Resolved 1990 Fuel sending unit phantom power.

[HallerDSM]

1990 Eagle Talon TSi fwd.

So a few weeks ago i left a buddy's house and barely made it home. When i pulled in, smoke had surrounded car and started to fill cabin. I was able to trace the source down to fuel pump wiring harness. (melted mess) Just happened outta the blue.

Figured i had a short and replaced harness with good one from donor car. Started car and bam instant smoke, harness fried all the way down to the cabin connector again. Cut open harness and found the culprit is the ground wire coming from the fuel level sending unit even though the gauge functioned properly until I shut off the car. I replaced the sending unit with a functioning one from another car and replaced ground.

Started car, and again burnt up the ground instantly. Next I replaced the ground with larger wire directly to chassis and still same result, burnt ground right after i start the car. I ran new sending unit wires and a new ground still over loading and burning ground. Replaced gauge with new one and same results.

No matter what Ive tried there seems to be too much resistance going through the ground wire instantly over heating it. The car runs fine when the connector for sending unit is unhooked, and the gauge works perfect with the key at the acc position. I know the sending unit, gauge, and wires (other than ground) are all good.

The problem arises when i start the car the ground instantly burns up. I'm thinking something could be causing a positive ground issue since there is very minimal resistance produced by the sending unit itself. (Although nothing else seems to be affected). I'm going to be checking my chassis grounds next but any other thoughts or suggestions would be appreciated.

This just isn't making sense to me, I'm scratching my head on this one. P.S. I have a Walboro 190-rewired with external relay that seems to be functioning normally.

 

[jimsgsx]

Maybe your pump is drawing too much power causing this problem. I would swap the pump and check it out.

 

[wrencher]

I would suspect the pump itself as the culprit.
So If yours is a FWD it has the pump unit w/2 wires & then the gauge unit with 3?

Sure the pump's ground is OK?
The pump is the only thing with the amperage in that area that can torch wires.

 

[HallerDSM]

Yeah thats correct about the wires. That was my thought process also. I checked the pump and relay grounds their good the car will start and run fine (so pump is grounding somewhere?) without the sender conector pluged in. After I'm done checking grounds I'm gonna pull pump and see if it's somthing with the pump itself. Thanks. Just can't wrap my head around how the sender would get the power from the pump.

 

[NHerron]

I honestly mean this in the nicest possible way- for the safety of yourself and others around you, invest time in learning proper repairs. You clearly have no idea what is going wrong. Electrical fires are nothing to scoff at and you could have a huge accident. Call me whatever you want or thank me, just understand the possible outcomes of what you are doing man.

With that said, here is a very good link to factory service manuals for your car.
Mitsubishilinks.com
I'd say your relay/modified areas are your problem, because the pump doesn't run until ecm sees cam sensor outputting a signal. So you wouldn't have the smoke happen until car was running or cranking. And stock wiring shouldn't smoke or burn like that, its properly fused, furthering my lean on (unfused) modified areas.

 

[HallerDSM]

Yes, thanks for the concern I understand this. I'll keep the names to myself. I come from a family of technicians and have been working on cars/engines and electronics since before I could ride a bike. So I know the dangers and practice proper safety in everything I do. If i didn't know fire+gas=bad things can happen, I shouldn't be around a car anyway haha. But thanks for the advice and the link, i have a copy of the service manual.

I'm somewhat new to DSMs and I don't know what's going wrong or I'd be driving my bomb around instead of having it up in the garage.

I'm under the impression that these sending units create their own resistance with the float? I may be misinformed but then there would be no power draw when the car is started causing only the SENDING UNITS ground to melt. Unless it's somehow shorting to something else. (Ie. possibly the fuel pump) So yes the fuel pump doesn't activate until the ecm receives the proper signal but it is in no way tied into the sending unit other than their near each other on the tank. And wouldn't affect it unless there's a fault somewhere which I understand just haven't found the reason yet.

I daily drove it for 7 months with no issues after the pump swap so that's why it just boggles me a bit.

Still open to helpful suggestions though.

 

[NHerron]

OK.

Diagrams are for a 90-91 FWD Turbo

It appears your pump is grounded inside the tank as a main ground, then a redundant ground is used since the connector is 2 wire

I would carefully check the walbro and relay connections again. If the larger than stock pump is grounding through a even less than stock size wire (ie gauge ground) then... ya know... test light will do wonders here. Disconnect connectors, check for ground, etc.

Good luck

 

[HallerDSM]

Thanks for the help issues been resolved.

The Insulator on the fuel pumps power connection through the pump mounting plate had worn through. This was causing a short to the tank. Put in new bolt with seal, good as new.

 

[NHerron]

Good to know and thanks for updating as resolved.