Experimental Modal Analysis
A Simple Non-Mathematical Presentation
Peter Avitabile, University of Massachusetts Lowell, Lowell, Massachusetts
Often times, people ask some simple questions regarding
modal analysis and how structures vibrate. Most times, it is
impossible to describe this simply and some of the basic un-
derlying theory needs to be addressed in order to fully explain
some of these concepts. However, many times the theory is just
a little too much to handle and some of the concepts can be
described without a rigorous mathematical treatment. This
article will attempt to explain some concepts about how struc-
tures vibrate and the use of some of the tools to solve struc-
tural dynamic problems. The intent of this article is to sim-
ply identify how structures vibrate from a nonmathematical
perspective. With this being said, let’s start with the first ques-
tion that is usually asked:
Could you explain modal analysis for me?
In a nutshell, we could say that modal analysis is a process
whereby we describe a structure in terms of its natural char-
acteristics which are the frequency, damping and mode shapes
– its dynamic properties. Well that’s a mouthful so let’s explain
what that means. Without getting too technical, I often explain
modal analysis in terms of the modes of vibration of a simple
plate. This explanation is usually useful for engineers who are
new to vibrations and modal analysis.
Let’s consider a freely supported flat plate (Figure 1). Let’s
apply a constant force to one corner of the plate. We usually
think of a force in a static sense which would cause some static
deformation in the plate. But here what I would like to do is to
apply a force that varies in a sinusoidal fashion. Let’s consider
a fixed frequency of oscillation of the constant force. We will
change the rate of oscillation of the frequency but the peak force
will always be the same value – only the rate of oscillation of
the force will change. We will also measure the response of the
plate due to the excitation with an accelerometer attached to
one corner of the plate.
Now if we measure the response on the plate we will notice
that the amplitude changes as we change the rate of oscilla-
tion of the input force (Figure 2). There will be increases as well
as decreases in amplitude at different points as we sweep up
in time. This seems very odd since we are applying a constant
force to the system yet the amplitude varies depending on the
rate of oscillation of the input force. But this is exactly what
happens – the response amplifies as we apply a force with a
rate of oscillation that gets closer and closer to the natural fre-
quency (or resonant frequency) of the system and reaches a
maximum when the rate of oscillation is at the resonant fre-
quency of the system. When you think about it, that’s pretty
amazing since I am applying the same peak force all the time
– only the rate of oscillation is changing!
This time data provides very useful information. But if we
take the time data and transform it to the frequency domain
using the Fast Fourier Transform then we can compute some-
thing called the frequency response function (Figure 3). Now
there are some very interesting items to note. We see that there
are peaks in this function which occur at the resonant frequen-
cies of the system. And we notice that these peaks occur at fre-
quencies where the time response was observed to have maxi-
mum response corresponding to the rate of oscillation of the
input excitation.
Now if we overlay the time trace with the frequency trace
what we will notice is that the frequency of oscillation at the
time at which the time trace reaches its maximum value cor-
SOUND AND VIBRATION/JANUARY 2001
responds to the frequency where peaks in the frequency re-
sponse function reach a maximum (Figure 4). So you can see
that we can use either the time trace to determine the frequency
at which maximum amplitude increases occur or the frequency
response function to determine where these natural frequen-
cies occur. Clearly the frequency response function is easier
to evaluate.
Most people are amazed at how the structure has these natu-
ral characteristics. Well, what’s more amazing is that the de-
formation patterns at these natural frequencies also take on
a variety of different shapes depending on which frequency is
used for the excitation force.
Now let’s see what happens to the deformation pattern on
the structure at each one of these natural frequencies. Let’s
place 45 evenly distributed accelerometers on the plate and
measure the amplitude of the response of the plate with dif-
ferent excitation frequencies. If we were to dwell at each one
of the frequencies – each one of the natural frequencies – we
would see a deformation pattern that exists in the structure
(Figure 5). The figure shows the deformation patterns that will
result when the excitation coincides with one of the natural
frequencies of the system. We see that when we dwell at the
first natural frequency, there is a first bending deformation
pattern in the plate shown in blue (mode 1). When we dwell
at the second natural frequency, there is a first twisting defor-
mation pattern in the plate shown in red (mode 2). When we
dwell at the third and fourth natural frequencies, the second
bending and second twisting deformation patterns are seen in
green (mode 3) and magenta (mode 4), respectively. These de-
formation patterns are referred to as the mode shapes of the
structure. (That’s not actually perfectly correct from a pure
mathematical standpoint but for the simple discussion here,
these deformation patterns are very close to the mode shapes,
from a practical standpoint.)
These natural frequencies and mode shapes occur in all
structures that we design. Basically, there are characteristics
that depend on the weight and stiffness of my structure which
determine where these natural frequencies and mode shapes
will exist. As a design engineer, I need to identify these fre-
quencies and know how they might affect the response of my
structure when a force excites the structure. Understanding the
mode shape and how the structure will vibrate when excited
helps the design engineer to design better structures. Now there
is much more to it all but this is just a very simple explanation
of modal analysis.
So, basically, modal analysis is the study of the natural char-
acteristics of structures. Understanding both the natural fre-
quency and mode shape helps to design my structural system
for noise and vibration applications. We use modal analysis to
help design all types of structures including automotive struc-
tures, aircraft structures, spacecraft, computers, tennis rackets,
golf clubs . . . the list just goes on and on.
Now we have introduced this measurement called a fre-
quency response function but exactly what is it?
Just what are these measurements
that are called FRFs?
The frequency response function is very simply the ratio of
the output response of a structure due to an applied force. We
measure both the applied force and the response of the struc-
ture due to the applied force simultaneously. (The response can
be measured as displacement, velocity or acceleration.) Now
1
 8

Output Response 5

Input Force 1 2 3

0
Figure 1. Simple plate excitation/response model. −3
8
7
Increasing Rate of Oscillation

−6

Figure 6. A 3 DOF model of a beam.

the measured time data is transformed from the time domain

to the frequency domain using a Fast Fourier Transform algo-
rithm found in any signal processing analyzer and computer
software packages.
Due to this transformation, the functions end up being com-
Figure 2. Simple plate response.
plex valued numbers; the functions contain real and imaginary
components or magnitude and phase components to describe
the function. So let’s take a look at what some of the functions
might look like and try to determine how modal data can be
extracted from these measured functions.
Let’s first evaluate a simple beam with only 3 measurement
Magnitude

locations (Figure 6). Notice that the beam has 3 measurement

locations and 3 mode shapes. There are 3 possible places that
forces can be applied and 3 possible places where the responses
can be measured. This means that there are a total of 9 pos-
sible complex-valued frequency response functions that could
Frequency be acquired; the frequency response functions are usually de-
scribed with subscripts to denote the input and output loca-
Figure 3. Simple plate frequency response function. tions as h out,in (or with respect to typical matrix notation this
would be h row,column).
Figure 7 shows the magnitude, phase, real and imaginary
parts of the frequency response function matrix. (Of course, I
am assuming that we remember that a complex number is made
up of a real and imaginary part which can be easily converted
to magnitude and phase. Since the frequency response is a
complex number, we can look at any and all of the parts that
can describe the frequency response function.)
Now let’s take a look at each of the measurements and make
some remarks on some of the individual measurements that
could be made. First let’s drive the beam with a force from an
impact at the tip of the beam at point 3 and measure the re-
sponse of the beam at the same location (Figure 8). This mea-
surement is referred to as h 33. This is a special measurement
referred to as a drive point measurement. Some important char-
Figure 4. Overlay of time and frequency response functions.
acteristics of a drive point measurement are:
a All resonances (peaks) are separated by anti-resonances.
a The phase loses 180 degrees of phase as we pass over a reso-
nance and gains 180 degrees of phase as we pass over an anti-
resonance.
a The peaks in the imaginary part of the frequency response
Mode 3
function must all point in the same direction.
Mode 1
So as I continue and take a measurement by moving the
impact force to point 2 and measuring the response at point 3
and then moving the impact force on to point 1 to acquire two
more measurements as shown. (And of course I could continue
on to collect any or all of the additional input-output combi-
nations.)
So now we have some idea about the measurements that we
Mode 2
Mode 4 could possibly acquire. One important item to note is that the
frequency response function matrix is symmetric. This is due
to the fact that the mass, damping and stiffness matrices that
describe the system are symmetric. So we can see that h ij = h ji
– this is called reciprocity. So we don’t need to actually mea-
Figure 5. Simple plate sine dwell responses. sure all the terms of the frequency response function matrix.

2 SOUND AND VIBRATION/JANUARY 2001

 One question that always seems to arise is whether or not it
0
is necessary to measure all of the possible input-output com-

Magnitude, dB
binations and why is it possible to obtain mode shapes from
only one row or column of the frequency response function
matrix.
−80
Why is only one row or column 0

Magnitude, dB
of the FRF matrix needed?
a
It is very important for us to understand how we arrive at
mode shapes from the various measurements that are available
in the frequency response function matrix. Without getting −80
0
mathematical, let’s discuss this.

Magnitude, dB
Let’s just take a look at the third row of the frequency re-
sponse function matrix and concentrate on the first mode. If I
look at the peak amplitude of the imaginary part of the fre-
quency response function, I can easily see that the first mode
shape for mode 1 can be seen (Figure 9a). So it seems fairly −80

straightforward to extract the mode shape from measured data. 180˚

A quick and dirty approach is just to measure the peak ampli-

Phase
tude of the frequency response function for a number of dif-
ferent measurement points.
Now look at the second row of the frequency response func-
−180˚
tion matrix and concentrate on the first mode (Figure 9b). If I 180˚
look at the peak amplitude of the imaginary part of the fre-
quency response function, I can easily see that the first mode

Phase
shape for mode 1 can be seen from this row also. b
We could also look at the first row of the frequency response
function matrix and see the same shape. This is a very simple −180˚
pictorial representation of what the theory indicates. We can 180˚
use any row to describe the mode shape of the system. So it is
Phase
very obvious that the measurements contain information per-
taining to the mode shapes of the system.
Let’s now take a look at the third row again and concentrate
−180˚
on mode 2 (Figure 9c). Again if I look at the peak amplitude of
0.5
the imaginary part of the frequency response function, I can
easily see the second mode shape for mode 2.
Real

And if I look at the second row of the frequency response

function matrix and concentrate on the second mode, I will be
a little surprised because there is no amplitude for the second −0.5
mode (Figure 9d). I wasn’t expecting this but if we look at the 0.5
mode shape for the second mode then we can quickly see that
Real

this is a node point for mode 2. The reference point is located c

at the node of the mode.
So this points out one very important aspect of modal analy- −0.5
sis and experimental measurements. The reference point can- 0.5
not be located at the node of a mode otherwise that mode will
Real

not be seen in the frequency response function measurements

and the mode cannot be obtained.
Now we have only used 3 measurement points to describe
−0.5
the modes for this simple beam. If we add more input-output
measurement locations then the mode shapes can be seen more 1
clearly as shown in Figure 10. The figure shows 15 measured
Imaginary

frequency response functions and the 3 measurement points

used in the discussion above are highlighted. This figure shows
the 15 frequency response functions in a waterfall style plot.
−1
Using this type of plot, it is much easier to see that the mode 1
shapes can be determined by looking at the peaks of the imagi-
Imaginary

nary part of the frequency response function.

Now the measurements that we have discussed thus far have d
been obtained from an impact testing consideration. What if
the measured frequency response functions come from a shaker −1
test? 1
Imaginary

What’s the difference between a

shaker test and an impact test?
−1
From a theoretical standpoint, it doesn’t matter whether the 0 40 0 40 0 40
measured frequency response functions come from a shaker Frequency, Hz Frequency, Hz Frequency, Hz

test or an impact test. Figures 11a and 11b show the measure-
ments that are obtained from an impact test and a shaker test. Figure 7. Response measurements of a 3 DOF model of a beam: a) mag-
An impact test generally results in measuring one of the rows nitudes; b) phases; c) real components; and d) imaginary components.

SOUND AND VIBRATION/JANUARY 2001 3

 a

1 2 3

1 b

a 2

3 c

h33 d

? ?
1 2 3

Figure 9. a) Mode 1 from third row of FRF matrix. b) Mode 1 from sec-
ond row of FRF matrix. c) Mode 2 from third row of FRF matrix. d) Mode
2 from second row of FRF matrix.
1

Mode #1
b 2 Mode #2
Mode #3

3
DOF #1

DOF #2

DOF #3

h32

1 2 3 Figure 10. Waterfall plot of beam frequency response functions.

of the frequency response function matrix whereas the shaker

1 ces describing the system are square symmetric, then reciproc-
c 2
3
h31
Figure 8. a) Drive point FRFs for reference 3 of the beam model. b&c)
Cross FRFs for reference 3 of the beam model.
test generally results in measuring one of the columns of the
frequency response function matrix. Since the system matri-
ity is true. For the case shown, the third row is exactly the same
as the third column, for instance.
Theoretically, there is no difference between a shaker test
and an impact test. That is, from a theoretical standpoint! If I
can apply pure forces to a structure without any interaction
between the applied force and the structure and I can measure
response with a massless transducer that has no effect on the
structure – then this is true. But what if this is not the case?
Now let’s think about performing the test from a practical
standpoint. The point is that shakers and response transduc-
ers generally do have an effect on the structure during the
modal test. The main item to remember is that the structure
under test is not just the structure for which you would like to
obtain modal data. It is the structure plus everything involved
in the acquisition of the data – the structure suspension, the
mass of the mounted transducers, the potential stiffening ef-
fects of the shaker/stinger arrangement, etc. So while theory
tells me that there shouldn’t be any difference between the
impact test results and the shaker test results, often there will

4 SOUND AND VIBRATION/JANUARY 2001

be differences due to the practical aspects of collecting data.
The most obvious difference will occur from the roving of
accelerometers during a shaker test. The weight of the accel-
1 2 3
erometers may be extremely small relative to the total weight
of the whole structure, but their weight may be quite large rela-
tive to the effective weight of different parts of the structure.
This is accentuated in multi-channel systems where many ac-
1
celerometers are moved around the structure in order to acquire
all the measurements. This can be a problem especially on
lightweight structures. One way to correct this problem is to a 2
mount all of the accelerometers on the structure even though
only a few are measured at a time. Another way is to add
dummy accelerometer masses at locations not being measured; 3
this will eliminate the roving mass effect.
Another difference that can result is due to the shaker/stinger
effects. Basically, the modes of the structure may be affected
by the mass and stiffness effects of the shaker attachment.
While we try to minimize these effects, they may exist. The
purpose of the stinger is to uncouple the effects of the shaker h 33
from the structure. However, on many structures, the effects h 31
of the shaker attachment may be significant. Since an impact
test does not suffer from these problems, different results may
be obtained. So while theory says that there is no difference h 32
between a shaker test and an impact test, there are some very
basic practical aspects that may cause some differences.

What measurements do I actually

make to compute the FRF?

The most important measurement that is needed for experi- b

mental modal analysis is the frequency response function. Very
simply stated, this is the ratio of the output response to the
h 13 3

input excitation force. This measurement is typically acquired

using a dedicated instrument such as an FFT (Fast Fourier
Transform) analyzer or a data acquisition system with software
that performs the FFT.
Let’s briefly discuss some of the basic steps that occur in the
acquisition of data to obtain the FRF. First, there are analog
signals that are obtained from our measuring devices. These
analog signals must be filtered to assure that there is no aliasing
of higher frequencies into the analysis frequency range. This
is usually done through the use of a set of analog filters on the
front-end of the analyzer called anti-aliasing filters. Their func-
tion is to remove any high frequency signals that may exist in
the signal.
The next step is to digitize the analog signal to form a digi-
tal representation of the actual signal. This is done by the ana-
log to digital converter that is called the ADC. Typically this
digitization process will use 10, 12 or 16 bit converters; the
more bits available, the better the resolution possible in the
digitized signal. Some of the major concerns lie in the sampling
and quantization errors that could potentially creep into the
digitized approximation. Sampling rate controls the resolution
in the time and frequency representation of the signals. Quan-
tization is associated with the accuracy of magnitude of the
captured signal. Both sampling and quantization can cause
some errors in the measured data but are not nearly as signifi-
cant and devastating as the worst of all the signal processing
errors – leakage! Leakage occurs from the transformation of
time data to the frequency domain using the Fast Fourier Trans-
form (FFT).
The Fourier Transform process requires that the sampled
data consist of a complete representation of the data for all time
or contain a periodic repetition of the measured data. When
this is satisfied, then the Fourier Transform produces a proper
representation of the data in the frequency domain. However,
when this is not the case, then leakage will cause a serious
distortion of the data in the frequency domain. In order to
minimize the distortion due to leakage, weighting functions
called windows are used to cause the sampled data to appear
to better satisfy the periodicity requirement of the FFT. While
windows greatly reduce the leakage effect, it cannot be com-
1
2
h 23
3
h 33
Figure 11. a) Roving impact test scenario. b) Roving response scenario.
pletely removed.
Once the data are sampled, then the FFT is computed to form
linear spectra of the input excitation and output response.
Typically, averaging is performed on power spectra obtained
from the linear spectra. The main averaged spectra computed
are the input power spectrum, the output power spectrum and
the cross spectrum between the output and input signals.
These functions are averaged and used to compute two im-
portant functions that are used for modal data acquisition – the
frequency response function (FRF) and the coherence. The
coherence function is used as a data quality assessment tool
which identifies how much of the output signal is related to
the measured input signal. The FRF contains information re-
garding the system frequency and damping and a collection of
FRFs contain information regarding the mode shape of the
system at the measured locations. This is the most important
measurement related to experimental modal analysis. An over-
view of these steps described is shown in Figure 12.
Of course, there are many important aspects of measurement
acquisition, averaging techniques to reduce noise and so on,
that are beyond the scope of this presentation. Any good refer-

SOUND AND VIBRATION/JANUARY 2001 5

 ence on digital signal processing will provide assistance in this
Analog Signals area. Now the input excitation needs to be discussed next.
Basically, there are two commonly used types of excitation for
Input Output
experimental modal analysis – impact excitation and shaker
Antialiasing Filters excitation.
Now let’s consider some of the testing considerations when
performing an impact test.

Autorange Analyzer
ADC Digitizes Signals What are the biggest things to
think about when impact testing?
Input Output
There are many important considerations when performing
Apply Windows impact testing. Only two of the most critical items will be
mentioned here; a detailed explanation of all the aspects per-
Input Output taining to impact testing is far beyond the scope of this article.
Compute FFT First, the selection of the hammer tip can have a significant
Linear Spectra effect on the measurement acquired. The input excitation fre-
Linear Linear quency range is controlled mainly by the hardness of the tip
Input Output
Spectrum selected. The harder the tip, the wider the frequency range that
Spectrum
is excited by the excitation force. The tip needs to be selected
Averaging of Samples
such that all the modes of interest are excited by the impact
force over the frequency range to be considered. If too soft a
Computation of Averaged
Input/Output/Cross Power Spectra tip is selected, then all the modes will not be excited ad-
equately in order to obtain a good measurement as seen in Fig-
Input Cross Output ure 13a. The input power spectrum does not excite all of the
Power Power Power frequency ranges shown as evidenced by the rolloff of the
Spectrum Spectrum Spectrum
power spectrum; the coherence is also seen to deteriorate as
well as the frequency response function over the second half
Computation of FRF and Coherence
of the frequency range.
Typically, we strive to have a fairly good and relatively flat
input excitation forcing function as seen in Figure 13b. The fre-
quency response function is measured much better as evi-
Frequency Response Function Coherence Function
denced by the much improved coherence function. When per-
forming impact testing, care must be exercised to select the
Figure 12. Anatomy of an FFT analyzer. proper tip so that all the modes are well excited and a good
frequency response measurement is obtained.
a 40 The second most important aspect of impact testing relates
Coherence to the use of an exponential window for the response trans-
ducer. Generally for lightly damped structures, the response
Magnitude, dB

of the structure due to the impact excitation will not die down
FRF to zero by the end of the sample interval. When this is the case,
the transformed data will suffer significantly from a digital
Input Power Spectrum
signal processing effect referred to as leakage.
−60
In order to minimize leakage, a weighting function referred
0 Frequency, Hz 800 to as a window is applied to the measured data. This window
b is used to force the data to better satisfy the periodicity require-
40
Coherence ments of the Fourier transform process, thereby minimizing the
FRF distortion effects of leakage. For impact excitation, the most
Magnitude, dB

common window used on the response transducer measure-

Input Power Spectrum
ment is the exponentially decaying window. The implemen-
tation of the window to minimize leakage is shown in Figure
14.
Windows cause some distortion of the data themselves and
−60 should be avoided whenever possible. For impact measure-
0 Frequency, Hz 200
ments, two possible items to always consider are the selection
Figure 13. a) Hammer tip not sufficient to excite all modes. b) Hammer of a narrower bandwidth for measurements and to increase the
tip adequate to excite all modes. number of spectral lines of resolution. Both of these signal
processing parameters have the effect of increasing the amount
of time required to acquire a measurement. These will both
tend to reduce the need for the use of an exponential window
Actual Time Signal and should always be considered to reduce the effects of leak-
age.
Sampled Signal Now let’s consider some of the testing considerations when
performing a shaker test.

Window Weighting What are the biggest things to

think about when shaker testing?
Windowed Time Signal
Again, there are many important items to consider when
performing shaker testing but the most important of those cen-
ter around the effects of excitation signals that minimize the
Figure 14. Exponential window to minimize leakage effects. need for windows or eliminate the need for windows alto-

6 SOUND AND VIBRATION/JANUARY 2001

gether. There are many other considerations when performing
shaker testing but a detailed explanation of all of these is far Autoranging Averaging with Window
beyond the scope of this presentation.
One of the more common excitation techniques still used
today is random excitation due to its ease of implementation.
However, due to the nature of this excitation signal, leakage is
a critical concern and the use of a Hanning window is com-
monly employed. This leakage effect is serious and causes dis-
tortion of the measured frequency response function even when
windows are used. A typical random excitation signal with a 1 2 3 4
Hanning window is shown in Figure 15. As seen in the figure,
the Hanning window weighting function helps to make the
sampled signal appear to better satisfy the periodicity require-
ment of the FFT process, thereby minimizing the potential
distortion effects of leakage. While this improves the distor-
tion of the FRF due to leakage, the window will never totally
Figure 15. Shaker testing – excitation considerations using random
remove these effects; the measurements will still contain some excitation with Hanning window.
distortion effects due to leakage.
Two very common excitation signals widely used today are Averaging
Autoranging
burst random and sine chirp. Both of these excitations have a
special characteristic that do not require the need for windows
to be applied to the data since the signals are inherently leak-
age free in almost all testing situations. These excitations are
relatively simple to employ and are commonly found on most
signal analyzers available today. These two signals are shown
1 2 3 4
schematically in Figures 16 and 17.
In the case of burst random, the periodicity requirement of
the FFT process is satisfied due to the fact that the entire tran-
sient excitation and response are captured in one sample in- Figure 16. Burst random excitation without a window.
terval of the FFT. For the sine chirp excitation, the repetition
of the signal over the sample interval satisfies the periodicity Autoranging Averaging
requirement of the FFT process. While other excitation signals
also exist, these are the most common excitation signals used
in modal testing today.
So now we have a better idea how to make some measure-
ments.

Tell me more about windows.

They seem pretty important! 1 2 3 4

Windows are, in many measurement situations, a necessary

evil. While I would rather not have to use any windows at all,
the alternative of leakage is definitely not acceptable either. As
discussed above, there is a variety of excitation methods that
can be successfully employed which will provide leakage free
measurements and do not require the use of any window. How-
ever, there are many times, especially when performing field
testing and collecting operating data, where the use of windows
is necessary. So what are the most common windows typically
used.
Basically, in a nutshell, the most common widows used to-
day are the Uniform, Hanning, Flat Top and Force/Exponen-
tial windows. Rather than detail all the windows, let’s just sim-
ply state when each is used for experimental modal testing.
The Uniform Window (which is also referred to as the Rect-
angular Window, Boxcar or No Window) is basically a unity
gain weighting function that is applied to all the digitized data
points in one sample or record of data. This window is applied
to data where the entire signal is captured in one sample or
record of data or when the data are guaranteed to satisfy the
periodicity requirement of the FFT process. This window can
be used for impact testing where the input and response sig-
nals are totally observed in one sample of collected data. This
window is also used when performing shaker excitation tests
with signals such as burst random, sine chirp, pseudo-random
and digital stepped sine; all of these signals generally satisfy
the periodicity requirement of the FFT process.
The Hanning window is basically a cosine shaped weight-
ing function (bell shaped) that forces the beginning and end
of the sample interval to be heavily weighted to zero. This is
useful for signals that generally do not satisfy the periodicity
requirement of the FFT process. Random excitations and gen-
Figure 17. Sine chirp excitation without a window.
2
h61
Input
1 4
6
3
Output
hout,in
5
Figure 18. Input-output measurement locations.
eral field signals usually fall into this category and require the
use of a window such as the Hanning window.
The Flat Top window is most useful for sinusoidal signals
that do not satisfy the periodicity requirement of the FFT pro-
cess. Most often this window is used for calibration purposes
more than anything else in experimental modal analysis.
The force and exponential windows are typically used when
performing impact excitation for acquiring FRFs. Basically, the
force window is a unity gain window that acts over a portion
of the sample interval where the impulsive excitation occurs.
The exponential window is used when the response signal does
not die out within the sample interval. The exponential win-
dow is applied to force the response to better satisfy the peri-
odicity requirement of the FFT process.

