Official Luthiers Forum!

As a guitar player (mostly finger style) the differnce between a good guitar and a great guitar is the separation of the frequencies. High's are well defined, mids are well defined and lows are well defined. Obviously dynamic range, tone, volume and playability come into play as well. But concerning note separation what do you think influences this particular quality. I have my own ideas but I am curious of other?

Thanks
Richard

Separation (taken apart from other qualities of tone and timbre) IMO is pretty much a function of damping, combined with lighter weight materials that give a fast rise time. But damping alone will do it. Take it too far, though, and you get a really dead guitar with great note separation. Recording engineers tend to like these.

I do not take strong note separation to be the hallmark of a great guitar. Too much depends on the goals of the player and builder to make that a blanket claim. But I do like to hear a continuity of voice throughout the guitar's registers. When you talked about separation in bass, mid, and treble I thought you might be describing a guitar that lacked this continuity.

_________________
Howard Klepper
http://www.klepperguitars.com

When all else fails, clean the shop.

To me the sound of a good J200 is what fits your description.

_________________
"It's a Tone Faerie thing"
"Da goal is to sharpen ur wit as well as ye Sgian Dubh"
"Sippin Loch Dhu @Black lake" ,Kirby O...

Thanks Howard, That makes sense. I come from a recording background so maybe that is why I like this particular quality in a guitar. So if I understand you correctly, damping is just describing the transient response of the instrument. More damping would be a fast attack and fast decay, like a snare drum. Slow would be something like a fretless bass. What factors in the building process would contribute to this?

I was coming at this from a different angle looking more at frequency response. Frequencies that are too close together tend to step on one another. It seems that if you could emphasize certain frequency and de-emphasize others you would gain more separation or "space" in the sound. Obviously this pure speculation. But I thought it would make an interesting topic

Richard

In terms of frequency response, I believe good seperation correlates with a reasonably 'peaky' spectrum: that is, a 'Goldilocks' number of peaks with about a 10 Db peak to dip ratio. But that's just my opinion.

OK Alan here is another question. If the 3 main modes of the top represent that top's frequency response (leaving out all other factos like back, sides, size of guitar)
then trying to manipulate these modes either by design or voicing will help with this?

Richard

Well, quick rise and quick fall tend to go together, as do slow rise and more sustain. But damping, I think, is more properly applied just to the decay.

_________________
Howard Klepper
http://www.klepperguitars.com

When all else fails, clean the shop.

I think that the thickness of the top and the stiffness of the top braces is a big factor in reducing or increasing the degree of separation. In my experience, guitars that are built more thinly tend to have a lot of overtones (I think this is similar to what Howard was pointing out with regard to 'dead' guitars having great separation).

I don't think you can look at the braced top alone, but that it is beneficial to tap tune (or voice, or whatever else someone may refer to it as) the braced top. The stiffness of the top changes a lot when you glue it up to the sides and back and this is why some manufacturers (as well as many small builders) like to sand the edges of the top to 'loosen' it up again in an effort to make the guitar more responsive.

Best,
Trev

_________________
http://www.PeakeGuitars.com

Bridge material? Separation with rosewood?

I'm not Alan, and hopefully he won't disagree, but the first three modes (free plate I assume you're referring to) is just a small part of the representation of the top. Those modes will give you some basic qualities of the top like potential bass response, and stiffness ratios. But most of the character and tonal quality of the top will come from higher modes. Unfortunately these modes are not easily (or at all) affected intentionally and predictably compared to the lowest five or six modes.

If, on the other hand, you're talking about the top on a completed instrument, then that's a whole different ball of wax, as you are talking about a system. Coupling interactions between components of the guitar complicate matters, and quite substantially change the behaviors witnessed in the free plate.

sprouseod asked:
"If the 3 main modes of the top represent that top's frequency response (leaving out all other factos like back, sides, size of guitar)
then trying to manipulate these modes either by design or voicing will help with this?"

If you're talking about the first three top modes of the _assembled_ guitar, then I think manipulating them sets up the basic tone of the box. The 'main top' mode produces most of the actual power, and the 'cross dipole' helps define the bandwidth of the 'main top' mode. As you go up you get more into tth erealm of 'tone color' rather than 'power', I think.

