79SC dying when rolling up to stop clutch in

[Jameel]

I'm back to struggling with this 79 3.0.

Here's what happened last time around (last post is the pertinent one)

http://forums.pelicanparts.com/porsche-911-technical-forum/991003-cis-experts-help-evaluate-couple-numbers.html#post10072546

And here's the changes since it was running well.

New M1 cams from William Knight
Rebuilt the CIS (new seals and rubber, checked every component, brand new airbox)
Ran all the pressure tests ala Dummies thread. All checked out.

Engine runs great, when actually driving but here's the problems.

Tach goes all the way to zero and engine dies when clutch in 80pct of the time when rolling up to a stop. Rest of the time it winds down and settles at 1k like it should.

Sometimes it won't start at all. I pop the air cleaner off and just barely push up on the sensor plate lever. Then it pops right off.

When driving and it starts winding down it's possible to keep it alive with a little heel and toe touch on the pedal .

Have not done a smoke test yet.

I did the adjustment like I did in the last post of my link above on Friday and I took it for a 10 mile spirited drive. Was perfect. Didnt die once. Parked it, drive it again later that day and it was back to dying.

Also, sensor plate position on that so stop screw seems to me to only affect starting for a split second. Once the engine is running the plate is riding on a cushion of air, so the stop is out of the equation. Right?

[jlex]

I'd look for a vacuum leak. Had the same problem with my '70 VW beetle. Had to replace the throttle shaft bushings to fix it. Do a smoke test to help locate the leak.

[mike sampsel]

I had a bad fuel injector and it caused similar issues. Maybe check the spray, compare the volume per on injectors and look for a leaking/dripping injector.

[Jameel]

Thanks guys. I should add all injectors are new and spraying equal amounts.

[SkiVT]

Since it’s easy to adjust, I would at least try adjusting your decel valve, assuming it’s connected: loosening the lock nut and screwing the shaft in, will make revs drop more quickly, and screwing it out will make revs drop more slowly. See if it helps hold idle high enough to not stall.

[walt]

Did you put in a lightweight flywheel by chance?

[Jameel]

Originally Posted by walt

Did you put in a lightweight flywheel by chance?

Yes! Forgot to add that

[Jameel]

Lightened pressure plate actually. Still running the stock fw

[Jameel]

https://photos.app.goo.gl/Ft8Hf6WfCejcrGuT7

Here's a short video

[Schulisco]

Originally Posted by Jameel

https://photos.app.goo.gl/Ft8Hf6WfCejcrGuT7

Here's a short video

That's very interesting!

You can see that the tachometer/the engine will hang a short moment before it dies completely. This is a desired behaviour (as it helps to smoothen the gear changes while accelerating) and shows that the deacceleration valve is working (which is very good!) and the vacuum connections here are tight. But unfortunately it doesn't proves that there's no additional vacuum leak at another place of the engine. But it also means that it could be a different culprit. Otherwise I heavily recommend to make a vacuum leak test with a smoke generator. And check mixture (CO) and the ignition timing, especially the vacuum acceleration and retard of the dizzy and if the vacuum lines are connected properly (even at the throttle housing!). The CO screw and the idle screw are working closely together. Will say - you can "tune" the idle to a too low CO value and vice versa. Also check and probaby adjust if necessary the injection begin with sensor plate height. Curt from Klassikats.com showed the procedure in this video close to the end, but I recommend to watch the whole video, also the following fine tuning video:
https://www.youtube.com/watch?v=K-OvjlkrgNo&list=PLkWFofpgz2OgF9FdliKnlzuY0pygqRkMj&index=4

Thomas

[Jameel]

Great reply Thomas. I'll check the video. Planning a smoke test next week.

I made a small adjustment to the deceleration valve yesterday, lowered the plate a little, and sped up the idle to 1k. Took it for a spin around the block and it didn't die. But that's happened before too, so maybe that did it, maybe not. But definitely some good points, thanks.

[Jameel]

Aha, I watched that video in detail when I rebuilt the the CIS a few months ago. Kurt has great videos.

[sigchuck]

I had a similar issue on my SC around 5 years ago - RPMs would drop to 400-500 rolling to a stop with clutch in, but then rebound. My recollection is that I was a bit too rich (10% chance I'm not remembering and it was too lean), but in any event, I was able to fix the problem by adjusting the idle mixture screw in connection with the test that involves slightly lifting the airflow sensor plate. Excerpt from link https://jimsbasementworkshop.com/CIS/pages/test_mixtures.html :

An approximation of the correct idle mixture can be made by the folllowing:
With the engine warm and running, slightly lift the air flow sensor plate.
If the RPM immediately falls off, the mixture is probably too rich.
If you pull down, and the idle falls off, you could either be too lean, or about right. Do the step immediately below to find out.
If the RPM slightly rises, then falls off as you continue to raise the sensor plate, it may be close to right.
You are trying for about 14.7:1.
The required movement of the sensor plate is slight, on the order of maybe a 1/16" or less.