MIMF

A guy on the F'Book claims that D'Angelico carved and braced his tops differently on the bass side vs the treble side so that different frequencies were emphasized. The idea being that even a non-cutaway archtop was "handed" and should not be made left handed if intended to be right handed. I think this is bunk, that he carved and braced his tops normally, and that this whole idea doesn't work anyway. But I have to ask - any ideas on this? Stiffer for trebles and floppy for bass, vice versa, or the whole top responds to all frequencies so do a good job everywhere?

As you say, the whole notion that there's a 'bass side' and 'treble side' to the top in terms of sound production is bogus. Basically, the whole top moves as a unit at lower frequencies, and it breaks up into smaller vibrating areas as you go up in pitch. Just exactly how that works depends a lot on on the wood of the top and on how it's carved and braced.

By comparison with flat tops, archtop guitars are very much 'treble balanced'. The carved arch adds a lot of stiffness moving the 'main top' resonant mode upward in pitch. The generally thicker top also tends to delay the breakup into smaller vibrating areas until a higher pitch is reached. Add to this the fact that the F-holes produce a higher 'main air' resonant frequency than is common on flat tops (often as much as a third higher) and the entire basis of the sound is shifted toward the treble end of things.

With all of that said, it's possible to affect the treble-to-bass balance a fair amount. The obvious place to start is by using smaller F-holes. From what I've seen 'parallel' bracing seems to produce a more 'bass balanced' timbre than X-bracing. It's likely that the top graduation scheme also can make a difference.

Back when I was taking voice lessons (from a fellow who also had a masters in jazz guitar from Berklee) he said that the way to improve your bass notes is to work on the trebles. Because of the way hearing works the energy of the high harmonics tends to be added to the fundamental of the note in your mind, so getting a clearer and stronger upper register can actually enhance the perception of bass tonality. This seems to me to depend more on how well the top is graduated, and how carefully the braces are balanced with the top, than the overall scheme of the top thickness and brace profile. Obviously, using a lower arch and thinner top graduations will probably help to emphasize the bass a bit, along with the afore mentioned smaller holes, and a deeper rib might help as well. Archtops seem to be even more a system than flat top guitars: you really can't change just one thing.

Tone happens

I think it's clear, as Alan said, that there's no such thing as a "bass side" and "treble side" to the top. However, the same is not necessarily true for the bridge. I haven't personally experimented with this, but I think you could carve a bridge to have less mass on the treble side, with a reasonable expectation that it would enhance treble response on that side. Also, the sound hole shape and placement makes a big difference. My guitars with a sort of pear-shaped sound hole in the upper bout have a much deeper bass resonance than the ones with slit-type sound holes in the lower bout. I can't prove this either, but I think leaving the top as flexible as possible also enhances bass response. I've shortened my X braces from what Benedetto recommends and tapered the vertical dimension considerably, so that they are pretty much providing structural support only under the central flatter area of the arch around the bridge, the idea being that the more curved peripheral areas are self supporting and should be as flexible as possible. Seems to work for me.

John, how interesting that your guitars with a upper bout sound hole have deep bass response, because I have the same experience. I also try to minimize the X braces on that guitar with a taper. We share some basic ideas, I think.

It also seems to me that cutting big holes in the lower bout must interfere to some extent with the monopole and cross dipole modes of vibration of the top. I'm wondering if anyone has studied this? It's kind of hard to produce chladni patterns on an archtop.

Moving the sound hole into the upper corner lowers the pitch of the 'main air' resonance, all else equal. Look up Allens' article on air resonance in American Lutherie #1 or the first 'Big Red Book' for more information.

From what I've seen 'parallel' bracing seems to produce a more 'bass balanced' timbre than X-bracing.

Wow, really? That goes against what's generally accepted out there as a kind of dogma since the '30s when Gibson made of it a thing and D'Agelico and later D'Aquisto followed.

Please, Alan. Would you elaborate a bit on the topic?

Thanks.

Paco Jimenez asked:
"Would you elaborate a bit on the topic?"

Of air resonances, I'm assuming? Be careful what you ask for: I could go on about this stuff for a long time. Much of the low end timbre of the guitar comes from the fact that it acts as a 'bass reflex' enclosure in that range; roughly from the lowest notes up to about the open G string or a bit higher on arch tops. Most of the experiments have been done on flat tops, but the principles are the same.

