Soot and performance issues in AEL engine / Turbocharger pressure profile

Hello everyone.
I'm having a problem with my Audi A6.
It's a 1994 Audi A6 C4 with a 2.5 TDI AEL engine (103 kW), manual transmission, and now has 265,000 km (approximately 165,000 miles) on the odometer. I've owned the car for 60,000 km (approximately 37,000 miles), and the problem hasn't changed since then. The previous owner drove it for about 80,000 km (approximately 50,000 miles) using fuel-oil mixtures with approximately 50% oil. About 60,000 km ago, he replaced all the fuel injectors, including a completely new NHG (noise, heat, and vibration) holder. The injectors were checked again 10,000 km ago, and they were still perfectly fine in terms of opening pressure, spray pattern, and deposits. A heat exchanger, a pre-filter, and a pre-feed pump (with a pressure difference of 0.4 bar before the filter) were installed by the previous owner. Disconnecting the pump and/or the heat exchanger does not result in any changes.

Here's the translation:

"The vehicle starts immediately and has a stable idle. Below 2000 rpm, it feels sluggish, which I suspect is due to the engine and turbocharger tuning, as I've read about this issue before. However, from 2000 to approximately 3000 rpm, it pulls strongly, but it smokes excessively, producing visible plumes of smoke that linger for a long time." Above 3000 rpm, the engine struggles, barely reaching 4000 rpm in fourth gear, for example. However, the black smoke decreases compared to the range between 2000 and 3000 rpm.
The fuel consumption is approximately 7 liters, even with very careful driving. I always try to drive smoothly, whenever possible. There are no differences whatsoever whether it's pure diesel or a diesel-kerosene mixture, roughly in a 50-50 ratio.

The following items have already been checked or completed:

- AGR: The previous owner sealed it with a "metal gasket".

- Checked for air leaks in the turbocharger pressure sensor and related components.

- Fuel supply: Bubble-free in both supply and return lines, filter replaced as a precaution, no vacuum thanks to the pump (measured with a pressure gauge).

- Air supply: Boost pressure measured manually, reaching just under 1 bar (the trend is described further below), air filter replaced, boost pressure hoses checked for leaks as much as possible.

- Read error codes: Unfortunately, my interface has been defective for the past 2 days. However, over the past few weeks, I have cleared the error code 00542: "Implausible signal, needle lift sensor" twice. Unfortunately, I haven't yet read the values and logged them; I still need to learn how that works.

- Measurements at the NHG (likely referring to a specific component): 124 ohms resistance with a warm engine, no changes observed when tapping the injector body while the engine is running. Based on what I've found in the archives, this resistance seems too high. Now, the question arises for me whether the potentially defective NHG (National Health Guarantee) can actually trigger the described errors.

I'll measure the static ignition timing again once I get my dial indicator back, check the timing belt, and if I can find an adapter, I'll also measure the compression. Additionally, I'll remove the intercooler hoses, the intercooler itself, and the intake manifold, and verschandeln everything.
I will also disassemble and test the fuel injectors myself using my testing equipment. Unfortunately, I can't test the NHG (likely referring to a specific component or system) based on its signal.
What's still making me think is the story about the added KAT (likely referring to a specific substance or process) that I read here. Could that also be a possible option?

Here's some more information about the boost pressure curve (everything was measured mechanically using a pressure gauge connected to the hose on the intake manifold, where the boost pressure sensor is also connected):

"When I accelerate from 1000 rpm with full throttle in 4th gear, I reach 1400 rpm at approximately 0.2 bar of boost, 1600 rpm at about 0.3 bar, and from 1800 rpm, the boost quickly rises above 0.5 bar towards 1.0 bar, which is reached just above 2100 rpm. The pressure then increases to about 1.15 bar for perhaps a second, before dropping back to 0.95 bar and staying there. Is this an overshoot that is being regulated back down by the boost limit?" I haven't encountered such aggressive behavior from my other engines (all 1.6 TD engines from VAG with purely mechanically controlled wastegate turbos).
"When I accelerate in 3rd gear, the boost pressure goes up to 1.2 bar, while in 6th gear it only reaches just above 1 bar. This makes sense to me, because in 3rd gear, the pressure increases much faster than in 6th gear (due to the faster increase in RPM in 3rd gear), and so, before the boost control system kicks in, the pressure simply continues to rise. Is this explanation correct?" Then I could focus more on the injection system and would prioritize checking and cleaning the intake system later.

