To prove their concept, the researchers designed three individually addressable groups of actuators on a single chip, as 14 pairs of long actuators in a 2-serial-7-parallel configuration. To increase the fill factor of the design, the researchers filled the remaining space to the edge of the rectangular shaped device with actuators of half the length.
Here the whole MEMS chip is under 1mm thick, with 400μm thick top and bottom capping wafers sandwiching the 75μm SOI layer and its patterned actuators. The active area of the device had a footprint of 9.3mm2. What the authors highlight is that since the height of the actuators equals the SOI layer thickness, it specifies the displaced volume for a given lateral actuator displacement. One could tune this height to increase the sound pressure without affecting the footprint of the device. What’s more, the MEMS can easily be mass fabricated using current silicon bulk micro machining processes.
The researchers at Fraunhofer IPMS anticipate that such tiny MEMS speakers could find their way in internet-enabled mobile terminals worn directly in the ear. This would be for applications such as simultaneous translation, payment services and marketing solutions, in effect, porting today’s voice-based home assistants into wearables. Such hearables would be predestined for what imec describes as the Internet of Voice, to be worn permanently in the ear, replacing bluetooth headsets.
The researchers are already working on improving the MEMS speaker for better linearity and higher sound pressure levels while reducing power consumption. Next, they will want to integrate the MEMS with its control electronics.
Fraunhofer IPMS - www.ipms.fraunhofer.de