I am trying to run a Convective heat transfer analysis on a double facade cavity that works during the winter as a thermal buffer zone with no air inlets/outlets (so in reality only slightly ventilated) and to see the temperature fluxes inside the cavity, by warming it up only with the solar radiation as an additional radiative source on the external reflective textile skin which I assigned to have a transparent radiative behaviour.
I used a value of 1000 W/m2 that I derived from a radiation analysis and assigned 13oC as an initial boundary temperature and 10oC as an ambient one. For simplification, I assumed that the rest of the boundaries are adiabatic. I ran different simulations with slightly changed settings for the outer layer to have either a layer wall thermal and account for the conductivity of the skin or no thermal wall.
However, in all the cases, the temperature results seem to be too high inside the cavity, so I would like to ask if I am understanding something wrong in terms of the settings or the behaviour of the model and what the boundary and material conditions should be.
To begin with, you can use the Isovolume feature to check which areas have really high values. As you can see below, only few random areas have values> 70 degrees Celsius, which means this could be mesh related.
So after checking the mesh, I see a very coarse result of 44k cells. I suggest you try to re-generate it using the Automatic mode, and a level 5 for fineness. This should help a lot. Keep in mind that this guide can help you a lot with the mesh quality: How to Check and Improve Mesh Quality | SimScale
Thanks a lot for taking the time to check the model.
Okay, I see, so it is not connected to the assigned boundary settings for the simulation. And is this message also related to the mesh then?: “Detected divergence of the linear system solver. Consider lowering the relaxation factor for field T.”
I will try to refine the mesh and check how it affects the results. I don’t have a specific temperature range in mind, but I was assuming that a surface temperature of 70oC for the outer boundary layer didn’t seem very logical.
