S500 Coupe Suddenly Dies, but Restarts
 

About a month ago, my (so far) completely dependable 1996 140 coupe stopped running at a light about 2 minutes after starting after sitting for two hours. No warning, no stumbling, just died, leaving all the warning lights on, like the ignition had just been switched on. About ten minutes later, again at a light, it almost died, but then did a little surge and stayed running.

I sought local advice, looked at the TSB's, did a Search and replaced the crank position sensor and added a .5 mm shim as advised.

The problem went away for a couple of weeks, and then it happened again just as I was coming to a stop. More consultation, and advice to try a different key. I also had all of the codes pulled from all of the modules, and nothing relevant showed up.

OK, so with a different, barely used key, the problem did not recur for a few days. Then, a couple of nights ago, driving at 25-30 mph, I could feel the engine lose power, twice, for less than a second each time.

The only other weirdness with the car is occasional ASR/ABS lights that go out when the ignition is cycled, but this did not happen in any of the recent dieing episodes.

I am considering a fuel pump relay. The fuel pump was replaced about 18 months ago and has about 8K miles on it.

So any other suggestions on what to test or what to try? This is a rather large car that you don't want to die on you in traffic.

Bump!

What were the codes??

I don't have the print-out with me, but as I recall:

- the window controller thought the switch direction was backwards (not true)
- a couple of "implausible" readings from one wheel sensor
- one other "implausible" from the transmission gear selection

The tech said they see this random stuff all the time, and none of it had any relevance to the stalling problem. It's not leaving any tracks.

The VSS (vehicle speed sensor) is very important . . . I'd re-think that conclusion. . .

OK, I can see the VSS code being related to the occasional ABS/ASR lights.

What is your hypothesis about the relationship to the engine suddenly shutting down?

Here are the codes:

HE1R
P1519 Right adjustable camshaft timing logic chain

ASR
C1101 RPM sensor RF -000 L6/2 Wire

ETC.6
114 Occasionally Selector lever implausible
120 Occasionally RF VSS from engine management implausible, CAN
126 Occasionally Altitude correction from engine management implausible, CAN

<What is your hypothesis about the relationship to the engine suddenly shutting down? > At this point, I don't think it's valid. . . . :)

Since you replaced the L5 (cksft sensor), it still could be electrical . . intermittant coil?? But the fuel relay is probably as good a choice as any other.

Have you checked the fuses in the Base Module . . maybe replace them. The Base Module is the "OVP" of W140 cars. . .

Hook up a fuel pressure gauge and go for a drive and watch fuel pressure when it dies.

The hose came in today, so here I go.

I have been driving around for a couple of days with the gauge. The pressure seems perfect.

But it did die on me at a light today. No miss, no warning, just died. The pressure was sitting at about 3.2 bar and went up to 3.8 as the car died. Then dropped back to 3.2 with the engine off. Car restarted easily, drove off fine.

The surge and the die happened simultaneously.

I have noticed that if you simply shut the ignition off, which it seems would simulate the fuel pump stopping, the pressure stays at about 3.2 and drops imperceptibly to about .8 bar overnight.

Does this ring a bell with anyone?

. . . it's not fuel related . . . I would bet?? What do you think?>

Agreed. More likely ignition/electrical.

I think the quick jump in fuel pressure is happening beceause the injectors are shutting down a fraction of a second before the fuel pump is cut off. Like the ECM is doing it for some reason, or the ECM or some other critical piece is losing power.

I did plug into the OBD port and the sensor plots and other readings looked normal. The only oddity was that the long-term fuel trim was 5.5% and 6.0% bank 1/bank 2.

Whatever it is, it's starting to happen more often.

Truth is stranger than fiction
 

I had a talk last night with my ultimate resource, a tech team lead at the area's largest dealership. He suggested first checking the basics: battery charging, grounds, clean leads, etc.

Then he mentioned three possible causes: bad base moduel, bad ECM, and the trans module locking up the torque converter and killing the engine. He also said to check the harnesses for oil "wicking" from the transmission and from the cam ajdusters.

So I took a look at the base module and noticed some oil in the bottom of the relay compartment. To get a better look, I unhooked the trans module and noticed oil around the connecting pins. Then I removed the trans module, and trans fluid spilled out of it as I turned it. I put it in a pan, and I would say a cup of fluid ran out.

So my working theory is that being full of oil caused the module to malfunction and occasionally lock up the TC.

