Sign in to follow this  
Peter K-G

Body modes as violin sound equalizer

Recommended Posts

Not really another opinion, but the same one.

 

It makes no sense to me, relying on the physics and acoustics that I know.

Share this post


Link to post
Share on other sites

What if it's possible to adjust violin sound by controlling relative dB level and Hz for A0, B1- and B1+

Those modes only cover the range of 260Hz to ~600Hz. A graphic equalizer for a violin would need to cover the range of 196Hz up to at least 10000Hz. A0, B1-, and B1+ just don't cover enough territory.

Share this post


Link to post
Share on other sites

Not really another opinion, but the same one.

 

It makes no sense to me, relying on the physics and acoustics that I know.

 

Thanks for your reply. You have posted a lot of great information on this forum that have been useful to me, also to this topic. I also have great respect for your knowledge in physics and acoustics. My goal is not to challenge scientific grounds but to help myself and hopefully other enthusiasts, to find the simple combinations that we people usually miss when searching for the path to results in a certain context. In this case; makeing violins sound the way we want for different violinists.

 

Body modes as violin sound equalizer is actually nothing new. It's just putting words different and visualizing with a picture (post #1)  to get the sense of it. Before someone can find a pattern to solve reletions between these modes and sound spectrum, the simple way is to test altering them on a completed violin. For example put some cotton wool in f-holes, attache lumps of clay in different places to see the changes.

 

The most important thing that I follow is that the sound really changes, in both low spectrum and high when doing this.

 

I will be back with more later...

Share this post


Link to post
Share on other sites

Those modes only cover the range of 260Hz to ~600Hz. A graphic equalizer for a violin would need to cover the range of 196Hz up to at least 10000Hz. A0, B1-, and B1+ just don't cover enough territory.

 

Hi Johnston and thanks for the replay. There's is a lot of example where low frequencis are mesured to predict high level spectras. The idea is not ment to be direct, but given body modes with in certain relations to each other, tells somting about the overall  relative stiffneses of different parts and how the whole soundbox produces sound.

Share this post


Link to post
Share on other sites

I thought we'd have some fun with the graph here. I enclose a graph showing Kreits fiddle and the Titian Strad in the same plot. I have done a simple assumption and normalized the curves just about for the A0 level of the two fiddles.

 

I know that the Titian has quite average A0 level response for Strads but it is slightly weaker than the average for some contemporary fine violin by some 3 dB, so I try that normalization in the second plot.

post-25136-0-82363500-1360310033_thumb.jpg

post-25136-0-85109900-1360310936_thumb.jpg

Share this post


Link to post
Share on other sites

Thanks Anders and Micheal

 

The snapshot that my example is based on (BodyModesEqualizer.jpg - Violin 3), is not directly comparable to Titian's spectrum. At the time the snapshot where taken I also did the same with Violin 1 & 2, I set up the mic to favourise body modes. This was done in Spring 2010 after correcting  Violin1 & 3, Violin 2 is not corrected and follows Hutchins principals.

 

 

post-37356-0-22546900-1360317719_thumb.jpg

 

post-37356-0-19261500-1360317735_thumb.jpg

 

post-37356-0-55561000-1360317745_thumb.jpg

 

 

 

Share this post


Link to post
Share on other sites

What if it's possible to adjust violin sound by controlling relative dB level and Hz for A0, B1- and B1+

 

I'd say you could do some adjusting. But maybe not "solve the problem" of adjusting violin sound. 

Share this post


Link to post
Share on other sites

Would you care Peter KG to share how you excite your instrument and how you hold it for your measurements. It seems to be pretty clear that the signal lack some high frequency energy and I think that must be due to the measurement input signal rather than the fiddle. Or at least, it could be.

Share this post


Link to post
Share on other sites

Would you care Peter KG to share how you excite your instrument and how you hold it for your measurements. It seems to be pretty clear that the signal lack some high frequency energy and I think that must be due to the measurement input signal rather than the fiddle. Or at least, it could be.

