- Email: [email protected]
Experimental study of an ammonia loop heat pipe with a flat plate evaporator
International Journal of Heat and Mass Transfer 102 (2016) 1050–1055
Contents lists available at ScienceDirect
International Journal of Heat and Mass Transfer journal homepage: www.elsevier.com/locate/ijhmt
Experimental study of an ammonia loop heat pipe with a flat plate evaporator He Song, Liu Zhi-chun ⇑, Zhao Jing, Jiang Chi, Yang Jin-guo, Liu Wei School of Energy and Engineering Power, Huazhong University of Science and Technology, Wuhan 430074, China
a r t i c l e
i n f o
Article history: Received 7 May 2016 Received in revised form 4 July 2016 Accepted 5 July 2016
Keywords: Loop heat pipe Flat plate evaporator Secondary wick Biporous wick Thermal characteristics
a b s t r a c t In the present paper, the combination of a primary wick and a secondary wick was put forward and applied to the loop heat pipe. The evaporator was constructed from 304L stainless steel, and anhydrous ammonia with a purity of 99.995% was selected as the working fluid. A secondary wick made of 400-mesh stainless steel wire mesh was inserted into the evaporator, and a blind hole was made in its radial direction. Some partial grooves at the end of liquid line served as the returning liquid exit, and it was wrapped by the blind hole for fear that it be directly exposed to the compensation chamber, potentially causing it to be blocked by the vapor. The experimental results demonstrated that the loop could start up successfully and operated without temperature oscillations under the given heat load range from 20 W to 110 W in the horizontal orientation. In addition, the system rapidly responded to heat load cycle. The thermal resistance of the loop and the evaporator heat-transfer capacity were analyzed, and two operation modes existed because of the different distributions of the operating temperature with the evaporator wall temperature above or below 60 °C. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Two-phase heat transfer devices have been widely studied owing to their high efficiency, high stability, and good compatibility with electronic equipment [1,2]. Loop heat pipes (LHPs) are two-phase heat transfer device that possess many advantages: long distance heat transport, small thermal resistance, low mass, and simple structure. Therefore, LHPs have great application prospects in the field of thermal control of electronic instruments. The factors affecting the operating performance of LHPs, including evaporator structure, working fluid, charge ratio, and operating conditions, have been investigated both theoretically and experimentally [3–5]. The evaporator plays prominent role in enhancing the heat transfer capacity of the LHPs, and the main evaporator shapes consist of two types: cylindrical and flat [6]. The evaporator structure is closely related to the location of the compensation chamber (CC) and the path of the working fluid supply, further determining the maximum heat transfer capacity. For the two types of evaporators, there are two different arrangements of the CC and evaporator. For cylindrical evaporators, the CC is located at the end of the evaporator; alternatively,
⇑ Corresponding author at: 318 Power Building, 1037 Luoyu Road, Hongshan District, Wuhan 430074, China. Fax: +86 27 87540724. E-mail address: [email protected] (L. Zhi-chun). http://dx.doi.org/10.1016/j.ijheatmasstransfer.2016.07.014 0017-9310/Ó 2016 Elsevier Ltd. All rights reserved.
for flat evaporators, the CC is integrated inside the evaporator. Numerous researchers have studied different evaporator structures for gaining the best operating performance. Maydanik et al. [7] tested 15 different variants of miniature LHPs with cylindrical evaporators with ammonia as the working fluid, and achieved a maximum heat flux of 135 W/cm2. For LHPs with the flat evaporator, Liu et al. [8–10] and Wang et al. [11–13] have conducted numerous fundamental experiments, especially regarding the optimized design of the evaporator structure. On the other hand, among the loop components, porous wicks are the key to improve the LHP operating performance because the capillary force developed on the evaporating surface of the wick drives the circulation of the working fluid. Numerous investigations about the thermal and hydraulic performance of the wick have been made. Aiming at improving capillary force and effectively organizing two phase flow in porous wick, efforts have been made to fabricate multi-layer wicks and biporous wicks with large pores and small pores in Ref. [14–17]. Xu et al. [18] combined modulated porous wick, secondary monoporous wick and third thermal insulation wick, which could provide good thermal and liquid link between the CC and the evaporating surface, enlarging capillary force and reducing heat leakage. In existing flat evaporators, most do not contain a secondary wick, and the primary wick is directly placed at the flange of the evaporator envelope [10,19], which increase the thickness of the
1051
H. Song et al. / International Journal of Heat and Mass Transfer 102 (2016) 1050–1055
Nomenclature A Ke Q app Rev ap RLHP
evaporator heating area, m2 overall heat-transfer coefficient of the evaporator, W/(m2°C) heat load applied to the evaporator active zone, W evaporator thermal resistance, °C/W thermal resistance of the LHP, °C/W
evaporator wall and heat leakage from the active zone. In the present paper, an annular secondary wick was inserted into the evaporator, and a blind hole was made in its radial direction. This idea of implementing a secondary wick into the evaporator of an LHP system was developed and tested. The secondary wick—made of stainless steel wire mesh—contributed not only to the tight contact between the primary wick and the inner surface of the evaporator active zone, but also supported the primary wick while providing a liquid link between the CC and the evaporator. In a horizontal orientation (the evaporator active zone below the CC) under different heat sink temperatures, the startup performance and stability of continuous operation within a heat load cycle were carried out. Specifically, the thermal characteristics of the LHP were analyzed in detail and comparison with other studies was also made.
T cond Te Tv
average temperature of the condenser inlet and outlet measured by Tc8 and Tc12, °C average temperature of the evaporator active zone measured by Tc1–Tc4, °C temperature of the vapor exiting the evaporator measured by Tc6, °C
the end was inserted into the blind hole for fear that it be directly exposed to the CC, potentially causing it to be blocked by the vapor. A cylinder with three embedded heaters, which were controlled by a voltage regulator and wattmeter, provided power input for the LHP system. A double pipe condenser was used to dissipate heat carried by the vapor. The external loop was made of stainless steel tube and wrapped by adiabatic material. Fig. 3 gives the system diagram of the LHP and the locations of the temperature measuring points. In order to minimize the effect of non-condensable gas, the loop was evacuated to a pressure of 3.7 104 Pa before charging with ammonia. The charging ratio was determined at 65% of the total volume of the loop: approximately 10.4 cm3 at an ambient temperature of 20 °C. During the tests, the temperature in the laboratory room was maintained between 24 °C and 28 °C, and the temperature of vapor exiting the evaporator was set below 35 °C for safety.
