Can a Hammer and Light Measure the Quality of Violins? L. H. Morset 1
Can a Hammer and Light
Measure the Quality of Violins?
The purpose of this article is to provide examples of research on violins based on research at NTNU. My current doctoral work is on studying emission from sound sources. My advisors at NTNU are from three different sciences: professor Asbjørn Krokstad from Acoustics, professor Ole Johan Løkberg from Applied Optics and professor Ola Kai Ledang from MI. I also collaborate with student Kjersti Hadland that is writing her thesis about violins at MI. My research interests are measuring methods and musical instruments, particularly the voice, the piano and the violin.
Basics of the Violin
This section contains simple facts about the violin mostly taken from ;Rossing-91, pp. 235-268；. For
further reading “the physics of the violin” ;Cremer-84； is recommended and has become the standard
on the subject.
Bowed string instruments have held a special place in music for many years. The violin developed gradually from the various bowed string instruments used in Europe during the Middle Ages. The instruments of the violin family, as we know it today, were developed in Italy from the sixteenth century and reached a peak in the eighteenth century by masters such as Antonio Stradivari (1644-1737) and Guiseppe Guarneri del Gesú (1698-1744) of Cremona. Even today the surviving instruments of these masters are regarded as some of the finest ever produced, although they have been altered to produce the more powerful sound required for large concert halls.
Figure 1 Parts of a violin ;Rossing-91 p. 238；
Figure 1 shows the essential parts of a conventional violin in an exploded view. The top plate is generally carved from Norway spruce and the back, ribs, and neck from curly maple. The fingerboard is made of ebony and the bow stick from pernambuco. Running longitudinally under the top is a bass bar, and a sound post is positioned near the treble foot of the bridge. The bridge is unnamed in the figure, but can be seen under the strings. The four strings of steel, gut or nylon are tuned to G, D, A 344
and E, where G is on the bass bar side. 53
Can a Hammer and Light Measure the Quality of Violins? L. H. Morset 2
The strings themselves do not transmit much sound because they are very thin compared to the wavelength of the sound waves. We might have noticed how little sound we can hear from an electric guitar being played without electric amplification. An acoustic guitar transmits much more sound since it has a resonance body that acts as an acoustic amplifier. The violin body has the same function. The aim of this paragraph was to stress the importance of the violin body. It is very important to find how the violin body vibrates in order to describe the quality of the violin. The vibrating violin body makes the air vibrate, which creates sound waves that we can hear.
How Can We Make the Violin Vibrate?
Vibrations of a violin can be measured using many different methods. What all methods have in common is that the violin somehow has to be put into vibration before the vibrations can be measured. To put the violin into vibration is called to excite the violin. For example when a violin is being bowed, the string being played is set into horizontal vibrations by the bow. These vibrations of the string will be transferred through the bridge to the violin body since the string is fastened to the violin bridge. In other words, when a violin is being bowed, the exaction source is the string and the source is situated where the string is fastened to the bridge. The reason why we emphasize this, it that other sources than the string can be used for excitation when measuring violin body vibrations. Examples of other sources used can be found in ;Jansson-82； and ;Bork-93；.
The goal of my master thesis work ;Morset-96； was to conduct measurements that approximately found
out how the violin vibrates when it is being played. This was not so simple since when the violin is being played, the player can disturb the measurements. To avoid this disturbance, the violin is usually supported in another, but similar way. An example of such a support is shown in Figure 2. The support is supposed to hold the violin in approximately the same way as the player holds it. As seen from the figure below, both a shoulder and a chin support are present.
Figure 2 Violin support ;Bork-93；
It can be difficult or time consuming to use the string as the excitation source. The Fourier theory ;Proakis-92 pp.152-158；, invented by the famous mathematician Fourier, tells us that as long as we have a linear medium that the violin body approximately is, other excitation sources than the string can be used. It is important that the excitation must be given at the same place. The excitation source thus has to be placed where the string is fastened to the bridge. In addition the excitation source has to have
Can a Hammer and Light Measure the Quality of Violins? L. H. Morset 3
enough vibrational power in the frequency range we want to measure. This frequency range is about 196-15000 Hz ;Morset-96 p.8； for the violin.
The paragraph above was intended to motivate the reader to accept the following. The vibrations of the violin can approximately be measured using a hammer as excitation source. The requirements for this are that the hammer hits very close to where the string is fastened to the bridge and that the hit contains enough power of the frequencies we want to measure. A picture of the hammer and the violin is shown in Figure 3. The signal that the hammer excites the violin with is called an impulse because it is of a very short duration. The excitation is thus called impulse excitation. A more detailed discussion of the excitation in present in the next chapter.
Figure 3 Impulse excitation of a violin ;Morset-96, p. 16；
More About Exciting the Violin
An important parameter for all kinds of vibrations is called resonance. We have a resonance at frequencies where the vibration is strong, se for example ;Morse-68；. This frequency is called the
Which resonances are exited and how much they are exited depends on where the excitation is placed. It is important that the excitation we use for measurements excite the same resonances as the player does when the violin is being played.
