Runningdog wrote:Ervin's a friend and builds wonderful guitars. But I'm not taken with this approach. I don't see any weight savings, it works against several concepts that I think are valuable, and it raises some long-term structural issues.
Al Carruth reminded me that braces account for no more than 1/3 of the mass of the whole X-braced soundboard. A small savings there doesn't amount to much. Now if there were a way to lighten the plate, that would be helpful. Lattice bracing has the potential to support the plate well enough to allow it to be thinner and thus lighter (this bracing appears to be a sort of lattice-lite) but the weight of the braces almost offsets that gain. In nylon-strung guitars, lattices are becoming more common and successful; so far, that's not true of steel-string instruments.
So, first, this seems over-braced to me. 5/8" height by 5/16" width is pretty massive for the main X. For the lower braces, it's very stiff even for the usual parallel tone bars; for an X, it becomes extremely stiff. That has to hurt response and resonance. I'm not sure that an upper X is more effective at resisting the pressure of the fingerboard extension than a simple transverse brace but it probably doesn't affect tone production so what the heck. If the ends of the upper X are firmly attached to both the main X and to the linings, it should be better than the transverse brace alone at resisting folding around the soundhole. Better than transverse plus soundhole braces? Probably, but one would have to test it to see.
Second, I'm not a fan of making the plate thin and regaining stiffness with overly-stiff bracing. It tends, all else being equal (which I admit it never is), to produce a thinner, less complex and rich sound. I prefer a somewhat thicker soundboard and "balanced" bracing -- the whole top, bracing and plate both, have to work together. It's a fine line. Ervin's guitars that I've played and heard are pretty forward (which I'd expect given the thinner plate) but not extreme. His flamenco guitars have the power and brashness you'd want for that style. For my taste, his steel strings tend to be too much like his flamencos.
Third, I like asymmetry. The soundboard must produce a range of frequencies evenly. Too much symmetry seems to me more likely to favor wolf tones, as evidenced by many double-X braced Larrivees. My thought is that many irregular, differently shaped and sized areas between braces (and, if you scallop, along each brace) are better at creating that even response.
If the ends of the upper and lower X's are not mortised into the central X, I am concerned about them loosening. They're quite stiff, especially as pictured, and any impact on the top will tend to pop them loose. Long, flexible tone bars are much less of a problem in this regard.
And then there's that bridge patch ... If you ever try tapping a top with all the braces but no bridge patch, you'll know how much it affects the response. A smaller patch would, in all likelihood, favor more response and richer tone as well as save some mass. Why does it extend all the way to the intersection of the X? That's the stiffest area, well above the bridge, and shouldn't need extra reinforcement.
Structurally, a triangular brace is 1/2 as stiff as a rectangular one of the same base width and height. The I-beam example is not applicable in this case since the brace doesn't have a wide base and top like an I-beam but rather tapers from base to top. Shaving braces to a triangular cross-section is a great way to reduce stiffness more controllably than lowering them since the stiffness changes in a linear manner as you reduce width. (Reducing height lowers stiffness proportional to the cube of the height.) The "hollow grind" cross-section can work against you if taken to an extreme -- you'll suddenly lose stiffness if the narrow peak flexes out of column. That's why braces were usually made in a bullet shape: you save some weight by rounding the top, maintain stiffness, and don't risk having the brace fatigue.
So it looks to me like a really well-executed soundboard but one that may be disappointing in performance and durability. I'll be interested to learn more as the guitar is completed.
Thank you for your logical and well executed response. Heck, it even made sense to ME! :)
I initially rejected your assessment of the bracing strength of the triangle shape, simply because one forms habits.
From an engineering standpoint, triangles are usually very hard to beat... but after carefully re-reading your statement and actually THINKING a little myself, it's obvious. :)
Again, thanks a lot for much thinking stimulus. I am sure there ARE yet undiscoverd ways to STILL improve our beloved guitar.
Being able to produce the sound and "Feel" of the Masters in a consistent and repetitive manner still escapes most of us...
Maybe one day.. (:0)>}
