86 911 refuses start below 65

[AMSRoadRunner]

My first attempt to remedy the problem worked for a few months, that is I moved from MD to TX. Car ran great all spring and summer. Now it is cold again and the problem didn't magically go away.

My car: Nearly completely stock 1986 911 (h4 head lights, momo steering wheel, i.e. not critical to the car enjoying cold weather as much as I do). 230k miles. No cat. Original exhaust.

Car runs great at temperatures above 65. It took a while to get her going after work today. Start it, a few back fires, it would die, start again. WOT would accomplish idle at 500 rpm.
took about 30 min and eventually it would idle and rev ok. didn't like over 4k (once oil was warmed) and it was rained the whole way home so I never got above 60.

First thought is intake gaskets? because they are exposed to the cold air, therefore they contract just enough...???
Any great way to test that?
Is there a computer tester for these cars? (that way I can get live data and see if the ________ misbehaves when cold...)

Last winter on one of those 20deg days when the car was thoroughly cold soaked I tested the crank position and speed sensors and they ohmed within parameters, same with temp sensor, IAC, MAF. I am sure I could use a new O2 but was contemplating a WBO2 with a gauge yet I have NFC on where to mount it and not look like some chopped up hoopde (sp?).

Thanks for any thoughts, suggestions and wishes for warmer weather!!

Andrew

[DanielDudley]

It was raining ? How old are distributor cap and plug wires ?

[cabmandone]

It seems to me like the cylinder head temp sensor would be the culprit but I don't know why it would run normal at above 65.

[AMSRoadRunner]

Originally Posted by DanielDudley

It was raining ? How old are distributor cap and plug wires ?

That was my initial suspicion as it acts like a dead miss. However, everything has been replaced or in the case of the distributor, completely rebuilt.

And yes, I too so badly wanted it to be the temp sensor but the sensor tests out ok and I don't really want to throw parts at this car (mostly I cannot afford it!)

I am sure the rain / humidity has nothing to do with it as it would run fine on those 100degree, 98% humidity days in MD and it ran fine in the deluge going to work (above 65)

Any other thoughts?

[cabmandone]

For intake leaks you can use a propane bottle with a rubber tube to direct the small flow of propane. If some gets in the intake it will make the rpm increase. I don't think intake leaks would show up as a problem only below a certain temp though.

When you do get it started, at W.O.T it will only hit 500 rpm?

[AMSRoadRunner]

I will try that propane trick.

Yes, while it is still cold sometimes I can get it to "idle" at 500 rpm, but only if I am WOT. Any less and it dies.

that is inconsistent though as it will die, on its own or me trying to pump the throttle. (I know it doesn't really have an accelerator pump)
then the next time I start it it will jump up to 2.5 or 3k for a few seconds then die...
then I can get it to do the 500 rpm idle...

[Quicksilver]

A vacuum leak in most of the intake ports will cause that exhaust port to heat up slowly when you startup. Check it with an IR thermometer.

Side subject -> I've never seen an explanation of this phenomenon on Pelican before, soooo...
Not all of the ports will show this behavior because of the way the injectors are timed. All of the injectors fire at once. The DME is not a sequential injection system. What this means is if the intake valve opens as the petrol is injected (or right before) that port will operate fine.
When the enjection event is during a different portion of the engine cycle the vacuum leak will allow the fuel charge to be drawn out of that intake runner where it will be consumed by another cylinder. So some ports will run lean and some will run rich. Over all it is rough and it can make the idle surge heavily.

------------------------------
Back to the posted issue.
I don't see a no start below 65° as an issue that can be caused by a gasket leak. It would run like crap above 65° so it would be perceived as a different issue.

Things I would look at would be:
- Testing the cylinder head temp sensor.
- Testing the flywheel sensors with an oscilloscope.
- Testing your DME. Put your DME in another car not the other way around!
- Testing the air temperature sensor in the AFM (barn door) (seems unlikely)
- Testing the fuel pressure retention for the injectors. (seems unlikely)
- Testing the idle stabilizer (IAC or ICV). It may ohm fine but be plugged. (seems unlikely)
- Test spark.
- Test fuel.

[cabmandone]

Originally Posted by AMSRoadRunner

I will try that propane trick.

Yes, while it is still cold sometimes I can get it to "idle" at 500 rpm, but only if I am WOT. Any less and it dies.

that is inconsistent though as it will die, on its own or me trying to pump the throttle. (I know it doesn't really have an accelerator pump)
then the next time I start it it will jump up to 2.5 or 3k for a few seconds then die...
then I can get it to do the 500 rpm idle...

I think a guy had a similar problem a few weeks ago. I think his ended up being the intake gasket now that I think of it. I'll go back and see if I can find the thread.

This isn't the thread but the problem sounds very similar. Check out post 9. He tells how to jumper the CHTS to see if it's the problem. I'm going to keep searching because I know I saw one where the owner had a problem with the intake.

Carrera 3.2 cold start problem

[cabmandone]

86 911 Hates the cold as much as I do!

