Experimental Study of Composite Heat Pipe Radiator in Thermal Management of Electronic Components

Abstract

Conventional straight fin (SF) radiators have difficulties meeting the cooling requirements of high-power electronic components. Therefore, based on the structure and technology of the detachable fin radiator, this paper proposes a kind of radiator embedded in the heat pipe base and uses the roll-bond flat heat pipe (RBFHP) to replace the traditional fin. The radiator has the advantages of modularity, easy manufacturing, low cost and good heat balance. In this study, the heat pipes (HPs)-RBFHPs radiator was tested in natural convection and forced convection to mimic the actual application scenario and compared with the conventional aluminum radiator. Heating power, angle, wind speed and other aspects were studied. The results showed that the cooling performance of the HPs-RBFHPs radiator was improved by 10.7% to 55% compared with that of the SF radiator under different working conditions. The minimum total thermal resistance in the horizontal state was only 0.37 °C/W. The temperature equalization of the base played a dominant role in the performance of the radiator at a large angle, and the fin group could be ineffective when the angle was greater than 60°. Under the most economical conditions with an inclination of 0° and a wind speed of 2 m/s, the input power was 340 W, the heat source temperature of the HPs-RBFHPs was only 64.2 °C, and the heat dissipation performance was 55.4% higher than that of SFs.

1. Introduction

The heat flux of high-power electronic components, such as IGBT, exhibits a significant increase with integrated design [1], thereby exacerbating the issue of heat dissipation. Research has indicated that exceeding the manufacturer’s specified operating temperature limit for electronic components can lead to substantial degradation in product performance and reliability [2].

Finned radiators are widely used in thermal management due to their simple structure and high efficiency [3,4,5,6,7]. However, traditional finned radiators have high thermal diffusion resistance. As the size of heating components decreases, heat flux increases and the influence becomes more prominent [8]. The use of phase change materials (PCMs) [9] and heat pipes can effectively improve the performance of radiators. Phase change heat dissipation technology can effectively absorb or release heat through the change of the physical state of phase change materials, so as to achieve passive thermal management. A heat pipe is an efficient heat transfer device based on the principle of vapor–liquid phase change, which requires no external power and only relies on repeated phase transformation of the internal working medium to achieve rapid heat transfer. It is usually composed of a tube shell and a liquid suction core. Flat heat pipes (FHPs) or vapor chambers (VCs) have been widely used in the thermal management of electronic devices [10,11]. The combination of a heat pipe and fins is one of the common heat dissipation methods of electronic components.

The working principle of a VC is similar to that of a heat pipe. It is composed of a sealed cavity with a vacuum inside. The inner surface of a VC is usually covered with a porous liquid suction core structure [12,13]. A VC is one of the phase change heat transfer devices that has attracted much attention at present, and it has been used to cool electronic devices in combination with finned radiators [14,15]. Li et al. [16] have proposed an ultra-large aluminum-based flat heat pipe to solve the heat dissipation problem of high-power communication equipment. Under the conditions of natural air cooling and vertical placement, the heat dissipation module can effectively manage 150 W of input power with the junction temperature close to 80 °C, which is about 20 °C lower than with a traditional aluminum-based fin heat dissipation module of the same size. Rezk et al. [17] designed an integrated vapor chamber–straight fins (VCSF) heat sink using the bottom of the fin as the condensing end, and the results showed that VCSF’s reduction of the temperature of electronic components under all operating conditions compares to that of SFs. Berut et al. [18] manufactured and tested a vapor chamber with integrated hollow fins, determining a triangle as the optimal fin geometry to prevent liquid hold-up and flooding on the condensing end. Ji et al. [19] designed an integrated flat heat pipe (IFHP) that consists of an evaporator and a condenser with multiple channels fabricated in the fin heat sink and compared it with a conventional flat heat pipe (CFHP). The results showed that IFHP eliminates the contact heat set between heat pipes and heat sinks, expands the condensation area and reduces the temperature difference between the base and the end of the heat sink.

