Elsevier

Applied Thermal Engineering

Volume 199, 25 November 2021, 117533
Applied Thermal Engineering

Investigation of thermal characteristics of an LHP evaporator with heat sources having different heating surface areas

https://doi.org/10.1016/j.applthermaleng.2021.117533Get rights and content

Highlights

  • Thermal characteristics of the evaporator of copper–water LHP with different heat sources were investigated.

  • The LHP was meant for cooling compact high-power components of electronics.

  • Components were simulated by heat sources with heating surfaces of 9 cm2, 16 cm2 and 25 cm2.

  • It is shown that the evaporator thermal resistance and flux depend on the heating area.

  • The same evaporator temperature was achieved with different heat sources at different heat fluxes.

Abstract

The paper presents the results of investigation of thermal characteristics of a flat evaporator 10 mm thick with an active zone of 27 cm2 that is part of a copper–water loop heat pipe (LHP) intended for cooling systems of high-power electronic components. The investigation of the device was conducted in a horizontal position with heat sources having thermocontact surfaces of 9 cm2, 16 cm2 and 25 cm2 at heat load from 20 to 900 W. It is shown that the evaporator thermal resistance and the heat flux depend considerably on the value of the thermocontact surface of the heat sources. The minimum values of thermal resistance were, respectively, 0.033, 0.021 and 0.016 °C/W at heat fluxes of 100, 56.2 and 36 W/cm2. The temperature distribution over the evaporator surface with different heat sources has been obtained in analytical way.

Introduction

The problem of cooling high-power electronic components, for instance, such as central (CPU) and graphics (GPU) processors used in computer technology, remains quite topical. These components can dissipate up to 300 W, and in immediate prospect even up to 500 W from surfaces that are usually in the range from 1 to 30 cm2. Thus, one can speak about heat fluxes with values up to 50 W/cm2 and more. In this case, the temperature of the thermocontact (heating) surface of such components must be maintained, as a rule, at a level of no more than 80 to 100 °C, whereas the temperature of the cooling medium (air or liquid) may reach or even exceed 40 °C.

Cooling systems can make good use of loop heat pipes (LHPs) as a passive thermal link between the heat source and the heat sink. LHPs are two-phase heat-transfer devices operating on a closed evaporation–condensation cycle and using “a capillary mechanism” for pumping the working fluid [1]. An LHP is a closed loop that includes an evaporator with a capillary structure (wick) and a condenser communicating by means of separate vapor and liquid lines. The evaporator, which is in a thermal contact with the heat source, is the main LHP component. It simultaneously acts as an evaporative heat exchanger and a capillary pump. Evaporators may have both a cylindrical and a flat shape. It is believed that flat evaporators are better adapted to objects being cooled which have a flat thermocontact surface. In their turn, flat evaporators are subdivided into evaporators with a longitudinal replenishment (ELR) and those with an opposite replenishment with a working fluid (EOR) [2]. Since LHPs with flat evaporators are quite sensitive to an excess internal pressure, working fluids with operating pressures that do not exceed the ambient atmospheric one are most often used for them. Among such working fluids is first of all water, which in combination with a copper evaporator makes it possible to achieve the highest thermal characteristics [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17]. The advantage of LHPs with water is also the fact that in cooling systems they are capable of providing an acceptable operating temperature of objects being cooled at heat-sink temperatures of 40 °C and over. A limitation for water is the fact that it cannot be used in LHPs if out of service they undergo the action of a negative temperature. In such cases use is made of alternative working fluids, for instance, methanol [17], [18], [19], [20], [21], acetone [22] or some freons [23]. They are also easily compatible with copper, but require an increase in thickness or an additional strengthening of the evaporator walls, or else limitation of the upper values of the operating temperature owing to a high pressure. In this connection, such LHPs with the working fluids mentioned above operate with heat sinks whose temperature usually varies in a range that does not exceed +25 °C.

