Jump to content
Maestronet Forums

How does a violin reproduce overtones? - Theorizing a model


Recommended Posts

1 hour ago, David Beard said:

It seems like the tension of the post setting matters greatly in terms of practical results.  But how can this tension be important but the string tension not be important.  That's confusing.

There will come a JASA article soon on this. The arch shape changes somewhat with the post tension. The article addresses how much the changes need to be before listeners are able to tell the difference.

Link to comment
Share on other sites

  • Replies 646
  • Created
  • Last Reply

Top Posters In This Topic

16 hours ago, Don Noon said:

Seems like there's the continuing concept of static stress somehow influencing body vibrations, which to first order it doesn't.

You are right in terms that once the stress has settled in there are no more changes. No problem with that. 
 

However I see a sound quality difference between instruments which need to stretch in and those which don’t (which has of course to do how solid they are made.) And this seems not to be measurable in the signature modes but rather anything what happens in the higher frequency regions. 
 

Furthermore it seems that the forces applying the stress to the body must be ‘balanced’ which means that the string angle can’t be too steep not too flat. 
 

Maybe a possible explanation for the difference in sound quality is that deflection (even in non-visible degree) must create in a plate stretch on one side and compression on the other side. I need to add that in the new concept violin some deformations were visible by eye notably the upper part of the body which curved upwards with the twist-pulling force of the neck. 

On 7/22/2021 at 11:49 AM, Don Noon said:

2) No, and No.  Marty's violins demonstrate that if you make something radically different, you can get radically different results... but not why.  A violin tuned to half tension will still have basically the same response profile*, and that's a much more direct test.

I would bet that if you use Marty’s tailpiece construction on any normal violin you will hear quite some difference in the sound. 
 

For the rest of his genial brain twisting construction I agree that it is too different to allow any comparison to normal violins. 

Link to comment
Share on other sites

15 hours ago, Anders Buen said:

There will come a JASA article soon on this. The arch shape changes somewhat with the post tension. The article addresses how much the changes need to be before listeners are able to tell the difference.

My general idea for sound post adjustments is that they serve the player but not the listener. In other terms it is much more about at which speed and which acceleration a good violinist can drive the violin. To make changes which can be perceived as ‘different sound’ by the audience pretty radical changes are necessary.

Link to comment
Share on other sites

8 minutes ago, Andreas Preuss said:

You are right in terms that once the stress has settled in there are no more changes. No problem with that. 
 

However I see a sound quality difference between instruments which need to stretch in and those which don’t (which has of course to do how solid they are made.) And this seems not to be measurable in the signature modes but rather anything what happens in the higher frequency regions. 
 

Furthermore it seems that the forces applying the stress to the body must be ‘balanced’ which means that the string angle can’t be too steep not too flat. 
 

Maybe a possible explanation for the difference in sound quality is that deflection (even in non-visible degree) must create in a plate stretch on one side and compression on the other side. I need to add that in the new concept violin some deformations were visible by eye notably the upper part of the body which curved upwards with the twist-pulling force of the neck. 

I would bet that if you use Marty’s tailpiece construction on any normal violin you will hear quite some difference in the sound. 
 

For the rest of his genial brain twisting construction I agree that it is too different to allow any comparison to normal violins. 

To me the difference is the thickness/stiffness of the back and how it is incorporated into the system and how it interacts with all those things. Ones that need to "stretch" are often thinner back that will slowly creep to final shape, whereas a thicker spine back is less prone to needing to settle as much as the back is much stiffer and resists the need to "settle"

Link to comment
Share on other sites

Link to the article abstract. https://asa.scitation.org/doi/full/10.1121/10.0005587

Perception of violin soundpost tightness through playing and listening tests

 
The Journal of the Acoustical Society of America 150, 540 (2021); https://doi.org/10.1121/10.0005587

ABSTRACT

This study involved playing and listening (using recorded sounds) experiments to investigate how changes in soundpost length (for a fixed soundpost position) affect the perceptual qualities of the violin and what the threshold of change is below which players and luthiers do not perceive differences. A length-adjustable carbon fiber soundpost was employed. During the playing experiment, subjects played a provided violin on which the soundpost length was modified by the experimenter to find their optimal soundpost lengths. Then the experimenter varied the soundpost length randomly in ten trials within ±0.11 mm around their optimal lengths and asked subjects to always compare with the previous setting. The results showed that subjects' optimal soundpost lengths varied from 0.132 to 0.616 mm relative to the original length (53 mm), but subjects could not recognize length variation of 0.11 mm or less at above chance levels. During the listening experiment, subjects listened to 16 pairs of recordings through a computer interface and were asked, for each pair, whether the violin setup was the same or different. The results showed that subjects could differentiate soundpost lengths with a difference of about 0.198 mm at better than chance level.
Link to comment
Share on other sites

