1989 3.2 Carrera--problems restarting after first start of the day

[89PS911]

Hi All,

Diagnostic:

1989 911 3.2 with 77,500 miles. Has always run perfectly. Motor burns a quart of oil about every 1,200 miles. Does not smoke on start-up or at other times.

Engine was out for a new clutch in March (75,000 miles). Also, all fuel lines in engine compartment replaced at that time. Ran well as always until this start-up problem developed (see below).

No smell of gas in the engine compartment or the front trunk with engine off or running.

New Bosch plugs, ignition wires (Magnecor black), Bosch distributor cap at 72,000 miles.

Have checked that all plugs and wires are seated properly.

Car starts right up in the morning with one push on the gas pedal, releasing gas pedal, and turning key. Idles at a steady 900 or so RPM.

For the first time last Saturday, after I had been driving the car during most of the day, I went to start it and it would not start up. Cranked with full power; but did not start up. I checked all plug wires. After a couple of minutes, started up again with no problem.

Situation now: Starts right up on first try; on second try, starts but seems starved of fuel, stumbles, and if I don't push the gas pedal it dies after 15 seconds. If I do push the pedal, it goes up to its usual steady idle. Occasionally does not start just after I turn motor off, but will do so after one or two more tries.

Any thoughts?

Thank you,
Jonathan

[DRACO A5OG]

Originally Posted by 89PS911

Car starts right up in the morning with one push on the gas pedal, releasing gas pedal, and turning key. Idles at a steady 900 or so RPM.

I am curious, why are you taking this step before you start? Our 3.2's does not require priming before start, in fact our fuel pump does not kick in until engine starts.

So, let's do some diagnostics:

You wrote you did not smell any fuel, but did you check the tail pipe? If so then it is ignition related. If no fuel smell it is fuel related.

Check voltage before and after start, please post readings.

Has your fuel filter been changed and when?

Jim

[89PS911]

Hi Jim,

Thank you for writing.

Indeed, pushing the pedal is an old habit... Did not push it today and car started right up.

Voltage measurements at battery terminals:

12.4 v car off

14.1 v after starting

Bentley indicates this is normal and would not suggest a voltage regulator problem.

No smell of fuel at exhaust pipe when car is running or just after turning off.

Fuel filter was changed at 61,300 and at 65,500. PO owned the car for 9 years and those filter changes were done at major services on either end of that tenure. I am the third owner and have owned the car since 7/2011. Car now has 77,500 miles.

More information:

On first start, with outside temp of 59 F, engine idled at about 1050 rpm for 1 minute, then dropped to just shy of 900 rpm for 23 seconds before dropping further and idling at a steady 750 rpm (too low relative to the 880 rpm Bentley states) until I cut it off at 3'23".

Immediately restarted car, it idled briefly at 750 rpm before gradually stumbling down the rpm scale and cutting off.

Waited 1 minute and restarted, engine settled into idle at 750 rpm, I revved it to about 1200 rpm then came off the pedal, engine dropped first to around 900 rpm then again back to 750 rpm and idled there without cutting off.

Cut off and restarted immediately several more times. Each time it settled in at 750 rpm and held there.

Jonathan

[steely]

Jumping in here - the idle is indeed low. Out of curiosity was any of this related to any colder temperatures lately?

I would suggest you look for some vacuum leaks as that could cause a low idle (sort of suddenly/recently).

Does she stall when you open the oil cap when at this low idle? Normally, it would drop like that from 880 to lower if all was Aok. If it doesn't drop, that might indicate you have an air/vacuum leak already).

Also consider cleaning your idle control valve (ICV).

A slight idle adjustment upward might be warranted, but I'd suggest only after you eliminate some of the other potential issues (as above and yet to be suggested) - so as to not cover up any root cause.

This is just my 2 cents. See what others suggest.

I had a similar issue a few years back and it was related to old vac lines and a bad (rare) ICV drive in the DME.

[89PS911]

Hi Dan,

Thanks for sharing your thoughts.

I just pulled the oil cap while the car was idling at 750 rpm and no discernible change in the idle by sound or on the tach. Also, if I put my hand briefly over the air intake, there is strong suction/pressure and the engine starts to cut off almost instantaneously, which, from my layman's perspective, indicates the vacuum system is in good shape.

Please let me know your further thoughts when you can.

Best,
Jonathan

[89PS911]

Dan,

Forgot to mention that this problem did coincide with a change in temperature, but nothing major--from average in the high 70s to average in the low 60s.

