TDI 2.0 German engine smoking and stinking

Okay, let's try to clarify things again.
My Audi TDI 2.0 (manufactured in 2007, approximately 180,000 km) started smoking at traffic lights and emitting a foul odor inside the car some time ago. I had just registered it then, but I had bought it much earlier. During the extensive test drive, I didn't notice anything, although it's possible that there was already a defect at that time.

Symptoms:
When the engine is somewhat warm, a strong, unpleasant odor quickly enters the interior of the car while it's idling (at a traffic light), and this smell gradually becomes a persistent fixture, which I believe is partially burned oil. It doesn't smell like the exhaust of an old Volkswagen Beetle, but distinctly like burned oil or something similar, and it's quite different. Sometimes it's more noticeable, sometimes less, but it seems to appear more quickly and strongly with a warm engine. It seems to be largely present only when the engine is idling. While driving, the airflow might distribute it. However, even in the headlights of the car behind, no visible smoke or fumes can be seen. In my previous B5, there was always a noticeable haze, although it was barely visible in bright light. The engine often runs unevenly at idle, and the nature of the unevenness also changes. It may or may not be related.

Questions:
After a prolonged and fruitless troubleshooting process, I have three questions.
a) What causes the smell?
b) Where does he exit?
c) How does it get into the interior?

I'm not a beginner when it comes to troubleshooting. Usually, I can find the solution, but I'm stuck here, so I'm hoping for tips from others who have experienced the same thing. I would like to explain my findings so far.

a): I suspect it's partially or barely burned oil. An invisible but barely perceptible film is accumulating inside, which must be coming from that. I've also considered unburned diesel (due to a faulty injector or a loose high-pressure pump seat), but it definitely doesn't smell like diesel soot, which is completely different. The smell is pungent and acrid on the outside, while inside (possibly due to filtration?), it's more heavy, sweetish, and unpleasant. For now, I'm sticking with the suspicion of "oil." I can rule out an oil leak to the outside. It's theoretically possible for it to be coming from inside the engine, but in that case, it would probably smell more typically burnt (like a "Beetle" engine). The coolant and oil are clean, and the oil is new (the good, expensive Mobil 1 ESP). Only the turbocharger remains: The condition of the storage area on the cold side appears normal, and it's smaller than the one on my other East German car, which has more kilometers and is "cleaner." There's a slight oil film in the lower part of both oil coolers, not a lot, but I expected less, especially considering that after 300 kilometers, it's the same as after 2000 kilometers. The steel pipe in between is practically dry, with only a slight film. The oil coolers may be acting as condensation separators and dripping when the engine is off. The oil consumption is likely between 0.3 and 0.5 liters per 1000 kilometers. With my low mileage (mostly short trips), it's difficult to determine this more precisely. The lower oil hose to the turbocharger is completely clear, so a high pressure due to backflow, as seen in other types, can be ruled out.It could also be an oil leak on the hot side of the charging station. On the other hand, the occasional idling roughness could be caused by oil droplets entering the combustion chamber on the cold side, acting as an additional "fuel."
I was also able to rule out KGE (see other thread).

b): The only visible area that seemed suspicious was the exhaust clamp located below the gear lever. I sealed it with high-temperature silicone some time ago. It's still attached, but the smell is still coming from inside the car, so it's not coming from there. Otherwise, there's nothing else visible. It could also be leaking from the VTG cover plate on the turbocharger. A certain amount of leakage is unavoidable there (it probably only takes a small amount to make it smell bad inside).
Otherwise, even with the hood open and the engine running, you can't really see anything in front of you with a flashlight. I don't want to remove the charger just based on a suspicion, as it's very difficult to access on this engine.

c) It smells quite strongly, even after just 10-15 seconds at the traffic light! Therefore, in my opinion, it must be coming from the ventilation system (although it also happens without the fan running). The only possible opening from below would be the gear lever housing; everything is connected there, and the rubber seal is fine. I haven't found any other openings. However, the fact that I installed an activated carbon filter speaks against it being the ventilation system, as it made no difference. Actually, the cross-beam in the engine compartment should prevent air from entering the ventilation system from the engine compartment.

