Technical I keep burning up GM Headlight Switches

First one was a week ago but I thought it was because the rheostat was old and didn't work well anyway- I had a thread on that one. I've owned the car for a year and it's been on the road since before then without this issue.
I snagged a new switch last week and swapped the wires over (it's a GM 8 pin switch) and the same thing happened today- the first time I turned the lights on since swapping the switch. What could be causing this?
I have a video but I can't seem to post it.

 

you have to post your video to youtube first, then post a link to it here.

Are you just burning up the dash light rheostat, or is another part of it not working?If it's the dash lights, there might be an intermittent short in one of the sockets, or something, that's causing excessive load on the rheostat.


And did you just get a new parts store switch, or did you get an old NOS switch?

 

Are your headlights running thru relays? Or pulling direct current thru the switch to the bulbs? Did you recently switch to halogens?

 

If it is the rheostat that's burning up, I am experiencing the same thing. I replaced a headlight switch on my 57 T-Bird maybe 10 years ago because the rheostat worked very poorly. I had to fiddle with it to find a spot that the dash light would work. The new one worked for maybe a year and had the same problem. I finally moved the dash light wire to the headlight output. It appears new headlight switches leave a little to be desired.

 

It was a brand new O’Reillys switch made by standard American, or some thing. I know that the NOS switch would have been preferred, but I just figured the old one was so bad the way that the Rio stat felt. This literally happened after three seconds so I tend not to blame this particular switch.
The headlights are halogen, but that did not change at all they have been since I’ve got the car. As far as I can see there are no relays. But again it hasn’t needed relays for the past year anyway.
The latest thought was to pull out the headlight switch and start pulling wires until it stops smoking, seems like a great way to burn myself though! Even with welding gloves.

https://youtube.com/shorts/_gLzxH7mulQ?feature=share

 

Why not remove all the wires except for the power wire and then start adding wires one at a time. When it starts to get hot you have found your bad circuit.

 

You didnt by chance oil or grease the headlight switch rod did you ? Or touch the resistor coil with oily hands ? Just some thoughts . Maybe the switch has the headlights themselves wired thru the reostat . You should have 2 power supplies for the switch .

 

I thought about unplugging all of them, but I’m sure I wouldn’t put them back on right!

I didn’t really touch anything, and this is a continuation of a problem that I had last week.

Should be fused and un fused, right?

The full story is that I was on vacation two weeks ago and we went to the beach. I’m not sure if salt air had something to do with it, or if general condensation from overnight had anything to do with it, but in the mornings the outside of the car would be soaked. Since it’s not necessarily watertight, I’m wondering if moisture got up under the dash and reeked havoc with this switch. There was an ignition switch problem that reared its head before this one, but I figured that one out relatively quick and it runs fine now.

 

Me thinks bad ground SOMEWHERE.

Ben

 

I think BJR is right...I need to label all of the wires with numbers to where they attach on the switch and remove them all and start plugging them in to see what smokes.

 

Sure looks like the wire going to the rheostat needs to be checked for a short to ground. Disconnect the wire from switch and measure resistance to ground with a volt meter. If the measurement is very close to 0 ohms, then a short is probably present. Or splice a 12 volt (57) bulb or a trouble light from a 12 volt source to the wire unplugged from switch. If it lights up bright then you have a short in that circuit.

 

I thought that's why we have phones.

 

It seems like you may have more than one issue going on, the combination of which is killing your switches.
The first thing I would do is route the headlights through relays. Halogen lights are a bigger load than good old fashioned sealed beams, and you might be on the ragged edge load-wise. Then trace out all other switch functions individually, one at a time, paying close attention to grounds and connections (including any splices-rewire to eliminate splices wherever possible).
There's a basic system problem that won't go away by simply replacing the switch on a regular basis. Good luck!

 

I am going to do some thinking "out loud" here...

I unhooked the feed to the gauge lights (wearing gloves this time) and the switch does not get hot or smoke with the lights on.
I followed that wire to what I would call a "junction block" where all of the light bulb wires for each of the 6 (Stewart Warner Wings) gauges came in.
I removed all of the wires from that junction block, thinking that it might have been shorting through the block. Every one of the wires going to the bulbs for the gauges has continuity with ground when checked one at a time.

If I am not mistaken, I always thought that if it was one wire for the bulb, that was 12v and they grounded through the mounting bracket.

I am running a jumper wire from a bolt on the firewall and getting continuity to the SINGLE wire to each bulb.

