Heat Transfer Analysis On PCM Based Heat Sink

International Journal of Scientific & Engineering Research, Volume 7, Issue 4, April-2016ISSN 2229-5518445temperature (e.g. in pure metals) or over a temperatureranges (e.g., in binary alloys). Instead of tracking the liquidsolid front explicitly, FLUENT employees an enthalpyporosity formulation. The liquid-solid mushy zone is treatedas a porous zone with porosity equal to its liquid fraction,and appropriate momentum sink terms are added to themomentum equations to account for the drop in pressurecaused by the presence of solid material. Sinks are alsoTABLE 840380831252998296029222900ºCCpTABLE 4TempMaterial property of Kconstant202.4Convergence settingsCpIJSERadded to the turbulence equations to account for porosityreduction in the solid regions. FLUENT uses volume controlapproach to solve the fluid flow problems. In finite volumemethod, flow domain is discretized into cells and analysis isdone by solving governing equations on control points onthe cells. The finite volume method represents and evaluatespartial differential equations as algebraic equations.The instantaneous continuity equation, momentum equationand energy equation for a compressible fluid can be writtenas:Momentum, mass and energy equations are monitored andthe solution is taken in such a way that the residuals areconverged to a value less than 10-6. . (1)Momentum equationGrid Independence studyDifferent configurations are made to get a grid independentsolution. In the steady 2-D problem the temperature valuesare plotted for various grid configurations and a gridnumber of 12500 rectangular grid is selected for the furtherstudy. In the case of 3D problems the previously obtained 2D grid is extruded along flow direction and discretization inthe flow direction is accordingly selected. A total number ofhexahedral of 625000 element used for 3D analysis.4Continuity Equation:NUMERICAL MODELING AND ANALYSIS . (2)Energy equation . (3)For a Newtonian fluid, assuming Stokes Law for monoatomic gases, the viscous stress is given by (4)4.1 Numerical SimulationFLUENT can be used to solve problems in fluid flowinvolving solidification and/or melting taking place at aWhere the trace-less viscous strain-rate is defined byIJSER 2016http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 7, Issue 4, April-2016ISSN 2229-5518 . (5)The heat-flux, q j , is given by Fourier's law: . . (6)Where the laminar Prandtl number Pris defined by: . . (7)To close these equations it is also necessary to specify anequation of state. Assuming a calorically perfect gas thefollowing relations are valid:446And but Ø 04.3Solidification ModelingFLUENT can be used to solve fluid flow problems involvingsolidification and/or melting taking place at one temperature(e.g., in pure metals) or over a range of temperatures (e.g., inbinary alloys). Instead of tracking the liquid-solid frontexplicitly, FLUENT uses an enthalpy-porosity formulation.The liquid-solid mushy zone is treated as a porous zone withporosity equal to the liquid fraction, and appropriatemomentum sink terms are added to the momentumequations to account for the pressure drop caused by thepresence of solid material. Sinks are also added to theturbulence equations to account for reduced porosity in thesolid regions.The enthalpy of the material is computed as the sum of thesensible enthalpy, h, and the latent heat, H: . (11)Where γ, C v, Cp and Rare constant.IJSERWhereThe total energy e 0 is defined by:h h ref (8)4.2AndFavre Averaged EquationsIt is not possible to solve the instantaneous equationsdirectly for most engineering applications. At the Reynoldsnumbers typically present in real cases these equations havevery chaotic turbulent solutions, and it is necessary to modelthe influence of the smallest scales. Most turbulence modelsare based on one-point averaging of the instantaneousequations. The averaging procedure will be described in thefollowing sections.dT .12)h ref reference enthalpyT ref reference temperatureCp specific heat at constant pressureThe liquid fraction, β, can be defined as.Let ϕ be any dependent variable. It is convenient to definetwo different types of averaging of ϕ:Classical time averaging (Reynolds averaging) (9) (13)Density weighted time averaging (Favre averaging): . (10)The latent heat content can now be written in terms of thelatent heat of the material:Note that with the above definitions, butIJSER 2016http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 7, Issue 4, April-2016ISSN 2229-55184.4 Analysis ProcedureThe modeling of flow domain has been completed usinggeometry