1 Introduction

Oscillating heat pipe (OHP), is a well cooling element, which can meet the demands of the increasing heat discharge density in some extreme conditions of electronic devices. It can be divided into three types: (1) open looped, (2) closed looped, (3) closed looped with flow control check valves. As a result of its satisfactory performance, cheap cost and high heat transfer capability, the OHP was introduced as an attractive choice to control the temperature of the electronic devices. The study of the complex internal flow of the OHP is still very challenging. Meanwhile, the heat transfer of the OHP also has not been fully understood.

Raffles and Takayoshi [18] have investigated the bubble generation in the OHP, the phase change phenomena in the evaporation section was studied in detail. The result showed that small driving force produced the liquid slugs and vapor slugs. The slugs movement led to the generation and growth of the bubbles, if the velocity of liquid slugs was lower 0.2 m/s, the tube-size bubble would not be generated. Yan et al. [26] studied the heat transfer behavior of the OHP in rolling motion, they mainly investigated three factors, including the span-wise, transverse and stream-wise additional forces. The first two factors were considered under the low Reynolds number. However, the last factors were considerably in low Reynolds number. Finally, they concluded that the effect of velocity oscillation period on the OHP performance is more limited than that of Reynolds number and oscillating velocity Reynolds number. Zhao et al. [29] studied the pressure distribution and flow characteristics of an closed OHP with different vacuum degrees. Yin et al. [30] discussed the operation limitation of an OHP.

It is important to control the heat transfer for the electronic components. Yang et al. [27] built a model to investigate the parameter optimization of the OHP for the electronic components. Their model can make well predictions about the startup characteristics of the OHP. Meanwhile, Yong and Luan [28] minutely investigated the startup characteristics of the OHP. Some of the known physics of this system were studied. The results showed that the model in this paper can produce many features of a real system. Cui et al. [4] studied the operation characteristic and heat transfer mechanism of the OHP in detail. They concluded that the thermal resistance was not decreased linearly, but slowdown or even change into rise first before decreasing. Some factors contributed the phenomenon happen in that experiment, maybe the material, structure and inclination of the OHP have influence on the operation characteristic and heat transfer mechanism.

Wang et al. [24] have studied the thermal performance of the OHP using different heating patterns. They mainly studied the unsteady vapor–liquid two-phase flow inside the OHP. The results showed that the OHP can easily startup under a lower filling ratio, meanwhile, the heating patterns have greatly influence on the thermal resistance and the startup process of the OHP. Mameli et al. [12] studied the thermal performance of a closed OHP under different gravity forces. They found that the thermal behavior can be greatly influenced by the gravity when the OHP positioned vertically. And the OHP can undergo a sudden temperature increase in the evaporator section under the microgravity. Such a behavior was not observed in hyper-gravity conditions.

It is generally known that the working fluid has significantly influence on the thermal performance of the OHP. The most common working fluid were water [31], ethanol [3] and acetone [10], or the two mixtures [33]. Riehl and Santos [17] used the water-copper nanofluid in an open loop OHP, adding some solid nanoparticles cloud enhance the thermal performance of the working fluid. They also concluded that the film evaporation effect was more predominant than nucleate boiling under the low heat loads.

In our previous work, the sintered heat pipe was [16] used for the battery thermal management and the experimental result showed that the maximum temperature of the batteries could be controlled below 50 °C, the heat pipes can be well applied to the battery thermal management. And then, we [15] designed an experiment to apply the OHP to the thermal management of power battery, the results indicated that the OHP can effectively decrease the temperature of the battery packs than the sintered heat pipe, the performance of the battery system was further increased. A phase change material (PCM)/OHP coupled system was built in our previous work [32]. Different influencing factors, including number of turn (3, 4 and 5 turns), thermal conductivity of PCM, flow rate, were considered. This work mainly investigated the heating and cooling process, the whole efficiency of the system.

In order to investigate the thermal performance of oscillating heat pipe in thermal management. The cooling section position, OHP placement and different heating powers were discussed in this paper. The start-up behavior and the temperature oscillations are critically discussed. The results of this experiment will provide a better understanding the performance of the OHP. Meanwhile, the results show that the OHP can be well used in the thermal management.

