Originally Posted by Gert-Jan (Post 734682)

You didn't read my post. I wrote:


"in a closed rigid compartment, density should not vary at all."


You can only use the bousinesq approach to include free convection if this is relevant.

Originally Posted by Gert-Jan (Post 734686)

Is the system stirred (forced convection) or do you only have free convection?

Originally Posted by oliveira1820 (Post 734684)

But why?

The density is 998 kg/m^3 at the beggining, but with the energy source it varies with time doesn´t it?

Originally Posted by Gert-Jan (Post 734690)

If you don't understand this, then you should not do CFD.

If your liquid mass remains the same, but your density decreases, then your liquid volume increases (I consider liquid as incompessibe). In real life your container will burst or increase in size as Glenn mentioned. In CFD your volume is fixed and rigid. So it cannot compensate or 'breath'. This leads to serious numerical troubles, because mass is not conserved. And that is killing your calculation immediately.

Originally Posted by oliveira1820 (Post 734836)

I noticed that the density only varies locally, the mean value is the same during all the simulation, so mass is conserved.

Originally Posted by oliveira1820 (Post 734836)

My doubt is, I´m heating the water with 1000W (not considering phase change or gas presence) and the pressure doesn´t increase when I use cosntant denisty and specific heat, but it increases allot with IAWPS water material.

I know that with the molecules vibrations increasing the force on the walls hould too (pressure), WhY DOSENT ICNREASES?

Originally Posted by Gert-Jan (Post 734837)

You might observe that mass is conserved on global scale. But on local scale it can be completely different. Already one element can screw up your whole calculation.



I don't know al the details of all your models. I would stay away of the IAWPS models and stick to temperature depedent water properties (cp, labda), but constant density. Then, outside CFX, determine the pressure using the average temperature inside your container, and the expansivity.

Bottom line, CFX is a flow solver where a lot of models can be added. But vibrating molecules...... Sure kinetic effects are somewhere incorporated in one of the submodels. But I am not sure what your idea and expectations are. I would contact your supervisor what to do.