FRporscheman

Here's a LINK to the first thread I started on this problem. I never really figured it out, and now I'm borrowing the car from my friend again to try to finally solve it.

The car is a 968 tiptronic with a supercharger. (Yeah, I can't think of a bigger headache of a pcar to diagnose.)

I'll just describe the current symptoms first:
When the engine is cold (about 22C), it starts up ok. Idle settles down to about 800-850rpm pretty quickly, but it should be 880. (On the 968, the idle is not adjustable - DME hard-coded to 880rpm).
The colder the outside temp, the rougher it idles, like 700rpm on a winter night.
While idling (and still cold), if i blip the throttle the revs will go up and then dip down below idle before coming back to idle. In the winter, it will actually dip so low that it stalls.
Driving around on a cold engine, when I come to a stop, the engine wants to stall. The quicker I stop (and the colder the weather) the more it wants to (and often does) stall.
When I try to accelerate from a stop, it has power for like a split second, then loses power severely, then after a couple of seconds the power surges back and it accelerates hard because by now I'm flooring it, and the damned thing is supercharged.
As the car warms up, the problem diminishes. At normal operating temp, the car behaves 100% OK. Idle is at 880rpm, revs never dip below idle, and the car has consistent power all of the time.

What I have done:
My friend has durametric so I hooked it up. The TPS, DME temp, MAF, O2, rpm, all sensors giving good signals. I watched injector duty cycle slowly decrease as temp rose.
Put in a new DME temp sensor (the NTC-II sensor). Even confirmed its readings against the WSM, at the DME plug, so I know the wiring is good too.
Also tried unplugging the sensor to make the DME use the default fuel map of a warm engine, just made it a little worse when running cold.
Also tried bypassing the sensor with resistors to make the mixture a little more rich or lean. Only made it worse.
New O2 sensor. Also tried driving around with it unplugged, no difference.
New spark plugs - NGK BPR6EY (copper, gapped correctly, factory heat range). I figured an engine with a blower might need a colder heat range, but at idle and driving at low rpms there should be little/no boost. No improvement. BTW, the plugs that came out were Bosch WR7DC+ which I believe is also the factory heat range.
Checked ISV and confirmed that it is functioning correctly (explained below).
Did a smoke test on the intake, didn't see any smoke. I might do this test again just to be sure.
Sprayed propane all around the intake while it idled, never heard the idle go up.
Swapped MAFs, no change.
Voltage doesn't really change from cold to warm - stays around 13V on the dash gauge. My friend says it has a new voltage regulator but sometimes the alternator makes a whirring sound like a bad bearing.

Some other things to note:
It has ford fuel injectors, F87E-D2B. Also an adjustable FPR which I know nothing about. I really should check the fuel pressure.

I'll post more if I think of anything. I'm totally stumped!

FRporscheman

Totally cold engine.





Warming up.





Totally warm engine.





Warm engine, in drive but stopped, AC on, lights and high beams on, steering wheel turned. (All loads possible.)




If if anyone wants to see, this is how I tested the ISV and figured out that it was working ok:
https://rennlist.com/forums/944-turb...l#post14106904

Jfrahm

I agree on the fuel pressure, those look like 14.5 lb/hr injectors? I run 32 lb/hr Volvo injectors in my black 968, naturally aspirated. These are for a Ford/Mazda pickup so not expected to deliver what a 968 needs at typical fuel pressures. Maybe they have been modded (decapped) or something.

The stalling sounds like what my blown 968 used to do before I relocated the MAF, too much turbulence with the recirulation valve return within inches of the MAF. I moved the MAF down by the air filter.

It sounds like it's set up to run really fat when cold and cleans up in closed-loop however that injector part number would be hard to get to run fat at idle and even harder to run safely under boost (with an RRFPR?).
-Joel.

FRporscheman

I checked the fuel pressure, and it sits at about 51psi steadily from cold to warm. Revving the engine only raises it to 54-55psi.

Interesting point about the MAF turbulence. This SC kit has the maf right at the filter, but I'll have to double check the recirc valve plumbing. Maybe I'll try having it dump to atmosphere and see if it improves at all.

FRporscheman

Could it be a problem with the ignition timing? If so, how would I troubleshoot that?

Jfrahm

can you monitor it, and knock, in Durametric? This:
"When I try to accelerate from a stop, it has power for like a split second, then loses power severely, then after a couple of seconds the power surges back and it accelerates hard because by now I'm flooring it"
sounds like it could be caused by knock and timing pullout.

FRporscheman

I'll check the durametric options next time I'm tinkering with the car.

So why would the engine be knocking when it's cold?

And would unplugging the knock sensors let me get around this (for testing purposes) or will it force the computer into some emergency mode with retarded timing?

Jfrahm

I do not know why it would knock when cold, maybe cap and rotor problems, wiring issue, crank trigger issue, etc. It really sounds like my blower car on a hot day when the blower kicks in, having been bypassing, and all that air hits, maybe when the car is not in a WOT map.

It could also be a bad knock sensor falsing, I had that in an Audi. Killed power and MPG. Unplugging it did improve things, probably put it into a conservative spark map. It was a pre-OBD-anything car. The 968 though might throw a code if it distrusts the knock sensor, it will if the sensor is disconnected but it'd still be a good test.

FRporscheman

I unplugged both knock sensors and drove it. The problem was exactly the same, except i now had a check engine light. Plugging the sensors back in made the CEL go away.

Shouldn't that have caused 6° of retard? It felt more or less the same. I tried unplugging the hall sensor on my own 968 to see what it feels like with 6° retard and it's just sluggish, not the bad hesitation like this s/c tip car.

According to the wsm, timing at operating temperature should be 10° +/- 3°, and that's what the car has. Maybe i should check that while driving...

I drove it the other night and it behaved 100% fine for the first 20 seconds, then had the rough idle and hesitation again. Makes me think theres a loose connection somewhere, or some kind of electrical gremlin.

Could it be the coil?

Also, the fuel tank just hit empty, and i got it with 3/4 tank and ive put like 100 miles on it, so it's guzzling gas. Could it be running way too rich?

So confused by this one.