Page 1 of 1

A guitar that rings - how?

Posted: Wed Nov 29, 2017 3:45 pm
by Doug Shaker
I am planning a steel-string for a friend of mine and he particularly likes it when his guitars "ring", that is, when they continue to make sound for a long time after his last strum.

I am trying to think of what characteristics will lead to a guitar that rings. I would say a light bridge and a light top. I try to do those in any case, but I suppose I would want to push the lightness a little more than usual in this case. Anything else I should be thinking about?

Thanks!

Re: A guitar that rings - how?

Posted: Wed Nov 29, 2017 6:48 pm
by Peter Wilcox
I'd think anything that bleeds the string's energy away from the top would decrease sustain, so my thoughts would be a stiff neck, solid neck to body joint, dense nut and saddle (like bone, not plastic), and properly cut nut slots.

Re: A guitar that rings - how?

Posted: Wed Nov 29, 2017 9:24 pm
by Mario Proulx
The term you want to use is "sustain".

Build it light and responsive...!

Re: A guitar that rings - how?

Posted: Thu Nov 30, 2017 8:59 am
by Brian Evans
I don't know on a regular flat-top, I usually think of "ringing" in the context of sympathetic overtones. I get a lot of that from the strings behind the fret and behind the bridge on my archtops, to me it's a big part of the acoustic tone. Really neat when you play a note that happens to coincide with a string pitch between the saddle and tailpiece, the note rings on well after you've damped the actual string with your fingers. Somogyi discusses sustain a lot, suggesting that it might be one of the more difficult things to achieve. I recall reading that he actually times the sustain when he is evaluating a guitar. Good luck with your project - building light and responsive is always a good thing!

Re: A guitar that rings - how?

Posted: Thu Nov 30, 2017 1:12 pm
by Freeman Keller
Brief anecdote. I own a Martin dreadnaught from the "over built" era of the 1970's. When it needed a neck reset I had the work done by a tech who is noted for doing what he calls "hot rodding" on these old guitar. I had the heavy rosewood bridge plate replaced by a small maple one and some scalloping done thru the sound hole. When the guitar returned my wife's comment was "you are playing louder". The guitar has been passed around the circle at some bluegrass jams and the feeling is that it stood up very nicely against some valuable vintage instruments.

Sustain is easy to measure but hard to quantify, but lightness seems to be your friend.

Re: A guitar that rings - how?

Posted: Thu Nov 30, 2017 2:53 pm
by Randolph Rhett
I don't know if you have made hundreds of guitars, but building for a specific tone is voodoo that only the most experienced builders ***maybe*** can do. If you have less than, say a dozen, under your belt don't worry about crafting a particular tone. Chances are any guitar you make, provided you make it with care and attention, will have more "ring" than any factory guitar your friend is likely to have played. Build him the finest guitar you can in the size and shape he likes and I can almost guarantee he will be satisfied.

Re: A guitar that rings - how?

Posted: Thu Nov 30, 2017 5:07 pm
by Doug Shaker
I would, in any case, try to build pretty light. I've made my beginner's overbuilt guitars and now I need to err on the other side so that I know where that line is.

The suggestions that are operational, at least for me so far, are:
-test the soundboard for stiffness (using tap methods from Gore & Gilette) and pick one that will build out as a light soundboard
-make the neck stiff - possibly using inlaid graphite struts
-use a bone saddle and nut
-be aggressive about trimming the braces
-make a few candidate bridges and use a pretty light one

That's enough of a plan for me now, though I am happy to hear other suggestions.

Re: A guitar that rings - how?

Posted: Thu Nov 30, 2017 11:28 pm
by Bob Gramann
To my ear, East Indian Rosewood imparts a chimelike sound to a chord that hangs on as the volume decays. Essentially, it adds some harmonics.

Re: A guitar that rings - how?

Posted: Fri Dec 01, 2017 1:06 am
by Trevor Gore
Doug, what matters are high impedance mismatches everywhere you don't want a wave to go, like out of the top and down the sides. Keeping the bending waves in the top keeps it ringing. High impedance mismatches reflect waves. Mass helps with an impedance mismatch more than stiffness does. So think about your linings, mass in the sides, low damping materials, live vs. non-live backs and all the other similar ideas and techniques explained in the books.