and mesh building software, GAMBIT. Generalsequence of operation involved is:ModelSettingsSpace2D/3DSolverPressure basedTimeSteady / Unsteady, 1stOrder ImplicitViscousLaminarHeat TransferEnabledSolidification and Melting1)TABLE 5Shows the basic solver settings provided.IJSEREnabledImporting grid.Checking grid.Setting units.Define solver properties (steady, unsteady, 2D/3Detc.).5) Define Model (Solidification and heating , turbulentproperties)6) Define material properties (density, viscosityvariation etc.).7) Define operating conditions.8) Define boundary conditions.9) Initialization.10) Setting convergence criteria.11) Iterating until the solution converges.4.5ANSYS Fluent 14 for numerical solution are listed in thissection. The discretized geometry is imported into Fluent.Fluent will perform various checks on the mesh and willreport the progress in the console. It is needed to make surethat the minimum volume is a positive number. Theimported grid is checked and proper scaling is done. Therequired units are selected.4.5.1 General settingsThe general settings such as solver settings, details oftemporal discretization, properties of materials andequations required solving and additional physics requiredetc. are selected depend on the problem.Create full geometry and decompose into mesh ablesections.2) Give meshes required.3) Continuum and boundary attachment.4) Export Mesh.Analysis is done using FLUENT software. General sequenceof operation involved is:1)2)3)4)4475RESULTS AND DISCUSSION5.1 IntroductionThe result from the various analysis is reported in thischapter. The various cases such as fin only, fin fully filledwith PCM with forced convection etc. are reported one byone. The characteristics of PCM on cooling and heating areanalyzed and discussed. The cases with different velocityboundary conditions are compared and a strategy ofintermittent use of convective heat transfer is made. Thegrid, boundary conditions, geometry and numerical settingsadopted etc. are discussed in the previous chapters. Theresults are presented in the form of contours of temperature,liquid fraction and heat transfer coefficientFluent SettingsGeometry is created using GAMBIT. Two dimensionalmodel is created. Discretization is done by using mappedquad mesh with boundary layer on solid fluid interface.Gambit file is exported as .mesh format. The settings ofIJSER 2016http://www.ijser.orgHeat sink with fin only

International Journal of Scientific & Engineering Research, Volume 7, Issue 4, April-2016ISSN 2229-5518448Fig. 7. Temperature (oC) contour at 300 s time In this case anunsteady simulation is carried out. The temperature contourat 300 s is shown in the figure 5.1. The maximumtemperature along the heat sink with time is plotted in thefigure. Since the walls are adiabatic the supplied heat fluxincreases the temperature linearly. From the temperaturecontour it is clearly visible that the isothermal lines areperpendicular with the adiabatic walls. A temperatureincrease in the rate of 0.03 unit.5.2 The heat Sink with Fin Fully Filled with PCMFig. 8. Grid used for fin with fully filled PCMIn this case an unsteady analysis is carried out for the heatsink with fully filled PCM. Boundary conditions includesheat flux value of 200W/m2 applied at bottom of PCM for3500s heating and 3500s cooling. All other walls are treatedor assumed to be as adiabatic walls. Domain interfacebetween PCM and air is considered as wall to avoid thematerial transfer. The initial condition set for entire domainis 180C. The time at which entire PCM changes to liquid isfind out (liquid fraction becomes one). The cooling is alsostudied by the implementation of reversing of heat fluxLocationBase of finGeometric center of PCMTop of PCM5.2.1 Maximum Temperature change with timeThe below figure shows the temperature verses time graphin both heating and cooling step for different locations P1and P2 as in the figIJSERNotationP1P2P3direction. The temperature characteristics at various locationof the computational domain is studied. The characteristicson PCM material on cooling and heating process is analyzed.It is observed that for a time of 53 minute the temperature ofthe domain can be controlled by PCM. At around 3200 secthe entire PCM is melted and further accommodation of heatin the PCM is not possible. After that period the domaintemperature increases linearly.Fig. 9.Table 6The different location selected for temperature comparisonthe figure shows the temperature variation with time fordifferent locations as mentioned in figure. The variation inthe slop indicates the presence of PCM up to 