3 minutes ago, Anders Buen said:

Link to the article abstract. https://asa.scitation.org/doi/full/10.1121/10.0005587

Perception of violin soundpost tightness through playing and listening tests

 
The Journal of the Acoustical Society of America 150, 540 (2021); https://doi.org/10.1121/10.0005587

ABSTRACT

This study involved playing and listening (using recorded sounds) experiments to investigate how changes in soundpost length (for a fixed soundpost position) affect the perceptual qualities of the violin and what the threshold of change is below which players and luthiers do not perceive differences. A length-adjustable carbon fiber soundpost was employed. During the playing experiment, subjects played a provided violin on which the soundpost length was modified by the experimenter to find their optimal soundpost lengths. Then the experimenter varied the soundpost length randomly in ten trials within ±0.11 mm around their optimal lengths and asked subjects to always compare with the previous setting. The results showed that subjects' optimal soundpost lengths varied from 0.132 to 0.616 mm relative to the original length (53 mm), but subjects could not recognize length variation of 0.11 mm or less at above chance levels. During the listening experiment, subjects listened to 16 pairs of recordings through a computer interface and were asked, for each pair, whether the violin setup was the same or different. The results showed that subjects could differentiate soundpost lengths with a difference of about 0.198 mm at better than chance level.

I'd like to see the same test done with a regular wood post that remains the same length but is tapped to various locations, thus changing the "squeeze" factor via location change, something I think is much more "real world" , but it is interesting that the increments are so small yet not surprising.

Link to comment
Share on other sites

I was surprised to see how small changes they made. I would have thought it would be possible to change the lenght by 1-1,5mm maybe even 2mm. But I am a bit rusty now on this. Moisture content variation may give arch height changes of order mm. I wonder what typical soundpost summer and winter season length differences are in average.   

Link to comment
Share on other sites

18 hours ago, jezzupe said:

Would that not be somewhat dependent on the backs stiffness? as the post is supported by the "floor" or back. This to me is one of the head scratcher's in that in my own instruments on occasion, as well as many examples out there, we can have some pretty dramatic graduation differences with some pretty thick spine'd backs, where the back would seem to be much stiffer,much more resistant to body curl from the pulling forces as compared to thinner more flexible backs that would seem to deflect much more than a thick spine'd one , yet I heard both thin and thick back violins sound good, which leads to I think the agreed on "well the top is doing most of the work"

I am getting more and more the idea that the stability of the back is made by the area around the border and not the center. This would more or less explain why center thickness can have from one instrument to another a pretty big difference and still both instruments would sound well.

However I see the central thickness as a sort of fly wheel mass. The more weight is accumulated in thickness and size the more a player needs to ‘dig’ into the strings to move it.

Link to comment
Share on other sites

1 hour ago, Andreas Preuss said:

1)  However I see a sound quality difference between instruments which need to stretch in and those which don’t (which has of course to do how solid they are made.) And this seems not to be measurable in the signature modes but rather anything what happens in the higher frequency regions. 

2) I would bet that if you use Marty’s tailpiece construction on any normal violin you will hear quite some difference in the sound. 

1) As I have mentioned many times, there is a documented transient effect (mostly damping) where damping is high after applying a static force (i.e. putting on strings), and takes some time to settle in or "stretch".  This would be most evident in the higher frequencies.

2)  I wouldn't bet against you... but I would bet that the sound difference would be mainly due to all of the modes of the cantilevered rod going under the bridge to take the static string forces, and not the different loadpath of the static forces themselves.

1 hour ago, Anders Buen said:

I don't have access to the full article, but was there any analysis or educated hypothesis of WHY soundpost length affects the sound?  Or of how much preload pressure varies with soundpost length?  The latter one is fairly easy to measure roughly, which I will do if I get around to it.

Link to comment
Share on other sites

1 hour ago, Don Noon said:

I don't have access to the full article, but was there any analysis or educated hypothesis of WHY soundpost length affects the sound?  Or of how much preload pressure varies with soundpost length?  The latter one is fairly easy to measure roughly, which I will do if I get around to it.

I can't find any by a fast read through of the discussion and conclusion. Rodgers and his co worker showed by FEA calculations of violin bodies that modes could move from changes of the arch shape. So that is my best guess. 

