- Email: [email protected]
Experimental passive electronics cooling: Parametric investigation of pin-fin geometries and efficient phase change materials
International Journal of Heat and Mass Transfer 115 (2017) 251–263
Contents lists available at ScienceDirect
International Journal of Heat and Mass Transfer journal homepage: www.elsevier.com/locate/ijhmt
Experimental passive electronics cooling: Parametric investigation of pin-fin geometries and efficient phase change materials Muhammad Junaid Ashraf a, Hafiz Muhammad Ali a,b,⇑, Hazrat Usman a, Adeel Arshad c,d a
Department of Mechanical Engineering, University of Engineering and Technology, Taxila, Pakistan Center of Research Excellence in Renewable Energy (CoRE-RE), King Fahd University of Petroleum & Minerals (KFUPM), Dhahran 31261, Saudi Arabia c Department of Mechanical Engineering, HITEC University, Taxila, Pakistan d Fluids & Thermal Engineering Research Group, Faulty of Engineering, University of Nottingham, Nottingham NG7 2RD, UK b
a r t i c l e
i n f o
Article history: Received 28 March 2017 Received in revised form 23 July 2017 Accepted 24 July 2017
Keywords: Phase change materials Thermal conductivity enhancers Staggered and inline arrays Circular and square cross-sectional pin–fin heat sinks Enhancement time Enhancement ratios
a b s t r a c t This experimental investigation focuses on the optimization of passive cooling system using extruded finned surfaces with phase change materials (PCMs) as the thermal conductivity enhancers (TCEs). The study develops comparison between fins of circular and square cross-sectional area, made of aluminium. Further classification is done in configuration in terms of staggered and inline arrays. The volume fraction of fins is kept constant at 9% of total volume of heat sink. The purpose is to single out the better arrangement with and without PCM. Six PCMs of varying phase change temperature and heat capacities, namely Paraffin wax, RT-54, RT-44, RT-35HC, SP-31 and n-eicosane are selected for thermal conductivity enhancement. The volume fraction of PCM is also constant at 90% of the heat sink volume, giving a 10% volume for expansion after melting. Moreover, power levels are used in a range of 4–8 W with an increment of 1 W. The analysis was carried out on graphical trends produced and explanations were given accordingly. The most effective PCMs were also discussed considering their enhancement time, enhancement ratios and other material properties. Finally, the results were justified by the scientific knowledge and found in compliance with the work of famous researchers as well. Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction Thermal control of modern electronics circuitry has become increasingly complex and it is very essential to ensure its reliability, durability and user comfort. Researchers are always in search of finding new effective solutions for thermal management of electronic devices. The choice of cooling techniques depends on number of factors. Active cooling and all other techniques have proven to be incapable of fulfilling the demands of modern electronic industries as it consumes power itself. So, passive cooling techniques employing PCM based heat sinks are used widely in these portable devices like computers, mobile phones, personal digital assistants, laptops and so on. Effective thermal management by incorporating PCM based heat sinks in these devices have led to increase in its functionalities, reliability, less probability to internal damages and failure, ultimately to stretch their useful life. Many studies both experimentally and numerically have been conducted to offer a deep insight into thermal management of electronic devices using PCM based heat storages. ⇑ Corresponding author. E-mail addresses: [email protected], [email protected] (H.M. Ali). http://dx.doi.org/10.1016/j.ijheatmasstransfer.2017.07.114 0017-9310/Ó 2017 Elsevier Ltd. All rights reserved.
Husseinizadeh et al. [1] examined both experimentally and numerically PCM based heat sinks to study the effect of various configurations of internal fins. The PCM RT-80 was filled in different heat sinks of constant overall dimensions. From the results obtained it was seen that by increasing number of fins, fin height and input power level had improved the thermal performance appreciably while increasing fins thickness had only marginal improvement. Similarly, Pakrouh et al. [2] presented numerical method for geometric optimization of pin finned heat sinks by coupling Taguchi and simulation method. For optimization, the effect of all critical parameters involving fins’ number, thickness, base thickness, height as well as PCM percentage were explored. All heat sinks filled with RT-44 and critical temperatures 50 C;60 C;70 C and 80 C were selected for analysing the results. The 2 mm thick fins performance was highest for 50 C while 4 mm thick fins performance was highest for all temperatures. Base thickness contributed less than the other parameters. Mahmoud et al. [3] investigated experimentally the effect of honeycomb structure in heat sink to compare its performance with finned heat sink. Six different types of PCMs and total six heat sinks, one with single cavity, two designed with parallel fins, two with cross arrangement and one with honey comb inserts were tested at
252
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
Nomenclature Roman symbols Symbol Quantity (Unit) PCMs Phase change materials SPTs Set point temperatures (°C) T PCM Melting temperature of PCMs (°C) T Al Melting temperature of aluminium (°C) t Thickness of fin (mm) h Height of fin (mm) vf Volume of the fin (mm3) Vf Total volume of fins Nf Total number of fins Total working volume of heat sink (mm3) VS m Mass of the PCM (kg) q Heat flux (kW/m2) tk Latent heat phase duration (sec) T Thermocouple inside the PCM
power level in the range from 3 to 5 W. The obtained results showed that increasing number of fins, insertion of honey comb and inclusion of low melting temperature PCM at higher power level had enhanced significantly the operation time of heat sink. Fins had significant role in improving thermal performance of heat storage unit. Recent findings considered identifying optimum distribution of these TCE (fins) in terms of dimensions and shape. Saha et al. [4] carried out their research to investigate the effective way of distributing fins in heat sinks i.e. to find the optimum volume fraction of TCE which maintains a low temperature of any component. Using n-eicosane as PCM in aluminium made heat sinks. Two types of fins (plate-fin and pin-fin) were analysed in heat sinks with base dimension of 42 42 mm2 and fins height of 25 mm. The case of 8% TCE volume fraction of heat sink or base plate was reported to give best results. Regarding fin dimensions and shape it was concluded that the large number of small cross-sectional area pin fins performed better. Baby and Balaji [5] experimentally investigated three different geometries of heat sinks employing different no of TCE. All heat sinks were made of aluminium employing 33, 72,120 number of pin-fins of corresponding volume fractions were 4%, 9%, and 15% respectively. The effectivity of volume fractions of TCE was examined at power level 4–8 w using n-eicosane with varying volume fractions of 0.3, 0.6 and 1. For SPT of 43 C and 8 W power, the highest enhancement factor of 21 was obtained for heat sink with 72 number of fins. Also, the performance was seen to be strongly dependent on PCM volume fractions instead of TCE volume fractions. The effect of orientation on thermal performance of porous matrix filled heat sink was investigated by Srikanth et al. [6]. Tracking system was employed to change the orientation of copper metal foam matrix embedded Al-heat sink filled with n-eicosane. The experimental results obtained in terms of enhancement ratios showed that the heat transfer enhancement was only comparable and effect of orientation has no significant impact on the mentioned heat sink. Experimental work performed by Fan et al. [7] determined the effect of internal fins and melting temperatures on performance of PCM based heat sinks under pulsating heat loads. Two organic PCMs (n-eicosane and 1-hexadecanol) having different melting temperature were tested in prototype heat sinks at different power levels and it was concluded that the PCM with higher melting temperature had resulted in improved thermal performance of electronic devises comparatively. Six heat sinks designs were tested by Mahrous [8] to investigate the effect of fins arrangement and fin number, on thermal performance. Identical plate finned heat
W H t cr
Thermocouple at side wall of heat sink Thermocouple at base of heat sink Time to reach critical temperature (sec)
Greek symbols Volumetric fraction of PCM wPCM mPCM Volume of PCM (m3/kg) qPCM Density of PCM (kg/m3) qAl Density of aluminium (kg/m3) ePCM Enhancement ratio at PCM kPCM Latent heat of PCM (kJ/kg) C PCM Heat capacity of PCM (kJ/kg) C Al Heat capacity of aluminium (J/kg) Thermal conductivity of PCM (W/m K) kPCM kAl Thermal conductivity of aluminium (W/m K) A Surface area ratio
sinks with four heat sinks of parallel fins arrangement, one with crossed fins and one with no fins were tested. A constant fin thickness of 1 mm and constant power level 4.84 W was used and it was concluded that the parallel and crossed fins arrangement both had nearly comparable performance but performance of all fins heat sinks was superior to that without fins. Numerical study carried out by Levin et al. [9] explored the effect of fin’s length, height and number for the optimal PCM percentage to be considered in designing latent heat thermal management system (LHTMS) for electronic devices. From the results, it was concluded that the optimal percentage of PCM is dependent on number and height of fins, heat flux and the difference between liquidus and critical temperature. Thomas et al. [10] carried out numerical study for the design of PCM based heat sink for average dimensions of a smart phone. Analysis were performed using ANSYS FLUENT 14.0 by providing constant heat flux to the base of heat sink with power input ranging from 4 to 6 w. Eicosane was selected as PCM for designed heat sink. High performance was obtained when PCM fraction had reached to its maximum. Selection of thermal performance enhancement method has a vital role in thermal management of components. Nagose et al. [11] used combined genetic and conventional simulation to get optimized configuration of heat sink which kept the temperature of microprocessor in acceptable limits. The designed heat sink consisted of fin array with depth equal to heat sink and a heat spreader inside the heat sink. From the results obtained with the assumption of constant heat flux from electronic device correlations were proposed relating heat sink operational time to the PCM fraction, heat sink depth and heat spreader thickness. It was found that the optimal spreader thickness was 2.5% of heat sink depth. Hatakeyama et al. [12] conducted their experimental and analytical research on PCM based transient cooling module employing pin fins. The setup used by them consisted of test module, heat spreader, chamber, substrate, power source, Data logger, three heat sinks with round fins of different number and diameter. The paraffin volume fraction used, was 81% of total space. Thermal network model of module was developed capable of predicting temperature transients. This model was reported to be effective design tool for thermal management of electronic devices. The research by Hassan et al. [13] investigated and compared the performance of three different types of PCMs namely salt hydrate, paraffin wax and milk fat in PCM based finned heat sink system. Experiments were conducted on finned heat sink at power input ranging from 4 W to 10 W. The conjugate heat transfer model was developed for each
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
PCM system that had predicted the effect of thermal conductivity, density, melting range of each PCM on heat sink performance. From the results obtained milk fat PCM was reported to have lower performance comparatively. Increased thermal conductivity and density of PCMs had resulted in increased thermal performance but was found to decrease for wider melting range. Fok et al. [14] performed experimental study on the application of phase change material (PCM) as coolant in portable hand-held electronic devices at power levels ranging from 3 W to 5 W, for extensive or light working conditions. The results summed up that TCE pin fins with PCM was feasible option for cooling hand-held electronic devices and enhanced the usage duration. But the success was dependent upon factors like the number of fins, the amount of PCM used, the power rating of the incoming heat concluding that the higher heat flux lead to higher temperatures and shorter operating time, and affected the usage mode of the device. Lastly, suitable PCM for a device was also dependent on the ambient temperature. PCM based heat sinks are becoming more and more attractive in thermal management of electronics in recent years. They can absorb and release large amount of heat during their melting and solidification process thus keeping the system at constant temperature nearly minimizing the occurrence of overheating or other damages. The present study aims at investigating and comparing the effect of two basic types of fins arrangement, inline and staggered on the thermal performance of PCM based heat sink. Additionally, the influence of two types of fin geometries, round and square shape on the thermal performance of PCM based heat sinks. Six different PCMs of wider range of melting temperature and latent heat values are used to determine the most suitable material relevant to the specific fins arrangement and geometry. 2. Experimental setup 2.1. Heaters and power supply To carry out the experimental analysis, a real-time assembly system is created to achieve efficient results. The archetypal heat input is used to impersonate and act as an electronic device. The schematic diagram and pictographic view of original setup are shown in Figs. 1 and 2 respectively. This setup is designed to get in conjunction with its theoretical structure. The mimicry of heat generation in electronics is done using 50 50 mm2 OMEGAÒ silicon rubber heater (SRFG-202/10-P-220V). A DC power supply module by Keysight TechnologiesÒ (6675A, 0–120V/0–18A) is used to provide the required power to the heater with programming accuracy of voltage 0.04% + 120 mV and current 0.1% + 12 mA at reference temperature 25 C, which is attached to
Fig. 1. Schematic representation of experimental setup.
