Page 1 of 1

Soundhole (f-hole) size

Posted: Sat Mar 16, 2013 11:18 am
by Dennis Duross
Bear with me here. These questions are not all strictly speaking related to each other. And I know this type of question comes up periodically...

What would be the effect of sizing the combined area of two f-holes with that of a (more or less) standard 4" round soundhole (12.5 sq in of opening)?

Take as an example the one suggested in the Benedetto book---the combined area of opening of f-holes that size and shape leave you with a soundhole opening that is 20% smaller than that of a 4" round soundhole.

And I guess there's a reason that this is sort of a standard, but I'm not sure why that'd be.

Do smaller openings mean that the sound of the guitar is more dominantly that of the soundboard? And that as you open up the box more (with larger or more holes) you hear more of... what? If anything?

Re: Soundhole (f-hole) size

Posted: Sat Mar 16, 2013 1:58 pm
by Beate Ritzert
To (rather approximate) first order such a change would rise the helmholtz frequency of the body. More here:

Which does not necessarily mean it will sound "lighter" or "thinner": short time ago i needed to raise the helmholz frequency af a bass cabinet from 40 Hz to 52 Hz in order to make its sound "bassier".

Furthermore, my thinline archtop sounds a lot fuller if the pickup holes are left open.

Basicly to first order the area of a soundhole is connected to the volume of the body. But reality is far more complex than that.

Re: Soundhole (f-hole) size

Posted: Sun Mar 17, 2013 1:42 am
by Michael Lewis
Beate is correct. Especially the part that "reality is far more complex". There is a simple concept that is 'generally' truish, that smaller f holes tend to bring more bass presence and make the sound more directional. This holds fairly true for mandolins, and so far it seems to also apply to guitars. A big body (air column) and small sound holes can make tremendous amount of sound for comping with a band. The original D'Angelico New Yorkers are a great example of this. Benedetto's sound holes always looked too large in my estimation, but it really depends on what sort of sound you are trying to acheive and how all the parts work together, including the musician. The instrument structure is capable of a certain range of dynamics, and the musician also has a range of possible dynamics, and hopefully both ranges overlap considerably.

There is no "best" archtop guitar because there are so many possible playing styles and varieties of music as well as situations for playing the guitar. There are a lot of variables to juggle in order to arrive at your preferred result. When making the guitar 'steer' each variable as best you can toward the outcome you are wanting. In the end you might be surprised how well it all turns out, or maybe you find you have taken it too far in one direction or another. Then you will have a better idea of what to do the next time you make a guitar.

Re: Soundhole (f-hole) size

Posted: Mon Mar 18, 2013 4:26 pm
by Alan Carruth
I remember reading that a narrow slot has about the same effect on the air resonance as a round hole about 1/3 the diameter of the length of the slot. Long narrow holes might very well have delusions of grandure. F-hole arch tops generally have 'main air' resonances that are much higher pitched than guitars with a single round hole, either arch top or flat top, and it's definitely part of the characteristic sound.

F-holes also 'hear' some resonances in the box that a round hole in the usual place does not, and vice versa.

Yes, it's complicated.

Re: Soundhole (f-hole) size

Posted: Mon Mar 18, 2013 5:37 pm
by Dennis Duross
What got me wondering was---I covered over completely my f-holes with low-tack painter's tape. Just to see.

It seemed that this had the effect of thinning out the sound, almost as if what I was hearing was more the sound of the strings.

Not sure if this had the effect of damping, or trapping, or cancelling any sound contribution made by the excited air inside the box, or if in addition to the sound of the strings what I was hearing was the sound of the soundboard.

Obviously, when I removed the tape, the sound opened up, deepened, got louder, etc, just as you'd expect. Like when someone is playing an instrument in a closed room, and then you open the door.

So I started wondering if this is the effect that opening up the box has on the sound.

Then I began wondering how the sound hole openings compared (area-wise) to a standard round hole. And as my particular size and shape of f-holes are roughly 25% smaller in area (can be determined with Illustrator and Photoshop) it seemed like I might have some room to tinker.

Or not.

But that's where the initial question above came from.

I taped off my f-holes.

Re: Soundhole (f-hole) size

Posted: Tue Mar 19, 2013 1:58 pm
by Alan Carruth
In the low frequency region, up to, say, the open G string or a little higher in pitch, the guitar acts like a 'bass reflex' speaker cabinet. The top (at least) and the air inside conspire to produce two major peaks in the output. Usually we call the lower pitched one the 'main air' resonance, and the higher pitched one is the 'main top' resonance. These are misnomers; both the top and the soundholes put out sound at both pitches, but at the lower frequwency most of the energy is in air motion, while at the higher pitch the top is producing most of the sound.

