Tankanzeige hängt nach - A4 Cabrio 2.5TDI (BDG) | Posts 16+

Of course, it could also be the case that you are willing to pay.
What prevents you from following Guste's suggestion above?
VCDS do you have? Also, also an RLF (which shows how to access the sensor, definitely not by removing the fuel tank), and maybe a multimeter.
hg
Herbert

Honestly, I still lack the RLF. With that, there definitely wouldn't be any problems.
But maybe I'll just take the back seat out and see if anything tank-like looks back at me.
However, I won't be able to get to that until just over a week from now.

RLF is available at Erwin, and you will probably find the cover over the fuel opening under the back seat. Likely foam, and underneath that a cover in the floor pan.
hg
Herbert

On the tachometer plate, stepper motors are responsible for the needle movement. In the picture is an 849 Passat plate. The stepper motors are clicked into the plate with plastic retaining clips. And of course, the contacts are soldered to the plate. I had to replace one of these motors once before. It actually went quite well. You could find the motor at an auction house. "VDO Stepper Motor VAG" would be a possible search term there...
If you want to replace the AI, you need to familiarize yourself with the keyless entry system. Another AI needs to be trained.
First, I would open the tank at the flange of the metering unit (without removing the tank) and then move the sensor. While doing this, you can observe the pointer in the KI and also the MWB on the laptop.

Hello everyone!

Steffen W wrote:

"I can also monitor the AI's progress and the MWB on my laptop."

Please note that these advertisements (Coolant temperature and (Tank capacity) are significantly reduced - and this is due to software-based AI
It is intended that these analog displays only react very slowly to changes.

However, using VCDS, you can of course query the sender or input values in real-time, but it is not an error if the associated display is not yet B. Approximately 1 minute delayed and very sluggish in response.
THIS is NORMAL and also INTENDED to be this way.

Best regards from Munich!

Thank you very much for your contributions!!!

I retested the behavior directly after ignition today, as I had installed my BT-FSE today anyway.
Car is parked, ignition on, fuel level sensor and display from the instrument cluster logged.

"The gauge repeated its peculiar behavior: it would rise to around 5/8 when the ignition was engaged, only to slowly decrease until the ignition was switched off (I didn't want to wait ten minutes), and at a rough 3/8 reading, I stopped the process.
Because: the sensor blocks for the fuel tank and fuel gauge were consistently (198 Ohms and 29 liters).

Therefore, I think the case is clearly defined: if the value that the AI measures (as stated above) is constant, but the display initially shows much too high a value before gradually decreasing to the target value, then the problem must lie with the AI, not the sensor.

Or, do you disagree with this, or do you see the information differently and not definitively attribute the error to AI?

Hello everyone!

guste100 wrote:

For me, the only option that really makes sense is AI. ...

Okay, Guste knew it already a week ago

Essentially, after excluding all other options, only AI remains, right?
However, I am concerned that the diagnosis for the valve was unremarkable.
That a defective. I don't think the motor itself is the cause, either, because this would already be apparent during diagnosis or testing.

Are the other ads working perfectly? Especially... Coolant Temperature Display and not similar "leaks"?

You could also test two things to be sure:
1.) Immediately start the control element diagnosis on a vehicle that has been completely cooled down overnight and has only about half a tank (or less) of fuel, and operate in 50% mode for an extended period, B. 10 minutes, allow to sit.
Do all the pointers stay verschandeln at the midpoint, or does something suddenly wander there too?
2.) W. o. (while the engine is still cool) turn the ignition on, and lower the fuel nozzle to its (here "3/8") position. Then, add approximately 10-20L from a canister and wait for 10 minutes, observing what happens. The read values should change immediately/simultaneously with the tank volume; the needle is VERY dampened and may take approximately 10 minutes to respond.
What about the ignition on/off immediately after?

Best regards from Munich!

Okay, a really interesting turn of events:

I just visited the specialist for car electronics: the dashboard is not broken.
They have checked the stepper motor and the electronics, both at cold and at warm temperatures - everything is functioning as expected.

Then, it seems, the only option left is the fuel dispenser...

Interesting... and where do you think you are?

I'm curious to see how the fuel level sensor, with constant resistance values, tells the combined instrument that the fuel level is changing, even though the sensor itself hasn't changed. Can you or your specialist explain this?

One can make mistakes or strange things can happen, no matter what, but one should a) stick to the truth and b) admit to making mistakes.

The usual case is unfortunately that nothing happens in such cases...