SOUND AND VIBRATION/JANUARY 2001 7

 Mode 1

2
4 aij1
1
ζ1
3 aij2 ζ2
6
aij3 ζ3
5

ω1 ω2 ω3

Figure 21. Breakdown of a frequency response function.

Figure 19. Plate mode shapes for mode 1– peak pick of FRF.
MDOF

SDOF
Mode 2

2
1 4 Figure 22. Curvefitting different bands using different methods.

3 6 How Many Points???

Residual
Effects
Residual
Figure 20. Plate mode shapes for mode 2 – peak pick of FRF. Effects

Each of the windows has an effect on the frequency repre-

sentation of the data. In general, the windows will cause a
degradation in the accuracy of the peak amplitude of the func-
tion and will appear to have more damping than what really How Many Modes???
exists in the actual measurement. While these errors are not
totally desirable, they are far more acceptable than the signifi-
cant distortion that can result from leakage.
Figure 23. Curvefitting a typical FRF.
So how do I get mode shapes
from the plate FRFs?

So now that we have discussed various aspects of acquiring

measurements, let’s go back to the plate structure previously
discussed and take several measurements on the structure. Input Time Force Output Time Response
Let’s consider 6 measurement locations on the plate. Now there f(t) y(t)
are 6 possible places where forces can be applied to the plate
FFT
and 6 possible places where we can measure the response on IFT
the plate. This means that there are a total of 36 possible input
output measurements that could be made. The frequency re-
sponse function describes how the force applied to the plate
causes the plate to respond. If we applied a force to point 1 and
measured the response at point 6, then the transfer relation
between 1 and 6 describes how the system will behave (Figure Input Spectrum Frequency Response Function Output Response

18). f(jω) h(jω) y(jω)

While the technique shown in Figures 19 and 20 is adequate
for very simple structures, mathematical algorithms are typi-
Figure 24. Schematic overviewing the input-output structural response
cally used to estimate the modal characteristics from measured
problem.
data. The modal parameter estimation phase, which is often
referred to as curvefitting, is implemented using computer quency, damping and mode shapes – the dynamic characteris-
software to simplify the extraction process. The basic param- tics. The measured FRF is basically broken down into many
eters that are extracted from the measurements are the fre- single DOF systems as shown in Figure 21.

8 SOUND AND VIBRATION/JANUARY 2001

Figure 25. Measured operating displacements.
These curvefitting techniques use a variety of different meth-
ods to extract data. Some techniques employ time domain data
while others use frequency domain data. The most common
methods employ multiple mode analytical models but at times
very simple single mode methods will produce reasonably good
results for most engineering analyses (Figure 22). Basically, all
of the estimation algorithms attempt to break down measure-
ments into the principal components that make up the mea-
sured data – namely the frequency, damping and mode shapes.
The key inputs that the analyst must specify are the band
over which data are extracted, the number of modes contained
in these data and the inclusion of residual compensation terms
for the estimation algorithm. This is schematically shown in
Figure 23. Much more could be said concerning the estimation
of modal parameters from measured data, the tools available
for deciphering data and the validation of the extracted model
but a detailed explanation is far beyond the scope of this ar-
ticle.
All structures respond to externally applied forces. But many
times the forces are not known or cannot be measured easily.
We could still measure the response of a structural system even
though the forces may not be measured. So the next question
that is often asked concerns operating data.
What are operating data?
We first need to recognize that the system responds to the
forces that are applied to the system (whether or not I can
measure them). So for explanation purposes, let’s assume for
now that we know what the forces are. While the forces are
actually applied in the time domain, there are some important
mathematical advantages to describing the forces and response
in the frequency domain. For a structure which is exposed to
an arbitrary input excitation, the response can be computed
using the frequency response function multiplied by the input
forcing function. This is very simply shown in Figure 24.
The random excitation shown excites all frequencies. The
most important thing to note is that the frequency response
function acts as a filter on the input force which results in some
output response. The excitation shown causes all the modes
to be activated and therefore, the response is, in general, the
linear superposition of all the modes that are activated by the
input excitation. Now what would happen if the excitation did
not contain all frequencies but rather only excited one particu-
lar frequency (which is normally what we are concerned about
when evaluating most operating conditions).
To illustrate this, let’s use the simple plate that we just dis-
cussed. Let’s assume that there is some operating condition that
exists for the system; a fixed frequency operating unbalance
will be considered to be the excitation. It seems reasonable to
use the same set of accelerometers that were on the plate to
measure the response of the system. If we acquire data, we may
see something that looks like the deformation pattern shown
in Figure 25. Looking at this deformation, it is not very clear
why the structure responds this way or what to do to change
the response. Why does the plate behave in such a complicated
fashion anyway??? This doesn’t appear to be anything like any
of the mode shapes that we measured before.
In order to understand this, let’s take that plate and apply a
simple sinusoidal input at one corner of the plate. For the ex-
SOUND AND VIBRATION/JANUARY 2001
Mode 1 Contribution Mode 2 Contribution
Figure 26. Excitation close to mode 1.
ample here, we are only going to consider the response of the
plate assuming that there are only 2 modes that are activated
by the input excitation. (Of course there are more modes, but
let’s keep it simple to start.) Now we realize that the key to
determining the response is the frequency response function
between the input and output locations. Also, we need to re-
member that when we collect operating data, we don’t mea-
sure the input force on the system and we don’t measure the
system frequency response function – we only measure the
response of the system.
First let’s excite the system with a sinusoid that is right at
the first natural frequency of the plate structure. The response
of the system for one frequency response function is shown in
Figure 26. So even though we excite the system at only one
frequency, we know that the frequency response function is the
filter that determines how the structure will respond. We can
see that the frequency response function is made up of a con-
tribution of both mode 1 and mode 2. We can also see that the
majority of the response, whether it be in the time or frequency
domain, is dominated by mode 1. Now if we were to measure
the response only at that one frequency and measure the re-
sponse at many points on the structure, then the operating
deflection pattern would look very much like mode 1 – but
there is a small contribution due to mode 2. Remember that
with operating data, we never measure the input force or the
frequency response function – we only measure the output
response, so that the deformations that are measured are the
actual response of the structure due to the input excitation –
whatever it may be.
When we measure frequency response functions and esti-
mate modal parameters, we actually determine the contribu-
tion to the total frequency response function solely due to the
effects of mode 1 acting alone, as shown in blue, and mode 2
acting alone, as shown in red, and so on for all the other modes
of the system. Notice that with operating data, we only look at
the response of the structure at one particular frequency –
which is the linear combination of all the modes that contrib-
ute to the total response of the system. So we can now see that
the operating deflection pattern will look very much like the
first mode shape if the excitation primarily excites mode 1.
Now let’s excite the system right at the second natural fre-
quency. Figure 27 shows the same information as just discussed
for mode 1. But now we see that we primarily excite the sec-
ond mode of the system. Again, we must realize that the re-
sponse looks like mode 2 – but there is a small contribution
due to mode 1.
But what happens when we excite the system away from a
resonant frequency? Let’s excite the system at a frequency mid-
way between mode 1 and mode 2. Now here is where we see
the real difference between modal data and operating data.
Figure 28 shows the deformation shape of the structure. At first
9
 Experimental Finite
Modal Element
Testing Modeling

Modal Perform
Parameter Eigen
Estimation Solution

Develop
Modal Mass Rib Stiffener
Model
Spring

Repeat
Figure 27. Excitation close to mode 2. Until
Structural
Desired
Changes
Characteristics No
Required
Are
Obtained Done Dashpot
Yes

Use SDM
to Evaluate
Structural
Changes

Figure 30. Schematic of the SDM process.

Frequency Finite
Response Corrections Element
Measurements Model
Figure 28. Excitation somewhere between modes 1 and 2.

Parameter Eigenvalue
Output Spectrum Estimation Solver

y(jω)

Modal Model Modal

Parameters Validation Parameters

Frequency Response Function Synthesis

of a
Dynamic Modal Model

Mass, Damping Structural Forced Real World

Stiffness Changes Dynamics Response Forces
Modification Simulation

Modified Structural
f(jω) Modal Response
Data

Input Spectrum

Figure 31. Overall dynamic modeling process.

Figure 29. Broadband plate excitation.

glance, it appears that the deformation doesn’t look like any-

thing that we recognize. But if we look at the deformation pat-
tern long enough, we can actually see a little bit of first bend-
ing and a little bit of first torsion in the deformation. So the
operating data are primarily some combination of the first and
second mode shapes. (Yes, there will actually be other modes
but primarily mode 1 and 2 will be the major participants in
the response of the system.)
Now, we have discussed all of this by understanding the fre-
quency response function contribution on a mode by mode
basis. When we actually collect operating data, we don’t col-
lect frequency response functions but rather we collect output Mass
Rib Stiffener
spectrums. If we looked at those, it would not have been very
clear as to why the operating data looked like mode shapes. Spring
Figure 29 shows a resulting output spectrum that would be
measured at one location on the plate structure. Now the in-
put applied to the structure is much broader in frequency and
many modes are excited. But, by understanding how each of
the modes contributes to the operating data, it is much easier
to see how the modes all contribute to the total response of the Dashpot
system.
So actually, there is a big difference between operating de- Figure 32. Modal model characteristics.

10 SOUND AND VIBRATION/JANUARY 2001

 how the structure behaves in service. This is extremely useful
information. However, many times the operating shapes are
confusing and do not necessarily provide clear guidance as to
how to solve or correct an operating problem (and modifica-
tion and response tools cannot be utilized on operating data)
as shown in Figure 33.
The best situation exists when both operating data and modal
data are used in conjunction to solve structural dynamic prob-
lems.

Figure 33. Operating data characteristics.

Summary
flections and mode shapes – we can now see that the mode
shapes are summed together in some linear fashion to form the Some simple explanations were used to describe structural
operating deflection patterns. But typically we are interested vibration and the use of some of the available tools for the so-
in the total deformation or response of the system. Why do I lution of structural dynamic problems. This was all achieved
even want to bother to collect modal data? It seems like a lot without the use of any detailed mathematical relationships. In
of work to acquire measurements and extract data. order to better understand more of the details of the data pre-
sented here, a theoretical treatment of this material is neces-
sary.
So what good are modal data?
Related Material
Modal data are extremely useful information that can assist
in the design of almost any structure. The understanding and
visualization of mode shapes is invaluable in the design pro- 1. Lecture Notes, Peter Avitabile, Modal Analysis I & II, University of
Massachusetts Lowell.
cess. It helps to identify areas of weakness in the design or areas
2. Seminar Presentation Notes, Peter Avitabile.
where improvement is needed. The development of a modal 3. The Modal Handbook, A Multimedia Computer Based Training and
model is useful for simulation and design studies. One of these Reference Guide, Dynamic Decisions, Merrimack, NH. For additional
studies is structural dynamic modification. information, contact: info@dynamicdecisions.com.
This is a mathematical process which uses modal data (fre-
quency, damping and mode shapes) to determine the effects of The author can be contacted at: peter_avitabile@uml.edu.
changes in the system characteristics due to physical structural
changes. These calculations can be performed without actually
having to physically modify the actual structure until a suit-
able set of design changes is achieved. A schematic of this is
shown in Figure 30. There is much more that could be dis-
cussed concerning structural dynamic modification but space
limitations restrict this.
In addition to structural dynamic modification studies, other
simulations can be performed such as force response simula-
tion to predict system response due to applied forces. And
another very important aspect of modal testing is the correla-
tion and correction of an analytical model such as a finite ele-
ment model. These are a few of the more important aspects
related to the use of a modal model which are schematically
shown in Figure 31.
And one of the final questions that is often asked is which
test is best to perform.

So should I collect modal data

or operating data?

Of course with tight schedules and budgets, do I really need

to collect both modal data and operating data? This is always
difficult to answer but it is always better to have both when-
ever possible. If only one of the two is available, then many
times some engineering decisions may be made without full
knowledge of the system characteristics. To summarize, let’s
point out the differences between each of the data sets.
Modal data requires that the force is measured in order to
determine the frequency response function and resulting modal
parameters. Only modal data will give the true principal char-
acteristics of the system. In addition, structural dynamic modi-
fications and forced response can only be studied using modal
data (operating data cannot be used for these types of studies).
Also correlation with a finite element model is best performed
using modal data. But of course it needs to be clearly stated
that modal data alone does not identify whether a structure is
adequate for an intended service or application since modal
data are independent of the forces on the system as shown in
Figure 32.
Operating data on the other hand is an actual depiction of

SOUND AND VIBRATION/JANUARY 2001 11

Martin Schleske – Empirical Tools in Contemporary Violin Making: Part I. Analysis of Design, Materials, Varnish, and Normal Modes
Empirical Tools in Contemporary Violin Making:
Part I. Analysis of Design, Materials, Varnish,
and Normal Modes
Martin Schleske
Workshop for Violin Making
Seitzstr.4
D-80538 Munich, Germany
ABSTRACT
The perfection of the violin by the famed old masters is attributed
to a long history of passionate innovative empirical work that has
been replaced in modern times by a devotion to traditional
prototypes. Improvements in violin design and performance during
the 17th and 18th Centuries can be correlated with the development
of progressively more demanding playing styles by composers,
including Monteverdi, Corelli, and Torelli. The successful
empirical approach of the old masters can be applied today using
careful measurements, recordings, and documentation that allow
us to relate specific changes in design to specific change in sound.
Our work has documented the major aspects of instrument design
(arching, thickness graduation pattern, and body outline shape) and
modal characteristics of about 90 fine reference instruments,
including many old Italians. We have also measured the materials
in old Italian instruments, which generally are similar to modern
materials. For example, the density of an entire top plate from a
Stradivari cello is 0.39 g/cm3, which is not unusual. Further, our
measurements indicate that the damping properties and modal
characteristics of old and contemporary fine instruments are of
comparable magnitude.
Tools available to the violin maker include selection of design,
materials, varnish treatments, sound analysis, and modal analysis.
Successive varnish layers can enhance or denigrate material quality.
Making “tonal copies” using modal analysis at many steps of the
working process allows the maker to change the geometry of the
copy to compensate for material differences between the reference
instrument and the copy.
INTRODUCTION
I first realized that the violin has a lot to do with experimental
physics during my training as a violin maker at the Mittenwald
50
school in the 1980s. Not only the physical fundamentals (being
taught by Helmut A. Müller) but alsoNumerous articles on the
subject published in Catgut Newsletters and Journals were of
particular importance to me. At that time my impression was that
many researchers tried to answer questions that no one had raised
before. On the other hand, they had few answers to the practical
questions raised by many violin makers. However, Erik Jansson’s
studies always had a definitive connection to the real matters in
violin making. This characteristic of his work has been maintained
to the present [1]. I am honored to contribute to this issue in
celebration of his 60th birthday.
“. . . the essence of the old approach was
continuous development, a history of
passionate empirical work.”
The following article is meant as a short introduction to some
“empirical tools” that I use in my violin shop. I use modal analysis
and sound analysis programs almost daily in my work as a violin
maker, as a complement to the usual tools: knives, planes, chisels,
saws, and so on. The new tools are used for support in tonal
adjustments of fine instruments as well as in the making of new
concert instruments. Over the years I have collected an extensive
database on design and acoustical properties, based on studies of
fine old instruments as well as evaluations during the making of
new instruments. In this paper, the use of “empirical tools” in
violin making is illustrated by examples that give the flavor of a
tour through my violin shop and laboratory. This paper, Part I,
gives an introduction to design, materials, varnish, and modes.
Part II will cover psychoacoustical analysis, acoustical tools, and
tonal copies.
CASJ Vol.4, No. 5 (Series II), May 2002
 Martin Schleske – Empirical Tools in Contemporary Violin Making: Part I. Analysis of Design, Materials, Varnish, and Normal Modes
TRADITION AND INNOVATION
How was it possible for the masters of the Italian violin making
tradition of the 17th and 18th centuries to produce such perfect
instruments? The culmination of the classical era of violin making
was the result of a living tradition, accumulated over generations.
Improvements in sound were kept while poor innovations were
discarded. One generation carried their experiences over to the
next—long, long history of trial and error. So it was that “centuries
of empirical research led to the maturity of the violin.” In order to
continue the spirit of the classical violin making tradition,
“tradition” must be broken. This is because the essence of the old
approach was continuous development, a history of passionate
empirical work.
“In order to continue the spirit of the classical
violin making tradition, “tradition” must be
broken.”
Probably there was a “major triad” of three interacting components
- composition, musical interpretation, and innovative instrument
making - that lay behind the development and perfection of such
impressive “resonance sculptures” as the master works of the 18th
century. Today we talk about “tradition in violin making” in
contrast to “innovation”. Nothing could be further from the history
of violin making. This history bears witness to passionate creative
power, as illustrated in the following examples:
• Claudio Monteverdi first called for 4th position violin playing
(E6, an octave above the open E-string; Marienvesper, 1610).
This piece demanded greater virtuosity than was required in the
works of other contemporary composers. His hometown was
Cremona, Italy, the famous violin-making town of the Amati
dynasty. It can be assumed that Monteverdi’s demands on play-
ing technique were soon transferred to violin makers and stimu-
lated them to develop an instrument with greater playing capa-
bilities.
• Arcangelo Corelli created his 12 Sonatas for Violin (Sonata a
violine e violone o cimbalo; Rome, 1700), which formed part of
the basic repertoire of violin players as recently as the 19th cen-
tury. Even more important was the year 1680, when he com-
posed the first “concerto grosso.” Now the violin was lifted
above the orchestra and became a solo instrument. For Antonio
Stradivari this must have been a musical revolution. At that
time he abandoned the model and arching style of his master,
Nicolo Amati, and developed his own “long pattern.” This
development was his response to the demands of the new music
style.
• It is easy to imagine how the creative power of Stradivari was
stimulated when Torelli, in 1698, laid the foundations of the
virtuoso violin concerto. Stradivari reacted to this innovation
CASJ Vol.4, No. 5 (Series II), May 2002
with another new model. It can be imagined that his so-called
“golden period”, starting about 1700, was influenced by this
soloistic revolution. Hence, the climax of Italian violin making
was in large measure a consequence of musical evolution. The
first virtuoso composers and performers helped bring the violin
to musical maturity.
• In a similar way, the development of the modern violin bow
through François Xavier Tourte, Paris (1747-1835) was acceler-
ated by the new violin schools led by Leopold Mozart and Josef
Haydn. As new concert styles required new instrument designs,
resulting in improved violins, now the enhanced playing tech-
nique called for new, improved bows.
Modern empirical methods can be used to pick up the “innovation”
trail. Acoustical analyses of characteristics of the sound of violins
by Antonio Stradivari, Guarneri del Gesú, and others make it
possible to establish a direct connection with these masters. In this
way they become contemporary teachers rather than simply
historical prototypes. Scientific methods in violin making include
careful documentation of all stages during the making of an
instrument, including changes in design and wood treatment. Such
a detailed “notebook” of the making process is necessary when
comparing the sound of finished instruments. Only in this way can
specific changes in design be related to specific acoustical features.
RECORDINGS OF SOUND
No scientific method can replace the trained ear, ultimate arbiter
in violin making. Tonal adjustment, an important part of violin
making, can be made more reliable using before and after recordings.
A systematically organized archive of such recordings comprises
an auditory “sound school”, illustrating the effects of possible
adjustments. Over the years we have compiled an extensive database
of sounds based on tests of an assortment of instruments, and tonal
adjustments in our shop (see Appendix A). Before the adjustment of
a fine instrument, we ask the musician to do the following:
1. “Play something so I can hear why you have come here” (i.e.
“What is the problem with the sound or playing properties?”)
2. “Play something so I can hear why you play this particular
instrument” (i.e. “What are your preferences in sound, playing
technique, and interpretation?”)
3. “Play open strings and scales” (i.e., “How even or uneven are
timbre, volume, and response of the instrument?”)
Answers to these questions help us to understand the kind of sound
the musician is looking for. Furthermore, this procedure allows us
to watch the interaction between musician and instrument. The
choice of musical passages, and in particular how they are
interpreted, helps us to answer the following fundamental questions:
What characteristics of the instrument must definitely be retained
in the tonal adjustment? What kind of “resistance,” which sound,
and which register is the musician looking for?
51
Martin Schleske – Empirical Tools in Contemporary Violin Making: Part I. Analysis of Design, Materials, Varnish, and Normal Modes

It must be emphasized that a fine professional musician is a highly

reproducible measurement system. The best method for Figure 1. Device for measuring arching profiles on an X-Y
determining the quality of violins is probably one in which the coordinate table.
musical relevance and reproducibility is highest. It goes without
saying that “the musical relevance” of a violin played by a musician
cannot be surpassed. However, the precision and reproducibility
that a fine professional may demonstrate in playing, for example by
repeating a passage many times in exactly the same way, is probably
little known among scientists (who often have limited contact with
professional musicians).

Example 1. An interesting experiment was made in our shop one

day when two Stradivarius violins were at hand for tonal adjustments
and associated recordings. The same passages were played and
recorded, alternating between the two instruments. Without the
musician’s knowledge, some of the recordings (Mozart’s Violin
Concerto in D-major, 1st movement) were later edited to reproduce
violin 1 on the left channel, simultaneously with violin 2 on the
right channel. The match of tempo, phrasing, articulation and
intonation is so close that for a long time a listener has the impression Figure 2. Longitudinal- and cross-arching profiles of top plates of
of hearing a single violin [Note: this recording is available for fine Italian violins by Carlo Bergonzi (17xx; date uncertain), Antonio
listening at www.schleske.de]. Stradivari (1727), and Domenico Montagnana (1729). Measurement
done using X-Y coordinate table and transducer (See Fig. 1).
ANALYSIS OF DESIGN Accuracy: +/- 0.1mm.
As stated above, the central empirical task for the violin maker is to
relate the sound of the instrument to its design. In particular, the
violin maker must learn to identify and judge the musical result of
each change in the design. The major aspects in the analysis of the
design are the arching, thickness graduation pattern, and the body
outline shape.

Arching
In our shop, the determination of an instrument’s arching is made
by measurements of six cross-sectional profiles of the plates
(maximum width of lower bout, lower corners, bridge position,
minimum width of C-bouts, upper corners, and maximum width of
upper bouts). In addition, longitudinal profiles are measured. All
profiles are obtained by digital distance measurements using an
X-Y coordinate table (Fig. 1). The advantages of this method are: (a)
The true shapes of profiles are measured and reproduced with high
accuracy; (b) The profiles can be reproduced with distorted
proportions, for example making the body short and fat by reducing Montagnana has a fuller character. The longitudinal arching of
the length coordinates more than the height coordinates. In this Stradivari has clearly been deformed (“pressed out”) due to the
way characteristic features of an instrument’s shape (including unusual thinness of the top plate. A depression in the top plate in
deformation) can be emphasized; (c) Comparisons between the bridge region can also be seen. Note that the calibration of the
instruments can easily be made by plotting two or more profiles on zero line for each profile is based on the level of the slightly irregular
the same diagram. glue-filled surface of the ribs at the corresponding measuring
positions. This is why the maximum arch heights in the cross and
Example 2. Cross and longitudinal profiles of three fine old Italian longitudinal profiles may differ somewhat.
violins (Antonio Stradivari 1727, Domenico Montagnana 1729, and
Carlo Bergonzi, 18th Century, date uncertain) are shown in Figure Thickness Distribution
2. The profile for the Stradivari is similar to that of his student Measurements of thickness graduation patterns are made using an
Bergonzi, especially in the region near the plate edges, whereas the electronic device developed in our shop. Thicknesses are measured

52 CASJ Vol.4, No. 5 (Series II), May 2002

 Martin Schleske – Empirical Tools in Contemporary Violin Making: Part I. Analysis of Design, Materials, Varnish, and Normal Modes

Figure 3. Device (prototype) for measuring the thickness Figure 4. Thickness graduations (in mm) of six violins made by
graduation. The thicknesses are shown on the analog scale and Guarneri del Gesú. Plates are viewed from the outside. Diagrams
processed by a computer to calculate the interpolation maps were created using our interpolation program. Data from
(Instrument: Cello by Domenico Montagnana, 1740). Biddulph [3].

by a magnetic sensor positioned on top of the plate of the assembled

instrument (Fig. 3). The values are monitored on an analog
instrument as well as being fed into a computer. In addition, the
device is equipped with a voltage-controlled signal generator that
produces a pitch which is directly proportional to plate thickness.
A difference of 1/10 mm is converted to a change in pitch of one
semi-tone. Using this equipment it is easy to get an overview of the
thickness pattern and to track contours of constant thickness (i.e.
follow paths giving the same pitch).

