function of F-holes in top vibration

[Brian Evans]

I've mentioned ad nauseum that I am designing my classical archtop build. I am planning a fan braced top, and no F-holes, rather a hole (or series of holes) in the upper bouts. Reminiscent of the Gibson Army-Navy guitar. I have always felt that the F-holes, along with the recurve, significantly define the vibration of the top plate of an archtop. I feel that the section of the top bordered by the F-holes is freed and able to vibrate openly, while the recurve frees (or doesn't free) the rest of the top. I wondered what other people thought about this idea of no F-holes, and their effect on the top. I don't think the shape of the F-hole matters much, I think that the length and area of the F-hole does matter, but I don't know why.

Brian

[Barry Daniels]

The f-holes definitely loosen up the most important part of the top and have a big effect. Sound holes also allow the body to resonate more than a closed box.

If you don't mind me asking, where are you coming up with your ideas of how a guitar works?

[Brian Evans]

Mostly reading here, Carruth, Benedetto, a few other luthier websites, a number of articles published and on the web, and there is a great guy on the Mandolin Cafe web site who is a professor of physics who has done and published a lot of stuff, but I forget his name. I take it all in, and try to make sense of it. My chief professional skill, in 40 years of data communications engineering and sales, was taking in a lot of technical and operational expertise and trying to make sense of it. I usually win in the end, but guitars are hard...

Brian

[Barry Daniels]

Yep, they are very complicated. Sounds like you are using good resources and your engineering knowledge should be very capable of processing all the info. The only thing you have to watch out for is conflicts between the various sources. Many people use a simplified model of how a guitar works and like most things, the devil is in the details.

[Brian Evans]

My experience so far, in many many fields of endeavor, is that if you think deeply on a subject and take in many expert points of view, you can come up with a path forward. For example, I was very impressed with the idea presented here recently of using a tone-right generator to vibrate the guitar and to feel for the expression of vibration across the top. I thought that had real possibilities. My current thinking is about how to engage as much of what I think of as the "big round circle" that is the lower bout of the instrument, centered on the bridge, which is where I think the tone is developed. I think the upper bout is a throw-away in most guitars, top-wise, and it only contributes volume to the interior of the instrument, but with appropriate fan bracing I think I can engage the entire middle section of the upper bout, right up to the neck block. I think a lot about how to engage that part of the top in tone production, but it seems you do have to cut a big hole in some part of it and throw it away. One of the key things I think about, and would value your opinion on, is the function of the braces (and I only build archtops, so keep that in mind) - are they for strength, or are they more primarily transmitting the vibration? In data communications and microwave, you transmit the "vibration" - the signal - down a signal path that is a tuned co-axial cable or a wave guide. How much does that transfer to how sound vibrations get transferred in a tuned, braced guitar top? In a flat top guitar it seems to me the back (and the entire upper bout of the top) is braced in such a way as to take it out of the sound creation and make it pretty dead. In an archtop, the back is tuned as a partner to the top and contributes mightily to the sound. Is that true, is it another area for thought? I don't have a clue.

Barry. I appreciate your feedback, btw.

Brian

[Barry Daniels]

One of the key things I think about, and would value your opinion on, is the function of the braces are they for strength, or are they more primarily transmitting the vibration?

Yes! In other words... braces do both. How could they not?

But one thing to keep in mind. The majority of archtops only have only two braces as opposed to something like ten on a normal flattop (steel string and classical). So if you put a bunch of fan braces on an archtop you need to be careful about over stiffness. Also fitting fan braces to an archtop is going to be a real challenge. Personally, I would not do fan braces on an archtop. The nylon strings place a lower load on the top so even an X-brace or two parallel braces may be too much. I would probably rely on the arched top to carry the load.

The design you are proposing combines several novel ideas which will likely lead to unexpected consequences. It is highly recommended to only make one change to a standard design so that you will know what is responsible for the tonal change observed. I am not saying don't build it, but I am saying you should be prepared for it to turn out to not produce the sound you are looking for and you may not be able to identify what was the real cause. Sorry to be such a wet blanket.

