Acoustic Testing of Instruments
Similarly to the tests I've been doing on tonewood, I have also been testing the instruments I build, both during construction and when they are completed. The purpose of this testing at this point is more to build up a database of measurements that I can later use to correlate physical properties with the resulting musical qualities of the finished instrument. This is a complex goal, and I have no expectations of deriving any coherent acoustical theory of musical instruments anytime soon, if at all. However, keeping this data has already proven its worth as a reference in building new instruments.
The sort of data I have been collecting are weights and dimensions of all component parts; compliance (deflection) measurements of tops, backs, and necks; natural resonance frequencies of the soundbox during various states of construction; vibrational patterns (Chladni patterns) of the resonances; and some graphs of the power output of the finished instruments. I have also built a couple of "matched" sets of three classical guitars: one set had backs and sides of bubinga cut from the same plank, bracing of the same stiffness and species, but different soundboard wood. The other set had cedar soundboard and spruce brace wood cut from the same billets, but backs and sides of different and quite different species.
At this point, the data is not in a presentable shape but I hope to put some graphs and tables here as time permits. However, here are some of my (somewhat subjective) observations to date of guitar acoustics.
- The soundboard is the most important of any of the components that go into a guitar. It contributes possibly as much as 80% of the final tone.
- The contribution of the back and side is only on the order of 10% to 15% of the final tone. Probably the most important consideration of the back is how well its vibrational resonances couple with the top resonances. The particular wood species of the back seems to contribute mostly "colour" to the tone, and little to volume or responsiveness.
- The neck can have a considerable importance to the bass response, with the important variables being density and stiffness rather than species
- There are many more usable wood species than spruce, red cedar, and rosewood and mahoganies that can be used to construct concert quality guitars, but customer expectations prohibit their wider use in luthiery.
- Finally, although the acoustic properties of each soundboard species vary widely from set to set, it is virtually impossible to find a set of one species, say cedar, that will match the characteristic properties of another species, e.g., spruce. So in one sense, we can say that there is a characteristic tone we can attribute to spruce or cedar, but the difference between these characteristic tones is of the same magnitude as the tonal difference between two different sets of the same species with wildly different characteristics.