Short irregular idle period on cold start - continued
 

Ever since I've had the car I've had this short period of rough/low idle when cold starting the car. It starts immediately most of the time, idles low for a second or 10 (600-800RPM), then picks-up to regular "warm-up-idle" (1600RPM). After that everything is perfect. Warm start is perfect as well. See here my previous thread about this topic: Short irregular idle period cold start.

Like every winter I methodically continue my search, solving every little electrical problem, vacuum leak etc one by one.
I had checked the cold start valve before and it works perfectly. No leaks when not powered, and a nice spray when powered. Thermo time switch (TTS) as well, good resistance (0 to ground on one half, 24Ohm to ground on the other, 24Ohm post to post). Closes off when powered, all be it a little quick. Problem is, I can't really find an official spec that says exactly how soon the TTS should close, only that in the coldest temperatures it should be max 8 seconds.

This year I decided to check the CSV in its' complete circuit, with TTS and powering it at the same time as powering the fuel pump, so that it would completely emulate a cold-cold start. I took a video of the cold start spray:

<div style="position:relative;"><iframe src="https://player.vimeo.com/video/676640160?h=9a05063894&amp;badge=0&amp;autopause=0 &amp;player_id=0&amp;app_id=58479" frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen width="750" height="422" title="ColdStartValve"></iframe></div><script src="https://player.vimeo.com/api/player.js"></script>
This looks like a very, very small puff of fuel to power the cold start. It looks like a combination of the fact that the fuel pump needs to get the pressure up which takes a while, and the TTS closing off really soon.

Now I'm wondering. Is my fuel pump getting old? Broken internal check valve? Should it be quicker to get the pressure up for a nice spray of the CSV? And/or is the TTS closing to soon? I checked the timing of the video; the spray is about ~1 second long.

Looking forward to read what you think.

The OP in this thread http://forums.pelicanparts.com/porsche-911-technical-forum/956477-cold-start-issue-my-newly-refreshed-cis-3.html seems to have had the same problem.

Bump

the CSV has nothing to do with idle once it starts.

that looked like quite a large burst of fuel to me.

start with basics,
new plugs, rotor and cap.
set timing.
set mixture
check fuel pressures (cold)
you said you fixed air leaks. what about the injector sleeves and Orings,
manifold bolts tight.

how are the injectors,
you might pull them
make sure none are leaking.
put them in containers and lift the sensor plate to put a good amount of fuel in there.
then compare the volume (or weight ) to make sure they all are the same.
should be easy if you have the flexible lines.

does your car have the thermos time valve for the WUR. is that working. (cold control pressures)

mixture. too lean/rich

does it do the same if it is warm outside or cold.


its been a long time since I had my 77. I know it always started right up but I cant remember if it did the same as yours.

Hi,
If you have a vacuum leak in the oil tank you will get an odd idle. Check the cap seal and the sender seal. Also maybe just send your cold start valve out and have it rebuilt. I used fuel injection products to do mine.
Turned out someone put a 930 part on my 911. Got the right one and now starts beautiful.
You should also have the auxiliary air regulator rebuilt by the same people- this affects idle, also I did that too- only thing I have left is rebuilding the fuel distributior.lol good luck-

That short time (approx. 1sec) of powering the CSV through the TTS is normal! They're intended to give a quick shot of fuel to help starting the engine. The TTS powers the CSV anytime while cranking the engine, it varies only on duration, max. 8sec to prevent drowning the engine depending on temperature. Newer TTS are fully electronic, they power anytime the TTS, older electromechanic TTS with bimetal strip inside - don't know. Some say they won't do it nymore on higher temperatures e.g. 35°C / 95°F. So - the spray in the video is ok! The shortest injection time is half a second on hot engine.

I experienced as well some start problems with low idle speed directly after starting the engine - after replacing the fuel pump on my SC (not original replacement part and - missing check valve!) they're gone! So - I recommend to replace the fuel pump (depending on how old yours is) with an original Bosch - those include a new check valve as well in the box! The CIS is extremely sensitive about fuel pressure! If the fuel pump is worn it might fit for normal driving conditions, but especially on start or WOT it might not deliver suffcient fuel pressure. If furthermore the check valve is gone or defective - then the fuel line can contain some air bubbles - and they are compressible, which the fuel accumulator isn't able to equalize over the whole fuel line across the car. Díd you verified the residual pressure of the fuel system after shutdown of the engine? The time you see low idle seems to me exactly that time the fuel pump needs to fillup all the line free of bubbles.

