- Email: [email protected]
Multiple orientations research on heat transfer performances of Ultra-Thin Loop Heat Pipes with different evaporator structures
International Journal of Heat and Mass Transfer 98 (2016) 415–425
Contents lists available at ScienceDirect
International Journal of Heat and Mass Transfer journal homepage: www.elsevier.com/locate/ijhmt
Multiple orientations research on heat transfer performances of Ultra-Thin Loop Heat Pipes with different evaporator structures Sihui Hong, Shuangfeng Wang ⇑, Zhengguo Zhang School of Chemistry and Chemical Engineering, South China University of Technology, Guangzhou 510640, China
a r t i c l e
i n f o
Article history: Received 14 October 2015 Received in revised form 11 March 2016 Accepted 13 March 2016 Available online 26 March 2016 Keywords: Heat transfer characteristic Looped heat pipe Multi-dimensions Operation stability
a b s t r a c t Two Ultra-Thin Loop Heat Pipe (ULHP) prototypes with parallelogram and trapezoid evaporator configurations were developed for battery thermal management system (BTMS). The dissimilarities between their heat transfer characteristics including the critical work angles, the start-up features, the thermal resistances and the flow instability were all explored and compared with experiments conducted under multi-orientations. The specific influences of these two evaporator configurations on the stable operation of ULHP have been fully acknowledged. The experiments results demonstrated that both the two ULHP prototypes displayed good performances with limited assistance from the gravity, meeting the demand of working under multiple orientations. Specially, the parallelogram configuration showed superior performance in resisting gravity by better suppressing the flow instability, the ULHP could not only start up under 15° inclination, but also the recession in heat transfer capability with placed angles was limited in 4%. Meanwhile, a reasonable mathematics model was established based on the steady operation state of ULHP, the predicted value fitted well with the experimental results. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Recently, with the rapid development of electric vehicle, higher performance requirement of electric power battery has been put forward and draws attention, the corresponding battery thermal management system (BTMS) is getting urgently demanded. Alvani-Soltani pointed out that the uneven temperature distribution or temperature greatly change would lead to the early damage or thermal runaway of the battery, even cause serious safety problem [1]. The battery thermal safety problem is a big obstacle for the wide application and usage of electric vehicles [2]. So far, the available technology applied in battery thermal management includes the air cooling method [3–5], liquid cooling method [6] and phase change material method [7–9]. However, a common problem exists in all these mentioned technologies is the complex structures as well as the huge weight and volume, which not only increases the extra energy consuming, but also goes against the development requirement of automobile lighting. In the future, the required battery thermal management should be conveniently installed, lightweight in both weight and volume, and minimized in secondary energy consumption. Thus, as a passive and effective heat transfer device, the heat pipe technology attracts more attention to be further applied in BTMS. ⇑ Corresponding author. Tel.: +86 020 22236929. E-mail address: [email protected] (S. Wang). http://dx.doi.org/10.1016/j.ijheatmasstransfer.2016.03.049 0017-9310/Ó 2016 Elsevier Ltd. All rights reserved.
Loop heat pipe (LHP) is a highly effective phase change heat transfer device linked by vapor line and liquid line. The heat transfer of LHP is realized by the phase change process of working fluid during circulating between the evaporator and condenser. A significant feature in structure of LHP is the separated vapor line and liquid line and the integration of the evaporator and the compensator, which determines the characters in heat transfer such as the small carrying resistance of steam, the quick start-up and multidirectional long distance heat transmission capability. With the fast development of electronic cooling, the micro flat loop heat pipe that can fit closely with the electronic components surface receives great attention [10–12]. However, the improvement of heat transfer capability of traditional LHP relies on the development of capillary core which requires not only the high capillary limitation but also the low flow resistance, yet the capillary core results in the heat leakage easily and increases the difficulty for LHP to startup at low heat load [13–14]. Moreover, the capillary core is generally relatively heavy, Valery M. Kiseev [15] pointed out with experiments that the evaporator of LHP worked best with capillary core in 5–7 mm thickness. As one of the heat transfer enhancement technologies, micro channel receives widely attention. Taking advantage of the characteristic of micro channel, the capillary core structure was replaced by it in the evaporator, and combined with the feature of separated vapor/liquid line of LHP, the Ultra-thin Loop Heat Pipe (ULHP) was developed, the design details can been referred in paper [16], yet its performance and application needed
416
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
Nomenclature a0 a1 b c d1 d2 h i j k m n u1 u2 b
ev qh qv ql r
A1 A2 Cp
the volume fraction of liquid at the entrance of evaporator the volume fraction of liquid at the exit of evaporator the number of tests the number of variables the equivalent diameter of single micro channel, m the inner diameter of loop heat pipe, m the height of the drawdown, m the sequence number of tests the sequence number of variables the coverage factor the evaporation rate, kg/s the total number of micro channels the speed of working fluid in the loop pipe, m/s the speed of working fluid in the micro channel, m/s the proportion of evaporation heat and total input heat the frictional resistant coefficient the density of high temperature vapor, kg/m3 the density of low temperature vapor, kg/m3 the density of liquid, kg/m3 the surface tension coefficient, N/m the cross section area of loop pipe, m2 the cross section area of single micro channel, m2 the specific heat at constant pressure, J/kgK1
to be discussed for further practical application in various operating conditions of BTMS. As to the structure, ULHP resembles Twophase Loop Thermosyphon [17–20] which however could not work under multiple dimensions as LHP and the thermal capability has been only discussed in vertical orientation so far. Hence, the research of ULHP’s performance under various inclinations plays a significant role in guiding the further practical application of ULHP, especially in BTMS. As the structure inside the evaporator is micro channels instead of capillary core structure, which though brought about the enhancement in heat transfer, but triggered the flow instability and the fluctuation of temperature/pressure. The fluctuation of temperature/pressure and the flow instability [21,22] were always caused by the fast transformation of two-phase flow limited by the micro-channel size. The recession in heat transfer ability usually including the Pre-Critical Heat Flux (PCHF) and the local dry-out, sharp rising of the surface temperature of devices and violent oscillation of pressure would be observed. Especially, when the placed angles decreased and the assistance of the gravity became weaker, the flow instability became more severe, the practical application of the ULHP would be limited. Aims to make sure that the ULHP can work under multiple orientations, the micro channels inside the evaporator brought not only the enhancement in heat transfer, but also the flow instability and the fluctuation of temperature as well as the pressure, which might result in the recession in heat transfer ability, especially when the flow instability became more severe caused by the various orientations, the practical application of the ULHP would be limited. Therefore, in order to improve the circulation efficiency of the working fluid inside the ULHP by solving the flow instability, two prototypes of ULHPs with a parallelogram and a trapezoid evaporator configuration were developed, the difference between their heat transfer capabilities and the effectivities in suppressing flow instability were compared under various angels, the operation characteristics of these two ULHP in antigravity conditions were well understood, and the effectiveness of ultra-thinning and lighting the loop heat pipe was verified.
FR Hf :g Le Lc Lv :l Ll:l DP f DP 0f DP g DP 0g DP s DP capillary Q in R Ts T ev ap: T cond:
DT e U X
the filling ratio of the ULHP the latent heat of phase change, J/kg the length of the evaporator, m the length of the condenser, m the length of the vapor line, m the length of the liquid line, m the frictional resistance of loop pipe, N the frictional resistance of micro channel, N the pressure load of gravity, N the inverse gravity loss of upstream, N the saturated pressure, N the resistance of capillary hysteresis, N the total input heat load, W the thermal resistance, K/W the saturated temperature, K the average temperature of evaporator, T ev ap: ¼ ðT 1 þ T 2 þ T 3 þ T 4 Þ=4, K the average temperature of condenser, T cond: ¼ ðT 6 þ T 7 Þ=2, K the temperature rise of evaporator, K the uncertainty error the Type B test error
Aims to make sure that the ULHP can work under multiple orientations and be adapted to the complex working conditions in BTMS, It is necessary and essential to examine the performance consistency and the operation stability for both the developed prototypes of ULHPs (with parallelogram/trapezoid configuration) under multi-orientation conditions. With experiments, the operation characteristics of these two ULHP under multiple orientations were well understood, and the effectiveness of ultra-thinning and lighting the loop heat pipe was verified. The exact influences of the configurations were clearly identified.
