Passive cooling of telecommunication outdoor cabinets for mobile base station Stephane Le Masson (1), Hasna Louahlia-Gualous (2), Ali Ahmad Chehade (2) (1) France Télécom, Orange Labs, 2 av Pierre Marzin, 22300 Lannion (2) Université de Caen Basse Normandie, LUSAC, 120 rue de l Exode, 50000 Saint Lô Abstract This paper reports a passive cooling of an outdoor telecommunication cabinet using a loop thermosyphon. The system performance is studied using different working fluids (n-pentane, HFC-365mfc, and SES-36). The experimental tests are conducted for different fill charges of the working fluids and various heat loads. SES-36 working fluid is retained as the best working fluid depending on its zero flammability, low boiling temperature, and thermal performance. 1 Introduction The high-speed telecommunication expansion and the requirement for high bit rate services both result a high increase in the generated heat inside the outdoor telecommunication cabinets. However, the thermal management of these components and their enclosures has become a worldwide critical issue since the electrical components reliability and efficiency depend on the materials temperature. Therefore, there has been extensive research in the field of thermal management for the cooling of such telecommunication cabinets [1-5]. Additionally, the greenhouse effect accompanied with climate change is a worldwide problem that is affecting people and environment. Nevertheless, the increase of the greenhouse gases emissions resulting from human activities would increase the heat trapped on the Earth, which in turn negatively affect the human life. Recently, the global association of mobile operators (GSMA) has published a report estimating the energy consumption and greenhouse gas emissions resulting from the global power consumption of mobile networks [6]. Table 1 shows the global consumption and energy cost for mobile communications. According to the "Mobile Green Manifesto" published by the GSMA in 2012 [7], the goal is to reduce the global greenhouse gas emissions per connection by 2020 in mobile communications industry by 40%. Subsequently, using cooling system with a low charge of the working fluid, high efficiency, and environmentally friendly. It is therefore necessary to worry about the energy consumption of telecommunication systems. Developing appropriate ways to contribute to energy savings and reduced CO 2 emissions must be taken into consideration, where it has been difficult for operators to find innovative solutions for such issues. This is why the OPERANET2 (Optimizing Power Efficiency in Mobile Radio Networks 2) project brings together several European partners in the conscious of improving the energy efficiency of telecommunication equipment. Under this project, a study concerning passive cooling telecommunication equipment is conducted. A loop thermosyphon is developed for cooling mobile telecommunication networks cabinets respecting the ETSI (European Telecommunication Standard Institute) [8], which necessitates that the air temperature at the inlet of the active materials should be between 5 C and 55 C. Region Energy Energetic cost Emission in consumed in in kg CO2/ kwh/connection $/connection connection Africa 24.4 2.2 11.1 South & Central America 13.3 1.9 4.6 Asia Pacific 29.3 3 17.6 East Europe 12 1.3 7 West Europe 17.1 2.6 7.6 Middle East 19.7 1.6 13.5 USA & Canada 38 3.2 25.3 Total 24.1 2.6 13.7 Table 1 Global estimation of consumption and energy costs and CO2 emissions [2].
This paper discusses the effects of working fluids and optimal fill charge on a passive cooling loop for the telecommunication cabinets containing active materials. Tests are conducted using different working fluids considering the environmental conditions. The thermal performance of the thermosyphon loop is investigated respecting the operating temperature of the telecommunication equipment. 2 Experimental setup A closed loop thermosyphon is an energy-transfer device capable of transferring heat from a heat source to a separate heat sink over a relatively long distance. Additionally to the usage of small amounts of working fluids, passive loop thermosyphon operates without active control instrumentation and any mechanical moving parts. These devices are thus particularly suitable for cooling applications where reliability and safety are of paramount importance without energy consumption. The working fluid plays a vital part in the cooling system, since it is the medium by which cooling or heating is carried out. The ideal working fluid therefore has a number of properties; a boiling point below the objective temperature, a high vaporization point, moderate liquid density, relatively high vapor density, high critical temperature, and low operating pressure. 9 7 8 5 6 1 4 3 2! Figure 1 Photo of experimental setup; (1) voltage regulator, (2) power meter, (3) vacuum pump, (4) heating equipment, (5) outdoor cabinet, (6) thermosyphon system, (7) thermocouples, (8) NI acquisition device and (9) computer. An experimental setup is built to conduct measurements for heat transfer and cooling enhancement while using thermosyphon loop at an outdoor telecommunication cabinet. It consists of an outdoor telecommunication cabinet, loop thermosyphon, heating equipment, vacuum pump, voltage transformer and power meter, pressure sensors and thermocouples, computer, and an