Although the use of a VC as a radiator base can greatly improve the performance of the radiator, its manufacturing process is more complex, and the cost is high. A lower-cost solution is the use of heat pipe-embedded heat sinks (HPeH) that, when properly designed, perform similarly to heat sinks with flat heat pipes, which can effectively reduce the thermal resistance of diffusion while having the advantages of low cost and simple process. Song et al. [20] embedded a heat pipe into the bottom plate of a traditional radiator, and the maximum temperature of a heat source was reduced by 15 °C compared to that of a conventional radiator. Wang et al. [21] studied the cooling performance of a heat pipe embedded in the bottom of a radiator with different arrangement and its influence on the temperature distribution of a CPU. The minimum thermal resistance of the heat pipe-embedded radiator studied was 0.15 °C/W, lower than the 0.22 °C/W of the reference conventional radiator. The overall performance of the H-type embedded radiator was also the best, with heat dissipation increased by 22.5%. At the same time, the weight was reduced by 30.1%, and the simulation showed that the use of a heat pipe-embedded radiator instead of a traditional radiator could reduce the power consumption of the fan by 66.2%.

Without changing the fin radiator structure, and further using phase change technology to strengthen the performance of the radiator, there is a solution to use flat heat pipes as fins to replace conventional aluminum fins. A kind of flat heat pipe called the roll-bond flat heat pipe (RBFHP), whose technology is used in refrigerator evaporators, is made of two aluminum plates through runner printing, rolling, expansion, liquid injection and liquid injection pipe welding [22], with low manufacturing cost, high reliability, and easy-to-obtain large area. Its thickness can be 1–5 mm. However, similar to the thermosiphon [23], there is no liquid suction core structure inside a RBFHP, and gravity has a great influence on it. Deng et al. [22] studied the thermal performance of RBFHPs with three typical structures (staggered, crossed and aligned) and four filling ratios (5%, 10%, 20% and 40%). The results showed that the internal boiling state was determined by the filling ratio and heating power, and the RBFHP with a staggered filling rate of 20% had the best performance. They [24] also conducted tests under the condition of multi-heat source natural convection, and the research shows that compared with the uniformly heated aluminum plate under a total heat load of 48 W, the total thermal resistance and temperature rise could be reduced by 15.3% and 10 °C, respectively. RBFHPs have excellent thermal performance, but it is rare to study it as a fin on the whole radiator system, and the comparison with conventional aluminum-fin radiators needs to be studied.

In this context, a comprehensive experimental study on the performance of a HPs-RBFHPs composite radiator designed for the cooling of high-power electronic devices was carried out. Phase change devices were introduced into the fin and base at the same time to strengthen the heat transfer capacity of the radiator. The heat transfer characteristics of the radiator were evaluated under a series of experimental conditions, and the factors affecting the performance of the radiator were revealed through careful comparison and analysis.

In subsequent sections, the experimental setup is described in detail (Section 2), the heat transfer characteristics of the radiator under various heating powers, inclination angles [25] and wind speeds are shown in comparison (Section 3), and the significance of future research is highlighted by summarizing the key conclusions (Section 4), which provide a reference for promoting the application of heat pipe radiators in high-power electronic devices.

2. Experimental Setup

2.1. Design and Manufacture of the Experimental Device

A high fin height would lead to a decrease in fin efficiency, and the fin ends would be unable to maintain a sufficient temperature difference with the surrounding air, resulting in a reduction in heat transfer. For ease of understanding, Figure 1 shows two kinds of fin cooling schematics. In Figure 1a, an excessively long fin would cause minimal heat exchange at the end, rendering that section of the fin ineffective. Without changing the height of the fins, by utilizing flat heat pipes instead of traditional metal fins, as depicted in Figure 1b, each part of the fin achieves a significant temperature difference environment, effectively enhancing temperature equalization and improving overall fin efficiency.