1. Experiment system 1.1. Designed samples The ULHP we studied was shown in Fig. 1. The evaporator was a flat plate with micro-channels inside it. The copper made ULHP can be divided into four parts—the evaporator, the condenser, the vapor line and the liquid line, the values were all listed in Table 1. The length of the entrance section and the length of the vapor/liquid lines were specifically explored and determined with series of experiments. Relevant results and conclusions can be referred to paper [23]. The evaporator of the ULHP was made of two pieces of copper plate. One plate was a 0.5 mm thick smooth plate using as the cover, the other one was 1 mm thick and used as the base board. The groove structure was milled in the base board. 25 rectangle grooves with 3 mm in width and 0.6 mm in depth placed in parallel. For Sample A, the configuration of the channels was parallelogram, the upper and lower edge paralleled to the lower right, the angle between the upper edge and the horizontal is defined as h1 ; h1 is measured as 3.434°. As to Sample B, the distribution of the channels is changed into the trapezoid. The upper edge keeps the same incline as that of Sample A whereas the lower edge inclines symmetrically and forms an isosceles trapezoid shape. The angle between the lower edge and the horizontal here is h2 , which is also 3.434°. Due to the special channel arrangement, a larger
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
Sample A
417
Sample B
Fig. 1. The structure of the designed ULHP samples. (a) The schematic diagram of structure of ULHP and the condensing fins. (b) The picture of ULHP prototype. (c) The schematic diagram of evaporator structure of ULHP.
chamber space at the entrance of the evaporator was formed and defined as ‘‘liquid pool”, related definition can be referred to in paper [24]. Water was chosen as the working fluid. The inner pressure of the ULHP was about 13,000 Pa, the corresponding boiling point was around 51 °C. A fin group was used for condensing. 1.2. Experimental apparatus and procedures The experimental apparatus of ULHP was displayed in Fig. 2. The operating orientation was adjustable. Three cartridges were
inserted into the equally spaced fixed positions, and heated a copper plate with 0.5 mm thickness. The copper plate and the base board of the evaporator bonded tightly by screws locking. The heat came from the copper plate was uniform and the evaporator could be evenly heated. As shown in Fig. 2, the apparatus were made up of the heating part and the condensation part, each part used a DC stabilized power supply (Zhaoxin KXN-645D) as the drive force, one was played as a heating load supplier, the other one was used to drive the fan. The OMEGA K-type thermocouples were installed to measure the wall temperature at different positions of ULHP. The detailed location of thermocouples was shown in Fig. 2.
418
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425 Table 1 The structure parameters of ULHP. Component
Parameters
Sample A
Evaporator
Length width height Arrangement forms Angles h Groove W D L Interval Number of channels Working fluid Filling amount
160 mm 115 mm 1.5 mm Parallelogram h1 = 3.434° 3 mm 0.6 mm 99 mm 3 mm 25 Water 5.0 g
Total length Length of vapor line Length of liquid line Length of entrance section Length of condenser ID/OD
559.65 mm 86.55 mm 235.1 mm 78 mm 160 mm 2.4 mm/3 mm
Outside loop
In the condensation section, an aluminum fin group was added onto the pipe with screws. All tests were conducted at an ambient temperature of 24 °C. The pipeline was insulated with glass fiber whose conductivity was around 0.034 W/(K m). In addition, wooden pallets with low conductivity material (0.05 W/(K m)) were placed under the evaporator and the condenser to support and insulate heat. The total heat loss was limited and less than 0.1%. The experimental error inevitably exists during the experiment process. According to Evaluation and Expression of Uncer-
Sample B Trapezoid h2 = 3.434°
tainty in Measurement JJF1059-1999, for the experimental measured value of direct observed parameters, such as the temperatures, the uncertainty in measurement can be calculated as type A evaluation of uncertainty. For the experiments under the same conditions, repeated tests were conducted to ensure the repeatability and effectiveness of experimental phenomena. The type A evaluation of uncertainty can be calculated as rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi P P ffi m n 1 UðT i Þ ¼ mðn1Þ j¼1 i¼1 ðT ij T i Þ ; thus the largest uncertainty
Fig. 2. The experimental set-up schematic diagram and apparatus. (a) Schematic diagram of experimental set-up. (b) Experimental apparatus.
419
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
of temperature was computed as 1.83 °C, and the largest uncertainty of thermal resistances was 1.533%. The uncertainty of a measurement caused by instruments belongs to the type B evaluation of uncertainty, and is defined as UðX i Þ ¼ a=k; where a is the nominal error range of instrument, k is the coverage factor based on the data distribution of X i . Using rectangular 2 nða0 þa1 Þ Le L p dd21 þ na1LLv :l þna0LLl:l þ distribution estimation, 2 nða0 þa1 Þ 2
Frictional loss DPf ¼ ev
u21 Lv :1 u22 Le ; DP 0f ¼ ev 2g sin a d2 2g sin a d2 ð5Þ
Resistance of capillary hysteresis 80 9 1 > > = 4r 1 n a1 > : ; 1 n 02 1
ð6Þ
LLc ¼ FR is defined as DP g . Hence, for the instruments in the experiments, the uncertainty in measurement of the thermocouples was ±0.289 °C, the uncertainty in measurement of DC stabilized power supply was below 5.77%, and the uncertainty in measurement of ambient temperature was less than ±0.8 °C.
b was defined as the proportion of evaporation heat and total input heat, according to the conservation of energy, Eqs. (7) and (8) were gotten.
2. Model establishment
Sensible heat transfer ð1 bÞ Q ¼
Since there was no capillary core structure but parallel micro channels inside the evaporator of the designed ULHP, the mathematic model based on the operation process was in essence the flow boiling heat transfer process inside the micro channel. The assumptions were as below. (1) The vapor was saturated and described by the state equation of ideal gas. (2) The liquid was incompressible fluid. (3) The heat losses of the vapor/liquid line were ignored. (4) The pressure losses at elbows were ignored. (5) Considered the system had reached the steady state that the working fluid had formed the stable one-way forward circulation, the transient processes of heating and condensing was ignored. (6) The phase change situation was the same for every channel, the evaporation rate was at certain at the certain heat flux.
2 nða0 þ a1 Þ Le d2 n a1 Lv :l n a0 Ll:l nða0 þ a1 Þ Lc p ¼ FR þ þ þ 2 2 L d1 L L L ð1Þ Based on the conservation of momentum, the driven force equaled to total pressure loss at steady state. The driven force included the gravity T ev ap: ¼ ðT 1 þ T 2 þ T 3 þ T 4 Þ=4;the phase change pressure difference T cond: ¼ ðT 6 þ T 7 Þ=2; while the total pressure loss included the frictional resistance of loop pipe DPf ; resistance of capillary hysteresis DPcapillary and the frictional resistance DP0f of the micro channels, inverse gravity loss of upstream
DP0g .
DPdriv e ¼ DPg þ DP s ¼ DPloss ¼ DPf þ DPcapillary þ DP 0f þ DP0g
ð2Þ
where the expressions of each force were described as below, a was the placing angle.
Gravity DPg ¼ Dm1 h sin a; DP 0g ¼ Dml Le g sin a
ð3Þ
B
Phase change pressure difference P s ¼ eAT s þC experimental
parameters,
ð4Þ Ts
was
Heat for evaporation Dmv Hf ;g ¼ n ½qv A1 u1 ð1 ai Þ
qh A1 ul ð1 a0 Þ Hf ;g ¼ b Q ð8Þ
Conservation of mass n ða0 ai Þ A1 u1 ¼ n
a0 þ ai A2 u2 2 ð9Þ
The basic model was achieved as simultaneous solving the above 9 equations. The solution was realized by iterative algorithm in MATLAB nonlinear multivariable equation system with given structure parameters and initial values. The calculation was ended till the deviation was below 0.1%. 3. Results and discussion 3.1. Start-up characteristic
According to assumption 5, Eq. (1) was achieved.