acquisition data device. The telecommunication outdoor cabinet is of 760 mm height, 600 mm width, and 326 mm depth. Four heating cards are placed in the heating equipment that is located at the center of the telecommunication cabinet. Heat flux generated there heats the telecommunication cabinet indoor air. The loop thermosyphon is installed at the outdoor cabinet as shown by Figure 1. It consists of an evaporator connected to a condenser by the means of two copper tubes. The loop s evaporator is installed inside the cabinet unlike its condenser, which is located outside. Heat transfer from telecommunication equipment could be dissipated by vaporization of the working fluid inside the evaporator, where the vapor would have the sufficient pressure to overcome the loss due to the vapor line distance to the condenser. The arriving vapor at the condenser changes into liquid flow due to the convection cooling. After that, condensate flow leaving the condenser moves under gravity along the liquid line. The vacuum pump is used to degas all the air and gases inside the loop before any test. The power supply is adjusted through a voltage transformer having an output voltage range from 0 to 220 V, and a power meter. Likewise, pressure sensors and 75µm diameter K type thermocouples are mounted on the entire system for the temperature and pressure measurements. Ambient air temperature is also measured during each experiment. National instruments data acquisition system is used to record all temperatures and pressures for each experimental test. Data sampling frequency is of one measure per second for all measurements. The acquisition of the data is entirely automated by using the Lab View software and of the National Instruments devices which facilitate measurements in real-time. 2.1 Experimental procedures Before starting the tests, the vacuum pumping and liquid preheating processes were performed to remove the gases dissolved in the loop thermosyphon and eliminate the influence of non-condensable gases. The fluid fill ratio, which is defined as the ratio between the working fluid volume and the loop thermosyphon volume (eq. 1), is also conducted by several tests using several fill ratios. The fill ratio is that delivering maximum thermal performance. Fill Ratio =!!"#$%&'!"#$%!!""# (1) During the experiments, various values of the heat loads ranging from 250 to 400W are experienced. For each test, the total power supplied to the heating equipment is set at the desired value. Heating process of the working fluid accompanied with temperature recordings begins. When all the temperature measurements at that power heat load reach steady state, the value of power supply is incremented by 50W. 3 Experimental results 3.1 Determination of the optimal charge The existence of an optimal fill charge is due to two physical phenomena that must be avoided to ensure an effective operation of the loop. The heat transfer in the evaporator is reduced due to the increase of thermal resistance of the working fluid in the liquid chamber, when the amount of fluid increases in the liquid chamber. On the other hand, drying in the liquid chamber can be triggered due to insufficient amount of liquid coolant in the liquid chamber, which also causes a decrease in heat transfer. The optimal fill charge is conducted using n-pentane as a working fluid. Several fill charges of n-pentane are tested in the thermosyphon loop (10g, 15g, 20g, 25g, and 30g corresponding to 7.4%, 11.15%, 14.9%, 18.6%, and 22.3% fill ratios respectively). For each fill charge, the heating equipment power is varied from 250W to 400W by an increment of 50W after reaching the steady state. Figure 2 presents the systems thermal resistance and response time as function of the fill charge for 350W heat loads. System thermal resistance is used as an indicator for determining the cooling thermal efficiency of the cooling system under various heat loads. R!!,!"! =!!"!!!"#!!"#$ (2) where R th,sys is the system thermal resistance, T op and T amb are the operating and ambient temperatures respectively, and Q load is the heat load at the heating equipment. Figure 2 shows that the system thermal resistance is strongly affected by the fill charge, where it is minimal at the optimal charge (20g of n-pentane). It tends to increase when the fill charge is more or less than 20g. The system thermal resistances and the response time are affected by the fill charge and have the same tendency. They attain the minimum value at optimal charge and increase as the fill charge is modified. It can be concluded from these results that the effect of the fill charge is obvious. Two regions can be identified; where the thermal performance of the thermosyphon loop is deteriorated because of the increase of the system thermal resistance causing the increase of the operating temperature and the loop response time.
