What produces sound?

[LKN2MYIS]

This question was generated by another thread.

I'm a recovering guitarist and fledgling mandolinist. I know that from research and understanding of guitar construction that:

1) the top wood of the guitar is responsible for about 80% of the tone
2) back and sides contribute only about 20%

On other forums, I've seen the sound engineer tests, proofs, etc., and I'm fairly comfortable that this is how it works for guitars. I always got a big kick out of people spending a fortune on Brazilian rosewood back and sides and not paying any attention to the top wood.

My question is this: is it the same for the mandolin? Is the top what generates the majority of the tonal quality, or, because of bracing and size, is it different?

Not trying to start any 'wars', just real curious.

It seems that most of the Cafe members appear to be bluegrass players, and the discussion is always on volume first, which I understand for that type of music. I'm more interested in tonal qualities and what PRIMARILY affects them. (I know there are many factors, but my interest here is in the wooden components).

Any thoughts and ideas on this are welcome - you luthiers know your craft.

Thanks -
John

[danb]

I think the back of a mandolin is absolutely critical to the bass response. If you lean the back against your belly when you play, the bass response is almost totally killed.

It might not account for 50% of the volume, but it's a critical part of the tonal response

[El Greco]

From what I've read so far and from logical deduction ( makes me say things) tuning the tone bars and taping the top will yield the sound you desire. Also top soundboards usually feature a more expensive wood (vs. dense maple backs), which makes me think that the top is the more important component in producing sound. But I wouldn't talk percentages...

[LKN2MYIS]

Meaning that the top is indeed the primary conveyor of tone?

[El Greco]

Meaning that the top is indeed the primary conveyor of tone?

---
That is what I've been inclined to think, based again on what mando construction/repair info I'm reading here and there.

[Celtic-Grass]

makes sense... research shows with violins, the vibrations of the string are transfered to the bridge which then "rocks" side to side making the top vibrate. because of the graduation (thinner wood) around the perimeter of the top (just like on a carved mandolin) the top flexes and moves, thus moving the the air inside the instrument. the difference with violins is, of course the sound post which additionally transfers the vibration of the top to the back as well. so the whole box moves the air out the f holes. almost all of the sound emanates from the top and back... the f holes more or less just allow the air to move in and out and equalize the pressures. allowing for more flex, more flex more sound.
Mandolins, i presume work much the same way. That's what Gibson was trying to archive when he started making carved top mandolins, anyway.
I"ve often wondered tho, why do arch top mandolins not have a sound post?

[first string]

I"ve often wondered tho, why do arch top mandolins not have a sound post?

This has been discussed a lot actually. I'm not an expert, so I may butcher this explanation, but the gist of what I have read is that in a violin the bow is constantly adding energy which can travel along the sound post and activate the back plates. Where as in a mandolin you only have one brief instant of excitation (when you pluck the string) so all a sound post would do is muffle the vibrations in the top. Hope that helps. I'm sure someone will chime in with a more scientific description.

[sunburst]

I'm going to wade right in here, though I might live to regret it, and probably quickly get in over my head, but here goes...whew...

"the top wood of the guitar is responsible for about 80% of the tone"

We can't know that. As far as I know, nobody has quantified the amount of "tone" that comes from different parts of a guitar. I suspect that that is speculation on someone's part.

Tapping tone bars and tops does nothing for the sound. It is a way that some people use to adjust their stiffness, but it isn't necessary to build a good mandolin. There are plenty of other ways to gauge and adjust the stiffness of the tone bar/top plate system.

Violin bridges rock because of the bow, the lightness and height of the bridge, the sound post, etc., that's true, but violin tops are not graduated particularly thin at the edges, and that is not a significant contributor to the transfer of energy from the bridge to the top. The sound post transfers energy from the top to the back and makes the violin body behave as an "air pump". Air moves in and out of the F-holes. The box is a modified Helmholtz resonator (there's plenty of information on that for the diligent seeker).
Mandolins work much more similarly to guitars than violins.

So how do mandolins make sound?
The strings drive the bridge, the bridge drives the top, the top drives the air in the box (as well as the air around it), the air drives the back plate, and the whole thing becomes a modified Helmholtz resonator with many complex interactions between top and back plate modes and air modes.

This question is similar, in many ways, to the thread running now on loudness. The workings of stringed instruments are pretty complex and not fully understood. It is much easier for someone to proclaim that "the top is responsible for 80% of the sound" than it is to actually figure out what the top, back, sides, neck shaft, etc. actually are doing. It's easy to jump to conclusions from observations and ascribe importance (or not) to things that we just don't know about.
In short, there's no simple answer, but there are many papers about how violins work, hundreds of papers on how guitars work, and a handful of papers on how mandolins work. They can often be found on line or in the library, and if you are curious enough, and able to understand the language of scientific papers well enough, you can bring your own knowledge almost to the level of the experts themselves by reading what's been written.