My Tip: Before you buy and install a new fuel tank sender, test a fixed resistor on the connector.

Hm, did you mean by 'staying true to oneself' me???

I visited an automotive electronics specialist.
They tested the AI using the resources available to them and were unable to find any errors.
I asked the exact same question as you. That's all they could say was that they couldn't rule out the possibility that the problem might be with the source.
And they should, hopefully, manage to correctly test the combination instrument.

Should.

As a layperson, I can't do much on my own with a workshop.

It is also possible that they (Endenga Company in Frechen) don't know what they are doing. But that's difficult for me to assess.

Then I guess I'll take a look at the resistors...

The specialists and you. I had already consulted "experts," and I took the time to prove the facts and measured the person myself to determine the results. Unfortunately, my time investment did not pay off, as it was clearly more acidic in the stomach than the joy of finally understanding and explaining something. Seems to be a human weakness.

That, of course, is also clear: working time during long-distance driving cannot be wasted.
Did you also say that the error only occurs after x minutes? Because if everything is fine after 2 minutes and you don't notice, you simply put it aside with "it's okay."

What does the donor actually cost? If everything needs to be removed for testing and the part only costs 30 EUR, you can also have someone else do the testing and immediately install a new one.

One must check the online marketplace for a combination instrument.

Hm, ok.
By doing so, you are accusing me of lying.
I find it regrettable and disconcerting that you would pass such a judgment on someone you don't even know.
Accusing someone of lying and refusing to admit a mistake (which one, actually?) is a really difficult situation. But this might be due to the Forum platform.

Perhaps you could clarify what I think you did wrong.
As far as I understand, we have concluded in this thread that the problem is likely due to the instrument cluster.
Now, it seems I made a mistake by going to a repair shop and posting their assessment here.
Does this lead you to your assessment?
How?
I have had no doubts whatsoever about dealing with experts like yourselves here.
And I can measure the sender, I think I can't repair the KI, that's why I went to the workshop.

I would really like to understand how you arrive at your conclusions.

P.S.: Of course, I have repeatedly pointed out that the error only occurs after a certain period of time.
Just like I tried again to get the VCDS measurement results across.
I'm not saying that the workshop is definitely right, I'm just posting the results of their investigation.
I don't consider this to be reprehensible, and I don't consider it to be a lie on my part.
And the donor is in the low three-figure range, according to Audi. Part-time professional: approximately 200€ + installation.

Edit: Formulation geändert, Schreibfehler korrigiert

Let's try to stay objective. No one has accused anyone of lying.
What exactly was checked in the workshop, and how? The result "KI in Ordnung" (KI = control unit) and your observation (unchanged resistance and fuel level measurements, but the display changes) don't match. Either one of them is incorrect, or the error has miraculously healed itself.
hg
Herbert

You are right, Herbert.

I was told that the steering motor has been checked and is running smoothly in every position, as it should.
Additionally, the electronics responsible for the fuel gauge were inspected. They probably cooled and heated the board at different times to simulate different temperatures.

They were unable to identify any malfunctions, as the fault memory was already empty.

Additionally, resistors were connected - and in each case, the expected value was displayed.

I really don't know what else can be checked anymore.

I'll try to get to the sensor and measure its value, then compare it with the value from the OBD. Maybe the connection is corroded. That would also be a simple option.

Quote:

I'll try to get to the sensor and measure the value on the sensor and compare it with the value from the OBD, maybe the connection is corroded. That would also be a simple option.

Okay, here's the translation:

Go ahead, practice makes perfect.
However, your observation (resistance and tank volume measurements remain unchanged, but the display changes) is worth considering.
Is it still like that (?), and if so, you can test again using a fixed resistor instead of the sender.

Quote:

Additionally, the electronics responsible for the fuel gauge were inspected. They probably cooled and heated the board at different times to simulate different temperatures.

Has the AI been upgraded and broken down?

hg
Herbert

Okay, unfortunately, it's proving to be more difficult than initially anticipated to get in touch with the donor.
He is, unfortunately, not under the back seat. There are indeed two metal covers on the left and right. However, if you remove them, you will find either the top of the tank or a heat shield underneath.
Under the cover in the trunk, before the partition to the interior, there is a metal cover on the floor.
To access those screws, however, the panel of the partition wall must be removed - I won't be able to do that today and will have to do it on another day.

I wasn't wrong during the repair, but it was presented to me in a way that made it seem like the AI was disassembled, taken apart, and inspected in that state.