MEMS speakers (Joe Hayes)
(too old to reply)
A3D Sound
2018-01-23 23:14:00 UTC
That would be pretty cool - an adaptive system where for lower frequencies
you use more audiopixels,
for higher frequencies you use less, and of course the higher the
frequency the more accurate localisation would be.
best, Justin
Justin Bennett
We want to use mems speakers in an array under the A3D algorithm. By using
say a 7 x 1 array of memes speakers you will create a highly energy
efficient speaker with maximal bass. This could be a TV or automotive
tweeter and use a normal piston bass driver below the frequencies the mems
cannot reproduce.

An A3D array can be either one or two dimensional and is based on any prime
number you like. So an audio pixel mems is 1cm x 1cm and has 32 columns
and rows of pixels. You could use a prime number like 31. I would A3D in
the lateral only. It could be 31 x 31 mems speakers but the vertical
columns would be all in the same phase/wiring. The number of channels
required is RND(Prime/2) = RND(31/2) = 16 channels. The bandwidth is
limited by the mems width + spacing in between. Say 1.2cm. So top end
wavelength = 1.2cm x 2 = 2.4cm say 14.3kHz. The bottom end of bandwidth =
top end/(prime-1) = 14.3kHz/(31-1) = 14.3kHz/30 = 476Hz. So you create a
great mid and high range driver (476Hz to 14,300Hz) with
perfect unidirectionality using a prime = 31 of mems drivers.
Joe Hayes
Acoustic3D Holdings Ltd
ACN: 150521092
Mb(AU): +61 423 172 933 <//423%20172%20933>
Mb(CN): +86 1 <//157%206763%209176>32 438 52412
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.music.vt.edu/mailman/private/sursound/attachments/20180123/fd7d448b/attachment.html>