(A,B,C were temperature)
a0 þ ai nq1 A1 u1 C p DT e 2 ð7Þ
saturated
As there was no capillary, the operation of ULHP relied on the gravity and phase change pressure. Under the conditions of nonvertical orientation, the driven force was decreased, it was hard for the one-way forward circulation of working fluid to be guaranteed, and it took more efforts for the system to start up. Namely, there should be a lowest placed angle for ULHP to start up and work. The critical working angle limited the actual application extent of the ULHP. The start-up situations under various inclinations of two samples were compared in Table 2. Sample A had already started up after 2600 s with 50 W heat load at 15° placement, while Sample B couldn’t work until the orientation increased to 30°, where 3000 s was still taken. By comparison, ULHP with parallelogram started up more quickly than the trapezoid one, the corresponding heat load was also lower, demonstrating that ‘‘liquid pool” made great contribute on the fast and stable establishment of the one-way forward circulation of the working fluid. Fig. 3 below showed the variations of the temperatures under different heat loads for the ULHP with the parallelogram and the trapezoid configurations respectively. When the working fluid in the evaporator was overheated and formed vapor, the vapor would rush out from the evaporator due to the high pressure of phase change. The vapor with enough driven force would conquer the
Table 2 The heat loads and the times for the two ULHP samples to start-up.
Sample A Sample B
0°
15°
30°
45°
60°
90°
/ /
40 W/2614 s /
30 W/2350 s 40 W/2996 s
30 W/2158 s 40 W/2630 s
30 W/1880 s 30 W/2546 s
30 W/700 s 30 W/1600 s
420
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
Fig. 3. The variation of temperatures for two samples placed at their critical angles. (a) The variation of temperatures for sample A placed at 15° angles. (b) The variation of temperatures for sample B placed at 30° angles.
Fig. 4. The start-up characteristic curves of two ULHP samples under the critical angles.
total pressure loss over the loop and get cooled in the condensation, and finally end up the circulation as subcooled liquid by flowing into the evaporator. However, if the energy of the vapor was deficient, the vapor that rushed out from the evaporator would fall back to the evaporator again, which embodied as the oscillation of temperatures at T8 and T6.
A significant feature of start-up was the sudden rising of temperature at the entrance of the condenser T6 and the sharp decrease of temperature at the exit of the evaporator T8. The establishment of working fluid’s one-way circulation can be reflected from the start-up curve which was drawn out as Fig. 4. The fluctuation of DT68 was defined as the temperature difference between
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
421
was 4% in the range of 30°–60°, even compared with the vertical condition, the maximum difference was still less than 10%, which demonstrated that the parallelogram configuration had significant performance in resisting the gravity, the placement had very little influence on the heat transfer capability, the consistency of performance under multiple orientations was remarkable. Yet, for ULHP with trapezoid configuration, the influence of orientation was much more negative. The temperature difference among angles was relatively larger, the average difference was enlarged to 7.8%, and the maximum difference was as high as 20% when compared with the vertical situation. Fig. 6 displayed the concrete difference between these two samples under various orientations, which reflected the influence of the placed angles on the difference in heat transfer characteristic between these two ULHPs. Seen from the figure, the temperature difference between these two samples decreased significantly with the increasing of placement angles. The temperature difference reached minimum under vertical condition where the evaporator average temperature almost kept the same below 100 W heat load, while the maximum temperature difference occurred at orientation 30° and reached 9.11 °C. That was to say, the performance in resisting gravity of parallelogram configuration was much superior, the heat transfer capability of trapezoid configuration declined more distinctly with the increasing of placement angles. The parallelogram configuration was more suitable for BTMS with larger scope of application and higher consistence in performance. 3.3. Thermal resistance According to the principle of LHP, the thermal resistance of ULHP was defined as:
R ¼ ðT Ev ap: T Cond: Þ=Q in
Fig. 5. The variations of the evaporator average temperature of the two ULHP samples along with the heat loads under different orientations. (a) The ULHP with parallelogram evaporator configuration. (b) The ULHP with trapezoid evaporator configuration.
T6 and T8, DT 68 ¼ T e6 T e8 , which indeed represented the oscillation of working fluid inside the system. When the system started up, DT 68 was positive and kept increasing with decreasing fluctuation. As can be seen, the ULHP with parallelogram configuration had more fast response, the positive temperature difference increased rapidly, the positive value of DT 68 was achieved at about 600 s, but the system hadn’t completely started up till 2600 s because of the working fluid’s insufficient thermodynamic potential. DT 68 of the trapezoid one finally reached positive value at 1100 s. Similarly, the system had formed stable circulation at 2996 s after more violent oscillation which was due to the more severe recoil phenomenon of the two phase flow.
3.2. The temperature difference Since the lowest orientation for both samples to start up and stably work was 30°, the comparison of other heat transfer characteristics was under 30°–90°orientations. Fig. 5 compared the variations of evaporator average temperature along with heat load under different placement angles. Obviously, for the ULHP with parallelogram configuration, the average temperature variation among different placed angles was small, the average difference
ð10Þ
Thermal resistance was one of the important and direct indices for measuring heat pipe heat transfer capability, the smaller the thermal resistance was, the higher efficiency of heat transfer was. The above discussion about start-up characteristic and operation temperature could be verified from thermal resistance too. Fig. 7 was the variation of thermal resistances. Thanks to the buffer effect of the ‘‘liquid pool”, the thermal resistance of Sample A was averagely 21% lower than that of Sample B. As the heat transfer characteristic of the ULHP with trapezoid configuration was affected more obviously by the placing angles, the difference of its thermal resistances between angles was larger. When placed vertically, the circulation was more smooth and the operation efficiency was higher due to the great assistance of the gravity, the thermal resistance of the samples, therefore, both declined sharply and the difference between them was much smaller. It was interesting to notice that there was an abrupt increase of thermal resistance at 30 W. Due to the serious heat leakage before start-up and the lower heat load, the temperature difference between Tcond and T ev ap: was relatively smaller, the thermal resistance was low at this time. When the heat load increased to 30 W, the ULHP system has already started up, the temperature difference increased while the heat leakage was relieved, the increase of thermal resistance was triggered as a result. The increase of thermal resistance at 30 W also showed some of regularity. Firstly, the extent of the abrupt increase decreased with the increase of the placed angle. It was very clear that the phenomenon didn’t appear when the two samples of ULHP were placed vertically. Secondly, the increase of thermal resistance was much more obvious for the trapezoid configuration that that for the parallelogram one. The regularity indicated that this phenomenon of abrupt change in thermal resistance was closely connected with the chan-
422
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
Fig. 6. The difference of the evaporator average temperature between the two ULHP samples under different angles.
Fig. 7. The variations of the thermal resistance along with the heat loads for the two ULHP samples under different angles.
nel configuration and placed angles. The ‘‘liquid pool” formed in the parallelogram configuration made contribution to reducing the thermal resistance by effectively suppressing the heat leakage caused by the vapor recoil.
3.4. Operation stability The biggest distinction between these two ULHP samples was the ‘‘liquid pool” structure that aimed at reducing the instability
of two phase flow, guaranteeing the stable one-way forward circulation and enhancing the heat transfer capability. T8 located at the entrance section of the evaporator where both affected by the condensed backflow liquid and the heat leakage and vapor recoil, was indeed a place with the most violent fluctuation and unsteady state. Comparing the fluctuation situation at T8 position under different angles gave further clear explication of the effect of ‘‘liquid pool” on the system operation. Fig. 8(a), (b) were the temperature fluctuation situations of the two samples respectively. The temperature fluctuation at T8 for Sample A was limited
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
423
Fig. 8. The variations of the temperature oscillation along with the angles for the two ULHP samples. (a) The ULHP with parallelogram evaporator configuration. (b) The ULHP with trapezoid evaporator configuration.
in ±6° while the range for Sample B was enlarged to ±12°, showing that the ‘‘liquid pool” in parallelogram configuration effectively suppressed the temperature oscillation by reducing the flow resistance for backflow liquid to enter the evaporator and accelerating the flow-out of vapor. Moreover, the temperature fluctuation of Sample A had better consistency under various angles than that of Sample B, namely, the ability to resist the gravity was superior, which also fitted with the results of the above discussion. What’s worthy noticing, the lowest fluctuation of Sample A occurred at 45° instead of vertical placing. Under vertical placing, the flow instability caused by the gravity and the friction of two-phase flow in the upstream was more prominent that led to a larger oscillation. As to Sample B, the fluctuation reached the lowest at 30° placement, which was relative to not only the above factors, but also the trapezoid configuration itself. Since the lower part of the evaporator of Sample B shaped like a divergent nozzle, when placed under the large orientations, the working fluid tended to flow to the entrance because of the height difference, and formed counterflow that increased the difficulty for the one-way forward
flow to be formed. The height difference of the evaporator’s right side reduced with the decreasing of the orientations, the counterflow and the flow resistance decreased as well, the backflow liquid supplied to the evaporator more easily, the fluctuation was eased consequently. 3.5. Model verification Fig. 9 showed the comparison validation of prediction values and the experiment results. Based on assumption 5, the system was under unsteady state when placed below critical angles or loaded less than 40 W heat load, the flow and heat transfer progress were too complicated to simplify with this model. Thus, only after the ULHP started up and formed stable flow, the steady state was calculated, namely, the calculation started from 30° and 40 W. Learned from the figure, the model fitted the experimental data well, the average error of was 5.7% while the largest one was 12%, indicating that the model basically depicted operation process and reflected the flow and heat transfer characteristic. The error
424
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
Fig. 9. The comparative results of the predicted values with the experimental data.
