We recently had the opportunity to speak with Stephen Day, Coto Technology Science and Engineering Consultant, and a key player in the development of Coto's Product of the Year Award-wining RedRock RS1-A-2515 MEMS-based reed relay. Here is an edited transcript of the interview:
Electronic Products: How did the idea for this product come about? Approximately when did it happen, and who were the people involved at the very beginning?
Stephen Day : In March 2003 the technical manager of Coto Technology’s switch manufacturing facility in the Netherlands reported that he could not build a reed switch with a capsule length of less than 5 mm. Prior to that, reed switch lengths had shrunk progressively from 50 mm around 1938 to 5 mm in 2003. As the VP of Engineering at that time, I had asked our technical manager to build a reed switch smaller than 5 mm since we need one to meet increasing demand for various applications.
The reason a switch less than 5 mm long could not be built was an unsurmountable physics problem; heat from the glass sealing process destroyed the platinum-group coating on the contacts. If the switch capsule was shrunk further, the contacts were too close to the source of sealing heat to prevent contact damage.
EP : When was it decided to go ahead with this project? What was the critical factor in committing to the product’s development?
Day : In May 2006 I met with the then president of HT Micro (HTM), a designer and manufacturer of MEMS devices, at a technical conference. At his request, we discussed the possibility of Coto winding the coils for a primitive electromechanical relay that HTM had developed. That was not feasible, but I immediately realized that the HTM microfabrication technology called High Aspect Ratio Microfabrication (HARM) could be used to build a reed switch with precious-metal contacts much smaller than 5 mm in length, since no high-temperature processes were necessary.
Such switches had huge potential for applications requiring a very small reed switch. At the time, those applications included hearing aids, insulin pumps, and lid closure detection in shrinking laptop PCs and PDAs. The switches also had the potential for use in new types of very small reed relays, since they could be operated by either a permanent magnet or an electromagnetic solenoid coil.
EP : What new technologies had to be developed?
Day : A radically different approach to fabricating the flexible nickel-iron blades and hermetic capsule of a conventional reed switches was needed. For 70 years, reed switches had been manufactured by stamping nickel-iron wire into a flat spring form, coating the tips with precious metal and then heat-sealing the blades into a glass tube. The process was expensive and labor intensive. In particular, control of the size of the contact gap was difficult, and as stated above, switches less than 5 mm long could not be built.
We realized that with the HARM process, the reed switch blades could be grown by electroplating rather than mechanical stamping, and that the thickness of the blades and the size of the gap could be defined with very high precision using the same photolithographic processes that were used to make integrated circuits. The glass capsule of a conventional reed switch could be replicated by an evacuated, hermetically sealed wafer cap over each HARM switch. The cap wafer could be bonded to the switch wafer at far lower temperatures than the 1000oC that had been destroying the precious metal in the reed switches.
EP : Who worked on the development?
Day : Initially myself, and the manufacturer’s [HTM's] CEO and Engineering VP. Later, a whole team of people at Coto Technology, including Coto’s Director of Engineering and VP of Sales & Marketing were instrumental in developing the “awareness” of the HARM technology and product offerings.
During the development process, solving difficult problems took much longer than originally planned. Referring to the actual HARM process, we learned there is more value than initially anticipated or designed, such as incredible robustness from wafer level packaging, magnet field directionality, etc. These were key attributes for applications not originally targeted.
EP : What happened in getting the product from design to manufacturing? Were there changes in the design along the way? If so, why?
Day : A pilot project was started in April 2007. Various initial attempts to build HARM reed switches resulted in devices that worked but had insufficient magnetic closure sensitivity for practical applications — a very large and powerful magnet would have been needed, which defeated the purpose of a small switch. Finally, a design that worked was developed — it had built-in magnetic field concentrators that focused the field from an external magnet in the contact gap, allowing high contact force to be developed from a magnet having similar dimensions to the switch. High contact force was the key to long contact switching life and high current switching capability, attributes that had not been achieved in previous MEMS switching devices. This development resulted in US patent 8,327,527, issued December 12, 2012.
EP : What was the market acceptance like? What feedback did you get from the marketplace?
Day : The market response was immediate and overwhelmingly enthusiastic, since the microfabricated reed switch (now called RedRock) enables many applications that simply had not been previously possible. Examples include ingestible capsule endoscopes that image the human intestine for early detection of colon cancer, and in-the-canal hearing aids that are too small to include a mechanical switch. Any application needing a combination of very small size, remote operation by a magnet or coil and zero power operation is an ideal target for the RedRock switch.
We also have a number of customers seeing the huge advantage in RedRock’s “robustness,” where improved product quality, reduced field failures and longer life are all enhanced relative to current solutions. RedRock may currently be a more expensive item but presents an overall cost of ownership reduction.
EP : Has this product’s development led to ideas for future products? How do you foresee the product family growing in the future?
Day : Though a product roadmap was developed as part of the initial process development process, since we released the initial switch, the positive field feedback has certainly influenced the latest changes leading to a more refined switch offering, followed by electromagnetic relays and moving into sensors. What we didn’t anticipate was the overwhelming demand for reference designs which, in itself, very well could become a product offering.
Advertisement
Learn more about Coto Technology
