Need interpretation!

[Dave Bagwill]

The following deflection setup, though crude, does give accurate and repeatable results.
The top is braced, and under it I have placed blocks as you can see in the pic.
I use a known 5 lb weight in different positions, and get very close results regardless of the position.

At 2.5 lbs the deflection is 4.5 mm
At 5 lbs the deflection is .9 mm

If I move the rear blocks further down in the lower bout, 5 lbs deflects the top 14.5 mm

I really have no figures as a target - would anyone here have an idea what a target deflection would be?

[Hans Mattes]

The only (possibly) relevant data that I recollect is Trevor Gore's suggestion that good tonal results occur when the angular deflection of the saddle/bridge is ~2 degrees as string tension is applied. Assuming 170 lbs of total string tension and 0.5" of string height above the soundboard, results in 85 inch-pounds (7 ft-lbs) of torque on the soundboard. String torque, of course, is entirely aligned with the neck and is measured when the soundboard is supported by the guitar sides, so translating his suggestion into your situation is a serious extrapolation.

A few years ago I built a metal framework that clamped a lever to the soundboard in the location of the bridge. I then mounted the guitar mold vertically in a vice and installed the soundboard (attached to the sides but with an open back) with the framework in place. As I added weight to the lever I could measure its movement toward the floor and find what deflection occurred when 7 ft-lbs were applied. Then I carved braces to achieve 2 degrees. It all worked; the results were fine; I no longer use that approach.

[Dave Bagwill]

Thanks Hans. I suspected Gore might have weighed in on this in one of his books.
Well all I can do is make a note of the results, and, if the guitar turns out exceptional one way or the other, I'll have the beginning of a data base.

[John Parchem]

Dave I think your idea is not a bad one. Do what you think is right and record the deflection. Depending how the guitar turns out look for more or less deflection in your next guitar. ALthough I also found Hans idea interesting; to test for bridge rotation before the back is closed.

[Dave Bagwill]

Brian Burns will put the top, joined to the body, in a form and then TAPE the back securely into place and then test for various resonances and other things. His test results with the back taped on are the same with the back glued on! So he runs his tests, undoes the tape and removes the back, and makes any changes to the top that he deems necessary. Repeat until happy.

[Hans Mattes]

Brian's technique sounds really interesting. Can you share any elaboration or photos?

I gave up on measuring bridge rotation when I transitioned to fan fretting (where the bridge is not perpendicular to the strings). Additionally, I now use a raised fretboard, a narrow oval soundhole, and internal headblock-to-tailblock bracing instead of a UTB, so I'm not depending on the toneboard for structure. As a consequence, learnings based on Martinlike X-bracing aren't much help. I do try to correlate spectral measurements with the resultant tone, but spectral measurements only count when the back is on, and my fat arms limit what I can do through the soundhole. But if I could remove the back, . . .

[Dave Bagwill]

Hans - sorry, I don't have photos any longer. I observed the process in Brian's shop but it was 20 years ago.
Brian is very generous with his procedures and information and would no doubt be happy to take some questions from you. If you decide to email him, here's the address, you can drop my name if you like.

brian@lessonsinlutherie.com