2. Description of experimental setup The investigated secondary wick was inserted into a 304L stainless steel LHP with a disk-shaped evaporator; the working fluid was anhydrous ammonia with a purity of 99.995%. Considering the thermophysical properties of the working fluid, ammonia has the advantages of a steep slope of vapor pressure to temperature and considerable latent heat of vaporization. However, the working pressure of ammonia is much higher than other fluids, such as water, methanol and acetone, which pose a challenge for the fabrication of the flat evaporator. Table 1 gives design parameters of the loop components. The tested evaporator comprises an upper tube structure, bottom cover plate, primary wick, and secondary wick, as shown in Fig. 1. A sample evaporator was used to test the pressure-bearing ability of the device prior to the beginning of the practical experiment. The upper tube structure and bottom cover plate were soldered by laser weld, which could bear an approximate pressure of 2000 kPa. The secondary wick was made of multiple layers of 400-mesh stainless steel wire mesh using wire-cutting technology, and a blind hole with a length of 32 mm was formed in its radial direction, as shown in Fig. 2. A biporous wick, which was introduced in Ref. [20], was adopted as the primary wick. The primary wick, secondary wick, and the bottom surface of the upper tube structure encircled a cavity as the CC. Some partial grooves at the end of the liquid line formed the returning liquid exit, and
3. Test results and discussion 3.1. Startup tests Successful startup is crucial for any type of LHP and the necessary precondition that demonstrates the LHP’s application potential. Startup of an LHP is closely connected with the CC and evaporator structure, initial conditions inside the evaporator, and operation termination prior to startup [21]. Good startup performance guarantees subsequent steady operation. When heat load is applied to the evaporator, the pressure difference across the primary wick drives the circulation of the working fluid in the horizontal orientation. The required pressure difference—affected by heat leakage from the heating zone and vapor penetration—corresponds to the temperature difference between the CC and the heating zone. Well-organized thermal distribution inside the evaporator and appropriate evaporator structure contribute to the establishment of this pressure difference. In this experiment, the CC was separated from the heating zone by a secondary wick that improves sealing quality and makes the primary wick tightly attach to the inner surface of the evaporator active zone, further leading to a fast startup. Fig. 4 illustrates the startup process with 10 W at a sink temperature of 5 °C. When the heat load was applied to the evaporator
Table 1 Main design parameters of the loop components. Evaporator
Overall height Heating surface diameter
12 mm 36 mm
Compensation chamber
Height Volume
5.1 mm 4.6 ml
Secondary wick
Thickness I.D./O.D. Hole depth
5.1 mm 27/37.4 mm 32 mm
Primary wick
Thickness Groove width/Groove depth Diameter Porosity
3.83 mm 2 mm/1.5 mm 37.4 mm 78.69%
Condenser
Length I.D./O.D.
910 mm 16 mm/20 mm
Liquid/vapor line
Length I.D./O.D.
515 mm/320 mm 2 mm/3 mm
1052
H. Song et al. / International Journal of Heat and Mass Transfer 102 (2016) 1050–1055
Fig. 1. Assembly diagram of the evaporator structure.
the condenser outlet and the evaporator entrance was approximately 4 °C, which was much less than that of the startup with 10 W. Comparing the two startup processes, the loop could start up successfully and possessed subsequent steady-state operation. The secondary wick was responsible for the separation between the CC and the evaporator active zone, laying a good foundation for LHP operation under high pressure. Furthermore, no temperature oscillation was observed during either startup test.
3.2. Operation with heat load cycle
Fig. 2. Schematic diagram of the secondary wick.
active zone, the generated vapor immediately rushed into the condenser, and the monitored temperature of the condenser inlet and outlet abruptly changed. The temperature of the evaporator wall gradually rose, resulting in an increase in evaporation intensity. The liquid in the condenser was pushed back to the CC with the help of the capillary pressure and phase-change driving force, subsequently accompanied by a temperature decrease at the evaporator entrance. The loop cycle quickly stabilized as normal circulation of the working fluid was established. Considering that the returning liquid exchanged heat with the environment, the temperature difference between the condenser outlet and the evaporator entrance reached up to 21 °C. Fig. 5 depicts the startup process with 100 W at a heat sink temperature of 5 °C. With the increase in heat load to the evaporator active zone, a faster startup process was achieved and the mass flow rate increased. The liquid exiting the condenser carried more subcooling and took less time to flow back to the CC, so the heat gain of the liquid decreased. The temperature difference between
Electronic instruments inevitably experience input power change, leading to inconsistent heat generation. However, variable input power for LHP operation produces different distributions of working fluid in the loop, some of which may lead to depriming. Continuous operation was tested under stepwise and random heat load cycles aimed at validating the reliability of the LHP, as shown in Fig. 6. Firstly, the heat load was varied with a step of 10 W or 20 W, increasing from 20 W to 110 W, and then decreasing from 110 W to 10 W. During the entire testing procedure, heat load changes were implemented every 30 min with steady-state heat load between changes. While the heat load increased from 20 W to 110 W, the LHP could quickly switch from the previous stable operation to the next, and no operation failure was observed. While the heat load decreased from 60 W to 40 W, the temperature of the evaporator wall dropped excessively before eventually stabilizing. However, similar behavior was not observed for the same heat load decrement from 110 W to 90 W. The change amplitude of heat load applied to the evaporator had an impact on the operating behavior of the LHP, depending to an extent on the redistribution of the working fluid inside the loop. For small heat loads, a large head load change easily presented unstable phenomena. For the evaporator with a combination of the primary wick and
Fig. 3. System diagram of the LHP and the locations of T-type thermocouples.
H. Song et al. / International Journal of Heat and Mass Transfer 102 (2016) 1050–1055
Fig. 4. Startup with 10 W at 5 °C.
