wae

I've got a 2012 GL350 that is new to me. Everything seems okay with it with no obvious fluid leaks or anything, however I do suspect an exhaust leak. I'm looking for a little guidance on where to be digging for a solution.

Occasionally, I will get some smoke under the hood that appears to be coming from basically the middle of the vee of the engine. I believe I'm looking at the fuel filter square in the middle and it's coming from around there somewhere. It's an acrid smoke that doesn't smell like oil, glycol, fuel, or plastic and burns my eyes. It only does it every once in a while and there's not enough of it to be visible unless the wagon's at a stop. Sometimes it's just a little wish, sometimes it's actually pretty thick. My assumption is that it's the DEF burning off, but I don't fully understand how that system works.

The other issue is that at idle there is a distinct diesel smell outside and inside the vehicle. It's more pronounced when I'm running the AC. The smell is basically the same as standing on the tarmac at the airport. There's nothing wet that I can see and my fuel mileage is what I would expect, so probably not a diesel leak?

My guess here is that there's an exhaust leak causing fumes and when the def cycles in, the smoke escapes the same leak. Am I on the right path? Any common spot I should be looking in? I'm a fairly decent mechanic, but I've never wrenched a diesel before!

​​

alx

it could be many things.

diesel puffing smoke on occasion can be the dpf regen cycle. if you have a leak (usually after the dpf) you will get what you are describing. there is no def burn off. it is the dpf burning off as fuel injectors run rich and fuel is accumulated in the dpf where it burns to bring the temps to about 1000 f.

raise vehicle, stuff (partially) some rags in both exhausts and have a helper gas the truck slightly. observe exhaust for leaks. if none, look under the hood. the pipe behind the turbo and the egr valve gasket are prime suspects.

007_e350

spot on!!

In bimmers the symptoms you're describing is usually the EGR Cooler leak (Its basically after EGR where coolant runs through a duct) - I dont know if our cars have it

wae

Thanks for the direction and the clarification on how the burn-off cycle works. I've got plenty to learn!

To make it easier to find the leak, is there anything that can sucked through the intake tract that will cause some smoke without damaging anything? To do this type of check on a gas engine without a catalyst, I've had it suck in a bunch of Seafoam which smokes up the exhaust really heavily so it's easy to see the leak. If that's just too much of a risk, I may be able to just find it by feeling for the pressure. There's a bunch of soot on the engine cover right in the general vicinity of the turbo, so I can start there and feel my way around.

alx

replace your two intake seals (the orange ones). check egr valve gasket. check pipe behind turbo. it is most likely your pipe.

wae

I planned to dig in to this some more in a little bit since it's almost due for fresh oil, air filters, and a fuel filter. But "It Happened" today while I was towing the race car. It's an open deck trailer with a total weight with car and tires of right about 4,500 pounds so well within the parameters. I had one adult passenger in the wagon with me as well as an additional 200ish pounds of tools and gear. Definitely not overloaded. On the way there, I had no problem getting up to speed and cruising on the highway, but as we approached the site where we were racing, I started crawling some hills and despite having the pedal completely depressed, we nearly came to a complete stop and were barely able to climb to the top of the hill. That same hill gave my 5.4-powered Ford E-150 pulling the same load a little bit of trouble, but never anything like that.

All the way home, it did the exact same thing, struggling to get to 70mph on the highway. Despite the pedal being floored, the "realtime MPG" display doesn't dip down to the 0 mark like I have previously seemed. With the trailer dropped off, the wagon is still super sluggish. It basically acts as though I have a block of wood under the accelerator. The most boost I could build was about 2.2 PSI.

When I restarted it for the trip home, the MIL tripped and gave me the codes P200A, P2006, and P2007. I'm assuming that the P200A code is something that I'd need STAR for, but for the other two I find it a little hard to believe that both intake manifold runners would fail at the exact same time, right?

It did a very similar thing to me once before: I had hooked up the trailer with a engine-less hull of a 3rd gen F-body to go to the scrap yard and as I went up the hill to exit the storage garage area it bogged waaaaaaaay down and just barely creeped up the hill. After that, though, it was just fine pulling the load around town. I got no codes thrown from that incident.

