Mandolin Back question

[Barbara Shultz]

I've noticed that mandolins come in one piece backs or two piece backs. What difference does that have on the mandolins... sound, tone, or just looks?

Barbara

[Jim Rowland]

Looks,usually price. I think the possible symmitries of the two piece are more attractive, but so awfully common,don't you know.
Jim

[Shawn Gambrel]

I heard that a one piece back has a darker tone to it

[Dave Cohen]

This topic has come up many times in the builders' section. The differences between one-piece backs and two-piece backs are entirely cosmetic.

A single glue seam vs no glue seam is an insignificant difference between back plates (or between top plates, for that matter). A good glue seam is between 0.002" and 0.006" thick. At that thickness, unless the joint has been glued with something like silicone caulk or rubber cement (both extremely bad ideas), the glue is hard enough that whatever excites the plate does not "see" the glue seam. In fact, the vibrational properties of plates are dominated by their masses, overall stiffnesses along the grain and across the grain, internal losses in the wood, etc. The excitations "see" only the overall properties of the plate. They do not "see" local inhomogeneities, etc. Differences in density, stiffnesses, and losses between plates render a good glue seam insignificant.

http://www.Cohenmando.com

[Capt. E]

Two piece backs are usually symmetrically matched, such as this: http://www.ellismandolins.com/Gallery06pages/page_2_06.htm

A single piece back tends to be from a piece of wood that has asymmetrical grain such as this: http://www.ellismandolins.com/gallery08pages/08pg_14.htm

[Bruce Evans]

They do not "see" local inhomogeneities, etc.

WoW!

"inhomogeneities"

I love it! Still trying to pronounce it, though. Are the last three syllables pronounced like, "deities?"

Is "inhomogenousness" a synomym?

[Dave Cohen]

...Except that "inhomogeneities" is a noun, whereas "inhomogeneousness" is an adjective.

Sorry about the geekspeak. It is, after all, my native language. :)

http://www.Cohenmando.com

[Mike Bunting]

...Except that "inhomogeneities" is a noun, whereas "inhomogeneousness" is an adjective.

Sorry about the geekspeak. It is, after all, my native language. :)

http://www.Cohenmando.com

Sorry, "inhomogeneousness" is a noun meaning the state of being inhomogeneous (which is the adjective).
The noun is "inhomogeneity".
Not that it matters but I am a language geek. :)

[man dough nollij]

It's a good thing that Waverly tuners are so well made...

I don't know if I could handle a thread about Waverlies' inhomogeneousnessness. :grin:

[Dave Cohen]

Sorry, I won't make that mistake again. In the future, I'll stick to generating the geekspeak; no more analyzing the geekspeak.

http://www.Cohenmando.com

[allenhopkins]

This thread is becoming increasingly inhomogeneous. Let's watch it, or we may have to rewrite the posting guidelines to discourage heterogeneity.

[Mike Bunting]

:))

[Joel Spaulding]

heterogenousness :whistling:

[man dough nollij]

heterogenousness :whistling:

I'm deathly afraid of that. I'm afraid I might have heterogenophobia!

What was the topic again? :popcorn:

[Mike Bunting]

I'd like to thank Dave Cohen for the generous (whether it be hetero, homo, or ideo) sharing of his knowledge regarding mando building.

[Jason Kessler]

Back to our regularly scheduled programming:

I've wondered about this question for some time. Logic would seem to dictate that a one-piece back would vibrate much more than a two-piece back. The two-piece would be, by definition, stiffer and less vibrant than a one-piece, no? Questions regarding the glue seam aside.

[Bill Snyder]

Back to our regularly scheduled programming:

I've wondered about this question for some time. Logic would seem to dictate that a one-piece back would vibrate much more than a two-piece back. The two-piece would be, by definition, stiffer and less vibrant than a one-piece, no? Questions regarding the glue seam aside.

I think the the answer is NO. BTW stiff when light is a good thing.

[Capt. E]

I was talking with Tom Ellis about this and he says that top thickness and shape is what determines degree and character of vibration. For example: thick topped mandolin will actually have more sustain (think electric guitar) with a loss of volume and complexities of tone. If a two piece back has any difference over a one piece, it is very very small.

Hope I got that right.

[Dave Cohen]

In the first place, what do you mean by "vibrate more"? Presumably, you mean that a one-piece plate can vibrate with greater amplitude. There is no physical reason why the center seam should restrict the amplitude. If a hard glue like hot hide glue, Titebond (a "PVA" glue), or even epoxy is used, and the glue seam is as narrow as it should be for a quality joint, the vibrations just don't see the glue seam at all. Reasoning is that the wavelengths of the vibrations are on the order of magnitude of the dimensions of the plate. Even the highest frequency plate vibrations (other than the physicists' curiosities above 1.5 or 2 kHz) have wavelengths on the order of, say, 10% of the plate dimensions. At those wavelengths, a glue seam thickness of 0.006" - or even 10 times that, for that matter - is invisible to the propagation of those waveforms. For a waveform to be affected by something as small as a glue seam, the wavelength would have to be on the order of 1 megaHertz (mHz = 1,000,000 Hz) or greater. Recall that your hearing cuts out at a maximum of 20 kHz (=20,000 Hz) if you are young and have very good hearing. PRobably more like 10 kHz or less for most of us.

