The Making of Moogfest 2017: How Bronto Built MIDI Controllers

Clif Jackson, Software Engineer

Clif Jackson, Software Engineer

This is the fourth article in a four-part series about how Bronto employees collaborated to transform raw, anonymized data into an interactive audio and visual installation for Moogfest 2017 called The Sounds of Commerce.

One reason I enjoyed participating in Bronto’s Sounds of Commerce is because I got to know new co-workers. Such was the case when fellow Bronto Ronnie Angerer approached me and told me he was interested in helping build controllers for our installation at Moogfest. After some discussion, it became clear he had experience building do-it-yourself (DIY) electronics and could really make the project his own. That was great, since this project was all about hands-on fabrication.

We agreed we wanted to build controllers that would have a steampunk-like look and feel. We didn’t want something store-bought. I started by passing around rough sketches of controller designs (like the one shown) that would make it possible for festival-goers to manipulate and interact with the sound and visual elements in our display.

Once we agreed on some basic ideas, I went with that and came up with five different controllers: four synthesizer controllers and one drum machine. With my background in industrial design, it was relatively easy for me to come up with different designs and execute them. I used Adobe Illustrator to create the cutting guides for different components. I found and ordered all the parts and then measured them to make sure the holes were correct and spaced properly on the boxes.

Then, I printed out the designs and used those as cutting guides for the bare, wooden boxes I bought. We had to overcome one pretty big hurdle: we needed a drill press to cut the holes. Luckily, a coworker had a friend who let us borrow his equipment. A drill press was critical, as a hand drill could really chew up the soft wood of the boxes.

Next, we started to get the circuitry together and test some of the components we purchased. I decided to go with the Teensy 3.2 for the microprocessors. It was a toss-up between the Teensy and the Arduino Uno. We weighed the pros and cons of both.

The Teensy can handle many more inputs than the Arduino, and it’s a lot smaller. It can also support 2 or 3 independent circuits because it has multiple power pins and grounds. However, it is dedicated USB, which limits flexibility. The Ardunio can output to 5-pin DIN or to USB depending on what you build. Moreover, you can simply not have MIDI input at all. With the Teensy, however, dedicated USB means both MIDI in/out all the time. This can be problematic under certain circumstances. We didn’t need MIDI input for these controllers, so we had to ignore any incoming MIDI signals in the Teensydunio sketch.

Soon, it was time to get the bread board and alligator clips out.  Fortunately, I have experience with soldering from wiring up my electric bass guitars. But I had never worked with a microprocessor, so I had to navigate a learning curve to understand how to program the Teensy. Fortunately, the Arduino IDE provided some great examples. After a few weeks of tinkering and learning, we were finally able to put some controller kits together.

As soon as I gave Ronnie the first kit, he asked for a second one. His excitement about the project was terrific. This was an exciting time in our timeline because we had to shift our efforts into high gear. Around Bronto, we felt momentum about the event building, but we still had a lot of work to do.

We were building and mapping controllers right down to mere hours before our Moogfest exhibit was scheduled to open. For me, the biggest challenge was the touch sensors. We found these “soft pots” which are theoretical equivalents of analog linear and rotary potentiometers.

What I didn’t realize when I ordered them was that they are not actual equivalents in practice. They work fundamentally differently than their non-touch counterparts and documentation is sparse at best. I really wanted these to work but was having problems with the floating values and getting the circuit right. Without enough time to research and prototype, I ultimately had to leave them out of the display.

(Later on, I followed up on this and figured out that you need to split the data line from the soft pot. One end goes normally to the Teensy and the other goes through a variable resistor (potentiometer or trim pot) to ground. This allows you to dial in and calibrate the soft pot and resolves the floating problem.)

I was really excited when Ronnie showed me the first pictures of the controllers he’d built. He had taken this part of the project and made it his own! And that felt fitting, because this project was so much about individual talents coming together to become more than the sum of their parts.

The controllers were a big hit. Festival-goers from around the world stopped by to see the display, ask questions, and lay down beats with the controllers. Musicians dropped by to ask about how we’d mapped the data to the controllers. Even a few representatives from Arduino visited our display and were delighted to see that we’d used their technology in our installation.

Overall, we were pleased with how the entire installation turned out. We’ve even talked about taking our installation on the road to share it with customers at our Bronto Spotlight events this Fall. One thing is certain: this project was an exciting way to showcase the sheer volume of the data we manage and the incredible creative talents of the Bronto team.