Vibration Enhanced Heat Transfer Of A Hot And A Cold Micro Tube With Staggered Arrangement In An Enclosure

In recent years, with the rapid development of the micro system and thenanotechnology, micro scaled heat exchangers (MSHE) have been widely used inelectronic components, micro-electronic mechanical system, advanced biotechnologyand tiny medical equipment and other fields, and their investigations haveincreasingly been given attentions. It is believed that micro scaled heat exchangertechnology is a future generation in heat transfer enhancement, which could play avery important role in thermal energy conversion and utilization, especially for thoseusing low enthalpy renewable energy resources.MSHE have many advantages over conventional heat exchangers, such assmaller size, higher heat flux, higher performance in mechanics, saving materials andspace, more flexibility. Since that generally MSHE holds a high ratio of heat transferarea to its volume, the passive ways, e.g. extended surface, to enhance its heat transferperformance become unnecessary. On the contrary, that how to organize the two side(hot and cold) fluid flows so as to overcome its intrinsic defects of flowmaldistribution, uneven temperature distribution, fouling and erosion damage becomesignificant to guarantee its high heat transfer performance and reliability.The believed scientific findings of this article are as follows: First, a new type ofmicrotube heat exchanger is put forward, i.e. cold and hot microtubes staggered in anadiabatic circular enclosure filled up with third heat transfer fluid, and the simplestone of this type, which has a pair of cold and hot tube immersed in cylindricalenclosure filled with water has been studied in detail; Second, the vibration ofmicrotubes relative to the cylindrical enclosure has been applied, and the heat transferenhancement mechanism between the micro cold and hot tubes are investigatedexperimentally. A dimensionless formula for the mixed heat transfer of microtubes ina circular enclosure is proposed. A performance assessment criterion, which is definedas the amount of enhanced heat transfer rate with respect to the theoretical powerinput, is also provided. Third, based on our numerous experiments, the correlations ofNusselt number, Nu, at various situations with the consideration of three-dimensionaleffect for natural convection outside micro cold and hot tubes are proposed.First of all, the natural convection heat transfer around a pair of hot and coldhorizontal microtubes in a relatively large concentric horizontal outer tube under thestationary state were investigated both experimentally and numerically. The numericalsimulation was implemented using Lattice Boltzmann Method (LBM) of convectionalBGK model. The heat transfer characteristics in micro tubes and the naturalconvection heat transfer performance between the cold and hot tubes at stationary state were studied in detail. The brief influencing factors include the relative positionof cold and hot tube, flow rate inside tube, temperature difference of the cold and thehot source and so on. The experimental results were compared with those of thenumerical simulated. According to the present work, some conclusions can be drawn,which are as follows:(1) Using water as heat transfer fluids both inside and outside microtubes, theoverall heat transfer coefficient of this configuration ranges from400-800W/(m2·K).According to the tests with orthogonal experimental design of mixed levels, it can beconcluded that the relative position of cold and hot tube has the greatest influence onthe overall heat transfer coefficient, the temperature difference between hot and coldsources takes the second place, and the flow rate inside microtubes is the last throughthe method of range analysis. The increase degree for the overall heat transferperformance changes with the temperature difference of cold and hot sources, and isslightly different under different tube arrangements, that hot tube under cold tube isthe optimal, hot tube above cold one is the second, and both tubes orientedhorizontally is the worst. It is noted that the thermal entrance effects of micro tubeshould be taken into account on the heat transfer inside micro tube. The frictioncoefficient inside micro tube are well consistent with the prediction values of theconventional correlations, but the critical Reynolds number of laminar transition toturbulence changes from2300to2000.(2) The experimental results, while the microtubes are at stationary state, alsoindicate that the Nusselt number of natural convection Nuoaround a pair of hot andcold tubes in the circular enclosure is not only related to the Rayleigh number Ra1defined by the temperature difference between the hot and cold tube transversely, butalso to temperature difference longitudinally along with the cold and hot tubes, i.e.two ends of the cavity, in addition to the Prandtl number Pr at the averaged fluidtemperature in the cavity. Nuoincreases with Ra1under different arrangements of coldand hot tubes, but decreases with Pr. For the two cases of hot tube above cold one,and the reverse Nuodecreases with increasing Ra2, and both trends are basically thesame, however, Nuois not sensitive to Ra2for the case of cold and hot tubes orientedhorizontally.The correlations of Nusselt number considering the three-dimensional effect fornatural convection outside the micro cold and hot tubes are obtained using the leastsquare