The main thermal parameters that characterize the efficiency of evaporators are the thermal resistance, which is determined by the formulaRe=Te-TvQand the heat flux at the heating surface of an evaporatorq=QFhsThe high efficiency of copper evaporators in LHPs with water is achieved owing to the high thermal conductivity of copper and water and also the extremely high values of the surface tension coefficient and the latent heat of evaporation of water. Thus, for instance, Singh et al. [4] developed and investigated an LHP with an evaporator having an active zone of 7.06 cm2. The heat source simulating a CPU had a thermocontact surface of 3.75 cm2. The maximum capacity of 70 W, corresponding to a heat flux of 18.7 W/cm2, was achieved at a temperature of the evaporator wall of 99.6 °C, which was measured next to the heat source. The thermal resistance determined by formula (1) was equal to 0.06 °C/W. The same LHP [5] with a heat source of 6.25 cm2 demonstrated the same maximum capacity and temperature at a heat flux of 11.2 W/cm2. Choi et al. [6] presented the results of testing an LHP with an evaporator having an active zone of 13.8 cm2 and a heat source of 9 cm2. The maximum heat flux of 20 W/cm2 corresponding to a capacity of 180 W was achieved at an evaporator temperature of 100 °C and a thermal resistance of 0.067 °C/W. In another paper Singh et al. [7] presented the results of testing an LHP with a heat source of 1 cm2 located on an evaporator with an active zone of 4.84 cm2. The maximum heat flux of 50 W/cm2 was achieved at a temperature Tj = 98.5 °C measured at the heating surface of the heat source. The corresponding value of the thermal resistance calculated by formulaRj=Tj-TvQwas equal to 0.3 °C/W. In Ref. [8] one can find the results of testing an LHP with a heat source having the same heating surface of 1 cm2 located on an evaporator with an active zone of 10.24 cm2. An evaporator temperature of 100 °C was achieved at a heat flux of 28 W/cm2. The corresponding value of its thermal resistance was equal to 0.14 °C/W. The same LHP with a heat source of 9 cm2 had an evaporator temperature of 100 °C at a heat flux of 87.7 W/cm2. In this case the evaporator thermal resistance was at a level of 0.012 °C/W. Xu et al. [9] developed and tested an LHP with an evaporator equipped with a two-layer composite wick. The heating surface of the heat source and the active zone of the evaporator had the same value of 7.07 cm2. At a maximum heat flux of 17 W/cm2 the evaporator temperature measured outside the heating zone was equal to 70.3 °C, and the corresponding value of the evaporator thermal resistance was 0.02 °C/W. Reference [10] gives the results of testing an LHP with an evaporator having an active zone of 14 cm2 and a heat source with a heating surface of 6.25 cm2. The maximum value of the evaporator thermal resistance equal to 0.01 °C/W was obtained at a heat flux of 72 W/cm2. The corresponding evaporator temperature, also measured outside the heating zone near the heat source, was equal to 78 °C. When testing an LHP with an evaporator having an active zone of 9 cm2 and a heat source of 6.25 cm2, Li et al. [11] obtained a temperature at the thermocontact surface of the latter equal to 98 °C at a heat flux of 64 W/cm2. The temperature of the evaporator wall next to the heat source was equal to 75 °C. The corresponding value of thermal resistance was at a level of about 0.025 °C/W. Reference [17] presents the results of testing an LHP with an evaporator having an active zone of 27 cm2 and a heat source of 25 cm2. At a heat flux of 24 W/cm2 the temperature of the evaporator wall measured under the heat source at the center of the heating zone was equal to 87.4 °C, and the temperature of the heat source was 93.8 °C. The corresponding value of the evaporator thermal resistance was at a level of 0.019 °C/W.

The efficiency of evaporators of LHPs with other working fluids is not so high, but in some cases, it may be quite acceptable. Thus, for instance, presented in Ref. [18] are the results of testing an LHP with methanol as a working fluid, which was equipped with an evaporator with an active zone of 10.68 cm2. In tests with a heat source of 9.6 cm2 an evaporator thermal resistance of 0.18 °C/W was obtained at a heat flux of 16.6 W/cm2. In Ref. [19] the same heat source was used to achieve a heat flux of 21.8 W/cm2 at an evaporator temperature of 90 °C measured outside the heating zone. The declared value of the thermal resistance was 0.068 °C/W. In Ref. [20] an evaporator with an active zone of 25 cm2 was used to obtain a heat flux of 1.16 W/cm2 at a temperature of 100 °C in the heating zone. The corresponding value of the thermal resistance was equal to 0.12 °C/W.

