W124 dying at speed - O2 sensor question

[taaboo]

Friends - 1989 W124 wagon, 220K, well maintained, new fuel relay and pumps/filter (relay had failed) and OVP (precautionary) recently.

Car is dying at speed and once or twice at idle. Like it's losing spark (e.g. different than when fuel pump relay would cut out). Sometimes if you coast it will 'catch' and you keep going, more often I pull over, turn off ignition, sit for 10-30 seconds, and it starts right back up and off we go. Averaging 1-3 times per day of this happening.

Car has original O2 sensor, check engine light comes on then goes off the past 2-3 months, I have a replacement but haven't had/made the time to put it in. Can a failing/failed O2 sensor cause the above? Based on my experience I'm thinking either the coil or crank/flywheel sensor. Where would you start?

THANKS many times for any advice.

Brian

[Duke2.6]

No I don't think a failed O2 sensor will cause these symptoms. An intermittent problem in the ignition system is a more likely culprit and the coil and crankshaft position sensor would be prime suspects.

A problem like this is very difficult to troubleshoot because by the time you get some test equipment hooked up it goes away.

Duke

[taaboo]

Thanks, that's what I've been gathering since I've read some more posts. There is no CPS listed on FastLane, is it the reference sensor located at the crank or the one on the flywheel - IIRC one is connected to the diagnostic port.

THANKS

Brian

[kramlavud]

Don't know if 89 MB is similar, but replacing the Hall Effect transistor in an 89 Volvo stopped the same problem.
Mark

[Duke2.6]

Originally Posted by taaboo

Thanks, that's what I've been gathering since I've read some more posts. There is no CPS listed on FastLane, is it the reference sensor located at the crank or the one on the flywheel - IIRC one is connected to the diagnostic port.

THANKS

Brian

The crank position sensor, reference sensor - whatever it's variously called located at the left rear of the block that reacts to the flywheel segments is the one that is part of the ignition system.

The TDC sensor at the front of the engine serves no purpose whatsoever for engine operation. The lead terminates at the diagnostic socket. Its signal is purely for some piece of test equipment. I suspect that since it provides a TDC signal it is a reference so the test set can read out spark advance.

Any permanent magnetic device that provides a voltage signal will slowly lose magnetic flux over its life and eventually will not generate enough voltage to operate the system, so even if the sensor doesn't fail in some way, it will eventually effectively "wear out" electrically.

I recall from reading service information that there is a voltage spec range at a given engine speed, and if the sensor doesn't generate voltage within that range it should be replaced. One would need a sensitive inductive pickup to read the voltage signal as I recall it is not more than a few volts.

Duke

[JimF]

Originally Posted by Duke2.6

The crank position sensor, reference sensor - whatever it's variously called located at the left rear of the block that reacts to the flywheel segments is the one that is part of the ignition system.

Any permanent magnetic device that provides a voltage signal will slowly lose magnetic flux over its life and eventually will not generate enough voltage to operate the system, so even if the sensor doesn't fail in some way, it will eventually effectively "wear out" electrically.

Duke

wear out?? Not enough so that you would notice, per what I read. In ten years, it MIGHT lose 1%, not exactly a big loss.

There's three main ways that magnets can degrade:
1) Heat: but it must be taken to its Curie temperature.
2) Demagnetizing field: As strong counter productive field could cause it to lose strength.
3) Shock: The magnet is violently struck to cause it to lose some of it's magnetic properties.

There's lots of heat in the engine but it doesn't get to . . . 1900degs F. . the curie temperature.

The magnet on the fly wheel can have a very slight effect, MAYBE the 1% degradation number.

The 'shock' levels in a car come nowhere close to changing the properties a CPS sensor.

You are adamant about NOT replacing a high mileage O2 sensor, yet you think that a CPS should be replaced. In the eight years that I 'put-in' at my tech's shop, I have NEVER seen a bad CPS.

For Taboo: " . . car has original O2 sensor, check engine light comes on then goes off the past 2-3 months". Check the codes from the DM.

[Duke2.6]

First Law of Thermodynamics: Energy is conserved!

Everytime the flux field is interrupted by the flywheel segments, the pickup produces a small voltage. The input impedance to the EZL is high, but not infinite, so a slight amount or current and thus, energy, is transfered and this will eventually consume the pickup's magnetic flux. "Permanent magnets" aren't really permanent, depending on how they are used.

Whether the reduction in magnetic flux in, say 150K miles of driving, is one percent, ten percent, or fifty percent I don't know, but it may be an issue.

I don't believe in replacing parts unless a reliable test shows that they are faulty, but it sometimes takes fairly expensive test equipment to run the tests that the average DIYer may not own or have access to.

A scope can be used to monitor the output voltage wave form of the 02 sensor, and given the cost of an O2 sensor, I believe they should be tested rather than replaced, apriori. A professional should have the equipment and testing the O2 sensor only takes a few minutes, but my suspicion is that many shops don't bother to look at the O2 sensor output waveform before replacing them. Many shops just want to replace parts rather than doing diagnostics. Even some dealers fall into this category. We here these stories frequently on this board.

Likewise, the Mercedes service literature has a voltage at engine speed spec for the crankshaft position sensor, and a scope should be able to handle this test, either with an inductive pickup on the CPS signal cable or directly connecting the CPS signal cable to the scope. And such a test will only take a few minutes.

I've never tested my CPS because I have no reason to suspect it may be faulty, but I have inspected the output waveform of my O2 sensor.

Duke

[JimF]

Originally Posted by Duke2.6

Everytime the flux field is interrupted by the flywheel segments, the pickup produces a small voltage. The input impedance to the EZL is high, but not infinite, so a slight amount or current and thus, energy, is transfered and this will eventually consume the pickup's magnetic flux.

