Jump to content
Maestronet Forums

Radical test - cutting slots in the back (with some surprising results)


Recommended Posts

So now the test is finished. My predictions were simply wrong. (I didn't display the pictures because it looks too much like a carnage)7939B0D4-3BA0-4DCD-81AF-2CDC72698BB2.thumb.jpeg.42494fddbadb6631eb32c0ddb3436f23.jpeg

First I cut a slot across the upper bout (see black line in diagram). First step was a slot approximately 3cm wide and then I increased the length on both sides. I expected some loss of sound volume, BUT...

... it didn't happen. Enough to think about. At the first stage (1) the sound became somehow rougher, but going further just evened it out again. If anything happened, the violin was more difficult to play. For the overall sound changes were just very subtle. In any case I don't think it is worth to compare graphs. In the process the neck projection dropped maybe 2mm. To execute the second test the upper bout slot was patched up rapidly with a small board.

Second test brought a kind of surprise. Because thinning down the c bouts affects the overtone range (how crispy the sound is under the bow) I expected a kind of muting effect when cutting through a slot along the C bouts (red line on diagram) I started at the narrowest points of the outline first bass side then treble side. Cutting the first 4 centimeters or so didn't change much, only when I extended the slot downwards in direction of the lower bout a change could be noticed. But not what I expected. Before it had a slightly boxy sound, deep and round, somehow a bit too bass heavy. 

With the slots open the sound became clearer loosing the bass heaviness and at the same time got some interesting texture. In a certain way just the opposite of what I thought would happen did happen. So in total a kind of very interesting improvement. 

So here are the graphs of the violin played in half note scales 5 times (before and after) I marked the areas where I see the biggest changes. The region between 900Hz and 1200Hz became weaker and the region between 1500Hz and 2800Hz stronger. The valley around 580Hz became flatter and broader, and the air resonance peak a little stronger. 

BEFORE1021796281_NCV120beforecuttingslotsinback.thumb.png.a3b2753d38c5ed1bd14e4ced2de6fddd.png

AFTER698506883_NCV120aftercutting7cmslotsatCboutsonback.thumb.png.31c2f5c446190f4be61a88f4e0bba7e3.png

 

Link to comment
Share on other sites

2 hours ago, Andreas Preuss said:

With the slots open the sound became clearer loosing the bass heaviness and at the same time got some interesting texture. In a certain way just the opposite of what I thought would happen did happen. So in total a kind of very interesting improvement.

Interesting experiment, somehow it can confirm my impression that thinning the back in the fluting area of the Cs especially the part facing the cornerblocks opens up the sound by moving it towards the bright register, also increasing the responsiveness to the bow. Of course it is difficult to tell when it gets too thin and the situation begins to deteriorate, but anyway this experiment can give some clues. Perhaps a series of holes instead of a continuous slot could be another interesting experiment to try and compare, maybe starting at the area facing the cornerblocks.

Link to comment
Share on other sites

2 hours ago, Davide Sora said:

Interesting experiment

Davide, thanks for the feedback from your experience.

I think there is a difference between ‘thinning down’ and ‘loosening’. Basically classic Cremonese makers left the area rather thick up the ribs and IMO the channels at the c bouts have to be treated with cautiousness. 
 

There is a good chance that the center cross arching has for the same reason a big influence too. The current arching on the ecoerimental back was rather stretched from one side to the other. (Not a curtate cycloid type of arching) 

i will know more when the new back is on the fiddle.

Link to comment
Share on other sites

29 minutes ago, Andreas Preuss said:

IMO the channels at the c bouts have to be treated with cautiousness. 

I agree, that's why I said that it is difficult to tell when it gets too thin and the situation begins to deteriorate;)

As usual there many other things to consider.

Link to comment
Share on other sites

While I do appreciate qualitative results of experiments, from an engineering standpoint some more physical information is needed.  What happens to the mode frequencies and amplitudes, and why?  Graphs of the response might help (if MN ever gets back to the state where such things can be displayed).

I can't think of any way that cuts in a structure would cause mode frequencies to go higher; they should go down.  However, mode shapes could change and become very different, especially above the signature mode frequencies, to where the modes and frequencies are unrecognizable from what they were.