Also, today the car restarted every time and settled in at a steady idle of 750 rpm. No second start today during which it stumbled down the rpm scale to a stall.

Jonathan

[steely]

Hi Jonathan,
I'm actually not sure how good a test that is of the vacuum system.
With the oil cap back on, and the engine running, could you test the vacuum at the trunk?
Pull the brake booster line off of the nipple on the trunk floor and see if you can feel for any suction there (the nipple is a hard line routed thru the tunnel to the vacuum system in the engine bay).

This may be a subjective test if you've never felt the vacuum there - but if there is no suction, I would try to check for any obvious leaks / cracked vac' lines, loose intake nuts. Some recommend spraying some flammable material (brake cleaner)around the bay to see if it gets into the intake (rpm increase), but that can be a bit dicey.

The theory I am working with here - and I could be wrong - is that you have an air leak, but as the car warms and adjusts the ICV position, the excess air makes for a lean mixture. Your rpm suffers, and then it may make a hot restart more difficult.

Another thing to check would be the ICV. There are some cleaning tips here in the forum that you might pursue, most I think involve swishing some alcohol inside, but check me out on this one. If the ICV sticks, this could affect its operation and your idle. Most of my early problems with the idle only became apparent in the fall - and this led me to the ICV.

That aside, you should check the idle position switch on the drivers side of the throttle body.
It's a micro switch that you should hear click when you move the throttle by hand (with the engine off). The switch should be closed when in idle.

no expert here, but I will try to help - I'll be off line tomorrow, so PM me if no one else can chime in and you don't hear back.

[r-mm]

I would bet it is a vacuum problem that was exacerbated by the motor going in and out. This line was cracked on mine: Repair vs Replacement options for this part?

It is not easy to see the crack, but some silicone "rescue tape" make a quick fix. I'm not positive how you'd go about diagnosing a vacuum leak, but I agree that seeing no idle drop whatsoever when the oil cap is opened does not seem entirely normal.

[steely]

good point r-mm - I had a crack in the rubber to that pipeline too. I slipped a small diameter vac' line into a larger diameter vac line. The crack was the same as yours, only evident if you gently bent it back. I had the intake out when I did the fuel lines and went over all of the vac lines.
pic below - part marked 'fixed' - hopefully right-side up.
@Jonathan - the idle position micro switch is in the pic center / black rectangle with red dot.

[89PS911]

Thanks r-mm and Dan.

I will check the suction at the brake booster line in the front trunk and that rubber part of the vacuum pipeline tonight or tomorrow.

Dan,

On the idle position micro switch, when the car is off, is the throttle plate supposed to be depressing the "clicky button" protruding out of the top of the switch housing (mine is), or is there supposed to be a very small gap (gap "D") as shown in the Bentley manual in the chapter on the fuel injection system where this switch is discussed (I am not at home, so I don't remember the page number)?

Jonathan

[Adrian M]

Check your mixture.. Sounds as if running lean and having related issues, ie "hunting" for 880 RPM at start up and quitting etc..It really should be at 1100 - 1200 upon start and settle at 880. Have a qualified mechanic run it through the check with a CO meter and tune accordingly. DON'T do it yourself if you can avoid. This can lead to bigger issues going forward. Could be as easy as the Mass air flow sensor or the Oxy sensor. There are multiple sensors on the car that talk to your Motronic to make it tune your car optimally. It only takes one to fail and the gremlins get whacky. That said, running at 750RPM both with the oil cap on and off is suspicious and a vacuum issue may be at hand. That said, if I read your posts correctly this is new and was not a factor initially.. Hmm.. Frustrating for sure! Good luck!!!

[89PS911]

Thanks Adrian. At start-up it goes to 1100-1200, then settles at 880, for minute, and then drops to 750. I will check that pipeline that the other mention and see.

By the way, yesterday I restarted many times and it never stumbled to a stall. Just settled at 750, oil cap on or off...

Question: Is it harmful for me to drive the car with this condition (low idle, possible vacuum leak, possible sensor problem feeding into Motronic)?

[Adrian M]

Well, that's a good question.. I'd say the biggest risk is being stranded in an inconvenient place other than home! Certainly running lean is not good for engine longevity..

[steely]

Originally Posted by 89PS911

Dan,

On the idle position micro switch, when the car is off, is the throttle plate supposed to be depressing the "clicky button" protruding out of the top of the switch housing (mine is), or is there supposed to be a very small gap (gap "D") as shown in the Bentley manual in the chapter on the fuel injection system where this switch is discussed (I am not at home, so I don't remember the page number)?

