What's that smell? Wait, and why isn't my car running?!
 

I'm driving along some country roads, enjoying my newly functioning blinkers, good engine temps, and dialed-in suspension, when I start to smell something slight...Smells like something electrical releasing smoke or a hint of plastic burning...I check around the cabin and in the rearview for any signs of smoke (none), and then the motor dies rather suddenly, in motion.

I pull off to the side of the road and start looking around, smelling electrical components like a crazy person. Nothing seems particularly fried. The car turns over with gusto, so I don't think it's alternator/battery. No fuel smell; tank half full. I got the car towed home so I can start poking around tomorrow.

I replaced the DME relay with a genuine Porsche sucker maybe a year ago. I guess it could still have failed, but I'd be surprised. It did feel warm, but not hot. Motronic did not feel particularly warm. I can't here a fuel pump turn on with the ignition, but I'm not sure I ever could. I suspect the fuel pump or fuel pump relay...My pump is below the gas tank.

The motor is a '95 3.6. Any guesses? Anyone gone through this with a good air-fuel-spark checklist?

Damn thing was driving so well. At least it chose a nice spot by a lake to break down.

http://ensaster.com/pics/vehicles/sm...-breakdown.jpg

Ugh...nothing to add but grief and sympathy. Your car is nice, hope you get it back on the road soon.

-Steve

Ignition units are a weak point, as well as bosch silver coils.

When my heater motor brushes wore out, I got smoke in the cabin.
It was that dreadful electrical smell that instills deep rooted fear.

If the fuel pump burned out, I think the smoke could make its way into the cabin.

Maybe you can try the fuel pressure test.




You welcome to borrow mine: http://www.arnnworx.com/catalog/index.php?main_page=product_info&products_id=279#. VNGbfHbokpw

As John said ignition final stages and coils are suspect but I think I'd start poking around near the ECU guessing it's mounted under the seat during the conversion and you had the smell inside
the car. If some "magic smoke" has escaped you'll see the cause.

When you say "ignition unit", what is that exactly? The switch with the key?

Turns out this car has no fuel pump relay—it's in the DME, duh. I pulled it out, sniffed it, checked for bad solder, and put power to it and it's clicking when it should. When I manipulate it manually in situ, the fuel pump spins up strongly. It does not when the key is in the position just before starting the car.

I pulled the Motronic, opened it, sniffed it...Looks clean and undamaged.

I'm about to wander down this list, though it makes no mention of coils...

Porsche 911 Engine Starting Troubleshooting | 911 (1965-89) - 930 Turbo (1975-89) | Pelican Parts DIY Maintenance Article

check to see if you are getting spark.

I smell fuel in the engine bay, which suggests that fuel is getting there when I try to start it...

DME wiring test...With the ignition off, I have power to pin 30 (direct from battery). With the ignition on run I have power to 30 and 86. I believe this means the ignition switch is good, right?

https://dorkiphus.net/porsche/attach...5&d=1206400894

I smell fuel in the engine bay

Like T77911S mentioned, how about the quick basic checks like spark (it sounds like you're getting fuel for the most part). Pull out a plug and see if it sparks.

I couldn't fit my spark plug tester on it, but I pulled one of the wires, stuck an extra spark plug in it, ground it against the AC compressor bracket, and turned the car over...No spark.

Inductive timing light.......
 

Tremelune,

A convenient and efficient way to test for ignition spark is using an inductive timing light. No need to remove or install spare spark plug for the test.

Tony

With the ignition on run, I have power to all four wires at the ignition coils...I guess I'll test those out.

Forgive my ramblings—Besides my astoundingly simple CB350, I don't think I've had a no-start condition in a vehicle in a decade. This is pretty much my day, so I'm researchin' and poking around...

Maybe the smell was a red herring and some passing car had their handbrake on...

Odds of both coils failing at the same time?

I'd think kinda slim.

That's a good point. I was just down a rabbit hole of investigating the possibility of a broken distributor belt, but the car would still run.

Perhaps it's just a blown $1 fuse. Have you checked them ?

Good luck !