Then I ran jumper cables into the cabin (carefully) and checked to see if the gauge lights worked, and they do...
Why would the headlight switch burn up but not the tiny little filament in each bulb?

The ONLY suspect wire I found is in the photo called "back of tach circled" and goes to the surround for the light on the tach...but that looks like a ground wire and isn't my problem that I have too much ground?

 

I think that you just found the problem. That heavy orange wire is not a ground! It is shorting to ground intermittently and causing your switch to heat up and burn out. Tape it up and move it away from the light socket it was rubbing on and see if that fixes it.

 

Bjr- you mean the wire where I circled? That is the ign to tachometer and the photo is an optical illusion- the wire is in good shape and it isn’t rubbing…also no continuity with that wire’s terminal and ground.

How does light have continuity with ground and the bulbs not burn out?

 

The ground completes the circuit back to the battery.

 

But is it supposed to ground through the one wire going to the light? Seems like it should ground through the gauge or gauge bracket- not the wire. Right?

 

Is your wire/wires to the dash/gauges fused after the rheostat as should be per how GM did it? I’ve had gauge sockets go to ground very easily. If a gauge wire goes to ground and the circuit is not fused as mentioned the coil on the rheostat will become a heating element for short bit until it burns up.

 

Do a forensic exam of the failed switch(s). Look for signs of arching and pitting on which contacts, spade connectors, rheostat, etc. That should give you a clue.
Personally, I just use relays on headlight switches as a rule. Rheostats are in a category by themselves. I usually set them once and rarely touch them after that. I tend to set the dash lights for comfort when it's dark and give up some visibility at dawn and dusk.

 

There was no fuse there, which I found odd. How many amps should that fuse be? 10?

It still seems (to me) like the problem is with the lights. Even with a fuse, the fuse would just blow, (only the switch wouldn’t burn up!).

Is it possible that all 6 gauge bulb sockets are shorting out?

Plan B would be to do some kind of dash mounted lights completely separate from the gauge…fused to either the headlight switch or some other switch of my choosing…

This attachment says that one of the 12v feeds (orange one) to the GM switch should be fused- mine was not.

 

Shift Wizard- this is the first switch- that I replaced a week ago and came with the car…
Terminals look ok but the rest burned up. I caught this new switch smoking before this could happen.

 

@Deyomatic yes fuse will blow but what if all other functions remain? Kind of begins to narrow things down?

 

Orange wire to switch should be fused and protects the tail lights and dash lights.

Still no accounting for why the rheostat would overheat and meltdown.

Since there is no problem with the switch with the gauge light feed wire disconnected and you futher tested that all the gauge lights work; is it a possibility you missed a worn spot or something in the gauge light feed wire between the switch and the junction block?

 

^^^^^ orange wire per GM color code is an out feed to fuse panel for things like glove box light, dome light… So again fuse after the switch not before.

 

Dash/ gauge lights are fed power into the center of the socket. Current flows through the wire to the center of the bulb then through the filament to the bulb case which is touching the socket then goes to ground which is why you get continuity there.

 

^^^^^ True. Also and will cause confusion, what if the br*** tip that makes contact with the lamp has made contact with the spring that keeps pressure against the bulb? Or as Op is suspecting “super ground”.

 

If you are frying the rheostat as soon as you hook things up or turn the lights on you most likely have a short in the dash light wiring. The Rheostat should only connect to the dash lights and switch the dome light off and on when you turn the knob all the way in one direction. It should have nothing to do with the headlights outside of turn the dash lights on when you turn the headlights on.
I'm thinking that if you manage to look up under the dash you will find a wire with a burn mark on it where it has shorted out to ground. That is unless you just flat hooked something up wrong.

The headlights don't dim by turning the knob to dim the lights unless you really screwed up hooking te lights up.

 

On a whim I just checked for continuity with the body i.e. ground, and every other wire going to that headlight switch. Every other wire except for the two that are labeled 12 V on that above diagram have continuity with ground. When that first switch burned up, could everything have melted somewhere else? I’m really confused right now. And this is well out of my wheelhouse!
I’m wondering if the first switch burnt up because of a bad Rio stat, and took everything else out with it.

 

What does "continuity" mean in this case? If they go to light bulbs there is contact to ground through the bulbs, some testers would call that continuity despite being perfectly correct for this circuit.

Many continuity testers will call anything below 50 ohms or something like that continuity. Quite useless, unless you have a wiring harness full of blue wires, and you need to find both ends of the same wire.