2 Experiment setup and procedure

Aluminum block was used to experimentally simulate a heating pack. It’s length and width was 115 mm and 90 mm, respectively. Heating rods, which were connected with a direct-current (DC) power supply (the uncertainty was ±0.5 W), were embedded into the aluminum blocks. Different heating powers (5, 15, 25, 35, 45, 55, 65 W) were supplied by the DC power supply. The OHP, was made of copper capillary tube with inner and outer diameters of 2 and 3 mm respectively, was bent into undulating tubes with four turn numbers. Figure 1 shows the charging and vacuum part [25]. The inside vacuum of the OHP was guaranteed by a vacuum pump. The liquid-filled rate of the OHP was maintained at 50 ± 5% (volume) in the experiment. The deionized water was chosen as the working fluid of the OHP. The theoretical inner diameter (Din) of the OHP in this paper was included in Dmax and Dmin (the Eq. 2). Figure 1 shows an elementary diagram of the OHP system. Figure 1a shows the OHP system that the OHP was sandwiched between the two aluminum blocks. Figure 1b is an altered version of the lower half of Fig. 1a. The evaporation and condensation section was 90 and 30 mm, respectively. The thermostat bath was used to ensure the inlet water temperature maintained at 25 ± 0.05 °C, the condensation section was immersed in it. However, the evaporation section was clamped between two adjacent aluminum blocks. In order to decrease thermal contact resistance between aluminum blocks and OHP, the thermal silica (ZC-801) was used to bond them. Three OHP orientations including vertical, 45 degree tilt and horizontal were studied. In order to ensure the uncertainty, the K-type thermocouples were used to measure the temperature of the ice water mixture before the experiment. The temperature of different locations were measure by the K-type thermocouples, which were connected to a Agilent (data acquisition module, 34970 A). CH1,2,3… indicate the temperature channels of the data acquisition module. Figure 1a shows the detail positions of the thermocouples. The temperature of the aluminum blocks were noted by T1–T5. The glass cotton was used to wrap the aluminum blocks to decrease the influence of the environment.

Fig. 1
figure 1

Details of the OHP experimental set-up

Full size image

Figure 2 shows the design of different OHP systems in thermal manegement. The position of the cooling section was clearly seen in this paper, L = 9.5, 8.5, 7.5 and 6.5. The geometric sizes and parameters are summarized in Table 1.

Fig. 2
figure 2

Design of different OHP systems

Full size image
Table 1 Geometric sizes and parameters
Full size table

It is generally known that the experiment results may be influenced by system uncertainty (equipment uncertainty and thermocouple uncertainty). Uncertainty e analysis [9]: the main uncertainties include system uncertainty e s and random uncertainty e r. The uncertainties of the direct measurement parameters such as U and I. The uncertainties of the indirect measurement parameters such as T, R OHP , L and D in . The maximum uncertainties of main parameters are summarized in Table 2.

$$e = \sqrt {e_{s}^{2} + e_{r}^{2} }$$
(1)
Table 2 Maximum uncertainties of main parameters
Full size table

3 Conservation equations

The theoretical maximum and minimum inner diameter (Din) of the OHP can be calculated by the following equation [8, 9], the actual diameter value must be meet this condition. Otherwise, the heat pipe could not start-up:

$$D_{\hbox{min} } = 0.7\sqrt {\frac{\sigma }{{\left( {\rho_{l} - \rho_{v} } \right){\text{g}}}}} \le D_{\text{in}} \le D_{\hbox{max} } = 2\sqrt {\frac{\sigma }{{\left( {\rho_{l} - \rho_{v} } \right){\text{g}}}}}$$
(2)

where ρ l and ρ v are the density of liquid and vapor, respectively. Din is the inner diameter of OHP. Where σ and g represent the surface tension and gravity acceleration.

