The dreaded engine fault light came on, although the car seemed to drive ok. The DPF fault light was out (Phew!)

Out came VAG-com. It told me P0473 Exhaust pressure sensor (G450) signal was to high. Clearing the fault code worked for a while then the fault came back. Now I know zilch about diagnostics (but learning) and zilch about wifeys 2007 EOS (but also learning). Now I'm pretty expert at electronics so here is some food for all.

First thing I learned was what the exhaust sensor was - in my language it's a differential pressure sensor, sitting either side of a filter in the exhaust (called a DPF) which is mega expensive and a pain in the butt!

First thing I did was put my water gauge (manometer) on each of the 2 tubes normally connected to the pressure sensor (see photo). I expected small but low pressure one side (the left larger bore hose), and a lot less on the right smaller bore hose, unless somebody had stuck a potatoe on the tail pipe! You can easily make a gauge with a U tube of pvc hose, water inside and colored with food dye.

At around 850 rpm idle the high side pressure was 2"WG (0.07psi). At 2k rpm the pressure increased to 5-6 inches (0.2psi) and at 3k rpm over 12" WG (0.4psi). Mine is the original DPF so I don't know what readings a new 'clear' DPF would give. Those pressures seemed about what I might expect. On the low side, the readings were all about zero WG. To prove the low side pipe wasn't broken, I stuck my hand over the tail pipes and watched the levels go up.

Next the sensor itself. I suspected the 3 wire sensor was either simple voltage output piezo crystal strain gauge type, or something clever digital and bus based. I poked fine needles into the back end of the sensor connector (see photo) and checked each made continuity to the pins with a multimeter. Then loosely pushed the sensor into the hoses, pushed the connector back and started the motor. When testing voltages on the test needles, TAKE CARE do NOT short them out!!

These were the first set of voltages: Pin 2 center (Brown) is the reference ground pin.

Pin 1 top (Black) +5 volts
Pin 2 center (Brown) 0 volts
Pin 3 bottom (Purple) +5.6 volts.

When the motor was running there was no change in measured voltages. I checked with a 'scope and there were no pulses either.

OK, so rang the Stealer about a spare - surprise, they have 2 on the shelf ! hmm must be a dodgy part.

Back with the new sensor (see photo) and repeated the above.

Pin 1 top (Black) +5 volts
Pin 2 center (Brown) 0 volts
Pin 3 bottom (Purple) +0.48 volts.

Then I pushed a rubber test tube on the high side of the sensor and did a quick blow whilst it was powered (ignition on). Yippee - the voltage increases more positive with pressure. So it is a straight forward differential pressure sensor responding to very low exhaust gas pressures. Zero pressure = +0.048 volt on pin 3, increasing voltage with pressure (usually linear). You can (and should) do the same thing to test the low pressure side. So now I'm confident that the exhaust pressure sensor is working. I also now understand that when VAG-COM says the fault is high voltage, it means the sensor chips have blown up and the output on pin 3 is stuck high at 5 volts, when it should be near zero volts.

I've attached some photos including a labotomy on the failed sensor. Inside there was a lot of squidgy gelatine - never seen that before. If you are curious, the photo shows what I think are the 2 pressure strain gauges near the top left and right corners, plus a load of chip components. Not something you would want to try fixing.

The failed unit at 21k miles was made in USA, the new unit is made in Malaysia. Not sure which is best!

I hope this post will save somebody a lot of money and help confirm an exhaust sensor is actually faulty, before shelling out cash and finding something else wrong. If the site can't upload my 5 pics, you'll have to work it all out from boring text.

I have something to post about the DPF too - more to come.

Have a nice day - Vox :)

That squidgy stuff is used to help dissipate heat and keep out moisture.

That squidgy stuff is used to help dissipate heat and keep out moisture.

I hope they are not using the same stuff in the new sensors coming from Malaysia.

If you look carefully at the pcb photo you will see the 3 pins of the connector and fine wires attached to the printed board on the right. Now it may be the wire on the bottom connector shown, came off during disection, but since then I've had a closer look at their surface mount board.