To prove that this is the fault: Before you normally start the engine, bridge the fuel pump relay before you first start the engine (sitting for more than 12hrs) to repressurize the fuel system (Howto to bridge the fuel pump) Let only the fuel pump run for a few seconds. Then remove the bridge, put the relay back in socket, and try to start the cold engine. If the cold engine then directly idles abve 1200rpms cold wihout limping on low rpms - congrats!

The rest has already been mentioned: check all pressures, especially fuel pressures (you find on the web some build plans for a fuel pressure check tool for CIS or purchase it), check with a smoke tool for air leaks (you won't find them all using brake cleaner spray when they're too small), injectors, O-rings etc.

Keep us posted!

Thomas

Do you have a thermovalve in your car? It is a small, round, black vacuum limiting device that has vac lines attached to it, the throttle body and the WUR. It limits the vacuum pulling against the lower chamber of the WUR for only about 30 seconds to allow for easier starting, but then shuts off to allow the correct vacuum to be applied.

After you have determined the ignition system is good and there are no vac leaks, you should check the CIS, incl the WUR, the AAR and AAV.

you can do flow tests on your FP.
the check valve is a good point but usually they show up bad with hot starts,
not to say your isn't going bad.

start with the easy basic stuff

Thanks for all the input.
I verified that I have no vacuum leaks left by pressurizing the intake with a smoke machine. It was high enough pressure to push open the pop-off valve so that I had to hold that one closed. After changing the injector sleeves, sleeve-o-rings, injector-o-rings and placing a little o-ring on the throttle shaft (rebushing the throttle body might be overkill) I couldn’t discern any smoke escaping from the intake anymore. I’ll double check the oil cap seal, but since I closed off the intake on the big side of the rubber boot that would have been part of the system and I didn’t see any smoke coming from there.

Control pressures I checked a while ago, they were fine then but I’ll do the test again and report back.

Thomas, I’ve been suspecting the fuel pump as well, I’m going to do the test with powering the fuel pump before start and see what happens. In the mean time I’ve ordered a replacement spare, a Pierburg 7.21659.70.0 which has a built-in check valve and should therefore be a simple bolt-on swap. The available Bosch pumps all come with external M12 check valves and I still have the M10 banjo connection original to the 78.

CIS troubleshooting.........
 

Niels,

You need to test the FP , CSV , and TTS independently to simplify your diagnostic. Watching your video does not help identify the culprit/s. There are too many unknown variables in the equation and reducing these variables would help you identify the culprit. If you are closer, I would volunteer to do the tests for you. The actual test for each component (FP, CSV, & TTS) would require only several seconds. It is the set up and connecting it to the tester that will need time. Maybe 10 to 15 mins. max.

BTW, the TTS does not supply power to the CSV . What it does is provide ground contact to complete the circuit when the engine is cold (45*C/113*F and below). Above these temperature, the TTS has an open ground for the CSV and disable the CSV. Starting the motor when it is warmed (45*C/1138F and above) does not require any help from the CSV.

I have a post about bench testing a TTS some where in the archive.

Tony

i`d to suggest for cis user this troubleshooting guide ..it has an excellent info in it how to go about it.

Ivanhttp://forums.pelicanparts.com/uploa...1645258766.jpg

here is one of them on ebay.com number 165005272938
it is all you need to solve your problems

Ivan

@Thomas, I did the fuel pump test like you said, bridge the relay, turn the key to run so that the pump runs for about ~5seconds. Key off, remove bridge, relay back, start and woohoo, it starts at once and goes directly to 1800RPM! So I guess that means the fuel pump check valve is gone. Since it's internal to the pump and I have my replacement ready I guess I know what to do!

Here's a video of my smoke test by the way; no smoke coming from anywhere:

<div style="position:relative;"><iframe src="https://player.vimeo.com/video/680100173?h=9a05063894&amp;badge=0&amp;autopause=0 &amp;player_id=0&amp;app_id=58479" frameborder="0" allow="autoplay; fullscreen; picture-in-picture" allowfullscreen width="750" height="422" title="ColdStartValve"></iframe></div><script src="https://player.vimeo.com/api/player.js"></script>

(and yes, I checked that the machine was creating smoke :) when I disconnected the pump there was a big release of smoke)