Are MEMS speakers ready to kick out the voice coil?
Vigna said the transition might take a “few years” and declined to say whether ST is planning to enter the market directly with such developments as autofocus units or MEMS speakers. However, ST is making sure it has the building block technology and manufacturing capabilities in place and is working with startup companies.
Vigna was speaking on the fringes of the European MEMS Summit organized in Stuttgart last week by SEMI. He cited the example of an alternative automatic focusing system being offered to the market by PoLight AS (Horten, Norway), a developer of piezo-MEMS auto-focus lens (see PoLight raises another $19 million ahead of IPO).
ST is PoLight’s manufacturing partner. “We are starting to see piezo-MEMS for autofocus and then it will migrate to optical image stabilization,” said Vigna. Vigna said that as smartphones start coming with multiple image sensors and use multi-framing and computational image reconstruction, this will likely play to piezo-MEMS actuators’ rapid response time.
Another example is the MEMS loudspeaker being worked on by USound GmbH (Graz, Austria) and AudioPixels Ltd. (Sydney, Australia). Vigna declined to say whether ST is manufacturing for USound. AudioPixels has selected Tower Semiconductor as its manufacturing partner (see Tower selected to make MEMS loudspeaker).
When asked if ST was preparing to enter the market place with MEMS-based speakers as a complement to the company’s strong position in microphones, Vigna declined to comment directly.
But in his keynote speech at the European MEMS Summit Vigna had spoken of the importance of piezo-actuators, microfluidic MEMS, and moving mirrors. “The next two or three years are very important and we are trying to find the right applications,” he said.
Next: It’s all about the bass
Speaking later on in the conference program Andrea Onetti, general manager of volume MEMS and analog at ST, had also stressed the importance of micro-actuators in general to ST’s diversification strategy as it creates a broad portfolio of technologies to ensure that ST catches whatever aspects of the Internet of Things that go to high volume. On the actuation side that includes piezoelectric, electrostatic and magnetic methods.
There are arguments about the efficiency of MEMS speakers and whether than can scale to larger room speakers or down to in-ear and headphone applications.
Generally balanced-armature speakers are used in headphones. However Onetti argued that there are applications where magnetic coils used to drive diaphragms present interference problems. Also the rapid response of MEMS speaker arrays can be well matched to output of MEMS microphones and this is useful in noise-cancelling applications.
“The transition of precision mechanical engineer to MEMS will happen,” said Vigna. “But you don’t simply replace one loudspeaker with a MEMS speaker. You have to engineer the complete solution.”
Andrea Rusconi Clerici, co-founder and CEO of USound also spoke at the European MEMS Summit claiming that USound’s MEMS speakers could provide 50 percent lower power consumption than voice-coil actuator speakers for portable and headphone applications. In volume they could provide a 40 to 80 percent cost reduction, he said. “For earphones we are comparable in power consumption with balanced armature,” Rusconi Clerici told eeNews Europe. USound’s MEMS speaker is based on lead zirconate titanate (PZT) piezoelectric material.
Rusconi Clerici said USound was keeping its focus on improving and reducing the cost of analog sound production and not trying to do digital sound wave construction.
Another advantage for a MEMS-based speaker is its reduced-height profile compared with voice-coil and balanced-armature speakers although initially USound is going for a drop-in replacement sale. The company has two parts sampling – Moon and Ganymede – that are aimed at the smartphone motherboard and headphones, respectively.
USound’s three founders all worked at SensorDynamics and left after it was taken over by Maxim Integrated Products Inc. Two of the founders – Rusconi Clerici and Ferruccio Bottoni had previously worked at STMicroelectronics.
Next: Declines to say
Rusconi Clerici declined to say who was manufacturing for USound. X-Fab Semiconductor Foundries is listed as a USound partner and its website does indicate sputtering of PZT layers as a capability. Two other foundry operations in Europe that can put down PZT piezoelectric layers are China-owned Silex Microsystems (Jarfalla, Sweden) and STMicroelectronics.
Rusconi Clerici also highlighted other potential advantages of the MEMS speaker that are part of USound’s R&D program. These include the potential to use MEMS arrays for audio beamforming and localizing audio in space, for use in ultrasound applications, and to use the same MEMS membrane as microphone element and as a diaphragm. Although it might be thought that the component optimizations might have to be different depending on the microphone or microspeaker function Rusconi Clerici said: “Speakers could work as microphones, eventually. We have IP on this.”
Related links and articles: