Performance of the single piezoelectric fan cooling system

To choose the best arrangement for the single piezoelectric fan cooling system, different orientations of the single piezoelectric fan cooling system were investigated.

The operating frequency was between 15 Hz and 40 Hz, while the operating voltage was between 50 V and 100 V. The simulation model was also used to examine the correlation between the flow field induced by the piezoelectric fan and the temperature drop on the inner fin surface. Additionally, the power consumption of the piezoelectric fan was investigated and the important dimensionless parameters were used to more efficiently analyze the experimental data.

5.1.1 The arrangement of the system

The arrangement of the piezoelectric fan may influence the generated flow field and the performance of the cooling system. To find the best arrangement of the piezoelectric fan, the vertical orientation (λ), the horizontal orientation ( Ê_¸¹º/Ë_ÌÍÎ), and the inclined angle of the cooling system (θ) were considered as shown in Fig.5-1.

35

According to previous studies [20], the gap between the fan and the heated surface strongly affects the performance of the cooling system. In this study, three different vertical orientations, where λ was equal to 5 mm, 10 mm, and 15 mm, were tested under the conditions of X /L =0.5 and θ=0°. As shown in Fig.5-2, the results indicate that the temperature drop on both the bottom and the side fin surfaces increases obviously when the piezoelectric fan approaches the bottom fin surface.

Thus, λ=5 mm was chosen as the vertical orientation. Tests of the inclined angle θ (at 0°, 15°, 30°, 45°, 60°, 75°, and 90°) were completed under the conditions of X /L =0.5 and λ=5 mm. The performance of the cooling system with θ=0° was superior to those of other inclined angles.

Fig.5-3 shows the dimensionless PZT-convection number ?M versus the horizontal orientation X /L under the condition of λ=5 mm and θ=0°. The experimental data demonstrates that the heat dissipation rate reaches its peak value between X /L =0.5 and X /L =0.6. Thus, the optimal arrangement of the piezoelectric fan embedded into a finned heat sink can be determined according to the experimental results. Further, these results also can be applied to the arrangement of the magnetic fans in the multiple-vibrating fan cooling system.

5.1.2 The flow field of the system

A three-dimensional model was built to investigate the flow field of the cooling system. Fig.5-4 shows that the maximum velocity near the bottom surface and the side surface both occur at approximately X /L = 0.3-0.5. Theoretically, a higher velocity leads to a larger temperature drop. Fig.5-2 shows good agreement with the anticipated phenomenon. The maximum temperature drop occurs at the position of maximum velocity. The three-dimensional simulation model may help us to decide the position of the heat source and the arrangement of the piezoelectric fan.

5.1.3 The power consumption of the system

In the fan cooling system, the power consumption of the PZT fan is also an important issue. Fig.5-5 shows the relationship between the power consumption and M for different vibrational frequencies of the piezoelectric fan. The power consumption is 0.511 W and M is 1.4 when the operating frequency is 24 Hz.

However, the power consumption decreases to 0.022 W while M increases to 1.53 at an operating frequency of 30 Hz, because 30 Hz is the resonant frequency of the fan, allowing the PZT fan to vibrate at its maximum amplitude [16]. According to this result, the power consumption should be considered first when selecting the operating frequency. An inappropriate operating frequency will increase the power consumption by a factor of twenty times.

5.1.4 The amplitude of the system

The fan amplitude plays an important role in the cooling ability of a piezoelectric fan. However, the operating frequency is also a factor that influences the cooling ability. Fig.5-6 [29] shows how M depends on the fan amplitude at different operating frequencies. In this figure, M increases as the amplitude and the operating frequency are increased. According to the figure, the influence of the amplitude on M is clearly larger than that of the operating frequency, especially at low fan amplitude operating conditions. The M can be increased from 1.05 to 1.3 at 25 Hz by increasing the amplitude from 4 mm to 9 mm. However, the M is only increased from 1.05 to 1.08 at a 5-mm amplitude by increasing the operating frequency from 25 Hz to 30 Hz.

Because a higher operating frequency only increases the cooling ability slightly when the fan amplitude is low, it is unnecessary to increase the operating frequency.

Thus, the resonant frequency of the piezoelectric fans was designed in a low range

37

between 20 Hz and 35 Hz.

5.1.5 The correlation between ÏÐ_ÑÒÓ and μ_ÑÒÓ of the system

According to Equation (3.8), the operating frequency and the fan amplitude are directly proportional to the Re . The two parameters can be easily observed, and they also play important roles in enhancing the Nu . However, the Nu cannot be calculated directly from the frequency and the fan amplitude, so the correlation between the Re and Nu may help in the estimation of the Nu conveniently. Fig.5-7 shows Nu and Re of the experimental data at different amplitudes and frequencies. When the Re is below 700, the Nu number increases almost linearly with the increasing Re number. Further, an approximate curve can be derived from these data. Thus, the Nu of the single piezoelectric fan cooling system can be inferred by its operating frequency and the fan amplitude.

5.1.6 The relationship between »_Ñ and Ri of the system

The dimensionless parameter is defined as Ri ?Gr/Re ^W) and represents the importance of natural convection relative to forced convection. When Ri<1, natural convection is negligible. However, when Ri>10, forced convection is negligible [45].

Fig.5-8 shows the dimensionless analysis of the experimental data at different frequencies and amplitudes. In Fig.5-8, two approximate curves are drawn according to the experimental data. M can be correlated with Ri and then generalized as the following equations. These equations mainly aid in judging which term should be considered seriously by a specific number. The equations may be used to analyze and assess the performance of the single piezoelectric fan cooling system more conveniently.