253
the base of heat sink. Table 1 shows power levels used in experiment and their respective input heat fluxes of are achieved at the contact surface of heater and heat sink. To control power, a manual switch is also introduced for safety purposes. 2.2. Heat sinks configuration In this study, circular pin fins of diameter 3 mm and square pin fins of dimensions 2 2 mm2 are selected. The total number of fins for circular and square cross sections are 42 and 72 respectively for both in-line and staggered arrays. The heat sinks are designed with overall dimensions of 71 70 25 mm3 with wall thickness of 7 mm around the boundary and 5 mm from base is also maintained. An optimum 9% volume fraction of fins is used in this system, as established by [15,16]. To calculate the number of fins, the following equation is used:
Nf ¼ 0:9
VS
ð1Þ
vf
Aluminium (6061-T6) is used for heat sink manufacturing using CNC machining. Its thermo-physical properties are shown in Table 2. A sectioned view representation of PCM base heat sink assembly is shown in Fig. 3 and dimensions of its components are given in Table 3. The orthogonal projections of square inline are shown in Fig. 4.1. The isometric view of the four pin fin heat sinks under study are also shown in Fig. 4.2. 2.3. Types of PCMs under investigation Several PCMs are analysed in this experiment. wPCM is kept constant at 90% and is calculated using Eq. (2) which is the ratio of the PCM volume to the difference of V S and V f :
wPCM ¼ 0:9
mPCM
VS Vf
ð2Þ
A total of six PCMs are used to explore the effect of melting temperature and corresponding latent heat capacities. The selection of materials is carried out keeping in view the various power densities of electronic devices. Paraffin wax (Mersck, Germany) [17] and n-eicosane [18] are used for comparison with other materials and their relative effect on each shape is studied. Further four materials from RUBITHERMÒ [19] are analysed. The RUBITHERMÒ RT-54, RT44, RT-35HC, SP-31 are also brought under study due to their distinctive thermal properties which are shown in Table 4. 2.4. Thermocouples positions Temperature is the key parameter being measured in this study. For this purpose, nine thermocouples by OMEGAÒ of highly calibrated ASTM standards [20] are used, within a temperature range of 0–100 °C and found maximum discrepancy of 0:1 C. This analogue data is converted to digital form by data acquisition system (Agilent 34972A, (100 °C to 1200 °C). This is then connected to laptop which attains the data through data acquisition software at an interval of 5 s. The Thermocouple placement is shown in Fig. 5. The designation H1 and H2 show position of thermocouples at a length of 35 mm along the base, measuring heat sink base temperature. The thermocouples from W1 to W4 show wall temperature or temperature at boundary wall. To visualize the phase change profile of PCMs, the thermocouples T1 to T3 are protruded at different heights from base. The thermocouples T1, T2 and T3 are placed at 10 mm, 15 mm and 20 mm respectively. To fix these thermocouples, Araldite is used.
254
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
Fig. 2. Pictorial view of experimental setup.
Table 1 Power values with their subsequent heat fluxes. Power (W) q ðkW=m2 Þ
4 1.6
5 2.0
6 2.4
7 2.8
8 3.2
Table 2 Properties of aluminium. Property
Value
Unit
kAl
180 2700 963 660.4
W/m k Kg/m J/kg K °C
qAl C Al T Al
Fig. 3. Sectioned view of PCM based Heat Sink Assembly.
Table 3 Material specifications used for making the assembly of heat sink. Sr. no.
Materials used
Dimensions (mm)
1 2 3
Perspex sheet Silicon rubber gasket Rubber pad for heat sink
4
Rubber pad for heat sink bottom
71 70 5 71 70 3 (with a cut out of 57 56) 220 220 25 (with a cut out of 114 114) 220 220 65
3. Results and discussions The purpose of this experiment is to single out the best operational circumstances for a heat sink to be effective. It is worth men-
tioning that all experimental values demonstrated in terms of graphical trends for pin fin comparison are for base temperature values for both with and without PCM. This is so because this study focuses on transfer of heat generated inside the ICs of electronics, which is gained and is maximum at the base of the heat sink. So, lower temperature at base means more heat is being absorbed by the PCM setup and this implies for its better heat transfer abilities. In consideration of the experimental constraints, the results are sorted out and explained as following. 3.1. The pin-fin configuration comparison (without PCM) As mentioned earlier, this study is performed for a range of power levels between 4 W and 8 W. For citation 5 W and 8 W are considered. First set of readings were taken without any PCM. The comparison between the square inline and staggered is shown in Fig. 6a and b. Whereas that between circular configurations is also shown adjacently in Fig. 6c and d. Temperature distribution of circular and square pin-fins is likewise compared collectively in Fig. 7. As it can be seen from the trends of temperature distribution in Fig. 6, staggered fin arrangement is found to be the better than its counterpart in both circular and square profiles. The reason is that the staggered configuration allows more disturbance due to development of natural convection currents. These convection currents develop due to the trapped air inside the heat sink. As a vacuum is not created in this case, so air remains trapped. The results also satisfy the numerical analysis performed by Yang et al. [21]. Results shown in Fig. 7 illustrate that square staggered pin-fin arrangement is a superior arrangement for heat transfer as compared to circular (inline, staggered) and square inline. The explanation for this sets with the established fact that square cross-section produces more turbulence of air due to its sharp edges, helping in heat dissipation. The trends of Fig. 7 comply with work of Muhammad [22]. Furthermore, if we take the surface area ratio, using Eq. (3), of square (inline or staggered) to circular (inline or staggered) pin-fins arrangements.
A ¼
Asquare Acircular
ð3Þ
It is obtained A ¼ 1:46, which shows that the total surface area of square configuration pin-fins arrangements is larger than the surface area of circular configuration pin-fins arrangements, which shows a significant reason of better outperforming of heat transfer rate for square pin-fins (inline or staggered) arrangements than circular pin-fins (inline or staggered) arrangements in the cases of without PCMs.
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
Fig. 4.1. Orthogonal views of square inline Heat Sink under study.
(a). circular inline
(c). Square inline
(b). circular staggered
(d). Square staggered
Fig. 4.2. Isometric view of Heat Sinks under study.
