My '85 190E (US 8v w/ only 88K mi) rough started one evening , but cranked and wouldn't start the next morning.

After going through the published troubleshooting routine to check the OVP relay, ignition component continuity and resistance values, and any obvious vacuum leaks without finding anything out of spec, I entrusted (i.e. towed) it to a local import shop that sells a lot of MBs and BMWs.

I had interviewed the owner, and he was confident the tech would find the problem. After three weeks and "a lot of work on it" they told me the timing chain was bad. Three weeks to learn this? I had seen enough and towed the car home.

To confirm their diagnosis, I lined up the timing marks at the distributor and the cylinder head to see how far off it was. It was dead on. I lost confidence in their opinion.

Concerned that there was bad fuel in the tank, I wanted to drain the 3-5 gallons that remained there and replace it with fresh fuel. I figured I'd disconnect the fuel line in the engine compartment, and use the fuel pump to deliver the gas to a container through some vinyl tubing.

I turned the key to excite the pump, but turned too far into the "start" position. THE MOTOR STARTED RUNNING WITH THE FUEL LINE DISCONNECTED.

It continued to run (for the first time in two months) at a smooth idle while the gas flowed from the tank into a five gallon bucket instead of the fuel distributor. (Thank God there were no errant sparks).

After adding some fresh fuel, and reconnecting the fuel line, the original problem remained. Now, I had to re-create the "No Fuel Line" scenario to see if the results would repeat. They did. The car starts easily and idles smoothly as long as the fuel line is disconnected from the FD. Otherwise it won't start.

I assume that the motor is running off of vapor drawn through the recovery canister. I also assume that the timing chain is not as bad as it was "professionally diagnosed" to be.

I hope to start troubleshooting the CFI soon. I did notice that the plugs are getting wet with fuel, and I assume that timing and spark are adequate due to the "NFL" running condition. There does appear to be some fuel accumulating below the air sensor plate.

Apparently fuel is getting delivered, but not under conditions to ignite. Perhaps its a fuel/air issue, but you would think that if it starts and runs on the vapor, that I could at least get a rough idle out of it under normal starting conditions.

Any insights into the KE II Jetronic tests without the use of a scope would be helpful. Also, has anyone had any luck using a dwell meter on the #3 terminal at the diagnostic socket?

Maybe you disconnected the fuel return line???
And maybe the fuel return is plugged & causing TOO high fuel pressure.

__________________
MERCEDES Benz Master Guild Technician (6 TIMES)
ASE Master Technician
Mercedes Benz Star Technician (2 times)
44 years foreign automotive repair
27 Years M.B. Shop foreman (dealer)
MB technical information Specialist (15 years)
190E 2.3 16V ITS SCCA race car (sold)
1986 190E 2.3 16V 2.5 (sold)
Retired Moderator

Yeah I doubt the car would continuously run and idle off fumes....it is getting fuel from somewhere. I second MB DOC's answer.

__________________
-diesel is not just a fuel, its a way of life-
'15 GLK250 Bluetec 118k - mine - (OC-123,800)
'17 Metris(VITO!) - 37k - wifes (OC-41k)
'09 Sprinter 3500 Winnebago View - 62k (OC - 67k)
'13 ML350 Bluetec - 95k - dad's (OC-98k)
'01 SL500 - 103k(km) - dad's (OC-110,000km)
'16 E400 4matic Sedan - 148k - Brothers (OC-155k)

"Maybe you disconnected the fuel return line???
And maybe the fuel return is plugged & causing TOO high fuel pressure."


Gentlemen,

Thank you for taking the time to reply. The fuel return line was not touched, but my initial hunch was that indeed there was too much fuel/pressure and the plugs were flooded before they had a chance to fire. Hopefully, I'll get the gauge on it this W/E and have some numbers to report.

MB DOC, my understanding is that the return line should be free flowing with a check valve at the tank connection. Would a shot of compressed air into the line (with the fuel tank filler cap opened to vent) be an acceptable test, or should I be somewhat more subtle? Any common causes to look for? Thanks again.

BTW, I was able to start and idle the car by removing the fuel pump relay, with all the fuel line connections in place. After idling for a while, oil started squirting from the PCV line, so I shut it down.


Apparently, the engine was able to run on fuel that was lingering around the injectors as well as fuel that had made its way into the crankcase.

