My younger sister has had a 1987 944 for quite some time without problems. Today she couldn't start it. She has a Haynes manual (yech) but it was enough to get me to the fuel pump relay (in the nicely layed out relay/fuse section(I know didly about Porsches)).

I jumpered 30 to 87B, fuel pump came on.

I pulled relay out, pulled the cap off the relay and verifed both relay's work with 12v, contacts are good etc, testing it on the bench.

Put relay back in vehicle with cover off and physically held the relay closed. Started and ran fine.

Checked voltage across the fuel pump relay solenoid, no voltage.

Checked Haynes manual diagram. Shows one side of solenoid goes to ground, the other from the ECU.

My question. Is this a common problem? Why does it go back to the ECU? Is the thing "smart enough" to not provide the solenoid voltage if another sensor is not correct? (she is throwing no codes) Or is it likely to be a simple wiring/ground problem coming to the relay solenoid. Or a bad ECU.. perish the thought. Just looking for experience here...

Thanks for any comments.

 

Lots of things can keep the car from starting. Bad dme relay. Bad dme. Bad reference sensors and/or connections. Next time it does not start, see if the tach is bouncing at all.

I keep an extra DME relay in the car at all times. They are 25 bucks, and tend to fail often.

 

Sorry, thought I made it clear. Nothing wrong with the relay.

Thanks anyway.

 

double post

 

right, and you haven't solved the problem. This leads me to believe that you have a problem with one of the formentioned items. But throw a spare relay in the car anyway, trust me on this, lol. It will not start again in the future.

 

Solving the problem is not a problem, solving the probably without shotgunning it, is the problem.

I'll pass on the tip about keeping a spare relay. Just thought I could get a bit more specific help, as opposed to banal generalities.

I've got about an hour and a half invested, not counting playing here. Car is on MY front driveway... thanks sis.

For instance, you imply that the DME(ECU) can indeed "decide" not to provide the solenoid voltage by virtue of a faulty sensor input, or a sensor not giving a valid input the ECU likes. What sensors would that be, any or all? And of course, you ARE saying a DME working perfectly can indeed "decide" not to provide the correct voltage, because of some other input to it not being "right"? It would be much easier if the ECU logic/schematic were available.

Next thing I will do, is to check that the solenoid gets a ground at the coil. If it doesn't I'll find out where that ground fails. If it has a ground, I'll trace the B+ side back to the ECU, and verify the ECU is actually not providing the + voltage it should.

Just looking for constructive criticism, as you guys have apparently seen this before, based on all the like threads (which I read, thanks).

Can't work on the car until tomorrow morning, but I'll pursue this and any shortcuts I get before then.

Thanks again.

 

Don't know if this will help, but my rotor spun when the screw fell out, giving me a no run situation. It cranked but nothing. I did a lot of testing and almost assumed that the fuel pump was bad. Another thing I would check is to make sure the timimg belt did not loose any teeth or break. Recommended intervals I believe is roughly 30K miles. I also own a 1987 944 NA.
Just my .02

 

If the dme does not get a proper pulsed signal from the speed and reference sensors, it will not signal the fuel system to do anything. If the dme itself goes bad, then it will not send any signals. These cars have a tendency to have cold solder joints on the board, they might not work, and sometimes the slightest vibration will let the cold joint make contact enough for the board to work. Check your speed and reference sensors for proper function, then closely inspect the wires and connectors if they check out ok.

 

Right about the potential for cold solder joints, I can see that on the relay board. Soldered too fast with not enough heat, although they all checked out. But clearly if that is representative, the DME would also have that potential.

I will check the speed and reference sensors. Could they malfunction without throwing a code? Also, you may remember that when I manually closed the relay (the other closed on its own) for the Fuel Pump, the car starts easily and purrs like a kitten. Tach works, everthing works in that situation. If I then take my finger off the contact, the car runs correctly, still with no codes, until the fuel runs out... several minutes as expected. Not sure why no codes are thrown in that situation. Wondering about how sophistacted the codes are. Maybe they only cover the Emissions stuff on the 1987's, so its possible no codes for malfunctioning non-emission related components.

Thanks again

 

When you say throw a code, are you using a blink tester on the diagnostic port?

There is no check engine light in these cars.