Behind everything that has ever been said or written about the guitar, it is in fact nothing more nor less than an air pump. As such, the air pumping efficiency of its design and materials are the most important factors a maker needs to consider in his work. Everything else -- the guitar’s history, its design aesthetic, its looks, the romance, the art, the techniques of its construction, its noble materials, the fame of its makers, or even its beauty -- is secondary.
Structurally, the guitar consists of a vibrating top and a vibrating back which are separated by a set of non-vibrating sides, and a non-sound-producing neck. Because the top and the back are the only two acoustically active parts of the guitar, the choices of top and backwood are the most important ones to be made in the selection of guitar tonewoods.Tonewoods are called such because they really make tone, or are capable of making tone, compared to more normal woods which are useful for making things like buildings, boats or furniture. Tonewoods can ring when you strike them, just like a bell or a piece of glass. Can you imagine a wood that rings to a musical note when struck? Brazilian rosewood can: it’s the material marimbas have been traditionally made of, and, even in guitar form, such wood can ring like a gong. Good quality face wood also can ring like crystal. Such materials, when studied by scientists or acousticians, are said to have a high degree of liveness, or "Q" [which stands for "quality"], and Brazilian rosewood is only one of many tonewoods that have high "Q".
Because of the dynamics of the guitar, tonewoods for faces need to be different than tonewoods for backs, if the instrument is to have the best and most even sound. The best guitar faces are made of high quality musical instrument grade softwoods such as spruce and cedar. The best guitar backs are made of high quality hardwoods such as rosewood, ebony, maple, walnut, koa, mahogany or any of a number of other suitable body woods. The consensus among luthiers is that face and backwoods need to be chosen from woods of differing densities because the resonant frequency of the back needs to be higher than the resonant frequency of the face, by at least a tone. The best wisdom on this matter is that if there is too great or too small a gap separating the fundamental resonant frequencies of the top and the back, then guitars have an uneven tone. That is, the sound becomes an uneven mixture of loud and quiet notes. Likewise, if the face and the back are most active at the same frequency or frequencies they’ll act in tandem to reinforce certain notes, but leave others weak. It does not matter what the sides are made out of, except that guitars in which the back and side woods don’t match are considered to look too strange and generally won’t be saleable: backs and sides need to match for aesthetic reasons.
How does one choose tone woods? Well, it depends on what the guitar is expected to sound like and how the face is expected to behave.
SPRUCES AND CEDARS
As far as top woods go, European spruce, on account of its cellular structure, is more brittle than American Sitka spruce: it cracks and splinters somewhat easily when sufficiently bent or stressed. Sitka spruce, in comparison, has superior tensile strength: it will bend a lot before it breaks. Because of these factors ships’ masts and airplane propellers -- which need to put up with lots of stresses -- are made from Sitka spruce. Before the advent of space-age materials, its stiffness-to-weight ratio even made it ideal for making airplane fuselages out of. On the other hand, no one uses European spruce for ships’ masts or airplane propellers: they’d snap from hard use. Nonetheless because of this internal brittleness, and when made into a guitar face, European spruce makes a beautiful sound rich in overtones -- a sound that is limpid, focused and full of nuance and tone color. Fingerpickers tend to like this sound, which is a little like having a choir of singing voices inside your guitar, or like listening to the clear fundamental and harmonics of a church bell. In comparison American spruce is supple and springy (in a ropy way) rather than brittle, as a function of its cellular structure. Because of these qualities, when it is made into guitar tops, it makes a sound that is not so much in focus as the European spruce is. Its sound is heard as not being so cleanly defined but, instead, as warmer, more fundamental, and largely free of overtones. It’s a good, solid sound and bluegrass flatpickers and folk-musicians tend to like it a lot. These are, of course, rules of thumb with many exceptions, because there is so much innate variability from sample to sample.
Cedars, as a vibrating material, sometimes have a better stiffness-to-weight ratio than spruces. Accordingly, the sound these can make is more quick and loud and, because it is so immediate, brighter and sharper than a spruce sound -- but without the European spruce overtone component. Because of its inner structure cedar is also a somewhat weaker wood than spruce, and it is more subject to cracking and fracturing. I recommend being a bit more careful in the care and handling of a cedar top guitar than a spruce one. These are also rules of thumb, with many exceptions.
Engelmann spruce has been brought into instrument making in the last few years, and is different from the above woods in several characteristic ways. First, it is very white. European spruce is white at first, but oxidizes and darkens over time so that after ten or so years a European spruce guitar face takes on a lovely and warm honey color which gives it a naturally aged look. If repair work needs to be done on such a face and the repairman sands some of this wood off it reveals a lighter color which won’t match the surrounding surface and needs tinting. Engelmann spruce seems to resist oxidation, and, in my experience, stays white for a longer time.
