Official Luthiers Forum!

Hey folks, Into the fray:
This is my first proper posting here. The reason I joined was to engage with serious luthiers on a project and concept I've been working on actively for the last year and one that started about 40 years ago and got blocked on and recently picked up.

Way back when I built a 12 string guitar, my first acoustic and I was incredulous about the idea of attaching 12 steel strings to that diaphanous piece of spruce, bracing notwithstanding. The destructive potential of 160-200 lbf of string tension is well known and indeed on 12 strings, often leads to grief. That instrument hangs on my wall waiting for repair a long time now.

It got my brain to thinking about how to systematically deal with the problem of static stress on a soundboard. The exact train of thought that brought me to my idea is now lost to me. I started to work on a prototype built on the carcass of a failed 12 string dreadnought from which the bridge had catastrophically separated, as they do. I got blocked for lack of tools, technology and time. All these years, it was hanging on my wall begging me to compete it after, this or that project. I'd started building guitars again starting with some electrics but thought: this is really more important than those fun baubles, maybe something significant and really new.

I was afraid to look to see if someone had beat me to it. Though a lot of folks have tried or at least proposed torque-compensation devices, novel bracing approaches and strategies, as far as I could, amazingly, nobody seem to have happened on *my* "one simple trick". I somewhat hastily fixed the problems of that old prototype with new technology, forged on and completed a baseline proof-of-concept to see if it was even remotely successful.

I was amazed that on first stringing and testing, the concept worked as intended. The soundboard is not subject to any net string tension forces, longitudinal, torsional or transverse (downward). The higher 25mm floating bridge has increased leverage to transmit acoustic energy into the soundboard which is free to be an acoustic amplifier and is not a structural member.

Since the first test, I've iterated on measurement and characterization of this control no-brace configuration, while improving compliance of the soundboard and other design aspects and have achieved a monopole mobility of 0.0231, a very responsive configuration though with some interesting resonance. increasing responsiveness and soundboard mobility does not come at the cost of delicate shaving of braces bringing the instrument closer to structural crisis.

I have only begun the process of exploring the options available to structure the soundboard acoustics (via bracing or other methods) and am working on some new instruments from scratch built to explore these. I've been stealth on this project so far, just sharing with family and friends and one well-known luthier on a private basis. Now that I have my IP ducks lined up, I'm excited to this method I call the "Evergreen Guitar" with luthiers.

As I said, I'm just getting started in exploring the potential of this approach. I'm not an experienced luthier with a lot of work to show for it. I'm definitely inclined to the Trevor Gore school of measurement/engineering/acoustic science-based interest and approach to what I'm doing.

I think I'll just leave this here and if you want to see a video of the prototype in action, you can check it out on my website at glyphstone dot studio.
Thanks for reading. I look forward to interesting discussions.

_________________
Breanna Anderson
https://glyphstone.studio

That is an interesting concept! Good job.

There are some good ideas there, I think. Certainly, the guitar sounded great in the video. And a monopole mobility of 23 is fantastic.

Using an internal brace (not touching the top) from the neck to the end, to take the string forces, makes a lot of sense. Why should the top do that work? The top just needs to be the right stiffness to vibrate at the right frequency. There's no reason it should also be structural. It requires anchoring the strings at the end, which may (?) reduce the long dipole mode around 400 Hz, compared to the usual anchor on a flat-top, but it sounded good to me.

I think Riversong Guitars does something similar, structurally. Except they use the traditional string anchoring.

Here's something to consider. If you can build a top that takes the forces (in this case, purely downward) at a certain thickness and with no braces, then you can always build a thinner top with braces that takes the forces and vibrates at the same frequency, but is lighter. Braces are just a more efficient use of wood. Like the joists in a floor.

Greg

I'm not quite seeing what the trick is... The tailpiece and wooden bar through the center to take the longitudinal string force off of the soundboard is one thing, but how is vibration transmitted to the bridge without downforce >= the maximum upward vibrational force?

15 years ago, Randy Muth posted this approach, which may be the same thing https://www.luthiersforum.com/forum/viewtopic.php?p=407422#p407422
But I never heard anything more about it, and never could find the original patent, so I've been waiting patiently until it's definitely expired 5 years from now.

