Jump to content
Maestronet Forums

No M2


FiddleMkr

Recommended Posts

The top plate has a strong m5 at F#-20cents, but the M2 node doesn’t form at an octave below or anywhere else that I can find. By ear M2 is about an E. the wt=66.9grams

the back has a M5=G, M2=G#-20cents. 
wt=115grams. Both nodal patterns are well defined for m5 and m2. 

considering the top again, it is 0.1-0.2mm too thick in the middle of the bouts, but it is below the weight of 68grams already.   
     Will reducing the thickness define a nodal pattern for m2? Will it matter? That is are the tops with well defined nodal patterns for m2 make better instruments?

Link to comment
Share on other sites

  • Replies 64
  • Created
  • Last Reply

Top Posters In This Topic

2 hours ago, FiddleMkr said:

Will reducing the thickness define a nodal pattern for m2? Will it matter? That is are the tops with well defined nodal patterns for m2 make better instruments?

Not to my knowledge. When parts of an instrument are assembled into working instruments, they vibrate quite differently anyway.

Link to comment
Share on other sites

6 hours ago, uncle duke said:

When I have trouble finding m2 the usual reason is a wrong max height for the wood I chose to use along with a non conforming inner grad pattern.

What are you using to weigh the plate?

I have a digital scale. 
           edit added: the outline is a Nicolo Amati (grand Amati) which is very close to a Stradivari. The height is 15.6mm and the graduation plan is a Stradivari, 2.8mm except around the edges in upper and lower bouts, 2.5mm. 
       The Amati poster doesn’t have a graduation pattern. 

Link to comment
Share on other sites

12 hours ago, David Burgess said:

Not to my knowledge. When parts of an instrument are assembled into working instruments, they vibrate quite differently anyway.

I understand that when you glue it all together you have a unit with its own parameters, and I don’t doubt your overall answer, that a well defined m2 pattern of the top doesn’t matter as far as the sound of that unit.  
     But, (of course) the natural frequencies of the plates don’t change because the parts are glued together, or connected via a sound post or whatever, (and I don’t think that you meant to imply that they do). 
        It would be interesting to reverse engineer a great sounding golden era violin, take it apart and measure the natural frequencies of each piece. And also measure the natural frequencies of the assembly. Then we could see how the plate frequencies correlate to that of the assembly, and see how all the natural frequencies affect the instrument’s sound. (Of course the bigger the sample size the better.)
          

Edited by FiddleMkr
Added last sentence
Link to comment
Share on other sites

2 hours ago, FiddleMkr said:

... the natural frequencies of the plates don’t change because the parts are glued together, or connected via a sound post or whatever     

If I'm reading this correctly, I'd agree with the opposite.

2 hours ago, FiddleMkr said:

It would be interesting to reverse engineer a great sounding golden era violin, take it apart and measure the natural frequencies of each piece. And also measure the natural frequencies of the assembly. Then we could see how the plate frequencies correlate to that of the assembly, and see how all the natural frequencies affect the instrument’s sound. (Of course the bigger the sample size the better.)      

This has been attempted a number of times with "tonal copies", with questionable results.  My attempt and Sam Z's copies I think lead to the same place:  if you take a specific instrument and carefully copy it, you can get some of the general character of it's tone.  But, with wood being wood with all of its large numbers of variables (including damping), you won't get an identical twin.  And I think age does some things that are likely impossible to copy.

Link to comment
Share on other sites

31 minutes ago, Don Noon said:

If I'm reading this correctly, I'd agree with the opposite.

What I was trying to say is that a part of metal or wood or whatever has a set of natural frequencies, and that attaching another part to it (that has its own frequencies) yields an assembly with a unique set of frequencies, but each part still has its original set of natural frequencies. 

Link to comment
Share on other sites

2 hours ago, FiddleMkr said:

...but each part still has its original set of natural frequencies. 

Nope.  Those frequencies only exist unassembled.  Everything changes in mode shape and frequency when assembled, although some modes change more than others.  More properly, some of the assembly mode shapes might look sorta kinda like a part's mode shape, but it is actually something different.

Link to comment
Share on other sites

3 hours ago, Don Noon said:

This has been attempted a number of times with "tonal copies", with questionable results.  My attempt and Sam Z's copies I think lead to the same place:  if you take a specific instrument and carefully copy it, you can get some of the general character of it's tone.  But, with wood being wood with all of its large numbers of variables (including damping), you won't get an identical twin.  And I think age does some things that are likely impossible to copy.