As Pete says, you're really looking at the results of some fairly complicated coupling, even when you think of the 'main top' mode on the completed guitar. That's really the upper part of the 'bass reflex couple', with the air in the box, and usually the back, influencing the activity and pitch of ewhat we call the top mode. It gets more and more complicated as you go up in pitch, so that, by the time you hit, say, 600-800 Hz, you're in the 'resonance continuum', where the modes are so close together that they overlap. At that point, even if you can get, say, a Chladni pattern on the top, you can't really say there's a 'top mode' at that frequency. It's just that, with all that's going on, the top happens to be what's moving, but the actual pitch will be greatly influenced by all the other parts of the guitar.

What you can say about the resonance continuum is that if there are ten peaks in the spectrum between 500-1000 Hz, there are ten resonances someplace in the guitar in that octave. They may not be at the frequencies where the peaks show up, and, again, if the sound is coming off the top that might be because the air and the back are pushing it. Also, you can look at the ratio of peak height to the depth of the dips and say something about the overall damping of the system, which is important.

It's almost impossoble to predict the frequencies of the assembled modes from the 'free' plate mode information. Too many variables, including the _shapes_ of the free plate modes, get into play.

Howard Klepper wrote:
"Well, quick rise and quick fall tend to go together, as do slow rise and more sustain. But damping, I think, is more properly applied just to the decay. "

Jansson, in his 'Acoustics for Violin and Guitar Makers', shows that it's possible to have long sustain along with fast rise and fall times. It all depends on how long the signal stays above the audible threshold.

Damping is not always a bad thing: it depends on what's doing the damping. If you've got a cardboard top that's eating sound, that's bad, and it will really cut into the high-end output. If, OTOH, you've got a nice, light, stiff top that's moving a lot of air the damping of the system will be high, but the energy that's 'lost' to the top shows up as sound, and that's what we want.

One measure of damping is 'Q value'. There are a couple of ways to measure this, but however you do one of the meanings of the term is the proportion of energy that's dissipated per cycle of vibration. That is, if you have a system with a Q value of 50, then 1/50th of the energy in the system is 'lost' for each cycle, and half the energy will be gone in about 40 cycles, if I remember my 'compound interest estimator' correctly. Thus, the higher the Q value, the lower the losses. If the Q value is really a fixed property, then this usually means that you're going to lose high end a lot faster than low end, since 40 cycles at 40 Hz is one second, and at 400 Hz it's 1/10 of a second. I'll note that high-Q systems can take a while to get going, too; so that's why high damping can be associated with fast rise times.

Hupaand asked:
"Bridge material? Separation with rosewood?"

Light bridges tend to favor the high end, and usually give better seperation, IMO. It has to do with impedance, as much as anything.

Except for that ebony favors separation because it has higher damping than rosewood.

_________________
Howard Klepper
http://www.klepperguitars.com

When all else fails, clean the shop.

[quote="Howard Klepper"]Separation (taken apart from other qualities of tone and timbre) IMO is pretty much a function of damping, combined with lighter weight materials that give a fast rise time. But damping alone will do it. Take it too far, though, and you get a really dead guitar with great note separation. Recording engineers tend to like these.

Hi, all:

The question about separation is an interesting one, and I like Howard's answer.

I would put it a little bit differently, or at least using different words. I'd frame my answer not so much in terms of damping as in terms of the rate of energy discharge of the system. This may sound like a technical difference, but bear with me.

When the strings are strummed a certain amount of kinetic energy is created-and-released by the tightness and mass of the strings as well as the energy behind the player's attack.
Such impulses are finite in the guitar: the sound stops as soon as the strings stop moving (compared with violins, where the player can saw away for hours). But until the strings do become inactive, they keep on pumping their energy load into the soundbox.

The soundbox can in turn discharge this energy budget stingily or immediately. The former will give a guitar a very sustaining sound; the latter will give a brisk, sharp, quick, zingy, or even harsh sound. Neither is better than the other except as a matter of taste. The thing about guitars that sustain a lot (simply make a chord and strum your guitar, and count how many seconds it takes for the sound to COMPLETELY disappear. This can be anything from five to fifteen seconds -- a huuuuuuuge dynamic and musical difference.

