William Cumpiano's
String Instrument
Newsletter #15

Go to newsletter number:

Questions, anyone? IV

William R. Cumpiano 1999, All Rights Reserved


Mr. Cumpiano-

I'm wondering if you can shed some light on the neck fitting issue? I see in your book where you propose that a neck made flat in its relaxed mode will pull up into a slight curve when up to concert pitch. This slight curve then tends to match the curve of the string vibration envelope. (Hope I'm not misquoting you in any way) Other people suggest putting the guitar in a neck setting jig to hold the neck in the same shape after the string tension is released and then planing the fingerboard flat, installing the frets and filing them flat. Bruce Johnson has built a holding fixture for mounting the neck assembly only and holding it in the pre-stressed condition. Then he carves the radius on the fingerboard and so on. Personally, I like your method best but wonder how there can be such a design divergence within the trade?

The guitar in this country has been repeatedly the victim of the quintessential American tinkering reflex. I never ceased to be amazed at the Byzantine inventions that some luthiers will devise to satisfy the dictates of some of their incompletely-thought-through notions.

Guitarmaking as a craft has been active for the better part of a thousand years. Does it make you wonder what all those thousands and thousands of makers did over the centuries without the benefit of Bruce Johnson's guitar rack and thumbscrew device? Where there was no need, Johnson filled it nonetheless. And now learners like you have bumped into yet another landmine in your path to efficient and effective guitarmaking technique.

Bruce Johnson. I don't know the gentleman but it sounds like he came into guitarmaking from the repair side of the business. Whereas Johnson's guitar torture device could conceivably have a useful function as a repair technique for very old, worn, lumpy or badly warped fretboards, it really is rather silly and counterproductive on new guitars.

The presence of a perfectly flat fingerboard AFTER the guitar is strung up to pitch requires that the action be raised unnecessarily to avoid buzzing when the strings, especially the wound strings, are plucked forcefully. There is no question that you can achieve a lower and more comfortable action for any given dynamic range with a slightly curved, rather than perfectly straight, fretboard. The curve that a previously straight fretboard naturally takes when placed under string tension usually does the trick. And if the neck naturally wants to take an excessive curve under tension -- that's precisely why the truss rod is adjustable--to bring it back to it's optimum. No muss, no fuss, no thumbscew torture rack. It seems clear that the Johnson rig defeats the purpose of the adjustable truss rod.



Dear Bill,

I am having what's turning out to be a rather heated debate with a few of my friends on the topic of re-stringing an acoustic guitar. They are saying that you should NEVER take off all the strings at once, but rather change them one at a time. They think that removing all the strings at once is somehow detrimental to the guitar. Having built a few steel-strings from your book, I say "hogwash".Take off all the strings if you want. It won't hurt a thing.

A note from you on this topic would settle it once and for all.

I doubt it. But it seems that everybody that promotes this myth is hard-pressed to suggest any factual justification as to why it is indeed bad to do so.

The myth most likely originated from a skewing of what is indeed sound advice on all TAILPIECE instruments (violins, cellos, arch-top guitars): if you take all the strings off, the bridge falls off! That good advice jumped from one instrument to another and became poor advice on flat top guitars. The skewed information just got handed down uncritically from teacher to student over the years and became enshrined as a Truism. By the way there are dozens of similar religiously-held myths among players/teachers/makers. Like, never cut the strings. It damages them. I should try to list them all some time.

I also love listening to the "reverse engineering" a justification for a myth by its proponents! That is, starting with a myth, and then creating an elaborate set of highly logical and credible reasons why it must be true. Let me know the "reasons" your friends made up to support it, please!



I am now on the part where I need to route a 3/16 groove in my neck blank, but I can't seem to find a 3/16, round nose router bit. I've looked in all my woodworking catalogs but can't find one. Any suggestions?

You need to look around some more for the 3/16 bit. It's out there. Did you try Sears, or any one of those huge home improvement chain stores that are popping up everywhere? It is definitely a consumer item. Keep trying. By the way it needs to cut 1/2" deep or almost 1/2". That makes it a little harder to find. But it's out there. Also, there are dozens of router-bit specialty stores that advertise on the net. Do a search for them.



I got a good deal on rosewood for the sides and back, but after planing the headstock veneer (also rosewood) I find that I am getting some tearout in random grains of the wood. Not real bad, but kind of annoying. Do you know of anything I can do to prevent this ?

Yes, planing figured hardwoods correctly calls upon skills which many beginning woodworkers haven't learned yet, and even many experts never perfected because they found easier and available ways around ever having to! Your plane has to be in perfect "tune" to shave figured hardwoods, and it takes several years of just perfecting that skill. Workarounds?

* If your work is close to finish dimensions, scrape it down to thickness. But your starting surface has to be reasonably smooth. Or plane down somewhere near finished thickness, and scrape down to dimension, hoping that your occasional tear-outs will disappear in the scraping process.

* purchase or make yourself a home-made wide-belt (or not as efficient sanding-drum) abrasive planer (called a "filletier") or purchase one. They can be rather large, and very dusty, so you can't rig one up in your kitchen. But if you intend to pursue this craft, some sort of motor-driven abrasive planer will be the most important piece of equipment in your entire shop. I've seen some very creative solutions done with bolted-tegether plywood and an electric motor that was very effective in taking a 1/8" bandsawed slat of curly maple and reducing it to a perfectly smooth .083" guitar side.

Its, that or work to improve your plane and planing skills, period.



A correspondent sent me the following excerpt from Steve Grimes’ guitar webpage for my comment.

