Sign in to follow this  
GerardM

Tap Tone or Plate Tuning Equipment

Recommended Posts

Are you measuring sound pressure on complete violin? Velocity or acceleration on a single position on a free top?..

I don't measure damping (yet).

Probably fitting modes on a complete structure such as a violin gives you a better representation of the phenomena, and allows you to estimate damping.

At least in correspondance of mode.

I think it is easier measuring damping in wood than in wood structures.

 

Share this post


Link to post
Share on other sites
On 9/19/2018 at 5:04 AM, Danube Fiddler said:

 

Would any of both software allow to do peak analysis - curve fitting ( for damping calculations ) ?

 

I don't know. The fancier Spectrplus version has many features which I have never explored.

Share this post


Link to post
Share on other sites
5 hours ago, francesco piasentini said:

Are you measuring sound pressure on complete violin? Velocity or acceleration on a single position on a free top?..

I don't measure damping (yet).

Probably fitting modes on a complete structure such as a violin gives you a better representation of the phenomena, and allows you to estimate damping.

At least in correspondance of mode.

I think it is easier measuring damping in wood than in wood structures.

 

I would like to measure ( total ) damping in a complete violin. My assumption is, that if there comes out something at all, there would be a wild field of quite different damping values (one for each peak, highly depending on the main nature/shape ( e.g. proportion of cross-grain or longitudinal bending/ cutting fibres in several directions of arching and numberless other factors ) of the related mode.).

To measure one simple mode in a wood blank ( tonewood wedge ) can be easier. The 2-0 ( lenght)-mode of a wedge could be a first indication, it normally has good shapes. However I am afraid, that the cross-grain damping is more important, because it is higher. Sadly it is more difficult to measure. I don´t have any idea, if "the" damping of an easy-wedge-mode is a good indication for the final damping-values of the assembled violin at all, which could be highly determined by a big number of several factors, lying in the archings/ shapes / grain-courses / varnish / combination of wood-properties of all oscillating parts/ graduations ....

My idea is to measure many peaks of a responsive - curve ( spectrum of a bridge-impact ) and eventually get a valuable info from this. 

Share this post


Link to post
Share on other sites
On 9/18/2018 at 12:54 PM, David Burgess said:

Prior to dropping some big bucks on Spectra Plus software, and learning  how to use it, I'd say that most of my earlier efforts didn't amount to much more than a waste of time.

Yes, i think that most people over time find it best to just listen to notes rather than look at them.

Share this post


Link to post
Share on other sites
17 minutes ago, Danube Fiddler said:

My idea is to measure many peaks of a responsive - curve ( spectrum of a bridge-impact ) and eventually get a valuable info from this. 

While I suspect you might get reasonable damping readings from the higher, isolated peaks, I think you run into trouble trying to get damping data from a broadband response spectrum.  The math is primarily for a single, simple oscillator.

I attempted to get damping of the complete violin by looking at the rise/decay of a sine burst input to the bridge from a voice coil driver.  Some frequencies had low-ish damping, others had high damping.  As there was no apparent pattern, I'd say the results were far from valuable.  Bissinger did some damping measurements of a complete violin.

Share this post


Link to post
Share on other sites
5 hours ago, Don Noon said:

While I suspect you might get reasonable damping readings from the higher, isolated peaks, I think you run into trouble trying to get damping data from a broadband response spectrum.  The math is primarily for a single, simple oscillator.

I attempted to get damping of the complete violin by looking at the rise/decay of a sine burst input to the bridge from a voice coil driver.  Some frequencies had low-ish damping, others had high damping.  As there was no apparent pattern, I'd say the results were far from valuable.  Bissinger did some damping measurements of a complete violin.

May be it is technically wrong - please tell me, if so - I would hope to have a good actual reading, when the peak is quite fitting to a Lorenz-curve a little bit more than 3 dB or a half-height in soundpressure - I would be satisfied even then, when only one side of the peak would fit quite well. Sadly the most peaks will not fit enough - and here I will try this : several micro - positions to eliminate detrimental directional radiation-effects - just an idea.

Your procedure sounds quite interesting. Did you compare an obviously bad violin to a quite fine sounding one in your damping readings ?

Share this post


Link to post
Share on other sites
5 hours ago, jezzupe said:

Yes, i think that most people over time find it best to just listen to notes rather than look at them.

It is always a very good idea to just listen. Especially because at the moment the technical approach to separate fine violin-sound from mediocre sounds seems to be not very good developped. However I think,  that most research still is based on time-averaged spectra. This view is to easy and seems not to give good results. Needed could be a more time-depending approach - damping would be one. Musicians sometimes speak of lingering sound or fastness of response = time-depending effects. You don´t have to see it, you can hear it while playing, but it is not easy to separate from bowing-technique.

