Official Luthiers Forum!

I am building a tenor guitar now and was into carving the braces when I noticed that the one lower transverse brace 'tone bar' was glued down flat sawn - Doh!. I guess I wasn't paying attention. This guitar is for myself and you know what? I'm just gonna leave it.

For the sake of science I made two more braces from the same stock each a square cross section of 1/4in and measured them using deflection in both flat and straight grained orientation and the in both cases the flat sawn direction was actually stiffer.

So there!

I'm guessing the usual argument is that a straight grained piece won't shrink and swell as much but I cannot imagine that on a brace that is 5/16th inch wide that would be significant.

It will be fine.

Yes - flat sawn spruce is nominally stiffer than vertical.

The general concern stems from braces cracking and misbehaving. Most likely, the perception is that they will peel along the grain lines. It's probably true if there's a big sap pocket or some hidden ring shake. That way - while the hidden defect is still present - it more or less stays glued down.

I worry more about curly grain or runout around the brace ends where you really don't want short grain to pop loose.

I think the other issue with flatsawn grain on braces is that it is easier to carve and shape with vertical grain. As far as the stiffness for your tone bar goes, you will be carving it where you want it anyway.

_________________
Bryan Bear PMoMC

Take care of your feet, and your feet will take care of you.

I'm with Bryan, thinking that the need for VG is mostly for ease of carving with bladed weapons, and, secondly, because flatsawn stock will introduce a lot of weak points via runout every place there is a scallop in the brace, which is usually quite a bit. If it was a 'load bearing' brace like the x, I'd be more concerned, but a lower transverse I'd worry less.

Yeah this is a scalloped brace too so it sort of looks like the topography of bedding planes in and around a desert mesa, if you know what I mean. I sort of wondered about hitting it with thin CA. I'm sure it will be fine as is. I remember being surprised to see flat sawn braces on an old Gibson but it lasted through the years too.

truckjohn wrote:
"Yes - flat sawn spruce is nominally stiffer than vertical. "

James Blilie disputes that in an article in the most recent 'American Lutherie' magazine. He bases this on measurements of a large number of samples of wood, and finds that flat cut samples average about 20% lower stiffness than quartered. Obviously there is wide variation in properties between wood samples, and any particular flat cut brace may well be stiffer at a given size than a given quartered one. In the end, you have to measure in order to know for sure.

There are so many varieties of spruce out in the marketplace that only time . testing and experience can deduct their strengths and weaknesses. I saw 1 article in the GAL where john greven broke down the spruces into categories of light med , and heavy, it/s so variable that outcomes are hard to predict. You can further breakdown the bracing samples ,with slash grain 60 deg angle vertical grain , 90 deg and flat sawn.

Didn't lutes use flatsawn braces?

Yes, lutes have flatsawn braces.
I question what large numbers of samples of which woods were tested. Unlike many hardwoods, every piece of spruce i have tested was stiffer when flatsawn. This is the same test outlined by the OP....make a perfect square and check stiffness in both directions. That eliminates the natural variability when measuring multiple samples.
Even resistance to splitting in the scoops is not better with vertical grain....the weakest plane is perpendicular to the growth rings. if you encounter ring shakes or pitch pockets, throw it away.

_________________
John

Blilie tested almost 200 samples, iirc. He goes into the test methods in the article. Keep in mind that Young's modulus along the grain in all softwoods varies directly as the density. He controlled for that. If you have data on more samples perhaps you'd be willing to publish it to establish your case?

It matters not at all if you are testing the same piece of wood in both directions. Unless the density changes in the few seconds it takes to turn a sample 90 degrees

_________________
John

A concern I have with the James Blilie article is that he uses a cantilever suspension method in his test procedure. It is notoriously difficult to control the clamping in such a way as to get repeatable results, with a tendency to crush the soft early wood at the clamp points of flat sawn samples once sufficient pressure has been applied to hold the sample firmly enough. This crushing effect makes flat sawn samples appear less stiff. The building industries have for years used flat sawn timber on edge (i.e. vertical grain) specifically to reduce the crushing effect when the piece is used as a beam with supports at each end.

When seeking a method for measuring damping in samples for the book I looked hard at cantilever methods, but could not develop a repeatable mounting system which clamped a sample firmly and precisely at a defined length. It sounds easy, but it isn't (like so many things). I reverted to a simply supported beam method, which is much easier to control. Consequently, I'd be reluctant to put a lot of faith in the Blilie results without seeing them duplicated using a beam method.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

I guess I should have taken pics and documented my method but I was only half hardheartedly joking about making a silly mistake and was just entertaining a curiosity. It was rather crass. I simply put the beam across two blocks with a dial gauge under it. I marked the ends on the blocks so that the very edge of the block when the beam was bending would touch in about the same place on each side. Then just put a weight dead center.

Trevor:
I agree with you on that. We had a long discussion about test methods with Ollie Rogers once at Carleen's, and felt that the free-free bar was best.

Trevor I did some small testing using centre loading of a beam of similar dimensions to braces. Within my experimental error I could see no difference in Youngs modulus with grain orientation. But my sample was small and experimental technique probably not great.

I wonder if you have done enough testing with centre loading to at least get a feel for the answer to this question...?

Dave

I haven't done enough testing to get a definitive answer (there's more important questions....) however, gut feel is there is not much in it either way. I suspect that the variability between pieces means that you'd have to test each piece prior to use if you really needed to know, and that the answer for flat sawn samples might depend on whether the surfaces were early or late wood. In use, there a a few reasons to use "vertical grain" over flat grain for the conventional bracing schemes but that's not true of all schemes.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au