A4 2.5 TDI: Temporary Power Loss at Startup

Hello everyone,

I'm experiencing a strange phenomenon with my A4 (275,000 km).
Currently, I occasionally experience a lack of power at the beginning of my journey, usually in the mornings. After a few hundred meters to a few kilometers, the car "wakes up" and then functions normally again.
The error memory contained errors such as intake manifold pressure being both too low and too high, as well as interruptions in the air intake temperature sensor.
I tested the threaded rod for the VTG, and it should be fine.
At least it moves when I create a vacuum with my mouth while the hose is disconnected.
My "master" of trust took me for an extensive test drive this weekend.
Of course, as one would expect, this was observed in a state where there was no noticeable lack of performance.
The measured values (air mass, throttle position, etc.) were all within the acceptable range.
Only the boost pressure was sometimes a little too high. However, it probably comes from my chip tuning .
By the way, I've had this chip installed since 90,000 kilometers and I've never had any problems!
The start of injection was at 10.4° instead of 14.2 ± 2° before top dead center.
I've now borrowed a laptop from my company, and yesterday morning I logged the first few minutes of my commute (see link: Click me).
I think the diagrams clearly show that, up to the first minute, the values for boost pressure and air mass are completely inconsistent.
The turbo isn't kicking in at all, and it feels like the turbo hose might be disconnected (which has happened to me before), or the MSG is in limp mode.
However, the engine control unit (ECU) doesn't go into a full emergency mode; at least, no engine warning light illuminates on the dashboard, and after about a minute, it starts running again.
After the first minute, it then runs as usual.
Following my master's advice, I installed a new mass airflow sensor last night, which he had given to me.
This morning, I unfortunately experienced the same problem again.
Could someone perhaps give me a tip about what might be broken?
Thank you in advance for your comments and suggestions.

Regards,
Stettler.

Hello,

Was the car parked outside at night, and were the outside temperatures below zero degrees Celsius during the night or in the mornings when the problems occurred?

Frozen condensate in the ATL (Air Turbine Linearizer) can cause the VNT (Variable Nozzle Turbine) mechanism to freeze. In ATL systems with an electric actuator, the error message "Target position of the electric actuator cannot be reached" may be suppressed up to a certain negative temperature due to this phenomenon.

I'm not sure what operating conditions are required to generate sufficient condensation.

While 100 meters of frozen shovels is still conceivable, 3 kilometers is really no longer possible.
The error message "Intake manifold pressure too high or too low" would be consistent with both the error log and your measurements. The mass airflow sensor (MAF) and, if not defective, the engine itself are not detecting any air.
I can't interpret the "outside temperature" error message, and I'm not sure if it's related to the problem or just a coincidental occurrence.

I think you should observe the PWM signal of the vacuum control valve for the vacuum actuator during the cold start phase. You might be able to identify significant differences compared to the "engine running" operating state.

I can imagine that this could also be done using a V-meter, similar to how fuel injection timing is adjusted, as described in the article "Testing and Adjusting the 1.9 Liter TDI Distributor Injection Pump."

But maybe you should first connect a vacuum hose to the vacuum fitting before starting the engine.

The problem with icing is that after a short period of driving, the phenomenon may no longer occur because the ice has melted by the next cold start.


Greetings.

Andreas.

Me again.

I just realized that there are not only measurement blocks 3, but also measurement blocks 10 and 11.
For me, the mass airflow sensor (MAF) is not defective because the mass airflow and boost pressure values are plausible in relation to each other.
The ATL doesn't provide any flow, and therefore it can't build up any pressure.
Either the air intake is obstructed, the VNT (Variable Nozzle Turbine) is stuck, or the ATL (Anti-Lag Turbine) shaft is not rotating.

How do you manage to cover 3 kilometers in 60 seconds (end of unusual findings in the measurement report)?

Greetings.


Andreas.

While my car is always parked outside, I don't think anything is (still) frozen until I start driving.
Luckily, I have a pre-heater, so the temperature in the engine compartment should have risen enough that nothing should be frozen.
On my daily 65km commute to work, the heating system starts at least 15 minutes before my departure.
With the measurement settings I used, I obviously didn't drive 3 kilometers.
It was just a few hundred meters, then he turned back.

However, it also happens (it happened this morning) that my car can take up to 3 kilometers to "wake up."
Today, on the federal highway after the town exit, the car's behavior was such that I was able to accelerate to 80 km/h in fourth gear (full throttle). Then, for about 500 meters, the acceleration stopped, and suddenly it started accelerating slowly again, and the performance was back to normal.

Everything is really strange.

Let's see what else I can try.

Is the VTG (Variable Turbine Geometry) closed when the engine is cold, and at least easy to operate when it's not?

Hello,

I printed out your measurement reports today so that we can see all the signals at the same time and discuss them with colleagues. (Are the three measurement blocks from the same measurement run?)

An open EGR valve, or if the intake throttle plate is open, which could increase the EGR rate, might explain the observed signal patterns.

