The Carbon Sound

Graphite Guitar Acoustics 101:

What Makes a RainSong Have Its Unique Sound?

A graphite guitar - specifically, a guitar with a graphite soundboard - has a unique "sound", as does a guitar with a spruce soundboard, or a cedar soundboard, or a soundboard made out of an old garbage can.

Excursus #1:

  1. 99% of the sound of a guitar comes from the soundboard - the top.
  2. To prove this point, the "Stradivarius of the guitar", Antonio de Torres (active 1830-1870), once made a guitar with a wood top and a paper mache body - which still exists in a museum in Paris.
  3. In blind tests people thought it was all wood!

The internal structure of any soundboard colors the tone, primarily by influencing which overtones are missing and to what degree.

Graphite, however, is different from wood, in its "acoustic damping" - vibrational energy which turns into heat instead of audible sound. Since damping is what gives a guitar its tonal character, a graphite guitar has a fundamentally different sound quality.

Damping in wood

Wood is very "lossy" - is heavily damping - in the sonic frequency range. Most woods used in guitars convert only about 50% of vibrational energy into sound. The rest goes into internal vibrations in the molecules of the complex organic chemicals that make up a piece of wood - that is, the other 50% becomes heat. That energy goes into raising - slightly - the temperature of the wood, not into sound.

More specifically, most woods show slowly-increasing damping with vibrational frequency. There is a "knee" in the vicinity of 1000 Hz - which varies with the wood species - at which 50% of the vibration is absorbed by the wood. Above this frequency, the damping increases rapidly, at around 20db per octave.

This is why the highest notes on a wooden guitar sound "muddy". You hear the "attack" - the percussive initial sound of the note being struck - but almost no "note", because the vibrations have been damped - been transformed into heat in the wood - so much that they're almost inaudible before your brain has time to hear them.

Non-linear mode interaction

It's more complicated than this, however. Most of our ability to tell a wooden guitar from a graphite guitar comes, as "hi-fi" fans know, from our brain's processing the band of sonic frequencies from roughly 2000 Hz to 12,000 Hz. The presence or absence of these very high overtones, as well as the presence or absence of "mix tones" - sum-and-difference frequencies from the nonlinear mixing of lower frequencies - give an instrument its acoustic character. They are what make a wooden guitar sound "wooden", a trash-can lid sound "tinny", and a RainSong sound clear, bright and pure.

Rigorous quantitative physics gets difficult at this point. The sort of acoustics taught in college Physics 101 assumed no losses - no damping - resulting in a nice regular series of higher harmonics, decaying exponentially with frequency, and producing no mix frequencies. Each frequency is independent of all the others, and they do not interact with each other. It is only in non-linear systems - that is in materials with sufficient damping that different non-harmonic frequencies interact with each other - that "missing" overtones, "enhanced" overtones and "mix tones" can occur.

Alas, in non-linear systems simple calculations don't work - the calculations more nearly work in a "2 + 2 = 8 or sometimes = 0" manner! Actually, one has to use the sort of non-linear wave-wave quantum calculus used to compute the energy levels in atoms. And since there are hundreds of frequencies, all interacting non-linearly, it takes a supercomputer to even approximately model an acoustic guitar! Currently, the state-of-the-art in guitar acoustics is to be able to model only the lower few modes, none of which are involved in "sound character".

On the other hand, one can attempt to measure accurately the sound waves of a guitar, looking for the frequency content and mode structure, in an anechoic chamber - a chamber with no echos nor sound-coloring reflections . However, discrete frequencies can be observed only for the lowest modes - below, say, 350 Hz. Above this, the frequency plot looks like "white noise" and one is unable to measure specific vibrational modes. The frequencies which the brain uses to determine sound character fall in the middle of this white noise and cannot be identified at all.

Excursus #2:

  1. There have been a number of experiments where people were asked to tell a Stradivarius violin, say, from a junk student fiddle or a Ramirez classical guitar from a junk guitar when they were played behind a curtain.
  2. Even "naive" subjects - people off-the-street - can almost always differentiate the quality instrument from the junk one.
  3. However, if one tries to identify them by their frequency spectrum or mode structure, even musical-instrument acoustic physicists who have spent their entire careers working in this field can't tell which is which.