That makes sense. Come to think of it, I believe Carleen Hutchins did something similar. I haven’t read how extensive her testing was, but it must have been substantial. At least she found that the intervals between the plates determine what type player will prefer the instrument. 

Link to comment
Share on other sites

2 hours ago, Don Noon said:

Nope.  Those frequencies only exist unassembled.  Everything changes in mode shape and frequency when assembled, although some modes change more than others.  More properly, some of the assembly mode shapes might look sorta kinda like a part's mode shape, but it is actually something different.

We still weren’t communicating. You can’t test a part’s natural frequency when it is attached to another part. As you say “everything changes in mode shape and frequency when assembled”. And the natural frequency of individual parts “only exist unassembled”. So, of course the assembly has its own natural frequency, BUT EACH PART STILL HAS ITS OWN NATURAL FREQUENCY (if it was by itself and was tested by itself). 

Link to comment
Share on other sites

21 hours ago, FiddleMkr said:

The top plate has a strong m5 at F#-20cents, but the M2 node doesn’t form at an octave below or anywhere else that I can find. By ear M2 is about an E. the wt=66.9grams

It seems very strange to me that the nodal pattern of your M2 does not form, usually this happens when it is very rigid (i.e. very high frequency), but the more flexible the plate becomes, the more clearly the nodal pattern forms. If your M2 is around E, that means it's more than one octave below your M5, so more flexible than if it were just one octave. Are you sure you placed the plate supports and excitation point correctly?

PS It's been many years since I did any glitter test, but from what I remember, the correct positioning of the plate is essential, and the power of the excitation is too, whether it is too much or too little, you have to find the right one.

Link to comment
Share on other sites

4 hours ago, Don Noon said:

This has been attempted a number of times with "tonal copies", with questionable results.  My attempt and Sam Z's copies I think lead to the same place:  if you take a specific instrument and carefully copy it, you can get some of the general character of it's tone.  But, with wood being wood with all of its large numbers of variables (including damping), you won't get an identical twin.  And I think age does some things that are likely impossible to copy.

I would think that it would be hard to copy all of the characteristics in one instrument, and it would take many instruments to get the data needed to know which parameters are important.  And it could be that some of the important things can’t be measured. As a violin making friend says, “if you can’t measure what’s important, what you can measure becomes important “.

Link to comment
Share on other sites

6 minutes ago, FiddleMkr said:

I would think that it would be hard to copy all of the characteristics in one instrument, and it would take many instruments to get the data needed to know which parameters are important.  And it could be that some of the important things can’t be measured. As a violin making friend says, “if you can’t measure what’s important, what you can measure becomes important “.

Martin Schleske (https://www.schleske.de/en/research/publications.html) has written several articles about copying instruments and two of them are attached.

Tonal copies.pdf similar violins .pdf

Link to comment
Share on other sites

13 minutes ago, Davide Sora said:

It seems very strange to me that the nodal pattern of your M2 does not form, usually this happens when it is very rigid (i.e. very high frequency), but the more flexible the plate becomes, the more clearly the nodal pattern forms. If your M2 is around E, that means it's more than one octave below your M5, so more flexible than if it were just one octave. Are you sure you placed the plate supports and excitation point correctly?

PS It's been many years since I did any glitter test, but from what I remember, the correct positioning of the plate is essential, and the power of the excitation is too, whether it is too much or too little, you have to find the right one.

The m2 of the BACK is (in fact) two octaves below its m5 (not just one).  
       The TOP has a well defined M5 pattern and a high excitation of tea leaves. But as I said the nodal pattern for M2 doesn’t form. The thickness is a little over spec; 3mm where it calls for 2.8mm, but the weight is at the midpoint (66.9g, midpoint is 68g). Do you think I might could get an M2 pattern by reducing the thickness (by 0.2mm) on either side of where the M2 node should be? Or should I try reducing the thickness at the nodes, since you suspect that the plate is stiff? I have a +-10gram tolerance on the weight, if that matters. 

Link to comment
Share on other sites

24 minutes ago, FiddleMkr said:

The m2 of the BACK is (in fact) two octaves below its m5 (not just one).  
       The TOP has a well defined M5 pattern and a high excitation of tea leaves. But as I said the nodal pattern for M2 doesn’t form. The thickness is a little over spec; 3mm where it calls for 2.8mm, but the weight is at the midpoint (66.9g, midpoint is 68g). Do you think I might could get an M2 pattern by reducing the thickness (by 0.2mm) on either side of where the M2 node should be? Or should I try reducing the thickness at the nodes, since you suspect that the plate is stiff? I have a +-10gram tolerance on the weight, if that matters. 

