Question for those who have voiced guitar tops

[Alan Carruth]

Braces are there mostly for structural reasons. A top that was thick enough to hold up under string tension would be too heavy to make much sound, so we make them thinner and then add bracing to get the stiffness up. That thick top would probably sound OK in terms of tone, though; the weight just keeps the power down. In one sense, then bracing doesn't really improve tone; it just gives you more of it.

Bracing can, however, make the tone worse, if it adds a lot of mass or stiffness in the wrong places. My friend Mark Blanchard, who makes a really fine sounding guitar, says: "the sound is in the top". His objective in working on the braces is to get them 'out of the way' in an acoustic sense, to allow the top to produce the sound, while still keeping the structural advantages.

From what I've seen of the systems the various makers use, we're each pretty much trying to do that in our own way, whether it's 'flex and feel', 'tap and listen' or something more 'technical'. If it works, we're all pretty much doing the same things.

Factories, of course, can't take the time to individually 'voice' tops to adapt to the wide range of variation in the properties of the wood. Instead they tend to take a standard design that has been developed over many years, possibly tweak it a bit to allow them to claim it's 'improved', and beef it up just a bit to guard against the possibility that the weakest top on the shelf got the weakest set of braces. The guitar you get is probably a bit over built, to guard against warrant work, and could sound better with a bit of work. The problems are to figure out where to remove material to make it 'better', and how much you can take off before you go too far.

Determining where to take off wood from a finished instrument can seem to involve some voodoo. In theory if you had some 'perfect' reference instrument to guide you, and a very complete set of measurements, you could get pretty close to that. Guys like Bryan and Dana used decades of experience to develop a sort of image of that 'ideal', but trying to distill that down to something that can be communicated via ASCII text on line is a dream.

The only objective measure I've heard of to tell you when to stop is the one proposed by Trevor Gore. He has measured how far forward the bridge tilts under string tension, and says that the 'right' amount is 2 degrees. If the tilt is much less than that the top is probably stiffer than it needs to be, and much more says it's too weak. A few people have started to use this as an indicator, but only time will tell if it's a good one. At least it's a number you can try.

[srick]

You might be interested in this post I made in 2018. I had the pleasure of attending the Woodstock Invitational Luthier's Showcase "Builders Coffee Talk" and was able to distill the 90 minute discussion of "What makes a good sounding guitar?" into three main points. The panelists were: Linda Manzer, Dick Boak, Chuck Emerson (The Duke of Pearl), Michael Gurian, Roger Sadowsky and Richard Hoover.

Here's the thread: https://www.acousticguitarforum.com/...d.php?t=526482

Enjoy.

P.S. - I'll second what Alan says directly above this post. It's the top and the build, plus the magic (read experience) of the builder. When you step back and look at look at a flat top guitars design, they are all basically the same. Bringing out those little differences and making the wood "sing" is not easily quantified.

[yellowesty]

For my last 20 (or so) builds over the last 4 years, I've been measuring acoustic spectra (using a "tennis ball on a stick" mallet and Audacity software) at various times during the build and beyond as the guitars are "played in." I started the process in hopes of finding a magic formula.

No joy. All of the spectra (~ 75) look fairly similar, and I found no dramatic characteristic distinguishing a lightly-braced smaller build from a jumbo intended for medium strings. So my testing approach uncovered no well-defined frequency targets for an unassembled braced top.

I did, however, identify some effects of various construction processes on the frequency of soundboard resonances:

- thinning the top, especially within an inch or two of the edge of the lower bout, reduced the frequency of several resonances by anywhere from 6% to 10%.
- applying finish to the top sometimes raised the frequency of resonances around 1% -- and sometimes did not
- gluing the bridge to the soundboard lowered the frequency of resonances by up to 4%
- "playing in" the guitar lowered the frequency of resonances, but the amount varied significantly
== in one case, the frequencies reduced by 1% to 2% in the first week and then didn't change in 9 more weeks of frequent playing
== in another case, the frequencies reduced by by 5% during 6 weeks of frequent playing

My bottom line: don't build to a formula; build to a sound.