Official Luthiers Forum!

The thing you have to realize about the guitar is that the designs we have are very highly evolved. This means, essentially, that if you do a reasonably good job of building and use decent materials, you'll end up with something that's objectively very nearly as good as the best. The trick is that, when you're dealing with such good designs, small differences become very important. The differences beween a WalMart box and the best guitar you ever heard may be 'small' in terms of measurements, but they certainly matter.

Another thing you need to understand is that the guitar is a whole thing, not just an assembly of parts. We often talk about how thinning a certain are of the top might bring up the bass, or whatever, but it's more complicated than that. For instance, if you wanted to get more treble you could:
1) enlarge the soundhole,
2) add a port,
3) make the top thicker,
4) make the bracing heavier and the top thinner,
5) make the body shallower,
6) use a smaller pattern,
7) use a different wood on the B&S,
8) use a smaller or lighter bridge,
9) use a longer bridge,
10) use scalloped bracing (see Proulx),
11) use tapered bracing (see Carruth),
12) thin the 'wings' of the top outward from the bridge ends,
13) etc.
Each of these things will get you more treble in some sense, but each one will effect other aspects of the sound as well. What do you mean when you say 'more treble', and how much do you want?

My opinion, FWIW, is that the 'best' results are gotten from achieving a sort of balance, so that everything works together. Again, the best designs have been arrived at through a lot of trial and error so that they get you pretty close to that more or less automatically. From there it's all fine tuning, and most of the time 'fine tuning' is _all_ you can do once the thing is together. Again, that's not to minimize fine tuning: it's often the difference between 'adequate' and 'great'. But, given 'the evils that lurk beneath', as has been said, ther'es only so much you can do most of the time. If the guitar is not at least 'pretty good' to begin with, you aren't likely to get it to 'great'.

You also have to remember what Dante said: "The closer a thing is to perfection, the more it feels of pleasure or of pain". Great guitars are often balanced on the head of a pin, so to speak, and even very good ones are more likely to be harmed by small interventions than helped. Adjustments that barely change lower-grade instruments would as likely as not be ham-handed on 'better' ones. For all these reasons, then, it's hard to learn much by modifying a less expensive instrument.

David Malicky wrote earlier in this thread:
>>>As a learning exercise, I did a similar experiment on a heavily-built import 000. I did find it valuable, and wrote it up in post #50 here--
http://www.acousticguitarforum.com/foru ... 598&page=4
(I wouldn't take my findings as generalizations, though. And I'd suggest Audacity for the frequency measurements.)

That said, if your RK 000 is indeed lightly built, I wouldn't suggest shaving it. The wolf tone can be changed by adding a small mass to the top -- search the archives for Alan Carruth's posts on this.

Is the Epiphone an AJ500? If so and similar to mine, those are definitely lightly built and I'd not suggest shaving it either. My AJ500 has a 0.105" top and deep scallops in the Xs (1/4" tall) -- that combo is pretty unusual for a 16" factory guitar.

As for direct ideas/instructions, there just isn't much out there in the sense of "shave ___ and you'll get ___", because there are so many variables in each guitar/design. Yes, the stiffer the top+bracing, the more treble and less bass. The archives have some great info from Alan C on selective thinning of the perimeter, the effect of X-scallops, and main air and body resonances (which affect tone and are affected by the braces). There do appear to be "sweet spots" in these resonances -- I can send more info if you want.[/quote]

I saw that thread and it was the genesis of my getting my hands inside the guitars, so thanks for the inspiration (I think ). Good stuff.

How did you get the camera in there with the strings on?

Alan,

I've been reading your past posts on this and other subjects and I'm overwhelmed by the breadth of it. Thanks for having it up for me to study . I've just begun reading the textbook you'd linked to in another thread and it looks as though it'll answer about a thousand of my questions. Here's the link in case anyone else missed it:

http://www.speech.kth.se/music/acviguit4/

Some of the things I've learned apply well to already finished instruments, such as the idea that the interior of the guitar should be finish sanded well to better focus the interactions of sound inside the instrument. Rough hewn woods won't reflect sound well, in an intended, directional fashion and that helped both of my guitars. Not that I did that at all perfectly, but they're much better now than they were before.

This particular post of yours I found is a great answer to my original question and I thought I'd repost it because it deserves a second showing:

>>>I think the location for the peaks on scalloped bracing have to do with controlling the frequencies of the modes above the 'main top' mode. Basically, you're working with the stiffness of the top/bracing system in different areas, and the effect on the mode frequencies will depend on whether the brace is bending or not in that location for that mode.

