Thermal performance of an open loop closed end pulsating heat pipe ...

Heat Mass Transfer (2012) 48:259–265 DOI 10.1007/s00231-011-0882-9

ORIGINAL

Thermal performance of an open loop closed end pulsating heat pipe Manabendra Saha • C. M. Feroz • F. Ahmed T. Mujib

•

Received: 30 November 2010 / Accepted: 31 July 2011 / Published online: 10 August 2011 Springer-Verlag 2011

Abstract This paper presents an experimental study of an open loop pulsating heat pipe (OLPHP) of 0.9 mm inner diameter. The performance characterization has been done using four working fluids at vertical and horizontal orientations. Water, Methanol, 2-Propanol and Acetone has been employed as the working fluid with 50% fill ratio. The experimental results indicate a strong influence of gravity and thermo physical properties of the working fluids on the performance of OLPHP. Considering all the working fluids used, Water has shown better thermal performance in vertical orientation while Methanol has shown better performance in horizontal orientation. All the working fluids perform better at horizontal orientation. List of symbols Eo¨ Eo¨tvo¨s number (mm2) Bo Bond number (mm) COP Coefficient of performance (mm) Q Heat input (W) R Thermal resistance (C/W) U Overall heat transfer coefficient (kW/m2C)

1 Introduction Thermal management of any system is one of the most significant parameter as it plays an influential role in device functionality, device safety and above all device failure. So the design of an effective cooling system is of utmost importance to enhance the system’s performance and M. Saha (&) C. M. Feroz F. Ahmed T. Mujib Department of Mechanical Engineering, BUET, Dhaka 1000, Bangladesh e-mail: [email protected]; [email protected]

reliability. Among other cooling techniques heat pipes have emerged as the most appropriate technology and cost effective thermal design solution due to its excellent heat transfer capability, high efficiency and structural simplicity. Although the conventional heat pipes (e.g. mini or micro) are one of the proven technologies, there are various limitations of the performance of heat pipes, e.g. capillary limit, boiling limit, entrainment limit, sonic limit, viscous limit. In both cases the limit will manifest itself by an unacceptable overheating of the evaporator due to lack of cooling working fluid (dry-out, burn-out) [1]. To overcome these difficulties researchers have come up with the idea of pulsating heat pipes (PHPs) which work on principle of oscillation of the working fluid and phase change phenomenon in a capillary tube. PHPs are passive two-phase thermal control devices first introduced by Akachi et al. [2]. In this application, reduced diameter channels are used, which are directly influenced by the selected working fluid. The vapor plugs generated by the evaporation of the working fluid push the liquid slugs toward the condensation section and this motion causes flow oscillations that guide the device operation [3]. One end of the PHP tube bundle receives heat, transferring it to the other by a pulsating action of the working fluids. Since each tube section between the evaporator and the condenser has a different volumetric distribution of the working fluid, the pressure drop associated with each sub-section is different. This causes pressure imbalances leading to thermally driven two-phase flow instabilities eventually responsible for the thermo fluidic transport without requiring any external mechanical power source [4]. A PHP must be heated in at least one section and cooled in another. Often the evaporators and condensers are located at the bends of the capillary tube. The tube is evacuated and then partially filled with a working fluid. The liquid and its vapor will become distributed throughout

123

260

the pipe as liquid slugs and vapor bubbles. As the evaporator section of the PHP is heated, the vapor pressure of the bubbles located in that section will increase. This forces the liquid slug toward the condenser section of the heat pipe. When the vapor bubbles reach the condenser, it will begin to condense. As the vapor changes phase, the vapor pressure decreases, and the liquid flows back toward the condenser end. In this way, a steady oscillating flow is set up in the PHP. Boiling the working fluid will also cause new vapor bubbles to form [5]. There are two possible configurations for PHPs, being as an open loop and as a closed loop. In the open loop configuration, one end of the tube is pinched off and welded, while the other end may present a service valve for evacuation and charging. The closed loop configuration has both ends connected and the fluid is allowed to circulate. On the open loop configuration, the liquid circulation is not possible. In this case, it is believed that a counter-current liquid/vapor flow occurs in order to promote the proper device operation [6]. Therefore, better understanding on this behavior is still a motivation for further investigations. Therefore, the purposes of this investigation are to study the heat transfer characteristics of an OLPHP and to find out the proper working fluids and other parameters for obtaining the optimum performance from an OLPHP by evaluating several issues related to its performance.

