Saturday, May 17, 2008

On Scalloped Braces

Everyone seems to make such a big deal about them, but what are your thoughts on scalloped braces?


I don't know what justification is advanced by builders who scallop their braces: I don't. Maybe you should ask
them. I suppose they would reply, "Martin used them on guitars made during the 20s and 30s, and since those guitars reputedly sounded so good, presumably it must have been the scalloped braces." Sounds like lore, doesn't it?

Few people know that actually, those early guitars were originally braced for gut strings, and when decades later people swapped steel strings onto them, their tops eventually collapsed from the extra tension. But boy, they sure sounded great before they collapsed!!

The peculiar scallop shape, it seems apparent to me, must have originated as a result of an early voicing technique where (in the days when
luthiers worked in factories, and were not just machine operators) builders reached inside through the soundhole with finger planes and removed material judiciously from the braces, in stages--and progressively listen to the changes after restringing, and stopped when they felt the compliance was "right."

There's nothing magic about the scallop shape itself, just that it was the result of the process of using finger planes to remove material through the soundhole. At least that's my best guess. But as it happens, that peculiar shape decades latered engendered enough lore and mystery to subsequently drive aspiring luthiers crazy trying to decipher its significance. I have been asked that same question by many of them.

Large, stiff, disappointing guitars are today "strutted" by technicians who reach inside through soundholes with finger planes to remove material from their "struts" to improve the sound. The result usually is to "hot rod" the guitar by making it sound somewhat louder and a bit deeper-voiced...while potentially hastening it's demise. The guitar survives only if the braces were way too large to start with or if the the strutting was done with great restraint. It takes an experienced eye to sense how much is too much.

The string's signal is nothing more than minute and rapid changes in tension (dynamic stress) amidst the constant background string tension (static stress). So the problem is to construct the top in a way that adequately supports the static stress without hindering the dynamic stress. It's the guitarmaker's dilemma: that the soundboard is really a trade off between structure and tone. If the brace heights are left high, the soundboard will resist the 200 lbs of steel-string tension with no problem at all, but with a price to pay insofar as the acoustical range that results: invariably a tight sound or a limited tonal response. Reducing the brace heights increases the compliance of the top to a wider range of signals coming from the strings and a more satisfactory acoustical response, but with a price to pay in its architecture. The builder is truly expert when he derives a sense of structure sufficient to dimension the top thickness and brace heights to achieve an optimum--say, minimally adequate structure.

So, the secret is not in the peculiar contour of the braces, it's in the acquired skill of the builder that senses the minimal structural requirements of the instrument and responds correctly when removing material. Note that Martin now again offers "scalloped braces"--as a marketing ploy, I suspect, because no one in the factory is graduating braces. They are likely to be using thicker, stiffer tops to hedge their bets, and make it a point to insist that low tension strings be used.

Some People Say...

A headstock thickness of 0,5", is that still safe for an acoustic steel string guitar (for strength and sound)? Or would it be advisable to add a veneer on the back of the headstock? Some people say the headstock dampens the sound when it's too flexible. Do you think that will be an issue in this case?


Don't ask me. I haven't tested that assertion. Ask those people what evidence they have to support that assertion, or at least how they came to that conclusion. Also, how flexible, precisely, is "too flexible." Talk is cheap.

Saturday, May 3, 2008

On the Significance of Soundbox Specie

I am curious what impact the different body woods has on the sound of the guitar. I have read that with mahogany one can expect a "warmer" sound. (less bright... less projection?) What I had in mind initially was something in Indian rosewood with a spruce top. (I simply have not had the pleasure to play a classical guitar made of anything else.)


I could give you the easy, pat response that luthiers usually give their customers when answering the common question, "what is the impact of different body woods?" --with the well-worn, and ultimately meaningless words: rosewood is "clear", mahogany is "warm" and "bright," maple is "dry" and attribute more or less of one or the other quality to the specific species selected for the back and sides. But I won't do that to you, mainly because it is useless and misleading. If given ten seconds to respond, I should answer: "the sound doesn't come from the wood, it comes from the strings."

35 years making guitars has brought me to the conclusion that the guitar can physically respond to only a distinct subset or portion of the the total string's sound spectrum, creating it's own "version" of that spectrum. The guitar's effect, so to speak, on the string's sound--and thus, what we actually can hear-- is determined by among other things, it's size; it's string length; the way it is constructed; the massiveness or resilience of its neck; the thickness and hardness of the walls of its soundbox, the placement and cross sectional shape of its braces; the size and shape of its bridge, and on and on. There are so many factors impacting on the performance of the guitar--some which the luthier is aware of and can control by choice; some which are not controllable by the luthier because they are hard-wired into the guitar's culturally-determined form; and some which are beyond the perception and comprehension of the luthier. Each of these factors affects the sound, some dramatically, some slightly. The specie-name of the wood used on the back and sides plates fall into the "virtually none at all" category. The actual sample of the given specie chosen rises to the just-perceptible category. Surprising?