What remains as a possible solution, what hasn't been ruled out yet?
Conclusion: Oil is entering the exhaust system on the hot side of the compressor (although there are no visible signs of a damaged bearing on the other side), then it's being forced onto the cover plate of the VTG and into the engine compartment (although there's no apparent reason for this), and from there it's making its way into the ventilation system (despite the activated carbon filter, and even without a fan, it happens quickly). Hmm, everything sounds very vague, and I'm not really convinced.
Based on this, I see little reason to remove the charger without concrete evidence. And in the end, it turned out to be a faulty power distribution element (PDE) all along.


I am grateful for any advice. Most other problems with the B7 are well-known - and unfortunately, there are many of them. It seems I've received a particularly defective copy.
If point a) were clarified properly, points b) and c) would largely resolve themselves. But I have described everything in detail to narrow down the cause.

Thank you very much, and Happy New Year!

Andreas.

Hi Heiner,

Based on your description, it sounds like there might be a leak in the exhaust gas recirculation (EGR) valve housing, which is located directly behind the engine and water-cooled. Take a sample of the smell near the open hood, between the engine and the firewall. Before doing this, turn off the air conditioning so the radiator fan doesn't interfere. Remove the underbody panel and check for signs of soot. If necessary, with the engine running, feel around to see if there are any leaks, also check the exhaust manifold, turbocharger, and flexible pipe connections.

Okay, I'll look for that valve. However, I seem to recall (from memory) that German cars (BRD) don't have something like that (?). The amount of AGR (Abgasrückführung) is controlled by a valve above the turbocharger, and I don't have a separate flap in mind right now. Perhaps it's the throttle valve on the intake manifold (the one used to shut off the engine)? It could very well be that the smell is coming from there. But unfortunately, that still doesn't explain the actual problem of why the exhaust smells so bad (see...). (Traffic light stop)

Here are a few more details: it's a B7 Avant A4, and there are no engine-related error codes in the diagnostic memory.
It was previously only a hypothesis that the exhaust fumes smell particularly strong when the vehicle is stationary. It's possible that it only draws inward optimally in that specific situation, and otherwise behaves the same way.
To rule out internal engine defects (as mentioned above, which I had forgotten): to my knowledge, when there are problems with the cylinder head gasket, piston rings, etc., the issue will manifest more noticeably under higher load or engine speed. I can definitely rule that out here.
I had been researching online for hours. While problems with smoke buildup when the engine is idling are mentioned more frequently, in some cases, it seems (at least temporarily) to be resolved by exposing the ventilation for the catalytic converter or replacing the cover (which didn't work for me, as mentioned above). However, in all other cases, the cause is never found. I can't be the only one with this problem.
I've already owned more than 10 Audis, some of which were quite old. But this B7 really has a lot of problems, I'm honestly disappointed.

Is the crankcase ventilation hose okay?

Okay, I've looked into it more closely: the smell seems to be coming from the ventilation system. So far, I haven't noticed anything when the ventilation is set to recirculation mode, but it's not certain yet. What's the point of an (almost new) activated carbon filter then?

Nothing comes out of the AGR valve (I made a mistake the other day; the two valves are located one above the other in this model). I checked there by sniffing and feeling with my hand, so I can rule that out.
@ Roger: Thanks for the tip, but the KGE hose is securely connected. Nothing is coming from there. I even replaced it already. After repeated smell tests, I keep coming to the conclusion that it must be coming from the direction of the charger. It smells slightly more pungent at the front than inside, but that might be due to the filtration.

"However, what would be really interesting are clues about the cause: where is something burning uncontrollably (see above)? Is it engine oil or diesel fuel? Surely, more people must be experiencing this problem..." I am grateful for any suggestions.

Regards,

Andreas.

Hi,

Check if the exhaust temperature sensor and the oxygen sensor are properly sealed.

Best regards, Rainer.

To locate the suspected source of the smoke in the engine compartment, you can try creating a side light by shining a bright flashlight into the area in the dark, as this may reveal rising wisps of smoke.

Hello Rainer and Roger,

@Rainer: The idea is good. "But I need to check it out in better weather to see if I can access it: A4 B7 2.0 TDI = everything is completely blocked (it's all located under the AGR cooler)."
@ Roger: I had already tried several times before, without success. A faint haze is visible from an undefined direction, but it's unclear what it is: I suspect it's coming from the top cover of the VTG coupling.

Since it sometimes runs roughly at idle, could there be a faulty lambda sensor that isn't being detected? For example, if the values are off, but not completely wrong? Or does the engine control unit (ECU) perform a plausibility check, for example, by recalculating values based on data from the mass airflow sensor, fuel injection quantity, etc.?

