Outside graduation

[Adam Edwards]

Well it has surfaced again......tuning or graduating from the outside after soundbox is together.

1 ) My thoughts at the moment is that most of the work is done on the back

2 ) String vibration determines when to stop removing wood

3 ) Can we pick this up on oscilloscope

This is just to get us started.......... I am yet to experiment with these ideas

please feel feel to add to the list of possibilities

If you can't make a positive contribution then DONT , this thread is only for those that want

to pursue this idea !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

[Mauricio]

2 ) String vibration determines when to stop removing wood

Thanks for starting this thread, Adam. Is point #2 based on Oded's idea or have you figured things mathematically? I ask about figuring out mathematically because I've always found it very intriguing that the vibrating length of the body (perhaps not accurate terminology being used here), say, the internal distance within top and bottom blocks corresponded with the original string length of the baroque era.

[Adam Edwards]

Thanks for starting this thread, Adam. Is point #2 based on Oded's idea or have you figured things mathematically? I ask about figuring out mathematically because I've always found it very intriguing that the vibrating length of the body (perhaps not accurate terminology being used here), say, the internal distance within top and bottom blocks corresponded with the original string length of the baroque era.

Oded's idea , I haven't read enough about what he does to fully understand.......

but I do think this concept has merit

[Oded Kishony]

I'm getting enough requests that I think it would be simpler if I just posted this manuscript version of my paper here. It's missing some pictures but I think it's very accessible without the them.

Oded

_______

Using String Vibration to Analyze The Acoustics of an Instrument

Abstract

A method used to analyze the acoustics of string instruments that

utilizes the principle of 'string reciprocity'. (6) In string

reciprocity when the surface of the violin is excited by tapping or

scraping, the vibrations cause the strings to vibrate in the same way

that the strings cause vibrations of the surface of an instrument.

This reciprocal relationship is used to analyze the acoustics of the

violin.

This paper explains how to isolate resonant frequencies, locate nodal

lines of those frequencies and identify air modes of instruments. The

shapes of the modes, their frequencies and location on the surface are

important determinant of the sound of an instrument

Method

All of the methods utilize the same basic technique. Begin with an

instrument that is strung up and in tune. To identify a discreet

frequency, three strings are dampened and only one is left to vibrate.

Normally the three lower strings (G,D,A on the violin) are dampened.

The top string is usually easier to tune to a wider range of

frequencies and due to it's lower mass is more responsive to

excitation. It's important to carefully choose the object used to tap

the surface because it will have an effect on the outcome. Tapping

with a hard metal object, like the back of a chinrest key, will excite

a wide range of frequencies, while tapping with something softer, like

a pencil eraser, will tend to excite only the lower frequencies. These

methods are meant to complement the use of computerized acoustical

analysis, they are limited in frequency range and the process of

gathering data is tedious in comparison. However, they can be useful

to the violinmaker as a convenient, accessible, workbench method to

gain insight into the vibrating guts of an instrument.

spectrum analysis: (1)

Objective:

to isolate those frequencies of a violin that will vibrate most

easily, the resonant frequencies. (2)

Procedure:

Dampen the three bottom strings.

Tune the open E-string down to G- 196Hz.

Tap the side of the bridge, treble or bass side (usually with a

chinrest key) while slowly raising the pitch of the E-string.

As the string nears a resonant frequency, it begins to noticeably

vibrate and hum.

Be sure to listen for string harmonics which indicate higher

frequency resonances

Variations:

use the fine tuner to help center over the frequency. Sweep back an

forth and listen for the most open and sustained vibrations. Another

method to help separate overlapping resonances is to sweep quickly

over a wide range by sliding the fingernail over the E-string (the

fleshy part of the finger dampens the string too much) while tapping

the side of the bridge. If there is a lot of overlapping resonances

then try using a pencil eraser or varying the force of the tapping.

A simple modal analysis: (5,3)

Objective: Having isolated a resonant frequency, locate the nodal

lines that separate areas of vibrations on the surface.

Procedure:

With the E-string tuned to a resonant frequency and the three lower

strings dampened, tap the surface of the instrument with the

fingernail or other hard object. The nodal lines are the areas where

the E-string will not vibrate as much.

Variations: Use a piece of chalk to tap an unvarnished violin and

make a dot where the vibrations decrease, the lines formed by the

connected dots are nodal lines.

To analyze air modes:

Objective:

To identify selected* frequencies of internal air movement.

Dampen the three bottom strings, tune the E-string to G (196 Hz) or

lower as desired. Repeatedly puff into one of the f-holes with a

plosive, as if saying the letter P, while slowly raising the pitch of

the open E string. When the string starts to vibrate, that indicates

an air mode. While this method can be used to identify the air modes

it is not strictly speaking reciprocal, since the air doesn't exit the

f-hole the same way it enters.

Variations:

When analyzing for an air mode you then tap the surface of the

instrument to search for any nodal lines that indicate a corpus or

plate resonance at the same frequency as the air mode.