was larger for low heat load and low orientation situations. As placed under 30°, the error between the predicted values and the experiment data was larger than other cases, and the maximum error appeared at heat load 50 W. Relatively, the error reduced along with the increase of the orientations, the model fitted best for vertical situation. The cause of this situation was that the ULHP system started up slower under lower orientations, the working fluid hadn’t established steady slug flow, instead, the bubbles might be at local flooded, collapsed and combined, or growth state, that’s why the model error was larger in these cases. Adding the pattern development process would make the model more close to the physical phenomenon and more precise.
(3) The ‘‘liquid pool” structure played important role in easing the flow instability, the temperature fluctuation decreased, not only the heat transfer capability was enhanced, but also the damage to the battery was effectively reduced, BTMS could work more stably. (4) The built model well predicted the operation temperature which fitted well with the experimental data. Limited by the assumption, the error of prediction values was larger under low heat load. Adding the development of flow pattern under different heat loads could better match the practical physical phenomenon and increase the accuracy of the model, which would be one emphasis of our work in the future.
4. Conclusion The heat transfer characteristics and their difference of two ULHPs that had different channel configuration were studied with experiments, a mathematic model based on the operation process was also established, the exact affects of the channel configurations on the flow and heat transfer process were fully understood, which was beneficial for further guide in the design and optimization of the ULHP. The research under multiple orientations gave further clues for ULHPs to be applied to different operation conditions in BTMS. Relative conclusions were listed below. (1) Both of the two ULHP samples had critical angle, but the angles varied with the different configuration. The ULHP with parallelogram channel configuration started up steadily at 15° while the trapezoid one didn’t work until the angle increased to 30°. The parallelogram configuration helped ensure the one-way forward circulation, the response of start-up was obviously faster. (2) The capabilities in resisting the gravity of these two ULHP samples differed. The ULHP with parallelogram configuration showed superior capability in resisting the gravity with smaller difference in heat transfer characteristic between angles, the average and largest difference were 4% and 10% respectively, while those for the trapezoid one achieved 7.8% and 20% separately. The superiority of parallelogram configuration structure stood out under small angles.
Acknowledgements This work was supported by National Natural Science Foundation of China (Grant No. 51476059) and the International Cooperation and Exchange Program from the Ministry of Science and Technology of China (Grant No. 2013DFG60080). References [1] S.R. Alvani-Soltani, T.S. Ravigururajan, M. Rezac, in: Proceedings of IMECE, American Society of Mechanical Engineers, 2006, pp. 383–386. [2] K. Smith, C.Y. Wang, Power and thermal characterization of a lithium-ion battery pack for hybrid-electric vehicles, J. Power Sources 160 (2006) 662–673. [3] Y.F. Chen, J.W. Evans, Heat transfer phenomena in lithium/polymer-electrolyte batteries for electric vehicle application, J. Electrochem. Sources 140 (7) (1993) 1833–1838. [4] M.S. Wu, K.H. Liu, Y.Y. Wang, C.C. Wan, Heat dissipation design for lithium-ion batteries, J. Power Sources 109 (1) (2002) 160–166. [5] R. Sabbah, R. Kizilel, J.R. Selman, S. Al-Hallaj, Active (air-cooled) vs. passive (phase change material) thermal management of high power lithium-ion packs: limitation of temperature rise and uniformity of temperature distribution, J. Power Sources 182 (2008) 630–638. [6] Zhijun Tang, Qunzhi Zhu, Research on thermal management technology for power batteries, Chin. J. Power Sources 137 (1) (2013) 103–106. [7] Siddique A. Khateeb, Shabab Amiruddin, J. Robert Selman, Said Al-Hallaj, Thermal management of Li-ion battery with phase change material for electric scooters: experimental validation, J. Power Sources 142 (2005) 345–353. [8] R. Kizilel, A. Lateef, R. Sabbah, M.M. Faridb, J.R. Selmana, S. Al-Hallajet, Passive control of temperature excursion and uniformity in high-energy Li-ion battery
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
[9]
[10] [11]
[12]
[13] [14]
[15]
[16]
packs at high current and ambient temperature, J. Power Sources 183 (2008) 370–375. X. Duan, G.F. Naterer, Heat transfer in phase change materials for thermal management of electric vehicle battery modules, Int. J. Heat Mass Transf. 53 (2010) 5176–5182. Y.C. Xiang, Z.Q. Hou, J.X. Zhang, The State-of-the-art of loop heat pipe, J. Eng. Thermophys. 25 (4) (2004) 682–684. J. Li, D. Wang, G.P. Peterson, Experimental studies on a high performance compact loop heat pipe with a square flat evaporator, Appl. Therm. Eng. 30 (2010) 741–752. S.C. Wu, J.C. Peng, S.R. Lai, et al., Investigation of the effect of heat leak in loop heat pipes with flat evaporator, IEEE Impact your future: Integration, Efficiency & the Eco-Friendly, 2009. K.J. Zan, C.J. Zan, Y.M. Chen, et al., Analysis of the parameters of the sintered loop heat pipe, Heat Transfer-Asian Res. 33 (8) (2004) 515–526. Xu. Jiyuan, Zou. Yong, Cheng. Lin, Pore structure optimization and properties of composite wicks for loop heat pipes, Proceedings of the CSEE 32 (23) (2012) 70–74. V.M. Kiseev, V.V. Vlassov, I. Muraoka, Experimental optimization of capillary structures for loop heat pipes and heat switches, Appl. Therm. Eng. 30 (2010) 1312–1319. Sihui Hong, Xinqiang Zhang, Shuangfeng Wang, Zhengguo Zhang, Experiment study on heat transfer capability of an innovative ultra-thin loop heat pipe, Int. J. Therm. Sci. 95 (2015) 106–114.
425
[17] Yu.F. Maydanik, M.A. Chernysheva, V.G. Pastukhov, Review: loop heat pipes with flat evaporators, Appl. Therm. Eng. 67 (2014) 294–307. [18] Alessandro Franco, Sauro Filippeschi, Closed loop two-phase thermosyphon of small dimensions: a review of the experimental results, Microgravity Sci. Technol. 24 (3) (2012) 165–179. [19] L.P. Grakovich, M.I. Rabetsky, L.L. Vasiliev, L.L. Vasiliev Jr., Polymer flat loop thermosyphons, 17th International Heat Pipe Conference (17th IHPC) Kanpur, India, October, 2013, pp. 13–17. [20] Rahmatollah Khodabandeh, Björn Palm, Influence of system pressure on the boiling heat transfer coefficient in a closed two-phase thermosyphon loop, Int. J. Therm. Sci. 41 (2002) 619–624. [21] S.S. Bertsch, E.A. Groll, S.V. Garimella, Refrigerant flow boiling heat transfer in parallel microchannels as a function of local vapor quality, Int. J. Heat Mass Transf. 51 (2008) 4775–4787. [22] Leonardo Carlos Ruspini, Christian Pablo Marcel, Alejandro Clausse, Twophase flow instabilities: a review, Int. J. Heat Mass Transf. 71 (2014) 521–548. [23] Sihui Hong, Xinqiang Zhang, Shuangfeng Wang, Zhengguo Zhang, Experimental Investigation on the Characters of Ultra-Thin Loop Heat Pipe Applied in BTMS, The 7th International Conference on Applied Energy (ICAE), 2015, pp. 3192–3200. [24] Sihui Hong, Xinqiang Zhang, Yongle Tang, Shuangfeng Wang, Zhengguo Zhang, Experiment research on the effect of the evaporator’s configuration design of an innovative ultra-thin looped heat pipe, Int. J. Heat Mass Transf. 92 (2016) 497–506.
