Learning Forum

[Buzzman]

I have been trying to recreate this paper(https://ieeexplore.ieee.org/document/5535483) with various setting in Ansys but failed.nThe problem most probably will be in the material property section in fluent. The paper describes the properties as - In the present work, a melting interval of 315k–319k is used. The other parameters are: latent-heat 141 kJ/kg, sensible heat capacity in solid/liquid state 2.4/1.8 kJ/kg K, thermal conductivity in solid/liquid state 0.24/0.15W/m K, constant density of 900kg/m3 in the solid-state, dynamic viscosity of 3.42×10-3kg/ms in the liquid state, and thermal expansion coefficient 0.0009 k-1.nNow using various settings in the FLUENT the results does not match. material properties are in page 2. (https://drive.google.com/file/d/1Ga7OMO_7e2pAEMQiFxkcQcu7Sy3af7iG/view?usp=sharing)nI have made over 20 attempts using different solution schemes like SIMPLE, PISO. different under relaxation factors, mesh settings & momentum-energy settings. but none of them yielded result as indicated in the attached file. So the problem is most probably in material properties.nMay I ask for a minute of your time to give me a hint to the solution to the problem? Are the material properties I have given as inputs correct?..B. :also from the assumptions 4. The densities of the solid and liquid are equal (i.e. no volume changes); 5. The Boussinesq assumption is valid for free convection i.e. density variations are considered only in so far as they contribute to buoyancy, but are otherwise neglected; n

[Karthik R]

Hello,nAs Ansys employees, we cannot download attachments from the forum.nCan you share a screenshot of the material properties you are using?nAlso, are you using the solidification / melting model? Is there some natural convection in the paper you are comparing results from? Did they also use a Boussinesq approximation? What are you using as your operating density?nAlso, how far off are your results? nTypically, there could be several reasons why you are not able to verify your results with the literature - different boundary conditions, properties, model did not converge properly. You have to absolutely make sure that you are repeating everything that the authors of the paper are doing.nPlease provide a response to all of my questions and, as much as possible, we'll try to help you.nKarthikn

[Buzzman]

Thank you for replying. Actually, the paper does not provide all the necessary info that is why I am trying every possible combination. it says The problem most probably will be in the material property section in fluent. The paper describes the properties as - In the present work, a melting interval of 315k–319k is used. The other parameters are: latent-heat 141 kJ/kg, sensible heat capacity in solid/liquid state 2.4/1.8 kJ/kg K, thermal conductivity in solid/liquid state 0.24/0.15W/m K, constant density of 900kg/m3 in the solid-state, dynamic viscosity of 3.42×10-3kg/ms in the liquid state, and thermal expansion coefficient 0.0009 k-1. so I have given these data as an input.nnnThe model activated was energy, viscous(laminar), and solidification/melting. The heat transfer mode in different range of PCM is discussed. Melting in the top part of shell is under convection dominated mode, and conduction dominated mode takes part in the bottom of shell. The middle zone is dominated by both convection and conductionin the following image the top 2 images are from literature after 7 and 10 min. and my results are presented below. they did not match clearly.nnI ran this under SIMPLE scheme with 1st order upwind in both energy &momentum relationship . I have also tried it with other scheme like PISO. etc.nnnnNB. also how can I How to create or extract custom graph from FLUENT? e.g. Liquid fraction vs time? like the following image. XY graph in fluent only has dimensional vector in X direction, not time. nThank you!nnn

[Karthik R]

Hello,nAh, gotcha! nIf your intention is to replicate the results presented in the paper, I'd strongly recommend that you connect with the authors of the paper and clarify this. This would be the best way for you to move forward in the correct direction. My guess would be as good as yours here.nAlso, I'm not completely sure about the problem you are attempting to solve and the boundary conditions used in the paper. But, it looks like there is a stronger plume-like flow (based on the contour plot). Again, this is just a guess from my side. To be on the right path, I'd definitely write to the authors of the paper and confirm if you are on the right track.nKarthikn