Body modes as violin sound equalizer

[Peter K-G]

The B1+ should now be right about where the dip was before, right?  It's hard to read off frequencies from your plot to that accuracy.

 

The B1+ did not change, (528 Hz today ), but the B1- went down from 445 to 439. Graphs are made with Excel from data imports from Studio Six digital, http://www.studiosixdigital.com/. great software and IPad is really nice, but it's a problem that all analyzing tools are on Windows. I also use Audacity and Spectrum lab.

[Don Noon]

The tailpiece resonance may have been splitting the B1- resonance rather than the B1+.  Hard to tell what's what just by looking at one series of peaks and dips.

[Peter K-G]

Puzzle solved it was TP split of B1-, thanks!

[Peter K-G]

It's so fun to play around with graphs and compare great Cremonese instruments with my own. I had to do put Violin 3 and Titian in the same graph. Their signature modes Hz are about the same but the dB is not. Titian is clearly more responsive!

 

 

Feel free to comment

[Peter K-G]

Violin 4 and Plowden

 

[Anders Buen]

I do not think that you can conclude on the level differences between two violin spectra unless these are measured using the same system. There will always be an assumption in a normalization done like that, as Michael have pointed out. What is your assumption?

[Peter K-G]

Yes Anders I realize that, my assumption is that as I now know how to control the Hz of the modes I'm beginning to understand how to control the dB:s. I overdid with Violin 4 (The goal was to copy Plowden) and got too great transition hill, and also bridge hill a little bit. but they look alike. Violin 3 have too low A0 and B1- due to too high/full arched top for the wood property.

[Anders Buen]

It's so fun to play around with graphs and compare great Cremonese instruments with my own. I had to do put Violin 3 and Titian in the same graph. Their signature modes Hz are about the same but the dB is not. Titian is clearly more responsive!

 

Violin3_Titian_Capture.PNG

 

Feel free to comment

This is a great tool! I do not think that you can conclude on the responsiveness of the Titian in relation to your own instrument, though. Because we do not really know what the calibration are in the two cases. In my plots I have compared the levels of the A0's or made an assumption based on how old and some fine modern violins compare for the A0 level, as presented in one of Joe Curtins Strad articles (I know more than only that). I am not quite sure if the Titian in general is much louder than most other violins. But it has a very even note to note playing volume which is quite unique, I think. This is related to the details in the distribution of the peaks in the spectra. 

[Peter K-G]

This is a great tool! I do not think that you can conclude on the responsiveness of the Titian in relation to your own instrument, though. Because we do not really know what the calibration are in the two cases. In my plots I have compared the levels of the A0's or made an assumption based on how old and some fine modern violins compare for the A0 level, as presented in one of Joe Curtins Strad articles (I know more than only that). I am not quite sure if the Titian in general is much louder than most other violins. But it has a very even note to note playing volume which is quite unique, I think. This is related to the details in the distribution of the peaks in the spectra. 

 

I do not think that you can conclude on the responsiveness of the Titian in relation to your own instrument

 

Responsivness is a wrong word I used here, I ment overall differences in the spectra. Titian shows more of everything except relativelly, B1+ in Violin 3 is higher. (Violin 4 shows more of everything than Plowden)

 

I have compared the levels of the A0's or made an assumption based on how old and some fine modern violins compare for the A0 level, as presented in one of Joe Curtins Strad articles

 

My beleive is that A0 level in relation to B1 modes is the key.

 

(I know more than only that).

 

Please tell

 

But it has a very even note to note playing volume which is quite unique, I think. This is related to the details in the distribution of the peaks in the spectra.

 

I totally agree

[Peter K-G]

Hope no one thinks that I compare my instruments to be as good sounding as the originals! We are all trying to copy Strads and DG:s beautiful appearance, so I think it's only natural to try to copy the sound also.

[Anders Buen]

I have compared the levels of the A0's or made an assumption based on how old and some fine modern violins compare for the A0 level, as presented in one of Joe Curtins Strad articles

 

(I know more than only that).

 

Please tell

I have the data set Curtin based his article on. Not much more to tell about that.

[Peter K-G]

I continued analyzing the split B1 modes phenomenon. Also went through Schleske's articles and Strad 3D again, and anything I could find on the subject.

 

Is there any more profound studies/explenations on what's really going on when tailpeice, fingerbord, chinrest splits the modes. I found earlier measurments not to be completelly right. It is actually B1+ split but something is splitting B1- also.

From my earlier different measurments (Spring/Summer/Vinter):

 

TP (Tailpiece)  125 125 
B0  273 273 
A0  270 269 
A1  460 459 
B1+  512/526 527 Double freq. due to Chinrest in touch with tailpiece! Summer 2012 - B1+ = 519
B1-  434/441 442 Double freq. due to Chinrest in touch with tailpiece! Summer 2012 - B1- = 436
CBR  397 398 

 

The chinrest was in touch with tailpiece, fixed that, but that was not all to it. When I measured in summertime I didn't notice that strong second B1+

 

The tailpeice has a very strong resonance (two modes 125 & 489) I can hear it when I play also.

 

=> So B0 is splitting A0, somthing is splitting B1-(FB?) and TP is splitting B1+. Really interesting!

