VW Vortex - Volkswagen Forum banner

The great KE-Jetronic mystery!

27K views 26 replies 7 participants last post by  t16vtricks 
#1 ·
I'm having troubles with my 16V PL engine for over a year already. I've read into how to tune this system, how it works and what every component should do. But I just can't figure out what's wrong...

The symptoms I have:
  • The car smells as if it's running rich
  • Starting is hard, once it runs it runs great at higher rpm's
  • When cold the car sputters a bit in higher rpm's
  • When I put the mixture leaner it won't start. playing with the idle screw doesn't help
  • Won't idle, when you come off the throttle the engine stalls, comes back to life and then stalls again and stops
  • Adjusting the idle screw makes the engine either not start, or come back to life after stalling a few times more before finally stalling
  • Sometimes the starter motor has problems turning the engine over, I plan on installing a diesel starter motor

To properly adjust the car it should be idling as both the DPR current and idle speed valve duty cycle have to be measured and adjusted at idle.
I tried to make a habit of writing down how much I turned the mixture and idle screws so I can always go back to and adjustment where it will start and run when you keep your foot on the throttle. I use a Bentley manual for all the work I do on my car, because they're the best!

Now here's what I've done too it:
  • Changed the injector O-rings and the plastic injector holders. Everything put together according to Bentley Publishers manual.
  • Cleaned the throttle body
  • Changed the O-ring around the idle screw
  • Replaced litteraly every seal in the intake with new ones
  • Installed a 4-2-1 exhaust manifold + new manifold to head seals
  • New lambda sensor
  • New spark plugs and spark plug leads
  • New distributor cap and rotor
  • Compression test: all four cylinders between 14 to 14.7 bar. That's more than it should originaly have, I guess that's why the starter motor sometimes has problems.
  • All hose clamps in the intake are properly tightend
  • On hoses which originaly don't have a hose clamp I put RVS or plastic zipties
  • Checked the full intake circuit for vacuum leaks
  • Checked the vacuum lines for leaks
  • Checked the injectors, at idle they have the same flow. They have good spray pattern but with air flow meter lifted up completely they should deliver 80ml/40s and they only deliver 60-65ml/40s. Shouldn't be a problem for idle I think.

I also changed the timing belt so the timing should be correct, I always doublecheck this. I will check this again if I have time. The ignition has been put in it's basic setting where the rotor lines up with the mark on the distributor housing. To adjust this properly the engine should be able to idle which it doesn't.

I plan on doing a fuel pressure test but I lack the equipment to do this so that's one thing I have to sort out.

I also did the complete CIS-E electrical test as described by Bentley's in the manual, there where some odd things but I did this half a year ago so I'll check this and post it here.

I'm completely lost and desperate on this, so if anyone is able to give some advice shoot it!
The biggest problem I'm facing here is that it seems that everyone who had a thorough understanding of the CIS-E system is dead... So I'm searching for a KE-Jetronic know-it-all who can give some advice or point me into the right direction!
 
See less See more
#2 · (Edited)
The Bentley is not written in a fashion to teach CIS-E, and in that regards the sections are out of order.

The CIS portion is mechanical injection. The mechanical mixture is controlled by the "idle mixture" screw on top of the airbox. The -E portion is electronic mixture control. In non-full throttle running the computer tries to maintain a stoichiometric mixture. Use the DPR test procedure to check the DPR value. Adjust the "idle mixture" screw to get the DPR value in range. Also make sure the CTS value is in range. And make sure the oxygen sensor is working.

But your injector flow is only 75% of what it should be. This is going to starve the engine of fuel. This is where you might want to do a fuel pressure test first. Trying to fix a fuel pressure problem by changing the mixture is not going to work. There are two quick maintenance items in the fuel system if you have not gotten hold of a fuel pressure tester yet.

1) The fuel filter.

2) The fuel mini filter. This is a tiny screen which was supposed to be removed from cars at the first maintenance. See Mini Fuel Filter in the Bentley.

