Contactless transportation systems based on near-field acoustic levitation have the benefit of compact design and easy control which are able to meet the cleanliness and precision demands required in precision manufacturing. However, the problems involved in contactless positioning and transporting cylindrical objects have not yet been addressed. This paper introduces a contactless transportation system for cylindrical objects based on grooved radiators. A groove on the concave surface of the radiator produces an asymmetrical pressure distribution which results in a thrusting force to drive the levitator horizontal movement. The pressure distribution between the levitator and the radiator is acquired by solving the Reynolds equation. The levitation and the thrusting forces are obtained by integrating the pressure and the pressure gradient over the concave surface, respectively. The predicted results of the levitation force agree well with experimental observations from the literature. Parameter studies show that the thrusting force increases and converges to a stable value as the groove depth increases. An optimal value for the groove arc length is found to maximize the thrusting force, and the thrusting force increases as the groove width, the radiator vibration amplitude, and the levitator weight increase.

Handling and transporting micro-electromechanical systems components or semiconductors is a challenging task due to their fragility and surface-sensitive characteristics.1 Therefore, contactless manipulation is necessary in many situations.2 Near-field acoustic levitation (NFAL) is a contactless technology that has attracted widespread attention in recent years. Compared to other contactless technologies, such as air cushion3 or magnetic levitation,4 the NFAL has two distinct advantages: (1) compact operation system and (2) no material restrictions for the levitator.5 In 1975, when Whymark6 investigated the ultrasonic levitation trapping small objects at the nodal positions of the standing wave field, he also found that a planar disk made of brass hovers above a piston operating at a vibration frequency of 20 kHz. This levitation phenomenon is referred to as NFAL. Hereafter, many researchers studied the mechanism and application of the NFAL. Hashimoto et al.7 presented a numerical model in which the radiation surface vibrates harmonically in the in-phase mode. Based on this model, they derived a relationship between the levitation distance, the vibration amplitude, and the weight per unit area. Later, Hu et al.8 pointed out that the sound field between the radiator and the levitator produces two forces which act on the bottom of the levitator. The first one is the acoustic radiation force that levitates the object. The second one is the acoustic viscous force that transports or rotates the object. These two forces play different roles in various applications. The acoustic radiation force is the main concerning force in contactless positioning or squeeze film bearing.9,10 In these applications, the acoustic viscous force is unwanted in order to guarantee stability of the levitator. However, in contactless transportation11 or ultrasonic motor12 applications, both forces are important.

The NFAL is mainly categorized based on the property of the vibration wave. Two major types of the NFAL are (1) the traveling wave type levitation and (2) the standing wave type levitation. Correspondingly, two methods are generally used to generate an acoustic viscous force in the NFAL.13 The first method uses a traveling wave sound field in the fluid between the vibrator and the levitator. This corresponds to the traveling wave type acoustic levitation. For example, Hashimoto et al.14 used two longitudinal vibration systems to excite the traveling wave along a beam. A 7.6 g duralumin plate was able to be transported along this beam. Based on this transportation system, other investigators studied its transportation characteristics15,16 or expanded its application field, e.g., for different cross-sections of the levitator.17 In order to acquire a better transportation performance, Li et al.18 introduced several engraved grooves on the surface of the levitator. An obvious characteristic of these traveling wave type-based transportation systems14–18 is that they require a guide rail to produce vibration. Therefore, Koyama et al.19 proposed a self-running sliding stage that produces a traveling wave vibration by itself. This stage not only produces self-levitation but also generates a thrusting force that produces forward motion.

The second method uses the asymmetric standing wave field to produce a gradient in the sound pressure field in the fluid domain. This corresponds to the standing wave type acoustic levitation. For instance, Hu et al.20 used a wedge-shaped stator to generate the standing wave vibration. The levitator can be driven to move in the horizontal direction due to the gradient in the stator vibration amplitude along the stator's length. Based on the structure from this study, Hu et al.21 developed a plate-shaped stator with triangular grooves to build a transportation system for long distance contactless transportation. These transportation systems20,21 with a single vibration source cannot ensure the consistent motion velocity of the levitator because the vibration distribution is not uniform. Wei et al.22 employed two longitudinal vibration sources to build a transportation system based on the standing wave type. These two vibration sources generate two non-parallel squeeze films with a pressure gradient between them. Similar to Ref. 19, Chen et al.23 reported a self-running and self-floating actuator based on the standing wave type. This actuator uses its coupled resonant vibration mode to produce an asymmetric standing wave field. Aono et al.24 proposed inserting the levitator between two opposing vibration sources to increase the levitator's holding force. This method can be used to acquire a high thrusting force for standing wave type transportation systems. Yet, this method decreases the load-carrying capacity of the transportation system since the two levitation forces oppose each other. Thus, this method is only suitable for light objects. However, the abovementioned transportation systems14–23 can only carry levitators with planar surfaces. Since some precision components, e.g., in microsystems, exhibit cylindrical surfaces, the problem of contactless transport and positioning remains.

