Having built two "practice" guitars (one for for each my right handed sons) my next build is for my left-handed self. I expect the guitar to outlive me and would like to think that one of my two guitar playing boys could change the bridge over fairly easily and make it a right hander. However the bracing cannot easiily be changed.
Should I brace for a left hander and not worry about the bracing being the wrong way round later or use a symmetrical bracing pattern (a la Gibson)???
Or is the symmetrical brace pattern a no-no in the serious luthier fraternity?? Every (steel string) guitar I have seen in my short time on this forum has had asymmtrical bracing, so it seems everyone here prefers it to symmetrical bracing.
Any thoughts??
A lefty builders legacy dilemma.....
-
Hippety Hop
- Blackwood
- Posts: 175
- Joined: Wed Apr 30, 2008 11:18 am
- Location: Moorabbin
-
graham mcdonald
- Blackwood
- Posts: 472
- Joined: Thu Oct 25, 2007 11:57 am
- Location: Canberra
- Contact:
I think that you will find the general consensus is that it will make no noticeable difference if the bracing if right or left handed however the bridge saddle slot is cut. There is no 'bass' or 'treble' side of the soundboard. The soundboard works as a unit with resonances as noticed by M. Chladni 200 or so years ago. The bracing adds stiffness which controls the frequencies at which those resonances occur, and there is some evidence to suggest that the bracing pattern influences the node and antinode shapes of the Chladni patterns. Have a look at the UNSW Acoustics website (can't remember the URL, but google will pull it up pretty smartly)
It is fairly simple to change a bridge over from right to left handed, but it will involve filling in the left handed slot and cutting a new right handed slot (or vice versa). That does require a router jig and a degree of set up and can be little nerve racking. Why not cut two bridge slots left and right handed in a shallow X when making the bridge. If the look of it is distressing the two halves of the slot you are not using could be filled in with press-fit plugs, but it would make a conversion easier down the track.
cheers
It is fairly simple to change a bridge over from right to left handed, but it will involve filling in the left handed slot and cutting a new right handed slot (or vice versa). That does require a router jig and a degree of set up and can be little nerve racking. Why not cut two bridge slots left and right handed in a shallow X when making the bridge. If the look of it is distressing the two halves of the slot you are not using could be filled in with press-fit plugs, but it would make a conversion easier down the track.
cheers
Graham McDonald
http://www.mcdonaldstrings.com
http://www.mcdonaldstrings.com
-
DarwinStrings
- Blackwood
- Posts: 1877
- Joined: Thu Nov 13, 2008 10:27 pm
- Location: Darwin
Graham - thanks again. last nights googling led me to the excellent USNW website and through the Chladni patterns to Ernest Somogyi's lecture on bracing. It was all very interesting.
What I took from this relatively short session was:
1. The first three or four modes of oscillation of the soundboard have nothing much to do with the bracing -in out, back of forth, side to side.
these modes are lower frequency and have more to do with bass sound than treble sound.
2. The upper modes of vibration produce the higher frequency sounds and appear to be a bit random at least to look at. There is no apparent correllation to bracing patterns.
So this begs the question, why is the "Martin" brace pattern so popular?
However.......
I think that it doesnt much matter which way round I put the bracing, or what pattern I choose, as it probably won't make a difference that I would be able to pick up without a microphone and a PC (or Mac!!) . For that reason I will probably go for a symettrical brace pattern.
As a structural engineer ( at least I was once) the X brace has now assumed more importance than the bracing pattern. I can make a better joint and make this a lot stiffer and lighter if I want to but.... is more stiffness necessarily the aim???
Does the fact that we hack the X brace in half at the cross joint make some of the guitars transverse response that we actually like??.
Something for me to think about - thanks again for all your adivce... I'm off to the shed to make some test X braces!!!
Cheers
Richard
What I took from this relatively short session was:
1. The first three or four modes of oscillation of the soundboard have nothing much to do with the bracing -in out, back of forth, side to side.
these modes are lower frequency and have more to do with bass sound than treble sound.
