How Does Soundhole Size Affect Tone? - created 12-22-2007

Behar, Joe - 12/22/2007.17:17:15
Toronto Canada

If I decrease the diameter of my sound hole by 10% will my OM be louder or softer? Will it have more or less treble and/or bass response?

Can anyone shed some light on this topic?

It's a little more complicated than you might think, so brace yourself. The size of the soundhole sets two different pitches, one high and one low. It also establishes the rate at which sound is radiated from the air in the box into the room. Making the hole 10% larger in diameter (which I'm assuming, rather than larger in area) will alter all of these things enough to hear.

The low pitched resonant frequency that is set by the sound hole size is that of the 'Helmholtz' mode. This is what you hear when you blow across the mouth of a wine bottle. The larger the hole is the higher that pitch will be. The problem is that you only get a real 'Helmholtz' resonance on something like a bottle that has rigid walls. The flexing of the top and back of the guitar changes the dynamics, with the upshot that the 'main air' resonant pitch on a guitar, which works the same way as the Helmholtz resonance, will generally be at a lower pitch than you'd expect. Making the hole on a guitar bigger may not raise the pitch of the 'main air' resonance much, simply because it changes not only the 'real' Helmholtz pitch, but also the interaction between that and the top and back. In other words, one would expect the 'main air' pitch to go up, but it's hard to say by how much.

The 'main air' resonant mode makes a nice peak in the spectrum of the guitar, usually right around G on the low E string. This gives you the fundamental of the lowest notes, contributing to the 'fullness' of their tone. Raising the pitch of the 'main air' mode shifts that peak upward. The lowest notes tend to become a bit more 'nasal', as they have less fundamental content. OTOH, the A string may come out more strongly.

Because there is less edge relative to the area of the larger hole, there is less drag for the air flow through it. This means that more air can be pumped in and out of the guitar, so the low range has more power. How this will sound in a subjective sense is hard to say in advance: on the one hand, the upward shift in the pitch 'should' make the sound less 'bassy', but the greater power might make up for that. All you can do is try.

The lower loss factor at the soundhole also cuts losses at the 'main top' resonant pitch, almost an octave higher, near the pitch of the open G string, since the two are strongly coupled. Subjectively this often comes across as an 'edgier' or more 'forward' sound. The lower losses also mean that the sound is less 'trapped' in the box, if you will; it gets out into the room more quickly, so there can be less sustain and a less 'smooth' or 'even' sound.

The other pitch that is set by the size of the soundhole is that of the 'upper cuttoff frequency'. A small soundhole is like a small speaker on a radio: it's not very efficient at radiating low frequencies. The only reason it is at all effective in setting up the low range sound of the guitar is that there is relatively a lot of power available: like swatting flies with cannon, effecive and efficient can be two different things. Anyway, as you go up in pitch the soundhole gets to be more and more efficient, until, at some point, essentially all of the power in the box at that frequency or above is radiated immediately. This is the upper cuttoff frequency, and for soundholes in the 'normal' range for guitars it's somewhere between 3000-5000 Hz. The larger the hole the lower the upper cuttoff frequency. This is a frequency range where your ears are very sensitive, and increasing the power output up there even a little could make a real difference in the timbre. We don't know much about this yet: it's hard to think of how to do experiments that isolate this parameter, and I'm not sure anybody has tried. Things are really complicated up in this range.

The upshot is that I usually expect enlarging the soundhole a bit to make the guitar a little louder, a bit less 'full' in the bass, and a bit 'clearer'. The sound is usually more 'forward' or 'carrying', but less 'even'. Some or all of these things might not happen for you, and, at any rate, your impression of what happens might be different from mine: we all have different ears. The physics is always the same, but that doesn't mean we all hear it the same way.

If you're really unsure, you could hedge your bets. Make the hole the 'normal' size, but leave a wide rim between it and the inside of the rosette. When the guitar is together trim material off the edge of the sound hole a little at a time until the sound stops getting 'better', or you run into the rosette.

Alan Carruth / Luthier

Alan is the guru of resonances, so if you can understand what he's saying, listen to him. Even though he writes very clearly... I used to sort of glaze over and run away in fright when he started using words like Helmholtz. but I think some of it is starting to soak in.

Joe, you did mean to say "decrease" right? I built a Jumbo size guitar with a 000 size 3-7/8" sound hole, not realizing that it probably should have increased commensurate to the size of the guitar. I think my soundhole is too small and my tone sounds a little "compressed" like a Dobro. So I have a negative impression of what a smaller soundhole will do. I have no room to cut a bigger hole and to get closer to the rosette, so I plan to cut in a port in the shoulder area.