After filling the heat pipe, the slug was formed inside the OHP due to the interaction between gravitation force and surface tension. However, The balance of the gravitation force and surface tension lead to the definition of the Eötvös number (Eo) and Bond number (Bo) [1], the relationship between Eo and Bo is:

$${{Eo = \left[ {\left( {\rho_{l} - \rho_{v} } \right){\text{gL}}^{ 2} } \right]} \mathord{\left/ {\vphantom {{Eo = \left[ {\left( {\rho_{l} - \rho_{v} } \right){\text{gL}}^{ 2} } \right]} {{{\sigma = \left( {\Delta \rho \,{\text{gL}}^{2} } \right)} \mathord{\left/ {\vphantom {{\sigma = \left( {\Delta \rho \,{\text{gL}}^{2} } \right)} {\sigma = ({\text{Bo}})^{2} }}} \right. \kern-0pt} {\sigma = ({\text{Bo}})^{2} }}}}} \right. \kern-0pt} {{{\sigma = \left( {\Delta \rho \,{\text{gL}}^{2} } \right)} \mathord{\left/ {\vphantom {{\sigma = \left( {\Delta \rho \,{\text{gL}}^{2} } \right)} {\sigma = ({\text{Bo}})^{2} }}} \right. \kern-0pt} {\sigma = ({\text{Bo}})^{2} }}}}$$
(3)

For successful pumping action the gravitational force should be lower than surface tension force. Therefore, the Bo should be less than 2L/D max. Here, L is the total length of the OHP.

The OHP is a well thermal component. For the evaporation section and condensation section heat exchange, OHP can transfer heat with as little as 1 °C temperature difference. However, the thermal resistance (ROHP) has influence on the heat transfer. Simultaneously, there is thermal contact resistance between heating block and OHP. They will disadvantage to the heat transfer. Therefore, some effective methods should be taken to decrease it. Thermal silica is used to decrease the influence of the thermal contact resistance in this study. The OHP thermal resistance can be obtained by the following equation [4]:

$$R_{OHP} = \frac{{(T_{e} - T_{c} )}}{{Q_{in} }}$$
(4)
$$T_{e} = \frac{{T_{4} + T_{5} }}{2}$$
(5)
$$T_{\text{c}} = \frac{{T_{ 1} + T_{ 2} + T_{ 3} }}{3}$$
(6)

Te and Tc are the average surface temperatures of the evaporation and the condensation section, respectively. T1, T2 and T3 are the temperature of different point in the cooling section. T4 and T5 are the temperature of different point in the evaporation. These points can be clearly seen in Fig. 1.

The OHP are applied to many fields due to its high thermal conductivity. The heat pipe can be considered as a very high thermal conductive element in energy storage systems [7, 21]. The heat transfer efficiency of the OHP is 400 times higher than that of solid copper. The effective heat conductivity can be calculated as the flowing equation:

$$k_{eff} = \frac{{L_{p} }}{{A_{1} R_{OHP} }}{ = }\frac{{Q_{in} L_{p} }}{{A_{1} \left( {T_{e} - T_{c} } \right)}}$$
(7)

Lp is the length from the center of the heating section to that of the cooling section and A1 is the total cross-sectional area of OHP. The effective heat conductivity decreases with the increase of the thermal resistance.

The total heat input is Q in this paper. However, there is heat loss Q1 in the heating section even though some necessary heat preservation measures are taken. The actual heat input (Q in ), is smaller than the total heat input, can be calculated. The heat is absorbed by the evaporation, then the fluid boils to vapor phase. All of the experiments are carried out under the room temperature. The experiment is considered under the natural convection in an infinite space.

$$Q_{1} = hA_{1} \nabla T_{1}$$
(8)
$$h = \frac{{Nu\,\lambda_{air} }}{H}$$
(9)
$$Nu = B\left( {GrPr} \right)_{\text{m}}^{n}$$
(10)
$$\Pr = \nu /a$$
(11)
$$Q_{in} = Q - Q_{1}$$
(12)

where Nu and Gr are Nusselt number (Nu is the ratio of convective to conductive heat transfer across (normal to) the boundary) and Grashof number (Gr is a dimensionless number in fluid dynamics and heat transfer which approximates the ratio of the buoyancy to viscous force acting on a fluid), Pr is Prandtl number (Pr is a dimensionless number, defined as the ratio of momentum diffusivity to thermal diffusivity), B, m, n are constants, ν is kinematic viscosity, λ is thermal conductivity, a is thermal diffusivity, h is convective heat transfer coefficient, H is characteristic length.

