I posted this on the 924/944/951/968 forum as well. Any suggestions would be greatly appreciated...

About a week ago I was driving along and my car just died...no sputtering or anything. While I coasted (still in gear) to the side of the road, I noticed that my tachometer was reading the engine speed accurately. If I pushed the clutch in, the engine revs and tach would drop to zero. If I released the clutch, then engine and tach agreed as the engine sped up again (not under its own power, of course).

After a few tries, the car started back up. This same scenario has played out about 5 times in the past week.

I have replaced my cap, rotor, DME/Fuel pump relay, and FPR. Recently checked the TPS (okay). I could use a new set of plug wires, but I don't see how that could cause this problem. (I was seeing arcing in total darkness, but replaced the cracked second layer of insulation on the wires with silicon tubing at key areas. This eliminated all of the arcing.)

This weekend I am installing a fuel pressure gauge, changing the fuel filter, checking out the resistance on the coil windings, looking for bad grounds, and reading any blink codes.

At first the problem only seemed to appear after hard driving, but I've driven the **** out of it lately and do not see a correlation. This morning when it died I swapped an ignition tester for one of the plugs, but the car cranked right away, nullifying the test.

Any ideas what this could be? What do coils behave like when they go out? Do they just stop working altogether, or could they cause an intermittant problem? I've heard of bad solder joints in the DME being problematic. Where EXACTLY should I be looking to identify this problem? Any other suggestions?

 

You could do what I've done, and install a fuel pressure gauge (Autometer sells a nice electronic fp gauge) inside your car. Then you can see if you're fp at the rail is falling off prior to your car dying. At least then you could figure out if your problem is fuel or ignition related. If it ends up being fuel related, perhaps your fuel pump is dying, or you've got a bad connection. Or maybe it's your DME, I remember hearing posts about DME circuit board solders coming undone and causing strange stalling problems.

If it's ignition related, maybe it's your spark sensor. I think it's bolted on to the back of the cam cover. Or perhaps one of your speed reference sensors is acting up. I don't know...just throwing out ideas.

 

I can't justify the dinero for an electronic fp gauge, but I just bought a mechanical unit for my fuel rail. When the car dies, I'll take a look at the residual pressure. I'll also try again with the ignition tester; maybe I can catch it and turn the engine over while the problem persists. Hope it's not the fuel pump, but not a bad idea to check that connection.

 

JLeake:

Sounds like the DME is loosing a signal. The car will still run with a bad TPS or idle control valve, but will not with bad speed/ref sensors, temp sensor, injectors, coil, grounds or lack of fuel, and something else I have forgotten to mention.

Check the connectors for the speed and reference sensors where they connect to the DME harness at the rear of the intake manifold. Pry the metal clip partially off, and wiggle the connectors loose. Look for broken pieces of the connector inside the female portion. This will cause intermittent failures. Other possible ideas have been expressed by others. While you are at the back of the manifold, check the two grounds for the DME harness to the back of the block and clutch housing.

 

I had this same issue with mine not too long ago. What seemed to fix it for me (for now anyway) was replacing the DME. I havn't had any other running issues except for it being VERY lean all the way up to 5K RPM (fingers are still crossed).

Hope this helps...

Dan

 

Check the dme for cold joints on the solder connections. I found a cold joint on my transistor that sends power to the ignition coil after i soldered it problem fixed.Same symptoms as yours.Look for the largest transistor. It has 3 terminals. You could just go over all of the joints that look questionable with a soldering iron and some solder. good luck.

 

Fixed it. The three solder joints on the large transistor in the DME were cracked, as jimbo1111 described. Thanks all for your suggestions along the way.