Jump to content
Maestronet Forums

How does a violin reproduce overtones? - Theorizing a model


Andreas Preuss

Recommended Posts

22 hours ago, Carl Stross said:

Well, somehow it's all of the above.... 

I think that's the spirit - asking questions, talking to people who might make good suggestions, lots and lots of critical listening etc etc etc. 

Wonderful post ! I must say I really enjoy your posts - always a different angle, something to think about.

thanks Carl

Link to comment
Share on other sites

  • Replies 646
  • Created
  • Last Reply

Top Posters In This Topic

21 hours ago, David Beard said:

For me, I'm entirely classically oriented, so it would be a violin that most of all sounds and plays like the violins people like Perlman and Vengerov treat as 'the best'.  So, like a very good and very happily set up and settled in Strad or DG.   If I could reliable deliver such instruments without the burden of an historical instrument price, I would consider myself to have achieve all the better/best I would care to aim for.

But, this goal is as much about the players feel and experience producing the sounds as it about the sounds produce.

Actual 'Better' would be about delivering wider range of player choice and command in tone color and articulation, greater player ease in reaching any expression they desire, and perhaps with better power, clarity, definition, etc.

 

 

Well, ya , what Don said

Link to comment
Share on other sites

15 hours ago, Andreas Preuss said:

I hope I didn’t say ‘better’ anywhere. All performers have personal choices and even on the top level of instruments there are instruments adored by some players and completely rejected by others. 

I am working on a sound which is rich and dense, flexible under the bow arm and crusted with overtones. Some performers will like this sound others won’t. I see it similar to the work of a composer defining his own unique style. I see in assymtric design new possibilities to define my own unique sound. (Or somehow I got tired in repeating Strad and DG.) 

improve generally implies better

Link to comment
Share on other sites

Again I will simply say that the best way to produce "improved/better" violins is to build as many violins as you can using the skills that you have learned over the years, using intuitions based off knowledge learned and that optimization of what ever is in your hands that you have made from scratch at that moment is what you should focus on.

I consider it like a race that starts and as it progresses you have to rely on your training {literally or figuratively} to try to optimize your pace and output, method and technique, arching has to be good, all the proportions have to be right , the glue joints tight, the glue mix right, the thickness/thinness, hand manipulation, flexing tapping, rubbing, listening, your ability to pick good wood, after all that there is no predicting the sound until it's strung, settled and cured 

with all the experiments, data and test's we're really "nowheresville" with having any universal information that "anyone" can use to "improve" a violin to make it "better" even if there are those that are convinced "there way" is "the way"

 so I say in the end just keep on building, what else can you do? the science of statistics will shine through as one who has built 150 violins has a much better chance of having "really good" ones and more of them that someone who builds 15

Link to comment
Share on other sites

On 8/5/2021 at 8:31 AM, Andreas Preuss said:

 

 

>

What I find more interesting is that the air volume is 25% smaller than a regular violin with a normal frequency around 270hz. this makes me wonder if less air mass can create a stronger sound. 

 

 

That's an interesting question.  

Suppose we accept the idea that a vibration variation in air pressure is what causes sound it then follows that in order to produce a loud sound  we should have a large variation in sound pressure.  Since the air pressure in a container is inversely proportional to its volume it follows that a proportional change in air pressure (deltaP/P) follows proportional change volume (delta V/V).

Assume the change in volume (delta V) is equal to surface area of the top plate A times the amount that it deflects d.

This suggests that the plate should be large in area and highly flexible.  And the rib height should be shallow to have a small violin cavity volume.

A large instrument with shallow ribs might be louder than a smaller instrument with deep ribs.

 

Link to comment
Share on other sites

7 hours ago, jezzupe said:

improve generally implies better

Yes, that’s correct. 
 

If there is anything which can be improved it is IMO the mysterious sound projection effect. In this sense I understand ‘improve’ as a better understanding how to make this work best for every instrument. If this is in the end an improvement compared to famed fiddles depends on judgement of the performers. As there are other aspects involved it is a pretty difficult thing. 
 

Following acoustic discussions I am not sure if the so called signature modes can give a complete answer:to this. Just by instinct I think it is better to focus on the whole overtone spectrum and how to ‘configure’ it in the making process. After many experiments now I am more and more convinced that wrong material choice for the top will automatically destroy good results, but good material choice alone is not the recipe to make the sound projection work. 
 

i see two main factors which need to be aligned. Top dimensions in thickness arching, breast width of the f holes and how string forces are acting on it. Only recently I started to think in which terms the weight of the top matters. Rather than thinking of the entire mass of the loose plate it might be more precise to think of the mass of the convex shaped portion as the main factor. (Likewise the curtate cycloid type arching diminishes the convex surface and at the same time stabilizes the convex part.)