[Randolph Rhett]

I wrote a long reply to your post, and realized you probably wouldn't have the patience or interest in such a rambling discourse. So here is the short version:

I don't think anyone can give you a definitive answer. I humbly encourage you to take a step into the unknown, build you design even if you don't know how all the variables play together, and tell us what worked and what didn't.

The long reply:

Trying to predict the tone of a guitar before it's built is a little like trying to predict the weather. Yes, scientist at NOAA use highly technical models and massive computational resources to make their predictions. The world relies on their very good predictions. There are also Polynesian mariners who through years of close observation and experience do as well in their predictions. Neither come even close to 100% accuracy.

I've noticed over dozens and dozens of student luthiers that some approach guitar building as if they could just get the "right" formula from NOAA they will be able to predict the weather with 100% accuracy. Others seem to look for that one pearl of wisdom from a wizened mariner will allow them always safe passage. The truth is that no one can tell what the guitar is going to sound like before you build it. Often not even a hint to what it is going to sound like, especially if you are departing from established norms. People like Benedetto have built 800 of near identical guitars, and no two Benedetto's sound exactly alike.

I love experimenting with guitar design. I doubt I will ever stop. However, I recently built 3 guitars. As best as I can tell, all identical. I don't even use wood for the soundboard, so I really do get closer to identical than most. One is brilliant (I'm keeping it!), one is definitely good and going to a buyer, the last is meh...don't know what to do with it. Since mine are also nylon string arch tops, I'll throw in a picture for good measure. As you can see, I don't just copy Benedetto. I am certainly not one to tell you to just copy a master.

However, building a unique guitar is by definition a lonely endeavor. No one can really tell you how it might turn out. There is a ton of pseudo science and pseudo wisdom in guitar building, but even if you could find the few real science/wisdom pearls out there, they probably won't apply to your design. Guitars, like the weather, are mind bogglingly complex systems that are near impossible to predict accurately. Move to another planet, and all bets are off. You can copy an existing design or bust out your own. You just have to build and be prepared to accept that the next guitar may not be all that great. As I mentioned in an early post to you, I tossed out a half dozen bodies before I got one that worked as a guitar (thank God I use bolt-on necks! ). That is not bragging, I REALLY didn't want to have to trash them. But, despite all my research I was unable to predict how it would all work together until I actually built it.

I don't want to stifle your quest for knowledge, and this is certainly not meant as criticism. In fact, I mean it as a fellow... whatever you call this affliction! I can't play a tune on the guitar if I don't understand it. I can't just copy the fingerings or play what my ear hears. I have to KNOW WHY. Why did they put that chord fragment there? Why those notes? Is that a substitution? What kind? Did he modulate? Where did he get those fingerings? Just three notes on the guitar and I won't be able to get past them until I understand why!

Many of my musician friends look at me like I'm crazy. "I don't know, that's just what he did. Isn't it cool?" Charlie Parker was a brilliant musician who fully understood the music theory at the highest levels. Still, he is quoted as admitting that he played what his ear told him to play and figured out the why after. I can't tell you how much I wish I was more like that!

So, as a fellow obsessive luthier I would suggest build your concept. If it doesn't work, ask here for help troubleshooting. Build the next. Closer? Good. Refine a little more on the next. Wash, rinse, repeat.

I'm hoping after you make a few you will share what worked and I can steal your ideas wholesale!

[Eric Knapp]

Great thread, folks. I wish I could try all the guitars you all make. I'm at the beginning of this "affliction", as Randolph calls it, and I'm already hooked. Wanting to play and hear a guitar I've made is a strong desire. I know my first one will probably be terrible as I am doing this on my own with no direct help. I'm hoping it will spur me on to the next one and not discourage.

