Official Luthiers Forum!

In a recent thread which was started on another topic, Hans Mattes raised the idea of structural bracing between the headblock and tailblock which is distinct from the soundboard bracing. The thread is here (viewtopic.php?f=10101&t=53505), but I will paste his comment below. I would love to have a conversation about this idea because I have been playing around with it myself.

Hans said this, and then posted the photo below:
"The unsaid background that drives "the luthier's balancing act" is the unfortunate tradition of using the guitar soundboard (probably the most critical element of the guitar) for two functions whose requirements are largely in opposition to one another: structure and tone. The need for structure appears because, in traditional construction, the neck/fretboard extension lays on the soundboard and the location and orientation of the headblock is, to a substantial degree, determined by the rigidity of the soundboard. With the duties of locating the headblock and supporting the fretboard being compromised by the soundhole, substantial bracing of the top is required and any contribution to musicality from the upper bout is generally sacrificed as a result. This is TOTALLY unnecessary.

An alternative approach to guitar construction uses braces between the headblock and the tailblock which don't touch either the soundboard or the back of the guitar and which rigidly fix the orientation of the headblock and, therefore, of the neck, with respect to the sides (or "rim") of the guitar. When this approach is coupled with an elevated fretboard (one whose extension is spaced away from the soundboard), the soundboard no longer needs an upper transverse brace or a popsicle brace or, in fact, any brace not useful for tone generation. The only exception to this is the need for braces to manage the twisting of the bridge under string forces. I've found that very light X-bracing that extends under the wings of the bridge is adequate to control bridge twist and that light X-bracing, along with light tone bars that are generally radially directed from the bridge will make a thin (< 0.100" thick) top perform well when not required to act as a structural member. Such a top will resonate over its entire area, including the upper bout, and can generate a richness of sound that many people will find pleasing. It will not, however, sound like a Martin guitar."


I completely agree with Hans that a design where the soundboard provides all of the structural stability for the box is probably sub-optimal for performance of the soundboard in sound/tone generation. I have been interested in methods developed by Rick Turner (he called it suspended bracing), Mike Doolin, and Hans Brentrup (who modeled his on old Larsen Brothers methods). Those three gentlemen are really accomplished and innovative luthiers, whose ideas I really respect. How come this has not caught on more widely? Using CF rods seems to be a good modern variation on this technique. I have tried it in 2 instruments and I will post photos below. One was a bouzouki-style instrument, and the other was an old drednaoght which I retopped (hence the messy looking kerf linings).



I would be keen to hear of the experience of others who have tried anything like this. Is there a disadvantage? If this sort of bracing is used how much lighter can the soundboard bracing be?

Tilton had a similar idea back in the 1850's, IIRC. He ran a metal rod from the neck block to the tail block. He also patented the guitar tailpiece. I had an old Zogbaum and Fairchild parlor guitar that used the tailpiece "Improvement"

I do agree using that type of a structure could eliminate the upper transverse brace making more of the top active. I not sure you could make the rest of the bracing all that much lighter. A lot of the bracing is to keep the bridge from over rotating. Also you need enough bracing to get the top resonating at a good frequency.

_________________
http://www.Harvestmoonguitars.com

Looking through my pile of derelict guitars I found the Zogbaum and Fairchild guitar and the Tilton Patent tailpiece. Unfortunately something has eaten every spec of glue that was holding the guitar together. Fun if you like building model airplanes!

Allowing the upper bout to fully contribute to sound generation requires more than eliminating the UTB; it also requires lifting the fretboard extension off the soundboard -- an "elevated" fretboard.* Preventing excessive bridge rotation is certainly a structural issue, but can be done without excessive bracing. Braces that are significantly north-south and that pass under the wings of the bridge needn't be particularly beefy over more than a few inches and can taper to more a more delicate profile past that section. Well-placed X-bracing with a modulated height can do the job yet not make the soundboard overly stiff.