A couple of electronic chips came away with just a little poke from a toothpick and worse still, the solder on some of the components did not appear to have fully flowed. The topside tracks were coated with what appeared to be an oxide which wouldn't easily tin until it was removed.

I hope the new sensor made in Malaysia is built to a different spec. and will last.

The dreaded warning light has come on after 23,000Km - the engine is running faultlessly and as we only do trips of 100+ Km, a clogged DPF is unlikely and there is no underbody damage to the exhaust system to increase back pressure.

Exhaust pressure differential sensor is the USA-manufactured version 036 906 051A which is identical to the faulty unit which is the subject of this thread.

We are a 2 hour drive from our preferred VW service centre and want to make a once-only visit to have the problem fixed - can someone with access to the current replacement part number for a TDI engine tell me what it is so I can make sure one is in stock at the dealer to reserve before making a warranty service booking.

There are references to replacement parts with numbers 076 906 051B and 076 906 051A in other posts I have read relating to this problem. I suspect these are the same unit with a minor change incorporated hence the 51B code. However, I have a nagging feeling that the A and B codes might refer to different versions for petrol and diesel engines and need to be certain.

Thank you in anticipation - silvershadow

Mine is a 2l TDI. I would have thought the numbers on the pic I posted had everything. I gave my stealer the numbers on the USA one that failed and they located the replacement part (made in Malaysia) from that.

It could be a 'side of the road' fix if you are sure it is the exhaust sensor- engine warning light 'on' could be many things? It's centre at the back with one torx screw holding it. The pain is getting the VW hose clips to release and slide down,then the hoses can be stuck on. Use a small screwdriver to work around the hoses to get them off. I refitted mine with a couple of fuel line screw type hose clips.

Mine is a 2l TDI. I would have thought the numbers on the pic I posted had everything. I gave my stealer the numbers on the USA one that failed and they located the replacement part (made in Malaysia) from that.

It could be a 'side of the road' fix if you are sure it is the exhaust sensor- engine warning light 'on' could be many things? It's centre at the back with one torx screw holding it. The pain is getting the VW hose clips to release and slide down,then the hoses can be stuck on. Use a small screwdriver to work around the hoses to get them off. I refitted mine with a couple of fuel line screw type hose clips.


Thanks for the prompt reply - given the time-wasting that will be involved if I use the wrong part number, I am applying the "to be sure, to be sure" approach. I am reasonably confident the 51B is the current part but need reassurance:). She who must be obeyed will not be pleased if I get this wrong :D.

This is a warranty item so I will not be getting my toolbox out for this one.....

Warranty is a lifesaver for everything.

Mine was used and still had a few months VW factory warranty left. Then had all the roof seals done and new carpets. Since then, no problems.

The dealer wanted me to take out the VW extended warranty, said it was Mondial, as good as factory warranty. What a load of puff. Did some checking and whilst the roof pump motor would be covered, roof leaks and sorting were not.

Decided their offer wasn't worth it so I bought a manual and some diagnostics kit instead.

................................

Decided their offer wasn't worth it so I bought a manual and some diagnostics kit instead.


Only way to go IMHO - works extremely well if you are disciplined enough to save the money you would have paid for the extended warranty and then use it to pay for any repairs outside scheduled servicing.

The insurer expects to make a profit on the extended warranty and it is highly likely you will too if you self-insure.

Is the warning light what looks like a little yellow engine if so mine came on friday :(

Is the warning light what looks like a little yellow engine if so mine came on friday :(


Yes - that's the one.

Are you experiencing any rough running or unusual engine response?

If not, the problem is most likely a failed exhaust pressure sensor and should be a warranty repair.

Yep no major change really in the driving although i had a bit of a scare as i pulled into the garage when i got a wiff of exhaust fumes turned out to be another car :(

Roof (intermittent fault opening and closing roof) and exhaust sensor both replaced :) did have to pay £45 (plus vat) investigation charge but happy now i know the cars ok :) Now just need to replace the tyres :)

............... did have to pay £45 (plus vat) investigation charge but happy now i know the cars ok...............