255
256
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
Table 4 Thermo-physical properties of PCMs. Material
kPCM (W=m K)
CPCM (kJ/kg K)
kPCM (kJ/kg)
TPCM (°C)
qPCM (kg/m3)
Paraffin wax
2.8
173.6
56–58
2.2(L) 1.9(S) 2
237.4
36.5
RT-54
0.167(L) 0.212(S) 0.160(L) 0.40(S) 0.2
200
54
RT-44
0.2
2
250
44
RT-35HC
0.2
3
240
35
SP-31
0.6
2
210
31
790(L) 880(S) 780(L) 820(S) 800(L) 850(S) 700(L) 800(S) 770(L) 880(S) 1300(L) 1350(S)
n-Eicosane
behaviour of square inline at 60th minute. However, in case of circular arrays, staggered has slower temperature gain, even below its inline arrangement. The difference between inline and staggered decreases and both plots overlap during time span of 30th and 60th mins. Then the staggered plot starts increasing. For paraffin wax square staggered is initially below the inline array pattern but turns to show a greater temperature gain after 60th minute. Circular pin-fin configuration also shows less temperature gain in case of its inline formation. Graphical trend for SP-31 show no visible difference in inline and staggered for both square and circular shapes. However, more temperature gain can be noticed for staggered temperature profiles at the end.
Fig. 5. Position of thermocouples across heat sink.
For a power level 8 W, Fig. 9a and b show the results for both square and circular geometries. Generally, the trend followed by PCM heat sinks under this power value depict the same result as that at 5 W i.e. inline is the better arrangement in case of both square and circular cross sections. The results are explained as follows for square geometry:
3.2. The Pin-fin configuration comparison (with PCM) After completion of analysis on heat sinks without PCM, different PCM are introduced in the four pin fin heat sinks both for inline and staggered fins arrangement. Results are shown in Fig. 8a as a comparison between inline and staggered pin fin configurations of square geometry at 5 W. Similar comparison for circular crosssection at 5 W is shown in Fig. 8b. Further, comparison between inline and staggered pin fin configurations of square and circular cross-sections is shown in Fig. 9a and b respectively. As it is evident from the Figs. 8 and 9, graphical trends suggest that inline is a better array geometry to consider. This is seen to be true for both square and circular pin-fin heat sinks. Here more temperature gain means less heat transfer as less heat is being dissipated. The reason why inline has less temperature gain is because it uniformly distributes temperature in the PCMs. This allows maximum heat transfer by PCM. Further, individual results for all six PCMs prove this finding. Let’s consider results at 5 W for instance: It is evident that there is visible constant difference between inline and staggered plots of square and circular cross section for RT-44 and RT-54 in both Fig. 8 (a) and (b). For n-eicosane the temperature difference in both configuration increases gradually in case of square whereas circular pin-fin shows gradual increase in change of temperature followed by abrupt increase between 50th and 90th mins. For RT-35HC in case of square, temperature difference is maintained in both array pattern until 45th minute where slope of square staggered heat sink increases rapidly followed by similar
RT-44 and n-eicosane show a fluctuating difference in their square inline and staggered trend lines but the dominance of square inline is apparent. Paraffin Wax and SP-31 allow a gradual increase till 45th minute where it follows a constant trend. RT-35HC shows staggered array losing more heat from base until 40th minute where its heat transfer is overcome by its inline counterpart. The temperature distribution of RT-54 for square staggered is slightly better till 45th minute, after which it attains similar trend with inline up to 70th minute, from where sensible heating is seen to occur. Similarly, the plots for circular configurations illustrate the following: RT-44 and paraffin wax show visible constant difference in the respective inline and staggered pin-fin formations used in this experiment after 15th minute where inline is again dominant. N-eicosane has no significant change in its trends as both curves overlap each other. SP-31 shows very minor change until 55th minute from where variation increases and then follows a constant difference between two temperature plots establishing inline pin-fin array again in leading position. RT-35HC has fluctuating temperature distribution along the span of 90 min. Still inline is better than the staggered pin fin arrangement.
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
a. Square inline vs staggered at 5W
c. Circular inline vs staggered at 5W
b. Square inline vs staggered at 8W
d. Circular inline vs staggered at 8W
Fig. 6. The temperature distribution on different configurations at 5 W and 8 W.
Fig. 7. Circular (inline, staggered) vs square (inline, staggered).
257
258
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
a: square Inline vs staggered
b: Round Inline vs staggered Fig. 8. Inter comparison between inline and staggered PCM materials at 5 W.
From above results it is known that inline fin arrangement is a better choice in both circular and square pin-fins heat sinks. To establish better of the two, a comparison for square inline and circular inline is illustrated in Fig. 10(a) and (b). It is clearly shown that circular inline heat sink is better for all PCMs with exception of n-eicosane. For n-eicosane square inline directs as more effective choice until at the end where circular inline takes over. Moreover, at 8 W n-eicosane and paraffin wax also deviate from the behaviour shown by all other PCMs. Conclusively, circular inline pin-fins arrangement surpasses the square inline pin-fins for all PCMs. It is evident that circular inline pin-fins are better because they enhance heat transfer effectively. Visualizing heat flowing in the centre of the fin, for a circular fin when heat flows from the base to the tip, at every part it can take whether a half diameter or half edge as a route out of the fin. Since half edge is shorter, it prefers it. This means that at every point across the radius of a circular fin edge, temperature is evenly distributed. The reason of better per-
formance of heat transfer rate in case of circular pin-fins in comparison of square pin-fins is because of optimum fin distribution and arrangement as well the number of fins. Despite of this, total surface area of pin-fins is not a limiting factor, in case of square pinfins, to increase or decrease the heat transfer rate, however, higher the number of fins even with optimum fins distribution and arrangement cause the rapid increase of heat transfer rate through the PCMs which resulting the increases the phase transition rate of PCMs [21,22]. To sum up, uniform temperature distribution at every segment of the circular fin leads to better heat transfer. 3.3. Comparison of different PCMs As circular inline is established in this study as the dominant and most effective arrangement for a PCM based heat sink, so it’s experimental reading will be considered in this section to study PCM behaviour relative to their properties and effect on overall performance of heat sink.
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
259
a: square Inline vs staggered
b: Round Inline vs staggered Fig. 9. Inter comparison between inline and staggered PCM materials at 8 W.
3.3.1. Effect of melting temperature and latent heat of PCM Melting temperature is one of the key aspects considered when choosing a PCM as a heat storage medium for passive cooling. Melting temperature affects the spectrum of PCM application ranges and the performance of the heat sinks integrated along with PCM. This experimentation focuses on PCM’s with a wide range of melting temperature from 31 C to 57 C. All PCM melting temperatures and corresponding latent heats are listed in Table 3. Considering circular inline, the effects of melting temperature are studied at 5 W and 8 W as it provides enough heat flux to melt all PCMs. As it is evident in Fig. 11, the point where all PCMs trends become less steep is the melting point of that PCM. These points are shown with a black encircled point. The changes in curve can also be observed at these points. But how does the melting point affect the system’s performance? The answer to this lies in the fact that it depends upon the heat flux generated in the system. For a
higher heat flux the corresponding temperature at base is also high. So, the PCM which is initially in the solid phase, absorbs heat with rise in base temperature. The temperature of thermocouple placed at H1 and H2 will reflect the external temperature until the PCM reaches its melting point. After the temperature ranges the melting point of the PCM, it starts to change phase and thus liquefies. As this liquefaction process occurs, the PCM absorbs in bulks of heat with almost no temperature variation. During this span, cooling effect is centred through the PCM. That span or amount of time the PCM usually provides a cooling effect is known as the latent heat of fusion. The enthalpy varies depending on the PCM material itself. In case of this experiment, the enthalpy is classically measured in kJ/kg. For higher number of kilo Joules per kilogram, the PCM will provide a cooling effect will be for a longer time. The PCM, which is now in molten phase, it releases the absorbed heat. The cycle reverses as soon as the base temperature
260
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
a: At 5W
b: At 8 W Fig. 10. Square inline and Circular Inline comparison.
cools. The PCM, now liquefied, releases the heat absorbed as the base temperature decreases. During this period, the PCM solidifies and generates a heating effect. Fig. 11a and b illustrate the PCMs with lower melting temperatures work best with lower power levels like 5 W and similarly PCM’s with higher melting points are considered for 8 W. Paraffin wax doesn’t reach its melting point in both 5 W and 8 W systems. Additionally, its latent heat is the lowest among the six PCMs. RT-54 like paraffin wax does not reach its melting point at 5 W but it does melt on the 35th minute at 8 W. Its latent heat is also low compared to other PCM’s used. RT-44 has highest latent heat amongst all. At 5 W, it provides a constant behaviour for most of its time and transfers heat with-
out even melting. But Fig. 11(b) shows that after surpassing its melting point at 20th minute, it starts to gain sensible heat by the 60th minute. So, its gradient becomes steeper. RT-35HC shows good heat transfer properties at 5 W temperature plots in Fig. 11(a), but changes this behaviour at 8 W. N-eicosane melts around the 35th minute but also advances to sensible heating by 80th minute. At 8 W, it shows similar character as that of RT-35HC i.e. moves towards sensible heating with sharp gradient. Both n-eicosane and RT-35HC have approximately similar melting temperatures and latent heats. SP-31 shows most heat absorption/dissipation in Fig. 11a and b. It has the lowest melting point among its all PCMs studied but also has lower latent heat than RT-44, n-eicosane and RT-35HC. The property allowing it the most heat absorption is its density which is 1300 kg=m3 in liquid phase, almost double the amount
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
261
Fig. 11. PCMs comparison of Round Inline at 8 W.
of all other PCMs. It is so because a lot of energy is required to overcome the strong bonds in the PCM. But it would also require a similar time span to cool the material as it was taken whilst heat absorption. So, based on its density its discharging time can be predicted as high as its charging phase. Although sensible heat is not studied in this experiment but the comparison of sensible heat and latent heat storage systems involving PCMs suggest that the latter has higher thermal energy storage density and so requires reduced material amounts in terms of masses and volumes. The sensible heat phase can be observed above the melting points in Fig. 11. 3.3.2. The effect of varying power level and enhancement in operation time In this section the enhancement in operation time is studied with respect to varying power levels, identifying the best PCM to use at a certain power level. This is depicted in terms of the time required to reach critical SPTs of 45 C and 60 C, and is shown in Fig. 12a and b. Here PCMs are divided into two groups, one with
those having lower melting temperatures whilst second with relatively higher melting temperatures. The first group consists of SP31, n-eicosane and RT-35HC and the second one has paraffin wax, RT-54 and RT-44. They are divided with respect to critical SPT. The critical SPT is defined as the maximum working temperature that an electronic device withstands without halt. From Fig. 12, it is apparent that for all PCMs the operation time decreases whilst the power input increases. It’s obvious that: SP-31 takes the most time to reach the critical SPT of 45 °C in rest of all other PCMs. n-eicosane and RT-35HC stand at close competition at all power levels but the latter ensures more heat transfer, showing much higher difference in enhancement time by 52 min at 5 W as shown in Fig. 12a. RT-44 dominates at SPT of 60 °C at 5 W but its operation time decreases rendering it least suitable for 7 W and 8 W. RT-54 has least operational time qualification for 5 W and 6 W but it proves to be more useful at higher power levels.
