Manfred

OK.

With both old and new defroster switch I get a dim light from 58 to 31 and a bright light from K to 31 with a 9V battery.

Manfred

Thanks for the encouragement.

I am measuring resistance from the potentiometer plug with the potentiometer removed. I have the red lead from the DMM contacting the Bk/Bl wire and the black lead from the DMM to ground (using coil ground in engine)

I'll check my resistance readings on the defroster switch again--I seem to remember I was having a hard time getting stable readings. The numbers I reported before are from my old switch. I never tested the new defroster switch.

Ed Scherer

Well, the relevant parts of the switches both work just fine then.

We may very well be back at the beginning again, but perhaps some of the stuff found and addressed along the way have actually taken care of the problem.



Maybe it's time to try to put it all back together, making sure the 58 and 31 contacts in the socket-side connector are in good shape, plug in one of the defroster switches (maybe leaving it dangling in front of the pod for now), and power it all up again.

Ed Scherer

The "illuminating the bulbs" test is actually a better test than resistance checking, anyway. Resistance tests involving terminal 58 are problematic (i.e., likely to be misleading), anyway, for the reason I mentioned a few posts back. I should have thought of that sooner.

Manfred

I have a new theory--and I think I have the problem sort of solved. I think the original problem was the bent connectors in the receptacle (picture in post 59). I think this resulted in the defroster switch itself becoming broken--post 31 is depressed and wiggles in the plastic (k post is center/left--depressed):



Because the connectors are mangled, when one pushes the defroster switch onto the connector it breaks the K post free and pushes it into the defroster switch causing the short. I wrongly assumed that because a new defroster switch did not fix the problem that the short was occurring somewhere else. What I failed to consider is that since I didn't identify and fix and the mangled connectors, that I broke the new defroster switch in the identical manner when I tried to plug it in. I hadn't paid very close attention but now I notice that post K is also depressed slightly and moving in the plastic in the new defroster switch.

So, I think I need to find new connectors and another new defroster switch. I just pulled post K out to it's original position in the new defroster switch and connected Bk/Bl to 58 and Brown to 31 while measuring resistance and did not get a drop. I might potentially be able to salvage the new defroster switch but I think even with new connectors, next time I plug it in, the post will probably be driven into the switch.

Alan

Well probably a terminology thing - when you say bridge I think you mean connect together, but it seem you mean you are simply testing with a DMM between those points. If so that's OK subject to my points before about cold filament testing.

Alan

Alan

No it doesn't work like this - all these are in parallel, so sequence is unimportant.

I'm beginning to think its something else in the defroster switch connection (other than the expected 58 connection) that is causing the short. It seems it isn't the switch - so it must be something in the pod wiring that gets completed by adding the switch in. I think we need to investigate this. Since you've done some testing with the switch in and the relay out - and still saw the problem - it seems to be just in the wiring - not in the relay.

Did you change anything else in the pod - or do any other work elsewhere preceding this issue?

If your DMM has a diode tester mode - can you test that you see the diode between 58 & K on the switch(es) I don't really believe this is the problem - but worth checking.

Alan

Alan

Manfred - interesting - hope you are on to something there. It does seem that if they break that easily (to have broken two switches) that we'd have seen this more commonly... Odd.

Anyway still hoping you are onto something here.

Alan

Manfred

Hi Alan. Did you see post #80? I think I may have this thing licked. I'm thinking the mangled contacts in the receptacle are breaking the defroster switches (first my old one and then my new one--even on the first plug in) and causing an internal short in the defroster switch. Although I'm a little puzzled why I can still get the bulbs in both the old and the new switch to light up when applying 9V to K/31 and 58/31.

In answer to your questions though. I didn't do the pod service myself, Greg Nettles was working on it. The pod had been ever so slightly askew ever since I bought the car. I thought it was because the pod height adjuster lever needed repairs (and it still does) but Greg thought it was something else. He pulled the pod off and fixed the slightly askew nature it. When I got the car back I noticed I had no HVAC lights which Greg hadn't noticed because he wasn't driving it at night. I subsequently found fuse 9 blown and started chasing down the short myself.