The heat transfer from inner wall of heat pipe to the outer wall, then vapor condenses to liquid phase, the liquid returns to the evaporation section of the OHP by the gravitation force. The heat (Q 2), releasing from the evaporation section, can be calculated as the following equation:

$$Q_{2} = 2\pi k_{eff} L_{l} \frac{{T_{o} - T_{in} }}{{\ln \left( {{\text{D}}_{\text{o}} - {\text{D}}_{\text{in}} } \right)}}$$
(13)

where L l is the length of the cooling section. Similarly, T in and T o are the temperature of the inner and outer wall of heat pipe,respectively. D o is the outer diameter of OHP.

The heat transferred from the heat pipe to cooling fluid. The heat (Q3) was diffused by the condensation section of heat pipe can be calculated as the following equation:

$$Q_{3} = \rho_{l} \cdot C_{l} \cdot \mu_{l} \cdot\Delta T$$
(14)

where C l and μ l are the specific heat and volume flow rate of water, ΔT is the temperature difference of water outlet and inlet, ρ l is the density of water.

The heat transfer between the heat pipe and cooling fluid was equal to the heat transfer from inner wall of heat pipe to the outer wall. The input heat also equal to them when the whole device condition is adiabatic.

$$Q_{in} = Q_{2} = Q_{3}$$
(15)

4 Results and discussion

Figure 3 shows the schematic of a oscillation heat pipe, the bubble/plug oscillation can be seen inside the OHP. The movement of the liquid and vapor slugs inside the OHP looks like the mechanical vibration. The flowing mathematical model can be utilized to predict the oscillation of the working fluid inside the OHP [10, 34]. The movement equation (Eq. 16) is similar to the forced damped mechanical vibration.

$$\frac{{d^{2} x}}{dt} + \left( {\frac{{8\mu_{l} W\phi_{0} }}{{\rho_{l} DA}}} \right)\frac{dx}{dt} + \frac{{2A^{2} RT_{sat} }}{{\left( {LA\rho_{l} \phi_{0} } \right)\left\{ {\left( {\frac{L}{2}} \right)A\rho_{l} (1 - \phi_{0} )/\rho_{v} } \right\}^{2} }}\left[ {\left( {\frac{L}{2}} \right)A\rho_{l} (1 - \phi_{0} ) + \frac{{Q_{e} }}{{h_{fg} }}t} \right]x = 0$$
(16)
Fig. 3
figure 3

Schematic of the OHP

Full size image

Here, x is the displacement of the movement liquid. Where D, W, L and A are the diameter, perimeter, length and cross sectional area of OHP, respectively, ϕ 0 is the working fluid fill ratio. Qe and h fg are the heat transfer quantity and the latent heat of vaporization of working fluid. ρ l and ρ v are the density of liquid phase and vapor phase under the operating temperature of working fluid; T sat and μ l are the saturated temperature of working fluid and the dynamic viscosity.

The oscillation and circulation of working fluid in the OHP undergo very complex displacements due to the phase change of the working fluid and interactions between the liquid and vapor plugs. The start-up performance, flow characteristics and thermal performance are three important problems of the application of OHP. The heat transfer within the OHP contain two parts: (1) the latent heat transfer between heating section and cooling section, (2) the sensible heat transfer between the OHP wall and the liquid slugs in the form of a single-phase heat transfer [19]. The interface of the liquid–vapor in the OHP can be divided into three regions: equilibrium thin film region, transition film region and intrinsic meniscus region [5]. It can be contribute to the heat transfer when the Qin is big enough. The local heat flux of the capillary tube at the interface is obtained as [23]:

