Natural Circulation loops
 

Hello everyone,
I am modelling a rectangular natural circulation loop with isothermal heat source and heat sink and the rest of the surfaces adiabatic.This whole system is isolated from the surroundings.
I am using carbon dioxide as a fluid and employing NIST real gas model in the subcritical region of CO2. The steady state simulations are converging.

My problem is that when I go to Flux Reports, and click on mass flow rate,it shows negative value,also it shows some heat transfer from the adiabatic walls even though I have specified zero flux.Is this the heat transfer value to the fluid from the walls by planar conduction? I have specified uniform wall thickness throughout and enabled the "Shell Conduction" option.

Any help is appreciated.Thnx