Contents lists available at ScienceDirect
International Journal of Heat and Mass Transfer journal homepage: www.elsevier.com/locate/ijhmt
Multiple orientations research on heat transfer performances of Ultra-Thin Loop Heat Pipes with different evaporator structures Sihui Hong, Shuangfeng Wang ⇑, Zhengguo Zhang School of Chemistry and Chemical Engineering, South China University of Technology, Guangzhou 510640, China
a r t i c l e
i n f o
Article history: Received 14 October 2015 Received in revised form 11 March 2016 Accepted 13 March 2016 Available online 26 March 2016 Keywords: Heat transfer characteristic Looped heat pipe Multi-dimensions Operation stability
a b s t r a c t Two Ultra-Thin Loop Heat Pipe (ULHP) prototypes with parallelogram and trapezoid evaporator configurations were developed for battery thermal management system (BTMS). The dissimilarities between their heat transfer characteristics including the critical work angles, the start-up features, the thermal resistances and the flow instability were all explored and compared with experiments conducted under multi-orientations. The specific influences of these two evaporator configurations on the stable operation of ULHP have been fully acknowledged. The experiments results demonstrated that both the two ULHP prototypes displayed good performances with limited assistance from the gravity, meeting the demand of working under multiple orientations. Specially, the parallelogram configuration showed superior performance in resisting gravity by better suppressing the flow instability, the ULHP could not only start up under 15° inclination, but also the recession in heat transfer capability with placed angles was limited in 4%. Meanwhile, a reasonable mathematics model was established based on the steady operation state of ULHP, the predicted value fitted well with the experimental results. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction Recently, with the rapid development of electric vehicle, higher performance requirement of electric power battery has been put forward and draws attention, the corresponding battery thermal management system (BTMS) is getting urgently demanded. Alvani-Soltani pointed out that the uneven temperature distribution or temperature greatly change would lead to the early damage or thermal runaway of the battery, even cause serious safety problem [1]. The battery thermal safety problem is a big obstacle for the wide application and usage of electric vehicles [2]. So far, the available technology applied in battery thermal management includes the air cooling method [3–5], liquid cooling method [6] and phase change material method [7–9]. However, a common problem exists in all these mentioned technologies is the complex structures as well as the huge weight and volume, which not only increases the extra energy consuming, but also goes against the development requirement of automobile lighting. In the future, the required battery thermal management should be conveniently installed, lightweight in both weight and volume, and minimized in secondary energy consumption. Thus, as a passive and effective heat transfer device, the heat pipe technology attracts more attention to be further applied in BTMS. ⇑ Corresponding author. Tel.: +86 020 22236929. E-mail address: [email protected] (S. Wang). http://dx.doi.org/10.1016/j.ijheatmasstransfer.2016.03.049 0017-9310/Ó 2016 Elsevier Ltd. All rights reserved.
Loop heat pipe (LHP) is a highly effective phase change heat transfer device linked by vapor line and liquid line. The heat transfer of LHP is realized by the phase change process of working fluid during circulating between the evaporator and condenser. A significant feature in structure of LHP is the separated vapor line and liquid line and the integration of the evaporator and the compensator, which determines the characters in heat transfer such as the small carrying resistance of steam, the quick start-up and multidirectional long distance heat transmission capability. With the fast development of electronic cooling, the micro flat loop heat pipe that can fit closely with the electronic components surface receives great attention [10–12]. However, the improvement of heat transfer capability of traditional LHP relies on the development of capillary core which requires not only the high capillary limitation but also the low flow resistance, yet the capillary core results in the heat leakage easily and increases the difficulty for LHP to startup at low heat load [13–14]. Moreover, the capillary core is generally relatively heavy, Valery M. Kiseev [15] pointed out with experiments that the evaporator of LHP worked best with capillary core in 5–7 mm thickness. As one of the heat transfer enhancement technologies, micro channel receives widely attention. Taking advantage of the characteristic of micro channel, the capillary core structure was replaced by it in the evaporator, and combined with the feature of separated vapor/liquid line of LHP, the Ultra-thin Loop Heat Pipe (ULHP) was developed, the design details can been referred in paper [16], yet its performance and application needed
416
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
Nomenclature a0 a1 b c d1 d2 h i j k m n u1 u2 b
ev qh qv ql r
A1 A2 Cp
the volume fraction of liquid at the entrance of evaporator the volume fraction of liquid at the exit of evaporator the number of tests the number of variables the equivalent diameter of single micro channel, m the inner diameter of loop heat pipe, m the height of the drawdown, m the sequence number of tests the sequence number of variables the coverage factor the evaporation rate, kg/s the total number of micro channels the speed of working fluid in the loop pipe, m/s the speed of working fluid in the micro channel, m/s the proportion of evaporation heat and total input heat the frictional resistant coefficient the density of high temperature vapor, kg/m3 the density of low temperature vapor, kg/m3 the density of liquid, kg/m3 the surface tension coefficient, N/m the cross section area of loop pipe, m2 the cross section area of single micro channel, m2 the specific heat at constant pressure, J/kgK1
to be discussed for further practical application in various operating conditions of BTMS. As to the structure, ULHP resembles Twophase Loop Thermosyphon [17–20] which however could not work under multiple dimensions as LHP and the thermal capability has been only discussed in vertical orientation so far. Hence, the research of ULHP’s performance under various inclinations plays a significant role in guiding the further practical application of ULHP, especially in BTMS. As the structure inside the evaporator is micro channels instead of capillary core structure, which though brought about the enhancement in heat transfer, but triggered the flow instability and the fluctuation of temperature/pressure. The fluctuation of temperature/pressure and the flow instability [21,22] were always caused by the fast transformation of two-phase flow limited by the micro-channel size. The recession in heat transfer ability usually including the Pre-Critical Heat Flux (PCHF) and the local dry-out, sharp rising of the surface temperature of devices and violent oscillation of pressure would be observed. Especially, when the placed angles decreased and the assistance of the gravity became weaker, the flow instability became more severe, the practical application of the ULHP would be limited. Aims to make sure that the ULHP can work under multiple orientations, the micro channels inside the evaporator brought not only the enhancement in heat transfer, but also the flow instability and the fluctuation of temperature as well as the pressure, which might result in the recession in heat transfer ability, especially when the flow instability became more severe caused by the various orientations, the practical application of the ULHP would be limited. Therefore, in order to improve the circulation efficiency of the working fluid inside the ULHP by solving the flow instability, two prototypes of ULHPs with a parallelogram and a trapezoid evaporator configuration were developed, the difference between their heat transfer capabilities and the effectivities in suppressing flow instability were compared under various angels, the operation characteristics of these two ULHP in antigravity conditions were well understood, and the effectiveness of ultra-thinning and lighting the loop heat pipe was verified.
FR Hf :g Le Lc Lv :l Ll:l DP f DP 0f DP g DP 0g DP s DP capillary Q in R Ts T ev ap: T cond:
DT e U X
the filling ratio of the ULHP the latent heat of phase change, J/kg the length of the evaporator, m the length of the condenser, m the length of the vapor line, m the length of the liquid line, m the frictional resistance of loop pipe, N the frictional resistance of micro channel, N the pressure load of gravity, N the inverse gravity loss of upstream, N the saturated pressure, N the resistance of capillary hysteresis, N the total input heat load, W the thermal resistance, K/W the saturated temperature, K the average temperature of evaporator, T ev ap: ¼ ðT 1 þ T 2 þ T 3 þ T 4 Þ=4, K the average temperature of condenser, T cond: ¼ ðT 6 þ T 7 Þ=2, K the temperature rise of evaporator, K the uncertainty error the Type B test error
Aims to make sure that the ULHP can work under multiple orientations and be adapted to the complex working conditions in BTMS, It is necessary and essential to examine the performance consistency and the operation stability for both the developed prototypes of ULHPs (with parallelogram/trapezoid configuration) under multi-orientation conditions. With experiments, the operation characteristics of these two ULHP under multiple orientations were well understood, and the effectiveness of ultra-thinning and lighting the loop heat pipe was verified. The exact influences of the configurations were clearly identified.
