Acoustic Guitar Construction

Hi All, Mike Doolin here. Great thread, here's my $.02 -

The main function of the soundhole is to tune the main air resonance of the body. The guitar body with its soundhole acts as a Helmholtz resonator (although it's certainly not a theoretically perfect one, as that would require an infinitely rigid box). The frequency of the main air resonance is determined by the box air volume, the size and shape and depth of the soundhole, the flexibility and resonances of the plates, and to a lesser extent the shape of the body and placement of the soundhole. At the resonant frequency, air is pumping in and out of the soundhole at highest velocity, which can also make the plates move more if their resonances are well coupled. This is the traditional way to make a loud guitar, by tuning the plates to couple with the main air resonance.

You can demonstrate to yourself how the soundhole tunes the main air resonance. Hold the neck of the guitar with your left hand so you're dampening the strings, and thump the top behind the bridge with the palm of your hand. You'll hear a low "boom" whose pitch is somewhere in the first few frets on the low E string on most guitars. That's the main air resonance. Now cover part of the soundhole with your hand and thump again. You'll hear the "boom" at a lower pitch. If you really want to lower it, stuff a roll of tape into the soundhole. This gives you a 3" soundhole that's 1" deep, which will result in a much lower air resonance.

The main air resonance is one of the loudest and lowest frequencies the guitar can produce, and it's almost entirely responsible for the instrument's bass response. This is why we use soundhole plugs when we amplify guitars. The main air resonance is the first thing that's going to feed back, so plugging the soundhole effectively kills that resonance. But there's a limit to how low the soundhole can tune the resonance. The lower the frequency, the more energy required to produce it, and the smaller the soundhole, the more it constricts air flow.

My personal experience with non-traditional soundhole placement began with my first acoustic guitar. I knew Richard Schneider and admired his ideas, so I put the soundhole in the upper bout of that first guitar. But with my double-cutaway design, that effectively put the soundhole at the end of a tube (like a bass reflex speaker), which lowered the main air resonance to about C# below the low E. That guitar might have worked as a baritone, but for standard guitar tuning, it effectively had no bass. I retopped the guitar with the soundhole in the usual place, and everything worked fine. But that lesson proved valuable later when I designed my acoustic bass guitar - I placed the soundhole in the upper bout to push the main air resonance down into the bass range where it needed to be. And put a trap door in the tailblock so I could still get my hand in there!

It would be great to get Al Carruth to weigh in on all this as he's done extensive experimentation with soundhole and soundport placement. I remember reading an article of his where he described the effect of moving the soundhole toward the edge of the top, and it did shift the resonant frequncy, but I can't remember which way, higher or lower.

Soundports add to the total soundhole area, so adding one raises the main air resonance. That can lessen the overall bass response, but on the other hand it does let the player hear some side-to-side air modes that normally would never get out of the box. That is to say, the player hears more of something, but not necessarily more of the normal sound of the guitar. I've only done one sound port, and because of my internal neck block bracing it had to be placed below the waist. For some reason, that made it affect the main air resonance a great deal, raising it too far for good bass response. I mounted a tube inside the sound port to tune it lower, which brought the main air resonance back down close to where it was with no sound port.

All of this reminds me of a favorite expression: "Experience is what you get, when you don't get what you want".

Thanks Mike -

Given that the function of the soundhole is to tune the resonance of the body (as opposed to what most people might think, that it "lets the sound out"), it seems I have to re-think things a little. Specifically:

1. Has the optimum resonance been determined? Is there a benchmark that holds for each body style? If so, what are those resonances?
2. How can a small shop luthier measure such resonance?
3. Would it be a good practice, given that one knows the target freqs and has a method to accurately measure them, to start with a smaller soundhole, measure, enlarge, etc. until the target freq is achieved? Or are standards already engraved in stone, and there is no good reason to tune them further?

Actually this leads to a whole lot of speculation on my part, it's a very interesting discussion, but answers to the simple questions above would be a big help.

