Official Luthiers Forum!

After reading this article on voicing by John Greven http://www.grevenguitars.com/pdfs/VoicingtheGuitar.pdf
which was mentioned in this post http://www.luthiersforum.com/forum/viewtopic.php?f=10101&t=36536
I thought it might be interesting to discuss the concept of "plate to brace ratio" in the soundboard, i.e. how much stiffness do you want to be provided by the plate itself, versus the bracing?

At one extreme, there's this Torres with no bracing in the bridge area at all http://www.vintageguitar.com/3434/antonio-de-torres-1863/, and at the other extreme you have the Smallman style, where almost all the stiffness is provided by the lattice, and the plate is more or less a membrane to fill the space.

And inbetween, you have the standards... for nylon strings, fan bracing with varying proportion of plate versus brace stiffness. For steel strings, X bracing ranging from the style Greven describes where the plate is fairly strong by itself but still needs at least the X brace to prevent collapsing into the soundhole, to the Somogyi style where the plate is quite thin and needs extra support immediately around the bridge, but still provides a fair amount of stiffness to the soundboard as a whole.

So... what style do you like, and for what reasons? On what size of guitars, and which string type?

I don't have enough experience to fully judge any type yet, but I'm presently leaning toward thick plate/less bracing on small bodies, and thin plate/more bracing on large bodies, of either string type. But it's a sort of polarizing effect, that a large body should logically have a thicker plate due to the longer span, so by thinning the plate even further than I would on a small body, the overall stiffness gradient is much more defined by the bracing than the plate. And on small bodies, even though I could go thin, there's no need since the weight is low by default, so then it's more like the Greven article where the top wood itself is the primary factor in the guitar's voice.

Nice link to John Greven's article. Thanks. I'm all ears on this one. Still learning.

I just tried a ladder braced, somewhat smaller (for me) bodied guitar on my latest and am still scratching my head about how to describe the difference. I think I'll go back to X bracing again, though when I play this new one a while it grows on me. So much for scientific accuracy I guess.

_________________
"Preoccupation with an effect gives it power and enhances the error"
from "Your Owner's Manual" by Burt Hotchkiss.

Yes that was a good read. He's one of the best, and he chasing/achieved the sound a pre-war D-45 he worked on in the 70's.
This will probably bore half this crowd to tears but my plan is to see if I can build really good Martin clones for the first 10 to 20 and see what I want to do after that.

But plate to brace ratio? Good question.

I like reading these articles, then I read something like, "the shorter the brace, the more it can flex". Isn't the opposite true? If the brace is shorter, that makes it stiffer. He's not referring to the height, because he describes that as "tallness" in the next section. so for a given height and width, the shorter brace is stiffer.
Unless he's REALLY describing something else.
The English language can be so confusing sometimes.

_________________
Old growth, shmold growth!

The flex (which I would define as deflection per unit length) should be fairly constant and not really related to brace length. A longer span is easier to flex, but the amount of flex per unit length should not change overmuch. I suspect this isn't entirely true, as some of the long-grain stregth due to length of fibre may impact this somewhat as well.

What he means with 'more flex', I think, is that the entire top can flex more if shorter brace lengths are used, because you don't have a continuous stretch of brace. i.e. uncoupled bracing.

A top structure need not be uniform across it . When you look at the top and how forces are applied you can use that info for your bracing. I am a traditionalist so I don't fall into a lot of fad bracing. Been there done that , With a top you have to have it on the guitar to know what you are going to get. I am talking steel string here.
After 150 guitars I can only say tapping a free plate will only tell you that you have a solid piece or a cracked piece. Once on the guitar , you can do a better job of voicing. I like to keep the top and back at least a semi tone apart. As for bracing , I do use some old style techniques. I prefer using fish or Hot hide glue , and I like to tuck braces. I also use a smaller bridge plate as Martin did pre war.
Most of my braces will fall into .275 to .300 in width. I also like them 9/16 high at the X. I don't think you can place a ratio on the top thickness and brace height at all points . The reason is that the forces applied change on the top. From the bridge to the neck block you have 2 main forces applied by the strings , A compression and B rotation. This is the area of highest load stress . From the bridge to the tail you have a tensional load. Structure can handle a tensional load better than compression so here you can be lighter.
The best thing you can do , is take information and then apply it and keep a building log. Many things about this hobby or profession is subjective , yes we have to build a box that can withstand the force applied to it, but it is the builder that will determine what and how that gets done . Keep the log and find the cause effect relationship that you like as the builder. This is what will set your guitar apart from anyone else.
we all have opinions about building but until you can actual replicate the effect you want you don't have the theory. Accept the fact we all hear things differently . After all when it comes to ears, and what we hear , it is subjective , if you want true factual information, get an oscilloscope they show the sound wave information without bias , human ears are not as exact. When we delve into subjective choices , it is in the end , like asking what is your favorite ice cream. Learning to apply numbers to the construction will help you get closer but finding what numbers are important is the journey.

_________________
John Hall
blues creek guitars
Authorized CF Martin Repair
Co President of ASIA
You Don't know what you don't know until you know it

Well put, John.
Another part in the mix here is using different woods.
That's where the inherent knowledge,
mixed with experience comes in.

John Hall wrote:
"After 150 guitars I can only say tapping a free plate will only tell you that you have a solid piece or a cracked piece."

Far all you can hear, that's about it. With Chladni patterns to show the mode shapes you can get a better idea of how the bracing and the top are working together, IMO, and try for the sort of balance that you think is likely to give you the sound you want. It's my opinion that there is a 'proper' ratio of plate to bracing, and that's the way to find it.

Of course, what the 'proper' ratio is will vary, not only depending on the wood used, but also with the bracing pattern. A pattern that uses a lot of braces, closely spaced, might well end up working best with a thin plate and tall braces, but it's hard to say. Like john, I tend to stay reasonably close to 'tradition' in my brace patterns most of the time, since that's what gives the sound people expect. When you _do_ want to take a flyer on something strange, or even just mosey on over to a type of guitar you don't normally make, Chladni patterns can help. Most of my guitars are 'fingerstyle' boxes, but when I wanted to replace a top on an HD-28 the mode patterns and shapes helped me get it right for that style.