Klaus Jr.

Hello everyone and many thanks in advance as new member but longtime reader and appreciate any input or feedback.

I have a 2002 ML 320 Sport, she just turned 108,000. Bought with 88,000 miles with full MB service record and upkeep up to that mileage.

Here is my story. I'm in the Bay Area and rolled down to visit my mom in Phoenix, about 1500 miles give or take. While there with ac on max I got a engine coolant light and upon inspection had a small pool under the car. Once under the hood the coolant reservoir shot me in the eye with a pinhole leak. The engine temp gauge never read hot, maybe two widths above dead level and at no time did the reservoir run empty.

As I knew of an excellent MB mechanic in the Palm Springs area I rolled her there and he quickly replaced the coolant reservoir and all was good. By the time I got back up to the top of I-5, car was running beautifully, great power and immediate acceleration and very good gas mileage, about 22.5 running at 80 to 100 the entire way back (91 oct).

The next day the car had more power than anytime I've owned it and I punched it and ran the tach up to 5,000 and then nothing. Accelerator
to the floor and she would slowly build up and accelerate again. And then the day after that driving into the city again suddenly loss of power and acceleration. Twice now and both solved by stopping and re starting the engine.

I've ordered both a new MAF sensor and CPS just because of the mileage and hopefully this will be the easy fix.

Do note the car starts immediately and idles steady and otherwise has been an excellent vehicle, great highway ride.
Anyone have any thoughts? I've read about limp mode but it sounds as if from prior tales that it is connected with poor or no starting.

My local MB mechanic is great but expensive, as said at the top, really enjoy the forum and my thanks in advance for your thoughts.

Klaus Jr.

art_arev

Hi Klaus. Good to have you here. The Crankshaft positioning sensor ( CKP not CPS) has nothing to do with the loss of power/stalling. It is a easy DIY. The MAF sensor maybe, maybe not. The best thing to do when there is a problem is to scan the car & post the codes. If you start buying parts based on guesswork it will be very expensive to maintain/fix your car.

Klaus Jr.

Art

Thank you for your timely and informative response, this forum is a godsend. I will pull the codes and respond back accordingly.

Klaus Jr.

To update, the vehicle spit out a PO 123 code. As described above, has only happen the two times but haven't been on the highway or above 35 mph since then.

My thanks in advance,

Klaus Jr.

Bezalel

I had a similar problem over-revving my ML320 too enthusiastically - the power would cut and the engine would shake a bit alarmingly as it limped along - solved by turning off and starting again. I just assumed it was the automatic gas cut-off at 6300rpm in 2nd gear. It seems okay now, but I've been treating it more gently since!

art_arev

Klaus, Code P0123 does not exist in the ML's. This code might exist in another M/B veh. or it is a generic code. Scan the car with a scanner that can read MB codes.

N_Jay

http://engine-codes.com/p0123_mercedes-benz.html

art_arev

Thank you N JAY.
The link you posted is hardly an authority in MB codes. The link gives a general description for P0123 in general ( not for ML's).
The said code does not exist for the ML. The scanner Klaus used has given a generic code. It is not able to read ML's correct codes. I am posting the latest MB 2,8 DTC codes for the ML. You can see P0123 is not present.

N_Jay

The code may be a generic code, but that would also mean that MB has to adhere to it.


Given his symptoms and the description of the code, it makes perfect sense.


If this code was skipped, it would not be model specific, but more likely engine specific.


Yes, an MB specific reader might give you more information, but that code sends you to check a handful of electronic parts between the throttle peddle and the throttle position sensor.

art_arev

MB did adhere to it only in single model.
W209 CLK .......
0161 M16/6 (Throttle valve actuator) : Actual value potentiometer 1 : Short circuit to positive or open circuit (P0123).
N Jay don't you agree that its in OP's best interest to have the car scanned with MB specific scanner to get proper info?

N_Jay

So are you saying it is "mysteriously" being set and should not?


Or saying it means something else because it is not listed?


Because, if it is a SAE standard code it would be very bad process to misuse it.

art_arev

I am saying it gets set because the scanner is not able to read the actual codes. This code gets set now & then on ML's with generic scanners. It means nothing concerning the ML. After rescanning with a pro level scanner the actual code dose show up.

N_Jay

Maybe you don't quite understand how the codes work.


The cars ECU (or another processor in the car) sets the code in memory, and the reader simply reads it.


The interpretation of the sensor inputs into an error code is done by the CAR not the scanner.

art_arev

I understand your point.....But look at it this way. OBDII is the "universal" language. But some vehicle brands have their own "dialect" which a general scanner will not recognise correctly. The "basic" scanner will read generic codes but not the "dialect" of lets say Benz. That's why you can find scanners for US vehicles/European/Japanese/ French etc. By buying this specific scanners you are buying a device that recognises a particular OBDII "dialect". As simple as that. That's why some scanners cost US$15-00 & some hundreds & thousands.

N_Jay

ODB-2 (ODBII) is a set of standards. Most of the codes are in the standards. (There is the ability for companies to add code, but not reassign them) There is no interpretation as to what those codes mean.
There are multiple standards, CAN being the types that most European cars use.
If your reader does not read CAN codes you get NOTHING from a CAN based car. (not wrong codes)

Mercedes cars recognize there own service system and report a great deal more information, however when possible they also will report errors to a CAN reader.
Any error that can cause excessive emissions MUST be reported using one of the standards. (and light the "Check Engine" light)

You seem to still be thinking that the car send a message that the reader interprets as a "P0123", when in fact that car is specifically sending a "P0123" using the CAN protocol.

