William Cumpiano's 
String Instrument



Tap tuning; guitar graphite, etc.
© William R. Cumpiano 1999, All Rights Reserved

In the early nineties, my shop was in Hadley, Massachusetts. I discovered that a customer of mine, besides being an avid guitarist, was also a senior production engineer for a large sports equipment manufacturer with a factory in our area. Richard Janes was in charge of the engineering design of their line of consumer- and professional-grade tennis racquets.

Our conversations invariably came around to the many conceptual similarities between tennis racquets and guitars. Although not immediately obvious, the similarities are indeed profound. At least in their basic form, they are both wooden artifacts under constant and considerable string tension, which must be optimized in architecture to be minimal in structure; and whose vibrations and resonant modes must be understood and controlled by their makers.

This confluence of interests and preoccupations kept us absorbed in conversation for months, and eventually Rich let on that he had become familiar with the guitar manufacturing scene and was surprised how technilogically backward it was, compared to most modern leisure industries. The tennis racquet industry was always on the cutting edge of the latest spinoffs from military and aerospace technology, particularly in the application of new polymer fibers and fiber composites--whereas the guitar industry was still at a level of World War II -era technology: The polyvinyl chloride and aliphatic resin glues we use are still only one step away from ancient hide glues; in 1948, nylon strings replaced ancient gut. And nothing much has changed. Although some manufacturers are now using lasers to select and verify the consistency of musical strings, and factories (and even smaller builders) are using CAD/CAM equipment to improve production level and accuracy, and pickup manufacturers are miniaturizing electronic sound processing components that now can fit on and in guitars, essentially everything else remains at an Industrial Revolution-level in the guitar industry. He felt strongly that many other leisure industries, especially the tennis racquet industry he was so familiar with, were light years ahead of our industry in exploiting current and ubiquitous technologies for the optimization of essentially similar structures.

For one thing, tennis racquet makers had long since abandoned wood in favor a wide variety of composites of different fibers--and combination of fibers. Thus the stiffness, weight, resilience, elasticity of the structure could be carefully tailored to all the different needs of the market. Gut racquet stringing was replaced by nylon for essentially the same reason that it was replaced on guitars (tension instability in varying moisture conditions)--but while the tennis industries went on to even more favorable polymers: boron, kevlar, tynex (which retained many of the favorable qualities of gut while superceding the bad qualities of nylon), the guitar industry is still mired in the outmoded nylon string, which although inferior in tone to gut was enshrined nonetheless as the industry standard for decades--to this day--simply because of its moisture stability.
The reasons for the slowness of the industry to change and its resistance to improvements in technology are another matter--something which Rich and I would eventually have to deal with head on.
Rich was convinced that soundboard spruce could easily be replaced by compression-molded fiber laminates (which he was using every day in his own industry), and that all the bad qualities of spruce could be lessened or eliminated by so doing: for one thing, it's tendency to crack and "creep" (plastic deformation) under load. Its natural tendency to widen and narrow in dimension as the relative humidity correspondingly increased or decrease. The difficulty in accurately predicting the acoustic response and structural quality of any given sample of spruce--from its visual characteristics, although a quasi-religious lore had for centuries grown around it related to predicting this qualities: by counting grain lines, by manipulation, by tapping and listening, by "candling" (inspection of hidden structures under a bright light). Rich was convinced that all these qualities could be tailored and predicted by the design of the particular lamination of each laminate soundboard. And the acoustical qualities? Spruce is essentially carbon fiber in a lignin (read: natural epoxy ) matrix. Its acoustical qualities come from its low weight, high stiffness and low sound damping. Just like compression molded carbon-fiber laminate.
So after all this commiserating, he shows up with a 1/16" thick sheet of compression-molded carbon fiber laminate, laid up in a way to make it stiffer the long dimension than the short (just like spruce). I took the spruce top off a bad-sounding factory guitar (an old Gibson LG-0 with a damaged top) , replaced it with the sample, and it sounded…grrrrrr—e—attt!!!.