How can the dealer get away with this charge for a warranty problem?

Most inappropriate and basically highway robbery IMHO.

Surely the UK consumer protection authorities would consider this charge unreasonable and order the dealer to refund it to you if you challenged it.

Well, partly expected, mine came on today and VCDS scan showed same fault code as voxmagna. Did a fault code re-set and took it for a blast in S mode to keep the revs up so now it's a wait and see what happens next.

What I can't really understand is, if the sensor was faulty in the first post, why did the light go out after a re-set and before the new one was fitted?

Second thing is, what is this warning actually for in this instance? The DPF has its own light so why have this light come on ahead of it?

Third one for voxmagna, is it still running ok since the replacement sensor went on?

Last, did the ECU need any re-programming?

Any further info would be greatly appreciated.

Regards,

John

http://www.volkspage.net/technik/ssp/ssp/SSP_336.pdf

Voxmagna, this will probably interest you. (if it works)

Regards,

John

I went into quite a lot of detail!

The sensor I described is a differential pressure sensor. That puts on the engine management fault light, confirmed by VCDS, NOT the DPF fault light.

I described how I didn't clutch at straws and just change it. I tested the differential pressure first and that I assume is either side of the dpf. Hell, I even blocked the exhaust outlet a few seconds just to make sure the pressure reading I got on both pipes to the sensor were the same and a pipe hadn't fractured.

You can even power it, put a voltmeter on the sensor output which should be near zero, then gently blow on the pressure port and watch the output voltage increase with a multimeter if it is working.

VCDS tells you it is faulty because (I think) the sensor output voltage is stuck at 5 volt logic high, when it should be close to zero with low pressure and a normal dpf. That stuck voltage seemed to me due to a circuit failure (due to heat?) and not something worn out in the sensor.

You should know when you reset diagnostics that all you do is reset the fault counter and it takes I don't know, say 3 fault events for the management light to come on. During this time the sensor could be intermittent, appears to work for a while then comes back on.

I didn't just change the sensor without first doing the tests I posted to make absolutely sure that I had a faulty unit. Please do not regard the VCDS fault reports as gospel truth without making sure another way.

Even though I knew the sensor was faulty, I still measured the differential pressure with a water manometer, to make sure I hadn't got a dpf that was close to being blocked. Because I got such low readings, I didn't just go and blast the guts out of the car to force a dpf regeneration or work some black magic.

I have had no further problems since I replaced the faulty sensor. The sun is shining, the top is down and the car with some attention in other areas I have posted about is running great.

That is an excellent article. I caught my sensor problem very early. I don't know if some miles down the road, the engine management light is followed by a dpf light. Does your VCDS measure the dpf loading? I haven't found it yet, what version are you using?

VAG have recently issued a pre-sales bulleting for the diesel Skodas telling prospective buyers how they should not drive their cars. Whilst I am supportive of cleaner burning engines, many are now considering the wisdom of buying a diesel engine fitted with dpf. Their service life is supposed to be 120K miles, but many who are doing short journeys are getting their dpf clogged up. The replacement dpf cost is far to high and the manufacturers need to bring it down if they want to sell more diesels. While this is happening, there are some 'specialists' starting up doing de-dpf and ECU mods, and most seem to be in the North of England!

I didn't do any ECU re-adaptation. If a wiring or connector fault appears on a sensor for a short time, then it is a sad thing if the ECU has to be re-programmed. As far as my fault was concerned I had the engine management light on for just a few miles, got it back, diagnosed the sensor fault and replaced it. As far as my car is concerned, it should still have the same dpf loadings stored. The sensor doesn't have a CAN module inside it so the diagnnostics won't know if it has been changed.

Hi again,
thanks for quick response. I appreciate how you tackled the job and just wanted a few more answers, didn't mean any offence or to question your methods.

I too will do a full diagnostics before changing anything, I'm trained that way too.

