# Melting model in fluent

 Register Blogs Members List Search Today's Posts Mark Forums Read

 May 5, 2014, 08:50 Melting model in fluent #1 New Member   Join Date: May 2014 Posts: 1 Rep Power: 0 I am simulating melting of paraffin wax in an enclosure with heat supplied from bottom (VOF active), all other walls adiabatic. Temperature, density and liquid fraction contours are as intended but i am getting velocity in solid regions and even in liquid region velocity is messed up. I am using viscosity and density functions as udf and gravity is turned on. I would be grateful if someone could suggest the possible solution. Please help. Thanks

 May 21, 2014, 09:58 #2 New Member   Nils G Join Date: Dec 2009 Posts: 2 Rep Power: 0 Hi, so far you described your problem very briefly, therefore I might miss some important details. What directly comes to my mind is that you might have a density that is temperature dependent in the solid region. IMHO this does not work / cannot work / should not work. The solidification / melting module in FLUENT tries to eliminate all velocities using brute force. But Continuity is more important. If the density changes (maybe decreases during heating), the solid will "spread out" in order to conserve the mass. In reality the shape would change, so your mesh should change to compensate that error. But as you mesh is not changing I strongly recommend using a constant density in the solid region. Best regards Nils

May 22, 2014, 03:28
#3
Member

Join Date: Dec 2012
Posts: 92
Rep Power: 9
Quote:
 Originally Posted by siegertyp Hi, so far you described your problem very briefly, therefore I might miss some important details. What directly comes to my mind is that you might have a density that is temperature dependent in the solid region. IMHO this does not work / cannot work / should not work. The solidification / melting module in FLUENT tries to eliminate all velocities using brute force. But Continuity is more important. If the density changes (maybe decreases during heating), the solid will "spread out" in order to conserve the mass. In reality the shape would change, so your mesh should change to compensate that error. But as you mesh is not changing I strongly recommend using a constant density in the solid region. Best regards Nils

Hi

I agree with Nils. I just have an addition to make. With constant density you will not have any buoyancy, so consider using the bousinesq approximation. I think there could be different approaches, too. For example you could try to integrate the density change in the melting zone and assign an appropriate velocity to the solid zone to change the shape of the solid zone.

Regards

 May 22, 2014, 03:44 #4 New Member   Nils G Join Date: Dec 2009 Posts: 2 Rep Power: 0 Hi, I think the bousinesq approximation "beer" suggested might work for you. If you got inlets that are able to compensate the increasing density, then I think you should be totally fine using a temperature dependent density in the liquid region. But, as I mentioned before, just use a constant density in the solid region. Regards Nils Last edited by siegertyp; May 22, 2014 at 03:44. Reason: forgot regards

 Tags fluent, melting, multiphase, vof