Previous investigation25 regarding the NFAL stability problem revealed that an engraved groove on the levitator can break the symmetry of the sound field, which produces a restoring force. Therefore, this study deals with introducing a groove on the concave surface of the radiator and its stabilizing effect on the problem of contactless transport of cylindrical components. This modified radiator generates the levitation force and the thrusting force simultaneously. A transportation system composed of multiples of these modified radiators can levitate and move a cylindrical object. The pressure distribution between the radiator and the levitator is described by the non-linear Reynolds equation, which considers the horizontal movement of the levitator. The eight-point discrete method is used to solve the discontinuous film thickness problem, which arises due to the groove. The levitation force and the thrusting force are calculated by integrating the pressure and the pressure gradient, respectively. Numerical predictions of the levitation force agree well with experimental results from the literature.26 Finally, the effects of the groove parameters, the radiator vibration amplitude, and the levitator weight on the thrusting force are studied.

This paper is organized as follows. Section II describes the working principle and the transportation system design. Section III presents the governing equations and the solution process for the pressure distribution. Sections IV and V discuss the pressure distribution of the squeeze film and the parameter study, respectively.

In the NFAL, increasing the vibration amplitude enhances the load-carrying capacity of the levitation system.27 The Langevin transducer is suitable for heavy loads since it produces high intensity vibration.28 A typical Langevin transducer26 mainly consists of five parts: the bolt, the back mass, the four piezoelectric transducer (PZT) plates, the horn, and the radiator with the concave surface, as shown in Fig. 1(a). Owing to the piezoelectric effect, the four PZT plates will generate the same frequency vibration along the through-thickness direction when a sinusoidal voltage with frequency f is applied. The horn transmits and amplifies the vibration produced by the PZT plates to the radiator. Since the radiator with the concave surface vibrates continuously, the air film between the concave surface and the outer surface of the cylindrical object is squeezed. Thus, the levitation force is generated by the squeeze film.

FIG. 1.

(Color online) (a) Schematic diagram of the Langevin transducer (Ref. 26) used to produce the vibration, (b) the engraved groove on the radiator, which is used to produce the asymmetric pressure distribution in the axis direction, and (c) three-dimensional diagram of the location of the groove within the concave radiator.

FIG. 1.

(Color online) (a) Schematic diagram of the Langevin transducer (Ref. 26) used to produce the vibration, (b) the engraved groove on the radiator, which is used to produce the asymmetric pressure distribution in the axis direction, and (c) three-dimensional diagram of the location of the groove within the concave radiator.

Close modal

In the case of a smooth concave surface, there is a symmetric pressure distribution in the axial direction.29 Therefore, no axial transport velocity is generated and the levitator's hovering position remains unchanged. Symmetry of the pressure distribution is broken by introducing a groove on the concave surface,25 as shown in Fig. 1(b). The groove's geometric parameters are prescribed by its width lg, its depth hg, and its arc length θg. Figure 1(c) is presented to clearly depict the location of the groove within the radiator. According to the gas-film lubrication theory,30 the established pressure gradient in the axial direction produces thrusting force Ft. Furthermore, levitation force Fl caused by the squeeze effect counteracts the weight of levitator W. Therefore, this method not only realizes the levitator suspension, but also drives its horizontal movement without any additional equipment.

Using multiple Langevin transducers creates a contactless transportation system for cylindrical objects, as shown in Fig. 2. This system can meet the demands for long-distance transportation. The levitator's transportation trajectory can be controlled by distributing transducers, by adjusting the radiator vibration amplitude, and by the orientation of the grooved surface. Changing the position of the groove can, for example, be used to slow down or stop the levitator at the end of the transportation system, cf. Fig. 2.

FIG. 2.

(Color online) Schematic configuration of the contactless transportation system, which contains multiple Langevin transducers.

FIG. 2.

(Color online) Schematic configuration of the contactless transportation system, which contains multiple Langevin transducers.

Close modal

The analytic model for one radiator is shown in Fig. 3. The center point, inner radius, and width of the radiator are denoted as O, R, and L, respectively. The center point and outer radius of the levitator are defined as Ol and r, respectively. Since the air film between the levitator and the radiator is symmetric in the circumferential direction and no other force exists to drive a rotation of the levitator, the levitator rotation is not considered. Further, the aerodynamic effect is neglected and therefore, only the squeeze effect exists in the air film. Hence, both center points are located on the x axis. The eccentricity displacement e is defined as the distance between O and Ol. The difference between R and r is denoted as the nominal clearance c. Since the levitator is simultaneously supported by several radiators (cf. Fig. 2), the levitator's inclination in the xOz-plane is neglected. In addition, the levitator is assumed to be a rigid body and well balanced.

FIG. 3.

(Color online) Analytic model for one radiator and corresponding coordinate systems.

FIG. 3.

(Color online) Analytic model for one radiator and corresponding coordinate systems.