2. The upper modes of vibration produce the higher frequency sounds and appear to be a bit random at least to look at. There is no apparent correllation to bracing patterns.
So this begs the question, why is the "Martin" brace pattern so popular?
However.......
I think that it doesnt much matter which way round I put the bracing, or what pattern I choose, as it probably won't make a difference that I would be able to pick up without a microphone and a PC (or Mac!!) . For that reason I will probably go for a symettrical brace pattern.
As a structural engineer ( at least I was once) the X brace has now assumed more importance than the bracing pattern. I can make a better joint and make this a lot stiffer and lighter if I want to but.... is more stiffness necessarily the aim???
Does the fact that we hack the X brace in half at the cross joint make some of the guitars transverse response that we actually like??.
Something for me to think about - thanks again for all your adivce... I'm off to the shed to make some test X braces!!!
Cheers
Richard
Richard
-
graham mcdonald
- Blackwood
- Posts: 472
- Joined: Thu Oct 25, 2007 11:57 am
- Location: Canberra
- Contact:
Richard,
I think you will find that all the modes of vibration will be directly related to the bracing. The lowest, the air/Helmholtz, resonance somewhere around the G on the bass E string is less to do with the bracing than the air volume of the box, but the stiffness of the soundboard and back does have some effect on the frequency. The first soundboard resonance, around an octave higher, and next three or four above that are very much influenced, both in frequency and shape by the bracing, effectively the stiffness the bracing is adding to the 2-3mm thick plate of spruce in the soundboard. There are lots more resonances about those first four or five, but they don't seem to have much effect on how the instrument works. It is the frequency and separation of the lower ones which have a critical effect on how the instrument works.
People use the Martin X bracing pattern because it works, and produces guitars which lots of people think sound good. To put it another way, it adds just the right amount of stiffness to the soundboard structure in the right places. It is almost infinitely variable in subtle ways. The angle of the X, the position of the bridge in relation to the X, the profile and height of the braces allows lots of variation. Check out the current thread on Allen McFarlen's 000 where there has been lots of informed discussion on the bracing layout.
As a matter of interest, how would you make the X brace stiffer and stronger. To make it the same stiffness with less mass would be an attractive idea
cheers
graham
I think you will find that all the modes of vibration will be directly related to the bracing. The lowest, the air/Helmholtz, resonance somewhere around the G on the bass E string is less to do with the bracing than the air volume of the box, but the stiffness of the soundboard and back does have some effect on the frequency. The first soundboard resonance, around an octave higher, and next three or four above that are very much influenced, both in frequency and shape by the bracing, effectively the stiffness the bracing is adding to the 2-3mm thick plate of spruce in the soundboard. There are lots more resonances about those first four or five, but they don't seem to have much effect on how the instrument works. It is the frequency and separation of the lower ones which have a critical effect on how the instrument works.
People use the Martin X bracing pattern because it works, and produces guitars which lots of people think sound good. To put it another way, it adds just the right amount of stiffness to the soundboard structure in the right places. It is almost infinitely variable in subtle ways. The angle of the X, the position of the bridge in relation to the X, the profile and height of the braces allows lots of variation. Check out the current thread on Allen McFarlen's 000 where there has been lots of informed discussion on the bracing layout.
As a matter of interest, how would you make the X brace stiffer and stronger. To make it the same stiffness with less mass would be an attractive idea
cheers
graham
Graham McDonald
http://www.mcdonaldstrings.com
http://www.mcdonaldstrings.com
I think that the bracing is by far the most interesting part of building for me. There are just so many variables ins style and materials. And it's just a total mystery as to what the sound is going to be like until those strings go on.
I had a chance to hear a well used Martin 0M guitar before and after it had been switched over for a south paw. I tried to hear a difference. But if I was to put it to a blind test, I'm positive that I wouldn't be able to tell you the difference.
I had a chance to hear a well used Martin 0M guitar before and after it had been switched over for a south paw. I tried to hear a difference. But if I was to put it to a blind test, I'm positive that I wouldn't be able to tell you the difference.
