Official Luthiers Forum!

I'm using the Gore and Gilet books to build an OM sized guitar. I'm at the point where the box is now closed, and I took my spectral analysis readings from it for the first time. Trevor says that a target should be around

103 hz fully coupled Helmholtz
205 hz Main Top
226 Main hz Back

I'm sitting at

107 hz fully coupled Helmholtz
223.8 hz Main Top
242.4 Main hz Back



So how do I lower these to get down to where I need be? My first thought is to take some mass off the top, but then that will weaken it a bit, right? I just don't want to screw it up. Any help appreciated. Thanks!

_________________
Jason McGowan
M&S Guitars
_____________________________
United States of America, looks like another silent night
As we're sung to sleep by philosophies that save the trees and kill the children....
Casting Crowns

when you say closed box do you mean strung up and ready to play???

_________________
MK5acoustics.com

Doesn't Trevor load mass onto the sides to lower frequency on a finished box? I remember him commenting that it had to be really firmly bolted to sides to have the desired effect. Check out page 15 here:

http://www.goreguitars.com.au/attachmen ... _paper.pdf

or find that place in his books.

cheers, SteveG

edit; I'm already way out of my depth here. I'm sure somebody smarter will be along to comment soon....

Closed box as in top and back already glued on. No really good access to the braces on the top.

_________________
Jason McGowan
M&S Guitars
_____________________________
United States of America, looks like another silent night
As we're sung to sleep by philosophies that save the trees and kill the children....
Casting Crowns

I did add the mass support blocks to my sides, so that is a possibility. I'm just so far off the target numbers to start, that I'm not sure that mass could lower it that much.. I will have to look deeper into that I guess.

_________________
Jason McGowan
M&S Guitars
_____________________________
United States of America, looks like another silent night
As we're sung to sleep by philosophies that save the trees and kill the children....
Casting Crowns

The bridge will lower the top main freq to the range you are looking for and it will shift the air res slightly lower". Your back freq is too high, you may have to brace carve to lower it. Wait until the bridge is on then measure again. Carve the back down first as it will slightly shift the top lower as you lower the back. Watch out for landing on scale tones! The back and top are relatively easy to shift. The air, not so much

In a simplistic sense, the natural frequency is sqrt ( k/m ). So taking some mass off the top (without altering the stiffness) would increase the frequency, not drop it.

Removing material is tricky. You will mostly likely affect both mass and stiffness at the same time. However, if you are strategic, you should be able to influence one more than the other. For the main plate modes, the middle of the plate will move a lot, but it doesn't bend all that much. So changes there will alter the effective mass more than the effective stiffness. Opposite case for the edges. This is why understanding the modes shape (through Chladni, perhaps) can be so powerful.

But so far this is all just theoretical for me, so take my input with a couple grains of salt.

I have not read Trevor Gore's books, but I am familiar with work done by Alan Carruth and other in the nineties and naughts. It would seem to me that your numbers are well within reasonable tolerances. 3hz? Have you done this spectral analysis with a variety of guitars to see how much variance is tolerable and still have a beautiful sounding instrument?


Sent from my iPhone using Tapatalk

As I recall thinning the top along the side will bring the top and main down. If you do not have the binding on that could be a possibility. Binding the sound hole will bring the main down some. That maybe possible with the top on. I have considered that at some point.

_________________
_____________

EddieLee

scoober jake wrote:
"In a simplistic sense, the natural frequency is sqrt ( k/m ). So taking some mass off the top (without altering the stiffness) would increase the frequency, not drop it. "

Right, but since the mass goes as the thickness of the top, and the stiffness goes as the cube of the thickness, thinning the top will always lower the pitch of one or more modes.

If you're out to lower the pitch of the 'main top' mode reducing brace height in the middle will get you farther faster than thinning at the edges. Ditto for back modes, of course.

Relationships are far more important than actual pitches. It's good to avoid having major modes on scale degrees, of course, but a semitone higher or lower won't be too audible to most folks. See where your top ends up with the binding/purfling in place and the bridge on, and, so long as it's not in a 'bad' place, tune the rest to match if possible. Keep inmind that Trevor's numbers reflect what Trevor likes; you might like something else....

Thanks, Alan. On that note, I'm a little surprised at the ratios Trevor has there. Main top / main air = 205 / 103 = 1.99 ~= 2. I'm surprised to see that, I would expect to see those further away from an even multiple, especially 2. I would think we would want to spread those modes out a bit more. I know one rule of thumb used in room acoustics, for instance, is to use a ratio of 1:sqrt(2):sqrt(3) for room H x W x L. On the other hand, a guitar with very evenly spaced modes might end up sterile, without character. Or perhaps it wouldn't even sound much like a guitar.