-Eric

[Brian Evans]

I had another thought on this whole f-hole thing. Makers of flat-top guitars are moving towards laminated sides - one reason for doing so to create a very stiff side that "robs" the least energy from the top, as I understand the explanation. My "lets think of sound/vibration waves as electrical" hat says that maybe the stiff sides are creating a reflection of the energy and returning it back into the top. Having half of the circumference of the lower bout top flopping in the breeze would not do that. The more I think about this, the more I think that maybe having no f-holes interrupting the top and paying a lot of attention to the recurve could be kind of interesting.

[Barry Daniels]

One eye-opening thing I learned from Trevor Gore's excellent book is that the soundhole allows the back to work with the top, which is also a big part of Benedetto's methods. This is not a subtle effect. If you plug the soundhole of a flattop, the back (if it is an active element) completely drops out of the mix. This not only shows up in the computer analysis of the recorded frequencies, but you can also hear it.

Not all guitars are built with an active back, but I think it adds a lot of dimension to the sound.

[Brian Evans]

I am seriously thinking of building this particular guitar with a removable back, so that I can adjust bracing and change the sound hole after I am playing it. Maybe a whole bunch of tiny screws around where the binding will eventually be. I plan a sound hole in the bass-side upper bout, close to the player, with an area typical of classical guitar sound holes.

[Alan Carruth]

As has been said; it's pretty complicated. Everything affects everything else, and it's hard to predict the outcome in detail. I you confine yourself to the main effects it's possible to say some useful things, though.

Arched plates vs flat braced ones:
It always seems to me that what flat tops do with bracing archtops do with the arch, more or less. Generally, if you look at the way the 'free' plates vibrate, the bracers on the top of an arch top are pretty much there to make up for the stiffness you loose when you cut the holes. There's no bracing on the back in part because you don't cut holes in it. The arch top is a 'distributed element' system; the stiffness comes from the arch height, arch shape, and thickness distribution, and bracing, along with the wood properties. Ideally you get them all to work together to make a system that is, perhaps, less 'lumpy' than a flat top, so maybe it works better in some sense, particularly in the high frequency range. Or so I imagine.

Because the arch top doesn't need an upper transverse brace (usually) the 'main top' mode can actually extend up to the upper block, so you get, in theory, a larger area moving than on a flat top. That was part of the reasoning when I made my own arch top Classicals. Of course, since I used plate thickness rather than bracing to get the stiffness I ended up using a thicker top than you normally would on a Classical, which reduced the effect of the added area. A thin arched top with fan bracing might end up lighter with the requisite stiffness. I'm not sure if you could bend normal sized fans to fit the arch the way they usually do, but they could be heat bent. Twist could be a problem. I'd have them fan out from the upper block and run more or less full length.

The main effects of the sound holes have to do with air resonances. In particular, F-holes have a larger effective area than a normal round hole, so they tend to move the pitch of the 'main air' mode up. This, as far as I can tell, is the major difference in the timbre of F-hole and round hole guitars. Iirc, a slot is equivalent to a round hole about 1/3 the diameter of the length of the slot. The width doesn't seem to matter as much. F-holes, because of their placement, also 'hear' internal air modes that a normal round hole doesn't, and that changes the timbre as well.

A normal round/oval sound hole introduces some interesting effects due to it's placement. Basically, because it's centered above the waist, which is itself just a bit higher than the middle of the length of the box, and with the help of top motion, it can cause 'splits' in some of the air modes. With careful measurement you can see a second 'Helmholtz' type air resonance above the normal 'air' pitch, as well as a pair of 'A-1' 'lengthwise bathtub' sloshing modes instead of one. Sometimes the A-1 modes can couple strongly with the top 'long dipole' mode, giving rise to an extra output peak in the spectrum. All of this has been more or less incorporated in the normal guitar design through the usual cut and try methods, and seems to be part of what makes the guitar sound different from,say, an Irish bouzouki or lute tuned in the same range. Again, you don't see this with F-holes, and also tend to miss it on guitars, like the Dreadnought, that don't have a pronounced waist. I said it was complicated!

