Official Luthiers Forum!

First off, Rob, you are imagining that there is quantifiable knowledge out there that just does not exist. The two sources that would come closest to what you want to find are David Hurd's Left Brain Lutherie, and Roger Siminoff's book. But if they provided formulae that could be readily put to use and really worked, we would all be using them. Read them, by all means, but then you be puzzling over why you didn't get what you had hoped to get. This is normal. If you want to be able to quantify and calculate, go into engineering.

I let go of the idea of controlling variables in my experiments long ago. One simply cannot build in one lifetime a fraction of the guitars that would be needed to treat building as an experiment in science, isolating variables. Some say that you then can't learn anything from a guitar. But that's wrong. If it turns out badly, you may well have no idea why. But if it turns out well, you learned that you have a viable combination. If a bunch of them turn out well, you may learn to trust your intuitions. If not, you might learn to go with standard plans.

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

When all else fails, clean the shop.

Oh please no! No engineering for me. Math and me get along as well as epoxy on shellac, but i do like to goof around.

It sure would be nice to be able to test a configuration, then pop off a brace and try something different.

I wonder if i could do something like shellac the inside of a top and the braces, then use a weakened formula of HHG that bonds enough to test with and maybe pop off under gentle heat (heat lamp?).

Maybe build a shop mule, why not put some of the top bracing on the outside?

Pat

_________________
formerly known around here as burbank
_________________

http://www.patfosterguitars.com

I think Howard nailed it , there is no real specific information out here on voicing,I started out the same way as Rob , thinkiing there was a magic formula, a certain hidden secret, or a geni in a bottle.there might be ,experience enlightens . Jody

Epoxy does go well with shellac provided that the shellac is on first.

I can be an enjunear after all!

I agree with everything that Al mentions regarding the ability of FEA to predict exactly how a guitar will sound. I agree it is impossible because of the fact that all pieces of wood are different. You have to make the most of what you have to work with (is there a greater lesson here...?).

I do think that FEA is applicable when comparing one design to another. Different builders with different philosophies would have different tonal goals....and each of us could come up with our own optimized recipe...

_________________
http://www.PeakeGuitars.com

alright , whats a FEA ? thanks Jody

if you are doing Chladni patterns, you can put braces on with double stick carpet tape (the thin plastic kind not the the fiber reinforced stuff) and get nearly the same frequency and pattern as if you had glued them. It is fairly easy then , to pop the brace off and move it. Note that taping braces on will kill tap tones, but the Chladni patterns will still be true (although it will take more energy to drive them).

Mark

Mark, I'll bet that has a few people saying "hhmmmmmmm"
I've also heard, although not mentioned here, about placing "sticky tack" (putty like stuff) at spots on braces, which acts as a mass increase (mimicing stiffness decrease) in an effort to improve nodal patterns. If it works, shave braces there! (It appealled to me, as non-destructive "test" before cutting anything.)
Hoped someone would expand on it.

_________________
Dave
Milton, ON

FEA is Finite Element Modeling or Method. It is a numerical technique for finding approximate solutions of partial differential equations as well as of integral equations. I have a degree in Mechanical Engineering and a little experience in FEA. I would agree with the post earlier that FEA is a very powerful tool and can be used to tell you something between different designs. Is it going to tell you what the guitar will sound like, never. FEA is a theoretical model, there are so many variables the make each guitar sound different that cannot be modeled with FEA. To that extent, you would probably need a PHD to create and run an accurate FEA model of a guitar. My FEA professor always said a fool with a tool is still a fool. If you do not have the background in Differential Equations, Mechanical Vibrations and Finite Element Modeling, you will not be able to create an accurate FEA analysis. There are so many boundary conditions that go in to a basic FEA model and when it is all said and done FEA does not always give you the exact results that a test will. That's my two cents for what it's worth. A book that I use regularly for FEA is Introduction To Finite Elements in Engineering by Tirupathi R. Chandrupatli and Asok D. Belegundu. But like I said if you don't have the background this book is worthless. Guitar building is much more an art than a science.
Travis

Travis

Unfortunately for me, Chladni patterns and such are far above my poor ability to comprehend. In my feeble mind, it would seem that once the top is glued down, it would change how the wood responds. And would the response be different depending on the type and thickness of the wood to which it is glued (kerfed lining, sides, neck and tail blocks, etc.)? Would it change based on changes in temperature, moisture content, etc.?
My lack of knowledge and experience in this area is legendary, but perhaps as I gain experience, I’ll be able to understand the science.