I will try to get the case off try to clean it up and see what that does for the stalling problem.

Hey there, been following your exploits on Benzworld. That base/trans module you talk about. Is it stuck to the bottom of the tranny?
Do you have any part numbers or pictures to identify which is which?

Just had a thought though, I've had an episode of this at over 2000RPM, would the Torque converter not already be fully locked up at this engine speed?
As an aside, I've noticed my engine speed when idling, (car in gear), is lower than normal. Last but not least it has performed the cut out thing while sitting in neutral too, so is the tx a red herring? (Start your car and just leave it sitting there idling in neutral and see if it happens).
Don't know really if these are just coincidence but they all came about at the same time.

Couple of pics added too. Found the 38 pin connector at last. For some reason I always assumed it was on the other side of the engine and could obviously never find it there. In the pic next to it there is a bracket. Is this where where the LED blinker might usually sit on US spec cars?
The fuse box is pictured also, can see none of the expected modules there, (base module).

Any help is much appreciated.

I was able to get the board out of the case and hose it off with electrical degreaser. Put it back in, drove the car around the block, and then let it sit and idle in gear for an hour. No stalling. Previously it would have stalled at least a couple of times. Not willing to declare victory at this point, but a very positive sign.

I also made barrier with a paper towel and sort of wove it through the dozen or so wires in the harness. Hopefully, this will keep things dry while I try to figure out what to do about the wicking.

Well, back to the drawing board for me. No grease in that box, no cracking or burnt cables so far either. No fuses blown.
How are you getting on?

I've not seen that problem . . did you take in pics?? How does oil get in there?? Bad seal?? Oil wicks up the harness??

I've seen it a lot, but mostly with 722.6

Isnt yours a 722.5?

Nope, it's a '96, first year with the electronic transmission. The plug on the transmission is faulty, and I find out, there is a redesigned part to seal it better.

Yes, trans fluid wicks up 48" or so of wiring, gets in through the connector and pins on the ETM and fills the case. I estimate that there were 2.5 ounces of ATF in the case.

You would think that this would be worth a TSB from Mercedes, since it apparently is not all that uncommon.

I drove the car yesterday, and I'm not out of the woods. It didn't die at idle, but it did stall in the middle of a rather sharp U-turn, which is a new problem.

Chuck, there is a bulletin. P-B-27.00/47a dated Oct 2004.

That bulletin mentions the possibility of fluid at the ECU.

Thanks for that information - Alldata did not include it on the 140 database. The ETM filling up seems to be a Mercedes family secret.

So, my next question to the experts is, would the ETM survive its ATF bath?

Would replacement be recommended, or are additional test steps?

Thanks,

How are you getting on, did the problem come back?

Still in the weeds. After the die-in-the-u-turn episode, it had one more surge-but-not-die episode on the way home that day.

I planned to swap the Base and Transmission Modules from my SL500 one at a time today and see if that made a difference, but the part numbers are different.

So my plan at this point is to replace the transmission electrical fittings with the updated pieces, and get any fluid out of the connectors. If that doesn't help, I will pay for a couple of hours of diagnosis at the dealership by the tech who gave me the lead on the trans fluid problem.

If anyone has a better idea, I'M CERTAINLY LISTENING!

How are you coming along?

Well, I know my car and my dads, (W210-E430), have the same Tx and climate control module. So I figured maybe some of the control boxes, fuses and relays might be the same. Nope. Except the fuses but that doesn't help me. I've noticed that some of the relays in my engine bay, and boot are the same so although it's unlikely to make any difference, its a no cost easy way of eliminating certain areas. That's to be my first avenue, after that. Not sure. I think I'll sit collating info for a while.
Couple of things I do know are;
a) it's probably not heat related as it can happen within 30 seconds of starting the engine from cold.
b) It's probably not transmission related as it's happened while sitting in neutral with the parking brake on.
c) It's probably not Air con related as although some others remarked how the AC settings made a difference. They didn't for me.
d) The car throws up no fault lights in the dash.
e) Its not the fuel filter. Just changed it a few days ago, (after the onset of these problems).
f) Spark plugs too had been changed not so long ago.
g) It almost failed the MOT test on emissions this time around, (was running very rich).
) whatever I change I should know instantly whether it's made the difference as there is a definite symptom I can see at the mo. (The car may well drop as far as 350-400 RPM while slowing to a stop).

Any of that lot point to anything for you?


ps. This is interesting too;
http://www.benzworld.org/forums/w140-s-class/1384007-140-545-7132-e-gas-computer-new-post.html
What is that box?