The trick is instead of having a vertical break angle over the saddle, there are diagonal slots in the saddle to form horizontal break angles, half left and half right so the overall torque on the saddle cancels out before the soundboard feels it. But that results unequal distance between strings behind the bridge, whereas yours looks like they run straight.

I've seen another zero-force concept where the bridge had 3 horizontal metal rods, two high and one low between them so the strings are bent down and then back up to cancel out the torque while creating an effective break angle. But I only see one line of saddle on yours.

Have you found a third way?

I didn't discuss *the trick* indeed in the original message. It was getting long and I didn't want to go all the way down the rabbit hole if nobody was interested.
The method for canceling out the vertical (about 36lbf downward force is that the tailpiece is terminated in stainless cables that go through brass (low friction bearings into the guitar interior and these cables go over a counter break angle over what I call "force compensators" in this iteration carbon graphite springs mounted on the underside of the bridge plate. This neutralizes the downward force at the bridge exactly as it is the same as the net of the string tension.

For the very first experiment to see if the force cancellation idea worked, I used a rigid "sub-bridge" and it worked effectively and simply to cancel the static forces but at the cost of also dramatically reducing the soundboard mobility. It sounded a bit boxy due to the snubbing of the monopole, but that was just a quick test.
Since then I've been tweaking the "force compensators" using different approaches and the most recent gives me good control of the K (spring constant) with minimal weight. The *previous* iteration of force compensators had a slightly higher K and was more complex. This iteration is easier to precisely engineer and i intentionally reduced the K with a goal of beating Trevor Gore's best published monopole mobility number and the engineering was right on . The overall weight of the new constructs plus the height adjustable bridge is a bit too much so the resonance kind of got out of hand but that was an interesting experiment anyway. the Air (helmholz) resonace is 82Hz and the Low body resonance is 145Hz so the wolf tone on the D is kind of ugly but the low resonant sound is quite something and the responsiveness is really quite shocking.

I built this instrument with a removable soundboard using a spline around the perimeter so it takes me just a few minutes to remove or re-attach it so swapping out components is trival which allow me to iterate very quickly and easily.

I believe that at minimum, some cross-grain bracing, very light, only enough to make the soundboard more orthotropic will be a likely good approach. The chladni patterns showed the lateral tripole of the previous experimental iteration at only 284Hz which is crazy low due to the "undisciplined" cross-grain. I'm going to keep this prototype-0 probably without bracing just to have a control example and experiment with some unusual topologies and acoustic focused bracing on my upcoming OMs.
The ability to engineer a very low K for the soundboard raises the idea of engineering a parlor or 0 style instrument with a conventional 100/200 resonance with some ease. Lots or things to work on.

_________________
Breanna Anderson
https://glyphstone.studio

A third way indeed! That will probably get around Ned's patent, as the bridge itself isn't zero-force, but rather counterbalanced by an underside bridge. You could even use a second set of tuned strings in there, to get some interesting reverb effects along with the downforce cancellation. I don't know exactly what Fred Carlson's sympitars are like inside, but I assume the internal strings counterbalance the torque of the external ones. But he doesn't use a tailpiece, so the soundboard is still heavily loaded in the axial direction.

There have been posts recently about a similar effort called the 'Turbo Tail'. He uses turnbuckles attached to posts on the bridge to counter the torque and take up the tension load. He has a web site with the obvious name. Look up also the 'Bridge Doctor', which uses a post inside attached to a rod that goes to the tail block. I've used internal columns to take up the string load on archtops and harp guitars. I know of lots of similar efforts over the past fifty years: the problem is not new, after all.

British designer and wood worker David Pye wrote:
" Where the problem is old, the old solutions will nearly always be best (unless a new technique has been introduced) because it is inconceivable that all of the designers of ten or twenty generations will have been fools"

I don't mean to knock your effort, but basically want to inject a reality check. By all means keep trying: you may be the person who unlocks the key to the problem. Just don't be surprised if you find that other people have tried the same things in the past.

I'm familiar with the mentioned efforts and the patents associated where there are any. I have done pretty thorough prior art search with professional support. Those are all valiant efforts but I, not to knock *them*, I think they miss the mark of paradigm shift. Most use the status quo conventional bridge and string termination as the starting point and take a direct line solution of a counter torque. It seems that most are though of as add-ons to conventional flat top instruments.