Link to comment
Share on other sites

3 hours ago, Anders Buen said:

I was surprised to see how small changes they made. I would have thought it would be possible to change the lenght by 1-1,5mm maybe even 2mm. But I am a bit rusty now on this. Moisture content variation may give arch height changes of order mm. I wonder what typical soundpost summer and winter season length differences are in average.   

 

 At Morel we measured the difference of before and after when setting a new post. Mostly it was around 0.5mm but occasionally 1mm and more.  Though in technical terms it is possible to squeeze a plus 2mm sound post into a violin. 

Link to comment
Share on other sites

20 minutes ago, Anders Buen said:

Rodgers and his co worker showed by FEA calculations of violin bodies that modes could move from changes of the arch shape. So that is my best guess. 

That would be my guess as well.  The detectable zone for change seems to be in the .1 - .2mm range and up, and since the back is far more rigid than the top, I would expect almost all of that to go into an arching change of the top.  And I would have guessed (even before seeing these results) that a top arch difference of a fraction of a mm, while not earthshakingly different, might be detectable.  1 or 2 mm difference in arch height I think would be very obviously different in tone.

Link to comment
Share on other sites

1 hour ago, Don Noon said:

2)  I wouldn't bet against you... but I would bet that the sound difference would be mainly due to all of the modes of the cantilevered rod going under the bridge to take the static string forces, and not the different loadpath of the static forces themselves.

Hmmm, really?

Just in terms of mechanics the downforce on the bridge is only governed from the point at the neck heel and upper block making it much more unstable.

Secondly Marty’s tailpiece setup does not allow a sideway  pivoting motion. This blocks more or less all the signature modes which need a rotation motion of the bridge. (though this doesn’t have anything to do with load path.)

Link to comment
Share on other sites

7 minutes ago, Andreas Preuss said:

Secondly Marty’s tailpiece setup does not allow a sideway  pivoting motion. This blocks more or less all the signature modes which need a rotation motion of the bridge. (though this doesn’t have anything to do with load path.)

Signature modes, or other modes, do react to excitation with the bridge foot in parallell too. Tapping the bridge straigt down does make sound..

Link to comment
Share on other sites

9 minutes ago, Andreas Preuss said:

Just in terms of mechanics the downforce on the bridge is only governed from the point at the neck heel and upper block making it much more unstable.

Ff hole width, holes diameter and possibly length along with bassbar surely has a say too.

Link to comment
Share on other sites

1 hour ago, Andreas Preuss said:

Hmmm, really?

Just in terms of mechanics the downforce on the bridge is only governed from the point at the neck heel and upper block making it much more unstable.

Secondly Marty’s tailpiece setup does not allow a sideway  pivoting motion. This blocks more or less all the signature modes which need a rotation motion of the bridge. (though this doesn’t have anything to do with load path.)

 

The bridge does rotate to give both horizontal and vertical movements.

I use something similar to the ancient Welsh crwyth where the sound post goes through a hole in the top plate.  My treble bridge foot rests directly on the sound post and it doesn't touch the top plate at all and this reduces the string tension downward force on the top plate by about half.

A very shallow string angle over the bridge also reduces the string tension downward load by another half so the total downward force on the top plate is only about 1/4 the normal amount.

The longitudinal string tension buckling compressive force on the top plate is eliminated by having the strings attached to the tailpiece extension of the fingerboard instead of the usual end pin /saddle.

These reductions of loads on the top plate allows the use of a thin flat plate made from low density (0.28g/cc) Pawlonia wood which reduces the plate's weight.  The elimination of the arching reduces the plate's stiffness and the combination of these two effects increases the plate's admittance which increases the instrument's sound output.

The reduction of plate stiffness also increases the modal density which means there are more resonance peaks in the frequency response curve which in turn makes the various notes in a scale more even in loudness.

Other body shape and material changes further make the frequency response curves quite different and therefore sound quite different from the usual "old Italian" character--

There's no sense in beating a dead horse.

 

2021_07_23_0503.JPG

Link to comment
Share on other sites

Just thinking about coiled metal springs as an analogy for other systems of mechanical oscillations, obviously there are physical boundaries to free oscillation behaviors, and related driven behaviors.

The conditions for free oscillation of a weight hanging from the string in two points.   1) the spring must be sufficiently load to stretch.  2) the load must not overwhelm or break or plastically deform the spring.

These can reduce to one condition that the spring and the mass reasonable balance each other.