As has also been advocated by Jeff Loen [2], we prefer to use

interpolation methods to transform the point-wise thickness
measurements into a map showing regions of equal thickness by a
color or gray code. This gives an immediate visual impression of
the thickness pattern.
Example 3. Generally, the “art” of empirical research work lies in
the ability to distill useful information from a large amount of data.
An example of the gain in information obtained through thickness
maps is shown in Fig. 4. Here the thickness distributions of six
violins by Guarneri del Gesú are displayed in gray-scale shading.
The data have been compiled from the extraordinary book on
Guarneri del Gesú by Biddulph [3], and entered into our computer
system. It is clear that the “Kreisler 1730” follows a different system
than the other five instruments. The top plate becomes thinner
towards the middle, while the back plate is generally thicker with a
rather uniform distribution. The back plates of the other instruments
show a more conventional “concentric” pattern, increasing in
thickness towards the center. A striking similarity in the thickness
distributions of the top plates can be observed for the “Stretton
1729” and the “Violon du Diable 1734”.
The measured thicknesses can also be used for a statistical analysis,
including averaging across several instruments, which makes it
possible to display general differences. Finally, the thickness maps
are used for evaluation of successive steps in the making and to
illustrate different working concepts of old masters.
Example 4. The thickness distributions of a “composite” cello
partly by Stradivari (top plate), partly by J.F. Lott (back plate) are

CASJ Vol.4, No. 5 (Series II), May 2002 53

Martin Schleske – Empirical Tools in Contemporary Violin Making: Part I. Analysis of Design, Materials, Varnish, and Normal Modes
Figure 5. Thickness graduation of a cello: Top plate made by
Stradivari, back plate by J.F. Lott.
shown in Fig. 5. This fine instrument, which possesses outstanding
sound, was brought to our shop for tonal adjustment, including
modifications of bass bar, fingerboard, soundpost, and bridge. The
thickness distribution of the top plate largely follows the body
contour, while that of the back plate is concentric.
Body Outline
The body outline is also measured by electronic determination of
coordinates on the X-Y table. Once again, it is useful to plot several
instruments on the same diagram in order to make direct
comparisons.
Example 5. The outlines of back plates of two violins, Antonio
Stradivari (1712) and Domenico Montagnana (1729), are shown in
Fig. 6 (see also Appendix B).
ANALYSIS OF THE MATERIAL
An established violin making practice, which is often applied when
trying to make a fine instrument, is to rely on “tradition”. This
means that the work is guided by examining an existing, fine
“reference” violin with desired timbral properties. An examination
of the reference limited to the geometrical properties will, however,
never lead to the goal, since it is the geometry in combination with
material properties that determine the vibrational characteristics
(normal modes), and hence the sound of the instrument.
The most relevant material properties for violin making are:
1. Sound velocity in the longitudinal direction of the grain
2. Sound velocity in the cross-grain direction
3. Density
4. Internal damping
5. Orientation of mechanical elements of wood (grain and rays)
54
Figure 6. Outlines of two reference violins: the “Schreiber”
Stradivari of 1712 (black) and Domenico Montagnana, 1729 (white).
Measured using X-Y coordinate table.
The importance of the last point follows from the fact that wood
shows a strong anisotropy with regard to sound velocity and internal
damping [4]. The orientation of grain and rays relative to the arching
determines what fraction of the maximum available sound velocity
will be attained in the two directions of the arching (longitudinal
and cross). In fact, the influence of the local orientation of grain and
rays across the plates is sometimes larger than the influence obtained
by varying the distribution of plate thickness.
Of course, no two pieces of wood are identical. Also, detailed
knowledge of the wood treatment of the reference instrument,
including primer and varnish, is never available. The treatment of
the wood particularly influences damping. The more the material
properties of the reference instrument deviate from those of the
instrument under construction, the larger the deviations from the
geometry of the reference instrument that must be made in order to
retain similar vibrational properties, and hence a similar sound. So,
what strategy should be followed?
• The geometry of the reference instrument must be known as
well as possible, because otherwise it will not be known what
to deviate from. This is the reason for detailed analyses of the
design of reference instruments (described above).
• A valuable preparatory task is to document the key acoustical
parameters of the wood that will be used. By making a series of
similar instruments from wood with varying material
CASJ Vol.4, No. 5 (Series II), May 2002
 Martin Schleske – Empirical Tools in Contemporary Violin Making: Part I. Analysis of Design, Materials, Varnish, and Normal Modes

parameters and treatments it is possible to learn indirectly

Figure 7. Device for measuring density δ [g/cm3] of a cello top-
about the material properties of the reference instrument. The
plate. Numbers are as follows: 1, scale; 2, metal handle; 3, water
hypothesis put forward by Schelleng [5] and Meyer [6], that the
(known density δw); 4, top plate; 5, plastic sack; 6, barrel; 7, weight.
acoustical quality of a piece of wood is determined mainly by
Symbols are as follows: Fb, downward force of device without top;
the ratio between sound velocity and density, certainly seems
Fres, force with top plate inside bag; Fa, upward force of top plate,
to be correct according to our experience.
VK, volume of the top plate.
• It is desirable to obtain the best possible information about the
material properties and wood treatment of the reference in-
strument. By using wood that is as similar as possible for the
new instrument, the deviations from the reference geometry
can be minimized. Among the many material parameters of
the reference instrument, only the following can be determined:

1. The sound velocity along the grain; and

2. The sound velocity across the grain.
(Note: these can only be determined indirectly as mean
values by averaging over the normal modes)
3. Density, including all treatments such as primer and var-
nish, can be measured in free plates (see Example 6).
4. Damping can be measured by determining the bandwidths
of the resonances in free plates.
5. The orientation of the structural elements of the wood
(grains and rays) can be determined, to some extent, by
optical methods.

• Finally, the question must be raised whether wood with acous-

tically comparable material properties is available. In other
words, are there any reasonable ways of deviating from the
reference geometry with the pieces of wood at hand, in order
to reach an acoustically comparable result? This question can
only be answered indirectly and after the new instrument has
been completed, using AB tests of recorded sounds from the
two instruments. If the tests give a positive answer (no signifi-
cant difference in sound) the answer is yes, it was possible. A
psychoacoustical evaluation of the radiated sound is an impor-
tant tool in all analysis and control.
“A psychoacoustical evaluation of the radiated
sound is an important tool in all analysis and
control.“
Some light will be shed on the strategic points mentioned above by
describing experiences from our experimental work. It is often said
that the old Italian master instruments exhibit material properties
that cannot be duplicated today. This could relate to a “secret”
treatment of the wood with particular mixtures of varnish, or to
chemical or other long-term, slow processes that have modified
wood properties.
Density
Example 6. What was the density of spruce used by Stradivari?
Work in our shop gives some answers. The bass bar of the fine cello
mentioned in Example 4 had to be replaced and this gave us an
opportunity to determine the density of the Stradivari top plate
without bass bar.
A very accurate method was used in which the weight of the top
plate was determined when immersed in a liquid (water) of known
density δW (Fig. 7). The top plate was put in a water-tight bag of
synthetic material. A measurement rig was designed consisting of
the plastic bag (5) with a weight (7) at its lower end, and a supporting
frame (2) resting on an electronic balance (1) standing on top of a
barrel (6). The weight was large enough to make the total force on
the completely immersed top plate (in the bag) directed downwards,
despite the buoyancy force Fa from the displaced water.
Equation I, gives Fb as the downward force on the measurement rig
without the top plate, and Fres as the corresponding force with the
top plate inside the bag. By this comparative measurement method,
the error due to the plastic bag is cancelled out. Fb and Fres were
determined from the readings mb and mres on the electronic balance.

CASJ Vol.4, No. 5 (Series II), May 2002 55

Martin Schleske – Empirical Tools in Contemporary Violin Making: Part I. Analysis of Design, Materials, Varnish, and Normal Modes

Fb = mb g and Fres = mres g (Equation I)

Figure 8. Scanning electron micrograph [7] of spruce top-plate of
where g = 9.81 m/s2 is the acceleration of gravity. a F. Gagliano violin (anno 1780) showing pits in tracheid walls
(2000 times magnification).
Further,

Fa = –G´ (Equation II)

where G´ is the so-called apparent weight of a fully immersed body.

Finally,
δΚ = δW * G / (G-G’) (Equation III)
where δK is the unknown density of the top plate, δW the known
density of water, and G the “real” weight of the top plate as measured
in air.

For this particular Stradivari top plate,

ma = Fa/g = (Fb-Fres)/g = (mb - mres) = (1480 - 870) gr = 610 gr

(Note: g denotes acceleration of gravity and gr the mass unit gram).

Table 1. Modal damping values averaged across all modes. “17xx”
This gives G´= –610 gr according to Equation II. The weight in air indicates that decade and year are uncertain.
was m = 390 gr, which gives G = mg. Finally, Substituting the
known G and G´in Equation III gives the density of the varnished
Stradivari plate as Maker Name Date Bass Bar Modal Damping
(Percent)
δΚ = 1.0 . 390 /(390 – (–610)) = 0.39 g/cm3. A. Stradivari 17xx without 1.04
A. Stradivari 17xx with 1.06
The volume of this top plate (Vk) happens to be exactly 1000 cm3 F. Rugeri 1669 with 1.22
(Vk = m/δΚ ). M. Schleske 2001 with 1.07
A density of 0.39 g/cm3 is not unusual, but clearly in the lower
range for spruce available today—even including wood from Figure 9. Damping of eigenmodes as a function of mode frequency
northern Italy. It must be emphasized that this value refers to the (eigenfrequency) of cello top-plates by Stradivari 17xx (with and
combination of wood, primer, and varnish, not to white, untreated without bass bar), Rugeri 1669, and Schleske 2001. Note that the
wood. The densities of the primer and varnish are certainly higher damping of spruce in the modern top plate is within the range of the
than that of the wood itself. Further, these measurements say nothing old wood.
about the original density at the time the instrument was made,
which may have differed from the present. The measured density
tells us only something about how this top plate compares to
contemporary top plates.

Anatomy of Wood
Many rumors exist in the violin making world regarding historical
methods for reducing the density of wood. A common hypothesis
is that the membranes of the bordered pits in the tracheid cells were
dissolved by attacks of microbes while the wood was immersed in
a river or lagoon. The water could then penetrate the cells. Our
studies, based on electron microscopy of wood obtained in repair
work from violins by old Italian masters Joseph Guarneri filius
Andreas, Domenico Montagnana, and F. Gagliano found no such

56 CASJ Vol.4, No. 5 (Series II), May 2002

 Martin Schleske – Empirical Tools in Contemporary Violin Making: Part I. Analysis of Design, Materials, Varnish, and Normal Modes

evidence. Similar microscope investigations of spruce fragments,
conducted by Barlow and Woodhouse [21, 22] on a range of old
instruments came to the same conclusion.
Example 7. Tracheid cell walls and bordered pits in a piece of wood
from the top plate of a violin by F. Gagliano (ca. 1780); Fig. 8. The
pits (diameter 0.012 mm) are clearly seen. They are not dissolved.
Damping of Resonances
An acoustically important material property is the internal friction,
which, in combination with radiation losses, determines the 3 dB
bandwidths of the resonance peaks. Does the internal friction of the
normal modes of fine old instruments with desired sound differ
from that of modern instruments? This question can be answered
by comparing the 3 dB bandwidths of pairs of similar modes (which
ought to have comparable radiation losses). Such differences have,
if they exist, been assumed to be caused by some initial processing
of the wood, a special composition of primer and varnish, or long-
term aging processes. Comparisons of transfer functions used for
modal analyses of assembled instruments as well as free plates give
the following examples:
Example 8. Modal damping (half the 3 dB bandwidth normalized
to the mode frequency) vs. normal mode frequencies was measured
for two old cello top plates (A. Stradivari 17xx; F. Rugeri 1669) and
a new instrument (M. Schleske 2001) (Table 1; Fig. 9). The plates
were freely supported. In the frequency range analyzed here, no
mysterious or anomalous damping values can be observed for these
300 year old cello top plates.
Example 9. Analyses of finished instruments also reveal no major
differences in damping between fine old instruments and
contemporary instruments. A comparison of the damping values of
three violins by Antonio Stradivari and three new violins built to
his model (Table 2; Fig. 10), shows that the modal damping factors
of the new instruments occupy the same range as those of the old
instruments. The individual regression lines for the six instruments
show the highest damping values for the Stradivari from 1712, and
the lowest for the Stradivari from 1721. Regression lines (shown in
white) for the new instruments are found in the range between these
extremes. A decrease in the damping values towards higher
frequencies, common to all instruments, is clearly seen. The
decrease is about a factor 0.7 per decade. This decrease in damping
with increasing frequency, which is reflected in sharper resonance
peaks, is a highly desirable property. In combination with the
increasing resonance density at higher frequencies this sharpening
of the resonances contributes to the “modulability” (sound control
by vibrato) of the instrument. In summary, our measurements
indicate that the damping properties of old and contemporary
instruments are of comparable magnitude.

Table 2. Mean damping values obtained by averaging across all THE “MUSICAL” VIOLIN VARNISH
modes of some old and new instruments. Note: “x” indicates In my opinion, the target of treating a violin with varnish is to
uncertain date. See Appendix C for a description of measurements. achieve low damping in the finished instrument, ideally a reduction

Average
Maker Name Date Damping Value Figure 10. Damping of eigenmodes of vibration in assembled
(Percent) violins: Three instruments by Antonio Stradivari are compared
with three contemporary violins. Note that damping values are
G. Grancino 1699 1.4 comparable. The regression lines of the new instruments (white)
are between those of the “Schreiber” Stradivari (1712) and the
J. Guarneri (fil.Andrea) 1706 1.3
“Hamma” Stradivari (1721).
A. Stradivari 1712 1.5
A. Stradivari 1721 1.1
A. Stradivari 1727 1.5
C. Bergonzi 17xx 1.3
D. Montagnana 1729 1.3
J. Guarneri del Gesu 1733 1.2
J.B. Vuillaume 18xx 1.3
S. Zygmuntowicz 1996 1.2
M. Schleske 1999 (37) 1.2
M. Schleske 2000 (39) 1.3
M. Schleske 2000 (40) 1.4
M. Schleske 2000 (44) 1.2
M. Schleske 2000 (45) 1.4
M. Schleske 2001 (51) 1.3

CASJ Vol.4, No. 5 (Series II), May 2002 57

Martin Schleske – Empirical Tools in Contemporary Violin Making: Part I. Analysis of Design, Materials, Varnish, and Normal Modes

in damping compared to the untreated wood. A subsidiary goal is

to achieve the highest possible increase in the ratio of sound velocity Figure 11. Step-by-step acoustical evaluation of the varnishing
to density. A reduction of the damping corresponds spectrally to a process. Note that the downward slope of the “material quality”
reduction of bandwidths and an increase in heights of resonance (Mq) line reveals unfavorable steps in the varnishing process.
peaks. As mentioned above this leads to a higher “modulability” of
the sound [for a detailed explanation see Part II]. The second goal,
an increase in the ratio between sound velocity and density, means
an improvement of the “acoustical quality” of the material. A varnish
with that second property will allow the instrument to be made
with slightly reduced thicknesses and hence lower mass load. The
decrease of mode frequency caused by reduced thicknessess will
be compensated by the application of such a favorable varnish. The
mode frequencies will be kept unchanged by the addition of the
varnish. In this way the acoustical efficiency of the instrument is
increased, as well as its dynamical range.

As the mode frequencies as well as the damping values may change

significantly due to the treatment of the wood with primer and
varnish [8, 9, 10], it is a good investment to learn about the influence
of your own treatments. The whole concept of “varnishing” can
include a number of treatments such as applying a filler and(or)
ground layer, followed by many layers of varnish, and it is not only
the final result which is of interest. The acoustics of the
intermediate steps in the varnishing process are equally interesting.
There is a danger that some single step in the process may have a
detrimental influence on the final result, although it may happen
that the loss is partly compensated or masked in a later step in the
varnishing process.
Such undesirable intermediate steps cannot be detected in a single
comparison between untreated and finished pieces. Each individual
step in the treatment of the wood must be evaluated. Since (a) each
treatment needs a certain drying time, and (b) the acoustical effects
of the treatment change with time, it is necessary to conduct an
evaluation of the varnishing process by making tests on a series of
wooden samples (strips). The number of samples must be equal to
the number of steps in the process, plus some untreated reference
samples. The first sample receives only the first treatment, the
second the first and the second treatment, and so on until the last
sample, which undergoes all treatments in the varnishing process.
The samples are allowed to dry for a substantial period of time,
after which the differences in sound velocity c, density δ, and
damping (quality factor Q) are determined and compared to the
original reference samples. The resonance properties are measured
using the standard method involving driving the free-free supported
sample strip and observing the first bending mode [10]. The change
in material quality Mq = c/δ due to the treatment is calculated for
each sample. These changes compared to the original, untreated
sample may be plotted vs. the successive steps in the varnishing
process.
perpendicular to the longitudinal direction. [Note: All values have
been normalized relative to the untreated reference values because
there is a possibility that the treated samples as well as the untreated
reference samples were influenced by changes in climate during
the treatment period. This way the effects of variations in humidity
and temperature are compensated].
Figure 11 shows that the varnishing system analyzed here has only
a limited positive influence. The two first treatments (ground) that
increase the sound velocity by 7.4 % and the density by only 2.7 %
give a noticeable increase (4.6%) in material quality. The following
step #4 (treatment by a layer of rosin oil) gives, however, a relatively
higher increase in density, and the material quality drops. In this
step some of the positive influence of the grounding in steps #1 and
#2 is lost. The same effect is found for the following treatments with
a fat oil varnish. After the final treatment (step #6) the material
quality is similar to that of the original wood. The sound velocity
has increased by 11.8% and the density by 11.4 %. It would have
been better if the initial trend could have been kept, with a faster
increase in sound velocity than in density. In that case, it would
have been possible to make the violin with slightly thinner plates in
order to reduce the mass load, and thereby increase the acoustical
efficiency. This way, the shift in mode frequencies and associated
change in sound that follows with thinner plates would have been
compensated in the varnishing process.

Example 10. Figure 11 shows successive steps in the treatment of The varnishing system we use in our shop consistently gives an
spruce strips with thickness of 3.0 mm, and grain oriented increase in mode frequencies of about 6 % compared to the untreated

58 CASJ Vol.4, No. 5 (Series II), May 2002

 Martin Schleske – Empirical Tools in Contemporary Violin Making: Part I. Analysis of Design, Materials, Varnish, and Normal Modes

Figure 12. Modal analysis of a violin showing impact hammer for
measuring excitation force and accelerometer (at the upper bridge
corner) for measuring response.
vibrations in the violin, and hence the sources of the radiated sound
of the instrument. Viewed in this manner the modes are the obvious
link between design, making and sound.
An Intuitive Method for Diagnosis
Modal analysis fosters an intuitive understanding of the function of
the violin, as the modes can be displayed animated on the computer
screen. The violinmaker can then watch how the violin’s corpus
bends, pumps, and twists. He does not only hear but suddenly begins
to see what will happen to the instrument when the musician excites
it. He thus learns to understand the function of the instrument, and
begins to develop a deeper understanding of the resonances his
making process produces. Even if the methods might look a little
technical, computer analysis does not tell him what to do. It just
shows him what happens. The art of violinmaking is to create a
certain resonance sculpture. Usually he cannot see resonances, he
can only hear what they are responsible for: the tone of the
instrument. But modal analysis is a method that reveals what he has
created: the resonance sculpture!

That is why my passion as a violinmaker is focused on resonances

and why I find modal analysis and sound analysis to be exciting
tools. They open a new sphere of intuition and experience as one
(white) instrument. In contrast to the disappointing example in begins to understand things that usually cannot be seen. Intuition
Fig. 11, a varnishing process has been reported which reduces the and experience will guide all further decisions on what
damping compared to the untreated wood [10]. modifications of the developing corpus will bring its resonances
closer to what you are aiming for (which may be to match the
MODAL ANALYSIS resonance profile of a fine reference instrument).
We will now turn to some important tools for acoustical analysis.
Methods for visualization of modes like hologram interferometry
[11] were used on violins by Jansson and others as early as 1970 [12]. Table 3. B1 frequencies of 11 old violins (sorted by increasing B1
Recently Bissinger started a comprehensive project in order to create frequency). “x” indicates uncertain decade or year.
a database on the design and acoustics of stringed instruments,
including modal parameters [13]. Modal analysis was first applied B1 Mode
to the violin in 1983 by Müller [14], and later by Marshall [15,16]. In Maker Name (Location) Date Frequency
1989 it was used for the first time during the making of a violin [17]. (Hz)
Still, few violin makers use it as an aid in their work. That is surprising
since modal analysis gives a very intuitive view of the modes of Giambattista Rogeri (Brescia) 16xx 500
vibration, which is the primary function of the instrument. Jean Baptiste Vuillaume (Paris) 1872 506
Antonio Stradivari (Cremona) 1712 513/524*
With the use of modal analysis it is possible to make an acoustical Carlo Bergonzi (Cremona) 17xx 519
“fingerprint” of a violin, showing the normal modes (often called Domenico Montagnana (Venice) 1729 519
“resonances”). The modes depend upon all the properties of the
Antonio Stradivari (Cremona) 1721 517/527*
vibrating structure, including geometry as well as material. With
Giacomo Rivolta (Milan) 1837 524
other methods, these two sets of properties have to be studied
independently. The particular combination of material properties, Antonio Stradivari (Cremona) 1727 533
arching, thickness distribution, body shape etc. (in other words all Giovanni Grancino (Milan) 1699 541
of the parameters that can be varied in violin making), define certain Joseph Guarneri del Gesú (Cremona) 1733 544
distributions of specific stiffness and mass, which in turn determine Joseph Guarneri fil Andreae (Cremona) 1706 565
the modes of the instrument. Modal analysis of mode shapes, mode
frequencies, and damping factors shows how the parameters of the * Split B1 mode due to torsional motion of the fingerboard [details
material and design combine into a single picture. At the same see PartII].
time, the modes are the basic components that determine the

CASJ Vol.4, No. 5 (Series II), May 2002 59

Martin Schleske – Empirical Tools in Contemporary Violin Making: Part I. Analysis of Design, Materials, Varnish, and Normal Modes

Figure 13. The typical lowest eigenmodes of a Stradivarius (1712) violin. In contrast to the high variety of different plate modes, these
four eigenmodes are in general observed on all violins. The mode frequencies and details in the mode shapes differ from instrument to
instrument. They give valuable clues about the acoustical function of the instrument. Together they form the individual “acoustical
fingerprint” of the violin.

Left-hand pictures (a through d): Contour diagrams of the mode-shapes. View from the outside. Gray-black areas vibrate in opposite phase
with gray-white areas. Nodal lines (lines of no amplitude) between black and white ring. Read rings of vibrating zones like height-lines of
mountains on a topographical map. Absolute Amplitude difference between two adjacent black (or two adjacent white) rings: see ∆ Mag.
(in m/Ns^2) above Fig. 13a...d.

Right-hand pictures: Frozen pictures of screen animation. Both maximum displacement and undeformed situation are plotted.

a) A0 Helmholtz resonance: 286 Hz. A mode with strong “breathing” of the whole corpus with very effective radiation of sound due to the
air vibrating in the f-holes. Notice it by playing a chromatic scale on the G string. The region around C# will sound a lot more sonorous
than the other notes. The reason: the fundamental of C# corresponds with the Helmholtz resonance. Note bending of the fingerboard.

b) C2 corpus mode: 409 Hz. A strongly twisting, weakly radiating mode. It is the only mode with almost identical mode-shape patterns on
top and back plates.

c) T1 corpus mode: 448 Hz. The lower of the two strong corpus modes. Strong radiation due to the pumping movement of the back plate
and the strong vibration area in the lower flank of the left top plate side. Note the strong (in plate) vibrations near the ribs at the bass
bar side: this is due to the strong maximum vibration of the back plate at the edges of the C-bout (bass bar side).

d) B1 corpus-mode: 524 Hz. Large vibration zone in phase on the top-plate due to the function of the bass bar. Back plate edges vibrate in
opposite phase to center of back plate, performing strong bending in cross direction. Strongly radiating mode.

60 CASJ Vol.4, No. 5 (Series II), May 2002

 Martin Schleske – Empirical Tools in Contemporary Violin Making: Part I. Analysis of Design, Materials, Varnish, and Normal Modes

Figure 14. An extensive acoustical protocol of a Stradivari violin (1712). The frequency functions are:

a) Black curve: Energetic averaged “Resonance profile” of sound radiation p/F (p = sound pressure; F = excitation force). Gray curve:
Energetic averaged mobility of both bridge feet (admittance v/F). Data points: Damping values as function of eigenfrequency (right
vertical axis).

b) The first 25 eigenmodes of vibration of the Stradivarius. Measured by experimental modal analysis. Mode frequency
(Eigenfrequency)values appear below each mode shape. White-gray areas vibrate in opposite phase with black-gray areas. See text for
details.