The overhead light in shed has blown up and its now dark in Godzone, so I am back on the forum. I seem to spending a lot of time here !
Graham, thanks again for your comments.
I hear what you are saying and agree that it is going to be the three dimensional stiffness of the top that is going to change the way it responds and thus sounds and that the bracing and pattern affects that stiffness. What I meant was that I dont know enough to see how changing the position/location of the secondary braces, if I can call them that, will affect the upper modes of vibration, not that they didnt matter at all.
These braces are pretty small and short of making them out of some wonder material that I dont have, then I cant see much I can do with them, structurally. As you say, these layouts have stood the test of time, so I think I will stick with conventional layout wisdom, be it "martin" or "symmetrical" bracing.
However, I could fiddle around with the cross brace and see if I can make it more efficient - that is increase its stiffness to weight ratio.
How to do this? Since you ask, here's what I was thinking about trying. Reading this back to myself before I post, it sounds like a bit of a lecture - apologies - I am just thinking out loud.
I am thinking that the cross brace is analagous to a stiffener on a piece of steel plate, and provides support in bending like a beam as the soundboard deflects in and out. It is in fact a T beam, because if you try and bend it, then some of the soundbard will act as a flange, because it is glued to one edge of the brace. The most efficient structural shape for pure bending is an I section (like an RSJ or UB steel girder), so I could glue a flange on the other edge of the cross brace, which can then act as the other flange of an I beam. The cross brace could then be a lot thinner to compensate for the added weight of the flange, but will be much stiffer. I can shape the cross brace so the "beam" tapers from centre to edges. This type of profiled beam is used in ply web beams for construction of rooves and in custom built steel portal frames. With modern glues there is no reason why I cannot make a small version in timber. You could machine the I profile with a router first and glue on a T - just like a welder would stiffen a ship hull with an angle.
However,
a) this would make the brace fairly finicky and less robust and it might break in time or glue joints fail, etc. (Fatigue loads) and
b) the centre joint would be a bit of a challenge!
If you took this to the extreme you could make the braces like trusses.
Another way to achieve the same thing might be to drill out a series of holes through the cross brace. In a beam, it is the outer fibres that do the real work - the bits in the middle are not doing much so you can afford to lose them. The stiffness of a rectangular beam is proportional to the cube of the height - so say you drill out 20% of the weight and allow the brace to be 20% higher, you will potentially increase stiffness by 1.2 to the power 3 = 1.7 or a 70% increase in stiffness, with no increase in weight. If you drill out 30% you can more than double the stiffness.
While we are on that subject, if I take a rectangular section brace and carve or plane it into a parabolic cross section, I am reducing its stiffness and (I need to check this but I am pretty sure I am right) I am reducing its efficiency. Why do we do that?
Next thing is the cross brace joint.
Mr Somogyi puts a strap across the half housing joint in the centre of the brace - I dont know what he uses but it looks pretty small. You could calculate how much spruce you would need to effectively replace the lost top half of the brace and splice glue it over the top of the joint. It would be much less than you have cut out, because you are putting it in a more effecient position - on the outer edge of the brace. The soundboard is doing the same thing for the other half of the brace. In fact a piece of offut soundboard about ten time the thickness of the top, long enough to get an effectively glued splice, would probably be ideal.
Come to think of it, thats the answer to the I section joint, just make one beam higher than the other by the thickness of the soundboard and fly one flange over the top of the other!. Or make them all but in the centre and glue a square plate over them all.
The best thing about all this is that its fairly easy to test - Soundboards can be made with different styles of brace and then simply apply the same load and see which one moves least, weight for weight.
However, it all comes back to the question - is a more efficient structural system going to give me a better sound or just more sound?
All very interesting
Graham, thanks again for your comments.
I hear what you are saying and agree that it is going to be the three dimensional stiffness of the top that is going to change the way it responds and thus sounds and that the bracing and pattern affects that stiffness. What I meant was that I dont know enough to see how changing the position/location of the secondary braces, if I can call them that, will affect the upper modes of vibration, not that they didnt matter at all.