I'm guessing your support for the top plate might not be right on the M2 node lines which will dampen the pattern shape.  How is the plate supported?

Link to comment
Share on other sites

39 minutes ago, Marty Kasprzyk said:

I'm guessing your support for the top plate might not be right on the M2 node lines which will dampen the pattern shape.  How is the plate supported?

It is supported on 4 foam blocks. I position the blocks where the m2 nodal lines are, at least I position them close as possible. If I am only close to the lines would it dampen the vibration enough to hide the nodal lines? (Pardon the dumb question. I am asking again because there’s no pattern at all. I would think that there would be some pattern forming even if there was dampening.)

Edited by FiddleMkr
Added last sentences
Link to comment
Share on other sites

5 hours ago, FiddleMkr said:

We still weren’t communicating. You can’t test a part’s natural frequency when it is attached to another part. As you say “everything changes in mode shape and frequency when assembled”. And the natural frequency of individual parts “only exist unassembled”. So, of course the assembly has its own natural frequency, BUT EACH PART STILL HAS ITS OWN NATURAL FREQUENCY (if it was by itself and was tested by itself).

Hi - Natural frequencies are free vibrations, i.e. you've excited the plate, say with an impulse, but there's no oscillator attached to the plate that forces it to vibrate - it vibrates at its 'natural' frequencies.  If you change the boundary conditions of the plate you change it's natural frequencies.

Natural frequencies of a plate do exist when the plate is in an assembled violin.  The plate's natural frequencies are different when in the assembled violin compared to those of the free-standing plate because of the boundary conditions.

Free-plate natural frequencies don't exist when the plate is assembled into a violin.  They only exist when the plate is free-standing.  Natural frequencies of a plate in an assembled violin don't exist when the plate is free-standing.

No-one to my knowledge (which is decidedly limited so I'm open to correction), has been able to convincingly relate free-plate natural frequencies to the natural frequencies of the plate in an assembled violin - although intuitively we'd think there should be some relation ceteris paribus.  Variation in boundary conditions may be the primary culprit.  Then there's the problem of relating the natural frequencies of the plate or the assembled violin to its tonal and playing characteristics.

 

Link to comment
Share on other sites

7 hours ago, Don Noon said:

Nope.  Those frequencies only exist unassembled.  Everything changes in mode shape and frequency when assembled, although some modes change more than others.  More properly, some of the assembly mode shapes might look sorta kinda like a part's mode shape, but it is actually something different.

What Don said.  'Modes', vibration  patterns exist in that form, pitch and shape because of the boundary conditions. If you hold a plate very tighly somewhere on the midline you won't get the M2 you're familiar with or possibly not M5. You will instead get a whole new series of modes which have a stationary point going through the place you're holding only some of which may be from the usual free plate series. 

The shape, pitch, amplitude of the free plate modes might tell us something about the material properties but those familiar free plate mode patterns only exist for that way of holding and exciting the plates whilst the material properties don't change. They are standing waves in 2D. 

 

Which begs a question - is there somewhere better we could be holding and tapping to find out something useful?

 

Edited by LCF
Clarify
Link to comment
Share on other sites

6 hours ago, FiddleMkr said:

The m2 of the BACK is (in fact) two octaves below its m5 (not just one).  
       The TOP has a well defined M5 pattern and a high excitation of tea leaves. But as I said the nodal pattern for M2 doesn’t form. The thickness is a little over spec; 3mm where it calls for 2.8mm, but the weight is at the midpoint (66.9g, midpoint is 68g). Do you think I might could get an M2 pattern by reducing the thickness (by 0.2mm) on either side of where the M2 node should be? Or should I try reducing the thickness at the nodes, since you suspect that the plate is stiff? I have a +-10gram tolerance on the weight, if that matters. 

It is very important that the supports for the modes are placed at the nodal lines of the modes. Mode 2 is not a very efficient mode to drive with a loudspeaker, but its heard next to the ear. Again, it will ring most easily if the holding point is on a nodeline, and the tapping at its most active area.

For every Hz change in mode 2 in the free plate, the average change in the B1 modes is in the range of 0,5 Hz, or so. The top plate shape of B1- is similar to the M2 free top, except for more activity near the center in the assembled fiddle and near the ends for the free plate. For the back it is related for the mode B1+, although the free back has a somewhat different shape of M2 due to the stiffer centre.