2 Experimental setup The device used during the experimental tests is shown in Fig. 1. A pulsating heat pipe configured as an open loop is built using a capillary copper tubing of closed ends with 2.30 mm OD, 0.90 mm ID, and 4.9 m long to form 13 parallel channels with 12 curves. The apparatus is placed

Heat Mass Transfer (2012) 48:259–265

on a support that allowed its adjustment on vertical and horizontal orientation. The internal diameter is a parameter which necessarily affects the definition of a pulsating heat pipe [5]. A theoretical maximum tolerable inner diameter, Dmax, of a PHP capillary tube is derived based on the balance of capillary and gravity forces by Akachi et al. [2]. sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi r Dmax ¼ 2 ð1Þ g qliq qvap If di \ Dmax surface tension forces tend to dominate and stable liquid slugs are formed. As the PHP tube diameter increases beyond the Dmax, the surface tension is reduced and all the working fluid will tend to stratify by gravity and the heat pipe will stop functioning as a PHP. For a given specified heat power, decreasing the diameter will increase the dissipative losses and lead to poor performance. Increasing the diameter much above the critical diameter will change the phenomenological operation of the device. It will no more act as a pulsating heat pipe. The critical Bond number (or Eo¨tvo¨s) criterion [7] gives the tentative design rule for the diameter. A high Bond number indicates that the system is relatively unaffected by surface tension effects; a low number (typically less than one is the requirement) indicates that surface tension dominates. Intermediate numbers indicate a non-trivial balance between the two effects. The Bond number is the most common comparison of gravity and surface tension effects. The Bond number (or alternatively, Eo¨) is the ratio of surface tension and gravity forces and defined as follows, D2crit g qliq qvap 2 P ðE€oÞcrit ¼ ðBoÞcrit 4 ð2Þ If the Bond number exceeds a particular critically value (Bocrit & 2) stable liquid slugs will not form and the device will not function as a pulsating heat pipe. For Bo [ 2, the heat transfer limitation comes from nucleate pool boiling and counter-current flow limitations. The pulsating heat pipe of 0.9 mm inner diameter has exhibited perfect harmonization with the stability criteria as the Bond Numbers of the OLPHP for all the working fluids used correspond to the critical value of 2. Experimental arrangements for the performance test of PHP are summarized in Table 1.

3 Test procedure

Fig. 1 Cross section of the OLPHP test apparatus

123

The OLPHP is built with 3 regions: evaporation, adiabatic and condensation, each one is 300 mm wide. The evaporator, adiabatic and condenser sections are 105, 100 and 95 mm long respectively with filling ratio of 50%. For the tests under


261

vertical orientation, the evaporation section is always below the condensation section. For the tests under horizontal orientation, all sections are at the same plane. Both evaporation and adiabatic sections are thermally insulated, while the condensation section is open to the surrounding air. The condensation section is cooled by forced air, using a fan able Table 1 Experimental parameters and their ranges Parameters

Condition

Outside diameter of pipe, do (mm)

2.3

Inside diameter of pipe, di (mm)

0.9

Length of total, L (mm)

300

Length of evaporator, Le (mm)

105

Length of adiabatic section, La (mm)

100

Length of the condenser, Lc (mm)

95

Working fluid Orientation

Water, methanol, 2-propanol, acetone Vertical and horizontal

Heat input range (W)

5–50

Fig. 2 Temperature profile along the length of OLPHP (a vertical, b horizontal) for methanol

to deliver air at a velocity of 4 m/s. The evaporation section is in contact with Ni–Cr wire which is coiled around two mica sheets to deliver the desired heat load. The heat loads are administered by an AC power supply. Eighteen K-type thermocouples are used to monitor the device performance during the tests. A total of 6 thermocouples are located at each section. The thermocouples are connected to a digital temperature controller (0–800C) to monitor wall temperatures. The temperatures are recorded after every 5 min. All experimental tests are performed on a heat load profile basis, where the perspective is to evaluate the behavior of OLPHP related to the orientation and applied power.