These braces are pretty small and short of making them out of some wonder material that I dont have, then I cant see much I can do with them, structurally. As you say, these layouts have stood the test of time, so I think I will stick with conventional layout wisdom, be it "martin" or "symmetrical" bracing.
However, I could fiddle around with the cross brace and see if I can make it more efficient - that is increase its stiffness to weight ratio.
How to do this? Since you ask, here's what I was thinking about trying. Reading this back to myself before I post, it sounds like a bit of a lecture - apologies - I am just thinking out loud.
I am thinking that the cross brace is analagous to a stiffener on a piece of steel plate, and provides support in bending like a beam as the soundboard deflects in and out. It is in fact a T beam, because if you try and bend it, then some of the soundbard will act as a flange, because it is glued to one edge of the brace. The most efficient structural shape for pure bending is an I section (like an RSJ or UB steel girder), so I could glue a flange on the other edge of the cross brace, which can then act as the other flange of an I beam. The cross brace could then be a lot thinner to compensate for the added weight of the flange, but will be much stiffer. I can shape the cross brace so the "beam" tapers from centre to edges. This type of profiled beam is used in ply web beams for construction of rooves and in custom built steel portal frames. With modern glues there is no reason why I cannot make a small version in timber. You could machine the I profile with a router first and glue on a T - just like a welder would stiffen a ship hull with an angle.
However,
a) this would make the brace fairly finicky and less robust and it might break in time or glue joints fail, etc. (Fatigue loads) and
b) the centre joint would be a bit of a challenge!
If you took this to the extreme you could make the braces like trusses.
Another way to achieve the same thing might be to drill out a series of holes through the cross brace. In a beam, it is the outer fibres that do the real work - the bits in the middle are not doing much so you can afford to lose them. The stiffness of a rectangular beam is proportional to the cube of the height - so say you drill out 20% of the weight and allow the brace to be 20% higher, you will potentially increase stiffness by 1.2 to the power 3 = 1.7 or a 70% increase in stiffness, with no increase in weight. If you drill out 30% you can more than double the stiffness.
While we are on that subject, if I take a rectangular section brace and carve or plane it into a parabolic cross section, I am reducing its stiffness and (I need to check this but I am pretty sure I am right) I am reducing its efficiency. Why do we do that?
Next thing is the cross brace joint.
Mr Somogyi puts a strap across the half housing joint in the centre of the brace - I dont know what he uses but it looks pretty small. You could calculate how much spruce you would need to effectively replace the lost top half of the brace and splice glue it over the top of the joint. It would be much less than you have cut out, because you are putting it in a more effecient position - on the outer edge of the brace. The soundboard is doing the same thing for the other half of the brace. In fact a piece of offut soundboard about ten time the thickness of the top, long enough to get an effectively glued splice, would probably be ideal.
Come to think of it, thats the answer to the I section joint, just make one beam higher than the other by the thickness of the soundboard and fly one flange over the top of the other!. Or make them all but in the centre and glue a square plate over them all.
The best thing about all this is that its fairly easy to test - Soundboards can be made with different styles of brace and then simply apply the same load and see which one moves least, weight for weight.
However, it all comes back to the question - is a more efficient structural system going to give me a better sound or just more sound?
All very interesting
Richard
Thanks for that Graham -Graham - Check out the current thread on Allen McFarlen's 000 where there has been lots of informed discussion on the bracing layout.
Allen - I have just looked at your Norman Blake thread and see that on the cross brace joint you have made one brace a bit higher than the other.
Have you cut half the depth out of each brace or have you cut most of the wood out of the one that is higher?.
Cheers
Richard
Richard
Both braces are exactly the same hight and the lap joint is exactly 1/2 the depth of each. I cut them at the same time with a dado blade in the table saw.. What you are seeing is the spruce patch that has been glued on top and spans the lap joint. Very important to do. If you don't then the intersection of the x-brace is much less stiff than if had not been cut in half for the lap joint, as you are well aware.