Fuses look good. I gave them all a twist...I dunno what they affect, though. Everything seems to be getting power.

I'm looking into doing a more comprehensive test of the DME/wiring, I'm just kinda poking around in the dark, though.

They definitely will run on only one coil... not well, and its not particularly good for it, but an engine that won't run on one coil has more trouble. Maybe a broken distributor belt or a bad coil. etc.

look up the pin out for the DME harness plug, and check for 12v power, with key on. If the ECU has power, then pull a plug wire and check for spark. If there is spark, jump power to the fuel pump and see if you can hear the pump running. If the fuel pump runs, leave it jumped and try to start the car.

Also, pull the caps off the distributor and make sure the belt is intact. It SHOULD run without the belt, but you could have compounding problems.

Check continuity in the wiring all the way to the coil. I had a wire break inside a connector on a car. It took a while to find it.
I made a Jumper wire long enough to go from the PCM to the coils to check. No continuity on #3, so I worked my way back until I found the problem.

I jumped the DME such that the fuel pump and DME were powered directly from 30 (the battery) and tried to start the car, just to double-check. No dice.

With power to all coils, power to the fuel pump and DME, no hint of rough running at any point, and then a sudden failure...It sure points to the DME, right? I'll pull the distributor cap and test the coils, but if there's a way I can bench test the DME somehow...Well, it's easy to get to and sure seems like the culprit, even though I can't see any visible issues.

I would be more systematic, but some things are much easier to verify than others...I also don't really know how the whole system works together. It seems to end with "at some point the DME spits out a spark pulse if conditions are right" but I don't know how to mock the conditions for which it would...

Troubleshooting link: http://forums.pelicanparts.com/2892572-post13.html

1984 pins:
http://forums.pelicanparts.com/uploa...1205355264.jpg

1995 pins:
http://forums.pelicanparts.com/uploa...1330090864.jpg

Have you checked the flywheel sensor. After all, the pulses from it "kick off" everything that happens in the DME ECU! A motor with a bad sensor would most likely show the same symptoms you have. just my $.02

I'm aware of three suspicious sensors: speed sensor, reference sensor, and cylinder head temp sensor. They all seem to be in rather tricky locations. If the wiring leads back to the DME harness, perhaps I can test them without taking anything apart...I just don't yet know how.

I'm armed and dangerous with a power supply, a multimeter, and a test light...

A 3.6 has just one flywheel sensor. It's the 3.2 that has two. A 3.6 has quite a different
system with sequential injection versus batch injection of a 3.2. Don't know what the exact readings should be on the flywheel sensor, but you need to test it with a scope
of some sort to see the pulses. Perhaps there is a continuity test. Do some searches!

Here's the procedure to check the flywheel speed/ref sensor.
http://forums.pelicanparts.com/uploa...1425789024.jpg

Part of the fun is that my system is neither a 993 nor 911, so I have to do some adaptation of search results...

The price we pay for a bit o' electronically fuel-injected torque...

Do you have fuel ? Is the fuel pump making its usual noises ? (don't know what that sounds like on a 3.6 but you should !)

Just wondering if that smell came in through the heater duct-in which case melted pickup wires
would shut things down. Anytime there's that smell there is melted insulation-somewhere.

Here's the order of things that happen when starting a 3.6.
1. Ignition turns on the DME relay first section which turns on the DME ECU and applies
12V to the coils and to the fuel injectors. (you will measure 12v on each side
of the coil as well as the fuel injectors)

2. While cranking the motor (at least 30RPM) the flywheel sensor triggers the ECU to
energize the second relay in the DME relay to turn on the fuel pump. Also at this
time, having synced the reference and speed pulses to the ECU, the ECU begins
supplying the ground signals to the coils and injectors to activate them. Thus firing
off the motor.

While cranking your motor, have someone listen for the fuel pump to energize, that will tell you if your speed/ref sensor is working.

I've had issues before where it was a result of something I did recently.
You just replaced bushings in the rear of your car correct?
Check the routing of your harness between the DME and the engine. Something may have gotten bumped or burnt.
Just throwing it out there.

