Magnetic Levitation is Key to Breakthrough Process for High-Volume Manufacture of LED Arrays


Innovative use of diamagnetic levitation is at the heart of a sophisticated new process being developed by high-tech startup SelfArray, Inc. to make a breakthrough advancement in high-volume production of LED arrays. While focused presently on the LED display market, the system might also be applicable to semiconductor, lighting, and solar array manufacturing.

Key to the companys progress are National Science Foundation (NSF) programs at the Center for Lighting Enabled Systems & Applications (LESA) based at Rensselaer Polytechnic Institute and a recent Small Business Innovation Research (SBIR) grant of $756,027.

SelfArray, based in Troy, NY, received the SBIR Phase II grant in March to continue its work in advanced manufacturing technology for the production of direct-view LED displays. Direct-view LED displays are much brighter, have a better contrast ratio, and are more energy efficient than traditional OLED and LCD displays, and are currently being used for video walls and large indoor displays. The displays use small, individual LEDs as pixels. There are 24 million such LEDs in a single 4K display.

With current manufacturing methods, it takes anywhere from three weeks to four months to assemble a single direct-view LED display because a robotic arm must place each LED into a pixel array individually. SelfArray, however, is developing technology that utilizes magnets, vibration, and levitation to self-assemble LEDs in an array that can then be used to make a display. By using this method, SelfArray can drastically reduce the manufacturing time from months to minutes.


Our technology enables the assembly of large LED subsystems hundreds of times faster and with lower capital equipment costs than is common today, and our new NSF grant will enable us to continue our research and development over the next two years, said Clinton Ballinger, CEO. Our goal is to create a process that facilitates the manufacture of a lower-cost direct-view LED display that will replace current methods and as we develop, will help displace LCD or OLED technology.

According to the SBIR grant proposal, the broader impact and commercial potential of the project is that it could revolutionize large area LED array assembly manufacturing, leading to unprecedented growth of the direct view LED display market. Direct view LED displays, or video walls, are becoming more ubiquitous in everyday life. First only appearing in sports stadiums, they have gradually started appearing in retail advertising, and now are starting to appear in cinemas and control rooms. Despite their advantages in brightness, contrast ratio, refresh rate, and efficiency, the rate of adoption of LED displays is slow because of the cost and time required to make them. Current manufacturing tools are facing a bottleneck as display resolutions (LED count) increase. A massively parallel assembly technology could drastically increase production speeds while reducing cost-of-goods. Additionally, the knowledge gained by the research into this self-assembly technology has the potential to unlock new opportunities in semiconductor packaging including lighting and concentrated solar.

Using diamagnetic levitation, the system speeds up the LED placement process by leveraging the parallel nature of directed self-assembly. Rather than placing individual LEDs into a grid-array as it is done today, this technology utilizes confining magnetic fields to quickly and simply arrange hundreds or thousands of LEDs into a neat and ordered grid-array. The objective is to achieve assembly speeds 100s of timesfaster than conventional tools. The research is focused on the effects of LED size and geometry on accuracy, rate, and yield. Additionally, new magnetic field stages to enhance assembly precision are being developed. The objectives are to develop a system that will be able to rapidly assemble thousands of LED die with the precision necessary for high-resolution displays, and a fully operational LED display panel constructed with this technology to demonstrate its manufacturing potential.

Research at LESA and Rensselaer Paved the Way

Behind the progress to date has been a seminal collaboration with LESA. Center Director Robert Karlicek proposed a challenge: Find a way to produce large sheets of precisely oriented unpackaged LEDs without having to pick up and place each one at a time.

james lu_IN

The challenge was accepted by Rensselaer Professor James Lu who created a radically new LED self-assembly technology — using diamagnetic levitation to create large arrays of LEDs without touching a single chip. This novel approach to solving large scale LED packing problems offered the potential to drastically reduce cost associated with manufacturing narrow pixel pitch LED displays, and was significant enough to spinout start-up company SelfArray. Lu now is also the Chief Technology Officer and a director of SelfArray.

Once SelfArray was joined by experienced CEO Ballinger, the start-up received its first SBIR award and hired Rensslaer doctoral graduates, Mark Durniak and Michael Conward, to drive the technology development work.

This is a great example of what engineering powerhouses like LESA are designed to do; create an entrepreneurial environment where researchers can innovate together, solve important technical challenges, and create real economic value, says Karlicek of the start-ups success. It is the fifth spinout company by LESA faculty and students.

Funded primarily by the National Science Foundation, LESA is an interdisciplinary, multi-university center. It engages faculty members, graduate students, research staff, and undergraduates to work on research leading to smart lighting systems with adaptive and controllable properties that will change the way society uses lighting. The center is headquartered at Rensselaer in Troy, New York, and partners with Boston University, the University of New Mexico, and Thomas Jefferson University.

To learn more, go to, and to