I've heard that it takes as little as 1/16 of an inch of patch to be effective in this. I generally use a bit of the brace wood off cuts and fashion a 3 mm patch. Glued in place after the x-braces have dried for a day or so. Usually the last piece to go on. I then give it some shape and a gradual transition into the underlying braces, simply because I prefer the look. I don't know if it makes a difference doing this other than aesthetics.
I've heard that it takes as little as 1/16 of an inch of patch to be effective in this. I generally use a bit of the brace wood off cuts and fashion a 3 mm patch. Glued in place after the x-braces have dried for a day or so. Usually the last piece to go on. I then give it some shape and a gradual transition into the underlying braces, simply because I prefer the look. I don't know if it makes a difference doing this other than aesthetics.
-
graham mcdonald
- Blackwood
- Posts: 472
- Joined: Thu Oct 25, 2007 11:57 am
- Location: Canberra
- Contact:
I will admit to never have put a cap on an X brace in my life. It seems to have become one of those 'it has to be done' things, but has anyone seen any hard information telling us that it does in fact make that joint that much stiffer. Remember, one edge of the X braces are firmly glued to the soundboard itself, and the area around the crossing point has to be about the stiffest area of the entire soundboard anyway.
Richard, the points you raise about more efficient brace designs are very valid, and over the years I have seen examples of I beams and braces with holes drilled through them and many other variants.
The great advantage of starting off with a length of 6x12mm (or whatever size is chosen) spruce is that by shaping the cross-section shape and lengthways profile the stiffness can be tailored to that required for that particular spruce (or whatever) soundboard.
cheers
graham
Richard, the points you raise about more efficient brace designs are very valid, and over the years I have seen examples of I beams and braces with holes drilled through them and many other variants.
The great advantage of starting off with a length of 6x12mm (or whatever size is chosen) spruce is that by shaping the cross-section shape and lengthways profile the stiffness can be tailored to that required for that particular spruce (or whatever) soundboard.
cheers
graham
Graham McDonald
http://www.mcdonaldstrings.com
http://www.mcdonaldstrings.com
-
DarwinStrings
- Blackwood
- Posts: 1877
- Joined: Thu Nov 13, 2008 10:27 pm
- Location: Darwin
I agree, this is this most interesting part of the guitar.
I have never added a cap like Graham ( I may in the future though) as I use cloth to reinforce the housing joint, However I believe that a cap has its merits, even if it is just to hold the thing together. The side of the X brace that is not fixed to the soundboard is in tension (also a fair whack of twist) at this point and as end grain won't glue something is needed to stop it opening.
I think when it comes to stiffening infront of the bridge that this may also be beneficial as it could help to stop the phase cancellation that occurs at this part or the soundboard.
Trusses sound good but where do you get those tiny little gang nails.
Jim
I have never added a cap like Graham ( I may in the future though) as I use cloth to reinforce the housing joint, However I believe that a cap has its merits, even if it is just to hold the thing together. The side of the X brace that is not fixed to the soundboard is in tension (also a fair whack of twist) at this point and as end grain won't glue something is needed to stop it opening.
I think when it comes to stiffening infront of the bridge that this may also be beneficial as it could help to stop the phase cancellation that occurs at this part or the soundboard.
Trusses sound good but where do you get those tiny little gang nails.
Jim
Hi Richard,
Fellow ex engineer here.
I cap the X-brace intersection, basically I see it as "free" extra strength and stiffness without any significant weight penalty.
I agree that the rounded or triangular crossection that most work towards is not the most efficient stiffness to weight compared to a simple rectangle or an I beam approach.
It does LOOK better though, and a luthier who left his braces uncarved would probably have his work rejected as crude and unfinished.
Probably an I beam arrangement would be unsuitabe through being too stiff and not allowing the stiffess to be readily modified along the length.
Resist the temptation common to engineers to make things as strong as possible within given parameters, a guitar top needs flexibility to vibrate and produce sound
Fellow ex engineer here.
I cap the X-brace intersection, basically I see it as "free" extra strength and stiffness without any significant weight penalty.
I agree that the rounded or triangular crossection that most work towards is not the most efficient stiffness to weight compared to a simple rectangle or an I beam approach.
