I know the topic of Check Engine (CE) fault codes has been beaten to death, but I have researched this thing over the past 6 months, checked with several MB mechanics, and no one has a solution that has worked. I can't be the only guy that has had this situation come up.

Here is the problem in a nutshell:
-1990 300te, W124 body style.
-CE light, fault code 17 flashes (Oxygen sensor is shorted to positive or ground)
-O2 sensor was replaced
-Codes cleared by holding button down and turning off ignition
-Codes cleared by disconnecting battery and KE control unit
-Code 17 always returns within a week after it is cleared

The wiring diagram below from Braingears.com confirms the following:
-O2 sensor in bottom right hand corner
-Green signal wire to Control Unit (with grounded shield)
-Power and ground wires (for the heating element) to the MAS control unit

Testing Procedures Completed
Car is off. I unplugged the connectors from underneath the passenger floor mat. Pulled out the Ohm meter and first tested the signal wire and the grounded shield for continuity. None was present indicating to me that the wire and its shield were not shorting against each other. Therefore, signal wire OK. (unless it is heating up and then shorting?)

I next ran continuity tests on the power and ground wires. Meter read .002 Ohms downstream (back to the O2 sensor) and .50 Ohms upstream (towards the MAS control unit). I spoke with Bosch and they indicated that there should be a small amount of resistance downstream in the O2 unit since there is a heater coil inside the sensor. They could not comment on the resistance in the upstream direction.

I was able to trace the power wire back to the MAS unit with a toner. I tested the wire with the MAS unit unplugged and it had no continuity with the other lead running to battery ground. I was not able to follow the ground wire since it disappeared under the dash and the toner test did not indicate where it came through the firewall.

I ran tests indicated by the MB 124 workshop manual with the car running and the O2 sensor is performing perfectly.

My bottom line question is this: Is this .5 Ohms of resistance from the plug connector up to the MAS control unit normal? If someone with an ohmeter is willing to check this measurement I would be most grateful.

The procedure to test is to pull up the bottom padded floor mat on the passenger side. There you will see two connectors. The first is a round single wire that is usually green. The second connector is a rectangle with two wires. One should be Black/Red/White (power) and the second is brown (ground). With the key out of the ignition, unplug this rectangle connector and put the leads of the ohmeter in the section that continues away from the O2 sensor (towards the outside of the car). Do you read any continuity?

Thanks in advance, and happy birthday USA.

Attached Images

O2x.bmp (64.9 KB, 103 views)

You must have voltage on the signal wire to ecu. It's a reference voltage.
Must read :0.5vdc or very close to.
Heater wire carries battery voltage.
Pin #7 on the ecu (N3.) is ground to O2 shield wire.( Brown/White.)

With engine warm, at idle, you should see a swing from 0.25vdc to 0.980vdc as max but ideal should be 0.35vdc to 0.870vdc.

Use a digital voltmeter with 10meg impedence.
If there is no swing of voltage usually the warm idle speed will fluctuate.

Unless c/o% is set to factory, the system seems to go into limp.

How old is the OVPR ?
OVPR sends battery voltage to pin#1 on ecu.

.A coolant temp' sensor reading of under 15c will not allow O2 to activate.
This is the two pin sensor at the rear of cylinder head.(B11/2.)
.

__________________
[http://languageandgrammar.com/2008/01/14/youve-got-problems-not-issues/ ]

"A liberal is someone who feels they owe a great debt to their fellow man, which debt he proposes to pay off with your money."

have you check the lambda duty cycle? i have seen some that have gone to full rich and set his same code. duty cycle is ok if it cycles between 38 and 50%
you need to get a multimeter with a duty cycle option and connect the negative to ground, and positive to pin 3 on the round pin commector on drivers fender. try doing a seark using lambda as your key word, you will find some good info.

Just tested again to make sure. With engine running and wire inserted into plug, I get a reference voltage of .49 dc. When I ran operational driving tests months ago, the results fell in the range of .25 vdc and .98 vdc. Per the MB Service manual, I remember that I simulated a "lean" condition by grounding the signal wire leading to the ECU. The O2 sensor then performed like it should by reading the max .98 vdc. Based on the above, I concluded that the O2 sensor is performing as designed.

