Commentary, opinions, and insights on all things MEMS.

June 2017: Six Actions to Improve Diversity in Tech

The lack of diversity in the staff of technology companies, particularly here in Silicon Valley, has become a hot topic lately. (Or finally, depending on your perspective.)

As a female Ph.D. engineer with more than 20 years in industry, and as a business leader, many colleagues have recently sought my opinion on how best to improve the diversity of their staff. I am inspired by their sincerity and earnestness to effect change within their companies, as well as in the wider industry.

I recommend six actions to improve the diversity of technical staff:

  1. Slow down the hiring process to make time to find diverse candidates. Qualified candidates are out there, but you may not have the network to find them right away. Give your job announcements time to diffuse outward to other networks. Startups should focus on diversity right away and beware falling into the easy trap of hiring one's friends, which inevitably leads to homogeneity.
  2. Invest in broadening your own professional network. Get out of your office and typical social circle and go meet some new people. Attend and sponsor events that cater to diverse demographics and cultivate connections with university student groups. Start today, because networks take years to develop.
  3. Let your diverse staff members be ambassadors for your company by giving them meaningful public opportunities, not by using them in your marketing materials. Send them to conferences and tradeshows to give technical talks or to represent your company's interests. It'll be great professional experience for them while they inspire and motivate those they meet. Don't just turn them into 'poster children' on your website, that's exploitative.
  4. Recruit diverse higher-level technical managers and executives. Senior people have their own networks of talent to tap, naturally draw the interest of those like them, and also act as aspirational models for their subordinates. Their positive leadership will also help to erode people's implicit biases.
  5. Hire for potential, not past experience. There's a lot of talented people out there who haven't yet had the opportunity to shine, or don't have experience in your exact field. Consider skills learned in adjacent fields which may translate well to your field. In your interviewing process, focus less on the resume and more on testing for smarts, coachability, positive attitude, and resourcefulness. Those are the qualities that make a really great employee. Teaching capable new hires about the specifics of your field is a relatively small investment.
  6. Create and enforce a culture of respect. Nothing is more demoralizing to a hard-working employee than to have their contributions marginalized or dismissed. Make sure your corporate culture is respectful and inclusive, and enforce it. Don't wimp out on firing bad apples, even if they are top performers. Their corrosive effect on your company culture will more than offset their benefits as contributors.
I have a lot more to say on this topic and would welcome the opportunity to speak with you further. -Alissa

Birds of a feather flock together, which is why it takes strategy and effort to improve diversity.

May 2017: MEMS shut out by century-old sensor

A few weeks ago, I joined a friend to see the San Jose Sharks play the Edmonton Oilers in Game 4 of the first round of the Stanley Cup playoffs. At each of our seats,as a fan appreciation gift, the Sharks had placed a PixMob LED bracelet. When the arena lights dimmed, the bracelets started to sparkle. We were treated to a dazzling and sophisticated light show: all of the fans' 17,000+ bracelets winked multiple colors, turned on and off by seating section, and flashed to the beat of thumping music. It was a thrilling example of how much fun could be had with wearable electronics.

After we left the arena, we noticed that our bracelets' LEDs would flash whenever we moved our arms. A motion sensor! I was eager to take the bracelet apart to discover which company's MEMS sensor might be inside.

When I extracted the circuit board from the bracelet, I turned it over a few times, seeking the MEMS chip. I identified an ABOV microcontroller chip, an EEPROM chip, a bunch of discretes, and an infrared sensor (for communications), but no MEMS chip.

Thinking my creeping farsightedness might be the problem, I placed the board under a stereo microscope for closer inspection. To my great surprise, I discovered that what I had initially thought to be an electrolytic capacitor was actually a BALL TILT SWITCH motion sensor. Unbelievable!

My goodness. This networked, wearable electronic device was using a 1920s-era motion sensor.* The ball tilt switch is nothing more than a metal ball in a tube that rolls when tilted or moved, shorting two electrical leads within the tube. The Rolamite sensor, a slightly more complex version of the ball tilt switch, was used to deploy airbags in cars until MEMS accelerometers came along in the 1990s.

