MIMF

The piece I'm referring to mentioned the shape (profile), cradling the string all the way down to the hole. I don't think he talked about individual saddle pieces, one per string. Gilbert has been doing that for a while and I've done a few guitars that way as well. Since much of this discussion was about the tiny bit of open string somehow robbing energy from the output, it seemed germane to the subject.

I'm familiar with the article, and the diagrams. Somogyi still leaves some string unsupported, albeit very little. It is cradled lengthwise , but not cradled side-to-side. Gilbert's saddles are almost minimalist. My theory is not just about saving every bit of open string energy "leak" at the saddle, but actually also "excite" the bridge in a way that it normally wouldn't. I'm not saying it's necessarily beter, just different in a way I like and can hear. This was actually just part of a total system, where I totally reinforce and immobilize as much as I can the area around the soundhole (I know Somogyi talks about this too) so much as to integrate it with the transverse brace and neck block, which allowed be the use of minimal bracing everywhere else.

I took the week long course from Somogyi in 2009 and saw several of his guitars up close. His saddles are about 1/4" wide and he cuts a oversized slot for each string similar to a nut slot. I tried slotting a bone saddle for one of my clients that played so hard that he was breaking strings at the saddle once or twice during every performance. I swear that this slotted saddle robbed the guitar of significant tone and volume. It just killed it. So we went back to a traditional saddle. I went along to one of his performances and sat in the front row to see what was going on. I could see the G string moving across the saddle at least 1/2" before the string broke. I finally told him that the only solution was to play easier.

Louie, one thing that I can see in the photo of your guitar is that the saddles appear to be quite high. The effect of this may be masking any tonal benefits that you are attributing to the unique design.

Barry,

I left them high to get the guitar strung up and playable for completing the challenge, with a lot a room to adjust once the strings settled in. I wasn't sure exactly how the new bracing pattern was going to work, especially with the heavier gauge bass strings. I have since lowered the string action at the bridge. I believe it's starting to come to it's own from settling in a bit, and maybe it sounds better since the action is ower, making it easier to play. It still is about 1/8"-13/16" higher than a Martin saddle intentionally

My thinking on your situation is that the high E became coupled to the entire saddle, so the "energy" so to speak of that string either gets absorbed by the saddle or sent back to the string. With my design, while the entire non-speaking length of the string at the bridge is restrained, since the saddle is indivual, and narrow, it should allow the top to respond more, especially side to side, since the taller saddles can generate more side to side torque than a flatter wider one (since torque is merely the force x distance). I tried to promote this more my making the bracing more parallel, yet restricting as much as possible torquing effects toward the sounhole, though I left the lower bout as free as possible. I actually over-baked it a little as despite the thinness of hte bracing at the lower bout, the top has not deformed at all, though time will tell...

You could think of the saddle design as two rectangular buildings, one taller and narrow and one shorter and wider. The strings could be a huge gust of wind. The wind would blow over the short stout building with no effect, though the effect could be seen in the tall buiding. When I was in the sign business we had clients at the WTC. Some cubicles had pendulums ahnging so they could see how much the tower swayed!

I wonder if you could even make a separate E saddle, and shape it such that there's just enough to keep it in the slot at it's sides, and elevate it in the middle similar to mine, and see if you get that string more responsive. Line I mentioned before, I tried as hard as I could to break a string, with a 2mm Dunlop Stubby, and a copper IcePix, and I'm ham-fisted to begin with... couldn't do it...

Thanks again for your (and everone's) comments and input...

Louie,
You make a number of assertions and suppositions that may or not be plausible. I wonder if it's possible to set up a test that would allow you (us) to prove or disprove them.

I've built bridges that have individual slide-able saddles and used six, three, two and one piece saddles on the same bridge. I've not been able to ascertain any difference in sound between the variants, to the point where I didn't think it worth doing a spectral analysis. It would not be difficult to do so, but if I can't hear a difference there's not much motivation trying to find one.

Barry, I'm confused by your story. It sounds like you are saying that a string sliding around on the top of the saddle provides better tone/volume than one where the string is anchored at the saddle. I wonder how this could be given that so much energy must be going into sliding the string around (1/2" ?) rather than directing it into the top ? Not to mention the odd intonation issues that must result. What you describe sounds similar to problems encountered when a string contact point is not right at the outside edge face of the nut. When it's just a little behind the edge you get what you describe, or a buzz. Am I missing something ?

