2005 C180K MOT Emissions failure - advice sought - help!

[b830ag2]

Hi,

I was hoping that some budding enthusiasts and general more experienced auto engineers out there might be able to offer some advice or thoughts on a tricky little emissions issue I seem to be having!

Having fancied the idea of owning a Mercedes for a number of years (but not having the funds to support that idea), I finally decided to take the plunge and buy an 05 C-Class C180K coupe in August 2015. It was a private sale and had MOT until April 2016. Looked in good condition, had two previous lady owners, 80k on the clock, panoramic roof and genuine reason for sale (new company car). £3000.

Love it. Great car. Not as nippy I was hoping, but handles great, nice interior, and much better than my previous Citroen Xsara!

However, it’s just had its first MOT under my ownership and it failed for emissions. No other issues.

The results I got were;

HC 177ppm
CO 3.93%
Lambda 0.886

I don’t have anything else for the initial test unfortunately, such as CO2, O2, NOx.
The results were also got after the car/CAT was FULLY warmed up after a good run.

The above results were also following a major service, which included an oil flush and change, new plugs, injector clean, and new air filter. Also during the service the mechanic noted, and cleared a P0170 fault code, which has not re-appeared.

Now, I’m new to the intricacies of emissions analysis, but I’m slowly learning!
The data suggests a rich mix(?), and apparently there are many things which could cause this.


I took it to an independent Merc specialist and he has now performed the following work over a period of 10 days or so;

- Thermostat and coolant replacement – No effect on emissions (however he noted a fault and said that if the ECU thinks the engine is cool when it’s not, then it would inject more fuel). £170
- Hose leak (on supercharger) – brought down CO by 0.5 to about 3.5%, £180
- Replaced both lambda sensors – brought down CO to 1.8% £tbd

He also noted that the ECU had a bit of oil in it and cleaned it up. £tbd

So, current emissions, after all that, are;

HC 26ppm – PASS
CO 1.8% - FAIL
Lambda 0.95 – FAIL (just)
CO2 14.4% - GOOD
O2 0.12% - GOOD?

Better, but still a fail.
He is now examining the idea of replacing the CAT. £tbd (>£400?)

All of the work done above does seem to have improved things, at an ever increasing cost!, and I’m just wondering where it’s going to end? If the CAT is good, then I could even be looking at a new ECU at a cost of over £1000!

I’ve done a little bit of research and found a diagram which shows the make-up of exhaust gas for a varying air:fuel ratio (lambda), and I’m struggling slightly with it. The lambda of 0.95 above suggests a slightly rich mix still, which has improved, however it’s not ideal and even if the CAT is found to be faulty, I’m concerned that if I replace it then;

1) it won’t affect the lambda and therefore still fail the MOT (doesn’t a CAT just reduce CO and HC? And therefore NOT affect O2 which is used to calculate lambda?), and
2) I’ll still be running a slightly rich mix, which would run the risk of damaging the new CAT?
3) Perhaps more fundamentally, I’m wondering how lambda is actually calculated? I’m presuming that the O2 in the exhaust gas is measured and lambda calculated on this basis? But there are two O2 sensors, pre- and post-CAT, so does it take an average? Or use just one of the readings? I’m fairly sure they are both working correctly, as they were both just replaced.

In fact, shouldn’t a lambda of less than 1 prompt the ECU to put less fuel in and get closer to 1 automatically?

Can anyone provide any advice? This is doing my head and bank balance no good at all.
As much as I love the car, this is putting me off somewhat and might just have to revert back to a Citroen Xsara!

Many thanks,
Andy

 

[190]

I have no specific experience to help other than to point out that it's possible to get well below 1.8% without any cat at all. In 21 MOT's on my old catless 190e I never had a CO figure that high.

Which suggests something is telling the ECU to provide too rich a mixture. Either that or the ECU itself is faulty. I'm sure a knowledgeable person would know a cheap way of fooling the ECU into reducing the mixture strength. Not a long term fix perhaps but enough to get through the MOT.

 

[Auto-mobile]

Oil in ecu? did they notice oil in lambda sensor to? You have oil in you harness if that's the case.

 

[estate-agent]

There are several things that can cause high emissions as well as items already stated. The low PPM (Parts per million) suggest the engine itself is in good order and running pretty good. If the PPM were higher then it may have suggested perhaps ignition timing is slightly retarded or cam timing incorrect or that there is a high crankcase breather issue.

