Very well explained, Phil. You've obviously cracked more than one or two IACs open in your time and given them a fair bit of thought. Your discussion gives me a rare opportunity to add my own thoughts developed after cracking open a couple of IACs many years ago. I posted them here many, many years ago buried in a reply that didn't make it into the archives. Some may dispute my conjectures here, but I hope you find them interesting nonetheless.

So these are a couple of tricks I've used to temporarily improve idle stability, enough that I was willing to tolerate my worn IAC or at least delay replacement. They're simple to do so at least worth a try. Of course there's not a lot you can do with a badly worn IAC short of replacement. I gave up trying to crack them open and polish out the wear groove, too much effort and too many failures.

To carry on from Phil's description of the contact wear groove on the commutator, the deep end of the groove will be where the IAC contact spends most of its life moving back and forth a little trying to achieve desired idle rpm. This will of course be at warm idle where the TPS is telling the ECU that the throttle plate is closed and it should now attempt to maintain idle using the IAC. When the engine is not idling, the IAC will be held steady at a closed position, which for the spring loaded ones is possibly end of travel, but more likely it's at some mid-point where the piston is rotated closer to the min flow position. No matter where it is above idle, at least the ECU won't be hunting and constantly rotating the IAC commutator back and forth to deepen the groove.

Trick #1 -better for LH 2.2

So my first thought was how I might change it so the contacts would spend more time away from the troublesome deep end of the groove in order to achieve better electrical contact. You can't and don't want to change idle rpm, but you can change base idle to alter the amount of fixed air bypassing the throttle plate without changing the operating idle rpm. This will require the IAC to compensate thus with the contacts on a different area. To move the normal contact area toward the shallow end of the groove, you would decrease the amount of bypass air at the throttle body forcing the IAC to be kept open wider to compensate (with the contacts more in the area where they ride at raised idle right after a cold start).

For LH 2.2 that's easy to do, just use the black thumbscrew at the throttle body to change base idle speed (on later engines I vaguely recall it's a large set screw, not a thumbscrew, but I may be confusing that with some other engine, like early LH 2.4). Base idle needs to be adjusted with the IAC kept in the closed position, accomplished by using a jumper wire to connect one of the open wire connectors on the fender to ground (blue/white or red/white wire, not the pink). Lower the base idle by perhaps 100-200 rpm (but staying comfortably above the engine stall point). Best to use a proper tach, not the inaccurate dash tach (if you try to use the dash tach, make exact note of the original needle position and mark the thumbscrew so you can get back to factory setting if you need to). An alternative is to do the opposite, raising base idle up by around 100-150 rpm, shifting the normal contact area to the other side of the deepest groove point.

For LH 2.4 without the air bypass, the only way to change base idle is using the throttle plate stop screw to open the plate a hair wider (you will then need to readjust the TPS to spec). You can try closing the plate a bit to lower base idle, but it's best to only try opening the plate to avoid air flow cavitation that can aggravate idle drop and hesitation.

In all cases, avoid changing the base idle so much that the IAC can no longer compensate under various engine opertating conditions (block temp being most significant). Going too high raising the base idle is normally witnesed by a warm engine no longer able to reliably get down to the set idle rpm. Going too low is witnessed by the rpm being too low right after a cold start in very cold weather.

Trick #2 -better for LH 2.4

The second trick is even simpler. It's perhaps not as long lasting, but for LH 2.4 it's easy to repeat whenever you want. The idea here is to try wearing down a bit of the corrosion and grunge that may have developed on the contact points and as well to try smoothing out the step that develops at the deepest point of the wear groove (IMO, that step may be responsible for a lot of erratic idle hunting). With LH 2.4 it's easy, just use the OBD dignostic socket to exercise the IAC repeatedly to far end of travel in both directions, like a dozen or more times, or as much as you can stand doing. It's a slow process waiting for the next tic-toc. This is ECU diagnostic test mode 3 (DTM #3) at socket #2 (box A), see the OBD section of the FAQ (Engine:OBD Diagnostic Codes). You can depress the button at any time to exit test mode.

For the spring loaded LH 2.2 IACs this trick is less easy to accomplish. The closest thing you can come to rotating the spring loaded commutator back and forth is removing the IAC and rotating it back and forth in your hand until your wrist is ready to fall off (your wrist will usually give out before you've accomplished much good). Another option with LH 2.2 would be to repeatedly ground out the open connector on the fender with the engine idling, but the engine won't enjoy it and you won't achieve full travel, so maybe don't try that.

Now having told you all this, I will admit the most benefit some may get trying these tricks is the satisfaction you at least tried doing something useful before deciding it is now worth coughing the bucks to get another IAC. For those of you about to go try these tricks, please post a reply if they helped.
--
Dave -still with 940's, prev 740/240/140/120 You'd think I'd have learned by now