LUTHIER’S MIND
When I build a guitar I ask three things of the completed instrument. Does it sound great? Does it feel great to hold and play? Is it beautiful to look at? All three criteria hold equal importance for me, because if they all make the grade, the guitar will inspire the player to greater things. In order to get there, a million intimate steps must be completed with care and what one of my mentors, Charles Fox—a giant in the world of guitar-making innovation and instruction—labels “Luthier’s Mind.”
Using Luthier’s Mind while building guitars, you wield all of your accumulated knowledge and skills like a keenly honed chisel, while focusing your intention and presence to the task in front of you. You’re able to visualize the project as a whole, cognizant of how each tiny task is framed by the step that led you there, and the anticipation of what will come next. All the while, being mindful of what the musical, tactile and visual consequences of your actions at the workbench will be.
Luthier’s Mind also relates to the design of the instrument. I can’t expect an instrument to be inspiring if I just throw a bunch of parts together without considering how they work together as a whole. When designed correctly, they function as a symbiotic system working in concert to produce beautiful sound. Understanding how all the component parts of a guitar relate to each other in the building process, as well as being coherent in the finished acoustic result is very important. If any part of the system is off, the answers to my three goals from above may be suspect.
A GUITAR’S VOICE
Plucked, stringed instruments of various styles have been around for nearly 500 years. For the vast majority of this period, makers have used a process to “voice” the instrument that primarily involves holding the soundboard to their ear during different stages of construction, and tapping it with a finger. The goal is to listen, and then make adjustments to the soundboard and its bracing until the tone is about right. The sound they’re striving for is as varied as the number of makers using this technique. It’s useful for determining some general tonal qualities of the soundboard, but there’s a lot of subjectivity in the interpretation.
Having watched many a lecture of talented luthiers demonstrating the process, I often walk away scratching my head at some of the ambiguous nature of it all. Historically, independent luthiers have used it to adjust guitar soundboards by virtue of having years of experience and handling many pieces of wood, using a sort of Shaman’s magic to get it right. It’s a learned thing, and a useful technique in a luthier’s tool bag, but ultimately, there seems to be a lot of room for improvement. There must be a more quantifiable way of doing the same thing and getting repeatable results…
ENTER THE 21st CENTURY
Another guitar maker who inspires me greatly, Trevor Gore—an Australian luthier and co-author of a two-volume set of encyclopedic books on guitar acoustic theory and construction—has developed an advancement in guitar making that approaches “voicing” the guitar scientifically and empirically, not subjectively. Using his innovative technique–in conjunction with the above mentioned tap testing and accurate measurement of the woods elastic properties–results in consistent-sounding guitars with deeply developed voices and beautiful tone.
Here’s a mouthful to explain it in a nutshell:
- Take acoustic measurements of the intended raw tone-wood using a microphone, a computer and Spectrum Analysis software to determine three of its natural vibrational frequencies. This takes place from the very earliest stages of construction, and is repeated throughout to guide the sonic progress.
- Determine the same woods’ physical properties (weight, volume, density etc.).
- Apply this and a few other bits of information to a complex mathematical formula—derived through careful analysis of acoustic principles—to obtain the optimal sizing of guitar parts. This leads to a finished instrument that has the frequency response—the sound—we are striving for.
The response you desire can be different depending on the intended style and use of the guitar, and thus tailored specifically using this method. At the same time, we design the internal structure of the guitar—the bracing patterns—so the finished instrument achieves a specific set of targeted resonant frequencies. This, in addition to precisely measuring and adjusting the mono-pole mobility (flexibility) of the soundboard and back, plus the age-old practice of intuitive thumping and flexing of the wood, leads to a tonally balanced and responsive guitar with greater consistency in sonic character from one guitar to the next.
It’s an evolution in guitar construction techniques that has otherwise remained stagnant for a very, very long time.
PLEASE INDULGE ME IN AN EXPERIMENT
I’d like to illustrate the difference between mass-produced and individually built guitars. Let’s go downtown to the corner music store and grab five of the same model guitar, all made at one of the mainstream guitar factories. You know the companies; they make thousands of instruments annually. If we placed these guitars next to each other, they would all look remarkably similar—aside from some variations in the color and grain of the wood (clue). You would expect this from such a company; their goal being to make good sounding guitars using repeatable mass production to reduce costs. The guitar parts are made through automation, and humans put them together, and because of the automation, visual and structural variation between guitars has mostly been eliminated.
If we sat down and played these five guitars you might be surprised at how differently they sound from each other. Some will sing sweetly and some will sound clunky, with the others ranging in sound somewhere between these extremes. This is curious, they all look alike, why don’t they sound the same? Because the factory doesn’t account for the inevitable variations in the tone-wood’s physical properties. Similar to how the wood color and grain patterns between these guitars differ, the physical makeup within the wood’s structure also differs. The wood that a factory uses to construct their guitars is all processed to equal specifications, without taking the time to specifically tailor each set of wood that is used for an individual guitar. It’s too cost prohibitive to do otherwise.
It’s very interesting to note that for a given species of wood—say Sitka spruce—there is wide variation in the material properties (density, flexibility, and an important quality called Young’s Modulus), even between samples milled from the same tree. It can range anywhere from dense and stiff to light and flexible. This means if the various soundboards and backs of these five guitars are sized in exactly the same way (factory style), not accounting for differences in the wood’s internal structure, they will vibrate at different resonant frequencies when excited by the guitar strings, thus producing the different sounds.
