SharkSkin

Hi Bill, sorry to see you're still tearing your hair out over this. I don't have time right now to trace out all of the circuits discussed above, but something occurs to me about the 12V on the door pin wires, something I haven't seen mentioned.

The alarm has to be able to sense the pin switches, so it must have a 12v sense line connected to that circuit. It's possible the alarm module is misbehaving. I don't see how that should interact with the window controller, but it is another bit of 18-year old tech connected to that wire that's puzzling you. How tough is it to bypass?

Bill Ball

Based on the wiring diagrams I have posted, I think the alarm could interface with the doorpins at the timing relay socket, but there are no other connections between them in the wiring diagrams So, I'll trace the rest of the wires from that relay socket (see next post). Thanks for the suggestion.

Bill Ball

I see no interface of the alarm with the door pin circuits. I think it monitors the locks instead.

Anyway, the more I look at this, the more suspicious I'm getting of something amiss around the timing relay socket wiring to which the door pins connect. It's located under the radio and I'll bet it's been tapped into by the radio installer and the tap has subsequently shorted. Who knows, I might have done it when I installed the keyless entry!

Bill Ball

This thread is crazy. I started with a passenger window spontaneously and unstoppably going down. Started looking at the window motor harness for a short. Found nothing there. Worked my way back to the window controller and found that the controller output to that window always had 12V. Found that the output itself was not 12V but it was coming from a jumper from another controller pin, function unknown. Then I found that pin was getting its signal from a door pin switch input to the controller. Then I found the door pin switch wires were hot all the time. That's where I am right now. Crazy. Next stop....timing relay under the radio.

Ed Scherer

It does sound like something is shorted somewhere. One course of action would be to disconnect everything at the end of harnesses in question and verify continuity from end to end of each wire and check for shorts across any two wires (i.e., check for continuity from a given wire to each other wire, one at a time).



One other thing comes to mind, although not related to anything in particular that you've posted yet, but just the weirdness of it all: a bad ground somewhere.

The circuits can get really bizarre if one or more grounds isn't connected (one of the ground wires lifted off a grounding point or whatever).

As much as you've been poking and prodding, is it possible you've disturbed one of the grounding points or perhaps loosened one of the ground wires enough that it's become intermittent?

It might be worth checking the ground points nearest to where you've been working and double check that all the ground wires are securely fastened to those ground points. And/or actually test all the ground wires and make sure they're really providing a good ground that will flow current.



And now that I think about it even more, there might even be cases where the reason a ground isn't connected is because you've intentionally disconnected a connector while diagnosing something. That might have strange consequences for another circuit that was getting its ground through that path. Just something to be careful about.

SharkSkin

Bill, you can validate whether the alarm operates off of the locks or the pin switches by closing the door latch with the door open, hold down the pin switch, activate the alarm then release the pin switch.

If it operates off of the locks rather than the pin switches, then my theory that the pin switches ought to always see 12v is invalidated, but that would strike me as an unfortunate design choice -- that would mean that someone could lever open the hatch with a prybar without setting off the alarm. If it does tie into the pins rather than the locks, being a "sense" wire it shouldn't sink much current when operating properly which would explain why you don't see the 12v on your car.

Also, I agree with Ed -- grounding issues can cause some very strange behavior.

Bill Ball

Ground issue does sound likely. I'll trace the ground for the controller and lights. 12V on the door pins is not right and the door pin circuit is involved in the controller and the interior lights. The door pins are supposed to be simple ground circuits. If I disconnect the door pin wire from the window controller, the door pin still has 12V. The only other source I can see then is the timing relay.

Alan

Bill & all - the alarm does sense the locks directly and the interior light relay output for the combined door/hatch opening (obviously int lighting has to work for the alarm to work though).

The pin switches should see 12v normally - via pull-up resistors - this is needed to see the voltage go low (its voltage sensed not current). The pull ups are at least in the window relay possibly in the timing relay (BTW this is the warning Buzzer relay - I have no idea why they label it this way).

