10th Speech in Noise Workshop, 11-12 January 2018, Glasgow

Biological inspired MEMS acoustic sensors

Andrew B Reid, Yansheng Zhang(a), James FC Windmill
University of Strathclyde, Glasgow, UK

(a) Presenting

We will present our current work on microelectromechanical systems (MEMS) acoustic sensors for applications ranging from novel hearing aid concepts to sound localisation in defence contexts. The developed MEMS microphone designs are inspired by the hearing mechanisms of two insect species, Ormia ochracea & Achroia grisella, which both have exceptional potential to resolve directional hearing cues. The general sensing concept is based on two mechanically connected membranes in which the two base resonance movement mode shapes, a tilt and bend around the connection, combine to create an apparent increase in resolving directional cues. To replicate this behaviour in an engineering solution we designed a range of silicon MEMS microphones with piezoelectric readout and design considerations looking at a low frequency coverage, a distinct resonance at 8 kHz and a broadband frequency coverage below 10 kHz. The fabrication is done using the cost effective multi-user silicon-on-insulator process PiezoMUMPs, commercially offered by Memscap Inc. The process uses a 10 ┬Ám thick single crystal silicon layer for the device structures and a 500 nm thick layer of aluminium nitride as piezoelectric active material, which is used as main acoustic sensing principle. Three current families of devices will be presented which each have a footprint between 1 mm x 2 mm and 1.2 mm x 3 mm. The mechanical and acoustic sensitivity characterisation of the devices will be presented, showing their directional acoustic response and good agreement with analytical estimations and acoustic-structure interaction finite element analysis simulations. Further steps to include the influence of the physical mounting and environment of the MEMS chips on the directional acoustic response of the microphones will also be shown.

Last modified 2017-11-17 15:56:08