A second, and much more important, difference is in the nature of its cellular structure. Sitka and European spruce and cedars tend to have dark grain lines which are hard; that is, they are areas of dense cellulose concentration. It is precisely this cellulose concentration which gives softwoods woods their longitudinal stiffness and strength: the white grain lines in between are mostly thin walled cells full of air (think styrofoam). It’s the dark material that does the work. Engelmann spruce seems to have dark grain that is less differentially concentrated from its own white grain. That is, it’s not all that much harder a material than the white grain next to it, like it is in the other woods. You can test this out yourself next time you’re in a position to compare these woods: dig your thumbnail into a few dark grain lines to see how hard they are. Or aren’t. The differences are pretty obvious. In consequence, because this concentration of linear "cellulose rebar" is decreased in Engelmann spruce I believe it is a softer and weaker wood in general. That doesn’t necessarily mean that it’s a bad wood, but it does mean that it has to be worked differently than the other spruces. I should add that I have noticed quite a range of this quality in Engelmann spruce, and some of it compares favorably in cellulose structure to the European and American -- but most of it is not like that, in my experience. Therefore, because it’s so varied a wood and there are so many exceptions to any general description of its quality, I’m inclined to believe that one has to be choosier in using it. Thus, saying that a particular top is Engelmann spruce becomes less meaningful than saying European or American spruce. To the extent that Engelmann is generally weaker and softer I’d expect that one would have to use it in thicker plates to work it with confidence; otherwise it would correspond, structurally, to the use of exceptionally thin (and correspondingly weaker) European and American spruces. And having a consistently thicker top would have to create some consistent, characteristic, difference in sound. But all in all, I’m not sufficiently experienced with this wood at this time to say anything more about it.
Rosewoods are more consistent from sample to sample than spruces and cedars and one piece is much more like another in behavior, if not appearance. Of the rosewoods used in guitarmaking, Brazilian rosewood has traditionally been the best wood of choice. This is partly due to tradition and partly due to its phenomenal "Q", which makes it a very acoustically active material. When struck, a properly cut sample rings like a plate of glass. This quality contributes to sustain and projection in a guitar, because those are the chief functions of the back. Sustain, because it rings a long time; and projection, because the back’s movement can be coupled in vibrating activity to the movements of the face, boosting the directional power of the activity of the guitar. Because of its high "Q", Brazilian rosewood is both vitreous and brittle, and therefore prone to cracking and checking. East Indian rosewood, the alternate wood of choice, is comparable to Brazilian rosewood but simply not as beautiful nor as "live", by a factor of some 10% to 20%. This is not a huge difference, and there are plenty of excellent sounding East Indian rosewood guitars around. Also, East Indian rosewood is an attractive choice because it is much less prone to cracking and therefore generally less problematic to work with. Other rosewood-like woods which have a high "Q" are wenge and padauk (both of which crack very easily) and certain Asian and Central American rosewoods, which do not have the beauty of Brazilian or East Indian. While I haven’t worked with all these woods I’d expect them all to be brittle in direct proportion to their liveness and be prone to the same mechanical failures. Removing their brittleness would in fact remove the factor that is responsible for their characteristic tone.
MAHOGANY AND KOA
Mahoganies and koas are very variable in physical properties. That is, whereas one piece of rosewood is much like another in this regard, these woods range from light to dense, and stiff to loose, while all looking the same. Accordingly, they will behave differently as tonewoods as they exhibit different degrees of "Q" and a guitar’s sound will be colored by the specific selection of koa or mahogany used. The denser and more brittle the wood, the more it will ring; the lighter and looser the wood is, the more it will be an acoustically passive part of the guitar. Heavy koa, mahogany and walnut are all comparable in their tone. Everything else being equal, it is generally recognized that mahogany and koa will produce a "warmer" sound in a guitar than the more brittle rosewoods can.
Maples usually have a low "Q" and tend to make passive backs in that they don’t ring, sustain, or further the vibrational activity of the face very much. In fact, they help to absorb the vibrational energies of the face and kill them. As an extreme example, consider the sound you’d get in tapping a guitar back made of cardboard. This is not necessarily a bad thing, however. Besides being beautiful, maples help to create a sound which is damped and short-lived and which is perfect for jazz style playing. The jazz musician will play many notes, and the music is such that it is not desirable for any of the notes to linger in the air. Such music does not need the sustain of Brazilian rosewood. For jazz, one wants quick notes that come out and then disappear -- because there are many more notes coming. This quality is also desirable for certain parts of the modern classical guitar repertoire, which has at this point somewhat abandoned the lush, dark and expressive tonalities so much appreciated in the Romantic classical repertoire. Again, there are exceptions to all these statements; but, as rules of thumb, these descriptions are accurate.