Edit: The mechanical part of the injection requires that the fuel pressure be correct, and the "idle mixture" (mechanical mixture) be set correctly when the fuel pressure is correct. All the electronic mixture adjustment is doing is fine tuning the mixture under different conditions.
 
#3 ·
I have already replaced the fuel filter. I'll check if the mini fuel filter had been removed or not. Could this mini fuel filter mess up the injector flow/fuel pressure? I will do a fuel pressure test as soon as possible! But I don't have the equipment atm.

After fuel pressure test, should I disconnect the DPR harness connector, use a CO measuring probe to put the mechanical mixture right and then connect the DPR to get the fine-tuning right? Or does it not work like that?

I suspect the lower injector flow is caused by low fuel pressure since all four injectors have a lower fuel flow. Will low fuel pressure also affect idle? I thought it would only affect full load running since that's when you need the most fuel. The strange thing is, it runs great in high RPM's but not on idle. And it smells like it's running waaay too rich, but when I adjust the mixture towards lean it doesn't want to start anymore.

One suggestion I got on another forum was to adjust the mixture at a workshop using a CO measuring probe to get around the problem of not being able to measure DPR current, and then adjusting the idle and ignition. But I'm not sure, will this work?

Thanks for the advice!
 
#4 ·
Sorry to ask some tough questions, so do not take it personal.

Have you tested any of the electrical signals for the CISe system? What is the DPR current reading? Have you tested the ignition timing and is it right? What does rich smell like? Do you know the injectors only spray fuel? The distributor is in charge for fuel flow/metering. Do you have any intake leaks? How did you determine that you have no leaks? Really? They're all dead?

I would first take some of your ideas and throw them out of your head. Smelling rich and then leaning the mixture out [and it now does not idle] does not sound like it's rich to me. Counting the amount of turns for the 3mm allen is not right and throw away your notes on that. You're way off by now.

The only way to adjust the idle now is to have a running engine with a fully operating CISe system. You can measure the current [ma] and determine where it should be. I'm more familiar with Mercedes, but I do believe the reading should be around 10ma [again, don't trust me, research to see if I'm right]. To have a fully operating CISe system, you need to have a good ECU, O2 sensor, coolant temp sensor and of course the wiring. If that is not working, you're really spinning your wheels with a whole bunch of assumptions.

Fixing cars issues is about diagnosis, not by swapping/replacing parts. More people should spend time with testing and less working on replacing parts. Just because it works with someone else, does not mean it will fix your issue. Once you supply the required test results, better decisions can be made. Better questions can be brought up for us to help you with an answer. You appear to be so lost that it's best to start over. There was a tool in the movie Men in Black that would erase what you saw, this is what you need. It's tough to start over and not want to do what you already did.

So, I would make certain the CISe current was working, if not I would find out why. That is your biggest problem you need to fix. I then would confirm you have no intake leaks. A smoke tester is a valuable tool that is way too expensive for a DIY'er, but ask around. ALL the air must be measured by the fuel system and the air that is not measure really effects the entire system much more than you think. The fuel pressure is important, but if it's low, the high loads will be effected much more than the low loads. In other words, idle will most likely not be effected much. Remember, the injectors will only open when it has 2-3 bar. If the fuel pressure is less than that, the injectors will not open and therefore the engine will not start.

Again, do not take it personal.
 
#7 ·
I did a full electrical test to check if all the sensors were still in good condition. I can't measure DPR current because you need to do this with the engine on idle. The last time I measured it it alternated between 12 and 8 mA, but this was with the idle set at 1600RPM's. It should be 5-6mA with the idle at 950 RPM. I know what rich smells like, the kind of smell you can get on tuned mopeds. Other people have also commented that it smells very rich.
I don't have a smoke machine so to check for leaks I visually checked it and sprayed some brake cleaner in the area's where leaks could occur and listened if the RPM's increased. I count the amount I turn the mixture screw to be able to go back to where I started, as a kind of backup.