I was giving two extreme examples that I have observed. The strings in the slotted saddle were very muted. But the strings sliding violently over the saddle was just a very heavy fisted player who was overplaying the guitar. He appeared to get the same volume even when he backed off a bit. I think the best situation is a normal saddle with strings unconstrained, but played at normal levels.

One area where I think multi piece saddles can help is with undersaddle transducers.
Cut the saddle between A and D and G and B and no more uneven output issues

Barry Daniels wrote:I was giving two extreme examples that I have observed. The strings in the slotted saddle were very muted. But the strings sliding violently over the saddle was just a very heavy fisted player who was overplaying the guitar. He appeared to get the same volume even when he backed off a bit. I think the best situation is a normal saddle with strings unconstrained, but played at normal levels.

Ah, OK, sorry to misunderstand. I haven't tried the wide slotted saddle with the string riding the entire length from hole to edge of saddle, but your experience makes me wonder why it should make any difference at all ? The only think I can think of is more force being transferred from the string directly forwards rather than downward. I guess I'll have to set up a test rig...

Chuck, thanks for your thoughts. But regardless what I hear physics is physics, and torque = force times distance. To make the saddles "taller" without making the stings ridiculously far from the top, I made the bridge a little thinner than I'd normally do, at least toward the back it tapers to about 1/16". Whether ths torque makes a muscial sound... I think it does. I had a Martin OOC-MAE that I compared the sound to, and while it would be tough to call one better, I ended up preferring the sound of mine.

I believe a lot of the super-"clean" sound came from the phenolic sides, so I'm in the process of building another guitar, this time of real wood. Already have the top back, and side wood resawn; just a matter of bending and bracing.

I geuss in short, I'm willing to potentially build a "dud" to test these theories out. I think there are a few simple experiments that can be done on a normal guitar. One would be to take a regular 1/8" saddle, and scallop in between the strings so the tops of the scallops are just wide enough to hold the strings (grooved for the strings naturally), and the bottoms of the scallops flush with the bridge. If the saddles don't stick ot much from the bridge this may not be possible. The second experiment wold be to make a wide saddle with a shelf that goes all the way to the bridge pin. I might have a beater laying around that I could use for 'tests'...

I think the idea that "if the string is not supported at all points behind the saddle that string energy is being wasted" does not make practical sense. That part of the string may be vibrating slightly but it is not doing much work other than moving a few air molecules. The majority of the string's energy will be adsorbed by the top with the remainder being reflected back into the string to enhance sustain. And I believe this is true whether the string is sitting on a skinny bone saddle or a super wide saddle.

Most the string energy will be adsorbed by the saddle at the leading edge because this is where all of the energy is available. Go back into a wide saddle a short distance and there is very little vibration going on. There just is no benefit to a wide saddle. In fact, I think a wide saddle will be much stiffer which makes more of an impedence mis-match with the string, and this can lead to more string energy being reflected back into the string instead of going into the top.

When I took Somogyi's voicing class, I was a bit bold and suggested to his apprentice that a thinner saddle might bring a little brightness to their guitars, which they seem to lack. But that suggestion went over like a lead ballon. Or well...

Barry Daniels wrote:I think the idea that "if the string is not supported at all points behind the saddle that string energy is being wasted" does not make practical sense. That part of the string may be vibrating slightly but it is not doing much work other than moving a few air molecules. The majority of the string's energy will be adsorbed by the top with the remainder being reflected back into the string to enhance sustain. And I believe this is true whether the string is sitting on a skinny bone saddle or a super wide saddle.

Most the string energy will be adsorbed by the saddle at the leading edge because this is where all of the energy is available. Go back into a wide saddle a short distance and there is very little vibration going on. There just is no benefit to a wide saddle. In fact, I think a wide saddle will be much stiffer which makes more of an impedence mis-match with the string, and this can lead to more string energy being reflected back into the string instead of going into the top.