Now, if you put a small but massive obstacle on the plate, such as a large brace, the lowest frequency vibrations of the plate will ignore it, but the highest frequency vibrations, having much smaller amplitudes, will see the brace as an obstacle, and consequently there will tend to be nodes, i.e., places of zero vibrational amplitude, at the brace location.

Oops, I did the geekspeak thing again, demonstrating that you can take the geek out of the classroom, but you can't take the classroom out of the geek.

http://www.Cohenmando.com

[BradKlein]

Putting aside the technical talk for a moment - it is often the case that one piece backs are plain sawn lumber, and two piece backs are more often nearly quarter sawn. (the latter method of milling lumber requires a larger tree, to put it simply) There is no reason that this HAS to be the case, but when it is, I think that most builders would agree that there is an affect on the sound of the finished instrument. I'll leave it to the builders to try and describe the differences, if they choose.

[Bertram Henze]

Reasoning is that the wavelengths of the vibrations are on the order of magnitude of the dimensions of the plate. Even the highest frequency plate vibrations (other than the physicists' curiosities above 1.5 or 2 kHz) have wavelengths on the order of, say, 10% of the plate dimensions. At those wavelengths, a glue seam thickness of 0.006" - or even 10 times that, for that matter - is invisible to the propagation of those waveforms. For a waveform to be affected by something as small as a glue seam, the wavelength would have to be on the order of 1 megaHertz (mHz = 1,000,000 Hz) or greater. Recall that your hearing cuts out at a maximum of 20 kHz (=20,000 Hz) if you are young and have very good hearing.

Summarizing this for the lesser geeks of us:

1. The influence of a glue seam is negligible compared to the impact of bracing.
2. In a Fourier analysis plot, that influence would only show at frequencies outside of human hearing.

Two assumptions go into this:
1. The glue seam is accoustically approx. as hard as the wood is (I suppose that is the case), or else there'd be reflections affecting also lower frequencies.
2. If the mandolinist's dog is present, he might or might not sing along, thus making the difference audible for his master :grin:

Bertram

[journeybear]

2. If the mandolinist's dog is present, he might or might not sing along, thus making the difference audible for his master :grin:

Or inaudible, if the dog won't stop! :)

BTW & FWIW, both Dave and Mike are correct: "inhomogeneity" and "inhomogeneousness" mean virtually the same thing, both appear in the dictionary, and the only error was Dave saying the latter was an adjective, for which he has been chided enough. Perhaps the two words should be put together in a blender and homogenized! ;)

Sorry, not only am I a language geek, I'm a meaning geek, and the former was invented and is supposed to serve the latter :grin:

[Capt. E]

Maybe it would be better if Tom Ellis threw in on this. All I really understood was that the thickness (or thinness) of the top has big effects.

[allenhopkins]

It's my understanding that the mandolin's back is mainly a sound reflector, and as such the difference between one-piece and two-piece would be pretty hard to detect. Unlike the violin, where the back is held free from the player's body, and the top's vibrations are specifically transmitted to the back via the sound post.

On the other hand, mandolin players report volume improvement by using attachments such as the Tone Gard, which hold the mandolin away from the body, allowing the back to vibrate more freely. I would speculate that a mandolin back held against the player's body would reflect sound waves equally well with one separated by a Tone Gard, so there may be more to the "back vibration" factor than I'm aware. If that's the case, thickness and composition of the back may be an important secondary variable.

[Jason Kessler]

OK, folks: if the mandolin back's vibrational and resonating qualities are negligible if not completely irrelevant, why are Tone Guards so popular?

[allenhopkins]

OK, folks: if the mandolin back's vibrational and resonating qualities are negligible if not completely irrelevant, why are Tone Guards so popular?

Question I just asked above. Could be that holding the instrument away from the player's body, just reduces a general dampening effect, as the body absorbs the vibrations of the back, which in turn reduces the vibration of the sides and top. It may not specifically be the back vibrations that produce the sound improvement Tone Guard owners report.

Also, I'd hazard a guess that the player hears more volume from an instrument that's suspended in front of him/her, rather than held against the body. Do members of the "audience" hear a difference, or is it just the person playing the mandolin?

What would be interesting, if some builder had the time and patience, would be to try a mandolin with several alternate back constructions -- one-piece, two-piece, thick or thin -- to see if a noticeable difference resulted.

Back in the 19th century, someone built a guitar with back and sides of papier-mache, just to prove the point that top vibrations were the crucial variable. Maybe so, maybe no...

[Dave Cohen]

The back of any stringed instrument is not mainly a reflector. The entire interior surface of the instrument body does some reflecting, but both the top plate and the back plate and even the ribs (at some frequencies) are also coupled vibrating members. The back plate does not simply reflect sound waves in air out of the sound holes. Rather, the air in the cavity acts as a spring to set the mass of air in the region of the soundholes in vibrational motion. The vibrating air in the soundhole(s) in turn sets nearby air in motion, and so on until eventually the motion is propagated to your eardrum. If you want to know more about how this works, the references to the Cohen & Rossing papers are on the "Acoustics" page of my website. Click on my url below, then go to the "acoustics" page for the references.

No one said that the vibrations of the back are unimportant; they are important. My point was simply that if any vibrations are going to be damped by a glue seam, they will have to be vibrations with wavelengths short enough (and hence the frequencies would have to be high enough) to be aware of the glue seam. Also, as I pointed out and Bertram reiterated, the prevalently used glues are hard enough that they do not result in significant damping as long as the glue seam is as small as glue seams in well-made instruments typically are.

http://www.Cohenmando.com