Thinning certain areas is a different thing from making cuts.  If the thinned area happens to be an antinode, the frequency could go up as well as the amplitude.

Link to comment
Share on other sites

8 hours ago, avandesande said:

I wonder if f-holes on the back would do anything.

We tested the sound with the slots taped and untaped. It doesn’t make a hell of a difference. Somehow the sound gets softer maybe weaker. 
 

i think it is best if the inside air pressure is directed to the outside from the top. Somehow I think the trick is to create a pumping motion with a big amplitude but at the same time the pumping motion must have punch. Any additional openings weaken the punch from the f holes.

Link to comment
Share on other sites

10 minutes ago, Andreas Preuss said:

We tested the sound with the slots taped and untaped. It doesn’t make a hell of a difference. Somehow the sound gets softer maybe weaker. 
>

Masking tape doesn't stop sound transmission very well so it is not surprising that the slots taped and untaped didn't sound much different.

 

 

Link to comment
Share on other sites

1 hour ago, Marty Kasprzyk said:

Masking tape doesn't stop sound transmission very well so it is not surprising that the slots taped and untaped didn't sound much different.

Well, we could notice a slight difference which was good enough to estimate what effect it has.

 

Link to comment
Share on other sites

I find myself highly curious what would happen if you cut slots in the lower bouts similar to how you did the C bouts.  I was always taught to make the edge fluting extra deep in the C bouts AND the lower bouts on the back.  The best sounding fiddle I've encountered has a comically deep fluting in the lower bouts.  If I acquire a test dummy sooner than anyone here, I'll let you know how it goes.  

Link to comment
Share on other sites

On 7/8/2021 at 12:27 PM, chiaroscuro_violins said:

I find myself highly curious what would happen if you cut slots in the lower bouts similar to how you did the C bouts.  

I tried to ‘tune’ instruments by making the final thickness of thee we back from the outside.I didn’t notice anything by deepening  the scoop on the lower c bout.
 

In general the sound changes much less than you we would expect. At least if tap time matching would be really crucial this should bring at some point the magical result. From my experience it doesn’t. I could observe sound color shifts by reducing thickness in rather large areas. 
 

on the other hand I think it is risky to make conclusion from abnormal  construction features of one instrument. We never know if the key point for the sound was not something else. 

Link to comment
Share on other sites

On 7/7/2021 at 7:43 PM, Andreas Preuss said:

Any thoughts  welcome.

Cuts in the back would most logically have the most effect on modes where the back plate is most active... i.e. signature modes and the "transition hill" centered around 1 kHz, so it it not surprising to see changes there.  Speculation about exactly what would happen would be mostly just that, unless you do modal analysis.  Even then, with mode shapes and frequencies, determining the AMPLITUDES is not straightforward, and involves coupling to the bridge and radiation efficiency.

The significant impact on the higher frequencies is more unexpected and worth some speculation.  It might be due to the C cuts allowing increased movement of the soundpost at higher frequencies, which in turn means lower impedance (i.e. more energy is transmitted from the string into the body). 

Even before these test results, I have been thinking somewhat along these lines, and avoiding the "Strad harp" band of thick grads on the back all across the C bout, and thinning along the rib line there (but still leaving the center very thick).

47 minutes ago, Andreas Preuss said:

I think it is risky to make conclusion from abnormal  construction features of one instrument. 

That's for sure.  Modifications that correct a defect on one fiddle can create a defect on a fiddle that started out good.

Link to comment
Share on other sites

59 minutes ago, Andreas Preuss said:

on the other hand I think it is risky to make conclusion from abnormal  construction features of one instrument.

Of course.  This is only a hypothesis based on that example and also what I've been taught by an experienced violin maker.  Not drawing any conclusions here.  

Link to comment
Share on other sites

I would think the primary effect of your cBout slots would be to unlink the center back mass' behavior from the sides.

Depending on how/where the slots are cut the could also reduce the integrity and strength of the sides structure through the cBout area.

But I think you mostly avoided that.

Link to comment
Share on other sites

7 hours ago, Don Noon said:

The significant impact on the higher frequencies is more unexpected and worth some speculation. 