When I took Somogyi's voicing class, I was a bit bold and suggested to his apprentice that a thinner saddle might bring a little brightness to their guitars, which they seem to lack. But that suggestion went over like a lead ballon. Or well...

Intersting points. But this is the reason I made the saddles in such a way that the vibrations are transmitted differently, actually more in line with the way the strings are normally strummed or plucked. I would say my guitar does not lach for brightness, but it's not shrill or thin-sounding either.

Louie Atienza wrote:Chuck, thanks for your thoughts. But regardless what I hear physics is physics, and torque = force times distance. To make the saddles "taller" without making the stings ridiculously far from the top, I made the bridge a little thinner than I'd normally do, at least toward the back it tapers to about 1/16". Whether ths torque makes a muscial sound... I think it does. I had a Martin OOC-MAE that I compared the sound to, and while it would be tough to call one better, I ended up preferring the sound of mine.

I try to get away from statements like "I think it does". While that may truly reflect my take on things, it doesn't really address whether this is verifiably true and repeatable. Comparing the sound to another guitar doesn't tell you anything unfortunately.

I believe a lot of the super-"clean" sound came from the phenolic sides, so I'm in the process of building another guitar, this time of real wood. Already have the top back, and side wood resawn; just a matter of bending and bracing.

I geuss in short, I'm willing to potentially build a "dud" to test these theories out. I think there are a few simple experiments that can be done on a normal guitar. One would be to take a regular 1/8" saddle, and scallop in between the strings so the tops of the scallops are just wide enough to hold the strings (grooved for the strings naturally), and the bottoms of the scallops flush with the bridge. If the saddles don't stick ot much from the bridge this may not be possible. The second experiment wold be to make a wide saddle with a shelf that goes all the way to the bridge pin. I might have a beater laying around that I could use for 'tests'...

Even on a "dud" you can test for differences in sound when you change something. This is really the only way to do it because it's the only way you can start comparing the sound with the base configuration and the changed configuration. There are still some testing glitches that can get in the way, but at least the top/back/sides/neck/strings etc. are the same. It's still possible to fool yourself, so being able to record (full spectrum) the sounds for later comparison and analysis beyond what you think you hear is important. To quote the point of The Point: "You see what you want to see, and hear what you want to hear" and we all do it.

I've built several test guitars with replaceable components. I'm not going to sell them and I certainly don't claim they have great sound compared to something I'd build for sale, but they serve the function of answering the question of "what happens to the sound if all I change is xxxxx ?". Very often the answer ends up being "nothing discernible", but you don't know that until you do a valid test case and remove the "want to" element from the analysis.

Chuck Morrison wrote:

Louie Atienza wrote:Chuck, thanks for your thoughts. But regardless what I hear physics is physics, and torque = force times distance. To make the saddles "taller" without making the stings ridiculously far from the top, I made the bridge a little thinner than I'd normally do, at least toward the back it tapers to about 1/16". Whether ths torque makes a muscial sound... I think it does. I had a Martin OOC-MAE that I compared the sound to, and while it would be tough to call one better, I ended up preferring the sound of mine.

I try to get away from statements like "I think it does". While that may truly reflect my take on things, it doesn't really address whether this is verifiably true and repeatable. Comparing the sound to another guitar doesn't tell you anything unfortunately.

I believe a lot of the super-"clean" sound came from the phenolic sides, so I'm in the process of building another guitar, this time of real wood. Already have the top back, and side wood resawn; just a matter of bending and bracing.

I geuss in short, I'm willing to potentially build a "dud" to test these theories out. I think there are a few simple experiments that can be done on a normal guitar. One would be to take a regular 1/8" saddle, and scallop in between the strings so the tops of the scallops are just wide enough to hold the strings (grooved for the strings naturally), and the bottoms of the scallops flush with the bridge. If the saddles don't stick ot much from the bridge this may not be possible. The second experiment wold be to make a wide saddle with a shelf that goes all the way to the bridge pin. I might have a beater laying around that I could use for 'tests'...

Even on a "dud" you can test for differences in sound when you change something. This is really the only way to do it because it's the only way you can start comparing the sound with the base configuration and the changed configuration. There are still some testing glitches that can get in the way, but at least the top/back/sides/neck/strings etc. are the same. It's still possible to fool yourself, so being able to record (full spectrum) the sounds for later comparison and analysis beyond what you think you hear is important. To quote the point of The Point: "You see what you want to see, and hear what you want to hear" and we all do it.