Anyone want to take a stab at answers?

Thanks
Dave

It turns out that the pitch of the main air resonance is relatively insensitive to changes in soundhole size and box volume. Most steel string guitars have a 4" soundhole, and even though the air volume varies tremendously between the largest and smallest (say, a Dread and an OM), the air resonance still falls within a pretty narrow range (around F# for a Dread and A for an OM). Classical guitars have a soundhole of about 3-1/4" and much lighter bracing, so their air resonance is usually around F# like the much larger Dread. That's an example of matching soundhole size to box size to get a desired air resonance. I haven't tried it, but I'd speculate that if you put a Classical size soundhole on a Dread, the air resonance might be too low, below the low E, so you'd effectively lose bass response.

I use a Classical size soundhole on my OM model to get the air resonance down to about G, but note that a 3/4" difference in soundhole diameter only moves the air resonance by about a whole step.

Conversely, my John Stowell model is identical to my Classical model, except the body depth is 2-1/2" instead of 4". And yet, cutting the box volume almost in half only raises the air resonance from G to A, again about a whole step.

The upshot of all this is, the main air resonance is pretty much built into the design of the guitar, and there isn't a lot you can do to tweek it once the guitar is built.

Also note that the main air resonance is a fairly "soft" resonance, with low Q. That means that while it does ring at a particular pitch, it will reinforce all the notes for a few half steps around it. For example, if it's at G, you'll get some support all the way down to the open E and up to A or Bb. On the other hand, if it falls too close to the actual pitch of a fretted note, that note can either be very short on sustain or even warble with a beat frequency (the dreaded "wolf tone").

To answer your questions Dave -

1. I guess the established sound of each style of guitar is determined in part by their typical air resonances. Dreads sound big abd beefy (if a little hollowed out in the mids) with their low air resonance, while OMs sound balanced and focused with their higher air resonance.
2. Thump the top with the strings muted, find the pitch.
3. It might be reasonable with a new design to experiment with soundhole size. But remember, it takes a pretty big change in size to significantly shift the air resonance, so it's probably going to be a matter of finding the size that fits the design and then sticking to it.

Thanks - that gives me quite a bit to work with.

Interestingly, in a troubling way, I have a Breedlove OM - traditional style - that has a warble in the open e string. Irritating as heck.
How to approach a fix for that?

There ya go. I bet if you thump the top of the Breedlove you'll hear an air resonance close to, but not quite exactly, the pitch of the open low E. How to fix it? You'd probably want to raise the air resonance since it's pretty low to start with. That would mean enlarging the soundhole, reducing the box volume, or stiffening the plates. Of those, the least destructive would probably be to add some bracing. The first thing I'd try would be to add a back brace. You could make a brace with a concave radius and double-stick tape it to the outside of the back to test the concept. You could do the same thing with top braces, although adding top braces will likely change the overall sound more than back braces would. Anyway, it's the easiest and most non-destructive way to see what happens, you'll certainly learn something if nothing else. Then if it does work, glue the braces inside.

You could also lower the air resonance by inserting a ring inside the soundhole. You can reduce the diameter, increase the depth, or both. This is also a non-destructive test you can do before making anything permanent. Make a ring out of layers of heavy cardboard that is a tight friction fit in the soundhole. You can make it the same thickness as the top to play with the diameter only, or just use one layer of cardboard and vary the depth. Once you determine what works, you can make a nice laminated wood veneer ring that either friction-fits in the soundhole or is glued inside to the top just out of sight.

Let me know what you find out, I'm sure you'll learn some really interesting stuff, even if it turns out there isn't a practical fix.

Outstanding. This will be fun. I meant to say it was the high "e" string, though. But you gave me plenty to work with, thanx.
Dave

I'm wondering how I will get my hands in the uke I am building. Maybe I can teach my sons how to tighten a wing nut! They have small hands! Or my wife could do it. But I suspect the training will take longer!