Now, let's apply this to the sound difference between wooden guitars and graphite guitars. Wood absorption is non-linear throughout the audible spectrum, and therefore exhibits a very complex structure of missing, enhanced and mixed overtones.

It turns out, however, that "graphite" (a composite material of carbon fibers in a resin matrix, usually epoxy) has almost no damping - no losses - at these frequencies. This means that a guitar soundboard made of graphite behaves in a nearly linear manner, displaying almost the regular overtone pattern of college physics.

Thus a RainSong's damping is almost constant across the acoustic spectrum, rather than increasing rapidly at high frequencies. Since the higher frequencies aren't damped as heavily as in a wooden guitar, the trebles are louder. High overtones are only slightly damped, and therefore display very little non-linear interaction. The overtone sequence is pure, with few mix tones or missing harmonics.

Projection Tuned Layering™

RainSong Graphite Guitars has quite recently developed a radical new technology for manufacturing its acoustic guitars, Projection Tuned Layering™. It is a proprietary process uniquely suited to RainSong's all-graphite construction, and patents covering it are pending. Until they issue I can't discuss the technique, but I can describe the basic acoustic physics of what we were trying to do and why we have succeeded.

Basically, we wanted to increase the acoustic volume of our guitars. Since guitar soundboards, like airplane wings and speaker cones, should ideally have zero weight and infinite stiffness, we reduced the weight of the soundboard while increasing its overall stiffness and response. We also did away with traditional soundboard bracing, which adds considerable weight and provides only localized stiffness. PTL provides uniform soundboard stiffness, enhancing the sound throughout.

I can affirm that the result is a warmer, richer bass which nicely balances RainSong's signature bell-like treble, and that the instrument as a whole is dramatically louder. Players accustomed to wooden-guitar acoustics are surprised by the light weight, delicacy and projection of the new RainSongs. Wooden soundboards simply cannot be made as light and stiff as PTL soundboards.

The new technology also allows for considerable improvements in production efficiency and quality control. Because of the unique nature of Projection Tuned Layering as a process, we can produce guitars faster and with a more consistent quality. That means we can drop the price while improving quality - a rare achievement!

This is a long answer to a short question. Physics is like that, I'm afraid, if one really wants to understand something.

There are, however, some short answers. -As compared to wood guitars:

Whether all this is "better" or not is ultimately a matter of personal taste - sort of like "Are the pure colors of a rainbow better than the blended colors of a painter's palette?" It does allow the musician a combination of projection, latitude and a breadth of expression that is simply not available with wooden guitars.

And judging from the reaction of owners of RainSong guitars, it is downright addictive.

Dr. John A. Decker, Jr.
Chairman, RainSong Graphite Guitars
Ph.D. Cambridge University, B.Sc./Aero Engg Massachusetts Institute of Technology


For those who want to delve deeper, the data on acoustic damping in wood was presented at the Symposium on Materials in Musical Instruments (1994), Materials Research Society, and published in MRS Bulletin, XX, No. 3 (March 1995). Mode studies and frequency analyses of guitars have been published by, among others, T.D. Rossing, in J. Guitar Acoustics, No. 4, p. 45 (1981); in

The Science of Sound, Addison-Wesley, Reading, MA (1982); and in GAL Quarterly, Guild of American Luthiers, 11 (4), pp. 12-18 (1983) and 12 (1), pp. 20-29 (1984).

For more general works on musical instrument acoustics, see F. Jahnel, Manual of Guitar Technology, Verlag Das Musikinstrument, Frankfurt am Main, Germany (1981);

L. Cremer, The Physics of the Violin, The MIT Press, Cambridge, MA (1984); and

A.H. Benade, Fundamentals of Musical Acoustics, Oxford University Press, New York, NY (1976), especially Section 24, pp. 527-553.