All the big companies get some standard 'tone' out of a guitar, as well as builders....its not that much of a mystery...there is alot of consistency in mass building...

_________________
Here is what a Parlor Guitar is for!
http://www.youtube.com/watch?v=oEa8PkjO6_I

Alan (not me) wrote:
"In my feeble mind, it would seem that once the top is glued down, it would change how the wood responds."

You've put your finger on the biggest theoretical problem with the whole enterpise. While it is certainly true that the structure of the top, and therefore the way it vibrates before you glue it down, has some bearing on the sound of the guitar, it is also beyond argument that getting from here to there in any rigorous mathematical way is no piece of cake. It takes a pretty hefty FEA model to get even close, and that assumes that you're working with a uniform piece of wood with known properties. Good luck finding one of those!

What guys like Mark and I do is just a higher tech version of the good old 'cut and try'. We look at the pretty patterns and see if any of them correlate at all with the way the guitar sounds when we get it done. If they do, then we have a better idea of what to do, or not do, next time. As you go along you get better at it. Some of the correlations suggest possible 'reasons' why things might work the way they do, such as the idea that 'closed' ring-type modes might work better because there are closed rings, but no open ones, on the completed instrument. The thing is doing what it 'has' to do because it 'wants' to, so it works better: hardly the stuff of a PhD thesis in physics. As with any system, tap tones or whatever, it takes some time to figure out what's happening.

The big advantage of this system is that I can send Mark a set of patterns and frequencies by e-mail, and he'll know just what I did, or vice versa. It's awfully hard to communicate what a plate should 'feel' like, or what 'dead' means to you, unless you can actually share some shop time. When we get together at a show we can sit down with our notebooks and learn a lot without having to make all of those guitars by ourselves. Pretty neat.

After all my reading on this forum my limited understanding leads me in my quest for good bass response, it mass seems to be the killer of bass so make the braces 1/4" wide instead of 5/16", keep to no higher than 9/16" and the top a little thinner than normal with only one finger brace on an OM seems to be the way forward in my thinking. I've read that a thinner back can help but my current thinking is that your body would damp that anyway so keep it thick 0.085"-0.100" for strucural reasons and hope it'll reflect a little.

Then use tapered/parabolic bracing with the ends thinned right back ala the John Mayes voicing DVD to get good note seperation on smaller guitars and use scalloped braces on Dreadnought and Jumbo bodys.