CASJ Vol.4, No. 5 (Series II), May 2002 61

Martin Schleske – Empirical Tools in Contemporary Violin Making: Part I. Analysis of Design, Materials, Varnish, and Normal Modes
Eavesdrop on Stradivari
Of the almost 90 instruments that we have examined by modal
analysis in our shop during the last few years, a “shining star” is the
“Schreiber” Stradivari of 1712 (Fig. 11), an outstanding example of
violin making at its highest level (for details of the measurement
method see Appendix C).
Example 11. The four most important low modes of this Stradivari
are shown in Fig. 13. Modes with these shapes are observed in
almost all violins. In contrast, the high-frequency plate modes
starting at about 650 Hz are very different between instruments.
We regard the B1 mode as a kind of “leading mode” for the tonal
color of the instrument; it acts as a “tonal barometer.” A B1 frequency
below 510 Hz is characteristic of a somewhat “soft” violin with
dark sound, lacking “resistance.” In contrast, a B1 frequency above
550 Hz is found in “stubborn” violins with bright sound, possibly
with a tendency to harshness, and with strong “resistance” to the
player. Hutchins has observed this effect and stressed the
importance of the frequency difference between the B1 and A1
modes [18]. The frequency of the A1 mode - which is a “longitudinal”
air mode with a single nodal line at half the body length [19, 20] -
varies very little around a value just below 500 Hz due to the almost
standardized length of the violin. The frequencies of the B1 mode
are, however, widely scattered in different violins (Table 3).
A detailed illustration of the acoustical properties of the Stradivari
violin from 1712 is shown in Fig. 14. The frequency responses show
the “resonance profile” of the radiated sound (black curve),
calculated as the RMS average of the normalized sound radiation
(sound pressure divided by excitation force) at different angles in
the room. The other frequency response (gray) is the quadratic
average of two admittance measurements, in which the excitation
force has been applied at the upper corner of the bridge (G-string
side) in the bowing direction. The detection of the resulting motion
was made at the two bridge feet.
It is clear that the admittance does not give a very good prediction
of the radiated sound from the instrument, although it is better
above about 1300 Hz. The reason is the increase in the ratio between
the bending wavelength in the plate and the wavelength of the air-
borne sound, which improves the sound radiation efficiency. The
damping of the modes is plotted with data points at the
corresponding mode frequencies. The lower part of Figure 14b
shows the mode shapes of the first 25 modes (up to about 1320 Hz).
The typical difference between body resonances and plate
resonances is clearly seen, among other things. In the body
resonances the top and back plates move as a single unit with strong
coupling via the ribs. The nodal lines run across the ribs from one
plate to the other, which means that the body bends and twists as a
homogenous body. This property is seen for all modes up to B1
(524 Hz).
62
The plate resonances, in contrast, are characterized by smaller
“islands of vibration” separated from each other, more numerous
the higher the frequency. The nodal lines run in many cases parallel
to the contour of the body. In these modes the violin behaves like a
membrane divided in many parts, vibrating in alternating phases.
The typical characteristics of plate resonances start with the mode
at 769 Hz. It has an efficient, asymmetric vibrational shape with an
antinode in the lower right part of the bottom plate. It is easily
identified as a peak in the resonance profile of the sound radiation,
and provides a substantial part of the sound radiation. Body and
plate resonances are always clearly separated by a border line at
about 700 Hz in the frequency responses of the sound radiation as
well as in the admittance. In the frequency range up to 2340 Hz, 48
modes of this particular Stradivarius violin were identified by modal
analysis (see Appendix C for details about measurements).
A collection of complete illustrations for different instruments is
very useful for making acoustical relationships between two
instruments visible. When making “tonal copies” the use of modal
analysis at many steps of the working process allows the maker to
compensate for differences between the reference instrument and
the copy caused by differences in the material, by making changes
in the geometry. As the modal analysis shows the vibration shape
of each single mode, it is possible to identify the “sensitive zones”
for each mode. In these zones, a modification of the plates or other
parts of the instrument gives maximal shift in the modal frequencies
and damping. In this way, modal analysis can be used to support
the “design of violin sound”, when striving towards a specific goal.
(To be continued in the next issue of the CAS Journal. “Empirical
tools in contemporary violin making: Part II. Psychoacoustical
analysis and use of acoustical tools)
ACKNOWLEDGEMENTS
Our “empirical approach” received “obstetrics” by the mentorship of
Helmut A. Müller and Paul Geissler who during a research project
(1988–89) with Müller-BBM had introduced me to the use of modal
analysis and by a fund of the Bavarian Government
(Innovationsberatung des Bayerischen Wirtschaftsministerium) in 1996
which helped to found my laboratory on violin acoutics. We would like
to thank Anders Askenfelt for translating the paper from German to
English and Jim Woodhouse and Jeff Loen who provided editiorial
assistance.
REFERENCES
[1] Jansson E. [in press], Acoustics for violin and guitar makers,
Kungl Tekniska Högskolan, Stockholm.
[2] Loen J.S., 2001, Thickness graduation systems of violin family
instruments: Preliminary statistics and conclusions, J. Catgut
Acoust. Soc., vol. 4, no. 4 (Series II), p. 42-44.
CASJ Vol.4, No. 5 (Series II), May 2002
 Martin Schleske – Empirical Tools in Contemporary Violin Making: Part I. Analysis of Design, Materials, Varnish, and Normal Modes
[3] Biddulph, P., Chiesa, C., Dilworth, J., Hargrave, R., Klein, P.,
Pollens, S., Rosengard, D., and Wen, E., 1998, Guiseppe
Guarneri del Gesú: Peter Biddulph, London, 2 volumes, 340 p.
[4] Schleske, M., 1990, Speed of sound and damping of spruce in
relation to the direction of grains and rays: J. Catgut Acoust.
Soc., vol. 1, no. 6 (Series II), p. 16-20.
[5] Schelleng, J.C., 1963,The violin as a circuit: J. Acoust. Soc.
Amer. vol. 35, p. 326-338.
[6] Meyer, H.G., 1995, A practical approach to the choice of tone
wood in the instruments of the violin family: J. Catgut Acoust.
Soc., vol. 2, no. 7 (Series II), p. 9-13.
[7] Unpublished electron microscopy undertaken in
collaboration with Olaf Gross (LM-Universität München).
[8] Schelleng, J.C., 1967, On the physical effects of violin varnish,
J. Catgut Acoust. Soc. Nov. 1965, p. 15-18; May 1967, p. 9-14
and Nov 1967, p. 17-24.
[9] Hutchins, M.A., 1991, Effects on spruce test strips of four-year
application on four different sealers plus oil varnish: J. Catgut
Acoust. Soc., vol. 1, no. 7 (Series II), p. 11-16.
[10] Schleske, M., 1998, On the acoustical properties of violin
varnish: J. Catgut Acoust. Soc., vol. 3, no. 6 (Series II), p. 27-43.
[11 ] Saldner, H.O., Molin, N-E., and Jansson E.V., 1996, Vibration
modes of the violin forced via the bridge and action of the
soundpost: J. Acoust. Soc. Amer., vol. 100, p. 1169-1176.
[12] Jansson, E.V., Molin, N.-E., and Sundin, H., 1970, Resonances
of a violin body studied by hologram interferometrie and
Acoustical Methods: Physica Scripta, vol. 2, p. 243-256.
[13] Bissinger, G., 1999, Modal Analysis of assembled string
instruments: The Viocadeas Project - A Progress Report. J.
Catgut Acoust. Soc., vol. 3, no. 8 (Series II), p. 19-23.
[14] Müller, H.A., and Geissler, P., 1984, Modal analysis applied
to instruments of the violin family (abstract); J. Catgut Acoust.
Soc., vol. 41, p. 12.
[15] Marshall, K.D., 1986, Modal analysis: A primer on theory and
practice: J. Catgut Acoust. Soc., vol. 46, p. 7-17.
[16] Marshall, K.D., 1985, Modal analysis of a violin: J. Acoust.
Soc. Amer., vol. 77, p. 695-709.
[17] Müller, H.A., Schleske, M., 1989, Is scientific research useful
for violin makers? (abstract): J. Catgut Acoust. Soc., vol. 1, no.
CASJ Vol.4, No. 5 (Series II), May 2002
4 (Series II), p. 48.
[18] Hutchins, C.M., 1989, A measurable controlling factor in the
tone and playing qualities of violins: J. Catgut Acoust. Soc.,
vol. 1, no. 4 (Series II), p.10-15.
[19] Jansson, E.V., 1973, On higher air modes in the violin: J. Catgut
Acoust. Soc., vol. 19, p. 13-15.
[20] Jansson, E.V., 1977, Acoustical properties of complex cavities.
Prediction and measurements of resonance properties of violin-
shaped and guitar-shaped cavities: Acustica, vol. 37, p. 211-
221.
[21] Barlow, C.Y., and Woodhouse, J., 1990, Bordered pits in spruce
from old Italian violins: Journal of Microscopy, vol. 160, p.
203-211.
[22] Barlow, C.Y., and Woodhouse, J., 1991, Was old Italian spruce
soaked? The Strad, vol. 102, no. 1210, p. 234-239.
[23] Moral, J.A., and Jansson, E.V., 1980,: Eigenmodes and the
function of the violin: J. Catgut Acoust. Soc., vol. 34, p. 29-32.
[24] Müller, H.A., Skriptum zum Physikunterricht, (unpublished
physics courses at the Violin Making School, Mittenwald,
Germany.
[25] Cremer, L., 1985, The Physics of the violin: Cambridge, Mass,
MIT Press, 472 p.
[26] Moral, J.A., and Jansson, E.V., 1982, Eigenmodes, input
admittance, and the function of the violin: Acustica, vol. 50, p.
329-337.
[27] Jansson, E.V., Niewczyk, B.K., and Frydén, L., 1997, On body
resonance C3 and violin construction: J. Catgut Acoust. Soc.,
vol. 3, no. 3 (Series II), p. 9-14.
[28] Goodkind, H.K., 1972,Violin iconographiy of Antonio
Stradivari: New York, 780 p.
[29] Doring, E.N., 1945, How many Strads, our heritage from the
master: Chicago.
NOTE ABOUT TERMINOLOGY
We use the terminology in Refs. [23-27], in spite of the fact that it
is inconsistent and partly misleading.
• A0 (Hutchins) = Helmholtz resonance (H.A. Müller) =
ff-hole resonance (Cremer)
63
Martin Schleske – Empirical Tools in Contemporary Violin Making: Part I. Analysis of Design, Materials, Varnish, and Normal Modes
• C2 (Jansson)
• T1 (Jansson) = B1–(Hutchins/Bissinger)
• B1 = B1+ (Hutchins) = C3 (Jansson) = main corpus
resonance (H.A. Müller)
In my opinion it would be sufficient to denote three categories of
resonances, using A# for air resonances, C# for corpus (body)
resonances, and P# for plate resonances. These three categories
are unambiguously defined, independently of the quality of the
instrument. The historically motivated use of several other terms,
like T1 for an often pronounced top plate mode, is strongly
dependent on the characteristics of the individual instrument. In
fine instruments with outstanding sound, the T1 mode (“1st strong
top plate mode,” Jansson) is not at all limited to the top plate. The
motion of the T1 mode is often of equal amplitude in the top and
back plates, sometimes even larger in the back plate.
APPENDIX A
Recordings of Sound
Microphone: Sound level meter B&K 2237. Distance from violin: 1
m vertically above the sitting player. Ceiling height: 3.5 m.
Reverberation time of the room (violin shop): 0.6 s. Recording
directly on hard disk. Software: digidesign. Near-field monitors:
EMES. The judgment of the recordings by many players runs like
this: “The recording is brutal. Everything is heard. But I have seldom
recognized my own instrument so clearly.”
APPENDIX B
About the Instruments
“Schreiber” Stradivarius, 1712
Made by Antonio Stradivari in Cremona, 1712. It belonged to Dr.
Schreiber in St. Petersburg after whom the violin is named. Prior to
1900 it was played by H. Wieniawski. About 1900 it was sold to
Oswald Möckel and Dr. Louis Ravené in Berlin and played by
Huberman. In 1937 it was passed over to Hans Basserman who
shortly afterwards emigrated to the U.S. The instrument was played
by Pinchas Zukerman between 1968 and 1972 and is documented
on many recordings, for example Concerto for Violin and Orchestra
No: 5 a minor Op: 37 by Henri Vieuxtemps (CD Sony Classics SBK
48274). In 1976 it was on loan to Daniel Heifetz. Today the violin
belongs to a CAS member and customer of our violin shop. It is
64
played by one of the members of the SWR Sinfonieorchester
Freiburg/Breisgau.
References: [28, 29], Certificates by J.A. Beare, London and A.F.
Moglie, Washington D.C., and others.
Dominico Montagnana, 1729
This violin has an outstanding Venetian varnish, and a warm,
voluminous sound. For a period of ten years it belonged to the
concert master Kolja Blacher of the Berlin Philharmonie. Today it
is played by Alban Beikircher, a young soloist.
Guarneri del Gesu, 1733
The instrument belongs to one of the concertmasters of the
Münchner Philharmoniker. It is the present reference instrument
for our work in the shop.
APPENDIX C
Measurement of Damping and Modal Analysis
Support of instrument: Foam rubber pillows (height 12 cm) at the
upper and lower end blocks.
Measurements: Transfer functions FRF (f) = a(f)/F(f); number of
samples 4096. Bandwidth: 3.2 kHz. Software: difa d-TAC
Hardware: difa Measurement Systems FA-100 (4 ch FFT analyzer)
Number degrees of freedom (DOF): 595 in each mesh.
Excitation at all DOF’s: Impact hammer PCB 86C80 with plastic
tip (red).
Sensor: Accelerometer PCB 352B22 (0.5 g) (fixed horizontally at
the top on the G string side of the bridge).
Analysis: STAR Structure Modal Analysis; Advanced Curve Fitting
(Advanced options: tolerance F = 1.00 Hz; D = 1.0 Hz; Model size
= 30; Bandwidth of curve fitted FRF is 512 lines; more than 3 separate
curve fits in the observed frequency range).
Mode shapes: Scaling of mode shape amplitudes: Residues (absolute
values). All residues were divided by 2πf (mobility representation)
in order to allow a comparison of the modes in a data set.
Visualization of the modes by interpolation of the residues (58,704
points) and computation of iso-amplitude contours.
CASJ Vol.4, No. 5 (Series II), May 2002
 Empirical Tools in Contemporary Violin Making:
Part II. Psychoacoustic Analysis and Use of Acoustical Tools
Martin Schleske
Geigenbaumeister und Dipl.-Physik-Ing.(FH)
Meisteratelier für Geigenbau
Seitzstr. 4; D-80538 München, Germany
ABSTRACT
A method for measuring and interpreting the sound radiation of
bowed stringed instruments is introduced. The aim is to extract
psychoacoustical information from “technical” data in an
instrument‘s resonance profiles. Measured transfer functions
(Lp=p/F; with p=sound pressure and F=exciting force) provide
the input data for calculating specific loudness and overall loudness
as a function of playing frequency. Examples are given for violins
made by Antonio Stradivari and Joseph Guarneri del Gesu. Although
similar in their overall loudness (a quantity relevant to their
dynamical potential) and their tone-to-tone fluctuations of loudness
(a quantity relevant to their dynamic balance), they clearly differ in
their specific loudness patterns, which seem to be useful to describe
their tonal color (timbre).
Practical benefits of such acoustical tools in violin making include
the following:
a) Understanding correlations (Example: Resonance profile and
the effectiveness of the player’s vibrato)
b) Parameter studies (Example: Acoustics of the fingerboard)
c) Controlling the process of constructing new instruments (Ex-
ample: making tonal copies)
PSYCHOACOUSTIC ANALYSIS OF THE RADIATED SOUND
If modal analysis can be thought of as an empirical diagnosis tool for
the violin maker in order to show the vibrational behavior of an
instrument (see Part I), psychoacoustic analysis could be called a
controlling tool to analyze the resulting sound. Psychoacoustic
analysis can shed some light on the question: What are the tonal
consequences of the instrument’s vibrational behavior?
Our method of analyzing and evaluating the sound radiation of a
bowed stringed instrument is based on the following six steps:
1. Measurement of the spatial radiation of sound Lp(f;α)=p(f;α)/
F(f). The sound pressure level p(f;α) is measured on a circle
around the instrument at 36 different angles αi (α=0...360°;
∆α=10°; number i of different angles = 36) between micro-
CASJ Vol. 4, No. 6 (Series II), November 2002
phone and instrument while the instrument is excited by a
force F(f) by means of a small impact hammer-pendulum at the
bridge. From the excitation F(f) and response p(f;α), the trans-
fer functions Lp(f;α)=p(f;α)/F(f) showing the frequency depen-
dent ratios of sound pressure to exciting force are determined
using an FFT-Analyzer (FA-100, difa). The measurements are
done under normal workshop conditions, so room acoustics
such as reflection, absorption, and room modes have to be
taken into account. The effect of room reflections being re-
corded together with the direct sound offers a certain advan-
tage: some averaging occurs over the various radiation direc-
tions, even for a measurement at a single microphone position.
In terms of the number of transfer functions necessary to calcu-
late a suitable average of the overall radiated sound energy of
an instrument, measuring in reverberant acoustics means that
a far smaller number is necessary than in an anechoic cham-
ber. There is a danger that the room modes, which have a modal
density (modes per frequency range) far higher than that of the
instrument’s modes, might cause trouble in a non-anechoic
measurement by producing many additional peaks in the trans-
fer functions. To avoid this, we adopt a simple procedure. For
each fixed angle αi between violin and microphone, the whole
array (tripod including violin, impact hammer and microphone)
is rotated through 36 different angles ϕ (0-360o at 10o incre-
ments), while the angle αi between violin and microphone re-
mains unchanged. In this way, 36 transfer functions for that
specific angle αi are measured and averaged (with respect to
energy). This drastically reduces the influence of room modes
on the desired transfer function (the measurement room is char-
acterized by a volume of approx. 180 m3 and a reverberation
time of approx. 0.6s). The distance between microphone (B&K
2237 SPM) and violin is 0.5 m, which is within the nearfield but
far enough to be out of the immediate canceling effects caused
by mode shape areas out of phase. The plane of the micro-
phone rotation is perpendicular to the longitudinal axis of the
violin at the position of the violin bridge.
The result of this measurement is a 3D contour diagram of the
radiated sound showing the absolute level of sound radiation
43
Martin Schleske – Empirical Tools in Contemporary Violin Making: Part II. Psychoacoustic Analysis and Use of Acoustical Tools

Figure 1.

A. Directional characteristics of sound radiation of a violin by Guarneri del Gesu, 1733. Sound radiation level L obtained by the ratio of
sound pressure p divided by bridge force F as function of frequency f (x-axis) and radiation angle α around the instrument (y-axis) in
the plane of the bridge at 36 different locations. α=0°= microphone perpendicular to the top plate; 90°=bass side; 180°=perpendicular
to the back plate; 270°=sound post side. Distance between microphone and violin 0.5 m. Levels L (dB, re 1Pa/1N) in absolute scaling:
L/dB= grey legend value/dB + 80/dB. Note the monopole-like radiation of the (corpus-) resonances below 600 Hz and the
increasing unevenness of the sound radiation field with increasing frequency.

B. Directional characteristics of sound radiation for some eigenfrequencies (data equal with fig. 1a). Note: Due to the measurement in
nonanechoic (chamber music) environment the unevenness of directional radiation is smoothed. Note strong directivity.

44 CASJ Vol. 4, No. 6 (Series II), November 2002

 Martin Schleske – Empirical Tools in Contemporary Violin Making: Part II. Psychoacoustic Analysis and Use of Acoustical Tools
Lp(f;α) as a function of direction α (y-axis) and frequency f (x-
axis; see Fig. 1a). Horizontal grey-scale variations in the diagram
indicate the frequency dependence of sound radiation in one
particular direction, while vertical grey-scale variations indicate
the directional dependence of sound radiation at a particular
frequency. Almost omni-directional (monopole-like) radiation
is seen below 1000 Hz, while increasingly complex radiation
patterns become apparent as we move to higher
frequencies, as expected [1]. Figure 1b, containing a subset of
the data of Fig.1a, shows the directional dependence of sound
radiation for the frequencies of various eigenmodes.
2. Energy-based averaging of the measured set of Lp(f;α) over all
room directions αi (with α=0-360° ∆α=10°). The result is
called the resonance profile of the instrument (Fig. 2A-D).
3. The levels of the resonance profile are determined at the
frequencies of all harmonics which belong to all playable
chromatic tones of the instrument. As an input parameter, the
vibrato shift is specified (in cents; 100 cent = 1 semitone). This
determines the frequency interval at which each of the named
levels is scanned. As a standard parameter here we usually use
an “average” vibrato of 25 cents. With an overall vibrato shift
of y cent the upper frequency limit hi+ (in Hz) of the ith vibrated
harmonic hi is given by
5.
+ (y/2400) (y/2400)
Ia) hi = hi*2 = i*f0*2
where the frequency hi = i*f0, with f0 being the fundamental
frequency (in Hz). The corresponding lower frequency limit
hi- is
- 2 + jective judgment of loudness. For example, a sound of 2 sones
Ib) hi = (hi) /hi .
The result of this step is a set of “harmonic levels”. Some
colored contour-maps of the harmonic levels can be seen in the
sound radiation section of the website:
http://www.schleske.de/06geigenbauer/
akustik3schall4musikdarst.shtml
Furthermore, comparisons between the harmonic levels of
various violins (Stradivarius, Guarneri del Gesu, tonal copies)
are given in the “Acoustical Handbook” of the website (Note:
An English version of the web site should be available by Jan.
2003):
http://www.schleske.de/09extras/
extras3handbuch03klangkopie.shtml
4. Multiplication of the harmonic levels by the following weight-
ing functions:
CASJ Vol. 4, No. 6 (Series II), November 2002
a) Bow excitation: The harmonic levels are corrected
corresponding to the relative strength of excitation of each
harmonic from the force spectrum of a bowed string. A
string vibrating in “Helmholtz motion” exerts a force on
the violin bridge with a sawtooth waveform, which requires
a weighting in which the nth harmonic is divided by n:
relative to the fundamental, a factor of 1/2 for the second
harmonic; 1/3 for the third harmonic etc. Correction
factors for various types of strings can be included, to allow
for waveforms which are not an ideal sawtooth.
b) Dynamic (pianissimo – mezzo forte – fortissimo):
Adjustment of absolute harmonic levels according to a
range of sound pressure values that would be produced by
real bowing of the instrument (this calibration is necessary
as the auditory filter functions of the inner ear depend on
the absolute value of input levels.)
After applying these weighting functions to the measured har-
monic levels, we have the spectral components that would oc-
cur if the instrument was bowed. In the following text and
figures this part of the procedure (step 1-4) is called “bowed-
calc.” The weighted and calibrated harmonic levels are the
input quantities for the next stage, the calculation of loudness.
Calculation of specific loudness (S) and overall loudness of all
playable chromatic notes of the instrument.
Remarks on the term loudness: Loudness (measured in sones) is a
psychophysical term describing the strength of the ear’s per-
ception of sound. The sone scale was created to provide a
linear scale of loudness, which correlates with a listener’s sub-
is twice as loud as a sound of 1 sone, a sound of 4 sones is twice
as loud as a sound of 2 sones, etc. The definition of the unit is
that 1 sone corresponds to the loudness experienced by a nor-
mal person hearing a 1kHz (sinusoidal) tone at 40 dB SPL. The
specific loudness S is the loudness per frequency bandwidth,
specifically the loudness per critical band (a psychoacoustical
frequency; see below). The distribution of specific loudness on
this scale is an essential indication for the perception of tonal
color. The overall loudness of a sound is obtained by summing
this specific loudness across the whole scale.
Remarks on frequency scaling in ERB: The frequency scale in
critical bands, expressed in ERB (Equivalent rectangular band-
width of the filter characteristic of the ear), corresponds to the
“frequency axis” of the human inner ear. Sound waves that
reach the ear are transmitted through the eardrum and the three
small bones of the middle ear to the cochlea, and set its basilar
membrane (the basic mechanical sensor) into oscillation rather
like a flapping flag. Around 30,000 receptor-units (hair cells)
that are located on the basilar membrane [2] respond to the
45
Martin Schleske – Empirical Tools in Contemporary Violin Making: Part II. Psychoacoustic Analysis and Use of Acoustical Tools
motion of the membrane and transduce this motion into a neu-
ral code in the auditory nerve. The “ERB-frequency-axis” in
the psychoacoustical diagrams (Figs. 3-9) can be thought of as
the unrolled basilar membrane of the inner ear. An interval of
1 ERB corresponds to the width of one critical band, and corre-
sponds to about 0.9 mm on the basilar membrane (total length
for adults, about 35 mm). A critical band characterizes the
ear’s ability to separate component tones, for example whether
a loud tone can mask a nearby quiet one [3, 4]. For the percep-
tion of a complex stimulus (like a “musical tone”) it is impor-
tant to know whether frequency components lie within one
critical band or if they are distributed over different critical
bands. This difference plays a major role for various character-
istics in our perception, like perception of tonal color, rough-
ness, consonance or dissonance of musical intervals etc. [3].
The transformation from the physical unit “frequency” f (ex-
pressed in kHz) into the critical band number (expressed in
ERB) is given by the following formula [4]:
II) Number of critical band = 21.4 log10 (4.37f + 1) .
Our loudness calculations are based on Moore, Glasberg et al.
[4-7], by means of their program loudaes (using parameters free-
field, diffuse, binaural, complex, and harmonic).
There are five steps:
a) Fixed filter for transfer from free field to eardrum
The outer ear (pinna and ear canal) forms a resonant
acoustic system (basically a quarter-wave resonator), that
provides approximately 0 dB gain at the ear drum below 1
kHz, rising to 15-20 dB gain in the vicinity of 2.5 kHz, and
then falling in a complex pattern of resonances at higher
frequencies.
b) Fixed filter for transfer through middle ear
The primary function of the middle ear is that of an
impedance matching system, designed to ensure that the
energy of the sound wave is transmitted smoothly with
minimum reflections from the air in the outer ear to the
fluid in the inner ear.
c) Transform spectrum to excitation pattern
The excitation pattern (descriptive as the vibration of hair
cells) of a given sound is calculated from the effective
spectrum reaching the cochlea. It is calculated from
auditory filter shapes. The auditory filter shape represents
frequency selectivity at a particular center frequency and
can be thought of as a stimulus-dependent weighting
function.
46
d) Transform excitation pattern to specific loudness S
The specific loudness is the loudness per critical band
ERB.
e) Calculate overall loudness (alternatively in sone or phon)
for each “musical tone”
The overall loudness of a given sound is assumed to be
proportional to the total area under the specific loudness
pattern S versus ERB.
Results:
Psychoacoustical diagrams showing
a) Specific loudness S of the instrument (in sone) as a function of
a bowed-calc. musical scale (in semitones) and critical band (in
ERB).
b) Overall loudness of the instrument (in phon) as a function of a
bowed-calc. musical scale (in semitones).
The whole procedure can be summarized as follows: Steps 1-2
record the physical sound radiation of the instrument and represent
it as a resonance profile. Steps 3-4 address the question: “In terms of
the resonance profile, how strongly will certain spectral components
be radiated when the instrument is bowed?” These steps help to
interpret the resonance profiles of violins, violas and violoncelli in
accordance with their playing condition, i.e. their musical relevance.
Step 5 goes a little further as it treats the question: “What is the
relevance of this radiation in terms of human perception?” This
step interprets and scales the sound radiation according to the
excitation of the basilar membrane and includes known effects of
psychoacoustical processing of sound, like human hearing
sensitivity and masking [7].
Example 1: Using the method for analyzing the radiation of a violin by
Antonio Stradivarius (1712)
Figure 3 could be called a “tonal color (timbre) diagram”. It shows
the specific loudness that results from the measurements and
calculations just described (for a description and eigenmodes of
this Stradivarius see Part I, May 2002 issue). The top curve in Fig. 3
is the measured “resonance profile” of sound radiation (step 2)
which gives the input data for calculating the specific loudness of
all musical tones (steps 3-5) as they are represented in the grey-value
diagram (middle). The darker the grey value, the higher the specific
loudness and thus the more neural activity is caused by the respective
bowed note (see vertical axis showing a chromatic scale with 60
semitones) at the respective location on the basilar membrane (see
horizontal axis showing frequency group 2-37 in ERB). In these
diagrams we decided instead of showing the plain specific loudness
S, to represent “specific loudness levels LS” as defined by LS=log2S.
The consequence of this transformation is that a given linear
difference of LS corresponds to an equivalent change of loudness
CASJ Vol. 4, No. 6 (Series II), November 2002
 Martin Schleske – Empirical Tools in Contemporary Violin Making: Part II. Psychoacoustic Analysis and Use of Acoustical Tools
Figure 2.
A. „Resonance Profile” of Guarneri del Gesu (black) and „Tonal
Copy Op.51, 2001" (white).
Typical similarities and differences become obvious:
• Similar characteristics of Corpus modes (eigenfrequencies,
prominent levels). Typical for del Gesu: T1-Mode below
500 Hz has stronger radiation than B1 above 500 Hz.
• Similar incision between the resonances
Differences: The tonal copy shows stronger level variations;
higher levels and more „brilliance-resonances” (around 32 ERB)
C. Comparison of „Resonance Profile” Stradivarius, 1712 (white)
versus Stradivarius, 1727 (black). Note highly different resonance
structure in the low-frequency corpus resonance region, but similar
resonance structure from 1500 Hz upwards.
CASJ Vol. 4, No. 6 (Series II), November 2002
B. Comparison of „Resonance Profile” Stradivarius 1712 (white)
versus Guarneri del Gesu 1733 (black). It shows the principle
differences in „corpus resonance structure”: Guarneri’s T1 (433
Hz) and B1 (544 Hz) are far more apart from each other than
those of the Stradivarius. Furthermore the T1 of Guarneri is
the strongest radiating mode in the lower frequency range.
D. Comparison of „Resonance Profile” Stradivarius, 1712 (white)
versus „Tonal Copy, 1999" (black). Note: Similar resonance
structure in the corpus resonance region and in the general
„covering” over the peaks. The “tonal copy” shows to have higher
specific loudness at 28...32 ERB which we would regard as having
more brilliance, focus and strength. Less loudness in that region on
the other hand results in a smoother tone.
47
Martin Schleske – Empirical Tools in Contemporary Violin Making: Part II. Psychoacoustic Analysis and Use of Acoustical Tools