Unfortunately, everything and anything that could've been done to my car was done at that time. Greg installed a sharkwoofer for me and rewired the stereo from scratch. He also replaced a leaking recirculating flap actuator so he was left, center, and right in the dash. Under the hood he did an intake refresh and replaced every engine sensor.

Manfred

Ah, I see you saw post #80 now; we must be chasing each other around. Yeah, you would think this was a common thing. It didn't take a whole lot of pressure to push the new switch onto the connector. Obviously, Greg broke the first one and I broke the second one, so we both felt like it was a reasonable amount of pressure to apply--the receptacles a pretty tight fit so even when they go on smoothly you have to press pretty hard...

Manfred

I'm trying to source new contacts for terminals 58 and K on the receptacle now (and probably another new defroster switch) so it'll probably be awhile before I post back with new info...

Ed Scherer

Not that odd when you see what they were being plugged into: the receiving contacts were pretty messed up based on the photo in post #59.

Now after seeing post #80 (especially the photo), I think this has all come together. What you wrote in that post sounds like a very plausible sequence of events leading to the symptoms you've experienced.

Is there no hope for bending the contacts (31 and 58) back to usable shape on the socket connector? If you need to get new ones, you might try Eagle Day: see their "Terminals for Connector Housings" page. Even if they don't have the right ones, they might be able to find a match if you can give them a good photo and maybe dimensions.

Likewise, you might be able to salvage one or both of the switches by pulling the loose blade contacts back out and filling the part of the nylon (or whatever the material is) housing immediately below the retaining part of the metal blade with epoxy, or, alternatively, melting it with a soldering iron so that it fills the channel that was likely formed as the blade was pushed into the housing.

BTW, on the switch you showed, the terminal K blade doesn't look that bad to me; if it hasn't broken free (i.e., isn't pushed down into the housing and/or allowing movement in/out), it might just need to be bent straight again. The 31 blade, though, obviously needs attention.

When you do reassemble everything, I'd recommend using a flashlight and watching the engagement of all the blades into the socket-side contacts as best you can (not easy, since you'll just get a little bit of an edge-on view) and making sure none of the blades is being pushed back into the switch housing. If something like that does start happening, then reinspect the contacts (socket side) and make sure the blades and contacts are all lined up nicely. Sooner or later, you'll get it.

Good luck, and keep us informed of your progress. I think you finally have this one figured out.

Manfred

Update. So I never could find a source for the original wire terminals. I wondered about getting a receptacle with terminals in good condition off a parts car and splicing it in to the harness but that seemed like an unclean solution.

I talked to my local Porsche mechanic (Jim Burton-Charlotte, NC--Hendrick Porsche) and he said the four prong wire terminals were discontinued due to their frailty and that they were replaced by two prong terminals which he had on hand. I was pretty much done with this problem and I had other things that needed to be done to the car so I had him do the work. I didn't waste any time seeing if I could repair my brand new but broken defroster switch--I simply ordered a new one to make sure everything was perfect. It's now working perfectly.

Just wanted to make sure everyone knew the problem and the solution. Thanks again for everyone's contributions on this thread.

Dan

Ed Scherer

Thanks for the follow-up. It's always nice to finish the story. Nice to hear that it had a happy ending.

NZ928S4

1991 S4

Dan,


Read through this. I too have #9 fuse blowing. Mine happens when I turn the headlights lights on. Putting in a fresh fuse nothing happens the clock don't go. Previous owners have had the dash pod out. (me too to replace a worn ignition switch). When it blows I hear a click under the passenger seat? I have checked all the wiring under the center console. Took the clock out and yep works fine if hot wired.

I will try disconnecting all the pod switches just incase one has a short in the little bulb.

In the end was it the terminal for the defroster that was what caused the problem on yours?

This problem has been there since I bought the car several years ago. Thought the clock was just dead. But noticed the light for the HVAC did not work. Took that apart yep light is fine in that. The pod lights all work.

So having 10 hours into it and the knowledge you and the other members have shared I have no clue.

Not sure if I will do more damage pulling the pod again. Got a feeling its something close.

Cheers, Dean