salukijac

I found this thread..maybe it can help you. https://rennlist.com/forums/911-foru...less-help.html

Joe6pack

OK, still fooling with this thing. I drove the car to work today. Ran great, but the idle still sucks. Soooo, tonight I decided to do some more troubleshooting. I decided to clean the ground behind the fuel filter. At some point while working back there, I knocked the cylinder head temperature connector loose. Turns out, the connector is now broken. I don't know if I did it, or if it was already like that. The sensor pigtail is what actually broke (good news) and I only see one of the two male plugs. I don't know where the other one went. I really don't think I made any major contact with the switch to damage it, so I am hoping it was already damaged and that is the source of my idle woes.

So, with all that said, can a broken CHT sensor with only half a circuit cause the symptoms I described at the beginning of this thread? Would this not show up off idle? I am hoping to get a new sensor from one of the local dealers tomorrow.

Thanks for all the help on this.

Joe6pack

The plot thickens. I am hoping that someone can help me out with this. I ordered a new CHT sensor from Pelican this morning. However, I decided to take a closer look at the old pigtail this evening. From what I have read in my Bentley manual, I don't think this is the right sensor. According to the manual, the pigtail should have two male terminals. My pigtail only has one (see pic). It's not like one broke off or anything. It appears to have been made that way. The female computer side has two receptacles. Is this for a different car? BTW, my car is a 1987 Carrera.

Thanks,

John

Ed Hughes

It almost looks like one of the spades broke off. I'll be unhooking all of mine tomorrow for my drop, I can snap a pic of mine.

Joe6pack

No, it isn't broken off. It was never there. It does appear that the connector was molded to be able to accept two prongs, but there is no opening for the second one. If the DME measures resistance, there is no way for it to do that with an open circuit. Mine is an early '87 (11/86), so I don't know if that would have anything to do with it.

Thanks,

John

Joe6pack

10-4. I went to the Pelican sight and did a search and sure enough the system was one wire on early cars. I guess mine is the original sensor. Oh well, the connector is broken anyway. Thanks.

John

Lorenfb

"Here is the problem. When sitting at a light the car will idle what I would call normally at around 800 rpms for awhile. Then, suddenly, the rpms will drop to about 750, the oil pressure will drop and the car will noticeably shudder. It will stay like this for a few seconds and then the rpms may climb a little."

The problem you describe is called 'hunting'. It can be caused by a rich mixture or a feedback
problem in the O2 system. Since you indicated that the problem still exists with the O2 sensor
disconnected, it mostly likely is related to a mixture problem. This can be caused by a bad
temp sensor (Just jumper the temp connector with a paper clip to eliminate it once the engine
is warm.) or too high of fuel pressure. Also, if the idle switch isn't closing, that will also cause
'hunting'. If the AFM is incorrectly adjusted either by the idle mixture screw or the internal
spring, a rich mixture will occur causing 'hunting' at idle.

It's normal to have the 911 3.2 'hunt' at little (50 RPMs) at idle with the O2 connected. This usually
verifies a functioning O2 system. Even the later 964/993 engines exhibit a 25-50 RPM 'hunting'
at idle.

Note: It's a good idea to keep a paper clip in the glove box to bypass a bad temp
sensor on the road, as an open temp sensor will cause major problems for a warm
engine (911 3.2). This bypasses any potential grounding problems with the early
single wire temp sensor. The connector to the DME ECM always had two wires.

theiceman

now there is some usefull info ...

Ed Hughes

That is indeed excellent info, thank you for posting that Loren.

Lorenfb

"Wouldn't it be better to switch to the CHTII 2-wire sensor?"

Yes, that's correct.

"Also, wouldn't the DME compensate for a rich mixture if the O2 sensor was healthy? Once it goes into closed loop, you'd sort of think that an incorrectly adjusted afm, or even too much fuel pressure would be compensated for. I agree every 3.2 hunts, but 50 rpm is bordering on excessive hunting."

