Experimental Objectives

Determine the governing theory

The correct theory behind Near-Field Acoustic Levitation is not yet completely determined and the phenomenon is often described from different angles. When the levitated object is very close to the horn (less than half a wavelength) the levitation is governed by Squeeze Film Levitation (SFL) or Acoustic Radiation Pressure, whilst when it gets further away, towards so-called Far-Field Acoustic Levitation (FFAL) it is the Standing Wave Acoustic Levitation (SWAL) that is keeping the object levitated. Even for strict NFAL the respective theories are divided. By subjecting it to hypergravity, we aim to see which theory holds best.

NFAL to FFAL

The behaviour of the object in FFAL, described by SWAL, suggests that the equilibrium position of the object at short distances (ie. high frequencies) should be constant with changing gravity. Our experiment aims to investigate the interaction between NFAL & FFAL and how the equilibrium position and resonance oscillations of the object changes with increasing gravity. This could provide valuable insight into the validity of the theories in different conditions.

Attractive force

An interesting phenomenon that we intend to investigate is the presence of an attractive force. Objects can actually be picked up using the aforementioned vibrating plates. By measuring the force on the object at different distances from the plate we will be able to determine whether this effect is present under hypergravity conditions and the location of the transition between attractive and repulsive forces. Even in the case that we cannot find this attractive force, the force profile we will obtain will provide us with information on the applicability of the theories at different distances.

Experimental Set-Up

Here you find information on each part of our experiment, with the different sections shown below. The experiment consists out of two main set-ups, designed to measure different properties of NFAL.

Near-Field Acoustic Levitation System

The common components of both set-ups needed to achieve NFAL.

Learn More

Hypergravity

The Large Diameter Centrifuge

The environment our experiment will be subjected to.

Learn More

Measurement

Set-up 1: Object displacement

The set-up to measure the equilibrium levitation distances.

Learn More

Set-up 2: Force profile

The set-up to investigate the force-field caused by NFAL & FFAL.

Learn More

Near-Field Acoustic Levitation System

This constitutes the components needed to achieve NFAL in the first place, measure the environmental factors it is subjected to and ensure functional optimisation & integrity.

Levitation system

The levitation system consists of a transducer, booster and horn, connected in that order. The transducer, being the object to create the vibrations using piezoelectrics, is driven using a signal generator and amplifier and will be driven at 20 kHz with amplitudes between 1-25 μm with a maximum power output of 200 W. The booster and horn are there to amplify these oscillations and provide a uniform radiating surface acting on the body. We are currently developing our own transducer using COMSOL.

Frame & Damping

To be able to subject our equipment to 20G whilst also maintaining high levels of precision, a sturdy aluminium frame has been designed. One of the challenges here is to design it as such that parts of the different set-ups can be quickly interchanged to maximise our time-usage during our experiment campaign. Our frame is being developed in collaboration with our Physics Mechanical Workshop. To dampen any vibrations present in the LDC, we will be lining the bottom of the gondola with layers of Sorbothane, with different types for different G-ranges.

Environment meters

There are many parameters affecting NFAL, all of which need to be measured and accounted for. This requires us to have a barometer, thermometer, hygrometer & accelerometer. Additionally, whilst we will be minimising vibrations, we still require to measure the vibration profile so its affect on our data can be corrected for. Above you see an image of the effect of a temperature error on alpha, that being a variable in one particular theory, used to determine the resultant error in the force.

The Large Diameter Centrifuge

We will be making use of the LDC to simulate hypergravity up to 20 G. Using such equipment does also come with extra requirements for our set-up.

Gondolas & Control

Our experiment will be placed inside a gondola at the end of one of the arms. As such, it will be subjected to the restriction of max. 80 kg and a workspace of 500x500x720 mm. Whilst the experiment itself is subject to as high an acceleration as possible, the processing and control devices will be placed either in the central gondola or one halfway. From a control station, the equipment can be operated remotely, though automation is a priority.

Force profile & vibrations

Though very close, artificially induced gravity is not perfectly uniform; the acceleration vector will be pointing slighlty outwards on the sides in the gondola and will be lower higher up in the gondola. Keeping our set-up centred and as low as possible is thus a goal to strive for. Furthermore, low frequency vibrations will be induced by components like the LDC motor and gearbox; this is a factor to consider given the micrometer range we are working in.