$${\text{q}} = \frac{{dQ_{in} }}{2\pi (r - \delta )dx} = \frac{{T_{w} - T_{ph} }}{{\frac{\delta }{{k_{1} }} + R_{ph} }}$$
(17)
$$R_{ph} = \frac{2 - f}{2f} = \frac{{T_{v} \sqrt {2\pi R_{g} T_{v} } }}{{h_{v} \rho_{v} }}$$
(18)

where R ph is the interfacial heat transfer resistance [20], T ph is the interface temperature. R g is the gas constant. f(0.02 ≤ f ≤ 0.04) is the thermal accommodation coefficient, it can be found in [13]. h v and T v is the latent heat of vaporization and the steam temperature, respectively. ρ v is the vapor density. k 1 is the thermal conductivity of liquid. δ is the thickness of liquid film. T w and r is the temperature of the OHP wall and the inner radius of the OHP, respectively.

The complex flow and heat transfer processes existed at the transition film region.

Heat transfer equation:

$$\left\{ {\begin{array}{*{20}l} {V\frac{\text{dV}}{dx} + \upsilon \frac{\text{dV}}{dy} = \frac{{V_{l} }}{{\rho_{l} }}\frac{{{\text{d}}^{2} V}}{{dy^{2} }} - \frac{{dp_{1} }}{{\rho_{l} dx}} + g} \hfill \\ {\frac{\text{dV}}{dx} + \frac{{{\text{d}}\upsilon }}{dy} = 0} \hfill \\ {V\frac{\text{dT}}{dx} + \upsilon \frac{\text{dT}}{dy} = \frac{\lambda }{{\rho c_{p} }}\frac{{{\text{d}}^{ 2} T}}{{dx^{2} }}} \hfill \\ \end{array} } \right.$$
(19)

Boundary condition [14]:

$$\begin{aligned} {\text{y}} = 0,\;V = 0,\;T = T_{w} \hfill \\ y = \delta ,\;\frac{dV}{dy} = 0,\;T = T_{ph} \hfill \\ \end{aligned}$$
(20)

The velocity distribution can be calculated by Eq. 19 and 20:

$$V_{{\left( {\text{y}} \right)}} = \frac{1}{{V_{1} }}\left( {\frac{{dp_{1} }}{dx} - \rho_{1} g} \right)\left( {\frac{{y^{2} }}{2} - \delta y} \right)$$
(21)

The interfacial equations at the liquid–vapor interface [2]. The mass conservation equation at the liquid–vapor interface can be seen as following equations:

$$\rho_{\text{v}} V_{v} - \rho_{l} V_{l} = \left( {\rho_{\text{v}} - \rho_{l} } \right)\frac{dN}{dt}$$
(22)

Momentum equation:

$$\rho_{\text{v}} \left( {V_{v} - \frac{dN}{dt}} \right)V_{v} - \rho_{l} \left( {V_{l} - \frac{dN}{dt}} \right)V_{l} = \rho_{\text{v}} - \rho_{l} + \frac{\sigma }{\text{r}}$$
(23)

Energy equation:

$$\rho_{\text{v}} \left( {V_{v} - \frac{dN}{dt}} \right){\text{h}}_{v} - \rho_{l} \left( {V_{l} - \frac{dN}{dt}} \right)h_{l} = q_{\text{v}} - q_{l}$$
(24)

where V is the velocity of the normal direction at the interface. N is position of the liquid–vapor interface. t and h are time and enthalpy. σ and r is the surface tension and the radius of curvature at the liquid–vapor interface, respectively.