I've built several test guitars with replaceable components. I'm not going to sell them and I certainly don't claim they have great sound compared to something I'd build for sale, but they serve the function of answering the question of "what happens to the sound if all I change is xxxxx ?". Very often the answer ends up being "nothing discernible", but you don't know that until you do a valid test case and remove the "want to" element from the analysis.

There's no such thing as verifying likes and dislikes in reference to sound. Heck, 99.999999% of guitarbuilders make a Martin clone. Is it necessarily because Martins sound so good, or because they sound so familiar? You can verify if others like the sound by having them play it and asking directly. As far as repeatability, there is absolutely no one who could make an exact duplicate of a guitar, not even a CNC. You can make something consistant, but due to the nature of wood, every guitar is unique. I can't say absoluely, because no matter how good a guitar sounds, there's always a better sounding one somewhere! And it can change day-to-day. Unfortunately, aside from getting billets of wood and making matched sets for multiple guitars, there'll never be a way to absolutely know.

But that's why I'm building another, to see if what I hear in the first will be there in some way with the second. I know that biases can sway judgment, but I'm a pretty analytical guy and I do record stuff to hear back... to the point of nauseum. Just ask my bandmates when we record. I notice everything... even how people breathe!

Almost every time I see someone try something different, it's always in these tiny insignificant moves, and they wonder why the differences in results are insignificant. It's like the guitar gods will strike them with a lightning bolt! And then others will say, "Don't do it, you won't hear a difference anyway." Of course not! Because what did they do? Shave a little off here, add a piece there, make this top .001" thinner... my eyes hurt from rolling them so far back! Kasha's bracing is odd, but cool nonetheless. But he DID IT! Or Smallman and his stiff back and lattice brace. He DID IT! Look at Fleishman's dual-wood tops. He DID IT! Or the interesting Batson bracing and bridge? I'm not saying that these are great guitars (though they have their following) but just giving examples of individuals doing things out of the norm and not afraid to try it, and sell it. I think if more people experiment DESPITE what others say instead of BECAUSE what others say, we'd have a lot more interesting things to talk about, listen to, and see... instead of walking lock-step with everyone else... but that is just me...

As I understand this discussion, we are talking about how to harness any energy that the section of string between the saddle and the bridge pin produces right? Louie’s theory is that vibrations in this section could be captured by the soundboard and produce a different tone or increase volume. I have never built a guitar but have been playing for quite a while. My question is this: What about the section of string between the nut and the tuners? Can any energy loss here be reflected back for either more sustain, increased volume or tone enhancement? My suspicion is that it can’t but like I said, I’m a player – not a builder.

Welcome to the new forum Michael.

Louie,
You're preaching to the choir as far as I'm concerned. Two points.

I completely agree that you can't test for whether you like something. That wasn't my point. You made claims that a specific change had a specific physical result. Since you make a specific physical claim, that result can be tested for to see if that's really what's happening. To do that you need to set up the "change only one thing" scenario and have a better way to measure than human ears.

On the other hand, if we are trying to learn how it all works, doing really big changes is a great way to go. The only thing is that you still have to verify which of the many things you changed (or what combination) had what effect. At some point we still end up having to verify our hypothesis. If we don't then we are just guessing. Personally I prefer knowing and not just speculating.

Chuck

Chuck,

I hope you don't mind me prodding you! My intentions are no other than to learn, and I appreaciate your input.

That said, I never made a claim that a specific change had a specific result. I did however explain the results the best I could observe (hear.) What I DO know is physics, and the simplest of ckassical mechanics tells me what I don't need to prove. Work is force times distance. In angular terms it's called torque. If the top surface of the bridge is the pivot point, then the top of the saddle would be the lever. If I can make one assumption, that the energy that one can "produce" on an individual string remains somewhat constant, regardless of saddle height, them the higher the saddle, the more sideways torque is applied to the bridge. Since a normal saddle is long and coupled to the bridge, then it's pivot point is the same as teh bearing point of the string, or the end of the 'lever' if you will. Therefore none of the sideways movement of the string can apply torque to the bridge in that direction. The only movement would be from the flex of the soundboard sideways or the bridge itself. I won't question Joule on his formula of work.