It is worth noting, however, that the idea of bracing the headblock directly against the tailblock has been around for a long time. A number of guitars have been built as "thru neck" designs where an extension of the neck heel, parallel to the fretboard, protrudes through the guitar and is glued to the tailblock. A good idea (IMHO) that never caught on.

I've used a number of materials as headblock-to-tailblock bracings: hardwood, Baltic Birch plywood, carbon-fiber tubes, and aluminum tubes. All add weight, none add very much. The most straight-forward build used a single, 1" diameter carbon-fiber tube. It was light and strong, but easily visible through the sound hole, so not an attractive solution. My current approach is to use two wooden braces, as "tall" as possible without interfering with front or back bracing but with multiple "lightening" holes to prevent the braces from sectionalizing the acoustic cavity.

* An elevated fretboard has a number of interesting attributes in addition to helping "free" the upper bout. It improves access to the higher frets, especially for guitars without cutaways. It also allows bolt-on necks to incorporate a simple mechanism to allow easy neck angle (and, thereby, action) adjustment by the user. This, in conjunction with light-weight bracing of the soundboard, allows the user to switch between medium, light, or even ball-end nylon strings without sacrificing tone quality.

Thanks for the further thoughts Hans and everybody.
I rediscovered a good discussion at the ANZLF from some years ago which first sent my thinking in this direction. It points out the history going back to the mid-19th century, and also has some input from Rick Turner:
http://www.anzlf.com/viewtopic.php?t=675

If you consider the modern X-braced steel string guitar - with its load bearing braces attached to the soundboard - it determines a number of the characteristic aspects of the sound of the guitar. The soundboard and bracing have significant mass, which means that it takes quite a bit of energy to get it moving and it also takes some time to respond to the energy input. But once it is activated it retains energy for a relatively long time (=sustain). Within the soundboard there are different regions established by the bracing pattern and they can respond with vibration at different frequencies, adding overtones and harmonic complexity to the sound. Compare this to a banjo which has all of the structural strength built into the rigid rim and the skin is extremely light and responsive to energy input, but cannot hold energy for long. Therefore the banjo is loud, has fast attack when plucked, but very little sustain. The banjo also lacks overtones and harmonic complexity.

So, Hans, what effect do you think your longitudinal braces and light top has on the sound. Taken to its limits are we going to make a guitar which sounds more like a banjo? I started thinking of this recently because a friend asked me to build him a Selmer Maccaferri style guitar. Researching them I have realized they are more banjo-like than most guitars. The laminated back and sides and solid linings produce a very solid rim, and the tops are very thin (often <2mm) and just horizontally braced in the lower bout. Their sound is dry, loud and barking, without complexity or sustain. This is not necessarily a bad thing. While it is not the pretty sound wanted by a solo fingerstyle artist, it can be just the thing to cut through in a band. In an ensemble it might be better if the guitar’s sound occupies a narrow sonic spectrum and is not too overloaded with harmonic complexity that makes a muddy mix with the other instruments. Being loud also helps. I wondered if longitudinal end-to-end CF tubes would be a fitting addition to this design, contributing to a good rigid frame? These are all just my uninformed musings. Am I off the track?

Gibson put an elevated fretboard on the L-00 in 1933, for just a year I think. Most of these guitars sound great, but I don't know if the elevated board made any difference. Here is the inside of one that I rebuilt showing it still has a UTB despite the elevated board.

I'm sure someone here will have an opinion on whether they sound better or not.

Ed

Mark: Thanks for the ANZLF reference. As has been long noted, "there's nothing new under the sun." Rick Turner's discussion and conclusions are entirely consistent with my experience; he makes a number of intriguing suggestions that I expect to incorporate, going forward.