Close modal

The Reynolds equation is widely used to describe the pressure distribution of the squeeze film in the NFAL.9,31 Some assumptions are introduced here to deduce the Reynolds equations. First, air is assumed to behave in a compressible manner and is treated as Newtonian fluid.32 Second, since the film thickness is much smaller than the radiator's radius, the squeeze film is assumed to function an isothermal film33 and the pressure gradient in film thickness direction is neglected.15 Third, since the transportation speed is usually low, around several centimeters per second,20,34 the Reynolds number is lower than the upper limit value for laminar flow.33 Therefore, the airflow in the squeeze film is treated as laminar flow. Finally, the magnitude of fluid inertia and body forces are smaller than the viscous force32 and they are neglected. The axial displacement of the levitator relative to its initial position is denoted as u. Based on these assumptions, the expression of the dimensionless Reynolds equation considering the movement of the levitator in the z-direction is25,29

θ(PH3Pθ)+Z(PH3PZ)=σ(PH)T+Λ(PH)Zα(P2H3)Z,
(1)

in which P=p/pa and H=h/c are the dimensionless air film pressure and air film thickness, respectively. T=ωt=2πft is the dimensionless form of time t and the dimensionless width position is Z=z/R. pa denotes the pressure of the ambient air, h is the thickness of the air film, and the squeeze number is σ=12μaωR2/pac2. The coefficients Λ=6μau̇R/pac2 and α=ρau¨R/2pa represent the effects of the levitator movement speed u̇ and its acceleration u¨ on the pressure. μa and ρa are the viscosity coefficient and the density of air.

As shown in Fig. 3, the arc length of the concave surface of the radiator is defined as θa. Due to the continuity of the pressure at the interface to the ambient air, the pressure on the boundary of the squeeze film meets the following conditions:

P(θ,z=±L/2)=1,
(2)
P(θ=π/2±θa/2,z)=1.
(3)

Since the radial displacement of the radiator can be treated as uniform,26 the normalized mode shape V(θ,z) equals 1. It is assumed that the existence of the groove does not affect the radiator's vibration amplitude distribution in order to conveniently conduct the parameter study. This assumption is reasonable if the dimension of the groove is much smaller than that of the radiator. Referring to Figs. 1(b) and 3, the grooved domain includes the front (z=L/2), the rear (z=L/2lg), the right (θ=π/2+θg/2), and the left (θ=π/2θg/2) boundaries. Thus, the squeeze film thickness is expressed as

h={cesin(θ)+ξV(θ,z)sin(T)+hg(groovedomain),cesin(θ)+ξV(θ,z)sin(T)(remainingdomain),
(4)

where ξ is the maximum value of the radial vibration displacement.

The computational domain is the area of the radiator's concave surface, as shown on the left in Fig. 4. This domain is divided into a grid of m×n sections, in which m defines the discretization in θ- and n in z-directions, i.e., Δθ=θa/m and Δz=L/n. The film thickness becomes discontinuous at the edges of the groove domain. Therefore, the partial differentials (H/θ)i,j and (H/Z)i,j in Eq. (1) are unsolvable.

FIG. 4.

(Color online) The mesh system of the computational domain, where the detailed eight-point discrete grid is presented on the right side.

FIG. 4.

(Color online) The mesh system of the computational domain, where the detailed eight-point discrete grid is presented on the right side.

Close modal

In the field of gas lubrication, the eight-point discrete method is adopted to solve this discontinuous film thickness problem.35,36 All nodes on the four boundary lines (i.e., z=±L/2 and θ=π/2±θa/2) meet the boundary conditions in Eqs. (2) and (3). The inside node Gi,j is surrounded by the area Ωij, which is spanned by the four red dashed lines Γij,14, as seen on the right in Fig. 4. The adjacent eight points Nij,18 in the eight-point discrete grid are evenly distributed on the four perimeter lines Γij,14.

By integrating Eq. (1) over Ωij, the dimensionless rate of airflow passing through Ωij can be expressed as

Ωij{θ(PH3Pθ)+Z[PH3PZ]}dθdZ=Ωij[σ(PH)T]dθdZ+Ωij[Λ(PH)Zα(P2H3)Z]dθdZ.
(5)

The surface integral on the left side and the second term on the right side of Eq. (5) can be transformed to a closed line integral by using Green's theorem,

Γij[(PH3PZ)dθ+(PH3Pθ)dZ]=Ωij[σ(PH)T]dθdZ+Γij[α(P2H3)Λ(PH)]dθ.
(6)

The path of line integration runs counterclockwise, which means from Γij,1 to Γij,4. Since the projections of Γij,2 and Γij,4 on the θ-axis equal zero, the first term on the left side of Eq. (6) reduces to

Γij(PH3PZ)dθ=Γij,1(PH3PZ)dθΓij,3(PH3PZ)dθ.
(7)

The film thickness of each perimeter line is approximately equal to the arithmetic average film thickness of the two corresponding points which lie on the perimeter line35 such as H3(Γij,1)=(Hi1/4,j+1/23+Hi+1/4,j+1/23)/2. Subsequently, using the center difference method and the trapezoidal rule, the right side of Eq. (7) can be expressed as

