OntarioTurbo

Sounds like you've covered the basics. This may sound silly, but have you checked for spark during these hard start incidents? I believe the fuel pump on our cars only starts pumping after the crank sensors pick up. Perhaps high resistance sensor wiring when hot? High resistance fuel injector wiring when hot? As bearone said it could be a fuel leak. Lindsey Racing sells a nice kit for installing a fuel pressure gauge at the end of the rail. With that installed you should be able to see if pressure is bleeding off. If it is, then it could still be your injectors since you replaced everything else. The leak-by could be small enough that the fuel in the oil is burned off while driving and is unnoticable, but large enough to give you a hard start problem. After all, you said it starts after about 10 cranks, so it's not totally flooded, but only bleeds off whats in the rail.

User 52121

I've got a new fuel rail end cap on order so I can make an adapter for checking fuel pressure (per the tip from Clark's Garage.)

I could probably convince myself to just do the little Lindsey kit - but I can't tell from the pictures: is the gauge liquid filled?

Past experience with on-rail fuel pressure gauges has shown me that if it's not liquid filled, the gauge will fail (or lose accuracy) fairly quickly as the motor vibrations tend to beat the gauge internals to death.

User 52121

Have not checked for spark. I suppose it's easy enough to pull the plugs and fuel pump fuse and then watch for spark.

If it's a sensor failing due to a heat related issue... well... I suppose (I'm still new to these cars, I guess it wouldn't surprise me if this is just one of those odd car-specific problems that defies logic, been there, done that!) But really, I'd think that if heat were causing a sensor to act up, preventing the vehicle from starting, it would be at it's worst right after shutting the car down, and NOT after the car has sat for awhile. On a 90+ degree day, after driving the car for a few hours (so it's good and hot), I can shut it off, and if I restart it within 5-10 minutes, will fire right away. I'd think that the immediate restart would be *worse* than waiting 30 minutes to give the car a chance to cool a bit.

quinnfiske

Have you replaced the DME relay? It enables the fuel pump as well as the DME itself. Perhaps it fails to close one or both of the contacts when hot? It does not hurt to have a spare anyway.

lee101315

The DME temp sensor should have about 300ohms hot.

A leaking check valve on the fuel pump would cause the OPPOSITE problem.

Checking for a good spark is an excellent start, I had a similar problem. The next time this happens to you, try wiggling the reference/speed sensor connectors to see if it makes a difference.

Rob

I keep thinking I've read about somebody else having a similiar problem and it turned out to be the speed and reference sensors gap was too large. Maybe you could check for fuel injector pulses with a noid light when it does it.

Luis de Prat

Wiggling the wires works for me every single time.

I guess I need to check the gap on my new speed and reference sensors, as I also have an erratic non start situation that pops up now and then.

User 52121

Ok... so I haven't had much free time to tinker with it since I swapped the fuel system parts in.

Oddly enough - the problem, as of this past Sunday, seems to have cured itself.

I am happy, and worried, at the same time. My old Shelby Dodges used to do this all the time. :/ They'd randomly fix themselves, and then proceed to strand me somewhere later.

It was super humid here on Sunday, and I put about ~200 miles on the car. It also downpoured for a few hours during the drive (PCA Rallye.) I can't imagine what the weather would have to do with any of this... but since the humid Sunday, the warm start issue is gone.

FRZ

Hey dude, did you find the problem?? I read on many forums about this problem (That I have too..) and MANY and MANY of Porsche 944 Turbo or not have this problem.

Theres 2 potential solution i've not yet tested but if you test it before me.. go ahead:

- Some ppl talk about the pressure line go to vaccuum when warm. This can explain why short time after stop the engine the car normaly start "normal".

- Other guys think, and I think this can be a good idea to test, the Reference Crank Sensor which give to the CPU or DME not sure about the crank position..

Let me know!

User 52121

*scratches head*

Man this was a while ago (2009!) But it's good you bumped it, as I can finally close the loop here. I've been trying to get better at doing that as I've since been frustrated reading other threads here where the solution was never posted. Sorry about that. Long story short: new AFM solved it.

The problem came back once the humidity dropped. Shortly after all of this, the car developed an odd driveability problem when in extended stop-and-go traffic. The car would kind of stumble/sputter juuuuust off of closed throttle and seemed to worsen by the day (along with prolonged warm starts.)

Turned out that the door inside the AFM was sticking... you could see marks on the "floor" of the sensor where the door was dragging and if I pushed the door with my finger ever so gently, I could actually get the sensor to stick in that position. I tried sanding/smoothing it where it was rubbing but it didn't completely solve it (though it shouldn't rub in the first place.) Wound up buying a used AFM off of somebody here on RL and that solved both the driveability issue AND the warm start issue.

My theory is that the car, when warm and attempting to start, sucked just that perfect amount of air through the AFM to put the internal door in the "sticky" spot and cause either too much or not enough fueling for a proper start.

rdwalker

Hey, OmniGLH, thanks for updating the results.

I am having a similar issue with my 928S - will check out your pointers.
Old thread, but always useful.
Robert.

dspill

It has not happened since but the conditions also haven't presented themselves. It was always very hot out with a hot engine and a brief stop. My car has been running well at 106,000 miles.
What were your conditions?
Keep us posted.

spanky

I would also use contact cleaner to clean the DME base along with each fuse slot as these can and will build up non conductive coatings as the yrs go by which will give a false DME failure condition. Turns out the DME is fine, the base that the DME plugs into is dirty and nonconductive. Ask me how I know this. I now remove each relay and clean both the base and the relays plugs and each fuse twice a yr.

ealoken

Just remember, the fuelpump does not start untill you turn the key on ign starter. ( stage 3 )

i just rebuilt my system, so the fuelpump starts on ign stage 2.