Official Luthiers Forum!

I have an order for a Parlor, so am looking at some adjustments for top deflection with a smaller guitar and came across this article looking for something else.

https://esomogyi.com/articles/specific-top-thickness-in-the-guitar/

Lengthwise I don't have a problem with his writings, but width-wise his advice/instruction seems strange -

Seems to me one would need a thinner top with a narrower guitar to maintain deflection?
Or is he taking the new thickness for the top plate after adjusting for the length?

Confused.

_________________
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.

He is so wordy that I sometimes have trouble following him, I read his books. From his books I think he is a very intuitive builder. He takes math and physics and puts it into a real world mental model to think about. I would rather just see the equations. But he has put together a model that works for him and the people he has taught.

He gets it right in his example:

Guitar A is 16/15 (106%) the width of Guitar B,
and Guitar B is 15/16 (94%) the width of Guitar A;
Therefore, as far as plate width influencing plate thickness goes:

Guitar A, being wider than B, needs to be thinned by the cube root of that 106%. Reciprocally, guitar B, being narrower than Guitar A, needs to be left thicker by the cube root of the percent of difference. These calculations will yield small numbers — something on the order of .002” .

His need to calculate the same result using the reciprocal is an example of why I call him wordy.

_________________
http://www.Harvestmoonguitars.com

My thinking is that the wider the guitar, the thicker the top should be to keep the same deflection.
The narrower the guitar, the thinner the top should be to keep the same deflection.
What am I missing?

_________________
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.

I'm with you, Colin. If I place a board across two sawhorses, the force needed to hit any given downward deflection is going to be greater when the two sawhorses are closer together than when they are further apart. I could leave the sawhorses where they are and thin the board, and that would lessen the force needed to hit the desired deflection.

I'm not a physicist, but my best guess is that Somogyi had the examples reversed.

Note that the article is also focused on how much deflection change is needed, depending on whether the top is wider or narrower, and gets into the cube root stuff. I can see how the wider/thinner differences can be affected by the cube root stuff. But in terms of the simple question of how width affects deflection (rather than how much), I think you and I are right.

I'm open to someone explaining to us in caveman terms why we are wrong.

Yep, that goes reverse to basic structural engineering principles.

No you are right I read it backwards, he has it backwards from my perspective.

_________________
http://www.Harvestmoonguitars.com

I'll be the odd man out here.
Don't think you really want the same amount of deflection on a Parlor as you would would on a Dreadnought.
You don't have as much air to move inside the box and if it is too floppy your sound will be muddy.

I agree with John that the numbers aren't going to be huge. But one should think about what sort of adjustment needs to be made to get the sound they are looking for.
If I remember right Somogyis' final tuning is done after the top have been mounted on the rims and placed inside the mold.

Yes, he tunes with the top on the rims.
So to clarify, you're saying the top on a parlor should be the same thickness (for the same piece of wood) as on a dread?
Or just a little bit thinner, but not too much, so that the actual deflection would be similar to what it would be on a dread?
Because that's what he says for a longer guitar.
If that works for lengthways (I'm fine with that) why does it appear he's saying the opposite for widthways?

_________________
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.

FWIW my parlour tops deflect more than my dread tops by about 20-/25%...

Colin,

If one were to do that then the decreased size of the template would tend to stiffen up the top. Hopefully.
So yes that is what I am saying.
What I am saying is ( just to pick a number ) If your deflection is .100 on a Dead you do not want your deflection to be .100 on a Parlor.
I'm not one to say that my top thickness will be .110 on everything though as I do tend graduate my tops on steel string guitars.

Hmm, on the one hand it seems logical for the deflection to scale in proportion with the size of the soundboard... but on the other hand, perhaps it scales in proportion with the frequency range of the instrument?

In general, the instrument size scales with its frequency range. But as we see with guitars (and ukuleles) there's a pretty wide range of sizes that will work for a given frequency. So does the optimal deflection track with the size, frequency, or some combination of both?

