I'm far from an expert on how LH 2.4 works, and it's been long gone from my car (it's megaquirt now). But I do know that the LH2.4 throttle switch doesn't have a potentiometer in it like many other throttle switches do, so the ECU is not given any information about how far the throttle is open, how quickly it's opening. IIRC it just has closed and WOT switches.

Not sure exactly what it does with the WOT switch (perhaps it triggers closed loop operation?) but the closed switch probably does trigger idle mode. The Bosch idle valve defaults to being 'mostly' open when no current is applied to it, which is how I would assume it operates most of the time when the car isn't actually trying to idle. Then, as the idle valve duty cycle rises the valve first closes, then reopens.

As far as I know, the idle control valve does not affect the fuel air mixture, all it does is allow a controlled amount of air to bypass the throttle valve, which nearly completely closes. The computer senses how much air is going into the motor via the AMM - Air Mass Meter - flow sensor. I don't think it really cares how the air gets into the motor, be it through the throttle valve or the idle valve.

And in a long, convoluted manner, I've managed to get to a suggestion. The flexible hose between the intake manifold and the AMM may have cracks or leaks in it thoat don't cause much problem at idle, but which might leak more air when the engine rocks over a little when you let off the throttle on the highway? Double check that - the computer has a limited ability to deal with air that gets into the engine without going through the AMM, expecially if the leak is inconsistent and erratic.
--
'63 PV544 rat rod, '93 Classic #1141 245 (now w/16V turbo)