Official Luthiers Forum!

Yes Hesh, this has been a very illuminating discussion. Downright blinding in fact. If any of you have any pics of what you consider to be a successful parabolic bracing effort, please share them. They would be invaluable to those of us like me who are trying to visualize these concepts. I was itching to start making some braces this morning. Now I’m just scratching my head <grin>.

_________________
Doug Mills
Chicago, Illinois

Thanks for the nice words, Hesh.

For a number of years, I thought that the rings you mentioned were centered in the bridge, but when I wroted my first article for "American Lutherie", Fall, 1996, I tried to map out the rings on my ol' D-28, and discovered that the center was actually at the cross point of the X brace. It kinda makes sense, in that this can be seen as acting as a natural fulcrum for the top. X bracing forces the top to work as a whole.

In contrast, classical guitars have a transverse brace below the soundhole, which forces the lower bout to act independently of the upper, and the center of its rings are actually close to the center of the bridge. I have only re-voiced two classical guitars, but the principles apply with them, too. But it's dang hard to get in there. Frank Ford says he can get his whole forearm and elbow inside a steel string. I'm jealous, and wear out skin from time to time getting those ends just right.

To clarify, another point is that these rings' diameters are inversely proportional to the input frequency--the higher the note, the smaller the ring, and vice versa. This is why I have found that the last inch or two of the X braces' legs are so important in delivering low end energy--they are farthest from the cross point.

There is a fairly detailed discussion of this theory on my web site, so I'll leave it at that here.

As far as sharing what I think I know, there are already too many guitars out there needing help for me to ever re-voice even a small portion of them. I don't need any new ones being built...

By way of penance for my earlier shameless self-promotion, I offer advise via email to those interested. Just don't send many photos at once, as I have slow speed internet at the end of an 8 mile extension cord up into the Gila forest.

If I could figure out how to download pictures, I have one or two that might give a few helpful ideas. I'm new at the computer stuff.

Scott

Scott, if my post seemed critical, it was. That wasn't my intention. I just
framed the question poorly. I'm very grateful to learn more about your
approach and philosophy to this. So many come at it from their own
experienced-based ways and for guys like me, threads like this almost
become text books to learn from. You guys take off in new territory and
then let us tag along after you did the work. It's something I thoroughly
appreciate AND admire.

Hi James,

I took no offense whatsoever. I didn't believe there could be such subtleties, but they have made themselves known time and time again.

I remember one of my earlief efforts was on a '65 J-50 Gibson. In 1988, I chased dead strings all over the top, not knowing why these dead spots I could feel over the belly braces and X legs could be influenced by the tiny imperfections I felt on the braces underneath. I sanded and moved the dead spots around a number of times. A dead D string, then a dead A, then, G, and back and forth. I spent many moments with that horrible adrenolin rush of thinking I had ruined the guitar, not knowing what was really going on. At least, I owned it. When all the braces felt "perfect", I had a balanced guitar. I had not had such troubles on my earlier efforts, which I began in 1983. Beginner's luck, I suppose.

A year later, playing and thinking about a Dobro I had recently purchased, in a blinding flash, the "Sound is Round" model came into consciousness full blown. Think of the guitar as a speaker cone, with many subtleties. After all, the Dobro is just a speaker cone energized by the strings directly. Look Ma! No Electrons!!

Scott

[QUOTE=Scott van Linge] I can't even come up with convincing bs as to why.[/QUOTE]
Ha! Spoken like a truely enlightened craftsman!!!    If it works, it works. Somebody else can figure out why, if he's got the time. After all, does the speed of light have to explain itself?

Thanks, Scott (and others), for taking the time to share your knowledge. It's an OLF tradition.

Thanks Scott, I always find your work interesting, and appreciate your input here.

Greg

_________________
Gwaltney Guitars

Thanks for posting those Hesh, and thanks Scott for letting us see them and sharing your extensive knowledge.

Colin

_________________
I don't believe in anything, I simply make use of a set of reasonable working hypotheses.

Thanks for the assist, Hesh, and for the nice words, Greg and Carlton.

This is a look at my latest OM, with a bear claw sitka top, koa back and sides. I'm trying to keep the bridge from bowing up more on the bass side with the transverse brace. I think it should be decoupled from the bridgeplate to lower that overall inertial mass, as the photo shows.   

The diagonal belly brace was then able to be a true parabola, because the upper brace evenly distributes the strings' force. I put two diagonal belly braces on bigger guitars, in addition to the transverse brace. You might also notice that it's braced backwards--as if it were a left handed guitar. In working on an earlier guitar, I ended up with a situation wherein if I put my arm over the top while playing, the volume was reduced noticeably. This spot is where the diagonal brace ended at the kerfing, and I noticed that holding the other side did not dampen the volume. So, I removed the brace and put in one going the other way. Problem solved. When braces get close to perfect, they sometimes get fussy, and behave strangely.

I once read about someone converting a right handed guitar to a left handed one, and that the sound did not change. And of course, if you believe in my "Sound is Round" model, it obviously doesn't make any difference which way the diagonal(s) go. They just need to support the top's belly, and help distribute energy from the X throughout the lower bout.

