Official Luthiers Forum!

The only basis I have for speculating the lesser effect on nylon strings is
the difference in the modulus of elasticity. At a point when I was
considering the effect of lateral waves vs longitudinal waves in
considering saddles being perpendicular to the top vs tilted back by a few
degrees I did a little number crunching. I recall the resulting longitudinal
waves in nylon strings being a small percentage of what they were in steel
strings, since they increased much less in tension with the same amount
of deflection from it's static position.

Of course in practice a nylon string is deflected much farther than a steel
string when playing, so I'm not sure how my speculation there adds up.
It's been long enough since I gave it much thought that I don't recall what
factors were considered. In any case, though I doubt I'll be delving in to
advanced research on this topic, it's an interesting seed of thought to
tuck away somewhere for future reasoning and problem solving.

_________________
Eschew obfuscation, espouse elucidation.

Neither Jean Larrivee nor Richard Hoover have noticed any change in the sound of their fairly traditionally made guitars in the switch to a tilt-back saddle. I'm not sure it would actually make a difference in the vectors acting on and through a bridge from strings to top, but it sure makes a difference in the strength of the bridge and in how well an undersaddle pickup works in the bridge.

[QUOTE=Rick Turner]Draw a line about 30 inches long on a piece of paper. Add a saddle witness point near one end of the line. Add a nut witness point at the other end. Add in a gently curved line from a bit below the nut witness point that represents the line of the fret tops with action you like. Those are your constants in the design process.   Now put in the back of the neck as you like that thickness...that's a variable. Now draw anything you want to and figure out how to make it work and imagine the forces working on what you've drawn. Imagine the vibrations from the strings starting to shake and twist the top.   Now rotate your view point 90 degrees and work on the guitar shape profile. You've started your design process at the very core of what's happening...the string vibration. Think about separating stress and strain from vibration as much as possible.   See if this doesn't make you rethink how to design a guitar.
[/QUOTE]

Thanks, Rick.  I like that description and I do think about guitar design that way.  I've experimented to some extent with all those factors and I have some working theories on them.  The only thing I'm confused about is when people talk specifically about string height over the top as if it's independent of all those things.  I keep thinking I'm missing something.  I see bridge/saddle height and the relative angle of the strings to the plane of the top as the things that determine string height and the relevent issues.  What I'm curious about is if you and others are just calling the conglomeration of those things "string height" out of convenience or if it is truly somehow an independent issue.

I'm also not sure I get... "Think about separating stress and strain from vibration as much as possible." I assume you're talking about the string tension as the stress and stain?  What do you mean by separating it from vibration?  The only variable I can see with a traditional bridge is torque.  Is that what you're getting at?

_________________
http://www.chassonguitars.com

Whew! (mopping brow here)... I think I get it.

Thanks, guys!

John

Sam,

As a fellow amateur, here are my thoughts.

I spent many years woodworking before I began using handplanes. Handplanes were so difficult. There's all that sharpening, fettling, and so on. So I used a power jointer exclusively. I often had to make some pretty intricate jigs to do things that really weren't that well-suited to power jointers, and some of those jobs weren't safe at all. I did them with a power jointer not because that was the most efficient or effective method, but because I was afraid of hand planes.

Eventually, I decided to quit futzing around and learned how to use handplanes. It took a while, but I got pretty good at sharpening, and it takes me no time to get a plane adjusted properly. I still have and use a nice 12" power jointer, but I use handplanes a lot more. They're usually quicker and easier, and often the only appropriate tool for the job. Now, when I choose the power jointer for a job, it's because it's really the better tool for the job, not because I don't know how to do the alternatives. Now, I make choices because I am confident that I know how to do it either way, not because I am trying to find a way around my shortcomings.

It sounds like you are afraid of resetting a dovetail joint. I can understand why you might be anxious now, if you haven't done them before. But why should you live a life of guitar-building afraid of dovetail joints because you might have to reset them? I'm not saying you need to use dovetail joints. I am saying that you should be able to. I am saying that you should not have to make your decisions based on a fear that you don't know how to do something. This is a great opportunity to learn. Steam the neck off that Avalon (if it is a dovetail), and learn something.

Perhaps afterward you will still want to invent some complicated adjustable neck system. But if you do so, it will be because you honestly find it better, stronger and faster, not because you are afraid of the alternatives.

Free yourself of the idea that the plane of the fret tops has to be locked into the plane of the top. (I'm purposely oversimplifying these lines...I'm aware that there is relief in the fingerboard and a dome in the top)   Those lines can be independent and therefore the space between the string and the top from the body to the bridge can be an acute taper with the narrowest part being either at the neck end (normal) OR at the bridge end (think Humphrey Millenium or my guitars).   The space between the underside of the fingerboard and the top can either be filled (with just glue as normal OR a block as per Millenium) or it can be open space (Howe Orme, McPherson, my own stuff). You do not have to build guitars with one thing necessarily determining another when you free up the geometry like this.

Stress and strain...Have you ever seen a guitar collapsing in on itself with the fingerboard and the top under it slipping like a tectonic plate when the top cracks on either side of the fingerboard?   It happens when the braces let loose in the upper bout. What I'm talking about is the fact that on 99.9999% of the guitars I've seen, the pressure on the upper bout from the string pull on the neck is resisted by the top itself and the braces under there.   Why? That's what I asked, and then I did something about it that deals with it all and relieves the top of the responsibility of supporting that load.   Of course there's still the tension and torque load on the bridge, and unlike Jeff Babicz or Ned Steinberger I think that this stress may just be essential to how we want flattop guitars to sound.

Rick, I think you're missing me here and maybe I'm missing you a bit but I think we're getting closer.  One more try perhaps?