 

These particular measurements where taken very primitively. The violins where one by one, lying on soft foam on upper and lower blocks, with mic hanging 30-40 cm above them. Hammering was done very gently with a pencil that had a small piece of eraser (rubber) attached to it, at the corner of G-string side of the bridge. It was on purpose to just excite the body modes gently.

 

I usually do measuring with real time FFT software on an IPad, while playing and use my senses to here the difference in sound (and also see the spectrum changes). I have planed to make a measuring system (Curtin style) but haven't had the time

Share this post


Link to post
Share on other sites

The most important thing that I follow is that the sound really changes, in both low spectrum and high when doing this.

 Yes, any change to an existing instrument will change many things, due to the fact that any given spot on the instrument participates in a whole truckload of modes across the frequency range.  And certain spots will affect one set of modes more than another, and so, theoretically, with enough experience, one could nudge the tone around slightly by selecting the areas to thin out... although my experience has been that the violin has much stronger opinions on what it wants to do.

 

I have no doubt that excellent-sounding violins can be made with plates tuned carefully, with signature modes of the completed instrument falling within the specified range.  I also have no doubt that very horrid violins can be made exactly the same way, and that excellent-sounding violins can be made many alternative ways which violate the specifications of "tuning".

 

So, my fundamental issue is that there is no foolproof recipe to bake the perfect cake.  Anyone promoting such a recipe had better show up with bulletproof evidence that it is true, and solid logic of why.

Share this post


Link to post
Share on other sites

These particular measurements where taken very primitively. The violins where one by one, lying on soft foam on upper and lower blocks, with mic hanging 30-40 cm above them. Hammering was done very gently with a pencil that had a small piece of eraser (rubber) attached to it, at the corner of G-string side of the bridge. It was on purpose to just excite the body modes gently.

That's probably why the high frequency response is weak, try tapping the bridge without the rubber eraser. The eraser does excite the bridge more gently, or slower, meaning less high frequency sound will be present.

Share this post


Link to post
Share on other sites

That's probably why the high frequency response is weak, try tapping the bridge without the rubber eraser. The eraser does excite the bridge more gently, or slower, meaning less high frequency sound will be present.

 

Yes I know, that was kind of the whole idea at that particular measurement; soften the impact to minimize influence on the high frequencies. I was looking for patterns from differencies in the 3 violins body modes compared to frequencies in range 1000 - 5000 Hz. Violin 1 is the sweetest sounding, Violin 3 is the brightest and Violin 2 has least "edge"

Share this post


Link to post
Share on other sites

I think almot any maker on this site might make instruments with a response similar to a fine cremonese violin, if we could look for it in a large enough database. There is nothing enigmatic about it, this can probably be done following a variety of methodes and procedures. There probably are no "right or wrong route" to follow. This is an intuition, but I feel it is right to mention, and I believe this is true.

Share this post


Link to post
Share on other sites

I think almot any maker on this site might make instruments with a response similar to a fine cremonese violin, if we could look for it in a large enough database. There is nothing enigmatic about it, this can probably be done following a variety of methodes and procedures. There probably are no "right or wrong route" to follow. This is an intuition, but I feel it is right to mention, and I believe this is true.

 

Yes, this is the most reasonable way to think of the fine art of violin making. But  it's very inspiring trying to understand how the Cremonese masters figured out how to make total excellence over 300 years ago.

 

My strength is to find patterns in small sample of information, using intuition and unconventional methods. (occasionally it's called action research, and is not allways recognized as a scientific method). Scientists usually don't like it  :)  

 

I am not in any way trying to prove to others how to make the best sounding violin, just want to share. I have gotten a pretty good sense on how to make violins sound in the direction I want them. If I manage to find the time I will try to publish Violin 3 & Violin 4 spectras with sound samples. They are as different as Titian & Plowden.