Fig. 5. Startup with 100 W at 5 °C.
the secondary wick, the loop had the ability to self-adjust, avoiding depriming. 3.3. Analysis of thermal characteristics 3.3.1. Temperature distribution The temperature distribution showed a typical trend during LHP operation, as shown in Fig. 7. There also existed two operation modes at the demarcation point of 50 W: variable conduction mode and constant conduction mode. During the test, the external loop was insulated to reduce heat-exchange with the environment. However, the liquid exiting the condenser gained a large amount of heat from the environment because of the low flow rate of the liquid in the liquid transport line. When heat load to the evaporator active zone was increased from 20 W to 50 W, the CC gradually got filled with liquid and the temperature difference between the evaporator entrance and condenser outlet diminished with the increase in heat load. The CC temperature gradually decreased and possessed the lowest temperature at 50 W, the loop operated at variable conduction mode. With heat load was increased from 50 W to 110 W, the operating temperature linearly increased and the condenser was fully utilized. The loop operated at constant conduction mode.
1053
Fig. 6. Continuous operation with stepwise heat load changes at 0 °C.
Fig. 7. Temperature distribution of the LHP at 0 oC.
From the temperature data of the evaporator exit and condenser inlet, the vapor experienced negligible heat-exchange along the vapor line. The temperature of the evaporator back (close to the CC) continuously decreased as heat load increased from 20 W to 50 W before stabilizing at 20 °C for heat loads from 50 W to 100 W. This implied that heat leakage to the CC was completely compensated for by the subcooled liquid, neglecting heatexchange with the environment. 3.3.2. Evaporator performance Evaporator thermal resistance is usually regarded as the index of evaporator heat-transfer capacity. The evaporator thermal resistance, Rev ap , is a comprehensive result of heat conduction through the evaporator wall and heat transfer from the inner surface of the evaporator to the working fluid. Conduction through the evaporator wall is mainly affected by the thickness of the evaporator wall and its material, whereas heat transfer from the evaporator to the working fluid depends on many factors, including the quality of the thermal contact of the wick with the evaporator wall, the thermophysical parameters of the wick, and the structure of the vapor-removal channels [7]. The evaporator thermal resistance can be calculated using
1054
H. Song et al. / International Journal of Heat and Mass Transfer 102 (2016) 1050–1055
Rev ap ¼
Te Tv : Q app
ð1Þ
The heat-transfer coefficient K e can be defined as
Ke ¼
Q app : AðT e T v Þ
ð2Þ
For conventional flat evaporators, the primary wick is directly located on the flange of the evaporator sidewall, which cannot guarantee the quality of the thermal contact of the inner surface of the evaporator with the primary wick. In order to address the problem, the primary wick may be tightly attached to the inner surface using the elasticity of the secondary wick. Fig. 8 gives the relations for thermal resistance and heat-transfer coefficient against heat load, the experimental results showed that the evaporator thermal resistance varied between 0.256 °C/W and 0.369 °C/W; the minimum value of 0.256 °C/W was achieved at 60 W. From Fig. 8, the evaporator thermal resistance gradually decreased with heat load increasing from 20 W to 60 W, and then increased until the applied heat load reached 110 W. Simultaneously, the heat-transfer coefficient, K e , varied from 3470 W/ (m2K) to 5430 W/(m2K), and the maximum heat-transfer coefficient was achieved at 60 W. The heat-transfer coefficient first increased and then decreased with the increase in heat load, whose trend was similar to that in Ref. [22,23]. For evaporative heat transfer in a capillary structure, the vapor–liquid distribution in the porous media varied with heat load change, and yielded different heat transfer mechanisms. While the heat load increased from 20 W to 60 W, the evaporation occurring on the wick structure surface intensified. More bubbles generated on the nucleation sites of the wick, and fled from the surface. Therefore, heat transfer on the capillary wick surface was enhanced. Vapor film formed on the active zone of the wick beyond 60 W, and increased thermal resistance from the inner surface of the evaporator to the wick, leading to a decrease in heat-transfer coefficient. 3.3.3. Thermal resistance of the LHP The LHP thermal resistance, RLHP , from the evaporator to the condenser represents the heat-transfer capacity of the LHP system:
RLHP ¼
T e T cond : Q app
ð3Þ
From the definition of the LHP thermal resistance, it is related to the evaporator heat-transfer capacity and the condenser cooling efficiency. Therefore, the thermal resistance of the LHP can be reduced by improving the evaporator structure and the condenser cooling efficiency. The overall thermal resistance, RLHP , ranged from 0.337 °C/W to 1.152 °C/W; the minimum value appeared at a heat load of 100 W, as depicted in Fig. 8. The evaporator thermal resistance occupied a small share of the LHP thermal resistance at heat loads less than 50 W, as the condenser cooling efficiency was relatively low. For heat loads over 50 W, the condenser thermal resistance was responsible for the LHP thermal resistance. For heat loads exceeding 50 W, the temperature difference between the condenser inlet and outlet became large while the condenser outlet temperature remained nearly unchanged.
3.3.4. Comparison with other studies Loop heat pipes with the flat evaporator have attracted wide attention from all over the world, because they have robust operation and a flat thermocontact surface. Thus, the comparison has been majorly made with loop heat pipes with the flat evaporator. Singh et al. [24] proposed a copper-nickel-water mLHP in which the CC was positioned on the side of the wick. In the horizontal orientation, the minimum LHP thermal resistance achieved at 50 W was 0.62 °C/W. Celata et al. [25] investigated a stainless steel LHP with flat evaporator, with water as the working fluid. The maximum transferred heat load was 75 W, and its LHP thermal resistance varied between 3.33 °C/W and 50.7 °C/W at horizontal elevation. Chen et al. [19] tested stainless-steel–ammonia loop heat pipes (LHPs) with flat evaporator, their total thermal resistance ranged from 0.33 °C/W to 1.47 °C/W. This research mainly studied the application of the combination of a primary wick and a secondary wick to a loop heat pipe with ammonia as the working fluid. With the evaporator wall temperature below 60 °C, the desired thermal performance was achieved, and the LHP thermal resistance from the evaporator active zone to the condenser distributed between 0.337 °C/W and 1.152 °C/W.
4. Conclusions In the present work, the combination of a primary wick and a secondary wick was applied to a loop heat pipe with the flat plate evaporator, and a series of tests were designed to examine its operating performance. The following conclusions could be drawn from the experimental tests. (1) The LHP could achieve a fast startup process at low heat loads and transfer 110 W (corresponding heat flux of 10.8 W/cm2) with the evaporator wall below 60 °C in the horizontal orientation. (2) The evaporator integrated with a combination of the primary wick and the secondary wick had the ability to selfadjust under the operation with heat load change. (3) Transition point of operation modes (50 W) and the evaporator thermal resistance (60 W) did not correspond to the same heat load. (4) The thermal resistance of the LHP was mainly affected by the condenser thermal resistance at low heat loads and limited by the evaporator capacity at high heat loads.