I see improvements in violin making in using new techniques in combination with new materials to make violins lighter with the same or better structural strength. In this field I see the neck, the bass bar shape and the ribs as largely unexplored fields. In the end I am convinced that total weight matters in relation to above mentioned acoustic challenges.

All this said, I don’t think one can ‘improve’ the concept of Strad, one needs to think bout a different concept from the beginning to make the most useful tool for musicians standing on the stage in front of the audience. 

Link to comment
Share on other sites

5 hours ago, Marty Kasprzyk said:

That's an interesting question.  

Suppose we accept the idea that a vibration variation in air pressure is what causes sound it then follows that in order to produce a loud sound  we should have a large variation in sound pressure.  Since the air pressure in a container is inversely proportional to its volume it follows that a proportional change in air pressure (deltaP/P) follows proportional change volume (delta V/V).

Assume the change in volume (delta V) is equal to surface area of the top plate A times the amount that it deflects d.

This suggests that the plate should be large in area and highly flexible.  And the rib height should be shallow to have a small violin cavity volume.

A large instrument with shallow ribs might be louder than a smaller instrument with deep ribs.

 

I see sound volume always in combination with the singer Formant. And so far I am sticking to the simplified idea that ‘air pumping’ combined with high amplitudes for overtones in the singer Formant area make a powerful sound. Somehow I Imagine that the air pumping waves carry the overtones to the ear of the listeners and in this game less mass in the air pumping mechanism works better under the condition that singer Formant frequencies are present. 
 

I posted the latest development of the new concept violin in the makers bench section.

Edited by Andreas Preuss
Addition
Link to comment
Share on other sites

46 minutes ago, Andreas Preuss said:

If there is anything which can be improved it is IMO the mysterious sound projection effect...
I am not sure if the so called signature modes can give a complete answer to this. 

Just by instinct I think it is better to focus on the whole overtone spectrum 

Here's a view of the sound of an open G string.  Two of the vertical lines are the result of signature mode response.

For the upper D string and low A string, it would be one line.  For the upper A string and all of the E string, zero.

691561777_Gspectrum.jpg.334745d867edcaaccfca54de213a1b7f.jpg

Link to comment
Share on other sites

5 hours ago, Don Noon said:

Here's a view of the sound of an open G string.  Two of the vertical lines are the result of signature mode response.

For the upper D string and low A string, it would be one line.  For the upper A string and all of the E string, zero.

691561777_Gspectrum.jpg.334745d867edcaaccfca54de213a1b7f.jpg

This would look more realistic as the listening system work by using a logarithmic scale for the frequency axis. The loudness of this tone can be predicted with fairly high precision by just including the first two harmonics. I think the second harmonic can be included in the signature mode range. 196Hz x 3 = just under 600 Hz. 

Link to comment
Share on other sites

5 hours ago, Anders Buen said:

This would look more realistic as the listening system work by using a logarithmic scale for the frequency axis. 

Yeah, but it looks more dramatic with the linear scale.  

"Loudness" might be mostly about the first few peaks, but clarity and projection I think are more in the next several peaks.

Link to comment
Share on other sites

17 hours ago, Marty Kasprzyk said:

Suppose we accept the idea that a vibration variation in air pressure is what causes sound it then follows that in order to produce a loud sound  we should have a large variation in sound pressure.  

 

 

12 hours ago, Andreas Preuss said:

I see sound volume always in combination with the singer Formant. And so far I am sticking to the simplified idea that ‘air pumping’ 

Marty didn't say air pumping, he said air pressure.

One day earlier this week I see a small swimming pool with one of those inflatable top rings.  Why during cooler overnight temperatures does the inflatable ring deflate some and then during a hot afternoon just seems to reinflate itself.  Observing that reminded me of playing outdoor basketball during the cooler parts of the year.  Why does a basketball bounce/dribble normaly at 53 F. but doesn't want to work well at 52 F.?  This was tested over a few years time period - 52 F. just isn't as good. for a basketball.

Now for a violin.  It just sets unplayed at a comfortable room temperature most days around 75 F.  Then it's pick up to play and eventually temperature may reach close to 100 F. - I'm thinking inner pore pressure of wood because of the higher temp. may be the main reason of overtone production after a bit of playing.

Blame my reply on observation of a cheap swimming pool.

