Cold start trouble - but no issues when warm...

[barnwerks]

The DME relay function is not likely the area of fault but this simple bypass test can quickly eliminate the inputs and outputs from the DME socket from suspicion. Replacing the DME relay with another does not do this. A fuse in the jumper is a good idea, particularly if you are not sure how to connect. Always connect to T30 last and disconnect fromT30 first.

[mikee30]

The DME relay function is not likely the area of fault but this simple bypass test can quickly eliminate the inputs and outputs from the DME socket from suspicion. Replacing the DME relay with another does not do this. A fuse in the jumper is a good idea, particularly if you are not sure how to connect. Always connect to T30 last and disconnect fromT30 first.

Thanks for the info. When I connected the three wires to the relay input, I immediately heard some cycling from the engine when pin 30 was connected last. When I tried to start the car (with the wiring still connected) the same issue occurred...started and died. I didn't start it a 2nd and 3rd time while giving gas like I normally do since I assumed I should be looking elsewhere for a solve.
Not sure where to go from here or what the test result confirmed. Should I be checking for something while the wires are connected?

[Tom]

Have you checked the ISV? If stuck, it might make it hard for the cold engine to idle?

What happens if you depress the gas pedal an inch or so when trying to start the cold motor?

[mikee30]

Have you checked the ISV? If stuck, it might make it hard for the cold engine to idle?

What happens if you depress the gas pedal an inch or so when trying to start the cold motor?

Depressing the gas pedal about an inch is how I keep the car running until it is warmed up. Once the car is warmed up and I shut it/start it again, depressing the gas is not necessary and it will run.

How would you suggest testing the ICV while in the car? When the car is running, I can hear it cycling. I did also spray some cleaner into one of the hoses going to the ICV in the event it was a bit stuck, but in retrospect that was probably useless When I did the smoke test, I did notice smoke coming out of the bottom of the ICV, right where there are two small indent type holes in the ICV body.

I have been trying to troubleshoot this issue without removing the intake manifold, hence the reason I have never actually removed the ICV for testing on a bench.

[cda951]

There should not be smoke coming out of an ICV body during a smoke test, that would be a vacuum/false air leak during idle condition/throttle closed.

I understand the hesitation to remove the intake manifold on a 951, if it hasn't been removed in a long time, you might be in for a snowball of a project with each and every crispy hose and vacuum fitting cracking, plus the usual "while you are in there" questions (venturi delete?).

If you can manage to persuade both ICV hoses off without either cracking, you can run a long loop of heater hose or something similar between the two manifold fittings, pinch the hose off slightly to approximate the cross-section of a ICV passage as commanded by the DME during cold start, and see if the engine achieves a high idle after you start the engine cold. If so, you know the idle control valve is the likely culprit.

You still need to make sure the throttle position sensor is adjusted properly (so that the DME recognizes that the idle microswitch) and that the DME is supplying the proper signals, but if the hose bypass changes things, you'll feel more confident in the inevitable intake manifold removal process.