3200 sec. Thespecific heat of PCM various with temperature, hence attime above 100s the PCM start phase change and which ishaving large heat storage capacity. After complication of thephase transformation the heat storage capacity becomesIJSER 2016http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 7, Issue 4, April-2016ISSN 2229-5518449linear in nature. In the cooling process the pcm materialreleases heat energy and transformed into solid phase.Fig. 12.Fig. 10. Temperature variation with time for differentlocations considered.Contours of Temperature at various T1 Figure shows thecontours of temperature and liquid fraction at 500 sec in thefin with fully filled PCM configuration. The PCM in solidface started liquefying near the interface region. The cornerportion liquid for because of the large interface area in thecorner region. The temperature contour is seems to beperpendicular to the top wall of the PCM, indicate aninsulated boundary condition. The maximum temperaturereaches at 19.10C at 500 sec.Figures. Contours of temperature and liquid fraction at 500sec Figure shows the contours of temperature and liquidfraction at 2200 sec in the fin with fully filled PCMconfiguration. Above 50% volume of PCM is having a liquidfraction more than 0.5. The maximum temperature reachesat 22.60C at 2200 sec.IJSERFig. 13.Fig. 11.IJSER 2016http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 7, Issue 4, April-2016ISSN 2229-5518450Fig. 16. Contours of temperature and liquid fraction at 3500sec5.3Heat sink with fin half filled with PCMIn this section the result of the half- filled PCM with forcedconvection is studied in detail. The domain created is threedimensional. The figure shows the transitional periodicmodel of grid used for this study.Fig. 14. Contours of temperature and liquid fraction at 2200secFigure shows the contours of temperature and liquidfraction at 3500 sec in the fin with fully filled PCMconfiguration. It can be seen that the minimum temperaturein the PCM domain is 25.20C (above the melting limit) andhence the entire portion of PCM is converted in to liquid andat/after that the variation in the temperature of systemchanged linearly.IJSERFig. 17. Computational domain used for half-filled PCMwith forced convectionFig. 15.The below figure shows the velocity vector in a crosssectional plane for an inlet air velocity of about 0 m/s. Anatural convection phenomena can be seen. The flowdirection indicates two major loop which meets at the centralportion of the domain with a high velocity magnitude. Thereexist 4 small loops along the corner portion of plane. Themaximum velocity of air is reached at about 0.15 m/s due tothe buoyant force.Fig. 18. Vector of the cross-section on air flow for an inletvelocity of zero.IJSER 2016http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 7, Issue 4, April-2016ISSN 2229-5518451The velocity vector on the cross section of air flow domain isdepictured in the figure below shows two recirculationloops. A temperature driven with a maximum velocity of0.14 m/s is obtained in the flow.Fig. 21. Air inlet velocity 100 cm/sIJSERFig. 19. Air inlet velocity 10 cm/sFig. 22. Air inlet velocity 150 cm/sTemperature contour at locations of inlet, 2,4,6,8 mm frominlet, outlet plane.Fig. 20. Air inlet velocity 50 cm/sIn the case of the forced convection flow, steady flowanalysis is carried out for an inlet velocity of about 0.1 m/s- 2m/s. The Figure below shows the temperature and liquidfraction for various inlet velocity at various axial locationssuch as 2, 4, 6, 8 mm from inlet and outlet boundary. It canbe noted that the air becomes hot as it flow through the path.As the flow velocity increases the maximum temperature isIJSER 2016http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 7, Issue 4, April-2016ISSN 2229-5518452reduced.Fig. 23. Temperature at the interface between PCM and airat different axial location 1,2,3,4,5,6,7 mm from inlet for airvelocity of 10cm/sat 7 mm from the inlet.Fig. 25. Volume averaged liquid fraction for various axialvelocitiesFigure below shows the temperature plot at the interfaceregion of PCM and air at various axial locations series1 from1mm from the inlet and series7 locations. Heat transfer ismaximum at inlet region of the flow. In the figure belowmass flow averaged temperature of the outlet for differentvelocities are plotted. It can be noted that increasing ofvelocity above 1.5 m/s doesn’t give any further temperaturechange (reduction) in the system. The velocity vs liquidfraction plotted in the figure below also proves this point. Ifthe velocity range is above 1.5 m/s the change in liquidfraction with velocity is less.The unsteady analysis is done for various velocity inlets.Initially the analysis is carried out for a time up to volumeaveraged liquid fraction of one with zero velocity at inlet.Later various analyses for various velocities are carried outby using the first result. Figure (a) and (b) shows theminimum domain temperature and liquid fraction of PCMalong with the time. The effect air velocity up to 0.2m/s isconsidered as negligible and further increase in velocitystabilizes the temperature and liquid fraction.IJSERFig. 