You're welcome for the tips.
Best regards, Matze.

Aloha.
I'm starting to run out of ideas.

So far, I have also tried the following:

- The needle valve lifter has been replaced with a new one, including a new nozzle and a re-magnetized core. The error message regarding the needle lift sensor signal has been cleared for over 1000 km.

- All other fuel injectors were removed and the spray pattern was checked for leaks; everything is okay.

- The intake and intercooler systems were disassembled and cleaned, but there was actually nothing inside.

- AGR - Checked the cover again, nothing is being recirculated from the AGR.

What else could it be due to now?
Here are two more possibilities: The completely clogged exhaust or catalytic converter. Can this explain the symptoms of excessive black smoke and loss of power?
Similarly, a ticking lifter. Could a gas exchange loss caused by a faulty turbocharger be enough to "disable" a cylinder to the point where it only produces soot?

I would really appreciate any tips you can offer.

Best regards, Matze.

Evening.

Okay, I would say, first make sure your diagnostic interface is working correctly. Then, log data block 3 and 11 while accelerating; that will show you what the boost pressure and air mass are doing.

Then, check the start of the subsidy with the timing checker and review document MWB 4. You can watch live what your NHG (presumably a device or system) is measuring.

Also, perform a full-load acceleration test with the MWB 4 to check if the speed limiter (SB) is functioning correctly.

Quote:

Now, the question arises for me whether the potentially defective NHG (National Health Grant) can actually trigger the described errors?

Sure, the NHG (likely an abbreviation for a specific system or sensor) captures the actual value in the injection start control. However, if the signal is detected as defective, the control system is deactivated, and the process continues under alternative control. It can definitely produce a lot of soot if the initial settings are too late.

Quote:

- The needle valve lifter has been replaced with a new one, with a new nozzle and newly magnetized.

What is it that magnetizes ?

Aloha.

Okay, I'm going to try logging some data today, if possible. The problem with the C4 is that it still has the 2+2 connector in the engine compartment; let's see if that works with the laptop...
Does anyone know where I can find the plugs and sockets for the 2+2 system, so I can build an extension cable?

The interface has been new for a long time, otherwise I wouldn't have been able to read the memory .

It's already clear how the NHG (National Health Grant) works and what it's for. The question was whether the symptoms I'm experiencing are consistent with an NHG that has a poor signal. What is probably not the case is that the symptoms were almost identical after the switch.

Here's a brief note about magnetization: During the adjustment of the NBFs at the factory, it is magnetized. This is achieved by applying a constant current of 230 mA for a few seconds. In effect, it is also electrically polarized and, when connected to the control unit and when the injector needle opens, it first provides the positive voltage pulse that the control unit requires.

During the overhaul of NBFs, in addition to installing a new nozzle, the process of remagnetization is repeated to achieve a stable signal.

Best regards, Matze.

Aloha.
I once tried a short logging trip, which only works with the laptop on my lap in the middle of Stuttgart, and the cable is too short for the passenger seat .
However, it was at least enough for a full-throttle acceleration in second gear, and that provides some useful information. Okay, I'll post some results now.

Air mass:
Bitte gib den Text, den du übersetzt haben möchtest.

Boost pressure:
Bitte gib den Text, den du übersetzt haben möchtest.

Spritzstart:
Bitte gib den Text, den du übersetzt haben möchtest.

I think it's clear that the problem lies on the intake air side. But how do I interpret that? Is the mass airflow sensor (MAF) reporting low values, or am I actually getting so little air? And why is so much fuel injected that it produces so much smoke?
And I'm asking myself: Why is the EGR valve being activated at 5.1%, even though the engine is at full load, which means no exhaust gas recirculation is required?

