Performance study of a valveless piezoelectric pump with built-in semi-arc bluffbody antique tower channel

When a fluid passes through a bluffbody, boundary layer separation occurs, which affects the flow properties. This vortex separation of the fluid flow through the passivate body and subjecting the passivate body to periodic hydrodynamic forces is known as bluffbody bypassing.¹ The effect of different structures of the bluffbody body on the flow bypass is not the same.² The cylindrical structure is simple, as a kind of flow-winding bluffbody body, its application is promising, and it is a good idea to apply cylindrical to the fluid device to improve the performance of fluid transport.³ For example, Hu et al.⁴ and Hou et al.⁵ used a hemispherical segmented group of fishtail-type bluffbody placed inside the piezoelectric pump cavity, respectively, to propose a piezoelectric pump without the use of motion valves and improve pumping performance. Hou et al. also designed a conical channel to further improve the output performance based on the bluffbody bypass.

Valveless piezoelectric pumps are driven by the inverse piezoelectric effect of the piezoelectric actuator, which converts electrical power into mechanical power movement of the piezoelectric actuator.^6,7 The volume of the pump chamber undergoes cyclical changes to pump the fluid.^8,9 The pump has some advantages, including a small structure size, low power cost, fast response time, high pumping accuracy, free from electromagnetic interference, low energy loss, and stable operation in the working process.^10,11 Mechanical lubrication,¹² biomedical study,¹³ precision instruments,¹⁴ agriculture, microelectronic systems,¹⁵ and other fields have a wide range of applications.¹⁶ Furthermore, the valveless piezoelectric pump structure is simple to manufacture at a low cost.

In recent years, based on the bluffbody bypass effect, more and more scholars have improved the valveless piezoelectric pump performance by changing the bluffbody structure as well as the shape of the channel.¹⁷ In 2020, Zhang et al.¹⁸designed a conical-head columnar bluffbody for the piezoelectric pump to transport fluid with a correlation arrangement to be arranged in a pump chamber. This structure of the bluffbody mitigated the reflux problem, and the output flow rate was found at 210 V and 49 Hz. In 2021, Yao et al.¹⁹ proposed a valveless piezoelectric pump in the Tesla valve structure. By designing this Tesla flow channel at both ends of the inlet and outlet, the backflow was effectively reduced, and the optimal output flow rate was measured to be 79.26 ml/min. In 2023, Wang et al.²⁰investigated a series of winged baffle-like bluffbody structures and improved the pump output performance by arranging the winged baffles in the output pipeline. Experiments showed that the NACA0015 winged baffle valveless pump had the best output performance, with a flow rate of 235.56 ml/min and a back pressure of 842.8 Pa. Zhang et al.²¹ designed an irregular bluffbody based on the condo effect, and this bluffbody was equivalent to a fluid deflector, which realized the bi-directional pumping characteristic according to the mounting position. Yu et al.²² designed a combination of a bluffbody and a conical channel for a valveless piezoelectric pump. The herringbone-shaped bluffbody was cleverly placed in the conical channel, and the structure effectively increased the flow rate due to the herringbone bluffbody mitigating reflux, with the best flow rate reaching 158.5 ml/min.

On the basis of previous research, this paper designs a semi-arc shape bluffbody to reduce the backflow and designs a tower-shaped channel to transport the fluid according to the shape of the ancient tower, improving the pumping efficiency. The effects of the bluffbody height and the tower-shaped channel angle on piezoelectric pump performance were investigated. The results show that the best output performance of the pump is 243.83 ml/min (220 V, 85 Hz) at 4 mm of bluffbody height and a tower channel angle of 20°. Section II describes the structural design and the fluid flow mechanism; Sec. III theoretically analyzes the causes of the ability of the pump to flow forward; Sec. IV simulates the fluid flow in the tower channel to verify the theoretical conclusions; Sec. V carries out experiments to obtain the optimal structure of the valveless pump; and Sec. VI concludes this paper.