Regarding the guitar-banjo issue, the banjo's "soundboard" is far simpler than the guitar's. The banjo has an unbraced circular diaphragm with a bridge pressing near its center as well as an open back, and so, displays a rather simple set of resonances, with node lines that are either circular or radial. (Check out page 15 of http://www.its.caltech.edu/~politzer/ai ... ad-exp.pdf) Because of its soundhole and shape, the guitar has a resonance structure that is far more complex, with its closed back adding a Helmholz-like resonance. Any guitar soundboard bracing will add to that complexity, as the braces become resonance node lines. (This isn't simple stuff. The "page 15" reference is to a paper analyzing banjo resonances by David Politzer, a theoretical physicist at CalTech and winner of the Nobel Prize in 2004 for his work in "asymptotic freedom in quantum chromodynamics." And the banjo is FAR simpler than the guitar. While a goodly effort has been invested in understanding the guitar (see https://www.hindawi.com/journals/aav/2016/6084230/) it is only loosely understood, and, I believe, has only been studied for guitars with the fretboard extension resting on the upper bout and, for steel-string instruments, with an upper transverse brace.)

So, "taken to its limits are we going to make a guitar which sounds more like a banjo?" No -- but we won't "take it to its limits" in any case.

The purpose of "flying braces" (I like Benoit's description in ANZLF) is not to remove all soundboard bracing in a guitar, but to allow the bracing to be focussed on tone rather than having to balance tone and structural issues. As Rick Turner notes in the ANZLF thread, "All this coincidentally takes a lot of pressure off of the upper bout of the top, and with the cantilevered fingerboard, it allows at least the possibility of getting tone and volume out of the upper bout." Adding the musically useful resonances of the upper bout to the guitar's response will move the guitar's sound away from the banjo's, not toward it.

What do the chladni patterns show?

I have not been looking at Chladni patterns, so I can't give you an answer directly. I have measured frequency spectra (using Audacity) on guitars that I've built following both traditional designs and designs using flying braces (with an elevated fretboard and light toneboard bracing). The major difference between the spectra of the two build styles is that the flying brace guitars have more resonances that couple well to the bridge, and so, appear more prominently in the spectal displays -- especially at higher frequencies. The differences between the specta of the build styles typically begins around 250 hz and becomes more pronounced above 1000hz. That said, there is not a massive contrast between the spectra of the two styles of build and I can't associate any specific resonance(s) with flying brace construction.

I would be very interested to hear how guitars built like that do sound.
Can you post sound-clips of standard fretted (not fan frets) guitars please?

_________________
The name catgut is confusing. There are two explanations for the mix up.

Catgut is an abbreviation of the word cattle gut. Gut strings are made from sheep or goat intestines, in the past even from horse, mule or donkey intestines.

Otherwise it could be from the word kitgut or kitstring. Kit meant fiddle, not kitten.

Colin: I'd like to accommodate your request, but I can't. Two reasons -- I have no mechanism to make quality recordings, and I have no "standard fretted" guitars built with "flying braces."

What I can do is share a short story -- I took an early version of a flying braces build (with fan frets) to a meeting with Bruce Sexauer, a top-tier luthier selling guitars in the $10,000 to $20,000 range (http://www.sexauerluthier.com). After I described the construction he played the instrument and commented, "I really expected this to be terrible, but it sounds terrific." Bruce is not wasteful of his praise.

I hope this is useful.

OK, I have decided to do the controlled experiment. I will build two essentially identical guitars, but one with flying buttress braces and one without. And standard frets. The only problem is that it usually takes me close to a year to get one guitar finished, so two simultaneously will take me a while. I’ll get back to you when I have some results. The answer might be 42, but I do want to know.

If you want it to be worthwhile, you'll need cantilever fretboard extension on both so the upper bout will be free to vibrate on the buttressed one. And of course non-identical soundboard bracing between the two. And the non-buttressed one should probably have some bracing between the headblock and upper transverse brace, because the fingerboard extension functions as a brace in the traditional style. Either A-frame braces or headblock extension going to the upper transverse brace.

Another note-worthy point about Hans' design is the oval soundhole, That's a good way to boost the output of a lightly braced upper bout, by not cutting across so much of the soundboard grain and isolating it from the bridge area.

