https://i.imgur.com/VfPNvIz.png ;)SmileWavy

The values I showed above are for the 930/10 (means ROW 3,0SC from '81-'83, no catalyst, no lambda control).

Location........
 

George,

Where are you located? USA or Europe?

Tony

@ Tony

Just completed FD leak test with pump running. No leaks.
See relevant video.

https://youtube.com/shorts/7usraTLfN3Q?feature=share

The weird thing is that after this test the engine fired at first crank. (7 hours since stop) Engine temp was around 28 c so CSV was activated for sure.

Next is to check sensor plate alignment, just to confirm it's within spec.

Waiting for your comments.

Thanks

Europe Athens Greece.

Thanks Thomas!

Weird, indeed.

I experienced a similar behaviour: I had to crank the engine much longer after the car was sitting a while due to an "emptied check valve" (no ball and no sppring inside anymore, therefore the fuel ran back to the tank; cranking that longer gave the fuel pump the time to refill the fuel lines) and the bad fuel accumulator as I wrote sooner. BUT: Still the engine started after a few seconds cranking....but your's won't start at all with the CSV connected...

You could check the CSV by plugging it manually with a switch to ground while the engine is running. If the CSV works, it will inject fuel and riches the mixture that much, that a warm idling engine will stall or at least start surging.
While driving / accelarating and powering the CSV you richen the mixture as well. Depending on the current mixture either the car gains or looses power. With this method I diagnosed (wihtout a CO tester) that the mixture on my car is too lean under load. The difference of power while accelerating with a lean mixture compared to a proper/richer mixture is tremendous...

Refer to the schematics on how to put the switch in the circuit for the CSV (similarly to replace the thermo time switch with a manual switch...)


https://i.imgur.com/bmFNL3x.png


Thomas

CIS troubleshooting............
 

George,

When the problem exists randomly, it would be challenging to diagnose the culprit/s. Double check the setting of the sensor plate before we go crazy looking for unknown culprit. It would be a challenge to align the sensor plate with the AFM (air flow meter) installed in the car. I always do it on the bench but you could do it with more patience.

If I were doing the troubleshooting, I would test for:

• Unmetered air going into the system. Test and verify.
• Residual fuel pressure.

After you have confirmed the setting of the sensor plate, test run the motor and run it for a few mins. and let it cool down. Wait 7 hours more or less and try to re-start the motor. Do not try to start after two (2) unsuccessful attempts. If it starts after a couple of attempts, we’re in trouble.

If the motor failed to start after a couple of attempts, try this:

• Pull out the FP relay.
• Bridge terminals #87A & 30 to test run the FP.
• Allow the FP to run a few seconds to “pressurized” the system.
• Install back the FP relay and try to start the motor.

Keep us posted. Thanks.

Tony

Change of plan.......
 

George,

Before doing any adjustment with the sensor plate, try the suggested test procedure. Does the motor start at the first attempt after a short stop? Could you test and verify that the motor would run after:

• 4 hours.
• 6 hours.
• 8 hours.

We need at least 2 data points for each test condition. We have to establish how reproducible is the occurrence of the problem to identify it. Thanks.

Tony

We will find the solution All together. I am sure about it. We are very close to narrow it down.

In order to summarize.Tests and spares:

1. Check valve and pump are new
2. Residual pressure will be maintained at 1.3 bar for more than 1 hour
3. Cold and warm pressures within spec. Verified many times
4. Csv is new. Tts tested with multimeter and test light with external 12v power supply.
5. New injectors
6. Fd not leaking from plunger or injector ports
7. Checked csv 12v power supply while cranking.
8 Adjusted CO to 2.5%
9. Set ignition timing
10. After 9 hours standstill i jumpered the fp relay to prime the system before i crank it. Same result. Will not start unless csv is disconnected OR after cranking for more that 10-15 seconds (within this time csv is deactivated)
10. Spark plugs are new W3CC
11. Engine will start half crank after a short stop. Even 1 hour later.

To be done:
1. Check sensor plate alignment
2. Test starting after 4,6,8 hours
3. Repeat intake leak test
4. Re confirm CO% level

Additional test.........
 

George,

Perform a residual pressure test for the system pressure (fuel distributor). Test run the fuel pump with the CIS gauge installed (valve open) and you are measuring the control pressure.