Link to comment
Share on other sites

9 hours ago, uncle duke said:

I'm thinking inner pore pressure of wood because of the higher temp. may be the main reason of overtone production after a bit of playing.

1. How do you know that the pores can store air pressure?

2. Overtone production does not increase over a time of minutes. If anything, the violinist (and here I mean a really well trained violinist) figures instinctively out how to trigger the overtones with his/her bow arm after a few minutes. 

3. Most immediately audible sound changes are more related to humidity. I heard a recital some years ago where it was clear that violinist was fighting with his instrument because the humidity in the small hall rapidly went up with 200 people in the room. 

Link to comment
Share on other sites

1 hour ago, Andreas Preuss said:

Overtone production does not increase over a time of minutes. If anything, the violinist (and here I mean a really well trained violinist) figures instinctively out how to trigger the overtones with his/her bow arm after a few minutes. 

Agreed.  I think there might also be some heightened perception of sound after some minutes of playing, or in other words, play-in of the human auditory system. Why would anyone be convinced that an organic electro-chemical system would be stable?

Link to comment
Share on other sites

On 5/25/2021 at 7:46 AM, Don Noon said:

...I think that the body will radiate sound from where it does so most effectively, and the the energy will come in from the bridge and circulate around the structure until it finds that place. 

Wow, that was over 2 months ago.  I have done some more thinking and research since that original statement, and while it's still close to what I think now, there are some differences.  By "circulate around the structure", there certainly must be some zones that are coupled more strongly to the bridge than others, thus can be better energized.  Secondly, the energy will find spots that vibrate most strongly at the frequency in question... but there is some more complexity here...

At the higher frequencies, there are multiple spots vibrating in opposite phases, and these spots are quite close together, often less than 1/2 air wavelength apart.  Being so close together and in opposite phase means some sound cancellation, or less net sound than the amplitude of the surface motion might imply.   You'd like to do something to minimize motion of the areas that are out-of-phase with the dominant sound-producing areas, if possible.  Although the pattern of vibration is different at different frequencies, there might be some way to use arching and graduations to help at least somewhat.  I have some ideas.

I feel an experimental top coming soon.  Sigh... it's not going to help my production schedule, but hopefully will help future production sound.  Or not.

Link to comment
Share on other sites

1 hour ago, Don Noon said:

At the higher frequencies, there are multiple spots vibrating in opposite phases, and these spots are quite close together, often less than 1/2 air wavelength apart.  Being so close together and in opposite phase means some sound cancellation, or less net sound than the amplitude of the surface motion might imply.   You'd like to do something to minimize motion of the areas that are out-of-phase with the dominant sound-producing areas, if possible.  Although the pattern of vibration is different at different frequencies, there might be some way to use arching and graduations to help at least somewhat. 

To know really what arching does it would need to make a low top and then reform the arching like in restorations to a higher arch. Making the experiment with different tops, the different wood properties might smudge up the result.

What I find strange is that Violins with a bigger projection are usually flatter arched and often pretty thinly built. Both would rather create a negative effect for what you are thinking about. 

It seems clear that a certain stiffness in that area makes a difference but I wouldn’t look on the area alone. For overtones I am more thinking of the entire structure, reason why things like neck angle do matter a lot. 
 

So far most of my experiments which could improve the clarity and crispness of the sound (what you perceive as overtones) helped stiffening the top in cross grain direction. This must not be in the area under the fingerboard and I would mostly look on the arching and thickness between the center bout above the f holes. 
 

 


 

 

Link to comment
Share on other sites

10 hours ago, Andreas Preuss said:

To know really what arching does it would need to make a low top and then reform the arching like in restorations to a higher arch. Making the experiment with different tops, the different wood properties might smudge up the result.

What I find strange is that Violins with a bigger projection are usually flatter arched and often pretty thinly built. Both would rather create a negative effect for what you are thinking about. 

It seems clear that a certain stiffness in that area makes a difference but I wouldn’t look on the area alone. For overtones I am more thinking of the entire structure, reason why things like neck angle do matter a lot. 
 

So far most of my experiments which could improve the clarity and crispness of the sound (what you perceive as overtones) helped stiffening the top in cross grain direction. This must not be in the area under the fingerboard and I would mostly look on the arching and thickness between the center bout above the f holes. 
 

 


 

 

It has been known for a long time that high arches and thick plates suppress the lower end of the violin's frequency response curve.  This gives the impression that the violin sounds bright.

Flat and low arched plates and thin plates have a similar amounts of high end sound but since the lower end has higher amplitudes the violin sounds deeper and louder.  

