Silicon speakers are ready for volume production

MEMS technology is a transfer function, capable of transfering “bulky” technologies into micromechanical Si elements. The two big moves of late have been MEMS’ transfer into the world of silicon microphones (by Knowles) and oscillators (led now by SiTime), which generated more than $1B last year.

The next big transition is happening now: for the first time, silicon speakers are ready for volume production, enabling the creation of a brand-new multibillion-dollar market for MEMS manufacturers. The last edition of Yole Développement’s best-selling report, Status of the MEMS Industry, mentioned that this market would arrive in 2017 – and here we are.

The transition into silicon speakers is led by USound GmbH, an Austrian company founded three years ago by several veterans of the MEMS industry. We recently spoke with the management team about their plans and expectations for this lucrative new opportunity.


MEMS Atransfertechnology Usound June2016

Yole Developpement (YD): What are the origins of USound?

USound: The founders are veterans of the MEMS industry and have an extensive knowledge of the market and the technology trends. During the creation of company, they identified the emerging field of MEMS speakers as one of the most promising. Thanks to the good relationships that the three founders built over the years with several research institutes, it was possible to gain insights into available but not-yet monetised technologies. A suitable patent and technology portfolio was identified, and an exclusive license taken. Once the application field and the technology were selected, the next challenge was securing knowledge and expertise in the acoustics-field. Luckily for USound, in the summer of 2014, a major supplier of micro-loudspeakers in Vienna substantially reduced its workforce. This enabled USound to hire very experienced acoustic engineers. Shortly thereafter USound hired a group of young engineers graduated IEM, a renowned University for acoustic engineering from Graz, Austria. The company was thus quickly able to assemble a strong team with expertise covering MEMS, operations and acoustics in its two offices in Austria and one in Germany.

The founders were able to convert their idea into practice thanks to the strong help and support of three famous high-tech industry veterans. Greg Galvin, Ignazio Rusconi-Clerici and Herbert Gartner are USound’s non-executive board members. They respectively founded the MEMS company Kionix, acquired by ROHM for $ 233 million, IRMS, later known as Hyperion Solutions, ultimately acquired by Oracle for $ 3.3 billion, and SensorDynamics, a MEMS and wireless sensor company acquired by Maxim Integrated for $ 164 million in 2011.

YD: Can you give us a quick overview of USound GmbH in terms of its background and technology?

USound: The three founders met in 2009 at SensorDynamics. Two years after Maxim Integrated acquired the company, they decided to try the start-up experience for themselves, and set off on that adventure. As Yole mentioned, speakers are one of the last mobile phone elements yet to be transformed into semiconductor/MEMS, so the founders seized the opportunity and set off to work on that technology. The initial idea of building directional speakers was soon abandoned after discussions with the first alpha customer. The technology was simplified to develop a balanced-armature replacement. This market is a very nice fit, as it only requires small MEMS (perfect for NPI) and supply is completely constrained due to manufacturing processes which are closer to mass handcrafting than HVM.

A second pivotal event was the discovery that the MEMS actuator displayed good performance in the ultrasound region, so a micro-tweeter useable for proximity sensing and positioning was produced. This potential space-saving, which should also help improve smartphone aesthetics, should be of great interest to the customers.

YD: What is the status of your technology and product development?

USound: Prototypes of the first balanced-armature replacement and the first micro-tweeter are currently being sampled to selected customers. Pre-production will start at the end of the summer, along with internal qualification. The technology is ready to for adoption, and will revolutionise the personal-audio market, just like what happened with the MEMS microphone.


Usound Yole Article June2016 products

USound micro-loudspeakers (Courtesy of USound)

YD: What is the key added value of USound’s technology?