24.Fig. 26. Minimum temperature in PCM region with time fordifferent inlet velocity (m/s).Mass flow averaged and maximum temperature in thedomain for different axial velocitiesIJSER 2016http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 7, Issue 4, April-2016ISSN 2229-5518LF 01050830700630500453Fig. 28. Heat transfer coefficient along the flow direction foran inlet velocity of 1m/sIJSERHeat transfer coefficient along axial distance at variouslocations is shown in figure. The heat transfer coefficient atthe center location of the air flow passage is higher than thatat the corner locations. The top air PCM interface is havinghigher than that at bottom air fin interface. This is due to thedifference in temperature gradients in their correspondinglocations.TABLE 6.Time requirement for solidification for different inletvelocitiesFig. 27. Liquid fraction in PCM region with time fordifferent inlet velocity (m/s).The time requirement for controlling the temperature of thePCM to various liquid fraction levels for various air inletvelocity is listed in the Table below. Based on the usage ofthe device the air flow velocities can be varied as per therequirements. From the tabular result it is clear that to obtaina liquid fraction of 20 % it is necessary to set a minimum airvelocity of about 1.6 m/s similarly for 30% liquid fraction it isabout 1.2 m/s. Increasing velocities above 1.6m/s gives onlya very small improvement in the performance. Hence theseresults can be used as an easy-to-use design guidance linefor the PCM based heat sinks, in terms of the forcedconvective conditions.IJSER 2016http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 7, Issue 4, April-2016ISSN 2229-55186. CONCLUSIONSThe PCM based cooling techniques have a great potentialapplication in electronic devices. In this work study andanalysis of PCM based heat sinks are carried out. Numericalanalysis is done by using FLUENT 14.0. The works includesthe study on characteristics of PCM and its applications arecarried out. Here we conducted three configuration of heatsinks such as fin filled with PCM material, fin filled with halfPCM material, fin only .The characteristics of Heat sink withPCM is analyzed in both melting process and solidificationand find out the time in which PCM controls temperature upto 2400s for full filled PCM. Performance of heat sink forcontinuous operation is carried out for different air flowconditions. The time requirements for control thetemperature of PCM for different forced convectionconditions to various liquid fraction levels are tabulated,which can be used as an easy-to-use design guide line forPCM based heat sinks with forced Engineering, Massachusetts Institute of Technology,77 Massachusetts Avenue, Cambridge, MA 02139,USA, Faculty of Engineering and Applied Science,University of Ontario Institute of Technology, 2000Simcoe Street North, Oshawa, Ontario, Canada.QIU Yifena, JIANG Nanb, WU Weia, ZHANGGuangweia, Heat Transfer of Heat Sinking Vest withPhase-change Material,China Aerospace Life-supportIndustries, Ltd., 2011.Paisarn Naphon, Laminar convective heat transferand pressure drop in the corrugated channels,International Communications in Heat and MassTransfer 34 (2007) 62–71Mónica Delgado*, Ana Lázaro, Javier Mazo, JoséMaría Marín, Belén Zalba, Experimental analysis of amicroencapsulated PCM slurry as thermal storagesystem and as heat transfer fluid in laminar flow,Applied Thermal Engineering 36 (2012) 370-377Mohammad M. Mansoor , Kok-Cheong Wong ,Mansoor Siddique, Numerical investigation of fluidflow and heat transfer under high heat flux ications in Heat and Mass Transfer 39(2012) 291–297S V. Garimella, A Novel Hybrid Heat Sink usingPhase Change Materials for Transient ThermalMan

A phase-change material (PCM) is a substance having a high heat of fusion. It is capable of storing and releasing large amounts of energy. It absorbs heat from the electronic component and melts. It is re-solidified by dissipating heat to the surroundings when device is not functional. PCM