The shape of M2 is predoiminantly related to the crossgrain stiffness and thus bending wave speed. If there is some pecularities with that, lets say the wood grainlines are at an angle on one side and square at the other, it are likely to show up in the M2 patters. Also wood with different stiffness and or density may show a similar effect. Usually the maturrity ofg the bar may influence too. Some of this are speculations. I do not monitor M2 shape as a rule for the free plates. 

Edited by Anders Buen
More info
Link to comment
Share on other sites

1 minute ago, Anders Buen said:

It is very important that the supports for the modes are placed at the nodal lines of the modes. Mode 2 is not a very efficient mode to dsrive with a løoudspeaker, but its heard next to the ear, again it will ring most easily if the holding point is on a nodeline, and the tapping at its most active area.

I have a friend who carefully tracks down the stationary points and pitches  for the signature free plate modes by holding and tapping and then makes the plate ring by singing the note loudly into the hot spots. The 20W speaker and oscillator is wasted on him. 

Link to comment
Share on other sites

1 minute ago, LCF said:

I have a friend who carefully tracks down the stationary points and pitches  for the signature free plate modes by holding and tapping and then makes the plate ring by singing the note loudly into the hot spots. The 20W speaker and oscillator is wasted on him. 

( but not mode 1!)

Link to comment
Share on other sites

1 hour ago, LCF said:

I have a friend who carefully tracks down the stationary points and pitches  for the signature free plate modes by holding and tapping and then makes the plate ring by singing the note loudly into the hot spots. The 20W speaker and oscillator is wasted on him. 

Impressing, specially as the note fundamental is 140-160 Hz or so below the lowest note on a violin. Some are more bassy than others! :-)

Link to comment
Share on other sites

`The simplest and most reliable way to hold a plate to run over a speaker is with 2 small pieces of 0000 steel wool placed as such,,,,,,,,,,

 

IMG_9224.thumb.jpg.423f6ed67084fac67a9e6512272af8f4.jpg  

So for mode 2,, which is one of the easiest to find, place the pads as such, lay the plate on them with the lower block directly over the speaker and scan,,,

IMG_9225.thumb.jpg.d49f86483464c28161518311a774e290.jpgIMG_9226.thumb.jpg.ec3b1d0a3d4d3c4bcea81da720b8196e.jpg

 

To find mode 5, place the pads as such,,,put the plate over the speaker, and,,,

IMG_9227.thumb.jpg.e360290a6a97152bc018b2c063633d7b.jpg

IMG_9228.thumb.jpg.2a5e8f0996d37d61ee29fa52423bc500.jpg

 

All of these were done with 2, 0000 steel wool pads, only the speaker position changes. The only time the pads are changed is M 6 and upwards, when the highest amplitude is in the central part of the lower or upper bout. . There is usually a node in the vicinity of the upper eyes somewhere, move one of the pads there.

IMG_2463a.thumb.JPG.a59407d48dd1f691d8294de552a06787.JPGIMG_2463.thumb.JPG.eae57ac6ffff005c359f9c27e0025e91.JPGIMG_2462a.thumb.JPG.d5d1cc7aff497025abc2016c6be59a19.JPGIMG_2462.thumb.JPG.04d05be49310010f947298e118484bfe.JPGIMG_2461a.thumb.JPG.66e1d87fd8c47bcfad77ce80ff884150.JPGIMG_2461.thumb.JPG.9814fddcf9ed20e4ce1cf89ca1249140.JPGIMG_2460a.thumb.JPG.c35399f364d85d0a553f9d82f7323a84.JPGIMG_2460.thumb.JPG.6a266d69f4e9856b1f6aa09e103adaca.JPGIMG_2457a.thumb.JPG.e8ba5de6befba847ff4fa7b883951010.JPGIMG_2457.thumb.JPG.b2896688bdcdd592a3cd0e159fad661e.JPGIMG_2241a.thumb.jpg.1092b68a43f9b1af2727e0b537839a71.jpgIMG_2242ajpg.thumb.jpg.4d1ddde557bf12f719d9b86ec746415e.jpg

It looks like that it could possibly work. It is by far the easiest and fastest with the cleanest lines, and I can't find one hertz difference, from any other method.

I think that too often people have the speaker opening way too large, and it cancels itself out. Also the thickness mounting of the hole is minimal, I have it beveled away to nothing, with a sharp edge at the surface. Too thick with straight sides creates phase cancellation and distortion, and hard to find clean lines.

Does it help, does it make a difference, to be able to see,, that discussion is in another world at the moment.

 

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...