Tests of an LHP with acetone, where the evaporator active zone and the heat source had the same area of 12 cm2, demonstrated a maximum heat flux of 6.25 W/cm2 and a thermal resistance of 0.12 °C/W [22]. Song et al. [23] tested an LHP with the working fluid R 235fa, which at a temperature of 60 °C has a pressure of 0.43 MPa. Nevertheless, the authors managed to develop a strengthened flat evaporator with an active zone of 54.76 cm2 which operated at vapor temperatures up to 55 °C. With a heat source of 26.24 cm2, a heat flux of 6.1 W/cm2 was achieved at a heat source temperature of 72 °C and a heat sink temperature of 25 °C. The corresponding thermal resistance of the evaporator calculated by formula (3) was equal to 0.11 °C/W. The minimum thermal resistance of 0.08 °C/W was obtained at a heat flux of 2.29 W/cm2.

The above analysis of developments and investigations of LHPs with flat evaporators makes it possible to draw the following conclusions:

  • -

    There is no uniform approach to determining the thermal resistance of evaporators, which complicates a comparative evaluation of their efficiency. Some authors use for calculations the evaporator temperature measured outside the heating zone, which essentially underrates the value of the thermal resistance. A considerable part of the authors uses for calculations the temperature of the heat source measured at its thermocontact surface, which, on the contrary, overrates the value of thermal resistance. A correct approach consists in measuring the temperature of the evaporator wall at the hottest point, which is usually located in the center of the heating zone under the heat source. Special comparative measurements of the temperatures of the evaporator wall in the heating zone and next to it made by the authors of Ref. [17] have shown that the difference between these temperatures, especially at a high heat load, may be quite essential. Correspondingly different are the values of the evaporator thermal resistance calculated by formula (1).

  • -

    There is no uniform definition of the evaporator active zone. Some authors regard it as a heating zone. In actual fact, the active zone of an evaporator is its surface under which a system of vapor-removal channels is situated. A heat load may be supplied to the whole of this surface or a part of it. The part of the evaporator active zone to which a heat load is supplied is the heating zone. As a rule, the heating zone is equal to the thermocontact surface of the heat source.

  • -

    The efficiency of evaporators may differ considerably even when they are made of the same materials, located in LHPs with identical design parameters and use the same working fluid, but operate with heat sources that have different thermocontact surfaces.

These conclusions were taken into account in setting the aim of this work: to show by experiment and analysis how the efficiency of one and the same evaporator, at other equal conditions, depends on the area of the thermocontact surface of the heat source and its relationship with that of the active zone. An additional aim was to show the relationship between the temperature of thermocontact surface of the heat source and the evaporator temperature at a correct location of the measurement points.

Section snippets

Description of the experimental device

An LHP with a flat-oval evaporator of ELR type with water as a working fluid was used for experiments. All the LHP components, including the wick were made of copper. The external view of the device and evaporator scheme are given in Fig. 1 and Fig. 2. Table 1 presents the main design parameters of the LHP.

Experimental setup and testing procedure

The main attention in testing the LHP was given to the investigation of the main thermal characteristics of the evaporator. The tests were conducted at a horizontal position of the device. Three copper blocks measuring 30 × 30 × 10 mm3, 40 × 40 × 10 mm3 and 50 × 50 × 10 mm3 were used as heat sources. The corresponding areas of the thermocontact surface in them were equal to 9 mm2, 16 mm2 и 25 mm2. The heat sources were placed in the evaporator active zone symmetrically relative to its center.

Heat flux and operating temperature

The value of the heat flux, which is one of the main evaporator characteristics, is determined by formula (2). From this formula it follows that the heat flux is directly proportional to the value of the heat load and inversely proportional to the area of the heating surface of the heat source. In its turn, the limiting or critical heat load depends on the thermophysical properties of the working fluid, the design parameters and the LHP slope angle. If the last two conditions are prescribed,

Simulation of thermal processes in an LHP evaporator

To study the effect of the dimensions of the thermocontact surface of the heat source on the thermal processes in an LHP flat-oval evaporator, a 2D evaporator model was developed, and results of numerical simulation were obtained for different values of the heat flux at the surface of the “evaporator – heat source” contact. Fig. 8.A presents the scheme of an evaporator for computer simulation. The simulation model takes into account all the main elements of the evaporator in its active zone,

Conclusion

Thermal characteristics of a flat-oval evaporator 10 mm thick with an active zone measuring 53.5 × 50.5 mm2 operating as part of a copper–water LHP with heat sources having thermocontact surface areas of 9 cm2, 16 cm2 and 25 cm2 have been investigated. The results obtained make it possible to formulate the following conclusions:

  • -

    Thermal characteristics of an evaporator, such as thermal resistance and heat flux, depend essentially on the area of the thermocontact surface of a heat source. They

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References (26)

Cited by (2)

View full text