Close the valve to measure the system pressure. And switch off the FP. Write down the pressure readings:

• Valve closed and FP running.
• Valve closed and FP off.

Take note of the pressure readings on the gauge:

• Immediately after turning off the FP.
• After 20 mins. (pump off).

Let us know the pressure readings. Thanks.

Tony

Ok will do.

System pressure with pump running (valve closed) is set to 5.1 bar. Reverting with remaining tests.

Thanks

Updates.

Today morning ambient was 30c. Engine did not start after over night stay for the first time. 13 hours have passed since engine was stopped.

Started after csv was disabled.

Seems that as ambient temp goes higher the problem is getting worse at cold starts. It is definitely an over rich mixture culprit.

Will try to start after 4 hours parked. Tomorrow will do the 6 hours start attempt.

Weekend will perform remaining tests.

Thanks!

You can play with the valve of pressure measuring unit to change the control pressure by half closing the valve. So you can lean the mixture by raising the control pressure....

is the CSV wired properly at the STARTER

Below video of start after 4 hours parked.

Started immediately. CSV should be active as i checked TTS ground contact is closed. Used multimeter to check contact to ground.

https://youtube.com/shorts/PWPc8lKGses?feature=share


Thanks

Good test idea!

And see at what control pressure it starts. Will DO

Thanks thomas

Yes.

Have 12V at TTS and CSV during cranking. Will make a lamp adaptor also to test the CSV plug this weekend. If you check previous msgs the engine will start easily when CSV is de activated. CSV is new. Old had also the same symptoms.

Thanks

Diagnostic test results........
 

George,

It is quite difficult to determine the root cause of your cold start problem because it occurs randomly. The “9-hour window” you have observed occurs at random. One day you observe it and other times it’s gone. Focus on solving the COLD START PROBLEM and ignore the 9-hour no start for the meantime.

Eliminating the CSV for cold start is a complete violation regarding CIS principles. You have tested and confirmed that the fuel distributor is NOT leaking. But you have to confirm the basic setting of the FD plunger using the fuel mixture screw. The initial setting is about 1/2 turn (CCW) from open/close setting. Followed by the final mixture setting with the engine at operating temperature.

Our goal is to identify the culprit/s preventing your motor from cold starting. Let’s make it start consistently and worry the other problems. Things I am interested to investigate:

• Spark plugs (brand, type, and appearance after all the tests).
• Is the CSV actually working?
• Residual pressure loss @ FD.
• Unmetered air.
• Etc.

Tony

Good day

1. Spark plugs 3wcc bosch. New. Last time removed brown tan color. Will pull out again.
2. CSV is new. Will connect test light to plug and confirm.
3. Residual pressure loss @ FD. Reverting with test results
4. Will do an intake test again

Do i need to do the basic setting of the plunger since i have access to gas analyzer? This will be final setting anyway.

George, did you already checked the rubber fuel injector sealing rings and the red injector sleeves? There might be a vacuum leak as well...hard to find if not checked manually.

I guess that we do not have only a single problem. When I recapitulate the facts of starting right away and next time not I guess that a vacuum leak in the past was corrected out by manipulating the CIS and now that or another leak will occur again from time to time and additionally preventing a proper cold start. Also a temperature dependency is given leads me to search for sth. where its getting hot...

Thomas

Hi Thomas

Injector pockets and all seals were replaced with new together with injectors.

Weird,
Almost like an electrical connection that is needed for spark is weakened when the tts is grounded and pulling current. The resistance to the power or grounds needed for the cdi being insufficient and changing with temperature/humidity/vibration, once the alternator is running the problem dissapears over all temperature, for now.

Phil

George,

At this point you can not use a gas analyzer if you could not make it START. If you are able to make it start and get the motor to operating temperature, then use the GAS ANALYZER. There is something basically wrong with your CIS if you could not make it to start.

If you have fuel, ignition, and the right amount of air, your motor should at least show sign of life. And disabling the CSV to make a cold motor start is a sure sign of trouble. Ask your self why would a working CSV is detrimental for a cold start? The problem is right in front of your/our noses but we are not seeing it. We could NOT solve this problem if we could not identify the culprit/s. Unless you get lucky.

Do you have the correct spark plugs? If they are, then one variable to eliminate. There are so many variables involve and eliminating some of them gives us a better batting average. Unmetered air comes #1.