1. Experiment system 1.1. Designed samples The ULHP we studied was shown in Fig. 1. The evaporator was a flat plate with micro-channels inside it. The copper made ULHP can be divided into four parts—the evaporator, the condenser, the vapor line and the liquid line, the values were all listed in Table 1. The length of the entrance section and the length of the vapor/liquid lines were specifically explored and determined with series of experiments. Relevant results and conclusions can be referred to paper [23]. The evaporator of the ULHP was made of two pieces of copper plate. One plate was a 0.5 mm thick smooth plate using as the cover, the other one was 1 mm thick and used as the base board. The groove structure was milled in the base board. 25 rectangle grooves with 3 mm in width and 0.6 mm in depth placed in parallel. For Sample A, the configuration of the channels was parallelogram, the upper and lower edge paralleled to the lower right, the angle between the upper edge and the horizontal is defined as h1 ; h1 is measured as 3.434°. As to Sample B, the distribution of the channels is changed into the trapezoid. The upper edge keeps the same incline as that of Sample A whereas the lower edge inclines symmetrically and forms an isosceles trapezoid shape. The angle between the lower edge and the horizontal here is h2 , which is also 3.434°. Due to the special channel arrangement, a larger
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
Sample A
417
Sample B
Fig. 1. The structure of the designed ULHP samples. (a) The schematic diagram of structure of ULHP and the condensing fins. (b) The picture of ULHP prototype. (c) The schematic diagram of evaporator structure of ULHP.
chamber space at the entrance of the evaporator was formed and defined as ‘‘liquid pool”, related definition can be referred to in paper [24]. Water was chosen as the working fluid. The inner pressure of the ULHP was about 13,000 Pa, the corresponding boiling point was around 51 °C. A fin group was used for condensing. 1.2. Experimental apparatus and procedures The experimental apparatus of ULHP was displayed in Fig. 2. The operating orientation was adjustable. Three cartridges were
inserted into the equally spaced fixed positions, and heated a copper plate with 0.5 mm thickness. The copper plate and the base board of the evaporator bonded tightly by screws locking. The heat came from the copper plate was uniform and the evaporator could be evenly heated. As shown in Fig. 2, the apparatus were made up of the heating part and the condensation part, each part used a DC stabilized power supply (Zhaoxin KXN-645D) as the drive force, one was played as a heating load supplier, the other one was used to drive the fan. The OMEGA K-type thermocouples were installed to measure the wall temperature at different positions of ULHP. The detailed location of thermocouples was shown in Fig. 2.
418
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425 Table 1 The structure parameters of ULHP. Component
Parameters
Sample A
Evaporator
Length width height Arrangement forms Angles h Groove W D L Interval Number of channels Working fluid Filling amount
160 mm 115 mm 1.5 mm Parallelogram h1 = 3.434° 3 mm 0.6 mm 99 mm 3 mm 25 Water 5.0 g
Total length Length of vapor line Length of liquid line Length of entrance section Length of condenser ID/OD
559.65 mm 86.55 mm 235.1 mm 78 mm 160 mm 2.4 mm/3 mm
Outside loop
In the condensation section, an aluminum fin group was added onto the pipe with screws. All tests were conducted at an ambient temperature of 24 °C. The pipeline was insulated with glass fiber whose conductivity was around 0.034 W/(K m). In addition, wooden pallets with low conductivity material (0.05 W/(K m)) were placed under the evaporator and the condenser to support and insulate heat. The total heat loss was limited and less than 0.1%. The experimental error inevitably exists during the experiment process. According to Evaluation and Expression of Uncer-
Sample B Trapezoid h2 = 3.434°
tainty in Measurement JJF1059-1999, for the experimental measured value of direct observed parameters, such as the temperatures, the uncertainty in measurement can be calculated as type A evaluation of uncertainty. For the experiments under the same conditions, repeated tests were conducted to ensure the repeatability and effectiveness of experimental phenomena. The type A evaluation of uncertainty can be calculated as rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi P P ffi m n 1 UðT i Þ ¼ mðn1Þ j¼1 i¼1 ðT ij T i Þ ; thus the largest uncertainty
Fig. 2. The experimental set-up schematic diagram and apparatus. (a) Schematic diagram of experimental set-up. (b) Experimental apparatus.
419
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
of temperature was computed as 1.83 °C, and the largest uncertainty of thermal resistances was 1.533%. The uncertainty of a measurement caused by instruments belongs to the type B evaluation of uncertainty, and is defined as UðX i Þ ¼ a=k; where a is the nominal error range of instrument, k is the coverage factor based on the data distribution of X i . Using rectangular 2 nða0 þa1 Þ Le L p dd21 þ na1LLv :l þna0LLl:l þ distribution estimation, 2 nða0 þa1 Þ 2
Frictional loss DPf ¼ ev
u21 Lv :1 u22 Le ; DP 0f ¼ ev 2g sin a d2 2g sin a d2 ð5Þ
Resistance of capillary hysteresis 80 9 1 > > = 4r 1 n a1 > : ; 1 n 02 1
ð6Þ
LLc ¼ FR is defined as DP g . Hence, for the instruments in the experiments, the uncertainty in measurement of the thermocouples was ±0.289 °C, the uncertainty in measurement of DC stabilized power supply was below 5.77%, and the uncertainty in measurement of ambient temperature was less than ±0.8 °C.
b was defined as the proportion of evaporation heat and total input heat, according to the conservation of energy, Eqs. (7) and (8) were gotten.
2. Model establishment
Sensible heat transfer ð1 bÞ Q ¼
Since there was no capillary core structure but parallel micro channels inside the evaporator of the designed ULHP, the mathematic model based on the operation process was in essence the flow boiling heat transfer process inside the micro channel. The assumptions were as below. (1) The vapor was saturated and described by the state equation of ideal gas. (2) The liquid was incompressible fluid. (3) The heat losses of the vapor/liquid line were ignored. (4) The pressure losses at elbows were ignored. (5) Considered the system had reached the steady state that the working fluid had formed the stable one-way forward circulation, the transient processes of heating and condensing was ignored. (6) The phase change situation was the same for every channel, the evaporation rate was at certain at the certain heat flux.
2 nða0 þ a1 Þ Le d2 n a1 Lv :l n a0 Ll:l nða0 þ a1 Þ Lc p ¼ FR þ þ þ 2 2 L d1 L L L ð1Þ Based on the conservation of momentum, the driven force equaled to total pressure loss at steady state. The driven force included the gravity T ev ap: ¼ ðT 1 þ T 2 þ T 3 þ T 4 Þ=4;the phase change pressure difference T cond: ¼ ðT 6 þ T 7 Þ=2; while the total pressure loss included the frictional resistance of loop pipe DPf ; resistance of capillary hysteresis DPcapillary and the frictional resistance DP0f of the micro channels, inverse gravity loss of upstream
DP0g .
DPdriv e ¼ DPg þ DP s ¼ DPloss ¼ DPf þ DPcapillary þ DP 0f þ DP0g
ð2Þ
where the expressions of each force were described as below, a was the placing angle.
Gravity DPg ¼ Dm1 h sin a; DP 0g ¼ Dml Le g sin a
ð3Þ
B
Phase change pressure difference P s ¼ eAT s þC experimental
parameters,
ð4Þ Ts
was
Heat for evaporation Dmv Hf ;g ¼ n ½qv A1 u1 ð1 ai Þ
qh A1 ul ð1 a0 Þ Hf ;g ¼ b Q ð8Þ
Conservation of mass n ða0 ai Þ A1 u1 ¼ n
a0 þ ai A2 u2 2 ð9Þ
The basic model was achieved as simultaneous solving the above 9 equations. The solution was realized by iterative algorithm in MATLAB nonlinear multivariable equation system with given structure parameters and initial values. The calculation was ended till the deviation was below 0.1%. 3. Results and discussion 3.1. Start-up characteristic
According to assumption 5, Eq. (1) was achieved.
(A,B,C were temperature)
a0 þ ai nq1 A1 u1 C p DT e 2 ð7Þ
saturated
As there was no capillary, the operation of ULHP relied on the gravity and phase change pressure. Under the conditions of nonvertical orientation, the driven force was decreased, it was hard for the one-way forward circulation of working fluid to be guaranteed, and it took more efforts for the system to start up. Namely, there should be a lowest placed angle for ULHP to start up and work. The critical working angle limited the actual application extent of the ULHP. The start-up situations under various inclinations of two samples were compared in Table 2. Sample A had already started up after 2600 s with 50 W heat load at 15° placement, while Sample B couldn’t work until the orientation increased to 30°, where 3000 s was still taken. By comparison, ULHP with parallelogram started up more quickly than the trapezoid one, the corresponding heat load was also lower, demonstrating that ‘‘liquid pool” made great contribute on the fast and stable establishment of the one-way forward circulation of the working fluid. Fig. 3 below showed the variations of the temperatures under different heat loads for the ULHP with the parallelogram and the trapezoid configurations respectively. When the working fluid in the evaporator was overheated and formed vapor, the vapor would rush out from the evaporator due to the high pressure of phase change. The vapor with enough driven force would conquer the
Table 2 The heat loads and the times for the two ULHP samples to start-up.