For a thought experiment, take an ukulele and an acoustic bass guitar. Obviously if the uke deflects as much as the bass, it will be paper thin and crumble. And if the bass deflects as much as the uke, it won't produce much sound at all. But what if you wanted to make an instrument the size of an acoustic bass guitar, but with the same frequency range as ukulele? Then you may actually want it to be stiff enough that the deflection matches the uke, so its resonant frequencies will be high. And use high tension steel cables as strings to produce enough power to move it.

Since the dreadnought and parlor are both tuned to the same pitch, maybe they really should have the same deflection. I'm not sure.

I find all this interesting. I think as a repair guy I've gotten to handle a ton of different instruments. One thing I've found is that there are instruments that are built feather light that sound good, and some that are brutally heavy but still sound good. The only Bozo Podnuvac instrument I ever touched was a 6 string dreadnought style one. It was so heavy but man it sang like an angel. Also had a 39 J35 in recently, barely weighed anything yet still sounded fantastic.

Now granted those are different instruments from way different periods/build style/wood but I always found it fascinating that there's really no 1 single method for making a good sounding instrument.

Not knocking the desire for measurements (and things like deflection studies) I just feel like so much of guitar making ends up being something you can do by feel as well as scientifically. I've heard some scientifically perfect guitars but they never quite sounded as "human" as the guitars built in the old style. Does that make sense?

Factory guitars were not made by feel. All the parts were the same and they were thrown together as fast as possible, romantic visions aside.

The fundamental issue is well-stated by Somogyi in his first paragraph:

"Top thickness is, along with bracing, the most debated and tinkered-with area of guitar making. It is so for two absolutely important reasons. The first is that the physical characteristics of the top set the stage for tone — along with the corollary that the lighter the construction of the top is, the better the sound. The second is that there’s a minimal top thickness/stiffness that must be respected if the plate is not to cave in under string load. If sound is one’s objective, then the luthier’s balancing act is in finding the correct balance point between the imperatives of ‘light construction’ and ‘not too light’."

The unsaid background that drives "the luthier's balancing act" is the unfortunate tradition of using the guitar soundboard (probably the most critical element of the guitar) for two functions whose requirements are largely in opposition to one another: structure and tone. The need for structure appears because, in traditional construction, the neck/fretboard extension lays on the soundboard and the location and orientation of the headblock is, to a substantial degree, determined by the rigidity of the soundboard. With the duties of locating the headblock and supporting the fretboard being compromised by the soundhole, substantial bracing of the top is required and any contribution to musicality from the upper bout is generally sacrificed as a result. This is TOTALLY unnecessary.

An alternative approach to guitar construction uses braces between the headblock and the tailblock which don't touch either the soundboard or the back of the guitar and which rigidly fix the orientation of the headblock and, therefore, of the neck, with respect to the sides (or "rim") of the guitar. When this approach is coupled with an elevated fretboard (one whose extension is spaced away from the soundboard), the soundboard no longer needs an upper transverse brace or a popsicle brace or, in fact, any brace not useful for tone generation. The only exception to this is the need for braces to manage the twisting of the bridge under string forces. I've found that very light X-bracing that extends under the wings of the bridge is adequate to control bridge twist and that light X-bracing, along with light tone bars that are generally radially directed from the bridge will make a thin (< 0.100" thick) top perform well when not required to act as a structural member. Such a top will resonate over its entire area, including the upper bout, and can generate a richness of sound that many people will find pleasing. It will not, however, sound like a Martin guitar.

I'd be very interested to see some of your guitars built like that, any chance of a link or pictures, or something?

_________________
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.

I, too, have found Somogyi's books hard to follow. Gore does a much better job, if you're up to the technical demands.

Iirc, Gore said that the crosswise stiffness of the top doesn't contribute much. It also seems to me that whatever contribution it does make diminishes over time and the top bellies up; 'cold creeping' under the sustained torque of the bridge. In the end, it seems that using the lengthwise Young's modulus of the wood, or measuring the deflection under load for lengthwise bending, should produce a usable, and somewhat conservative, thickness, in the sense that any (small) benefit from the cross stiffness will be a safety factor.

There was an article in 'American Lutherie' some time back on scaling.