When the violin is being played, the strings are usually excited by a bow, but can also be plucked by a finger. The strings do not radiate much sound because of their small volume, so the vibrations of the violin body cause the sound we hear. These vibrations are due to the force from the strings transferred through the bridge to the violin body. This can be measured, but before discussing the measurements, we state the properties the excitation should have ;Krokstad-96 ；.
Can a Hammer and Light Measure the Quality of Violins? L. H. Morset 4
Sufficient power must be generated in the frequency range where the measurements are intended to be accurate.
The excitation source should be placed as close as possible to where the strings are fastened to the bridge.
The main part of the excitation power must be transmitted into the bridge in the same direction as the force exerted when bowing a violin (this is if we want to find the quality of the violin with respect to bowing, not plucking).
The signal of the excitation source must be measurable or known
;Raman-18； invented a machine that performs automatically bowing of violins, and this has been used for measurement. This excitation is almost identical to the excitation of a violin being bowed by a musician. Unfortunately, only a few frequencies are exited for each tone played with this method, and we want to find the vibrations of the violin for all the tones (frequencies). This means that many tones must be played one at a time, which makes it a slow measuring method.
If the strings have been used to excite the violin, the type of strings used could influence the measurements. By exiting the bridge directly with a hammer, for instance by exiting on one side of the bridge as suggested by ;Bork-93；, the type of strings used will not influence the measurements.
Measuring Violin Body Vibrations Using Laser Light When the violin is excited and put into vibrations we would like to measure the vibrations. A smart way of doing this is a method called holographic interferometry, which use laser light. By using laser light very accurate measurements can be obtained, and we don’t need to touch the violin to detect the vibrations. Professor Ole J. Løkberg at Applied Optics, NTNU and Erik Jansson from Stockholm has measured the vibrational behavior of the violin using such methods. An example of a holographic measurement is seen in Figure 4.
Figure 4 Modal shapes for six modes in a violin. The heavy lines are nodal lines (where there is no vibration) and the phase of motion is indicated by + or - ;Jansson-82a；
The modal shapes of violins differ, but we assume the principal shapes of deformation are nearly the same for all instruments if they are built the same way. The frequency of the resonances varies and has been measured for many violins ;Jansson-82；. The resonance frequencies were found to be within the
ranges seen in Figure 5 where the modes are divided into air-, body-, top- and backmodes. These names reflect where the modes are situated on the violin body.
Can a Hammer and Light Measure the Quality of Violins? L. H. Morset 5
Figure 5 Frequency ranges for different types of resonances in a violin ;Jansson-
The way of identifying the modes of the violin described in Table 1 was employed by Erik Jansson and his co-workers and has become a standard in the field.
, A, A,… Air modes Motion of enclosed air. A012
T, T, T,… Top modes Motion primarily of the top plate. 123
B, B, B,… Back modes Motion primarily of the back plate. 123
C, C, C,…, Body modes Similarly motion of top and back plate. Mode N (neck) and C 1231
also N (corpus) are one-dimensional bending modes, modes C, C,… 23
are two-dimensional flexing modes.
Table 1 Nomenclature for modes in a violin ;Jansson-82a；
Some of the modes in Table 1 can be seen in Figure 4.
Measuring the quality of violins using objective methods is not an easy task. But we have seen that different methods can be used. One of the problems has been that different methods do not give the same results. In the near future, as the measuring methods become better, I think this problem will be solved. The remaining problem is to find which factors in the huge amount of measurement data are really important for the quality of violin sound. This can be done by correlating objective measurements with subjective measurements, and is a task that can only be solved by using knowledge from several sciences.
Can a Hammer and Light Measure the Quality of Violins? L. H. Morset 6
Bork, "A Fast Method for Measuring the Vibrations of Violins", reprint from a work thpresented at the 94 Convention of the Audio Engineering Society, Berlin (1993). ;Cremer-84；
L. Cremer, The physics of the violin, The MIT Press (1984).
J. Alonso Moral and E. Jansson "Eigenmodes, input impedance, and the function of the violin", Acustica, vol. 50, pp. 329-337 (1982).
Conversation with Prof. A. Krokstad, Norwegian University of Science and Technology, Trondheim (1996).
P. M. Morse and K. U. Ingard, Theoretical Acoustics, p.47, Princeton University
Press, Princeton, NJ (1968).
L. H. Morset, “Impulse excitation of violins”, M. S. Thesis, Norwegian University of Science and Technology, Trondheim (1996).
C. V. Raman, “On the mechanical theory of the vibrations of bowed strings and of musical instruments of the violin family, with experimental verification of the results”, Bull. 15, The Indian Association for the Cultivation of Science (1918). ;Rossing-91；
N. H. Fletcher and T. D. Rossing, The physics of musical instruments, Springer-
Verlag New York Inc. (1991).