Weird Electrical and CIS problems

[emcon5]

I am having some trouble with the CIS system on my 82 911SC. It seems like the idle mixture is bouncing around, causing varying idle speed. Idle hunts when cold, swinging from ~500 to ~1500 RPM before settling to 900ish when it starts to warm up. There is no warmup high idle, although I examined the aux air valve when the engine was out and it was about halfway open. Once up to operating temp, idle speed will periodically drop to around 500. I was thinking that the problem may lie with the power relay for the lamda system under the passenger seat, so I got a new relay from a local Porsche dealer. When I installed the relay, the ignition was off, and the key was in my pocket. As soon as I plugged in the relay, I could hear a buzzing from the rear of the car (frequency valve, maybe?), and a dash light came on (either the seat belt or brake warning light). After 2 or 3 seconds, the fuse blew. After replacing the fuse, and installing the old relay, the engine started and ran like it did before. The new Bosch relay is identical to the old one.

My first thought was the new relay was stuck closed, so I tested it for continuity across the switched terminals. For reference, a diagram of the relay is here: http://www.pelicanparts.com/911/911_Parts/Electrical/911_electrical_82SC_Part6-2.jpg There is no continuity between 30 and 87 or 87b, which is expected with the relay not energized. There is continuity (with some resistance) between 85 and 86, which powers the relay. This also seems normal. I will check the harness for the relay this evening. From the diagram, it looks like the only thing that may be hot with the ignition off would be 30. Does anyone have any ideas where else to look?

I pulled out the Jetronic box under the seat and took the cover off, and didn't see anything obviously wrong. Everything looked to be securely connected, the circuit board looked fine, and most importantly it didn't stink of fried electronics.

I also checked the O2 sensor and it was making voltage with a running hot engine. I checked to see if there was a duty cycle coming from the freq valve, and there was a signal from the test points. I don't know what the duty cycle was, because I couldn't remember what scale to use on the dwell meter. I will verify the duty cycle this evening.

Is the duty cycle at the test connection the signal from the controller to tell the frequency valve what to do, or does that come from the valve? If it is from the controller, is it possible my frequency valve is bad? If the freq valve is bad, what the buzzing noise from when I plugged in the relay? I put my hand on the valve to see if it was vibrating, and I don't think it is. It doesn't feel like it is buzzing with the old relay plugged in and the car running. I realize I may have fuel pressure problems as well, but I would like to get the electrical screwieness sorted out first.

Where do I go from here?

Thanks,

Tom Aiken
Fremont, CA

[lagoon]

Tom, you should hear the F/V buzz. That tells you it's working. If you don't, you mixture could be too rich. Set your dwell on 4 cylinder and read it when hot. should be averaging 45 degree dwell.

[emcon5]

Well, I think I may have figured out part of it, why everything managed to get power with the ignition off.

On the new relay, terminal 85 is loose. When I pushed it into the socket, the spade pushed into the body of the relay. When I remove it, the terminal is pulled back to where it is supposed to be, so it not obvious what the problem is.

The problem is, when the relay gets pushed into the body, it shorts out with terminal 30. Terminal 86 is powered from the fuel pump relay, terminal 30 is always hot. Essentially plugging in the relay put current to the pin that triggers the relay, energizing the freq valve and the brain.

Who would have figured that. Like I said, wierd.

Tom

edit: typo