He had already replaced the OVP relay so he is good there, the fuel pump relay isnt a bad idea but I am buying on it being a bad cold start valve. Try going to www.benzworld.org and looking up cold start valve. If you say once running its fine I am betting your cold start valve has a leak. Since it gets shut off when its at operating temp and you say after it cools off 30 min after shutdown it happens again It really sounds like that.

Good luck hope this helps.

I'm beginning to wonder about the cold start valve as well. Early in the thread I said I have been chasing this problem since I bought the car in March. When I think back, that's not accurate. It started perfectly the first couple of weeks, though it had a slight miss at idle. It was only after that, that the start/stall thing started happening.

Since the weather is better, I've been driving the car almost every day and the behavior is very consistent. Hot or cold it always does the start/stall thing. The only time it doesn't, is if I restart it within a very few minutes. Most of the time it take 4 try's for it to keep going. When it does there's always a few seconds of buck & shudder until it settles down to a smooth idle.

If I hold the throttle open a bit it will start and not stall as long as I keep the revs up. If I let off the gas, it will stall, even if I've held it down for a minute or more. It seems like it needs to be left to its own devices in going through the buck & shudder stage before it will settle down and behave.

The fuel pump relay seems unlikely. At one point in my chase I jumpered it, and the behavior was the same. Seems like, if it was the relay, the start/stall behavior would have gone away when the pump was jumpered.

The new airflow potentiometer should be here in a few days. Hopefully it will do the trick. In any case I'll report on the results. Thanks to everyone for the advice.

__________________
2000 Mercedes S500
1990 Mercedes 560SEL
1970 Triumph Spitfire

Pull the OVP and with a single strand wire, jumper/tie the pins 1 [30] and 2 [87e] together at the bottom of the relay pins [ wrapping around the pin bases]..you can then plug the the relay back into it's connecor so the other pins have their completed designated circuits..you will just be jumpering the suspect terminals at those two pins.
Retry the car...............

** Note
..this is just a TEST. Do not leave it b/c the ign sw is also bypassed [ pin 3, circuit 15] with this jumper test..............

__________________
A Dalton

I did the test and this is what happened:

• When I put the jumpered relay back in I heard a click from somewhere under the hood, in the area of the fuel distributor.
• The CEL was on when I got in to start the car.
• The engine cranked a long time before starting, with a strong smell of fuel.
• When it finally did start, it was stumbling pretty much as it always does, and the CEL stayed on.

I removed the jumper and when I got in the car to start it the CEL was off and it started right up and settled to a good idle right away with the CEL off.

Later, when I tried to start it, it did the usual start/stall thing.

Does this behavior give any clue as to what's going on?

__________________
2000 Mercedes S500
1990 Mercedes 560SEL
1970 Triumph Spitfire

It just eliminates the main feed from the OVP as suspect.

Got a 12v test lamp? [or anaolog meter]

Pull the relay again and stick a test lamp + into the smaller pin on the connector..this will be 15 Ign [ pin #3. between pin 1 and 5]

Go in the car where you can still see the test lamp under the hood and turn the key to RUN and jiggle it around...test lamp stay steady of Flicker?
If flicker, ign sw is going bad.
If steady, problem is elsewhere....
Suspects is a poor conection somewhere........

__________________
A Dalton

Sounds just like my '90 300SE when the coolant temp sensor went bad.

The temp sensor is a four-prong brass plug near the #6 cylinder. The unit contains two actual sensors, one for the fuel computer and the other one for I forget which system.

A while back I posted the temp/resistance curves for that sensor. Check your sensors with a DMM. If one or more of them is bad or off, it will act just like you are describing, though I never saw the CEL light. The CEL light could just be a bad or fouled out O2 sensor.

Oddly, I had a bad temp sensor a while back and the brand new one I put on only lasted 6 months before one of the elements went open.

Of course, I checked EVERYTHING else because I just KNEW I had a good temp sensor. Not so.

At any rate, they're $30 and a good one will improve performance and MPG so you might wanna replace it regardless, if it's the original one still in there.


Good luck!