Figure 3. Psychoacoustic Evaluation of the sound radiation of a violin (violin: Stradivarius, 1712) – Demonstration of the steps and results
of the method as described in the text:
On top: “Resonance Profile” of sound radiation (step 2).
Below right: Grey-Value-Diagram (contour plot) of the specific loudness as function of “musical tone” and number of ERB. Calculated
from the Resonance Profile (step 3-5a). Vertical scale = chromatic scale of 60 musical semitones, starting with the open g-string as tone-
#1- Horizontal grid lines represent musical fifths.
Horizontal axis = frequency, being scaled using a) “technical” frequency (Hz) (upper edge of Resonance Profile) and b) using the
„frequency axis of the inner ear”: number of ERB. Vertical grid lines represent each frequency group on the basilar membrane (number
of ERB).
Note: The darker the grey values the stronger the excitation at this region on the basilar membrane caused by the respective “musical
tone”. Further explanations on this “timbre-diagram” see text.
Below left: Overall loudness level (in phon) as function of musical tone. It results from the summing across the specific loudness-values of
all frequency groups (step 5b).

48 CASJ Vol. 4, No. 6 (Series II), November 2002

 Martin Schleske – Empirical Tools in Contemporary Violin Making: Part II. Psychoacoustic Analysis and Use of Acoustical Tools

Figure 4. Psychoacoustical differences between the Stradivarius, Figure 5. Specific loudness variations as function of musical
1712 and a Guarneri del Gesu, 1733. The diagram represents the tone (vertical scale) and number of ERB (horizontal scale) caused
differences of specific loudness-level [∆LS=log2(SA/SB) with SA by playing each tone of the chromatic scale with a 25-cent
being the specific loudness of Instrument A, SB being the specific vibrato. It shows that the variations in different frequency groups
loudness of Instrument B] as function of the „bowed-calc. musical and for different tones are relatively unequal. The diagram shows
tone” (vertical axis) and the number of ERB (horizontal axis). to which extent the vibrato allows to modulate not only loudness
Note: A difference of ∆LS =1 (s. key) represents twice the specific but timbre of the violin. Instrument: Guarneri del Gesu, 1733.
loudness. White areas of a certain note show the amount to which Calculation not from the real bowed string but from “Resonance
the „white instrument” (here: Stradivarius) causes higher specific Profile” using steps 1-6 (see text). See fig. 4 for explanation.
loudness at this particular region on the basilar membrane as
compared with the „black” instrument (here: Guarneri). Analogous
is true for the black areas: Here Guarneri creates higher specific
loudness than Stradivarius. So the diagram from tone to tone shows
a psychoacoustic comparison of timbre as it visualises to which
amount the different instruments excite different parts of the inner
ear.

CASJ Vol. 4, No. 6 (Series II), November 2002 49

Martin Schleske – Empirical Tools in Contemporary Violin Making: Part II. Psychoacoustic Analysis and Use of Acoustical Tools
Table 1.
Sound Specific Loudness S Specific “Loudness Level LS” Perceived
(in sone) LS=log2 S loudness-
(as used in our topographic change
contour diagrams)
#1 1 0
#2 2 1 = twice
#3 4 2 = twice
#4 8 3 = twice
#5 16 4 = twice
perception that is particularly helpful when dealing with this sort
of topographic-style maps: each grey-value change represents an
equal change of perception if LS instead of S is plotted (see table 1).
The specific loudness levels around the musical tone’s fundamentals
are located on the first (from left) diagonal bent grid line. The
respective diagonal grid lines that follow adjacent on the right-hand
side give the positions of the specific loudness levels that are caused
by the following harmonics. Furthermore a white data point for
each musical tone in the grey-value diagram represents the position
of the tone‘s “spectral center of gravity”: the sums of specific
loudness S across the critical bands above and below that point are
equal. The “spectral center of gravity” gives an indication of the
tonal color, “dark” versus “bright”: the nearer this point is to the
left-hand side, the more the sound of that musical tone tends to be
“sonorous” or “dark”.
It should be noted that Fig. 3 shows which resonances of the
instrument are responsible for causing the various regions of
“concentrated” specific loudness of musical tones. Thus the degree
of importance of various resonances in the resonance profile can be
interpreted. Moving from the Stradivari‘s strong corpus resonances
(see resonance profile around 500 Hz, respectively 10-12 ERB)
vertically downwards into the grey-value contour diagram we find
the “concentrated” specific loudness in the region of the
fundamentals of the lower part of the A-string (tone# 15-20), giving
those tones “strength” and “depth”. Changes of tonal color
(“timbre”) from tone to tone are made evident by the vertical change
of grey-value structure along the chromatic scale. The G-string notes
start with only little specific loudness at the region of their
fundamentals, followed by a clear timbre change around tone #7
(#C). Here the effective radiation caused by the Helmholtz resonance
A0 causes a strong increase of “fundamental loudness” around 7-8
ERB.
Comparing G- and D-string with A- and E-string, a significant
weakness of violins in general becomes obvious. Compared with
the higher strings the lower strings lack excitation potential in the
region where their fundamentals excite the basilar membrane. Due
to the comparative human deafness in the low frequency region
and due to the existence of only one single useful resonance (the
50
Helmholtz resonance A0) in the region of the first octave, the violin
sounds comparatively bright in the lower range compared with the
timbre of the A- and E-strings. In agreement with this, the lower
strings show centers of spectral gravity shifted to rather high ERB-
values relative to their fundamental frequencies (on the G-string
around 14 ERB, on the D-string around 16 ERB). Only from the
open A-string upwards do the fundamentals start to benefit clearly
from the resonance strength of the corpus resonances (T1 and B1)
#1
#2
and the centers of spectral gravity of the respective musical tones
#3 (#19...) start to move rather steadily upwards with increasing pitch.
#4 For this Stradivarius, from the open A-string (tone #22) up to #F3
(tone #36) the centers of spectral gravity are relatively constant
around the frequency of the respective 2nd harmonic. For higher
musical tones on the E-string the centers of spectral gravity start to
shift downwards. Due to the low-pass filtering effect of the violin’s
bridge which progressively affects the lower harmonics of musical
tones, the specific loudness of these high musical tones on the E-
string above tone #36 are increasingly dominated by the strength of
their fundamentals.
The curve in the left part of Fig. 3 shows the overall loudness-level
(in phon) as a function of the chromatic scale. The overall loudness
of each of the musical tones results from “horizontally” summing
the specific loudness across all frequency groups. The maximum
spread of overall loudness-level, an important indicator of the
dynamic balance of the instrument, is about 10 phon. Also here we
see that the “acoustical potential” clearly decreases from the higher
strings to the lower ones, which “challenges” the player to balance.
This challenge (to avoid the term weakness) is even stronger in the
violoncello.
PSYCHOACOUSTIC COMPARISONS
As a final step of our method it has proved valuable not only to
analyze the specific loudness of a single instrument but also to
calculate the differences between two instruments. It has been found
that even when using a high resolution color scale for plotting
absolute loudness patterns, the visual appearance of specific loudness
pattern diagrams of different sounding violins look almost the same.
Obviously our eye is poor at perceiving and weighting such small
differences, but our ears are more acute. My impression has been
that our sense of hearing seems to perceive even the smallest changes
in tonal color when listening to different sounds, whereas our visual
sense almost seems to ignore those differences on the respective
colored loudness diagrams. Only if colored maps are created that
visualize the loudness differences of instruments instead of
visualizing their absolute loudness patterns do the tonal differences
seem to correspond with the visual representations. This finding
led to the last step:
6. AB-comparison of two instruments
a) Difference of the specific loudness levels LS”=log2S of two
instruments A and B. Result: Psychoacoustic differences of
CASJ Vol. 4, No. 6 (Series II), November 2002
 Martin Schleske – Empirical Tools in Contemporary Violin Making: Part II. Psychoacoustic Analysis and Use of Acoustical Tools
Figure 6. Stradivari did not copy himself: Psychoacoustical
differences between the Stradivarius, 1712 (white) and
Stradivarius, 1727 (black) caused by differences in the resonance
structure (see fig. 2c). High similarity in specific loudness above
18 ERB. (see fig. 4 for explanation)
their tone. Note: Due to the use of specific loudness level
LS” in the comparison diagrams by forming the logarithm
of the specific loudness S, each distance between two grey-
values in the contour-diagram represents an equal
perceptual change in the specific loudness of the two
instruments.
b) Representation of overall loudness (in phon) of two
instruments A and B (see Fig. 10a-c)
CASJ Vol. 4, No. 6 (Series II), November 2002
Figure 7. Psychoacoustical differences between the Stradivarius,
1712 (white) and Schleske, 1999 (black). Higher specific loudness
(particularly between 16-24 ERB) of the new instrument (see fig. 4
for explanation).
Example: Psychoacoustical differences between the violin by
Antonio Stradivarius (1712) shown above, and a Joseph Guarneri
del Gesu (1733) are shown in Fig. 4. This plot represents the
differences ∆LS” of specific loudness as a function of the chromatic
tone and the “frequency axis of the inner ear” (in ERB). A difference
of ∆LS =1 (see legend) corresponds to a perception of twice the
specific loudness caused by one instrument compared with the
other in a particular region of the basilar membrane. The white
areas in the diagram show the regions and the amounts of higher
51
Martin Schleske – Empirical Tools in Contemporary Violin Making: Part II. Psychoacoustic Analysis and Use of Acoustical Tools
specific loudness of the “white” instrument (the Stradivarius)
compared with the “black” instrument (the Guarneri del Gesu).
The horizontal scale reveals where on the basilar membrane these
different levels of excitation occur. The corresponding is true for
the black areas, showing where the “black” instrument has higher
loudness. So this diagram gives a psychoacoustic comparison-tool
for visualizing tonal color from tone to tone (Note: For a colored
version of this diagram see http://www.schleske.de/picshoch/
a0015_vgl_a0044_abstr_psy.gif).
AB-comparisons of different pairs of masters can be done, based on
a database of some 40 violins analyzed using the above techniques.
In contrast to other masters the comparison of Stradivarius versus
del Gesu might reveal typological differences, since the following
difference-pattern can be observed (more violins by these two famous
masters should be analyzed and compared before strict conclusions
can be drawn): The del Gesu causes a relatively strong “firing” of
nerves in the lower (sonorous) ERB-region (8-20 ERB) and at the
same time in the higher (brilliance) ERB-region (27-32 ERB), while
the Stradivarius causes stronger excitations in the intermediate
region of the basilar membrane. For very high musical tones (from
about one octave above the open E-string) the Stradivarius has a
more fundamental-dominated sound: from D3 (1175 Hz) on, the
calc-bowed tones create stronger specific loudness in the
fundamental region of each tone (left region of white areas) compared
to those of the Guarneri del Gesu. Despite these differences that can
clearly be perceived as different tonal colors when listening to both
instruments, their calculated overall loudness (as an indicator of
the “dynamical reserves” of each instrument) is almost identical
(this follows from Fig. 10a). Furthermore, both violins show a
comparable degree of fluctuations of overall loudness going through
the chromatic scale from tone to tone. As mentioned above these
fluctuations are in the range of 10 phon which corresponds to a
doubling of loudness perception. They are an indication of the
degree of “dynamic imbalance” of the instruments, for which the
player must compensate.
Comparisons made with the aid of these psychoacoustic diagrams
prove to be a discriminating empirical tool
• when investigating tonal color differences of various instru-
ments
• when controlling the sound of newly made instruments
• when performing parameter studies on the tonal influences of
certain construction parameters or adjustments of the instru-
ment.
Recall that these results are based on the data of the measured
resonance profiles of the instruments (step 1) and the ensuing
calculations (steps 2-5), not on the subjectively real bowed
instrument. The method tries to extract musically relevant
information (“harmonic levels”) and psychoacoustically relevant
52
information (“excitation of the basilar membrane”) from the
“technical” data of the resonance profiles of the instruments and
thus suggests how the specific resonance profile could be interpreted.
Nevertheless some limitations (that may have noticeable effects on
our perception of sound) should not be overlooked:
• The hammer-excitation of the instrument (step 1) is performed
only in the main excitation direction of the bow (perpendicu-
lar to the string)
• All phase information of the radiated sound is lost (due to
averaging out the room modes)
• All spatial information of the radiated sound (still present in
step 1) is lost with the further steps (due to energetic averaging
across the radiation angles in order to calculate a resonance
profile as input for the following psychoacoustic evaluation)
• No time-varying effects are taken into account.
WHY ACOUSTICAL TOOLS?
Together with Part I (concentrating on construction and modal
analysis), our acoustical workshop tools have now been described.
We would like to end this introduction with some examples of the
benefit of their use. In terms of workshop practice we would state
that acoustical tools can be helpful for
A) Learning to understand correlations between construction and
sound
B) Learning to evaluate parameter studies
C) Learning to control the making process of new instruments
D) Learning to diagnose faults and weaknesses of instruments (be-
cause of length limitations for this paper, this last point must
be kept for a later publication)
A) Understanding correlations – Example: resonance profile
and the effectiveness of the player’s vibrato
Each serious violin maker tries to learn which controllable
parameters in the making process of a violin are responsible for
certain tonal attributes. This section focuses on a “triad” which is
particularly important for the musical success of a violin: the
harmony of construction parameters (here: violin varnish), the
acoustical relevance (here: resonance damping) and musical perception
(here: vibrato and modulability):
When the resonance profile of an instrument shows significant level
differences in narrow frequency bands, vibrato from the player
(which causes a periodic change of pitches of all harmonics of the
played tone due to the oscillating motion of the left hand) will
create a distinct amplitude modulation of the harmonics. To achieve
significant level differences between narrow frequency bands
requires a reasonable density of resonances per frequency range
and low damping values of the single resonances, leading to a low
“modal overlap factor” [8]. When there are significant narrow-band
level differences in the resonance profile, the small periodic
CASJ Vol. 4, No. 6 (Series II), November 2002
 Martin Schleske – Empirical Tools in Contemporary Violin Making: Part II. Psychoacoustic Analysis and Use of Acoustical Tools
Figure 8. The specific loudness differences between a Guarneri-
Copy and a Stradivari-Copy are very similar to those between the
two original instruments (compare with Fig. 4).
frequency shift from vibrato is sufficient to make the harmonics of
the played tone rush up and down over the (fixed) resonance peaks
of the instrument. The work of Gough [9] and, back in the 1970’s,
McIntyre and Woodhouse [10] has considered this “vibrato effect”,
and aimed “toward a psychoacoustically realistic violin physics”.
The latter work arose out of the experiments by Max Mathews with
electronically simulated violin resonances. Similar
psychoacoustical experiments by Weinreich [11] using electronic
synthesis showed that purely frequency-modulated vibrato was
perceived as sounding synthetic or vulgar, whereas a pure amplitude
modulation of harmonics (with the frequency-modulation
CASJ Vol. 4, No. 6 (Series II), November 2002
Figure 9. Effect of thickness modification (see Fig. 14) on specific
loudness pattern. It shows: You can not win everything. But in most
regions (black areas) the dynamic resources could be developed.
(see Fig. 4 for explanation)
artificially removed) was perceived as almost unchanged, as
compared with a natural real vibrato.
In our opinion a reason for this phenomenon might be based on the
fact that a pure frequency modulation only causes a periodic local
shift of the unchanged excitation pattern on the basilar membrane,
whereas an amplitude modulation leads to a periodic change of
excitation pattern shape. This change of shape would be due to a)
periodic changes of the overall area of the excitation pattern, and b)
the nonlinear fanning out of the upper flanks of auditory filter
53
Martin Schleske – Empirical Tools in Contemporary Violin Making: Part II. Psychoacoustic Analysis and Use of Acoustical Tools

functions with increasing levels [4]. It can be assumed that such

periodic changes of excitation pattern shape give rise to more Figure 10. Comparing overall loudness (calculated as described in
complex neural signals to the brain than simple periodic frequency steps 2-5) as function of „musical tone” of various violins – showing
shifts of excitation patterns. It is well known that fluctuations or the „dynamic resources” of the instruments:
changes in all kinds of sensory stimulation create higher attention
than constant stimuli, even when being of high intensity.
A. Stradivarius, 1712 (white) and Guarneri del Gesu, 1733 (black)
For the quality of the instrument this means: the higher the
resonance density and the less damped the resonances of the violin,
the smaller the variations in the playing of the instrument (vibrato;
change of bowing parameters etc.) which will affect the neural
excitation level and will thus increase the noticeability of the sound.
It can be presumed that the “fieriness” and “liveliness” in the tone
of high quality violins is (besides other things) based on such an
effect. This effect might have to do with what workshop experience
may call “perception by quality” in contrast to a pure “perception by
intensity” when judging violins.

Example: As described above, the minimum and maximum sound

pressure levels of the harmonics of a musical tone during one vibrato
period cause different excitation patterns on the basilar membrane
and thus different specific loudness. For a typical musical tone (A1;
440 Hz fundamental frequency) Fig. 11 shows the maximum (black
curve) and the minimum (grey curve) specific loudness as they are B. Guarneri del Gesu, 1733 (black) and Schleske, Op.51, 2001 (white)
created by the sound pressure levels of the harmonics during one
vibrato period (medium vibrato with an overall shift of 25 cent).
Levels are “calc-bowed” according the method described earlier.
The instrument used is the Joseph Guarneri del Gesu 1733 violin
(see Fig. 4). The sound pressure (dB) showing the maximum levels
of harmonics is indicated on the left scale, while the specific loudness
curves (sone) corresponding to the maximum and minimum levels
are indicated on the right scale. There are very obvious differences
in specific loudness curves caused by vibrato (or more precisely by
the amplitude modulation of the vibrated harmonics). These
differences are particularly distinctive in the region of the
fundamental (10 ERB) which for this particular violin is due to its
strongly radiating T1 corpus mode in the immediate neighborhood
of the fundamental frequency of that tone.

