Floating archtop bridge on flattops

[Fernando Esteves]

Hello there!
I was thinking here, as the traditional glued bridge in flattops has a lot of trouble, is there any reason why people don't make that style of bridges of archtops in acoustic?

Has anyone experience in one guitar like this?
Cheers

[Freeman Keller]

Fernando, you have to remember that flat tops with pinned or tied bridges and archtops with floating bridges make their sounds in quite different ways. The pinned bridge "rocks" the top around the bridge axis pulling the lower bout up and pushing the area towards the sound hole down where a floating bridge more or less pushes he top straight up and down. Archtop tops are usually thinned as they approach the rim to allow them to move more freely (there is a discussion about this currently running at the archtop subforum).

A floating bridge on a typical archtop has upwards of 50 pounds of down force from the strings breaking over it to the tail piece, archtop bracing (X or parallel) along with the thickness of the top and the location and size off f-holes is all designed to withstand that and still let the top move freely. Flat top bracing (fan, X, lattice...) is designed to counter the rotational torque. In addition, floating bridges are quite often considerably taller than pinned bridges.

Bottom line, the entire instrument has different forces acting on it and you would probably have to make a lot of redesign to make it work. Not that it couldn't be done.

I'll add one more thought to that. I was asked to build "an acoustic guitar that looks like an ES175" which as you know is a laminated top hollow body electric guitar. I built something with the same shape as a 175 with a spruce top that was not carved but braced at 16 foot radius (so a bit more than a normal flat top but not nearly as much as an archtop). I gave it a simple X brace, put a bridge plate in in case we decided to pin the bridge after all, here is the inside

And the finished guitar

It works, sounding more like an archtop than a flat top and my friend is happy with it.

[Fernando Esteves]

Very cool!
If you ever get him to record a clip, post here! Sounds interesting a mix between the two worlds!

[Freeman Keller]

Very cool!
If you ever get him to record a clip, post here! Sounds interesting a mix between the two worlds!

I haven't seen the guitar for quite a few years, if it ever comes back to the shop I will not only take some sound clips, I'll run the spectrum analyzing software and see if I can see anything significant. What I remember is that it was pretty bright, but had the good note separation and clarity. So yeah, like an archtop but not quite...

I did do one other thing with this guitar - because I was uncertain of how it might sound acoustically I did install a set of K&K piezo sensors on that bridge plate before I closed the box (there wouldn't have been any access after). The idea is that if it totally sounded like crap (a real possibility) he could plug it in and fiddle with effects to make it sound better. As far as I know it has never been plugged in.

[Alan Carruth]

Freeman Keller wrote:
"...flat tops with pinned or tied bridges and archtops with floating bridges make their sounds in quite different ways. The pinned bridge "rocks" the top around the bridge axis pulling the lower bout up and pushing the area towards the sound hole down where a floating bridge more or less pushes the top straight up and down."

Not quite. It's true that the tailpiece takes up the tension change in the string, so that the arch top bridge has no force rocking it fore and aft along the top. However, that turns out to have only a minimal impact on the actual sound.

First, you need to understand the forces that the string exerts on the top of the saddle as it vibrates. I've spent altogether too much time measuring this over the past several years, so all of what I say in based on actual data, which I've posted on my web site as a 'pdf entitled 'String Theory' (I couldn't resist..). You can also look all of this up (with eqations) in the standard reference by Fletcher and Rossing: "The Physics of Musical Instruments".

If you push the string down toward the top it forms a triangle with a 'kink' at the plucking point an two straight lines out from there. Because the string makes a bit of an angle downward at the bridge it's now pulling the top down. When you release the string the kink the was formed by the pick or your finger runs down the string in both directions toward the nut and saddle. In between the two kinks the string makes a straight line, and outside of them it's still straight out to the bridge or nut and at the same angle as it was before you released the string. The change in momentum of the string moving away from the top pulls the ends down, and the bridge doesn't 'know' you released the string until the first kink reaches it. Then the pull switches suddenly from 'down' to 'up'. The kink at the other end does the same a little while later (assuming you plucked nearer the bridge) . The reflected kinks run back toward each other, eventually (1/2 cycle later) meeting at a spot that's as far from the nut at the initial pluck was from the bridge, with the triangle now pointing 'up' instead of 'down', forming an inverted mirror image of the original string just before the pluck. The whole process repeats for the next half-cycle of the string until the kinks meet again at the initial pluck point. The force on the saddle top is this a 'square wave' with a duty cycle that depends on where you plucked the string. If you plucked 1/5 of the way up the string it will be 'down' 1/5 of the time and 'up' 4/5. The spectrum of the sound of the string will contain energy at all partials except the 5th and multiples of that (10th, 15thj and so on): at those frequencies the string would have to have a stationary 'node' at the 1/5 point, and since you made it move there with the pick it can't.

Of course, if you pulled the string sideways to pluck it, all of this would be happening at a 90 degree angle, but otherwise the forces would be the same. Even if you move the pick sideways it tends to drive the string at least partially in a 'vertical' direction relative to the soundboard; it acts as a cam. It's actually pretty hard to get 'pure' horizontal or vertical string motion with a pick.

When you push the string aside or down the tension rises. Once you release it the tension starts to drop as the string gets shorter, and it falls until the first kink reaches the end (usually the saddle). Once the first kink reflects from the saddle the two kinks are running the same direction, and the string stays the same length until the second kink reaches the nut. so the tension stays the same. Once the second reflection happens and the kinks are running toward each other the string is getting longer, and the tension rises until the meet (at the 1/2 cycle point), and the whole dance begins again. The tension signal pulling the top of the saddle (or tailpiece) toward the nut is thus a gapped sawtooth wave, with two peaks per cycle, and frequency components doubling all the ones in the square wave.

