Electrical fire! Fuel pump the cause?

JohnJL · 08-29-2009, 06:40 PM

Yesterday was a sad day. The 1971 let all the smoke out of its wires. Luckily the fire was small and went out quickly. There was no further damage, the car was in a safe-ish place, and I was able to tow it home with the Patrol. It doesnt even stink today!

I guess it could have been sadder.

I am perplexed by the symptoms and cause and am interested in ideas.

Last weekend the car died on the highway. At first I suspected a battery or alternator failure as in my recollection I lost all electrics but no fuses were blown. I towed the car home and went about looking everywhere for a popped fuse. I found that the fuel pump wasnt getting voltage and confirmed it worked by jumpering it directly to the battery. I traced the fault back to pin 50 on the ignition switch not feeding the pump relay, but when I removed it, cleaned it and reinstalled everything worked again.

So I took the car out for a spin around the block, everything is normal. I get about 5 miles and the car dies again. This time all the circuits work except the fuel pump. I removed the ignition switch plug again and notice the red #50 wires are a bit warm. I figure maybe the fuel pump is dying and drawing too much current, maybe it needs to cool down...but it wasnt long before white smoke starts billowing out of the front of the car. I shut everything down, grab my extinguisher, open the hood and disconnect the battery.

Lots of fried wiring, from down by the fuel pump and back all the way to fuse panel. Funny thing is, no fuses popped. You can see here in this pic the wires are fried back to the fuse panel itself but the fuses are intact!

So I'm going to embark on a complete re-wire and I'll separately post on some questions I have on that. I am interested in help on why this would have happened? Is it possible or common for a fuel pump to draw more current as it is dying?

Thanks!

Hugh R · 08-29-2009, 06:55 PM

From the title of the thread, I expected much worse! Slow burns like that are often the cause of dirty connections. Clean all connections. Good luck.

boyt911sc · boyt911sc's Garage

John,

It's difficult to judge the 'blue fuse' (5th from left in pic) but it looks different from the other blue fuse. What's the amperage rating for this fuse? 25 amps? Glad you were able to prevent further damage on your wiring. A FP drawing too much current than specified is an early indication of failure. Keep us posted.

Tony

911pcars · 08-29-2009, 09:18 PM

In the photo showing the fuse box vertically, what side of the fuse box is source voltage? If it's on the right side, then the red source wire to the fuel pump is attached to the side of the fuse leading to the battery (the source side). And if so, circuit current to the fuel pump is bypassing the fuse box.

Confirm circuit current is going through the fuse first before the fuel pump.

A short circuit in the fuel pump will draw more current through the system.

Hope this helps,
Sherwood

JohnJL · 08-29-2009, 10:43 PM

Thanks guys.

I should note this car ran perfectly for 2 years, including many grueling Targa stages, with no faults. Absolutely nothing was changed when this occurred, hence my suspicion of something "wearing out" like a fuel pump dying and drawing too much current or abraided insulation and a short. I'm just shocked a short didnt blow a fuse. Even a 25 amp fuse should have blown before 12-gauge copper stranded OEM wire melts.

feelyx · 08-30-2009, 12:57 AM

You might want to check your grounding points throughout the car. It could be possible that rust /corrosion in the body could be causing enough resistance to heat up the wires. Especially if the circuit has to seek for ground. When I owned my '71, I had to run jumper wires from different body parts to get a good ground.
You can test with an ohm meter to see what you got between your panels.
Tim

viking911 · 08-30-2009, 04:42 AM

I've only had my 911 for a few months and I'm not qualified to talk about much mechanical yet (first engine drop 2 days ago) but I do know electricity. 911pcars is correct that you should verify that there is really a fuse in the circuit for the fuel pump. The terminal on fuse block might be a simple connection point and not really using a fuse at all. An OHM meter should be all you need, you don't need to have the battery connected to troubleshoot. I would use an OHM meter to track the wiring through the car from point to point and also check for shorts to ground as you go. Pay special attention to wire insulation where wires are routed near mechanical movement. I'd look at how the "hot side" of the fuse block is connected to the battery. If the hot side just consists of a series of jumpers, did one of the jumpers ground out? Is fuse grossly over sized? Just take your time and draw out the circuit as you go on paper, it really helps.