The assembly of the heat pipe composite radiator is shown in Figure 2a. The top surface of the 200 mm × 200 mm × 15 mm aluminum plate was machined with a 1.2 mm-wide and 5.5 mm-deep groove, spaced at a distance of 10 mm for accommodating RBFHPs (200 mm × 100 mm × 3.2 mm), while the bottom surface was processed to incorporate four inlay slots for heat pipe insertion. The Φ8 mm diameter and 160 mm length heat pipes were bent and inserted into the grooves, with their bottoms smoothed to minimize contact thermal resistance. Simultaneously, an SF radiator of identical specifications was constructed as a control group. Real images depicting both radiators are presented in Figure 2b.

Realistic images of the two test devices in this study are presented in Figure 3. They could effectively mimic environments characterized by natural convection and forced convective heat transfer, respectively. Figure 3a shows a device capable of adjusting the platform inclination within a range of 0° to 90°, while Figure 3b depicts a device that allows for the adjustment of wind speed. These experimental setups enabled the simulation of various working conditions encountered by electronic components, thereby facilitating a comprehensive understanding of the heat transfer properties exhibited by HPs-RBFHPs. The device labeled as “⑨” in Figure 3 consisted of a copper block inserted with three electric heating rods, serving as an input source for constant heat flux into the system. The contact area between the copper base and other components was 50 mm × 50 mm. To regulate input power, a DC Power Supplier was employed alongside power indicator readings to ensure accurate control over energy supply levels. To minimize contact thermal resistance, thermal grease was applied on the contact surface with a thermal conductivity value set at 3 W/m·K for optimal performance enhancement. Furthermore, black polyurethane foam was utilized as an effective thermal insulation material to mitigate heat loss from both the heated copper block and underside region of the radiator.

The data acquisition module included a data acquisition instrument (Hangzhou Mike paperless recorder (Hangzhou, China), model MIK-R6000C) and a K-type thermocouple (model TT-K-30-SLE, measurement error ± 0.3 K). The thermocouple was installed on the bottom of the base and the fin, and the distribution of measuring points is shown in Figure 4. T₁ measured the temperature of the heat source, T₂, T₃ and T₄ measured the temperature of the base and T₅–T₁₂ measured the temperature distribution on the fin.

Table 1. Heat pipe parameters.

Name	Material	Fluid	Wick	Liquid Filling Ratio
Heat pipe	Copper	Water	Sintered copper powder	30%
RBFHP	Aluminum	R1233ZD	/	20%

summarizes the main parameters of the blowout plate and the copper-water sintering heat pipe.

2.2. Data Processing and Uncertainty Analysis

To control the ambient temperature, the test was carried out in an air-conditioned room. During the whole experiment, the ambient temperature was maintained at 17 ± 1 °C. At the beginning of the experiment, the heat flux was adjusted by controlling the DC Power Supplier. Each time the temperature of the measuring point reached a steady state, the next step began. Steady state was defined in this experiment as a state in which the temperature changed less than ±0.5 °C over a period of 100 s. After one set of experiments, the system naturally cooled to room temperature before proceeding to the next experimental step. Due to the symmetrical characteristics of the radiator, the measuring point was arranged on one side of the radiator in this experiment to monitor the temperature change of the radiator.

The mean plane temperature at the bottom of the fin (${\overline{T}}_{f,b}$), the mean plane temperature at the top of the fin (${\overline{T}}_{f,t})$ and the temperature difference between the two ($\mathsf{\Delta }{T}_{f,b-t}$) were calculated by the following expressions:

$${\overline{T}}_{f,b}=\frac{T_5+T_7+T_9+T_{11}}{4}$$

(1)

$${\overline{T}}_{f,\mathrm{t}}=\frac{T_6+T_8+T_{10}+T_{12}}{4}$$

(2)

$$\mathsf{\Delta }{\overline{T}}_{f,b-t}={\overline{T}}_{f,b}-{\overline{T}}_{f,t}$$

(3)

The most direct way to evaluate the performance of the radiator is discussed by the temperature rise of the heat source. In addition, the quantitative evaluation of the heat dissipation capacity of the radiator can be analyzed and judged by calculating its thermal resistance value. In this paper, the heat dissipation capacity was measured by analyzing the total thermal resistance from the heat source to the space environment, that is, the total thermal resistance R_tot. The formula is as follows:

$$R_{tot}=\frac{\mathsf{\Delta }T}{P}=\frac{T_1-T_{air}}{P}$$

(4)

where T₁ is the average surface temperature of the heat source, T_air is the ambient temperature and P is the heating power of the heat source.