Notice the side braces, with the upper one skewed, and the lower one, a true parabola. I have spent many hours modifying these braces since assembly.   I later shaped the legs of the X concave, starting from the cross point, not your usual scallop (described in more detail earlier) and it is quite fun, alive, loud, dynamic, and balanced. But it took this reshaping for the alive feeling to emerge, along, especially, with the bottom end.

I think I'll be making the bridge plate smaller in the future, after reading and contributing to that recent discussion. (And noticing how flat the top is below the bridge.) The transverse brace was lowered in height as well, to get the wood above it to vibrate.

You can also see that, instead of fabric, I used mahogany pieces to strengthen the sides from top to back, as well as transfer the energy between them. A lot of work, but I think this may have contributed to the success of this guitar's sound. Couldn't say for sure, but when I tap over these pieces, the sound is a deeper thunk than in between them. That's gotta count for something.

Scott

Wow! Lots of meat here! Really appreciate all the shared info.

Hope this isn't a hijack, but since we're talking about bracing....

One thought that has been floating around my head the last few months is that we see the height of braces vary due to shaping, but not the width. If the effect on stiffness in bracing is so much greater with variations in height than in width, why don't we see the bracing change width as the height changes? Seems like we could reduce mass in areas where the braces height is low, especially at the brace ends. Granted, this would be tough to accomplish once the brace is glued on, but if we have an idea of how the brace will be shaped along its length, we could taper the width before the braces are glued. It would certainly add complexity to the process, though. Scott, have you looked into this possibility? Anyone else with any thoughts?

_________________
now known around here as Pat Foster
_________________
http://www.patfosterguitars.com

Thanks guys, I have copied this entire thread and put in my Bracing file. I have to go back and study things a little slower, trying to absorb all this info off the thread can easily make ones head explode. Great stuff!

Mike

Just so we don't start thinking there is only one to do this I thought I should tell you about a 1936 D-28 a friend of mine has. Top is caved in around the pick guard, several cracks in the top. The X brace is heavely scalloped, with no cap over the joint. The braces are anything but smooth. It needs a neck reset so the saddle is very low. The bridge plate is so chewed up he has to put a second ball end over the strings. The tuners are worn and feel a little loose.
But the sound------It will make a grown man cry. If I ever make a guitar that sounds that good I think I will retire. I'm not saying I would try to duplicate this, just that a lot of great sounding guitars have been made using many different methods. Peace, Paul

Paul,

Next time you get around your friend's D-28, take a look at the back braces and the side braces on the top. I'll bet you will see parabolic shapes. I once saw a 1912 small bodied Martin. These braces on it were parabolic. The X was scalloped, as well as the upper belly brace. But the lower one was half and half--a peak on the end toward the X, but half a parabola towards the kerfing. According to the "Sound is Round" model, that lowest peak dampens the low E, and this guitar had amazing bottom end.

And Alan, I agree that the braces must be adjusted to the top to work together. That's why I spend days and weeks tweaking the braces after I've assembled a guitar.

I was looking at the picture of my OM top, and noticed that the diagonal belly brace's outside end is about 1/8" high. But before I assembled the box, I decided to cut and feather the end down to zero. I don't want to steer anyone wrong here.

I've done some other experiments, and have come to the conclusion that the X brace needs to have height where it buries into the kerfing, but the side and belly braces need to be down to zero to keep from dampening/stiffening the top around its perimeter. This is in keeping with most factory guitars, including the OM-28EC currently under the knife at Parabolic Braceworks. But the ends need to flow in smoothly with the rest of the brace, and not have the scallop at the end that almost all builders use.

I re-voiced a '54 OOO-18 last April, and was surprised to see the side and belly braces cut down (drum sanded at the factory) to close to nothing for the last one and half or two inches before the kerfing. Not too good an idea, in my opinion, from the aspect of the braces' task of delivering energy throughout the soundboard.

To make matters worse, someone had already been into the braces earlier in its life. Perhaps thinking that the legs of the X should receive the same treatment, he cut down the last two inches of the legs to zero, very crudely with a chisel. Splinters of the end were sticking out of the kerfing. This resulted in the low E being much weaker than is usual, since, as I said earlier, I believe that the ends of the X are where the lowest frequencies are delivered to the top and sides.

I was able to fashion spruce pieces to restore the original ends of the X, and then shaped them as I've described in an earlier post. No easy task, but it brought the return of the low E. I also added a piece to the upper belly brace's end, so that the shape could flow smoothly to the kerfing, but the area over that end did not start to vibrate until I cut it down to zero at the kerfing, reaffirming my decision to do this to the OM top before assembly. I couldn't reach the end of the bottom belly brace well enough to add anything. But the side braces all got a supplement, for to make them parabolic would have left nothing there, and what was there, to begin with, was dampening the high end.

The owner was beyond belief (as was I--it was incredible. If I could only build one that sounded that good...) and left his HD-28 for work. I mentioned that one's problems earlier. What I didn't mention is the side braces. Someone at the factory had taken time to rather crudely hand sand and shape them from the peaks that were stock at that time (1989). But only one was close to a parabola, and the other three were in between. Strange that such a company seems to have forgotten what they once knew.

Scott