Fig. 12. Time to reach critical SPT at different Power levels for various PCMs.
262
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
Fig. 13. Enhancement ratio of different PCMs.
Paraffin wax shows consistency in its enhancement in operational time throughout all the power levels. 2. 3.3.3. Enhancement ratio Fig. 13a and b present the enhancement ratio at two critical SPTs of 45 C and 60 C. For SPT of 45 C, PCMs of low melting point namely SP-31, RT-35HC and n-eicosane are considered, whereas for the SPT of 60 C, PCMs of high melting point like RT-54, RT-44 and paraffin wax are compared. The enhancement ratio, ePCM , is the SPT time ratio of heat sink with PCM to that of the same heat sink without PCM, calculated using Eq. (4).
ePCM ¼
tcrwith PCM tcrwithout PCM
3.
4.
ð4Þ
Here it becomes obvious that Sp-31 has the highest enhancement ratio for SPT of 45 °C. The enhancement ratio of RT-35HC is above n-eicosane at a power level of 5 W, but tends to become similar at 6 W and 7 W. At 8 W n-eicosane has a slightly better enhancement ratio than RT-35HC. For critical SPT of 60 °C paraffin wax stands apart as best option whilst RT-44 is better for low power levels and RT-54 for higher power levels. So, this makes it obvious that higher enhancement ratios are achieved on lower SPT of 45 C due to the higher time taken by PCM to complete its latent heat phase. This decline in the enhancement ratio at SPT of 60 °C as compared to SPT of 45 °C is caused by the abrupt rise in temperature after completion of the melting phase.
5.
6.
7.
array is found to be the most effective heat sink of all four geometries without PCMs. The second discussion was the pin-fin configuration comparison with PCMs. The studies of six PCMs, two geometries and varying power levels shows a unanimous result of inline as the dominant configuration for both square and circular cross-sections. Further comparison between circular inline and square inline helped in establishing circular inline as the most efficient choice for a PCM based heat sink. Using circular inline as dominant geometry, the effect of melting point and latent heat was studied. Latent heat is found as the major reason of more efficiency of PCMs heat transfer properties. RT-44, n-eicosane, SP-31 are dominant coolants at lower power levels whilst RT-54 was best option at higher power levels due to its mid-range thermal properties. The effect of varying power level and enhancement in operation time is also studied in terms of critical SPTs of 45 °C and 60 °C. The results suggested that SP-31 requires the most time to reach the critical SPT of 45 °C whilst paraffin wax is consistently better choice for SPT of 60 °C due to its high melting point. Finally, the enhancement ratio is considered as it becomes obvious that SP-31 has the highest enhancement ratio of 9.28 at 5 W followed by RT-35HC and n-eicosane. For 8 W power level systems paraffin wax is the best PCM with highest enhancement ratio.
4. Conclusions
Conflict of interest
Successful experimental study is conducted to single out an efficient working setup by passive cooling using PCM based heat sinks. Circular and square pin-fins are used for experimentation, with configurations varying between inline and staggered arrays. Power levels of 4 W to 8 W were switched keeping the volumetric fraction of PCM constant at 90%. Six PCMs were broadly studied for each configuration of heat sinks. From all these set variations and conditions, following deductions are made:
Authors declare no conflict of interests regarding this paper.
1. The pin-fin geometry and configuration comparison without PCM yielded staggered is the better arrangement in both square and circular cross-sections. Overall the square staggered pin-fin
References [1] S.F. Husseinizadeh, F.L. Tan, S.M. Moosania, Experimental and Numerical studies on performance of phase change material based Heat sinks with different configurations of internal fins, Appl. Thr. Engg. 31 (2011) 3827–3838. [2] R. Pakrouh, M.J. Hosseini, A.A. Ranjbar, R. Bahrampoury, A numerical method for phase change material based Heat sinks optimization, Energy Conv. Manag. 130 (2015) 542–552. [3] Saad Mahmoud, Aarun Tang, Chin Toh, Raya AL-Dadah, SeinLeung Soo, Experimental investigation of inserts configurations and Pcm Type on the thermal performance of Phase change material based Heat sinks, Appl. Energy 112 (2013) 1349–1356.
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263 [4] S.K. Saha, K. Srinivasan, P. Dutta, Studies on optimum distribution of fins in heat sinks filled with phase change materials, J. Heat Transf. 130 (2008), 034505-1. [5] R. Baby, C. Balaji, Thermal management of electronics using Phase Change material based pin fin Heat Sinks, 6th European Thr. Sciences Conference, J. Phys., IOP Publishing, 2012. [6] R. Sarikanth, Pavan Nemani, C. Balaji, Multi-objective geometric optimization of a Pcm Matrix Type Heat Sink, Appl. Energy 156 (2015) 703–714. [7] Li-Wu Fan, Yu-Qi Xiao, Yi Zeng, Xin Fang, Xiao Wang, Xu Xu, Zi-Tao Yu, RongHua Hong, Ya-Cai Hu, Ke-Fa Cen, Effect of melting temperature and the presence of internal fins on the performance of Pcm Based Heat sink, Int. J. Thr. Sci. 70 (2013) 114–126. [8] A.F. Mahrous, Thermal performance of phase change material based heat sinks, Int. J. Mech. Energy 2 (2013) 35–42. [9] Peleg P. Levin, Avraham Shitzar, Gad Hetsroni, Numerical optimization of phase change material based heat sink with internal fins, Int. J. Heat Mass Transf. 61 (2013) 638–645. [10] Jesto Thomas, P.V.S.S. Srivatsa, S. Ramesh Krishnan, Rajesh Baby, 2010, Thermal performance evaluation of phase change material based heat sink a numerical study, Global colloquium in recent advancement and researches in engineering, Science and Technology 2016, Elsevier Ltd, Procedia Technology 25 (2016) 1182–1190. [11] Atul Nagose, Ankit Somani, Aviral Shrot, Arunn Narasimhan, Genetic Algorithm based optimization of Pcm based heat sinks and effect of Heat sink parameters on operational time, J. Heat Transf. 130 (2008), 011401-1. [12] Tomoyuki Hatakeyama, Masaru Ishizuka, Sadakazu Takakuwa, Shinji Nakagawa, Kanjii Takagi, Experimental and thermal network study on the performance of a pins studded Pcm in electronic device cooling, J. Thr. Sci. Tech. 6 (2011) 11–0030.
263
[13] Ahmad Hassan, Hassan Hejase, Shaimaa Abdelbaqi, Ali Assi, Mohammed O. Hamdan, Comparative effectiveness of different phase change materials to improve cooling performance of Heat sinks for electronic devices, Appl. Sci. 6 (2016) 226. [14] S.C. Fok, W. Shen, F.L. Tan, Cooling of portable hand-held electronic devices using phase change materials in finned heat sinks, Int. J. Therm. Sci. 49 (1) (2010) 109–117. [15] R. Baby, C. Balaji, Thermal optimization of PCM based pin fin heat sinks an experimental study, Appl. Therm. Eng. 54 (1) (2013) 65–77. [16] A. Arshad et al., Thermal performance of phase change material (PCM) based pin-finned heat sinks for electronics devices effect of pin thickness and PCM volume fraction, Appl. Therm. Eng. 112 (2017) 143–155. [17] EMD Millipore is a part of Merck KGaA, D., Germany, 2016, Available from http://www.emdmillipore.com/US/en/product/Histosec-pastilles,MDA_ CHEM-111609. [18] ‘‘Eicosane 442673.” Sigma-Aldrich Italy. N.p., n.d. Web. 04. 2016, Available from http://www.sigmaaldrich.com/catalog/product/aldrich/219274/. [19] ‘‘Rubitherm GmbH.” Rubitherm GmbH, Gemany. N.p., n.d. Web. 2016, Available from https://www.rubitherm.eu/. [20] G.W. Burns et al., Temperature-electromotive force reference functions and tables for the letter-designated thermocouple types based on the ITS-90. NASA STI/Recon Technical Report N, 1993, p. 93. [21] Yue-Tzu Yang et al., Numerical simulations of electronic cooling with in-line and staggered pin fin heat sinks, IJMME 10 (2016) 1172-1177. [22] Eaman Hassan Muhammad, A comparison of the heat transfer performance of a hexagonal pin fin with other types of pin fin heat sinks, Int. J. Sci. Res. 4 (2015) 1781–1789.