Γij,1(PH3PZ)dθΓij,3(PH3PZ)dθ=(Pi,j+Pi,j+1)(Pi,j+1Pi,j)(Hi1/4,j+1/23+Hi+1/4,j+1/23)Δθ4ΔZ(Pi,j+Pi,j1)(Pi,jPi,j1)(Hi1/4,j1/23+Hi+1/4,j1/23)Δθ4ΔZ.
(8)

All terms in Eq. (6) can be expanded in the same form as Eq. (8). Thus, the nonlinear differential equation that describes the relation between the pressure and the film thickness over time is stated as

f(Pi1,j,Pi,j,Pi+1,j,Pi,j1,Pi,j+1,H,T)=0.
(9)

The Newton–Raphson method is used to linearize Eq. (9),37 which acquires the pressure distribution in the computational domain. The mean levitation force F¯l and the mean thrusting force F¯t in one vibration period follow as11 

[F¯lF¯t]=12π02ππ/2θa/2π/2+θa/2L/2L/2[(ppa)sin(θ)h2pz+μau̇h+ρahu¨3]dzdθdT.
(10)

Under the steady conditions, the mean levitation force counterbalances the weight of the levitator, i.e., F¯l=W. The mean thrusting force is mainly influenced by four variables: film thickness h, pressure gradient in z-direction p/z, levitator movement speed u̇, and its acceleration u¨.

A basic design of this radiator is used to quantify the effect of the groove on the pressure distribution. The parameters for the basic design and its operating condition are listed in Table I. In addition to this basic setup, transportation velocity u̇ and acceleration u¨ are assumed to be zero to conveniently analyze the effect of the groove parameters on the thrusting force by using control variates. The influence of the mesh size on the numerical results has been checked in previous studies,25,29 which is not displayed here for the sake of brevity. A mesh resolution of m×n=60×40 has been identified for sufficient results with respect to calculation time and convergence. All parameters remain as listed in Table I, if not specified elsewhere throughout the following studies.

TABLE I.

Parameters of the radiator's basic design and its operating condition.

ParameterValueParameterValue
Vibration amplitude ξ 9 μVibration frequency f 20 kHz 
Radiator concave radius R 10 mm Levitator outer radius r 9.96 mm 
Nominal clearance c 40 μRadiator width L 20 mm 
Radiator arc length θa 120° Groove width lg 3 mm 
Groove depth hg 1 mm Groove arc length θg 80° 
Air dynamic viscosity μa 1.81 × 10−5 Pa s Air pressure pa 1.013 × 105 Pa 
Air density ρa 1.204 kg/m3 Weight of the levitator W 0.5 N 
ParameterValueParameterValue
Vibration amplitude ξ 9 μVibration frequency f 20 kHz 
Radiator concave radius R 10 mm Levitator outer radius r 9.96 mm 
Nominal clearance c 40 μRadiator width L 20 mm 
Radiator arc length θa 120° Groove width lg 3 mm 
Groove depth hg 1 mm Groove arc length θg 80° 
Air dynamic viscosity μa 1.81 × 10−5 Pa s Air pressure pa 1.013 × 105 Pa 
Air density ρa 1.204 kg/m3 Weight of the levitator W 0.5 N 

Since the radiator vibrates continuously, the film thickness and pressure vary over time. When the squeeze system reaches a steady condition, the transient film thickness and pressure are periodic functions in time with vibration frequency f. The dimensionless pressure distributions of the radiator's basic configurations with a smooth and a grooved surfaces at the onset time of one steady period are shown in Fig. 5(a). The corresponding distributions of the dimensionless pressure gradient are displayed in Fig. 5(b).

FIG. 5.

(Color online) Comparison of the dimensionless pressure (a) and pressure gradient (b) for configurations with a smooth and a grooved surfaces at the onset of one steady period, computed for W=0.5 N and ξ=9 μm.

FIG. 5.

(Color online) Comparison of the dimensionless pressure (a) and pressure gradient (b) for configurations with a smooth and a grooved surfaces at the onset of one steady period, computed for W=0.5 N and ξ=9 μm.

Close modal

For the radiator configuration with a smooth surface, the pressure distribution is symmetric with respect to the xOz- and the xOy-planes, see top of Fig. 5(a). Therefore, only levitation force Fl is produced and no thrusting force acts on the levitator. For the grooved configuration, corresponding to the bottom of Fig. 5(a), the pressure value in the groove domain equals the ambient pressure, i.e., P=1. This results from the contact between groove domain and ambient air such that the squeeze effect is not evidenced in the groove domain. Hence, symmetry of the pressure distribution about the xOy-plane is broken due to the existence of the groove. The asymmetric pressure distribution in the z-direction produces thrusting force Ft, and the available domain with squeeze effect is smaller than for the smooth surface configuration. Thus, given the same weight of the levitator, the dimensionless maximum pressure value in the grooved condition (Pmax=1.1443) is higher than that in the smooth condition (Pmax=1.1333).

Figure 5(b) shows that the pressure gradient value is antimetric in the z-direction for the smooth surface configuration. Since the positive part and the negative part are of equal magnitude [Σ(P/Z)=0], the thrusting force equals zero, which corresponds to the results presented in Fig. 5(a). However, the balance between the positive part and the negative part is changed by the groove. The pressure gradient value in the groove domain equals zero. The negative part dominates in this case [Σ(P/Z)=0.3813], which means that the thrusting force points toward the negative z-direction.