I think you should do the fuel pressure check as I described earlier. You would look for a fuel pressure drop just before engine stalls.

Your rich running could be a number of things like pressure regulator, O2 sensors, MAF, coolant temp sensor, etc.

My update is that since the u-turn incident, the engine has not stalled. It "staggered" a couple of times, but did not stall. I drove it in traffic today and it was fine.

My next step is replacing the old-style connectors on the transmission harness with the new ones. I will also clean the harness at the transmission end. My theory is that since the lockup signal wire and the ground are next to each other in the plug, that the "wicked" oil may be causing enough of a connection to cause the TC to lock up.

I'll have to get myself a hose and a pressure guage then. Exactly where did you connect the thing?

Just my 2c - I've just now read your post for the first time. First thing that went though my mind was WIRING to the crank and cam sensors. Wiring, connectors. They short all the time, melt, degrade etc. You can often see it as a frayed wire, cracked connector etc.

Look around, trace a few wires, do a simple continuity ck on a few - let us know.

Also - the little mag bumps on the flywheel HAVE to be clean in order for the crank sensor to see them. Dirt/grease between sensor and flywheel tabs will kill it in a hurry. Remember, crank sensors almost always shut off the motor, sometimes briefly. Air sensors (like MAF) and fuel sensors will just make it run awful, but usually runs.

Look at the exhaust stream - namely O2s sensors. AND their wiring.

Would a detonation sensor/advance sensor cause a shutdown? I've always noticed the proverbial rough/sloppy running motor when that occurs

Finally - look at the cam advance connectors - they're right up front in all the dirt. I have to replace mine once every few years. Oil soaks them and they get brittle. Again, never had one shut down the motor though, but may be a place to look.

Cam and Crankshaft sensors almost always shut down the motor. Always when you least want it to (like downtown Harlem)

Many thanks for that. I shall get to it, wiring first and then the sensor. I'll have to do some research to see if there is a way to test any of those sensors, 'on the bench' so to speak.

The connector for the pressure test is right at the front center of the fuel rail. You can't miss it if you remove the little cover in front of the air cleaner.

No tabs on the flywheel, though. The crank sensor on the M119 actually generates a high/low signal from each tooth on the flywheel. Crank position is determined from two missing teeth at 270 degrees. The computer uses this and the cam position sensor to determine TDC on #1. But I am told by a very experienced tech that missing cam sensor input will not stop a running engine.

You know I've noticed that before but didn't know what it was.

Where on earth do you get something like that?
 

Went out today looking for some hose and a guage. Turns out that fitting is a thread of 11x1.25
Nobody but nobody seems to have them.

Of all places, a Mercedes dealer
 

The PN is 119 589 04 63 00. $56.12.

Now that is surprising.
In the mean time I have invested in Carsoft and it tells me that the 02 sensor readings are odd. I looked under the car and lo and behold the right pipe was dripping fluid and the joint looks loose. I shall have another closer look at the piping and software later-just wanted to give it a test run you know.

Update
 

The updated connectors at the transmission end of the harness have been installed, and the harness was flushed and blown out. "Quite a bit" of trans fluid was there.

Codes cleared, the car spent an hour hooked up to Star Diagnostics and still has the "RF speed sensor implausible," and a new one "Air flow mass sensor inoperative."

So I have a new air mass sensor (MAF) on order. I was told that this is not a problem that can be fixed by cleaning the sensor.

I think that I will try cleaning the speed sensors to see if that resolves their problem.

The car's behavior has changed since I cleaned out the ETM. Except for when I was making the low-speed u-turn, it has not died. But it has had four incidents where it started to die, then lurched and caught itself. Nobody is sure at this point what's going on, but the thinking is:

1. The speed sensor code could be confusing the engine module
2. The MAF could be sending a signal that makes the engine module want to go lean

At this point, I personally think that there may be more than one problem, the fluid in the trans unit/harness and something else.

I told a friend about that 'wicking UP the cables bit', it just sounds so unreal.

Anyway. I've asked about that part at the dealer and it is ££33.20+VAT, be with me Tuesday.
Now can you look over the attached photo, (BAY) and confirm or correct where applicable;
1 is the Fuel Pressure regulator?
2 is the .........? (talked about in the next photo)
3 is the connector for the hot film?
4 is a temp sensor of sorts?
5 is a combination temp sensor of sorts.
6 is crankcase breather?

