Stalls after warm-up cycle / No start warm

Good afternoon Everyone,

I've got some theories about what's going on with my car, but I'm still working on troubleshooting and looking for thoughts from those more familiar with the vagaries of K-Jetronic with Lambda control.

The Car: US-spec 85 308 QV, stock configuration, except for deletion of pulse air injection system and catalytic converter.

Set up: driving the car moderately hard last weekend at an autocross (and not giving up much, if any time to newer, faster cars, by the way--see gratuitous photo below). After my last run of the day, just as I coasted down out of the exit shoot, the car backfired and stalled. Couldn't re-start until the car was cool enough for the open-loop cold start cycle to kick in. At that point, I drove back to the paddock parking lot and ... just as my car finished it's warm-up cycle and returned to regular idle, it stalled out again.

Current status: the car starts cold in the normal manner. But after the aux air valve closes to return the car to normal warm idle, the car does not seem to go into close-loop Lambda cycle. I can feel the metering valves buzzing at what I take to be the expected 70hz, but the usual up-and-down rich-lean-rich idle hunt of +/- 100 RPM doesn't happen, and the engine just bogs down and stalls instead. The proximal cause of the stall seems to be an over-rich condition--hence the backfire on initial occurrence.

At current, I have two leading theories:

1. Stuck/failed throttle microswitch telling the CIS system that the car is at WOT and needs full enrichment, and thus taking precedence over signals from the O2 sensor.

2. Bad O2 sensor, (or wiring) transmitting no voltage, which would translate into a over-lean condition that the ECU would try to correct through the metering valve by enriching the mixture. (Note: I am not sure that the metering valve, even at full duty-cycle, has the ability to enrich the mixture SO much that the engine stalls.)

But of course, there could be a combination of other problems, including some I haven't even thought of yet. Any suggestions on how to troubleshoot? Anyone encounter this same problem?

I think someone posted procedures for testing the throttle microswitch here at some point. Right now, I think that is my first step--unless someone has a better suggestion.

Thanks in advance!
Image Unavailable, Please Login

 

Update:

After about an hour of probing and prodding, (breaking connections, cleaning, reseating) and replacing one of the two relays on the ECU (the one for the Throttle Microswitch, I believe) due to a very minor discrepancy of resistance from spec, I started the car again and was not able to replicate the problem. Hmm. I experimented with disconnecting the throttle microswitch, but it didn't really do much of anything. Which suggest that the idle is not set quite right, or the microswitch isn't making full contact with the throttle butterfly closed. I took some turns off the idle stop because I realize that having it in too far might cause the idle-side contact in the microswitch to not make contact as currently adjusted. That, and some adjustment to the bypass air screw got her to a normal-sounding idle.

I won't be sure I've solved my problem until I replicate the conditions fully (or at least more so than I can without leaving my own driveway) but I hate not finding a real smoking gun.

 

It is NOW, but it wasn't. Agree on adjusting the microswitch position after setting a good idle, but first I had to get the idle stop someplace close to right. I think we're making contact at the idle position now, because the idle fluctuations seem normal from what you'd expect on a lambda car--but I haven't tested the contacts yet. Thanks for posting that again Steve.

 

I could buy the backfire resulting from a bad coil, but I had PLENTY of power up until seconds before, so it would have had to have failed right then and there--and I can't see the other effects note resulting from that. I've lost a ignition power to a cylinder bank before--it just idled really rough and only stalled under low-RPM, moderate load. Then again, this car has surprised me more than once, so what do I know.

 

Update 2: working late last night (because sleeping at night isn't something I'm particularly good at) I tested the resistance across the three pins on the throttle microswitch while moving the throttle plate through its full range of travel.

I was hoping to see that the WOT switch was showing some amount of inappropriate conductivity (IE: less than infinite resistance at a position other than WOT). Not the case, it turns out. However, when the contact did close at WOT, the resistance was surprisingly high: 5.0 Ohms. Not ideal, certainly, but I don't see that being the source of the problem I am currently having.

Similar story on the idle condition pin. Infinite resistance through the full range of motion except closed throttle, and a bit more resistance than I would like to see when contact is made (0.8 Ohms).

Now that's it daylight, I'll investigate further. The fact that the contacts are opening and closing at the appropriate points in throttle shaft rotation would seem to indicate that the switch position isn't *completely* out of alignment, but I will experiment with that a little to see if I can make those resistance figure a little better before I open the switch body itself.

 

I think you might be onto something there...