It's peculiar that the air mass doesn't change when the engine speed reaches 2800 RPM. Without the working turbocharger (ATL), the engine would operate like a naturally aspirated engine, and the air mass should increase.
In your case, the intake manifold pressure is decreasing without an increase in air mass.

Why does your engine briefly rev up to 1500 RPM after reaching the idle speed, when the pedal position sensor reads 0% within the first 20 seconds?

Greetings.

Andreas.

Thank you very much for your contributions and for the great forum!

Regarding your questions...

I just took another look.
The linkage/rod for the VTG is quite difficult to operate manually.
What is considered heavy or light is quite subjective.

The measurement data blocks originate from the same measurement run, and I logged all three measurement data blocks simultaneously.
The problem is that the measurement blocks are logged with a time delay, and the measurement values are only recorded every 1.8 seconds.
By interpolating the graph, it's possible that the measured values may not perfectly align with each other.

When I started the engine a little while ago while it was cold, and I adjusted the throttle slightly, the linkage didn't move at all.
I once disconnected the vacuum hose to the EGR valve and sealed it with a screw. Therefore, nothing should be happening in that area.
I'm not sure how to test the EGR valve myself.
With the vacuum hose disconnected, the linkage of the VTG (Variable Turbine Geometry) also did not move!

After that, I took it for a short test drive.
Once again, the same phenomenon of initial "fatigue" occurs.
After the test drive, I experimented with the accelerator pedal again while the car was stationary.
It was then possible to observe that the rod had moved slightly!

I assume that my problem is related to the VTG, which seems to be a bit difficult to operate.
It's strange, though, because I'm not driving that slowly, and recently I've been pushing my turbo pretty hard on the highway.
Is there a way to make the mechanism easier to operate using simple methods?
Or have I just created a bigger problem, which I really hope not!?!

Hello,

To truly feel whether the VNT (Variable Nozzle Turbine) mechanism is operating smoothly or with difficulty, you should disconnect the pressure rod from the pressure sensor. To do this, you need to remove the retaining clip at the bottom of the pressure rod from the pin on the adjustment lever. Be careful not to let the clip spring away, as you'll never find it again. Then, disconnect the pressure rod from the...
Slide the pin to the side. If you then don't have enough clearance, you can unscrew the pressure vessel from its bracket. However, DO NOT loosen the lock nut on the pressure rod itself.
If you adjust the lever manually (without the engine running), it should move freely from one end to the other without any catching, scratching, or jamming.
"With this procedure, no adjustment was necessary after the disassembly. The bracket holding the pressure capsule has holes with a diameter of 7mm for M6 screws. Try to align the capsule so that the pressure rod is as free as possible; you can tell this by how much you have to deflect the pressure rod to get it back onto the pin. Only attach the clips at the very end."

The pressure rod inevitably makes contact with the opening into the pressure chamber due to its design. There's likely already some of the yellow chromate coating worn away on one side. You can lubricate it, just like the connection point between the pressure rod and the pin where the clip attaches.

Possible causes for malfunctioning variable nozzle turbine (VNT) mechanisms can be varied:
Shovels can become stuck due to reduced clearance, Russian debris buildup, deformation of the turbine casing, or nozzle deformation.
Evidence of wear on the blade surface inside the turbine casing.
Erosion of the impeller blades within the nozzle, loose nozzle ring screws that are rubbing against the adjustment lever, and damaged blades caused by foreign objects.

So, you can already see that there is no general, simple way to free the VNT mechanism. However, I lack the experience to know what is most likely.

Quite a different matter, to what extent has the chip tuning affected the turbine inlet temperature?

Turbine housings might not necessarily be designed for higher temperatures, and even a 10°C difference can be significant in the range of 800°C. That's probably why the performance of cars from manufacturers is often lower than that of cars modified by tuners. Maybe you not only have a few tuned cars on the track, but you've also pushed your own car's turbocharger (ATL) too hard.

But before you panic, first check the ease of operation of the VNT mechanism.

There's a thread somewhere in this forum with pictures of the opened ATL, but I can't find it right now.

Greetings.

Andreas.

Found an entry with photos.

"Spin the pressure vessel at GT2O..."

That should be your ATL.

Let me know before you disassemble it.

Greetings.


Andreas.

Hello,

What is your grumpy morning person doing?

Have you made any progress or gained any new insights?

Andreas.

"Check for any vacuum leaks in the system. I once read about someone who had a leak in the hose that runs to the vacuum reservoir in the wheel well. This caused the vacuum to build up very slowly, which meant it took a long time for the correct vacuum pressure to be reached in order to fully engage the VTG."
You can test this by having someone start the engine while you observe the VTG rod. Normally, due to the residual vacuum in the system, the rod should immediately move upwards after the engine starts. If it doesn't, you either have no vacuum or the control valve isn't releasing any vacuum. However, since the engine runs correctly after a certain distance or time, I suspect there's a leak somewhere.