I totally get what you're saying about the plane of the frets being independent of the plane of the top.  Like I was saying earlier, I messed with that some by adding a wedge under the f/b extension.  I'm just trying to make sure I understand exactly what you want to effect by changing the "string height".  Is it the angle of the strings in relation to the plane of the top?  In order to effect bridge torque?

Thanks for your patience and sorry if this is getting off topic from the original post but it's something I've long been curious about.

_________________
http://www.chassonguitars.com

Oh, and freeing the top of compression from the neck makes sense too.  Again, just wasn't clear before if it was that or the bridge you were talking about.

_________________
http://www.chassonguitars.com

Kent, all I am saying is that once you free the issue of action from being dependent on saddle height, you can start to experiment to see what height of the strings off the face works best for you. This makes that distance changeable independent of string height off the fingerboard.

I do not have an answer re. what bridge saddle height is optimal. Obviously the height affects the torque load on the top, so too high is going to do quite the load. I don't know if there's a too low. But with this approach to building, it would be possible to experiment within a reasonable range to see how a guitar responds acoustically to different saddle heights while keeping the playability at wherever you like it.

From my point of view, it's all about opening up choices for us to work with. It's about not being locked in by the physical constraints of traditional design while respecting that there is much to learn about why people like the sound of those traditional instruments. At the same time, I'm deliberately going for a different sound, but I'd like to know exactly what makes that sound different...if that is possible. It's like saying that I'd like to be able to make a great 1938 D-28 sounding guitar...and then not do that.

Got it.  Thanks.

_________________
http://www.chassonguitars.com

I had been considering the idea of a stress free top for quite some time. After watching the Les Paul special on PBS recently, it re-ignited this idea. His passion was making a guitar out of a 4x4, then attching wings to it that were hollow to allow for resonance.

I cant help but wonder if there isnt a better way to make an acoustic guitar using some of this same type of thinking. By extending the tension load completely through the back of the guitar by way of some internal structure and then cantilevering the bridge backwards toward the normal bridge placement you should relieve the top from all of the normal bracing and from the upper bout's non-resonant bracing too. I havent decided how to attach the  bridge to the top at this point, but I'm sure with enough cogetating over it I could decide how it is done.

One guitar style that is almost the same thing is the double top guitar or the semi-doubletop. If you think about this, the torque of the bridge and bridgeplate is relieved by CF material sandwitched between the tops. This also spreads out the lateral tension placed there by the strings in such a way that it is equally spread throughout the top of the guitar. Internal bracing is reduced and resonance is increased. At least that is the theory. Im my model, the top would be solely the resonance of the wood only and would not rely on CF implanted in the top to give strength. By relieving all torque and tension in the wood, resonance should be at it's maximum potential.

The down side to this line of thinking is that spruce, or any top wood for that matter, may not resonate at an acceptable level if it is not under some sort of tension. This is where experimentation and calculations come into play.

_________________
Reguards,

Ken H

Don't throw the baby out with the bath water...

Okay Rick, I was backward on my thoughts and recollections earlier, but I
did find what I was looking for. At the shop this evening I browsed
Rossing's "Science of Sound" and Fletcher/Rossing "The Physics of Musical
Instruments" and found what I was referring to earlier.

The longitudinal waves I was concerned with earlier would have no
effect because their speed (therefore their frequency) does not change
with tension. In other words, they are not in any way related to the
frequency of a string, only it's material and length and die out fairly
quickly with little influence.

The transverse wave is what carries the entire pitch, which exerts a
force perpendicular to the string. The increase in tension however will
exert a force parallel to the string, which seems to be the area of focus in
the projects you mentioned earlier. One wave, two seperate resulting
forces. Distinguishing a wave (vector) from a force (scalar) is important in
thinking about it, for me at least. The longitudinal, or parallel force is not
a wave, but rather a force pulse.

Here are some of the interesting parts.

The Physics of Musical Instruments, chapter 9.8

"In Section 9.3, we considered the parallel and perpendicular
forces a string exerts on the bridge when it is plucked. The parallel
force at twice the fundamental string frequency was found to be small
at ordinary playing amplitudes but increases quadratically, so it can
become a factor in loud playing. This force exerts a torque whose
magnitude depends upon the bridge height and which could be a factor if
it occurred at a frequency near a resonance of the (0,1)(0,2) or similar
mode having a node near the bridge."

So there is at least a brief mention of the longitudinal/parallel force
pulses occuring at twice the frequency of the fundamental frequency. I
didn't have time to read much further, but it seems to be at least briefly
established in academia.

Back in section 9.3 we have,

"For a typical high-E nylon string ....(formulas containing
characters not available here)... the maximum transverse force is roughly
16 times greater than the maximum increase in longitudinal force; more
importantly the amplitude of the transverse force pulses is about 40
times greater than the longitudinal pulses and they couple more
efficiently to the top plate. However the longitudinal force pulses are
proportional to d2 compared to d for the
transverse pulses, so the difference diminishes with increasing amplitute.

The elastic (Young's) modulus for steel is about 40 times greater than
for nylon, and static string tensions are about 50% greater, so the
longitudinal and transverse force amplitudes will be more nearly
equal"

This is where my thoughts of the lesser influence on nylon must have
come from. I knew I read it somewhere.

Sorry to stray so far from the neck joint topic everbody, but at least it's
pretty interesting stuff anyway, eh?

_________________
Eschew obfuscation, espouse elucidation.

David, very interesting stuff there, and you got it right about what I was up to. This was not picking up longitudinal waves.

BTW, to hear them in a material like a steel rod, hold the rod at some node and whack the end of the rod. You'll hear a "ping"...that's the longitudinal wave coursing back and forth.   Do it with a mild steel or iron rod aimed north, and you'll turn it into a magnet...   Weird science...