Conflict of interest Fig. 8. Thermal characteristics dependence of heat load at 0 °C.
None declared.
H. Song et al. / International Journal of Heat and Mass Transfer 102 (2016) 1050–1055
Acknowledgement The current work was supported by the National Natural Science Foundation of China (Grant Nos. 51276071 and 51376069).
References [1] Vladimir.G. Pastukhov, Yuri.F. Maydanik, Combined LHP and PHP based heattransfer system, Int. J. Therm. Sci. 74 (2013) 81–85. [2] Yury.F. Maydanik, Valery.I. Dmitrin, Vladimir.G. Pastukhov, Compact cooler for electronics on the basis of a pulsating heat pipe, Appl. Therm. Eng. 29 (2009) 3511–3517. [3] Benjamin Siedel, Valérie Sartre, Frédéric Lefèvre, Literature review: Steadystate modeling of loop heat pipes, Appl. Therm. Eng. 75 (2015) 709–723. [4] Mariya A. Chernysheva, Vladimir.G. Pastukhov, Yury.F. Maydanik, Analysis of heat exchange in the compensation chamber of a loop heat pipe, Energy 55 (2013) 253–262. [5] Yu.F. Maydanik, Review: Loop heat pipes, Appl. Therm. Eng. 25 (2005) 635–657. [6] Yury F. Maydanik, Mariya A. Chernysheva, Vladimir.G. Pastukhov, Review: loop heat pipes with flat evaporators, Appl. Therm. Eng. 67 (2014) 294–307. [7] Yury F. Maydanik, Sergey V. Vershinin, Development and tests of ammonia miniature loop heat pipes with cylindrical evaporators, Appl. Therm. Eng. 29 (2009) 2297–2301. [8] Zhichun Liu, Dongdong Wang, Chi Jiang, Jinguo Yang, Wei Liu, Experimental study on loop heat pipe with two-wick flat evaporator, Int. J. Therm. Sci. 94 (2015) 9–17. [9] Zhichun Liu, Dongxing Gai, Huan Li, Wei Liu, Jinguo Yang, Mengmeng Liu, Investigation of impact of different working fluids on the operational characteristics of miniature LHP with flat evaporator, Appl. Therm. Eng. 31 (2011) 3387–3392. [10] Zhi Chun Liu, Huan Li, Bin Bin Chen, Jin Guo Yang, Wei Liu, Operational characteristics of flat type loop heat pipe with biporous wick, Int. J. Therm. Sci. 58 (2012) 180–185. [11] Dongdong Wang, Zhichun Liu, Song He, Jinguo Yang, Wei Liu, Operational characteristics of a loop heat pipe with a flat evaporator and two primary biporous wicks, Int. J. Heat Mass Transfer 89 (2015) 33–41.
1055
[12] Song He, Zhi Chun Liu, Dong Dong Wang, Jing Zhao, Wei Liu, Jinguo Yang, Investigation of the flat disk-shaped LHP with a shared compensation chamber, Appl. Therm. Eng. 104 (2016) 139–145. [13] Dongdong Wang, Zhichun Liu, Jun Shen, Chi Jiang, Binbin Chen, Jinguo Yang, Tu Zhengkai, Wei Liu, Experimental study of the loop heat pipe with a flat diskshaped evaporator, Exp. Therm. Fluid Sci. 57 (2014) 157–164. [14] X.L. Cao, P. Cheng, T.S. Zhao, Experimental study of evaporative heat transfer in sintered copper bidispersed wick structures, J. Thermophys. Heat Transfer 16 (4) (2002). [15] Wei Zhou, Weisong Ling, Lian Duan, K.S. Hui, K.N. Hui, Development and tests of loop heat pipe with multi-layer metal foams as wick structure, Appl. Therm. Eng. 94 (2016) 324–330. [16] Wu. Shen-Chun, Dawn. Wang, Yau.-Ming. Chen, Investigating the effect of double-layer wick thickness ratio on heat transfer performance of loop heat pipe, Int. J. Therm. Sci. 86 (2014) 292–298. [17] Randeep Singh, Aliakbar Akbarzadeh, Masataka Mochizuki, Effect of wick characteristics on the thermal performance of the miniature loop heat pipe, J. Heat Transfer 131 (2009). 082601-082601-10. [18] Xu Jinliang, Xianbing Ji, Wolong Yang, Ziwei Zhao, Modulated porous wick evaporator for loop heat pipes: experiment, Int. J. Heat Mass Transfer 72 (2014) 163–176. [19] B.B. Chen, W. Liu, Z.C. Liu, H. Li, J.G. Yang, Experimental investigation of loop heat pipe with flat evaporator using biporous wick, Appl. Therm. Eng. 42 (2012) 34–40. [20] Huan Li, Zhi Chun Liu, Bin Bin Chen, Wei Liu, Chen Li, Jinguo Yang, Development of biporous wicks for flat-plate loop heat pipe, Exp. Therm. Fluid Sci. 37 (2012) 91–97. [21] Jentung Ku, Operating characteristics of loop heat pipes, 29th International Conference on Environmental Systems, Denver, Colorado, July 12–15, 1999. [22] Q. Liao, T.S. Zhao, Evaporative heat transfer in a capillary structure heated by a grooved block, J. Thermophys. Heat Transfer 13 (1999) 126–133. [23] F.J. Hong, P. Cheng, H.Y. Wu, Z. Sun, Evaporation/boiling heat transfer on capillary feed copper particle sintered porous wick at reduced pressure, Int. J. Heat Mass Transfer 63 (2013) 389–400. [24] Randeep Singh, Aliakbar Akbarzadeh, Chris Dixon, Masataka Mochizuki, Novel design of a miniature loop heat pipe evaporator for electronic cooling, J. Heat Transfer 129 (2007) 1445–1452. [25] Gian Piero Celata, Maurizio Cumo, Massimo Furrer, Experimental tests of a stainless steel loop heat pipe with flat evaporator, Exp. Therm. Fluid Sci. 34 (2010) 866–878.