Like I said, it will degrade about 1%, so it's not bad even after 10 years.

Originally Posted by Duke2.6

A scope can be used to monitor the output voltage wave form of the 02 sensor, and given the cost of an O2 sensor, I believe they should be tested rather than replaced, apriori.

Originally Posted by taboo

1989 W124 wagon, 220K . . . .Car has original O2 sensor

It's the original O2 sensor and if there's EVER a sensor that needs to be replaced, it is that one. As I've said before, Bosch rates them to 60k miles but at 220k miles, it SHOULD be replaced! No questions asked!

Originally Posted by Duke2.6

I've never tested my CPS because I have no reason to suspect it may be faulty, but I have inspected the output waveform of my O2 sensor.

A CPS can be easily tested, at least to a first order, by using an ohmmeter; should measure about 800- 900 ohms. So it does, it probably is good. It's nothing more than a passive magnetic amplifier consisting of a coil of wire around magnetic material. Certainly a scope is the much better test but as I said, I NEVER seen one go bad in normal engine usage especially in the older Mercedes.

[pmckechnie]

If I were working on this car, the first question I would have is: Is the tach working when this problem occurs? If it is, forget about the CPS. If not then it is a possiblity. The other part of this discussion, I have never seen an O2 sensor cause a car not to run. At 220K it should be replaced but not because of this cutting off problem.
Report back about the tach when the failure occurs.

[taaboo]

Thanks everyone, didn't have computer access this weekend.

So we are talking about the reference sensor at the rear of the engine, OK, that helps (the more expensive one of course).

When the car dies at speed (does it at idle sometimes too), the tach definitely show the engine RPM - half the time I can coast and feather the accelerator and it will 'catch' and off we go, the RPM's fall with speed obviously. If I put it in neutral at speed, the tach will fall to zero of course.

I replaced all fuel system components last year, and have not checked the fuel relay, but I think I can rule it out because when the previous relay was failing, it was a very different type of 'engine dying' - removing the fuel made was a much more dramatic 'dying' than what I'm seeing now, basically right now the it just 'loses power' in kind of a smooth way if you know what I mean.

I've got the famous oil leak at the rear of the head, it comes and goes, usually only comes on long trips, which we don't use the car for anymore. I will crawl under there this evening to see if the sensor is all oily.

Tech Forum searches indicate either the ignition module, coil, or maybe the temperature sensor may be a problem. The coil does not feel hot after the drive to work (if that's worth anything diagnostically), and the ignition module is portrayed as a "it either works or it doesn't" type of failure.

Please keep the feedback coming.

THANKS

Brian

[taaboo]

On a DMM I'm getting 0.6VAC when cranking, book says 1.5 but not sure the DMM can be directly compared?

Brian

[JimF]

Originally Posted by taaboo

When the car dies at speed (does it at idle sometimes too), the tach definitely show the engine RPM - half the time I can coast and feather the accelerator and it will 'catch' and off we go, the RPM's fall with speed obviously. If I put it in neutral at speed, the tach will fall to zero of course.

Tech Forum searches indicate either the ignition module, coil, or maybe the temperature sensor may be a problem. The coil does not feel hot after the drive to work (if that's worth anything diagnostically), and the ignition module is portrayed as a "it either works or it doesn't" type of failure.
Brian

that it's something esoteric but in my experience, it is an intermittant coil due break inside. It can vary w/ temperature usually but it could be due to vibration, etc..

I do agree w/ your diagnosis about the ign module, so not sure where that leaves you . . .

[carymcooper]

Taboo:

I recently had an ignition module on my 300CE fail. It may not be related, but I experienced several "loss of power" episodes prior to the complete failure and have not had any since. I have had ignition modules fail in other cars that were proceded by partial failures.

With respect to testing the crank sensor, I tried to check mine. I got voltage readings of .06+/- volts with the engine spinning by the starter. I was told that it should read closer to 1 volt "at engine speed". I replaced mine to no avail - it was the ignition module.

There are not many inputs to the module that cause no spark.

The good news is that the module for your car should be available and relatively inexpensive if that is what you need (mine was $2700+ from MB new, $1250 or so rebuilt, and for $975 could be rebuilt - very luckily I found a good used one for much less). If you need it,you should be able to find new, rebuilt, or have yours rebuilt or a good used one.

Cary.

[taaboo]

Thanks everyone, I'm going to grab a tech with a scope and check the reference sensor over lunch. JimF, I'll test the coil if the reference sensor checks out OK, may be worth replacing at 220K anyway. Cary, I appreciate your humor about the inexpensive ignition module, I'll be calling some salvage yards!

Brian

[JimF]

Originally Posted by taaboo

Thanks everyone, I'm going to grab a tech with a scope and check the reference sensor over lunch. JimF, I'll test the coil if the reference sensor checks out OK, may be worth replacing at 220K anyway. Cary, I appreciate your humor about the inexpensive ignition module, I'll be calling some salvage yards!

Brian

my page, you will see that the DI Module (Ign Mod) was the cause for the existance of the complete site!

My DI, out of the blue, turned on the CE light; the code was "Crankshaft Sensor (CPS) bad or faulty". A local dealer said ". . bring your check book"! That's was bad enough but after meeting with the techs including the chief, they had no more idea of what it may be than I did. A new one at that time was $1100.

That was the birth of my page: I fixed it by a junk yard part from NYC for $225.00. That was five years ago and it's still in the car.

The DI worked PERFECTLY; ie it was a functional DI as far as producing output to drive the coils, controlling fuel mapping, etc. But a bad MEMORY location in the module that caused the CE light.

If you can find a friend with a same DI, it's a five minute job to wire it in for test; ie 'overboard' test. You also can remove your DI and put it in the other MB for test.