W211 270CDi - Thick black smoke

[Trigger8]

Hi guys.

I know, you´ve seen this kind of posts a million times, but this is not the usual EGR / injectors / swirl flaps case. Please, if you know a bit about these engines, read through. I did a fair bit of google research, searched through this and other forums, but I haven´t found anyone with the same problem as I have. You guys are my last hope.

As the title says, I have a W211 270CDi 647.961, 130kW . This may be a bit lenghty, but I will describe the whole process what I did and didnt do, so please bear with me. A couple of months ago, I noticed RPM fluctuation when I revved the engine (the same position of gas pedal and the revs fluctuated a bit). I assumed that it can be an injector singing its swan song. So I pulled them all out and I had them inspected. Two of them were actually faulty and were replaced with new ones. Due to very poor mechanic work of previous owner (badly repaired borken injector bolts...) when I put the injectors back in, one of them was not sealing well with the combustion chamber (I was not able to tighten it enough). The engine ran a little rough, but it ran, didnt smoke and the fluctuation of revs was gone. I drove the car in this state just around the block, just a testdrive, I didnt want to further damage the engine. So far, so good. To much of my swearing, I pulled the cylinder head out and I had the head professionally done in a head-workshop (milling, reseating valves, resealing stems, repaired threads for injector bolts, pressure tested). Now the fun part begins. I put it back together with new gaskets and timing chain, started the engine and it SMOKES like crazy. And I dont mean a little smoke, or smoke on acceleration, the thing smokes like an old locomotive . Heavy thick black smoke on idle, on revving, it just smokes like hell. What have I done from now is important.

Changed EGR valve
Chenged MAF
Chenged the fuel rail pressure sensor
Checked the air pressure sensor, even tried to change it from another 270CDi, but both read the exact same values
Swirl valves are taken out (this was done earlier)
Double checked the timing
Tripple checked for air leaks
Pulled the injectors out agian and have them tested again (they are all good)
Inspected every single sensor I could find, to no avail
Ran the computer diagnostics with no fault codes wahtsoever

I found the cause of the black smoke, but not the culprit of the problem. The diagnostics shows, that the total injected ammount of fuel is 2x higher then it should be (computer says that the value should be 8 -12, and it is more like 22). Is there someone here that can point me in the right direction? Does anyone know exactly how does the ECU calculate the injected fuel ammount and what sensors can influence the resault? I feel that Im close, but desperate .

Thanks for all suggestions (apart from those "just burn it down" ones )

 

[mersum1es]

MB engines are pressure controlled, throttle pedal asks pressure, ecu tries to reach that by giving fuel. Is MAP reading in range? MAF reading is range? Is turbo arm moving when revving? Absolutely no air leaks in boost lines? Did you code injectros to ecu (should not cause such difference anyway)?

 

[Trigger8]

Thank you for your answer.

1) MAF reads within the range althouth it is near the limit at idle. I assume this could happen because the swirl flaps are not present, so the engine can theoreticaly inhale a bit more. But still in range. I tried to change the MAF from another 100% functional 270CDi, to no avail.
2) MAP reads within specs. As you can imagine, by now I made a lot of (smoky) test drives and on the last of them, the MAP read "boost pressure too high". But I assume that this could occure because the catalyst is starting to get clogged from the sooth.
3) About the air leaks and turbo arm, it smokes like crazy even at idle, so this should not be the problem. But yes, I tripple checked for pressure air leaks and Im 99% certain that it is tight. But I will check the wastegate actuator if it moves.
4) I did not code the injectors, as I was told that it is not necessary on these engines and the ECU will calibrate itself. And as you mentioned, it shouldnt be that much of a difference. If it blew a bit of smoke on the acceleration, I would say that it is the calibration, but not thick clouds like this one does... even at idle.

Im more of a petrol engine guy and dont know much about these modern diesels. Please elaborate a bit about the pressure controlled engine... What pressure does the throttle pedal ask for? Manifold air pressure? How does this work off load / off boost? Thanx

 

[Trigger8]

Update from my colleague who has done the computer diagnostics: He ran the MAP test through the diagnostic tool. This asks you to settle the engine at multiple RPMs and it checks if the MAP value is in specs. And this test went OK, no errors, everything within specs.