The lowest four modes of the top do most of the work in producing sound, and particularly the 'main top' mode, which is thought by some to be the most important sound producer all the way up to 1000 Hz. That's the mode that works like a loudspeaker, with the whole lower bout moving in and out.

On loudspeakers they make the edges really loose, and try to get the center to move like a piston, with no bending. We can't do that: you need a certain amount of stiffness at the edges, and anyway, we're not trying for the sort of 'flat' frequency response the loudspeaker people are. One researcher, Richardson in Wales, has shown via computer models that a top that is loose in the center will actually move more air in the 'main top' mode than one that is loose around the edges, assuming both tops are stiff enough to take the loads. Scalloping reduces that central area stiffness, and thus accentuates the power of the 'main top' mode, as well as the 'main air' mode that it drives.

The next three modes up also contribute to the tone, even if they are not as effective at producing power as the main top mode.

From what I can see, the 'cross dipole', which is usually the next higher one in pitch, tends to be a 'loser': it's so much less effective than the 'main top' mode that you'd rather not have it stealing energy. You do this by making the areas of the top on either side of the centerline, about halfway to the edge, stiffer, to move the cross dipole pitch up away from that of the 'main top' mode. Meyers (iirc) showed that guitars that had a narrow 'main top' peak in the spectrum tended to sound 'harsh', and too low a cross dipole pitch contributes to that by stealing energy and cutting off the main top mode. So, having the X braces peaked outside of the bridge wings can raise the cross dipole pitch, and make the 'main top' spectral peak broader. I've noticed on the guitars that I've measured that this seems to contribute to a 'full' or 'solid' sound.

The 'long dipole' could be a 'waster' like the cross dipole, but it often couples strongly with one of the inside air resonant modes, and ends up producing a significant amount of sound; often around A on the high E string, where nothing much else is going on. Loosening the area in the center of the top behind the bridge with scalloped tone bars will drop the pitch of the top 'long dipole' mode.

The fourth mode, the 'cross tripole' may or may not show up on any given steel string guitar. It's a power producer, though; generally up around 500 Hz (C on the high E). The peaks on the lower end of the X ,and the tone bars, are in places that probably help to move that one upward in pitch.

The important thing to think about is that each system of shaping the bars tends to establish a certain relationship between the mode pitches, and thus to form the spectrum of the guitar's output. That's why scalloped, tapered and flat bracing tend to sound different.

Also, of course, there's no free lunch. Loosening up the top in the center is structurally risky, and also can, if not done right, give too much bass and less sustain. All of this is pretty complicated, and good, stable, balanced, and 'clear' sounding instruments can be built with all sorts of bracing. Still, knowing the strengths and weaknesses of each pattern, and how they tend to sound, can help you build what the customer wants, even if that's just you.<<<<

That's only true until you get those, "lower-grade instruments" tuned to the point of a thoroughbred. Then they'll .....

Does this mean you're not going to sponsor the anual "Hot rod your Esteban competition?" It would be great. Simple rules, you can't change parts, just shave braces, plates on the inside, and add dead weight. All the guitars will look the same. Nobody will know who perfected them. Judging will be fair. You could run it like the TV show "pink slip", winner gets to keep all the guitars!

John

Alan, Thanks for another great post.

Danny, Glad you liked that thread. That 000 was way overbuilt so an easy candidate for shaving. All but the one of those pics were with strings off. And that's indeed an terrific post of Alan's from the archives. On your RK, can you describe the problem more specifically ("doubling up of harmonics at B through E") -- that sounds different than a classic wolf (a lack of sustain for a single note).

_________________
David Malicky

David,

The sound I'm assuming is a wolf tone is a sort of runaway doubling up of the harmonic content of the plucked B string on the third fret. It's the main body (main air?) resonance pushing the note much more loudly than the rest (maybe a 5db jump on a middling force pluck). The guitar is 'lively' and loud, with loads of harmonic content but is, on that note especially, lacking in balanced control. If there is a way to make it more tilted towards the bass end of the spectum, I'd be ecstatic with it.

Playing with the quarter today, I found that putting the weight on the top pretty much right on top of the main-X brace intersection tamed the forward jump of the sound pretty well (It didn't quell it even nearly entirely though) and made the bass sound more evenly mixed and the guitar sound more as I expect it should. That's the area between the soundhole and the front of the bridge, under the strings.

