Eulerian 2 phase boiling
 

I am trying to simulate boiling using the Eulerian multi-phase treatment. I am running a transient analysis with a domain of liquid water initially at 372 K and secondary vapor phase with an initial volume fraction of .01 and temperature of 373 K. At the base of the domain there is a small heater of fixed temperature: 383 K. At the top, there is a pressure boundary of 1 bar and the liquid is initially at 1 bar. The zero gradient switches for both phases are turned on across the pressure boundary. I am using the user subroutine code for flunj.f and sorent.f found in the standard demo problem for the E2P boiling model. When I run the analysis with a constant liquid density of 958 kg/m^3, it runs, but natural convection does not occur in the liquid above the heater because the liquid density is constant. In addition, the volume fraction of the vapor never truly rises, even though there is liquid directly above the heater at greater than Tsat. Do you have any suggestions as to why there is no vapor production.

If I run the analysis with an Isobaric density change, (Beta = 7.501e-4, Dref = 1000kg/m^3), the pressure of the liquid at the base of the fluid is less than the pressure of the liquid at the top. Any ideas as to why the pressure at the base would be lower instead of higher when I change the density calculation? Even though the pressure change is small, it should not occur, because the pressure at the base of the fluid should actually be 103675 Pa (P = Po + row*g*h, where h = .25 m), not 101200 Pa. Could the problem be in m choice of either Beta or Dref? If so, what values would you suggest?

Any suggestions for the above problems? Perhaps a sample flunj.f or sorent.f files that are more appropriate for the above situation than the code listed in the standard demo problem for the E2P boiling model?