* the string is responding to mechanical vibrations modified by

coupling to the confined air. The method does not register air modes

that do not exhibit pressure variations at the F holes.

Testing Sound post Tension

Objective:

A convenient method for testing the tension of the sound post. This

method only indicates the extremes of loose/tight although in the

author's experience most instrument will be most open and resonant if

the sound post is within the middle range.

Procedure:

Tap the area above and below the treble side f-hole. I usually use my

fingernail. Listen to the vibrations of all the open strings. If the

area above the f-hole vibrates but the lower area vibrates less then

the sound post is tight (mnemonic- 'Up tight'). If the post is loose,

then tapping below the f-hole will cause the open strings to vibrate

more than tapping above the F hole

While this test indicates the tightness of the sound post, it is not

necessarily the ideal tension for every player. Some musicians will

prefer either a looser or tighter post setting.

Analysis of Acoustical Problems (2)

Objective:

To use string reciprocity as an aid in isolating and resolving

various acoustical problems of instruments.

Procedure:

Using the methods outlined above the problem frequency can be

isolated and then analyzed. For instance, a "wolf note", once

identified by playing, can be mapped on the surface of the instrument

using the modal analysis technique. With three string dampened the

open string tuned to the errant wolf note, the surface is then tapped.

The string will then respond and this response can be used to

determine areas of greatest or least vibrations. Strategies can then

be used to selectively dampen the areas of interest. In attempting to

analyze problematic notes, using the air mode and corpus mode

techniques in tandem might reveal the reason for the problem. By

retuning several strings, simultaneous analysis of multiple

frequencies can be done. (6)

Conclusion:

Presented here is a simple, novel method for examining an instrument

acoustically, using only the open strings, without computers, or any

other tools. While I've experienced instances of great precision in

the outcome, I don't expect that this method will equal the speed and

accuracy of high tech equipment and powerful computers. Nevertheless,

it can provide the working violinmakers with a new, simple tool to

gain acoustical insights. I hope in future publications to expand the

repertoire of useful techniques, to introduce the use of string

reciprocity in new instrument construction and to also develop methods

to help in resolving acoustical issues of older instruments.

ACKNOWLEDGMENTS

The author would like to express his deep appreciation to Oliver

Rodgers for his early encouragement , continued support, tutoring in

acoustics and patience during the development of these ideas. Thanks

also to Jim Woddhouse, Martin Schleske, George Stopanni, Michael

Darnton, Fan Tao, Joe Curtin, the VSA Oberlin Acoustical workshop,

Norman Pickering, Members of Tobi-L list serve.

REFERENCES

1. Arthur Benade, Fundamentals of Musical Acoustics, second revised

edition 1990.

Dover 0-486-26484-X

2. J. Woodhouse, Body vibration of the violin – What can a maker

expect to control?

Catgut Acoustical Soc. J., Vol. 4, No. 5 (Series II)

3. Hermann Helmholtz, On the Sensation of Tone 1885 note Pg 87 Dover

1954 ISBN 0-486-60753-4

Mr. Helmholtz describes his method for analyzing classic Cremonese

instruments using tuning forks.

4. Martin Schleske, Empirical Tools in Contemporary Violin Making

Part 1. Analysis of Design, Materials, Varnish, and Normal Modes, CASJ

Vol.4, No5. (Series II) May 2002

5. Oliver Rodgers, Pamela Anderson, Finite Element Analysis of a

Violin Corpus, CASJ Vol 4 (Series III) November 2001

6. Oliver E. Rodgers, The effect of the musical key on perceived

violin tone quality, CASJ Vol2, No5 (SeriesII) May 1994

[Melvin Goldsmith]

That's very cool Oded. Thanks!

[Fjodor]

Thanks Oded,

I can't comment on outside graduation but I read your article you sent me earlier and I think this is a simple and hands on way to analyze an instrument. For a newbie like me it's also a great way to find out where a violin vibrates at different frequencies.

[Oded Kishony]

Adam wrote:

Can we pick this up on oscilloscope

Yes, my teacher and mentor Oliver Rodgers had something he called a 'mode sniffer' that used a 'scope and tone generator. The tone generator was set up to vibrate the bridge and the 'scope was hooked up to a lapel mic. He would slide the tone generator from low to high, as you passed through a resonant frequency it got louder whem you passed the lapel mic over the surface of the instrument and as you passed over a nodal line the display on the 'scope flipped- this also gave you phase information. You can download a computer scope as well as a tone generator. And if you have two sound cards you can run both simultaneously on a computer (thanks to Michael Darnton for that bit ;-)

Oded

[Adam Edwards]

Oded .......thanks for your imput

I am in deep thought about what I've read , so at the moment I can't think of anything else to add to the discussion

[Oded Kishony]

The idea in the paper is really very simple.