It seems possible to me that both the string tension and the post tension may need to be rather high to load the twisting of the treble side bridge table and allow that dynamic to act as an admittance path for high frequencies as deflections of the plate.

Also, a spring loaded by a free weight is not as good an analogy for this twisted bridge table analogy as would be something trapped between two oppossing springs that are loaded by stretching. 

In this analogy, the weight of the trapped something is not necessarily a primary factor. And, in fact, the something could be more of transmission point than the main thing being oscillated or driven. It could even be something like a transmission lever.

But, the interesting thing here is that a close good balance between the opposing springs would greatly contribute to the freedom in this kind of system.

 

Perhaps that is what's happening, why the tensions matter.   The twisting dynamic is present at all until you introduce the opposition of the post and loaded bridge foot.  And, the dynamic isn't basically free unless the post tension and string tension are both sufficient.  And, the dynamic isn't very very free until the tensions find a good balance.

 

It's an idea. And at least it roughly confirms to setup experiences.

Link to comment
Share on other sites

4 hours ago, Don Noon said:

of how much preload pressure varies with soundpost length?  The latter one is fairly easy to measure roughly, which I will do if I get around to it.

OK, I got around to it.  I won't go into the gory details of my measurements and calculations, but...

The top is very wimpy compared to the back.  When strings are put on (11 pounds vertical due to A and E only), the treble side sinks only .32mm relative to the endblocks, due to the very stiff soundpost and back.

The added vertical force due to even a tight soundpost isn't much.  If you put in a post .32mm taller than the gap (which would raise the top to counteract the sinking when strings are put on), it only takes 1.8 pounds vertical.  The back doesn't move, but the top does.

Link to comment
Share on other sites

12 minutes ago, David Beard said:

But, the interesting thing here is that a close good balance between the opposing springs would greatly contribute to the freedom in this kind of system.

As I understand your thinking, no.  

The natural frequency of any system is governed by (k/m)^.5, where the spring rate k is the addition of the rates of the "opposing" springs.  The don't oppose... they add together for the net spring rate.

Also, the spring rate, and therefore the natural frequency, are independent of how much the springs are pre-compressed, or loaded, unless you happen to be using non-linear springs.

Link to comment
Share on other sites

Just now, Don Noon said:

As I understand your thinking, no.  

The natural frequency of any system is governed by (k/m)^.5, where the spring rate k is the addition of the rates of the "opposing" springs.  The don't oppose... they add together for the net spring rate.

Also, the spring rate, and therefore the natural frequency, are independent of how much the springs are pre-compressed, or loaded.

That can't be a complete picture.  If I take a spring that barely begins to stretch with 10lb load, say it stretch 5mm.  And, if I pair this with another spring that stretches say 60mm with an 8ounce load.  Well, if I trap something of negligible weight between these, the lighter spring will only barely be able to engage the heavier spring.  You will be technically correct. But the combined system will behavior very much as the unloaded heavier spring would.  That is that light pulls on the trapped object would only barely stretch.

I guess the core intuition of my earlier suggestion, that part that is either right or perhaps wrong, is the intuition that if you trap something between two of the heavy springs, and then stretch them a total 60mm, it should now take less than 10lbs force to move the trapped object 5mm.

Hmmm?

Link to comment
Share on other sites

1 hour ago, Don Noon said:

As I understand your thinking, no.  

The natural frequency of any system is governed by (k/m)^.5, where the spring rate k is the addition of the rates of the "opposing" springs.  The don't oppose... they add together for the net spring rate.

Also, the spring rate, and therefore the natural frequency, are independent of how much the springs are pre-compressed, or loaded, unless you happen to be using non-linear springs.

 

53 minutes ago, David Beard said:

That can't be a complete picture.  If I take a spring that barely begins to stretch with 10lb load, say it stretch 5mm.  And, if I pair this with another spring that stretches say 60mm with an 8ounce load.  Well, if I trap something of negligible weight between these, the lighter spring will only barely be able to engage the heavier spring.  You will be technically correct. But the combined system will behavior very much as the unloaded heavier spring would.  That is that light pulls on the trapped object would only barely stretch.

I guess the core intuition of my earlier suggestion, that part that is either right or perhaps wrong, is the intuition that if you trap something between two of the heavy springs, and then stretch them a total 60mm, it should now take less than 10lbs force to move the trapped object 5mm.

Hmmm?

I think I'm seeing.  Seems I have some bogus notions floating around in my thoughts about springs and oscillation.