This is my current thinking please take pot shots to correct it

_________________
Formerly JJH

I learn more from my mistakes than my successes

Ok...here's my 2 cents..then it's back out to ye olde woodshed....as before I'm speaking more from engineering knowledge than I am from guitar building experience (some of you have way more than I'll ever have, I'm sure!).

~~~~~~~~
Mass is not necessarily a killer of bass. It only kills it if you increase the stiffness quicker than you increase the mass. Think of a Low E string on a guitar....they're thicker than the treble strings. They are overwound because this adds mass without adding tension/stiffness (the core is similar in diameter to the plain strings).

All vibration can be related in some way to a string vibrating. Increase the stiffness (tune it up) and the pitch goes up. Decrease it and the pitch goes down. Increase the mass at the same stiffness and the pitch goes down...and vice-versa.

for better than average bass, you want a top that has less stiffness and/or more mass than your average guitar. Mass can be an overall tone killer...the strings can only pump so much energy into the top, and if it is too heavy they won't be able to move it much. So in practical terms, make it floppier for more bass. Too floppy and it will fall apart. You get to draw the line in the sand where you think this point exists...!

_________________
http://www.PeakeGuitars.com

I'm not saying all tops must be paper thin, just I over sanded my top and it seems to of worked out for the best not strung yet so can't be 100% sure it ended up between 0.090" and 0.110" rather than the 0.125" I missed. The top I've just bought for my second top is sitka rather than central european spruce and it very floppy so will aim for 0.125" then tweak the bracing. It's an OM size so I may only use one finger brace each side ala John Mayes dvd and possibly add the brace behind the bridge patch to keep the centre stiff and edges loose. Only thoughts nothing firm

_________________
Formerly JJH

I learn more from my mistakes than my successes

'Tuning' the back to work with the top is an effective way to get more bass. It works well even though your body does damp the back vibrations some. IMO the 'right' frequency for the back will be a semitone higher than the 'main top' tap tone (plug the soundhole to hear these). This gives a good combination of effectiveness and low risk of a 'wolf' note. It also has the big advantage that you don't have to make the top weaker, and it also seems to preserve the clarity of the high treble notes.

Are you measuring the 'main top' tap tone with the guitar strung or un-strung?

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Alan, I've heard about the advantages of having the top and back's main resonance (if that's a correct term to use here) should optimally be a semitone apart.

Some have said that the top will drop a semitone once the bridge is added, so should I be shooting for having the top and back roughly the same after bracing but before assembly?

Thanks.

-Clint

_________________
-Clint Beacock

The reason I ask is this: I think you're after having the top and back main tap tones close but not perfectly the same. I would think that putting the strings on would raise the top's main tap tone. So I guess I'm asking are you shooting for that half step difference before stringing i.e. are you shooting for a top and back frequency that are closer than a half step when strung or are you shooting for that half step difference once the instrument is strung up or am I missing something entirely?

Does the back's main tap tone go up when the guitar's strung?

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com

Alan the only way I would have to do this is to hum the notes into a tuner would that be accurate enough?

_________________
Formerly JJH

I learn more from my mistakes than my successes

I find that the 'main top' tap tone drops about 1/2 semitone on the average when I glue on the bridge, and it drops another 1/2 semitone in the first month of playing. Thus I usually shoot for having the top and back tap tones pretty much the same when I get the body together. I have not found that the 'main top' mode pitch drops simply from putting the strings on: the tension seems to have no effect on that mode on my guitars. YMMV.

Humming the note into a tuner should work. If you can record sounds on your computer there are free programs that will do a 'Fast Fourier Transform' and show you the resonant peaks. You have to be careful when interpreting this information, but it can give you some exact numbers quickly. I use an ancient DOS program called 'FFT4WAV3': it's quirky but it does what it does quickly. 'Wavesurfer', which can be downloaded free from:
http://www.speech.kth.se/software/
is more sophisticated, runs under all the versions of Windows I've tried it on, and doesn't require a seperate program for recording. I'm not a Mac guy, so I can't tell you what might be out there for that platform.

This is a little off the immediate discussion on resonance and matching etc.., but definitely related to tuning the top.
Dana Bourgeois has published several articles in American Lutherie over the years on voicing guitar tops and are certainly worth while reading IMO. Dana gave a very good presentation at this years GAL conference covering the subject outlined in his paper. I have been using his method for a while now and find it pretty intuitive while being coupled with some objective goals.
You might check it out.
Here's the link: http://www.pantheonguitars.com/
Select "Tonewoods & Voicing" off of the left hand menu.
I don't believe I've seen this referenced here, at least not very often, it's good stuff. At least food for thought.

_________________
Jim Watts
http://jameswattsguitars.com

That's very interesting. I haven't actually done any experiments on that but I totally thought that stringing it up would have a change on the top tone. I thought though that it would raise the tap, not drop it though. I'm building a guit right now and will try and keep track of the tap tones through the process.

I also wonder if the no-change thing will be different on a tailpiece type of instrument like the bandura I normally build.

I've already recorded the top tap tone before gluing to the top. I can keep a sound log of all the taps as I build and post them if anyones interested. If anyone has decent mac software for the FFT thing I can run it through there as well and post.

_________________
http://www.birkonium.com CNC Products for Luthiers
http://banduramaker.blogspot.com