In other words, I could say that Indian Rosewood results in a "brighter" sound than Mahogany, but there are extremely soft samples of Indian and extremely hard samples of Mahogany that would contradict this statement--the variations between samples of the same specie are dramatic. So it may be smug and easy to talk of the "warmth of mahogany" or the "punch of rosewood" but really, they are ultimately meaningless statements.

Besides, what sound am I eliciting in your mind when I say that Indian Rosewood results in a "clearer" sound than Mahogany? Does my statement that Mahogany results in a "warmer" sound than Indian really communicate accurately to you the sound difference that I perceive in my ear? Given all the myriad known and unknown factors in any given guitar that result in it's distinctive tone, am I safe to say that the "warmer" sound that I'm perceiving is a direct result of the wood specie? Or some other solitary factor or combination of factors?

So to conclude this overlong message, I find it /not/ helpful at all to preoccupy myself or my clients about what the sound difference between two hypothetical guitars might be, one rosewood, one mahogany. That's because the perception of tone is a personal experience and not something that can be talked about, or written about. It must be experienced, it can't be described.

One thing is clear: it costs me about $20 dollars to purchase the mahogany used on a guitar, and $100 to purchase the Indian rosewood used on another; and $800 to purchase the Brazillian rosewood on yet another. And that difference will be reflected in the value of the guitar, you can be sure of that! The fact is, the cost difference biases people's perception of the guitar's sound "quality" in many people's minds (or ears). One must be very careful of falling into that trap. The perception that the Brazilian rosewood guitar is per se "better" than the Mahogany guitar, evidenced by the difference in cost, is incorrect. The Brazilian rosewood guitar is more valuable because its materials are rarer, and to some eyes, more beautiful in appearance. But one is not "better" than the other--indeed, they may be simply different, or you may not be able to perceive the difference at all. And if they are different in sound, surely there would be no way to safely say that the difference is because the specie name of the back and sides is different.

On Avoiding "Acoustical Problems" with Bridges

I'm a beginner builder of steel string guitars. I've used you wonderful book as instruction to build 3 guitars so far. I've noticed on many mass-produced as well as popular small-builder product lines, that one of the distinguishing things is the shape of the head plate and the bridge. I've done a good bit of searching the internet to find information on the pros and cons of bridge shapes/designs. I'd like to design a head and bridge that would be unique to my guitars in the future, once I get a little more experience.

Are there any general guidelines for bridge shapes to avoid acoustical problems? I notice some are very eccentric while othes (like Martin-style shapes) are very simple and straightforward. Can you advise me or point me to any infromation on this subject?


Keep in mind that guitarmaking is an art, and not a science. Your question is like, if I change that blue area in my painting, will there be a "visual problem"?

People who like to put batman-shaped bridges on their guitar do so because after they did, they found that no acoustical "problems" resulted, so they kept doing it. It certainly is not that they calculated the precise bridge's form to produce some kind of anticipated sound. I guarantee you that no one has that power. It's more probable that the motivation was simply to make the guitar stand out.

It stands to reason, however, that if the bridge footprint is too small, there may be insufficient gluing area relative to the stress--and the bridge will be prone to eventually flying off. If it is too large, there will probably be a change, most likely a deadening, of the tone. But since you are breaking away from the traditional proportions, no one will know if that shape results in an acoustical "problem"--or has no discernible effect whatsoever--until you actually do it. Or whether it's favorable or not.

I'm reminded of someone bringing me a cheap ukelele and asked me to do something to "wake it up." It was indeed, rather reticent in its sound. I examined it, and the bridge seemed oversized. It was decidedly different from the smaller bridges that you ordinarily see on those instruments. So I took the risk and carved off some length on both sides of the bridge. Voila! The instruments sound now clearly opened up and became much more complex. The owner looked at me as if I were a wizard. And as my partner Harry always says, "never refuse an undeserved compliment."

Conversely, I also recall a Mexican Bajo Sexto that came into the shop. It had an enormous, grotesque, bat-wing bridge that covered fully a quarter of the real estate of the soundboard. Yet it had a mammoth sound, precisely the one coveted by Tex-Mex fanatics. I dared not suggest that I could improve the sound by carving material off its bridge.

So clearly, there is no magic key to world-class consistency in instrument making. The designs handed down to us by history and tradition are the sum product of the efforts of hundreds of very thoughtful and committed makers in the past who have progressively, and slowly, refined that instrument's form. So the form itself is actually the DNA for the standard sound of that instrument. If you go too far afield in your tinkering with the traditional model, you'll end up creating something...far afield. And you run the risk of it sounding "peculiar" to the listener. And be broke.

On the other hand, your tinkering may create a new paradigm, a change so drastically superior to what came before, that everybody will flatter you by stealing it. This is what happened to Antonio de Torres in Spain at about the time of our CIvil War, when he created the modern classical guitar. But be warned: this has only occurred once or twice in modern times, Torres being one and arguably, Martin's foreman taking a stab at an x-brace being the other. But it is not at all a scientific process, not a process of somehow knowing what the sound waves are going to do and then deftly harnessing them to some desired end.

Its a try it and see process, and anyone who claims that power of acoustical omniscience is either a faker or a fool. Or an X-man. Or Roger Siminoff.