That the surface causes the strings to vibrate in the same way that the strings cause the wood to vibrate.

This means that you can now explore any given point of a violin and discover what effect that point has on the strings.

It's like playing the violin backwards.

Instead of exciting the string to make the wood vibrate, you excite the wood to make the strings vibrate.

See? simple.

Oded

[Adam Edwards]

Thanks Oded....................my thoughts now are about finishing the fluting area using this method

and getting the back and belly into a good relationship....................

At the moment I just thin plates to my graduation pattern and glue soundbox together and hope

it will turn out...........I would like a little more control over the finishing of the soundbox

[Magnus Nedregard]

Oded, can you think of a way to use your method to control the so called "nasal range" (about 1000-1800 Hz I beleive?) when making instruments? I find this is one of the more difficult things to control. You want it to be there, to give a strong presence to the tone, but you want it to sort of blend in and not dominate. I think most often it is too dominant in new instruments.

I like your article!

[Oded Kishony]

Hi Magnus,

I like your article!

Thanks!

can you think of a way to use your method to control the so called "nasal range" (about 1000-1800 Hz I beleive?) when making instruments?

Yes, I assume that you're starting with an instrument that's well set up with strings and a bit too stiff, where wood will have to be removed.

Dampen the two bottom strings and tapping with something hard like a scraper edge or finger plane listen carefully for the strings to resonate at their harmonics. These harmonics can go up to very high frequencies-the limit is usually one's hearing As you go around the instrument find the area with the greatest amount of high frequency and locate the nodal lines, ( string vibrations stop or reduced) then carefully scrape adjacent to the nodal line. The idea is to increase the vibrating area of that high frequency. You can tap with a piece of chalk and when you get to a nodal line make a mark then connect the dots. You do not want to go around the entire instrument but rather concentrate on distinct areas, because going all around will defeat your effort (phase canceling)

If you want to use a computer to enhance this process then do a spectrum analysis by tapping the bridge and identify a resonant frequency within the range you're interested in, then tune one of the high strings to that frequency or an octave below and follow the same procedure as above.

This process has a learning curve, don't expect instant results, listen carefully in a completely quiet room. Be sure to tap the ribs as well. When you decide where you want to scrape keep listening to the vibrating strings as you're scraping, this will tell you where the sound is going. As you're scraping you'll want to hear that high harmonic frequency sustaining and getting louder. Stop periodically and play the instrument, see what's going on. (or do a spectrum analysis to monitor the changes)

Granted it's not easy and there's some trial and error involved but it's a lot better than just going at it blindly IMHO. And over time it will become easier and more efficient.

Oded Kishony

PS This method can also be applied to regraduating an instrument by mapping the outside and marking the relevant areas with a non permanent marker. This will at least identify some possible areas to thin. However it's not as predictive as when the instrument is assembled but it will yield some clues.

[Selim]

I like the approach and the presentation. I have not worked it in details yet.

Thanks for pushing the limits and sharing it with us.

One issue coming to my mind;

Removing wood is exactly the same as re-arranging the nodal map on the plate and implicitly entire violin body, I mean shifting the positions of vibrating or non-vibrating spots for certain frequency group. The final objective is to catch the best combination and positions in for the intended frequecy group(note), and the combination will change for every other note on the fingerboard.The combination may be good for G notes, but not necessarily for treble, which is the actual case in my experienece. Reaching a good blanace is critical, I think.

I think, taking the vibration may be one dimention, there are other dimensions neded for attributes such as response, playability and the most importantly, preserving the complex frequency group that makes the sound unique.

Doagnosing the problem is the most important stage, and it seems reverse playing approach is a very creative step to get out of abstractions.

[Roger Hill]

Hi Oded:

Suppose you have carved your plates to Curtate Cycloid outer shape. After applying your method, how much (typically, in your experience) are the resulting arches distorted from CC's? Is the distortion from CC's visible?

[Oded Kishony]

The arching is as distorted as you make them meaning that the arching does not need to be distorted at all. You can decide which is more important, perfect undistorted curtate cycloid arching or a better sounding instrument? While the overall arching of the best Cremonese instruments is often very beautiful it is never mathematically perfect.

Take a look at this Guarneri. Notice that the arching 'breaks' between the bridge and the F hole. There is a ridge running North and South the entire length of the F hole. It's possible, in my opinion, that this photo is showing exterior graduation.

(edit) Another intriguing feature of this photo is that the texture of the spruce/varnish is different along the side of the F hole. To me this suggests the possibility that the surface may have been scraped after a sealer had been applied.

Oded

[Berl Mendenhall]

Oded,

Thank you for posting this article and all the information. I haven't had a chance to read it yet but looking forward to it. From reading your post over the years I'm sure it will very be informative.

Berl

[Roger Hill]

That is pretty convincing, Oded. Do you have any feelings as to whether it might also have been common to do some adjustment near the ff holes by scraping the inside? Special tools shouldn't be all that difficult to make.