Log Trip Emergency: High Turbo Pressure

Dear fellow enthusiasts,

After hopefully having worked through all (there are already quite a few ) technical articles on the topic of 'excessive boost pressure, sticking VTG, and boost pressure control,' and not having figured out where the problem lies, I would like to bring my problem up for discussion here.

The vehicle is an Audi A4, manufactured in December 1996, with an AFN engine and approximately 167,000 kilometers (103,000 miles) on the odometer. The vehicle has been modified with a chip and new 216 injectors, increasing its power output by approximately 10,000 kilometers (6,200 miles). A used ATL turbocharger (GT1749VA) was later installed because the original turbocharger was consuming excessive amounts of oil.
'Because the vehicle was emitting a significant amount of smoke during rapid acceleration, I conducted a diagnostic drive (by the way, the smoking strangely disappeared after the first emergency mode event, which was a sudden power loss but without the glow plug warning light flashing!).'

During this test run, there was a sudden loss of power while operating at full throttle, and an error message was recorded.
'00519 - Intake Manifold Pressure Sensor (G71), 28-10 - Short to B+ - Intermittent'.
The error entry '00575 Intake manifold pressure control differential' is NOT recorded (it is usually used here ).

Subsequently, I temporarily installed a boost pressure gauge (0-1.6 bar). This gauge confirms the reading from the engine control unit (ECU). During full throttle, the boost pressure is exceeding 1.5 - 1.6 bar!

Therefore, I have checked/replaced the following parts:
- VTG adjustment (pressure canister released, the adjustment lever moves very easily).
- Pressure chamber (return spring checked; by manually moving the rod back and forth, you can clearly feel the spring force).
- Pressure vessel and VTG (tested multiple times using a vacuum hand pump with a pressure gauge, moving very slowly each time, and running smoothly from start to finish).
- Check valve (tested for unidirectional flow with the mouth ).
- Replaced the N75 solenoid valve (converter) , but it didn't work (and I'm now 77 euros poorer }).
- Checked (vacuum hoses to the N75 valve, to the vacuum reservoir) or replaced (hoses to the air filter housing and oil filler cap).
- Vacuum pump (it generates approximately 600 mmHg, and the chamber is initially pressurized with approximately 500 mmHg in the laboratory).
- Mass airflow sensor (MAF) replaced approximately 5,000 km ago (for informational purposes only, the MAF is not suspected as the cause of the problem).

In the attached log, it is clearly visible that the actual LD (LD(ist)) is significantly higher than the target LD (LD(soll)) during the full-throttle phases. Furthermore, the N75 valve is being closed further. Nevertheless, the actual LD remains high.
If I hadn't checked/replaced the Canister Purge Valve (CPV) and the Venturi Tube (VTG), and the N75 sensor, it would immediately be suspected as the source of the problem.

By the way: Why wasn't error code 00575 saved? There's a significant difference in the regulations (leading up to a fault/sudden power loss, but without a flashing preheating lamp).

I'm really at my wit's end!

Here's a crazy idea, but it hasn't been tested yet: Short-circuiting the cable to the N75 valve to either the positive or negative terminal (whichever fits...) during hard acceleration, and observing the resulting motor movement (my test track is uphill, with a 6% gradient).

Do you have any other ideas (or even several)?

Best regards,
Kassen.

Can you rule out a VTG_HEISS terminal?

You probably didn't perform those extensive tests with the turbo running at full boost... but it's "strange" when it's hot (under full load).

Just a thought.
He's going to start messing with his turbocharger now.

Hello dieselmartin,

I prefer to let my turbo cool down properly before stopping.

Of course, I can't rule out a short circuit.
It's less than 2 km to my test track. The engine was just barely at operating temperature from a previous trip when I started the test run.
During the first 'emergency run' (is it really a 'true' emergency run if the preheating indicator doesn't blink?), it was probably about 300 meters.
Is the turbo already so hot that it could seize?

'Has there been any documented cases of 'hotlinking' in the forum?'

Best regards, Kassen.

Let's just say it like this:
"Was ist der Grund, warum the loader is for sale? Do you know that it was in good condition before?"

DIY Idea: Is it possible to monitor the N75 valve during driving using an LED, and can flickering or changes in the light indicate a fault? Or would that overload the control module?

To determine the clamping position and ensure proper adjustment, the adjustment rod needs to be connected to the driver with a Bowden cable and positioned towards the inside.

Digicam, reed relay... but the effort involved.

Hello teileklaus,

I cannot confirm or deny your assumption.
A lot of things are sold on eBay, and of course, nothing was mentioned about potential 'hot connector' issues.
The loader looks good from the outside, but it appears to be used, and it's supposed to have only run for about 80,000 hours.