Is that nearly 2:1 ratio of main top : main air commonly seen in guitars? Is that kind of data available anywhere?

Let's be clear about what I wrote in the book. The figures Jason quoted (103,205,226) are not a target but are tabulated for the T(1,1)x modes from a guitar fresh out of the building mould, top just attached and I say that the guitar at that stage is well placed at that point in the process to hit target frequencies of 95, 180 and 226Hz as a finished guitar. The guitar in question is a falcate braced medium sized guitar [my shape], not a X-braced OM, which Jason is building, for which you would want different targets if you want it to sound like an archetypal OM. The processes to get to the target frequencies are fully documented in the book. Jason's problem (communicated to me via PM) was that he copied over the bracing sizes for a large bodied guitar, hence it coming out stiffer. That, of itself, is not a bad thing if he wants his guitar to sound more like an OM. I write at length about the relationships between mode frequencies and why certain relationships seem to work better from the scientific point of view, backed up by data from historic guitars that we preferred the sound of.

Yes, it is commonly seen and recommended by some, but not me. Those two resonances (measured as fully coupled resonances of a guitar held as you would play it - how you measure them matters) naturally fall about an octave apart. That's the nature of guitars. One of my objectives is to make sure that they don't fall exactly an octave apart. High mobility resonances are the cause of wolf tones and dead notes. Having them stacked an octave apart attacks two partials of the same note. Table 22-1 in the book documents the range of preferred resonance values for each of small/medium/large SS guitars, classical guitars and flamenco guitars. Numerous scientific papers list resonant frequencies (Rossing often appears as an author) but you have to be cognisant of how they were measured. Many are measured with parts of the guitar immobilised (by burying them in sand for example) in an attempt to measure uncoupled resonances. But those numbers are only useful if you intend to bury your guitar in sand. Usually the guitars are "random guitars" i.e. not selected specifically for some area of excellence or preferential sound quality.

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

http://www.goreguitars.com.au

I did exactly like Trevor said. I thought the sizes would be close enough to have similar bracing, but I'm a rookie and will have to learn the hard way this time. I think I'll start by thinning the edges of the soundboard, and installing my binding. Then I'll tape a bridge to the soundboard, and take new readings to see how close I am to final target. If I'm no where near it, I'll just have to find away to shave braces through the sound hole.

_________________
Jason McGowan
M&S Guitars
_____________________________
United States of America, looks like another silent night
As we're sung to sleep by philosophies that save the trees and kill the children....
Casting Crowns

Are you sure you really have a problem? Even Trevor's reply "hence it coming out stiffer. That, of itself, is not a bad thing if he wants his guitar to sound more like an OM." indicates that you may not have a problem. I am not sure you have a hard lesson to learn here. It looks like you are in a position to hit around 200 on your top, your back will be about 4 semi tones higher. I would built the guitar you made and see how it turns out.

_________________
http://www.Harvestmoonguitars.com

When the guitar is finished, broken in a little bit with the string guages you want to use and the action adjusted to a fine degree, you can compare it to fine sounding guitars of similar size, woods and bracing. Then it is time to think about fine tuning the wood and the air resonances. I start by playing chromatic scales in the lowest octave and a half and listening for notes much louder than their neighbors, especially if the same thing happens playing the same pitches on different strings (otherwise it could be an uneven fret height problem). When a note is too loud, look for resonances on that note or a fourth, fifth or octave away. Look for resonances tuned too exactly to 4ths, 5ths or octaves of each other.
It is not hard to shave braces through the soundhole. I use a small curved chisel with a rounded wooden handle, also single edged razor blades. Sadly though, it is hard to make the braces look as good as they did before you put the box together. Removing just a tiny amount of wood makes noticeable changes. Take off a weight of wood equal to one split pea, string the guitar back up and try it. If you take off too much you will be very unhappy.
To adjust the air resonance to a lower note you can tape a tiny cresent of thin cardboard on to make the soundhole just slightly smaller. Unfortunately you can't try a higher air note without sanding the hole bigger.
For both bracing and air changes, keep comparing your guitar to a really good comparable guitar. I strum loud full chords and listen close to the gutsy bottom. All the different chords should have a similar impact or density in the low range.
Bring the guitar to life, then enjoy tinkering to get the best out of it.

Thanks for the clarification, that all makes much more sense to me. I am just gearing up for my first build now, but I certainly plan to buy the book after this initial cost outlay. It really sounds like a fantastic text for those interested in the physics at work within the instrument.

I hate to barge in here, but I just closed up the box on an OM sized guitar, a lightly built X-braced. Would love Trevor's comments on this. Back is thicknessed to .075"