DIY Network-Ready Polyphonic Music Controller

Hans Peter Portner’s Chimaera project is a touch-less, expressive, network-ready, polyphonic music controller released as open source hardware. It is a mixed analog/digital offspring of the Theremin. An array of analog, linear Hall effect sensors make up a continuous 2-D interaction space. The sensors are excited with Neodymium magnets worn on fingers.

Portner's Chimaera project

Portner’s Chimaera project

The device continuously tracks and interpolates position and vicinity of multiple present magnets along the sensor array to produce corresponding low-latency event signals. Those are encoded as Open Sound Control bundles and transmitted via UDP/TCP to a software synthesizer. The DSP unit is a mixed-signal board and handles sensor read out, event detection and host communication. It is based on an ARM Cortex M4 microcontroller in combination with WIZnet W5500 chip, which takes care of all low-level networking protocols via UDP/TCP.

First Prize — Chimaera: The Poly-Magneto-Phonic Theremin, Hans Peter Portner (Switzerland)

The poly-magneto-phonic Theremin

In his project write-up, Portner explains:

With its touch-less control (no friction), high update rates (2-4 kHz), its quasi-continuous spatial resolution and its low-latency (<1 ms), the Chimaera can react to most subtle motions instantaneously and allows for a highly dynamic and expressive play. Its open source design additionally gives the user all possibilities to further tune hardware and firmware to his or her needs. The Chimaera is network-oriented and configured with and communicated by Open Sound Control, which makes it straight-forward to integrate into any setup.

The hardware of the Chimaera consists of two types of printed circuit boards and an enclosure. Multiple sensor units are daisy-chained to form the sensor array and connected to a single digital signal processing (DSP) unit.

Sensor unit

Sensor unit

A single sensor unit consists of 16 linear hall-effect sensors spaced 5mm apart and routed to a single output through a 16:1 multiplexer which is switched by the DSP unit. Downstream the multiplexer, the analog signal runs through an amplification circuitry.

A modular hardware design consisting of identical sensor units and a single DSP unit embedded in a wooden case allows building devices with array sizes of 16-160 sensors.
A modular hardware design consisting of identical sensor units and a single DSP unit embedded in a wooden case allows building devices with array sizes of 16-160 sensors.

The DSP unit is a mixed-signal board and handles sensor read out, event detection and host communication. It is based on an STM32F303Cx ARM Cortex M4 microcontroller in combination with WIZnet W5500, a hardwired 100Mbit IPv4/PHY chip taking care of all low-level networking protocols via UDP/TCP. The board’s analog part features 10 analog inputs providing connection points for the sensor units, leading to a maximally possible array of 160 sensors. Those analog inputs connect directly to three in parallel running 12bit analog-to-digital converters.

Schematic of the DSP unit (STM32F303Cx part)

Schematic of the DSP unit (STM32F303Cx part)

Networking technology in a zero configuration setup has advantages in respect to long-distance transmission, operating system independence and inherent ability for network performances. We thus use the Open Sound Control (OSC) specification via UDP/TCP as low-level communication layer.

Schematic of the DSP unit (WIZnet W5500 part)

Schematic of the DSP unit (WIZnet W5500 part)

Portner’s project won First Prize in the WIZnet Connect the Magic 2014 Design Challenge. The entire project and its associated files are now available.

Microcontroller-Based Markov Music Box

Check out the spectrogram for two FM notes produced by FM modulation. Red indicates higher energy at a given time and frequency.

Cornell University senior lecturer Bruce Land had two reasons for developing an Atmel AVR micrcontroller-based music box. One, he wanted to present synthesis/sequencing algorithms to his students. And two, he wanted the challenge of creating an interactive music box. Interactive audio is becoming an increasingly popular topic among engineers and designers, as we recently reported.

Land writes:

Traditional music boxes play one or two tunes very well, but are not very interactive. Put differently, they have a high quality of synthesis, but a fixed-pattern note sequencer and fixed tonal quality. I wanted to build a device which would play an interesting music-like note sequence, which constantly changed and evolved, with settable timbre, tempo, and beat… To synthesize nice sounding musical notes you need to control spectral content of the note, the rise time (attack), fall time (decay), and the change in spectral content during attack and decay.  Also it is nice to have at least two independent musical voices. And all of this has to be done using the modest arithmetic capability of an 8-bit microcontroller.

Land’s students subsequently used the music box for other projects, such as an auto-composing piano, as shown in the following video.

In early 2013 Circuit Cellar will run Land’s in-depth article on the Markov music box project. Stay tuned for more information.