Indeed.

To me any frequency as a played note is inked to what moves on a violin and can be partly visualized with the signature modes and/ or holographic pictures.
 

Overtones (or high frequencies) is HOW everything moves (comparable to sine wave and sawtooth wave). So in this sense it would not be visible with signature modes, the reason why signature modes can be matched to old Italian instruments without having the same sound effect. 
 

For HOW it moves I think the lengthwise stiffness from top nut to end block plays an important role and in a schematic model we have the spring force of the strings acting against the spring force of the entire body. When the string pulls lengthwise the body has to stretch, this stretch somehow stores the energy to pull itself back again creating the vibration. So, the more energy can be stored the quicker it will go back. (?) Accordingly light weight can be risky in this scheme. In this setup  it is only logic that the string angle is of importance.

So what does this have to do with the slots on the back?

My interpretation is that the crosswise link to the ribs hinders how the whole violin can move in the longitudinal direction. In fact it does not take much away from the entire lengthwise stiffness. This goes along with my tests that thick back linings are not necessary especially on the c bouts. 
 
Maybe this can be measured by tapping on the head or on the endgrain of the back and analyzing that spectrum? 

More surprising in this sense is that the crosswise cut didn’t show a significant effect. I can only assume that it is not the most sensitive spot for the entire lengthwise stiffness. If I had cut it closer to the upper block it should have a had a bigger effect.

Link to comment
Share on other sites

One thing... Since the B1+ mode is characterized by longitudinal flexing of the top and crossgrain flexing of the back, I would have thought that the C-bout cuts would reduce the B1+ frequency significantly.  That does not appear to be the case.  So either the top stiffness dominates in determining B1+ mode stiffness, and/or the connection of the back to the cornerblocks (uncut) keeps the back crossgrain flexing stiffness from being reduced.

One other thing:  Andreas, it looks like you have in mind that longitudinal string tension variations are a big thing in creating sound.  I'm quite sure that's not the case, and lateral vibrations at the bridge are what we need to think about.

Link to comment
Share on other sites

Posted (edited)
3 hours ago, Don Noon said:

One other thing:  Andreas, it looks like you have in mind that longitudinal string tension variations are a big thing in creating sound.  I'm quite sure that's not the case, and lateral vibrations at the bridge are what we need to think about.

Not creating sound. Definitely not any more. ;)

We have all the lateral movement which for sure drives the vibrations. But I do think that the longitudinal overall stiffness governs somehow the ‘punch’ of the lateral vibrations. Otherwise I couldn’t explain that changing a soft material tailgut with a rigid material tailgut definitely does enhance overtones. Very simplified, I think overtones equal the punch of motions.

Another thing is that increasing the string angle should increase the sound volume. It does increase it sometimes in the wrong way (especially on thinly built instruments) with a sort of over resonance of the lateral vibrations and reducing the ‘punch’ at the same time. Repeatedly I could fix this with lowering the string angle. (Somehow the more upward pulling force there is on the upper nut the less stable this becomes.)

On the entire structure the weakest point for this is where the neck leverage acts on the body. In other terms if the area around the top block gets too flimsy this will take away the ‘punch’. This is IMO confirmed by restoration measures which reinforce this area with patches either on the top or on the back or just by stabilizing the neck with the double lining method on the upper bout. The Result is in general a clearer sound with more definition. Depending on the method and how bad it it was before this can have a dramatic effect.

This could also explain why a higher top arching adding stiffness in the area around the top block can increase overtones. Your thought on that is that it increases the antinodal vibration areas. I would rather think narrower nodal lines have something to do with vibrations having more punch.

Therefore I was thinking of another constructional experiment by making a super wide upper block. EE2AEB92-319B-4CCE-A39A-3509567F63FF.thumb.jpeg.62b6b7e45780ea4bdf620641804d5de6.jpeg

At least I can say that I arrived with some constructional alterations on the new concept violin at a sound which is not absurdly different from normal violins despite its light weight. But still a lot of work ahead.

Edited by Andreas Preuss
Add picture
Link to comment
Share on other sites

46 minutes ago, Andreas Preuss said:

I was thinking of another constructional experiment by making a super wide upper block. 