 

Best guess, for measured B1 modes is now 445 & 522, by putting wedges in different places and hammering bridge with a light home made hammer.

 

 

 

[Anders Buen]

Ted White is into research on stuff like this. He is manufacturing tailpieces with a moveable mass under the tailpiece. I think this is commercialized now.

[Peter K-G]

Violin 4 and Plowden

 

Violin4_Plowden_Capture.PNG

 

How close is this to tonal copy? I scaled down the whole spectra of Violin 4 and voila:

 

 

The biggest difference is the Plowdens high B1- dB. This is also noticable in spectrum analyzes of Meditation from thais, played on the Plowden (from Strad 3D)

 

Everytime player hits A (444 Hz) there is a peak:

 

[Peter K-G]

The B1+ is also about the same frequence, right between the two peeks of B1+ (Tailpeice split, se post #37)

[Marty Kasprzyk]

How close is this to tonal copy? I scaled down the whole spectra of Violin 4 and voila:

 

Violin4_Plowden_Capture_dB_scaled.PNG

 

The biggest difference is the Plowdens high B1- dB. This is also noticable in spectrum analyzes of Meditation from thais, played on the Plowden (from Strad 3D)

 

Everytime player hits A (444 Hz) there is a peak:

 

Plowden_Meditation_B1-.PNG

Have  you done a spectrum analysis of the A (444 Hz) peak?

[Peter K-G]

Yes Marty, and I also compared it to impact hammering test

 

 

Avg of all impact hammering at 446 Hz:

 

[Peter K-G]

I would like to post the mp3 files also, but I think it might violate copy rights (Strad 3D)

 

Noticed that I had spectrum analyze at 1024. If I set them to 8192 on both Meditation playing and hammering test,

the peaks are 442 Hz in both cases.

[Peter K-G]

I have put a lot of hours analyzing Strad 3D documentation. I think they did not get B1- modes right! I might be wrong but Bissingers Excel Sheet, where B1- is 465 Hz (also some splitting is described), does not fit with the Meditation and hammering mp3 files. I beleive that the Plowdens real B1- is around A (442 Hz). I also have another document showing that.

[Anders Buen]

The mode frequencies will depend on what data one uses. Bissingers and Curtins set are not the same, and are probably not done the same day. Bissingers likely without the chinrest, Curtins set could be both. Also remember the influence of the humidity! :-)

[Marty Kasprzyk]

Yes Marty, and I also compared it to impact hammering test

 

Plowden_Meditation_444Hz.PNG

 

Avg of all impact hammering at 446 Hz:

 

Plowden_Meditation_HammerTest.PNG

Thanks Peter.

 

It looks like the A note's fundamental its first overtone harmonic are both quite large.  Added together they cause the extreme loudness.

 

Many players seem to like even loudness notes.  I'm not good at probability mathematics but I'm guessing that there might be some spacings of a violin's resonance peaks (A0, B1-, B1+ etc) that avoid overly strong or weak notes.  Maybe great violins are just a result of great luck.

[jezzupe]

Thanks Peter.

 

It looks like the A note's fundamental its first overtone harmonic are both quite large.  Added together they cause the extreme loudness.

 

Many players seem to like even loudness notes.  I'm not good at probability mathematics but I'm guessing that there might be some spacings of a violin's resonance peaks (A0, B1-, B1+ etc) that avoid overly strong or weak notes.  Maybe great violins are just a result of great luck.

" Maybe great violins are just a result of great luck"

 

Shh' we can't let the "secret" get out.

[Violadamore]

 Maybe great violins are just a result of great luck.

Given that everything else (workmanship, design, finishing, etc.) is equal, i sort of favor that hypothesis. No two pieces of wood are ever identical, and neither is the handwork involved.  Didn't Strad leave something like 96 unsold violins behind when he died?  "Gee, Tony, it's pretty and all, but it just seems kinda dead........."  One wonders.

[Peter K-G]

Well, I want to be lucky every time

 

Just me doing Action reaserch here so now I'm going to trow in some more non verified stuff, right out of the blue.

Don Noon is also pritty lucky and gets quite predictable results so I'm going to use his spectras for Violin 10 & 11.

 

Here we go:

 

I think Hutchins got the Wood resonance wrong but got the analyzes of how deltas in a good violin should be. She kind of mixed up with the B1- and A1. B1- and B1+ acts together (Top and Back plate). The dipp right between them is the relative strength of Hz and Db. This has an indirect effect on peaks and dipps in transition hill and later in th bridge hill.

 

Because Don Noon gets his plates arched the right way (they are light and strong) he is also "lucky" and gets very clear peaks even if he sets them on the higher end, like Il Cannone and the Ex Vieuxtemps DG's.

Shoot this down with a cannon if you like:

 

 

This is more distinct when the dipp between B1- and B1+ is higher (Hz) when it's lower (Hz) the transition hill becomes more "fuzzy" like the Plowden, Willemotte and my Violin 4. Sorry if the text is sounding like a statement of truth/facts, but it's easier and shorter to write that way.

 

 

[Anders Buen]

A thought experiment: Lets say we have a 1000 master built violins at hand. Then we pick the 10 best out of these through a "very smart process". The experts say that there are about ten really great Strads tonally around. I'd say these ten instruments would aappear as great much by chance and not much more than that.