I discovered at least one ATL with hot-clamping issues while browsing the forum. It all sounded very similar .

I will definitely still try to measure the vacuum inside the container during the journey before I start looking for a new charger.

Best regards, Kassen.

Hi,
Before I would do that, aside from the money, I would first want to know the VTG's position if it were to go into emergency mode.
und wenn es mit mechanischen Mitteln wäre ( s. O.)
Why measure the can pressure anymore? I think the hoses are sealed, and the N75 valve is new?

Hello teileklaus,

You're right, the hoses/N75 valve are new.
However, if the cables to the valve have a short circuit to either positive or negative, the pressure would still not be correct (as mentioned in my previous post).
Perhaps that could also trigger an error message (the MSG would detect a disconnected N75 sensor!)?
However, if there were a simple way to measure the mechanical position, that would certainly be a good thing as well.

Best regards, Kassen.

Quote:

However, if there were a simple way to measure the mechanical position, that would certainly not be bad either.

My BMR has something similar... but I can't figure it out either.

Buy yourself some welding gloves (thick leather) for a few euros.
Then you'll probably be able to start working on the VTG (vehicle tax gross) relatively quickly after a test drive.



And don't let yourself be observed when you first blast onto the hardware store parking lot at 4000 RPM and 2 bar of boost, and then, with thick gloves, you start fiddling with your engine .

But an A4 AFN should be the dream of achieving VTG (Very High Gain), right?


m;

Hello dieselmartin.

But an A4 AFN would surely be the dream of VTG accessibility, right?

My VTG is, frankly, practically exposed in front of me , after I removed the air filter snorkel and the intake pipe.
Especially compared to the engine compartment of my summer car (with an air-cooled six-cylinder in the rear ), it's very spacious.
I will try it once with welding gloves (a good tip , I even have some, and I don't even need to buy them ).

Best regards, Kassen; the one who is going to sleep now.

oldewurtel wrote:

During this test run, there was a sudden drop in performance while operating at full throttle, and an error message was recorded: "00519 - Intake Manifold Pressure Sensor (G71), 28-10 - Short to B+ - Intermittent". The error entry "00575 Intake manifold pressure control differential" is NOT recorded (it is usually used here ).

To comment on this further:
The LD signal error ranges "LD too high" and "sensor short circuit" must be separated in software such that a high LD value never triggers the "short circuit" error condition with the intended sensor.
Therefore, I would definitely recommend adding a copper worm to your setup, which will, of course, disrupt the normal behavior depending on the type and amount of signal distortion.
Therefore, it's very good that you're also checking the LD using LDA to ensure its "simple reliability."

Hello Ulf,

Thank you for your reply.

What surprises me is that the LDA shows very similar maximum values (observed and noted during the trip ) as the log that was later analyzed.
The chip sets the upper limit for the LD pressure at 2250 mbar, which is currently 1.25 bar above atmospheric pressure.
My LDA shows values up to approximately 1.5 to 1.6 bar (you can't stare at it for too long, you still have to drive the car ).
The error code 00519 only appears if the described emergency mode/performance drop occurs beforehand when the engine is running at full throttle.

Where is the end of the measurement range for the sensor in the MSG?
Is it not possible that the upper limit of the measurement range (e.g., 2600 mbar = 1.6 bar LDA) is equivalent to a short circuit to the positive voltage for the software?

Otherwise, I could also start looking for a new or used MSG (Mass Airflow Sensor). Or, is there a suitable sensor available online? (I'll start searching and using Google shortly).
Do you happen to have the exact name or part number? That way, I wouldn't have to remove and disassemble the MSG.

Best regards, Kassen.

oldewurtel wrote:

The error code 00519 only appears if the described emergency mode/performance drop occurs beforehand when the engine is running at full throttle. Where is the end of the measurement range for the sensor in the MSG? Is it not possible that the upper limit of the measurement range (e.g., 2600 mbar = 1.6 bar LDA) is equivalent to a short circuit to the positive voltage for the software?

After the official error explanation, not:

Quote:

Error code: 519, 17563. Manifold pressure sensor - G71, short circuit to positive. Possible causes: Short circuit to positive. G71 defective

LD rule errors "must" trigger this text:

Quote:

Error code: 575, 17958. Intake manifold pressure, control differential. Possible causes: Faulty boost pressure control valve (N75). Hoses connected incorrectly, not properly attached, leaking. Air leak between the turbocharger and the engine (charge air path). Turbocharger failure

However, if I consider faults such as a partially broken ground connection for the laser displacement sensor, or an excessively high supply voltage, the signal value at the ADC input could potentially rise higher than what corresponds to the end of the sensor's measurement range, and thus potentially trigger a short-circuit error.

