Urgent help, Smoked ECU ground traces...

[UglyBoost91]

Hello again,

Thanks to those who helped with my questions last time pertaining to a friends car I'm helping setup. It's a Chevy Vega with a 92 S10 2.8 v6 and stock control system.

Anyhow, the car was running great last year with the stock 2.8, I helped him switch to a 3.4. In getting the car to start yesterday for a break in tune and the new motor, nothing would turn on- no check engine light, fuel pump prime, etc. All of a sudden sitting in the car, having touched nothing, the pump primed and the CEL came on, while smoke poured out of the top of the ECU! I immediately hit the battery kill switch, but it was too late, the ecu was toast. Opened it up to find every ground trace on the ecu Melted! The only thing wrong with the car we found was a ground strap he forgot to hook up in the engine bay. This wire was the main ground wire for the ecu, D1 (TAN/WHITE) on the large ECU connector.

Im picking him up a new ECU from the junkyard today and re-solder a G2 adaptor into it. I just want to ensure this doesn't happen again! Any insight? All the rest of the harness was 100 percent ok, minus that ground lug undone and just resting there. Could that actually cause the ecu to fry, when it has to take all it's current through the case and foil traces? Why didn't the fuse blow? If it fried the traces in the ECU, wouldn't that be too much current for one ground wire anyway?

So confused and he's frustrated. Nothing on the harness was changed, just a larger displacement engine since last time..

All the best,
Eric

[brian617]

Smoke is voltage shorted to ground. I would look for pinched wire etc. Start with checking ECM grounds at ECM connector verifying no voltage with good continuity to ground. Then check 12v inputs for short to ground.

[notime2d8]

...yeah, I had that happen before and it was the grounds for me too. The best advice is to double check all of your grounds and make sure your wiring is correct - no 12v to any grounds & vice versa, maybe take the ecm out and use a simple 12v probe light to each pin on the ecm connector. Also, maybe get a spare ecm while you're at the junkyard.

[UglyBoost91]

Thanks guys, I'll make sure to take a look over everything again. I probed all of the connectors last night to make sure, and all the grounds and power feeds in the right locations. I'm just scratching my head here. It's like 12 Volts was sent through the ground plane from somewhere as you said...

[lionelhutz]

Also make sure the battery to engine and body grounds are all solid. Main lead from battery to the engine block and then good ground leads or straps from the engine block to the body. You could add an engine block to frame ground too but generally it's not used for any power circuits. However, if anything is grounded to the frame then it's required.

[EagleMark]

There was a time when you let the smoke out of an ECM then it was no longer usable. But now thanks to Lucas you can service the item by replacing the smoke!

http://www.gearhead-efi.com/Fuel-Injection/attachment.php?attachmentid=6819&stc=1&d=1396060899

[RobertISaar]

and there is PLENTY of Lucas smoke to go around.

[ony]

I would just try the 2.8 chip and make sure everthings working then install the adaptor.

[Bigdav160]

I'm going to disagree with Brian's assessment slightly. Melting of the wires is from too much current flow. That *could* mean a direct short to ground but not necessarily.

It just means the expected resistance in the circuit was less than specified. It sounds like a couple loads found their grounds through the ECM. Very similar to my buddies Corolla that would melt his clutch cable because of a bad ground connection between the battery and engine block.

[UglyBoost91]

Thanks for the replies. Once we fixed the ground issue and tried the fresh ECU, everything was functional. The old ECU was smoked for sure.

[brian617]

I'm going to disagree with Brian's assessment slightly. Melting of the wires is from too much current flow. That *could* mean a direct short to ground but not necessarily.

It just means the expected resistance in the circuit was less than specified. It sounds like a couple loads found their grounds through the ECM. Very similar to my buddies Corolla that would melt his clutch cable because of a bad ground connection between the battery and engine block.

I agree, wire carrying to much load because proper wire is broke or missing is a likely problem as well.

[UglyBoost91]

Thanks for all the help Gents,

Ended up getting the engine running well (and emulating in my other thread). To the best of my knowledge, the block grounds for the main wire harness that fed the computer were disconnected. All of the current was "sunk" through the case and ground traces on the board, instead of the proper wires, causing the traces on the board to burn up.

I appreciate all the quick insight, it's rare to find a forum that is so helpful. Hopefully someone else can learn from our miss-fortunes!

That Lucas smoke is great by the way, best quality I've found ; ) hahaha