Sample A Sample B
0°
15°
30°
45°
60°
90°
/ /
40 W/2614 s /
30 W/2350 s 40 W/2996 s
30 W/2158 s 40 W/2630 s
30 W/1880 s 30 W/2546 s
30 W/700 s 30 W/1600 s
420
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
Fig. 3. The variation of temperatures for two samples placed at their critical angles. (a) The variation of temperatures for sample A placed at 15° angles. (b) The variation of temperatures for sample B placed at 30° angles.
Fig. 4. The start-up characteristic curves of two ULHP samples under the critical angles.
total pressure loss over the loop and get cooled in the condensation, and finally end up the circulation as subcooled liquid by flowing into the evaporator. However, if the energy of the vapor was deficient, the vapor that rushed out from the evaporator would fall back to the evaporator again, which embodied as the oscillation of temperatures at T8 and T6.
A significant feature of start-up was the sudden rising of temperature at the entrance of the condenser T6 and the sharp decrease of temperature at the exit of the evaporator T8. The establishment of working fluid’s one-way circulation can be reflected from the start-up curve which was drawn out as Fig. 4. The fluctuation of DT68 was defined as the temperature difference between
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
421
was 4% in the range of 30°–60°, even compared with the vertical condition, the maximum difference was still less than 10%, which demonstrated that the parallelogram configuration had significant performance in resisting the gravity, the placement had very little influence on the heat transfer capability, the consistency of performance under multiple orientations was remarkable. Yet, for ULHP with trapezoid configuration, the influence of orientation was much more negative. The temperature difference among angles was relatively larger, the average difference was enlarged to 7.8%, and the maximum difference was as high as 20% when compared with the vertical situation. Fig. 6 displayed the concrete difference between these two samples under various orientations, which reflected the influence of the placed angles on the difference in heat transfer characteristic between these two ULHPs. Seen from the figure, the temperature difference between these two samples decreased significantly with the increasing of placement angles. The temperature difference reached minimum under vertical condition where the evaporator average temperature almost kept the same below 100 W heat load, while the maximum temperature difference occurred at orientation 30° and reached 9.11 °C. That was to say, the performance in resisting gravity of parallelogram configuration was much superior, the heat transfer capability of trapezoid configuration declined more distinctly with the increasing of placement angles. The parallelogram configuration was more suitable for BTMS with larger scope of application and higher consistence in performance. 3.3. Thermal resistance According to the principle of LHP, the thermal resistance of ULHP was defined as:
R ¼ ðT Ev ap: T Cond: Þ=Q in
Fig. 5. The variations of the evaporator average temperature of the two ULHP samples along with the heat loads under different orientations. (a) The ULHP with parallelogram evaporator configuration. (b) The ULHP with trapezoid evaporator configuration.
T6 and T8, DT 68 ¼ T e6 T e8 , which indeed represented the oscillation of working fluid inside the system. When the system started up, DT 68 was positive and kept increasing with decreasing fluctuation. As can be seen, the ULHP with parallelogram configuration had more fast response, the positive temperature difference increased rapidly, the positive value of DT 68 was achieved at about 600 s, but the system hadn’t completely started up till 2600 s because of the working fluid’s insufficient thermodynamic potential. DT 68 of the trapezoid one finally reached positive value at 1100 s. Similarly, the system had formed stable circulation at 2996 s after more violent oscillation which was due to the more severe recoil phenomenon of the two phase flow.
3.2. The temperature difference Since the lowest orientation for both samples to start up and stably work was 30°, the comparison of other heat transfer characteristics was under 30°–90°orientations. Fig. 5 compared the variations of evaporator average temperature along with heat load under different placement angles. Obviously, for the ULHP with parallelogram configuration, the average temperature variation among different placed angles was small, the average difference
ð10Þ
Thermal resistance was one of the important and direct indices for measuring heat pipe heat transfer capability, the smaller the thermal resistance was, the higher efficiency of heat transfer was. The above discussion about start-up characteristic and operation temperature could be verified from thermal resistance too. Fig. 7 was the variation of thermal resistances. Thanks to the buffer effect of the ‘‘liquid pool”, the thermal resistance of Sample A was averagely 21% lower than that of Sample B. As the heat transfer characteristic of the ULHP with trapezoid configuration was affected more obviously by the placing angles, the difference of its thermal resistances between angles was larger. When placed vertically, the circulation was more smooth and the operation efficiency was higher due to the great assistance of the gravity, the thermal resistance of the samples, therefore, both declined sharply and the difference between them was much smaller. It was interesting to notice that there was an abrupt increase of thermal resistance at 30 W. Due to the serious heat leakage before start-up and the lower heat load, the temperature difference between Tcond and T ev ap: was relatively smaller, the thermal resistance was low at this time. When the heat load increased to 30 W, the ULHP system has already started up, the temperature difference increased while the heat leakage was relieved, the increase of thermal resistance was triggered as a result. The increase of thermal resistance at 30 W also showed some of regularity. Firstly, the extent of the abrupt increase decreased with the increase of the placed angle. It was very clear that the phenomenon didn’t appear when the two samples of ULHP were placed vertically. Secondly, the increase of thermal resistance was much more obvious for the trapezoid configuration that that for the parallelogram one. The regularity indicated that this phenomenon of abrupt change in thermal resistance was closely connected with the chan-
422
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
Fig. 6. The difference of the evaporator average temperature between the two ULHP samples under different angles.
Fig. 7. The variations of the thermal resistance along with the heat loads for the two ULHP samples under different angles.
nel configuration and placed angles. The ‘‘liquid pool” formed in the parallelogram configuration made contribution to reducing the thermal resistance by effectively suppressing the heat leakage caused by the vapor recoil.
3.4. Operation stability The biggest distinction between these two ULHP samples was the ‘‘liquid pool” structure that aimed at reducing the instability
of two phase flow, guaranteeing the stable one-way forward circulation and enhancing the heat transfer capability. T8 located at the entrance section of the evaporator where both affected by the condensed backflow liquid and the heat leakage and vapor recoil, was indeed a place with the most violent fluctuation and unsteady state. Comparing the fluctuation situation at T8 position under different angles gave further clear explication of the effect of ‘‘liquid pool” on the system operation. Fig. 8(a), (b) were the temperature fluctuation situations of the two samples respectively. The temperature fluctuation at T8 for Sample A was limited
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
423
Fig. 8. The variations of the temperature oscillation along with the angles for the two ULHP samples. (a) The ULHP with parallelogram evaporator configuration. (b) The ULHP with trapezoid evaporator configuration.
in ±6° while the range for Sample B was enlarged to ±12°, showing that the ‘‘liquid pool” in parallelogram configuration effectively suppressed the temperature oscillation by reducing the flow resistance for backflow liquid to enter the evaporator and accelerating the flow-out of vapor. Moreover, the temperature fluctuation of Sample A had better consistency under various angles than that of Sample B, namely, the ability to resist the gravity was superior, which also fitted with the results of the above discussion. What’s worthy noticing, the lowest fluctuation of Sample A occurred at 45° instead of vertical placing. Under vertical placing, the flow instability caused by the gravity and the friction of two-phase flow in the upstream was more prominent that led to a larger oscillation. As to Sample B, the fluctuation reached the lowest at 30° placement, which was relative to not only the above factors, but also the trapezoid configuration itself. Since the lower part of the evaporator of Sample B shaped like a divergent nozzle, when placed under the large orientations, the working fluid tended to flow to the entrance because of the height difference, and formed counterflow that increased the difficulty for the one-way forward
flow to be formed. The height difference of the evaporator’s right side reduced with the decreasing of the orientations, the counterflow and the flow resistance decreased as well, the backflow liquid supplied to the evaporator more easily, the fluctuation was eased consequently. 3.5. Model verification Fig. 9 showed the comparison validation of prediction values and the experiment results. Based on assumption 5, the system was under unsteady state when placed below critical angles or loaded less than 40 W heat load, the flow and heat transfer progress were too complicated to simplify with this model. Thus, only after the ULHP started up and formed stable flow, the steady state was calculated, namely, the calculation started from 30° and 40 W. Learned from the figure, the model fitted the experimental data well, the average error of was 5.7% while the largest one was 12%, indicating that the model basically depicted operation process and reflected the flow and heat transfer characteristic. The error
424
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
Fig. 9. The comparative results of the predicted values with the experimental data.
was larger for low heat load and low orientation situations. As placed under 30°, the error between the predicted values and the experiment data was larger than other cases, and the maximum error appeared at heat load 50 W. Relatively, the error reduced along with the increase of the orientations, the model fitted best for vertical situation. The cause of this situation was that the ULHP system started up slower under lower orientations, the working fluid hadn’t established steady slug flow, instead, the bubbles might be at local flooded, collapsed and combined, or growth state, that’s why the model error was larger in these cases. Adding the pattern development process would make the model more close to the physical phenomenon and more precise.
