involved. Otherwise, not much apparently if it is a true trace.

I'm boning up on PID controllers, don't know if that will help or just confuse me but I thought it could be relevant. Doesn't seem to be much on idle oscillation on youtube, maybe I just need better keywords.

My cam has 212 duration @.050" and .436 lift so not at all radical. It does have a long high volume inlet tract with the blower and intercooler.
Also cable throttle, manual transmission.

Jim

[Jim Blackwood]

Flattened the idle area of the advance map at 15° and flattened the idle quadrant of the VE map a good bit as well. Way rich but the oscillations have dropped below 700rpm to initiate, next step is to lean it out.

Jim

[Jim Blackwood]

Here are this mornings results. I think it is clear now that the IAC is not the problem.

I've flattened out the VE table considerably and am now a good bit closer on the AFR values. I furthere reduced idle timing to 13° and increased the overspeed and underspeed spark control to 10°, also lowering the minimum spark limit another 2° to -2.

The oscillation is still there, slightly submerged. It shows up most clearly on throttle close and startup. If I could get that smoothed out I think I'd be pretty happy with it. But as it is now I think it's driveable enough to take out on the streets and run a more full datalog.That should happen sometime this weekend.

Jim

[Jim Blackwood]

I found and downladed an idiot's guide to PID controllers by Finn Peacock. About $8, and well worth it in my opinion. However the units don't seem to match anything in the HPTuners universe so how much use here specifically? Maybe only in terms of understanding the concepts of what is going on. BUT, if I could settle on the right parameter to adjust the Derivative value, I just might be able to jack that up a bit and get rid of that overshoot. It looks to me like an increase in the numbers of the fast and slow RPM filters of the Idle RPM Derivative might do it though.

Turns out these values were out of range, over 1.0, in one case about 1.6, so I will try them at 1.0 and see what it does.

A couple of things seem clear: First, my engine does NOT make 15psi of boost. It'll do well to hit 5, so the MAP trace is very wrong. I don't know how to fix that so that's another thing to learn.
Second, my commanded AFR is consistently 4 leaner than my AFR reading which is 15 and matches my Innovate MTX-AL. So probably something is going on there too.

Jim

[Jim Blackwood]

I may have muddled this thread with a scattered approach, but after a short test drive the issue seems clear enough.
The oscillation cycle is about 4 seconds long and that is plenty long enough for any kind of PID controller to intervene. Certainly the IAC is fast enough to take corrective action. Heck, I was able to do it with my throttle foot.

What I'm seeing now is that the IAC is not doing anything or very little about it. I have good IAC action during startup so it is clearly working. The adaptive idle control does not seem to be doing much though. May have to crank up the proportional and see what that does.

Jim