(I just did some repairs on a Guild Jumbo 12 that had a 4-3/32" sound hole)

So I have a question too. I want to install a port in the shoulder with an area which increases total area of the two soundholes (main + port) to that of a 4-1/16" circle. Will it have the same effect or nearly the same, as if I had cut the main soundhole to 4-1/16? I want to get rid of that compressed sound. I plan to start with a smaller port and make it bigger if it sems to be helping.

Santa Cruz makes one called the tony rice pro model that has a 10% larger soundhole and they claim it has a stronger response in the upper ranges.

I don't make anything with a soundhole smaller than 4-1/4". Not even my size 0. I go up to, and a bit over, 4-9/16" at times. I just shipped out an OM with a 4-3/8" soundhole, and a large soundport. Didn't suffer the least for it...

Larry, I find over time I understand Alan's posts better and better. Keep reading and they'll start to open up the way repeated practice makes knowing the notes above fret five open up.

Alan, I've a couple questions, perhaps basic.

First are the Helmholz, rumjug, and main air the same measurement? You're saying the main air increases more slowly than the Helmholz, but if you sing across the soundhole are you also hearing the main air or is it lower than what you are hearing?

Second, can you explain the reason why the pitch goes up when the hole gets bigger. That seems counter-intuitive to me. I think I understand why a stiffer soundboard increases in pitch and also why a lighter soundboard increases, but not this.

If you started with a pinprick and ending with a really big hole, would there be a sweet spot that was detectable? As you mentioned could one keep making the hole bigger until passing the sweet spot,then bind the edge of the soundhole to get it back?

Thanks for taking the time to post, you've got a big following out here.

can you explain the reason why the pitch goes up when the hole gets bigger

Sing a low note. Now, sing a high tenor note. Notice what you had to do with your mouth to achieve each note.

If I try I can sing both notes without moving my mouth at all. My throught however seems to squeeze smaller on the high note and open up on the low note. Which Is why it seems counter-intuitive to me. I know resonance is not exactly the same thing as generating the tone, though.

Is it just that I'm losing mass on the soundboard with the bigger hole?

I can sing both notes without moving my mouth at all ///snip//// squeeze smaller on the high note and open up on the low note

Dude, tha's just plain, weird

The rest of us make our mouth smaller for low notes and larger for high notes.

If I had little guitar strings in there I'd start to see the relevance, but since I am a wind instrument you'll have to spell out what my mouth and the guitar soundhole have in common.

I'm glad no-one is around while I sing scales to the computer, but it seems that the shape of my mouth changes according to what vowel I am singing more than what pitch. Maybe that's why they didn't tell me when they changed the time for choir practice.

I used to run away screaming when this stuff came up too. Then, one day while I was at one of Carleen Hutchins' workshops, a couple of us got talking about 'resonance'. One of the people there asked us to define the term, and we couldn't. Embarassing....

Bill Lierer asked:

":First are the Helmholz, rumjug, and main air the same measurement? You're saying the main air increases more slowly than the Helmholz, but if you sing across the soundhole are you also hearing the main air or is it lower than what you are hearing? "

I guess it depends on how you define things... ;o)

Strictly speaking, you only get a real Helmholtz resonance if you've got a bottle with rigid walls, preferably a spherical one, and it has a neck. A rum jug comes pretty close, so that counts. The classic model of the Helmholtz resonance is that there's a certain amount of air in the bottle that can be compressed, so it acts like a spring. There is also a mass of air in the neck of the bottle, which acts like a piston, and bounces on the spring. The resonant frequency depends on how heavy the 'piston' is and how strong the spring.

Since a guitar is not a rigid spherical bottle with a neck it doesn't have a 'real' Helmholtz resonance that you can hear all by itself. The 'air spring' pushes against the walls, and since they are flexible, they change the strength of the spring. In fact, since they have resonant pitches of their own, they work together with the air moving in and out of the soundhole. The result is that the frequencies all get shifted, or, more correctly, you end up with a new family of resonances that are related to the original ones, but in a more complicated way. The 'main air' resonance on a guitar is one of those shifted ones: it acts like the Helmholtz resonance, in the sense that there's air moving in and out of the hole, and the pressure changes inside the box the way you'd expect, but it's not anywhere near the 'right' frequency, owing to the motion of the walls. That's one reason I usually put 'Helmholtz' in apostophes in these discussions: to indicate that it's not really that, but close.

What you hear when you sing into the soundhole is the 'main air' resonant pitch, so long as you keep your face at least one soundhole diameter away from the hole. you can also change the pitch by putting your hand on the top or the back.

