Official Luthiers Forum!

Leo - what I need to do is pick up a copy of Trevor's book.

As in any field, there will be disagreements on topics which appear on the surface to have a singular black and white answer, but which upon closer inspection end up being contingent on some "it depends" clauses.

What I know of Trevor's work appears brilliant and thorough. On this topic there appears on the surface to be some irreconcilable conclusions, and observations which seem hard to reason could coexist. From my perspective this is a curiosity which seems simultaneously terrible and wonderful, because while I would not doubt the thoroughness of Trevor's work, I have yet to find an explanation as to what differences in conditions or criteria may have led to our disparate observations and conclusions.

My guess is that there are some "it depends" clauses buried somewhere, yet to be clearly exposed. I don't hope for verification or repudiation, but rather assume there may be a clarification somewhere in there. Of course as I argue my side to the best of my experience and understanding, I will always try to remain open to errors or elements I may have missed.

Given the work Trevor has put in to publishing his positions writings, in all fairness I should acquire and read them when I get the chance.

Regarding the Evertune systems, no, I've never worked with them. I do recall noticing however, that even with a constant tension spring system, their saddles still end up compensated differently with string thickness changes, which of course lends argument to the influence of stiffness.

I do believe the stiffness issue is left unresolved, in part because in application it is so greatly variable on conditions that it becomes difficult to resolve to clear applied formulas. Even the Fletcher/Rossing books, while listing clear formulas for elongation and elasticity, seem to describe stiffness as a vague variable which exists to some undefined degree.

I find myself absent of clear formulas to describe the influence of stiffness, and they may likely be so complex as to change drastically depending on materials, tensions, angles, etc. The results I have consistently observed in application however, have never been fully satisfied by equations of elasticity alone. The differences I find lacking between simple elasticity formulas and real world observations however, do seem quite perfectly placed to be explained by influence from stiffness.

_________________
Eschew obfuscation, espouse elucidation.

You're quite right David. I've been telling people that for years, but still wrote it down wrong. I need to wake up some time soon! To be clear, for a guitar with a fret-plane level nut and no relief, the break angle at the nut doesn't change with fretting position, because between nut and fretted position the string lies on the frets. The real life problems with setting the nut that low, though, are nut life and back buzz. The back buzz can be solved by either increasing the nut height or playing with the relief. Playing with the relief buys you another set of problems so the nut height is usually lifted and you get back to the break angle problem. I edited the original post, hopefully so as not to add confusion.

The rest of my post I stand by, so we'll have to agree to differ.

Your later post came in as I was writing. There are all sorts of "it depends". What weight/type of strings are modeled (longitudinal stiffness rather than bending stiffness), relief settings (more significant than one might think), action at each fret, (which relief contributes to), how finger pressure is modeled, nut height (or not) etc. etc. And we each have our own experiences. My experience is that nut compensation is a very positive benefit to getting a guitar to play equal temperament accurately. If it wasn't making a positive contribution, I wouldn't be doing it.

Thanks for the kind comments regarding the book etc.. It's currently sold out and I'm flat out organizing the re-print. That's been a whole lot harder than it should have been... and a lot of late nights.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

Two authors (that I know of) have produced formulae for strings with bending stiffness, Morse and Fletcher (of Fletcher and Rossing fame). Beyers, when he did his intonation analysis (for nylon strings) used Morse's formula, approximating a clamped end condition. Fletcher's formula allows for different end conditions, pinned and clamped. You'll know the simple Mersenne formula, of course. I did a whole lot of tests measuring the string frequency as it was shortened using guitar type terminations and then looked at how each formulae predicted the outcome. The simple Mersenne formula with no bending terms was the best predictor, hence my view that bending effects have little impact on the fundamental in guitar applications.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

I've read Greg Byers' work, but was completely unaware of Morse and Fletcher's. I'll have to dig that up - Thanks!

_________________
Eschew obfuscation, espouse elucidation.

This is simple--the Coriolis effect (spins the strobe wheel to the right in the northern hemisphere, and to the left in the upside down countries).

How do you deal with back buzz when the guitar is capoed?

Would that not be the relief? (Presuming you are talking about fretted note above the capo)

_________________
The name catgut is confusing. There are two explanations for the mix up.

Catgut is an abbreviation of the word cattle gut. Gut strings are made from sheep or goat intestines, in the past even from horse, mule or donkey intestines.

Otherwise it could be from the word kitgut or kitstring. Kit meant fiddle, not kitten.

When I'm playing, I use two capos if I'm getting back buzz. There's a few tunes I play capoed at 7 which need a 2nd capo on 3 (I put it directly on top of the fret) to stop the back buzz. Of course, that's not a solution my customers will necessarily accept and back buzz is more prevalent the more responsive the guitar. So you're left playing with nut height and neck relief and hope you don't have to compromise the set-up too much.

Re: the problem of back buzz between the fretted note and the capo, that's not one I've come across.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

Maybe they'd accept the Van Epps damper (Patented 1950, used by Jim Hall, Herb Ellis, Scotty Moore-- more versatile than a second capo--allows damped open chords with partial voicings).

I haven't either. Why not make a long scale guitar, tune it down a half step, and capo at the first fret?

I finally found the 26 minutes and 35 seconds to watch. Educational and inspirational to say the least. Thanks very much for taking the time.

I found it interesting that you feel a standard set up is best to work with. Almost from the beginning I have been taking 1 mm off the nut end of the FB (on acoustics) and have been happy with the results. I'll give it a try on the next on without.

Thanks again!

I have constructing an acoustic guitar after some electrics and started following plans in a book, but discovered how complex it was to tune the sound. So I looked for a more scientific method and thanks to the forum discovered the Trevor Gore book. I am at the stage of glueing down the bridge after deciding to follow the original compensation in the book I have. After reading the section about compensation in Trevor's book, this video was released, then the comments between David and Trevor. So I decided to check 2 factory acoustics and my first guitar a semi hollow 25.5 I will call 335 because that is the body shape, now with flat wound strings.
I did not check all frets on all strings because it takes some time, however
The factory guitars follow the same trends Trevor's graph shows, but worse, being 12 cent sharp at fret 1 and 14 flat at fret 20. Weirdly the flat strings seem closer to correct.
I will do more testing now, and am considering compensating the nut as well.
Thanks to both of you for sharing your findings.

Those sound like some pretty odd and extreme errors. I would suggest this to help in diagnosis - if you capo at the first fret and adjust it in to tune, what does it read around your 2nd and 3rd fret, and up around your 13th? Sounds suspiciously like a nut slot not only being high, but possibly even breaking over from toward the back side (which would amount to reverse compensation).

It's worth noting that flatwounds can be a wild card, varying widely in intonation and consistency among different makers.

_________________
Eschew obfuscation, espouse elucidation.

Bumping this topic to call your attention, David, to a recent issue of American Luthierie which has an article about intonation which I'm sure you would find very interesting.

It's by Mark French entitled A Simple Modification to Reduce Frequency Errors in Guitars in AL#125


Also very relevant is an article in AL#116 by Gary Magliari and Don MacRostie called Beyond the Rule of 18.


I don't know if you have access to these issues but I'd highly recommend them and I'd love to hear your thoughts on them if you do get a chance to read them.

Cheers!
Leo