Where to look next? Feedback/advice?

[spacecad3t]

If you are able to smoke test it, I still would do that. I can tune on the fly and my car will maintain the idle at 840 (or whatever it's set to) no matter how rich I run it.

Interesting...I must be wrong about that. I could have sworn I saw a high idle when I induced a rich condition by clamping the fuel return hose, but maybe that was an extreme measure. I might be just adding noise with the whole rich condition thing in that case.

For what its worth, while hooking up my vac gauge, I pulled the line to the FPR and it started to idle faster, which I think is suppose to happen because it cause excessive fuel pressure, i.e. rich?

[dr bob]

From 928 experience enhanced by exactly zero actual hands-on 944 work. With the 944 based on the 928 engine though, perhaps some of this will help you. Have a large grain of salt handy just in case.

My initial focus would be on the idle stabilizer. It's usually active only when the TPS shows no pedal .and. engine speed is below some threshold like 1200 RPM.

By now you've checked the throttle position switch functions at the controller end. Sometimes checking just at the switch doesn't catch accidental connection swaps or a broken wire, while checking function at the controller verifies the switch, the connections, and the wire in between.

On the 928, there are two different versions of the idle stabilizer. I'll speculate that the same is true of the 944. One version gets a proportional analog signal, IIRC the cars with AFM. MAF cars get a stabilizer that receives a PWM signal from the controllers. I'm not sure which you would use with your conversion.

A way too common problem on the 928 is the deterioration of hoses under the intake. That includes crankcase ventilation from the cam cover vents to the AOS, the vent from the AOS to the intake, and the hoses that connect the idle stabilizer above and below the throttle. All are subject to oil contact and contamination, and eventually soften and split. 944 have the same? Carefully inspect those hoses. A small leak in any will allow false air into the intake, all but one below the throttle. I've worked on clinic cars where the owner has replaced the MAF, had the throttle and MAF out of the car multiple times, and only later found that a split hose was the culprit. A split hose may not show up with a smoke test unless you add enough intake pressure to open an axial split. Physical/manual inspection is often necessary. If any of the hoses are soft and especially if they are a little gooey, plan on replacing them all. Beyond the high idle issue, the car will never run just right with damaged hoses.

If you've had the throttle apart for maintenance (new bearings, seals, etc.) verify that the shaft is actually sealing at the ends, and that the throttle plate is installed so it's centered in the bore. A bright light in one side in a dark room is a good way to look for spaces around the plate. The stop screw should hold the plate just at that closed point in its movement.

If you had the throttle plate out, the new screws get staked carefully so they won't come loose. It's critical that the shaft is supported completely while doing that. Even a slight bend risks air leaks, but through the bore and via the bearings and seals on the shaft. Ditto if you've been prying on shaft fittings. I've tested for leaking shaft bearings and seals with a mitivac after all the shaft fittings are removed. Leaks there are unmetered so more critical than air passing the closed throttle plate.

HTH!

[johnb]

If you are able to smoke test it, I still would do that. I can tune on the fly and my car will maintain the idle at 840 (or whatever it's set to) no matter how rich I run it.

Interesting...I must be wrong about that. I could have sworn I saw a high idle when I induced a rich condition by clamping the fuel return hose, but maybe that was an extreme measure. I might be just adding noise with the whole rich condition thing in that case.

For what its worth, while hooking up my vac gauge, I pulled the line to the FPR and it started to idle faster, which I think is suppose to happen because it cause excessive fuel pressure, i.e. rich?

Well that's where it gets tricky - if you opened the line, then you bumped fuel pressure but also created a vacuum leak, so it's hard to say which one affected the idle.

I just did a quick test where I disconnected the FPR vacuum line. The vac leak made the idle sound a little rough but didn't affect rpm that I noticed. I didn't see much of a change in AFR either. When I plugged the leaking vac line it seemed normal (that's why I say it was the vac leak causing the roughness and not the AFR).

Then I reattached that vac line and tried clamping the fuel return hose. That bumped my AFR up into the 12s and raised my idle rpm to around 1000rpm. So it's definitely possible to get a high idle with a rich condition. That said, to get it up to what you're seeing, I don't know. It would have to be pretty rich. I don't know if it's even possible - if it gets rich enough I imagine it'll lose power and come back down again.

