This is a repository for the scanning microphone setup used as a possible way of imaging a phononic topological metamaterial. We are essentially using a CNC controlled microphone to record audio samples from different locations in our acoustic metamaterial (similar to the StM). This allows us to get high-resolution images of the acoustic dynamics of our system.
In this example we scan 28 kHz for an acoustic metamaterial. We expect modes down the boundary in the center (the region between thick and thin pillars) to be topologically protected and everything else in the bulk to insulate. Here is what the temporary setup looked like (all of the parts for this initial test were just borrowed from Science Center 102).
Then, after scanning with a microphone that could pick up the 28 kHz sine wave we broadcasted, we scanned a rectangular slice above the metamaterial to see what is going on.
We see that the sound is conducting through the middle, and getting insulated everywhere else! Nice!
As seen from the picture, our setup is divided into a few major sections, all of which have to be connected to your computer.
- CNC Machine (Setup Instructions)
- Microphone (Setup Instructions)
- Signal Generator (Setup Instructions)
The two main types of CNC machines that I experimented with was a 3D printer and a CNC mill. I felt like the 3D printers were safer to use, since you can send commands to turn off the heat and extruder. The setup instructions can be found here.
The two main types of microphones that were used were ones connected to the serial headphone jack, and oscilloscopes connected directly to the computer. The oscilloscopes were used because some of the frequencies used went over 22 kHz, which normal sound cards don't handle. The setup instructions can be found here.
There are also control scripts for a RIGOL Signal Generator. This is useful if you want to vary the sound waves you send in or scanning different frequencies as you go along. The setup instructions can be found here.
The main library that you use for controlling everything and scanning will
be scanner.py.
I find the easiest way is to use an ipython terminal and do everything
from there interactively.
Some troubleshooting and small starting example can be found
here.
There are a few different ways for performing the measurements, so I just documented the few ways I tried measuring. Some of the notebooks are available here.
For any inquiries about the setup or packages, contact
wfu (at) college.harvard.edu. Will be happy to help if there are any questions
about how to set up something similar.