When I came up with this concept 40 years ago I was *not* aware of those tricks/techniques and most of these were developed since the 1980s. I think I was inspired by some book I don't have in my library and have only hazy recollections that critiqued, validly, technical issues with the state of the flat top guitar, relative to classical instruments with their long useful life, powerful delivery. The high bridge of the violin family for example providing strong delivery of acoustic energy into the soundboard. Those instruments of course use the arch as their defense against the more modest string forces but those forces are in some ways simpler: longitudinal and transverse and not the combined longitudinal *and* torque forces applied directly at the center of the soundboard that we have with the acoustic, flat-top guitar.

By not questioning the bridge and string termination and working around it as the starting point, those solutions are boxed-in IMHO.

The closest IP we found was a now expired (but still relevant) patent by a French luthier in 1994 for a *lute* that featured drone strings interior to the neck but passing under an internal bridge. It was not a very detailed patent and I seriously doubt it was ever actually implemented and there is no further record of the patent holder or an examples but the idea of the sub-bridge and strings passing under it to counter the downward, transverse string force is super similar to my approach. I suspect he was thinking about the drone strings used in the lute family and how to leverage them in a unique way. There was no mention of compliance (spring) necessary to maintain soundboard mobility and the strings require manual tensioning with their own set of tuning pegs (I assume). This is similar to the weaknesses of almost all other *torque-counteracting* strategies: they require manual adjustment of the counter force: A tricky and fraught issue in many ways. Further none of them address the mobility issue save one or two, that included springs and such. Most counteracting strategies also add substantial mass to the soundboard and that's not a good way to go.

All that said, one of my motivations for talking about this while in the development and intra-patent phase I admit is looking for some prior art, patented or not, that overlaps the claims I have laid out in my provisional so thanks for the tips and references.

_________________
Breanna Anderson
https://glyphstone.studio

"The high bridge of the violin family for example providing strong delivery of acoustic energy into the soundboard."

Drawing parallels between violin and guitar practice is generally hazardous. The primary driving force for the guitar top comes from 'vertical' displacement of the string: motion perpendicular to the soundboard. Violin strings are driven by the bow horizontally: more or less parallel to the soundboard plane. To produce sound effectively the force on the soundboard has to be 'vertical'. The bridge and sound post of the violin work together to produce this. The post produces a hard spot on the top at the treble foot, forcing the bridge to rotate around it. That gives a vertical force at the bass foot over the bass bar. The violin bridge is a 'bell crank'. The taller the bridge the more leverage it has to move the top, but this comes at a price. Forces on the top at the bass foot can also produce a lot of transverse motion of the bridge top. This happens most notably with 'cellos, where a strong 'air' resonance can line up with one of the lower notes. When this happens the bridge top is not stationary at that pitch, and the string length is no longer well defined. Since the working of the bow depends on the reflected wave from the bridge to break it loose the string no longer gets driven reliably at that pitch. Instead it tends to shift upward an octave to where the bridge is stable. Once the fundamental that was driving the offending resonance is no longer there, that vibration dies out, the bridge becomes stationary, the wave reflects, and it shifts back down in pitch. The rapid shift up and down in pitch produces the well known 'wolf tone'.

I'll note, BTW, that it's generally accepted that the guitar is far more efficient at turning string energy into sound than the violin. Guitars are seriously under powered, and we need to make the most of the little bit of power we have available in a pluck. IMO fiddles have the opposite problem: they get in so much power that the issue is how to handle it. That's one reason they're so much more heavily built. Many of the problems we run into occur precisely at the pitches where the guitar is most efficient, and this poses limits.

"...the idea of the sub-bridge and strings passing under it to counter the downward, transverse string force is super similar to my approach. "

I think it was undertaken for acoustic reasons, and owed more to the 'viola d'amore' and the Swedish 'hardanger fiddle'. These have strings that run in a hollow neck, crossing through the bridge well below it's top. They can't be bowed, of course, but they do 'steal' energy from the top to set the drones in motion. This works against any sort of sharp attack or decay of the sound: the drone ramping up in amplitude reduces the force the string can put on the top. Once it's going, the drone continues to drive the top even when the bowing stops at that pitch. The timbre is, strongly legato, to put it mildly: no jigs and reels...