method, and their deviations range from-10%to+10%and the absoluteaveraged errors are between4.3%and14%.The comparisons between the experimental and the numerical simulated resultsindicate that they are in good agreement for the case of oriented horizontally tubearrangement, however, not very well for the other cases. This is most possibly due to the two-dimensional limitation for the numerical simulation, while the actual naturalconvection in the enclosure is, in fact, three dimensional, and another reason is theheat/cold leakage through the enclosure walls even they are well thermally insulatedby using advanced aerogel materials with0.02W/(m·K) or even less in thermalconductivity.Second, the enhanced heat transfer mechanism for the periodically vibrating hotand cold microtubes immersed in a third fluid is experimentally studied. The overallheat transfer coefficient for the mixed convection between the hot and the cold tubesat various situations are obtained. Compared with the limited relevant works in theprevious literature, in respect of the influences of vibration frequency, amplitude,arrangement of cold and hot tubes, temperature difference of cold and heat sourcesand their flow rates etc. on heat transfer, a few conclusions following can be drawn:(3) The overall heat transfer coefficient of this vibration system ranges from500to1200W/(m2·K). The mixed convection Nusselt numbers for all cases increasewith the vibration frequency while the tubes are vibrated up and down and thefrequency is low; however, their increments are imparity. It is the best while the tubesare oriented horizontally at the same level, and the worst while the hot tube is belowthe cold one. The Nusselt numbers for all of the three arrangements increase with theamplitude while the vibration frequencies are the same.(4) A dimensionless formula for calculating the heat transfer coefficient betweena hot and a cold vibrating tubes is proposed. It is noticed that the ratio of vibratingtube heat transfer coefficient to that of the tube in stationary, h/ho, appears as anS-shape curve with respect to a dependent defiRe v(A/D)0.4Pr0.6ned asRa0.26, which issimilar to the function firstly proposed by Lemlich and Anandha in1964. It showsthat when the vibrational disturbance is strong, the relative effect on a given naturalconvection is strengthened, and when the vibrational disturbance is weak the relativeeffect of vibration is weakened as well. For the vibrational disturbance too weak incomparison with the natural convection, no significant increase in heat transfercoefficient is observed. This is confirmed by our experimental results, which showthat below an abscissa of roughly20the increase in heat transfer coefficient is quitesmall and h/horemains unity. However, h/hoincreases remarkably when the abscissanumber is larger than about20. The absolute averaged error of the values h/hocalculated with those of experimentally measured of is6.2%.Comparing our experimental results with those in the previous literature for thecase of single tube, we found that the best vibration strength effect (extrapolated) isabout6.16, which is less than the value of15.49of single vibrating heat transfer tube. The S-shaped curve obtained in our case is relatively flat and the starting point ofrising h/hois located at20-30of the abscissa number, which also lags the value in theliterature2-3times. The mainly two possible reasons are: first, the fluid flow and heattransfer processes are different from those of single heated/cooled tube; second, boththe vibration frequency (0-10Hz) and amplitude (0.2to0.8mm) in our case are muchsmaller than those values of20-225Hz and0.64-22.4mm, respectively, given in theliterature.(5) In order to give a comprehensive evaluation of heat transfer performance forthis vibrated microtube heat exchanger, we define an index, Sy, of the vibrationstrengthening heat transfer, which equals to the ratio of the heat transferred withvibration to the theoretical input power, Psr, for overcoming the fluid resistance due tovibration. The correlated formula of the index, Sy, with respect to the input power, Psr,shows that the enhanced heat transfer could be1.7×104W at input power Psr=1W.However, the income of transferred heat decreases with the increase of the vibrationinput power in our experiment range.In addition, natural convection around a pair of hot and cold tube in a closedsquare cavity under constant wall temperature boundary and adiabatic boundary werenumerical simulated by multi-Lattice Boltzmann Method. Several influencing factorson the heat transfer performance were specifically discussed, including Rayleighnumber, arrangement of cold and hot tube, boundary conditions and so on. Then thenatural convection around two pairs of hot and cold tubes in a closed adiabatic squarecavity was carried out, taking the Rayleigh number, the cavity rotation angle, tubespacing of the two pairs of hot and cold tube into account. The natural convectionaround a pair of hot and cold tubes in an adiabatic circular enclosure was numericallysimulated by using direct force embedded boundary Lattice Boltzmann Method. Theeffect of Rayleigh number and the arrangement of hot and cold tubes were especiallydiscussed. The Nusselt number correlations were proposed and compared with theexperimental results.Finnaly, some proposals and the improvement needed for the future study aregiven according to the problems and the insufficiencies found in this study.