None of the parts I replaced where replaced without reason. Everything I replaced was either at the end of it's life or broken. I agree that too many people just assume things and swap stuff. I study automotive engineering so I get told every day to test first and then replace. So yeah, everything that was replaced was due for replacement!

And I know that there's only fuel coming out of the injectors because I did a flow test. Fuel is lighter than water so if there was water in it you would see two layers in the container where the injector sprayed into.

I don't take it personal, don't worry ;)

For basic "idle mixture" and airflow sensor plate adjustments see:

6. CIS-Electronic (CIS-E) and CIS-E Motronic > 6.1 Air Flow Measurement and Fuel Metering > Basic Adjustments

This should get you in the ballpark.

Then to dial it in see:

6. CIS-Electronic (CIS-E) and CIS-E Motronic > 6.7 Idle Specifications > CIS-E

Take a look at all of the stuff mentioned there. This includes the CO test you mention.

Fuel injector volume is a function of two things: fuel pressure and how much the airflow sensor plate lifts the fuel plunger in the bottom of the fuel distributor. Since you do not have access to a fuel pressure tester yet skip to looking into the height of the airflow sensor plate at rest and the distance between the airflow sensor plate and the fuel plunger. See "Basic Adjustments" mentioned above.

If "idle mixture" is way off the ECU still tries to bring mixture into range by using the O2 sensor. Problems if it is way off show up at full throttle when the ECU disables electronic mixture adjustment and relies totally on the fuel pressure and idle mixture.
I did center the sensor plate but didn't check the height adjustment. There was a bit of freeplay in it when I did the injector flow test.
I think I'm going to first check and adjust the height of the sensor plate and then do a CO test to adjust the mixture. I will also check the fuel pump in the tank to see if this is working properly and once I get my hands on a fuel pressure tester I will test the fuel pressure.

I think I have everything covered then, since the only part of the system I haven't checked completely is the fuel system.
 
#6 ·
For basic "idle mixture" and airflow sensor plate adjustments see:

6. CIS-Electronic (CIS-E) and CIS-E Motronic > 6.1 Air Flow Measurement and Fuel Metering > Basic Adjustments

This should get you in the ballpark.

Then to dial it in see:

6. CIS-Electronic (CIS-E) and CIS-E Motronic > 6.7 Idle Specifications > CIS-E

Take a look at all of the stuff mentioned there. This includes the CO test you mention.

Fuel injector volume is a function of two things: fuel pressure and how much the airflow sensor plate lifts the fuel plunger in the bottom of the fuel distributor. Since you do not have access to a fuel pressure tester yet skip to looking into the height of the airflow sensor plate at rest and the distance between the airflow sensor plate and the fuel plunger. See "Basic Adjustments" mentioned above.

If "idle mixture" is way off the ECU still tries to bring mixture into range by using the O2 sensor. Problems if it is way off show up at full throttle when the ECU disables electronic mixture adjustment and relies totally on the fuel pressure and idle mixture.
 
#8 ·
This is the aquiles heel of cis based systems now.

The technicians that worked on tgese back in the 70s,80s and early 90s are all retired, moved onto other jobs or have not worked on this technology and the tools to work on them at dealership levels are almost non existent.

So owner operators of such cars have to collectively help each other.

I have owned and cared for a cis e motronic passat since 2005.
I hace had sinilar issues and still have a smaller up hill battle left with tge poor running condition on cold starts.

My problem was also fuel system pressures related.
 
#9 ·
I started working on cars in 1980 and worked with strictly German cars since '84. 30 years working at a dealership. I'm still here and own my own shop now. There are others around.

My professional stand point is that the customers that own them do not want to pay the money for a proper diagnosis and cannot afford the repair either. All they want is a free diagnosis and if you could install the part for free, that would be ok with them. When you get my age, you see the same thing again and again. That's why you cannot find someone that can work on it.
 
