solidification and melting with VOF: how to sink solid PCM as it melts?
Hi all, I have an enclosure spherical glass that is contain with phase change material (PCM) that is in its solid state initially. Heat will be applied above the melting temperature of the PCM around the spherical glass. During the melting process of the PCM there will be 2 phases present in the enclosure, the liquid and solid state. The solid PCM will drop to the bottom due to the gravitation force.
A 2d mesh was created and exported out from Gambit. In Fluent, I used Solidification and Melting function in my simulation with VOF. All Material properties of the PCM and glass have set accordingly. The Gravity was also enable with Y=-9.81 in the Operating Condition and the Variable-Density Parameters was also set to an operating density of PCM, that is 777kg/m^3. The Body force weighted under Discretization in Solution Control was also selected. After iterating for 200 sec, the values for the Liquid fraction was calculated but the simulation did not show the solid PCM sink to the bottom of the enclosure. How can I make the solid PCM to sink to the bottom as it melts? Thanks.:D |
I have done that. It's easy, you may have done a great mistake.
|
Quote:
Thanx |
It should be checked
|
1 Attachment(s)
Thank you for ur replay.
I have attached the full procedure and strategies. I am eagerly waiting for ur replay. Thanx |
Please can anyone help me???
|
Hi
Of course the solid region doesn't move, the moment source holds it back. I would try to add a source term for gravity and use (U(n)-(n-1)) instead of U, where n is the timestep. I think with this modifications the source your solid reagion will sink to the bottom. Greetings |
solidification and vof
I have a similar problem, can you send me your file . cas, to check on my side
|
Dear Saeed Sadeghi,
It seems that you are so happy knowing every thing and not helping others! |
Quote:
|
Hello dear friends,
I try to simulate melting process using VOF model. Pcm is solid phase initialy. Upper part of the solid there is air. They do not interpenetrate eachother. Natural convection is active. My primary phase air, secondary phase is pcm (phase change material). 85 % of volume is solid pcm initialy, 15% of total volume is air. I have some questions to be sure about my model. First, I do not define interface between air and pcm. Mean, There are 3 cell zone: air, fin ,pcm but I do not create name selection for this interface in meshing. Is it ok? Initial condition of Wholesystem temperature is 293K, but air temp is 300K. how to set air temp ? For cell zone condition, all cell zone (air, pcm, fin ) show mixture. Also for example, if I change pcm to phase2, all of cell zone change to phase 2 Only for secondary phase (pcm), I patch within fluent by adapt-region-mark then patch. Ok? Value of volume fraction should be 1 or 0 while patching?? I am confused. Initialy pcm is solid. For initial values phase-2 volume fraction is 0. Ok? I did some iteration for a short time to check model. For results, in graphics-countors I can not see total temperature for phase 1 and 2 only entalphy values I see for phase1 and 2. Also countors of solidification/melting it shows only mixture. Mixture show what? Any interaction should I use like surface tension etc. Thanks for your help. ı am in stuck. |
A large mushy zone constant enforces the velocity in solid phase to go to zero (or to pull velocity). If you set it to zero the solid phase will move with a velocity obtained from the solution of the N-S equations. However, it will deform like a liquid.
|
Quote:
|
Hi, would you please tell me more detail about how to acheive your idea to move the solid part, thx
Quote:
greetings Quote:
Quote:
|
Hi zbc19910519
Sure. Just use a very strong force, which is velocity dependent. For example 1e10*(vx-1) if you want your velocity component in x-direction to be 1. Just remember to tell the solver the derivative of this equation (I think it is called Source Coefficient in Fluent), otherwise the source will blow up your calculation. The solver uses the derivative to include the force in the implicit part of the solver (see Patankar). Cheers |
Quote:
Thank you again ! |
I didn't really get what you're trying to say. Can you show a picture or something?
|
Question
how I can define the parameters of the liquid phase of PCM in Fluent initially solid.
Please help!!!!!!!!!!!!!!!!!!!!!!!! |
Hi Hamada,
Depending on the melting temperature that you define for your PCM in Materials tabs and the initial temperature you set in the "Solution Initialization" tab, the initial state of PCM is automatically determined. For example, if your PCM melting point is 309<Tm<311, if you set the initial temperature to 308, the PCM will start from solid phase. If your initial temperature is 312 the initial PCM phase will be liquid, and if the initial PCM temperature is 310, the initial volume fraction of liquid phase will be 0.5. I hope this helps. Hamid |
Quote:
But i have a question, i have two data one for solid phase and the other for the liquid phase, for example there is a liquid density and a solid density, the same for the Specific heat capacity, and the Thermal conductivity. but in fluent, I can enter the parameters of a single-phase. please, what can i do? |
Quote:
Did you find any solution to this? If so, please help me. Regards Arash |
Quote:
Thanks in advance, Parsa |
Quote:
Do you know the reason for usual selection of an initial temperature of the PCM higher than the liquid temperature in the case of fully liquid or lower than the solidus temperature in the case of the fully solidified rather than using the exact PCM solidus and liquidus values in FLUENT? As I know the FLUENT considers the liquid fraction = 1 for the liquidus temp and 0 for the solidus temp. Is this just related to the accuracy of the results and just to ensure convergence? |
Quote:
|
solidfication & melting VOF function
can any one have have the steps for melting of ice in ansys fluent .. please share the steps
|
In order to give different thermal properties for solid and liquid phases. I think the following description may do the task.
If you take density not as a constant value but as a function of temperature (say piecewise linear function), you can set one value for below melting point (i.e. for solid) and one value for above. |
Quote:
In order to give different thermal properties for solid and liquid phases. I think the following description may do the task. If you take density not as a constant value but as a function of temperature (say piecewise linear function), you can set one value for below melting point (i.e. for solid) and one value for above. |
Quote:
I am trying to solve a similar problem,like ice melt. In order to solve the move of solid, i try to add a gravity force source to the y momentum. DEFINE_SOURCE(solid_source, cell, mix_th, dS, eqn) { real source; cell_t c; real x[ND_ND],r; real m_lg,a,b; Thread *pri_th, *sec_th; pri_th = THREAD_SUB_THREAD(mix_th,0); sec_th = THREAD_SUB_THREAD(mix_th,1); m_lg = 0.0; if (C_LIQF(cell,pri_th)==0) { m_lg = -9.8*C_R(cell,pri_th); } else { m_lg = 0; dS[eqn] = 0; } return m_lg; } but still it did not move. Can you help me? |
heat exchangers
1 Attachment(s)
Hi all, I want to do a project about Numerical simulation of heat exchangers with phase-change material with the help of Fluent.
For example, air from the inside of phase-change material between two parallel plates are passed, and We want more cool air to keep output As shown below: http://www.cfd-online.com/Forums/att...1&d=1478289496 What options should be activated and how that process performed?Please please help me , If it is possible…. thank you. |
Quote:
Hamada, You need to specify temperature dependent properties for your PCM by using a UDF. I suggest you consult UDF manual for "property" related UDFs. Hamid |
Quote:
Rami, If you set the initial temperature exactly equal to the liquidus temperature a very small temperature fluctuation due to round-off errors or any other numerical swing may result in dramatic change in the source terms related to the phase change and creates numerical instabilities. That's the only reason. Good Luck, Hamid |
Quote:
<i'm new in the domain can you help me thank you |
Quote:
That is useful.... Thanks |
All times are GMT -4. The time now is 01:41. |