In addition, in order to better measure the dispersion degree of the temperature difference between the fins, the standard deviation (STDEV.P) between them was introduced, and its calculation formula is as follows:

$${\overline{T}}_{a,b}=\frac{T_a+T_b}{2}$$

(5)

$${\overline{T}}_f=\frac{{\overline{T}}_{5,6}+{\overline{T}}_{7,8}+{\overline{T}}_{9,10}+{\overline{T}}_{11,12}}{4}$$

(6)

$$STDEV.P=\sqrt{\frac{{({\overline{T}}_{5,6}-{\overline{T}}_f)}^2+{({\overline{T}}_{7,8}-{\overline{T}}_f)}^2+{({\overline{T}}_{9,10}-{\overline{T}}_f)}^2+{({\overline{T}}_{11,12}-{\overline{T}}_f)}^2}{4}}$$

(7)

In the process of experimental research, different experimental results have different uncertainties. The total uncertainty of variable X was calculated using the following formula:

$${\omega }_{\mathrm{x}}={\left\{{\Sigma }_{\mathrm{i}=1}^{\mathrm{n}}{(\frac{\partial X}{\partial x_i}{\omega }_{x_i})}^2\right\}}^{\frac{1}{2}}$$

(8)

where ${\omega }_x$ is the uncertainty of the variable X, ${\omega }_{x_i}$ is the uncertainty of the parameter $x_i$ and $\frac{\partial X}{\partial x_i}$ is the partial derivative of X with respect to $x_i$. The uncertainty of the experiment is listed in

Table 2. Uncertainty.

Quantities	Notation	Uncertainty
Temperature of thermocouples	T1–T12	0.3 K
Input heating power	Pin	1 W
Hot-wire anemometer	Vair	0.01 m/s
Dimension measurements	L, W, H	0.1 mm
Total Thermal resistance	Rtot	5.93%
STDEV.P	/	11.77%

. The uncertainty of the temperature measured by thermocouples at lower power was the main reason for the large error.

3. Results and Discussion

By using HPs-RBFHPs to cool high-power electronic components and equipment, several experiments were conducted to study the effects of different heat fluxes, different system gravity directions (0° to 90°) and different wind speeds (0 m/s to 4 m/s) on the whole system under forced convective heat transfer. The performance of an SF radiator was compared.

3.1. Radiator Performance in Horizontal State

In HPs-RBFHPs, heat transfer is a complex interplay of multiple processes that mutually influence each other. The heated copper block simultaneously conducts heat to both the aluminum base and the heat pipes, which in turn facilitate the transfer of heat to the remote end, effectively spreading it laterally across a flat surface. Subsequently, longitudinal heat conduction through the aluminum base enables thermal energy to be transferred to the RBFHPs. With gravity aiding its operation, passive phase transition cycles involving evaporation–boiling–condensing reflux occur within the 2-dimensional aeration loop of the RBFHP, resulting in efficient flattening of heat distribution within its fins. By analyzing temperature response patterns, one can gain insights into the thermal characteristics exhibited by this composite radiator.

As depicted in Figure 5, overall temperature variations observed in different parts of both radiators exhibited similar trends and generally showed an incremental rise with increasing heat flux levels. Under identical heating power conditions, over time, temperature curves experienced significant initial climbs followed by gradual stabilization tendencies. Notably, contrasting SFs performance-wise, HPs-RBFHPs demonstrated significantly smaller temperature differences among their respective components, indicative of superior temperature equalization capabilities.

