Chasing electrical bugs

[czwbkx]

One by one the electrical circuits on my 911S are shutting down. Now it's serious - the most recent to stop working is the right-side lighting so I can't drive!

I've checked the fuses and relays, cleaned the contacts, and verified that there is power at the fuse, but still can't get lights. The total list of circuits that've shut down in the last few weeks includes power windows, fog lamps, passenger side driving lights and side markers and tail lights. Brake lights and turn signals work fine. Bulbs all check out good.

BTW - amazingly, both power mirrors work. I thought they were usually the first to go!

Anyone else have to diagnose such a problem before? Any hints? Otherwise I'm back into the garage with lots of wire and a VOM.

Thanks,

Ray

[davis911s]

Ray what year? If there are more than one circuits I would suspect a poor ground. Chech to see that your grounds are tight and clean too.

Shawn

[911pcars]

I agree with Shawn. Locate and check all the ground circuits/connections. Take them apart and clean the base metal; spray with contact cleaner, then assemble.

Sherwood Lee
http://members.rennlist.org/911pcars
www.seinesystems.com

Look at the circuits and see if they have any common grounds or power sources.
Also since your car is old enough use a test light for checking for voltage. If you have a corroded connection you may read good volts with a meter and still have a bad connection. So load the circuit with the test light to verify it.
Or in simple terms connect your voltmeter and then touch the connection with the test light. If the volt meter still reads 12v or so the connection should be good. If it goes to zero or 1 or 2 volts then you are looking at a bad connection.
Good luck
Chris

[ClayMcguill]

Check out this article:

http://grassrootsmotorsports.com/zap.html

[911pcars]

The article provided some helpful troubleshooting hints. There are two schools of thought in electrical diagnosis; using a test light or a voltmeter. A test light is a go-no go device. Connected to a circuit with voltage, the bulb will light; no voltage and it won't light. Pretty simple. However, there's no way to tell how much voltage is reaching the accessory with a test light. Automotive electrical accessories are designed to operate most efficiently with 12 volts. Anything less and the accessory (load) may not work as well or at all. If a circuit only has 10 available volts (instead of 12), the test light will still light which mistakenly indicates a functioning circuit. Not neccesarily so.

A voltmeter connects to a circuit in the same way as a test light and will tell you exactly how much voltage is reaching the load. Not getting 12? Trace back toward the source (battery) and find out where you lost 2 volts; usually a corroded, dirty or loose connection between source and load. You can also connect a voltmeter to measure the amount of voltage loss (voltage drop) in a particular segment of the circuit.

In addition, modern vehicles use several electronically-controlled systems. Many of these systems use a 5 volt signal to their sensor circuits. Probing these circuits with a 12 volt test light may/will fry the electronic control unit (expensive). Even an analog voltmeter can cause damage. A good digital voltmeter (VOM) can prevent this from happening.

For pinpointing the cause of a battery drain (parasitic current drain), I'd suggest connecting the ammeter (part of the VOM) in series with the positive battery cable, then follow his suggestions (pull fuse, etc.) for isolating the offending circuit.

Sherwood Lee
http://members.rennlist.org/911pcars
www.seinesystems.com

[czwbkx]

Thanks for the help so far, Team. I'm in the process of disassembling the offending ciruits. Battery out and fuse block loose so far. The right-side parking lamp circuit got so hot that it has melted some of the wire insulation!

First task is to remove all the patched-in wiring from the PO. He added two power amps for the stereo and *shudder*, it looks like there may have been a CB in there! There is a PL259 hanging from some coax under the dash! Lookin' for "smokies" back in 77? Even though the car ran well for several weeks it is going to hell fast - these circuit hacks have to go before I can test the rest.

"How long before the warp drive is back on line, Scotty?" "It'll be at least several days, Cap'n." While the battery is out I may as well go ahead and clean up the box and coat with POR15. There's no rust yet, but why wait? Then all the lamp assemblies come out for connection testing and clean-up.

This is a good thread to show the problems a DIY person can run into.
Using the the apropriate tool for the job is very important as 911pcars sugests.
A good DVOM will work for testing circuits that require it's use, computer type circuits etc. But in lighting or non sensative circuits a properly used test light could be the better tool. A circuit may read 12volts on the Fluke 88 and seem good, but when tested with a test light which loads the circuit it will fail to light.
I guess it comes down to experience and familairity with the circuits envolved.

Chris Sherwood ASE Master Tech

[911pcars]

"A circuit may read 12volts on the Fluke 88 and seem good, but when tested with a test light which loads the circuit it will fail to light. "

I don't purport to know everything about electrical troubleshooting and I'd to learn as much as I can which is why I ask this question:

If a voltmeter (analog or digital) is connected to a "working" circuit (circuit switch is ON), in what case would a voltmeter read 12 volts (or source voltage) while the test light will not light? IMHO, as long as the circuit switch is ON and the meter is connected to the load input (e.g. light bulb hot wire), the meter will indicate source voltage up to that point, whatever that is. The test light should illuminate unless the voltage isn't high enough to light the bulb.

Put another way, if there is enough circuit resistance to prevent the test light from going ON, then this will show up as something less than 12 volts on the voltmeter. Yes?

Thanks,
Sherwood Lee

Think of it like a garden hose.
Example If you have 12psi of dead head pressure to the valve.

And you have two hoses one has an ID of 0.05" and the other 1.0".

Little Johny is out playing with political prisoner Barbie giving her the old blow torch and puff up goes Barbie and the lawn.
With the little one you have an impressive stream but not enough volume of water to put out Barbie. Now the other hose has the volume to do the work.

Both hoses have the same "pressure" but the larger one has the volume needed to perform the work. Putting out Flaming Barbie and the lawn.

Same idea with the wiring issue, for what ever reason the circuit cannot support the requirments of the test light but it can the Fluke meter.
Burned contacts, corrosion or a load higher than the capacity of the wire or my favorite one strand of wire left when there should be 20 or so.

The Fluke meter does not load the circuit which is important for working on circuits that are load sensative IE working in uAmps verses mAmps or Amps.
Where as the test light would smoke a circuit that opperates in uAmp's and mAmp's because of it's Amperage draw on the circuit. That is the advantage of the test light it loads down the circuit and will provide a visible indication of the circuit condition. Example being bright good, to dim something going on here, to not lighting we have a problem. At this point you would use the DVOM to further troubleshoot the problem. It works quite well in complement to the DVOM depending on the problem and circuit in question.
The key is picking the correct tool for the job at hand.

Keep on trying ask questions and learn from other peoples mistakes.....it's cheaper and less embarrising that way.
Chris