Thermal resistance (K/W) 0,12 0,11 0,1 0,09 0,08 0,07 Thermal resistance Response time 0,06 0 5 10 15 20 25 30 35 Fill charge (g) Figure 2 System thermal resistance and response time for different charges for 350W heat load. 3.2 Impact of the working fluid on the loop performance The most important properties of the working fluid are non-flammability, low toxicity, suitable thermophysical properties, economic availability, zero ODP, and a low GWP. The performance of the thermosyphon loop is examined using three fluids: n- pentane, HFC-365mfc, and SES-36. Table 2 shows the environmental and thermal characteristics of each fluid. Concerning the environmental impact; all the tested fluids have a zero ODP. SES-36 has the highest GWP followed by HFC-365mfc, and n- pentane has a low GWP. Unlike SES-36, n-pentane and HFC-365mfc are flammable. Concerning the thermal part, SES-36 has the lowest boiling point and highest density. Moreover, n-pentane has the highest latent heat of vaporization. Characteristics n-pentane HFC-365mfc SES-36 ODP 0 0 0 GWP 11 840 3126 Flammability High Low 0 Boiling Point ( C) 36 40 35.64 Latent heat (kj/kg) 366 198.8 152.49 Liquid density (kg/m 3 ) 620.78 1257 1365.6 Vapor density (kg/m 3 ) 2.0583 3,4964 4.96 Liquid specific capacity (kj/kgk) 2.3154 1.3768 1.1672 Vapor specific capacity (kj/kgk) 1.7043 1.0058 0.6419 Table 2 Properties of the three working fluids at 25 C. 1800 1600 1400 1200 1000 800 600 400 200 Response time (s) Figure 3 shows the operating temperature of the telecommunication equipment while subtracting the ambient temperature during the steady state for different heat loads. The operating temperature is minimal using n-pentane which has a high latent heat and specific heat capacity. On the other hand, the operating temperature obtained using SES-36 and HFC-365mfc is increased by 2 C than that obtained with n-pentane. Depending on the zero flammability of SES-36 and its thermal performance, it can be selected as the best working fluid. Difference between operating and ambient temperatures [ C] 45 40 35 30 25 20 15 n-pentane HFC-365mfc SES-36 10 0 100 200 300 400 500 600 Heat load [W] Figure 3 Operating minus ambient temperature for different working fluids. 3.3 Performance of the system using SES- 36 in the thermosyphon loop Figure 4 shows the operating temperatures of the telecommunication outdoor cabinet for different heat loads. Tests are conducted with and without the thermosyphon loop inside the telecommunication cabinet. Ambient temperature is also measured during tests. SES-36 is used in the thermosyphon loop as the working fluid. The maximum heat load is defined as the heat load value below which the operating temperature becomes higher than the ETSI temperature (55 C). Using only the classical cooling system, the maximum heat load is about 250W. Cooling with the thermosyphon loop implies that the heat load of the telecommunication equipments must be lower than 500W; value for which the operating temperature is equal to the limit value fixed by ETSI. This result confirms that this cooling system is very attractive at practical viewpoint.
Difference between operating and ambient temperatures [ C] 50 45 40 35 30 25 20 15 10 Figure 4 Comparison between the equipment operating temperatures for cabinet with and without thermosyphon loop versus heat load 4 Conclusion: cooling with thermosyphon loop without thermosyphon loop 0 100 200 300 400 500 600 Heat load [W] This article presents experimental results of passive cooling of outdoor telecommunication cabinets using a loop thermosyphon. Three fluids are tested in the system; n-pentane, HFC-365mfc, and SES-36. A thermosyphon loop was studied and tested under different heat loads. It showed that the passive cooling loops using SES-36 as the working fluid in phase change are the useful systems for cooling the telecommunication equipments. SES-36 is retained as the best working fluid depending on the zero flammability, low boiling temperature, and its thermal performance. The telecommunication outdoor cabinet of France Telecom fitted with thermosyphon can operate under heat load lower than 500W by respecting the ETSI norm. 6 REFERENCES [1] A. Samba, H.Louahlia Gualous, S. Le Masson, D. Nörtershäuser, Two-phase thermosyphon loop for cooling outdoor telecommunication equipments perimental, Applied Thermal Engineering, In Press, Accepted Manuscript, Available online 23 May 2012. [2] M.J. Marongiu, Issues in the Thermal management of Outdoor Telecommunications Cabinets/Enclosures, in Proc. Telecommunication Energy Special Conference, 1997, pp.379-386. [3] S. Le Masson, D. Nörtershäuser, D. Mondieig, H. Louahlia-Gualous, Towards passive cooling solutions for mobile access network, Journal of Annals of telecommunications, pp: 125-132, Numbers 3-4, Vol.67, 2012. [4] H. Louahlia-Gualous, B. Mecheri, D. Nortershauser, S. Le Masson, Transient characteristics of a two phase thermosyphon loop for cooling telecommunication cabinets, Proceeding of the 14th International Heat Transfer Conference IHTC14 ASME, Washington D.C., USA, 2010. [5] R.R. Schmidt, H. Shaukatullah, Computer and telecommunications equipmentroomcooling; a r eview of literature, IEEE Transactions on Components and Packaging Technologies, 26 (1) (2003), pp. 89 98. [6] Télécommunications et consommation en énergie, Strategies, Telecoms & Multimedia, http://www.strategiestm.com. [7] Mobile s green manifesto, GSM Association, 2012, www.gsma.com/mee. [8] ETSI EN 300 019-1-3, Environnemental conditions and environnemental tests for telecommunications equipment; Part 1-3: European Telecommunication Standard Institute. 5 ACKNOWLEDGEMENT The authors of this article would like to thank the French Ministry of Industry and Commerce (DGCIS) for the funding of this work, which is integrated in the European project OPERANET-2 labelled by Celtic-Plus.