That really motivated us. Two years later we received a bonafide US patent. Since then I've built several guitars, steel string and classic, each one with lighter and lighter bracings and thinner and thinner laminations. I sold one for $4200, another $3800. I think my market accepted the concept. At shows, people marveled at its high-tech appearance ("it looks like black marble tweed!"). Others commented favorably at its surprising sustain and clear sound that was totally sprucean. The next step was to show it to the movers and shakers of the guitar industry.
We're still trying to sell it to the guitar industry, but with little luck so far. The sales people love it. They love the looks, the whole idea of it. But the manufacturing people don't want to deal with it. Fender turned us down, Martin (who developed their own graphite composite soundboard patent—and put cardboard inside it--and then didn't develop it) turned us down too. Taylor tried it, never got back to us. We're still developing it, and I'm making custom instruments on special order with the material that I get from Rich. We're still working on making it a soundboard for the next century. 

A list member sent me the following thread on another list and asked me to comment. The names have been deleted to protect the easily embarrased.
> >> There are references in the back to various sources of information but hereis what Cumpiano's discussion regarding tap tones, etc. is limited to:
> >> "...final graduation of the braces (Fig. 7-20) is a step that invokes
> >> and ultimately tests the luthier's mastery of the medium. As such, its
> >> successful execution cannot be reduced to a mechanical formula, but rather
> >> must be learned over the course of building many instruments. We strive for
> >> a familiar combination of resilience and tap response that is difficult, if
> >> not impossible, to describe in words alone."
> >Come on, surely you understand Bill's point better than this! Can
> >you please write us a description of the taste of a tomato? If you can't
> >articulate it, is it because your taste is Zen-like, or is it just non-verbal?
> >Tell us how to hear the nuance you like in your favorite music! Can't do
> >it? Do tell! In Zen, there is the notion of the gateless gate. The gate
> >is not a barrier. It is a portal. And yet it cannot be passed unless one
> >has resolved some matter that can only be lived, not taught. Sometimes
> >descriptions of the experience are more confusing than a simple
> >invitation to go through the experience. The best description of a
> >tomato is, "Go eat a tomato."
>Thanks for the background on Bill, it does add depth to
>the discussion, and gives me perspective on your impatience with
>him. But, to grind a metaphor further, isn't Bill saying, "We've
>come to the moment in making tomato paste where you must tune
>it by adding just the right tomato. It takes a little experience
>to choose the right one, but on the other hand, that's _all_ it
>takes! Go out and taste tomatoes till you have perfected your
>paste. Meanwhile, take up any tomato you think is good now, and
>make a journeyman tomato paste.
>I've read Alan Carruth talk about the process of rubbing a finger
>across a top, listening and feeling for just the right tone and drag
>on his fingertip. What he says is clear as a bell, but I still won't
>know what he is precisely trying to teach unless I stand beside him
>and hear and feel under his direct guidance, or go spend a lot of time
>putting his instructions to practice. He can't just talk me through
>an experience which is essentially non-verbal.

>Please oh please tell us the shape, weight, density, young's modulus,
>grain spacing, color, species, specific gravity, shear strength, compression
>ratio, timing advance, and mpg of your magic guitar top!
> No, no, no, everyone knows a tomato tastes just like a Fire-Engine
> Red Telecaster. But this still doesn't tell how to tune tops.
> Yet, with absolutely no experience at this, I will now reveal my
> own secret for top tuning (or is it tap tuning?). Make guitars until
> you end up with one that sounds really good. Remove the top, and then
> use that top as a reference when making other tops (HOW to use it as a
> reference? Sorry, I can't give away ALL my secrets). It's very
> scientific and reproducible, and can't fail.

> So, does anyone have any .wav files of a good top being tapped, so
> we know what a good tap tone sound like? Or is it only available on a
> CD with 30 seconds of tapping at cdnow.com for 'only' $29.95? What's
> that saying, a .wav file is worth a thousand words?>
> >> I interpret this to mean that Cumpiano feels that everyone needs to pay
> >> their dues in terms of making many, many instruments so that when that
> >> Zen-like experience of knowing how to make a great sounding guitar arrives
> >> it will have been worth the trouble. Until that time, you're more or less
> >> making a nice box with a stick on it.