4 Results and discussion 4.1 Temperature profile The temperature profiles along the length of OLPHP at vertical and horizontal orientations for different working

Fig. 3 Temperature profile along the length of OLPHP (a vertical, b horizontal) for 2-propanol

123

262


fluids have been exhibited from Figs. 2, 3, 4 and 5. For Methanol at vertical orientation as shown in Fig. 2a it is observed that the intensive oscillations have begun from 20 W. Temperatures of the adiabatic and the condenser sections are similar at low heat loads. Again, at horizontal orientation as shown in Fig. 2b the evaporation section temperature is higher showing poor performance. Intensive oscillations begin from 35 W. When the heat load reaches its maximum value of 50 W, the temperature of the evaporation section is about 95C. The increase in temperature is gradual. The vertical orientation is preferable for OLPHP using Methanol as working fluid as the temperature of the evaporation section is lower at higher heat loads. For 2-Propanol at vertical orientation as shown in Fig. 3a oscillation begins from low heat loads and becomes intensive after 25 W. Upon comparison with Methanol it is observed that at higher heat loads the evaporation section temperature of both working fluids is similar. At maximum heat load of 35 W, the evaporation section temperature is about 90C. At

horizontal orientation as shown in Fig. 3b the evaporation section temperature is higher at all heat loads as compared with the vertical orientation. Temperature distribution is more even as compared to vertical orientation. Intensive temperature oscillations occur at both low and high heat loads. At intermediate heat loads oscillations are absent. When testing the OLPHP with Acetone at vertical orientation as shown in Fig. 4a, it is observed that a more intensive plug/slug pumping action is initiated at 15 W. For this working fluid, OLPHP presents even temperature distribution along the evaporator, adiabatic and condenser sections. At horizontal orientation as shown in Fig. 4b Acetone exhibits more drastic pulsations and intensive oscillations start from 15 W. At horizontal orientation, the effects of surface tension and shear stress are very expressive because of capillary dimension of the channels. For Water at vertical orientation as shown in Fig. 5a oscillations begin from 20 W. At the highest heat load of 45 W the evaporation section temperature rises to 110C. So Water is more suitable at lower heat loads. At horizontal

Fig. 4 Temperature profile along the length of OLPHP (a vertical, b horizontal) for acetone

Fig. 5 Temperature profile along the length of OLPHP (a vertical, b horizontal) for Water

123


263

Figure 6a represents the comparison of thermal conductance of OLPHP using different working fluids at vertical

orientation. In case of 2-Propanol it is observed that from about 16 W onwards, thermal conductance increases gradually. In case of Methanol, overall thermal conductance decreases gradually with increasing heat load. For Water the thermal conductance curve doesn’t show any definite trend. For Acetone at vertical orientation the thermal conductance doesn’t change much with increasing heat load. At vertical orientation for all heat loads possesses the lowest thermal conductance when Acetone is used. When Water is used as working fluid the OLPHP exhibits the highest thermal conductance at higher heat loads. It can be concluded that Water is the best working fluid for vertical orientation on the basis of thermal conductance. The comparison of thermal conductance of OLPHP at horizontal orientation using different working fluids has been exhibited in Fig. 6b. Using 2-Propanol as the working fluid it is observed that the thermal conductance increases up to 20 W heat load, and then at higher heat loads it decreases gradually. In case of Methanol, the curve does

Fig. 6 Comparison of thermal conductance of OLPHP at a vertical, b horizontal orientation

Fig. 7 Comparison of COP of OLPHP at a vertical, b horizontal orientation

orientation as shown in Fig. 5b temperature distribution of Water is more uniform as compared to vertical orientation. From comparison of temperature profiles of all the working fluids at vertical orientation it is observed that Methanol and 2-Propanol shows similar performance at all heat loads. Acetone exhibits the higher evaporation temperature and Water has lower evaporation section temperature at all heat loads. Hence, Water is the better choice for use at vertical orientation. 4.2 Thermal conductance Thermal conductance of the heat pipe is calculated by using the evaporation and condensation section temperature. Gcalc ¼

Q ðTE TC Þ

ð3Þ

123

264


not show any particular trend. In case of Acetone thermal conductance gradually increases except at 30 W heat load where a sudden drop in thermal conductance occurs the reason of which is not clear yet. Again, the curve found by using Water as the working fluid does not show any definite trend but mostly the performance is better at horizontal orientation. In case of horizontal orientation, OLPHP shows the highest thermal conductance for Methanol at almost all heat loads except at 25 W. It can be concluded that Methanol is the best working fluid for horizontal orientation judged on the basis of thermal conductance.