The variation in thrusting force Ft in one steady period is shown in Fig. 6. The value of the thrusting force alternates between negative and positive. The maximum negative thrusting force absolute value is 1.6355 mN, which is larger than the positive absolute value 1.3341 mN. Therefore, a mean thrusting force of F¯tF¯t = –0.1182 mN is generated, since negative contributions for the thrusting force dominate.

FIG. 6.

Temporal variation of the thrusting force in one steady period, where the dashed line represents the mean thrusting force of one period, computed for W=0.5 N, ξ=9 μm, hg=1 mm, θg=80°, and lg=3 mm.

FIG. 6.

Temporal variation of the thrusting force in one steady period, where the dashed line represents the mean thrusting force of one period, computed for W=0.5 N, ξ=9 μm, hg=1 mm, θg=80°, and lg=3 mm.

Close modal

The measurement method of the levitation force stated in Ref. 26 is represented by the simplified diagram shown in Fig. 7(a). When the levitation system reaches a steady condition, the levitation force generated by the squeeze film is counteracting the external load which contains the weight of the levitator and two objects. Therefore, the value of the levitation force is adjusted by changing the weight of the two objects. At the same time, the levitation height is measured by the displacement sensor. Consequently, experimental data which describes the relationship between the levitation height and the levitation force is obtained. In order to validate the proposed calculation method, numerical calculation results are compared to with these experimental data from Ref. 26. Some calculation parameters are adjusted compared to the basic configuration. The adjusted parameters are listed in Table II. All other parameters remain as prescribed for the basic configuration.

FIG. 7.

(Color online) (a) Measuring principle of levitation force stated in Ref. 26 and (b) comparison between numerical and the experimental results from the literature (Ref. 26), where the solid line represents the numerical results, and the dots are the published experimental data.

FIG. 7.

(Color online) (a) Measuring principle of levitation force stated in Ref. 26 and (b) comparison between numerical and the experimental results from the literature (Ref. 26), where the solid line represents the numerical results, and the dots are the published experimental data.

Close modal
TABLE II.

Adjusted parameters of the radiator, according to Ref. 26.

ParameterValueParameterValue
Vibration amplitude ξ 10.8 μm Vibration frequency f 16.11 kHz 
Radiator concave radius R 10.015 mm Levitator outer radius r 9.985 mm 
Nominal clearance c 30 μm Radiator arc length θa 110° 
Total mesh grid number m×n 55×40 Radiator width L 20 mm 
ParameterValueParameterValue
Vibration amplitude ξ 10.8 μm Vibration frequency f 16.11 kHz 
Radiator concave radius R 10.015 mm Levitator outer radius r 9.985 mm 
Nominal clearance c 30 μm Radiator arc length θa 110° 
Total mesh grid number m×n 55×40 Radiator width L 20 mm 

Figure 7(b) shows the comparison between numerical calculation and experimental results in terms of the levitation force. The levitation force decreases with increasing levitation height ce. This is due to the weakening of the squeeze effect for higher levitation heights.7 The calculated results match well with the experimental data.

For a given radiator, the thrusting force is mainly determined by four variables as mentioned in Sec. III C. The first two variables, i.e., the film thickness and the pressure gradient, are mainly affected by the groove parameters, the radiator vibration amplitude, and the levitator weight. The other two variables, i.e., levitator movement velocity and its acceleration, are assumed to be zero as mentioned in Sec. IV. In the following parametric study, only the absolute values of the forces are presented, since the directions of the thrusting forces are the same.

According to Li et al.,38 the groove depth has an obvious influence on the squeeze effect. In order to determine the effect of the groove depth on the thrusting force, a configuration with three different radiator vibration amplitudes (ξ=8, 9, and 10 μm) is analyzed for varying groove depths. Figure 8 displays the variation of the thrusting force with the groove depth for all three vibration amplitude configurations. The thrusting force obviously increases when the groove depth is lower than 0.9 mm. This phenomenon occurs because the magnitude of the groove depth approximates the magnitude of the nominal clearance in low groove depth. Thus, the squeeze effect still exists in the groove domain, which in turn leads to a deviation in the pressure in the groove domain from the ambient pressure. The squeeze effect weakens with increasing groove depth. Finally, the thrusting force converges to a constant value when the groove depth exceeds 0.9 mm. Therefore, the groove depth is set to 1 mm in this study to acquire a high thrusting force. This trend is visible for different amplitudes. The relationship between the thrusting force and the amplitude will be discussed in Sec. V D.

FIG. 8.

(Color online) Thrusting force versus groove depth at three vibration amplitudes, computed for W=0.5 N, θg=80°, and lg=3 mm.

FIG. 8.

(Color online) Thrusting force versus groove depth at three vibration amplitudes, computed for W=0.5 N, θg=80°, and lg=3 mm.