Update #2: the wastegate arm is moving

 

[John Laidlaw]

a few things off the top of my head, as it seems you've covered most everything:-
were the swirl flaps deleted correctly /completely ?
Oil quality/Type?
Did you change injector seals?

 

[mersum1es]

Ok here short description, thanks and courtecy of member 'tuikku' in Finnish MB forum:
Pedal defines wanted power level = boost pressure level. Boost pressure is controlled via map where ecu takes target pressure value by function of engine rpm and injected fuel. This target pressure can be seen by diagnostic, too. Then ecu controls turbo vanes (not waste gate arm) and tries to keep pressure in predefined window. When that is not possible, error code triggers.

edit: I also found some max readings for 270CDI: injected fuel around 75mm^3, max boost pressure 1.4bar, max air flow 1500mg/hub

edit2: check back pressure, that catalysator is flowing free...

 

[Trigger8]

Thanks guys.

John Laidlaw:
1) The swirl flaps were deleted by the previous owner. Its done several thousand (maybe tens of thousands) kilometers since with no issue.
2) Oil is new (obviously), according to specs, MOBIL 5w30
3) Yes, the injector seats were freshly cut and new sealig washers installed. However there was a problem with one of the seals as this clearly was not the first time that the seats were cut and standard washer on one cylinder was not enough. So one of the injectors is installed with two washers on top of each other. But it seems that it seals OK. Anyway, could a leaking injector mess up the ammount of injected fuel?

mersum1es:
Hmm, thats an interesting way of regulating an engine . As I said, Im more of a petrol engine guy and new to these diesels... But since the throttle input on idle is 0, what does affect the fuel ammount at idle? Or at light loads with no boost pressure?

 

[mersum1es]

there is always pressure, as you know nuturally aspired engine will draw vacuum and if turbo vaes are closed, same happens in diesel. I think there are parameters for idling and all other conditions (all included in 'maps' like said - there 'remapping' comes when tuning diesels). I am not expert of ecu maps so can't comment much more

 

[ajlsl600]

You just replaced cyl head. I would start with a valve clearance check compression test .timing check and go from there,being as u say no issues of excess black smoke before.then try unplugging 1injector at a time .

 

[Trigger8]

Thanks guys. OK, I will meassure the compression and we´ll see from there. Yes, essentially I just replaced the head, so there could be a mechanical issue somewhere. The timing is 100% OK, double checked that. I believe that the ECU is also aware of the timing and if its good. It should read an error if it isnt.

 

[joderest]

Don't know if this is any help.
My W163 ML270 ( same engine I think) suddenly started belching black smoke out the back, on idle and trying to get up a hill when it happened, no power, could not see people behind me. Limped to a petrol station and lifted bonnet, found the pipe that runs to Turbo actuator had come off, pushed it back on, started up, black smoke for about 30 secs, then fine and has been ever since.
I know this is not the same, but could it be something to do with turbo, as engine will be pushing fuel in, and if turbo not doing its job, engine will be very rich.

 

[V6Matty]

Don't know if this is any help.
My W163 ML270 ( same engine I think) suddenly started belching black smoke out the back, on idle and trying to get up a hill when it happened, no power, could not see people behind me. Limped to a petrol station and lifted bonnet, found the pipe that runs to Turbo actuator had come off, pushed it back on, started up, black smoke for about 30 secs, then fine and has been ever since.
I know this is not the same, but could it be something to do with turbo, as engine will be pushing fuel in, and if turbo not doing its job, engine will be very rich.

Different engine in the 211, all electronic compared to vacuum operated in the other models. So OM647 compared to OM612.

my only thought is that when recut the new washer aren’t quite high enough in the head now, so mixing isn’t optimal, everything else seems to have been tried. Potential an O2 sensor giving duff information but would normally eventually throw an error code and drop you into limp mode

 

[Trigger8]

Joderest: Thanks for the comment, but as V6Matty said, it is a different engine and I have checked for air pressure leaks several times.