Regarding the boost pressure curve: The increase is delayed compared to the target value. However, this could be an error resulting from an issue with the intake air. The adjustment to the target value seems quite accurate, and the value corresponds well with the additional boost pressure gauge (manometer). The value also remains constant during prolonged operation under load, although this is not clearly visible here. Therefore, I would rule out problems with the boost pressure sensor and the solenoid valve. The question arises again: Why a 5:1 aspect ratio, even if the target pressure has not yet been reached?

Regarding the start of the injection: In my opinion, this looks quite good. However, I'm wondering why we have such a large deviation in the near-idle range (first measurement), even though the other values seem correct. (For additional information: In an idle measurement, this value was also too high, similar to the first measurement here). And why is sufficient spark advance not achieved in the higher RPM range?
I cannot assess the tempo. What does the esteemed audience think about that?

Okay, I have some values here now, and I'm happy to provide more if needed. Especially if I can figure out how to get Excel to recognize the numbers from the log file directly as numbers. Unfortunately, they are separated by periods, and I couldn't get Excel to recognize them as commas.

And the question remains, what could possibly be wrong?
What should be done, looked at, and measured next?

Best regards, Matze.

Quote:

Why is the EGR valve activated with a value of 5.1, even though it's at full load, which shouldn't require exhaust gas recirculation?

5.1 = AGR to

Quote:

Why a 5:1 aspect ratio, even if the target pressure has not yet been reached?

5.1 = max. Boost pressure.

Quote:

Only why we in the near-idle range (1. Measurement) the deviation is so large, even though the others fit here.

The initial settings are adjusted relatively early, and the injection pressure regulator doesn't need to be adjusted much later.

Otherwise, the start of the injection is fine. You can disregard the 0.5° deviation in that measurement point.

Quote:

How do I interpret this? Is the LMM showing values that are too low, or do I really have so little air? And why is so much injected that it produces so much smoke?

So, if you accelerate like it says in the log, will it make a lot of noise?

I would say that your EGR valve is not sealing properly, or that it is stuck open or partially open.

Please log the basic settings for 45 seconds in MWB 3.

EDIT: I just read that the previous owner sealed the EGR valve with a metal plate. I would still check to see if the sheet metal is still sealed and hasn't possibly corroded through.

Otherwise, I have no explanation for the high Russian readings when such a low air mass is measured. (Unless the intercooler hose is disconnected and the mass airflow sensor is defective, or there is mechanical damage to the engine). The SB (likely referring to a sensor or system) is working properly.

Regarding the boost pressure curve. The increase is delayed, compared to the target value. It could, however, be an error that arises from the intake air issue.

Hello,

That's where I would start looking.

I think you have a VTG loader (I'd rather ask, as I've been wrong about things here before...).

Your LD configuration is highly unusual for a VTG loader. In fact, you might already expect a 'regulation difference' error in the LD system. However, the engine seemed to be just about to fail, but it didn't because you hadn't read its error codes.
The engine is reaching full power much too late. Ideally, it should be fully engaged around 1900 rpm. However, in your case, it's not fully engaged until over 3000 rpm (!). Before that, you're just lazily drifting through the smoke barrier.
Overshoots are normal in VTG chargers, unlike bypass chargers.

The LLM also outputs values that don't align with the LD structure. In principle, one should expect a maximum LM value when there's a maximum LD value. Both values are different.

I don't know the smoke curve of your engine, but I can imagine the following scenario:

The LD is not building up -> smoke containment.
At the same time, LMM <False> The smoke limit is operating with inappropriate LMM values.

Together, these factors result in an excessive amount of material with insufficient lubrication, leading to increased friction and wear, while simultaneously producing low exhaust gas volumes.
Eventually, the exhaust gas volume will be sufficient, and the lambda sensor will reach its target value and the system will be regulated.

My suggestion:

Log the injection amounts so we can see what the MS is injecting and where the limits are being set.
Then, replace the LLM with a diode to rule out the possibility that the LLM is providing completely incorrect values.
Now we should be able to see how the turbo behaves.
This component can also be defective, which would require a large exhaust volume to build up the pressure needed for proper operation.

Best regards, Jochen.

Aloha.

Okay, so 5.1 is the lowest clock rate. The only question that arose for me was why bother timing it at all?