Attached are some graphs showing these effects from George Bissinger's  2019 VSA Papers translation of Hermann Meinel's 1937 study which is also attached.

This works suggests two other combinations should have been studied: High arched thin plates, and low arched thick plates. What is finally selected is a matter of personal taste regarding the low end/high end balance.

 

Screen Shot 2021-08-10 at 8.58.08 AM.png

Screen Shot 2021-08-10 at 8.59.31 AM.png

Bissinger's Tran. of Meinel .pdf

Link to comment
Share on other sites

1 hour ago, Marty Kasprzyk said:

Flat and low arched plates and thin plates have a similar amounts of high end sound but since the lower end has higher amplitudes the violin sounds deeper and louder.  

While I agree about the low frequencies not liking high arches and/or thick plates, it has been my experience that higher arching gives slightly more power in the high frequencies.

But the main effect of high arching I think is to suppress the midrange, a significant tonal effect (which also reduces net power).

Link to comment
Share on other sites

18 hours ago, Don Noon said:

Wow, that was over 2 months ago.  I have done some more thinking and research since that original statement, and while it's still close to what I think now, there are some differences.  By "circulate around the structure", there certainly must be some zones that are coupled more strongly to the bridge than others, thus can be better energized.  Secondly, the energy will find spots that vibrate most strongly at the frequency in question... but there is some more complexity here...

At the higher frequencies, there are multiple spots vibrating in opposite phases, and these spots are quite close together, often less than 1/2 air wavelength apart.  Being so close together and in opposite phase means some sound cancellation, or less net sound than the amplitude of the surface motion might imply.   You'd like to do something to minimize motion of the areas that are out-of-phase with the dominant sound-producing areas, if possible.  Although the pattern of vibration is different at different frequencies, there might be some way to use arching and graduations to help at least somewhat.  I have some ideas.

I feel an experimental top coming soon.  Sigh... it's not going to help my production schedule, but hopefully will help future production sound.  Or not.

this is a more sciency kinda way of describing what I was trying to describe.....yes I agree 100% imo graduations along with arching "benchmarks" create the ability to "control" particularly those very close highs where "action" is happening. This starts to bump into "wolfs" a bit and good cancellation vs bad, or constructive or destructive forces. 

And or if we can locate specific regions that we want to minimize out of phase motion counter acting negatively motions we do want, that it is simply a matter a fattening that area* to prevent as much motion as compared to one next to it, assuming at this point the arching is "done" and your working with what you have established as a benchmark

So, what I've been doing is trying to see if I can create "braile maps" of really good violins and compare them to bad ones, this is very early stage stuff with limited pool of instruments, but to the best of my knowledge I have not read much on anyone trying to approach it from my angle.

I think in simple terms I'd like to know if all very good violins have similar "patterns" of activity when examined with the hand during playing in order to create "vibration maps"  and that if bad ones are also similar. all with the intent of trying to see if there is not some way to "feel" if there are areas that would be better thinner or thicker in order to establish some method of repeatability as well as teachabilty so other could do the same 

I do think of it like trying to establish a form of braille for violin making that allows a builder to

1. know where a good violin should be moving/ vibrating the most, as well as know the dead zones, for all played frequencies 

2. to use the simple yet highly effective sense of touch to be the tool to determine this {anyone can do it} and it can be extremely accurate 

3. with the hope that not only can amplitude be maximized , or a high ceiling can be installed, and volume can be called if needed but perhaps more importantly , hoping advanced understanding of a "vibration map" can help  one "blend" all the regions together cohesively with some level of deterministic control. 

I think we'd all love our own 3d map maker so we could "look" at our progress, I "think" that "feeling" it {literally} may not only be a cheaper {free} way of doing a very similar thing that may even be more accurate and detailed region to region.

It's early stage stuff, but there seems to be something to it.

*there is no simple, by thickening one area to counteract an "issue" may create another one somewhere else

 

Link to comment
Share on other sites

42 minutes ago, jezzupe said:

I think in simple terms I'd like to know if all very good violins have similar "patterns" of activity when examined with the hand during playing in order to create "vibration maps"  and that if bad ones are also similar. all with the intent of trying to see if there is not some way to "feel" if there are areas that would be better thinner or thicker in order to establish some method of repeatability as well as teachabilty so other could do the same.

I think in complicated terms... because it is.

