Strange electrical problem

[Crazy_pilot]

My car decided to start acting up today.

I'll describe the symptoms as best I can, they're odd.

First, waiting at a red light the radio (stock) started to make random thumps from the speakers, still playing tunes. Light goes green, I pull ahead to the next light. The thumping gets worse, so I turn the radio down. The speakers start putting out a short electic noise and the occasional thump, then the radio shuts itself off.

Next the real trouble starts. The light goes green, but when I push the pedal the engine nearly dies. It repeatedly gave short bursts to ~1200 RPM, then stopped firing, but restarted before the engine had stopped turning. It did this over and over, taking maybe 1-2 seconds each time. I managed to get it pulled over and parked.

The car kept running, and after about a minute of this the engine began running normally again. Revving normally, idling normally...I turned it off and started again, all normal.

I get the rest of the way home, but as I'm pulling into the driveway the speakers start thumping again. In the driveway I roll the windows up and close the sunroof, and they're barely moving. Revving the engine to 2-3K RPM has no effect on the window/roof speed. Also, the air bag and ABS light started to light dimly.

I just checked the battery, it reads 12V bang on.

So I'm at a loss as to what this could be. Any ideas?

[cruisen_15]

hmm, i had a problem kinda like that, my ecu was messed up, got a new one. maby not your ecu but hey, thats what i can think of...

[dannmarr]

Check the battery with the engine on. It should read 13-14Volts. If not, the alternator should be checked.

[crazyhorse]

Being intermittent in nature, if you can use some alligator clips to hold your meter leads, you can monitor the voltages. My bet is that when you get the thumps, bad running, the voltage drops.
It COULD be the crank pulley separating.

[Trevor]

You specifically describe a thump, rather than crackles or other noise. By any chance do you have a high power amplifier and possibly with a back up capacitor installed? If so try disconnecting it.

[Crazy_pilot]

Originally Posted by dannmarr

Check the battery with the engine on. It should read 13-14Volts. If not, the alternator should be checked.

I'll check that first thing tomorrow.

Originally Posted by crazyhorse

Being intermittent in nature, if you can use some alligator clips to hold your meter leads, you can monitor the voltages. My bet is that when you get the thumps, bad running, the voltage drops.
It COULD be the crank pulley separating.

My understanding is that the pulley is two peices, so by separating do you mean the outer circle slipping? I think it's fine, the steering never got heavy.

Originally Posted by Trevor

You specifically describe a thump, rather than crackles or other noise. By any chance do you have a high power amplifier and possibly with a back up capacitor installed? If so try disconnecting it.

Several thumps, as well as a whine. I do not have any aftermarket audio components.

Thanks for the replies so far guys, any continuing thoughts would be appreciated!

[Trevor]

This advice is significant ---- “Also, the air bag and ABS light started to light dimly.”

As the first up option, everything now indicates an intermittent connection and or shorted turns within the alternator. A bench test with the machine cold, may not show this up. It must be accepted that the fault is not permanent when making tests.

[crazyhorse]

Originally Posted by Trevor

This advice is significant ---- “Also, the air bag and ABS light started to light dimly.”

As the first up option, everything now indicates an intermittent connection and or shorted turns within the alternator. A bench test with the machine cold, may not show this up. It must be accepted that the fault is not permanent when making tests.

That's why I suggested "monitoring" the voltage with his meter. That way he can take note of where the voltage is when the car "acts up"

This also adresses one of the very few things I dislike about the SVX, a lack of insrumentation. Every other car with sporting intentions has a full guage package. (Tach, Speedo, Temp, Oil pressure, Volts, Fuel level) Even My Olds has those, & it makes no sporting pretense.

[Trevor]

Originally Posted by crazyhorse

That's why I suggested "monitoring" the voltage with his meter. That way he can take note of where the voltage is when the car "acts up"
.

This will be a worthwhile test which should pick up a fault, but it can not be absolutely conclusive. The slow response of a DC meter which does not incorporate a bar graph, may not show up errors of short duration much less one involving reverse voltage.

[Crazy_pilot]

Alright, tested the voltages again. While running the battery reads 14V steady. I had my brother give me a quick hand. Engine RPM from idle to 3500 RPM had no effect on the voltage. I also turned on most of the electrical systems, which reduced the voltage to ~12.5V. It's not a digital multimeter, so I can't be decimal accurate.

There was one oddity. Before starting the car I checked the battery. It didn't register a voltage. In fact, the needle on the multimeter swung as far below zero as it could. The car was able to start on it's own though, although the cranking speed was slower than normal. After being run for 5 minutes the battery read just over 12V with the car off.

[dannmarr]

Probably the leads from your meter didn't have good contact on the battery terminals. Check it again, also check that the battery cables are tight on the battery terminals. It's ok if it reads 12V with the engine off.

[Trevor]

It would appear that all was well when the voltages were checked. Was this in fact the situation?

Surely it is obvious that no form of testing will indicate a fault if the fault is not present at the time. This is the nature of a very difficult scenario.

[Crazy_pilot]

Originally Posted by Trevor

It would appear that all was well when the voltages were checked. Was this in fact the situation?

Surely it is obvious that no form of testing will indicate a fault if the fault is not present at the time. This is the nature of a very difficult scenario.

Except for the one anomalous reading before I started the car, all appears normal. I'll have to do as crazyhorse suggested and run some leads to the battery so I can watch the multimeter from inside the car. I'm busy with work until monday, so any testing will have to wait until then. I'll be sure to update once I've got some more information.

[Crazy_pilot]

Quick update: I took the SVX for a quick spin with the multimeter attached to the battery to see if I could get some more information. Nothing happened. I drove around the neighbourhood for about 10 minutes, but didn't get a single odd signal from anything. Everything works fine, all the numbers check out.

Kind of annoying. Until I can figure out what this is I don't feel confident using the SVX as my daily driver. I still have the Taurus as the backup car, but I'd rather be driving my Subaru.

I'll keep testing it and post any new information.

[Trevor]

Chris,

I wonder if after ten minutes running the engine compartment was not as hot as when the original fault occurred.

I still suspect an alternator problem. Shorted turns within windings very often require heat, before a breakdown will show up. The same goes for intermittent connections involving heat and expansion.

If the fault returns, disconnect both alternator leads as a possible means of getting things going.

I fully commiserate with you. Trevor.