USound: USound brings MEMS technology to the speaker world. Similar to what happened with accelerometers, microphones, and pressure sensors, the same standard advantages of MEMS over traditional technology will occur: + Smaller actuator size/weight, for lower power and higher reliability + Batch processing, for lower volume pricing and lower product variation + Higher-performance materials and architectures, for lower power consumption + Electronics integration, for more intelligent products

YD: What is your technology’s competitive advantage over a current smartphone’s micro speaker, other MEMS speaker developments, and the bone conduction transducer used by Google Glass, for example?

USound: As mentioned earlier, the speakers can efficiently operate all the way to the ultrasound band. This higher-frequency operation is generally a weak point of receivers, pushing manufacturers to implement complex solutions to improve quality of sound reproduction. Additionally, control electronics can be integrated straight into the package, thereby further increasing performance without size penalty. The principle is based on analog audio signals, so the complex audio processing required in competing MEMS technologies is avoided. USound has already produced prototypes for personal audio (virtual headphones) that provide great sound quality to earphones and are transparent to the outside world, but without the sound quality loss commonly associated with bone transducers.

YD: In terms of key materials, you’re using piezoelectric material (PZT), right?

USound: Yes, and we are also testing other materials in order to prepare next-generation products with further-improved features and quality. But the main material is currently PZT.

YD: Is it true that, through disassociation of the vibrating structure and the membrane which is amplifying the soundwave, the packaging structure is much more simplified, and so too the integration of the driver within the package?

USound: That is correct. The packaging integration is now very simple and easy to optimise. Normally, a manufacturer looking at increasing sound levels with a larger membrane would need to increase the magnet size, which reduces space available for back-volume, which in turn requires a larger magnet to compensate for the loss of sound level, and so on and so forth. By comparison, the MEMS solution enables independent tuning of the actuator and the membrane dimensions to reach customer specifications. Additionally, the large membrane is very cheap (costing about one cent to produce), meaning that silicon area can be minimised, leading to optimized costs and yields.

YD: And today, you have solved most of the problems in order to move into volume production?

USound: That’s correct. The actuators are ready for implementation. In order to facilitate use by our future customers, we will also develop a dedicated amplification solution. Basically, the MEMS needs higher voltages than an electrodynamic driver. Currently-available amplifiers for piezo-actuation are not fully optimised for mobile applications, that’s why we have decided to develop the amplifier by ourselves. We have already tested a couple of concepts, and right now we are testing a concept that’s very simple, and which will be available at the beginning of next year.

YD: How do you think you will compete with existing technologies which are established, and better adapted in terms of cost? What you’re bringing is better performance, better integration, and lower cost, isn’t it?

USound: Yes, you’re absolutely right. What we have right now, is able, in many ways, to address the ear speaker and the speaker for mobile phones, which are two different markets. We will first attack the earphone or earbud market, where we offer better performance in terms of acoustics and power consumption. We are also offering much better cost structures and scalability in comparison to existing technologies. Once we have successfully entered the earbuds market, we will move to mobile phones, and there we will definitely offer better performance, lower power consumption, lower cost, and much more functions. MEMS has matured a lot over the last years, really enabling a fabless business model which maybe five years ago or six years ago was not the case. So we are very carefully working with our partners in the supply chain to build a fabless business. We are addressing the complete personal-audio market, and are challenging the incumbent technology for a piece of this multibillion pie.

YD: I suppose you’re already working with your future customers. How would you describe their interest? Are they seeing this as a nice add-on? Is it perceived as a breakthrough technology? What is their feedback?

USound: We would describe their feedback as mixed optimism. Some would like to see more data before committing to a new technology. At the same time, others are very excited that somebody is finally disrupting the audio-transducer market. They know that microphones have brought quite a few nice things to the market, so the potential is clear for similar gains and advantages to be provided to the speaker world. Customers are currently evaluating the parts and figuring out what will be the best fit for them. But right off the bat, considering the fact that our first product is a replacement for balanced-armature receivers, they think that there is tremendous potential to solve the current constraints on the supply chain, and in general are checking what that means to them in term of organization and product optimisation.