The archtops I've made with round holes, such as the arch Classicals, have had more of the usual 'guitar' sound. You do need some sort of thin cross grain patch above the hole. I used X bracing on mine, but if I did use A bracing I'd put a cross grain patch below the hole as well. This integrates nicely with the fans.

I found it hard to get an acceptable timbre in my arch Classics. It was a matter of choosing the top wood carefully, and using the right arch height and thickness, along with a round sound hole. The tone still strikes me as a bit 'lute-like, or maybe like a 'small' guitar. Perhaps if I do another I'll make the body 17" wide instead of 16" to get a 'bigger' sound. I think A bracing would have helped. Few experiments work out as well as you'd hoped, but I guess if you learn something from them they can be worth the effort.

[Peter Wilcox]

I am seriously thinking of building this particular guitar with a removable back, so that I can adjust bracing and change the sound hole after I am playing it. Maybe a whole bunch of tiny screws around where the binding will eventually be.

That's how I did a removable back for an experimental guitar. http://www.mimf.com/phpbb/viewtopic.php?f=56&t=3074

I never have got around to actually finishing the guitar (maybe this year), but the plan is for binding to cover the screw holes. I left a 1/8" lip on the back to avoid it splitting at the holes - the holes are not as far from the guitar edge as they look on the back view. I put in more screws after the pic was taken, and re-enforced the holes in the lining with CA glue.

[Patrick Hanna]

Many of their functions have already been mentioned. In addition to those functions, sound holes allow the instrument to breath. By this, I mean they allow air to enter and exit the sound box as the top and back flex and vibrate. Air moving in and out plays a role in coupling the top and back vibrations, as Barry has said. It effects resonances and other sound qualities, as Alan has said. I recommend the Benedetto DVD /video construction series to you if you have not already watched and listened. His method relies on a very active back and he demonstrates scraping the back recurve in a way you can actually hear as it opens up. He says the back is the diaphragm of the instrument. I don't believe his method would create instruments this responsive without a "breathing" sound box. You only get a theoretical understanding of this effect by reading. I read the book long before I watched the video. The video was a real "Ahaaaa!!!" moment for me because I could hear the instrument coming alive.

[Alan Carruth]

The top moves more than the back, and pumps more air, but if you can get the back to work with the top that's a real bonus. There's a lot out there on this in various references. The trick is to get everything 'tuned' just about right before you put the box together, and then do some fine tuning afterward.

[Beate Ritzert]

I am reading everything regarding archtops and their tuning because my plans to start with the bass are just delayed (until i can live in my new old house)... and of course the same questions come up.

I am seriously thinking of building this particular guitar with a removable back,... Maybe a whole bunch of tiny screws ....

A bunch of screws might add a significant mass to the guitar and change the vibrational pattern. (I tuned the sound of my skeleton violin which is now in repair by adding mass to the sides, and that was drastic...)

What about tiny gaps which might cause unwanted rattle? I experienced this when the back of my old acoustic violin came loose - for quite a long time, the gap was unvisible, but the rattle on some resonances was annoying. This might explain why i am so skeptical against screws to hold the back.

What do You as the experts think?

Would a violin like construction be an alternative? Or were screws better? Remove them and glue the back on after finding a "sufficiently good" tuning?

[Barry Daniels]

Jimmy D'Aquisto made an archtop guitar using violin-like designs. It was beautiful.

[Peter Wilcox]

Remove them and glue the back on after finding a "sufficiently good" tuning?

That was my understanding of his plan (and mine.)

[Beate Ritzert]

Again, how to deal with tiny gaps which i consider hard to avoid?

[Peter Wilcox]

Make the back and sides fit as smoothly as possible, but I suppose there will always be gaps of several thousandths. My back is sprung into a longitudinal convex shape, which keeps it stiff, so there is no buzzing, but there is a gap of 0.008" (I just measured it) for several inches on one side of the lower bout. I suppose I could put an extra screw in there, but I'm just assuming (possibly erroneously) the gap won't contribute significantly to it's sound. I am also assuming that removing the screws, gluing on the back (glue adds mass too), and adding binding will not significantly alter the sound.