On 09/17/2016 10:56 AM, Stefan Schreiber wrote:
Fernando Lopez-Lezcano wrote:

Hi all,

I've been working on this project since November 2015, and at the time
I thought I would be done by Christmas, hence the subject line... (I
was very naive). The main motivation of the project was to have easy
to build and cheap microphone arrays for my students to use in class
(@ CCRMA, Stanford)...

So, you can choose what you can use this for: Ambisonics themed Xmas
tree ornaments, 3d puzzles of platonic solids, big earrings for your
loved ones or, perhaps, microphone arrays.

I've been working on designs that are 3D printable as flat pieces on
cheap or medium priced printers and are assembled and glued together
like 3d puzzles, starting with a regular tetrahedral first order
microphone and then moving on to Eric Benjamin and Aaron Heller's
Octathingy (8 capsules) and a few more "platonic solid" designs (12
and 20 capsules, these last just to test the concept of even bigger 3d
puzzles - it works).

It is a bit frustrating to read about (probably)  interesting projects
which are< completely unintroduced >.

https://ccrma.stanford.edu/search/node/Octathingy

Didn't find any real info. Is this supposed to be some 2D design? If so,
Orange Research Labs also had some prototype, and also some 12-capsule
(2nd order-3D) prototype but which seems to be to big. (Capsule
distances on sphere implied an spatial aliasing limit of about 4.5kHz,
which would be (too) low for any serious recording.)

So, < before > talking about 3D printed holders: What actually IS an
8-capsule Octathingy?

Sorry about that, yes, the Octathingy is the informal name and it would not be found by a search. As I say, there is a LOT of documentation missing from the pages. I will slowly fix that, thanks for pointing this one out!

Waiting for some enlightening infos...     :-)

I have to add a link and/or the pdf of the paper. Look for the AES Audio Engineering Society Convention Paper 8728 (2012) by Eric Benjamin, "A second-order soundfield microphone with improved polar pattern shape"... the abstract says:

"The soundfield microphone is a compact tetrahedral array of four figure-of-eight microphones yielding four coincident virtual microphones; one omnidirectional and three orthogonal pressure gradient microphones. As described by Gerzon, above a limiting frequency approximated by f c = πc/r, the virtual microphones become progressively contaminated by higher-order spherical harmonics. To improve the high-frequency performance, either the array size must be substantially reduced or a new array geometry must be found. In the present work an array having nominally octahedral geometry is described. It samples the spherical harmonics in a natural way and yields horizontal virtual microphones up to second order having excellent horizontal polar patterns up to 20 kHz."

The capsules are located in the vertices of a square antiprism...

(the reference is also in the paper about the project that I have in my home page)
Best,
-- Fernando


All models are written in Openscad (a 3d modeling programming
language), with most of the dimensions being parametric - the models
are, after all, just software.

What does this mean? "Parametric dimensions" for some 3D printer???

I spent a couple of weeks doing plain old geometry on paper to try to
get everything to fit just right...

I wrote a paper on the progress of the project so far for AES SFC
(which I regretfully was unable to attend), you can find it for now in
my web page - jump to the publications link[*]. I have a first working
prototype (calibration and measurements in the paper), I'm currently
working on two more and looking forward to testing the 8 capsule design.

A lot of work ahead (coding and hardware design, documentation, etc).
This turned out of be a black hole for any time I can throw at it.
Contributions welcome...

GPL Openscad code and Creative Commons licensed 3D models are
available here:

   https://cm-gitlab.stanford.edu/ambisonics/SpHEAR/

(there is also a low volume mailing list available, so far 0 messages :-)

You can also find a Kicad PCB design for the phantom power interface
for each capsule (they fit into the body of the latest design) and the
preliminary calibration software (GPL, written in Octave) for the
tetrahedral design. But of course everything needs better
documentation. Take a look, I included a few more pictures...

If you are tempted to build one be forewarned that it is a LOT of work
:-)

Many in this list helped a lot (you know who you are, thanks!!), I
would not have gotten this far by walking alone.

Enjoy!
-- Fernando

[*] https://ccrma.stanford.edu/~nando/

_______________________________________________
Sursound mailing list
Sursound@music.vt.edu
https://mail.music.vt.edu/mailman/listinfo/sursound - unsubscribe here, edit 
account or options, view archives and so on.

Reply via email to