Contents lists available at ScienceDirect
International Journal of Heat and Mass Transfer journal homepage: www.elsevier.com/locate/ijhmt
Experimental passive electronics cooling: Parametric investigation of pin-fin geometries and efficient phase change materials Muhammad Junaid Ashraf a, Hafiz Muhammad Ali a,b,⇑, Hazrat Usman a, Adeel Arshad c,d a
Department of Mechanical Engineering, University of Engineering and Technology, Taxila, Pakistan Center of Research Excellence in Renewable Energy (CoRE-RE), King Fahd University of Petroleum & Minerals (KFUPM), Dhahran 31261, Saudi Arabia c Department of Mechanical Engineering, HITEC University, Taxila, Pakistan d Fluids & Thermal Engineering Research Group, Faulty of Engineering, University of Nottingham, Nottingham NG7 2RD, UK b
a r t i c l e
i n f o
Article history: Received 28 March 2017 Received in revised form 23 July 2017 Accepted 24 July 2017
Keywords: Phase change materials Thermal conductivity enhancers Staggered and inline arrays Circular and square cross-sectional pin–fin heat sinks Enhancement time Enhancement ratios
a b s t r a c t This experimental investigation focuses on the optimization of passive cooling system using extruded finned surfaces with phase change materials (PCMs) as the thermal conductivity enhancers (TCEs). The study develops comparison between fins of circular and square cross-sectional area, made of aluminium. Further classification is done in configuration in terms of staggered and inline arrays. The volume fraction of fins is kept constant at 9% of total volume of heat sink. The purpose is to single out the better arrangement with and without PCM. Six PCMs of varying phase change temperature and heat capacities, namely Paraffin wax, RT-54, RT-44, RT-35HC, SP-31 and n-eicosane are selected for thermal conductivity enhancement. The volume fraction of PCM is also constant at 90% of the heat sink volume, giving a 10% volume for expansion after melting. Moreover, power levels are used in a range of 4–8 W with an increment of 1 W. The analysis was carried out on graphical trends produced and explanations were given accordingly. The most effective PCMs were also discussed considering their enhancement time, enhancement ratios and other material properties. Finally, the results were justified by the scientific knowledge and found in compliance with the work of famous researchers as well. Ó 2017 Elsevier Ltd. All rights reserved.
1. Introduction Thermal control of modern electronics circuitry has become increasingly complex and it is very essential to ensure its reliability, durability and user comfort. Researchers are always in search of finding new effective solutions for thermal management of electronic devices. The choice of cooling techniques depends on number of factors. Active cooling and all other techniques have proven to be incapable of fulfilling the demands of modern electronic industries as it consumes power itself. So, passive cooling techniques employing PCM based heat sinks are used widely in these portable devices like computers, mobile phones, personal digital assistants, laptops and so on. Effective thermal management by incorporating PCM based heat sinks in these devices have led to increase in its functionalities, reliability, less probability to internal damages and failure, ultimately to stretch their useful life. Many studies both experimentally and numerically have been conducted to offer a deep insight into thermal management of electronic devices using PCM based heat storages. ⇑ Corresponding author. E-mail addresses: [email protected], [email protected] (H.M. Ali). http://dx.doi.org/10.1016/j.ijheatmasstransfer.2017.07.114 0017-9310/Ó 2017 Elsevier Ltd. All rights reserved.
Husseinizadeh et al. [1] examined both experimentally and numerically PCM based heat sinks to study the effect of various configurations of internal fins. The PCM RT-80 was filled in different heat sinks of constant overall dimensions. From the results obtained it was seen that by increasing number of fins, fin height and input power level had improved the thermal performance appreciably while increasing fins thickness had only marginal improvement. Similarly, Pakrouh et al. [2] presented numerical method for geometric optimization of pin finned heat sinks by coupling Taguchi and simulation method. For optimization, the effect of all critical parameters involving fins’ number, thickness, base thickness, height as well as PCM percentage were explored. All heat sinks filled with RT-44 and critical temperatures 50 C;60 C;70 C and 80 C were selected for analysing the results. The 2 mm thick fins performance was highest for 50 C while 4 mm thick fins performance was highest for all temperatures. Base thickness contributed less than the other parameters. Mahmoud et al. [3] investigated experimentally the effect of honeycomb structure in heat sink to compare its performance with finned heat sink. Six different types of PCMs and total six heat sinks, one with single cavity, two designed with parallel fins, two with cross arrangement and one with honey comb inserts were tested at
252
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
Nomenclature Roman symbols Symbol Quantity (Unit) PCMs Phase change materials SPTs Set point temperatures (°C) T PCM Melting temperature of PCMs (°C) T Al Melting temperature of aluminium (°C) t Thickness of fin (mm) h Height of fin (mm) vf Volume of the fin (mm3) Vf Total volume of fins Nf Total number of fins Total working volume of heat sink (mm3) VS m Mass of the PCM (kg) q Heat flux (kW/m2) tk Latent heat phase duration (sec) T Thermocouple inside the PCM
power level in the range from 3 to 5 W. The obtained results showed that increasing number of fins, insertion of honey comb and inclusion of low melting temperature PCM at higher power level had enhanced significantly the operation time of heat sink. Fins had significant role in improving thermal performance of heat storage unit. Recent findings considered identifying optimum distribution of these TCE (fins) in terms of dimensions and shape. Saha et al. [4] carried out their research to investigate the effective way of distributing fins in heat sinks i.e. to find the optimum volume fraction of TCE which maintains a low temperature of any component. Using n-eicosane as PCM in aluminium made heat sinks. Two types of fins (plate-fin and pin-fin) were analysed in heat sinks with base dimension of 42 42 mm2 and fins height of 25 mm. The case of 8% TCE volume fraction of heat sink or base plate was reported to give best results. Regarding fin dimensions and shape it was concluded that the large number of small cross-sectional area pin fins performed better. Baby and Balaji [5] experimentally investigated three different geometries of heat sinks employing different no of TCE. All heat sinks were made of aluminium employing 33, 72,120 number of pin-fins of corresponding volume fractions were 4%, 9%, and 15% respectively. The effectivity of volume fractions of TCE was examined at power level 4–8 w using n-eicosane with varying volume fractions of 0.3, 0.6 and 1. For SPT of 43 C and 8 W power, the highest enhancement factor of 21 was obtained for heat sink with 72 number of fins. Also, the performance was seen to be strongly dependent on PCM volume fractions instead of TCE volume fractions. The effect of orientation on thermal performance of porous matrix filled heat sink was investigated by Srikanth et al. [6]. Tracking system was employed to change the orientation of copper metal foam matrix embedded Al-heat sink filled with n-eicosane. The experimental results obtained in terms of enhancement ratios showed that the heat transfer enhancement was only comparable and effect of orientation has no significant impact on the mentioned heat sink. Experimental work performed by Fan et al. [7] determined the effect of internal fins and melting temperatures on performance of PCM based heat sinks under pulsating heat loads. Two organic PCMs (n-eicosane and 1-hexadecanol) having different melting temperature were tested in prototype heat sinks at different power levels and it was concluded that the PCM with higher melting temperature had resulted in improved thermal performance of electronic devises comparatively. Six heat sinks designs were tested by Mahrous [8] to investigate the effect of fins arrangement and fin number, on thermal performance. Identical plate finned heat
W H t cr
Thermocouple at side wall of heat sink Thermocouple at base of heat sink Time to reach critical temperature (sec)
Greek symbols Volumetric fraction of PCM wPCM mPCM Volume of PCM (m3/kg) qPCM Density of PCM (kg/m3) qAl Density of aluminium (kg/m3) ePCM Enhancement ratio at PCM kPCM Latent heat of PCM (kJ/kg) C PCM Heat capacity of PCM (kJ/kg) C Al Heat capacity of aluminium (J/kg) Thermal conductivity of PCM (W/m K) kPCM kAl Thermal conductivity of aluminium (W/m K) A Surface area ratio
sinks with four heat sinks of parallel fins arrangement, one with crossed fins and one with no fins were tested. A constant fin thickness of 1 mm and constant power level 4.84 W was used and it was concluded that the parallel and crossed fins arrangement both had nearly comparable performance but performance of all fins heat sinks was superior to that without fins. Numerical study carried out by Levin et al. [9] explored the effect of fin’s length, height and number for the optimal PCM percentage to be considered in designing latent heat thermal management system (LHTMS) for electronic devices. From the results, it was concluded that the optimal percentage of PCM is dependent on number and height of fins, heat flux and the difference between liquidus and critical temperature. Thomas et al. [10] carried out numerical study for the design of PCM based heat sink for average dimensions of a smart phone. Analysis were performed using ANSYS FLUENT 14.0 by providing constant heat flux to the base of heat sink with power input ranging from 4 to 6 w. Eicosane was selected as PCM for designed heat sink. High performance was obtained when PCM fraction had reached to its maximum. Selection of thermal performance enhancement method has a vital role in thermal management of components. Nagose et al. [11] used combined genetic and conventional simulation to get optimized configuration of heat sink which kept the temperature of microprocessor in acceptable limits. The designed heat sink consisted of fin array with depth equal to heat sink and a heat spreader inside the heat sink. From the results obtained with the assumption of constant heat flux from electronic device correlations were proposed relating heat sink operational time to the PCM fraction, heat sink depth and heat spreader thickness. It was found that the optimal spreader thickness was 2.5% of heat sink depth. Hatakeyama et al. [12] conducted their experimental and analytical research on PCM based transient cooling module employing pin fins. The setup used by them consisted of test module, heat spreader, chamber, substrate, power source, Data logger, three heat sinks with round fins of different number and diameter. The paraffin volume fraction used, was 81% of total space. Thermal network model of module was developed capable of predicting temperature transients. This model was reported to be effective design tool for thermal management of electronic devices. The research by Hassan et al. [13] investigated and compared the performance of three different types of PCMs namely salt hydrate, paraffin wax and milk fat in PCM based finned heat sink system. Experiments were conducted on finned heat sink at power input ranging from 4 W to 10 W. The conjugate heat transfer model was developed for each
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
PCM system that had predicted the effect of thermal conductivity, density, melting range of each PCM on heat sink performance. From the results obtained milk fat PCM was reported to have lower performance comparatively. Increased thermal conductivity and density of PCMs had resulted in increased thermal performance but was found to decrease for wider melting range. Fok et al. [14] performed experimental study on the application of phase change material (PCM) as coolant in portable hand-held electronic devices at power levels ranging from 3 W to 5 W, for extensive or light working conditions. The results summed up that TCE pin fins with PCM was feasible option for cooling hand-held electronic devices and enhanced the usage duration. But the success was dependent upon factors like the number of fins, the amount of PCM used, the power rating of the incoming heat concluding that the higher heat flux lead to higher temperatures and shorter operating time, and affected the usage mode of the device. Lastly, suitable PCM for a device was also dependent on the ambient temperature. PCM based heat sinks are becoming more and more attractive in thermal management of electronics in recent years. They can absorb and release large amount of heat during their melting and solidification process thus keeping the system at constant temperature nearly minimizing the occurrence of overheating or other damages. The present study aims at investigating and comparing the effect of two basic types of fins arrangement, inline and staggered on the thermal performance of PCM based heat sink. Additionally, the influence of two types of fin geometries, round and square shape on the thermal performance of PCM based heat sinks. Six different PCMs of wider range of melting temperature and latent heat values are used to determine the most suitable material relevant to the specific fins arrangement and geometry. 2. Experimental setup 2.1. Heaters and power supply To carry out the experimental analysis, a real-time assembly system is created to achieve efficient results. The archetypal heat input is used to impersonate and act as an electronic device. The schematic diagram and pictographic view of original setup are shown in Figs. 1 and 2 respectively. This setup is designed to get in conjunction with its theoretical structure. The mimicry of heat generation in electronics is done using 50 50 mm2 OMEGAÒ silicon rubber heater (SRFG-202/10-P-220V). A DC power supply module by Keysight TechnologiesÒ (6675A, 0–120V/0–18A) is used to provide the required power to the heater with programming accuracy of voltage 0.04% + 120 mV and current 0.1% + 12 mA at reference temperature 25 C, which is attached to
Fig. 1. Schematic representation of experimental setup.