The equipment and an exploded view of a valveless piezoelectric pump with a semi-arc bluffbody and a tower channel are shown in Figs. 1(a) and 1(b). Figure 1(c) shows the semi-arc bluffbody structure. In Fig. 1(d), the tower-shaped channel is then designed in the shape of a Chinese classical tower. The pump primarily consists of a piezoelectric actuator (vibrating in TE mode), a pump body, a semi-arc bluffbody, a tower channel, and a pump cover. Four bolts and nuts are used to connect the pump body to the pump cover. Grooves in the pump cover and pump body are designed to place the rubber seals. In order to attenuate the backflow of the pump, three groups of a semi-arc bluffbody with tower-shaped channels are designed in the chamber. The bluffbody has a rounded headwater surface and backside in the vertical rectangle. When a sinusoidal AC voltage is supplied to a piezoelectric actuator, the actuator vibrates cyclically back and forth.

The pump height is 10 mm, and the cavity depth is 4 mm. The inlet and outlet are both 3.5 mm. The bottom of the bluffbody is 2 × 2 mm² square, and the headwater surface of the bluffbody has a radius of 3 mm. The channel angle and bluffbody height are two variables in this paper, and the detailed geometric parameters are in Fig. 2.

Figure 3 shows the operating mechanism of the pump. The supply and pumping phases of the process are realized by reciprocating oscillations of the piezoelectric actuator, which cause periodic changes in both volume and pressure in the cavity. The positive direction is defined as the direction of flow from the inlet to the outlet, and the negative direction is from the outlet to the inlet. The pump in the supply phase is shown schematically in Fig. 3(a). When the piezoelectric actuator vibration causes the chamber volume to increase, the cavity draws in fluid through both inlet and outlet. Influenced by the semi-arc bluffbody and the tower channel, the flow resistance in the positive direction is much less negative; hence, the volume of fluid sucked in from the inlet is greater than that sucked in from the outlet. Figure 3(b) shows the schematic of the pump phase. When the piezoelectric actuator compresses the cavity, the pressure increases and excludes the fluid. As the vertical rectangle generates a greater flow resistance in the negative direction, the amount of liquid discharged from the inlet is smaller than that from the outlet. The piezoelectric actuator vibrates repeatedly and realizes the one-way flow of the pump, achieving the effect of fluid pumping.

As the flow resistance at the headwater surface and the backside of the bluffbody is different, so ξ_n2 ≠ ξ_n1, ξ_p2 ≠ ξ_p1, the cavity produces a different resistance to flow in the forward and reverse directions to achieve the pumping effect.

The pump cavity with a combination of the semi-arc bluffbody and the tower-shaped channel flow field simulation is established and analyzed by finite element software, and pressure differences in the forward and reverse directions were analyzed. The software used for finite element analysis is ANSYS Fluent module. Through the finite element analysis, the flow velocity and pressure distribution characteristics of the pump chamber can be obtained with a channel angle of 20° and a bluffbody height of 4 mm. Distilled water is selected as the fluid, the fluid velocity at the inlet is 0.1 m/s, the pressure is 100 Pa, and the conditions at the outlet are opening in 0 Pa. The piezoelectric pump used for finite element analysis has a piezoelectric actuator metal part with a diameter of 35 mm and a piezoelectric ceramic part with a diameter of 24 mm, with thicknesses of 0.5 and 0.3 mm, respectively.

Figure 4(a) shows the velocity change in the positive direction of the flow field. In the positive direction, the fluid mainly meets the resistance from the headwater surface of the semi-arc bluffbody, which generates a low velocity area in the front of the first bluffboody, flows to the sidewall of the tower-shaped channel, and then flows back to the center of the channel in a low velocity. As the fluid meets the next bluffbody where the channel is narrow, the velocity rises again. Around the third bluffbody, the channel is slightly wider than the previous constriction and generates the highest velocity in the cavity. At the backside of the third bluffbody, the velocity drops again, and the fluid converges at the outlet with a high velocity. Figure 4(b) shows the velocity change in the negative direction. When the fluid flows backward into the cavity, the fluid is mainly impeded by the backside in the vertical rectangle, which produces a low velocity area that bigger than it generates by the headwater surface of the semi-arc bluffbody in Fig. 4(a), and the tower-shaped channel slows the velocity with the suddenly enlarged space. As a result, the velocity at the outlet in the negative direction is slower than in the positive direction. At the part where the tower-shaped channel starts, a higher velocity is generated in the positive direction than in the negative direction, which also contributes to the velocity difference in the positive and negative directions.

