Official Luthiers Forum!

I'm having fun planning what I want to do on my first build......but I need to narrow a few things down quickly since I would like to purchase by Christmas.

Trying to decide on a bridge design and have a few conceptual questions I hope you can help with. Not looking for precise answers necessarily as I know there are many variables but generally there are trends. Here are my questions:

- How does the width of the bridge (side-to-side) affect the transfer of vibration on or across the soundboard and how does the bridge width affect the sound? For example, if I had the same style bridge that spanned 1/3 the width of the lower bout and another bridge that spanned 1/2 the width of the lower bout, what differences would be noticed?

- How does the depth of the bridge (measured from side nearest the sound hole to the side nearest the tail block) affect the transfer of string vibration to the soundboard and how will it affect the sound or sustain? Is there a minimum depth that should be maintained so too much stress isn't placed on the soundboard from the string pull?

- Any thoughts on how bridge plate wood choices affect the guitar sound or how the bridge plate size relative to the bridge size affects sound would also be appreciated. I've read where some folks use a particular wood for bridge plates on "large" guitars and another wood for "small" guitars. Is an OM considered large or small? I'm used to dreadnaughts so it seems middle of the road to me.

_________________
Formerly known as Adaboy.......

Read Ervin Somogyi's article on principle of guitarmaking. He addresses some of your questions about three-quarters through the article.

http://www.esomogyi.com/principles.html


Larry

Darryl also consider that for an X braced guitar the bridge has a function that is really important that we often do not think about. The bridge is a brace that needs to span the X brace legs and become part of the bracing structure for the top. With this said we don't have the option on X braced guitars of making the bridge shorter than would be required to span the X legs when the bridge is properly positioned in respect to the scale of the guitar.

Bridge weight and mass are important. Some time back I took 10 pre-made martin bridges in ebony and EIRW and weighed them. Most seemed to come in the 28 to 42 gram area. With this knowledge, unless you are trying to reinvent the wheel I would suggest a bridge design that stays within the above weights.

Also my pal JJ has said it best with one of the criteria that he uses to choose tone woods for bridges and bridge plates - if it pings good it may make a good bridge. So you want a bridge material with low damping. Some folks such as Colin will not even consider using ebony for this reason, it has higher dampening and it more mass too.

Many of the rosewoods would make good bridges and folks have used many other woods as well. I have had good luck with ebony even though I agree with Colin's logic completely.

My guess is that some folks use particular woods on certain sized guitars looking to match the mass of the bridge to the size of the top that needs to be driven. Often is is simply aesthetics.

I would consider an OM to be a medium sized guitar IMHO.

Just to expand a bit on what Hesh pointed out. The bridge design length is not a ratio of the lower bout width. The bridge length is deternined by the need of the bridge wings to extend across each Xbrace arm the full width of the bridge wings for structural reasons. The wider the Xbrace spread the longer the Ideal bridge length. This aids in reducing the outward belling behind the bridge and the inward belling behind the sound hole caused by the pull of the strings over time.

As has been pointed out, the bridge is the heaviest, and one of the stiffest, braces on the top, and that's the aspect of it that has the most effect on the tone of the guitar. It's main job, though, is to define the vibating length of the string. Naturally, none of the different aspects of bridge design can be seperated out neatly: you can't really talk about the tonal effect of making it deeper along the line of pull, for example, without also talking some about what effect that will have on the other aspects, such as mass and stiffness.

The bridge does it's main job of defining the string length mostly through presenting a large 'impedance mismatch' to the end of the string. Mass and stiffness are the key words here: impedance is simply the square root of stiffness times mass, and you want that number to be much higher for the bridge than the string. The trick is, of course, that if the bridge was _perfectly_ rigid and _massively_ heavy the guitar would make no sound, since the top would not be moving. So you end up compromising, as usual. The range of weights given for steel string bridges, from about 28-40 grams works well. As you get more massive the total output will probably go down, and particularly the trebles, all else being equal (which it never is). Of course, the mass and stiffness that the string 'sees' is not just that of the bridge, it's the whole top.

Making the bridge deeper, that is wider along the line of the string pull, helps it do it's length-defining job by keeping it glued to the top longer. Suppose you overlap two sticks by, say, 2", and glue them together. If you pull on the ends the glue line will be in shear. The stress in the glue line will not be the same throughout the length of the joint: it will be at some uniform low level through, say, the middle inch, and rise to a maximum toward each end of the joint. The area under that stress curve will equal the total force on the joint. The maximum allowable stress on the joint will occur when the peak at either end rises to the point that it starts to peel the end loose.

The height of the uniform part of the curve will be established by the nature of the wood and the glue (assuming you did a good job). The longer the overlap of the two sticks the more of the total stress will be taken up by that level part of the curve in the center, and the lower the peaks on either end will be. That's why 'belly' bridges tend to stay glued down longer than straight ones; the max stress along the rear edge is lower.