This is my point fellas. It’s actually far simpler than you’re all making it out to be! Not easier. Simpler. After you make half a dozen guitars, or maybe a dozen guitars—you just….start to get the IDEA of it. It’s not ZEN. It’ll happen to you, to anybody, with a mind to do it. I’ve seen it happen to many of my early students who stuck with it. You just have to keep an open, receptive mind, and not take it all like some kind of personal TEST. After a couple of dozen guitars, the good ones start coming more often and every once in a while an extraordinary one pops out. If you haven’t gone broke or nuts by then and you get to number fifty or seventy five, "extraordinary" starts to happen more often. After a hundred, you forget what it was you were doing when they were coming out mediocre, you just know that you’re doing things differently now. It’s kind of seamless. The only reality is that you’ve made more guitars than you can remember—and you don’t make dogs any more. I like to tell my students to start with a "mental model" and then refine that model progressively with the experiences you get from making each guitar. Everything you find out, read, are told, discover on your own using your senses, all your past mistakes--serve to refine this personal, imaginary model which gets slowly perfected in your mind. This model then serves as your guide to the myriad decisions that you have to make with insufficient information or with the lack of "true" knowledge during each actual guitarmaking process.
Now that’s just one way of getting from amateur to master. There may be a quicker, less painful way. I honestly don’t know. I only know that I was not preoccupied with making a master instrument: just with making an instrument the best way I knew how at each point in time. I was confident that for a person of modest resources like myself (both personal AND financial), mastery would only come with patience, persistence. with tenacity. Maybe it can also come from finding out, ahead of time, the "shape, weight, density, young's modulus, grain spacing, color, species, specific gravity, shear strength, compression ratio, timing advance, and mpg of the magic guitar top." That information was simply not available to me. So, that’s not the way I did it. Some of that information is available now, but alas!, it still won’t tell me when to stop carving the braces. I now stop carving when they…look right. And when the top…sounds like the one on the last good guitar I made. That’s what I meant when I said that you can be taught to build a guitar, even a really good guitar. But consistent mastery is something that can’t be taught, but can be learned, given time. If you haven’t the time--but still want to achieve mastery, you’re in the wrong business.


My email friend Bruce Lee (NOT the late, great martial arts expert!) and I had a recent conversation that would make a perfect addendum to the above commentary on tap tones. I thought you all would be pleased to listen in on it. He's given me permission to send it on to you.

> Are you looking for a tap tone (and overtones) in a range of notes, not one
> particular note? That is, the typical bass voice range starts at one octave
> and a 6th below middle C (E). Sub-bass would be below this note.
> Concidentally, this E is the lowest note the guitar makes at standard
> tuning.
> At the risk of acquiring 'luthier disease' (per your recent newsletter),
> that the soundboard on a good guitar can naturally vibrate below the
> guitar's normal range makes sense but I don't know why. Going way out on a
> limb and extrapolating (always dangerous), if the player is going to drop
> tune the base string to 'D', the top tap tone should also be below D.

Just the opposite. You are making a basic conceptual (and widespread) mistake that the way the guitar top sounds as a free plate will somehow rule how it behaves when its glued onto the guitar. Lose that concept from your mind.

You do not "tune" a top like you "tune" a radio receiver to receive signals of certain frequencies. You are not tuning a tuning fork. You tap a top to receive a very broad general sense of how stiff it is. You can't "see" how stiff it is. You can't "feel" how stiff it is (if you flex a free braced top you might break it!) You must "listen" to how stiff it's become. You must listen to it as you pare it down, until you reach a familiar "sound" that it makes when it has reached the point that you have left it before with good results. That's why you must make several guitars before you get the "idea." Your aim is not to "tune" the top to a specific bell-like note or range of notes.

Remember that when the edges of the top get fixed onto the rim of the soundbox, it is not a free plate anymore, it becomes a bound plate, a totally different animal from an acoustical systems standpoint.

In fact, rather than "tuning"the top, I actually "de-tune" it, working it until the sounds it makes when tapped sound dispersed and indistinct. But just until. If I hear a clear bell-like note when I tap it, that's telling me that it's way too, too stiff and I have to bring it down. You hear that clear note best just after it's freshly (an thus, massively) braced, and its waiting for brace carving and shaping with the plane, chisel and sandpaper.