253
the base of heat sink. Table 1 shows power levels used in experiment and their respective input heat fluxes of are achieved at the contact surface of heater and heat sink. To control power, a manual switch is also introduced for safety purposes. 2.2. Heat sinks configuration In this study, circular pin fins of diameter 3 mm and square pin fins of dimensions 2 2 mm2 are selected. The total number of fins for circular and square cross sections are 42 and 72 respectively for both in-line and staggered arrays. The heat sinks are designed with overall dimensions of 71 70 25 mm3 with wall thickness of 7 mm around the boundary and 5 mm from base is also maintained. An optimum 9% volume fraction of fins is used in this system, as established by [15,16]. To calculate the number of fins, the following equation is used:
Nf ¼ 0:9
VS
ð1Þ
vf
Aluminium (6061-T6) is used for heat sink manufacturing using CNC machining. Its thermo-physical properties are shown in Table 2. A sectioned view representation of PCM base heat sink assembly is shown in Fig. 3 and dimensions of its components are given in Table 3. The orthogonal projections of square inline are shown in Fig. 4.1. The isometric view of the four pin fin heat sinks under study are also shown in Fig. 4.2. 2.3. Types of PCMs under investigation Several PCMs are analysed in this experiment. wPCM is kept constant at 90% and is calculated using Eq. (2) which is the ratio of the PCM volume to the difference of V S and V f :
wPCM ¼ 0:9
mPCM
VS Vf
ð2Þ
A total of six PCMs are used to explore the effect of melting temperature and corresponding latent heat capacities. The selection of materials is carried out keeping in view the various power densities of electronic devices. Paraffin wax (Mersck, Germany) [17] and n-eicosane [18] are used for comparison with other materials and their relative effect on each shape is studied. Further four materials from RUBITHERMÒ [19] are analysed. The RUBITHERMÒ RT-54, RT44, RT-35HC, SP-31 are also brought under study due to their distinctive thermal properties which are shown in Table 4. 2.4. Thermocouples positions Temperature is the key parameter being measured in this study. For this purpose, nine thermocouples by OMEGAÒ of highly calibrated ASTM standards [20] are used, within a temperature range of 0–100 °C and found maximum discrepancy of 0:1 C. This analogue data is converted to digital form by data acquisition system (Agilent 34972A, (100 °C to 1200 °C). This is then connected to laptop which attains the data through data acquisition software at an interval of 5 s. The Thermocouple placement is shown in Fig. 5. The designation H1 and H2 show position of thermocouples at a length of 35 mm along the base, measuring heat sink base temperature. The thermocouples from W1 to W4 show wall temperature or temperature at boundary wall. To visualize the phase change profile of PCMs, the thermocouples T1 to T3 are protruded at different heights from base. The thermocouples T1, T2 and T3 are placed at 10 mm, 15 mm and 20 mm respectively. To fix these thermocouples, Araldite is used.
254
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
Fig. 2. Pictorial view of experimental setup.
Table 1 Power values with their subsequent heat fluxes. Power (W) q ðkW=m2 Þ
4 1.6
5 2.0
6 2.4
7 2.8
8 3.2
Table 2 Properties of aluminium. Property
Value
Unit
kAl
180 2700 963 660.4
W/m k Kg/m J/kg K °C
qAl C Al T Al
Fig. 3. Sectioned view of PCM based Heat Sink Assembly.
Table 3 Material specifications used for making the assembly of heat sink. Sr. no.
Materials used
Dimensions (mm)
1 2 3
Perspex sheet Silicon rubber gasket Rubber pad for heat sink
4
Rubber pad for heat sink bottom
71 70 5 71 70 3 (with a cut out of 57 56) 220 220 25 (with a cut out of 114 114) 220 220 65
3. Results and discussions The purpose of this experiment is to single out the best operational circumstances for a heat sink to be effective. It is worth men-
tioning that all experimental values demonstrated in terms of graphical trends for pin fin comparison are for base temperature values for both with and without PCM. This is so because this study focuses on transfer of heat generated inside the ICs of electronics, which is gained and is maximum at the base of the heat sink. So, lower temperature at base means more heat is being absorbed by the PCM setup and this implies for its better heat transfer abilities. In consideration of the experimental constraints, the results are sorted out and explained as following. 3.1. The pin-fin configuration comparison (without PCM) As mentioned earlier, this study is performed for a range of power levels between 4 W and 8 W. For citation 5 W and 8 W are considered. First set of readings were taken without any PCM. The comparison between the square inline and staggered is shown in Fig. 6a and b. Whereas that between circular configurations is also shown adjacently in Fig. 6c and d. Temperature distribution of circular and square pin-fins is likewise compared collectively in Fig. 7. As it can be seen from the trends of temperature distribution in Fig. 6, staggered fin arrangement is found to be the better than its counterpart in both circular and square profiles. The reason is that the staggered configuration allows more disturbance due to development of natural convection currents. These convection currents develop due to the trapped air inside the heat sink. As a vacuum is not created in this case, so air remains trapped. The results also satisfy the numerical analysis performed by Yang et al. [21]. Results shown in Fig. 7 illustrate that square staggered pin-fin arrangement is a superior arrangement for heat transfer as compared to circular (inline, staggered) and square inline. The explanation for this sets with the established fact that square cross-section produces more turbulence of air due to its sharp edges, helping in heat dissipation. The trends of Fig. 7 comply with work of Muhammad [22]. Furthermore, if we take the surface area ratio, using Eq. (3), of square (inline or staggered) to circular (inline or staggered) pin-fins arrangements.
A ¼
Asquare Acircular
ð3Þ
It is obtained A ¼ 1:46, which shows that the total surface area of square configuration pin-fins arrangements is larger than the surface area of circular configuration pin-fins arrangements, which shows a significant reason of better outperforming of heat transfer rate for square pin-fins (inline or staggered) arrangements than circular pin-fins (inline or staggered) arrangements in the cases of without PCMs.
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
Fig. 4.1. Orthogonal views of square inline Heat Sink under study.
(a). circular inline
(c). Square inline
(b). circular staggered
(d). Square staggered
Fig. 4.2. Isometric view of Heat Sinks under study.
255
256
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
Table 4 Thermo-physical properties of PCMs. Material
kPCM (W=m K)
CPCM (kJ/kg K)
kPCM (kJ/kg)
TPCM (°C)
qPCM (kg/m3)
Paraffin wax
2.8
173.6
56–58
2.2(L) 1.9(S) 2
237.4
36.5
RT-54
0.167(L) 0.212(S) 0.160(L) 0.40(S) 0.2
200
54
RT-44
0.2
2
250
44
RT-35HC
0.2
3
240
35
SP-31
0.6
2
210
31
790(L) 880(S) 780(L) 820(S) 800(L) 850(S) 700(L) 800(S) 770(L) 880(S) 1300(L) 1350(S)
n-Eicosane
behaviour of square inline at 60th minute. However, in case of circular arrays, staggered has slower temperature gain, even below its inline arrangement. The difference between inline and staggered decreases and both plots overlap during time span of 30th and 60th mins. Then the staggered plot starts increasing. For paraffin wax square staggered is initially below the inline array pattern but turns to show a greater temperature gain after 60th minute. Circular pin-fin configuration also shows less temperature gain in case of its inline formation. Graphical trend for SP-31 show no visible difference in inline and staggered for both square and circular shapes. However, more temperature gain can be noticed for staggered temperature profiles at the end.