The length of the bridge, measured across the top, is more important in terms of it's 'brace' function. The wings of the top, between the bridge ends and the edge, flex a lot in certain types of vibration, and making the bridge longer (and the open area of top in the wings less) can really stiffen the top up in those modes.

There's certainly a lot more that could be said about bridge design, but as you go further along it gets harder to seperate fact from conjecture. At some point I'm hoping to get a chance to sort some of it out, but don't hold your breath.

Michael, I'm not sure I'd agree with that. As you point out it should extend over the X braces as a minimum. However, beyond that I believe it is an important design consideration. The length would affect the function of the bridge as a brace, providing structure to the top, and would also affect the weight. Personally I do vary the length depending on the lower bout width.

_________________
David White, Toronto

"All my favourite singers can't sing."

Thanks everyone for their replys......and thanks for the link to the somogyi article, good info.

The bridge wants to rock forward due to teh string tension so the front part of the bridge is being pushed into the top of the soundboard and the rear of the bridge is being lifted. Seems you want the bridge as free as possible while keeping things structurally sound so does anyone locate the bridge where the front corners straddle the X-bracing (to support the front of the bridge which wants to rotate into the soundboard) and the rear corners of the bridge aren't over the X-bracing so the rear of the bridge is more free to rotate transferring vibration? Not sure if this would be structurally sound or not.

_________________
Formerly known as Adaboy.......

Another related question the fits into this very informative post - what considerations are there for the tonal and structural qualities of bridge design in terms of the grain direction of the wood (i.e. rift sawn, flat or quarter sawn)?

john

IMO the rear corners of the bridge wings should not be inside the X. If you do that the corner will lift the top too much. I guess you could compensate by extending the bridge plate back, but my rule of thumb has been that the rear corner right on the inside edge of the brace is as far as I'll go. Rear corner more toward the outside or completely outside the brace makes for a tighter sounding, and structurally stronger guitar.

Some of the older Gibson AJ's had the rear corners well inside the X. But they tend toward excessive bellying and eventual self-destruction.

Like David, I vary the bridge width (the lateral dimension), which also allows the spread of the X to be varied. My bridges are wider than a Martin's. Their one size bridge fits all never made sense to me.

I go from quarter to rift. Never flat. If rift, I prefer the annular ring lines to run from front bottom to rear top. It probably doesn't really matter much, but I figure this would present a radial line to the typical saddle split, which is less likely to break than a tangential line.

_________________
Howard Klepper
http://www.klepperguitars.com

When all else fails, clean the shop.

Let's say your bridge is profiled like most - thick across the center section (where the saddle and pins go) and thin at the wings. With typical X bracing, the thick center section sits directly over the legs of the X at the front edge of the bridge. Otherwise stated, the legs of the X pass under the front "corners" of the thick center section. If you position the intersection of the X so far north of the bridge that the legs pass under the wings only - not under the center section at all - the structure of that is much weaker. The wings are obviously not nearly as rigid, and they may twist a bit as the bridge torques forward. Moreover, in that scenario, there is a fairly large area of unsupported top directly in front of the bridge, where the downward force is strongest. Unless your top is quite thick, you may see excessive distortion (dipping down) of that area. If you look at guitars where the X is further north than usual (where the legs pass under the wings only), looking at reflected light on the surface of the top, you are likely to see a triangle of concavity in that area between the front edge of the bridge and the X. Arguably, some bellying of the top behind the bridge is not as much trouble to the guitar as dipping down in front of the bridge. I would take a good hard look at how the structure you're creating (all elements - the bridge, b plate, the top itself, the bracing - and how they work together once glued up into a unified structure) supports the area in front of the bridge.

No matter how you slice it, it comes down to the balancing act between making it structurally sound on the one hand and acoustically responsive on the other. Simplistic though it may sound, a good guiding principle is to make it just stiff/strong enough, and no more. Thinking about specifically how the force of string tension acts on the top - where the forces are strongest and in what direction the forces are working on the various areas of the top - tells you where the structure needs to be stronger and where it can be more flexible, so that you can aim for that "just enough and no more" point in each area and component of the top. Then, beyond that, you can do whatever you want with attempts at shaping the sound by arranging the braces in various ways, putting peaks on them in various places, or what have you - always bearing in mind the aforesaid principle.

_________________
Todd Rose
Ithaca, NY

https://www.dreamingrosesecobnb.com/todds-art-music

https://www.facebook.com/ToddRoseGuitars/

Here's the layout of a 1936 D-18. I don't have any thing to add, I just like the picture.

