Saddle design, more thoughts for discussion...

[Louie Atienza]

Al,
Many thanks for your response.

Mario's observations are reasonable: notching the bridge rather than the pin would result in a sharper break angle, shifting
the centroid and obtain generally positive results.

It all depends on what kind of instrument you're trying to build of course.

Again, I'm looking forward to your paper, and grateful as hell to you for sharing a lot of hard work.

Tom

Tom thanks for joining in... I do feel notching the bridge can be more beneficial as far as effecting more break angle, though of course this is relative to pin placement. One could set the pins farther back and then notch the bridge, and that wouldn't have an effective change on the break angle!

I have some theories on the pins: Since the bridge is slotted rather than the pin, it would stand to reason that there's more mass or density at the string's anchor point since the bridge pin material could be more dense than the bridge material. Maybe this wouldn't have as much an effect with an ebony bridge/ebony pins, but maybe with a rosewood bridge/bone pins.

It's pretty obvious that any change will have some type of effect. The debate I believe is whether it's discernable, and if so, does it produce a 'musical' sound.

I do find it hilarious that some luthiers claim to take off .001" of the top thickness here, shave .001" of cross bracing there, use wood with a Young's modulus that's 1% higher than comparable wood... and claim that these microscopic differences havev a huge effect. Yet if a large design change is made, extensive testing is needed to see what effect it has! Would Segovia tell Mauser to remove a sliver of wood on the fan bracing?

Or better yet, you pick two identical guitars up at the local music store. One plays and sounds like a dream. The other is a dud. It's not because they had a bad luthier at the factory. Heck they probably don't have luthiers at the factory! The woods, amount of glue and finish, etc., are all in a sense randomly done or chosen, no matter what computer cut them. Great guitars have been made with woods with crappy tap-tone, and vise versa. How does one test that? You couldn't do this by taking one guitar, then configuring it 10 ways. You'd have to do it with 1000 guitars configured 10 different ways, and evevn then, you'd always find one exceptional guitar and one dud.

So one could conclude through reason alone that these exercises are pointless, and we can go about building guitars the way we do, since these changes are insignificant [musically], or one could say that since the changes are insignificant [musically] why not do things a bit different? The truth is, it's human nature for the curious ones, myself included, to look for answers to "why" and "how" instead of saying, plainly, "Hey, I like the way this sounds!" or "I can get a vibe playing this one!" or "This guitar sucks!" (usually spoken after playing one of mine!)

In fact we do know that the break angle at the bridge can have a significant impact on how a gutar sounds. We see this regularly when a guitar comes in for a neck reset. The bridge is planed or shaved down so that the saddle can be lowered. The break angle is reduced (if the pin holes are not slotted). The customer complains that the guitar has lost some of it's oomph or volume. In fact one type of quick repair for this sort of problem, if the customer doesn't want to install a new bridge, is exactly to slot the pin holes, to create more break angle. But then when the neck moves too much, the angle past the saddle decreases, and the bridge needs to be replaced since it's too thin at that point to hold the saddle. The other case is when the soundboard starts warping behind the bridge. I'm sure this happens to many guitars built lightly to some extent. But when it happens to a significant amount, the break angle decreases at the saddle, and affects the sound.

[Alan Carruth]

Louie Atienza wrote:
"I do find it hilarious that some luthiers claim to take off .001" of the top thickness here, shave .001" of cross bracing there, use wood with a Young's modulus that's 1% higher than comparable wood... and claim that these microscopic differences havev a huge effect. Yet if a large design change is made, extensive testing is needed to see what effect it has! Would Segovia tell Mauser to remove a sliver of wood on the fan bracing?"

Did you mean Hauser?

I think it's as much a mistake to say small changes _can't_ have an effect as to say that they always _do_, or that a given change is always beneficial. When something's wrong, you have to fix what's wrong, whatever that might be, whether the needed change is big or small. You would not say that having a fret that's just a little too tall is not a problem, or that dressing it down would not be the right fix.

"Or better yet, you pick two identical guitars up at the local music store. One plays and sounds like a dream. The other is a dud. It's not because they had a bad luthier at the factory. "

Right: it's because wood is a laughably variable material to be using in mass production, where you have to make every unit the same. Guitar factories have to work to the average, with the provision that they have to assume that the weakest top is going to end up with the weakest braces, so they need to over build a bit from what the 'average' would dictate. Given good designs (and the standard designs are pretty good) they'll end up making acceptable instruments most of the time, with occasional duds and rare great ones. There's no mystery to it.

"In fact we do know that the break angle at the bridge can have a significant impact on how a gutar sounds."

As a humorist said, it's not the stuff you don't know that gets you into trouble, it's the stuff you do know that isn't true...

"We see this regularly when a guitar comes in for a neck reset. The bridge is planed or shaved down so that the saddle can be lowered. The break angle is reduced (if the pin holes are not slotted). The customer complains that the guitar has lost some of it's oomph or volume. In fact one type of quick repair for this sort of problem, if the customer doesn't want to install a new bridge, is exactly to slot the pin holes, to create more break angle."

Nobody is going to argue that you you don't need at least some break angle: the string has to stay in contact with the top of the saddle, as I've said. You could get into a discussion about how much is 'enough', and that's certainly legitimate. The whole 'centroid' thing shows that changing the break angle does alter the static forces on the top (which surprised me), and that certainly could change the sound. I've seen some evidence of that. I think it's a stretch to say that any particular change is _always_ going to be beneficial: I like ketchup on French fries, sometimes put it on eggs, and don't particularly care for it on ice cream. It does seem to me that once you get past 'enough' break angle, more is not automatically better. That's like the idiot with the pepper shaker: a little bit is good, so a lot must be terrific.