Contents lists available at ScienceDirect
International Journal of Heat and Mass Transfer journal homepage: www.elsevier.com/locate/ijhmt
Experimental study of an ammonia loop heat pipe with a flat plate evaporator He Song, Liu Zhi-chun ⇑, Zhao Jing, Jiang Chi, Yang Jin-guo, Liu Wei School of Energy and Engineering Power, Huazhong University of Science and Technology, Wuhan 430074, China
a r t i c l e
i n f o
Article history: Received 7 May 2016 Received in revised form 4 July 2016 Accepted 5 July 2016
Keywords: Loop heat pipe Flat plate evaporator Secondary wick Biporous wick Thermal characteristics
a b s t r a c t In the present paper, the combination of a primary wick and a secondary wick was put forward and applied to the loop heat pipe. The evaporator was constructed from 304L stainless steel, and anhydrous ammonia with a purity of 99.995% was selected as the working fluid. A secondary wick made of 400-mesh stainless steel wire mesh was inserted into the evaporator, and a blind hole was made in its radial direction. Some partial grooves at the end of liquid line served as the returning liquid exit, and it was wrapped by the blind hole for fear that it be directly exposed to the compensation chamber, potentially causing it to be blocked by the vapor. The experimental results demonstrated that the loop could start up successfully and operated without temperature oscillations under the given heat load range from 20 W to 110 W in the horizontal orientation. In addition, the system rapidly responded to heat load cycle. The thermal resistance of the loop and the evaporator heat-transfer capacity were analyzed, and two operation modes existed because of the different distributions of the operating temperature with the evaporator wall temperature above or below 60 °C. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Two-phase heat transfer devices have been widely studied owing to their high efficiency, high stability, and good compatibility with electronic equipment [1,2]. Loop heat pipes (LHPs) are two-phase heat transfer device that possess many advantages: long distance heat transport, small thermal resistance, low mass, and simple structure. Therefore, LHPs have great application prospects in the field of thermal control of electronic instruments. The factors affecting the operating performance of LHPs, including evaporator structure, working fluid, charge ratio, and operating conditions, have been investigated both theoretically and experimentally [3–5]. The evaporator plays prominent role in enhancing the heat transfer capacity of the LHPs, and the main evaporator shapes consist of two types: cylindrical and flat [6]. The evaporator structure is closely related to the location of the compensation chamber (CC) and the path of the working fluid supply, further determining the maximum heat transfer capacity. For the two types of evaporators, there are two different arrangements of the CC and evaporator. For cylindrical evaporators, the CC is located at the end of the evaporator; alternatively,
⇑ Corresponding author at: 318 Power Building, 1037 Luoyu Road, Hongshan District, Wuhan 430074, China. Fax: +86 27 87540724. E-mail address: [email protected] (L. Zhi-chun). http://dx.doi.org/10.1016/j.ijheatmasstransfer.2016.07.014 0017-9310/Ó 2016 Elsevier Ltd. All rights reserved.
for flat evaporators, the CC is integrated inside the evaporator. Numerous researchers have studied different evaporator structures for gaining the best operating performance. Maydanik et al. [7] tested 15 different variants of miniature LHPs with cylindrical evaporators with ammonia as the working fluid, and achieved a maximum heat flux of 135 W/cm2. For LHPs with the flat evaporator, Liu et al. [8–10] and Wang et al. [11–13] have conducted numerous fundamental experiments, especially regarding the optimized design of the evaporator structure. On the other hand, among the loop components, porous wicks are the key to improve the LHP operating performance because the capillary force developed on the evaporating surface of the wick drives the circulation of the working fluid. Numerous investigations about the thermal and hydraulic performance of the wick have been made. Aiming at improving capillary force and effectively organizing two phase flow in porous wick, efforts have been made to fabricate multi-layer wicks and biporous wicks with large pores and small pores in Ref. [14–17]. Xu et al. [18] combined modulated porous wick, secondary monoporous wick and third thermal insulation wick, which could provide good thermal and liquid link between the CC and the evaporating surface, enlarging capillary force and reducing heat leakage. In existing flat evaporators, most do not contain a secondary wick, and the primary wick is directly placed at the flange of the evaporator envelope [10,19], which increase the thickness of the
1051
H. Song et al. / International Journal of Heat and Mass Transfer 102 (2016) 1050–1055
Nomenclature A Ke Q app Rev ap RLHP
evaporator heating area, m2 overall heat-transfer coefficient of the evaporator, W/(m2°C) heat load applied to the evaporator active zone, W evaporator thermal resistance, °C/W thermal resistance of the LHP, °C/W
evaporator wall and heat leakage from the active zone. In the present paper, an annular secondary wick was inserted into the evaporator, and a blind hole was made in its radial direction. This idea of implementing a secondary wick into the evaporator of an LHP system was developed and tested. The secondary wick—made of stainless steel wire mesh—contributed not only to the tight contact between the primary wick and the inner surface of the evaporator active zone, but also supported the primary wick while providing a liquid link between the CC and the evaporator. In a horizontal orientation (the evaporator active zone below the CC) under different heat sink temperatures, the startup performance and stability of continuous operation within a heat load cycle were carried out. Specifically, the thermal characteristics of the LHP were analyzed in detail and comparison with other studies was also made.
T cond Te Tv
average temperature of the condenser inlet and outlet measured by Tc8 and Tc12, °C average temperature of the evaporator active zone measured by Tc1–Tc4, °C temperature of the vapor exiting the evaporator measured by Tc6, °C
the end was inserted into the blind hole for fear that it be directly exposed to the CC, potentially causing it to be blocked by the vapor. A cylinder with three embedded heaters, which were controlled by a voltage regulator and wattmeter, provided power input for the LHP system. A double pipe condenser was used to dissipate heat carried by the vapor. The external loop was made of stainless steel tube and wrapped by adiabatic material. Fig. 3 gives the system diagram of the LHP and the locations of the temperature measuring points. In order to minimize the effect of non-condensable gas, the loop was evacuated to a pressure of 3.7 104 Pa before charging with ammonia. The charging ratio was determined at 65% of the total volume of the loop: approximately 10.4 cm3 at an ambient temperature of 20 °C. During the tests, the temperature in the laboratory room was maintained between 24 °C and 28 °C, and the temperature of vapor exiting the evaporator was set below 35 °C for safety.