The "GS" guitar is the result of a design collaboration with Ned Steinberger, the innovative creator of the Steinberger bass, guitar, and violin. Ned designed a bridge which, under full string tension, would not exert any stress on the soundboard! Utilizing this stress-free bridge, the strings do not pull up on the top, as in a conventional steel string or classical design, or push down on the top, as in standard arch top instruments. The strings interact with the bridge and attach to a tailpiece, leaving the top unencumbered by the heavy bracing needed to counteract the string tension in a normal guitar. The soundboard can be braced for tonal purposes without regard for structural considerations.

In this design the soundboard becomes to the luthier what a blank canvas is to the painter, limited only by the imagination. The guitars we've produced thus far have turned quite a few ears! Most players have commented on the surprising clarity of all the notes- from the low bass to the high trebles. Because the soundboard can be lightened up structurally, its possible to produce a wonderfully deep, pure bass.

This model also features other innovations such as neck-thru-body construction where the neck extension goes through the body about an inch below the top and attaches to the tailblock. Both the action and intonation are adjustable through an access panel built into the tailblock. By using one allen wrench the position of the neck to the body can be adjusted up or down, fore or aft. The tuning machines are another technical breakthrough, using a gearless, direct drive system that eliminates backlash and slippage.

The same correspondent then included a recent response about the system from Grimes:

Dear Marc,

Thanks for your email. Below is posted the price/options list for all the models I make. I no longer make the GS with a neck through body design. It didn't work all that well as it turned out. The picture and info on this model in the brochure were done right after we made about five prototypes. Since that time, we have found that what works best is a smaller body guitar such as a OM style or slightly larger, and a center soundhole instead of offset. The increased activity/vibration of the top due to smaller braces tends to enhance bass response, and the offset soundhole also does this. We found that with the offset soundhole the guitar was a bit too strong in the bass, so we went back to the center hole. When we sold the patent for this model to Gibson in '93, they also did some research and found they preferred the center hole. (I still retain the rights to build 20 of these guitars per year).

If you play in any lowered tunings, the Hapa and Beamer models may suit your needs too. These guitars have really rich tonality in the bass, even when the low E is tuned down to C. Trebles seem to sustain longer.

  I feel a bit badly for Steve Grimes, a friend and colleague: this was a misbegotten project from the beginning. The string’s tension is the prime driver of the top, besides being it's slow destroyer. After all, as the string swings back and force, it minutely increases and relaxes the "background tension." This IS the string’s signal. Any scheme to dump the bulk of this tension onto the tailblock would of course result in gross disappointment: just listen to any flattop guitar that has a tailpiece: they all sound dreadful, particularly the best-made of all of them, the early Martin D-2's which, like Steinberger's was discontinued. He should have taken heed of those old lessons. Those old Martins survived by having their tops and tailpieces torn off by my friends Matt Umanov and Ivon Schmukler, and replaced with traditionally-braced soundboards: a move that presaged the Martin M series.

The sound of a guitar is intimately related to its anatomy. "The sound," that is, the pleasing high-quality guitar sound that we have come to appreciate as such, is a product of a traditional anatomy. By changing the anatomy massively, you change the sound so it sounds like...something else. So it's bound to be a disappointment, unless you're incredibly lucky.

And I love all those self-assured after-the-fact rationalizations of essentially invisible phenomena. Steve "found that with the offset soundhole the guitar was a bit too strong in the bass, so we went back to the center hole." The lack of doubt, the total certainty of cause and effect is really quite impressive. That dumping the tension from the top "leaves the top unencumbered by the heavy bracing needed to counteract the string tension in a normal guitar. The soundboard can be braced for tonal purposes without regard for structural considerations." They sure sounded like they were on top of it, that the keys to world-class consistency were firmly lodged in their hands. So what happened to the scheme, to the certainty? Well it sounds from the letter he sent you that they reverted back to the traditional forms, without owning up that their original premise was careless and in the end, a lesson that anatomy is destiny.

I'm not saying that the traditional forms can't be improved: my point has always been that pursuing the soundboard bracing pattern as the holy grail of guitar sound is a waste of time. I place a greater weight on materials, plate thickness and dimensional proportions, and disregard bracing patterns, soundhole placement or soundhole shape. But that's my opinion, and that is why guitar design is an Art, and there are many artists doing it differently. I say it better in the Guitar Bracing Designs section of my book, written in 1985:

" Unfortunately it is extrememly difficult to prove a concrete causal relationship between any specific design modification and tonal improvement. Our observations have led us to a conclusion that some readers may find surprising: Specific elements of brace design, in and of themselves, are not all that important! One has only to look at the myriad designs employed on great guitars to recognize that there is no design secret that will unlock the door to world-class consistency. A great maker will probably build a great guitar no matter what brace design is used. What the great builders have in common is the ability to integrate the proportionate weight and stiffnesses of all the parts relative to one another and relative to the volume of the soundbox and the scale length. The brace design, in short, can only be as successful as the whole of which is [just a small] part."

Recently I experienced a rare epiphany, when I heard my old graphite-soundboard guitar from which all the soundboard braces had fallen off (the guitar sounded wonderful!). This dramatic demonstration just reinforces my conviction that the precise pattern is meaningless.

In fact 1/32" thick carbon fiber soundboard in and of itself accomplished what Steve was essentially trying to do: not have to consider string tension as an architectural factor, and allow the instrument to be unencumbered by braces. But his scheme required a massive change in the fundamental anatomy of the guitar, which spoiled the experiment. 

But I should reserve all MY certainty until I get the results of my own upcoming experiments. I may indeed find that 1/32" may not be the ideal thickness of a (graphite) soundboard which is plenty strong to withstand the string stress without distortion. I will be making two guitars with the new top, one without braces at all, and one with an exceedingly light x. Stay tuned.