 

You probably have my violin data allready but here comes again:

 

Violins_Analyses.pdf

Share this post


Link to post
Share on other sites

My strength is to find patterns in small sample of information, using intuition and unconventional methods. (occasionally it's called action research, and is not allways recognized as a scientific method). Scientists usually don't like it  :)  

There are infinite paths to obtaining a desirable result.  I don't dislike any of them, including yours, whether based on conventional acoustics, intuition, tradition, or phases of the moon.  The problem I have is when unconventional methods are presented with scientificky numbers and measurements without the solid, conventional (i.e. real) physics and logic to support it. 

 

"Here's what I did and this is the result" ... needs no support. :)

"Here's what I did and this is the result and why it's that way"... needs the physics and logic. :huh:

"Here's what I did and the result is wonderful and this is why it so good"... is beyond physics and logic. <_<

 

In the violinmaking world, I'm certainly one of the science types... but I'm a firm believer in good old trial-and-error as the most powerful method for learning what works and what doesn't.  I also think science and technology can be a help... but not a substitute.

Share this post


Link to post
Share on other sites

"Here's what I did and the result is wonderful and this is why it so good"... is beyond physics and logic. <_<

 

 

I beleive this is the key issue to overcome (actually I know because I practice this every day with software scientists at work. "I know" is also a strong word that can be attacked here, sorry about that.).

 

It's about different peoples emotions and how we discover new things differently.

 

Within a certain frame of reference measured results is the proven results but outside the frame it may not be. This you can explain better. I have read and listened to a lot of your posts on this forum so "I know" that you are much better than me to highlight measureable things.

 

In the past when I was younger I could sometimes be quite insensitive and say something to punch hole in a large study that somone had done, by just scrolling through the documentation. And this just based on my own gut feelings, not nice! Hope I can be a little less enthusiastic on this forum and  put text more reasonable.

 

What I really would like to post to this topic is my discoveries and all the wonderful results and why it's so good, from years of experimenting with Violin 1, 15 years ago, and how this can be coupled to results today with corrections of Violin 1 & 3 and making Violin 4, linked to the results following a certain book, in combination with Strad 3D, Ex Vieuxtemps project and all Il Cannone recordings that I have and listen to every day. All this combined with yours and Anders posts, none of this can be proven scientifically - I wouldn't Dare

Share this post


Link to post
Share on other sites

Yes, this is the most reasonable way to think of the fine art of violin making. But  it's very inspiring trying to understand how the Cremonese masters figured out how to make total excellence over 300 years ago.

 

My strength is to find patterns in small sample of information, using intuition and unconventional methods. (occasionally it's called action research, and is not allways recognized as a scientific method). Scientists usually don't like it  :)  

 

I am not in any way trying to prove to others how to make the best sounding violin, just want to share. I have gotten a pretty good sense on how to make violins sound in the direction I want them. If I manage to find the time I will try to publish Violin 3 & Violin 4 spectras with sound samples. They are as different as Titian & Plowden.

 

You probably have my violin data allready but here comes again:

 

attachicon.gifViolins_Analyses.pdf

I just want to thank you for sharing this data. I can use that information along with data extracted from your spectra of the same fiddles in my correlation research between free plate data and signature mode properties. 

Share this post


Link to post
Share on other sites

Violin 4 appears to show the mode-splitting effect of the tailpiece resonance, with the resonance tuned to the B1+ frequency.  Put a wedge between the tailgut end of the tailpiece and the top (or chinrest)  and compare what you get then.

Share this post


Link to post
Share on other sites

Put a wedge between the tailgut end of the tailpiece and the top.

 

That was it!

 

I did try several times to put a lump of modelling clay on different places on the TP with no effect, I also sort of glued the TP to chinrest but that lowered B1+ to around 500 Hz.Modelling clay between the gut and top did the trick. B1+ now 528 and B1- 439.

Share this post


Link to post
Share on other sites

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.

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.