It does LOOK better though, and a luthier who left his braces uncarved would probably have his work rejected as crude and unfinished.
Probably an I beam arrangement would be unsuitabe through being too stiff and not allowing the stiffess to be readily modified along the length.
Resist the temptation common to engineers to make things as strong as possible within given parameters, a guitar top needs flexibility to vibrate and produce sound
And there in lay the attraction to lutherie. Not 'just' structure, not 'just' art, not 'only' a sense of when something is about right and knowing when to stop. Rather lutherie is a carefully balanced, a sweet combination of art, form and intuition.jeffhigh wrote: Resist the temptation common to engineers to make things as strong as possible within given parameters, a guitar top needs flexibility to vibrate and produce sound
Hence the depth of addiction, these basic ingredients just like the sweet, salt and acid of our foods can produce and endless array of things that taste right, but never anything that could not taste better if not for a little more of this or maybe a little less of that.
I think when all is said and done, above all else most who involve themselves in this craft, be they scientifically minded, or artistically minded, are at heart optimist. They seem to me to be people who worship potential. They see and hear potential in each new piece of wood, they see potential in each new idea or method and remain optimistic that these things may add to their lutherical spice rack and some how lead to the perfect recipe.
Yep, no two ways about, you guys are all nuts.
Cheers
Kim
Graham, I have done some numbers on the cross brace joint. Using a 16mm by 8mm brace, the brace that has the bottom notched and is continuous at the top is 5 times stiffer than the other one, that is not continuous at the top. This assumes that the glue does not transfer the full tensile or compression load in the brace through the joint, which seems reasonable. If one glues a piece of spruce on the top of the brace, equal to the thickness of the sound board and same width as the brace, then the stiffness is fully restored, hence the logic behind Allen's "patch". This all stands to reason, as you are mirroring the situation of the other brace, with the soundboard bridging the gap of the notch.has anyone seen any hard information telling us that it does in fact make that joint that much stiffer
I suspect that the glue can actually provide a lot of transfer in the beginning but I dont know if I would rely on it long term, so it seems to me to make sense to put the patch on the top and bridge the gap.
Jim, I think that a cloth covering is probably only cosmetic and wouldnt provide the stiffness that a patch will.
Getting back to the question of whether stiffer is better, clearly it aint necessarily so.... And I think several people have made a good point that exotic braces like I beams, "ventilated" ones with holes in and trusses (you must be able to get those little gangnails at Bunnings, surely...) dont give you the same ability to change the stiffness of the top when you are building. So maybe I need to trade a little stiffness for some flexibility!!
As for building a lefty bracing pattern - I am not sure such a beast exists, so am none the wiser as far as thats concerned. That's probalby a good outcome, as reversing or not reversing the bracing pattern is not likely to make a huge difference.
But I WILL be glueing a little patch over the top of my cross brace half-housing joint!!
Thanks again for all the comments and advice.
Cheers
Richard
-
DarwinStrings
- Blackwood
- Posts: 1877
- Joined: Thu Nov 13, 2008 10:27 pm
- Location: Darwin
Yes that is reasonable as glue will not hold end grain. I don't think anyone has come up with a successful end grain glue yet.seeaxe wrote: This assumes that the glue does not transfer the full tensile or compression load in the brace through the joint, which seems reasonable.
I suspect that the glue can actually provide a lot of transfer in the beginning but I dont know if I would rely on it long term, so it seems to me to make sense to put the patch on the top and bridge the gap.
Jim, I think that a cloth covering is probably only cosmetic and wouldnt provide the stiffness that a patch will.
"cosmetic" does that mean that the cotton has no tensile strength at all?
Jim
-
DarwinStrings
- Blackwood
- Posts: 1877
- Joined: Thu Nov 13, 2008 10:27 pm
- Location: Darwin
-
DarwinStrings
- Blackwood
- Posts: 1877
- Joined: Thu Nov 13, 2008 10:27 pm
- Location: Darwin
Who is online
Users browsing this forum: Google and 250 guests