AJT Smoke: Causes, Risks & Prevention

Hello forum members!

Background:

My friend and I went on vacation together with our vans, and I noticed that his T4 AJT from 1999 was emitting a lot of black smoke, especially under heavy load.

Upon reaching the destination, I suspected that the EGR valve might be stuck. Since a piece of metal had gotten lodged there, we proceeded to inspect it. We discovered a corroded, thin aluminum plate. All the soot that had accumulated in the EGR pipe was directly entering the intake manifold. After some disassembly, we took a look inside the intake manifold. A layer of soot at least 8mm thick had completely clogged the intake manifold.

A 3mm stainless steel sheet, which happened to be a perfect fit, was used to permanently seal the EGR valve. The amount of soot emitted during the journey home was then kept within acceptable limits.

A few weeks have passed, and the amount of soot being produced has been increasing significantly. What's a little concerning is that after warming up the engine at idle, white smoke comes out of the exhaust pipe. This smoke clearly smells like unburned diesel fuel. When accelerating, it gets worse – there's a devastating amount of soot production. However, this is happening WITHOUT any loss of power.

The AJT is chipped and uses an engine control unit (ECU) from the company Tuneline, which is switchable. When turned off, it performs slightly better.

The following actions were attempted and verified:

LMM has been renewed.
Air ducts checked for leaks.
Boost pressure controlled.
Vacuum hoses checked.
All the target values from the table were checked using VAGCOM (throttle valve control voltage, air mass, injection quantities, fuel injection timing, etc.).
VAG timing check: OK, all values are within the specified range.

The only thing that was noticed was:

The idle speed control showed very high and varying values:

Cylinder 1: -1.3
Cylinder 2: ,5
Cylinder 3: -1.5
Cylinder 4: -0.7
Cylinder 5: ,5

I am already aware that the values can be between +2 and -2, but do you think this is correct? In my case, all the values for ACV are around +/- 0.5.

Okay, here are my guesses:

Fuel injectors defective??? Dripping afterwards? Spray pattern distorted.
Are the valves leaking due to excessive carbon buildup?

As mentioned before, the engine has its full power output, it doesn't require any oil or water, all temperatures are within normal ranges, the diesel filter has been replaced, and the air filter is new. This T4 hasn't even reached 100,000 kilometers on the odometer.

Since I'm now completely at a loss and standing next to a huge cloud of smoke, please help me.

Greetings from Graz,

Peter.

Does anyone have any ideas on how I should approach this problem?

Hi everyone,

Peter.

Please thoroughly review the logs from files 003-008-011 and show them to us.
Maybe something can be recognized.

When it comes to mechanical problems, we can only guess based on the information available online.

Okay, please provide the German text you would like me to translate into English. I will only provide the translation.

You know how it is; when someone has a similar problem, they just say, "Oh, that was... this and that." This was just an attempt to post something here. Of course, it's incredibly difficult to find anything without completely taking apart half the bus.

I won't tell you how much time I've already spent sitting here, watching the logs. But everything is okay; no values are out of place.

We visit the Bosch service once and start checking the injectors, then the pump, and so on. This costs a lot of money, which we actually wanted to save .

Hi everyone,

Peter.

Hello everyone!

Okay, so the fuel injection pump has been repaired, costing a hefty €800, but unfortunately, the problem still persists. The AJT is smoking like crazy.

The entire flow control valve was replaced, and we were allowed to keep the old one. According to the pump specialist, he always replaces the flow control valve without a second thought, which you can imagine caused quite some trouble.

Okay, now let's get to the problem. What could still be causing such excessive soot production? As mentioned before, we have repeatedly checked all measurement data blocks, conducted test drives, but found nothing conclusive.

My question:

Is it possible that a hanging needle injector is the cause?
Or perhaps the crankshaft sensor is providing incorrect readings, which would cause the injection timing to be off?
Or, finally, could the engine control unit (ECU) be scrap?

All errors that were controlled (as much as possible), but whose function cannot be verified without replacement.

I am grateful for any advice you can offer.

Greetings, Peter.

Mc_Givertechnik wrote:

Is there a possibility that a hanging needle valve is the cause?

I'm less inclined to believe that: if the signal arrives late, the MSG adjusts the injection timing to an earlier point, which reduces soot instead of increasing it.

Quote:

Or perhaps the crankshaft sensor is providing incorrect values, which would mean that the injection timing is no longer correct?

It's more likely that the donor wheel is misaligned on the crankshaft, or that the sensor is positioned "too late" in terms of rotational direction, perhaps due to a partially broken mounting bracket.

Quote:

Or, finally, could the engine control unit be scrap?

Alternatively, the tuning might have an excessive richness, which can lead to significant soot production even with small turbocharger air leaks (-> check!).

Speculation: A lack of air in one cylinder (e.g., due to an EGR sludge blockage in the intake manifold or on the valve) is masked in the log data by the other four cylinders, but the soot from that cylinder is not absorbed by the exhaust gases of the others; it is only diluted.
I'd like to run a compression test; maybe something will show up there.

Hi Ulf!

Well, that's interesting. Your last statement really surprised me.

Warum? So...

The AJT started smoking slowly, and the cause was a corroded metal piece in what we can call the EGR area. This allowed large amounts of accumulated soot to enter the intake manifold. I'm also wondering where smaller pieces of that metal might have gone.

The sheet metal was then replaced with a 3mm V2A stainless steel plate. Since then, things have only gone downhill with the good AJT engine. Consequently, the intake manifold looks very sooty. My recommendation to remove and verschandeln it, or to clear the intake channels, has not been successful.

The engine control unit (ECU) has two software levels and is from Tuneline, which I believe is one of the most experienced tuners (specializing in Golf Group N TDIs and developing solutions for VW racing TDIs).