2. Description of experimental setup The investigated secondary wick was inserted into a 304L stainless steel LHP with a disk-shaped evaporator; the working fluid was anhydrous ammonia with a purity of 99.995%. Considering the thermophysical properties of the working fluid, ammonia has the advantages of a steep slope of vapor pressure to temperature and considerable latent heat of vaporization. However, the working pressure of ammonia is much higher than other fluids, such as water, methanol and acetone, which pose a challenge for the fabrication of the flat evaporator. Table 1 gives design parameters of the loop components. The tested evaporator comprises an upper tube structure, bottom cover plate, primary wick, and secondary wick, as shown in Fig. 1. A sample evaporator was used to test the pressure-bearing ability of the device prior to the beginning of the practical experiment. The upper tube structure and bottom cover plate were soldered by laser weld, which could bear an approximate pressure of 2000 kPa. The secondary wick was made of multiple layers of 400-mesh stainless steel wire mesh using wire-cutting technology, and a blind hole with a length of 32 mm was formed in its radial direction, as shown in Fig. 2. A biporous wick, which was introduced in Ref. [20], was adopted as the primary wick. The primary wick, secondary wick, and the bottom surface of the upper tube structure encircled a cavity as the CC. Some partial grooves at the end of the liquid line formed the returning liquid exit, and
3. Test results and discussion 3.1. Startup tests Successful startup is crucial for any type of LHP and the necessary precondition that demonstrates the LHP’s application potential. Startup of an LHP is closely connected with the CC and evaporator structure, initial conditions inside the evaporator, and operation termination prior to startup [21]. Good startup performance guarantees subsequent steady operation. When heat load is applied to the evaporator, the pressure difference across the primary wick drives the circulation of the working fluid in the horizontal orientation. The required pressure difference—affected by heat leakage from the heating zone and vapor penetration—corresponds to the temperature difference between the CC and the heating zone. Well-organized thermal distribution inside the evaporator and appropriate evaporator structure contribute to the establishment of this pressure difference. In this experiment, the CC was separated from the heating zone by a secondary wick that improves sealing quality and makes the primary wick tightly attach to the inner surface of the evaporator active zone, further leading to a fast startup. Fig. 4 illustrates the startup process with 10 W at a sink temperature of 5 °C. When the heat load was applied to the evaporator
Table 1 Main design parameters of the loop components. Evaporator
Overall height Heating surface diameter
12 mm 36 mm
Compensation chamber
Height Volume
5.1 mm 4.6 ml
Secondary wick
Thickness I.D./O.D. Hole depth
5.1 mm 27/37.4 mm 32 mm
Primary wick
Thickness Groove width/Groove depth Diameter Porosity
3.83 mm 2 mm/1.5 mm 37.4 mm 78.69%
Condenser
Length I.D./O.D.
910 mm 16 mm/20 mm
Liquid/vapor line
Length I.D./O.D.
515 mm/320 mm 2 mm/3 mm
1052
H. Song et al. / International Journal of Heat and Mass Transfer 102 (2016) 1050–1055
Fig. 1. Assembly diagram of the evaporator structure.
the condenser outlet and the evaporator entrance was approximately 4 °C, which was much less than that of the startup with 10 W. Comparing the two startup processes, the loop could start up successfully and possessed subsequent steady-state operation. The secondary wick was responsible for the separation between the CC and the evaporator active zone, laying a good foundation for LHP operation under high pressure. Furthermore, no temperature oscillation was observed during either startup test.
3.2. Operation with heat load cycle
Fig. 2. Schematic diagram of the secondary wick.
active zone, the generated vapor immediately rushed into the condenser, and the monitored temperature of the condenser inlet and outlet abruptly changed. The temperature of the evaporator wall gradually rose, resulting in an increase in evaporation intensity. The liquid in the condenser was pushed back to the CC with the help of the capillary pressure and phase-change driving force, subsequently accompanied by a temperature decrease at the evaporator entrance. The loop cycle quickly stabilized as normal circulation of the working fluid was established. Considering that the returning liquid exchanged heat with the environment, the temperature difference between the condenser outlet and the evaporator entrance reached up to 21 °C. Fig. 5 depicts the startup process with 100 W at a heat sink temperature of 5 °C. With the increase in heat load to the evaporator active zone, a faster startup process was achieved and the mass flow rate increased. The liquid exiting the condenser carried more subcooling and took less time to flow back to the CC, so the heat gain of the liquid decreased. The temperature difference between
Electronic instruments inevitably experience input power change, leading to inconsistent heat generation. However, variable input power for LHP operation produces different distributions of working fluid in the loop, some of which may lead to depriming. Continuous operation was tested under stepwise and random heat load cycles aimed at validating the reliability of the LHP, as shown in Fig. 6. Firstly, the heat load was varied with a step of 10 W or 20 W, increasing from 20 W to 110 W, and then decreasing from 110 W to 10 W. During the entire testing procedure, heat load changes were implemented every 30 min with steady-state heat load between changes. While the heat load increased from 20 W to 110 W, the LHP could quickly switch from the previous stable operation to the next, and no operation failure was observed. While the heat load decreased from 60 W to 40 W, the temperature of the evaporator wall dropped excessively before eventually stabilizing. However, similar behavior was not observed for the same heat load decrement from 110 W to 90 W. The change amplitude of heat load applied to the evaporator had an impact on the operating behavior of the LHP, depending to an extent on the redistribution of the working fluid inside the loop. For small heat loads, a large head load change easily presented unstable phenomena. For the evaporator with a combination of the primary wick and
Fig. 3. System diagram of the LHP and the locations of T-type thermocouples.
H. Song et al. / International Journal of Heat and Mass Transfer 102 (2016) 1050–1055
Fig. 4. Startup with 10 W at 5 °C.