OVP relay is probably original (I purchased the car at 37k mi and I am now at 120k). Can I test if voltage to pin 1 is functioning with the ECU unplugged?

Why do you mention the coolant temp sensor? Would this factor into the fault code?

This is normal operation......

Normal operation again.......


You can back probe pin #1 in the ecu plug by first dismantling the black plastic cap and refit female register back in ecu, or you can cut back the loom sleeve and find the correct wire...

If coolant temperature is reported as below 15c, ecu will not allow the O2 to start operating....this is warm up cycle when engine exhaust will not be warm enough to operate the O2.

To test OVPR, pull plug from CIS-e ecu and measure voltage on pin #1 with
ignition on.
Yes? Fuse okay.
No ? Fuse bad...check 10amp fuse.

Pull off OVPR from it's plug...Bridge socket #1 to socket #2...12volts ?

( Pin #2 of OVPR socket is Batt+ supply voltage to ecu pin#1 )

Test pin#1 ecu for 12 vdc..... confirms continuity of Batt+ to ecu from OVPR.




.

__________________
[http://languageandgrammar.com/2008/01/14/youve-got-problems-not-issues/ ]

"A liberal is someone who feels they owe a great debt to their fellow man, which debt he proposes to pay off with your money."

I must be having holiday brain lock. Can't find schematic for CIS-E Control Unit (N3).

I hate to ask, but can you tell me where can I find diagram of Control Unit?

More holiday brain lock.

The schematic I keep coming up with for N3 shows 2 rows of pins (1-13 & 14-25). The unit in the car has 3 rows?

did you see thread #3 above? have you checked that? that is the easiest and from what i have seen in the past in a mercedes repair shop environment that could be your cause. if lambda is full rich or lean, the computer thinks the o2 sensor is shorted to pos or ground and sets a code accordingly. until you check that, you may just be spinning your wheels with all this testing.

Yes, I did see the post above. Started to search the forum for Lambda testing and found way too much info. I guess I was looking for a simple step-by-step. My multimeter does not have a duty cycle option, so I was reading up on how to do the calculations. If you know of a post that is laid out real well on the subject, let me know.

I needed to replace my rear shocks, so I have been out doing that for now. I will re-address the duty cycle when I get the other shock installed.

Thanks.

Look at the OVPR....trace the small bundle of wires coming from it's plug socket...they will disappear into a thick loom ...pull on loom to trace it back to the front ECU...yes ? This is the 1-13/14-25 ecu.



.

__________________
[http://languageandgrammar.com/2008/01/14/youve-got-problems-not-issues/ ]

"A liberal is someone who feels they owe a great debt to their fellow man, which debt he proposes to pay off with your money."

i don't know of any specifically, if you are anywhere near san diego, i would be happy to meet you and do it for you.

Thanks for the offer Carson356. Very kind of you. I live in Orange County, so it's tempting. I will take some time today and research how to perform the duty cycle. We'll see how it goes. After replacing my rear shocks yesterday, I did test the OVP relay and it was functioning perfectly.

In my first attempt to determine the duty cycle, I am following the article at:

http://www.landiss.com/mixture.htm

I have a CA 1990 300te. He indicates the following:

"For California cars, starting in 1988: Press the push button switch on test connector X92 (passenger side, between the two firewalls) for two to four seconds. The LED will flash once, indicating no fault code is stored in the system. Now press the switch again for two to four seconds. The LED will be on continuously, and the injection control is switched over to an on/off ratio output."

I am following the above procedure and every time I depress the button for 2-4 sec, I get 1 flash. I am unable to get the LED to stay on continuously.

How to proceed?

i visit oc quite often. when is a good time for you?

Thanks for the offer, but now that I have invested all this time and $ in tools I want to see this one through myself.

Do you have any idea why I am not able to get the LED into the "continuous ON" mode?

Duty cycle reads fixed 49.9% with key on/engine off. After 2 minutes of warm up, it kicks in and reads normally (alghough a little bit lean). I don't want to start adjusting the mixture till I find out why I can't get the 85% California reading at key on/engine off.

Any suggestions?