My main interest was how it's going after some miles down the road.

I'm running VCDS 10.6.4. I think you can read DPF loading under Engine building blocks or some name similar to that. I'll have to wait for my son to come home so I've got a sensible passenger on board to do full testing.

For info, VW did a replacement campaign called 26E6 which replaced these sensors but my dealer seems to think it was for Passats only. Not sure how that works if they're all the same thing,,,

Thanks again for sharing,

Regards,

John

Hi John, something I picked up in that article is when a fault occurs in the temperature/pressure/air mass sensor the car runs its dpf management on 'time and distance'. When it then decides the dpf is full it puts on the filter light.

In my case I fixed the problem very quickly, so I hadn't got a dpf with no real dynamic management for that long.

The other thing I learned is if a sensor fails, is left failed until the dpf light comes on, then the sensor is replaced, the dpf would have to go through a regen cycle, even though the sensor was fixed.

VCDS has a download bulletin about forcing the dpf to regen, but the procedure looks scary- keep fire extinguisher handy type of thing!

Well I got into VCDS today and read the pressure diff across the DPF as a low 10 millibars and the DPF at 24% loading if I'm reading it properly, so off to TPS for a new sensor. Pig of a job getting the hoses off but once they shifted it was simple enough. Re-set the fault and all ok now.Took it for a test run and it actually seemed more eager to go which might just be wishful thinking BUT my average MPG which I've recorded over about 3000 miles went very quickly up to 48 after dropping steadily over the last month from 50. Maybe it's been getting un-necessary regen commands from the failing sensor,,,,
I'm wondering if the heat intensity coming off the turbo pack is part of the problem with failures, it's like a furnace once you stop and get the bonnet up. I've read the same stuff on various forums and we're not alone!
Anyway, big thanks to voxmagna for telling us about it, could've meant paying for a dealer's summer holiday otherwise!

Regards,

John

Glad you got it fixed. You will spot in my post that I didn't think much of the hot location where they put it. If I have any more trouble I will move it.

I replace the VW clips with screw type fuel line hose clips. I don't believe in keeping VAG original if I think I need to go back there again.

I'm also pleased you did take a technical approach, because there are too many things that can be changed and get wrong, unless you think it through.

How much do you think you saved over taking it into the VAG dealer as a dumbo?

I'll send you a PM.

Regards - Vox

Where is this sensor located? I have just paid $418.00 in labor to replace an $82 part!
Is it that difficult to access?

Regards - Peter

It is just about the easiest part on the car to replace. However, if the car was driven with a failed sensor for some time, they may have had to do some diagnostic reset work to get the car back to running with the dpf at the point before the sensor failed.

The sensor is located on the top of the engine towards the rear, above the turbo charger. It's clearly identified by the two small pressure hoses and an electrical connector. The sensor is just larger than a matchbox and the fixing screws are easy peasy to get at. I think all I took off was the gray plastic cover over the oil filler which comes off easy. If I have another failure, I shall relocate mine away from the (HOT) manifold and turbo charger.

It's a stupid place to put quite a complex box of electronics containing pressure sensors, that whilst having some temperature compensation, aren't designed for extremely hot environments.

Assuming they knew what they were doing and didn't have to keep consulting the mothership to do any tricky resets, it should be no more than an hours labor and that includes taking off the seat protectors and driving the car back to the lot!

On my 2007 Passat the voltage is ranging from 0.8v at idle to 1.2v at 2000rpm. Would this suggest that my filter is blocked & this is the reason for my engine light being on?

You may be clutching at straws until you do a diagnostics check. Engine light 'on' can be several things or one real thing causing another 'phantom' fault.

Remember what I said way back that driving around with an engine light on with some faults and on not knowing what's wrong, might get the car into 'limp home' mode - that's a pretty slow driving car!

The sensor is a differential type which means an increase in pressure across the DPF will produce an increase in voltage at the sensor output. I don't know what point they set the fault trip pressure or its equivalent voltage at the sensor. Although with a water manometer, a known good sensor off the car connected, and diagnostics running it should be possible to find out. Even then, I think the ECU is quite clever because it is comparing the actual range of DPF back pressures (voltages) with rpm.