My somewhat uneducated guess is that I'm looking for some common component that would feed both runners, right? Is there a single actuator motor or device that could be malfunctioning? Would it have anything at all to do with the likely exhaust leak or am I chasing two separate problems? Am I doomed?

clvincent

it sounds like your truck is in limp mode
so something is wrong and the truck is trying to stop you from damaging it more

there are a lot of things that can kick the truck into limp mode
guessing you need a star scan

sure alx will give more info

wae

Well, I pulled the air filters and their associated tubes and pipes out of the way, and found at least one problem: The orange seal that goes between the turbo inlet and the intake tube was crushed, pretty much like how the WIS document says it can be if installed incorrectly. So, there's that.

There is also a good coating of oil in the piping that appears to be entering from the PCV system. Some of that oil dribbled out and is on the wiring cover that runs under the turbo inlet. I'm not sure if that's causal or even related to the problems I'm having, but it is definitely something that I'll need to fix!

I agree that the wagon is going in to limp mode most likely. Right when the vehicle was just coming to a stop as I was going uphill with the pedal fully depressed, it acted as though it downshifted and powered up the hill - not like normal, but it did make it. I had already commanded a downshift to D1 and unless my counting was off, I'm nearly certain it was there, so that seems like some sort of electronic limiter for sure.

clvincent

I have a CTS monitor and transmission temperature gauge on my GL and can tell you you need to be very careful driving or towing in limp mode
don't push it

The temps went up quite a bit the last time I had a light 6 x 12 trailer behind me and went into limp mode because of my differential pressure sensor

wae

Good advice! I think I'll move my transmission service up a few thousand miles and do it with the 100k service.

As a quick tempafix, I went ahead and unplugged the swirl motor and put a 10k resistor in the wiring harness. I don't think it's got the computer completely fooled since the MIL came on again, but this time there's no change in drivability and it's making 20+ pounds of boost. The specified 4.7k Ohm resistors have been ordered and when they come in, I'll swap that out for the 10k and see if that completes the charade.

The soot that is in the engine compartment seems to all be towards the front:




You can see that the heat shielding is still shiny and a paper towel rubbed across it comes back mostly white. What I suspect is the EGR valve there in the top-right seems to be pretty well covered.

I'm guessing that this oil wasn't doing my any favors, either, and I'll probably add a catch can to the PCV system:




This is that seal all crushed up with the oil everywhere:



The tube itself is a little malformed at this point and the groove that the little nub on the seal fits in to is smashed together. You can also see the amount of oil in the intake plenum there.

Next steps are:

1. Get the 4.7k Ohm resistor and install it
2. Track down the correct fuel filter
3. Get a replacement intake seal
4. Research the EGR system and common sources of soot/leaks around it
5. Do the 100k service plus do the transmission service and replace that orange seal
6. Add a catch can to the PCV system
7. Check the EGR for sources of leaks
8. Clean up all the soot and oil
9. See what comes back where and how fast

wae

Because I hate stories that don't have an ending, here's how things shook out:

In addition to having these problems, I also developed a problem with the fuel rail pressure sensor. It was sort of intermittent and after it would go back into limp mode, I could sort of wiggle things around and everything would be fine for a while. When I finally got to replacing it, the whole plastic part of the sensor as well as the wiring harness connector were both melted. I had to cut off the connector, adapt a connector that I had laying about from another wiring harness, and splice it in so that I could connect the new pressure sensor.

Yesterday, I finally had the time and the parts to go ahead and pull the turbo off to replace the swirl motor and to replace gaskets and seals. As it turns out all the problems I had were directly related to the exhaust leak. Like the pressure sensor, the swirl motor was melted to the point that one of the sides of the module was completely missing. I had to pick it out of the area between the intake manifolds with tweezers because the melting plastic had just dripped down everywhere.

The exhaust leak was pretty easy to find. I suspect that I'll be the only person to have this particular problem, but on the back of the turbo there are three bolts that connect the turbo to the exhaust system. Yeah, I had zero of those. So that's a pretty big exhaust leak.

It appears that the swirl motor may have already been replaced as the one I pulled out didn't have Mercedes markings on it but did have a Pierburg sticker (lightly scorched). My guess is that someone replaced the swirl motor and either tried to re-use the bolts, didn't torque them properly, or just plain forgot to install them.

clvincent

thanks so much for letting us know what happened

i think that connection does have some issues with exhaust leaks but not usually because someone left all the bolts out!

hope the truck runs trouble free for a long time

wae

Well, as sagas often do, my own saga with this continues!