1. The O2 system is very limited in its ability to correct for mixture problems:
a. an incorrectly set AFM can make a huge mixture change,
b. an open temp sensor with a warm engine makes extreme 'hunting',
i.e. Just unplug the temp sensor with a warm engine and see the black smoke.,
3. Try disconnecting the vacuum line to the fuel pressure regulator (3 - 5 psi increase)
and see how the idle is affected. The fuel pressure is very critical for proper O2
system functionality.
4. That's one of the benefits of the '96 & later OBDII cars in that they have
a TRIM capability to correct when the O2 system is out of its range the result
of small variations, e.g. fuel pressure.

rusnak

Thanks Loren. I'll try disconnecting the CHT on a warm engine. I enjoy learning about the L-Jetronic system, since it is very basic and works well.

I did disconnect the vac line on the fuel pressure regulator. It went very rich, and then quickly the dme sensed this, took over, and it went back to .99 to 1.01 lambda. I was very intrigued. I re-connected the vac line, the opened the oil filler cap, and it went very lean, then compensated back to .99 to 1.01.

I agree that the afm makes a huge difference. I was trying to adjust mine rich for more power, and was dismayed to "see" on the LM-1 how the dme locked me out of this tweaking. I think the only way the dme doesn't take over is if the O2 sensor is disconnected.

I think mine hunts in the range of about 10-12 rpm once in closed loop at most. It's been dead steady since I changed all of the hoses and fixed all of the vacuum leaks.

As an aside, I found that there is a California only bridge to ground under the driver's seat that must be connected in order to easily pass smog. You can still pass without it connected, but it's closer to failing.

Joe6pack

Thanks all - Since the weather was so nice here this weekend, I played around with the car some more. My new 2-wire CHT sensor should be here on Wed. I measured the resistance at the old one per the Bentley manual. I got a reading that was in range, but it's hard to tell if it is correct since they only give you a range. I've got to change it anyway due to the broken connector.

I decided to go throught the FI part of the Bentley manual step by step checking each component. The fuel injectors seemed to check out per the vibration test. Checked the Air Flow Sensor and everything seemed to work and be in range including the temperature sensor. I also set the base idle per some instructions in another post. What's interesting, is that when I removed the jumper from the test port, the idle seemed to drop back down to around the 800 rpm range even though I had set it closer to 880. Maybe that's a clue.

That's as far as I got.

I did notice one other thing. sometimes, I will get a surge above 3,000 rpm. Like something kicks in. I don't think it is the WOT switch as I am not getting on it that hard. It only does this occassionally, but it has been doing it since I got the car.

Maybe I am being unrealistic with this idle thing. I guess I'm thinking I should be able to put a glass of Chardonnay on the intake manifold and not have it spill. It's just that the car is so smooth off idle that I think it should be smoother when idling.

The search continues.......

John

Lorenfb

"I agree that the afm makes a huge difference. I was trying to adjust mine rich for more power, and was dismayed to "see" on the LM-1 how the dme locked me out of this tweaking."

Actually, once you go beyond a certain point (richening) by reducing the
spring tension, the O2 system can no longer correct. Also, you can make
make the engine more responsive during acceleration by loosening the
spring and adjusting wiper.

dshepp806

Yes: a big thanks to you, Loren......

Good stuff (as always....)

Best,

Doyle

Joe6pack

Still working this thing. I had the battery and alternator checked at the local parts store this evening - both checked ok. My new theory is that it is indeed the idle control valve. This based on the fact that when I jumper the ports on the the test connector and set the base idle, everything seems fine. When I pull the jumper, the idle slowly but surely deteriorates. It is my understanding that jumpering the ports sets the icv to its middle "home" position. Is this correct? Also, does the icv correct for additional load from the alternator or the A/C similar to an idle control solenoid on an old carburetored car?

Anyway, $150+ is going to be an expensive experiment to find out if I am right.