Due to the heat pipe embedded at the bottom, the T₂ point was at the far end, but due to the role of the base-embedded heat pipe, its temperature was the highest with the exception of T₁ (heat source temperature). It can be considered that the heat of the composite radiator diffuses in a cross-like plane at the bottom of the base. SF mainly relies on heat conduction, the heat diffuses from the center to the four sides on the base. Its temperature equalization was poor, which led to the lowest T₂ temperature measured at the base and the largest temperature difference with the heat source. Specifically, at 220 W stability, the temperature difference between T₁ and T₂ reached 39.7 °C, while with the composite heat pipe radiator it was only 15.5 °C, which was only 39% that of SFs.

ΔT_f,b−t is the difference between the average temperature of the fin bottom and the top of the fin, which is used to observe the average temperature of the whole fin. Figure 5 shows that ΔT_f,b−t changes of the two radiators had great differences, and with the increase of heat flux, the difference in fin temperature average became more and more obvious. The ΔT_f,b−t of SF basically showed a step-up trend with the increase of heating power. At 220 W, the ΔT_f,b−t of SF reached about 7 °C, while with HPs-RBFHPs it was only about 2 °C. When starting from 60 W heating power, the ΔT_f,b−t of the two radiators showed a trend of rapid rise first and then stability. The temperature difference between the upper and lower ends of the fins of HPs-RBFHPs was 1.5 °C, while that of SF was 2 °C. Under the heat flux of 60 W, it can be considered that the RBTHP group did not start. The main reason for this slight difference was the difference in the average temperature of the base, and the secondary reason was the difference in the surface morphology of RBTHPS and the conventional aluminum fins. When P ≥ 100 W, the ΔT_f,b−t of HPs-RBFHPs basically fluctuated at about 2 °C, at which time it can be considered that the RBTHP group was fully activated.

The solid line in Figure 6 shows the comparison of heat source temperatures between HPs-RBFHPs and SF under variable power conditions. From a safety point of view, the junction temperature of electronic devices at work usually cannot exceed 85 °C. The results show that the maximum cooling power of HPs-RBFHPs (about 190 W) was 18.75% higher than that of SF (about 160 W) when the heat source temperature was 85 °C. The dashed line in Figure 6 shows the change of the total thermal resistance of the composite heat pipe radiator and SF under different heating power. With the increase of heating power, the overall variation trend of the total thermal resistance decreased first and then became stable. HPs-RBFHPs had the highest slope in the heating power range of 100 to 140 W because the fin group was fully activated in this power range, reducing the thermal resistance at the fin end. From 140 W onwards, the total thermal resistance of the two radiators began to decrease slowly. After 180 W, the total thermal resistance of the two radiators hardly changed. When the heating power increased from 60 to 220 W, the total thermal resistance of HPs-RBFHPs decreased from 0.49 to 0.37 °C/W, a reduction of 24%. The total thermal resistance of SF was reduced from 0.53 to 0.43 °C/W, a reduction of 19%. When the heating power reached 180 W, the difference between the total resistance of HPs-RBFHPs and that of SF was the largest, and the total thermal resistance of HPs-RBFHPs at this power was 16.3% lower than that of SFs.

3.2. The Effect of Different Gravity Directions on the System

In practical application, especially in electronic components that need to work in special conditions, such as navigation, aviation, photovoltaic systems, etc., electronic equipment needs to operate under inclined conditions. Therefore, several sets of experiments were designed to study the effects of different gravity directions on the system’s performance at intervals of 30° from 0° to 90°.

Figure 7 shows the startup process of each part of HPs-RBFHPs at angles of 0°, 30°, 60° and 90°. Figure 8 depicts the distribution of the working fluid inside the blown plate under different angles of gravity action and the internal flow diagram based on a reasonable assumption of the heat source. Due to the role of the heat pipe on the base, T₂ still maintained a high temperature and the heat pipe had a strong anti-gravity performance. It can be considered that the change of inclination angle had little influence on the temperature average of the heat pipe composite base.

