I'm trying to simulate a heat pipe from a solar power plant in Fluent. The geometry is basically composed of two concentric cilinders made of a glass shield at the external part, a vacumm region (filled with a material similar to air) a steel shield, and an internal region filled with Therminol-VP1 (which its characetristics depends on the temperature of the fluid). The fact is that I'm introducing the therminol at a rate of 0.04 Kg/s and everything works well until I activate a UDF that computes the temperature Mass-Weighted-Average at the outlet of the Therminol pipe and uses the result of this calculation as Thermal boundary condition for the outlet. It works well with the UDF for a inlet mass flow rate of 0.2 kg/s, when it is reduced to 0.1 kg/s the simulation doesn't reach convergence and at 0.04 kg/s it stops with temperature divergence. Someone could give me some tips to apply the UDF as boundary condition and use 0.04 kg/s as boundary condition without having divergence?
Thank you
Most probably I think you need to change the convergence criteria for energy equation. It may work
A few things to do to achieve convergence:
1. loose the under-relaxation factor, then change it back at a later point
2. Change second-order back to first-order calculation. Keep it simple. Then later if it converges, you can switch to second-order again for a more precise result.
3. Conduct a mesh sensitivity analysis since it could be due to poor mesh that the divergence happens.
- Badges
- Science topic