The PixMob engineers had clearly done their jobs well. The board's design indicated that they had deliberately engineered the bracelet for "good enough" performance at bare minimum cost. Their discipline extended to selecting a century-old mechanical sensor for this most modern of products.

There's a lesson here for all of us MEMS business people who aspire to sell sensors for cost-conscious or disposable products: the cheapest, "good-enough," simplest sensor will win the socket, not the smallest, nor the most accurate, nor the most sophisticated. MEMS may seem like the obvious choice for wearable electronics, but this time it got beat by an old mechanical sensor.

The other big surprise of the night: the Oilers also got shut out, falling 7-0 to the scrappy older Sharks.

*The oldest patent I could find after a quick search. If you know of an older one, please let me know!

April 2017: Development of a High Performance Micro-mirror Array

For the past several years, AMFitzgerald has been developing the fabrication process for a novel MEMS micro-mirror array designed by Dr. Robert Panas's research group at Lawrence Livermore National Laboratory. The technology has been developed specifically to serve LIDAR, laser communications, and other demanding applications where existing MEMS mirror array technologies are insufficient. The novel mirror architecture offers exceptional speed and tilt range, with three axes (tip-tilt-piston) feedback control and 99% fill factor.

In this video, Dr. Panas provides an overview of the Light-field Directing Array (LDA) technology. The technology is available for license from the LLNL Industrial Partnerships Office.

At the upcoming MEMS & Sensors Technical Congress, on May 11, Dr. Carolyn D. White will present a case study on how she developed this complex prototype and leveraged AMFitzgerald's ecosystem of partners to integrate specialty processes not available in our fab. Her presentation will offer a unique look inside one of our client projects and how AMFitzgerald develops novel MEMS devices. She will also offer insights on form factor-driven process challenges and development strategies for proof of concept budgets.

In-process photo of the LDA prototype during fabrication by AMFitzgerald

March 2017: Collaboration with Millar, Inc. to Enhance OEM MEMS Pressure Sensor Capabilities for the Medical Device Industry

AMFitzgerald and Millar, Inc. are pleased to announce an official collaboration agreement to serve OEM customers seeking a complete sensor solution using micro-electromechanical systems (MEMS) technology. This agreement arises from a long history of MEMS technology development between the two companies and furthers each company’s vision to better serve the growing interest in the medical device and clinical communities for increased in vivo physiologic data.

The clinical market has experienced an upward trend of sensor integration into existing medical technologies, especially in the cardiovascular market, to enable improved patient monitoring, diagnostics and outcomes. Due to their small size, low power consumption, and low cost, MEMS sensors offer exciting new capabilities to gather in situ data at locations in the body which cannot be otherwise accessed. Pressure, temperature, motion, and flow parameters may now be directly measured.

By leveraging Millar’s ISO-13485 quality control certification and 45+ years of MEMS-enabled product manufacturing expertise, AMFitzgerald can now provide advanced and proven solutions for sensor testing, wire attach to small sensor die, and biocompatible encapsulation, delivering OEM customers a fully-packaged sensor module solution for medical device integration. The end result will be a faster path to market for companies that work with the AMFitzgerald and Millar teams.

AMFitzgerald has been developing innovative MEMS sensors and actuators for medical devices for more than 10 years. Full-service engineering capabilities include custom MEMS design, RocketMEMS®; semi-custom pressure sensors, selection of commercial sensors, electronics and packaging, foundry transfer, and supply chain creation and management.

About Millar, Inc.  Since 1969, Millar, Inc., headquartered in Houston, Texas, has led the development of catheter-based, solid-state pressure sensors and is known worldwide as the leader in sensors that advance medical understanding. Millar OEM serves the medical device and life sciences industries through its MEMS pressure sensors, ISO 13485 precision manufacturing and wireless power technology, resulting in cost savings and rapid time to market for device integration.

Millar offers proven solutions for integrating MEMS pressure sensors into invasive medical devices (Source: Millar OEM)

Prior Newsletters: 2016 and earlier