This is all helpful, thanks! So:

- The DME relay behaves properly outside of the car.
- The DME relay socket has good ground.
- The DME relay socket has 12v on pins 30 and 86 with the ignition off.
- There is 12v to the ignition coils.
- Test light flashes on the ignition coil inputs while cranking (barely).
- The fuel pump kicks on while cranking.
- No visible indication that the Motronic DME is fried.

but no spark.

Current theories:

- DME is toasted somewhere.
- Both ignition coils are toast.
- Some combination of problems (one dead coil, broken distributor belt).

I wish my test light flashed more prominently during cranking on those ignition coils...I'm not certain it's not just power fluctuations due to load. That said, according to the post from uwanna, my fuel pump wouldn't spin if the speed/reference sensors weren't good, sooo...

It sure seems like the DME is toast, but it's friggin' immaculate inside. I'm gonna find a way to test the ignition coils and put an eyeball on that distributor belt.

I realize there need be no visual indication for a circuit board or chip to be bad, but damn. Opened it back up, pulled up the little white pin insert to allow the board to pop out of the case, and got shots of both sids this time.

http://ensaster.com/pics/vehicles/sm...n/dme-clip.jpg

http://ensaster.com/pics/vehicles/sm...wn/dme-top.jpg

http://ensaster.com/pics/vehicles/sm...dme-bottom.jpg

http://ensaster.com/pics/vehicles/sm...n/dme-chip.jpg

In terms of recent work...In the last 100 miles I've done the rear suspension bushings, added front blinkers, a fan, controller, and relay, swapped out the temp gauge, installed a handbrake and heater controls...I supposed I could have jostled something, I just can't think of what. There's certainly nothing obvious from the outside.

To make sure your injectors are pulsing you need to attach a "noid" test light to one or more of the injector plugs. Was just at my local O'Reillys store this morning and they have a set of 6 NOID lights that you can borrow from their loan tools.

...Progress? A hot tip on Rennlist pointed me towards the existence of an external ignition module for this motor. If it wasn't bad before... I think it is now. I found this under my seat, resting on metal after some interior work. I guess it got detached, and since it was under the seat, I didn't notice.

Current (hopeful) hypothesis is that it was bouncing around, metal moved inside, and it shorted itself out...But I would think that the ignition coils wouldn't flicker my test light if this thing was the real problem, right?

One edge of the metal plate was detached, so I gave it a bit of a tug and it came apart. What a weird piece of electronics. It smells a bit burnt, but it also might just be the peanut butter smell of all the gunk this thing is coated in.

I guess I'm waiting on parts, now.

http://ensaster.com/pics/vehicles/sm...ion-module.jpg

Sure looks like a bad ignition module, obviously. But you should have two modules. one for each coil. If one is kaput, don't know why the second one wouldn't fire the engine. Perhaps the burning smell you noticed was from the broken module and
could have possibly damaged the ignition output driver circuit in the DME ECU.
Before installing new parts I would try to test the ignition output circuit using the
procedure below. If you don't have a scope, maybe you could use an LED test lite
to check for the pulses. Just a thought.

http://forums.pelicanparts.com/uploa...1314133733.jpg
http://forums.pelicanparts.com/uploa...1314133751.jpg

Well. It was the ignition module (I believe there was only one in 1995). Kind of. I bought a new cheapo ignition module, the car leapt to life, and I howled like a maniac!

Then about two miles down the road, it died suddenly again. I pulled the ignition module and it was hot. Like crazy hot. Almost burned my hand. I let it cool down, but the car won't catch. I assume this one is toast too, or was no good to begin with (though that seems unlikely based on the circumstance).

So. There is clearly a condition that is frying the ignition module. What could cause that??

Didn't realize a 993 only used one ign module. I have a 964 3.6 in my car with two modules and just assumed a
993 was the same. You do have two coils and twin plugs,
correct? Wonder why 993 uses just one ing module.
Inquiring minds want to know!

Two coils, two distributors, one ignition module...I'm not positive my setup is factory, but I can only find one, and the parts fiche only has one. This module has two chips inside, so I'm guessing they just combined the two circuits into one module. I wonder if the same chip fried this time around, too...