A note on this -- my oval soundhole was, indeed, intended to couple better to the upper bout. But, as my construction skills are consistent with my amateur status, my manual attempts at oval soundhole, rosette, and rosette channel cutting were seriously inadequate. That's when I bought my first CNC router. It cuts excellent oval soundholes, rosettes, and rosette channels. It also gave me the courage to attempt fan fret designs and is useful for necks and bridges as well as bas-relief headplates and tail wedges. And it cuts very accurate nuts.

Hans (and Dennis , and others) - could you share some pictures or other details of how your elevated/cantilevered FB extension is done?
mm

Mark asked, "could you share some pictures or other details of how your elevated/cantilevered FB extension is done?"

Here's a picture of the neck heel with the hole near the bottom for adjustment of neck angle/action. and an elevated fretboard extension over the (non-centered) soundhole. This example is of a harp guitar, but I chose it because the neck arrangement shows up best here. Disregard the carbon fiber tube; it holds the headstock for the unfretted harp strings. The round plug insert with the small hole in the neck heel allows a 3/32" hex wrench to access the head of a 10-32 cap screw that controls the neck angle. The plug can be chosen and finished so that it is not noticeable.



The picture that Mark copied to start this thread give a good view of the headblock from the inside of the guitar.

Below is a "copy" (except that I had to convert line drawings to character "drawings") of a Word document that I wrote to remind myself of the steps needed (and the order of taking the steps) to set up the adjustable bolt-on elevated FB neck. I hope it's useful.

Start with the rough-cut neck, and before prepping for adjustable, bolt-on attachment, this needs to be the status:

-- the fretboard is not yet attached to the rough-carved neck. This allows the neck surface, ready for the fretboard, to be a good reference for all cuts.
-- the neck heel is 2" wide where it acts as a tendon against the headblock. You may want to choose a narrower heel; I use a wide fretboard because I have fat fingers, so a 2" heel works for me.
-- if it hasn't been already done (and it should have been done before the sides were glued to the headblock), an appropriate mortice (2"?) needs to be cut in the headblock from top to bottom. A 1/4" deep mortice is adequate.
-- the neck heel face, where it will press against the headblock in the mortice, is cut 90 degrees to the fretboard surface (the adjustment screw will assure proper action/neck angle when all is done).
-- the neck heel should be drilled with a 3/4" hole, centered in the heel, from the bottom of the heel to 1/4" from the fretboard, and a 3/4" maple dowel glued in, with the grain running crosswise to the neck direction. That gives the metal screw inserts a good grip and strengthens the neck heel. That's important, as string tension will try to bend the neck heel.

Now for the "copy" of my notes:

ADJUSTABLE BOLT-ON NECK

1. Drill 1/4” hole in headblock ~ 3/4” from back of guitar, centered in mortice
2. Install 10-32 T-nut in headblock on the "inside"
3. Fit neck heel into headblock mortice & mark a hole in the heel face using pointed 10-32 screw in the T-nut
4. Drill 13/64” hole through heel at the mark, from the heel face through to the "front" of the neck
5. Drill neck heel as shown in drawing
6. ============================================== < fretboard
. . . --------- . . . . . . . . . . . _______________----------------
. . . . . . . . . . . .|. . . . . . . . . . /
. . . . . . . . . . . .|. . . . . . . . . /
. . . . . . . . . . . .|. . . . . . ___/
. . . . . . . . . . . .|. . . ____| ./
. . . .13/64" ->|==|___ . /<---- As needed for plug (e.g., 1/2” dowel)
. . . . . . . . . . . .|. . . . ^ |_|
. . . . . . . . . . . .|_____|___|
. . . . . . . . . . . . . . . . |
. . . . . . . . . . . . . . 15/32” hole – start w. conventional drill, finish w. brad point drill for flat bottom near the middle of the dowel