ΔT_f,b−t is the difference between the average temperature of the rib base and the top of the rib that can be used to evaluate the operation of the blown fin. As shown in Figure 7a, when the heat flux was 100 W at 0°, ΔT_f,b−t reached a stable state, around 2 °C. As shown in Figure 7b, when the inclination was 30°, a step change was shown from 60 to 140 W, and a peak appeared at 180 W, with an evolution time of about 250 s. This was because, as shown in Figure 8, when the inclination was 30°, the edge of the liquid pool of RBFHP was close to the heat pipe in the center, and the heat load was insufficient at low heat flux. At 180 W, nuclear boiling took away a lot of heat, and the heat flux was basically the same from 180 W to 220 W. The results show that the blown plate group could activate normally at 30°inclination, but the heat flux required by ΔT_f,b−t was 40 W higher than that required in the horizontal state. As shown in Figure 7c, when the inclination was 60°, ΔT_f,b−t basically presented a step trend, but when the heat flux was greater than 140 W, ΔT_f,b−t had a large disturbance but the upward trend slowed down because the liquid pool could not operate stably due to constant heat flow. As shown in Figure 7d, when the inclination was 90°, ΔT_f,b−t showed basically the same performance as SF did under the horizontal state (Figure 5b). It can be considered that the inflating sheet group was not activated, and the temperature difference of the heat source was 8 °C when the temperature of the heat source was stable at 220 W, which was mainly caused by the temperature uniformity of the base.

Figure 9a reflects the relationship between the temperature of the heat source and the inclination angle under different powers. The temperature of the heat source did not change greatly with the increase of the inclination angle. In the horizontal state, HPs-RBFHPs had good performance. It can be seen from Figure 9b that the thermal resistance of the system was the highest at a lower power angle of 60°, but the thermal resistance decreased the most with the increase in power from 0.6 to 0.39 °C/W. At the same time, the thermal resistance in the horizontal state, which was only 0.12 °C/W, changed the least with power. In addition to the horizontal working state, the T₁ difference under the other three angles could be almost ignored, and it can be considered that the T₁ difference of the heat pipe radiator and SF under angled working conditions was mainly due to the temperature uniformity of the base. It can be concluded that the temperature uniformity of the base had the greatest effect on the heat source temperature in the case of large dip angles.

3.3. Effect of Forced Convection Heat Transfer on the System

The thermal characteristics of HPs-RBFHPs under forced convection conditions were investigated by comparing them with those of SF. Wind speed options included 1, 2, 3 and 4 m/s.

Figure 10 illustrates the comparison of heat source temperature between HPs-RBFHPs and SFs at different wind speeds and power levels. When subjected to a maximum heat load of 340 W in a forced convection environment, the heat source temperature of HPs-RBFHPs remained below the junction temperature threshold of 85 °C. Furthermore, an inverse relationship was observed between the heat source temperature of HPs-RBFHPs and wind speed; significant reduction in heat source temperature occurred when wind speed increased from 1 to 2 m/s. However, once the wind speed reached or exceeded 2 m/s, there was minimal change in the heat source temperature due to incomplete fin passage at lower speeds causing greater influence among fins. Additionally, at a wind speed of 1 m/s, there existed a maximum temperature difference of approximately 28.2 °C between two radiators and their respective heat sources; however, this difference decreased to a range between 23.0 and 25.5 °C after reaching a wind speed above or equal to 2 m/s with little variation thereafter. The increase in wind speed further contributed to reducing the heat source’s temperature; nevertheless, the impact on its overall performance became negligible beyond wind speeds exceeding 2 m/s as diffusion thermal resistance within the radiator base emerges as its primary limiting factor.

