[devilmaycare]

Had the dreaded MIL lamp come on the other day - first time ever in over 7 years. I suppose it's inevitable with so many sensors being monitored.

Hooked up Torque Pro, which came up with the above + "Range/Performance".

This is what the service manual says:

"The Inlet Pressure Sensor provides the ECM with the Barometric/Atmospheric Pressure signal. The ECM compares
the Atmospheric/Barometric signal to the signal of the Exhaust Pressure Sensor during low engine rpm (non-turbocharger
boost conditions) to determine the plausibility of these sensors.
- When Monitored:
Engine speed is below 670 rpm. There are no other Exhaust Pressure Sensor DTC’s. There are no Atmospheric/
Barometric Pressure Sensor DTC’s..
- Set Condition:
The difference between the Exhaust Pressure Sensor Signal and the Atmospheric/Barometric Pressure Signal
is greater than 400 hpa (5.8 psi.) for at least 3.0 seconds."

Now, 400 hpa is a huge difference between two sensors which are supposed to be measuring similar pressures, so I decided to check (with Torque Pro) the air inlet pressure. This came up as 980 hpa (or actually, 98 kpa, as per Torque Pro). I then checked my calibrated indoor barometer at 1000 hpa, so it does appear that the Exhaust Pressure Sensor is at fault.

The service manual contains a great deal of info on electrical diagnostics for this sensor, but I don't know whether I wish to go down this route and especially if it involves crawling under the car. I don't even know where the sensor is located, as the location shown in the manual assumes a single silencer.

I attempted to clear the DTC several times. On each occasion, Torque Pro stated that this couldn't be done, although the MIL lamp did extinguish, only to re-appear on the next start-up. So the fault looks genuine.

Has anyone seen this fault before and/or has any suggestions? My inclination is just to replace the sensor.

 

[jackwalton]

The sensor you need is the one next to the EGR valve (See attached picture)

I would suggest as a first step you remove the sensor and check the hole in the sensor, (Underneath where the pressure is sensed) is clear and not blocked up with crud. Just clear it out with any small screwdriver or drill making sure you dont damage the internals.

The sensor and wiring can easily be checked from under the bonnet.
Its the one I've labelled as EGR Pressure.
If everything seems OK I would check for damaged cables in the area.



Jack

 

[devilmaycare]

Thanks Jack, for the quick response. Since posting this query I have found earlier threads on this topic in which you have provided helpful comments.

It may be just coincidence, but this fault only appeared after I had been poking about on that side of the engine in connection with my on going glow plug issue. At least I now know that the exhaust pressure sensor is easy accessible and not as shown in the diagram that I came across in the service manual.

 

[jackwalton]

Could easily be the wires having rubbed through then as it happened just after you were working in that area.

I assume your aware of the common area where the wires sometimes rub through on ours ?

 

[devilmaycare]

Yes Jack. I was specifically inspecting the cabling on that side of the engine in connection with Cyl 4 glow plug fault and I didn't see any signs of chaffing. However, I will have another look. I'll also check that I didn't do anything silly when removing and replacing the plastic sound insulation. It does seem very suspicious that the exhaust sensor fault appeared immediately after putting the covers back in place.

Just to clarify, EGR Pressure is Exhaust Pressure?

 

[devilmaycare]

Checked the wiring again and everything looked OK, so decided to remove the sensor and inspect. First problem: didn't have a 24mm spanner. Managed to borrow a spanner from my neighbour. Seems a very large nut for such a small component, but the threaded part is much smaller and easy to remove by hand once loosened.

Couldn't find anything obviously wrong with the sensor. The thread was blackened with carbon, but the orifice was clear. Tested the terminals for resistance:

Pin 1 (5V) to Pin 2 (Ground) = 3K Ohms (the signal is tapped from this resistor)
Pin 1 (5V) to Pin 3 (Signal) = 5.5K Ohms
Pin 2 (Ground) to Pin 3 (Signal) = 6K Ohms.

From these measurements, I deduced that the signal output must have about 4K Ohms in series with the tap. Whether this is normal or not, I don't know.

Replaced sensor and connector, started engine (I can't remember whether I tried to clear DTC first), but the MIL lamp was now extinguished and stayed that way until switching off after a couple of minutes. Re-started engine and MIL lamp extinguished. Whoopee, I thought, must have been a connector problem. However, after starting up a third time, the MIL lamp came back on. I have now established that although I cannot clear a fault code with Torque Pro, the process does seem to re-set the MIL lamp for two start-ups only. Very strange.

My next move is to try a new sensor, when I've found a suitable supplier. I did establish that the sensor has a Mercedes part number printed on it: A0061531128, so I might try a local Merc dealership

 

[jackwalton]

I don't understand the "4K in series with the tap" bit and how you've worked that out but, I checked my car a long time ago and have a note that the voltage on the Exhaust Gas Sensor Signal wire was approx 0.62Volts at idle.

Could be worth a check ?

Re MIL lamp re-appearing after 2 start ups. The MIL lamps don't come up on the first detection of a fault. (Unless its a high priority "One Trip" fault). Other Codes are logged as "Pending" or "Matured" and the MIL light only comes up after a few start ups/trips.

Jack

 

[devilmaycare]

One diagram is worth a thousand words:

3K
1-----------/\/\/\/\/\/\/\/\/\----------2
^
|
|
R
|
|
3

Resistance between 1 and 2 is 3K
Resistance between 1 and 3 is 5.5K
Resistance between 2 and 3 is 6K
​

This suggests that the potentiometer tap is near the mid point and that the "unknown" value of R is about 4K. Were a voltage to be measured at Pin 3, then this would be about 1/2 the supply volts, ie 2.5 V. Of course, all these measurements assume that the device is linear, which is probably unlikely as I think it may be a semiconductor formed on a ceramic substrate. However, reversing the polarity of the test meter did not reveal any directional effects.

