Copying a guitar without removing the top: Is it possible?

[Adrian Lopez]

I have a flamenco guitar from which I'd like to draw some plans, but opening up the guitar so I can measure things is simply not an option. Has anybody come up with a means of accurately locating and measuring a guitar's struts and other internal components without having to remove the top? Is it just too impractical?

[Mark Swanson]

Hello Adrian and welcome to the MIMForum. Well, you can place a mirror inside, and that helps. You can also place a bright light in the guitar, and dim the lights in the room and you'll see the bracing pattern shining through the top. That should allow you to take the measurements you need.

[Alan Carruth]

Remember that the copy will only sound and play like the original if you've got 'the same' wood. It's really hard to get that data from a completed guitar. OTOH, I've made a reasonable 'tonal copy' by duplicating the main low order resonant modes of the completed instrument. I started with the mode and spectrum info on a particular '43 Martin OM, and managed to get acceptably close to the sound in the copy. Of course, maybe it's just that I got lucky once, but I like to think it was 'skill'...

Alan Carruth / Luthier

[Tom Harper]

I've had pretty good luck taping a piece of paper to the top and using a pair of tiny tiny supermagnets like Radio Shack sells. Stick one on the inside, holding it in place with one on the outside resting on the surface of the top.
Move the pair until the inside one runs into the edge of a brace, then fiddle with the top one until it feels centered. Mark the location on the paper. Repeat process until you have enough marks to make lines where the braces are. Don't press real hard when marking locations... Now that I think about it, I guess I should say that I think I've had good luck because I've never pulled a back off to check my work. I second Mark's suggestion for a mirror to make sure that you don't get surprised and get an idea where to roam around.

You can use low melt plastic to measure brace height and width by pressing a glob oto a brace and letting it coo. Once it cools, you can pull it off and measure the notch created by the brace. So far, its come off cleanly for me once it has totally cooled. I think its sold as "Friendly Plastic". StewMac used to sell it and perhaps still does.

[David Malicky]

Al has great advice, as usual, on copying the main resonant modes -- copying geometry isn't nearly enough.

Here's the light technique in action (called a "Henkogram"):
http://theunofficialmartinguitarforum.y ... GihcK7X9Bk

A few other ways to measure brace heights:
- Press a loosely crumpled ball of aluminum foil onto the brace. Or poster putty.
- Make some 1" long sticks, with heights from 1/4" to 9/16" at every 1/16". Then by feel, find the one that matches the brace.
- Hold a webcam (plus light) inside to study cross-sections from various angle.

You'll also need the top thickness distribution -- there's a magnetic thickness gauge for that.

The lining style and side thickness (mass and stiffness) can also be important.

I would copy the Henkogram and make the braces ~1/16" taller than the original. Measure the 'main air', 'main top', and 'main back' modes. Shave until within a Hz on the 'main air' and a few Hz on the others.

[Steve Senseney]

I made some gauges for measuring brace height (and thickness) out of plastic. They work, and are not expensive. You use them by feeling which slot the brace slips under.

[Alain Bieber]

We all copy existing models, to a large extent. Professional builders try to reproduce, from one guitar to the other, the sound for which they are known, and this both for Flamencas and classical guitars. But focusing on bars and struts is not enough, this is what I firmly believe. You have to reproduce the more you can. Any detail counts. As Alan as underlined, you even have to reproduce, and the most you can hope is a so and so reproduction, the kind of woods involved. Do not try to reproduce a white Flamenca with something other than spruce and cypress for instance.
The problem of thicknesses is important, more than strutting alone, I am firmly believing that too. So you need an evolved thickness gage, an Hacklinger or some other ones, going from simple DIY solutions to more industrial solutions. A pair of recent American Lutherie numbers give newer and cheaper solutions to that long lasting problem.
Some features such as internal, at first glance secondary, pieces of the structure are very important, to give an example, I would avoid changing the internal linings or blocks (peones) as on your model.
Then remains a "luck factor". Last advice; it seems good to me to build with an eye on the total weight of the instrument you are attempting to reproduce. It is not good news when you drift apart.. and this is easy.
Good luck. One thing is sure you have better results attempting to reproduce, than inventing one or several "new" features, even worse when several at the same time. I learned that the long and painful way. <g>

[Adrian Lopez]

Lots of great advice. I had some rather impractical approaches in mind, but this goes to show that simple is best.

My goal is to draw a plan of a Gerundino Fernandez spruce/cypress flamenco guitar to share with those who'd like to follow the pattern but don't have access to an example of the maker's work. I think I'll practice on my mass-produced guitar before attempting to copy the Gerundino.

Thank you all for your help.

[Adrian Lopez]

I just found out about profile gauges. Looks like a good way to measure and reproduce the various struts that make up a guitar's bracing.

The ones made from plastic are probably safer for the instrument than the metal ones.

[Chuck Tweedy]

If you can get them into the guitar.
The profile gauges that I have (or have seen) are all too big to fit in a guitar body.

[Michael Lewis]

Profile (contour) gauges are a simple concept, and could be made any size you like.

[Barry Daniels]

I wouldn't get too hung up about the profile shape of the braces. More important is their position, height and width.

[Alan Carruth]

I'd even say that the most important thing to copy about the braces is the layout, and width. Start with them a bit tall, and trim them down so that they work with the top you've got. It's pretty unlikely that you're going to be so lucky as to get wood for the top and braces that's exactly like the original, and there's no way to know for sure in advance. If you're trying to copy the sound, you have to make the new top vibrate like the old one, and if the wood's not as stiff (say), the new to might need to be thicker or have taller braces. Of course, you will also need to know the 'tap tones' of the original, at least, so that you can copy them.

Alan Carruth / Luthier

[Ryan Mazzocco]

How 'bout a magnetic resonance image? It's very clear and precise. just make sure you remove ALL metal hardware before running it!

[Alain Bieber]

X-ray tomography has been often used by very exacting measurers, generally Museum guys. To give an example an interesting written introduction to this technique, in English, is found in the Journal devoted to violin research in the US. A paper by Borman and Stoel (summer 2009), if my memory still works.
As first reaction I have doubts about the possibility to use IRM for rather uniformly dry wood items. But who knows?
The problem is to have access to such big tools, and you will need some help to start with them. Except if you are already using them of course.
"Borman violins" is an useful entry point.

[Chuck Tweedy]

How about magnetic-coupling imaging??

You get two small neodymium magnets and couple them across the top. Drag the top one around until it hits a brace. Mark its location to generate the image.

[Michael Lewis]

Pad the outside magnet or you might mark the finish of the top. Wrap the magnet with a couple layers of masking tape.

[Ron Belanger]

What Chuck and Michael said.

[Randy Roberts]

Excellent use of your engineering acumen Chuck <g>, ... is that what they used to call ferro electron polarization modeling?

If you place a sheet of paper on the top/back (you will need to cut out a spqce for the bridge if it's still on the guitar), and do what Chuck and Michael are suggesting, you've got your whole bracing scheme staring right at you. Just mark with a pencil on the paper where the edges of the braces are at their ends and junctions, as indicated by the magnets, and connect the points. lightly trace the edges of the guitar and soundhole with pencil and you've got your template.

Just remember you will need to make a mirror image copy of this template when you lay out your bracing, because when you go to brace your top, you will be looking at the braces from the inside out, not outside in as you were when you marked where the magnets were..