Gravity levels

The LDC is able to accelerate and deaccelate from 1 to 20 G in 60 & 55 s respectively and is able to provide us with a range of G-levels with small increments between them. Though we will only have 8 hours of experiment time per day, the LDC could theoretically run for 6 months straight.

For more information regarding the Large Diameter Centrifuge, see our Spin Your Thesis page

Set-up 1: Object displacement

The set-up investigating NFAL in particular, looking at the equilibrium position of free-floating objects as well as the presence & extent of resonance.

Displacement sensor

With an expected range of a few milimeter, we will be using a displacement meter with sub-micrometer precision and micrometer accuracy. The meter will be placed just above the levitating object to measure its levitation height as well as the amplitude of its resonance. Additionally, the meter will be used to measure the horn amplitude itself to obtain the radially symmetric displacement profile and calibrate the amplitude at different G-levels. The image above shows the set-up to be used in our own laboratories to obtain this profile.

Levitated object

A homogeneous surface aluminium plate of around a milimeter thick is to be floated freely above the horn. The object is designed to reflect the acoustic radiation well, thus optimising the levitation force but also be light to allow for the fullest exploration of high G-levels whilst still enabling NFAL. Whilst the squeeze film provides a restorative centripetal force, a cage of wires is designed to be placed around the levitation system to provide a mechanical boundary. The image above shows the set-up as to be used in hypergravity. Note, the wire-cage is not shown.

Set-up 2: Force Profile

The set-up bridging the gap between NFAL and FFAL, investigating the forces as one moves away from the levitation system. Also, the attractive force, where it might be and how strong it is, will be investigated using this system.

Force sensor

The force profile is expected to increase exponentially moving closer to the radiating horn surface. However, we expect peaks of a few Newton at integer multiples of half-wavelengths. The force sensor will be subjected to various effects to account for, including high baseloads, from the live mass, as we move up to higher G-levels, inducing creep; hysteresis caused by the connection of the electronics to the metal, temperature fluctuations and more. The system will be handling tension and compression based forces as we move through the force profile. The force sensor will be screwed onto a gantry above it and the "levitated" object below.

Linear actuator

A system of two stepper motors & ball screws will be used to precisely move the force sensor up and down. Sticking out from the top, holding the object through a gantry and supported on the bottom by bearings in support stands. We have chosen motors that give us micrometer stepsizes, with the possibility of going for smaller steps with microstepping due to the excess torque they provide. This system will be controlled using a stepper motor driver connected to a microcomputer, taking feedback from the force sensor.

Experimental Plan

Our experiment campaign is made up out of three days using the Large Diameter Centrifuge; consisting out of five 4-hour slots. Whilst we have some time flexibility at our own laboratories at Durham University, when present at the ESTEC facility we need to maximise our time-efficiency, which is balanced against experimental risks. This means changing between the set-ups above as little as possible but also obtaining as much data as we can before going to high G-levels, to account for the risk of components breaking due to the higher strain.

Set-up 1

Voltage calibration

As we can vary the voltage provided by the amplifier signal to the transducer, this output needs to be calibrated to obtain the right amplitudes at different G-levels. Without the object present, the voltage range will be gone through, noting down the exact voltages providing the right amplitude increments. Additionally, with the transducer off, a measurement of the distance between the meter and the horn is used as calibration, to account for factors like bending of parts.

Set-up 1

Object measurement

With the now pre-determined voltages, these amplitude increments will be cycled through with the levitation height and oscillation amplitude of the object measured.

Set-up 2

Baseload calibration

With the transducer turned off, a measurement of the baseload will be made at each G-level. This will be done every time before and after the measurements are taken at that level. This gives a measure of the zero-output relative to the acoustic force, and creep of the load. By taking measurements of the expected baseloads over longer time-spans in Durham, we aim to obtain the creep and recovery profile of the force sensor.

Set-up 2

Force measurement

In between these calibrations, the force sensor will be moved down to the transducer and up again with measurements of the force acting on the object made throughout. Doing two measurements, allows us to reduce the error caused by hysteresis & repeatability. This will be done using the set amplitudes determined by the calibration made with set-up 1.

Our Experimental Set-up & Plan