To extend that, I would infer that stiffening or loosening the soundboard bracing in that direction would affect the way the soundboard moves as a result of the torque on the bridge exerted by the strings. My attempt was to utilize that by removing as much sideways bracing as I though it I could without the top collapsing. I made a sort of 'hybrid' brace that's neither x or parallel. But I also attempted to minimize torque in fromt of the bridge by stiffening the soundhole area. I would say part of my inspiration comes form Mark Swanson's parallel-braced guitars. It also derives from the archtop world, where both parallel and x bracing are both used, and both sounds and effects are well known.

But even if I made one guitar, and swapped out the bridge, then who's to say it wasn;t the wood itself that caused the change? How is that quantified then? Or the time of day, humidity, glue mix? Point is, no matter what, you'll never ever really know what really happens. We all observe the world in the past tense, even if it's only a fraction of a second. Even recordings, slow motion video, oscilloscopes, are observed and read in the past tense.

But such is the beauty of humanity. It is our gift of veing able to judge for ourselves what we like and don't that makes us who we are. It's our sense of exploration and wonder that made us saw the world is round, broke the 4-minute mile, and landed on the moon. I've always been curious, willing to explore, take risks, be different. Many people feel I get along and converse with anyone, yet I can be a bit of a loner. This type of stuff keeps me going, my hands moving, my mind working. Sadly, I tend to leave many projects unfinished, because I have already completed them in my mind, and know they work, and don't feel the need to bring them to fruition. I'm really trying to change that. I own Benedetto's, Cumpiano's and Somogyi's books. But if I followed the instructions in them, I don't think I'd be here now because I wouldn't need to complete them; the pictures in the book show me they're already done...

Louie Atienza wrote: That said, I never made a claim that a specific change had a specific result. I did however explain the results the best I could observe (hear.) What I DO know is physics, and the simplest of ckassical mechanics tells me what I don't need to prove. Work is force times distance. In angular terms it's called torque. If the top surface of the bridge is the pivot point, then the top of the saddle would be the lever. If I can make one assumption, that the energy that one can "produce" on an individual string remains somewhat constant, regardless of saddle height, them the higher the saddle, the more sideways torque is applied to the bridge. Since a normal saddle is long and coupled to the bridge, then it's pivot point is the same as teh bearing point of the string, or the end of the 'lever' if you will. Therefore none of the sideways movement of the string can apply torque to the bridge in that direction. The only movement would be from the flex of the soundboard sideways or the bridge itself. I won't question Joule on his formula of work.

.

Sorry, but that is in no way correct Physics, classical mechanics or analysis of the bridge
For starters, work and Torque are in no way equivalent.

Jeff Highland wrote: Sorry, but that is in no way correct Physics, classical mechanics or analysis of the bridge
For starters, work and Torque are in no way equivalent.

Jeff, thanks for your comments. I have to admit, my physics knowledge had been buried in my head around 1993, when I took my last physics course. That said, I tried to draw vector diagrams of the bridge and saddle, to grasp how a force such as a string's vibration can act upon it. Maybe I missed something, and you can shed some light on the subject, or point me in the right direction.

I did misspeak about work and torque. Though if torque is force times the length of the lever arm, then the work produced by the lever arm moving radially from the 'pivot point' woud be the torque times the angle in radians. Does this still not show that increasing the lever arm increases the work done if the force remains constant?

A friend and I are in the process of finalizing (we hope!) a paper for submission to the on-line Savart Journal on the subject of the influence of string height off the top and break angle on the sound of a guitar. There's a lot to it, but I can say that in the listening tests (recordings of 'standard plucks' plasyed back through earphones) what people could most reliably hear were changes in string height. Physical measurements gave us some information on what they were probably hearing. As usual, it's all a lot more complicated that you think it ought to be. I went in thinking I knew what the pivot point was, and now I'm not so sure, for example. What I usually learn in these experiments is how to do the experiment that might have a chance of actually proving the point.

Alan Carruth / Luthier