Mark that. If the tension change signal was the main driver on an archtop guitar the strings would all sound an octave higher than they do on a flat top. Do they?

Measuring the forces involved helps make sense of this. The 'transverse' force, the one pulling the string (and top) up and down when the pluck is perpendicular to to the plane of the top averages about seven times as much as the tension change force, varying from string to string. F&R give the equation for this, and it's confirmed by my measurements. On a flat top guitar it's a lot easier to push the top up and down than it is to rock the bridge forward. We build tops to resist this bridge torque, so a given force produces less 'rocking' motion that it would 'vertical' motion. In addition, when the bridge rocks forward it's pulling half the lower bout 'up' while it pushes the other half 'down', so a lot of the motion simply cancels out. The vertical motion produced by the transverse string force moves the whole lower bout the same way (and a lot further, remember), helping to push air out of the hole as well a making sound off the top. The combination of a:
1) much larger force,
2) applied to a much less stiff structure,
3) that moves a larger area,
just makes more sound.

Plucking a string sideways, so that there is no 'vertical' force pushing the bridge up and down on a flat top guitar laves only the tension change signal to drive the top. The sound output is much less: initially about 20 dB less, which is 1/100th of the power.

Using a taller saddle on a flat top guitar gives the 'tension' signal more leverage, so it can produce more sound. I measured this in another set of experiments using a guitar. With the string height moved up from 11mm to 18mm off the top (don't try this at home!) there was more energy in the even-order 'harmonics', the pitches that make up the tension change signal, and people could hear the difference. . However, there was no more sound produced (maximum amplitude and sustain were the same). Changing the break angle without changing the string height off the top made no audible or measurable difference in the sound.

Plucking a string exactly in the center in theory produces a 'transverse' signal with no even order 'harmonics'. That is, the low A string would have energy only at 110, 330, 550, and so on. The 'tension' signal, being doubled, would have energy at 220, 440, 660, etc. Plucking a flat top that way gives output on all of the 'harmonics', since both the 'transverse' and frequency-doubled 'tension' signals drive the top. On an archtop you don't get the even-order 'harmonics' of the tension signal, since the tailpiece takes up the tension change.

So much for that.

[Alan Carruth]

The main reason you don't see tailpieces used much on flat tops is that it's harder to get enough break angle over the saddle. You need to have something on the order of 6 degrees (more or less) to keep the strings from 'hopping' off the saddle top when they're plucked 'vertically' with respect to the plane of the top. That's no problem. When they're plucked 'horizontally' you need something like 12-15 degrees to keep them from sliding or rolling across the saddle top. Notching the saddle can meet that need, to a point. A high break angle can be harder to get. One solution is to use a tie block, like on a Classical, and pass the strings through the holes on their way to the tailpiece. This puts torque on the bridge, so it still has to be glued down, but it does take up most of the shear load from string tension, so the bridge is less likely to peel up. You can 'crank' the top as they do on mandolins, or simply go to an arched top.

[Fernando Esteves]

Interesting
I was thinking in some kind of tailpiece alike the archtops

[Beate Ritzert]

Maybe there is one really simple reason why floating bridges are rare on flattops:
namely that there have been times they were not rare, but actually pretty common in ultra cheap but robust guitars. Here in Germany called "Wandergitarre". I remember instruments which sounded like crap - like what You would expect from a guitar with a relatively small body, a thick top of plywood, not even spruce, and massive bracing. Here one of the better models - but You get the point: compact size, robustness in order to be able to carry it in a simple back made of some cloth on Your back:



All "better" guitars had fixed bridges.

BTW: recently i had to ask a few questions regarding a project falling in that range: i converted a cheap acoustic fretless bass into a fretted one with a floating bridge. Here its current state (playable, touchup of the finish lacking):

As Alan mentioned above: notching the bridge was necessary to hold the strings in place. I also hat to plane down the fixed former bridge a bit and to round the string contacts in order not to damage the strings.

The sound did not change significantly - of course You'll hear the difference between fretless and fretted tone and the effect of the shorter scale, and the effect of the additional sound hole in the rim -- i removed the preamp box which obviosly has become useless. The bass is more or less as loud as it would have been with the fixed bridge alone (and unfortunately it is slightly overbraced).

[Alan Carruth]

Back when I started making arch top guitars I believed (erroneously) that more break angle = more down bearing = more sound. To get more down bearing I used 'hook' tailpieces. These were wood roughly in the shape of an 'L'. The strings hooked onto the top of the stem, and the lower end of the foot went down the side of the guitar, providing a pivot point well below the edge. The line of string pull goes from the top of the bridge to the pivot point, so the string in effect went through the top somewhere between the bridge and the lower end. As you can imagine there was considerable stress in the 'heel' of the L, and it was a challenge to make a wood tailpiece that would hold up for any length of time. Of course, if you didn't feel the need to use only wood in the tailpiece it would be fairly easy to implement such a thing on a flat top.

There are banjo tail pieces with adjustable arms that push each string down individually at a point near the bridge. Something like that could be a good project for a machinist you know.

I stopped using those hook tail pieces after some experiments indicated that too much down bearing, at least on an archtop, can actually kill the sound. Since then I've pretty much tried to minimize the break angle on my guitars, with good results so far.

[Kurt Temreto]

Throughout history, many guitar concepts have already answered your question, Fernando. Consider, for example, the Selmer-Maccaferri.
I am currently in an experiment to build a “dome-top” guitar in a 7 foot radius dish, along the lines of Andrew White, but with a different bracing pattern (based on the Golden ratio)..

https://www.andrewwhiteguitars.com/gree ... top-build/