Through each vibrato period of the bowed musical tone the specific
loudness changes periodically between the two curves shown. C. Stradivarius, 1712 (white) and Schleske, Op.37, 1999 (black)
However, the individual harmonics do not evoke their minimum
and maximum values simultaneously, but normally at different
times. The reason for that is that each “vibrated” harmonic is shifted
along the frequency axis (in a local frequency range determined by
the degree of vibrato shift) and thus each harmonic moves up and
down over the resonance profile of the instrument and reaches its
local level-maximum and local level-minimum at different times.
So, for each harmonic its spectral proportion of maximum and
minimum sound pressure per vibrato period is radiated at different
times. As a consequence the local maxima and minima on the basilar
membrane are not created synchronously. In other words the motion
from maximum to minimum loudness pattern and back within one

54 CASJ Vol. 4, No. 6 (Series II), November 2002

 Martin Schleske – Empirical Tools in Contemporary Violin Making: Part II. Psychoacoustic Analysis and Use of Acoustical Tools
Figure 11. The figure shows the effect of the amplitude modulation
of harmonics caused by the player’s vibrato: It creates a periodically
differing excitation of the basilar membrane. Maximum (black
curve) and minimum (grey curve) specific loudness pattern during
a vibrato period (vibrato frequency shift: 25-cent; tone; a1;
Instrument: Guarneri del Gesu, 1733. The Sound-levels of the
harmonics (belonging to that tone) that create the maximum pattern
are represented by data-points. Calculation not from the real bowed
string but from “Resonance Profile” using steps 1-6 (see text).
vibrato period does not happen in phase but (depending on the
characteristics of the resonance profile) with significant phase
differences. It can be guessed that this effect may intensify the
“liveliness” of the perceived musical tone, since with these non-
simultaneous effects the neural processor presumably is kept busier
evaluating the neural code than with a uniform change of specific
loudness. Only if all harmonics reached their local maximum in
the resonance profile at the same time would the vibrato-caused
modulation of overall loudness for this example (with unchanging
bow dynamics) vary between L=29.2 sone and L=24.3 sone as
indicated in the figure. Effects of time-masking are not considered
here. Both the fluctuation of overall loudness caused by the vibrato
and also the local fluctuations of specific loudness could be
important for the perception of the “liveliness” of a musical tone –
for some critical bands and tones, as shown in Fig.5, these specific
fluctuations reach a factor of two! The amount by which the specific
and overall loudness of a musical tone is modulated within one
vibrato period depends on the quality of the instrument’s resonance
profile in the region of the harmonics belonging to that tone.
In order to investigate this quality Fig. 5 shows the differences of
specific loudness for all playable musical tones (vertical axis:
chromatic scale) caused by the vibrato (same vibrato and same
instrument as in Fig. 11). The heights of white areas represent the
difference of specific loudness level ∆LS=log2(Smax/Smin) where Smax
is the specific loudness caused by the maximum levels of harmonics
and Smin is the specific loudness caused by minimum levels of
harmonics during one vibrato period. A difference of ∆LS =1 (see
legend) corresponds to a perception of doubling the specific
loudness.
CASJ Vol. 4, No. 6 (Series II), November 2002
The diagram shows that over the whole playing range the maximum
fluctuations of specific loudness caused by the vibrato are
considerable. It is obvious that these fluctuations are significantly
different in different critical bands. The fluctuations show a slight
tendency towards higher magnitudes at higher critical bands. As
the fluctuations are not equally distributed over the ERB-scale it is
obvious that not only the overall loudness but also the tonal color
(timbre) is modulated by the vibrato. As can be seen, this musical
attribute, which one might call “modulation”, is not equally
pronounced for all musical tones. The differences in shape and size
of white areas show that the musical tones are far from reacting
with uniform sensitivity to the vibrato of the player. Some tones
react more sensitively (i.e. create more excitation pattern changes
per given “playing change”) than others. It can be presumed that
these processes (namely an invariable resonance profile and
frequency-variable vibrated harmonics) demand highly complex
processing activities of the human sense of hearing, occupying the
neural processor. The testimony of musical listeners who “feel
thrilled” by such “lively violin tones” might be based on the effects
described.
In terms of the quality of the resonance profile, the account
developed here emphasizes the importance of a high resonance
density and small resonance damping values in order to increase
the modulabilty of the instrument. This underscores our statement
(see Part I) about “musical violin varnish” concerning the target of
achieving low damping of eigenmodes and an increase of the wood-
quality-ratio c/δ (with c = speed of sound of longitudinal waves;
δ = density).
With regard to their experiments on electronically simulated violin
resonances” [10] Woodhouse points out [12] that “if the damping is
too light the result becomes unpleasant again. The reason seems to
be that the high-Q resonances are still ringing on at their own
frequency when the pitch of the vibrato note has shifted significantly
away, and the resulting composite sound has discordant
ingredients”(Note: Q is defined as the ratio of resonance frequency
divided by half power bandwidth; Q is the inverse of loss factor η,
or the inverse of twice the damping). This raises the potential danger
of damping becoming too low, but Woodhouse admits that although
the damping values of conventional instruments may come close to
such a ringing situation their resonances do not seem to reach such
low damping. So it should be safe to say (unless a violin is made of
a material other than wood) that the aim of using any treatments
(wood pre-treatment, primer, varnish, etc.) is to create a resonance
profile with the lowest possible damping values and the highest
possible resonance density.
In view of the fact that the density of violin resonances as a function
of frequency (in Hz) is approximately constant whereas the
frequency shift of each harmonic caused by vibrato increases with
harmonic number, the effect described should become more
noticeable in the higher frequency range of the resonance profile.
55
Martin Schleske – Empirical Tools in Contemporary Violin Making: Part II. Psychoacoustic Analysis and Use of Acoustical Tools

Figure 12. Eigenmodes of the fingerboard being mounted on the violin.

On top: Input admittance at driving point A (free fingerboard corner) showing the resonance peaks.
Below: Mode-shapes belonging to those peaks (experimental Modal Analysis).

56 CASJ Vol. 4, No. 6 (Series II), November 2002

 Martin Schleske – Empirical Tools in Contemporary Violin Making: Part II. Psychoacoustic Analysis and Use of Acoustical Tools
The higher the harmonic number, the wider is the absolute frequency
shift (in Hz) by which that harmonic fluctuates over the violin’s
resonances and the higher is the number of resonance peaks of the
instrument that are “crossed” by that harmonic within one vibrato
period. Fig. 5 shows the maximum differences of specific loudness
during one vibrato period but not the locally different frequency of
fluctuations within that period. The increasing resonance density
relative to the frequency shift of each harmonic is the reason why
this fluctuation increases with increasing harmonic number. There
is also a spatial effect. Each resonance peak of the instrument is
characterized by its own “radiativity” [1]. The vibrato on a fine
violin (fulfilling the criteria described above) is not only
characterized by an amplitude modulation but also by a “radiativity
modulation” – a modulation of spatial radiation which is more
noticeable when more resonance peaks are crossed by the harmonics
of the vibrated tone. This might be the reason why musicians
sometimes claim that the tone of a fine violin “lives” in the hall or
has some kind of “spatial authority”. These considerations
emphasize the importance of the higher frequency range of the
resonance profile for the modulability of the instrument.
Note: The modulabilty diagram Fig. 5 must not be confused with
the AB-comparison diagram Fig. 4. While the modulabilty diagram
compares the differences of maximum and minimum specific
loudness caused by the vibrato on one instrument, the AB-
comparison diagram compares the maximum specific loudness
(within a certain allowed vibrato shift) between both instruments.
B) Parameter studies – Example: Acoustics of the
Fingerboard
From Part I of this paper (table 3) it can be seen that two violins by
Antonio Stradivarius happen to show two frequencies of the B1-
mode in the resonance profile. The eigenmode-map of the
“Schreiber”-Stradivarius 1712 given in Fig. 14b of Part I indicates
that indeed two modes (at 513 Hz and closely above at 524 Hz)
show the typical B1 mode-shape. The difference between these
modes lies in the fact that they show opposite phases between the
free end of the fingerboard (twisting around its longitudinal axis)
and the top plate. This obviously indicates a resonant coupling
between the “corpus system” and the “fingerboard subsystem”.
Modal analysis of the glued-on fingerboard indeed shows a number
of fingerboard modes that act largely independently from the modes
of the corpus. Fig. 12 shows the resonance peaks of these fingerboard
modes in the input admittance (measured at driving point “A” at
the left corner of the free fingerboard end) and below that the mode
shapes corresponding to these peaks. In the frequency range up to
2500 Hz the fingerboard subsystem shows eight modes, where mode
#2 is a bending mode of the free end (“B0-mode”); mode #3 and #4
are torsional modes (“Tf-modes”) of the free end; mode #6 and #7
high frequency bending modes with nodal lines already within the
free fingerboard’s end and finally mode #8 a torsional mode with a
nodal line cross within the free end.
CASJ Vol. 4, No. 6 (Series II), November 2002
Investigations on adjusting violins with new fingerboards in our
studio showed that with a carefully graduated fingerboard it is
possible to create a coupling of one of the torsional modes Tf of the
fingerboard (modes #3 or #4 in Fig. 12) with one of the corpus
modes, especially with the B1-mode. What follows from that is a
“spectral splitting” of that corpus mode. As this splitting causes a
“widening” of the frequency region of significant radiation from
the corpus modes (see Fig. 14a of Part I, May 2002 issue) by
transforming one well radiating mode into two well radiating
(adjacent) modes, this effect is probably desirable. It will be
particularly noticeable in the region of the first position of the G-
and A-strings, where the second harmonics (G-string) and
fundamentals (A-string) fall in that frequency region.
More familiar is the so-called A0-B0 coupling. In this case (also
present in the violins by Stradivarius shown in Part I) mode #2 (Fig.
12), here at 283 Hz, couples with the Helmholtz mode A0. The
mechanism of A0-B0 coupling is described in detail by Woodhouse
[13]. From our workshop practice we have observed that “resonance-
coupled” instruments (with A0-B0 coupling and preferably also Tf
coupling) are regarded by most players as being more “lively” or
“resonant”. We share this judgment with Hutchins [14] and
Woodhouse [13].
An example of a fingerboard made in such as way as to give both Tf
coupling and A0-B0 coupling is shown in the frequency response
curves (acceleration a divided by force F) in Fig. 13. The black
curve is a transfer function between the two corners of the free end
of the fingerboard (glued in place on the violin). For this
measurement of fingerboard modes the ff-holes are covered with
foam in order to damp the A0-mode and minimize coupling, to
isolate the real eigenfrequency of mode #2 (“B0”). The grey curve is
an input accelerance (a/F) of the corpus (driving point at the left
Figure 13. Example of a successful coupling of fingerboard modes
with corpus modes. Black: Frequency Response Function (FRF) of
fingerboard; grey: FRF of corpus. There occur two couplings; A0-B0
and TF-T1 and the instrument is well adjusted to perform a more
„lively” and „resonant” feeling.
57
Martin Schleske – Empirical Tools in Contemporary Violin Making: Part II. Psychoacoustic Analysis and Use of Acoustical Tools
bridge foot). For this measurement the free fingerboard end was
held tightly with one hand in order to avoid coupling with the B0
and Tf modes, so that the real eigenfrequency of the Helmholtz
mode (A0) could be determined. The figure shows that two close
matches of eigenfrequencies have been achieved (vertical broken
lines): (a) Helmholtz mode (A0) with corpus mode (B0), and (b)
fingerboard torsion (Tf) with the lower corpus mode (T1). This
fingerboard is well tuned to give a twofold coupling with the corpus
of the instrument. In our experience this gives the instrument a
“living” and “resonant” feeling when being played.
Tuning the fingerboard:
In order to achieve the best tuning of the fingerboard for a given
instrument we have determined some empirical workshop rules.
The mode shape of the first bending mode (“xylophone mode”) of
the fingerboard under free boundary conditions is somewhat similar
to how the fingerboard will vibrate in the B0 mode when it is glued
on the violin. To hear this xylophone mode frequency, hold the free
fingerboard at one of its nodal lines at approximately 1/4 of its
length and tap and listen at one end or the centre. In the case of our
newly made violins (which have relatively similar scrolls and necks)
the eigenfrequency of this free fingerboard mode turns out to fall at
1.67-1.68 times the eigenfrequency of the eventual B0 mode. As this
frequency ratio is almost a major sixth, the free fingerboard should
thus be tuned to that interval above the frequency of the A0-mode,
in order to achieve a good A 0-B 0 coupling later. The A 0-
eigenfrequency must be determined (e.g. by blowing across the ff-
holes) with the sound post inserted, as the sound post shifts A0 to a
significantly higher frequency.
Note: Depending on the individual graduation of neck and scroll
the suggested frequency ratio (1.67-1.68) can vary as the B0
eigenfrequency depends also on the stiffness of the neck and on the
mass of neck, scroll and pegs. Each maker should establish their
own frequency ratio by simple tests.
A certain amount of fine tuning of the fingerboard can be carried
out after it is glued on, by shortening its length. An experiment
involving successive shortening of a fingerboard in small increments,
measuring B0 and Tf eigenfrequencies and curve-fitting the resulted
data cloud, resulted in the following empirical formulae for
eigenfrequency shifts of B0 and Tf modes:
(III): Increase of B0 eigenfrequency (in %)=
∆L*(0.1531*∆L+1.3097)
where ∆L = length reduction of the fingerboard (in %); length=270
mm being 100% for the violin. The curve fit has high accuracy of
R=0.9979.
Example: Shorten the fingerboard from 270 mm to 265 mm, which
is a reduction ∆L of 1.85%. Insert in formula III:
58
Increase of B0 eigenfrequency (in %)
=1,85*(0.1531*1,85+1.3097)=2.95%
So if B0 used to be 250 Hz, by shortening the length of the finger-
board by 5 mm it will rise by 2.95% to 257.4Hz.
(IV): Increase of Tf-eigenfrequency (in %)=1.6246*=∆L
Again the curve fit of this empirical function has high accuracy of
R=0.9972.
Example: Shorten fingerboard by 5 mm and insert this percentage
of length reduction (∆L=1.85%)
Insert in formula IV: Increase of T f-eigenfrequency (in
%)=1.6246*1.85%=3%
So if Tf used to be 510 Hz, by shortening the length of the finger-
board by 5 mm it will rise by 3% to 525.3 Hz.
These formulae provide a certain latitude for later adjustments of
the fingerboard. Nevertheless it should be noted that the main
frequency-sensitive working step when making the fingerboard
comes from its thickness and the concave hollowed out area
(particularly length and depth of that area) on the underside.
Of course, as well as fingerboard studies many other parameter
studies focusing on other elements of the violin can be useful. Good
examples of such “simulation experiments” are to be found in the
work of Rodgers [15;16], using the Finite-Element Method for
investigating the acoustics of the bassbar, thickness graduation etc.
C) Controlling the making process of new instruments
When acoustical tools are used in the making process of a violin, an
obvious approach is to learn from the acoustical properties of
existing fine instruments whose tonal quality is highly regarded.
Modal analysis as a diagnosis tool and sound analysis as a controlling
tool can be very valuable here. They can help to match the modes
being modified by the various steps in the making process as closely
as possible to those of the reference instrument [17, 18]. At the end
of the making process psychoacoustical sound analysis can control
(and potentially optimize) the final result. While one aim could be
to make a “tonal copy” (more precisely a “resonance copy”) of one
particular reference instrument, another (perhaps more artistic)
aim would be to understand the typical characteristics of various
reference instruments from their modal behavior and tone and to
create an individual “modal composition”.
The term “tonal copy” must not be misunderstood as creating
complete identity, but rather as guaranteeing a clear affinity of the
two instruments. Confirmation of the success (through the
similarity between original and “tonal copy”) can only be given by
the musicians who know and play both instruments.
CASJ Vol. 4, No. 6 (Series II), November 2002
 Martin Schleske – Empirical Tools in Contemporary Violin Making: Part II. Psychoacoustic Analysis and Use of Acoustical Tools

Example: A television team of ‘Norddeutscher Rundfunk (NDR)’
recently filmed our studio and some details of our methods for a
documentary program. They also filmed a young German soloist
playing in the Münchner Max-Joseph-Saal on the Joseph Guarneri
del Gesu 1733 mentioned earlier, and on a “tonal copy” of that
instrument which had recently been made in our studio.
Immediately afterwards they recorded the owner’s opinion, which
was “Both instruments have the same qualities. But it is not just
that. I have played quite a number of ‘Strads’ and ‘del Gesus’, but
never have I had such a comparable feeling and tone as between my
‘del Gesu’ and that new violin.”
As shown in Fig. 2a the two resonance profiles are not identical.
But it has been possible to recreate some typical tonal characteristics.
These are especially:
• The quality (eigenfrequencies, ratios between levels) of corpus
resonances. Modes below 500 Hz clearly show higher radia-
tion than the higher-frequency B1-mode which for both violins
is around 540 Hz.
• Gaps between resonance regions.
• A general similarity of the “envelope shape” of the resonance
profile.
Comparing the overall loudness of both instruments as a function
of musical tones, Fig. 10b shows the slightly higher values of the
“tonal copy”. This figure reveals the dynamical reserves of both
instruments. The maximum differences of loudness between the
musical tones are 10 phon for the original and (because of a slightly
higher radiation of the T1-mode) 12 phon for the copy.
shows that the acoustical differences between the two violins by
Stradivari are not inconsiderable compared with the differences
between Stradivari (Fig. 2c) and tonal copy (Fig. 2d). As an indication
of the “dynamical reserves”, the (comparable) overall loudness of
both instruments is given in Fig. 10c. A comparison of the specific
loudness differences between Stradivarius and tonal copy (Fig. 7)
shows that the tonal copy accentuates the middle frequency groups
(14-24 ERB) even more than the original Stradivarius – an
accentuation which seemed to be typical for Stradivarius as
compared with Guarneri (see above).
Example: Since the tonal copy of the Guarneri seemed to be a
successful attempt to emulate typical Guarneri characteristics, it is
not too surprising that a comparison of the specific loudness
differences between the two tonal copies (Fig. 8) show the same
“genus-differences” as those between the two original violins:
compare Fig. 4 (original) with Fig. 8 (tonal copies). Musicians
recognized each instrument immediately as being typical violins of
the respective “genus”. A major factor in achieving successful results
when making “tonal copies” is the parameter studies mentioned
earlier. Empirical studies of the shift of eigenfrequencies and
modifications of mode shapes show the advantage of modifying the
corpus rather than the free plates [17, 18].
Example: Such parameter studies can broaden one’s experience
about which modifications will be successful when aiming to
implement a certain tonal task. As an example in the working process
the thickness graduation is modified in various steps and the
acoustical consequences documented. Fig. 14 shows the
modifications of the back plate of a developing “tonal copy” (the

Example: In order to judge the magnitude of differences between

Figure 14. Thickness modifications in the working process. The
these two violins, Fig. 2b shows a comparison of the resonance
grey values represent the amount of further thinning (max. 0.8mm)
profiles of the Stradivarius 1712 (“Schreiber”) and the Guarneri del
of the plates.
Gesu 1733. A general difference in the frequency range of the corpus
resonances is obvious. The frequency difference between T1 and B1
mode is smaller for the Stradivarius. Not only are these modes
pulled further apart in the Guarneri, but the latter also shows a
higher radiation of the lower frequency T1 mode compared with
the higher B1 (for detailed explanations of these modes and
description of their mode shapes, see Part I).

Example: Similarly revealing could be a comparison of the resonance

profiles of two violins by Antonio Stradivari (1712 and 1727) as
given in Fig. 2c. In the lower frequency range (particularly around
the corpus resonances 400-600 Hz) there occur considerable
differences which support the idea that even Stradivari did not copy
himself (at least this is true for the present-day acoustics of those
instruments, but it says nothing about how much they have been
modified over the centuries). Nevertheless, clear similarities occur
in the higher frequency region. This is also obvious from the loudness
differences of musical tones in the higher ERB-regions (Fig. 6). It

CASJ Vol. 4, No. 6 (Series II), November 2002 59

Martin Schleske – Empirical Tools in Contemporary Violin Making: Part I. Analysis of Design, Materials, Varnish, and Normal Modes

Figure 15. Effect of thickness modification (fig. 14) on the Figure 16. Measuring sound radiation on a turntable tripod by
„Resonance Profile” and thus on the change of psychoacoustic means of an impact-hammer-pendulum.
perception (fig. 9). Corpus resonances have changed their ability to
radiate sound. A0 and C2 increase their radiation level. Increase of
levels and level-differences in the brilliance region and thus increase
of brilliance and modulabilty (explanations see text „resonance
damping and vibrato”).

darker the area, the more wood was taken away, see scale). The
psychoacoustical effects of the modified resonance profile are given
in the loudness-difference diagram (Fig. 9). The effect is particularly
noticeable in an increasing specific loudness in the lower critical
bands up to 14 ERB and for the higher musical tones in an increase
above 28 ERB. A comparison of the resonance profiles before and
after the named modification is given in fig.15. It shows the changes
of radiation levels and eigenfrequencies of single modes: Particulary
the increased radiation of both the Helmholtzresonance A0 (by
3.5dB) and the torsional corpus mode C2 (by 6dB) is obvious.
Furthermore the eigenfrequencies of the T1- and B1-corpus modes
appear to be slightly shifted down. The radiation level of the T1-
corpus mode is reduced by 2 dB

Although such studies can undoubtedly be helpful, the use of the

“empirical tools” described here (Fig. 16) will in our opinion never
replace the “normal” working processes of the violin maker, which
will still be characterized by many “trials and errors”, nor will it
replace the emotions during the working process of making an
artistic object. An advantage of those acoustical tools might be that
they allow one to gain more information from the inevitable “errors”.
Quite often a whole series of instruments that are intended to follow
the resonance profile of one particular reference instrument will be
constructed before one may reach a result that could justifiably be
called a “tonal copy”.
with this Stradivarius resulted. The projection of that tonal copy
had to be demonstrated in the soloistic situation, when during the
last concert period its owner, the leader of the Philharmonisches
Orchester Ulm, played Béla Bartók’s concerto No. 2 “against” the
orchestra on that new violin. He states, “Although I have played a
nice 18th century Italian fiddle for the last seven years, it is only
since I began to play this “young Strad” that people have come to
me after the concerts to ask about the instrument I was playing.”

In my opinion the immediacy of such a success still has a lot to do
with how emotionally one “understands” the reference. With
Guarneri del Gesu it was the very first (out of a two-year series) that
seemed to score a hit comparing it with the 1733-genuine ‘del Gesu’,
while a troublesome series of several violins following the
“Schreiber” Stradivarius model had to be made (trying numerous
material and construction modifications) until a strong relationship
Obviously it is not important to play a violin made in the 18th
century, but rather one which may be called a successful “resonance
sculpture”. What a blessing that we may benefit from some of the
outstanding sculptures that already exist! Empirical acoustical
methods can be tools for collecting evidence. The deciphering of
fascinating acoustical ‘secrets’ of outstanding instruments can yield
a vast store of experience. I believe we can carry on where they left

60 CASJ Vol. 4, No. 6 (Series II), November 2002

 Martin Schleske – Empirical Tools in Contemporary Violin Making: Part I. Analysis of Design, Materials, Varnish, and Normal Modes
off. That is why (after some preliminary exercises) currently we try
to create violins that combine some beloved essential attributes of
two of our reference violins: the soloistic power and passion of the
‘del Gesu’ (1733) and the unsurpassed richness of varying colors,
depth and warmth of a Domenico Montagnana (1729).
CONCLUSION
One of the major messages of my work is that the terms “tone” and
“acoustics” must not be confused. Acoustics is characterized using
physical language (“mode shapes”, “eigenfrequencies”, “sound
pressure”, etc.), whereas tone is an aesthetic, rather than a physical,
quantity. Aesthetics is not a discipline of physics but belongs to the
field of art and philosophy. Although “tone” has its origin in
acoustical processes, the sensation and quality of tone cannot be
described using physical terms. Acoustics demands an intellectual
attitude, while tone demands instead an existential realization. If
tone is thought of allegorically as a painting, then acoustics is
comparable with the colors of that painting. Of course the painting
consists of a certain distribution of colors, just as tone consists of a
certain distribution of sound pressure. But it would be absurd to
claim that the aesthetic quality of the painting and its artistic content
would be understood by a pure analysis of the frequency distribution
and the spectrum of its colors. If the musician is a painter and the
composition his motif, then the resonances of the instrument are
the colors that he can use to express himself. The art of violin making
is to provide an instrument whose palette of colors allows the
musician to express what he feels and hears. The resonance profile
of the instrument is nothing but this palette. I hope the examples in
this paper have thrown some light on this palette.
Martin Schleske, München, den 27. Feb. 02
ACKNOWLEDGEMENTS
I would like to thank B.C.J. Moore and B.R. Glasberg for making
available their loudness calculation program “loudaes” which is the
basis for the fifth step of our method. Jim Woodhouse and Jeff Loen
provided editorial assistance. Some parts of this investigation were
possible by the help of a research project of the European Commission
(for details see: http://www.schleske.de/06geigenbauer/ecproject.pdf).
REFERENCES
[1] Weinreich, G., 1983, Violin radiativity: Concepts and
measurements: Proc. SMAC 83. Royal Swedish Academy of
Music, Stockholm, p. 99-109.
CASJ Vol. 4, No. 6 (Series II), November 2002
[2] Bredberg, G., Lindemann H.H., Ades, H.W., West, R.., and
Engstrom, H., 1970, Scanning electron microscopy of the organ
of Corti, Science, vol. 170, p. 861.
[3] Roederer, J.G., 1977, Introduction to the Physics and
Psychoacoustics of Music, Springer New York.
[4] Moore, B.C.J., Glasberg, G.R., and Baer, T., 1997, A model for
the prediction of thresholds, loudness, and partial loudness: J.
Audio Eng. Soc., vol. 45, no. 4, p. 224-240
[5] Glasberg, B.R. and Moore, B.C.J., 1990, Derivation of auditory
filter shapes form notched-noise data: Hearing Research, vol.
47, p. 103-138.
[6] Moore, B.C.J., and Glasbert, B.R., 1996 A Revision of Zwicker’s
Loudness Model: Acustica Acta Acustica, vol. 82, p. 335-345.
[7] Moore, B.C.J., 1995, Hearing: Academic Press, Inc., London.
[8] Woodhouse, J. 2002, Body Vibration of the Violin – What Can
a Maker Expect to Control?: J. Catgut Acoust. Soc. vol. 4, no.
5 (Series II), p. 43-49.
[9] Gough, C., 2001, Physical aspects of the perception of violin
tone: in Bonsi, D., Gonzalez, D., and Stanzial, D., (eds.), ISMA
2001, Proceedings, International Symposium on Musical
Acoustics, Perugia, Italy, published by Fondazione Scuola di
San Giorgio, Venezia, p. 117-122.
[10] McIntyre, M.E. and Woodhouse, J., 1974, Toward a
psychoacoustically realistic violin physics: Catgut Acoustical
Society Newsletter 22, p. 18-19.
[11] Weinreich, G., personal communication, Ann Arbor 2000.
[12] Woodhouse, J., personal communication, Cambridge 2002.
[13] Woodhouse, J., 1998, The Acoustics of “A0-B0 mode matching”
in the violin: J. Acoust. Soc. Amer. vol. 84, p. 947-956.
[14] Hutchins, C.M., 1985, Effect of an air-body coupling on the
tone and playing qualities of violins: J. Catgut Acoust. Soc.,
vol. 44 (Series I), p. 12-15.
[15] Rodgers, O.E., On the function of the violin bass bar, J. Catgut
Acoust. Soc., vol. 3, no. 8 (Series II), p. 15-19.
[16] Rodgers, O.E. and Anderson, P., 2001, Finite element analysis
of a violin corpus: J. Catgut Acoust. Soc., vol. 4, no. 4 (Series
II), p. 13-26.
[17] Schleske, M., 1996, Eigenmodes of vibration in the working
process of a violin: J. Catgut Acoust. Soc., vol. 3, no. 1 (Series
II), p. 2-6.
[18] Schleske, M., 1996, On making ‘tonal copies‘ of a violin: J.
Catgut Acoust. Soc., vol. 3, no. 2 (Series II), p. 18-28.
61
 SOUND WAVES
Man and the
machine
Your office computer can be easily transformed into a vital
sound-analysis tool. By Joseph Curtin and Martin Schleske
Most violin shops have a computer in their office. Can it
double as a workshop tool?
As most computers have soundcards they stand
ready to process audio signals. With the addition of a
microphone and about around €100 of sound-analysis
software we now have a piece of workshop equipment
of the same day-to-day usefulness as a thickness
caliper. Violin makers interested in plate tuning can
determine the tap-tones of a free plate by holding it
near the microphone and tapping the appropriate anti-
nodes; after a few keystrokes or cursor movements, the
frequencies can be read off the monitor. This method
does not show the mode shapes – the Chladni patterns
made visible by the loudspeaker-and-glitter method –
but it is considerably faster and requires neither ear
protection nor a broom to sweep up glitter.
It is equally simple to assess the fingerboard and
Helmholtz resonances, with a view to matching them –
you just hold the instrument near the microphone and
tap the back with your knuckle. The Helmholtz reso-
nance – the lowest air mode, also known as A1 – is the
broad peak on the chart appearing around 286Hz.
‘Plucking’ the end of the fingerboard with your thumb
yields a smaller peak above or below the air resonance,
assuming they don’t already coincide. Many simple
352
diagnostics can be performed in this manner. For
example, it became clear from a few taps and thunks
that a slow-rising, rather annoying resonance on the
open G of a certain violin was a tailpiece resonance
(see figures 1 and 2), which was then lowered to
unobtrusiveness by the installation of a heavier fine
tuner on the E string.
An instrument’s wolf note is most often caused by
the main-corpus resonance – the so-called B1 mode
appearing around 500–550Hz for violins, 170–190Hz
for cello (see figure 3). The bass-bar region is the most
sensitive area for this resonance and tapping here
will yield its exact frequency and also gives us an idea
of the effect of our adjustments to the soundpost,
fingerboard or tailpiece.
We have seen how some general features of violin
frequency response affect both projection and tonal
balance. Old Italian violins, in particular, tend to
concentrate their energy in two distinct frequency
regions and this sets them apart from other violins, at
least on a statistical basis. In the light of this, the ability
to measure frequency response in the workshop takes
on a real value. Instruments can be compared in terms
of overall power or by using ‘quality criteria’, such as
those discussed in the October 2003 issue, or in
The Strad April 2004
 SOUND WAVES