[LKN2MYIS]

John -

That all makes sense.

My comment about the 80% comes from an acoustical engineering web site that had all of this broken down - schematics, test results, etc. Can't find the link (of course!) but when I run across it I will post it. Better minds than mine, and certainly not meant to come across as a proclamation of any sort. The testing, controls, and the rest made a lot of sense. I was comfortable with the validity of what I was researching. Of course, that doesn't mean it is 100% correct, but it was certainly not something to be dismissed as out of hand.

I believe what you are saying about how mandolins produce sound is very valid. In the mandolin world, has anyone ever done controlled testing to determine the contribution of the top vs. the back and sides for tone production? I am sure that the entire sound chamber contributes to the sound, but I am very curious as to the distribution properties of it.

A recent example from the guitar world: A close friend of mine (full-time professional, touring musician) has a guitar. He wanted to purchase a second guitar (the same if possible) as a back-up. In Mandolin Bros., we found the exact guitar, however the difference between his and this new one was that the new one had an Adirondack spruce top. All the other specs were the same, even down to the sunburst. The difference between the two was simply phenomenal, and the only difference was the Adi top.

I realize, of course, that they have not aged the same necessarily, and there are other factors, but it truly caught my attention. Night and day in tonal quality.

The 'loudness' thread you spoke of is what prompted my delving into this. I always hear bluegrass players talk about their instruments and the majority of times the word "loud" dominates the conversation. I'm more interested the the tonal aspects of the instrument. I would think (could be wrong) that 'loud' is easier to get than a certain tonal quality.

I came across this site:
http://www.phys.unsw.edu.au/~jw/guitar/intro_engl.html

Not the one I was referring to, but interesting as well.

[Patrick Killeen]

I"ve often wondered tho, why do arch top mandolins not have a sound post?

This has been discussed a lot actually. I'm not an expert, so I may butcher this explanation, but the gist of what I have read is that in a violin the bow is constantly adding energy which can travel along the sound post and activate the back plates. Where as in a mandolin you only have one brief instant of excitation (when you pluck the string) so all a sound post would do is muffle the vibrations in the top. Hope that helps. I'm sure someone will chime in with a more scientific description.

The biggest effect of the soundpost is to create a firm spot on which the bridge can pivot driving the bass bar up and down. This makes the bridge more efficient at transfering the energy of the side to side motion of a bowed string into the up and down motiong of a soundboard.

Here's a little experiment you can try with a violin bow and a plucked instrument with a sound hole under the strings. If you bow the instrument by going side to side, like a violin, then you will get a very thin sound. If you poke the bow into the sound hole and bow the instrument up and down you get a much richer sound (my bouzouki sounded a lot like a cello when I tried this).

Mandolin strings vibrate up and down anyway, so they don't need a soundpost to transfer their energy efficiently.

Patrick

[sunburst]

...we found the exact guitar, however the difference between his and this new one was that the new one had an Adirondack spruce top. All the other specs were the same, even down to the sunburst. The difference between the two was simply phenomenal, and the only difference was the Adi top.

As I'm sure I don't have to tell you, that is an anecdote, not a scientific experiment, and certainly far from a representative sample. I'd say any conclusion drawn from it is "jumped to".

Here, then is my own anecdote.

Quite a few years ago, after hearing it proclaimed many times as fact that 'you can't build two instruments the same even from the same wood', I decided to test the hypothesis and built two mandolins from the same wood. I cut the tops, backs, sides, fingerboards, head blocks, tail blocks, point blocks, linings, everything from the same wood as close as possible from the same stock. I carved the arches the same, graduated them the same, stained them the same, finished them the same, etc.. In short, built them as nearly identical as I could. When they were done, they sounded the same. I couldn't hear a difference when I played them, I couldn't hear a difference when someone else played them, and nobody that listened to them or played them while I still had them could reliably tell them apart by sound (it wasn't easy to tell them apart by sight!).

I learned that, in fact, you can build two instruments that sound the same.

A few years ago, I did the "experiment" again, but this time I did everything the same except for the top wood. I built one with sitka spruce and one with red spruce. The sitka came from some very old "aircraft spruce" that I bought from a friend many years ago, and had probably been cut for close to 50 years, and the red spruce came from a tree I cut in West Virginia in 1989.
When the mandolins were done, they sounded almost the same. There were only subtle differences in the tone and both were about the same in terms of loudness. I took them to the Galax fiddlers convention when they were newly completed and got lots of mandolin players to play them, without telling them that they were different at all, let alone what the difference was, and asked them which one they preferred. They were divided in which they preferred, but slightly more preferred the sitka topped one.
Anyway, when I hear the properties of one species or another of top wood touted as amazingly better than some other, I'm skeptical because that has not been my experience, and I tend to assume that it is a conclusion drawn from limited evidence (just like my own conclusions from my own limited "experiments").