_________________
Jeremy Douglas

Well I do find that the rear corners of the bridge crossing the X is desirable to reduce rotating. i think Howard and i are saying the same thing. Isee the D-28 did not but that does not disprove the that having the rear corners crossing the X increases the resistance to bridge rotation. The rear of the bridge is in sheer load. the front of the bridge is in compression.

Thanks for posting that, Jeremy. Very interesting. I'd like to look at that guitar and see how thick the top is, how thick and broad the bridge plate is (it's odd that you can't really see the shadow of the bridge plate in this photo), what the dimensions of the braces are, how much the top has distorted and in what areas it's distorted most, etc. Whether the bridge and/or bridge plate are original, and how thick the wings of the bridge are, would be revealing also. It may well be an example of a design with that relationship between the positions of the bridge and the X that has worked fine, albeit assuming the guitar has had at least one neck reset and possibly some other structural repairs as well.

Although this example appears to challenge what I was saying, and what Howard and Michael are saying as well, there are a couple of things to note that are in line with my points. One is that the wide angle of the X puts the intersection closer to the front of the bridge than it would be if the angle were closer to 90 degrees with the legs crossing under the bridge at the same points on the wings. So, there's not so much unsupported top in that area directly in front of the bridge. The angle of the X is certainly a factor which I didn't mention in my post above.

The other thing is just a more general observation about the brace layout and the need for support in the area in front of the bridge. If you draw an imaginary line across the guitar at the saddle, and another one across the guitar at roughly the middle of the soundhole, just look at how much bracing there is between those two lines, compared to the area below the lower line (in the belly area). I would suggest that that illustrates Martin's experience with how much support is needed in front of the bridge to keep that area from caving in excessively.

_________________
Todd Rose
Ithaca, NY

https://www.dreamingrosesecobnb.com/todds-art-music

https://www.facebook.com/ToddRoseGuitars/

I place my bridges somewhere very close to the Martin photo .. but I also do have a larger bridge plate that is trapezoidal and extends along the rear edge of the belly to the X. Plus I use a Laskin/larrivee bracing which has a brace right behind the bridge plate, and parallel .. simliar to Marios add on (however with larrivee, its always been there)

_________________
Tony Karol
www.karol-guitars.com
"let my passion .. fulfill yours"

I got radial and tangential reversed here. But the grain direction is right.

_________________
Howard Klepper
http://www.klepperguitars.com

When all else fails, clean the shop.

Sure thing. From my understanding those bridge plates were pretty small. I think that shadow below the wings of the bridge is it! All those pre-war D-18's had an X angle of about 97.5-98 (so I'm told) so the bridge should fall in about the same spot on all of them. The braces should be about 5/16" thick and 5/8" tall at their peak (including the UTB!).

_________________
Jeremy Douglas

Jeremy my friend I think that the picture is very telling as well - thanks again for posting this!

When I look at the photo I see that the back edges of the bridge are not over the X. Although the bridge is free to rotate more is it also not unlike a crude form of suspension? Just as it is important to not over brace the inside of a guitar top is it not also important, since the bridge is a brace, to not over brace the outside of a guitar top? Let's not get into pick guards........

Also in my mind when considering the pre-war and highly prized Martins I think that it is important to consider the historical context. The guitar had not yet earned the fame and popularity that it has today. Since there were no electric guitars the guitar had a difficult time being heard over or even with the other band instruments of the day. Projection was probably the holy grail when it came to acoustic guitar design in 1937.

With these things in mind in 1937 perhaps tone and projection had a seat closer to the chairman of the board at Martin than the risk-reduction/warranty claims folks.

Check out this Norman Blake model OM.

On the treble side, the back of the bridge is supported by one of the tone bars. On the bass side, the rear of the bridge doesn't appear to have support other than the bridge plate. I've heard good things about the way these guitars sound but I've never heard one in person. I'm no expert but it seems this would spread the vibration of the higher notes across a large area of the soundboard.

I think you can see where the bracing is taller from the wider and darker shadows.

_________________
Formerly known as Adaboy.......

Actually, I want to do something along the lines of the Norman Blake on my first build. I may go with a 13 fret neck but use the longer 25.5" scale which would place the bridge near the same location. (the Blake has the shorter 24.9" scale....or whatever the shorter measurement is....and a 12 fret neck) I'm waiting on a tracing from John Hall now.

Thoughts?

_________________
Formerly known as Adaboy.......

Well, now, I reckon you're right. Darnedest thing.

I know this is taking a tangent from Darryl's original question, but does anybody know how thick those prewar Martin D tops tended to be? I assume they were generally made of some seriously stiff Adirondack spruce.