As for cost, the readers can do a lot more but a $10 USB/Bluetooth or a $20 cheapo scanner (CAN Compatible) or a multi-thousand dollar service system would all show a "P0123" if that is what the car said.

The MB system, would provide a lot more detail as to why the car's controller thinks that a "P0123" is occurring.

art_arev

And if the P0123 doesn't exist( in ML's)? I am sure you have looked at the PDF I posted.
It means the scanner doesn't recognise the MB code stored & interprets it as P0123 for whatever reason.

N_Jay

1) The reader CAN NOT make up a code based on unreadable data. It dose not "guess".


It simply reports the code AS GENERATED BYU THE CAR.


2) Where on your list does it say (or even imply) that the list is complete and all inclusive?


Maybe you can Google (or even just search this forum) to see just how many times people have asked questions about THIS SPECIFC CODE, on w/X164 as well as many other MBs?

art_arev

Come on..........
This is the header:
All tests of the electrical system of the ME-SFI fuel injection system on engine
112/113 are described in STAR DIAGNOSIS.The following list contains the fault code
description for the corresponding fault code.

This is not my opinion, this doc is a fact.
I can bet non of P0123 codes were read by a MB specific scanner

N_Jay

It says:
"The following list contains the fault
code description for the corresponding
fault code"

Not;
The following list contains ALL the fault
code descriptions for ALL the corresponding
fault codes


But, you have convinced yourself you are right.


There is nothing more to say.

Maj. Dundee

N Jay, the code P0123 does not correspond with an ML model, and as Art said before it might be a generic code for an altogether different manufacturer or another year and model M/B, but not to an ML.

The correct code for an ML, should be P0120 if the model year was an ME 2.0 98-2000. But since the advent of the ME 2.8 ML's, 2001 and above, a new coding system was born which is more precise in breaking down codes.

The M/B list that artrev posted is correct and it clearly shows that P0123 does not exist in an ML and I have also gathered info from Alldata which also shows that code P0123 does not exist for an ML, in any year and model.

The whole object of reading codes is first to know what the problem is and the second and most important is to get the correct code and that is why it must be read by a professional scanner or a M/B scanner.

If the proper scanner was used, the code should have read P20DC, 20DD, etc, as shown below. And in addition, why didn't the scanner display the right code (P0120) instead of the generic code P0123?

http://www.alldatapro.com/alldata/PR...4763/177934310

N_Jay

If someone can explain LOGICALLY how the scanner comes up with a code that is not in the cars system, I would love to hear about it.


My bet is the documentation MB puts out is not 100% complete. Why the car's system chose P0123 instead of P0120, we can guess at, but it was not the scanner, it was the car's decision.


(and it seems to be a common one from the number of threads calling out that code on this site and others)

Klaus Jr.

Thanks for all the responses, a lot of information here and I am trying to find someone locally who can read the car with a better scanner. The local MB guy wants USD110. to plug it in.

Otherwise I did replace the MAF with a Bosch part and the improvement to the vehicle is noticeable - smoother idle and immediate power (prior there was some lag). I also bought the crankshaft position sensor and will attempt a DIY later this week.

My situation is what Bezalel described, a high rev limiter engaging at around 5800 +/- rpm when shifting from 2nd to 3rd.

I will update upon receiving the better scan.

Maj. Dundee

No one is doubting what you posted previously Njay. The software that recognizes problems and the codes that are shown for that problem are built into every ECM, but with a proviso.

Mostly all code readers are softwared for the American and Asian markets, as popularity prevails. These vehicles use standard OBDII codes, e.g. P0123,,but for whatever vain reason, the European market vehicles do not.

This is where the aforementioned proviso comes in. Euro vehicles definitely need a dedicated scanner to read their codes and those other mystery codes that do not have generic codes assigned to them. Look at the list of M/B codes (I only highlighted some in orange) that do not have standard generic codes assigned to them. These are the money makers for M/B vehicles.

So in essence, what I think is that the code reader/scanner that the OP used, was no doubt set up for the American/Asian market, recognized the problem but could not post the correct code, but hunted for the generic code which was P0123.

This might very well be the problem, but a dedicated scanner will, in this case, pin point exactly where the problem lies instead of conducting numerous tests for high and low voltage, etc. and also show other codes which will only show with a dedicated scanner.

N_Jay

The problem is your description is your assertion that the reader "hunted" for a code.


The reader simply asks the car is and codes are stored, and the car responds with the codes.
The reader does not have the ability to "translate" codes it does not "understand" (using your analogy) into other codes.


My guess is either the code is their intentionally, and since it is used EXACTLY AS DESCRIBED IN THE STANDARD, MB felt no need to provide further information. (Most likely reason)
Or;
The Code is there because the car tripped a legacy software routine in the ECU that MB never expected to be tripped. (Unlikely, as the ECU software goes through quite a lot of regression testing during development)

Here is a description of the process.
ODB defines multiple modes to get the data.
Modes[edit]

Mode 3 is the manner (all) ODB readers extract error codes.

As you can see, all the reader is doing is displaying the digits provided to it. (Some readers have internal look up tables to display additional information, such as the code's standard meaning in English)


The MB specific reader may have processes that when receiving an error code re-query the car for additional data using other modes and then display a summary or combination of that data hiding the actual Mode 3 error code from the technician. (My guess at this point)

Maj. Dundee

I'm sorry I even got involved.