I have no experience with the C180k engines but most engines nowadays run a PCV (positive crankcase ventilation) system. UK cars are designed to allow any crankcase emissions to return back into the inlet system which can cause the engine to run slightly rich as any unburnt fuel or oil fumes get drawn back into the engine to be re-burnt in the combustion cycle. As your engine is relatively low mileage and the fact it has been serviced probably means that this is unlikely to be a cause of fuel enrichment but maybe worth trying as it is a "free" item to do. If you can find out where the crankcase pipe enters the inlet system you can remove it from the inlet and leave it open so any crankcase emissions will just vent to atmosphere instead of into the engine. If the pipe is connected into the inlet manifold then you will need to blank the inlet manifolds pipe hole otherwise it will allow an air leak into the manifold and make the engine hunt or idle to fast. If the crankcase breather pipe is in your rubber intake pipe or airbox before the throttle body then removing the crankcase breather pipe will make no effect to the engines running speed so can be just left open or block it to stop any chance of debris getting in (highly unlikely). With a car as new as yours it is unlikely that this is causing the emission issue but it is well worth discounting by doing the above, if it does reduce the emissions then you need to investigate this further.

https://en.wikipedia.org/wiki/Crankcase_ventilation_system

Modern car quite often have no problem with ignition timing but if for some reason your cars ignition is retarded then it can cause a slightly rich mixture.

Did you MOT tester try revving the engine quite a lot to help with getting cats hot? Also did the MOT fail it on fast idle emissions?


.


.

 

[b830ag2]

Just another little update. The indie's courtesy car is the same model and he got it back today. Then he swapped his cat onto my car (his only passed mot a couple of weeks ago), and absolutely no difference to emissions! So looks like it's not the cat.
Why is it so damn difficult and costly to diagnose a little CO problem which would have next to zero effect on the environment? So frustrating!
Gonna buy a Nissan micra I think.

 

[BlackC55]

If there is oil in the ECU then it needs a new engine loom and the leak fixed before any else is done. If there is oil at the lambda sensors then new ones won't work unless the oil contamination is removed (new loom and oil leak fixed).
There is a breather and one way valve under the inlet manifold that fails too that would need to be looked at.

 

[BlackC55]

Oh and the indie should know all this as its as common as a cold.

 

[b830ag2]

OK, thanks for that.

He did say that the oil didn't seem that bad in his experience and was unlikely to be a problem at the moment (but would need sorting eventually). I guess he may be wrong though.

It's there anyway to check this categorically? I can't keep chucking money at it trial and error. Wouldn't there be codes present if this was the case? He's not seeing any.

 

[Lyserman]

There are breather pipes that go brittle and crack on the 180k.
They are under the inlet manifold between the blower and the engine block.
I had the same issue with mine and renewing all the old pipes sorted it.

It can be done with the supercharger in place but it is quicker in the long run to get it out of the way.

Get all the oil cleaned out of the loom and the top of the ecu. Check all the ecu pins are ecu plug is clean.
It may need a new loom but cleaning it out may work as well.

Do a search on here, as Olly (BlackC55) says. A very common problem.

Tony

 

[b830ag2]

There are breather pipes that go brittle and crack on the 180k.
They are under the inlet manifold between the blower and the engine block.
I had the same issue with mine and renewing all the old pipes sorted it.

It can be done with the supercharger in place but it is quicker in the long run to get it out of the way.

Get all the oil cleaned out of the loom and the top of the ecu. Check all the ecu pins are ecu plug is clean.
It may need a new loom but cleaning it out may work as well.

Do a search on here, as Olly (BlackC55) says. A very common problem.

Tony

Hi, I believe that this was the pipe he already changed (was leaking). He's also cleaned up the ecu and harness as best as he could. Still no joy. CO stuck at 1.8% and lambda at 0.95. Won't budge. He's thinking only option now is ecu/harness. Big bucks. Although initially he said might be worth refurbing the ecu at another company?

Nightmare this.
Bought the car for 3k. Had it 8 months, runs fine.
Run risk now of having to spend 2k altogether (already cost about 700 so far) to get it back on the road! Wish I'd never bothered 'living the dream' now.

Thanks for all the advice though [everyone].

 

[Auto-mobile]

Clean the ecu, that all you need to do, you won't need to recondition it.

 

[Lyserman]

There is more than one pipe that can leak.
I know from experience.

Has he inspected the pins where the ecu plug connects to it?
One of mine was covered in crud as it was not making good contact so I bought and replaced the pins in the ecu plug.
When mine was playing up we had it on the gas analyser and had the loom covering stripped back.
When the offending connection was touched the CO came back into spec very quickly.

Have you tested the O2 sensors with Star live data?
A brand new genuine MB sensor was defective on mine giving me all sorts of issues whilst I was having the same problem as you with my 180k.
Bought a Bosch one much cheaper and all was well, plus MB refunded my money.