Other things help also with "apparent" sustain. Like the sound starting loud and hanging in. So high monopole mobility is your friend there. Other things that are less obvious are "intuneness". If you can get more fretted notes and harmonics to be truly in-tune, that helps, too. So look at the sections on nut and saddle compensation.

Building to a "target" is always a fun challenge and is why I like making the wide variety of guitars I do. Pushing out e.g. dreadnoughts all the time isn't my idea of fun. I'm sure you'll have fun with this one!

Re: A guitar that rings - how?

Posted: Fri Dec 01, 2017 12:14 pm
by Barry Daniels
That might explain why a Martin D-35 that I recently made a fully compensated nut and saddle for now sounds louder to the client.

Re: A guitar that rings - how?

Posted: Fri Dec 01, 2017 5:40 pm
by Alan Carruth
One definition of sustain is how long the note stays above an audible threshold after you play it. If you think about it, there are two kinds of sustain; what you could call 'Les Paul' sustain, and 'banjo' sustain. A Les Paul solid body gets a lot of sustain by keeping the energy in the strings, only allowing it to leak out slowly. It does this by using a very rigid and heavy body and a massive bridge. The sound is quiet, stays at the same level for along time, and it's quiet. Did I mention that it's quiet? Banjos have a sound that builds up fast after the pluck and falls off fast, but it can reach such a high level at the peak that it stays audible for a fairly long time. They get this by using the lightest possible bridge and soundboard, and making the soundboard flexible as well. The sound is punchy, edgy, and loud.

The guitars we make don't approach either extreme, but you still have a choice of which way you want to lean. Scalloped braced tops tend toward the 'banjo' end, while a top with 'tapered' bracing will be more on the Les Paul line. Scalloped bracing tends to favor the bass and 'punch', although when it's properly done you can also have very nice trebles. Often they need to use a fairly heavy bridge to control the tendency to a low-frequency wolf at the 'main air' pitch because of the tight coupling between the top and the air. A top with tapered bracing concentrates it's stiffness in the center, and so avoids that problem. Because of the stiffness of the top and the fact that you can use a light bridge the tone favors the mid range and treble, and can deliver quite long sustain. One of my customers, Ken Bonfield, says that his biggest problem in recording is holding his breath at the end of a number until the sound dies away.

So there's lots of good advice so far, and actually most of it fits together well, if you think about it. Look at the style of music your friend wants to play, and see what sort of 'sustain' that calls for. Then go for a top that's light, with the bracing and bridge mass tailored to give the sort of frequency response you'll want. Try to get the top working as well as it can within itself; no big lumps or dips in the bracing, for example, beyond what the style calls for. Don't go for the lightest wood on the B&S; stay in the rosewood class, rather than something light and soft. Low damping can be a real help too, I think, which also says 'rosewood' or a surrogate, such as locust or Osage. As Trevor says, get the intonation right, too. I would not worry so much about an extra-stiff neck, but do make it stable.

Re: A guitar that rings - how?

Posted: Sat Dec 02, 2017 10:54 am
by Clay Schaeffer
As Trevor mentioned:
"Mass helps with an impedance mismatch more than stiffness does."
As Alan mentioned:
"Don't go for the lightest wood on the B&S"

I do two things that seem to help with sustain - I use a pyramid style African blackwood bridge (heavy but with smaller footprint) and stiff sides and a reflective back. The bridge seems to give enough mismatch to where the energy doesn't leave the strings all at once, and the reflective (as opposed to "live") back and sides don't suck the energy out of the top so the guitar is still pretty loud. Extremely lightly built guitars seem to vibrate everywhere - including places that don't push the sound out.
For a guitar to "ring" it should have volume and sustain.
On the other hand for a fingerstyle guitar you might build with a live back for more overtones and less sustain - you don't want it to ring on like a piano with the dampers held off.

Re: A guitar that rings - how?

Posted: Sun Dec 03, 2017 4:48 pm
by Alan Carruth
Keep in mind that impedance is frequency related. Technically it's the ratio of force/velocity at a particular frequency. If you want to raise the impedance of a particular structure, you can either add mass or increase stiffness. Both will ,ake it harder to move the thing, but the effect will be different at different frequencies.