You can think of the 'main air' resonance as being sort of like the familiar 'Helmholtz' resonance that you hear when you blow across the mouth of a wine bottle. Indeed, if you encase the guitar box in sand with just the holes open, you can get a nice Helmholtz resonance going, but it will be a bit higher in pitch than the 'main air' on the same guitar because of the interaction with the top. Similarly, the top by itself can move like a speaker diaphragm, but without the air in the box working with it, it would be at a bit lower pitch. On the assembled box the two resonances working together tend to push each other apart a bit.

When you cover the soundholes you're cutting out the Helmholtz resonance, which can't then interact with the top in the usual way. The result is that the 'main top' resonance moves down a bit in pitch, but not nearly enough to fill in the sound that's no longer coming out of the soundholes. On archtops the 'main air' resonance can be as high as C on the A string, although they are normally lower. Flat tops usually run between G and A, and Classical guitars can be as low as F or F#. This low frequency 'formant' peak is a big part of the characteristic timbre of the guitar. The missing low frequency peak means that you've lost a lot of the fundamentals of the lower notes, which is why they might not sound as 'full'.

On the plus side, it's usually the 'main air' resonance that feeds back the most strongly, so blocking the holes can cut down on that problem.

Re: Soundhole (f-hole) size

Posted: Tue Mar 19, 2013 2:59 pm
by Beate Ritzert
Now take an archtop with pickups mounted into the bridge. Remove the Pickups and play it with the additional holes. The Helmholtz resonance should go up - not too much: the area of an F hole roughly corresponds to the area of two holes for PAF like PUs, and the Helmholtz frequency rises with the square root of the area of the openings.

I tried this in my thinline archtop guitar and observed a fuller and louder sound - despite the increased Helmholz frequency. And that surprises me.

BTW: shouldn't it be possible to tune the resonance of a guitar down by increasing the length of the cavity (i.e. adding "walls" to the cornes of a soundhole)?

Re: Soundhole (f-hole) size

Posted: Tue Mar 19, 2013 7:41 pm
by Dennis Duross
ok I'm going to go ahead and admit this right up front---I understand very little of what either of you just said.

Don't get me wrong---I appreciate your thoughtful and generous answers.

I just don't understand them.

But I'll keep trying.

How are you measuring resonances and frequencies and such?

Maybe just point me at a resource?

Re: Soundhole (f-hole) size

Posted: Wed Mar 20, 2013 1:41 pm
by Dave Stewart
Dennis, if you download "Audacity" (or "Visual Analyzer" as the guys on ANZLF seem to prefer) you can record "taps" and then "analyze spectrum" which plots the peaks of resonant frequencies. Roughly speaking, you can then tap (for main air), plug the soundholes and damp the back (for main top) or damp the top & tap the back (for main back). (This is probably a gross over simplification and I welcome if Alan jumps in to correct me). I also like to use a tone generator to audibly confirm tap pitches.
One of the wrinkles with f-holes is their inefficient shape, hydraulically. The most efficient "duct" shape is a circle. The edge effects of alternate shapes makes them less efficient. You can calculate the "hydraulic diameter" of a non-circular shape as (4 X actual area) /perimeter (So, an f-hole with an area of only 5 sq in. but a perimeter of 20" has an hydraulic diameter of a 1" round hole........ most are better than this, hence the 1/3 that Alan mentions)
I do agree that it's more complicated than this (for example soundhole position can have a huge effect), but I often play with hydraulic diameter to get an idea of the effective area (ie the area of the hydraulic circle) of various soundhole shapes.
BTW Beate, I've sometimes thought of flaring the inside carving right around the f-hole so the edges stayed at 1/4" for that reason ....never tried it though.

Re: Soundhole (f-hole) size

Posted: Wed Mar 20, 2013 2:27 pm
by Alan Carruth
Some years back there was an article in the Journal of the Catgut Acoustical Society on altering the air modes of the viola, where the author tried extending the walls of the f-holes to drop the 'main air' pitch. He used stiff, thin, clear plastic pieces taped on, and reported that it worked well. He had a couple of other tricks that helped make small violas sound like big ones.

Anything I've tried that lowered the 'Helmholtz' type resonance also reduced the output and increased the losses. Generally the things you do to make the 'Helmholtz' pitch go up reduce losses and make it louder.

It's at least plausible that changing the length of the F-hole will alter it's 'upper cutoff frequency' . Like a small speaker, a small sound hole is inefficient at low frequencies, and the efficiency rises as the frequency goes up. At some point, as hole/speaker size approaches the wave length of the sound, it becomes essentially 100% efficient. This holds for round holes, but I'm not versed enough on the theory to say how it relates to long narrow ones: it's possible that the length of the hole establishes the cutoff frequency. In that case a long, narrow hole could have a noticeably different timbre from a short wide one with the same area. The relationship is probably not simple: it never is!

I do know that putting something like a 'rose' in a soundhole can add enough drag to reduce the height of the usual 'main air' peak to almost nothing, without altering the pitch much. Those parchment roses have about as much open area as the round hole they inhabit, due to the their 'inverted wedding cake' structure, but have lots of edge. This is, of course, the inverse of the usual archtop, but it's nice to have a notion of what the other end of the line looks like.