Fig. 5. Position of thermocouples across heat sink.
For a power level 8 W, Fig. 9a and b show the results for both square and circular geometries. Generally, the trend followed by PCM heat sinks under this power value depict the same result as that at 5 W i.e. inline is the better arrangement in case of both square and circular cross sections. The results are explained as follows for square geometry:
3.2. The Pin-fin configuration comparison (with PCM) After completion of analysis on heat sinks without PCM, different PCM are introduced in the four pin fin heat sinks both for inline and staggered fins arrangement. Results are shown in Fig. 8a as a comparison between inline and staggered pin fin configurations of square geometry at 5 W. Similar comparison for circular crosssection at 5 W is shown in Fig. 8b. Further, comparison between inline and staggered pin fin configurations of square and circular cross-sections is shown in Fig. 9a and b respectively. As it is evident from the Figs. 8 and 9, graphical trends suggest that inline is a better array geometry to consider. This is seen to be true for both square and circular pin-fin heat sinks. Here more temperature gain means less heat transfer as less heat is being dissipated. The reason why inline has less temperature gain is because it uniformly distributes temperature in the PCMs. This allows maximum heat transfer by PCM. Further, individual results for all six PCMs prove this finding. Let’s consider results at 5 W for instance: It is evident that there is visible constant difference between inline and staggered plots of square and circular cross section for RT-44 and RT-54 in both Fig. 8 (a) and (b). For n-eicosane the temperature difference in both configuration increases gradually in case of square whereas circular pin-fin shows gradual increase in change of temperature followed by abrupt increase between 50th and 90th mins. For RT-35HC in case of square, temperature difference is maintained in both array pattern until 45th minute where slope of square staggered heat sink increases rapidly followed by similar
RT-44 and n-eicosane show a fluctuating difference in their square inline and staggered trend lines but the dominance of square inline is apparent. Paraffin Wax and SP-31 allow a gradual increase till 45th minute where it follows a constant trend. RT-35HC shows staggered array losing more heat from base until 40th minute where its heat transfer is overcome by its inline counterpart. The temperature distribution of RT-54 for square staggered is slightly better till 45th minute, after which it attains similar trend with inline up to 70th minute, from where sensible heating is seen to occur. Similarly, the plots for circular configurations illustrate the following: RT-44 and paraffin wax show visible constant difference in the respective inline and staggered pin-fin formations used in this experiment after 15th minute where inline is again dominant. N-eicosane has no significant change in its trends as both curves overlap each other. SP-31 shows very minor change until 55th minute from where variation increases and then follows a constant difference between two temperature plots establishing inline pin-fin array again in leading position. RT-35HC has fluctuating temperature distribution along the span of 90 min. Still inline is better than the staggered pin fin arrangement.
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
a. Square inline vs staggered at 5W
c. Circular inline vs staggered at 5W
b. Square inline vs staggered at 8W
d. Circular inline vs staggered at 8W
Fig. 6. The temperature distribution on different configurations at 5 W and 8 W.
Fig. 7. Circular (inline, staggered) vs square (inline, staggered).
257
258
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
a: square Inline vs staggered
b: Round Inline vs staggered Fig. 8. Inter comparison between inline and staggered PCM materials at 5 W.
From above results it is known that inline fin arrangement is a better choice in both circular and square pin-fins heat sinks. To establish better of the two, a comparison for square inline and circular inline is illustrated in Fig. 10(a) and (b). It is clearly shown that circular inline heat sink is better for all PCMs with exception of n-eicosane. For n-eicosane square inline directs as more effective choice until at the end where circular inline takes over. Moreover, at 8 W n-eicosane and paraffin wax also deviate from the behaviour shown by all other PCMs. Conclusively, circular inline pin-fins arrangement surpasses the square inline pin-fins for all PCMs. It is evident that circular inline pin-fins are better because they enhance heat transfer effectively. Visualizing heat flowing in the centre of the fin, for a circular fin when heat flows from the base to the tip, at every part it can take whether a half diameter or half edge as a route out of the fin. Since half edge is shorter, it prefers it. This means that at every point across the radius of a circular fin edge, temperature is evenly distributed. The reason of better per-
formance of heat transfer rate in case of circular pin-fins in comparison of square pin-fins is because of optimum fin distribution and arrangement as well the number of fins. Despite of this, total surface area of pin-fins is not a limiting factor, in case of square pinfins, to increase or decrease the heat transfer rate, however, higher the number of fins even with optimum fins distribution and arrangement cause the rapid increase of heat transfer rate through the PCMs which resulting the increases the phase transition rate of PCMs [21,22]. To sum up, uniform temperature distribution at every segment of the circular fin leads to better heat transfer. 3.3. Comparison of different PCMs As circular inline is established in this study as the dominant and most effective arrangement for a PCM based heat sink, so it’s experimental reading will be considered in this section to study PCM behaviour relative to their properties and effect on overall performance of heat sink.
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
259
a: square Inline vs staggered
b: Round Inline vs staggered Fig. 9. Inter comparison between inline and staggered PCM materials at 8 W.
3.3.1. Effect of melting temperature and latent heat of PCM Melting temperature is one of the key aspects considered when choosing a PCM as a heat storage medium for passive cooling. Melting temperature affects the spectrum of PCM application ranges and the performance of the heat sinks integrated along with PCM. This experimentation focuses on PCM’s with a wide range of melting temperature from 31 C to 57 C. All PCM melting temperatures and corresponding latent heats are listed in Table 3. Considering circular inline, the effects of melting temperature are studied at 5 W and 8 W as it provides enough heat flux to melt all PCMs. As it is evident in Fig. 11, the point where all PCMs trends become less steep is the melting point of that PCM. These points are shown with a black encircled point. The changes in curve can also be observed at these points. But how does the melting point affect the system’s performance? The answer to this lies in the fact that it depends upon the heat flux generated in the system. For a
higher heat flux the corresponding temperature at base is also high. So, the PCM which is initially in the solid phase, absorbs heat with rise in base temperature. The temperature of thermocouple placed at H1 and H2 will reflect the external temperature until the PCM reaches its melting point. After the temperature ranges the melting point of the PCM, it starts to change phase and thus liquefies. As this liquefaction process occurs, the PCM absorbs in bulks of heat with almost no temperature variation. During this span, cooling effect is centred through the PCM. That span or amount of time the PCM usually provides a cooling effect is known as the latent heat of fusion. The enthalpy varies depending on the PCM material itself. In case of this experiment, the enthalpy is classically measured in kJ/kg. For higher number of kilo Joules per kilogram, the PCM will provide a cooling effect will be for a longer time. The PCM, which is now in molten phase, it releases the absorbed heat. The cycle reverses as soon as the base temperature
260
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
a: At 5W
b: At 8 W Fig. 10. Square inline and Circular Inline comparison.
cools. The PCM, now liquefied, releases the heat absorbed as the base temperature decreases. During this period, the PCM solidifies and generates a heating effect. Fig. 11a and b illustrate the PCMs with lower melting temperatures work best with lower power levels like 5 W and similarly PCM’s with higher melting points are considered for 8 W. Paraffin wax doesn’t reach its melting point in both 5 W and 8 W systems. Additionally, its latent heat is the lowest among the six PCMs. RT-54 like paraffin wax does not reach its melting point at 5 W but it does melt on the 35th minute at 8 W. Its latent heat is also low compared to other PCM’s used. RT-44 has highest latent heat amongst all. At 5 W, it provides a constant behaviour for most of its time and transfers heat with-
out even melting. But Fig. 11(b) shows that after surpassing its melting point at 20th minute, it starts to gain sensible heat by the 60th minute. So, its gradient becomes steeper. RT-35HC shows good heat transfer properties at 5 W temperature plots in Fig. 11(a), but changes this behaviour at 8 W. N-eicosane melts around the 35th minute but also advances to sensible heating by 80th minute. At 8 W, it shows similar character as that of RT-35HC i.e. moves towards sensible heating with sharp gradient. Both n-eicosane and RT-35HC have approximately similar melting temperatures and latent heats. SP-31 shows most heat absorption/dissipation in Fig. 11a and b. It has the lowest melting point among its all PCMs studied but also has lower latent heat than RT-44, n-eicosane and RT-35HC. The property allowing it the most heat absorption is its density which is 1300 kg=m3 in liquid phase, almost double the amount
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
261
Fig. 11. PCMs comparison of Round Inline at 8 W.
of all other PCMs. It is so because a lot of energy is required to overcome the strong bonds in the PCM. But it would also require a similar time span to cool the material as it was taken whilst heat absorption. So, based on its density its discharging time can be predicted as high as its charging phase. Although sensible heat is not studied in this experiment but the comparison of sensible heat and latent heat storage systems involving PCMs suggest that the latter has higher thermal energy storage density and so requires reduced material amounts in terms of masses and volumes. The sensible heat phase can be observed above the melting points in Fig. 11. 3.3.2. The effect of varying power level and enhancement in operation time In this section the enhancement in operation time is studied with respect to varying power levels, identifying the best PCM to use at a certain power level. This is depicted in terms of the time required to reach critical SPTs of 45 C and 60 C, and is shown in Fig. 12a and b. Here PCMs are divided into two groups, one with
those having lower melting temperatures whilst second with relatively higher melting temperatures. The first group consists of SP31, n-eicosane and RT-35HC and the second one has paraffin wax, RT-54 and RT-44. They are divided with respect to critical SPT. The critical SPT is defined as the maximum working temperature that an electronic device withstands without halt. From Fig. 12, it is apparent that for all PCMs the operation time decreases whilst the power input increases. It’s obvious that: SP-31 takes the most time to reach the critical SPT of 45 °C in rest of all other PCMs. n-eicosane and RT-35HC stand at close competition at all power levels but the latter ensures more heat transfer, showing much higher difference in enhancement time by 52 min at 5 W as shown in Fig. 12a. RT-44 dominates at SPT of 60 °C at 5 W but its operation time decreases rendering it least suitable for 7 W and 8 W. RT-54 has least operational time qualification for 5 W and 6 W but it proves to be more useful at higher power levels.