_________________
Todd Rose
Ithaca, NY

https://www.dreamingrosesecobnb.com/todds-art-music

https://www.facebook.com/ToddRoseGuitars/

This thread has spurred me to do some more homework on different bracing designs, looking especially at the relative positions of the bridge and the X, along with the design of the bridge plate, as well as the rest of the picture. I just looked at one of the pages on Martin's web site, which I'd never looked at before -

http://www.martinguitar.com/guitars/fea ... index.html

The second to last one, called A-Frame "X", of which there is also a photo on Frank Ford's site, is kind of a mind-blower to me. I didn't know Martin was using this bracing design, and I'm amazed at how minimal it is. There appears to be hardly anything supporting the belly, and there's not much to speak of by way of finger braces, either. But apparently this works (structurally, anyway; I don't know how they sound), even on dreadnoughts.

I've also been looking at a variety of other plans and photos of guitar tops and seeing plenty of other examples of things that step outside of the models I was familiar with and which I've been using as springboards for my own design ideas (and I'm just talking about variations of X bracing, not Kasha or Klein or lattices or other more radical designs, which I was already familiar with).

Anyway, I'm not about to abandon all my current thoughts and working hypotheses on bracing/bridge/bridge plate/top design, but at this moment, I am inclined to say, "What the heck do I know?"

Actually, though, I find the minimal Martin bracing scheme as inspiring as it is puzzling. Maybe it says something very significant about just how little bracing can be used if it's put in just the right places, which, maybe, points back to the basic principle I was trying to articulate in my first post, about "just enough and no more".

_________________
Todd Rose
Ithaca, NY

https://www.dreamingrosesecobnb.com/todds-art-music

https://www.facebook.com/ToddRoseGuitars/

Here are some top measurements I got from the UMGF (John Arnold originally posted them I think).

"My 1937 D-28 measures 0.130" to 0.135".
1937 D-28 #66282....0.120".
1935 D-18 sunburst #59665....0.110" to 0.115".
1935 D-28 #61180....0.105" to 0.112".
1941 D-28 #79361....0.115" at soundhole.
1931 D-2 #47055....0.127" below bridge, 0.122" at soundhole.
1929 Ditson 111 #37857.....0.118" at soundhole, 0.094" bass side below bridge, 0.105" treble side below bridge.
(Chuck Enger also noted that the top on his 1937 D-18 is .118" thick)"


Here's another pic of that A Frame bracing

http://martinrep.com/tb.htm

You have to wonder how they do it. Thick tops and bridge plates?

_________________
Jeremy Douglas

That A frame X brace layout is wild. I'm guessing angling the grain on the bridge plate helps the stiffness considerably....but you would think it would also hamper the forward and aft rocking of the bridge while the strings vibrate. Bridge plate is fairly good size as well. Sure be nice to have time and material to play with these ideas.

Anyone ever heard one of the Martins with the A-frame X bracing like shown in the picture? I've not but I'm curious how they sound.

_________________
Formerly known as Adaboy.......

Jeremy, thanks for posting the pic and the list of top thicknesses.

"I'm guessing angling the grain on the bridge plate helps the stiffness considerably....but you would think it would also hamper the forward and aft rocking of the bridge while the strings vibrate." Darryl, the thing is, you've got to keep the bridge from rotating excessively under string tension one way or another. Any way that you do that is going to also stiffen the structure such that it will affect the motion of the bridge. If anyone with advanced degrees in physics or engineering wants to correct me if I'm wrong, please do, but it seems to me that no matter how the stiffening of that area is accomplished, if the bridge rotation under string tension is equal between two guitars, then the effect on the rocking oscillation of the bridge will also be equal.

_________________
Todd Rose
Ithaca, NY

https://www.dreamingrosesecobnb.com/todds-art-music

https://www.facebook.com/ToddRoseGuitars/

Todd's right.

If you look at the sound output of a guitar in response to a 'rocking' signal applied to the bridge, you'll see that it's pretty low until you hit the frequency of the 'long dipole' mode of the top. This is the resonant mode in which the area behind the bridge rises as the area in front goes down, and it's usually somewhere around 375 Hz (say F# on the high E string, 2nd fret) on most guitars. This is not a very efficient sound producer, since the motion of the two areas cancels out to some extent: it's easier for the air to 'slosh' along the top than to actually make sound. Finally, the 'rocking' signal in the strings is not very powerful compared to the 'up and down' transverse force from the vibration of the string. Typically the tension change force will be something like 10-30% as strong as the transverse, and it only approaches the same strength on a plain G string.

In short, the tension change signal rocking the bridge is likely not going to be responsible for a lot of the sound output of the guitar. It probably contributes something to the tone color, and it's one reason changing the saddle height alters the tone. The higher the strings are off the top the more leverage they have to torque it, and the more sound the tension change signal makes. The bridge also tends to come off sooner, meaning you have to reinforce things to keep body and bridge together, which is likely to cut down on the effectiveness of the tension change signal in producing sound.... So, yeah, Todd's right.