Re: Soundhole (f-hole) size

Posted: Fri Mar 29, 2013 9:58 am
by Dave Stewart
...and then I sometimes see examples like Jim Triggs beautiful D'Angelico 18" acoustic, and think how much more work I have to do to understand f-hole sizing. Comments?? ... post311892

Re: Soundhole (f-hole) size

Posted: Fri Mar 29, 2013 1:58 pm
by Alan Carruth
Well, if I'm right about the equivalent circle thing, then length trumps width for the most part. A long narrow F-hole might well have a larger effective area than a short wide one, and a long wide one might not be all that different from one the same length that's narrower. One of these days I need to do the experiment...

Re: Soundhole (f-hole) size

Posted: Fri Mar 29, 2013 5:48 pm
by Jason Rodgers
So when John Monteleone puts one or two huge oval soundports in the side of his archtops, big enough to fit a flat hand inside, in addition to various-shaped soundboard holes, what exactly is he going after? I mean, his "Quatroport" feature probably opens up the equivalent of about 1/4 the area of the soundboard! No losses, super-loud, but what happens to the tone?

Re: Soundhole (f-hole) size

Posted: Fri Mar 29, 2013 7:20 pm
by G.S. Monroe
One of the issues that I keep getting asked about is feedback/distortion from monitor speakers. I've been experimenting with alternatives to the traditional sound holes in the soundboard. It seems to me though that f-holes produce less feedback than circles even though they are not as efficient acoustically. As more and more acoustic guitar players are "plugging in", the feedback issues will probably become more of a concern than raw frequency response. Currently I'm testing a guitar with the sound hole in the back, and a double bottom like a mountain dulcimer to vent the sound around the guitar edges.

Re: Soundhole (f-hole) size

Posted: Fri Mar 29, 2013 9:23 pm
by Mark Swanson
Sorry to be skeptical here, but I am of the opinion that as long as an instrument is "acoustic" that you will have feedback issues. The same exact things that make a good acoustic instrument what it is are the same exact things that make feedback happen. I don't think it matters not one bit how or where you put the soundholes.

Re: Soundhole (f-hole) size

Posted: Sat Mar 30, 2013 11:15 am
by Dennis Duross
I did a rough measurement of the New Yorker soundhole in the link that Dave included in his post above, and given the formula he gave for finding the hydraulic diameter of a complex shape, each the New Yorker f-hole would ROUGHLY have the hydraulic diameter of 1.55".

Re: Soundhole (f-hole) size

Posted: Sat Mar 30, 2013 7:52 pm
by Alan Carruth
G.S.Monroe wrote:
"It seems to me though that f-holes produce less feedback than circles even though they are not as efficient acoustically."

The lower efficiency is the reason they produce less feedback. It's the drag of all that 'edge' in relation to the soundhole area that's making them inefficient, and that makes the 'main air' resonant peak lower and broader than it would be with an equivalent round hole.

Mark Swanson wrote:
"Sorry to be skeptical here, but I am of the opinion that as long as an instrument is "acoustic" that you will have feedback issues."

Sure, but its a matter of degree. As long as the room can hear the guitar, the guitar can hear the room. When the sound level in the room gets high enough to drive a resonance of the guitar strongly enough to replace whatever loss there is in the system, you get feedback. It's a matter of gain vs loss. The higher drag of F holes adds to the loss, and requires more gain to instigate feedback at the main air frequency, which is one of the worst offenders. 'Feedback busters' in round soundholes work by adding drag.

Dennis Duross wrote:
"...each the New Yorker f-hole would ROUGHLY have the hydraulic diameter of 1.55"."

That's a much smaller combined area for two F-s than the hole of almost any flat top acoustic. Very interesting. That, and the heavier and stiffer top of your average archtop, goes a long way to explaining why they have more feedback resistance than flat tops.

Re: Soundhole (f-hole) size

Posted: Sat Mar 30, 2013 9:08 pm
by Dennis Duross
I just rechecked my figures and I was wrong.

There appear to be barely 3 sq inches of opening per New Yorker f-hole, and roughly 14.5" of perimeter.

So the hydraulic diameter PER f-hole would be (2.82 * 4)/14.5 = 0.78".

The 1.55"" I stated before was the COMBINED value.

How likely is THAT to be true? Very? Not very? Beats me. But that's what I got.

Bear in mind that my measurements are based on a tracing I did of the picture provided on that web site.

Re: Soundhole (f-hole) size

Posted: Sat Mar 30, 2013 9:38 pm
by Dave Stewart
That's better. I got an actual area/hole of 2.57" & perimeter of 14.5". Clearly they are very restrictive (fluid wise), can see it, which is what grabbed me initially & prompted me to post the link. (The other thing that makes them look tiny is the 18" body.) BUT they are about 6.35" long as near as I can scale, which is a length that seems to appear frequently and has me increasingly thinking it may have a greater effect than equivalent area, as Alan speculated.