The parameter ΔT_f,b−t represents the temperature difference between the average temperature at the base and top of the rib, which serves as an indicator for evaluating the overall performance of the blown fin group. STDEV.P refers to the standard deviation of average temperatures measured across four groups of fins, providing a measure for assessing temperature dispersion among fins. STDEV.P is primarily influenced by two factors: firstly, variations in base temperature equalization leading to differential heat conduction among individual fins; secondly, mutual interactions between fins with greater impact observed closer to the center of the radiator. Increasing wind speed effectively mitigates this effect. These two sets of data collectively evaluate radiator performance, with smaller values indicating better performance.

Figure 11 and Figure 12 illustrate changes ΔT_f,b−t and STDEV.P over time under variable power conditions. With increasing heat flux, there was a stepwise increase trend observed in STDEV.P; however, compared to SFs (standard fins), HPs-RBFHPs (heat pipe-embedded radial basis function heat pipes) exhibited approximately 70% lower STDEV.P due to superior embedded temperature equalization on its base plate compared to that of SFs. As wind speed increased, there was a general decreasing trend observed in STDEV.P parameters along with weakened influence from fin interactions. When wind speed increased from 1 to 2 m/s, both HPs-RBFHPs and SFs experience significant decreases in STDEV.P—HPs-RBFHPs’ decreased by 45%, while SFs’ decreased by 34%. This further confirmed that at a wind speed of 1 m/s, airflow could not fully penetrate through the fin group, resulting in substantial impact on internal fins from both sides. When wind speed exceeded 2 m/s, although reductions in STDEV.P became less pronounced between these two cases, increased wind velocity still effectively enhanced uniformity in average temperatures among the fins. Figure 11 shows that the ΔT_f,b−t of HPs-RBFHPs stabilized in a range under relatively high power, similar to that under natural convection. When stable, it could be considered that the blown fin group was fully activated. However, as wind speed increased, the heat flux required to stabilize within a range increased. For example, at a wind speed of 1 m/s, ΔT_f,b−t stability required 140 W, and at a wind speed of 3 m/s, it required 220 W. In the case of high power, the ΔT_f,b−t of HPs-RBFHPs was basically stable between 1.5 and 2 °C, which reflected the good temperature equating performance of the fin group. The ΔT_f,b−t of SFs showed a step-up trend with the increase of heat flux, and the effect of wind speed was not significant, but with the increase of wind speed, it showed a change of decreasing first and then increasing. At low wind speed, the average temperature of the fin group could be improved, but because the heat was not transmitted to the upper end of the fin when the wind speed increased, it led to ΔT_f,b−t increasing. The stability of the two systems was better with wind speed increases, which is reflected in Figure 11 and Figure 12. The greater the wind speed, the smaller the fluctuation and the smoother the curve.

4. Conclusions

Under both horizontal natural convection and forced convection, the utilization of HPs-RBFHPs demonstrated a substantial reduction in temperature for electronic components compared to that with SFs. The heat dissipation efficiency was improved 10% to 55%. The performance of HPs-RBFHPs was significantly affected by the inclination angle, leading to noticeable failure in the fin group when inclined at 60° and complete failure when inclined at 90°. When subjected to high heating power and inclination angle below 30°, the temperature difference between the upper and lower ends of the fin group remained stable without significant changes as heat flux increased. In a forced convection environment, there was a notable improvement in average temperature among fins, while the temperature difference between their upper and lower ends remained unaffected. The minimum total thermal resistance in the horizontal state was only 0.37 °C/W. The maximum heat dissipation power of natural convection was 180 W. At an optimal wind speed of 2 m/s (optimal economic wind speed), air could effectively pass through the fin group; further increasing wind speed had a minimal impact on heat source temperature due to thermal resistance limitations imposed by base materials. Under the most economical conditions with an inclination of 0° and a wind speed of 2 m/s, the input power was 340 W, the heat source temperature of HPs-RBFHPs was only 64.2 °C, and the heat dissipation performance was 55.4% higher than that of SFs. These findings validate that HPs-RBFHPs effectively enhance fin efficiency and are suitable for managing thermal issues associated with high-power electronic devices.