Fred Carlson picked up this idea from his wife, Suzi Norris, for his 'musical sculptures' based on plucked instruments. It would not surprise me at all if the lute in question had a similar inspiration. Fred has also adopted the sitar style of bridge in some of them. So far as I know he has not used the drone strings to apply a countervailing torque.

The tailpiece on an arch top guitar effectively eliminates the torque on the bridge, but, of course, leaves one with a down load. String pressure lengthwise on the top from the end pin and it's saddle push upward on the arch, and many people believe the 'up' and 'down' forces can be made to cancel out. Certainly increasing the down load on the bridge by increasing the break angle can 'choke' the sound if it's carried too far.

I'm beginning to see where you're going: I may be slow but I'm not too quick... Pay attention to any resonances your setup can introduce into the system. See the 'cello 'wolf suppressor' as an example.

I'll be interested to see what you come up with.

The admiration of the advantages of the violin are inspiration and insight into what makes them good but my intent is not to make the guitar a violin but simply to understand why and how the makers came to the conclusion of what works there and what lessons are to be learned. Certainly the lineage of the guitar overall is split between the inspiration and influence of the lute by way of the classical and flamenco guitars of de Torres via Martin and the violin and the arch top instruments via Gibson. I'm not really that educated an historian of the guitar so I'm sure that many other threads and influences have also contributed.

As you noted, the division between the flat top, primarily X braced instrument with great and versatile qualities is saddled, for the most part, with the challenge of mastering the torsional forces so it can do it's job of making beautiful music.

Gibson started out by applying the influences of classical instruments with floating bridges, tailpieces and arched tops with f holes. Both the bracing-oriented or the arch oriented approaches to managing string forces in their various ways all led to a stiffening of the soundboard and all that flow from it. In both cases however, the sound producing diaphragm is still left under unrelenting pressures, a constant force offset that the acoustic energies are overlayed on top of.

The challenge of the luthier, as I understand it, is to execute the structural prime directive of the instrument's survival on one hand and the aesthetic raison for the instrument: to be a responsive partner to the musician on the other; to amplify and resonate the motion of the strings, while maintaining things like sustain, acoustic response as appropriate to the playing style.

Way back then, i had the hubris of youth to think that I could come up with something new that would break the stalemate, well practiced by many generations of luthiers with wonderful results very often. Exceptional claims require exceptional proof however and I did and do expect some skepticism. My intent is to be careful, methodical and show my work and hopefully make some worthy instruments that help to validate the potential of this approach.

_________________
Breanna Anderson
https://glyphstone.studio

Over the years I have worked through a number of ideas to 'improve' my guitar making with the result that I'm firmly in the traditional construction camp - ha. Doesn't mean I'm against new ideas at all; I've tried many different ideas and I built a falcate braced dred that's out in the wild that is well regarded by some of our local players. The demo sounded good, I'm looking forward to following this to see where it leads.

_________________
Steve Smith
"Music is what feelings sound like"

I've grown less fond of traditional steel string construction over time. Particularly the acceptance of neck resets. Unreasonably expensive with dovetail (as proven by Brad's truss rod catastrophe), and still painful with bolt-on, which gives up some of the romance of tradition and adds weight and often a chunky heel that wastes a lot of the benefit a cutaway gives for playing at higher positions without unhooking your thumb from the back of the neck.

But most of all, zero-force bridge concepts interest me for instruments with more than 6 strings. Harp guitars in my case, and 12-strings in Breanna's case. If 150lbs of tension is already a tightrope between structure and tone, 250lbs+ is sacrificing tone for sure. Especially on harp guitars where you need lower frequencies than usual.

Harp guitars are fun, for sure. I've made a few forays into that realm, including some fairly 'odd' ones. I even came up with a design that Fred Carlson hadn't tried, where the sub-bases pull 'up' (or, in this case, 'down') on their own soundboard: a real 'harp' guitar, at least in part. (It's now in Greg Miner's harp guitar museum, and you might be able to see it on line). I've used that separate soundboard, with minimal static loads, for the sub-bases ever since. And, yes, I've used an internal pillar on some of those.