It may not be just a simple pressure threshold trip. If diagnostics reports a pressure high and NOT a sensor fault like I had, then I think it should be possible to clear the fault, start the car at low rpm and see if the fault comes back at higher rpm. Diagnostic can also report on the percentage of blockage in the DPF. If I was seeing a high percentage of blocking AND the pressure sensor was not confirmed as faulty, I would suspect the engine management was telling you the right things.

Until you do some proper work with a diagnostics tool, you are in the dark. You may even have a fault code that can be reset, although often this is only temporary. Sorry, but few faults on modern cars are simple to understand and fix without taking a methodical fault finding approach.

Hi

I had to have the exhaust sensor replaced on my 2006 EOS TDI Sport and since the repair was done at an approved VW Dealership/Repair centre I have noticed a drop in my MPG, I'm now getting between 80-90 miles less out of each tank of fuel, has anyone else experienced this?

I had the garage check it when the car went in for its service in January and they say its running as it should be but I'm not happy at losing 80-90 miles per tank of fuel especially with the price on disel these days! :(

I'm not driving any different to how i have always driven so it has to be down to something mechanical on the car, only problem being female blokes at garages do tend to try and fob us off, so if anyone can offer any advice or guidance on what may be wrong or whether anything needs to be adjusted so that I look like I know what I am talking about when I go back, it would be appreciated.

Thanks

#24
I've also noticed a worsening mpg this winter. I have a couple of suggestions: (1) during the cold weather it takes longer for the car to warm-up, (2) you're using additional power for the heater-fan and lights, (3) tyre pressures - I was running my car for a while with the full-load pressures and since adjusting to the correct pressure noticed a 4/5 drop off in mpg.
Just a thought!

Moon Goddess, being from Cheshire you aren't constantly passing through Poynton by chance? That Road works at the junction of the Macclesfield Road and Park Lane will add much to your fuel consumption.

Glad you got the main problem sorted - the exhaust sensor. What did they charge you?

The VW diagnostics is so comprehensive on these cars that I'm sure some of the data the garage can read will tell them if your fuel mapping is average or not. The emission controls are also so tight these days that I would expect those to show up a little worse.

I'm no expert on these car ECU's, but many can learn the way a driver drives a car. For example, lets say I bought a car from a lady who only drove it to the shops every day and never gave it a good sporty blast. Then I buy that car and find it has no 'go' or zip like my friends identical car. On some car models it is just a matter of leaving the battery disconnected for half an hour, reconnecting it and the ECU re-learns a new driving style and fuel map. But on the EOS if you do this, you then have to drive a hundred yards or so for the steering to get re-calibrated and the warning light off.

I can't say if this will work. What does annoy me though is once you take a car into a dealer, they can be doing all kinds of things to the electronics, updating software etc. After only one dealer visit I'm sure that's why my clone radio now shows up a fault code. If my EOS ever goes near a dealer again for anything simple, I'm going to padlock a plate across the diagnostics port or disconnect the CAN wires! I own my car not VW.

Another factor to consider is Winter diesel fuel. It may all come from the same fuel station, but during Winter they put in more additives to reduce 'waxing'. Cold diesel fuel is also less likely to be as efficient. Check your fuel economy on a longish run on say motorways when the weather warms a little. Even being quite hard on top speed cruising I can get around 47mpg.

Well after owning the car for a month (just turned 3 years old) our exhaust pressure sensor light has come on too:(.

Luckily my mate has VCDS so managed to get the fault codes.
009299 & 001137.

Points towards the pressure sensor so bought one today from TPS £39 all in.
Thanks vox for showing us the piccis showing where it's position is.

Will change it over tomorrow and get my mate to clear the fault.

Will keep you posted on the results:o

Changed the sensor and light on dash went out on it's own!

Will still get my mate to erase the fault code in case something else goes wrong:rolleyes: