Thursday, August 30, 2012

MEMS Product Development Challenges – Sweet Dreams and Nightmares

MEMS product development is not for the faint of heart. Though you will see many success stories in the industry, you will also see many failed ventures (did anyone say “telecom bubble?”), several bankruptcies (TeraVicta to name one), gray hair (or no hair), and divorces (sad, but true). And then there are the companies that are just middling along, waiting to break through – we have a duty to help them break out –  now!

That’s why I aptly titled our upcoming 2nd annual MEMS Industry Group (MIG) workshop with BSAC, on September 19, “MEMS Product Development Challenges – Sweet Dreams and Nightmares.” We have a lot to be proud of in the MEMS industry, but we still have a lot to learn and a lot to improve on in order to grow. We may be a $10B/year industry now; but to get to my dream of “MEMS frickin’ everywhere,” we need to do more.

All year long, MIG’s theme for content and programs has been focused on addressing MEMS product development and commercialization challenges. Our annual technical members meeting, M2M Forum, focused on MEMS new product development and we invited Len Sheynblat of Qualcomm to give a keynote on the real truth about what makes integrating MEMS and sensors into end-use mobile devices so darned hard and complicated: a lack of MEMS standardization. We teased out the differences and nuances between MEMS technology push and market pull; when, what and how it matters and why we should care. We developed a MEMS Technology Development Process Template to help managers navigate the gating process to determine when and if a MEMS device is a GO or NO GO. Additionally, MIG has worked closely with our MIG Technical Advisory Committee (TAC) to ensure the content of our MEMS Education Series webinars is focused on MEMS product development.

We also hosted a fantastic one-day pre-conference symposium at Sensors Expo & Conference in June – all focused on, you guessed it – MEMS commercialization challenges. This time we called it “MEMS in the Mainstream: Commercialization and Product Realization – Leveraging the MEMS Infrastructure” (I was feeling less creative that day, I guess).  I am really proud of the folks we invited to speak and present at the Sensors Expo pre-con and MEMS tracks. The presenters all spoke honestly about the sweet dreams and nightmares they experienced while taking a MEMS device through the many stages of development along the MEMS supply chain. Thankfully, the supply chain is a heck of a lot more robust to help shift the odds more towards the “sweet dreams” side of things rather than the nightmares…but it’s still critical to keep that honest conversation going and give those case studies about MEMS product development challenges, specifically about MEMS fabrication models.

That is why our September 19 workshop at BSAC is so unique. We will give attendees an inside peek at how some of the best and most impressive MEMS companies have overcome their nightmares and are enjoying some sweet dreams (with occasional scary monsters under the bed and in the closet at times, I am sure). I am thrilled that one of MIG’s Governing Council members, and a really nice guy, the esteemed Dave Monk, MEMS Automotive Sensor Product Manager, Freescale Semiconductor, will present a case study on Freescale’s hybrid model of fabrication: “To Integrate or Not to Integrate: A Case Study on System-in-Package Integration for MEMS-based Products.”

This workshop is especially unique – because for the first time – drum roll please – we will have Knowles Electronics share a case study on their perspective of having a totally fabless model for their MEMS microphones. This is big stuff people. I didn’t have to change the name of my first born to get Knowles to agree to come and present; but it did take some convincing so I hope you will join me to hear Angelo Assimakopoulos Director, New Business Development and his colleague, Eric Lautenschlager, MEMS Engineer Manager, talk about how the magic happens. The third case study will be given by my friend and colleague, Peter Himes, VP Marketing of Silex Microsystems, the worlds’ biggest pure play MEMS foundry (I feel like have been programmed to say that every time I say the word “Silex”). Peter has been working on this presentation all summer – “Foundry: MEMS Product Proliferation and Time to Market: A Foundry’s Perspective on Process Standardization vs. Full Customization”  - and I can’t wait to see it (especially the part with the Swedish chef – right, Pete?).

After the case studies we’ll allow for a very short break and then the fun will begin again. Leslie Field, Consultant, Manager and Founder, Small Tech Consulting will moderate the panel “Successful MEMS Commercialization – Lessons Learned.” Panelists include MIG members  Evgeni Gusev Sr. Director Technology R&D, Qualcomm MEMS Technologies; and Marcellino Gemelli, Senior Marketing Manager, Bosch Sensortec; and BSAC spin-outs Christine Chihfan Ho CTO, Imprint Energy; and Octavian Florescu President, Silicon BioDevices. These four bring very diverse backgrounds and perspectives on what it takes to successfully navigate the commercialization process, including considerations for planning the development of new products. I look forward to hearing how each of them will define “successful commercialization” and how they address the challenges of integrating MEMS into existing applications vs. new products. I also think it will be fun to hear what questions the audience will ask…

I encourage you to join me on September 19 on the UC Berkeley campus – for the workshop, our joint session in the afternoon with BSAC, and finally our mixer/social/cocktail party in the evening (sponsored by ClassOne Equipment – thanks guys!). The early bird registration discount ends on September 6 – so don’t delay! The time for MEMS really is now.

And, if you’re a MIG member, start the party early in Mountain View on September 18 at a Happy Hour we are hosting at Tied House Brewery & CafĂ©, sponsored by one of our distinguished industry partners MIPI Alliance.  RSVP today!

Thursday, August 16, 2012

A Visit to GE Global Research, Niskayuna, NY

Perched on a bluff overlooking the Mohawk River in Niskayuna, NY is a powerhouse of industrial R&D; GE's Global Research Center (GRC). GRC just celebrated its 110 year anniversary. Thomas Edison's original desk is on display in the entry lobby to prove this point!

(Click on any of the images below to enlarge them.)





The Niskayuna facility is the largest of several GRCs. GE also has centers in Munich, Germany; Bangalore, India; Shanghai, China; Rio de Janeiro, Brazil and San Ramon, California.

The history of invention and innovation that has taken place at GRC to create major new businesses was on display as we walked along the entry hall. Some highlighted examples include x-ray medical imaging, jet engines, magnetic resonance imagers, digital x-ray panels and a number of other world firsts. The Research Center in Niskayuna is one of the world's largest corporate R&D centers that conducts focused, strategic research and development. This is GE -- after all -- and GE has the demonstrated ability to identify new business opportunities, utilize its research capacity to develop the required technologies in collaboration with GE businesses, and then to grow these businesses globally. As such, the master plan needs significant R&D capability to back it up.

The GRC R&D programs are divided between longer term projects (Advanced Technology programs) and projects that develop technology for one or more GE products in a two to five year time frame. GE business divisions (i.e. GE Aviation, Healthcare, Energy, Oil and Gas, etc.) fund the shorter term, targeted R&D projects and researchers at the Center therefore have an internal business customer. GRC is responsible for meeting the needs of that business, and the businesses frequently visit the Center to review technical progress.

GRC has been conducting research in MEMS since 2002. The objective of this research is to create new MEMS components that are not commercially available and that enable a breakthrough feature or strengthen the value proposition of a GE product. One of the most impressive MEMS devices that I learned about at GRC is a MEMS microswitch -- GRC researchers claim this switch to be the fastest switch that acts as a mechanical relay (vs. solid state, semiconductor switches). Through applied and focused R&D, the brilliant scientists at GRC have created a MEMS microswitch that handles over a kilowatt of power (240 Volts AC and 7 Amps). For its industrial applications (just for starters) this MEMS microswitch reduces the time required to interrupt a fault current (short circuit) from milliseconds (the switching speed of conventional circuit breakers) to a few microseconds (for a breaker made from MEMS switches).

It is equally amazing that these researchers were able to utilize a novel set of materials to construct the MEMS switches (GRC's "secret sauce"). The switch materials and the process flow are compatible with both silicon and quartz substrates, and it may be possible to fabricate these switches on completed CMOS wafers. The novel materials play a key role in enabling the fabrication of the robust, reliable and mega-power-conveying MEMS microswitches. In addition, GRC also does all the failure mode analysis, reliability testing, and most of the packaging for the microswitches; as well as most of the other MEMS they R&D and fab onsite. It's quite an impressive operation.

Chris Keimel of GRC presented an abstract on the MEMS microswitch at Hilton Head Workshops in June 2012. His paper created a lot of buzz; and as more folks learn about it I am confident that the buzz will turn into a loud roar of applause (maybe this blog will help; I can only humbly hope so!).

Beyond the research being done at universities and research institutes in the US and abroad, I can think of only one other example (in the MEMS field) of a company doing this level of sophisticated commercialization-focused research; Robert Bosch GmbH (and they are privately-held; GE is not).

But what makes GRC unique is that for MEMS, not only do these guys do the R&D at GRC, they also do the pilot volume fabrication. With their new expanded cleanroom facility (nearly 30,000 sq. ft.); they are doing amazingly cool stuff in MEMS. From what I learned on the full head-to-toe bunny suit tour that I had with fab manager, Ron Olson, GRC is fabricating both silicon carbide based power MOSFETs and MEMS for GE at an impressive rate. By no means is this a high-volume fab; but these guys have a sophisticated operation that is able to accommodate both the R&D and the pilot volume needs of GE's businesses in a single facility.

Lastly, no visit to the facility can be complete without a mention of the beauty of the place. It truly is gorgeous, especially in July (maybe a bit cold in January?). The rate of turnover at GRC is low and I can see why; why would you want to leave this place? You get to work with brilliant scientists, doing great commercially focused R&D, actually manufacture products and live in a beautiful part of the country. As Thomas Edison once stated: "I never perfected an invention that I did not think about in terms of the service it might give to others." Clearly at GRC Niskayuna, his legacy lives on.