One of the issues with all of this is that there's a circular discussion going on. Guitars are tools for making music. Players usually buy an instrument to play existing music, or something close. The existing designs produce that sound pretty well, but most of us would like more of it. Any changes you make, however, tend to produce a somewhat different sound. Les Paul and Leo Fender wanted louder guitars to play with Jazz and Country bands, respectively. The solutions they came up with engendered Rock and Roll; an outcome they didn't foresee.

The longer I study this the more I believe that the existing designs have been pretty well optimized. They're not perfect, of course, and the deficiencies are well known, but it's very hard to make significant improvements without basic changes. Those tend to throw out the baby with the bath: you end up with something that is not particularly good for the traditional use. Playing 'Recuerdos de la Alhambra' on a solid body guitar is more a novelty than anything else.

To get an instrument that works for existing music you need it to produce a spectrum that is similar; that is, the lower pitched main resonances, at least, should be in the 'right' ranges. The 'Violin Octet' is a case in point. The instruments cover the entire range of the orchestra in about half octave increments. They were carefully developed over several decades to have the same resonant placement as the violin in their own range. Yo Yo Ma used an 'Alto', with a 20" long body in the same range as the viola, to record the Bartok-Serly Viola concerto. The recording won a Grammy. Critics said that it didn't sound like a viola, but like a big violin. That was the point...

When I look at what you're trying to do I see a whole cascade of things that will need to be addressed to solve the 'simple' issue of making a louder guitar. If history is any guide I suspect that the it will take about 50-75 years to work out all the kinks, and it will end up someplace we don't expect.

I'm a lover of a finely tuned and responsive instrument for fingerstyle playing. I'm a singer songwriter myself and for so many years now my acoustic guitar is my intimate partner in my musical pas de deux. I understand the comfort with the devil you know but I also have a restless mind and you all know the issues associated with the status quo. Reasonable solutions, even excellent solutions have been worked out for that format. My "Evergreen Guitar" approach does open up challenges and problems to solve so stay tuned as I explore some of those avenues. Thanks for sharing. This is what I came here for: an educated conversation about the acoustic guitar.

_________________
Breanna Anderson
https://glyphstone.studio

I find it inspiring to see people experiment and measure their results and try more iterations.

Thanks very much for sharing your work and I very much hope that you'll feel welcomed here and will continue to share.

My friend Martin Brunkalla developed a system he called the Freedom Top and applied for a patent in 2006. He is now semi retired and his website is no longer up. He used carbon fiber tubes between the neck block and tail block, a tail piece and kind of a dual saddle bridge to couple the saddle to the top. The instruments sounded quite good but I think the tailpiece is a hard sell to a traditional crowd. The band Tripping Lily played Brunkalla instruments, in fact their name was a nod to the tipped lily logo Marty inlaid on the headstock. I think Marty prefers to build 5 string fiddles these days.

Kent

I could only find one image, but I think it is another different approach, having the strings suspended between the tailpiece and nut so they're right about saddle height, and then screw down a bar over them to clamp them to the bridge rather than using a break angle to transmit force.

I think the diagonal break angle approach is still my favorite, not requiring metal hardware on the bridge and not having to unclamp to change strings, or thread them through the 3-bar down-and-up style I mentioned in my first post, but the end result should be largely the same with all of them.

Here's a pic from Marty's Facebook page. I think this one has Kauri back and sides

That approach would result in no torque or down force on the top if the string anchor at the tail is the same height off the top as the saddle top. He's used far more break angle over the saddle than is necessary; reducing that would lower the mass needed for the hold-down. It would be fairly easy to reduce the bridge mass by a fair amount as well.

So suppose you take all of this to the extreme, with a veneer thickness of wood for the soundboard and the bridge that has minimal weight. What you've got is effectively a wood banjo. Would it sound like a guitar?

So it is the 3-bar down-and-up style! Excellent, that means the core claim of zero-torque in the Ned Steinberger patent is invalid due to prior art, and we can all have at it next year. Although I'm not sure about achieving it via counterbalanced diagonal saddle slots, they might still be covered for another 5 years.


Probably not with veneer thickness, but maybe with 2-3mm. My Coral Snake sounds like a guitar despite having only the bridge as a brace in the active area.