#10 ·
So quick update! I took the air flow meter assembly out of the car and did all the basic adjustments.

The rest position of the lever arm was a bit too low, probably 1mm out of tolerance. It is now set to 2mm.
The basic adjustment of the lever arm is now 19.1mm so also within tolerance.

I also found that the mini fuel filter wasn't swapped for a normal union bolt. I ordered the union bolt (N 021 071 2) + the two sealing washers at the dealership. The car had been sitting for years before I got it, I begin to suspect the mini fuel filter could be the cause of my troubles. Maybe fuel residue from evaporated fuel was partially blocking the mini fuel filter reducing fuel flow and pressure..?

I also took the coil out of the car since that was the only part of the ignition system I didn't look at yet. The primary resistance measures 1.0 ohms, the secondary measures 7100 ohms. The primary resistance should be between 0.6-0.8 ohms and the secondary resistance should be between 6500 - 8000 ohms. So the primary resistance is out of tolerance, but can this be the cause of the problems I have?
 
#11 ·
Many years ago I would visit some guys who worked at a shop which worked on VW/Audi/Porsche. One of the mechanics was developing a collection of those mini filter screens. Supposedly they can be pushed out of the banjo bolt. An old one will definitely hurt your fuel flow.
 
#12 ·
Soooo everything is installed back into the car. The result is: the car runs a lot better with that mini fuel filter removed. Adjusting the rest position of the air flow sensor lever arm seems to have helped with starting the engine. Starts really fast now, from the first turn of the key.

BUT. I still can't make it idle. With throttle the car runs great. But on idle it stalls. The closest I can get is to a point where the RPMS will drop below idle to like 300rpm or so, then the car revs up to restore idle and after that the rpms drop again. It does this two or three times and then the engine stalls. Engine temperature doesn't seem to affect this.

I took a look at the ECU's, here's what I got:

BOSCH 0 280 800 180 (VW 811 906 264 F) KE-Jetronic controller


Triumph-Adler knock control unit (VW 811 907 397 E)


Hüco 138004 ignition module


They seem to be the correct computers for my car, except for the ignition control module. When I google this one it says it's for a VW T3 caravelle 1.9l.
Could this be the source of the problem? Would seem odd to me since the car runs great in higher RPM's.
 
#13 ·
Second update today.

I doublechecked the timing, here are the position of the camshaft, cranckshaft and ignition distributor.



Everything seems fine.
The camshaft is halve a tooth out of alignment but this is acceptable, the Bentley manual says this:
Some movement of the sprockets and their marks is to be expected as belt tension is adjusted. Keep in mind that the smallest possible increment of adjustment is one whole tooth of the belt or sprocket.
I also did the control input tests again as described in the Bentley manual



The electrical systeem seems to be ok. I only can't find a green oxygen sensor wire. My sensor is connected with a brown, a black and a red/white wire.
I also noticed that terminal 9 is missing from the connector for the ECU. Any ideas about this?

I also found a few loose connectors I couldn't find a connection for.

A red, white and orange one. They're in front of the ABS pump, under the ignition distributor.
red: green wire
white: white/blue and green wire
orange: white/blue wire


White 2 pin connector with a green and a blue wire. Same location as the three other connectors.


A black 3-pin connector which is next to the connector for the ignition module above the KE-Jetronic ECU. It's taped to the wire loom. The colors of the wires:
2 times green/blue
1 brown blue wire


Quick list of the symptoms:
  • Under load (= applying throttle) the engine runs fine
  • The engine stalls on idle, engine temperature doesn't affect this.

To me it seems like the problem is somewhere with the idle function. the idle valve and idle switch both work. I really don't know where to look anymore.
 
#14 ·
Have you cleaned the inside of the idle stabilizer valve with some sort of electrical parts cleaner? I have read the ISV can gum up over time. I think my engine might have this problem with idling.

Try turning the A/C on. Running the A/C puts the ECU into a different idle mode. Do you have the same idle issue? This makes my idle work correctly.
 