For an individual guitar then, the goal is to determine the optimal “target thickness” or sweet spot for the particular pieces of wood I am using for the soundboard and back. Using the Spectrum Analysis software gets me there in a definitive way. Factories use the same generic thickness for every guitar model they make; it leans towards the overly robust side (read: over built & less responsive) to reduce the amount of warranty work that might be required down the road. This produces a mostly OK sounding guitar; some great, some good and some not so good. For me–with the time and inclination–it’s a matter of customizing each instrument according to the specific materials we choose to build your guitar.
RESONANT FREQUENCIES
Think of the term resonant frequencies as synonymous with the sound that a guitar makes. This sound is very complex in nature, but it’s primarily a function of two resonant frequencies, and for truly responsive guitars, a third one can be built in. All three are controlled by the luthier.
The vibrating strings act on the top (soundboard) of the guitar via the saddle and bridge. The top in turn oscillates and imparts motion to the air cavity inside the body of the instrument. The motions of these two elements produce resonant frequencies, and are measured in hertz, or oscillating cycles per second. The separate soundboard and air cavity frequencies largely determine how the guitar will sound. During various phases of the construction process there are opportunities to check the sonic progress with the Spectrum Analysis software and adjust the structure to meet my resonance goals.
As the instrument nears completion, if I’ve been mindful (Luthier’s Mind) of all the elements I’ve talked about thus far, the primary frequencies are very close to the goals I set before beginning. The beauty is, there are still various methods to nudge them higher or lower to achieve my targets.
Of interest is the third resonant frequency mentioned above, because by building in a relatively simple but less common way, a good sounding guitar can be made into a great sounding guitar. On many guitars, the back of the instrument is a relatively thick and heavy componant used primarily to close the body and make a box-like structure. Because it is so rigid, it doesn’t contribute much to the production of sound—acting to mostly reflect the tops vibrations.
Instead, if I give the back of the guitar the same consideration as I do the top, going through the Spectrum Analysis process and bracing it with care to give it the full potential for mobility, it will have the freedom to vibrate and harmonize with the tops resonant frequency so they act in unison and complement each other. This coupling increases the number of harmonic peaks in the sound spectrum, which adds color, interest and complexity to the sound of the guitar. This is called an active or “live” back.
As mentioned earlier, the process of how guitars are “voiced” hasn’t significantly changed for many centuries. My approach retains the time-honored craftsmanship of making a guitar by hand—which I love, while using twenty-first century innovation to determine how to best utilize the precious resources that go into their construction. The Guild of American Luthiers recently published a research paper analyzing the consumption of wood products in the timber industry. The good news is guitar makers—both independent and factory—use a tiny slice of the pie. Unfortunately, the traditional rosewoods that (arguably) make the best sounding guitars are endangered and their distribution is highly regulated. Using Spectrum Analysis software to model tone-wood allows a wider range of woods to be used—thus taking the pressure off of traditional rosewoods—yet still produces extraordinary sounding instruments.
INTONATION
Intonation is simply the pitch accuracy of a musical instrument. For fretted, stringed instruments like lutes, mandolins, banjos and guitars, it’s commonly known that having all the strings and all the fret positions in tune simultaneously is a real challenge. There will always be a compromise somewhere along the fret board. The primary culprit comes from pressing the string down to a fret, which causes the fretted string to stretch when compared to the non-fretted (open) string. This stretching of the string causes its tension to increase–thus increasing the pitch–and every fretted note will sound a bit sharp; even with the open strings in tune. The amount of pressure the player exerts while fretting a string greatly influences the degree of sharpness that results. Two other variables involved with this sharpening are the lengthwise stiffness of the string, and the distance it’s stretched when fretted. The bass string will have a larger increase in pitch than the treble string because its thicker diameter is inherently stiffer, and a guitar with high string action will have a greater increase in pitch over a lower action setup because the strings are stretched further when pressed to the fret.
Compensation is what luthiers do to help fix this problem. We increase a string’s length–the distance between the nut and saddle–to drop the pitch back down to where it should be. You’ve probably noticed the saddle of your guitar is slanted, that’s the compensation added to the string length. Since the bass strings are thicker—more longitudinally stiff—than the treble strings, they need more compensation, hence the slant. To fine-tune things even further, we file the saddle so each string touches it at a precise point. This fine tuning of extra string length is accomplished by adjusting the string/saddle contact point until the note played at the 12th fret (octave) plays in tune with its harmonic played by plucking the string while barely touching the string at the 12th fret position. If the fretted note is sharp or flat compared with the harmonic, the contact point is adjusted until they both play in tune. Most guitars are set up in this fashion.
NUT AND SADDLE COMPENSATION
The big problem with the above technique is that it accounts for intonation at only one position—the octave. The rest of the fret board is left to sort itself out, and the first few fret positions near the nut will usually play a little sharp. This is generally accepted by the playing community, but there is a way to get most of the frets to play more in tune along the full length of the fret board. I treat each string as if it has its own individual scale length by compensating the nut in addition to the saddle.
I set the saddle compensation so multiple frets play in tune, not only at the 12th fret like mentioned above. This leaves a significant problem though; the un-fretted open strings now play out of tune. What to do? Way back a thousand steps ago (Luthiers Mind again), I purposely made the fretboard too short between the nut and the first fret. This allows me to now mill the nut away from the first fret the proper amount for each individual string, until each open string plays in tune.
It’s necessary to go back and forth between these two steps a few times for each string until the intonation is optimized. In effect, the saddle intonates the fret array and the nut intonates the open strings. It’s a complex dance—and the main reason most guitars use saddle-only compensation—but one I feel is well worth the extra effort to get the full potential out of the guitar.
You might notice I don’t use a slanted saddle, like mentioned earlier. Because the nut is compensated, the saddle doesn’t require as deep a compensation, and because I use a wider than normal saddle, it gives me enough “real-estate” to work within when filing its compensation points.