The quality of the notes gotten from a very sustaining guitar is not a very differentiated one. It's as thought one is observing a herd of animals all the same size, going the same speed and in the same direction. And there's a sort of stretched out and not very acoustic quality to these notes as well: sort of as though the player had stepped on an effects pedal. In food terms, one might even say that these notes have a sort of aftertaste.

But a guitar that discharges its energy load more quickly -- say in the five to eight second range rather than the ten to fifteen second range -- brings more differentiation of its notes to the listener's ear. The notes don't muddle each other up nearly as much.

These things have everything to do with how one builds the soundbox, of course.

Ervin Somogyi

Hi Ervin and all,
thanks for the input, interesting thread.
So for good separation kind of aim for a "sweet" sound but not too much sustain?
Makes sense.
Would be nice for a beginner to hear from others what they think the magic ingredients of a guitar's sound are. Separation, tone, volume etc.
Thanks

_________________
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 All!

My apologies Richard if this hijacks your thread - don't mean to.
It's a newbie question for Howard since he mentioned it first, but others can chime in for me too. The term 'damping' has been referred to throughout the discussion - is this a function of stiffness controlled by the combination of top thickness and bracing?

Thanks - a very interesting thread Richard!

Rick

No problem at all, the more the merrier.

R

Damping translates tonally into a material that has more or less resonance.
A sheet of brass will have very little, if any, damping. Whereas a sheet of cardboard will have a lot of damping, especially when wet…
Ebony can sound like carboard, it has a high level of damping. But it also favours the bass, perhaps because of its weight. IME guitars with ebony B&S tend to have very good bass response, but little sustain in the bass registers. That's the material I would choose for a guitar with good separation and clarity. Malaysian blackwood has the same characteristics as well.

_________________
Laurent Brondel
West Paris, Maine - USA
http://www.laurentbrondel.com/

Damping on the level of the guitar (rather than a particular part) is anything that draws mechanical energy out of the system. It can be radiating the energy as sound, or turning it into heat by internal friction. It's affected by pretty much everything in the instrument. A piece of the guitar that itself has low damping can damp the guitar by radiating sound more efficiently. More often people are talking about materials that have more internal friction and damp the system that way.

_________________
Howard Klepper
http://www.klepperguitars.com

When all else fails, clean the shop.

In most instruments internal damping counts for a lot more of the energy 'loss' than sound radiation. That's why we tend to spend time looking for materials, like BRW, that have low damping in themselves.

It occurs to me that, as is so often the case, we might be using that word 'separation' with slightly different meanings. I've heard plenty of guitars that had a sweet sound and good sustain that also had good separation in my view, but maybe other folks would not feel the same way. In those instruments you'd have to look for the cause in the spectral response: the frequency domain, rather than the time domain view of damping factor. There's a lot to sort out here.

A comment and a question. First, an excerpt from Somogyi's website:



With these comments in mind, would using good quality European spruce (as opposed to American spruce) help with note seperation?

_________________
Formerly known as Adaboy.......

Laurent, Howard and Alan,

Thanks so much for the response - I have much to learn!

Cheers!
Rick

I think, probably, not. What I expect, and hear, with Eurpo Spruce, is lots of overtones and sparkle, which may be percieved to be poor separation, but may present the notes well, and the guitar may have a balanced sound, but because of the overrtones, may not be good separation. I think separation would be a guitar with strong fundamentals, and not much in the way overtones.. But, that's just what I think.

_________________
Waddy

Photobucket Build Album Library

Sound Clips of most of my guitars

Thanks for that response Waddy.

I guess I was thinking about this statement regarding American spruce:



If it's not as clearly focused not as cleanly defined.....maybe the seperation won't be as defined. I'm asking as I've not heard Euro spruce for comparison (though my current build is using Carpathian spruce which would be classified as Euro spruce I'm guessing).

_________________
Formerly known as Adaboy.......