[Nukengine]

Hey all,

I am relatively new to CFD.

I am simulating natural circulation in a small box (~6inx6inx16in tall box) with 2 heater rods placed inside along the height and gravity. The rods have 'heat source' thermal conditions; one rods is at 400W, the other at 250W. The top wall of the box has a heat source condition as well of -650W. The fluid in question is water (using a polynomial in T density because the IAFW water thing generally gave me errors with T and P out of bounds).

I have run the simulation with various physics continua and solver settings and finally have a solution that appears realistic. I have also refined mesh and I don't think that is an issue. The problem is that under all of the continua and solver settings I have tried (whether turbulent/laminar or steady/unsteady) my energy residual flattens out at just over 10. The XYZ momentum residuals are 10^-5, continuity 10^-3. Under turbulent models (K-Omega) TKE is about 10 and SDR is about 0.01.

My question may simply be answered by telling me that a high energy residual in a natural convection simulation in STAR CCM is normal and acceptable. Otherwise I would appreciate some opinions; I can list more simulation parameters.

Thanks!

[siara817]

I do not think that it is normal. I think there is a problem. Are you sure that the solution is converged? check if the solution is converged