V6Matty: Yes, that seems possible. But would the sub-optimal mixing cause the engine to inject more fuel?

Additional information: The engine seems to be running quite happily at idle, no excess noise or vibration, except for tons of smoke of course. Once it is actually driving, there is not much power (but the turbo is working) and it seems that the engine is trying. By now my strongest suspect is either a mechanical failiure of some sort (loss of compression / bad mixing...) or some kind of a timing issue. If the timing is in fact off, it can inject too soon or too late and compensate for it by increasing the ammount... likewise if the compression is low, it can compensate the ammount of fuel to make it run at desired RPM.

 

[Taffy7hfa]

I appreciate that your confident its not an injector issue but think it would be worth doing a leak off test anyway, If its over fueling as badly as you say I'd be careful as it will clog up the catalytic converter pretty quickly (if it hasn't already)

 

[ajlsl600]

Can that motor not run with an injector upplugged.if so u can identify a dud injector.no ?

 

[Doug1234]

You mentioned you did not code the injectors , get them coded into ecu then proceed , here is why.
My om642 in my Jeep had an injector go bad , just driving along and it threw a fit with running rough and trying to stall and black smoke.
Bonnet up and checked everything was in place and ok and started it up and it ran nice when over about 1200rpm but leave it idling and it poured out white smoke and generally leapt about on its mounts.
So driving is ok albeit a thick cloud of blackish smoke when flooring it (constant smoke for as long as throttle down)
Tick over about ten seconds and it starts stinking of diesel and white smoke everywhere and cannot even see cars behind me.
Raise it to a fast idle around 1000rpm and no more white smoke.

So Jeeps done about 113k and is an 2006 so stands to reason if I chase up and replace the bad one/s then the others that ain't so bad are eventually going to go.
So two brand new injectors and the other four are remanufactured.
So one Saturday morning replace 4,5,6 with new injectors , primed it twice on the key and it fires up in about ten seconds of cranking. Drives better / smoother but still smokes at idle badly.
Sunday replace 1,2,3 and again it starts up as before , but this time its lumpy and has a rough tick over , no more white smoke but its not right and there is quite a haze when its floored , not black but running quite rich.
Number 1 injector had a thick black coating on it and number 3 was similar.
So after a week go and get the injectors coded to the ecu and its instantly smooth running again.
When getting coding up on the screen all injectors showed AAAAEE , I expected to see the original codes from replaced injectors but was told that as soon as you plug in another injector it cancels out the code that was originally there ??.
So moral of my story is my Jeep ran worse when not idling with new injectors not coded , ok it didn't stink of fuel and only left a visible haze in daylight when flooring it but it shook at tick over so bad I avoided traffic until I got it coded.

 

[bladecrazy]

hi, you say the maf was on the limits at tickover, what g did you record, as has been said injector height will make a difference and will cause smoke if not correct, maf should be about 2gs or thereabouts.

 

[ajlsl600]

Generally in my experience.non benz.cept my own... injectors if correct part nu will run ok on stored map.but at best when coded in .
Tho it wud not surprise me if benz have made this a non starter .???

 

[V6Matty]

Joderest: Thanks for the comment, but as V6Matty said, it is a different engine and I have checked for air pressure leaks several times.

V6Matty: Yes, that seems possible. But would the sub-optimal mixing cause the engine to inject more fuel?

Additional information: The engine seems to be running quite happily at idle, no excess noise or vibration, except for tons of smoke of course. Once it is actually driving, there is not much power (but the turbo is working) and it seems that the engine is trying. By now my strongest suspect is either a mechanical failiure of some sort (loss of compression / bad mixing...) or some kind of a timing issue. If the timing is in fact off, it can inject too soon or too late and compensate for it by increasing the ammount... likewise if the compression is low, it can compensate the ammount of fuel to make it run at desired RPM.

As it’s pretty consistently black smoke it would indicate a cold late burn rather than unburnt fuel, it’s hard to say if them not being the right height would be the root cause but that along with not having them coded may just be enough for it to adjust incorrectly.
Whats it like if it gets up to temperature?

 

[John Laidlaw]

And you did mention it was reading overfuelling