Close modal

Figure 9 presents the relation between thrusting force and groove arc length for three different vibration amplitudes, namely, ξ= 8, 9, and 10 μm. The thrusting force initially increases and then decreases as the groove arc length increases. For a groove arc length of θg=0°, the concave surface is a smooth surface. In this case, the thrusting force equals zero as discussed in Sec. IV. For a groove arc length of θg=120°, the concave surface becomes a new smooth surface with the width Llg. This new surface is also symmetric in the z-direction. Thus, the thrusting force is also zero. As the groove arc length increases, the available domain with the squeeze effect decreases. At a constant levitation force, the maximum pressure value and the pressure gradient increase while the film thickness decreases. According to Eq. (10), the thrusting force is simultaneously affected by the pressure gradient and the film thickness. Therefore, there is an optimum value for the groove arc length. Since the high-pressure air is located in the area near the middle line θ=90° in the θ-direction as shown in Fig. 5(a), the influence of the groove on the pressure gradient and the film thickness attenuates as the groove arc length increases. Thus, the variation of the thrusting force is steeper for small groove arc lengths (θg<40°) than for large groove arc lengths (θg>40°). The optimal groove arc length falls on the low-value side and is nearly 40° for the given parameters and operating condition.

FIG. 9.

(Color online) Thrusting force versus groove arc length at three vibration amplitudes, computed for W = 0.5 N, hg = 1 mm, and lg = 3 mm.

FIG. 9.

(Color online) Thrusting force versus groove arc length at three vibration amplitudes, computed for W = 0.5 N, hg = 1 mm, and lg = 3 mm.

Close modal

Wθghg In this subsection, the groove arc length is set to θg= 40°. Figure 10 shows the relationship between thrusting force and groove width for three different vibration amplitudes (ξ= 8, 9, and 10 μm). The thrusting force increases with increasing groove width. This result agrees with the numerical and experimental results presented by Liu et al.,25 who used an engraved groove levitator to replace the levitator's misalignment. However, the available domain with the squeeze effect decreases if the groove width increases. Thus, the levitator cannot be suspended when the groove width is too great. Consequently, the study of the thrusting force is only meaningful within a small range of the groove width.

FIG. 10.

(Color online) Relationship of the thrusting force and groove width at three vibration amplitudes, computed for W = 0.5 N, θg = 40°, and hg = 1 mm.

FIG. 10.

(Color online) Relationship of the thrusting force and groove width at three vibration amplitudes, computed for W = 0.5 N, θg = 40°, and hg = 1 mm.

Close modal

The change in vibration amplitude is assumed to stem from changes in the driving voltage,39 while the vibration frequency remains unchanged in order to keep the same vibration mode shape of the radiator. In this subsection, the groove arc length is set to θg= 40°. Figure 11 depicts the thrusting force as a function of the vibration amplitude at four different weights of the levitator, namely, W = 0.4, 0.5, 0.6, and 0.7 N. The thrusting force increases as the weight increases, which is consistent with the results of Refs. 13 and 25. A higher levitator weight indicates a smaller levitation height at a constant vibration amplitude. According to Li et al.,16 the thrusting force increases under these conditions. Additionally, the thrusting force follows a positive linear trend with increasing vibration amplitude. This is due to the fact that the squeeze effect is enhanced by increasing vibration amplitude.25,40

FIG. 11.

(Color online) Thrusting force as a function of the vibration amplitude at four different levitator weights, computed for hg = 1 mm, θg = 40°, and lg = 3 mm.

FIG. 11.

(Color online) Thrusting force as a function of the vibration amplitude at four different levitator weights, computed for hg = 1 mm, θg = 40°, and lg = 3 mm.

Close modal

This paper introduced the design of a contactless transportation system for cylindrical objects. The levitation and transport are based on near-field acoustic levitation. The thrusting force for the transported items is generated by the radiator that has an engraved groove. Theoretical analysis of the influence of the groove on the pressure distribution is also presented. The nonlinear Reynolds equation which describes the relationship between the pressure and the motion of the levitator has been solved by combining the eight-point discrete method and the Newton–Raphson method. The levitation force and the thrusting force are calculated from the solved pressure distribution. The calculated levitation force agrees well with similar results in the literature.26 Moreover, the effects of groove depth, groove arc length, and groove width on the thrusting force have been investigated to determine suitable groove parameters. The predicted results yield the following conclusions:

  1. The thrusting force initially increases and then remains at a stable value as the groove depth increases.

  2. There is an optimal groove arc length to maximize the thrusting force.

  3. Increasing the groove width can improve the transportation performance within a small range of groove width.

  4. Increasing the vibration amplitude of the radiator can increase the thrusting force.

  5. A higher levitator weight can lead to higher thrusting forces.

Future work addresses the manufacture of a prototype for this transportation system. Furthermore, precisely controlling the motion trajectory of the levitator by adding a control algorithm is planned. This paper and our continuing research supply a helpful guide for designing a practical contactless transportation system for cylindrical objects in industrial applications.

The first author acknowledges the support provided by the China Scholarship Council (CSC) (Grant No. 201808340068).