Fig. 12. Time to reach critical SPT at different Power levels for various PCMs.
262
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263
Fig. 13. Enhancement ratio of different PCMs.
Paraffin wax shows consistency in its enhancement in operational time throughout all the power levels. 2. 3.3.3. Enhancement ratio Fig. 13a and b present the enhancement ratio at two critical SPTs of 45 C and 60 C. For SPT of 45 C, PCMs of low melting point namely SP-31, RT-35HC and n-eicosane are considered, whereas for the SPT of 60 C, PCMs of high melting point like RT-54, RT-44 and paraffin wax are compared. The enhancement ratio, ePCM , is the SPT time ratio of heat sink with PCM to that of the same heat sink without PCM, calculated using Eq. (4).
ePCM ¼
tcrwith PCM tcrwithout PCM
3.
4.
ð4Þ
Here it becomes obvious that Sp-31 has the highest enhancement ratio for SPT of 45 °C. The enhancement ratio of RT-35HC is above n-eicosane at a power level of 5 W, but tends to become similar at 6 W and 7 W. At 8 W n-eicosane has a slightly better enhancement ratio than RT-35HC. For critical SPT of 60 °C paraffin wax stands apart as best option whilst RT-44 is better for low power levels and RT-54 for higher power levels. So, this makes it obvious that higher enhancement ratios are achieved on lower SPT of 45 C due to the higher time taken by PCM to complete its latent heat phase. This decline in the enhancement ratio at SPT of 60 °C as compared to SPT of 45 °C is caused by the abrupt rise in temperature after completion of the melting phase.
5.
6.
7.
array is found to be the most effective heat sink of all four geometries without PCMs. The second discussion was the pin-fin configuration comparison with PCMs. The studies of six PCMs, two geometries and varying power levels shows a unanimous result of inline as the dominant configuration for both square and circular cross-sections. Further comparison between circular inline and square inline helped in establishing circular inline as the most efficient choice for a PCM based heat sink. Using circular inline as dominant geometry, the effect of melting point and latent heat was studied. Latent heat is found as the major reason of more efficiency of PCMs heat transfer properties. RT-44, n-eicosane, SP-31 are dominant coolants at lower power levels whilst RT-54 was best option at higher power levels due to its mid-range thermal properties. The effect of varying power level and enhancement in operation time is also studied in terms of critical SPTs of 45 °C and 60 °C. The results suggested that SP-31 requires the most time to reach the critical SPT of 45 °C whilst paraffin wax is consistently better choice for SPT of 60 °C due to its high melting point. Finally, the enhancement ratio is considered as it becomes obvious that SP-31 has the highest enhancement ratio of 9.28 at 5 W followed by RT-35HC and n-eicosane. For 8 W power level systems paraffin wax is the best PCM with highest enhancement ratio.
4. Conclusions
Conflict of interest
Successful experimental study is conducted to single out an efficient working setup by passive cooling using PCM based heat sinks. Circular and square pin-fins are used for experimentation, with configurations varying between inline and staggered arrays. Power levels of 4 W to 8 W were switched keeping the volumetric fraction of PCM constant at 90%. Six PCMs were broadly studied for each configuration of heat sinks. From all these set variations and conditions, following deductions are made:
Authors declare no conflict of interests regarding this paper.
1. The pin-fin geometry and configuration comparison without PCM yielded staggered is the better arrangement in both square and circular cross-sections. Overall the square staggered pin-fin
References [1] S.F. Husseinizadeh, F.L. Tan, S.M. Moosania, Experimental and Numerical studies on performance of phase change material based Heat sinks with different configurations of internal fins, Appl. Thr. Engg. 31 (2011) 3827–3838. [2] R. Pakrouh, M.J. Hosseini, A.A. Ranjbar, R. Bahrampoury, A numerical method for phase change material based Heat sinks optimization, Energy Conv. Manag. 130 (2015) 542–552. [3] Saad Mahmoud, Aarun Tang, Chin Toh, Raya AL-Dadah, SeinLeung Soo, Experimental investigation of inserts configurations and Pcm Type on the thermal performance of Phase change material based Heat sinks, Appl. Energy 112 (2013) 1349–1356.
M.J. Ashraf et al. / International Journal of Heat and Mass Transfer 115 (2017) 251–263 [4] S.K. Saha, K. Srinivasan, P. Dutta, Studies on optimum distribution of fins in heat sinks filled with phase change materials, J. Heat Transf. 130 (2008), 034505-1. [5] R. Baby, C. Balaji, Thermal management of electronics using Phase Change material based pin fin Heat Sinks, 6th European Thr. Sciences Conference, J. Phys., IOP Publishing, 2012. [6] R. Sarikanth, Pavan Nemani, C. Balaji, Multi-objective geometric optimization of a Pcm Matrix Type Heat Sink, Appl. Energy 156 (2015) 703–714. [7] Li-Wu Fan, Yu-Qi Xiao, Yi Zeng, Xin Fang, Xiao Wang, Xu Xu, Zi-Tao Yu, RongHua Hong, Ya-Cai Hu, Ke-Fa Cen, Effect of melting temperature and the presence of internal fins on the performance of Pcm Based Heat sink, Int. J. Thr. Sci. 70 (2013) 114–126. [8] A.F. Mahrous, Thermal performance of phase change material based heat sinks, Int. J. Mech. Energy 2 (2013) 35–42. [9] Peleg P. Levin, Avraham Shitzar, Gad Hetsroni, Numerical optimization of phase change material based heat sink with internal fins, Int. J. Heat Mass Transf. 61 (2013) 638–645. [10] Jesto Thomas, P.V.S.S. Srivatsa, S. Ramesh Krishnan, Rajesh Baby, 2010, Thermal performance evaluation of phase change material based heat sink a numerical study, Global colloquium in recent advancement and researches in engineering, Science and Technology 2016, Elsevier Ltd, Procedia Technology 25 (2016) 1182–1190. [11] Atul Nagose, Ankit Somani, Aviral Shrot, Arunn Narasimhan, Genetic Algorithm based optimization of Pcm based heat sinks and effect of Heat sink parameters on operational time, J. Heat Transf. 130 (2008), 011401-1. [12] Tomoyuki Hatakeyama, Masaru Ishizuka, Sadakazu Takakuwa, Shinji Nakagawa, Kanjii Takagi, Experimental and thermal network study on the performance of a pins studded Pcm in electronic device cooling, J. Thr. Sci. Tech. 6 (2011) 11–0030.
263
[13] Ahmad Hassan, Hassan Hejase, Shaimaa Abdelbaqi, Ali Assi, Mohammed O. Hamdan, Comparative effectiveness of different phase change materials to improve cooling performance of Heat sinks for electronic devices, Appl. Sci. 6 (2016) 226. [14] S.C. Fok, W. Shen, F.L. Tan, Cooling of portable hand-held electronic devices using phase change materials in finned heat sinks, Int. J. Therm. Sci. 49 (1) (2010) 109–117. [15] R. Baby, C. Balaji, Thermal optimization of PCM based pin fin heat sinks an experimental study, Appl. Therm. Eng. 54 (1) (2013) 65–77. [16] A. Arshad et al., Thermal performance of phase change material (PCM) based pin-finned heat sinks for electronics devices effect of pin thickness and PCM volume fraction, Appl. Therm. Eng. 112 (2017) 143–155. [17] EMD Millipore is a part of Merck KGaA, D., Germany, 2016, Available from http://www.emdmillipore.com/US/en/product/Histosec-pastilles,MDA_ CHEM-111609. [18] ‘‘Eicosane 442673.” Sigma-Aldrich Italy. N.p., n.d. Web. 04. 2016, Available from http://www.sigmaaldrich.com/catalog/product/aldrich/219274/. [19] ‘‘Rubitherm GmbH.” Rubitherm GmbH, Gemany. N.p., n.d. Web. 2016, Available from https://www.rubitherm.eu/. [20] G.W. Burns et al., Temperature-electromotive force reference functions and tables for the letter-designated thermocouple types based on the ITS-90. NASA STI/Recon Technical Report N, 1993, p. 93. [21] Yue-Tzu Yang et al., Numerical simulations of electronic cooling with in-line and staggered pin fin heat sinks, IJMME 10 (2016) 1172-1177. [22] Eaman Hassan Muhammad, A comparison of the heat transfer performance of a hexagonal pin fin with other types of pin fin heat sinks, Int. J. Sci. Res. 4 (2015) 1781–1789.