Using NI FlexRIO to Control Ultrasonic Arrays for Sonotweezing


"The complete dexterous manipulation system was a fraction of the cost of a traditional multichannel controller and the LabVIEW integration greatly increased its versatility and the future possibilities for the hardware."

- D Lines, Diagnostic Sonar Ltd, Livingston, UK

The Challenge:
Driving a multichannel array with continuous wave and independent channel phase excitation to manipulate particle clusters and objects as small as 10 µm in real time using ultrasound, and permitting the evaluation of applications like focused ultrasound surgery.

The Solution:
Using NI FlexRIO hardware to produce repeating digital waveforms to drive a multichannel transmitter with individual phase control and using a multielement ultrasound source to create standing waves to trap particles and provide precise manipulation of the phase for accurate positioning and dexterous particle control.

D Lines - Diagnostic Sonar Ltd, Livingston, UK
C Demore - University of Dundee
R Habeshaw - IMSaT, University of Dundee
H Wu - University of Dundee
B Drinkwater - University of Bristol
C Courtney - University of Bath
Z Qiu - University of Dundee
S Cochran - University of Dundee

About the Teams

The Institute for Medical Science and Technology (IMSaT) at the University of Dundee was founded in 2006 as an interdisciplinary institute positioned at the interface of physics, engineering, and clinical and life sciences. It comprises three active research groups, with ultrasound being a major focus of the institute’s work.

Diagnostic Sonar Ltd. (DSL) was founded in 1975 with the intent of successfully introducing the first real-time medical ultrasound scanner manufactured in the UK. Over the following four decades, the company expanded into other areas including industrial nondestructive evaluation and medical phantom supply.

Multichannel ultrasonic array controllers are often designed for imaging, with pulser units creating successive high-voltage, but very short pulse-length, excitation. Other varieties include large, cumbersome machines that consist of independent function generators, physical output switching circuitry, and power amplifiers intended for high-power, long-burst duration applications such as underwater sonar, industrial processing, and focused ultrasound surgery.

This project uses a DSL-designed imaging transmitter card that interfaces with NI FlexRIO hardware and is modified with an active cooling system for continuous wave operation at midlevel output voltages. This creates a compact, versatile, and modular excitation system with controllable waveform generation and individual channel phases configured with NI LabVIEW system design software.

Our Experiment

The basic method fixed amplitudes for all elements and applied phase delays based on two principles, the first principle created a field using a low-pressure node at the centre of the array, surrounded by a high-pressure antinode. This field was produced when the phase of the signal applied to each transducer increased successively around the array. The second principle relied on negligible reflection and stated that the field near the centre could be moved by further changing the phase of the elements according to the distance from each element to the desired target position.

IMSaT fabricated the device we used, which consisted of 16 individual transducers and featured real-time trapping and a focus control method. A 2.45 MHz square wave signal at 10 Volts (peak-to-peak) with 70 percent pulse-width modulation (PWM) drove each element. The PWM was equivalent to three-level waveshaping to approximate a sine wave root mean square (rms) value from the digital output. That is, if the voltage was at a positive value for 70 percent of the time, then it would return to zero for 30 percent of the time. If it was at a negative value 70 percent of the time, then it would return to zero for 30 percent, and so on. This significantly reduced the third harmonic content in the signal.

We calculated the phases according to the principles mentioned above and applied them to each individual element. The increment that the phases could be varied at was based on the sample waveform word length. In the present case, we built each waveform from 26 bits giving 26 possible phases for each element.

To complete the experiment, we used 10 µm diameter fluorescent green polystyrene beads as the targets. These beads sediment relatively slowly. Additionally, for the tests, a layer of agar raised the beads to the centre plane of the ultrasonic beam and provided a relatively frictionless base where the beads could be moved. As an alternative, a novel transparent piezoelectric device could be used to lift the beads against gravity.

To image their motion, we observed the particles under a microscope and connected a Schlieren imaging system to help us observe the ultrasonic pressure fields directly through acousto-optical modification of the refractive index of the water.


We successfully ordered and trapped agglomerates of particles out of suspension and moved them with real-time control by more than 1 mm from the centre of the ring. We used LabVIEW to write the focus code and calculate the successive channel delays for each small change in focal position. We captured the result on video from the microscope.

Figure 1. A video through the centre of the 10 mm diameter ring of transducers shows an agglomerate of 10 µm particles (in light green) trapped in the ultrasonic standing wave field, in suspension in water. The image shows the centrally positioned trap and the first surrounding ring, which is approximately 0.63 mm in diameter. The next ring out, which is 1.26 mm in diameter, is visible at the corners of the image.


The Schlieren imaging highlighted our success with the full ultrasonic pressure field clearly visible within the array cavity. Requesting the focal position to move showed smooth changes in the field and the trapping focus remained stable to a distance of around 1.3 mm from the centre. Again, we captured this on video. 


Figure 2. The Schlieren video shows the centre of the 10 mm diameter ring of 16 transducers. The lighter areas show the areas of high pressure. The low pressure area in both the centre and the surrounding rings (the darker areas) trap any small particles present in the water.


The complete dexterous manipulation system was a fraction of the cost of a traditional multichannel controller and the LabVIEW integration greatly increased its versatility and the future possibilities for the hardware.

Future Work

Ultimately, the Sonotweezers project aims to achieve dexterous, noncontact manipulation of cells for applications in life sciences and medicine. The tests reported here successfully demonstrated the theoretical background of the project, the application of the NI FlexRIO hardware adapter, and the simplicity of incorporating the theory within LabVIEW. It will be relatively straightforward to implement a control system that empowers an operator to move cells under a microscope through a touch screen interface or with a mouse.

We also recently began to explore the idea of using the same system for focused ultrasound surgery and targeted drug delivery, which is ideal for bench-based research.

Author Information:
D Lines
Diagnostic Sonar Ltd, Livingston, UK
Baird Road, Kirkton Campus
West Lothian EH54 7BX
United Kingdom

Bookmark and Share

Explore the NI Developer Community

Discover and collaborate on the latest example code and tutorials with a worldwide community of engineers and scientists.

‌Check‌ out‌ the‌ NI‌ Community

Who is National Instruments?

National Instruments provides a graphical system design platform for test, control, and embedded design applications that is transforming the way engineers and scientists design, prototype, and deploy systems.

‌Learn‌ more‌ about‌ NI