ABOVE hitting the side of

the bridge with a small
hammer is a simple way
to excite the violin across
a broad frequency range

countless other ways. At another level, the coordination
of the objective data with the subjective experience of
playing and listening provides a kind of ear training,
enabling you to listen to violin sound in a more
analytic fashion. Plotting frequency response using
computer-generated coloured maps lends an addi-
tional, visual sensation to the experience of listening to
violin sound and this helps sharpen what is surely a
maker’s most important tool – tonal intuition.
To measure frequency response, we need to do at
least four things: excite the violin across its frequency
bridge is normally driven artificially – by everything
from bowing machines to record-cutters. Ideally, the
driver should be able to excite measurable amounts of
sound over the violin’s entire frequency range, without
coupling significant amounts of mass to the bridge,
which would effectively mute the instrument and skew
the results. The driver is typically fed with a computer-
generated signal, which can take the form of a sine
wave, swept across the frequency range in a kind of
continuous glissando. Alternatively, a mixture of all
frequencies, such as white noise, can be used.
▲

range; pick up the resulting sounds; compare the

excitation forces with the sound output; and deal with
Figure 1
the acoustics of the room in which the measurements 70
were made. Each of these can be addressed in a variety
Wolf note
of ways; indeed, researchers often build equipment resonance
All photos and diagrams: Martin Schleske
Log Magnitude

and write software specifically suited to an individual

experiment. Here we will consider only equipment and
Tailpiece
software that is commercially available, reasonably resonance
priced and suitable for a variety of uses.
When a violin is played, the vibration of the bowed
string drives the bridge in a predominantly side-to-
side motion. Scientific experiments require a repeatable -10
0 2,500
and consistent means of excitation, but players rarely Frequency (Hz)
play a bow stroke exactly the same way twice, so the

The Strad April 2004 353

 A third approach, known as impulse Figure 2
excitation, is to deliver all frequencies at
once in the form of a sudden blow.
Imagine a small hammer, the handle
made of six inches of soundpost stock,
the head from a piece of hardwood
dowel weighing several grams. If the
head of the hammer were infinitely hard,
the energy of its blow would be equally
distributed across all frequencies; in
practice, the relative hardness of the head
and the surface it is striking limit high-
frequency excitation. This is easily
demonstrated by tapping a table top,
first with your fingertip and then with a
house key. The brighter click is due to the
greater excitation of the table’s high-
frequency modes and is made by the key.
If we tap the side of the violin bridge
with our wooden hammer, the resulting
sound – a kind of resonant thud – will
contain contributions from all of the
instruments’ resonances, or at least those
Figure 3
which radiate sound, can be excited by a
side-to-side motion of the bridge and are
within the hammer’s frequency range.
The difficulty with this approach lies
in achieving the same result twice.
Hitting the bridge a little harder or in a
slightly different place, or from a slightly
different angle, will excite a somewhat
different response. This consistency can
be increased by building a jig to hold the
instrument and another to hold the
hammer. The hammer can be suspended,
for example, in the manner of a pendu-
lum, then lifted to touch a reference bar
of fixed height before being released to
strike the bridge. The blow needs to be
strong enough to produce sounds that
are well above the level of ambient noise,
yielding a good signal-to-noise ratio, but
not so strong as to knock the bridge out
of position. In practice, a blow with
roughly the energy of a robust pizzicato
note is quite sufficient. Provided the jigs
are used in consistent ways, much can be learned the tip of the hammer during impact. This signal TOP the movement of
a resonating tailpiece
from these kind of relative measurements. If, however, allows the computer to subtract the frequency
you want to compare your results with those of characteristics of the hammer-blow from the total ABOVE the main-corpus
or B1-mode resonance,
professional researchers or colleagues using different acoustic measurement. Commercial impact hammers which occurs at around
equipment, it may be worth investing in an are calibrated – that is, a given force at the tip will 500–550Hz in violins.
Tapping the region
impact hammer. create a signal of specified magnitude. In this way, above the bass-bar will
Impact hammers come in a variety of sizes and are measurements made by different hammers, or indeed yield its exact frequency

widely used for vibration testing – in everything from
car parts to aeroplane wings and suspension bridges. A
piezo crystal embedded in the hammer head creates an
electrical signal relative to the rapidly changing force at
by other calibrated excitation methods, can be
meaningfully compared. The following page shows a
PCB Piezotronics impact hammer suitable for violin
testing. At roughly €1,000, it is not cheap. Still, it is a
▲

The Strad April 2004 355

 professional tool suited to a wide variety of tests
and will last a lifetime (see the contacts box for
other suppliers).
A microphone is the obvious way to pick up sound,
and surprisingly clean, relative readings can be made
with inexpensive ones; the bumps in their frequency
response tend to be smooth and relatively small
compared with those of a violin. Most professional
sound recording and measurement is done with electret
or ‘condenser’ microphones, which require phantom
power, a 48 volt DC supply usually provided by the
pre-amplifier. Computer soundcards don’t typically
provide phantom power, but external devices that do –
ABOVE the PCB including sound card/pre-amp combinations – are
Piezotronics impact
hammer has a sensor
widely available. Some microphones are designed to
in the tip which can make particular voices or instruments sound good,
be calibrated to allow
accurate measurements
claiming to add ‘warmth’ or any number of other intan-
gibles. For measurement purposes, however, we want a
neutral microphone with a frequency response that is as
flat as possible. Fortunately, calibrated microphones are
available at reasonable prices (several hundred Euros
and up). These come with their own frequency-
response chart, which can be put into a microphone
compensation file in the software, allowing measure-
ments of absolute, rather than relative, sound pressure.
Microphones are also sensitive to sound direction:
omnidirectional ones pick up sound from all directions
with about equal sensitivity; whereas cardioid micro-
phones, by contrast, mainly detect sound from a single
direction. This can be useful when measuring in normal
listening environments as some of the reflected sound is
ignored by the microphone. The directionality comes at
the expense of bumpier frequency response and the
microphone has to be calibrated before it can be used
for measurement purposes.
Once we have managed to excite the instrument’s
resonances with an impact hammer, even a highly
trained ear would have difficulty identifying more
than a few of these resonances. Fortunately, in 1822
Jean Baptiste Joseph Fourier, a French mathematician
and former classmate of Napoleon, developed a
mathematical tool that today’s engineers use probably
more frequently than any other. The Fourier
Transform yields analysis of any wave form in terms
of its constituent sine waves, the basic units of sound
analysis. The Fast Fourier Transform (FFT), a version
356
specially adapted for computers, is the core of most
acoustical software.
Separating the acoustics of a violin from the acoustics
of the room in which it is tested is a significant chal-
lenge. Many researchers use anechoic chambers, whose
lack of reverberation allows for clean measurements.
A small shop-built anechoic chamber is not unfeasible,
but, lacking this, the reflected sound can be reduced by
building an absorbent shell that partially encloses the
test violin. Another shell could be built around the
microphone, or a directional microphone can be used.
Large, acoustically dead rooms are more suited to
acoustic measurement than small, live ones. Another
approach is to average out the room’s acoustics by
taking readings in a number of different positions
around the room. We will examine this and the practi-
calities of a custom anechoic chamber in the next article.
Acoustic research is a fascinating undertaking, but
also a demanding and often frustrating one. If you
watch an acoustics experiment in progress, you will
probably conclude that researchers spend most of
their time finding the right cables, trouble-shooting
equipment and fixing software glitches. Computers,
originally designed as willing drones, turn out to be
temperamental creatures, ill at ease with the tasks we
assign them. Violin makers new to the field are likely to
make as many mistakes as acousticians building their
first violins – everything from runaway purfling to
leaky varnish. A relationship with a professional
researcher is therefore invaluable and organisations
such as the Catgut Acoustical Society, now part of the
Violin Society of America (VSA), were founded to
bring makers and researchers together. The VSA-
Oberlin Acoustics Symposium is a week-long summer
workshop that provides both theoretical and practical
support for violin makers. S
CONTACTS
The Violin Society of America
Tel: +1 845 452 7557 www.vsa.to
Impact Hammers:
Kistler Instrument Corporation
Tel: +1 716 691-5100, fax: -5226, www.kistler.com
Dytran Instruments, Inc.
Tel: +1 818 700-7818, fax: -7880, www.dytran.com
PCB Piezotronics
Tel: +1 716 684-0001, fax: -0987, www.pcb.com
Microphones:
Bruel & Kjaer
Tel: +454 580-0500, fax: -1405, www.bksv.com
You can also locate equipment suppliers by visiting
www.globalspec.com or www.directindustry.com
The Strad April 2004
 SOUND WAVES

A world
apart?
Are there tonal differences between
old Italian violins and those built at Frequency response function
other times in other places? Can these The frequency response function (FRF) is so fundamental to violin research
differences be measured? In the last that it is helpful to become familiar with its appearance. The first thing you
article (October 2003) we saw how notice about figure 1 is the multitude of jagged peaks – several hundred in
a violin’s ability to radiate sound in all. Those to the left, at the low frequency end, are fairly distinct and well
spaced, but looking to the right we can see they overlap more and more
two particular frequency regions
until they form a kind of spiny continuum. Note that there is almost no
determined its projection. In this
sound radiated at 196Hz, the fundamental of the open G string. The first
article we will examine the concept
significant peak occurs at 275Hz, very near to C sharp: this is the Helmholtz
of frequency response – how it is
or f-hole resonance. The prominent peak about an octave higher, created
measured and plotted, and how it
by a resonance of the violin body, is most often responsible for wolf notes.
can be used to address some of the
Standing back and squinting, the strongest resonances fall in two broad
questions that have haunted violin
clusters or formants – the first between about 250Hz and 1,000Hz, the
making for the past two centuries. second between about 2,000Hz and 4,000Hz. This general shape turns
Imagine a line tracing the amount out to be characteristic of old Italian violins.
of sound made by a violin as it is
swept through its frequency range Figure 1
5
in a kind of electronically induced 2x10
glissando. While the typical note-to-
note unevenness experienced by
Amplitude (Pa/N)

players suggests the line might not be

straight, there is little to prepare you 5
10
for the jagged landscape observed
when acousticians plot the
frequency response.
Frequency response is an account
0
of how much sound is radiated at 0 5,000 10,000
each frequency in the instrument’s Frequency (Hz)
range, for a given input at the bridge.

16 The Strad January 2004

Joseph Curtin and Martin Schleske quantify some of
the sound characteristics that distinguish classic Italian
instruments from the rest of the pack
A violin’s frequency response is highly complex, partly
because of the hundreds of resonances that contribute
to its many peaks and valleys. If we are to account for
all the radiated sound, then measuring a violin’s
frequency response is greatly complicated by the
instrument’s directionality at higher frequencies.
While low-frequency radiation is roughly equal in
every direction, at high frequencies the sound radiates
as beams whose angles shift even within a single
vibrato cycle. This presents researchers with a
problem: identical microphones placed in different
positions will pick up different signals – suggesting
different frequency responses.
Even a single microphone will receive different
signals if the bridge is driven from a different position
– indeed, the measurements will change if you simply
wait until the humidity alters. All this makes the
comprehensive charting of a violin’s frequency
response an exhausting enterprise. Instead, researchers
often rely on a more conditional, limited measurement
– a frequency response function (FRF). This can be
thought of as a single ‘snapshot’ of the instrument’s
acoustical characteristics – one particular view of a
three-dimensional landscape. Its meaningfulness
depends on how well the measurement conditions
are specified; its usefulness on how much light is
shed on the question at hand.
If the question is, ‘How do we distinguish old Italian
violins from other violins?’ then it is worth looking at
The Strad January 2004
the work of German physicist Heinreich Duennwald. ABOVE analysing the
response of the violin to
He measured a great variety of violins, from factory a range of vibrations at
ones to old Italians. He measured them in an anechoic the bridge is useful when
comparing instruments
chamber, driving them with an electromagnetic
transducer designed to rock the bridge from side to
side, in much the same way as a bowed string does.
A computer-generated signal was fed into the
transducer and the resulting sounds from the
instrument were picked up by a microphone and
fed into a computer for comparison with the original
signal. The ratio of the output signal to the input
yields the frequency response function.
Instead of dealing with the fine details of several
hundred curves, Duennwald divided each of them
into six distinct frequency regions, then looked at the
average levels in each. This is a fairly common strategy
among acousticians – the ear tends to hear these
▲
Figure 2
log Amplitude (dB (ref 1Pa/N))
120
110
100
Photo: Clarissa Bruce
90
80
0 1,000 10,000
Lower fundamental frequency of 1/3 octave interval (log Hz)
17
regions independently, and the proportions of energy
in each band to some extent determine tone colour.
Home stereo systems with graphic equalisers take
advantage of this, allowing the overall tonal balance
to be adjusted by means of 1/3 octave filters. Here are
Duennwald’s frequency regions, along with the tonal
effects he ascribed to high relative levels in each sector:
Category Qualities
A 90–650Hz fullness and depth
B 650–1,300Hz nasal quality
C 1,300–1,640Hz brilliance and projection
D 1,640–2,580Hz brilliance and projection
E 2,580–4,200Hz brilliance and projection
F 4,200–6,400Hz harshness and inclarity
In our experience the nasal region extends
from about 1,200Hz to 1,600Hz, while the region
600–1,000Hz is important for the openness of the
sound, lending it an ‘aaah’ quality. If you want to
experience this yourself you can visit
www.schleske.de/hoerbeispiele/playlist10.html and
listen to a single musical example, artificially modified
for higher and lower values in each frequency region.
Note that while a high level of B, for example, might
create an unpleasantly nasal sound, too low a level can
lead to an equally undesirable overcast quality. This is
true for the other regions: balance is everything.
Precise subjective attributes, however, are not crucial
to Duennwald’s results. Rather, he was concerned
with comparing violins, using the
measured values for each region, along with the
strength of the the violin’s lowest radiating resonance,
the so-called Helmholtz or f-hole resonance. He first
formulates the following quality parameters:
● high relative level of the Helmholtz resonance
● high percentage of un-nasal notes versus nasal notes
● high percentage of clear notes versus unclear notes.
These parameters were applied to each violin on a
note-by-note basis, making allowances for some of the
characteristics of the bowed string and the human ear.
Violins which satisfy these parameters have a strong
Helmholtz resonance and produce relatively little
sound in regions B and F. Duennwald tested 700
violins, comprising 53 old Italian instruments, 75 by
pre-1800 masters such as Klotz and Stainer, 42 by
hobbyists, 300 by post-1800 masters and 180 factory
instruments. In each category we can see the
percentage of instruments showing good values
for all quality parameters.
Category % or lowering the entire curve until the highest peak in
Old Italian 92.5
Pre-1800 masters 30.7
Hobbyists 26.2
Post-1800 masters 19.1
Factory 8.4
The Strad January 2004
Figure 3
5
2x10
Amplitude (Pa/N)
5
10
0
100 1,000 10,000
Frequency (log Hz)
Plotting a frequency response function
Amplitude is plotted on the vertical or y-axis, using either a linear or logarithmic
scale. Plotted linearly, as in figure 1, a doubling of the signal doubles the
height of the curve. In figure 2, amplitude is plotted logarithmically using
the decibel scale (see October 2003, p.1107) and this produces a more
compact graph that roughly mimics our subjective sense of loudness. Linear
scales, by contrast, highlight small differences in amplitude. Frequency is
plotted horizontally, on the x-axis. Again, a linear scale can be used, where
each 1,000Hz increment is given equal space, or a logarithmic one
(see figure 3), where each octave is given equal space – as it is on the
piano keyboard. The many ups and downs of an FRF can be smoothed to
emphasise the overall shape of the curve rather than individual details.
It can also be plotted as a bar chart, where the height of the bar is the
average amplitude across, say, each 1/3 octave band (see figure 2).
Clearly, violins from all categories make the grade,
though none so consistently as the old Italians. In
terms of actual numbers, and given the enormous
quantity of them in existence, factory violins satisfying
all the parameters far outnumber old Italians, assuming
Duennwald’s instruments are representative. And it
is hard to know what to make of the violins built by
hobbyists. Who are these makers who do so signifi-
cantly better than the masters after 1800? Duennwald,
sadly, does not say.
What conclusions may be drawn from his results? If
one can extrapolate from the quality parameters back
to perceived tone, old Italian sound seems to combine
fullness and depth along with brilliance, clarity and
projection. In terms of frequency response, this implies
a broad peak below 1,000Hz and another centred at
about 3,000Hz. This nicely satisfies the two paradigms
for projection we looked at in our last article.
There is a need for caution, however. Duennwald
does not compare the total amount of sound each
instrument produces. This information is lost when
he normalises the curves to an arbitrary value (raising
band B reaches 25dB). For this reason, a violin could
satisfy all Duennwald’s quality criteria and yet be
almost inaudible!
The concept of old Italian sound is necessarily an
abstraction from generations of playing and listening
▲
19
 Phase difference Figure 4
A comparison of two oscillating 110
signals, in this case the input and the

log Amplitude (dB)

output, phase difference measures
the difference between their
respective positions on the 90
peak–trough–peak cycle, as a
proportion of the whole cycle. This
lag or lead is expressed as an angle
between -180º and 180º, where the
70
midpoint, zero, is perfectly in phase
and 180º or -180º is completely out 180
of phase.
Imagine two identical pendulums 90
Phase (º)

one behind the other. When they are

moving as one, they are said to be 0
in phase. If they are both at the
vertical and one pendulum swings -90
to the right while the other swings
left, they are 180º out of phase. If,
-180
however, one pendulum reaches 100 1,000 10,000
the top of its swing just as the other Frequency (log Hz)
reaches its lowest position, there is Figure 5
a difference of 90º. 110
For violin measurements, you can Phase
0º 60º 120º 180º 240º 300º 360º
compare the side-to-side oscillation -180º -120º -60º 0º
log Amplitude (dB)

of the bridge with the in-and-out 100

motion of the microphone
diaphragm. Phase difference
requires a separate graph, usually
90
placed directly beneath the FRF, as
in figure 4. The angle is plotted on
the y-axis and the frequency on the
80
x-axis. This data can be displayed on 0 500 1,000
a single chart (see figure 5). Frequency (Hz)

experience. It is complicated by the vast differences

Anechoic chambers
in quality among the old Italian instruments them-
Heinreich Duennwald, like many acousticians, performed his experiments selves, which range from the barely playable to the
in an anechoic chamber – literally, a room without echoes. Sophisticated sublimely expressive. And if there is such a thing as
sound-absorption techniques eliminate almost all reflection off the interior old Italian sound, can we assume it is measurable in
surfaces, thus creating an acoustically ‘dead’ space, whereas all normal an empirical fashion? Can the beauty of a Matisse
listening environments are reverberant to some extent. This reverberation adds painting be understood by studying the way light
to the sound emanating from the instrument, leaving researchers with the bounces off the canvas? Research like Duennwald’s
problem of disentangling the two. is valuable precisely because these questions are so
Yet banishing reverberation creates its own complications. The directional difficult to answer or even formulate in meaningful
patterns of the sound radiating from the violin vary sharply with frequency, at least ways. Duennwald demonstrates the existence of a
above 1,500Hz or so. In typical listening environments the portions of the sound statistical community of violins linked by specific,
that travel away from the listener are effectively ‘gathered and sent back’ by the measurable parameters. These are at best only
surface reflection, thus providing the listener with a fairly complete tonal effect. partial definitions of old Italian sound – still, if one
In anechoic chambers, however, sound not heading directly towards the micro- is building violins today and the violins happen to
phone is not heard. Thus, for any given microphone placement, some of the show high levels in regions B and F, and they are
radiated sound is lost. At high frequencies even a 50mm shift in microphone proving difficult to sell, then Duennwald’s
position can a produce dramatically different FRF curve. This problem can be parameters seem a useful point of reference.
solved by surrounding the violin with numerous microphones or by rotating the With this in mind, we will next consider practicable
instrument in relation to a single microphone and averaging the readings. ways in which frequency response can be measured
in the workshop. S

