The Process
Initial Experimentation
How might we plug a drum straight into an amp?
In the initial test phase, I experimented with a practice pad by attaching a guitar pickup with a spring screwed onto the surface. The hypothesis was that each drum strike would cause the spring to vibrate, producing a sustained sound, which could be adjusted by tightening or loosening the spring. To compare results, I also connected a contact microphone to capture the sound alongside the guitar pickup.
Next, I focused on improving the build and started testing both the contact microphone and guitar pickup through a mixer. I used both drumsticks and brushes to explore the variety of sounds produced. I discovered that the two microphones captured complementary sound characteristics: the contact microphone delivered a short attack and decay, while the guitar pickup provided a medium attack with a longer decay.
To blend these distinct sound profiles, I researched and decided to integrate an MN250K blend potentiometer, commonly used in guitars and basses with two pickups. This allowed me to explore different ways to harmonize the output of the two mics. Finally, I moved on to soldering the components to enhance the setup and prepare for further testing.
Prototype 1
How might we make a plug-and-play drum with an analog signal?
I began by modeling the body in Rhino, focusing on the internal structure and precision of the design. I sourced Sapele wood for CNC routing and explored the idea of using a mold to achieve more accurate cuts. I modeled the internal structure in greater detail and proceeded to laser cut various components.
To achieve the desired aesthetic, I sandblasted the metal frame to give it a starry appearance. Manually modifying the frame allowed for easier attachment of the spring, while using tape helped eliminate unwanted rumbling caused by metal-on-metal contact. I also welded the frame for the guitar pickup and spring to ensure stability.
The next steps involved CNC'ing the body and gluing two parts together, followed by laser cutting a new variation of the metal structure—though one attempt at laser cutting did fail. After assembling the body, I reviewed any gaps from the gluing process, sanding, and cutting inserts to fit the metal structure properly.
With all components ready, I assembled the springs and soldered the electrical parts, finalizing the build. For the first test, I sought feedback from Dru Masters, a composer known for his BAFTA-winning work on the drama Damilola. Dru has a broad understanding of composition and audio production, with his own studio filled with instruments and effects, making him well-positioned to offer valuable feedback.
He noted that Version 1 of the drum was effective for FX in film as well as emulating concert percussion instruments like timpani, toms, and taiko drums. Dru highlighted the drum’s ability to plug directly into guitar effects, but suggested that more control over the tone using the knob would improve the instrument’s versatility. The raw sound of the contact mic and spring mechanism wasn’t as polished as I had hoped, prompting me to explore digital signal processing (DSP) to enhance and expand the sound palette beyond tom-like sounds.
Prototype 2
How might we leverage a Digital Signal Processor (DSP) to expand the Messier 57 drum's sound palette and give users greater creative control?
For this phase of the Messier 57 drum project, I decided to incorporate digital signal processing (DSP) to refine the sound. I chose to use a Bela Mini, as it supports PureData audio programming patches and offers very low latency, which is crucial for a drum-based instrument.
I began by programming a PureData patch and learning how to integrate it onto the Bela Mini, adding a potentiometer to alter the sound. Since drummers have their hands occupied while playing, I aimed to offer customizable sound through a simple, single-knob interface. After experimenting with different sound characteristics, I settled on Sustain, as drums typically don’t have variable sustain like other instruments, such as keyboards or guitars. This feature would be novel to test in a percussive instrument.
For user feedback, I tested the drum with Ayo Salawu, a touring drummer for the band Kokoroko, known for blending acoustic and electronic sounds. Ayo appreciated the drum's potential and offered insightful feedback. He felt that variations of the drum, each with different materials and sound profiles, could be collected like snare drums. He wanted a sharper, quicker sound from the drumhead, as the prototype mainly produced sound from the rim. He also enjoyed using the knob to alter the sound, though he suggested that the control should be more intuitive and clearer for performance use.
Prototype 3
How might we create a drum that responds naturally to a drummer's hits, while providing intuitive, momentary control over sound adjustments?
&
How might we design a drum that expands sound possibilities while feeling natural to play and works seamlessly in live performances?
In the final phase, I encountered a significant challenge: the sound of Prototype 2 did not align spectrally and timbrally with the drummer’s hits. The main issue was the guitar pickup, which captured vibrations from the rim through a spring. It only picked up low frequencies with low energy, as the spring couldn’t vibrate at higher frequencies. To address this, I revisited the problem from first principles and opted to place a Lavalier mic under the drumhead, paired with a noise gate controlled by a contact mic on the drum body. This setup fed into a PureData patch with a physically modeled snare drum, excited by the full sound spectrum captured by the Lavalier mic. This solution resulted in a more natural and dynamic sound.
Ayo also faced difficulty quickly adjusting sounds with the small knob, frequently losing his previous settings. To improve this, I experimented with spring-loaded mechanisms, starting with a quick hairpin prototype. I then progressed to a laser-cut, plastic-welded acrylic model with an external spring, and finally, a hand-fabricated brass version with an internal spring mechanism. This design allowed for momentary sound control, which reverted back to the original value, enabling drummers to make rapid, artful adjustments without needing precision.
However, the Bela Mini struggled to handle the CPU-intensive physical modeling patch. As a result, I am now considering a new solution: a headless Windows computer running a Max/MSP patch, which offers greater versatility and customization.
To be continued!