Share this post


Link to post
Share on other sites
1 hour ago, Danube Fiddler said:

May be it is technically wrong - please tell me, if so - I would hope to have a good actual reading, when the peak is quite fitting to a Lorenz-curve a little bit more than 3 dB or a half-height in soundpressure - I would be satisfied even then, when only one side of the peak would fit quite well. Sadly the most peaks will not fit enough - and here I will try this : several micro - positions to eliminate detrimental directional radiation-effects - just an idea.

Your procedure sounds quite interesting. Did you compare an obviously bad violin to a quite fine sounding one in your damping readings ?

I'm no sure about what a kind of damping measurement is important.   The violin's total damping includes the damping of the holding device (player or fixture), the violin's internal wood damping which generates heat, and damping which comes from moving air around some of which produces useful sound while some just produces hot air often seen on MN.

Share this post


Link to post
Share on other sites
10 hours ago, Danube Fiddler said:

 I don´t have any idea, if "the" damping of an easy-wedge-mode is a good indication for the final damping-values of the assembled violin at all, which could be highly determined by a big number of several factors, lying in the archings/ shapes / grain-courses / varnish / combination of wood-properties of all oscillating parts/ graduations ....

My idea is to measure many peaks of a responsive - curve ( spectrum of a bridge-impact ) and eventually get a valuable info from this. 

Good thoughts, but nothing new. Really talented people have been chasing this around for decades.

2 hours ago, Marty Kasprzyk said:

I'm no sure about what a kind of damping measurement is important.   The violin's total damping includes the damping of the holding device (player or fixture), the violin's internal wood damping which generates heat, and damping which comes from moving air around some of which produces useful sound while some just produces hot air often seen on MN.

Yup.

Share this post


Link to post
Share on other sites
20 hours ago, GerardM said:

It’s a fascinating subject this violin acoustics, never gave it a thought till reading the article. I have quarter sawn pine in stock. Will make a violin belly with some of it, then do the tea leaf test and carve away to watch the effects. I have also downloaded an app that tells the pitch of a tapped sound. So will be playing around with that also. Is the idea to have  the same pitch in all areas of the belly?  I realise that the pine I have ( it’s not spruce , it’s yellow pine cannot recall the Latin name ) may not be ideal acoustically but I will certainly learn a lot from testing. Have been tapping pieces of it and seems to make a reasonable noise, having said that not shure what I should be hearing. It’s good to learn new skills , so will more than likely go on to construct a violin. Meany thanks again Gerard 

Watch Davide Sora's Videos, you cannot get a book better than his excellent Videos! his last 2 weeks of vidios show a lot about the top and back.

Davide Sora Video's
https://www.youtube.com/channel/UCggqGrdYUEovaC9UKnSMD3g/videos

 

Share this post


Link to post
Share on other sites
5 hours ago, David Burgess said:

Good thoughts, but nothing new. Really talented people have been chasing this around for decades.

I didn´t have the impression, that it ever was a really big topic, explored by many researchers. One or two reports( may be of Bissinger ) I had read. Perhaps I should read again. However why not to do own measurements, the focus could be a little bit different and anyways one needs an own approach for any practical purposes. The same way as you probably make your own ( time-averaged ) spectral analysis, in spite of this was extremely often done by many researchers.

Share this post


Link to post
Share on other sites
8 hours ago, Marty Kasprzyk said:

I'm no sure about what a kind of damping measurement is important.   The violin's total damping includes the damping of the holding device (player or fixture), the violin's internal wood damping which generates heat, and damping which comes from moving air around some of which produces useful sound while some just produces hot air often seen on MN.

The question can also be, which kind of measurements are able at all ( with common equipement ). For me that is only total damping. Radiation damping and internal/external damping are parts of it. Somewhere I read, that internal damping would not be an important part of the total damping. I think, that total damping should be the critical factor for the player. He must compensate by bowing efforts, when a violin has a tendency to finish notes quite fast and abrupt. A further point would be the string-exitation by bow/colophon : my assumption is, that this is not a continous process but rather more a process of many little errors/accidents ( depending on the local eveness of hair surface including the regularity of barbs ). If the player has " enough bow" , this will not play a big role. But if he is forced to bow very slowly ( long note ) and quiet, these little errors become much more important - than he needs something, which carries him over the time-span of error - this would be a sufficient low damping ( total damping ). That means, during the error - time span the string must not loose so much energy of its regular oscillation, that the change gets audible - otherwise the sound becomes unsteady or fragile.

Share this post


Link to post
Share on other sites
5 hours ago, carl1961 said:

Watch Davide Sora's Videos, you cannot get a book better than his excellent Videos! his last 2 weeks of vidios show a lot about the top and back.

Davide Sora Video's
https://www.youtube.com/channel/UCggqGrdYUEovaC9UKnSMD3g/videos

 

Really great video - showing Davides high - developped way of mode-exploring the plates. 