#15 ·
I cleaned it with brake cleaner, then sprayed a little WD40 in it for lubrication. This was over a year ago, it has always worked perfectly. Electrical tests are okay, and it hums/buzzes when the ignition is on.

I don't have A/C so won't be able to do that.
I have measured the transfer pump fuel delivery rate, and the main fuel pump fuel delivery rate. The transfer pump delivers 500cc in 10 seconds, the main pump delivers 750cc in 30 seconds. So both are well within specifications. I think it's safe to assume fuel pressure will also be okay since pressure depends on the flow (delivery rate). I will test fuel pressure to be sure
 
#17 ·
After testing the fuel pressure I am now quite sure the injectors are bad.

System pressure: 5.4 bar
Differential pressure with DPR disconnected: 5.2 bar
Differential pressure with DPR connected, Amp meter in series with the DPR and 15K ohm resistor over the coolant sensor connector: 4.5 bar, with 60.9mA of DPR current
Residual pressure after 11 minutes: 2.8 bar

Since the problem arose after pulling the injectors out to change the O rings, and since I did need a lot of force to pull them out because the old O rings were super hard, I think the injectors may have suffered from that.

At low load the spray patern is okay but they don't chatter and create a nice mist. More like little jets continuously spraying. Under high load they do chatter and create a good mist of fuel.

Also, the engine only runs bad under low load. Up until 2500rpm it runs rough and very erratic. Once above 3000rpm it runs very smooth and is super responsive.

So all that, together with the fact I checked almost every other possible cause, leads me to believe the injectors need to be replaced
 
#18 ·
Oh yes the fun isn't over yet, this story hasn't reached its end...

So I replaced the injectors, which made a huge difference. But I still can't get the car to run right. I also replaced the ISV so I'm now a 100% sure that wasn't the cause of my problems. I still can't get the idle speed below 1500rpm.

Yesterday I hooked the multimeter up and started fiddling with the mixture, I did this with a warmed up engine. So I got the mixture to where it was going back and forth between 4mA and 6mA, at an idle speed of about 1400rpm. When I touch the throttle the DPR current goes up, when I lift off the throttle it falls to zero. So that's working right. At this point it ran good, good throttle response over the whole RPM range. None of this lazyness under 3000rpm like before, that's where the new injectors come into play I think.

But, now the engine has cooled down I can't get it to start anymore. I managed to get it started, but when it starts it doesn't want to idle. Lifting off the throttle makes the engine cut out.


Because of the high idle speed I'm still thinking about vacuum leaks, but I have checked every vacuum line, every breather hose, every gasket is new,... I can't see where the leak could be. Also when it was idling, spraying around with deodorant didn't do anything. So what the heck could be going on here?


The weird thing is that these problems only started after pulling the injectors out, and changing the old O-rings. And yes I changed the plastic inserts etc aswell. So I'm thinking it must be mixture related.
 
#19 ·
The ONLY way your engine is going to idle at 1,500 rpm is if there is enough air for it to idle that way.

You need to smoke test it. Either the idle control system is not working right or you have a large air leak. There is not too many options. Smoke test the intake, that is the only way you'll find there is no leak. You cannot see a vacuum leak, but with smoke, you will.
 
#22 ·
Okay so big big big update!

I built my own smoke tester and did a smoke test. No vacuum leaks! I use a test port on the intake boot to bring the smoke into the intake system, when reinstalling the plug in the test port I used some sealant and after doing this the idle rpm dropped so I guess there was a vacuum leak there.

But the engine still idled very rough. It constantly dropped rpms, then the ECU kicks in and brings the rpms back up. This repeats itself.

