Gradual Power Loss at High Mileage - Help!

Hello everyone,

I own a first-generation TDI. Before I describe the problem, let me tell you a bit about the car, as it differs somewhat from modern TDIs in terms of its technology:

Audi 100 2.5 TDI, type 44, engine code 1T, 120 horsepower, no AGR valve, no catalytic converter, no variable turbine geometry (VTG).

The engine has at least 300,000 km on it, and possibly more if the odometer has been tampered with (which I can't rule out). Since I've owned it (at 200,000 km), it's been getting weaker. Recently, I drove another 1T for comparison and almost cried. Currently, it only reaches a speed of 185 km/h (instead of 225), and it accelerates very, very slowly (almost not at all when cold). In a power test (which I haven't tried yet), it will take at least twice as long as it should. The power build-up is really very poor; an old MB 200D pulls better than 2500 rpm. Also, it smokes (only when cold, up to 50 km of driving) above 3000 rpm, like a Russian steam locomotive.

Here's what I've done: The fuel injection pump is new, as the old one was defective. No difference. The timing belt is new (the crankshaft pulley is okay), and the fuel injection timing has been adjusted, but there's still no difference. The diesel filter and air filter are also new. I even installed a new turbocharger as a test, but there was no change. I've installed an analog LDA (load detection actuator), which shows an overpressure of approximately 0.9-1.0 bar. So, the pressure seems to be okay. The error memory is taunting me with a nice '4444' error code. Now I'm at a loss. Here are some other things I'm considering:

* Occasionally, I see bubbles in the (transparent) hose connecting the filter to the ESP, especially when accelerating. Could this be causing the gradual but constant loss of power?
* The injectors are likely very old (at least 300,000 km). However, the injectors in the comparison vehicle (which is running perfectly) are also demonstrably at least 290,000 km old.
* Could the boost pressure sensor be faulty and be sending a signal to the control unit indicating a lower-than-actual boost pressure, causing it to release less fuel? Should there be an error code stored in the memory?

Does anyone have any ideas? I'm open to all suggestions.
Sure, no problem. Here's the translation:

'Thank you in advance.'
Björn.

Hmm, so if there wasn't enough fuel, it wouldn't produce so much soot.

I would suggest removing the nozzles and testing their spray pattern (there's a thread about this here). I think they are either worn out or severely clogged, which is the most likely cause.
*LMM, alternatively, exchange.
Is the LLK clean?
*Replace/test coolant temperature sensor and diesel temperature sensor.
*It might also be a good idea to check the engine's compression.

Good luck!

Hello Boernhardt,
* Occasionally, you can see bubbles in the (transparent) hose connecting the filter to the ESP, especially when accelerating.
This is likely caused by a blocked fuel vent or a leak in the diesel lines.

* The injectors are probably very old (at least 300,000 km). However, the injectors from the comparison vehicle (which runs perfectly) are also demonstrably at least 290,000 km old.
300,000 km is quite a lot; I would -especially given the smoking you described- go to Bosch and have them checked. (It's not that expensive). The fact that the other ones lasted 290,000 km doesn't prove that yours are still good . I think that's the most likely cause of the problem.

* Could the boost pressure sensor be faulty and suggest a too low boost pressure to the control unit, causing it to release less fuel? Should there then be an error code stored?
Then an error message 'Charging pressure below regulation limit' should be stored. Furthermore, there should definitely be no excessive smoking.

Although the engine doesn't have an EGR system, I'm wondering if it has a mass airflow sensor (MAF) that is known to cause gradual performance loss. (But even then, it probably shouldn't be producing smoke.)

I don't know if such an old engine has that feature, but I would try to read the idle speed control data.

I don't know if a small compression test would be appropriate given such high mileage, but that's why we have experts here in the forum . It certainly can't hurt.

Regards,
Private.

Edit: Julian was faster...