YD: In terms of business model, if we make the parallel between what happened in 2003, with the introduction of the first silicon microphone by Knowles, do you have a strategy to step out of the device business, take benefit of the technology to leverage sales, and move away from very high price pressure by delivering more value/functions? Have you chosen your business model? Will you be a device maker, or do you think you can also leverage your technology by selling it to companies that are already in the speaker business?

USound: Though we can serve the market as a device supplier, we want to extend our offer by selling not just the speaker, but the speaker together with the amplifier and in future also the codec and the DSP. We also plan to add firmware to develop unique features. That’s our strategy. We know that there are some specific companies, which represent a nice market, that want to develop their own products and will just buy the loudspeaker from us. So our main business model would be closer to audio-solution suppliers like Knowles, but considering that there aren’t that many MEMS speakers suppliers, we wouldn’t mind getting some business by offering it on the side, for example similar to Infineon’s business model in microphone.

YD: So that means you’re confident about adding all the environments in order to sell functions to your customers?

USound: Yes. We’ve seen how quickly the product gets commoditized; if you look at microphone or accelerometers, they’re worth just twenty-something cents. The money is made in the sensor fusion between all the environments. We want to be one step ahead, or even two, three, four steps ahead of the competition.

YD: Sound quality and power are big challenges for MEMS-based speakers. How do you solve these issues?

USound: USound approaches these aspects from two sides. Internally, our various teams (acoustics, MEMS, and semiconductor) are pushing the limits of the current technology, with support from alpha customers. Externally, partnerships with leading research institutes will give us access to new materials and architectures of even higher performance. We have applied for and been granted a total of twelve patents specifically in the field of sound quality.

YD: Do you have a road map of products or specifications you can share with YOLE?

Usound Yole Article RoadmapOK June2016.jpg


YD: Which applications is USound targeting?

USound: Our current focus is on mobile applications, with high-end earphones as the first target. As our technology matures, smartphones and portable speakers will be addressed. And of course, now that there’s interest in the fingerprint sensing world of ultrasonics, this is something that we are also considering.

YD: Aside from the obvious outside consumer market, do you see other opportunities in medical, automotive, smart homes, industry, etc.?

USound: Since USound is still a young company, we have decided to concentrate on consumer audio. We’ll address additional markets as the company grows and the technology matures.

YD: Do you already have partnerships with potential integrators?

USound: Various OEMs are currently evaluating our products.

YD: When do you plan to ship the first products?

USound: With internal qualification on pre-production units starting post-summer, we expect to ship our first production units in Q2 2017.

YD: What are the next steps for USound’s product: reducing cost, increasing performance, adding intelligence, combining with other sensors/transducers?

USound: Of course, USound will work on all of those aspects. The company’s primary focus is on achieving the best performance, but the most dramatic factor for the industry is our opportunity to add sensors to the actuators for self-monitoring, which could also be used to detect other physical effects.

YD: Do you plan to sell devices or more complex functions?

USound: More complex functions like equalization, 3D audio, etc. will be developed and marketed.

YD: What are your long-term objectives?

USound: USound will evolve into an audio-system developer, offering complete solutions spanning from hardware to firmware, in order to simplify technology adoption and help our customers achieve optimum product performance. We are learning a lot in collaboration with our partners and prospects, and we know that in order to have a new technology, you need to be twice as good at the same price. We’ve demonstrated that the technology works, and now our efforts are focused on entering the markets.

You want to know more ? Discover Yole Développement recent report Status of the MEMS Industry

Couv MIS

Amid a commodization paradox, Chinese players taking off, sales stabilization and new applications, how can the MEMS industry regenerate value?

Key features of the report:

•Explanation of MEMS market dynamics
•Application landscape
•Industrial landscape and player evolution
•Focus on China
•Technology trends
•New devices





Source: / 

Related presentations

Cet article vous a plu ?

Partagez-le sur vos réseaux sociaux