There's no more smoke, no more acrid odor making eyes water in the cabin, and overall the vehicle is a bit quieter. There seems to be some diesel clatter audible at all temps at lower RPMs, but that goes away at faster revolutions. I'm paranoid of the whole timing chain thing, but I think I'm just hearing normal diesel noises that were masked by all the raw exhaust note I had previously.

But that's the good news.

I still have my check engine light illuminated and the vehicle went into limp mode immediately. The exact code escapes me -- I wrote it down and can't find that scrap of paper now! -- but using the MBII iCarSoft scanner and looking at the CDI system DTCs, it translated to the M55 motor being shorted to ground. In order to take it out of limp mode, I put my 4.7k Ohm resistor back in to fake out the ECU and I'm now just left with the two codes for each bank of intake actuators being stuck open. This, fortunately, does not engage the limp mode, so everything runs fine. But I hate looking at that light!

I have one of two things happening here: First, there is a short somewhere in the wiring harness or second, the replacement swirl motor I got is bad. It's a Peirburg that I got from an eBay seller with good ratings, so while possible I'm ranking that as the 2nd possibility. My most likely culprit is a short in the harness, but I'm having a hard time figuring out how that could be. Help me think through this:

With the resistor in place, I am simply routing pins 2 and 3 back on each other but putting resistance on the line to fool the ECU into thinking that the motor is doing something. According to WIS, the pinout of the harness is:

Pin 1 - Brown wire that goes to the Z6 junction which is a shared ground.
Pin 2 - Red/Yellow wire that goes to the Z7 junction. I am not 100% sure, but I believe this is the shared source voltage for sensors/modules.
Pin 3 - Grey wire that goes to pin #20 on the CDI control unit.

I don't have a wiring diagram for the internals of the M55 motor, but my quasi-edumacated guess on it is that the positive and negative run the motor and the control signal is a PWM signal (guessing here!) that instructs the motor to turn forward or backwards a number of degrees. Maybe it isn't a PWM signal, but I would think it would only need two wires if it was just an on/off or open/closed thing.

In either case, right now, it's sensing resistance on the signal which makes the CDI control unit believe that the motor is doing its thing. Of course, the position sensors aren't tripping so the control unit knows that the flaps are stuck open and it's not happy about that, but it isn't so unhappy as to pull the limp mode alarm.

When the motor is plugged in, it immediately knows that it's shorted to ground. Don't even have to start the engine. In theory, the only connection to ground should be through the motor and its brown wire connection. If the ground wire was bridged to another pin in the harness connector, wouldn't it have the same failure regardless of if the motor was connected or not? If pin 3 was shorted to ground in the harness, I'd be in the same boat, right?

If this was just a normal case of replacing a part like this, my first assumption would be that I just got a bad part. It wouldn't be the first time. But because of the exhaust leak there were massively high underhood temps that literally melted wires, sensors, modules, and connectors. My fuel rail pressure sensor was literally melted to the harness and I had to replace about 3 inches of wiring and put a new connector in place along with the new sensor. Because of all that damage, my first thought is that the motor is fine but something upstream is melted. But I can't think through how I would be getting a "shorted to ground" code when it's plugged in, but not when I'm looping the two center wires back through a resistor.

That and I really don't want to have to take the turbo off again to replace the stupid swirl motor again!

Thoughts?

alx

Take the turbo off and replace the stupid swirl motor. Trust me. It will go much faster now and youb retain your sanity.

piuric

2009 ML320 (W164) diesel, 90K miles.
I have a problem of strong exhaust smell, outside the car.

It does not come from engine bay, it all comes from the exhaust tip and from outside the car.

Exhaust smell is really strong, I can not stay beside the car at idle after 1 minute. There is no smoke only smell.

The only smoke I get is once a day when I accelerate hard. It comes out a black smoke from tail pipes and It "cleans" the motor. Even if I floored again after this first acceleration I get no more smoke untill the next day (I drive 120km per day).

Any suggestion would be appreciated.
Thank you!

piuric

In time, the car runs fine, no MIL, consumption on highways is 8 to 9L/100km. The only problem is this smell...