Note: unfortunately, this drawing, which is a freehand sketch in my Word document, doesn't translate well to OLF presentation. Try to ignore the "periods", but it is the only method I found to give any control of spacing (multiple spaces are collapsed into a single space by the OLF formatter.) The arrangement for the 10X32 screw has holes of three diameters. Starting at the flat heel face is a 13/64" hole that passes the shaft of the 10X32 screw. The 15/32" hole (drilled from the open side, not the heel face side) provides a flat face for a washer on the 10x32 screw (the 15/32" diameter is just larger than the washer OD). The washer is a bearing surface for the screw head and sits in the center of the maple dowel. The next hole, opening on the open face, is sized for a plug to hide the head of the 10X32 screw. The plug should have a hole that is centered on the screw to allow a hex wrench to access the recess in the head of the 10x32 cap screw. I get all the needed hardware from McMaster (mcmaster.com).

Note: the drawing only applies through step 6.

7. Drill a 1/4” hole in headblock, 1/2” from top edge
8. Put the neck in place, and fully screw in the 10X32 screw. Use a pointed 1/4” rod in the headblock to mark the neck heel for drilling (a 1/4x20 screw, sharpened on a bench grinder, works well)
9. Drill the neck heel 9/16” deep w. 5/16” drill, chamfer the hole, & install a 1/4x20 softwood screw insert
10. Tap/clear the insert – verify w. 1/4x20 screw
11. Enlarge the 1/4" hole in the headblock to 3/8”
12. Insert a 10-32x1.25" socket head cap screw screw/washer in 15/32” hole & verify length to T-nut in headblock
13. Install a wooden plug (w. 7/64” hole for hex wrench) in front side of heel – verify hex wrench fit to the screw

The final steps are to install the fretboard, finish carving and sanding the neck, and cut the neck heel to appropriate length and angle and install a cap.

I use a 1/4"x20 cap head screw with two wavy washers (separated by a flat washer) in the 3/8" hole in the headblock to hold the neck to the box. This screw should NOT be fully tightened. This screw and the 1/4"x20 insert aren't absolutely needed; 150+ pounds of string pull will hold the neck heel FIRMLY against the headblock mortice face. But without the 1/4" screw, the neck wobbles around during string changes, and some folks might not want that. The wavy washers allow the neck to be fairly firmly positioned while final positioning is done by adjusting the 10-32 screw after the strings are on and at a loose tension (which allows proper action to be determined).

I know this is complicated and I hope my explanation is helpful. If clarification is needed, let me know.

Like Colin, I'd like to hear some of the results of these experiments. The discussion is certainly interesting.

I'd also be interested in how much lighter the top with bracing is compared to a "standard" X brace guitar. I have found that T. Gore's statement is generally true. Namely, about 80% top plate and 20% bracing. Yet, 80% of the top's stiffness (structural capacity) is in the bracing and only 20% in the top plate. This would seem to say that to make the top system significantly lighter the top plate must be made thinner to reduce mass since most of the mass is in the plate.

Hans - thanks for the photo and the information. I think I get most of the concept. But I am a bit uncertain, when the neck angle is being altered by the screw towards the bottom of the heel - what is the neck pivoting on? And I imagine that the depth of the body at the neck end is less than a guitar with a conventional neck joint. so that the fingerboard can project that extra bit higher than the soundboard?

The neck heel pivots on the upper corner of the mortice in the headblock. 150 to 170 pounds of string tension assures good contact. (It's important to assure that no damping material compromises that contact.) I don't reduce the depth of the body at the neck. My current build, a "small jumbo" with a lower bout that is 16" wide and a 4"-5" deep (Manzer wedge), is 3 7/8" deep at the neck. The neck heel projects 4 3/16" below the fretboard.

I think one reason the "Tilton Improvement" didn't catch on was that it obstructed doing repairs through the soundhole. Similar to the Tornavoz that also had a support function to free up the top (in addition to it's "sound enhancing" function) the benefit wasn't seen to be as great as the disadvantages it created.
An improvement to the "Improvement" might be to use a small turnbuckle or similar device and fit the rod into pockets drilled into the head and tail blocks, so the obstruction could be removed to effect repairs through the soundhole when necessary.
Or Not.....
Sometimes the cure is worse than the disease.