Hello fellow simulators,

at the moment i'm trying to do some flow analysis with airflow and a closed waterloop inside an internal analysis.
It's basically a cube with fans inside which generate an airflow loop.

This selaed box is connected to a outside by a "wall of constant temperature" BC.

Inside this cube is some other hardware whch generates heat and of those parts is equipped with a closed water loop (necessary for scientific reasons, not cooling).
You can see a pretty easy principle sketch in the attachments.

I made some first runs without the waterloop. Everything is working fine so far.

Now the next step is to implement the water loop. I did this as as fluid subdomain but the problem is the piping of the water.

I often read about "use a fan as pump", so did i. I placed a "internal fan" inside the pipe defined with a specific volume flow rate.

Now i have two issues:

1. i get the error "mass flow out of fan curve's range"
maybe this is a problem caused by the different densities of air and water and the "internal fan" is programmed to work only with air?...idk

Btw. the fan curve reaches from 0 to 3.5 bar with specific volume flow rates, the pressure inside the water loop should not exceed that range...

2. the estimated sim time rises from previous, fine 7h to massive 60h (~500k cells)

(i know, that there's still the option to replace the heat exchanger (low performance fin cooler -> thick and few fins) with a porous media, but i gues this won't serve a big advantage this time.)

The aim is to see the temperature distribution of the inner cube and how fast it's getting homogeneous, so i think i can't get rid of simulationg that water loop...

Hopefully there's someone out there who has a good idea...
Thanks in advance, guys!