Stationary / time dependent solution differs

Hello everyone,

my model is a U-shaped pipe in a concrete block. Cold water (1°C) flows through the pipe and the warmer concrete (11°C) surrounding it warms up the water inside the pipe.
I used the "pipe flow" and the "heat transfer in solids" physics. Everything works ok in stationary mode and the outlet temperature of the water is about 1,6°C, which is what was expected and modeled with another program.
(check study 1 and results-->temperature)

Now to the problem:
I also started a time dependent study just to see how long it takes that nothing changes in the temperature anymore. Everything starts with a temp of 11°C so it should take some time. But no matter how long I set the time and how small I set the time steps the final outlet temperature is much higher ~3.3°C. It should end up at 1,6°C though.
(check study 2, I could not post the file with results because it would exceed the allowed file size to upload)

The file I attached contains only a big step size and a short time, but I also calculated for very small stepsizes (60secons) for one year and it ends up at around 3°C.

I hope somebody can help me out.

Thanks

Christian