Unusual and potentially fatal ABS/ESP Issue

[Bellow]

Oh, there's no question about that. When it does the job it's supposed to do (rather than malfunctioning) I liken the MB ESP to a very large God-like hand grabbing hold of the car and guiding it back on track.

There's been at least two occasions when it's saved my neck. Hard to believe that those early fortwos left the factory without ESP.

However...

An unwarranted and, by definition, unexpected intervention by malfunctioning automation can and does have catastrophic consequences, not least because the driver will tend to fight against the unexpected intervention. This is a well-understood problem in commercial aviation (B737 MAX, anybody?) because the person at the controls is a) taken by surprise at the intervention, b) they take time to assimilate what is happening, and c) in the intervening time take what they believe is corrective action that may actually make the situation worse. In the aviation world that tends to result in a smoking hole ending.

That's what happened with the Roadsters - when they tried to correct.

Returning to automotive automation...

Faults that could result in uncommanded directional control inputs (brake or steering intervention, etc.) should result in functionality of the intervention system being disabled and that disabling being flagged up to the driver by way of a warning. That the OP is experiencing uncommanded inputs without prior warning of a malfunction is seriously worrying.

That the system itself isn't recognising and recording any faults makes me think it is acting on spurious inputs which it views as genuine and I'm guessing a significant mechanical aspect is at play. A G-sensor adrift from its moorings could throw in a G spike, a split or slipping ABS reluctor ring, a detached steering wheel angle sensor, etc.
There's also the (more remote) possibility of induced voltages such as happened with early Porsche 964s when the drivetrain moved on its mountings under hard acceleration and brought the new dual distributor set-up into close proximity to the ABS unit and activated it.

Always pulling to the right suggests it's trying to correct a perceived oversteer event incurred in a left hand turn.

 

[st13phil]

That the system itself isn't recognising and recording any faults makes me think it is acting on spurious inputs which it views as genuine and I'm guessing a significant mechanical aspect is at play. A G-sensor adrift from its moorings could throw in a G spike, a split or slipping ABS reluctor ring, a detached steering wheel angle sensor, etc.

I agree that a mechanical fault is most likely. An electronic fault would normally be detected, while a mechanical fault such as a loose sensor can easily result in a plausible but erroneous input that is then acted upon.

Whatever the cause, the OP really needs to get it diagnosed and corrected before driving the car again.

 

[Teego]

I agree that a mechanical fault is most likely. An electronic fault would normally be detected, while a mechanical fault such as a loose sensor can easily result in a plausible but erroneous input that is then acted upon.

Whatever the cause, the OP really needs to get it diagnosed and corrected before driving the car again.

I just had similar(ish) problems on my CLS55. It turned out to be the infamous SBC pump.

 

[CarFan79]

If you haven't already had the car scanned I would find a mercedes independent with star/xentry and see if there's any codes stored. If that has been done and there's nothing stored or current then a full visual inspection of all brakes/suspension/tyres should be carried out.

 

[bybuys]

Wow guys, that a lot of information to get my head around! Many thanks to you all for your input. I fully realise the risks in driving the car and have had suspension and brakes checked....in fact it passed its MoT mid May and the first event happened 2 days later. My MoT tester who is a mate has had it on the ramp and confirmed that there is nothing obviously amiss.

I have had the codes read and the guy took a piccy of them but unfortunately the last one isn't on the pic...bit of a blow. The codes I have got are
C1022
C1120
C1140
+ another
C1120 is a YAW rate sensor issue I believe. I note that Bellow said he would change out the Lateral G Sensor....is the the YAW sensor?

The problem with changing the YAW or Lateral G sensor is that it doesn't prove to me that the car is any safer to drive. Hopefully it would resolve the problem but the only way to find out is to drive the car. If it is not the root cause of the issue... I will get another intervention in the future. I don't see a way round that.

I will take the advise and change the lateral G if Bellor would be kind enough to confirm it is the YAW sensor or otherwise and I will feed back on progress.
Thanks again

 

[Bellow]

C1120 is a YAW rate sensor issue I believe. I note that Bellow said he would change out the Lateral G Sensor....is the the YAW sensor?

Yes, that is the one I was referring to and more accurately, it is a yaw sensor.

The problem with changing the YAW or Lateral G sensor is that it doesn't prove to me that the car is any safer to drive. Hopefully it would resolve the problem but the only way to find out is to drive the car. If it is not the root cause of the issue... I will get another intervention in the future. I don't see a way round that.

I will take the advise and change the lateral G if Bellor would be kind enough to confirm it is the YAW sensor or otherwise and I will feed back on progress.
Thanks again

I suspected it had broken from its mountings and was rattling about but that it has thrown up a code shows a problem with the unit itself (or connections to it). Unless corrosion, dirty contacts, etc is found then replace the sensor and that should be the end of the problem. Ultimately, the only way you will know it is fixed is by driving it - unavoidable. Any repetition though could be failure in the wiring unless the error code specifically identifies the unit not 'lost connection to..'. If the other codes are relevant resolve them also.
By the sounds, the yaw sensor is throwing up a signal compatible with the car about to spin anti-clockwise and is applying the front OS brake to pull the car out of the (non existent) spin.
I think you have this solved. I hope so - good luck!

 

[DSM10000]

The fact that it passed an MOT and that your "mate" says there is"nothing obviously amiss "doesn't make it safe to drive though.

I wouldn't drive it anywhere except at low speed to a MB specialist ASAP.

 

[bybuys]

Thanks again guys. I found a video on youtube of a guy changing his YAW sensor on an identical vehicle (even the colour) When he is talking he had pretty much the same problem and scary experience as me. He did it 4 months ago and I have communicated with him andhis problem has not returned. Its a bit long winded but worth a watch for anyone who doesn't know where the sensor is on a 2001 CLK. I will change it and feed back in the weeks to come. Thanks again.
YAW sensor change

 

[Petrol Pete]

When you think about it this is a pretty serious design fault . If that unit fails in any way it should stop working altogether and light up the dash like a Christmas tree and stop the car from being driven....they do it for AdBlu.

As they get older this will become more prevalent. Not sure what the answer is , just hope it does not happen to you I suppose...or the guy coming towards you , or driving next to you in an old MB with a dodgy YAW sensor .