Low-frequency modes are pretty much the same pattern in all violins.  In the "transition hill" there are some common modes, and some differences.  Above that, things are very chaotic, and violins have different mode shapes.  Using touch to detect vibrations might work at low or even some mid-frequencies, but I don't know how you'd detect phase.  At higher frequencies, the displacements are so small I don't think you could feel them.  Even if you could, the mass of your finger would kill off local vibrations and the mode shape would move around to compensate.  And even if you COULD feel out the vibrations of everything, so what?  The Strad3D animations show in pretty good detail how the Titian vibrates, but how can that information be used effectively?

Link to comment
Share on other sites

9 hours ago, Don Noon said:

I think in complicated terms... because it is.

Can’t be too complicated. If I really believe one thing it is that makers in Cremona in the 18th century and before didn’t think in complicated terms. Their results speak for themselves. We just need to get rid of the ‘data approach’.

 

Link to comment
Share on other sites

2 hours ago, Andreas Preuss said:

Can’t be too complicated. If I really believe one thing it is that makers in Cremona in the 18th century and before didn’t think in complicated terms. Their results speak for themselves. We just need to get rid of the ‘data approach’.

 

Engineering technologies are PACKED with simple rules of thumb which, while seemingly unrelated to reality work wonderfully and (would) require a couple of PhD's in Mathematics to be placed on solid theoretical grounds. Once the rule of thumb is lost all we have left is the data approach.

Link to comment
Share on other sites

16 hours ago, jezzupe said:

I think in simple terms I'd like to know if all very good violins have similar "patterns" of activity when examined with the hand during playing in order to create "vibration maps"  and that if bad ones are also similar. all with the intent of trying to see if there is not some way to "feel" if there are areas that would be better thinner or thicker in order to establish some method of repeatability as well as teachabilty so other could do the same 

Two violins can gave identical "patterns of activity" and still sound completely different. 

Link to comment
Share on other sites

1 hour ago, Carl Stross said:

Engineering technologies are PACKED with simple rules of thumb which, while seemingly unrelated to reality work wonderfully and (would) require a couple of PhD's in Mathematics to be placed on solid theoretical grounds. Once the rule of thumb is lost all we have left is the data approach.

Interesting!

could you give an example for one of those engineering ‘rule of thumb’?

Link to comment
Share on other sites

2 hours ago, Carl Stross said:

Two violins can gave identical "patterns of activity" and still sound completely different. 

Maybe? I'm not so sure, I do think that perhaps intensity would be the determining factor? again I wish when they did the 3d imaging that they also had some consensus on some gawd awful violins so as to compare them to the Titan.

Regardless of the hairbrainedness of my notion I do look for simple ways to utilize these things and feel areas of excitation are clues as to where thinning/thickening may be useful, depending on the benchmark.

If I have a violin that sounds very good, and I pay attention to the "braille" and have fore knowledge of the graduations because I made the instrument, on the next violin, I can attempt to alter "that" area in several ways and start to get some idea of the importance of that location, as I Penrose tile may way across the field, hopefully eventually having intimate knowledge of all quadrants and their role in the big picture.

I do not disagree that it is very complicated, yet there it is, and has been for 500 years with the basic/simple concept being the same and in all my reading of anything "sciency" I don't see it helping much, or as Don put's it "and then what" But just like trying to come up with some simple unification theory in physics, some basic simple and elegant formula, I see this very much like that, that in all it's complexity there does lie simplicity, a "cliff notes" to basic success, that I feel is already there, you just have to do it, Basically you have to spend your life "doing it" slowly trying slight different things here and there over all the years with all the instruments one makes paying close attention to everything, including all those that came before who had some level of recognition to the "greats". 

Again assuming that is the tonal scape your shooting for, I'm not sure who the most famous bluegrass fiddle maker ever was but something tells me they are / were not going for the same thing as say Andreas or Don is.

so my approach has always been about developing extreme tactile sensitivity with my hands and trying to sync that to my ears, and regardless of what "science" may be there I'm pretty sure that everyone past and present including myself is really just "guessing" to a certain extent letting trial and error lead the way as the years roll by and the incremental knowledge base builds up, hopefully improving as time goes by.

I feel one of the saddest parts about the violin is what it has become because of the economics, I truly feel that the best way to really learn and preserve this is to work very closely with apprentices who are eager to learn and then eventually do the same thing. It saddens me to think of how many of us work alone turning our minds into lockboxes of information that very well will get lost with the key thrown away when we die. I guess that's why MN is still here , it can act as a data dump for anyone who ever showed any interest.But alas, it's not the same as working with someone side by side year after year.

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...
  • Recently Browsing   0 members

    • No registered users viewing this page.



×
×
  • Create New...