The more material you remove from the braces and from the top's actual thickness, the lower in pitch its tapped sound. If you can hear any focused musical tone when you tap a top, it is still too stiff, too massive. Most amateur builder's first guitars are impossibly massive, because they simply have not developed a sense of proportion that comes with refining their awareness of the precise resilience of the material. My aim, the aim that works for me, is to remove all the material from the top of the guitar which is not needed to support the string tension and its accompanying physical distortion. If you still want to learn about tapping the top until you get a certain note, go to another luthier to explain it to you. I don't do that. I'm not interested in that, because I know how widely and how unknowably the tops acoustics changes when you add the rest of the guitar to it.

Please allow me to use this communication for the benefit of others. Thanks.

William R. Cumpiano
William R. Cumpiano, Guitarmakers


"Lee, Bruce" wrote:

> Once the top is braced, where do you hold it and where do you tap it? e.g.

Put a finger through the soundhole and let it hang from your finger. Put it up close to your ear and tap it with the fleshy pad of the middle finger of your other hand.

> If I hold it at top center and tap anywhere in the lower bout, I get a low
> tone that is sort of fuzzy (for want of a better word).

You are listening for very, very low indistinct tones along the bottom rim. When you tap it about where the bridge is you will hear a ringing note (never mind which one) when the braces are massive but rough-shaped. As you remove material that note will a) go down in pitch and b) become more sustained and focused. Keep removing material from the braces (start by removing mass from the cheeks (sides) of the brace, and reduce it in height as a last resort) until the pitch goes down to a low note, and then just gets so low it kind of disperses and becomes indistinct. I don't hit it very hard because I then get a bunch of high-pitched noises in the mix which complicates things.

Be well, Bruce
William R. Cumpiano
William R. Cumpiano, Guitarmakers

"Lee, Bruce" wrote:

> In a nutshell: thin the top until the fundamental tap tone is "dispersed and
> indistinct". Then after adding the braces, do it again?

Right. You should stop right at that point. Don't keep working it after you've reached that point. I any case, to stay safe, the top shouldn't fall thinner than .1 " (spruce/steel on a 000 size) or .080" (spruce/classic) , because you'll get wrinkling under tension. Add about 15% to both if you're in cedar.
Hi. I have tried a laminated X brace using Graphite as the core. I have found that it is difficult to work and ruins my edges .Do you know of a way work with this stuff ?
Graphite will burn right through a carbide router bit, any thing. The only thing that cuts it are abrasives and rasp bits. Think of it as fibrous brick, not wood. It will wipe out your band saw blade the first time you cut it. But that blade will keep cutting it--but won't cut anything else. So save it for graphite. You can get "rasp" bits and bits with carbide chips welded around it from industrial supply houses under "laminate" and "composite" material cutting tools.

I’ve noticed that I’ve got slight gouges in both sides. one just sort of appeared on the show face (i would assume from careless handling as i only have a few hours I can work on it late nights during the week), but the other, larger one is on the back face. they're both not terribly deep, but deep enough so that I don't want to scrape, plane, or sand any more to get them out. I’m afraid of the whole "messing it up by trying to get too good a product" and ending up with serious thinning of the soundboard.
So I guess my question is this: will these gouges develop into cracks over time from stresses put on the guitar by tensioning the strings? and also, will the gouge on the underside affect gluing the braces if one of the main "x" braces runs right over the top of it?
If something has pressed an indent into the soundboard, you can "pop" it out again--as long as the surface has not been cut or material removed--just compressed. Take an electric iron and turn it to the coolest setting: the mark just above "off'". Dip a corner of a clean old cotton tee shirt in water and place the wet corner on top of the dent or gauge, and gently press the warm tip of the iron on the tee shirt. It should send a small shot of steam into the little dent and after a couple of tries, pop it out. Let it dry thoroughly before sanding it flush. It should just go away.
Gauges or dents on the underside of the top are inconsequential, especially the ones that get braces over them. Don't worry about them, as long as they don't go, say, deeper than a third of the way throught the top. Just glue a flap of stabbed wood down, by working some glue into the flap, putting a square of wax paper over it, and a block of wood over that, clamp it down, and scrape the excess glue of flush after about an hour. Good as new. Not to worry.
I’ve noticed that while planing this soundboard (unlike any other one I’ve done, even while planing with the grain with a suitably adjusted plane) that some light chipping occurred no matter what. and I sanded as much of it out as I could but stopped when I noticed it began to thin down one of the upper bouts a tiny bit at it's edge. any ideas as to why this happened?
As far as planing the sound board, some minor chipping is inevitable if you're planing cross-grain, but the idea is to scrape them out before you get to the final thickness. If your getting more than just minor chipping, chances are your plane is just not sharp enough, or badly adjusted, period. Chipping is always minimized if you can close down the "throat" of the plane, that is, the open gap in front of the blade. Make sure that the throat is adjusted so that there is not much more than a 1/32-1/16" gap in front of the blade, and that the plane iron is right up close to the plane blade edge. If you still get bad chipping, then it's probably too deep a cut for the sharpness of the blade. Sharpness, sharpness. Sharpness is the great lubricant, the great facilitator in this line of work.