But there's no reason you have to change from your usual bracing. It just gives you the option to carve farther than usual, use more strings than usual without beefing up the bracing, or experiment with entirely different patterns with no structural requirements.

Try looking at it from the other way around: What if you remove the rear saddle on that bridge so the strings do apply a torque to the soundboard? Would there be any tonal benefit, or pointless stress on the wood?

DennisK wrote:

"Probably not with veneer thickness, but maybe with 2-3mm. "

But that's the thickness of a normal top! All of the bracing together only makes up about 30% or so of the total weight of a standard top, not counting the bridge. A lot of that is, of course, the UTB, that takes up much of the load of the neck. You need to deal with that somehow. Some folks, of course, use some sort of 'flying buttress' for that, but I'm not sure that can take the whole load. Usually they include at least some sort of UTB. A through bar to the tail block can do it. At any rate, if you're trying to cut down on the weight of the top to make it easier to drive, the top itself is the best place to start. That's the focus of both the 'sandwich' and 'lattice' top designs.

" What if you remove the rear saddle on that bridge so the strings do apply a torque to the soundboard? Would there be any tonal benefit, or pointless stress on the wood?"

As far as I can tell, nothing that happens behind the saddle should make any difference in the sound, so long as it doesn't change the weight. The forces the string puts on the saddle top as it vibrates are well defined, and those are what feed the sound. The main driver is the 'vertical' polarization of the transverse force: as the string goes 'up and down' it pulls the top along with it. The lower bout acts like a loudspeaker. The horizontal transverse force can't really move the bridge to speak of.

As the string vibrates it gets tighter twice per cycle; when it's 'up' and 'down'. This tugs the top of the bridge toward the neck and rocks it fore and aft. This force varies depending on the string, but it averages about 1/7 of the 'transverse' force. Since tops are made to be stiff in that direction, so that they won't fold too soon, this can't drive the top very much, although it's more effective when the strings are higher off the top. Also, as the bridge rocks it pulls half of the top 'up' and pushes the other half 'down', so much of what motion there is cancels out at most pitches. Any sort of setup that takes up the bridge static torque tends to negate this 'longitudinal' signal signal.

There is another longitudinal signal, the 'zip' tone, that can also drive bridge rocking. It's a pressure wave running the length of the string. It's pitch is determined by the properties of the string an it's length, and is very little affected, if any, by changes in tension. It's usually somewhere between the 7th and 8th partial of the string: on a plain steel string of the usual length somewhere around 4000 Hz. If it's pitch agrees exactly with a string partial it can cause some odd behavior. As with the twice-per-cycle tension change, it's more prominent in the signal when the strings are higher off the top, and, of course, the bridge has to be able to rock.

So; removing the rear saddle would get you back to where a Classical bridge with a tie block, or a pin bridge, is. That's the thing that cancels the torque, as you point out.

So far as I've been able to measure things, swapping bridge pins alters the sound if they change the weight of the bridge. A set of six pins can weigh as little as a couple of grams or as much as 29 grams, and maybe more for brass. A Martin style 'belly' bridge in ebony weighs about 30. Glue another bridge on the top, anybody? Anyway, I don't think there's any 'signal' transmitted through the bridge pins, although that seems to be 'common wisdom'. Leaches, anybody? I keep asking people to get a measurement of that 'pin' signal, but no takers so far.

I mean 2-3mm thick with minimal bracing would probably still sound guitar-like, wheres 0.020"/0.5mm veneer with minimal bracing would sound very weird. If you meant lattice bracing on a veneer-thickness top, I doubt it would be too much different than a 1-1.5mm lattice braced top. I don't think there's a whole lot to be gained in this direction, or any direction with 6 strings. Just a little bit lower stiffness than usual (very much lower and it will sound worse rather than better), and perhaps better longevity depending on the internal structure.


Breanna's system with a tall bridge may be able to make some use of the transverse vibration, especially if the bracing is stiffer on one side than the other to reduce cancellation. Asymmetrical bracing is worth experimenting with on regular-height bridges too. Usually it results in greater total stiffness since you can't loosen up the bass side too much without excessive deformation, so instead you have to stiffen the treble side beyond its minimum survivable value. But in this case we can do whatever we want.