Quote:

Or is there a suitable sensor available on the web (I will soon be searching, or consulting Google). Do you perhaps have the exact name ready?

No, you probably have to tighten the screws...

Hello Ulf,

I need to make a slight correction. The error code 00519 also appears even without the engine running, if you only have the 'ignition' turned on.
This error occurs due to pressure exceeding approximately 1.55 bar (which corresponds to a reading greater than 2250 mbar).

I tested the pressure sensor in the MSG as follows:

Using a compressor and a downstream pressure regulator (including a pressure gauge), I generated an overpressure of approximately 0.8 bar to approximately 1.7 bar.
At the same time, I read this 'charging pressure' on the pressure gauge (from the pressure regulator), LDA, and with VAGCOM on the MSG .
All three pressure sensors were consistent within the measurement accuracy! I performed approximately 10 measurements within this pressure range.
Everything was fine until the MSG displayed a pressure of 2550 millibars.

As soon as I exceeded approximately 1.6 bar, the MSG/VAGCOM no longer displayed anything meaningful (I no longer remember whether it showed 0 mbar or the ambient pressure ).
After that, error code 00519 was stored, along with another error message (indicating an implausible engine speed or something similar).

'This suggests that error code 00519 is triggered when the maximum pressure is exceeded (in this case, > 2550 mbar), without it necessarily indicating a defect in the sensor itself (as claimed in the repair manual), right?'

Normally, one would design such a circuit (sensor + ADC) to utilize the entire measurement range, and in the event of a significant exceedance of the normal operating range (normally, the maximum operating pressure without overshoot is around 1985 mbar for an AFN sensor?), the system would simply generate a 'short circuit to positive' error message.
I'm sure no one was thinking about people with a faulty GT1749VA turbocharger.

BTW:
Do you know exactly when (duration and size of the LD overrun) error 00575 occurs?
Assuming a pressure of 300 mbar and a duration of 6 seconds (as described by you regarding the emergency mode), the MSG (assuming a tuning for a maximum LD of 2254 mbar) would need to measure 2554 mbar in order to clear this error.
Could it be that the maximum measurement range has already been exceeded, making it unreadable and therefore preventing it from being logged as error 00575?

Best regards,
Kassen.

oldewurtel wrote:

However, once I exceeded approximately 1.6 bar, the MSG/VAGCOM no longer displayed any meaningful information (I no longer remember whether it showed 0 mbar or the ambient pressure ). After that, error code 00519 was stored, along with another error message (indicating an implausible engine speed or something similar). That seems to indicate that when the maximum pressure is exceeded (in this case, > 2550 mbar), error code 00519 is logged, without it necessarily meaning that there is a defect in the sensor (as claimed in the repair guide), right?

That assumption is indeed likely...

Quote:

Do you know exactly when (duration and size of LD overrun) error 00575 occurs?

No.

Quote:

Assuming there are 300 mbar and 6 seconds (as you described in relation to emergency mode), then the MSG (with a tuning for a maximum LD of 2254 mbar) would have to measure 2554 mbar in order to clear this error. Perhaps the maximum measurement range has already been exceeded, i.e., it is not displayable, and therefore cannot be recorded as error 00575.

I don't have any further information on this, and I need to pass the question on.

Hi Kassen,

Quote:

What is the end of the measurement range for the sensor in the MSG?

2600 mbar. A replacement sensor is available in the shop, but they rarely fail.
The error message described seems illogical. How did you read the error code (VAG-COM version)?

If the magnetic valve of the LD control and the hose connections are okay, then in 99% of cases, the excessive boost pressure is caused by the turbocharger.

Best regards,
Rainer

Hello Rainer, hello Ulf.

Thank you for the answers!

I read the values using the full version 409.1-S of VAGCOM.

The LD control valve is new, and the hoses are new or have been inspected and pressure-tested.

My log (see the first post) shows that the regulation reduces the vacuum in the can (TV 94%) in order to limit the boost pressure. However, it doesn't quite manage .

It will be the loader!

@ Rainer:
Do you have a suitable turbocharger in your shop? Please provide a quote as a part number or via email. The turbocharger type is GT1749VA, 038 145 702 G, or a compatible model for an Audi A4 (AFN).
Would it arrive at my place before Christmas?

BTW: The internal sensor reads 250 kPa, which is equivalent to 2500 mbar (see the shop).
Can the original sensor still measure up to 2600 mBar, or is that just the extreme end of the measurement range, where the readings become quite non-linear?

Best regards,
Kassen.