One could add, that the reported points are not always at the same places in all plates. The location can change considerably because of wood-properties/ archings/ bass bar - even by graduation differences. So if one wants to do it exactly, the points are always needed to get re-found ( not so much the tapping points, here little deviations make not a big difference - but the nodal-lines/ fixing - points ).

Share this post


Link to post
Share on other sites
On 9/21/2018 at 1:37 PM, Danube Fiddler said:

Really great video - showing Davides high - developped way of mode-exploring the plates. 

One could add, that the reported points are not always at the same places in all plates. The location can change considerably because of wood-properties/ archings/ bass bar - even by graduation differences. So if one wants to do it exactly, the points are always needed to get re-found ( not so much the tapping points, here little deviations make not a big difference - but the nodal-lines/ fixing - points ).

Of course, although I would not describe that change as considerable, more or less the grip points are always close to the indicated area, but as I say in a caption of the video : "......If necessary, slightly move the fingers that grip the plate until you find the best point to hear a clear and resonant note".

 

Anyway, the grip points are possible along the whole nodal line, so as by finding and marking them with a pencil it would be possible to accurately trace the complete nodal lines, as long as someone do not considers this a waste of time.....

Some spots are better than others (more volume) and you can choose a point where you hear one mode only, gripping on its nodal line but at the same time stopping all the others (same grip point in antinodal areas).

 

 

 

Share this post


Link to post
Share on other sites
41 minutes ago, Davide Sora said:

Anyway, the grip points are possible along the whole nodal line, so as by finding and marking them with a pencil it would be possible to accurately trace the complete nodal lines, as long as someone do not considers this a waste of time.....

I like to waste this time and make copies on sheets of the received nodal-line-courses. My observation was, that especially the M5 nodal-lines in the back - upper-bouts moves interestingly while graduating. If the plate was about 1-2  mm more than final condition the M5 line in some of my plates didn´t close at all in the upper plate-end. In this state it still has a more M2-like (transversal) character in the upper bouts.

However my explorations are only for M2 and M5 and I admire very much your extended way to use much more modes, which surely gives you much more valuable informations about several local plate-conditions than I can get. 

Share this post


Link to post
Share on other sites
21 minutes ago, Danube Fiddler said:

My observation was, that especially the M5 nodal-lines in the back - upper-bouts moves interestingly while graduating. If the plate was about 1-2  mm more than final condition the M5 line in some of my plates didn´t close at all in the upper plate-end. In this state it still has a more M2-like (transversal) character in the upper bouts.

That's the big problem... finding any meaningful connection between the free plate observations and the final assembled result.

In looking at the free plate mode shapes, M2 and M5 look sortof-vaguely-maybe like the way the plates flex in the B modes of the assembled instrument, so it makes sense to pay attention to them, if any of them warrant attention at all.  The other free plate modes are all flapping edges, and get completely eradicated when you glue ribs to them.  These edge flapping modes do indicate something about local stiffness/weight properties... but again, the connection to assembled modes is obscure at best.  Thicknesses taken by a caliper give good local information, and it's easier.

Share this post


Link to post
Share on other sites

Far from having found any holy grail or mystical recipt, I track changes using Modal Analysis.

This take time, but as soon as you learn how to do it quickly, this does not add an enormous amount of time. A matter of minutes, and you find that modes #5 often trade its place with neighours.

Alternatively, you could place a mic close to the upper edge (point of maximum upper width) and hit in the middle with a felted wood stick. An fft would give you the first modes.

Moving the stick around and averaging the fft will allows you to be more confident.

Ok: easier to show than to explain! :-)

(Plate can be sustained by elastic bands, or sitting on a morbid elastic foam such as a pillow foam)

Again: very useful for tracking changes, not as an absolute characterization.

Share this post


Link to post
Share on other sites
1 hour ago, Danube Fiddler said:

I like to waste this time and make copies on sheets of the received nodal-line-courses. My observation was, that especially the M5 nodal-lines in the back - upper-bouts moves interestingly while graduating. If the plate was about 1-2  mm more than final condition the M5 line in some of my plates didn´t close at all in the upper plate-end. In this state it still has a more M2-like (transversal) character in the upper bouts.

However my explorations are only for M2 and M5 and I admire very much your extended way to use much more modes, which surely gives you much more valuable informations about several local plate-conditions than I can get. 

 
44 minutes ago, Don Noon said:

That's the big problem... finding any meaningful connection between the free plate observations and the final assembled result.

In looking at the free plate mode shapes, M2 and M5 look sortof-vaguely-maybe like the way the plates flex in the B modes of the assembled instrument, so it makes sense to pay attention to them, if any of them warrant attention at all.  The other free plate modes are all flapping edges, and get completely eradicated when you glue ribs to them.  These edge flapping modes do indicate something about local stiffness/weight properties... but again, the connection to assembled modes is obscure at best.  Thicknesses taken by a caliper give good local information, and it's easier.