At some point, things will stop giving returns.  If a normal block and "Sacconi Plateaus" make the area rigid enough to prevent inefficient flexing, then more won't do anything.  But if you don't test it, you won't know.  Go ahead if you have the time to spend on experiments, and let us know.

Link to comment
Share on other sites

8 hours ago, Don Noon said:

If a normal block and "Sacconi Plateaus" make the area rigid enough to prevent inefficient flexing, then more won't do anything

So far the new concept violin works better with stronger linings on the top side and reduced rib height, a new type of bass bar and a guitar type lightweight neck. I wouldn’t be surprised if a stronger and wider top block merging out into the linings changes things again.

However, I don’t say that this works as well on classical built instruments. Maybe, maybe not.

In any case the last hurdle I have to take on the new concept violin is the overtone range.

Always with thanks for your critical thoughts and comments.

Link to comment
Share on other sites

On 7/9/2021 at 6:56 PM, Don Noon said:

>

One other thing:  Andreas, it looks like you have in mind that longitudinal string tension variations are a big thing in creating sound.  I'm quite sure that's not the case, and lateral vibrations at the bridge are what we need to think about.

The longitudinal string tension variations have frequencies twice that of the strings bowing frequencies.

Starting at a zero starting position with zero longitudinal variable tension the string is stuck to the bow hair and is dragged to one side (left side for example) which increases the longitudinal string tension. The string eventually slips off of the bow hair and flies back in the opposite direction to the right to its starting position with zero additional tension.

The string continues to move in the right hand direction which again increases its longitudinal tension for the second time until it stops and begins to move left back to its starting position with zero additional tension again. This completes one total cycle of the string.

So during this one string vibration cycle the tension has hit two two peaks and two zeros hence the longitudinal tension has twice the frequency of the bowed string.

Thus the longitudinal tension vibrations always contribute to the violin's high frequency range.  But a question is how much?  It can be shown (1) that the longitudinal tension variation is dependent upon how close to the bridge the bowing occurs--playing close to the bridge increases it which explains why playing close to the bridge increases the loudness of the note's higher harmonics thereby making the note sound brighter.

 

1.  N. Fletcher, T. Rossing,  "The Physics of Musical Instruments" 2nd ed.  Springer, 1997,  Chapter 9.3 Force Exerted by the String

Link to comment
Share on other sites

On 7/13/2021 at 9:48 AM, Marty Kasprzyk said:

What do you mean by the term "overtone range"?

 

That’s a very unscientific broad description of what a player can hear in the quality of sound. 
 

Bad sound somehow equals a weak overtone range. 
 

in a spectrum I am mostly interested in the frequencies that can’t be played on a violin so above 3000hz (eventually above 1600hz) My unscientific impression just from looking repeatedly at spectra is that good violins have an beautiful arc pattern spanning from 1600 hz to 4-5000hz, with many peaks and not too deep valleys (around 2000 resolution on an audacity graph), the peak of this arc seems to be good around 2500hz  and all should be quite high compared to low frequencies. 
 

If there anything I really believe in it is how the violin sound ‘feels’ when being played. Accordingly I see in sound spectra only an abstract part of the reality, best to find problems but not really giving the ultimate answer to what is a really good sound. 

Link to comment
Share on other sites

16 hours ago, Marty Kasprzyk said:

Thus the longitudinal tension vibrations always contribute to the violin's high frequency range.  But a question is how much?

Marty, thanks for providing the scientific explanation. (Or a part of it) 

Just by my gut feeling as a violin maker, there is much more than we probably know about this. 
 

In any case, all alterations which are made in the chain of ‘strings-tailpiec-tail gut’ have a clear influence on overtones or the clarity of sound. (On cellos this is sometimes sacrificed to kill a wolf note.)

Likewise I think that alterations in the lengthwise direction of the violin structure play a role. (I am really curious now what result brings a much larger and structure stiffening top block) 

Secondly it seems to me that the ‘stretch effect’ stiffens the top and back to the maximum in this direction with the result of bringing out a stronger overtone range. 

————————-

In this context I am just guessing that the string angle plays a role too.

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...