During playing of a bowed string instrument, the rosin-coated bow hair is drawn against the string, causing the string to repeatedly stick to and be suddenly released by the hair, a phenomenon known as stick-slip oscillation. These oscillations create vibrations in the bridge and instrument, which radiate sound into the air. For a clear tone to be generated, the stick-slip oscillation must generally occur at the fundamental frequency of the string. It is widely believed that bows strongly affect a violin’s tonal properties. Despite nearly a century of study, how the different tonal qualities of violin bows are related to their physical properties is poorly understood. To investigate the role of the bow’s physical properties in sound generation, we constructed a bowing machine capable of playing the violin with a well-defined velocity, downward force, and position. A MEMS accelerometer mounted to the bow measures the vibrations of the bow. To modulate the bow vibration, we attach a piece of rubber tubing to the bow shaft, increasing its mass and damping. In this manner we test the effect of bow vibrations on the violin sound. We performed these tests in different regimes: (1) Raucous sound due to high bow force, (2) normal sound due to moderate bow force, and (3) surface sound due to low bow force.

We hypothesize that vibrations of the bow influence the stick-slip oscillation of the bow-string system, causing a difference in sound amplitude and/or spectrum. Moreover, a note with a raucous-like sound should exhibit a more complex spectrum than a normal note, and a surface sound should exhibit a higher harmonic to total sound ratio than that of a normal note.

As a result, dampening of the bow using rubber tubing consistently decreased vibrations in all directions. Dampening bow vibrations changed acoustic power and the harmonic ratio in a note-dependent fashion. Vibration shifted stick-slip oscillations in a complex way and its effect on sound depended on the note being played. These observations supported the idea that bow properties indeed influence violin sound. A raucous-like sound showed a smaller harmonic power ratio, and a surface sound showed a higher harmonic power ratio. Future work will focus on understanding how bow vibrations interact with string oscillations. Variation in harmonic dampening with tubing among different bow materials (wooden, carbon fiber, fiberglass) will be studied in the future. Follow-up work will artificially change the vibration of a bow with a speaker to examine how it affects the motion of the string, thereby further testing how vibrations change Helmholtz motion. A future project may address how bow properties affect subjective perception of tonal qualities.