Exactly.

Actually I mainly use Mode 5 as a reference, the other modes are used to have a more global idea of the stiffnes in the comparison between two plates of the same model but not much conclusion can be drawn. Mode 2 is the second of some interest in the list, but frankly I do not care that much lately.

M3 and M4 seem to me dependent on the M2, if you lower the frequency of the latter there is no way to keep up their frequencies, they are basically uncontrollable in a direct way. M6 is bassbar dependent (I call it bassbar mode) but it is as elusive as a ghost to see.

M1 is the lowest mode and so it must have some meaning, but I'm still not sure which one.:P  F holes spacing seem to have some effect.

Knowing how to distinguish the various modes is important to avoid making mistakes, for example on cellos for some years I was convinced to hear the M 2 and instead I was hearing the M 3 that was louder, then I started trying to understand something and I noticed it ....
I think they are simply one more thing to evaluate, and I'm sure that some useful feedback can be found, but being sure of the effect on the finished instrument is another story : a consideration would be that if you use wood with good properties and good arching shapes it is quite difficult to have modes frequency very outside the boundaries so that your violin does not work.:)

Share this post


Link to post
Share on other sites
6 hours ago, Don Noon said:

That's the big problem... finding any meaningful connection between the free plate observations and the final assembled result.

In looking at the free plate mode shapes, M2 and M5 look sortof-vaguely-maybe like the way the plates flex in the B modes of the assembled instrument, so it makes sense to pay attention to them, if any of them warrant attention at all.  The other free plate modes are all flapping edges, and get completely eradicated when you glue ribs to them.  These edge flapping modes do indicate something about local stiffness/weight properties... but again, the connection to assembled modes is obscure at best.  Thicknesses taken by a caliper give good local information, and it's easier.

My idea is not to calculate/predict any concrete mode in the assembled instrument or any (ideal) B1-frequencies. This is, because I don´t believe in any ideal target modes or frequencies. My idea is, to get some infos about the movability and mass/stiffness relations of a single free plate - to document these nodal - infos and to see in a later assembled condition, what was before. As some people said, M2 and M5 don´t represent the edge-regions of a free plate very good. Here some other modes could help. Thicknesses as indication for local stiffness could be a too isolated factor, difficult to bring in connection with arching/graincourse/raycourse and actual plateshape. However I agree - all total difficult and not clear, what finally can really help. I just like it, to make this little "manual modal-analysis" in a similar way as Davide explained it. And additionally I get some documentable stiffness-values to compare e.g. with values given by Dr. Harris or Joseph Curtin in old Cremonese instruments.

Share this post


Link to post
Share on other sites
6 hours ago, francesco piasentini said:

Far from having found any holy grail or mystical recipt, I track changes using Modal Analysis.

This take time, but as soon as you learn how to do it quickly, this does not add an enormous amount of time. A matter of minutes, and you find that modes #5 often trade its place with neighours.

Alternatively, you could place a mic close to the upper edge (point of maximum upper width) and hit in the middle with a felted wood stick. An fft would give you the first modes.

Moving the stick around and averaging the fft will allows you to be more confident.

Ok: easier to show than to explain! :-)

(Plate can be sustained by elastic bands, or sitting on a morbid elastic foam such as a pillow foam)

Again: very useful for tracking changes, not as an absolute characterization.

Which is your equipment for modal - analysis ? 

Share this post


Link to post
Share on other sites
7 hours ago, francesco piasentini said:

HW:

-pcb miniature hammer

-dytran smallest accelerometer

-2x pcb old signal conditioners

-M-audio M-track II

-Dell precision M4400

SW:

George Stoppani Suite for Modal and Sound Analysis

 

Thanks !

Your equipment sounds very professional but also quite expensive. I assume, all together some thousands of dollars ( without Pc or laptop ). Do you feel to have sufficient practical output ?

Share this post


Link to post
Share on other sites
59 minutes ago, Danube Fiddler said:

Thanks !

Your equipment sounds very professional but also quite expensive. I assume, all together some thousands of dollars ( without Pc or laptop ). Do you feel to have sufficient practical output ?

You can read FRANCESCO PDF 

 

REPORT_STSM_Piasentini.pdf

Share this post


Link to post
Share on other sites

I made some test with piezo transducers. I mean that kind of thing you find in chip watches as a "buzzer". It works.

Since I have the expensive transducer, I prefer them.

You still can find used stuff, especially in US (pcb and dytran). I considered it a good investment. I use the hammer for modal analysis, input admittance, sound radiation and internal pressure measurements.

I achieved more consistent results in terms of sound and playability.

And Iove making fancy coloured modal shapes. :-)

Share this post


Link to post
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...
Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.