Figure 7a and b indicate the variations of COP of OLPHP using different working fluids at vertical and horizontal orientations respectively. The trends observed from the curves of the COP (Fig. 7a, b) are similar to the trends of the thermal conductance (Fig. 6a, b). So, from the Fig. 7a it can be concluded that OLPHP has the higher COP at vertical orientation when Water is used as the working fluid because of the larger latent heat of vaporization of Water is larger than all the other working fluids being used. So, Water is better among all the working fluids for vertical orientation. On the other hand, Fig. 7b points out Methanol to be the better working fluid at horizontal orientation for OLPHP.

4.3 Coefficient of performance 4.4 Diameter effect The coefficient of performance (COP) is determined by the ratio of the thermal conductance of copper tube with working fluids (Gcalc) and empty copper tube (Gref). COP ¼

Gcalc Gref

ð4Þ

The findings of this experiment have been compared with the experimental results of Roger R. Riehl [8]. Roger R. Riehl conducted the similar experiment with copper tube of 1.5 mm ID. Figure 8 signifies the effect of diameter on the

Fig. 8 Diameter effect of OLPHP for a 2-propanol, b methanol, c water, d acetone as working fluid

123


performance of OLPHP for different working fluids. For 2-Propanol and Water, as shown in Fig. 8a and c respectively, the 0.9 mm inner diameter copper tube has operated successfully and shown good thermal performance for both the orientations. On the contrary, while using Methanol and Acetone, as shown in Fig. 8b and d respectively, the 0.9 mm ID tube shows good thermal performance only for vertical orientation. For horizontal orientation the 1.5 mm ID tube shows better thermal performance.

5 Conclusion •

•

•

Thermal conductances of OLPHP with Methanol and Water as working fluids are found to be superior at horizontal and vertical orientation respectively. So conclusions can be drawn that Water and Methanol should be given priority among all the working fluids being used for OLPHP at vertical and horizontal orientation respectively. OLPHP using Acetone shows drastic pulsations at all heat loads for both the orientations. The plug and slug pumping action occurs even at low heat load of 5–15 W. So, OLPHP using Acetone is found to be better for low heat load level. The findings of this experiment have been compared with the experimental results of Roger R. Riehl [8]. Roger R. Riehl conducted the similar experiment with

265

copper tube of 1.5 mm Inner diameter. The 0.9 mm inner diameter copper tube in this experiment operated successfully and showed good thermal performance compared with the copper tube of 1.5 mm inner diameter.

References 1. Yang H, Khandekar S, Groll M (2008) Operational limit of closed loop pulsating heat pipes. Appl Therm Eng 28:49–59 2. Akachi H, Pola´sˇek F, Sˇtulc P (1996) Pulsating heat pipes. Proceedings of the 5th international heat pipe symposium. Melbourne, Australia, pp 208–217 3. Zhang Y, Faghri A (2002) Heat transfer in a pulsating heat pipe with an open end. Int J Heat Mass Transf 45:755–764 4. Khandekar S, Groll M, Charoensawan P, Rittidech S, Terdtoon P (2004) Closed and open loop pulsating heat pipes. 13th International Heat Pipe Conference (13th IHPC), shanghai, China, pp 21–25 5. Karimi G, Culham JR, Review and assessment of pulsating heat pipe. Microelectronics Heat Transfer Laboratory, Department of Mechanical Engineering, University of Waterloo, Waterloo 6. Riehl RR (2003) Evaluation of the thermal-hydro-dynamics behavior of an open loop pulsating heat pipe. National Institute for Space Research (INPE) Report, p 35 7. Khandekar S, Groll M (2003) On the definition of pulsating heat pipes: an overview. Proceedings 5th minsk international seminar (heat pipes, heat pumps and refrigerators), Minsk, Belarus 8. Riehl RR Characteristics of an open loop pulsating heat pipe, SAE paper #2004-01-2509

123