As for the other photo there appear to be two devices hidden within my MAF, d'you know what the other is?

Oh, the reason I'm looking at the temp sensors is because the PC says my coolant temp is -40 degs. The bigger combination sensor has already been changed.
I know it won't be what the problem is but I figure it can't be helping matters.

1 is the Fuel Pressure regulator? YES
2 is the .........? (talked about in the next photo) THE CONNECTOR FOR THE INTAKE AIR TEMP SENSOR
3 is the connector for the hot film? YES
4 is a temp sensor of sorts? DUNNO, WILL RESEARCH
5 is a combination temp sensor of sorts. DUNNO, WILL RESEARCH
6 is crankcase breather? YES

As for the other photo there appear to be two devices hidden within my MAF, d'you know what the other is? THE LARGER IS THE AIR MASS SENSOR, THE SMALLER THE AIR TEMP SENSOR.

If the brain is getting a coolant temp of -40, that would drive it pig-rich or at least generate an "implausible" and it would do the best it could with bad input.

Progress, sort of ...
 

This weekend I cleaned all four speed sensors. The RF was pretty grungy with about 1/4" of calcified stuff on the tip. See picture. The RR was not much better. I used the car for all manner of errands this weekend and got only one minor stutter at a stop in ten drive cycles. So this is a step forward.

Today I replaced the MAF sensor unit. Took a couple of test drives, one with the laptop hooked up. All the sensor and fuel trim readings seem fine. No stalling.

Next step is to get the codes read, cleared, drive the car and see if the stalling is gone, and see if any of the codes come back.

In that case, I'll look into my temp sensor readings first, then as you have done I'll take a look at the speed sensors after that, not sure, other logical stuff I suppose.
Thanks and good luck.

Oh, just had a closer look at my carsoft readings. That coolant temp reading of -40 degrees is wrong. It was the oil temp I was looking at, the coolant temp is fine and the car registers it without trouble. So do you know where the oil temp sensor is perchance? Also you say you've changed the MAF, now as you have a 96 with an M119 so your car should be very, very similar to mine.
Tell me, if you switch the ignition on, (turn off blower motor, stereo and such), can you hear the MAF buzzing?
Kind of like a louder version of the interior blower aspirator in the headlining but at about 3 times the volume.

An update. I unplugged the MAF and the car has not cut out since. Though it doesn't run quite as well as usual.

Maybe out of the woods?
 

I now have a weeks worth of commuting and errands on the car, and no issues with stalling, and the car is running super. I am hopeful that fixing the three issues has solved the problem(s).

My indie adviser thinks it was the MAF all along, but that doesn't explain how the problem got noticeably better in three distinct steps.

Gumsie - I have no clue what happened when you unplugged the MAF. Did you ever find out what was dripping from the exhaust?

Not yet. I'm just going to run it for a few days sans MAF to see if the problem returns. My problem is space, I don't have a garage to take a real good look at the car. So I'll replace the MAF first and then tackle the exhaust. It's probably just a very small leak from the joint.

An antedotal story: A late model MB "E" drove in to his shop and it was running rough with a poor idling, etc. Pat said to me, ". .watch what happens when I disconnect the MAF". The car became smooth as glass.

One new MAF later, and all was well!

Glad you've got your car problems resolved.

Now I've got to find a place in the UK that sells a good, (Bosch), MAF for less than £160.00

Jim, I've been watching your page for some time now, (and once sent you an email regarding my next point), you do a chip for your car but never did one for the 96 on models. I'm assuming nothing has changed in that respect?

Epilogue
 

It's now been a couple of months since I put the new MAF on, and the car has not died, surged, or otherwise misbehaved. I am not sure exactly what fixed the problem. Here is a summary:

- No change after new crank sensor.

- Fuel pressure was fine.

- After the fluid in the ETM was cleaned out, it never died again, but it did surge.

- Cleaning the harness and installing the updated plug and seal at the transmission end had no effect

- Cleaning the four speed sensors had an unknown effect. No surging, but the amount of driving was very limited.

- After installing the new MAF, no stalling/surging problems have occurred. Power, smoothness and gas mileage seem better.

You know I'm 99.99% recurring that the MAF is my problem too. Since unplugging it I've had no stalling issues. The reason I've not replaced it yet is because I want a genuine one but don't want to deal with the outlay. Just can't bring myself to spend that amount of money on a plastic tube with a glorified thermometer in it.