A rigid guitar shaped box with a hole (which you can get by burying the guitar in sand leaving the hole open) will have a low pitched 'Helmholtz' air resonance, analogous to what you get when you blow across the mouth of a wine bottle. Air rushes in and out of the mouth, changing the pressure in the bottle.The classic model of this is to think of the air in the neck of the bottle as a rigid 'piston' of a certain mass, and the air in the body of the bottle as a spring with a certain stiffness. The air piston bounces on the air spring, and the relationship between them sets up the pitch of the resonance.

The wider the mouth of the bottle the larger the surface area of the piston. This means that the pressure in the bottle will change more for a given amount of motion of the piston, so the 'spring' is stronger. That raises the pitch of the 'Helmholtz' mode. A longer neck on the bottle makes the piston heavier, and drops the pitch.

In this model the walls of the bottle have to be 'rigid' in order for the model to be of any use in predicting the pitch of the air mode, so that's why you have to bury the thing in sand to find the 'real' Helmholtz mode. Helmhotz' original resonators were spherical, too, but guitar bodies are not. Allen showed in his article that the location of the hole on a box that was roughly guitar sized made a difference in the pitch. With the hole in the middle of the top the pitch was higher, and it dropped as you moved the hole toward the end of the box, and then even lower if you made the hole in the side rather than the top. The total drop in frequency from doing this is about 30%; say from about A=220 to C~131. Moving the sound hole up into the corner of the upper bout is not that much, but makes a big difference.

Since the walls of the guitar, and particularly the top, are not rigid motion of the top can affect the air flow, and the air can influence the top too. Air pressure changes in the box push on the top, and top motion takes some energy out of the air, slowing the flow and lowering the the pitch of the Helmholtz-type 'main air' mode. At the same time the loudspeaker like 'main top' resonance will be raised in pitch by the added force on the top of pressure changes in the box. These two modes, which would start out about a fifth apart in pitch if they're isolated from each other, end up about an octave apart on most guitars, with the 'air' mode dropping from it's 'real' Helmholtz pitch, while the top mode rises in frequency.

These two resonances tend to enhance the response of notes close to them in pitch. Often, on flat top guitars, they're at around G on the low E string for the 'air' resonance an the open G string pitch for the 'top'. On archtops they tend to be higher in pitch, an that's one of the major differences in the timbre. With the hole up in the corner the 'real' Helmholtz mode starts out lower in pitch than it does with the hole in the 'normal' place, and then gets dropped some more by the couple with the top. It can end up all the way down at low E or even lower. That can boost the output for that note, but leaves a 'hole' in the response higher up; there's no free lunch.

Allen also talked in his article about 'higher order' air resonances in the box. Mostly these are in the nature of 'sloshing' modes, where the air moves back and forth between more or less parallel walls of the box. The lowest pitched one is the 'A-1' air resonance, sometimes called the 'lengthwise bathtub mode', as it works like sloshing the water along the length of the tub. The pressure changes a lot at the neck and tail blocks, and pretty much not at all in a line across the middle of the guitar. If you think about it there are several other similar ways the air can move, such as sloshing across the lower bout, and so on.

On a normal round hole guitar the hole doesn't 'hear' these modes very much, since it's not in a place where the pressure changes much, and sound is a pressure change. F-holes should pick up more of the A-1 mode, and even more of the crosswise 'A-2' in the lower bout, ans so on. A hole in the upper corner can pick up the A-1 very strongly (up around 350 Hz, close to the open high E), so that make a difference in the sound.

As you can see, this stuff gets involved as you get into it.

Be careful what you ask for: I could go on about this stuff for a long time.

Well, I don't see the problem. Always ready for a good reading LOL.

But you are right, I should have been more precise. I was trying to ask about bracing as per the original quote.

From what I've seen 'parallel' bracing seems to produce a more 'bass balanced' timbre than X-bracing.

Anyway, that's exactly the subject I was wondering about right before I read that particular sentence: Soundhole placement, and specially moving it away from the typical locations, even to the sides. So you nailed it. This explanation is the most enlightening about this topic I've ever found so far.

Thanks so much, Alan. Now, if you don't mind to give it another blow on the parallel vs X bracing bass balance effect...

Well, I should say that 'what I've seen' is not a very large sample, so I could just be talking through my hat.