Tony

The funny thing is that if i let it rest for a couple of days it will start cold first crank!
How about that?

Will do some tests during the weekend and if i have the time i will pressure test the intake.

Phil i conected the ignition timing light and there is spark during cranking.

Thanks

I am missing something and i put you in all this trouble while it is surely something simple.

Have started with the fundamentals.

Checked sensor plate alignment. Is good.

http://forums.pelicanparts.com/uploa...1653597889.jpg

I would also check both UFO/disc vacuum valves on the right side of tge engine ... If you have only one, then the second is on the left side close the throttle body.
One is the additional air valve. This valve is open until the engine creates vacuum on start. Then it shuts the channel when the engine runs. Its the one with only two connectors.
The other one is the deacceleration valve. It is controlled by vacuum. It prevents popping in the exhaust when going off from the throttle. This valve also let the engine keep 2500-3000 rpms to make easier/smoother shifting before idles.
If one of both is going crazy and remains open or closed from time to time
this confuses the CIS and as well the machanics working on that engine...


https://imgur.com/Zn1bkXn.png


For a proper test take them out and test them with a vacuum pump manually...

Pretty expensive if they need to replaced.

Edit: I watched your vodeo again - I cannot see any of both valves at your engine...is that right? Or do your engine has the "vacuum limiter" next the throttle body (911 110 220 00) ??

Thomas

Hi Thomas

The setup you present is for US spec cars.

See below Euro configuration.

All are in position on my engine. AAV should be closed once 5" vacuum is applied. I will test that also, but when this valve goes bad the engine gets crazy. Surging / high idle ect...

Decel valve opens, bypassing air around the throttle plate to avoid rpm sudden drop and stalling when high vacuum induced. (foot of the throttle)

http://forums.pelicanparts.com/uploa...1653640256.jpg

CIS troubleshooting.........
 

George,

Making your motor STARTs at first attempt is where you should focus. You need the CSV for cold start. A properly set-up CIS regardless it is a Porsche, Mercedes, BMW, VOLVO, BENTLEY, Ferrari, etc. should fire and run........and stall if you have a significant air leak or lack of fuel or weak ignition signal. The motor needs all three (3) critical elements:

• Fuel.
• Ignition.
• Air.

Test and verify that you don’t have significant air leak. They all leak, it is a question if it is significant to affect ignition. The lesser leak the better and become less troublesome. Air leak is the Achilles heel of CIS. Good luck.

Tony

i am preparing the equipment for relevant tests.
Will do everything from scratch.

I have one theoretical question.

Since disabling CSV makes it start, wouldn't an air leak make it start easier with the CSV connected? it would compensate the rich start condition by inducing more air.

Has anyone faced cold start fail by vacuum leak?

In my experience (NOT CIS engines) even when vacuum lines are missing the engine will start but idle will be erratic or stall. I am also referring to carbureted engines w/o O2 sensor to compensate the mixture.

Thanks

George,

The more you have unmetered air going into the system, the less chance you could get the motor to start or idle. You need a smoke generator to locate these hard to locate air leak source/s. Some of these leak sources could be overcome by setting the fuel mixture too RICH to make the motor start but would not run well.

A disabled CSV would make the condition more LEANER plus additional unmetered air makes the ignition more difficult to attain. That is why it is critical to eliminate unwanted unmetered air in a CIS motor.

I will give you a very good example. Find me a good running CIS motor and I will make it not start or run at all by simply introducing an air leak. You could spend all the time you need to make it run again and it won’t run again. Unless you locate and identify this culprit. It is not how hard you work but how smart you tackle the job. This is not rocket science. There is something wrong with your CIS set up and simply not aware about it.

You are an engineer and knows how to check and verify things. Wish you the best and good luck.

Tony

I guess the very first and foremost thing to decide is to find out whether the mixture is either to lean or too rich. And George already found out that it's too rich. So my conclusion is, the engine didn't get enough air with the current settings and the connected CSV. Either the setting of the CIS is bad or it get's too less air. Yesterday I wrote also that we probably face here not only one fault...this is also my experience. Most CIS cars suffers on more than one problem. And this confuses everyone.