I was lucky enough to be able to all the work myself so there was no labour costs.
It was a frustrating experience but it was sorted in the end.

If you are anywhere near PCS in Horndean then I recommend letting Olly sort it for you.

Tony

 

[b830ag2]

There is more than one pipe that can leak.
I know from experience.

Has he inspected the pins where the ecu plug connects to it?
One of mine was covered in crud as it was not making good contact so I bought and replaced the pins in the ecu plug.
When mine was playing up we had it on the gas analyser and had the loom covering stripped back.
When the offending connection was touched the CO came back into spec very quickly.

Have you tested the O2 sensors with Star live data?
A brand new genuine MB sensor was defective on mine giving me all sorts of issues whilst I was having the same problem as you with my 180k.
Bought a Bosch one much cheaper and all was well, plus MB refunded my money.

I was lucky enough to be able to all the work myself so there was no labour costs.
It was a frustrating experience but it was sorted in the end.

If you are anywhere near PCS in Horndean then I recommend letting Olly sort it for you.

Tony

Hi. Not sure whether he used star live data or not, but he probably did, says he's hooked it up to diagnostics several times. Popped in yesterday for a chat. General consensus it is the ecu. It's a sealed unit so he can't take it apart easily to check for oil. However the pins were contaminated (through the loom as it has an 'external' seal. He's cleaned the pics up as best as he can but there's bound to be a certain level of contamination within the ecu itself. In getting the ecu sent off for refurb. Will be 5-7 working days away. Fingers crossed!

Oh , he also did another smoke test. Said it was leak tight.

 

[b830ag2]

The ECU has now been at BBA for over a week and finally heard today that they have found no fault. Slight trace of oil in component but not causing an issue in their opinion. Apparently, however, they only have sufficient test equipment to test 'half' of the ECU. So even though they found no fault I'm still non the wiser as it wasn't a complete test. Great. This is not funny any more.
Going to try out ecutesting.com I think, to see if they're any better. Anyone used this company before?

 

[b830ag2]

Looks like ecu testing.com can't guarantee a full test either!
Decided to try a reflash of the ecu at Mercedes, and if that doesn't work, a new ecu :-(

 

[markjay]

Have you tried a new ECU loom as BlackC55 suggested?

If BBA found traces of oil in the ECU.... it would have gotten there via the loom?

 

[b830ag2]

Hi, not I haven't. The mechanics at the indie think it's more likely the ecu as they've had loads of contaminated looms apparently and never had to replace one to get emissions down. Also, another thing that they say points to the ecu is that there have not been any error codes on diagnostics. They suspect the ecu is faulty and that because it is faulty that it doesn't realise it's faulty, if you know what I mean. Had a quote for new loom and it's 400 quid just for the loom. Already spent a grand on all this stuff so reluctant to spend another 500 on something that I've been advised is less likely to be the problem. Might have to in the end though, knowing my luck. Bought the car for 3k 9 months ago and looking like I'll be pumping another 2k in now. Wish I'd had a crystal ball...

 

[b830ag2]

FINAL UPDATE!

Well, finally, I am the proud owner of an MOT!!The saga is over {until next year Atleast}.

Opted not to send the ecu away to another place. The indie rang ecu testing.com and they couldn't guarantee a full test either. So went for a reflash at the local dealers. 35+vat for an initial assessment and another 45+vat for the reflash, so just short of 100 quid. Anyway, an update was also available and they did the reflash. The indie picked it up the next day and retested emissions. IT PASSED !! Holy Jackanora!! Speechless. CO 0.0% HC 2PPM, lambda 0.998. Bob on!

Looks like (and i may be wrong), that a few little bits were causing the engine to run rich and the ecu tried to compensate so far and eventually couldn't cope any more. The bits were fixed but the ecu got stuck and couldn't normalise without a reflash.

All in all cost me just short of 1200, but could've been worse as I was preparing to buy a new ecu (with programming would've been another 1200), and now emissions are spot on and most of the system is now new or checked. Because of this I'll be keeping the car as it should in theory be ok for a while now. Won't stop me sweating a bit next year though!

Thanks for all the advice from this forum, most appreciated! I hope that someone else might benefit from this experience in the future.

 

[StMarks]

Well done b80ag2. After all that grief I was beginning to think you may have thrown in the towel.
You stuck with it, and (even retrospectively) made the correct decisions along the line it seems.

 

[PrzemoF]

I'm sorry for reviving an old thread, but you gave me some hope. I have same problem with SLK200, M271 engine from 2004, already replaced half of the engine and now I'm facing ECU replacement. I'll try software update first Wish me luck...

Edit: I feel I broke a record of time between registering and the first post