Figure 4 shows the movement of the vapor slugs and dynamic contact angle hysteresis [10]. In order to maintain the vapor slugs inside the OHP, the condition inside the OHP should meet the Eq. 25 [11]. Only in such condition the steady movement inside the OHP could be developed. Through the Eqs. 25, 26 and 27, the theoretical inner diameter of OHP can be calculated.

$$\Delta P \ge \Delta P^{{\prime }}$$
(25)
$$\Delta P^{{\prime }} = \rho_{\text{l} } gD_{in}$$
(26)
$$\frac{{\Delta P\pi D_{in}^{2} }}{4} = \sigma \pi D_{in}$$
(27)

The high pressure inside the heating section is caused by the input heat, which is the power source to provide the driving force to drive the fluid flow [22]. Due to the different capillary pressure between plug front and plug end lead to the capillary resistance, the three important influence factors on the different capillary pressure are turn numbers, dynamic contact angle hysteresis and gravity. When the OHP start-up, the gravitation force would be dominant over the surface tension with the OHP placed vertically. On the contrary, there would have an opposite effect when the OHP placed horizontally. The contact angle depends on the combination of working fluid and the tube wall of OHP. The fluid in the heat pipe has to overcome the capillary resistance force (F cap ), which can be concluded as the following equation [6]:

$$F_{cap} = \Delta P_{\text{cap}} \cdot \frac{\pi }{4}D_{in}^{2}$$
(28)
$$\Delta P_{\text{cap}} = 4\sigma \left( {cos\theta_{front} - cos\theta_{back} } \right)/D_{in}$$
(29)

where ΔP cap is capillary resistance force per unit area. cosθ front and cosθ back are front contact angle and back contact angle.

Fig. 4
figure 4

Movement of the vapor slugs and dynamic contact angle hysteresis

Full size image

Figure 5 shows the temperature variation of the OHP with different directions when the L was 6.5 cm. The the trend of temperature variation likes the forced damped mechanical vibration. When the heating power was 15 W, the OHP did not oscillate, it was just a heat transfer unit. The temperature of the working fluid did not reach the evaporation temperature. As can be seen from Fig. 5a, when the heating power reached 25 W, the OHP begin start-up, the temperature of the start-up was 78.5 °C,the temperature variation with large amplitude and high frequency, the temperature of the evaporation section increased slowly. The surface tension varied with the temperature, therefore, the Bond number also varied. When the OHP start to oscillate, the working fluid would move inside the OHP, which represents the energy transfer, the flow of the working fluid and the heat transfer was complex inside the OHP, the equations of results and discussion section can explain the flow and heat transfer inside the OHP. With the increment of the heating power, the scope of the swing became bigger than before, so did the oscillation frequency. Figure 5b shows the temperature variation with the OHP placed tiltly (45°), the time of the start-up was later than that of the time when the OHP placed vetically, the temperature of the start-up was 91.6 °C. However, we can clearly see the intermittent temperature change, ΔP < ΔP ′, the stable oscillation did not formed inside the OHP until the heating power reached 45 W. With the increment of the heating power, the maximum temperature did not increase quickly, the temperature changed in a range. However, the OHP did not oscillate until the heating power reached 35 W when the OHP was placed horizontally, the temperature of the start-up was 98 °C and the oscillation amplitude was obviously decreased. The heat conductivity of the OHP was low before working fluid evaporated. However, it would be rapidly increased when the OHP start to oscillate. Meanwhile, the temperature difference between Te and Tc decreased, the R OHP also decreased.

Fig. 5
figure 5

Temperature variation of the OHP with different directions when the L was 6.5 cm

Full size image

With the increment of the heating power, the maximum temperature did not change much. The reason might be that the heat input at the evaporation section and the heat output at the condensation section achieved a balance. Figure 6 shows the maximum temperature of the OHP with different directions when the L was 9.5 cm. The heat pipe was most easy to oscillate when the OHP placed vertically, the temperature of the oscillation was the lowest among these conditions, and the gravity played a great role in this process. When the working fluid evaporated, the gravitation force would be dominant over the surface tension. However, when the OHP placed horizontally, the backflow of the working fluid would be hindered by the gravitation force, the fluid in the heat pipe has to overcome the capillary resistance force. Therefore, the oscillation amplitude decreased, the temperature of the oscillation increased. During the oscillating process, Q 1 should be avoided. Therefore, the Q in would be increased. The k eff would be also increased. The heat could be transferred to the outside quickly.