Jonathan

that is page 240-21. Not sure I understand the intent of the "D" dimension unless it is to take up slack in the linkage. if you adjust "E" - it looks like the sw won't be depressed. First para seems to explain intent better than figure note. The sw should be closed (depressed) when off/and or in idle.
I have a hunch this didn't suddenly change, but worth a look. I was thinking maybe it could be grossly off (0.2mm isn't easy to measure there in situ), or the internal sw contacts were faulty. Bentley shows you how to check the continuity if you have it.

[89PS911]

Hi Jim, Adrian, Dan, and r-mm,

Prior to doing anything else I inspected all the vacuum hoses carefully with a light and mirror, looked carefully at the rubber end on the pipeline going to the AFM, found plenty of suction at the brake booster nipple in the front trunk, checked voltage before and after start. All hoses/rubber look and feel new (not hard, no dryness, no cracks) and voltage, as noted above, seems to be spot on. So, I then decided to start replacing things. Today I put a new fuel filter in the car and it is running well. No restart problems, idle is at 850-875 according to the tach, power feels right across the power band.

Jim asked about the fuel filter in his post. It had about 12,000 miles on it over a period of 3 years. I drained the filter onto a mound of paper towels, flipping it back and forth end to end. I tapped it as well as I drained it and some black flecks, very tiny, came out onto the paper. Fuel itself looked completely clear as it drained out.

I have a new O2 sensor on the way from our host and will install it as well. The PO owned the car from 2002-2011 and there is no record in the service records he gave me of that sensor ever having been replaced so it would seem a good idea to replace it in any case.

After I replace the O2 sensor, I'll have my mechanic test things with a CO meter.

I'd appreciate any further thoughts you may have.

Thanks again,
Jonathan

[Adrian M]

Looks like Jim nailed it! Great news! The other steps are good to do anyway so you know what has and hasn't been done..

Post your results!

Cheers,

Adrian.

[steely]

Congrats Jonathan - good on ya'.

[89PS911]

Hi All,

An update, with a sign of some trouble.

After driving with no incidents and what has seemed like normal power to me for a total of 330 miles since I changed my fuel filter on 10/18, and my car passing Maryland's emissions inspection with flying colors a week ago, my car stalled today as I was driving. It just lost power quite quickly--felt like it was starved of fuel--and stopped running. Fortunately, there was no one behind me late on a Sunday afternoon and I coasted off into a parking lot.

As was the case several weeks ago, my car started back up after a few minutes sitting and I drove home with what felt like a normal power band.

One other piece of information, I put one container of Techron fuel system cleaner in my near empty gas tank yesterday and then filled up my tank. This is the second time I have used Techron--first time was about 3,000 miles ago early this year.

After this experience today, I decided to install my new O2 sensor this evening. That went fine and it is in.

Car started up fine after the sensor went in and I took it for a 14 mile drive about an hour ago. No incidents while driving. However, when I arrived home, turned my car off, and then tried to restart, my car started but then stumbled and cut off. After a few minutes, it started back up as usual and idled normally. About 30 minutes later I restarted it several time with no problem.

Any theories on my problem and what do you recommend as next steps?

Thank you.
Jonathan

[scarceller]

The 89 3.2L idles at 880RPMs, it's baked into the chip. If it's idling at 700RPM something is wrong, I'd start with these tests.

- Check fuel pressure
- Set Base mixture and base idle, I'd set mixture at 14.2AFR and idle at 900RPMs.

A stall and no start after sitting could be a lot of things but Here's some to think about:
- Test CHT sensor, at room temp should read 2000 - 3000 ohms
- How old are the ref and speed sensors? These have been known to cause issues like this.
- DME could have developed solder cracks, can you test this DME in another car? I'm seeing more and more DMEs fail these days from years of road vibration issues.
- Test fuel Pressure and particularly the ability for the system to hold pressure for 20 minutes after you shutdown the engine.

When the engine stalls does the tach act up? any wild tach bounce?

I also recommend that you leave the O2 sensor disconnected till you find the source of the problem. These engines run just fine without the O2 and having it connected can mask the underlying issue.

Do you know what your mixture is currently set at? You can use a WBO2 sensor/controller to check it, at idle it should be set at about 14.2AFR.

[Gregt909]

May want to check the filter again. Maybe the Techron loosened some gunk in the tank or lines and plugged the filter again.