The issue before with what you said was comprehension on my part. You mentioned ohms to ground and I said infinite with doors closed - this is technically correct - but you can only measure it to be true with the battery disconnected - with the battery connected you will essentially be measuring the pull-up resistor value (because battery internal impedance between ground & B+ is <<1 ohm). So I'd say a red herring - though why the pin switch input took so long to get to 12v is rather interesting - but may be a reading anomaly (or done with window relay pulled?). Sorry to have caused some confusion.

The ground theory is interesting mainly because you do seem to have some odd interactions between circuits.

SharkSkin

Pin switches connect to CE panel L12(Dwg. loc. M3). You can pull this pin from the harness end at the CE panel to isolate the circuit. On the harness side, this connects to pin 5 on the window controller and some pin on the timing relay(hard to read on my dwg -- it's at O24). Inside the panel it just connects to the inside light relay, XIX. Have you tried swapping out that relay?

Bill Ball

OK, thanks for the clarification on the pin switches being 12V, although I don't see that in my 89.

I've pulled plug L on the CE and the pin switches stay 12V. I swapped in the inside light relay from my car. Same. I found the "timing" (buzzer) relay and pulled it. Pin D on the relay socket (the pin wires) shows milivolts until I push in a door pin, then it goes 12V.

So, is all this 12V stuff a red herring? Then should the window controller be seeing 12V through pin 15 with the doors closed? As I push the pin switch in and out I can hear a relay inside the controller trip. Then pin 38 goes 12V and the permanent jump wire between pin 38 and 33 sends 12V to the passenger window motor [CORRECTION: Pin 38 is 12V all the time]. I cut that jumper and I have everything normal except no window operation with the ignition off and the doors still closed.

With the battery disconnected, the pin switch feed to the window controller is ground and breaks when the door is closed, as you described, Alan. With the battery connected, the the window controller is powered anytime the doors are closed. So, it stays powered all the time. At one point we found it was draining 150ma all the time, draining the battery. I haven't confirmned that it still is.

Which grounds should I be checking?

Alan

I believe so.

Yes - but for USA its only when it goes to 0v that anything happens - thats the signal to turn off the power relay (on/off) in the controller. Are you in ROW mode (e.g. jumper pin 22 to pin 17unplugged) if so it should indeed go on and off with the door pin switch - in USA mode (plugged) it will turn the relay off but the relay needs pin 36 (ignition) high to turn on. Does it work correctly again with the door open?

Well I'd check GP V the ones above the CE panel.

Alan

Bill Ball

Pin 15 goes 12V with the doors closed. It goes to milivolts with a door open and I get a ground continuity now! The jumper on 22 is connected to a wire plug in the harness that does route back to 17. I confirmed continuity. Regardless of whether that is plugged in, I hear the relay switch with the door pin.

THANKS FOR BEARING WITH ME. I feel like all I've been doing is chasing red herrings for the last few days.

Alan

The only thing that sounds odd here is the relay clicking when in USA mode (jumper unplugged) with the pin switch in/out - don't understand that - only in ROW mode is that the normal operation.

I don't really know what this pin 38 stuff is all about - that pin should not go to 12v in the "power maintained mode" - all that should happen then is that the module stay powered up and monitoring its inputs ready to respond as normal - no other functional change.

I thought I had an idea what this pin 38 was for - but I may have been totally off base - I was assuming (for some reason) it was merely a monitoring input...

Alan

Alan

Bill - OK a wild idea here - what happens when you hold the key in the lock for the window/sunroof auto close thing - does this mode work? - does the passenger window go up or down?

Alan

Ed Scherer

You know, Alan, Bill is probably at his wit's end here and — if he's pulling his hair out, as he claimed in earlier posts, IIRC — will look like me pretty soon. Scary, indeed.

It occurs to me that we can really start the fun now.

Like:

"Bill, I know this is a long shot, but what happens if you strip naked, stand on your head, press the window down button with the big toe on your left foot, put your tongue on a wire connected to CE panel L12, and pee on the passenger side pin switch? Can you feel any current?"

He might just do it!



Seriously, though, I've actually kind of lost track of where things stand now. Bill, you might want to summarize current status. We need to refocus after the red herring stuff, but I've lost track of what the real remaining issue(s) is/are.