Figure 4(c) shows the pressure distribution diagram of the cavity in the positive direction. When the fluid at the entrance knocks into the bluffbody, velocity drops while the pressure increases. At the back of the first bluffbody, relatively high pressure pushes the fluid out of the tower part of the channel, and the velocity increases at the constriction part after the first tower part of the channel. At the second tower part of the channel, an enlarged space raises the pressure and pushes the fluid to the third bluffbody with a relatively high velocity. After the third bluffbody, the pressure rises again and remains in the channel close to the outlet, pushing the fluid to flow out at a high velocity. Figure 4(d) shows the pressure distribution in the negative direction. The fluid flows into the cavity with a high pressure and high velocity, and the backside of the third bluffbody blocks the flow, causing a high pressure and low velocity at this area. Inside the cavity, the distribution of the pressure is the same as in Fig. 4(c), with a high pressure pushing the fluid out of the tower part of the cavity. As fluid flows out after the first bluffbody, the pressure inside the channel is much lower than it is in the positive direction, resulting in a flow with less pressure to drive, and the velocity of the fluid reduces. As a result, the negative velocity at the outlet was slower than the positive velocity, proving the existence of positive and negative velocity and pressure differences, as well as the pumping effect of the cavity.

The samples are shown in Fig. 5. To verify the characteristics of the pumps with unequal resistance in the positive and negative directions and to test the performance of the pumps, nine pumps with different geometrical parameters were fabricated in PLA using a 3D printer (Makerbot Replicator Mini+). The piezoelectric actuator consists of a single crystal 0.3 mm thick piezoelectric ceramics with a diameter of 24 mm that vibrates in TE mode and a brass base layer with a diameter of 35 mm and a thickness of 0.5 mm. The experimental equipment includes an AC signal generator (CUH, SDVC40-S, Nanjing, China), a pressure meter (Sanliang, DP390, Dongguan, China), and an electronic balance (Yueping, YP50001, Shanghai, China). Table I presents the detailed parameters of the nine different prototype pumps.

To investigate the resistance difference caused by the bluffbody and the tower-shaped channel in the positive and negative directions, in Fig. 6, the experimental equipment is set up. The unit consists of a beaker that applies fluid to the pump, a pipe, a stop valve, a lifting platform, an experimental beaker, an iron frame table, another beaker that collects the fluid flowing out from the pump, and a clock. Distilled water is selected as the fluid.

To demonstrate the unequal resistance, the beaker set on the lifting platform is filled with 800 ml fluid, leading the fluid to the pump via the horizontal pipe. Before the experiment, the stop valve on the pipe is closed, and the clock is cleared. Then start timing while opening the stop valve, and record the height of the fluid level in the beaker for every 2 s. Pipes were connected to the inlet and the outlet of the pump, and resistance tests were carried out in the forward and reverse directions, with four experiments in each direction. The effect of error was minimized by calculating the average of the experimental results. In Fig. 7, pump 5 is selected as the sample pump to conduct the experiment and the results. From the results, the fluid volume in the positive direction reduces faster than the fluid volume in the negative direction, and the flow rate in both directions slows down as time increases, because the lower pressure caused by the lower fluid level cannot promote a high flow rate as fluid flows into the pump. In addition, a 10.8% drop in fluid volume in the positive direction is achieved compared to the negative direction, proving the character of resistance difference for the cavity.

Figure 8(a) shows the experimental equipment. An AC signal generator, a beaker, a lifting platform, a bracket, a clock, and an electronic balance were used. The experimental prototypes are shown in Fig. 8(b), with the upper and lower prototypes demonstrating different tower channel angles. The experimental test voltage was set to 220 V, and the frequency varied from 40 to 100 Hz. By weighing the liquid collected from the outlet in 1 min, the flow rates of the pumps with different geometrical parameters were tested.