I speculated earlier that the uneven idle might be caused by oil leaks leading to unwanted fuel injection, but a sensor malfunction is also a possibility. However, the unevenness isn't reproducible like a misfire on three cylinders; it's more of a random, erratic behavior. Generally, it's worse when it's hot, but better when it's cold.

Hello,
After a long time, here's an update on this matter – to help other affected individuals. While I sold the Audi as defective back then, I'm still bothered by this "defeat" without a solution being found. Furthermore, I still have another vehicle with a similar "BRD" (likely referring to a specific component or issue), which now, with 215,000 km on the odometer, occasionally emits a slight odor when idling. However, it produces significantly less emissions than the old one, and (still) without any visible smoke.

Besides the somewhat vague possibility of a crack in the Z-head at the time, I now have a hypothesis that seems much more plausible: the tandem pump. This is a common issue with many TDI engines, although back then I only came across posts about the pump leaking externally or about diesel fuel being transferred into the oil, which raises the oil level. Now, however, I've also found posts mentioning black deposits in the oil filter and, allegedly, larger puddles of oil on the bottom of the fuel tank. The symptoms could also vary depending on the engine type or pump manufacturer. At least, so far, I haven't read a clear description in any post about the pump that mentions it producing a lot of smoke and smelling like oil. And the posts with "smoke" as the main topic still end without a resolution.
So, it appears that there are several vehicles that are introducing oil into the diesel fuel in small droplets. This would certainly explain the smoking. At the same time, it also explains the simultaneous fluctuations in idle speed that were described previously. And the fact that this is only observed when the engine is idling might be due to the fact that this leakage depends on the pressure conditions across the (worn) seals, which in turn are naturally speed-dependent.

Now, I don't have that old car anymore, otherwise I would first check the diesel filter and the fuel tank, and then possibly replace the fuel pump. And with my current Audi, the problem is so minor that I don't want to do anything further about it yet. I will continue to monitor it and get back to you.

Based on my experience, the latent problems and issues that arise after turbocharger failures, often followed by "oil surge," are due to the damage caused by the turbocharger itself. When the turbocharger shaft breaks, oil is drawn in continuously until the engine stalls due to the excessive compression in the combustion chamber.
In this process, one or two connecting rods become slightly bent, which then leads to incomplete combustion of the diesel fuel.

Heiner111 wrote:

... So, it seems there are several vehicles that are introducing oil into the diesel fuel in small droplets. This would certainly explain the smoking. At the same time, this also explains the simultaneous fluctuations in speed that were described back then during idle. ...

Hello Heiner,
I don't believe that oil in diesel fuel (whether in small droplets or even in milliliter quantities) causes smoking and/or DPF (Diesel Particulate Filter) issues. There are many people who intentionally add relatively large amounts of diesel fuel to their tank, claiming that this results in a smoother engine running and protects the fuel injection pump/nozzles.
I am unable to translate "haithamina" because it does not appear to be a word or phrase in the German language. It may be a misspelling, a proper noun, or a term from a specialized field. If you can provide more context or clarify the intended meaning, I would be happy to assist you with the translation.

Thank you for the suggestion. I had also thought about that before; ultimately, it's a matter of the oil concentration. It's undeniable that a significant amount of oil is being burned (due to the smell, etc.). Whether this oil is entering the combustion chamber due to an internal engine issue (e.g., a crack in the cylinder head) or is being mixed with the diesel fuel and injected through the nozzle is something I can't say for sure. However, I don't think it matters much; it all comes down to the amount. If I intentionally mix oil with the diesel, I might be adding 5% or even 10% – and it would be well mixed. But if even a single drop of oil leaks from the fuel pump into the line, it's unlikely to mix properly over the short distance to the injectors. Then, suddenly, I have a drop of almost pure oil in the nozzle, and it takes quite a few strokes to get it out.
The uneven appearance is at least an indication that it is not distributed uniformly, but rather in small amounts (presumably individual droplets).