Having read quite a few posts on different manufacturers forums, it does appear that these sensors fail with alarming regularity and often just outside warranty.

I have reserved a new sensor (the last remaining one in stock) from Mercedes-Benz, Brooklands and will collect it tomorrow. If that doesn't work then it's looking like a dealership job.

Please note that a new sensor has now been fitted and all is OK. The above measurements refer to the faulty sensor.

 

[jackwalton]

See what you meant but it won't be as you first described. Not just a simple potentiometer with a resistor in series with the wiper. Putting a resistor in series would serve no purpose at all as current drawn will be negligible or zero. (High impedance input)

jack

 

[devilmaycare]

You're probably correct Jack, but my diagram is taken straight from the service manual and I've only introduced R=4K to explain the measurements I've taken without power being applied (other than that supplied by the test meter). I've no idea whether I'm dealing with a duff sensor or not. Perhaps I'll find out tomorrow.

 

[devilmaycare]

All sorted! Fitted new sensor this morning, MIL lamp off and DTC cleared. Tried 5 separate start-ups and all OK.

I did query the part number, as it was quite different from the original and the MB service guy invited me round the counter to see all the info and part history. I reckoned that this part had been updated at least 6 times in recent years. I asked what the number 2,127 referred to at the bottom of the screen; this turned out to be the total number of EG sensors sold in the last 12 months. Tells you something.

I also noticed that the new sensor (made in China) looked better engineered than the old one (made in Germany). The sensor orifice looked better finished, nicely chamfered and looked more aerodynamic.

The new MB part number is: A007 153 43 28 and costs £43 + VAT.

Now to sort out Cyl 4 glow plug problem, PO674.

 

[devilmaycare]

Exhaust Differential Pressure Sensor.

During my recent activities concerning the EG Pressure Sensor, I noted that the service manual also describes an "Exhaust Differential Pressure" Sensor. Does anyone know where this is located? Also, I cannot find any reference to a DTC for this sensor.

 

[jackwalton]

During my recent activities concerning the EG Pressure Sensor, I noted that the service manual also describes an "Exhaust Differential Pressure" Sensor. Does anyone know where this is located? Also, I cannot find any reference to a DTC for this sensor.

Cant remember the codes it brings up off hand, but there are 2 pipes running from before and after the Diesel Particulate Filter (DPF) to the sensor. Its the sensor that decides when the DPF needs a regen by measuring the differential pressure across the DPF. (Higher differential, the more its restricted)

Jack

 

[devilmaycare]

Spot on, Jack. There is a group of relevant DTCs, of which, this is one:

P2453-DIESEL PARTICULATE FILTER DIFFERENTIAL PRESSURE SENSOR CIRCUIT
PERFORMANCE.

Would still like to know where the sensor is, though. Would need to find out where the two pipes from the DPF end up.

 

[jackwalton]

Spot on, Jack. There is a group of relevant DTCs, of which, this is one:

P2453-DIESEL PARTICULATE FILTER DIFFERENTIAL PRESSURE SENSOR CIRCUIT
PERFORMANCE.

Would still like to know where the sensor is, though. Would need to find out where the two pipes from the DPF end up.

According to drawings, the sensor is mounted very low down in the bottom of the engine bay at the drivers side. (Approx below the brake master cyl. area)

I assume its easier to find it when the under-tray splash shields are removed.
It doesn't appear to give much trouble. I've never seen any posts on this forum about sensor failure, but if I had DPF problems its the first item I would check.


jack

 

[Terry3333]

Just wondered

Hi great info
Dads just bought one and getting same fault code with pending once cleared ! Did you measure the resistances on the new sensor at all before fitting ?
Terry

One diagram is worth a thousand words:

3K
1-----------/\/\/\/\/\/\/\/\/\----------2
^
|
|
R
|
|
3

Resistance between 1 and 2 is 3K
Resistance between 1 and 3 is 5.5K
Resistance between 2 and 3 is 6K
​

This suggests that the potentiometer tap is near the mid point and that the "unknown" value of R is about 4K. Were a voltage to be measured at Pin 3, then this would be about 1/2 the supply volts, ie 2.5 V. Of course, all these measurements assume that the device is linear, which is probably unlikely as I think it may be a semiconductor formed on a ceramic substrate. However, reversing the polarity of the test meter did not reveal any directional effects.

Having read quite a few posts on different manufacturers forums, it does appear that these sensors fail with alarming regularity and often just outside warranty.

I have reserved a new sensor (the last remaining one in stock) from Mercedes-Benz, Brooklands and will collect it tomorrow. If that doesn't work then it's looking like a dealership job.

Please note that a new sensor has now been fitted and all is OK. The above measurements refer to the faulty sensor.

 

[jackwalton]

Hi Terry,

Which code is he getting cos there are 2 different codes mentioned on these threads with 2 totally different problems.

P0471 or P2453 ?


Jack

 

[Terry3333]

Sorry Jack

He is getting P0471

I cleared it and the scanner then showed Pending for P0471 but the Dash warning stayed off until the next day, maybe a few starts later ?

 

[devilmaycare]

Did you measure the resistances on the new sensor at all before fitting ?
Terry

I did think about doing this, but decided against it for the following reasons:

1. Although extremely unlikely, I didn't want to risk damaging it by applying a test voltage.

2. As I believe the device is semiconductor based, I wouldn't know for sure how to interpret the readings.

My original measurements were mainly to check for open or short circuits.

 

[Terry3333]

Hey no probs, just thought I would ask

Going to remove the sensor in the earlier picture tomorrow for him and see how it looks and maybe clean it if coked up and check resistance against your values ? May give us a clue before buying a new one..

Thanks