The "bus" performs the same function at both software levels; however, with modified software, the symptoms are significantly more pronounced.

Greetings, Peter.

Mc_Givertechnik wrote:

The STG has two software levels and is from Tuneline, which I believe is one of the most experienced tuners (for Golf Group N TDIs and developments for VW racing TDIs)

.
Caution: Drawing conclusions about a tuner's software for everyday customers based on their motorsport activities can backfire badly.

For example, two tuning files (VP and PD) from a "premium address" in southern Germany somehow ended up on my computer.
Both files are made so simple and tame (to put it mildly), and they show no deeper or even premium understanding of the software, so that with them, at best, you could compete in disciplines like "Whose tank will be empty last when running on full throttle?"

Hi Ulf!

"No, no, the ones from Tuneline are already fine. Their engine control units (ECUs) are even sold and recommended by local VAG dealerships. They've been in business since the 80s, and most of the cars here in Imfeld use their ECUs. They also develop ECUs for Volvo locomotives and various other projects, with quite good success. I personally have driven two TDI cars with their ECUs and was more than satisfied."

@everyone
Subject:

Okay, so on Friday, a lot of time was spent working on the AJT system. A new injector control valve was tested, and the check valve on the pump for the return flow was replaced. The engine control unit (ECU) was removed to try a different one, but it couldn't be found.

Initially, there was no change, but after a certain amount of driving time, there was a sudden jolt and the power suddenly returned. When I pressed the accelerator firmly, a large amount of soot shot out of the exhaust pipe.

Over the weekend, I drove several kilometers on the highway at full speed. The result was a respectable 15 liters per 100 kilometers. The soot emissions have decreased, but are still slightly visible.

Okay, so now someone has to figure out what that was. Was it a contact error in the engine control unit (ECU) that was fixed by disconnecting and reconnecting it? Was it the check valve? Or was it a blockage in an intake manifold that suddenly cleared itself?

What I've noticed since the beginning of this problem is that there's a certain unevenness coming from the exhaust. One cylinder seems to be running slightly differently. Of course, it's a five-cylinder engine, which sounds different, but believe me, I can hear the difference, so please no remarks .

Okay, so my friend's bus was back with me again, and I waited until he drove off because I wanted to hear the exhaust fumes and the noise one more time. When he accelerated, you could clearly hear that one cylinder wasn't working properly. Now I think it's probably a leaky valve or a faulty injector. The injectors will be checked next week. Then there's the compression test; let's see what we find.

The search continues, I will report back later...

Greetings, Peter.

The solution!

After a long period of troubleshooting, including checking the ESP, injectors, filter, various sensors, replacing the mass airflow sensor, and so on... the following issue was finally identified.

Actually, it's the simplest thing imaginable. Just dirt, yes, literally just dirt. The classic oil and soot sludge in the intake manifold. One of the intake channels was definitely too dirty, which is why there were such large discrepancies in the idle speed control, insufficient airflow, and therefore extreme soot formation.

However, the T4 owner's solution wasn't exactly the most refined English approach. The gentleman used brake cleaner and similar substances quite aggressively to remove the debris, then started the engine and blew the dirt out with a massive, unpleasant expulsion from the poor AJT engine.

The engine seems to have handled it quite well at first, but we'll see how it holds up in the long run.

Best regards, Peter.

Mc_Givertechnik wrote:

The classic oil and soot sludge in the intake manifold. One of the intake channels was definitely too dirty.

Thank you for the clarification!
(At least once, I've thought ahead about how technology might trick us ).

Hello!

It seems like the whole thing is starting all over again.

The AJT has started smoking again, and the performance is very poor. Since the last interventions, 20,000 kilometers have been driven. In the meantime, the cylinder head has been removed and cleaned, the injectors have been checked, diesel return lines and other potential sources of error, such as vacuum hoses, have been replaced. The fuel injection module was already replaced with a new one at the beginning. The ESP (Electronic Stability Program) was serviced (during which the water pump gasket was also replaced).

Okay, I connected this device to VCDS and made a discovery that I hadn't observed before. The control signal generator (block 19) always shows 0.000V in both fields, regardless of whether the engine is running or not, or under any conditions. When comparing it to the target value table, there's nothing listed there either. However, the voltage in block 1, field 3, *is* at the target value.
That might be true.

In the idle speed control system, one cylinder shows a value within the range of ,7 - +2.0, which is technically within the normal range, but deviates significantly from the other cylinders, where the values are typically between -0.5 and +0.5.


The measurement block 16 in the EDC15, which is not listed in the target value table, permanently displays "Load 96.8%". I have no idea what this measurement block indicates. Is there a connection to anything else?

Best regards,

Peter.

Hello,

Quote:

Now I've connected this device to the VCDS and made a discovery that I haven't been able to observe before. The control unit shows 0.000V in both fields, regardless of whether the engine is running or not, no matter what.

In measurement block 1 or the values for the end positions?

What is the VAG part number for the control unit, and what is the software version?

Best regards, Rainer.

Measurement block 19, fields 1 and 2 always at 0.000V. End positions: start/stop.
Measurement block 1, field 3: values are OK according to the target data.

Unfortunately, I didn't note down the serial number from the EDC or the software version. I will ask the bus owner to stop by again.

Best regards and thank you,

Peter.

Mc_Givertechnik wrote:

Measurement block 19, field 1 and 2 are always at 0.000V, end positions start/stop.

IIRC (at AFN - and ALH-times), this can sometimes happen with the VP37 depending on the motor starting conditions, without anything (noticeably?) being defective.

Hello Ulf!

And what, based on experience, might be found to be a hidden defect?

Best regards,

Peter.