At any rate, as I believe I've pointed out in this thread, some very careful listening tests showed that people could not reliably hear a change in break angle, but could hear a difference in string height off the top when they did not know what they were listening to. It's all too common for people to hear what they expect to hear, or want to hear, and this colors our experience any time we do a repair. That's why careful tests have to be done, and _especially_ in those cases where we 'know' what to expect. As Feynman said, the easiest person for you to fool is yourself.

Alan Carruth / Luthier

[Louie Atienza]

Did you mean Hauser?

I think it's as much a mistake to say small changes _can't_ have an effect as to say that they always _do_, or that a given change is always beneficial. When something's wrong, you have to fix what's wrong, whatever that might be, whether the needed change is big or small. You would not say that having a fret that's just a little too tall is not a problem, or that dressing it down would not be the right fix.

Yes1 I type faster than my mind thinks! Anyways this is my point exactly. Then again who's to say what's right or wrong? It's more a question I believe as to what's acceptable and not. Obviusly, a microscopic change will at the very least have a microscopic effect, whether the mass or size is reduced. That can be measured. There are other factors that are impossible to measure. For example, I've owned guitars with crappy frets that I thought sounded great, and played guitars with impreccable fretjobs that sounded horrible. I know what I'd rather play.

Right: it's because wood is a laughably variable material to be using in mass production, where you have to make every unit the same. Guitar factories have to work to the average, with the provision that they have to assume that the weakest top is going to end up with the weakest braces, so they need to over build a bit from what the 'average' would dictate. Given good designs (and the standard designs are pretty good) they'll end up making acceptable instruments most of the time, with occasional duds and rare great ones. There's no mystery to it.

I completely agree. As an independent luthier, you have the "luxury" if you will of custom-tailoring the measurements of your materials based on a combination of some formula, some testing, and some experience. There are certain "rules" builders follow to accomodate for the variations in wood strength and mass. The thing is, what I might consider a "dud" you might consider an exceptional piece, and vice versa. Heck, even the fact I mention a guitar a dud to you, could temper your judgment.

At any rate, as I believe I've pointed out in this thread, some very careful listening tests showed that people could not reliably hear a change in break angle, but could hear a difference in string height off the top when they did not know what they were listening to. It's all too common for people to hear what they expect to hear, or want to hear, and this colors our experience any time we do a repair. That's why careful tests have to be done, and _especially_ in those cases where we 'know' what to expect. As Feynman said, the easiest person for you to fool is yourself.

This is I believe where some confusion lies. The break angle can be changed by a few factors: changing the saddle height relative to the bridge, slotting the pin holes, moving the pinholes relative to the saddle, or decreasing the bridge height relative to the saddle. These can all be done independant of the string height from the top. But to change the string height off the soundboard, do you keep the same bridge and just change the saddle, or do you make a thicker bridge and keep the same saddle? One changes the break angle more significantly than the other. If only the saddle is changed, then the pin hole must be movevd proportionately to keep the same string angle.

Maybe there is an optimal string angle or height based on the soundboard thickness or stiffness. I might suppose that a stiffer soundboard could withstand having the strings higher off it, for example.

[Alan Carruth]

Louis Atienza wrote:
"The break angle can be changed by a few factors: [snip]"

I did my experiments with a classical guitar, so the break angle was changed for a given saddle height by tying the strings differently on the tieblock. I did tests with the same saddle height and different break angles, and then put in a really tall saddle to alter the string height and raise the break angle from the 'low' value of about six degrees to the 'normal' one of about 25 degrees. Since this could not be doe quickly and silently, I used recordings of 'standard plucks' for the listening tests. I suppose you could do the same test with a steel string. The one I did was enough work!

Alan Carruth / Luthier

[Louie Atienza]

Louis Atienza wrote:
"The break angle can be changed by a few factors: [snip]"

I did my experiments with a classical guitar, so the break angle was changed for a given saddle height by tying the strings differently on the tieblock. I did tests with the same saddle height and different break angles, and then put in a really tall saddle to alter the string height and raise the break angle from the 'low' value of about six degrees to the 'normal' one of about 25 degrees. Since this could not be doe quickly and silently, I used recordings of 'standard plucks' for the listening tests. I suppose you could do the same test with a steel string. The one I did was enough work!

Alan Carruth / Luthier

Interesting... I guess you could affect the angle even further by tying the strings in reverse so they come out the back...

Thinking about this now, I wonder if I would achievev similar results to yours, using a steel-string as opposed to a classical. I wonder if there are optimal angles/string heights for different bracings (fan, ladder, x, parallel)?

Hopefully my CNC will do most the work. But since a steel string can be both plucked, picked, strummed, (and slapped?!) maybe tests should be done for those types of playing styles?

[Alan Carruth]

Louis Atienza wrote:
"Interesting... I guess you could affect the angle even further by tying the strings in reverse so they come out the back..."

That's what I did. I converted the bridge to an 18-hole tieblock, which allowed me to tie the strings on in two ways: the 'normal' one, where the string comes off the saddle and goes through the center hole in the front of the tieblock, and the 'low break' tie, where the string goes over the tieblock and into the center hole in the back. This gave break angles of about 25 degrees and about 6 degrees, respectively, with the strings 11mm off the top. Then I put in a tall saddle, which got the break angle back to about 25 degrees, but with the strings 18mm off the top(don't try this at home!).

Alan Carruth / Luthier