As an example, we can look at swapping out bridge pins. If you change the stock plastic pins (3 grams or so) for a set of brass pins (30 grams), you will have raised the impedance of the bridge/top system: it will be harder to move at all frequencies. However, the effect will be a lot grater at high frequencies, which is why it' often said that adding mass at the bridge gives 'more bass'. Relatively that's true; there's less sound all the way up, but since most of the loss ins in the treble the guitar sounds 'bassier'. Now, if you swap those brass pins out for bone (8 grams), you'll gain output all around, but more in the treble. Stiffness, of course, works the other way; adding stiffness at the bridge tends to cut down on sound overall, but more in the bass, so it gets 'brighter'.

Impedance is always lowest at resonant frequencies. Since it's generally easier to measure those rather than impedance we use resonances as a sort of proxy. Since the strings generally have lower impedance than the top most of the energy will tend to stay in the strings except near top resonant pitches. One thing you want to avoid is having the top impedance drop too close to the string impedance at any played pitch: this is the formula for a wolf note. That's why you avoid having low-order top or air resonances close to played pitches. Trevor goes into this in some detail in his books. The top resonance the strings 'see' the most is the 'main top' or 'monopole', but others can get into the act from time to time. Typical scalloped bracing, with the braces low at the bridge location and with peaks near the outer ends, tends to drop the 'main top' or 'monopole' pitch and raise the pitches of the long and cross dipoles, as compared with 'straight' bracing. On a scalloped braced guitar you tend to use a heavy bridge to raise the impedance that the strings 'see' to avoid wolf issues. 'Tapered' bracing does more or less the opposite; the stiffness is high in the middle and thus so is the impedance, and the 'main top' resonant pitch can be higher, so low frequency wolf notes tend to be less of a problem, even if you use a light bridge. The response curve is different at different pitches. Either setup can give you a nice ringing sustained sound, but emphasizing different pitch ranges. You put down your money and take your choice.

Re: A guitar that rings - how?

Posted: Sun Dec 03, 2017 8:33 pm
by Doug Shaker
Trevor,

I'm thinking about your suggestion to increase the impedance of the sides. However, I usually build with a live back. I took a look at the design book and tried to figure out the connection between high impedance sides and a live back.

I had thought the possibilities might be:
1) whoa, useless. If you have high impedance sides, the energy will never get to the back. But I suppose you could do it if you wanted.
2) counterproductive. If you are trying to keep energy in the top, build with high impedance sides and a dead back.
3) perfectly valid. The high impedance sides will help, live back or dead back. Go ahead and try high impedance sides and a live back.

To my surprise, I read the design book as saying that high impedance sides and a live back could work. Am I reading this right?

Re: A guitar that rings - how?

Posted: Mon Dec 04, 2017 5:21 am
by Trevor Gore
Doug Shaker wrote:To my surprise, I read the design book as saying that high impedance sides and a live back could work. Am I reading this right?
Well, everything works to some extent, but, yes, high impedance sides and a live back work well, compared to run-of-the-mill guitars.

As ever, you can't look at anything on a guitar in isolation. So let's check what's going on.

Dense, deep linings will help keep the wave in the top, due to the impedance mismatch. Heavy sides will work the momentum equilibrium of the top, (Fig 1.7-3 and Section 2.3.12.2) to make the guitar louder. A live back at the right pitch separation from the top will couple pneumatically with the top due to the pressure changes in the box. So you can both have your cake and eat it. You won't get quite as much volume as you would with a non-live back but, well executed, you'll have more than the majority of factory guitars, and you will gain the advantages of a live back. If you use CF in the bracing, you can lose the wood mass that you would have had to put in to guard against cold creep and also gain a bit of stiffness, so you can keep the monopole mobility high (another way of saying high average admittance, measured in a particular way by observing the monopole). That will give you high responsiveness and volume potential.

So you can see how much performance the more conventional designs "leave on the table". If you execute the books ideas well, your customer should end up with a fine guitar. Be sure to let us know how it all goes!