Also are you sure the ISV is installed in the right orientation? I honestly don't know what difference it makes but it has an arrow on it. Maybe it only seals properly against air in one direction?

[spacecad3t]

Well that's where it gets tricky - if you opened the line, then you bumped fuel pressure but also created a vacuum leak, so it's hard to say which one affected the idle.

I just did a quick test where I disconnected the FPR vacuum line. The vac leak made the idle sound a little rough but didn't affect rpm that I noticed. I didn't see much of a change in AFR either. When I plugged the leaking vac line it seemed normal (that's why I say it was the vac leak causing the roughness and not the AFR).

Then I reattached that vac line and tried clamping the fuel return hose. That bumped my AFR up into the 12s and raised my idle rpm to around 1000rpm. So it's definitely possible to get a high idle with a rich condition. That said, to get it up to what you're seeing, I don't know. It would have to be pretty rich. I don't know if it's even possible - if it gets rich enough I imagine it'll lose power and come back down again.

Also are you sure the ISV is installed in the right orientation? I honestly don't know what difference it makes but it has an arrow on it. Maybe it only seals properly against air in one direction?

Sorry, I was unclear. I have a T in my vac gauge so I can run it inline without creating a leak. I just noticed the change while making the switch, but I see what your saying- I just saw the momentary leak increasing RPM. I thought I remembered from another car unplugging it would cause the FPR to go full rich, but I could be mistaken.

I guess this is why I'm also at a loss, the car doesn't loose power, hell it doesn't even have that "off idle stumble" or hesitation people mention. It has strong acceleration and power all the way to 6000rpm. No backfiring, or anything. This entire thread is because I hate idling at 1200-1400rpm, its hard on the clutch, shut down, wastes gas, etc.

I triple checked the ISV during install, arrow is toward manifold.

[johnb]

Do you have a stock chip BTW? I know you said you had the FTech MAF but apparently it can be used with stock and aftermarket chips.

[spacecad3t]

Do you have a stock chip BTW? I know you said you had the FTech MAF but apparently it can be used with stock and aftermarket chips.

I have the FTech Sport DME that I ordered with the MAF.

[johnb]

Do you have a stock chip BTW? I know you said you had the FTech MAF but apparently it can be used with stock and aftermarket chips.

I have the FTech Sport DME that I ordered with the MAF.

Oh I assumed you still had to supply a chip. Does it come with one installed?

[Tom]

Did it idle fast before the MAF?

[spacecad3t]

Do you have a stock chip BTW? I know you said you had the FTech MAF but apparently it can be used with stock and aftermarket chips.

I have the FTech Sport DME that I ordered with the MAF.

Oh I assumed you still had to supply a chip. Does it come with one installed?

I think they reuse existing chips. I cant speak to it, from what I read, it isnt much different from stock.

The high idle was present before. I had hoped switching from the old/monkied with AFM was going to solve it.

[944er]

Here is something to think about. You absolutely CANNOT have a too high idle without enough air to mix with fuel, in order to create the horsepower necessary to raise the idle. This means you either have a vacuum leak, your ISV is open too much of the time or does not close completely, or your throttle plate isn't closed enough, or some combination thereof. If making the mixture more rich increases the idle even more, that just means that the too much air you already had was running lean, and the additional fuel increased the horsepower even more. The DME computer tries real hard to put enough fuel in to match the airflow you have, in order to keep the engine running.

The first step in solving this is to find the source of too much air entering the intake manifold. You squeezed off the ISV intake line, and the RPM's dropped to normal. I don't know if there is a minimum ISV duty cycle, but if there is then intake leakage and/or throttle idle bypass is overwhelming it. If there is no minimum duty cycle (i.e. it can close and stay closed,) then the ISV valve is not working correctly, AND you have too much leakage and/or throttle plate is too open as I would believe Porsche/Bosch would expect at least some activity of the ISV at idle in any case, resulting in a too low idle during your test.

Some of the other comments would lead me to believe the ISV just does not close completely. Tom pointed you to his video of testing the ISV, and you should do this test. Another possibility is too high of a resistance in the close wire path to the DME that causes it to move slowly in the close direction, resulting in it not completely snapping shut before the open signal is given to it. This would not be apparent in a bench test of the ISV.


Fair warning: I have an early car, so my knowledge of late cars is highly influenced by internet disinformation.