First, the bluffbody is set at 3 mm and varies the channel angle at 15°, 20°, and 25°. As the results show in Fig. 9, the pump with a channel angle of 20° has the optimal flow rate of 145.9 ml/min at 60 Hz, followed by the pump with a channel angle of 15° with the flow rate of 120.8 ml/min at 60 Hz. The channel angle at 25° has the lowest flow rate of 95.6 ml/min, a decrease of 34.5% compared with the a channel angle of 20°. The flow rate of the channel angle at 20° rises fast before 60 Hz and then drops to 0 ml/min at 85 Hz, and the highest flow rate is 20.8% higher than the channel angle at 15°. The pump with a channel angle of 15° has a slow flow rate increase and then drops fast after 60 Hz. The pump with a channel angle of 25° has a smooth trend of rising and falling in the flow rate, with a peak at 55 Hz. From the results, a big or small channel angle in a tower-shaped flow rate has a noticeable impact on the flow rate of the pump, as the space of the channel affects the pressure distribution in the cavity, and the velocity is also influenced.

Then, the channel angle is set at 20°, and the bluffbody height varies as 3, 4, and 5 mm. Figure 10 illustrates the experimental results. The pump with a bluffbody height of 5 mm has a flow rate rising of 11.8% compared to the pump with a bluffbody height of 3 mm, and the fluid keeps being pumped out until 100 Hz. The pump has the highest flow rate of 243.83 ml/min at 85 Hz (bluffbody height is 4 mm) and then drops obviously. The pump with bluffbody heights of 3 and 5 mm has a similar trend, but the pump with a bluffbody height of 4 mm has a flow rate increase of 67.1% compared to the pump with a bluffbody height of 3 mm. The rounded headwater surface and backside in the vertical rectangle of the bluffbody at different heights have a significant influence on the flow rate of the pump, because the resistance generated by the head water surface and vertical rectangle is different, and a smaller height of the bluffbody has a small effect on the flow rate. However, a high bluffbody increases the resistance in the positive direction, which also has a negative influence on the performance. From the comparison of 3 and 5 mm bluffbody heights, the higher the bluffbody, the greater the resistance to flow in the positive direction. In addition, the negative flow resistance is greater than the positive direction, the 5 mm flow is greater than the 3 mm, while the 4 mm bluffbody height has the best effect on the pump performance. Table II compares the output performance of the valveless piezoelectric pumps that have been studied so far.

A valveless piezoelectric pump with a semi-arc bluffbody and a tower-shaped channel is introduced in this paper. The theoretical analysis divides the flow rate into two parts of the positive and negative direcions. The combination of a semi-arc bluffbody and a tower channel produces unequal resistance to flow in the positive and negative directions will affect the fluid flow to achieve the pumping effect. The simulation results demonstrate that the pressure distribution within the cavity drives the fluid, resulting in different flow rates in different parts of the cavity. Experiments also demonstrated a 10.8% decrease in fluid volume in the forward direction compared to the reverse direction. The effect of the tower channel and the semi-arc bluffbody on the output flow rate of a valveless piezoelectric pump was investigated by performing flow tests on nine pumps with different geometrical parameters. The pump with a channel angle of 20, and a bluffbody height of 4 mm has the highest flow rate of 243.83 ml/min (220 V, 85 Hz).

This work was supported by the Education Department of Jilin Province (Grant No. JJKH20220678KJ) and the Opening Project of the Key Laboratory of CNC Equipment Reliability, Ministry of Education, Jilin University (Grant No. JLU-cncr-202206).

The authors have no conflicts to disclose.

Renhui Hu: Investigation (equal); Methodology (equal); Writing – original draft (equal); Writing – review & editing (equal). Chensheng Wang: Formal analysis (equal); Methodology (equal); Software (equal); Visualization (equal). Yi Hou: Data curation (equal); Formal analysis (equal); Resources (equal); Validation (equal). Dianbin Hu: Data curation (supporting); Project administration (lead); Resources (equal); Software (lead). Lipeng He: Funding acquisition (equal); Methodology (equal); Project administration (equal); Supervision (equal).

The data that support the findings of this study are available from the corresponding author upon reasonable request.