[foldedpath]

1) the top wood of the guitar is responsible for about 80% of the tone

Which guitar? They're all different, so it's impossible to generalize to a specific number like that.

I have a Guild acoustic built in the mid 70's that I could probably drive my truck over, and it would still be in tune. I also have a Santa Cruz that's the most lightly built guitar I've ever owned. When the Santa Cruz is amplified and it starts to run away into body resonance feedback, I can kill the feedback by pressing on the soundboard, or I can also kill it just by pressing against the back of the guitar (not the best way to deal with feedback... just an observation about resonance). That's a guitar with a very "active" back. The Guild isn't. The arched (plywood?) back is more of a reflector than an active participant in the tone.

I suspect it's the same with mandolins. Some are going to be more "active" than others.

[LKN2MYIS]

Obviously the guitar experience I pointed out was indeed that, an isolated experience. (Huss & Dalton's, by the way).

I don't know if the type of guitar would influence it that much (speaking acousticaly). The point was that the top would was, as opposed to the back and sides, the part that produced the majority of tone. Could it be 65% in one make and not another? I would assume so. However, I think we're getting sidetracked a bit. I am sure that ALL of our experiences differ, and I certainly have no intent arguing any of what has been said. My guitar story cited earlier was indeed in isolated experience and not meant to be a bell-weather in any way, shape, or form.

What I am trying to find out is if anyone has run any tests or knows of any tests done that point to either the top being the primary tone producer on a mandolin, or the back and sides, and, if either, what the percentages would be?

[sunburst]

What I am trying to find out is if anyone has run any tests or knows of any tests done that point to either the top being the primary tone producer on a mandolin, or the back and sides, and, if either, what the percentages would be?

If you go to Dave Cohen's web site I think he has links to his papers. So far, he and Rossing have done the most research on mandolins and how they produce sound.

[LKN2MYIS]

Thanks - I'll GOOGLE him.

Appreciate the info. This stuff fascinates me.

[billhay4]

No need, it's here
Bill

[steve V. johnson]

There is a lot of good stuff in the 'loudness' thread that speaks to these questions, too.

As a practical matter, I accept the premise that the top of a guitar, the wood type, the grain, the carving/construction of it, the bracing, are a primary variable in the sound and performance of the guitar.

The top wood and the characteristics of the other woods used, are also important in the language and parlance that we use to describe and choose among mandolin-family instruments.

It's also true that these, and other commonly discussed aspects of guitar and mandolin construction are truly oversimplified, even gross measures of what really makes these wonderful instruments work.

Having said all that...

"Waitress! #I'd like a cedar over walnut, oval-hole two-point, please."



stv

[LKN2MYIS]

GREAT response!

Would it then be a logical extension of thought that the mandolin produces tonal qualities in a way very similar to the guitar?

[John Flynn]

This has been posted before, but it bears repeating. It is from "The History of the Classical Guitar"

"Manual Torres changed the design and construction of the guitar through experimentation. He once built a guitar with a wooden top and paper mache back and sides to prove his theory that it was the top that produced most of the volume. He is considered the 'father' of the modern guitar by many."

[LKN2MYIS]

Yes, however volume and tone are two different beasts.

Love to see that guitar, by the way.

[sunburst]

Would it then be a logical extension of thought that the mandolin produces tonal qualities in a way very similar to the guitar?

Not only logical, but also demonstrated with holography. Mandolins behave similarly to guitars, and a lot can be learned about how they work from reading the guitar studies.

[Bruce Clausen]

But we're not told whether Torres really did prove his point. If I'm not mistaken, that guitar still exists, but no one seems eager to play it. Along the same lines, a builder called Denis Grace once showed me a guitar he'd built with no upper bout, to prove that the upper bout contributes nothing to the guitar's sound. It was a good-sounding guitar, but hard to love.

BC

[sunburst]

A well known company made (makes) guitars (and mandolins) with a fiberglass bowl rather than wooden back and sides. They sound like guitars. It's a matter of opinion whether or not they sound like good guitars.

[LKN2MYIS]

Paper mache is a lot different than fiberglass tonally, I would imagine.

Any idea where Torres' guitar (or photos of it) can be seen?

[John Flynn]

I remember reading elsewhere, although I can't find the reference, that Torres did the experiment to prove the point to his fellow luthiers. He made the paper mache-backed guitar to take to some event where a lot of them would be present and he did blind "taste tests" with them A/B'ing t one of his regular guitars. Supposedly, both guitars came out about even in the taste tests. It's been a while since I read that, so I may not be 100% accurate on it.