Here's another thread with similar problem. I haven't found the one I'm thinking of but some point to the intake runner. It's worth a check. Also some good suggestions on things to check.

Whatever you find, don't forget to post what fixes it. Sorry I can't be more help.

I'm going to go out on a limb and say it's not your speed and reference sensor as they aren't temp sensitive. If they were failing you'd have the same problem hot or cold. IACV, CHTS are the ones people seem to point to the most.

If you can get the car in a warm garage and start it (door open obviously) you can pull the oil fill cap and see if the idle goes down. If you have vacuum leaks I think it won't change. If your vacuum system is tight it will idle down a bit IIRC.

[CCM911]

Mine was bad fuel lines, and bad intake manifold gaskets. But mine would at least idle(in between the backfires). It also would not be the worst idea to check the Idle COntrol Valve(ICV) to make sure that it is working well. You should feel it vibrate when the key is turned on.

[AMSRoadRunner]

Originally Posted by CCM911

You should feel it vibrate when turned on.

My car won't run.
My boss who has left me alone has suddenly decided I need micromanagement.

yet you have made my day!


In all seriousness, I suppose the motor is coming out to do the intake manifold gaskets?
Good time to adjust valves and _______?

Thank you all!!!!

Andrew

[cabmandone]

Originally Posted by AMSRoadRunner

My car won't run.
My boss who has left me alone has suddenly decided I need micromanagement.

yet you have made my day!


In all seriousness, I suppose the motor is coming out to do the intake manifold gaskets?
Good time to adjust valves and _______?

Thank you all!!!!

Andrew

I don't think the engine absolutely has to come out to do the intake. I was considering giving my top side a gasket refresh this winter and was considering doing it with the engine in. If you decide to pull the engine I'd definitely do the valves and check the head studs. If you have any oil leaks that would also be the time to address those. The "triangle" on the top of the engine is a big one guys talk about. I'd check your oil return tubes and if they look like they're seeping I'd put in new returns while it's out. It's MUCH easier to do them while the engine is out.

Good luck and let us know what you figure out.

[AMSRoadRunner]

I remember!

BTW, when I turned her it she was vibrating.

At some point I had determined that in fact it was the intake manifold gaskets. I got the gasket set, the valve cover gasket set and then I ____(got a job with a heated garage for parking and was focusing on building a Volvo 242 turbo for my cousin?)_______.

Now I am in Texas and left my 8 pallets of parts at the folks farm in MD, so this weekend looks like I am making gaskets...

[AMSRoadRunner]

New intake manifold gaskets. (all 12 of them)
and it would hit on all cylinders once then quit.

I figured with the insulation falling off the crank speed and position sensors it was time to replace them, too.
Just finished that up and....
nothing. No spark. Nada. Zilch

and my 60 year old tube oscilloscope finally died.

All sensors are plugged in, the connector behind (in front of) the engine is plugged in the ground on the #1 runner and behind (in front of) the fuel filter are good.

Without a scope to check signal from the position sensors I feel like I am kinda stuck. Suggestions?
*edit*
Have arduino, (actually Several) and in accordance with the majority of people being AWESOME I now have a 0-5v oscilloscope!
One voltage divider coming up, need to find a working laptop and we should be in good shape.

[AMSRoadRunner]

*update*
The beloved bently manual just says that the crank speed and position sensors need to be within .030" of the teeth. (I just assumed they would be unless you were running a different flywheel or something bad had happened...)

The Haynes manual is much more helpful in this regard:
apparently the bottom bolt for the bracket-that-holds-the-crank-speed-and-position-sensors (remember that is one word auf deutsch is slotted to act as a pivot. (sexy diagram and everything!) They recommend using the old sensor and gluing a washer that is .8mm on the lower sensor then adjusting bracket for the clearance. (if you don't have an old sensor just make sure you aren't using jb-weld to glue the sensor)

off to find a .030 washer and some super glue!

[AMSRoadRunner]

having removed the washer and super glue from my thumb I now have amazing spark. Good fuel pressure (even bypassed DME to remove from equation) and still no start. It will kick over, run like it should then cut, cough and die. So I am really no better off now than I was before.

boogers.

[AMSRoadRunner]

Isolated the problem to the connector at one of the crank sensors.

If I pull it out of the bracket and wiggle it a bit while the GF is cranking the motor I can get her to start. (the car that is)

I have tried a bit of di-electric grease on the contacts, no help
I have bent the contacts closed a bit so they "grab" the male contacts from the sensor a bit better, no help
I have pushed back the protective sleeve and examined and tried wiggling and tried inserting the wire a bit further into the house but apparently they are molded in place.

My thought at this point is to replace the plug. Ideally I would like something more like this:
http://www.summitracing.com/parts/smp-sk25?seid=srese1&gclid=CP768IG5pcICFZSHaQod7q8Afw
With the exposed wire that you push down. I have found they give a much more positive connection.

Any thought or suggestions?

Thanks!!

Andrew