I think it would depend on the friction between the string and saddle, and in the case of Marty's "excessive" break angle that would be quite strong. Probably enough to transmit the longitudinal tug.

You could also dramatically increase the bridge rocking motion by orienting the soundboard grain across the width of the body instead of in-line with the strings, but I think it would be easy to overdo it and cause problems from the string length varying as the bridge rocks forward and back.

DennisK wrote:
" I doubt it would be too much different than a 1-1.5mm lattice braced top. "

The Smallman we had in the shop had a 'soundboard' that was .8mm thick (I'm pretty sure it was redwood). That's about the same a the old US standard veneer thickness of 1/28" (.9mm). Basically, from what I've seen, Smallman's lattice takes all of the loads, and the soundboard is just there to bridge the gaps and move air. That guitar, BTW, had been sold at a heavy discount ('only' $12,000!) because somebody had punched a hole at the edge of the top, and patched it over with some CF.

A 1mm thick top would be nearly twice as stiff as his veneer, and 1.5 almost eight times as stiff with the same material. In thinning out the top like that it seems to me that the key is getting the unsupported spans between the lattice short enough to avoid 'oil canning' at low frequencies: thinner top, closer lattice.

I made asymmetric braced tops on Classical for a number of years, with the idea that the 'cross dipole' mode might be more effective at producing sound. When I got my 'anechoic closet' together I was able to check that. It turned out that the larger area vibrated with lower amplitude, so the actual output was about the same from both sides, but the sound radiation was somewhat asymmetric. This happens anyway due to the way the monopole and cross dipole modes add up in use, but with the asymmetric bracing it was more pronounced.

I use the 'Chladni' method of 'tuning' tops; sort of a 'tech' version of tap tones. Generally speaking, I've found that the tops that vibrated more symmetrically in the 'free' plate tests made better guitars, and that was harder to get with asymmetric bracing. I switched over to symmetric bracing on both Classicals with fan bracing, and steel strings, using a 'double-X' pattern. In both cases it was an upgrade. My 'take' is that the higher-order vibration modes are more efficient with a more symmetric soundboard: you get clearer trebles. And, as my voice teacher used to say: " The way to get good low notes is to work on your high notes"

"I think it would depend on the friction between the string and saddle, and in the case of Marty's "excessive" break angle that would be quite strong. Probably enough to transmit the longitudinal tug."

?

The only way you'd get the longitudinal tug past the saddle would be to have zero friction between the string and saddle top, or nearly so. I ran some experiments with a monochord that has a piezo pickup at the 'bridge' end which only responds to longitudinal forces. Any saddle in front of the bridge was sufficient to stop those signals so long as it defined the vibrating length of the string. That's the saddle's 'job', so if it's working there should not be any tension or 'zip' signal at the pins, so far as I can tell. It would be nice to get some independent measurements of that.

"You could also dramatically increase the bridge rocking motion by orienting the soundboard grain across the width of the body instead of in-line with the strings, but I think it would be easy to overdo it and cause problems from the string length varying as the bridge rocks forward and back."

I did an experiment looking at the way the sound changed (or didn't) with changes in the break angle and string height off the top. Changing the break angle made no measurable or discernable difference in the sound so long as the string height was not changed. Changing the string height off the top while keeping the break angle the same produced a sound with a slightly different timbre, but no change in output power or duration. Raising the strings from 11mm off the top to 18mm (don't try this at home!) produced a sound with relatively more energy in the 2nd and 4th partials, and also at the 'zip' pitch of the string. Both of these were tied to the increased leverage. Again, there was no difference in either the amplitude or duration of a 'standard' pluck (force controlled within 2% +/-, average over six plucks per case).

So, if you want more 'rocking' motion the easy way is to raise the string height off the top, assuming the top and bridge can react to that. Canceling out the torque negates that, so far as I can tell, no matter how you do it.

I think what Dennis means is that there will be too much friction at the saddle (as you point out Alan) for the tension change signal to get past the saddle, so it is reacted by bridge rotation and the stiffness of the soundboard, thus driving (to maybe a lesser extent) the long dipole mode. I'm sure Dennis will correct me if I got that wrong!

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

http://www.goreguitars.com.au

I prefer the tone from traditionally braced guitar tops that are under the full stress from string tension.