Please don't think I am trying to talk you out of it.... It would be totally hypocritical of someone who just can't leave well enough alone himself.... John Hall tried to talk me out of it twice.... He didn't succeed.... I was just trying to share what I learned from my own experiments..

Since you seem like the curious sort -- I would encourage you to have a shot at it....

What's the worst that happens? You wreck the instrument, and then get the opportunity to rebuild it..... That's not actually a negative thing, really.... It can be quite fun and the results can really surprise you....

As they say, you can't make a cake without breaking some eggs.

Now... On the idea of re-voicing an Esteban.... I won't vote for keeping the original top.... My top was thick and heavy plywood... Full of glue filled voids and hollow spots... Not exactly the ideal basis for learning about voicing.... Retopping it was the best thing that ever happened to it!

Thanks

John

Danny like John and a number of others on the forum I confess to being a chronic fiddler but instead of a bow I use a fingerplane I was not happy with some aspects of the sound of my 3rd build so proceeded to carve both back and top bracing . The guitar went from ok to dead and back to sort of good , over a period of six months. I must say that I did gain some understanding through this process but not sure I want to do this regularly. When you do make changes and John said this in an earlier post, I recommend waiting a number of weeks for the instrument to fully settle before changing further.
All the best.

Craig.

something like this might help:

http://www.youtube.com/watch?v=pC-49Gnj ... re=related

_________________
Walter Lay
"It's taken me so long, but now that I know
I can see. All that I do or say, is all I ever will be"
- Billy Joe Shaver

"The sound I'm assuming is a wolf tone is a sort of runaway doubling up of the harmonic content of the plucked B string on the third fret."

My pitch/frequency chart says that's up around 277 Hz; well above the 'bass reflex couple' between the top and air on most guitars, and generally out of the usual wolf territory. Which is not to say it's not a wolf, it's just not the usual one. I can think of a few things it could be, but would really need to get the gutiar in hand to start working on this one.

What could it be?

Here's one guess....

Main Top or back resonance that's a few cents off..... causing "Wobble" at that frequency.. but it's a blind guess (Just that sometimes, the main top or back is around this frequency)

Honestly, your best bet is some Poster putty, Modeling clay, or some double-sticky tape and a few coins...

The weight will lower the frequency of the plate where you stick it... and you can drop the frequency off of the "Wolf" once you find which part is causing it....

Another thing to do is "Hum" into it and physically feel where it vibrates.... An extremely light touch will tell you where it is vibrating...

Once you know where it is -- you can either stick a weight to that area, or shave a little off... Both drop the frequency a bit when done right.

Thanks

John

At 277 Hz?

Danny, There are better people than I for these issues (Alan C is the best, at least that I know of), but you might try this test to see if it is a air/top/back resonance issue, or something else like a string buzz. If I'm reading my chart right, the 3rd fret of the B string is a D at 294 hz -- Do you get a similar problem with the 7th fret of the G string (same D)? Also, try tuning the B string up or down a few half steps and see if it is consistent on that D frequency, or related to that fret position.

_________________
David Malicky

Sure, why not? It's within the plausible range you might get out of a factory instrument...

The Retop Esteban Dread had a main back around 260 hz and a main top in the 160's
My Tak GS330S had a main back around 290 hz and a main top in the 190's

Both of these were taken through an Apex USB studio mic through Audacity.

What's your suspicion?

If I was trying to fix it -- I would be busting out the poster putty, though...

Thanks

I just haven't seen anything like a wolf note up that high, which doesn't mean that it can't happen. I've come to expect something more around the first few frets of the low E string or maybe up a bit on the A. When I get funky sounds up further it often turns out to be a saddle or nut issue.

I've only seen Estebans on TV though. If you're playing with one again I would love to get about a 3 second wav file (44K or so) of the low E string and a similiar recording of a tap on the bridge with the strings slack and damped. I would be curious to compare that with other guitars.

John

John Platko had to ask:
"What could it be?"

Beating between some higher-order top/back/air modes would be my first guess, and yes, that's pretty non-specific. It's possible that the second Helmholtz-type mode, the one that's only active in the lower bout, is that high in pitch, and that could be coupling with something like a back mode, or even the top cross dipole if that's asymmetric. The point is that, without the guitar in hand to test, I'm just making WAGs like everybody else. In the absence of data it's hard to even make useful suggestions.