(3) The ‘‘liquid pool” structure played important role in easing the flow instability, the temperature fluctuation decreased, not only the heat transfer capability was enhanced, but also the damage to the battery was effectively reduced, BTMS could work more stably. (4) The built model well predicted the operation temperature which fitted well with the experimental data. Limited by the assumption, the error of prediction values was larger under low heat load. Adding the development of flow pattern under different heat loads could better match the practical physical phenomenon and increase the accuracy of the model, which would be one emphasis of our work in the future.
4. Conclusion The heat transfer characteristics and their difference of two ULHPs that had different channel configuration were studied with experiments, a mathematic model based on the operation process was also established, the exact affects of the channel configurations on the flow and heat transfer process were fully understood, which was beneficial for further guide in the design and optimization of the ULHP. The research under multiple orientations gave further clues for ULHPs to be applied to different operation conditions in BTMS. Relative conclusions were listed below. (1) Both of the two ULHP samples had critical angle, but the angles varied with the different configuration. The ULHP with parallelogram channel configuration started up steadily at 15° while the trapezoid one didn’t work until the angle increased to 30°. The parallelogram configuration helped ensure the one-way forward circulation, the response of start-up was obviously faster. (2) The capabilities in resisting the gravity of these two ULHP samples differed. The ULHP with parallelogram configuration showed superior capability in resisting the gravity with smaller difference in heat transfer characteristic between angles, the average and largest difference were 4% and 10% respectively, while those for the trapezoid one achieved 7.8% and 20% separately. The superiority of parallelogram configuration structure stood out under small angles.
Acknowledgements This work was supported by National Natural Science Foundation of China (Grant No. 51476059) and the International Cooperation and Exchange Program from the Ministry of Science and Technology of China (Grant No. 2013DFG60080). References [1] S.R. Alvani-Soltani, T.S. Ravigururajan, M. Rezac, in: Proceedings of IMECE, American Society of Mechanical Engineers, 2006, pp. 383–386. [2] K. Smith, C.Y. Wang, Power and thermal characterization of a lithium-ion battery pack for hybrid-electric vehicles, J. Power Sources 160 (2006) 662–673. [3] Y.F. Chen, J.W. Evans, Heat transfer phenomena in lithium/polymer-electrolyte batteries for electric vehicle application, J. Electrochem. Sources 140 (7) (1993) 1833–1838. [4] M.S. Wu, K.H. Liu, Y.Y. Wang, C.C. Wan, Heat dissipation design for lithium-ion batteries, J. Power Sources 109 (1) (2002) 160–166. [5] R. Sabbah, R. Kizilel, J.R. Selman, S. Al-Hallaj, Active (air-cooled) vs. passive (phase change material) thermal management of high power lithium-ion packs: limitation of temperature rise and uniformity of temperature distribution, J. Power Sources 182 (2008) 630–638. [6] Zhijun Tang, Qunzhi Zhu, Research on thermal management technology for power batteries, Chin. J. Power Sources 137 (1) (2013) 103–106. [7] Siddique A. Khateeb, Shabab Amiruddin, J. Robert Selman, Said Al-Hallaj, Thermal management of Li-ion battery with phase change material for electric scooters: experimental validation, J. Power Sources 142 (2005) 345–353. [8] R. Kizilel, A. Lateef, R. Sabbah, M.M. Faridb, J.R. Selmana, S. Al-Hallajet, Passive control of temperature excursion and uniformity in high-energy Li-ion battery
S. Hong et al. / International Journal of Heat and Mass Transfer 98 (2016) 415–425
[9]
[10] [11]
[12]
[13] [14]
[15]
[16]
packs at high current and ambient temperature, J. Power Sources 183 (2008) 370–375. X. Duan, G.F. Naterer, Heat transfer in phase change materials for thermal management of electric vehicle battery modules, Int. J. Heat Mass Transf. 53 (2010) 5176–5182. Y.C. Xiang, Z.Q. Hou, J.X. Zhang, The State-of-the-art of loop heat pipe, J. Eng. Thermophys. 25 (4) (2004) 682–684. J. Li, D. Wang, G.P. Peterson, Experimental studies on a high performance compact loop heat pipe with a square flat evaporator, Appl. Therm. Eng. 30 (2010) 741–752. S.C. Wu, J.C. Peng, S.R. Lai, et al., Investigation of the effect of heat leak in loop heat pipes with flat evaporator, IEEE Impact your future: Integration, Efficiency & the Eco-Friendly, 2009. K.J. Zan, C.J. Zan, Y.M. Chen, et al., Analysis of the parameters of the sintered loop heat pipe, Heat Transfer-Asian Res. 33 (8) (2004) 515–526. Xu. Jiyuan, Zou. Yong, Cheng. Lin, Pore structure optimization and properties of composite wicks for loop heat pipes, Proceedings of the CSEE 32 (23) (2012) 70–74. V.M. Kiseev, V.V. Vlassov, I. Muraoka, Experimental optimization of capillary structures for loop heat pipes and heat switches, Appl. Therm. Eng. 30 (2010) 1312–1319. Sihui Hong, Xinqiang Zhang, Shuangfeng Wang, Zhengguo Zhang, Experiment study on heat transfer capability of an innovative ultra-thin loop heat pipe, Int. J. Therm. Sci. 95 (2015) 106–114.
425
[17] Yu.F. Maydanik, M.A. Chernysheva, V.G. Pastukhov, Review: loop heat pipes with flat evaporators, Appl. Therm. Eng. 67 (2014) 294–307. [18] Alessandro Franco, Sauro Filippeschi, Closed loop two-phase thermosyphon of small dimensions: a review of the experimental results, Microgravity Sci. Technol. 24 (3) (2012) 165–179. [19] L.P. Grakovich, M.I. Rabetsky, L.L. Vasiliev, L.L. Vasiliev Jr., Polymer flat loop thermosyphons, 17th International Heat Pipe Conference (17th IHPC) Kanpur, India, October, 2013, pp. 13–17. [20] Rahmatollah Khodabandeh, Björn Palm, Influence of system pressure on the boiling heat transfer coefficient in a closed two-phase thermosyphon loop, Int. J. Therm. Sci. 41 (2002) 619–624. [21] S.S. Bertsch, E.A. Groll, S.V. Garimella, Refrigerant flow boiling heat transfer in parallel microchannels as a function of local vapor quality, Int. J. Heat Mass Transf. 51 (2008) 4775–4787. [22] Leonardo Carlos Ruspini, Christian Pablo Marcel, Alejandro Clausse, Twophase flow instabilities: a review, Int. J. Heat Mass Transf. 71 (2014) 521–548. [23] Sihui Hong, Xinqiang Zhang, Shuangfeng Wang, Zhengguo Zhang, Experimental Investigation on the Characters of Ultra-Thin Loop Heat Pipe Applied in BTMS, The 7th International Conference on Applied Energy (ICAE), 2015, pp. 3192–3200. [24] Sihui Hong, Xinqiang Zhang, Yongle Tang, Shuangfeng Wang, Zhengguo Zhang, Experiment research on the effect of the evaporator’s configuration design of an innovative ultra-thin looped heat pipe, Int. J. Heat Mass Transf. 92 (2016) 497–506.