1.
G.
Reinhart
and
J.
Hoeppner
, “
Non-contact handling using high-intensity ultrasonics
,”
CIRP Ann.
49
(
1
),
5
8
(
2000
).
2.
P.
Liu
,
J.
Li
,
H.
Ding
, and
W.
Cao
, “
Modeling and experimental study on near-field acoustic levitation by flexural mode
,”
IEEE Trans. Ultrason. Ferroelectr. Freq. Control
56
(
12
),
2679
2685
(
2009
).
3.
F.
Erzincanli
,
J.
Sharp
, and
S.
Erhal
, “
Design and operational considerations of a non-contact robotic handling system for non-rigid materials
,”
Int. J. Mach. Tool. Manuf.
38
(
4
),
353
361
(
1998
).
4.
O.-S.
Kim
,
S.-H.
Lee
, and
D.-C.
Han
, “
Positioning performance and straightness error compensation of the magnetic levitation stage supported by the linear magnetic bearing
,”
IEEE. Trans. Ind. Electron
50
(
2
),
374
378
(
2003
).
5.
V.
Vandaele
,
P.
Lambert
, and
A.
Delchambre
, “
Non-contact handling in microassembly: Acoustical levitation
,”
Precis. Eng.
29
(
4
),
491
505
(
2005
).
6.
R.
Whymark
, “
Acoustic field positioning for containerless processing
,”
Ultrasonics
13
(
6
),
251
261
(
1975
).
7.
Y.
Hashimoto
,
Y.
Koike
, and
S.
Ueha
, “
Near-field acoustic levitation of planar specimens using flexural vibration
,”
J. Acoust. Soc. Am.
100
(
4
),
2057
2061
(
1996
).
8.
J.
Hu
,
K.
Nakamura
, and
S.
Ueha
, “
An analysis of a noncontact ultrasonic motor with an ultrasonically levitated rotor
,”
Ultrasonics
35
(
6
),
459
467
(
1997
).
9.
T.
Stolarski
and
W.
Chai
, “
Load-carrying capacity generation in squeeze film action
,”
Int. J. Mech. Sci.
48
(
7
),
736
741
(
2006
).
10.
D.
Ha
,
T.
Stolarski
, and
S.
Yoshimoto
, “
An aerodynamic bearing with adjustable geometry and self-lifting capacity. Part 1: Self-lift capacity by squeeze film
,”
Proc. Inst. Mech. Eng. J. Part J
219
(
1
),
33
39
(
2005
).
11.
K.
Feng
,
Y.
Liu
, and
M.
Cheng
, “
Numerical analysis of the transportation characteristics of a self-running sliding stage based on near-field acoustic levitation
,”
J. Acoust. Soc. Am.
138
(
6
),
3723
3732
(
2015
).
12.
Z.
Dong
,
M.
Yang
,
Z.
Chen
,
L.
Xu
,
F.
Meng
, and
W.
Ou
, “
Design and performance analysis of a rotary traveling wave ultrasonic motor with double vibrators
,”
Ultrasonics
71
,
134
141
(
2016
).
13.
J.
Hu
,
K.
Nakamura
, and
S.
Ueha
, “
Stability analysis of an acoustically levitated disk
,”
IEEE Trans. Ultrason. Ferroelectr. Freq. Control
50
(
2
),
117
127
(
2003
).
14.
Y.
Hashimoto
,
Y.
Koike
, and
S.
Ueha
, “
Transporting objects without contact using flexural traveling waves
,”
J. Acoust. Soc. Am.
103
(
6
),
3230
3233
(
1998
).
15.
A.
Minikes
and
I.
Bucher
, “
Noncontacting lateral transportation using gas squeeze film generated by flexural traveling waves—Numerical analysis
,”
J. Acoust. Soc. Am.
113
(
5
),
2464
2473
(
2003
).
16.
J.
Li
,
W.
Cao
, and
W.
Zhang
, “
A nonlinear model for the layer between plates in acoustic noncontact transportation
,”
J. Appl. Phys.
116
(
21
),
214905
(
2014
).
17.
T.
Ide
,
J.
Friend
,
K.
Nakamura
, and
S.
Ueha
, “
A non-contact linear bearing and actuator via ultrasonic levitation
,”
Sens. Actuat. A Phys.
135
(
2
),
740
747
(
2007
).
18.
W.
Li
,
Y.
Zhu
,
K.
Feng
, and
K.
Zhang
, “
Effect of surface grooves on the characteristics of noncontact transportation using near-field acoustic levitation
,”
Tribol. Trans.
61
(
5
),
960
971
(
2018
).
19.
D.
Koyama
,
K.
Nakamura
, and
S.
Ueha
, “
A stator for a self-running, ultrasonically-levitated sliding stage
,”
IEEE Trans. Ultrason. Ferroelectr. Freq. Control
54
(
11
),
2337
2343
(
2007
).
20.
J.
Hu
,
G.
Li
,
H.
Chan
, and
C.
Choy
, “
A standing wave-type noncontact linear ultrasonic motor
,”
IEEE Trans. Ultrason. Ferroelectr. Freq. Control
48