Of course the car can also have a vacuum leak problem (which probably occurs from time to time only without any reliable pattern). But it's also imaginable that one or more problem exist. Or that the following problems cam from a previous problem like vacuum leaks. Then you can "tune" any CIS car to run better and hiding / not solving the problem...if the next problem occurs come on top, this can confuse totally...

An example: My car also still had serious cold start issues. I examined every part of the CIS...almost everything was either bad (airbox, TTS, FD, injectors, vacuum lines, switches, temp sensor) or a wrong part (in my case the WUR). I replaced them (TTS, WUR, airbox) pr let overhaul them (FD), injectors, temp sensor, switches), adjusted the sensor plate etc. And still the car won't have a proper idle on cold....
The reason I found later - that red plastic flange for the CSV was one of that false series with no hole - it was closed for the air line coming from the vacuum valves and the AAV....so the engine could not start either when cold. It immediately ran on too rich mixture every time when cold. I could keep it running while playing with the throttle. But then the mixture was quickly too lean because the engine was cold...it took me everytime several minutes until then engine was warm enough before I could get it out of the garage...very frustrating and annoyed neighbours...

I identifed the problem with a smoke generator filling the vacuum lines from one of the vacuum valves and closing each other - after a while the filling line of the generator popped out due to overpressure...:eek:
I already recommended to use such a device at my first posting here as Tony did as well...

That's what I mean when working on a going crazy CIS car...you cannot rely on any part of it unless you examined and proved it...

Thomas

Thomas

I suppose that when you had the cold start problems the engine would not keep high idle during the cold start due to air from AAR and AAV not passing through. Correct???

Thanks

Correct, that's why I had to play with the throttle to keep it alive as long as the engine was too cold. When it was warmed up a bit (after 2-4mins depending on the ambient temperature), it ran at an idle speed about 900-1100. But this sticked together with the other problems the car had...it was a challenge over years, because I'm not a professional car mechanic...in the beginning after the purchase I put the vahicle to a professional workshop, they replaced the airbox and made other minor things, but they didn't solved the real problems the CIS system had. So I studied CIS myself and learned a lot about it over time. The fact that the engine also had other camshafts installed (WebCam20/21) did not make the whole thing any easier.

Thomas

Good day to everyone.
1. Performed a thorough smoke test. NONE leakage found.
2. Connected a test lamp to csv plug. Working.

http://forums.pelicanparts.com/uploa...1653744402.jpg
http://forums.pelicanparts.com/uploa...1653744528.jpg

Going back to fuel rerated tests.
Reverting

Smoke Test...........
 

George,

How did you isolate the system (airbox) from atmospheric condition? A picture of your set-up would be helpful. Thanks.

Tony

I put a plastic bag on top of sensor plate body and installed the boot back into position. This way i also check the intake boot integrity.

The smoke device outlet hose was installed below throttle plate where the bottom hose of decel valve is normally connected.

http://forums.pelicanparts.com/uploa...1653758394.jpg
http://forums.pelicanparts.com/uploa...1653758508.jpg

Additionally i checked spark advance during cranking with strobe light. Well eatablished around 5-7 degrees BTDC.

tomorrow will remove all spark plugs for inspection and connect fuel gauge to check pressure during start and re check residual with valve closed.

Thanks!

Good day

Spark plugs photo below. Condition looks optimum to my eyes. Some oil on the circumference but ground electrode is tan brown.

http://forums.pelicanparts.com/uploa...1653845752.jpg

1. Residual pressure with valve open 1.5 bar after 30 min.
2. Residual pressure with valve closed drops instantly from 5 bar to 2.5 bar and will stay to 1.8 bar after 30 min.

Please check the video of cold start after 24 hours. I think something wrong with WUR?

https://m.youtube.com/shorts/GIcySvwJhXo

Let me know what you think. If someone can do same test and upload video it would be appreciated.

Thanks

Watching the last video, is the control pressure supposed to be so low for the first 10 seconds? (looks like zero bar at first then very slow build up)
I am not sure how much difference it might make, but when you attached the pressure gauge, was it the day before and you ran it to clear out the air and make it ready for the test 24 hours later?



Phil

Your observation is very accurate.

Gauge was not installed previous night. Was installed just before cold start attempt.

Will connect tonight and check tomorrow.

What i dont understand is that when i disconnect the WUR to FD line after 1 day there will be not even a single drop of fuel although residual pressure is within spec.


Thanks