Official Luthiers Forum!

I bent ribs for a violin and a Staufer guitar yesterday. The same piece of wood, some quarter sawn, curly Padauk. It looks cool. I rough carved the violin back, and have the guitar back read to put the three ladder bars on, and had no problems. I wondered about bending the ribs though.

So I did the violin ones first. The dark ones bent fine, The light ones didn't. The curl was almost already broken through. I cut some more, and I got the two to bend. I'll have to put some CA in at least the one crack that you can see in upper bout on the left. It was there before I started. Like checking, but cross grain with the curl.

The guitar ribs bent pretty well. They are 1.5 to 2.0 mm. The ribs on a Gofriller cello I have a plan for has ribs that are 1.7 on the bouts, and 1.3 in the c bouts. I went thicker than that. The ribs are only 3" max. There were a couple spots on one where the curl wanted to pop out instead of bending. Not all the way across, and not all the way through. Easy enough to fix that.

I thought of gluing on some linen with HHG to add a bit of stiffness, especially over the curl to keep it from even thinking about being bad. I've seen people recommend that before, but they always say not the entire rib. Any reason not to?

The wood does have amazing resonance.

_________________
Why be normal?

A full layer of fabric and glue will kill the resonance.

and would this be a bad thing?

serious question...

Somogyi and all of his apprentices use laminated sides...doesn't this also kill resonance?

I truly understand what you're saying, but what I wonder is does "killing" the effect of the sides in the equation give the builder more control?

by this I mean that modeling what's really going on with an instrument is very complex math, and eliminating one of the variables (resonance of the sides) can only make the math easier to understand/control...

it certainly seems that everyone concurs on the concept that it's the top that is most important, followed by the back, and certainly the sides impart some effect...

now, if how the sides interact in it all could be quantified and accounted for (I'm not saying they can't be, rather that is makes the math far more complex) then I'd be fully in the camp of solid is better (which I used to be until I thought about it all in a mathematical formula sense)...

the whole point I'm making is that perhaps a better sounding instrument can more easily achieved by focusing on the top and back and eliminating the variable of the sides by making them non-resonant...

Except for the small, nearly flat, area below the waist any vibration in the sides will tend to be at a high frequency due to the curvature. Eve that area probably normally has it's resonant pitch well above most of the 'main' resonances of the top, back, or air. It's possible that it could couple with some of the higher order resonant modes of the air, say those involving 'sloshing' across the lower bout, but it's a long shot on that affecting the sound.

As for damping in the material itself, there doesn't seem to be much agreement about the effects of damping even in the top wood. It's hard to see how you'd demonstrate anything like that in the sides.

I have heard of people using a layer of cloth (often silk) between side laminates, and that has not seemed to produce problems with the sound, at least according to the folks who do it. Cloth on the surface would produce a texture, and there is debate over whether surface roughness influences sound. Again, I don't think that has ever been resolved to everyone's satisfaction.

Side stiffness might be a factor in sound transmission between the top and back. Side mass certainly is; Gore goes into that in his books.

I've used cotton-polyester tapes, about 5/8" wide and spaced every 3-1/2"-4" or so, to reinforce the sides. These increase the force necessary to crack a side, and tend to stop cracks from running if they get started. It would, again, be hard to determine whether they hurt the sound at all; any effect much be minimal, and the increase in durability seems well worth any putative cost in tone. Stradivari used similar tapes (in linen, of course) on his 'cellos, and nobody complains about the way they sound.

I used carbon fibre as a center lamination between two pieces of wood veneer on a back and side set and aside from a "tupperware" tap tone the combination worked fine and the guitar sounded good.
But if you are building a replica of a romantic guitar you may want to stick to traditional materials and construction to get a traditional sound. Some lutes had parchment linings - haven't seen it in a guitar, but that doesn't mean it never happened.

I will probably just use strips. I haven't really tried to figure out how guitars work. I don't understand violins either! I don't use modern methods; I'm not saying that they are wrong. I just don't want to do all that. I just use simple area turning that seems to work for those who build large soundboards like pianos and harpsichords, where you couldn't measure thicknesses easily. It seems to work well for getting even thicknesses, and must account for differences in stiffness.

I see how the back of a violin would easily add some to the sound; it is coupled with the sound post. Guitars only have the blocks and the ribs. On an arch top, there may be more going to the back because of the tailpiece peg, and the saddle on the lower block; but it isn't like a sound post, where there is a constant driving force. So would stiffer ribs allow more driving force to the back? I don't know. Benade said that weight was somewhat equal to stiffness; so maybe the weighted sides Gore suggests, (I haven't read Gore at all) could be done with stiffness instead? I don't know.

I don't know is my key phrase!

I haven't noticed that much difference in different woods and different sounds. It seems that if you thin the stiff woods down more, and make the flimsy woods thicker, you eliminate a lot of the difference already. A very stiff Birch back with a .38 sg Englemann top violin sounds like a violin. It is very easy to play (I don't play, but it is easy to play) and maybe because the stiff back supports the flexible top. It does like lighter strings. But the perception of the user; especially the violin buyer is probably the biggest problem. Only a rebel would by a Walnut/Alaskan Yellow Cedar, or a Padauk/Port Orford Cedar violin. But if you don't make them, how do you know?

The Padauk is not that hard. It is only .65 sg, so not any different than some Maple is.

_________________
Why be normal?

The back can enhance output in the lowest range of the guitar. Higher in pitch it tends to 'steal' energy from the top, so it reduces the output, but helps to produce 'tone color'. Guitars are very complicated acoustically, IMO deliberately so, but, at least to begin with, you don't really need to worry to much about that. A lot of what's supposed to happen is pretty much taken care of in the design, which has been worked out over centuries. Just stick with 'standard practice', use decent materials, and do a careful job, and you'll end up with a better than average guitar.

Padauk is a good wood for B&S. It has very low damping, and bends pretty well. On the down side, it is prone to splitting when bent across the grain, so don't get too carried away with flexing the back. I like to glue a layer of paper on the outside of woods like that when bending the sides; it helps a lot, and is no too hard to scrape off later. The open pore structure can take some time to fill in the finishing stage.

"Benade said that weight was somewhat equal to stiffness"

IIRC Benade was using weights to find areas of the soundboard to reduce the stiffness of to bring them to the characteristics he was looking for. So in a sense weight was inversely comparable to desired stiffness, i.e. more weight - less stiffness.

Yeah. You thin spots where the weight was, I've done that before. So ribs as thin as possible! Makes more sense than 600 gram ribs. Good catch Clay, I was thinking about weight being stiffer. Because if the weights work, why not just thin them???

_________________
Why be normal?

I tend to agree with Alan - guitars are complicated acoustically. I think there are several ways to build them with different but equally valid outcomes. Some guitars use lightly built sides and work well, while other guitars have very stiff sides and work well, still others have heavy, but not stiff sides and work well.
The old Martin size 1 guitar I have uses 5 cross braces on the back. On some of Torres's guitars he only used two. Dreadnought guitars use 4. They all work.
If you follow a proven model, as Alan said, you will probably make a decent guitar - of whatever type you are making.