Difference between Simulation and Calculated solution?

Gregor Keil

Hi All,

I had some issues with setting up a simulation that features mostly radiation heat transfer in a 300 K environment.

After playing around a bit, I set up a DO Transient (flow) simulation with SimLab (2023.1) using AcuSolve as a solver. For my test case, I used two small disks facing each other (Dimensions: radius = 5 mm; thickness = 1 mm and spacing = 10 mm, made out of tungsten). By keeping one of the disks at a constant 2000K (I used the "Wall" boundary condition here) and enclosing both disks in one big solid vacuum (modeled as a vacuum with very low density and thermal conductivity, high heat capacity, and 0 for absorption coefficient). I then applied an external-radiation-surface on the outer surface of the vacuum body (emissivity of 1 to an environment of 300K) and a radiation-surface-wall on the surfaces of both disks (emissivity of 0.3). I ran that simulation for 60 s, and the result was that after 1 s, the target disk was heated to ~352K, and after ~30 s, it reached a constant ~1150K.

To verify this simulation, I also calculated the problem analytically using the simple Lumped Capacitance Analysis for radiation only (setting dT/dt = deltaT/delta t) with delta t = 1 s. For this, I got ~318K after 1 s, and it leveled off at ~1035K after about 100 s.

So, why is the radiation heat transfer in the simulation higher than in the calculation? Could it be that the "cooling" of my disk is not implemented correctly?

best regards

Gregor

ydigit

Could you try running without any external BC to the vacuum volume? 

That way all heat will go from one disk to the other and rule out heat flux across external walls.