BUT! all of a sudden this was gone, the engine idled around 950rpms and started smoothly. It did hold back above 4000rpm when driving. When I went to adjust the ignition I noticed the ignition was severly retarded, and when I turned the distributor to bring it back it would retard again after a while. This led me to the knock sensor, which was bad. I replaced this (and torqued it down properly) and now the ignition was adjustable. I brought the engine up to temperature, connected my multimeter and stroboscope and got adjusting. I got the DPR current to 'search' between 4 and 6 mA, and the ignition to 6° before TDC. Idle rpms where too high now, but I brought these down by adjusting the idle air bypass screw. After every adjustment of this screw I checked and adjusted mixture and ignition again.

Finally I got the idle rpms to sit around 950rpm, the mixture at 5mA +/-1 and the ignition at 6° before TDC. The exact specs as given by the Bentley manual. Throttle respons was great. I tightend down the distributor, checked the ignition one last time and removed all the test equipement. I shut off the engine.

I cleared out my driveway with the intention of doing a test drive, so half an hour later I come back to the car to do a test drive. All of a sudden the car is very hard to start, and when it starts the idle problem is back. Throttle response is still great, but below about 2000rpm the car runs extremely rough. The problem doesn't go away when the engine warms up, oil temp was 96°C so the engine was warm.

I did an electrical check of the knock sensor system as explained in the Bentley manual and everything came back fine. Hall sensor on the distributor recieves power, and the ignition module is fine according to the electrical tests in the Bentley manual. There was some oil in the hall sensor connector, but cleaning this out did not eliminate the problem.

What am I missing???
 
#23 ·
Another update: I'm now quite positive the problem can be found somewhere in the ignition system. It's almost certainly something electrical anyway.

I did some checks of the ignition system as described in the bentley manual. When the ignition is switched on, the voltage between pin 1 and pin 15 on the coil should jump to at least 2V and then drop back to 0V. It jumps to about 1.5V. When pin 12 of the knock sensor module connector is briefly connected to ground, the same thing should happen. The voltage jumps to about 1.5V but doesn't go back to 0V. It stays floating between 0.6V and 0.8V. The bentley manual tells me this points to a faulty coil or ignition module. I measured primary and secondary resistance on the coil:
  • Primary: should be 0.8-0.6 ohm, I measured 1 ohm
  • Secondary: should be 6500-8500 ohm, I measured 7500 ohm
So basicly, secondary resistance is within spec, primary isn't. However a multimeter isn't accurate enough to measure such small resistance, so I believe the coil is fine.
This led me to believe the ignition module was faulty, since it's only 30€ for a new one I replaced it. I ripped the old one open and found a stain on the board:


I don't know if there was supposed to be a component there, and if so which. I also don't know if this was the cause of the roug idling and the strange engine behavior. But all I know is that the module was due for replacement anyway, judging from that stain. Which makes me happy because untill now I haven't replaced anything that didn't have to be replaced, and I want to keep it that way!
 
#24 ·
Replacing the ignition module fixed the rough idle issue. It was a long road and I've learnt a lot about the K-jet system, but I'm happy I kept searching and didn't bring the car to a shop. Knowledge about these systems is valuable.

In the end the original problem was probably a combination of bad injectors, bad knock sensor and an ignition module which acted up from time to time. The injectors were probably okay at first, but the sheer force which was needed to pull them out when I replaced the O-rings probably destroyed them. The knock sensor was bad but I think it worked half of the time. Same goes for the ignition module.

I'm happy I didn't replace things which were not due for replacement. It took me longer to find the fault but cost me less money.
Now I'm off to have fun with my 16V :laugh:
 
#26 ·
You have a great attitude towards your journey. You did a lot of testing that helped you go towards your issue. Even some of the tests that did not give you the results you wanted, those test will help you in the future about most any fuel injected car [CIS or EFI].

Congratulations!
 
#27 ·
Ok I can tell you from my experience and having almost the same issue and doing just about everything you’ve done. I found that my fuel distributor is bad it was leaking internally causing all of my issues. If you’ve got a spare I would throw that in and see if that fixes the issue but make sure it is exactly the same as the one you have because they’re not all the same good luck
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top