Fig. 5. Startup with 100 W at 5 °C.
the secondary wick, the loop had the ability to self-adjust, avoiding depriming. 3.3. Analysis of thermal characteristics 3.3.1. Temperature distribution The temperature distribution showed a typical trend during LHP operation, as shown in Fig. 7. There also existed two operation modes at the demarcation point of 50 W: variable conduction mode and constant conduction mode. During the test, the external loop was insulated to reduce heat-exchange with the environment. However, the liquid exiting the condenser gained a large amount of heat from the environment because of the low flow rate of the liquid in the liquid transport line. When heat load to the evaporator active zone was increased from 20 W to 50 W, the CC gradually got filled with liquid and the temperature difference between the evaporator entrance and condenser outlet diminished with the increase in heat load. The CC temperature gradually decreased and possessed the lowest temperature at 50 W, the loop operated at variable conduction mode. With heat load was increased from 50 W to 110 W, the operating temperature linearly increased and the condenser was fully utilized. The loop operated at constant conduction mode.
1053
Fig. 6. Continuous operation with stepwise heat load changes at 0 °C.
Fig. 7. Temperature distribution of the LHP at 0 oC.
From the temperature data of the evaporator exit and condenser inlet, the vapor experienced negligible heat-exchange along the vapor line. The temperature of the evaporator back (close to the CC) continuously decreased as heat load increased from 20 W to 50 W before stabilizing at 20 °C for heat loads from 50 W to 100 W. This implied that heat leakage to the CC was completely compensated for by the subcooled liquid, neglecting heatexchange with the environment. 3.3.2. Evaporator performance Evaporator thermal resistance is usually regarded as the index of evaporator heat-transfer capacity. The evaporator thermal resistance, Rev ap , is a comprehensive result of heat conduction through the evaporator wall and heat transfer from the inner surface of the evaporator to the working fluid. Conduction through the evaporator wall is mainly affected by the thickness of the evaporator wall and its material, whereas heat transfer from the evaporator to the working fluid depends on many factors, including the quality of the thermal contact of the wick with the evaporator wall, the thermophysical parameters of the wick, and the structure of the vapor-removal channels [7]. The evaporator thermal resistance can be calculated using
1054
H. Song et al. / International Journal of Heat and Mass Transfer 102 (2016) 1050–1055
Rev ap ¼
Te Tv : Q app
ð1Þ
The heat-transfer coefficient K e can be defined as
Ke ¼
Q app : AðT e T v Þ
ð2Þ
For conventional flat evaporators, the primary wick is directly located on the flange of the evaporator sidewall, which cannot guarantee the quality of the thermal contact of the inner surface of the evaporator with the primary wick. In order to address the problem, the primary wick may be tightly attached to the inner surface using the elasticity of the secondary wick. Fig. 8 gives the relations for thermal resistance and heat-transfer coefficient against heat load, the experimental results showed that the evaporator thermal resistance varied between 0.256 °C/W and 0.369 °C/W; the minimum value of 0.256 °C/W was achieved at 60 W. From Fig. 8, the evaporator thermal resistance gradually decreased with heat load increasing from 20 W to 60 W, and then increased until the applied heat load reached 110 W. Simultaneously, the heat-transfer coefficient, K e , varied from 3470 W/ (m2K) to 5430 W/(m2K), and the maximum heat-transfer coefficient was achieved at 60 W. The heat-transfer coefficient first increased and then decreased with the increase in heat load, whose trend was similar to that in Ref. [22,23]. For evaporative heat transfer in a capillary structure, the vapor–liquid distribution in the porous media varied with heat load change, and yielded different heat transfer mechanisms. While the heat load increased from 20 W to 60 W, the evaporation occurring on the wick structure surface intensified. More bubbles generated on the nucleation sites of the wick, and fled from the surface. Therefore, heat transfer on the capillary wick surface was enhanced. Vapor film formed on the active zone of the wick beyond 60 W, and increased thermal resistance from the inner surface of the evaporator to the wick, leading to a decrease in heat-transfer coefficient. 3.3.3. Thermal resistance of the LHP The LHP thermal resistance, RLHP , from the evaporator to the condenser represents the heat-transfer capacity of the LHP system:
RLHP ¼
T e T cond : Q app
ð3Þ
From the definition of the LHP thermal resistance, it is related to the evaporator heat-transfer capacity and the condenser cooling efficiency. Therefore, the thermal resistance of the LHP can be reduced by improving the evaporator structure and the condenser cooling efficiency. The overall thermal resistance, RLHP , ranged from 0.337 °C/W to 1.152 °C/W; the minimum value appeared at a heat load of 100 W, as depicted in Fig. 8. The evaporator thermal resistance occupied a small share of the LHP thermal resistance at heat loads less than 50 W, as the condenser cooling efficiency was relatively low. For heat loads over 50 W, the condenser thermal resistance was responsible for the LHP thermal resistance. For heat loads exceeding 50 W, the temperature difference between the condenser inlet and outlet became large while the condenser outlet temperature remained nearly unchanged.
3.3.4. Comparison with other studies Loop heat pipes with the flat evaporator have attracted wide attention from all over the world, because they have robust operation and a flat thermocontact surface. Thus, the comparison has been majorly made with loop heat pipes with the flat evaporator. Singh et al. [24] proposed a copper-nickel-water mLHP in which the CC was positioned on the side of the wick. In the horizontal orientation, the minimum LHP thermal resistance achieved at 50 W was 0.62 °C/W. Celata et al. [25] investigated a stainless steel LHP with flat evaporator, with water as the working fluid. The maximum transferred heat load was 75 W, and its LHP thermal resistance varied between 3.33 °C/W and 50.7 °C/W at horizontal elevation. Chen et al. [19] tested stainless-steel–ammonia loop heat pipes (LHPs) with flat evaporator, their total thermal resistance ranged from 0.33 °C/W to 1.47 °C/W. This research mainly studied the application of the combination of a primary wick and a secondary wick to a loop heat pipe with ammonia as the working fluid. With the evaporator wall temperature below 60 °C, the desired thermal performance was achieved, and the LHP thermal resistance from the evaporator active zone to the condenser distributed between 0.337 °C/W and 1.152 °C/W.
4. Conclusions In the present work, the combination of a primary wick and a secondary wick was applied to a loop heat pipe with the flat plate evaporator, and a series of tests were designed to examine its operating performance. The following conclusions could be drawn from the experimental tests. (1) The LHP could achieve a fast startup process at low heat loads and transfer 110 W (corresponding heat flux of 10.8 W/cm2) with the evaporator wall below 60 °C in the horizontal orientation. (2) The evaporator integrated with a combination of the primary wick and the secondary wick had the ability to selfadjust under the operation with heat load change. (3) Transition point of operation modes (50 W) and the evaporator thermal resistance (60 W) did not correspond to the same heat load. (4) The thermal resistance of the LHP was mainly affected by the condenser thermal resistance at low heat loads and limited by the evaporator capacity at high heat loads.