Yes, when accelerating, a wall of mist forms and it takes a good 20-30 seconds to dissipate.

The AGR sealing ring has just been inspected and is ready for use this weekend. The lead plating is completely intact. The entire intake system and intercooler system have been thoroughly cleaned and inspected.

I'm starting to run out of ideas.

Regarding mechanical damage, I wanted to measure the compression, but I don't have a suitable adapter for my tester, so I'll probably have to improvise one from an old fuel injector. I'm not entirely convinced that's the problem, though. When driving in a high gear with full throttle from idle, and with the window open, you can hear a smooth combustion on all cylinders, and there are no irregularities. Also, no cylinder misfires during startup. The only thing that doesn't quite fit is a ticking sound that is independent of load, which I suspect is coming from a hydraulic lifter. That's why my question is whether he could be responsible for such a bad filling.

And what about the exhaust system? Can a completely blocked catalytic converter, which has also occurred in A6 AEL models before, cause these symptoms?

Best regards, Matze.

Aloha.

@ Jochen:

The part has a KKK K14 turbocharger with a wastegate, which is why the power increase only starts just before 2000 RPM. The overboost protection is achieved because the boost pressure is not always directly applied to the pressure sensor of the wastegate, as in a purely mechanically controlled engine, but is partially regulated via a valve.

I'm still going to do a few more logging trips, but as I said, it's not that easy with the Audi A6 because of the 2+2 connector in the engine compartment, and the cable is too short. Does anyone know where I can buy connectors and sockets to extend this cable?

Best regards, Matze.

The part has a KKK K14 turbocharger with a wastegate, which is why the power increase only starts just before 2000 RPM. The overboost protection is achieved because the boost pressure is not always directly applied to the pressure diaphragm of the wastegate, as in a purely mechanically controlled engine, but is partially regulated via a timing valve.

Okay, then something still doesn't fit.

Even with the bypass loader, the maximum load should be slightly above 2000 rpm.
Certainly not, and definitely not at 3500 rpm.

I haven't observed any overshoot in my AHU yet.

Best regards, Jochen.

Quote:

This part has a KKK K14 turbocharger with a wastegate, which is why the increase only starts just before 2000. The overboost protection is achieved because the boost pressure is not always directly applied to the pressure sensor of the wastegate, as in a purely mechanically controlled engine, but is partially regulated via a timing valve.

Absolutely correct. The additional cascaded controller (MSG) with an integral component, designed for fast response and good tracking performance, also reduces overshoot in systems with disturbances.

"Only wastage = P-controller, and with that, there will be no more overshooting."

My EGR valve starts to engage at 1900 RPM.

The problem I'm having with this thing is that the air mass is so low, the boost pressure eventually gets quite high, but it still sputters.

He's measuring 600mg/H of air, with a boost pressure of approximately 1900mBar, but even less than 600mg/H must be reaching the cylinders for it to produce black smoke. That could be nicely explained with an open agricultural policy...
Also, because the air mass hardly follows the low-pressure system, and also because of the shift in the maximum... Boost pressure after higher engine speeds would be a perfect addition.

It seems like the only possibilities are some combination of different errors, but I find them quite unlikely (e.g., a faulty mass airflow sensor AND a leak in the intercooler, or a faulty mass airflow sensor AND a hole in the piston, etc.).

Quote:

The AGR sealing ring has just been inspected and is ready for use this weekend. The lead plating is completely intact. The entire intake system and intercooler system have been freshly cleaned and checked.

Could there be a hose in the intake system that is soft and collapses under higher flow rates? This could also lead to inaccurate (overestimated) readings of the mass flow rate due to the change in cross-section and the resulting increase in flow velocity.
Please double-check the air intake system, from the stub to the air filter and then to the compressor.

Quote:

Okay, so 5.1 is the lowest clock rate. The only question that arose for me was why, then, is it even clocked?

The lowest and highest clock rates vary depending on the engine and software.
I have never seen a "0" error, and one possible reason is that it might no longer detect a broken connection at the "0" level.

Regarding MOSFETs, it can be said that the lower the minimum duty cycle, the smaller the gate capacitance.
With my AGR, the lowest duty cycle is 34%, with an APA it's something around 15%. If yours is allowed to go down to 5% (you can imagine how 5% of the LD would affect the WG dose), then that's already a sign that it must be quite a sluggish device.

