Ashamed to admit that I'm an electrical engineer, v.BMW electronics

Ohhhh BMW electrical gremlins.

I've been chasing around a stray current draw and charging issues for almost a year in my car. I FINALLY got time and an area to go through it, and thanks to 2 multimeters and a E36 M3load of test leads, I finally figured out where the current draw was coming when shut off (electric windows and a faulty door switch). I also replaced the alternator which remedied my charging issues.

SCORE! Everything is mostly fixed. The only issue I am having is an intermittent gauge issue. I have an accessory oil temp and oil pressure gauge in a DIN pod in the dash. They're nice-ish Autometer units with the shmancy new LED internal light (i.e. no plug in lightbulb anymore). It's now just 1 mini harness for each gauge.

Basically what was happening was that 1) the lights wouldn't work, and 2) my oil pressure gauge worked very sporadically, like <10% of the time. It wasn't repeatable and I couldn't figure it out or replicate the issue on demand. It was maddening the crap out of me.

I figured bad grounds, loose connections, or improper wire-splicing. Last night I spent 6 hours just doing gauges. I figured out the lighting issue quickly (I cut into the wrong wire from the stereo harness), and that was that. I now have backlit gauges. OH SO PRETTY!

So finally after a bunch of tearing the car apart, I'm down to 1 single issue- why is my oil pressure gauge working intermittently? I started from the gauge and worked all the way through to positive battery. Nothing was loose, I was getting 12v, and I even powered both signal wires from the oil temp sender to ensure connectivity, which I had. What in the berkeley is going on... Rechecked. Re-re-checked... Nothing. I could hotwire the gauge and get functionality, but I couldn't for the life of me get the existing ignition +12v wire to power it, even though it was verifying switched +12V on a multimeter.

Basically, repeat that for 4 straight hours last night. Hammers were thrown. About 100' of spooled wire was sacrificed for jumpers...

I go into work this morning with the car, with the intent of working on it at lunch, and asking a couple coworkers to double-check my work. I get in, explain the situation to my buddy, and he goes 'Did you check to see if you maybe pushed a contact out of the harness going into the gauge?'

Mother of Mary, is it that simple...?

Pull the panel, fire up the car, look at the crimped pins in the back of the connector and they look fine. They arent exposed or anything...? Use my pin probe and push down on the +12v wire... it moves like 3mm and makes and audible 'CLICK'... turn the gauge around, and I have oil pressure. I immediately hang my head in shame at the realization that I overlooked the SINGLE SIMPLEST AND MOST PLAUSABLE EXPLANATION for my issue, and now I'd have to hear it from my coworker for the next 15 years about how he solved my problem.

Just remember. If it seems too simple to be complicated, 99.999% of the time it is.

Autometer- 1, Me - negative 6 hours.

Don't feel too bad about it. I once parked a car for 8 months because I disturbed an iffy ground hiding under the coil bracket and made it into a bad ground, and apparently couldn't be bothered to trace the 12 inch wire grounding the ignition module. Worked on it for like a week, parked it, had it running in like half an hour when I finally got back to it and realized what I berkeleyed up.

One of our techs once ripped the entire cab of a truck apart trying to figure out what was going on with the airbags, completely stripped down. He's getting mad, finally asks another tech what is going on because the diagnostics don't deal with what he is finding and nothing is making any sense. Turns out that his test light had a break in the cable and he was chasing his tail.

In reply to NickD:

I was berkeleyed by a faulty test light once. Ever since, I've pretty much only used DMM set to beep on continuity.

BrokenYugo wrote: Don't feel too bad about it. I once parked a car for 8 months because I disturbed an iffy ground hiding under the coil bracket and made it into a bad ground, and apparently couldn't be bothered to trace the 12 inch wire grounding the ignition module. Worked on it for like a week, parked it, had it running in like half an hour when I finally got back to it and realized what I berkeleyed up.

Did that too... that was a bad day.

No spark... why?

Meter coil, looks good. change coil yup it was good alright.

change points (this is on a Lada) nothin

change condenser, nothin

change cap and rotor, nothin

meter HT leads, all good....

check coil positive for 12v, check ground for continuity. yep all there...

Wait a stinking tick! why is there a good ground when the points are open? where does this wire go? oh god I'm an idiot...

Turns out there was the usual ground wire running from the coil to chassis (through the points), but someone had an aftermarket tach installed at one point and left that wire going from the coil - terminal to behind the dash. Apparently it just never touched anything until that day...

In reply to AngryCorvair:

I just always check the test light before I start testing wires. But I use a DMM as well.

Similar issue in a 1985 Bronco, car would run fine, sometimes. others it would not start just crank until it flooded. did new coil, plugs, wires, rebuilt the edlebrock carb on it, had a buddy take a look at the two lead wires going TO the coil, one was broken, green and corroded. took me two weeks to figure it out

At least you heard the click and got satisfaction. Try not to remember the lost time or the ribbing and think about how great it is to have a working gauge. It may take 20 or 25 years but the strife will be a dim memory someday...

Did similar on my Opel GT. Started when I replaced the broke cigarette lighter to have a plug for GPS. Didn't have any dash lights after that so started digging into that and found that the very last light in the series circuit was indexed and I didn't index it. Indexed it and had lights but then fuel level & oil pressure gage wasn't working. Dug through the whole system and couldn't find anything wrong. Double checked and triple checked every wire and ground in the system. Finally started going through the fuse panel and found a burned out fuse. Replaced that and still didn't work and then found that one of the plugs in the fuse panel was loose. Thought I had checked that numerous times. Snapped it in and everything works. I'm now real good at removing and installing the instrument panel in an Opel GT.

There's a reason we call them gremlins . . . it's never easy, until you have already found out what's wrong.

If it makes you feel any better, I'm a mechanical engineer and can't figure out how to adjust the E36 convertible top to get everything working smoothly, either.

I had an early 80's Volvo wagon with L-jetronic injection and an intermittent issue with the lift pump. I pulled the access panel for the lift pump and everything looked good. I ended up pulling up the carpet along the drivers side and tracing the wire all the way to the fuse box. It turned out that the push-on connector for the hot wire from the fuse box had vibrated loose. After that I learned that if I have a power issue, start at the source, not the destination....,