Inconsistent starting behavior, '86 3.2
 

Hello all,

My '86 3.2, all stock, doesn't really have a problem but shows an inconsistency in starting that I'd like to address before it becomes an issue.

Sometimes it will fire up immediately with the first bump of the key and settle into a steady high idle (if cold) or regular idle (if warm).

Other times it will require two or three seconds of cranking before it catches, and will either settle into a normal idle (regardless of warm/cold) or hunt for a second (one or two surges to 1100 rpm) before settling into a normal warm idle.

I can't tell whether this is ignition or fuel related, though it does seem to run a bit rich for a few moments when it exhibits the longer cranking cycle. There's no rhyme or reason around ambient temp, cold or warm engine, length of time sitting, etc. Seems to be completely random.

Obviously it seems happiest when it fires immediately. No other driveability issues.

Crank position sensor going out of spec? Fuel pump or relay?

Thanks for any thoughts.

Tim

I have a similar issue on my '86 3.2 that I've been tracking down for the past year. I've been posting about it here: http://forums.pelicanparts.com/porsche-911-technical-forum/745746-hard-warm-start-rich-idle-3-2-a.html

I haven't posted an update in a while but I think I'm very close to tracking it down to the fuel evaporation system. I recently did a valve adjustment and tune-up and ran a couple bottles of techron thru a tank of gas to clean the injectors, and then I bought an LM-1 and nailed down the A/F ratio. These things really just got my engine running like a champ and narrowed down the possible causes. It seems now if my gas tank is near empty then the car cranks right up everytime as soon as I turn the key.

Sorry no definite answer for you <yet>, but maybe you can read everything I've done and get some ideas of what to look into...

I would guess fuel pump or fuel system problems. Just a guess.

The old vacuum leak issues can be a problem as well. I recently re-sealed my intake manifolds. All were nice and tight and not leaking EXCEPT one. The nuts were just finger tight on just one intake. Not a big vacuum leak but it was there. I suspect as the engine got hot things tightened up. Some of the plastic hoses and connections were rather sloppy as well.

It does run better now but I did a lot of other things while the engine was on the engine stand. The old "while I am in there" routine.

I think you are on the right track with the reference sensor. When were they last replaced? I replaced mine this spring on my '86 and mine looked awful. All the insulation was gone, etc. If you end up replacing one swap out the other while you are at it. Not a hard job but irritating enough to only go in once.

Bosch 0261210002 is the part number for an alternative to the Porsche part. Same part just a little longer wire.

Link with testing procedure and other good info below. -J

http://forums.pelicanparts.com/porsche-911-technical-forum/769810-reference-sensor.html

Hi Tim,

Are the times when it takes longer to start when it has been sitting for a week or more? My car longer to catch if it has been sitting for a bit. My guess is that the check valve gives up pressure after a decent bit of time. It seems to idle about the same either way. I did recently find that a vacuum hose end was rather hard and brittle so I cut back a little and now I think it is a better seal.

Franny

Thanks all---great info so far.

It started right away this morning, and I'm at a quarter tank. I'll have to see if there's a pattern there.

Reference sensors are original based on all the paperwork I have (back to '90). I may change them out just as a preventive measure.

Sitting for a while doesn't bother it. I've always been amazed that it'll start instantaneously after not running for a weeks or even months at a time when there's salt on the roads.

Your fuel system pressure is probably bleeding off. In the future, bump the key a few times without hitting start position. If it fires up consistently, check valve is bleeding off. If it's still inconsistent, there's other things to look at.

I'd start there though.

Following up--tried the "priming the pump" idea, no difference. Similarly, the car sat for five days and fired instantaneously, so I don't think fuel pressure is bleeding off. Also had the delayed-fire condition with slightly less than half a tank, so not sure fuel level is driving it.

I ordered two replacement speed/reference sensors, sounds like a good step anyway after 27 years even if it doesn't solve it.

Thanks.
Tim

Also, a faulty CylHeadTemp sensor can cause issues like this. Do you have the old 1 wire CHT sensor? If so replace it with a new style 2 wire sensor.

The older style had 1 wire for signal and got it's ground via the threads. The newer one has 2 wires, 1 for signal other for ground.

On my 3.2, the fuel pump doesn't run until the DME sees a few pulses of reference sensor (engine turning), ie it does NOT prime the pump in the ON position, even for a few seconds like many other cars of the same era do.

To properly test this, jumper the fuel pump on at the fuse block, sorry I don't remember which one it is, but they are labeled.

Chuck.H
'89 TurboLookTarga, 365k miles

On the 3.2 the fuel pump only runs when the starter is active (key in START, starter cranking) or if the flywheel is turning and speed sensor sees motion. There is no priming like on some other DME-controlled cars.

I would start with simple things:
- verify the idle microswitch on the throttle body engages properly
- make sure the check valve on the fuel supply is working (else hot-start issues due to vapor in fuel rail)
- Verify the CHT is the two-wire version, if not upgrade it

Then come the usual suspects: vacuum leaks on the intake, condition of plugs, cap, rotor, fuel filter, and good working ICV, good speed and reference sensors, AFM, DME relay, DME.

Good luck,
Ingo