(
3
),
699
708
(
2001
).
21.
J.
Hu
,
K.
Cha
, and
K.
Lim
, “
New type of linear ultrasonic actuator based on a plate-shaped vibrator with triangular grooves
,”
IEEE Trans. Ultrason. Ferroelectr. Freq. Control
51
(
10
),
1206
1208
(
2004
).
22.
B.
Wei
,
R.
Shaham
, and
I.
Bucher
, “
Theoretical investigation and prototype design for non-parallel squeeze film movement platform driven by standing waves
,”
Tribol. Int.
119
,
539
548
(
2018
).
23.
K.
Chen
,
S.
Gao
,
Y.
Pan
, and
P.
Guo
, “
Self-running and self-floating two-dimensional actuator using near-field acoustic levitation
,”
Appl. Phys. Lett.
109
(
12
),
123503
(
2016
).
24.
K.
Aono
,
M.
Aoyagi
,
H.
Kajiwara
,
H.
Tamura
, and
T.
Takano
, “
Increase of holding force in near-field acoustic levitation of tabular object inserted between opposing vibration sources
,”
J. Appl. Phys.
58
,
SGGD11
(
2019
).
25.
Y.
Liu
,
M.
Shi
,
K.
Feng
,
K.
Sepahvand
, and
S.
Marburg
, “
Stabilizing near-field acoustic levitation: Investigation of non-linear restoring force generated by asymmetric gas squeeze film
,”
J. Acoust. Soc. Am.
148
(
3
),
1468
1477
(
2020
).
26.
H.
Li
,
Q.
Quan
,
Z.
Deng
,
Y.
Hua
,
Y.
Wang
, and
D.
Bai
, “
A novel noncontact ultrasonic levitating bearing excited by piezoelectric ceramics
,”
Appl. Sci.
6
(
10
),
280
(
2016
).
27.
Y.
Hashimoto
,
Y.
Koike
, and
S.
Ueha
, “
Acoustic levitation of planar objects using a longitudinal vibration mode
,”
J. Acoust. Soc. Jpn. E
16
(
3
),
189
192
(
1995
).
28.
S.
Zhao
,
S.
Mojrzisch
, and
J.
Wallaschek
, “
An ultrasonic levitation journal bearing able to control spindle center position
,”
Mech. Syst. Signal. Proc.
36
(
1
),
168
181
(
2013
).
29.
Y.
Liu
,
X.
Sun
,
K.
Sepahvand
, and
S.
Marburg
, “
Theoretical analysis on the static and dynamic performances of a squeeze film air journal bearing with three separate pads structure
,”
Int. J. Mech. Sci.
200
,
106442
(
2021
).
30.
F. A.
Morrison
,
An Introduction to Fluid Mechanics
(
Cambridge University Press
,
New York
,
2013
), Chap. 5, pp.
369
379
.
31.
T.
Stolarski
, “
Running characteristics of aerodynamic bearing with self-lifting capability at low rotational speed
,”
Adv. Tribol.
2011
,
973740
(
2011
).
32.
T.
Stolarski
and
W.
Chai
, “
Self-levitating sliding air contact
,”
Int. J. Mech. Sci.
48
(
6
),
601
620
(
2006
).
33.
K.
Czolczynski
,
Rotordynamics of Gas-Lubricated Journal Bearing Systems
(
Springer
,
New York
,
2012
). Chap. 1, pp.
11
14
.
34.
S.
Ueha
,
Y.
Hashimoto
, and
Y.
Koike
, “
Non-contact transportation using near-field acoustic levitation
,”
Ultrasonics
38
(
1
),
26
32
(
2000
).
35.
S.
Chen
,
H.
Chou
, and
Y.
Kang
, “
Stability analysis of hydrodynamic bearing with herringbone grooved sleeve
,”
Tribol. Int.
55
,
15
28
(
2012
).
36.
K.
Feng
,
Z.
Huang
, and
Z.
Guo
, “
Design of spherical spiral groove bearings for a high-speed air lubricated gyroscope
,”
Tribol. Trans.
58
(
6
),
1084
1095
(
2015
).
37.
H.
Heshmat
,
J.
Walowit
, and
O.
Pinkus
, “
Analysis of gas-lubricated foil journal bearings
,”
ASME J. Lubr. Technol.
105
(
4
),
647
655
(
1983
).
38.
W.
Li
,
Y.
Liu
, and
K.
Feng
, “
Modelling and experimental study on the influence of surface grooves on near-field acoustic levitation
,”
Tribol. Int.
116
,
138
146
(
2017
).
39.
K.
Feng
,
M.
Shi
,
T.
Gong
,
Y.
Liu
, and
J.
Zhu
, “
A novel squeeze-film air bearing with flexure pivot-tilting pads: Numerical analysis and measurement
,”
Int. J. Mech. Sci.
134
,
41
50
(
2017
).
40.
E.
Matsuo
,
Y.
Koike
,
K.
Nakamura
,
S.
Ueha
, and
Y.
Hashimoto
, “
Holding characteristics of planar objects suspended by near-field acoustic levitation
,”
Ultrasonics
38
(
1–8
),
60
63
(
2000
).