Hi:

I have a question regarding the cfx solver and I'm hoping someone can give me an answer

When I run a simulation in CFX with isothermal flow of air, I get a converged solutiuon. When I add a uniform heat flux to one of the wall boundary conditions but do not invoke buoyancy and use the results from the isothermal case as initial values, the momentum residuals change. I dont understand this as there is nothing in the momentum equations that should invoke a change in velocity. Does the CFX solver do something that I'm not aware of. Does anyone know the answer to this?

Jeff Berg

Jeff,

If you are adding a heat flux to a flow, this is going to alter the density of the fluid, which in turn will affect the fluid dynamics.

Cheers

Paul

Hi,

Which Air are you using in your simulations? In CFX there is Air at 25 C, Air at STP and Air Ideal Gas. Air at 25 C and Air at STP are treated as fluids with constant properties (including density), while Air Ideal Gas is obviously treated as an ideal gas (density is function of pressure and temperature).

When you change the Heat Transfer Model from Isothermal to Thermal Energy or Total Energy (which one are you using?) and add a heat flux to a wall boundary, obviously the fluid temperature will change. If your fluid is Air Ideal Gas its density will also change, and even if you are not adding a buoyancy term, the momentum equations will change as the fluid density makes part these and the continuity equations.

If your fluid is Air at 25 C or Air at STP, the fluid density is independent of temperature, and therefore I see no reason for the momentum equations and residuals to change, unless you haven't chosen the Initialisation options Automatic or Automatic with Value.

Regards,

Rui