__________________
-tp


1990 300SE "Corinne"- 145k daily driver - street modified differential - PARTING OUT OR SELLING SOON - PORTLAND OR. AREA - PM ME FOR DETAILS
1988 560SEL "Gunther"- 190K passes anything except a gas station
1997 S420 - 265k just bought it with a rebuilt trans. Lovely condition

I like the logic of this. I'm still waiting on the new airflow potentiometer to arrive, so last night I adjusted the EHA clockwise 1/4 turn to see what would happen. While it didn't make the problem go away it did make things a little better.

The car starts best when cold, first start of day, and it restarts perfectly if you shut it off and do a restart in a few minutes. After that it takes more try's to get it going.

• This morning's first start took one try. Before it usually took 2.
• The second, after sitting about 10 minutes, took 2 try's.
• The third, after about 30 minutes, took 3 try's.
• the forth, after about an hour, took 4 try's.

Does the idea of enriching the EHA a little and having it work a slight bit better fit in with your coolant temp sensor theory?

Anyway, considering the price, I'll order one and see what it does. Will I need to drain the cooling system to replace it? Also, is this the right part?



Thanks for the new direction. I'm not crazy yet, but I will be soon. I tried
the idea of the problem being in the ignition switch, but the test yielded a nice steady test light. :-(

__________________
2000 Mercedes S500
1990 Mercedes 560SEL
1970 Triumph Spitfire

That's the correct part. I didn't drain the coolant to replace it, but I did get a little spillage, not much though.

I have a new coolant temp sensor on the way, but decided to measure the resistance of the current one just to see what it's doing.

With the car fully warmed up, I touch the diagonal pins on the sensor and see a reading that almost instantly disappears. When I take the probes off, then touch the pins again, it's the same thing, a reading that flashes on then quickly off. It the same with both sets of pins. Is this a sure sign that the sensor is bad?

__________________
2000 Mercedes S500
1990 Mercedes 560SEL
1970 Triumph Spitfire

Does your meter work properly on other components?

__________________
2012 E350
2006 Callaway SC560

As far as I know. The last thing I used it on was to set the voltage on the airflow potentiometer and it seemed to work correctly. Before that it was to check battery voltage and it was fine then. Do you suspect that what I saw may be the result of a faulty meter?

__________________
2000 Mercedes S500
1990 Mercedes 560SEL
1970 Triumph Spitfire

< __________________
1989 Mercedes 300 SE
>>

It's a sign your battery is low on your meter.

Hold the leads together and see if you zero...........

__________________
A Dalton

WOW! I'm impressed. I wanted to say my battery was fine as I replaced it a month or so ago with a new Duracell alkaline, but I was wrong, it was low. Holding the leads together gave a reading of "1.". I replaced the battery, and when I measured its output with the new battery in place it was about 6.5 volts. Holding the leads together now gives a reading of ".001".

I went back and measured the sensor again. At 37C. it was ".99". I started the car and let it warm up until the dash gauge showed normal and it measured ".344". That's not what I expected to see, maybe I'm even more meter-challenged than I thought. This is the setting I was using to get the readings. Is it correct?

__________________
2000 Mercedes S500
1990 Mercedes 560SEL
1970 Triumph Spitfire

No...

That is Continuity Buzzer and Diode test position..

Read the book.

__________________
A Dalton

Dang, how embarassing! OK, I moved the meter to the 20K range, and got these readings:

It was 2.00 measured before starting the engine this morning. Using a digit thermometer with the tip placed right next to the sensor the engine was at 62F/16.66C.

After starting the engine and letting it warm to normal according to the dash gauge, I got a measurement of .26. Engine temp measured 172F/77.77C.

I turned it off and waited an hour. It then measured .91 at 104F/40C. I took it for a drive then measured .18 at 183F/83.88C.

I don't if this is significant but when it does the start/stall thing, it fires as soon as the key is turned, the engine revs up to a little over 1,000 then drops straight to zero and stalls. Once it has started and stayed running, you can turn the key off and restart it without the start/stall thing, and it will crank for about a second before it fires. It revs to to a little over 1,000 but instead of dropping to zero, it drops to about 400 then steadys down to a normal idle. It does this even if the car is not warmed up yet.

__________________
2000 Mercedes S500
1990 Mercedes 560SEL
1970 Triumph Spitfire