Acoustic Guitar Construction

And the answer is --- drum roll please ----- leaving the side material too thick.

The absolute fact is --- Martin thins their sides to .075"

Think about that, many if not most guitar "clone" renderings (they are not plans since I have yet to see one where the math adds up and makes sense) so the side thickness to be .10 or so 25% thicker than optimum, that is a huge error when it comes to persuading wood not to be "tree like." $.02

Nicely explained Ken. If someone really wants a .10" side (and there may be good reasons for such rigidity according to Somogyi and others), the best approach is to make two .05" sides and laminate them together. As a bonus, this would be stronger than "solid" wood, and might help prevent cracking. For aesthetics, most would match the outer layer to the back, but the inner layer can be any wood with similar physical properties.

To make it really crack proof, laminate a thin layer of cross-grain veneer in the middle, and a third with the grain going long inside, and use veneer for the outer layer. This would be similar to the approach Michael Collins uses and may be the best way to achieve the extremely tight radius in the gypsy cutaway side. Or even the less radical Venetian cutaway.

This said, I have successfully bent .11" EIR sides using a hot pipe while trembling with fear of breaking them (but only for the loose waisted dreadnought outline). It is much easier to see what is going on when you do it by hand, which is always a good idea for the waist even when using a bender.

I thickness mine to bet .080 & .085". Now bending cracks yet. I figure by the time they are leveled, they are near the .075" mark.

.075" is the thickness of the Martin side blanks as they come out of their thickness sanders. Try it you'll like it -- I believe Martin make a pretty nice product. $.02

I've been bending at around .080 also and so far things have gone very well.

Perfect timing for this information - thank you.
I'm about to begin hand thicknessing some Flamed Maple and Ziricote - time to get the scrapers tuned up - .075" here we come.

Well if it's best to thin the wood to 0.075 to facilitate bending, how about thinning the sides in the area where bends are tight, for example the waist of an OM or cutaways, and leaving a bit of extra thickness elsewhere? The localized thinning can be done with a scraper, sanding block or by sticking a few staggered index cards under the side as it is run through a thickness sander.

This way you have the option of additional thickness for the rest of the sides, if your design goal is a rigid stable box for a light resonant top. It also leaves a bit more leeway for sanding and leveling. Also, the tight curve adds stiffness which offsets the effect of the local thinning.

Jake Booth