"The engine idles with the fuel pump relay removed"


Without pressure from the fuel pump, I have to assume that the injectors are not opening. Therefore, the fuel would be arriving through the intake. Is it possible that a leaking FD could provide a path to the return line which makes fuel available to the vacuum of the intake, allowing the engine to run without the pressurized fuel pump feed?

If so, adding pressure to the equation (assuming that the 3 bars necessary to open the injectors is maintained by the faulty FD) would mean fuel from two sources. A potential flooding situation. Is this logic sound? Is it enough to create a non starting situation?

I was able to put the fuel pressure tester on the vehicle, here's how it reacted:

Initially, the pressure shoots up quickly to about 6 bars, then the regulator seems to bring it down to about 3.5 bars. If I continue to crank the motor, the pressure slowly rises.

I tried it with the return line both connected, then disconnected and fed to a gas can, the results were the same both times, so I am assuming that there is not a clog in the return line.

The system pressure seems to hold for several minutes after the ignition is turned off, but I can see fuel dripping inside, beyond the air sensor .

The "lower chamber" pressure seems to be near the .4 bar difference that is mentioned in the shop manual.


With good fuel pressure numbers and wet plugs, perhaps I have to look at the possibility of a weak spark? Keep in mind, the engine WILL idle smoothly when the fuel pump relay is removed, or the fuel inlet line is disconnected. So there is enough spark to ignite whatever it's burning under that circumstance.

Another question I have regards the auxiliary air pump. My understanding is that it provides the extra air necessary when the air sensor is in the resting position. If the air pump is not functioning, would this have the effect of choking out the engine during start up ? I tried cranking the motor with the sensor plate slightly depressed, but the it reacted the same.

Anyone test their auxiliary air pump lately?

This week I removed the fuel distributor as well as the housing attached to the throttle opening at the intake.

Upon close inspection, it appears that the lambda screw was run almost all the way in. Besides that, the spring loaded hex adjuster in the lambda tower was stripped at the point where it meets the adjustment screw.

I repaired the tower assembly, and would like to think that I set the air sensor zero (rest) position properly, but the service manual diagram seems different than what I'm looking at on the bench.

Assuming the zero position is set properly, the next step is to establish 1-2 mm of air sensor movement (play) from the zero point to the point that the assembly contacts the FD's control plunger. I used the lambda adjustment to bring this in from the 4mm play that existed after setting the zero position.

With the FD assembly connected to the fuel line and pressurized, a drop of gas forms on the tip of the control plunger. There is a serviceable seal beneath the threaded insert that surrounds the control plunger. A slotted tubular wrench is used to access it. However, even though the procedure (with a tool number) is described in the manual, there is no part number for this plunger seal.

After bridging contacts 7&8 at the fuel pump relay connector, pressure was maintained at about 5.3 bars. I removed the injectors and mounted them in plastic bottles. The spray pattern seems acceptable, and manually deflecting the air sensor plate controls the flow evenly.

I ordered some non-resistor plugs from Phil. 5k ohms less resistance than what was on the car.

Yes, you can use a dwell meter on the 4cyl. scale:

>45 degrees/50% = running lean
<45 degrees/50% = running rich

__________________

1993 190E 2.3
2001 SLK230
1971 LS5 (454) Corvette Convertible

Oil through the PCV outlet was caused by gas in the crankcase which thinned out the engine oil and created too much fluid for the system to handle. I am assuming that the gas was from all the unsuccessful attempts at starting the vehicle while the lambda screw was pegged out against the control plunger of the FD. I hope that no permanent damage was done to the internals. I have replaced the oil, and the PCV system is functioning normally. Oil pressure has been restored.

After sitting overnight, there did not seem to be any gas in the oil. I'll have to watch that.

I am going to spend one more session setting the positions of the air flow sensor plate and control plunger, in order to create the 1-2 mm of play when pressurized at idle. In order to start the car at the current settings, over 8mm of play was generated by the new lambda screw position

After that, I have to determine whether the electronic controls of the CIS are functioning. Initially, it appears that I am getting only 0.6 volts at terminal 3.

I've ordered a fresh O2 sensor.

Is checking the reference resistor a straight-forward task? That thing looks like something from a Tesla experiment.

Thank you, my dwell meter is digital, so I'll verify whether the scale is in degrees or per cent.