My Norwegian correspondent writes:
A small comment on the Article-site: "How good is the guitar".

I read your article on Guitar evaluation accorded to buying oneself a guitar.
> It was generously packed with good advice, but as a former Guitar salesman
>something crossed my mind (Which also made me think of the name of another
>article of yours: "Love Your Guitar To Death".) Following your advice on the
>wrong terms could be: "Test your Guitar choice to death, a guide to REAL
> My point is: Most people (passionate guitar players included for that
>matter) are buying serial-produced instruments. And most buyers has a
>limited budget, with, thanks God, a few exceptions.
> However: The guidelines in your article are most relevant to hand and
>custom-made guitars.
> (I’m not referring to "Spot the plywood" parts, which are quite useful in
>all categories.)
> Point: From serial/mass-produced instruments, is there hard to exempt al
l>those required features too much. And from my Shop-experience do I know that,
(Unfortunately) People that runs an "Expert-cross-check" on their first,
budget nylon acoustic, ends up with no guitar, or even worse, the most
stupid choice ever made. Up to a certain price level, playability and a
decent sound is what you can expect.
> Telling my customers this fact, was in 95 out of 100ed, NO DEAL "What? Are
YOU telling me that you know better than the EXPERT???" HAH!!! And:
> "what i try to say is that, to cover your terms of what...etc. You have to have
ten-times-your Guitar budget."
> (The customer leaves, to buy a Casio keyboard, in another store.)
> So in every article on guitar buying there should be a part that tells
(forgive me this comparison, please)
> "Guitars, are like sneakers. You Buy a pair of sneakers for 15 bucks, and
they will lass trough summer, if you’re lucky. You buy yourself, or INVEST
150 bucks, you'll probably have 'em for years, EVEN if you’re a regular
> You'll get what you pay for, regardless of expert-advice. What’s your need?
> Much of the work in front of picking out the definitive guitar is done by
playing the one that you already got, asking yourself: What are my BASIC needs?"
Being conscious to that, will make the choice easier to make, and save one from
LOTS of confusion.
> And IF the shop one is in has a guitar with a loose bracing-rib or
visually bad gluework out fore sale WITHOUT, a HUGE Discount label attached
to it. Your first chose will be: Find yourself another store. These guys
ain't too serious 'bout it.
> Seriously hope I didn’t offend anybody, I think I got a point, you see.
> Stay in tune, I will for sure.
> Erland Eikestad, Norwegian Guitarslinger & Aspiring Luthier
Dear Erland,
I get your point, and you're absolutely correct:
a) the article did not cover how to check out inexpensive guitars as
carefully as it covered how to check out more costly guitars.

b) because of this, a reader of the article might get unreasonable
expectations when looking for an inexpensive guitar, and that this could
drive someone ín the business of selling inexpensive guitars to be concerned
that the article could create "confusion" and cause problems between sellers
and buyers.
c) It would indeed have been better if I had made the point that if you're
only spending $100 on a guitar you shouldn't expect the refined
construction, intonation, durability and sound quality than you would on a
$1000 guitar.
On the other hand,
a) I think most people would naturally understand that quality features are
less likely on an inexpensive guitar. On NEW guitars, my article described
quality features to look for--regardless of cost, and which features you
would normally find on guitars of different prices. On USED guitars, my
article focused on steering people away from guitars with problems, such as
high action without the means to lower it easily. Or bridges that are coming
b) It is probably worse to discover action or structure problems on an
expensive guitar you just bought, than on a cheap guitar. Thus:

c) The article would be indeed have been more helpful to more people if I
had focused on how to tell if an inexpensive guitar was bad. But I told them
THE LEAST to expect was:
That it play reasonably in tune
That the action not be hard
That all the seams be tight and well glued
That you don't scrape your fingers on the fret ends.
That there not be any unusual bulges on the top.
That it not buzz or rattle anywhere when you play.
This list is the least to expect of any guitar regardless of cost. Or would you disagree? If someone reads my article and then complains that the inexpensive guitar you are selling has high action, or a gap under the bridge, I don't think you'd tell them: "Oh, you can't expect low action and a firm bridge on an inexpensive or used guitar. That article just confused you!" Not unless you were trying to take advantage of him. And which lower-quality features are acceptable on an inexpensive guitar?
How's this list:
Laminated veneer construction
A catalyzed lacquer finish which is a bit wavy and just shiny, not glossy.
NOT having a real ebony fingerboard/ bridge
Simple decoration or decoration of imitation marquetry or pearl.
Tuning machines which will need to be replaced within five years.
Plastic nut and saddle
An opaque, somewhat muffled tone, but sweet and pleasant nonetheless.
That list summarizes the article pretty well. So I think I educated more than I confused. I hope you agree.
I was getting ready to laminate a block of mahogany to make a headblock from and realized my single board of mahogany is not wide enough. I don't have anyplace to by more unless I order it (there are disadvantages and advantages to living out here in the hills of Kentucky). I do happen to have a board of padauk that is pretty well quarter sawn. I was wondering what you may think about using that or should I just go ahead and order a mahogany board?
Any even-textured, homogeneous hardwood will serve for a headblock, as long as it seasoned and reasonably quartersawn. Clear Poplar will do, if it is a small guitar (it is not as strong, and small guitars are usually under lesser tension: early guitars used pine (it was denser and of better quality then, than what's commercially available today, through). Mahogany is preferred because it is dimensionally stable--indeed, more stable than any other wood (in fact a dimensional stability quotient is set with mahogany as the theoretical standard, that is, one. Other woods are .8 , .5 of the standard). Flat-sawn maple is very, very low on the scale of dimensional stability, and quarter-sawn East Indian Rosewood is actually about .9, very close to Mahogany. Oak is fairly high--which is why it is used in good furniture as drawer parts. Padouk is around .8 so it will probably serve all right, as long as it's quartersawn and dry. I would lightly sand and lightly wipe clean the mating surfaces with naptha just before gluing. But it seems a waste. It is such a rare and beautiful wood, hardly appropriate for lowly internal instrument parts!
Does a smaller bridge effect the guitar's ability to withstand the forces of a medium set of strings on it? also, is there a limit to how small you can make a bridge of that sort? I was thinking about making it just under 1-(1/8)". is that too small?
There are other factors beside the "footprint area" of the bridge that affect whether it is adequate to the load of the strings. The height of the bridge is crucial because that is the principal determinant of the amount of leverage the strings exert on the bridge. That, of course is determined by the amount of neck angle you've imparted to the guitar. I can't guarantee that your bridge will fail or not fail--the quality of the glue joint is critical. Also the design: the forward half of the bridge is pressing into the top; and the rear half is pulling away from it. If most of the rear half is gone because of the string holes, you're going to compromise its integrity. That's why the Dreadnaught/Martin bridge flares at the bottom: to provide gluing area BEHIND the string holes. If you make a bridge rectangular, you reduce its chances of staying on...unless you use light gauge strings AND make it a low bridge. That's why you find low, classic-like rectangular steel string bridges on small Martins, like old timey 0s and 00s: they were originally designed for gut strings (half the tension of steel) and light gauge steel.
So if you ARE making a rectangular bridge that is going to be strung up with medium gauge strings, a) leave about 1/2" of bridge to extend behind the bridge pin holes b) design your neck angle so you'll get low action with a bridge which is 1/4" to 5/16" high c) make your bridge patch out of rosewood, about 1/8" thick and to extend 1/2" at least below the bridge, to insure that the top doesn't flex away from the bridge when under tension d) don't sock the saddle slot all the way up to the front edge of the bridge, the front lip can fracture easily under the tension of medium strings. Bring it back 3/16" to 1/4" from the front edge of the bridge.