Fig. 6
figure 6

The maximum temperature of the OHP with different directions when the L was 9.5 cm

Full size image

Figure 7 shows the temperature variation of the OHP with different directions when the L was 9.5 cm. The oscillating heating power was higher than that of when L was 6.5 cm under the same working condition. The oscillation frequency slowed and oscillation amplitude decreased. As can be seen from Fig. 7, the OHP was easy to oscillate heating power when it was placed vetically. On the contrary, it was hart to oscillate when it was placed horizontally. In previous literature [35], it is also difficult to continue oscillating in the OHP with the angle of 0º. The indirect oscillation in Fig. 7b appeared frequently, the reason might be that the pressure inside the OHP instability and the flow inside the OHP was very complex. A phenomenon can be seen that the average temperature of the evaporation section did not change significantly. However, the average temperature of the condensation section increased gradually. If the heating powers continue to increase, the average temperature of the evaporation and condensation section would be very close. If the heating power was high enough, the dry-out phenomenon inside the OHP would appear. This condition would be harm to performance of the OHP. Therefore, the OHP should be applied in the corresponding field. The OHP should be placed vertically when it was applied to the thermal management that the gravitation force could overcome the F cap .

Fig. 7
figure 7

Temperature variation of the OHP with different directions when the L was 9.5 cm

Full size image

Figure 8 shows the temperature variation of the OHP with different cooling section position when the OHP placed vertically. When the L was 9.5 cm, the oscillation amplitude was the smallest in these four cases. The condensation section was closed to the evaporation section, once the liquid evaporation, it would be cooled immediately. When the L was 8.5 cm, the oscillation amplitude increased. However, the oscillation frequency became sparse. When the OHP started, the temperature of the OHP decreased immediately, the temperature oscillation in a range around. The maximum temperature of the start-up can be seen in Fig. 9. With the decrement of the L, the oscillation amplitude increased, the temperature of the start-up decreased except when the L was 9.5 cm. Even though the copper was good thermal conductivity material, the cooling and evaporator section was too close, the absolute adiabatic section did not exist in fact, ultimately resulting in the temperature of the evaporation section could not increase immediately. The cooling effect will be better when the cooling section away from the evaporation section. The temperature could be controlled by the OHP, which could be applied in the field of the thermal management.

Fig. 8
figure 8

Temperature variation of the OHP with different cooling section position when the OHP placed vertically

Full size image
Fig. 9
figure 9

The maximum temperature of the OHP with different cooling section position when the OHP placed vertically

Full size image

5 Conclusions

An experimental investigation was undertaken to study the thermal performance of oscillating heat pipe. This paper presents some new transient experimental data of the closed oscillating heat pipe in terms of condensation and evaporation temperatures. The gravitation force, surface tension, cooling section position and inclination angle were considered in this study. The oscillating behavior and the temperature oscillations are critically discussed. The phase change within the working fluid and the axial oscillation in the OHP prompted the heat transported from the evaporation section to the condensation section. The working fluid inside the OHP was constantly evaporated and cooled. Therefore, the temperature oscillated constantly like the forced damped mechanical vibration. The movement of the working fluid was also similar to the forced damped mechanical vibration. In order to form a stable oscillation, the condition inside the OHP should be ΔP ≥ ΔP ′. The oscillation frequency slowed and oscillation amplitude decreased with the increment of the inclination angle of the OHP. However, the temperature difference of the OHP increased. Therefore, the thermal resistance increased. With the increment of the heating power, the average temperature of the evaporation and condensation section would be very close. When the heating power is large enough, dry-out phenomenon inside the OHP would be appeared. However, 65 W in this paper cannot able to dry out the working fluid, which is needed further researched. With the decrement of the L, the stable oscillation would be formed easy inside the OHP and the oscillating heating power was decreased. The condensation section should not be close to the evaporation section. The temperature could be well controlled in a range by the OHP. When the OHP was applied in the thermal management, the OHP should be placed vertically, the working fluid could return by gravity to the evaporation section. The cooling effect could be better than other placement. Although a lot of research has been done to study the OHP, the study of the complex internal flow of the OHP is still very challenging. However, the OHP can be well applied to thermal management. In order to thorough understanding the OHP, the work about the OHP will be further studied in following-on researches.