Couldn't agree more about Gore being easier to follow - the maths is challenging, but I'm fairly handy writing a spreadsheet.
And as for searching for references in Somogyi - he had me lost initially - what's wrong with page numbers for crying out loud?
For different lengths of body I've used stiffness adjustments based on my standard OM size, ignoring width adjustments, and have played with stiffness/thickness vs bracing height on some builds.
I think any plate tuning for stiffness width-wise could be handled by thinning the outer part of the lower bout..
I remember too you said a fellow builder kept the tops with less stiffness width-wise for smaller bodied guitars.

_________________
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.

Hi Colin,
One of my favorite types of guitars to build is the Martin size 1 template. It is only about a half inch shorter in length than an OM (I stretch it out a little so mine are only about 1/4 inch shorter - I also use the long 25.4 scale). Since it is a 12 fret design the bridge winds up in a better position on the soundboard, and having a narrower (12 3/4 inches) lower bout it meets the requirements of a parlor guitar. I really feel the longer body improves the bass response and gives a better balance than the typical parlor guitar. I generally use a light X bracing scheme, but on the past few have used an X fan scheme. Taking a page from Torres on the one I am building now I'm using an open bar brace and extending the fans into the upper bout. The fans are fairly heavy because I will use steel strings on it.

Nice one Clay S, very interesting,

_________________
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.

I don't mean to be "hijacking" the thread, but, as Colin initiated the thread and then asked for a picture, here is one:



The headstock-to-tailstock braces are cut from Baltic Birch plywood. This, coupled with an elevated fretboard, means that no UTB is necessary and the entire soundboard is available to contribute to tone. The soundboard needs only to deal with tension between the bridge and the tailblock and to control bridge twist. Use of a fanned fretboard and the resultant slanted bridge and bridgeplate help this issue. None of the braces are tucked and none intersect with other braces. Only the diagonal brace from lower left to upper right in the photo has any structural duties, helping to control twist by running under a wing of the fan fret bridge.

The rim supports a Manzer wedge, with the string 6 side about 1" narrower than the string 1 side, when measured at the lower bout. The two holes in the headblock accommodate a bolt-on neck. The hole nearest the soundboard holds the neck to the box using a ¼"x20 hex drive capscrew with stacked wavespring washers that allow some angular adjustment. The hole with the T-nut supports a 10-32x1.5" hex drive capscrew with its head and washer buried in the heel of the neck. The hex socket of that capscrew is accessible through a small hole in the heel, allowing the neck angle (and, thus, the string action) to be adjusted easily by the user at any time.

The structural rigidity provided by the headstock-to-tailstock bracing permits the very light soundboard bracing. This, in turn, allows the guitar to be perform well when using strings with a wide range of tensions, ranging from ball-end nylon strings to medium gauge steel. The range of action needed by that variety of strings can be accommodated via the adjustable neck angle.

Sweet. Thanks. Nice to see different ideas put into practice, and this one looks relatively simple to construct

_________________
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.

The whole width/thickness/stiffness thing (and I'll add to that density and resonant frequency) over different sizes of guitar has challenged people for years. Hence Equ. 4.5-7 which sorts all that out for different woods and different body sizes. Using the "f" number (f for frequency) you can also tailor the equation to your own building style if you prefer not to use mine. As it happens, I always use f=75 for steel string guitars, which covers all top woods and my range of body sizes with lower bout widths ranging from 390mm down to my "super parlour" at 340mm and delivers a thickness number to suit.



A 50% change in the crosswise stiffness of the top panel (cross grain Young's modulus), with other parameters held constant, only changes the resonant frequency of the panel by 3%, so pretty insensitive, as Alan confirms. The vast majority of the sound produced by a guitar is radiated by the resonant modes (standing waves in the top that produce a piston-like action on the air), so their resonant frequencies are critical to the nature of the sound a guitar produces.

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

http://www.goreguitars.com.au

I am really interested in the longitudinal structural bracing that Hans illustrated. I would like to discuss that idea some more so I started another thread here:
viewtopic.php?f=10101&t=53515&p=702668#p702668