Best of luck,

Bob

RoninLB · 08-30-2009, 05:15 AM

motor windings "sag" over time drawing more current.

personally if the FP is over 100,000 miles it has one foot in the grave and the other on a banana peel

if a wire abrsion or loose connection caused sparking the 9,000F degree plasma is an issue

try to track down source of fire for all of us to learn from

and I wish arc fault circuit breakers were available for cars

Formerly Steve Wilkinson · 08-30-2009, 07:16 AM

I think one of the best things you can put into an "old" 911 is a good Bosch (or equivalent) master switch. It's cheap and easy to install; a 10-year-old could do it.

The advantage of the master switch is two-fold: if you do have a fire, you can instantly shut off the current rather than standing there without tools (perhaps), unable to disconnect a battery cable. Second, you can shut down the entire electrical system when the car is parked overnight, or whatever, and have no risk of a fire burning without your knowing it.

I suppose it's anti-theft contribution is also useful, although anybody who wants to steal a 911 is probably going to do it master switch or not. (My best defense is chokeless PMOs; at least the casual booster will never get the car started, not knowing anything but FI.)

ossiblue · 07:57 AM

On a 71, the fuel pump is not fused. It gets power from terminal #7 on the supply side when the ignition is turned on, that's why you didn't blow a fuse. It does sound like a dying fuel pump that is drawing too much current, so now is the time to check it out and possibly add a fuse between the terminal and the fuel pump. Other suggestions about cut off switches are great, also.

Edit: Here's a suggestion. Fuse #5 is typically unused (no wires coming out of the left hand side) and it is a switched terminal. If that fuse is unused on your car, you can connect the feed to the fuel pump to the load side (left) of terminal #5, replace the existing fuse with one that will carry the load of the fuel pump, and you will have a fused fuel pump. Later, you may wish to add a relay to the set up to take the load off the ignition switch (good idea regardless if you change the terminals.)

RoninLB · 08-30-2009, 08:51 PM

Quote:

Originally Posted by Formerly Steve Wilkinson

if you do have a fire, you can instantly shut off the current rather than standing there without tools (perhaps), unable to disconnect a battery cable.

FD car fire sop

JohnJL · 08-30-2009, 11:17 PM

Quote:

Originally Posted by Formerly Steve Wilkinson

if you do have a fire, you can instantly shut off the current rather than standing there without tools (perhaps), unable to disconnect a battery cable.

I boldly (or stupidly) opened the hood with one hand, extinguisher in the other, and disconnected the ground strap.

Agree on the master cut-off, not just a "coil kill switch" for these same reasons.

JohnJL · 08-30-2009, 11:21 PM

Quote:

Originally Posted by ossiblue

On a 71, the fuel pump is not fused.

This is news to me. If this is true, its a good example of why I refuse the occasional statements like "Why change anything, don't you know Porsche knows better?" Why on earth would the factory not fuse a high-current drawing device that manages highly flammable liquids and is exposed to conditions likely to encourage shorts like water, vibration and moving suspension/body bits?

And why would Porsche choose to pass the full current for the headlights through the stalk?

Arg.

ossiblue · 08-31-2009, 05:50 AM

Quote:

Originally Posted by JohnJL

This is news to me. If this is true, its a good example of why I refuse the occasional statements like "Why change anything, don't you know Porsche knows better?" Why on earth would the factory not fuse a high-current drawing device that manages highly flammable liquids and is exposed to conditions likely to encourage shorts like water, vibration and moving suspension/body bits?

And why would Porsche choose to pass the full current for the headlights through the stalk?

Arg.

...and, why (in the early cars) would the pump be wired to pump raw fuel even if the engine should die, say in an accident?