"Spanish" cypress is the traditional wood of choice for flamenco guitars. It’s a formerly cheap and plentifully available Mediterranean wood which is wonderfully aromatic and easy to work. For much of the Spanish guitar’s early life this cypress was the poor man’s wood for backs and sides; it was the most common default material for anyone who could not afford the more expensive rosewood. For all its humble origins, though, and in spite of its comparative lack of density, properly cut and selected Spanish Cypress has at least as good a "Q" as East Indian rosewood and a much better one than maple. Even though it isn’t used for making steel string guitars, I know of no reason at all why it wouldn’t work very well on them.
CELEBRITY WOOD AND EXOTICS
From time to time a new wood surfaces that captures everyone’s imagination as being "superior" in one way or another. It’s an interesting phenomenon that is part and parcel of the guitar making community’s cyclic attempts to find The Holy Grail. It has its counterparts in many other departments of human endeavor such as the search for a better President, the best athlete or athletic team, the ultimate racing-boat design, the best next actor/actress, the sure-fire cure for cancer or urban decay, etc. In lutherie, as an example, African blackwood has emerged as a popular alternative to Brazilian rosewood in the last few years: it’s a perfectly good wood, and getting quite pricey as demand for it rises. Likewise, Adirondack and red spruces have been getting a lot of press lately on the strength of their having been the "original" spruces used by the Martin Company. I sometimes wonder what African or Asian wood will be discovered next year as the answer to my acoustical problems -- both those continents being so well known for their long and rich traditions of guitar making. The fact is that the use of celebrity wood -- that is, simply because it’s popular all of a sudden -- is always driven as much by marketing and wishful thinking as by experience and the laws of acoustics. I want to underline the obvious: namely, that one can overbuild or underbuild with celebrity wood as easily as with anything else: just using itwill not be a guarantee of anything.
Parenthetically, the corollary to the myth of "the best wood" is the myth of bad wood. The fact is that, within reason [for instance, I wouldn’t hold out much hope for a balsa wood guitar], there aren’t many really "bad" woods: one simply has to know how to work with the materials. Personally, I prefer the stiffest and most lightweight tonewoods woods to work with. But perfectly good guitars have been made with stiff woods, floppy woods, heavy woods, lightweight woods, tightly-grained woods, widely-grained woods, domed woods, flat woods, quartersawn woods, off-quartersawn woods, etc. etc.; you get the idea. It’s very largely in what one does with them. In fact, that’s what this book is about.
And as far as exotic woods go, keep in mind that one man’s exotic is another man’s boring domestic product. Today, many American guitar players and makers believe that European spruce is the best guitar top wood: European spruce comes, of course, from Europe. Prominent Swedish luthier Michael Sanden reports that he has great demand for Sitka spruce on his guitars; Sitka spruce is, of course, a Northwestern American and Canadian (and Alaskan) wood, and Sanden’s clients consider this wood superior. Each group of end-users considers its preferred wood to be an exotic.
TONAL POTENTIAL AND TONE CHANGES OVER TIME
It is common knowledge that wooden string instruments -- whether they be pianos, mandolins, lutes, or guitars -- benefit from being "played in". Older instruments have tonal qualities of mellowness and smoothness that newer ones lack, the latter often sounding somewhat brittle and harsh in comparison. The analogy of making a stew is often used to describe the quality of transition of a sound which is initially a bit rough, "green" and unsubtle but which gradually blends its elements into something more integrated and smoothly pleasing. In the guitar, also, different woods take different amounts of time for getting "played in". Why this is so is not fully known but, obviously, it has to do with changes in the cellular and fibrous structures of the woods over time.
Some of these changes have to do with the adaptation of the woods to the stresses of being strung, after possibly centuries of being unencumbered by such forces. A main physical indicator of these changes is seen in the doming in the area behind the bridge which almost all older guitar tops show, but which new ones won’t yet have. Extreme distortion is problematic, but a merely visible amount of it is absolutely normal and even desirable; in fact, guitars which are so overbuilt (through thicknessing, doming, bracing, etc.) or understrung that this distortion of the wood is prevented will never manage to have the developed sound every player wants. The act of actually playing on a guitar, over and above simply stringing and stressing it, seems to have a decisive and accelerating effect on this blending; as with muscles, stretching and "warming up" seems to loosen things up significantly. I make my guitars yielding enough to have some top pull-up, and tell my clients to play them a lot for at least the first few weeks.
Finally, all of the woods described above have a certain tonal potential rather than a fixed quantity of tone. That is, they can be worked with to enhance or suppress certain portions of their potential response spectrum. However, like a plank of wood that can only yield usable pieces shorter than itself when it is cut, and never a longer piece, guitar making woods benefit from the outset only in having the most and best potential tonefor their intended use. You can work with any wood to make it sound a lot worse than its potential; but you can only work with it to make it a little, if any, better. Once you’ve figured out what you want your next guitar to sound like, go out and buy the best wood you can find for it.