"Second, can you explain the reason why the pitch goes up when the hole gets bigger. That seems counter-intuitive to me. I think I understand why a stiffer soundboard increases in pitch and also why a lighter soundboard increases, but not this."

A small soundhole means a small 'piston' that doesn't have much area. That small area doesn't compress the air in the box much for a given amount of movement, so, in effect, you've got a soft spring. For a big soundhole, you've got the opposite condition: the air in the box gets compressed a lot for a small movement of the 'piston', so it acts like a hard spring. It turns out that the mass of the 'piston' doesn't go up as fast as the stiffness of the 'spring' when you make the hole larger, so the pitch rises. I'll note that that spring and piston stuff is Helmholtz' own approximation, and it's just that; approximate.

Another way to think of it, that William Allen brought up, is to think of something like a can with no top. If you sing into the open end there will be a resonant pitch that depends on the length of the can: it's a quarter wave length long. If you plotted the sound pressure level in the can as you did it, you'd find that it was high at the bottom of the can, and drops off as you approach the open and, falling to zero at the mounth. After all, at the open end, there's nothing to confine the air, so it can't build up any pressure, and sound is a periodic change in pressure. If you put a top on the can, with a hole in it, you can build up pressure just inside the opening; the pressure wave doesn't drop to zero in the length of the can any more, and it acts like it's longer. Longer wave = lower pitch. The smaller the hole, the more pressure build up, and the longer the effective wave.

This actually relates to the whole subject of ports, too, in a way. If you punch another hole in the can in effect you make it even shorter. The greater the sound pressure level at the place where you make this port, the bigger the change in the effective length. Of course, you get more pressure change, a higher sound level, inside the box as you get farther from the end and closer to the bottom. It's the same with making a port in a guitar; the further from the main soundhole it is, the greater the pitch change for a given size hole.

"If you started with a pinprick and ending with a really big hole, would there be a sweet spot that was detectable? As you mentioned could one keep making the hole bigger until passing the sweet spot,then bind the edge of the soundhole to get it back? "

If you start with a tiny hole, like the 'sound hole' on a sitar top (about 1/16" diameter), it's pretty plain that the pitch would be really low, and anyway you couldn't squeeze much air in and out. The losses due to drag would be terrific, like trying to breath through a little straw. As the hole gets bigger, the pitch gets up to something you can actually hear, and it becomes possible to move some usable quantity of air through the hole. The loss may still be high, though, and the resonant peak will be low and spread out over a farly broad frequency band. The bigger the hole, the higher the pitch, the higher the spectral peak, and the more sharply defined it is. In terms of enigineering, it just gets better and better.

The problem is that we're looking for music here, not efficiency. At some point that air resonant peak gets so sharp and tall that the note at that pitch is overwhelmingly strong, while the ones next to it are weak. It has no sustain on that note, but just 'thumps'. As usual, while we'd like to make these things as efficient as we can, there is a balance that has to be struck. One way to find the balance is to go past it, and then back up, as you say.

BTW, that person's definition of 'resonance' was 'the pitch at which the capacitative and inductive reactances cancel, leaving only resistive impedance'. Yeah, right....

Alan Carruth / Luthier

The can thing is great, like the theory of relativity, it begins to dawn on me. But Mario and Alan.... neither of you adressed my question about a "compressed" tone and whether a larger soundhole or a port might might help that. Or if you did, I am obtuse and missed it. (My closest friends often point out my obtuseness to their great delight)

Thanks for the clarification. I'm not going to say I have a 'feel' for the concept yet, but I'll keep thinking about it. Another site was saying you could actually measure the height of the 'piston' by feeling the air pump out of the soundhole when a string is plucked, keep lifting your hand until you feel the air drop off to find the height.

Larry:

One of my students built a reso with a very involved inlay on the top, and the soundhole in the upper bout was part of the design. I thought it looked a little small when he proposed it, but he went ahead. Sure enough, when it was done, it sounded 'choked'. He ended up using a handshaft to cut a small hole in the side, and enlarge it until the sound stopped getting better. Then he cleaned it up and made a surround that finished off the hole a little smaller, and it turned out great. It even looked as though he meant it!

Alan Carruth / Luthier

Thanks for getting back to me on that.

Compressed, choked, squished... yes this is what I mean.. On a single cone spider bridge dobro, it's got that tambourine looking soundwell. I've been told that soundwell acts as a compressor, and accounts for some of the dobro's distinctive tone. It sounds metallic because of the aluminum cone, but kind of choked because of the soundwell. So yes the guitar in question sounds a bit like that choked or squished dobro tone. I love that tone on a Dobro but not from a jumbo acoustic.

I'll try the sound port thing and see what happens. BTW did you notice that Joe was asking about decreasing the size of his soundhole?