Conflict of interest Fig. 8. Thermal characteristics dependence of heat load at 0 °C.
None declared.
H. Song et al. / International Journal of Heat and Mass Transfer 102 (2016) 1050–1055
Acknowledgement The current work was supported by the National Natural Science Foundation of China (Grant Nos. 51276071 and 51376069).
References [1] Vladimir.G. Pastukhov, Yuri.F. Maydanik, Combined LHP and PHP based heattransfer system, Int. J. Therm. Sci. 74 (2013) 81–85. [2] Yury.F. Maydanik, Valery.I. Dmitrin, Vladimir.G. Pastukhov, Compact cooler for electronics on the basis of a pulsating heat pipe, Appl. Therm. Eng. 29 (2009) 3511–3517. [3] Benjamin Siedel, Valérie Sartre, Frédéric Lefèvre, Literature review: Steadystate modeling of loop heat pipes, Appl. Therm. Eng. 75 (2015) 709–723. [4] Mariya A. Chernysheva, Vladimir.G. Pastukhov, Yury.F. Maydanik, Analysis of heat exchange in the compensation chamber of a loop heat pipe, Energy 55 (2013) 253–262. [5] Yu.F. Maydanik, Review: Loop heat pipes, Appl. Therm. Eng. 25 (2005) 635–657. [6] Yury F. Maydanik, Mariya A. Chernysheva, Vladimir.G. Pastukhov, Review: loop heat pipes with flat evaporators, Appl. Therm. Eng. 67 (2014) 294–307. [7] Yury F. Maydanik, Sergey V. Vershinin, Development and tests of ammonia miniature loop heat pipes with cylindrical evaporators, Appl. Therm. Eng. 29 (2009) 2297–2301. [8] Zhichun Liu, Dongdong Wang, Chi Jiang, Jinguo Yang, Wei Liu, Experimental study on loop heat pipe with two-wick flat evaporator, Int. J. Therm. Sci. 94 (2015) 9–17. [9] Zhichun Liu, Dongxing Gai, Huan Li, Wei Liu, Jinguo Yang, Mengmeng Liu, Investigation of impact of different working fluids on the operational characteristics of miniature LHP with flat evaporator, Appl. Therm. Eng. 31 (2011) 3387–3392. [10] Zhi Chun Liu, Huan Li, Bin Bin Chen, Jin Guo Yang, Wei Liu, Operational characteristics of flat type loop heat pipe with biporous wick, Int. J. Therm. Sci. 58 (2012) 180–185. [11] Dongdong Wang, Zhichun Liu, Song He, Jinguo Yang, Wei Liu, Operational characteristics of a loop heat pipe with a flat evaporator and two primary biporous wicks, Int. J. Heat Mass Transfer 89 (2015) 33–41.
1055
[12] Song He, Zhi Chun Liu, Dong Dong Wang, Jing Zhao, Wei Liu, Jinguo Yang, Investigation of the flat disk-shaped LHP with a shared compensation chamber, Appl. Therm. Eng. 104 (2016) 139–145. [13] Dongdong Wang, Zhichun Liu, Jun Shen, Chi Jiang, Binbin Chen, Jinguo Yang, Tu Zhengkai, Wei Liu, Experimental study of the loop heat pipe with a flat diskshaped evaporator, Exp. Therm. Fluid Sci. 57 (2014) 157–164. [14] X.L. Cao, P. Cheng, T.S. Zhao, Experimental study of evaporative heat transfer in sintered copper bidispersed wick structures, J. Thermophys. Heat Transfer 16 (4) (2002). [15] Wei Zhou, Weisong Ling, Lian Duan, K.S. Hui, K.N. Hui, Development and tests of loop heat pipe with multi-layer metal foams as wick structure, Appl. Therm. Eng. 94 (2016) 324–330. [16] Wu. Shen-Chun, Dawn. Wang, Yau.-Ming. Chen, Investigating the effect of double-layer wick thickness ratio on heat transfer performance of loop heat pipe, Int. J. Therm. Sci. 86 (2014) 292–298. [17] Randeep Singh, Aliakbar Akbarzadeh, Masataka Mochizuki, Effect of wick characteristics on the thermal performance of the miniature loop heat pipe, J. Heat Transfer 131 (2009). 082601-082601-10. [18] Xu Jinliang, Xianbing Ji, Wolong Yang, Ziwei Zhao, Modulated porous wick evaporator for loop heat pipes: experiment, Int. J. Heat Mass Transfer 72 (2014) 163–176. [19] B.B. Chen, W. Liu, Z.C. Liu, H. Li, J.G. Yang, Experimental investigation of loop heat pipe with flat evaporator using biporous wick, Appl. Therm. Eng. 42 (2012) 34–40. [20] Huan Li, Zhi Chun Liu, Bin Bin Chen, Wei Liu, Chen Li, Jinguo Yang, Development of biporous wicks for flat-plate loop heat pipe, Exp. Therm. Fluid Sci. 37 (2012) 91–97. [21] Jentung Ku, Operating characteristics of loop heat pipes, 29th International Conference on Environmental Systems, Denver, Colorado, July 12–15, 1999. [22] Q. Liao, T.S. Zhao, Evaporative heat transfer in a capillary structure heated by a grooved block, J. Thermophys. Heat Transfer 13 (1999) 126–133. [23] F.J. Hong, P. Cheng, H.Y. Wu, Z. Sun, Evaporation/boiling heat transfer on capillary feed copper particle sintered porous wick at reduced pressure, Int. J. Heat Mass Transfer 63 (2013) 389–400. [24] Randeep Singh, Aliakbar Akbarzadeh, Chris Dixon, Masataka Mochizuki, Novel design of a miniature loop heat pipe evaporator for electronic cooling, J. Heat Transfer 129 (2007) 1445–1452. [25] Gian Piero Celata, Maurizio Cumo, Massimo Furrer, Experimental tests of a stainless steel loop heat pipe with flat evaporator, Exp. Therm. Fluid Sci. 34 (2010) 866–878.