Hello, I'm writing again about this topic to avoid creating any false expectations. I replaced the tandem pump with a new one in the fall, but even after several tankfuls of fuel, it still smells. And apparently, it's not just when the vehicle is stationary, as I initially thought. So, this isn't the problem where a leaking tandem pump is causing oil to enter the fuel tank. This does happen occasionally, as some reports suggest (oil residue in the tank), but not with the consequences I'm experiencing.
So, it's possible there might be a problem in the cylinder head area. In which case, like with the last Audi, the only option left eventually will be to sell it "as is," meaning with the defect. A new ZK (pump) is uneconomical if it costs more than just replacing the ZK seal.

Does anyone else have any new insights? This should be affecting an increasing number of vehicles with a 2.0-liter engine and 170 horsepower, assuming they are still running. Otherwise, the car is great...

Heiner111 wrote:

Hello, regarding this issue, I'm writing again to avoid creating any false expectations. I replaced the tandem pump with a new one in the fall, but even after several tankfuls of fuel, it still smells. And apparently, not just when the engine is off, as I initially thought. So, this isn't the problem where a leaking tandem pump is causing oil to enter the fuel. This does happen occasionally, as some reports indicate (oil residue in the tank), but not with these consequences. So, it's possible there might be a problem in the cylinder head area. In which case, like with the last Audi, the only option left eventually will be to sell it "as is," meaning with the defect. A new ZK (pump) is uneconomical if it costs more than just replacing the ZK seal. Does anyone else have any new insights? This should be affecting an increasing number of vehicles with a 2.0-liter engine and 170 horsepower, assuming they are still running. Otherwise, the car is great...

Hi,

If the engine is emitting bluish smoke while "stinking," it is likely due to burned motor oil.
It's also possible that an injector might drip instead of spraying. In that case, the exhaust will smell strange, but you won't see anything. In this case, the corresponding cylinder should be clearly identifiable in the idle speed control system.

Are the lambda values for the injectors still within acceptable limits? Have they been cleaned before? Certain PPD injectors were prone to clogging of the injection nozzles.

Best regards, Rainer.

Hello again.

Due to a question regarding my old YouTube video on this topic (https://www.youtube.com/watch?v=ChFNGxr3K9g; where I discussed my previous BRD-TDI), I've added some notes about the oil smell, which I'm summarizing here:
Unfortunately, the new tandem pump didn't make any difference. I think I now know where that comes from.

Take a look online at offers for cracked cylinder heads from 16V engines (e.g., with the identifier BRD). They tend to crack starting from the glow plug hole and likely extend into underlying oil channels. Just hairline cracks, but that's probably enough.

Once the engine warms up, this thin oil can be forced into the combustion chamber. And this "supplemental fuel" also contributes to the idle fluctuations, which are not yet noticeable when the engine is very cold.

"In my current TDI, it's relatively mild and almost not visible, despite having 250,000 km, but you can smell it, especially at traffic lights. It hasn't noticeably worsened in the last 60,000 km. The old one in the video probably had significantly more mileage." They had cheated me and "adjusted" the speedometer, but I couldn't prove it.

Considering some other signs of wear and tear (like the seats, etc.), which are nowhere near as apparent in the current one, I estimate the mileage of the older car to be between 400,000 and 500,000 kilometers. And then it's advanced to the point where it's smoking heavily.

However, it's also somewhat a matter of luck, as the quality of the casting can vary.

However, it still seems that there is no definitive confirmation anywhere online, for example, by cutting open an old Z-head. Customers are being sold new turbochargers, cylinder head covers, etc., but this doesn't actually solve the problem. In principle, it can be welded, but that would require removing the head and disassembling it, machining it, etc. And in the worst-case scenario, the valve seats would also need to be replaced, which would be a huge undertaking.

Or a new head, which would be a complete financial disaster at this age. And all of this without 100% certainty that it was actually the cause (although I am now convinced that it was).

Otherwise, I think these engines are great, and they can also last a long time (provided the oil pump drive has been improved).

Has anyone else gained new insights on this topic in the meantime?

Hi,
It's possible that cracks could be causing oil to leak into the cylinders. To achieve this, an oil channel would naturally need to be present in the corresponding location within the ZK (presumably referring to a specific component).
Cracking (which can be checked using the dye penetrant method) or, at the very least, oil residue, should be visible on the gearbox shafts.
In my opinion, welding is not a viable option for repairing ZK lightweight metal. Besides, such a crack is only accessible from the flat surface, if at all. Changes in the microstructure of the aluminum casting, material stresses, and inhomogeneities introduced into the microstructure would quickly lead to premature failure of the ZK component in operation.
hg
Herbert.