Undergraduate Research

Violin Playing-in Experiment

The main goal of my undergraduate research is to quantify the variation in mechanical and acoustic response of violins as they are played, an extremely common belief among violin players and violin makers that is not backed by any substantial body of research on the subject. 

An important part that has been missing from past experiments attempting to prove or disprove this phenomenon is an analysis of the uncertainty in response measurements. It is ontological to any highly-responsive and sensitive system that uncertainty is analyzed before the results of a change in response are recorded. Yet most studies of violin playing-in either omit this step, and the studies of violins that include uncertainty measurements are not specifically measuring the effects of playing the instrument over time. 

I have completed this step using a Laser-Scanning Doppler Vibrometer (LSDV) and microphones in an anechoic chamber, presenting my first set of measurements at the 182nd meeting of the Acoustical Society of America in Denver, Colorado. The article for this presentation can be read here. This was done through the use of Polytec software in the vibration lab, LabView NXT to process the signals in the anechoic chamber, and MATLAB programming to perform the calculations and display the data as necessary. Figures of various datasets can be viewed below.

Initial Uncertainty Analysis

Big graph showing all of the data taken

Each of the lines on the graph is a separate measurement I made with the LSDV. These measurements would be characterized as the "mobility" of the violin at a specific point. You can see that the units are dB over frequency, so this is actually a Fast Fourier Transform (FFT) of the vibrational signal detected by the LSDV. 

Graph showing only relevant range of data

To put it simply, FFT is an efficient method of analyzing sound signals with a stupidly high amount of clarity. The peaks you see here correspond to frequencies that are more common than others in the vibration of the point after the violin is tapped. Despite being the most accurate method of measuring violin response, this method is not well-researched, so efforts were made to reduce uncertainty in this process.

Reducing Uncertainty

At the 184th Meeting of the Acoustical Society of America, I presented the results of my reduction in uncertainty with the previously mentioned measurements, as well as the results of various experiment, such as the effect of hammer position, laser position, moisture content, and string tension on the response of the violin. The slides for this presentation can be found here.

Overall, I managed to reduce the uncertainty of the measurements by about 60% by keeping the hammer and laser positions more consistent, as well as increasing the signal-to-noise ratio. These procedures will be used in future measurements to better evaluate the results of various experiments.

Results

The lay version of the product of my research can be found here:

https://acoustics.org/busting-the-myth-that-new-violins-sound-better-after-a-period-of-playing-in/

The result of stimulating the violins through mechanical vibration and simulated playing over a 1600-hour period was, essentially, nothing. There were no meaningful differences nor trend in differences between the control violin and the two stimulated violins, accounting for factors such as aging, moisture content, and experimental uncertainty. This confirms our hypothesis that the playing-in of a violin does not change the baseline auditory and vibrational response of the instrument, and the idea of "breaking in" a violin is not supported by my research.

Future tests will include replicating the experiment with other violins, and using psychoacoustic analysis to see if listeners can identify a difference in the tonal properties of violins before and after they are subjected to similar stimulus. However, I do believe that this research is currently the most in-depth analysis on the topic, and hopefully will serve as a stepping stone for busting the breaking in myth not only for violins, but other instruments, as well.

Presentations

Publications

Awards

Future of Physics Days Best Presenter - APS March Meeting 2023

Best Physics Poster - Reconnect Regionally AAAS-PD/NWSA

Research Achievement Award - Central Washington University Department of Physics

Acknowledgements

This research is made possible by the Central Washington University College of the Sciences and performed under the mentorship of Dr. Andy Piacsek. Thanks to Peter Zencak for providing the test violin and constructing the testing apparatus. Thank you to Hammond Ashley Violins for providing the three main experiment violins.