Quote:

With my AHU, I haven't yet registered any overshoot...

Simply due to the way the controller works, the TV doesn't turn on until the target pressure is reached, meaning that the boost pressure only builds up properly once that target is achieved. Therefore, there *must* be some overshoot. Unless your control systems, including the wastegate, operate at infinite speed.

Or perhaps the system is designed differently for AHU units, which would mean the adjustment process takes longer.

Hi everyone.
There have been developments regarding the soot problem...

In the meantime, I had removed the catalytic converter and discovered that increased back pressure from the exhaust system was NOT the cause of the insufficient gas exchange. The previous owner of the car had also had this idea and had simply completely emptied the catalytic converter. A short drive with the exhaust system completely removed revealed that the problem wasn't related to the catalytic converter or the mufflers.

"Last fall, I took the intake manifold off, but it was surprisingly clean; I suspect the previous owner may have already cleaned it. Cleaning it again didn't make a difference, but it was actually clear from the beginning that it wouldn't. However, I noticed that all the intake passages in the swirl area, just before the valve plate, were extremely coated with sludge." Since I had removed the exhaust manifold on the street back then, I couldn't do anything about it in that direction.
Yesterday, I was at the workshop for the service, and I was on the lift. I took the exhaust manifold off again and, with the valves closed (I removed the valve cover and checked the camshaft position), I scraped out the channels using a flathead screwdriver and blew them verschandeln with compressed air. I estimate there was about 5 mm of buildup on all sides. In total, I scraped out at least half a glass full of sludge from the air filter box and the floor. The first two owners probably drove it mostly with light loads.
After reassembling the exhaust manifold and intake manifold, the engine sputtered briefly when starting, likely due to some dirt particles still on the valve seats. A short test drive provided initial insights, and it became clear that the engine responds much better to acceleration.
Today, I took the car on a long test drive of 150 km, including 80 km on the highway, and performed some full-throttle tests. The difference between the target air mass and the actual air mass is now only about 10% at full load in 6th gear. The engine now achieves speeds of over 205 km/h in 6th gear (it couldn't go any faster, as the rev limiter often kicked in), compared to around 180-185 km/h before the cleaning, which also included thick plumes of smoke.
The smoke plume is still visible, but it has become significantly smaller. When the overvoltage protection trips, a significant amount of smoke still appears. If the load is properly adjusted, it's only a small puff of smoke at full power, and eventually, the power is restored.
Unfortunately, I couldn't find a suitable tool to properly scrape out the swirl channels behind the valves, so there are probably still some very narrow spots there, and the valve discs are likely still full of dirt.
I'm going to build a small cleaning system so that I can spray small amounts of a cleaning agent into the intake manifold while the turbo is at full boost. I'll start by trying it with Flashlube, since it's designed for that purpose, so that the dirt might be chemically dissolved.
Otherwise, the head would have to be removed, and the valves taken out, to fix it properly. However, since the belts have only been replaced 40,000 km ago, I will wait another 2 years until the next technical inspection (TÜV) before doing that. If this car passes the technical inspection (TÜV) without any problems again, I'll take the engine out and verschandeln everything properly.
Until then, I'll try using chemistry.

Okay, all you 5-cylinder TDI owners who have contacted me in recent months (it seems this problem occurs more often), get a new exhaust manifold gasket, the right tools (a 6mm Allen key in various sizes for the exhaust manifold screws is the most important), compressed air, and gloves (!!! I still have dirty fingers from the grime and can't get it off!!!), and clear out your intake channels.

To reduce carbon buildup until the intake is properly clean, we'll first install smaller nozzles, which are already available. Then it will only have 115 horsepower, but hopefully it will be smoke-free for now.

Best regards, Matze.

Hi,

Congratulations on successfully debugging the issue.
I accidentally damaged the AGR valve today at a different AEL (Advanced Engine Learning) session. It definitely improves the throttle response.

compared to my AA1ZYXUDRA, however, that's a rather lame solution .
Best regards, Michael.