Link to comment
Share on other sites

8 hours ago, Andreas Preuss said:

I am getting more and more the idea that the stability of the back is made by the area around the border and not the center. This would more or less explain why center thickness can have from one instrument to another a pretty big difference and still both instruments would sound well.

However I see the central thickness as a sort of fly wheel mass. The more weight is accumulated in thickness and size the more a player needs to ‘dig’ into the strings to move it.

It's both, the outer edge along with the re-curve , the way you approach the gluing table for the ribs DEFINITELY has to do with torsion motion, and the thicker spine is what will resist the nut break to endpin pull that wants to fold the violin in half 

to me that need to dig in is because the "floor" that the post is sitting on is very stiff and will resist syncing with the tops undulations, it will take more bow energy to "get it moving" 

again back to the original "reproduction of overtones"....I had made post that Don responded to, but after getting roughed up by Noon the Science Goon, of course realizing he was right based on how I was trying to express myself....I was talking about "hyperflapping"  and "event horizons" and "nodal areas"  Of course it made perfect sense to me at the time, but I now realize I did not express myself properly...."hyperflapping" should be "Flutter" "event horizons" should be "outer edges of regions of air turbulence driven by frequency vibration" and "nodal regions" should be changed to "surface face"

I think the problem with what we are interested in is that it is a system that is highly subject to fractal Mandelbrot like mathematics related to examination of regions and their reactions and interactions within the entire system

Meaning first we look at the motions of the entire violin, the undulations, then we could zoom in, then pick a quadrant to examine ,and then zoom in further and further to specific localities to examine the factors there, then of course knowing we are now only examining one tiny area, we need to zoom out and look here, there and then "there" , and then of course once we have brought ourselves down to a lower level we loose sight of the connections of the other areas and how interactions may contribute to results that can not be known unless looking at a bigger picture and then we need to zoom out ...

and so at a certain point, knowing that "tone" is subjective enough to have a broad consensus of no consensus pretty much across a broad spectrum of musical styles one has to make a choice of what kind of sound they are shooting for, guys who build bluegrass fiddles are not really shooting for the same tone quality as someone going for a powerhouse romantic violin for classical , like a Strad  or someone who build period instruments for Barouqe, all kind of different tone goals ....


And then once you decide what kind of sound you want to attain, you have to decide if you want to actually build violins and try to get them into musicians hands, or do you want to spend most of your time trying to figure out what makes it tick.

I hate to say it but I think if Don pulled a Methuselah and ended up living for a thousand years , we would have a thousand years of fantastic analysis with tests and results but not much closer to "it"

because there is no it

Most all violin, or violin like objects will produce and then reproduce overtones, we must choose how much, in what way and then really "why and how" also , and then ask how much time we want to put into finding that out vs being in the shop building 

Link to comment
Share on other sites

3 hours ago, David Beard said:

 

I think I'm seeing.  Seems I have some bogus notions floating around in my thoughts about springs and oscillation.

There's a lot of that going around these days, and as yet a vaccine for it hasn't been developed.

Link to comment
Share on other sites

13 hours ago, Anders Buen said:

Link to the article abstract. https://asa.scitation.org/doi/full/10.1121/10.0005587

Perception of violin soundpost tightness through playing and listening tests

 
The Journal of the Acoustical Society of America 150, 540 (2021); https://doi.org/10.1121/10.0005587

ABSTRACT

This study involved playing and listening (using recorded sounds) experiments to investigate how changes in soundpost length (for a fixed soundpost position) affect the perceptual qualities of the violin and what the threshold of change is below which players and luthiers do not perceive differences. A length-adjustable carbon fiber soundpost was employed. During the playing experiment, subjects played a provided violin on which the soundpost length was modified by the experimenter to find their optimal soundpost lengths. Then the experimenter varied the soundpost length randomly in ten trials within ±0.11 mm around their optimal lengths and asked subjects to always compare with the previous setting. The results showed that subjects' optimal soundpost lengths varied from 0.132 to 0.616 mm relative to the original length (53 mm), but subjects could not recognize length variation of 0.11 mm or less at above chance levels. During the listening experiment, subjects listened to 16 pairs of recordings through a computer interface and were asked, for each pair, whether the violin setup was the same or different. The results showed that subjects could differentiate soundpost lengths with a difference of about 0.198 mm at better than chance level.

Dear Anders,

I'm trying to sell my violins at very modest prices so I can afford to purchase these ASA violin research publications which hopefully will teach me something of value so I can raise my violin prices so I can buy more of these publications.

 

 

 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share

  • Recently Browsing   0 members

    No registered users viewing this page.




×
×
  • Create New...