20 The Strad January 2004

 SOUND WAVES
Domestic
bliss
In the last article in the series Joseph Curtin and Martin Schleske
look at affordable ways to transform your home into an acoustics lab
RIGHT AND OPPOSITE One way to evaluate the sound
you can objectively
evaluate the sound of a of a violin is to give it to a good
violin with measurement violinist then sit back and listen. The
systems like this one,
which is made of wood trouble is that good violinists tend to
and aluminium tubing make almost any instrument sound
good – it’s their job. Instinctively,
they compensate for unevenness,
coax out hidden reserves of power,
smooth over rough spots and bring
everything together with vibrato.
They impose their own voice on an
instrument, so that what you hear is
not so much the violin as violin-
playing – a kind of magic trick they
do, using the instrument as a prop.
Violin makers are used to dealing
with all this, but most would
welcome some sort of objective
scale for violin sound – some way to
measure what it is that violinists
and their audiences most want from
an instrument. Earlier in the Sound
Waves series we saw that two
much-sought-after qualities –
projection and what might be called
‘old Italian tonal balance’ – are
largely determined by an instru-
ment’s ability to concentrate sound
into two particular frequency
regions. This ability can be assessed
All photos: Joseph Curtin
by measuring frequency response;
and here we look at measuring it in
a workshop setting and then
consider a way of plotting
frequency response so as to give a
696
quick sense of overall power and
tonal balance.
The pictures show a workshop-
built measurement set-up using an
impact hammer (see The Strad,
April 2004). It is built with plywood,
scraps of hardwood and aluminium
tubing – held together by bolts and
wing nuts. This makes it easy to
dismantle and transport, assuming
that it is often more feasible to bring
the equipment to a valuable violin –
in a museum or a soloist’s hotel
room, for example – than the violin
to the equipment. The laptop
computer runs SpectraPLUS on a
Windows 2000 operating system.
While almost any computer will do
the job, more recent vintages
provide better support for devices
using a USB interface – in this case,
the sound card.
Laptops are typically supplied
with very basic sound cards and the
inside of a computer is a noisy envi-
ronment, electrically speaking, so
external sound cards are generally
used for high-quality audio record-
ing and measurement. Fortunately,
external cards are widely available
in all price ranges. The one pictured,
a Creative MP3+, connects to the
USB port, which also powers the
device. The card does not provide
the phantom power needed by most
The Strad July 2004

condenser microphones, but other-
wise functions admirably – and at
around €35 the price is hard to beat.
Details of all the component manu-
facturers are given on page 700.
The microphone is by Earthworks
– a calibrated, omni-directional,
measurement microphone with an
impressively flat frequency
response across the violin’s range.
Because it runs on an internal
battery it does not need phantom
power, and its switchable 14dB pre-
amp gives the signal a sufficient
boost to feed directly into the line
input of most sound cards. The
microphone can be moved up and
down the support rod, which in
turn can be moved along the
aluminum rail, over a range of
about 600mm from the centre of the
instrument (here defined as a
The Strad July 2004
vertical line ascending through the
endpin). A self-adhesive scale on
the rail makes it easy to establish
the exact distance between the
microphone and the instrument.
The instrument itself is supported
from underneath and at the neck by
thin elastic thread, available from
sewing stores.
The impact hammer is the small-
est model made by PCB Piezotronics
and together with a power supply
and cables, it costs about €1,000.
The hammer has been mounted on a
ball bearing to ensure a free swing
along a consistent path. Flicking the
back of the hammer with your
fingertip will swing it towards the
violin bridge. If the pivot point is
placed about a centimetre behind
the hammer shaft, the hammer will
hang at about 30 degrees from the
vertical and ‘park itself’ clear of the LEFT this measurement
set-up is small enough
bridge after impact. In practice, to be used in a house
aligning the tiny hammer tip with or violin workshop
the corner of the violin bridge can BELOW the instrument is
be ticklish; in order to speed things supported by thin elastic
thread, to isolate it from
up, the whole assembly is attached the rest of the equipment
to an adjustable X–Y–Z mount,
using three linear slides from
Edmund Optics, costing about €100
each. The violin and the impact-
hammer assembly can be rotated
through 360 degrees with respect to
the microphone.
A microphone distance of around
500mm is large enough to avoid
most near-field cancellation, and yet
close enough that sound coming
directly from the violin is signifi-
cantly stronger than the sound
reflected by the room surfaces, at
least in a normally reverberant
room. Because of the violin’s
directional characteristics, a number
of readings must be taken around
the instrument in order to get a
reasonable estimate of the total
radiation. In an anechoic chamber, a
great many would be needed, but in
normal environments the room
reflections help moderate the
directionality. We have found that
six readings, taken at 60-degree
intervals as the instrument is rotated
with respect to the microphone, are
usually sufficient.
Room acoustics are probably the
biggest obstacle to getting clean
measurements. One way around
this is to build an anechoic chamber
– not unfeasible if you have the
space. Something the size of a
pantry or walk-in closet could work
for violins and violas. Most profes-
sional chambers use foam wedges to
absorb sound, but acoustic foam can
be prohibitively expensive.
Fortunately, there are products that
work very well at a fraction of the
cost. Owen-Corning type-703 fibre
glass panels are semi-rigid boards
available from building suppliers.
Four-inch thick panels attached to a
wall provide virtually perfectly
absorption to below 200Hz –
adequate for violins and violas,
remembering that almost no sound
is radiated below the frequency
697
 then average the results. It is true LEFT the measurement
set-up is mounted on a
that the overall amplitude of the trolley and can easily be
measurements will be increased wheeled out of the way

by the room reflections, but most

of the Helmholtz resonance. Cellos
and basses, on the other hand,
require larger chambers and thicker
absorbent layers. Master Handbook
of Acoustics by F. Alton Everest
offers an excellent non-technical
account of room acoustics and
much practical advice about the
treatment of rooms for a wide
variety of purposes.
Even without an anechoic
chamber, several things can be done
to reduce the effect of room
acoustics: absorbent panels or
acoustical tiles hung from the
ceiling can minimise reflections;
soft wall hangings and upholstered
furniture also help; alternatively,
absorbent materials can be hung to
form a tent around the measure-
ment area. As long as the apparatus
is kept in the same position in the
room, meaningful relative measure-
ments are possible. But perhaps the
most practical approach for violin
makers is to repeat the six-reading
measurement cycle in a number
of positions in the room and
of the peaks created by certain
room resonances can be smoothed
out this way. We have found that
20 measurement cycles, with both
the instrument and microphone
rotated 18 degrees with respect to
the room for each cycle, gives a
reasonably repeatable measurement
from room to room. This does
mean 120 separate readings, but
remember that an impact hammer
reading takes only a few seconds
and the software takes care of
the averaging.
It is easy to become discouraged
by the many obstacles on the way to
getting clean measurements –
or to fall into a never-ending
search for the perfect measurement
system. There is no such system.
It is simply a matter of finding the
kind of workable compromise that
allows you to keep moving forward.
While a complete measurement
system may be out of the range
of an individual maker, makers’
groups and associations might
invest in the equipment and
organise workshops on its use.
Figure 1 shows the frequency
response for the ‘Schreiber’
Stradivari violin, measured in a
workshop using the 120-reading
approach suggested earlier. The
graph shows a jagged line typical to
all violins. If you are looking for a
detailed account of how sound

Useful software
An acoustical-analysis program is needed to compare the impulse from the hammer with the signal from the
microphone and present the results graphically. A variety of programs can be downloaded free from the web. These are
mainly single-channel analysers and, though useful for plate tuning and many other measurements, two channels are
needed for impact hammer measurements – one each for hammer and microphone. SpectraPLUS is a mature
program for the PC platform. It is capable of a wide variety of measurements, including real-time spectrograms and
room acoustics, and costs about €500; Spectrafoo is a similar program for the Macintosh. WinMLS, a program written by
Norwegian researcher Lars Morset, was not available in its latest form when this article was going to press, but it
promises to be relatively inexpensive (less than €200) and comes with a number of pre-installed configurations for violin
measurements, including impact hammer, bridge tuning and plate tuning. Morset has also developed a system for
measuring frequency response by using a coil to drive a tiny magnet attached to the bridge – well worth closer
examination. Details of this can also be found on his website. Cool Edit from Syntrillium Software, for many years one of
the least expensive two-channel analysis programs, was recently acquired by Adobe Systems and is now available only
in a more powerful and expensive ($299) version called Adobe Audition.

The Strad July 2004 699

 Figure 1
100
Sound radiation (dB)
80
0
0 1,000
Figure 2
100
Sound radiation (dB)
90
80
202 339
Figure 3
110
100
Sound radiation (dB)
90
80
70
60
50
0 5,000
output changes with frequency, you
need look no further, but if you want
a quick sense of how the instrument
sounds, you may feel you have been
given a satellite photo when you
asked for a road map. The informa-
tion is there, but it is buried in the
detail. The quality parameters
devised by Heinrich Duennwald
(The Strad, January 2004) can be
used to extract useful information –
in fact to separate the old Italian
instruments from the others, at least
on a statistical basis. Duennwald
used some fairly sophisticated
analytical tools, however, and there
is a simpler approach.
Diagrams: Martin Schleske
Acousticians have long employed
third-octave bands when looking at
everything from room acoustics to
speech defects. While the human
ear is capable of the subtlest dis-
crimination in the pitch of notes,
700
10,000
Log frequency (Hz)
571 960 1614 2714 4565 7678
Log frequency (Hz)
Stradivari 1712
Guarneri ‘del Gesù’ 1733
10,000
Log frequency (Hz)
third-octave bands roughly capture
the resolution with which the ear
perceives tone colour – the width of
the brush-stroke, so to speak.
In Figure 2, the response curve
of the violin is plotted as a third-
octave bar chart, the height of each
bar representing the total energy
in each frequency band. Though
much detail is lost, we are able to
get a quicker sense of the general
distribution of energy.
The contribution of each
frequency region to tone colour is
indicated above the graph. When
several of these graphs are overlaid,
it becomes difficult to sort out one
from the other. For this reason we
prefer to plot the data as a series of
super-imposed smooth curves.
In figure 3, a smoothed third-
octave graph of the ‘Schreiber’
Stradivari is overlaid with that of a
1733 Guarneri ‘del Gesù’. Both the
Strad and the ‘del Gesù’ show the
two broad peaks typical of old
Italian violins – the difference
between their curves reflects their
individual tonal characters. It is a
mistake, however, to assume that
two instruments with identical
third-octave profiles would sound
the same. Firstly, the ear is highly
sensitive to sounds in the 2,000–
4,000Hz band, so small differences
of emphasis within this region
are likely to be tonally significant.
Secondly, this kind of representation
suppresses any sense of the ‘spiki-
ness’ of the response curve – the
very spikiness which contributes to
the amplitude modulation of
harmonics during vibrato (The
Strad, October 2003), an important
element of tone colour and, in all
likelihood, projection.
Science is still some distance from
a complete understanding
of the violin, but it already
offers violin makers many
unprecedented insights into the
inner workings of the instrument.
It also offers measurement tools
that enable meaningful and
relevant comparisons between
new instruments and those of
past masters. Should makers
today try to emulate the response
curves typical of the old Italians?
Should we try to match or even
exceed their overall power? Well,
it’s what good violinists want.
CONTACTS
Earthworks
www.earthworksaudio.com
PCB Piezotronics
www.pcb.com
Edmund Optics
www.edmundoptics.com
SpectraPLUS
www.spectraplus.com
Spectrafoo
www.spectrafoo.com
WinMLS
www.winmls.com
Adobe Audition
www.adobe.com
The Strad July 2004
 Corrections to Curtin/Schleske articles
“Can You Hear Me?” and “Domestic Bliss”

Can You Hear Me?

The Strad, October 2003

Figure 3 caption should read as follows:

“The ears sensitivity varies across its frequency range. The yellow, red, and brown curves
represent perceived loudness (in response to sounds of equal intensity) as a function of
frequency. Thus at pianissimo levels, to produce the same sensation of loudness at 100 Hz as at
4,000 Hz, it would take an increase in sound level of 22 dB. Note that the curves are normalized
to an arbitrary relative sound level of 0 dB at 4,000 Hz.
 Domestic Bliss
The Strad, July 2004

Figure 3: frequency in the center of the horizontal axis should be 1,000 Hz, rather than 5,000 Hz.
 Anmerkung: Der nachfolgende Artikel wurde als Festvortrag anlässlich des 100-jährigen Bestehens des Verbandes Deutscher Geigenbauer
(VDG) auf dessen Jahreshauptversammlung am 22.Mai 2004 in Wiesbaden gehalten.
ZEITGEIST UND GEIGENBAU
MEILENSTEINE AUS KUNST UND WISSENSCHAFT - EINE KLEINE ZEITREISE
Martin Schleske,
Geigenbaumeister und Dipl.-Physik-Ing. (FH)
M
it Blick auf die großen italienischen
Meister der Geigenbaukunst des 18.
Jahrhunderts herrscht heute die An-
sicht vor, jene berühmten Vorbilder hätten ihre
epochemachenden Werke „aus dem Bauch her-
aus“ geschaffen. Moderne Wissenschaft habe
es noch nicht gegeben, daher sei die Entwick-
lung der Geige eine rein ‚intuitive’ Angelegen-
heit gewesen. Diese - möglicherweise auch hier
mehrheitlich kultivierte - Anschauung möchte
ich im Folgenden in Frage stellen, indem ich
versuche, einen kleinen Abriß zum Thema Wis-
senschaft und Forschung im Bereich des Gei-
genbaues zu skizzieren.
BEGNADETE EMPIRIKER
Wer die Geschichte der Kunst und die
der Wissenschaft gleichzeitig liest, der kommt
mit Blick auf die Entwicklungsgeschichte der
Geige zur Überzeugung: Die berühmten Groß-
meister des Geigenbaues können sich unmög-
lich in ihrer Kunst eingeschlossen haben. Sie
haben sich nicht - fernab vernünftiger Überle-
gungen in ihr geheimnisvolles Werkstattkäm-
merlein -„eingeschlossen“. Vielmehr waren sie
hochgradig aufgeschlossen, offen für Metho-
den, Überlegungen und Entdeckungen von
Wissenschaft und Künsten ihrer Zeit – Archi-
tektur, Naturwissenschaft, Komposition und
Malerei. Der Klangkörper Geige als ein – wie
wir heute erkennen - hochgradig optimiertes
akustisches System kann nur in einem Milieu
begnadeter Empirie entstanden sein. Hier waren
großartige Empiriker am Werk, in deren
© Meisteratelier für Geigenbau Martin Schleske 2004. [www.schleske.de]
Selbstverständnis das Spannungsfeld aus Kunst
und Wissenschaft niemals einen inneren Wider-
spruch hätte darstellten können.
Die großen Meister der Geigenbaukunst
waren nicht nur Künstler. Sie waren zwei-
felsohne auch empirische Wissenschaftler.
Denn ganz offenkundig waren sie in der Lage,
aus „Versuch und Irrtum“ die richtigen Lehren
zu ziehen. Dieser „empirischen Kunst“ ver-
dankt die Geige ihre Entwicklung. Sie entwi-
ckelte sich nicht in einem Milieu spontaner
Willkür, sondern ganzheitlicher Intuition - zwei
Begriffe, die bisweilen undifferenziert verwen-
det werden. Intuition wird häufig mit Willkür
gleichgesetzt. Während jedoch willkürliches
Arbeiten nicht zwischen richtig und falsch zu
unterscheiden weiß, läßt sich das intuitive Ar-
beiten stets von einem inneren Erfahrungsreich-
tum leiten. Aus unermüdlichen Versuchen und
notwendigen Fehlschlägen wurden die richtigen
Lehren gezogen.
DAS ZEITALTER DER WISSENSCHAFT
Die Geige entstand in einer Epoche „wis-
senschaftlicher Revolutionen“, die die Erfah-
rungswelt der ersten Geigenbauer tief geprägt
haben muß:
• In den Kindheitsjahren Andrea Amatis ereignet sich
im Bewusstsein der Menschheit ein „geistiges Erd-
beben“. Kopernikus: Die Erde dreht sich mit den an-
deren Planeten um die Sonne (1512). Was bislang
gültig war, wird schwer erschüttert. Das Zeitalter der
Wissenschaft bricht auf.
• Nicolo Amati ist noch nicht geboren, da begründet
Johannes Kepler (1571 – 1630) die theoretische Ast-
Artikel „Zeitgeist und Geigenbau“ Seite 1
 ronomie. „Harmonices mundi“ - die „Weltharmonie“
mit dem dritten Keplerschen Gesetz der Planetenbe-
wegung.
• Sein Zeitgenosse, der Italiener Galileo Galilei (1564
– 1642), begründet die moderne Experimentalphy-
sik. Im Jahr 1589 – sieben Jahre später wird Nicolo
Amati geboren werden - erhält Galilei einen Lehr-
stuhl für Mathematik in Pisa. Berühmt geworden ist
seine „Goldene Regel der Mechanik“ mit dem me-
chanischen Energieerhaltungssatz (1594).
• Wir haben nicht einmal das Geburtsjahr Stradivaris
erreicht, als im Jahre 1640 der französische Priester
Mersenne die Schallgeschwindigkeit in Luft be-
stimmt. (Er tut dies durch einen Zeitvergleich zwi-
schen dem Blitz und dem Knall einer Kanone. Ein
ähnliches Experiment gelang Bacon bereits im Jahr
1600).
Unzählige große Wissenschaftler jener Zeit
wären zu nennen, die uns etwas vom Pionier-
geist des Wissenschaftszeitalters erahnen lassen
– jenes Zeitalter, in dem die Entwicklung der
Geige ihre Wurzeln hat. Einige weitere Beispie-
le:
• Dass Santorio (1561 – 1636) den modernen
Feuchtigkeitsmesser erfindet, gibt uns guten
Grund zur Annahme, dass dem Geigenbau eine
differenzierte Kenntnis über Größen wie Holz-
feuchtigkeit alltäglich war und er sich mit Fra-
gen nach günstigen und ungünstigen Materialei-
genschaften auseinander zu setzen in der Lage
war.
• Vergegenwärtigen wir uns, dass Isaac Newton
im Jahr 1672 - Stradivari ist Mitte Zwanzig -
das Sonnenlicht mit einem Prisma in die Spekt-
ralfarben zerlegt, dann kann man sich leicht vor-
stellen, dass Begriffe wie „Klangfarbe“ durch-
aus alltäglicher Sprachgebrauch der damaligen
Welt des Geigenbaues waren. Es war klar: Far-
ben haben naturwissenschaftliche Ursachen…
• Mit dem Paradigmenwechsel des neuen Welt-
bildes Anfang des 16. Jahrhunderts beginnt die
Entwicklung der Wissenschaft förmlich zu exp-
lodieren. So wird im Jahr 1662 die Royal Socie-
ty („Königlich Wissenschaftliche Gesellschaft“)
in England und vier Jahre später die Pariser A-
kademie der Wissenschaften gegründet
© Meisteratelier für Geigenbau Martin Schleske 2004. [www.schleske.de]
BEEINFLUSST DURCH MUSIK UND MATHEMATIK
I
hre wesentlichen Impulse erhielt die Geige
m.E. durch die Auseinandersetzungen mit
der Wissenschaft und Kunst ihrer Zeit. Ei-
nige interessante Wechselwirkungen zwischen
Entwicklungsschüben der Geige auf der einen-
und Meilensteinen der Musikgeschichte auf der
anderen Seite sind offenkundig:
• Claudio Monteverdi war der Erste, der die vierte
Lage (e3 Marienvesper, 1610) verlangte und
damit der Geige ein höheres Maß an spieltech-
nischer Virtuosität abverlangte als andere Kom-
ponisten seiner Zeit. Monteverdis Heimatstadt
war Cremona. Sicher darf man davon ausgehen,
daß Monteverdis spieltechnische Forderungen
angesichts der örtlichen Nähe zur Amati-
Dynastie sehr unmittelbar eine spieltechnische
Weiterentwicklung ihrer Instrumente angespornt
haben wird.
• Arcangelo Corelli schuf mit seinen 12 Violinso-
naten (Sonata a violine e violine o cimbalo) im
Jahr 1700 ein Standardwerk der Violinliteratur,
das bis ins 19. Jahrhundert hinein zum Grund-
studium jeden Geigers gehörte. Noch wichtiger
könnte das Jahr 1680 sein: Erstmals in der Mu-
sikgeschichte entsteht durch Corelli ein ‚concer-
to grosso‘. Damit hebt sich die Geige aus dem
Orchester heraus und wird (freilich vorerst nur
phasenweise) zum Soloinstrument. Dies muss
nicht nur für den jungen Stradivari eine musika-
lische Revolution gewesen sein. Er begann, sich
konsequent von Wölbung und Modell seines
Vorbildes oder Lehrmeisters Nicolo Amati zu
lösen und entwickelte sein „langes Modell“. So
stellte er sich der neuartigen musikalischen Her-
ausforderung.
• Noch stärker muß Torelli im Jahr 1698 die In-
novationskraft Stradivaris angestachelt haben.
In diesem Jahr begründet Torelli das virtuose
Violinkonzert! Stradivari reagiert auf diese re-
volutionäre Entwicklung durch ein neues Mo-
dell. Es beginnt das, was spätere Generationen
seine „goldene Periode“ nennen werden.
• Auch die Entwicklung des modernen Streichbo-
gens (durch Francois Xavier Tourte, Paris 1747
- 1835) wäre ohne die neue Violinschule eines
Leopold Mozarts oder Josef Haydns sicher nie
„erlitten“ worden. Wie zuvor die Konzertsitua-
tion neue Instrumente verlangt hatte, so war es
nun die Spieltechnik, die neue Bögen verlangte.
Artikel „Zeitgeist und Geigenbau“ Seite 2
 Die Blütezeit des italienischen Geigenbaues
wäre ohne das unmittelbare Reagieren auf mu-
sikalische Entwicklungen nicht denkbar gewe-
sen. Es bestand eine lebendige Wechselwirkung
zwischen der geforderten Virtuosität der Kom-
position und der ständigen, kreativen Offenheit
und Entwicklungsbereitschaft der damaligen
Geigenbaukunst.
KUNST UND GEOMETRIE
Aber auch Anstöße durch Mathematik und
Architektur haben ihre Spuren im Geigenbau
hinterlassen:
Eine der Renaissance-Architektur in Wöl-
bungen, Bögen und Kuppeln häufig zugrunde
liegende Kurvenform ist die sog. Zykloide.
Eine Zykloide entsteht, wenn ein Punkt auf
dem Umfang eines Kreises auf einer Geraden
abrollt. Liegt dieser Punkt im Innern des Krei-
ses, dann entsteht eine Zykloide „mit Hohlkeh-
le“. In jüngster Zeit wies Quentin Playfair
[Cremona’s Forgotten Curve, Zeitschrift The
Strad, November 1999] auf die beeindruckende
Übereinstimmung klassischer italienischer Gei-
gen-Wölbungsprofile mit Zykloidenkurven hin.
Auch meine eigenen Wölbungsmessungen
an typischen Cremoneser und Venetianer In-
strumenten zeigen, dass durch die Wahl des
Einstichpunktes und Raddurchmessers nicht
nur jede der fünf Querquintas gezeichnet, son-
dern auch der grundsätzliche Wölbungscharak-
ter festgelegt werden kann. So liegt typischer
Weise bei Amati das Zykloiden-Minimum in-
nerhalb des Geigenumrisses, bei Stradivari
i.d.R. im Bereich der Außenumrisslinie und
schließlich bei Guarneri ‚del Gesu‘ außerhalb
der Umrisslinie.
Die Konstruktionen und Zeichnungen von
Bögen und Wölbungen mithilfe der Zykloide
erfreuten sich aufgrund ihrer „harmonischen“
Ästhetik in der Renaissance-Architektur großer
© Meisteratelier für Geigenbau Martin Schleske 2004. [www.schleske.de]
Beliebtheit. Wenngleich die Form der Zykloide
bereits den alten Griechen bekannt war, fehlte
ihnen noch die Mathematik, sie exakt zu be-
schreiben. Der Zusammenhang zwischen Kunst
und Geometrie war in der Renaissance eine
Selbstverständlichkeit.
• So setzt sich etwa Albrecht Dürer (1471-1528) in
seinen drei veröffentlichten Abhandlungen über Ge-
ometrie, Befestigungskunst und menschliche Propor-
tionen auch mit den theoretischen Grundlagen seiner
Kunst auseinander. In seinen Büchern gibt er detail-
lierte Erläuterungen über geometrische Grundprinzi-
pien und Kurvenkonstruktionen und zeigt, wie diese
dem Künstler und Handwerker dienlich sein können.
In der darstellenden Geometrie ist er bis an die
Grenzen der damals bekannten Mathematik vorge-
stoßen und hat bedeutende Impulse für die Arbeiten
Galileis und Keplers geliefert.
• Die Brücke Ponte di Mezzo in Florenz geht auf
Zykloiden-Berechnungen Galileis (1564-1642) zu-
rück. Neben Galilei befassen sich im 17. Jahrhundert
eine große Zahl berühmter Mathematiker, darunter
Pascal (1623-62), Leibniz (1646-1716) und Newton
(1643-1727) intensiv mit der Mathematik der Zyc-
loide und ihrer physikalischen Anwendungen.
Von Albrecht Dürer bis Issac Newton (der
Zeit also vor Andrea Amati bis Antonio Stradi-
vari) gehörte die Zykloide zum Grundwissen
eines jeden gebildeten Menschen. Es ist nicht
denkbar, dass die Geigenbauer dieser Zeit bor-
niert oder isoliert an wesentlichen wissenschaft-
lichen Erkenntnissen vorbei gearbeitet haben.
Die Wölbung der Geige ist eines ihrer we-
sentlichen klangformenden Elemente. Am Bei-
spiel der Zykloide wird m.E. deutlich, dass die
Geige in ihrer großen Entwicklungs- und Blü-
tezeit alles andere als „aus dem Bauch heraus“
gefertigt wurde. Ihre Entwickler beherrschten
die geometrischen Grundregeln und Konstruk-
tionsmethoden ihrer Zeit und wendeten sie kon-
sequent an. Sie waren beeinflußt von den Ent-
deckungen der Mathematik und Physik, An-
wendungen der Architektur, der Kunst- und
Musikentwicklung.
Artikel „Zeitgeist und Geigenbau“ Seite 3
DIE FOLGEZEIT – DAS 19. JAHRHUNDERT
E
s scheint mir, als entwickelten sich Gei-
genbau und Wissenschaft in der Folge-
zeit eher auseinander. Im Geigenbau tritt
im Zuge der „industriellen Revolution“ an die
Stelle einer kreativen Auseinandersetzung mit
Impulsen aus Kunst und Wissenschaft mehr
und mehr das Bestreben, bewährte Modelle
durch rationellere Produktionsmethoden zu
kopieren. Im Bereich der Akustik könnte man
das 19. Jahrhundert als eine Epoche der Grund-
lagenentdeckungen bezeichnen. Unter den „Vä-
tern der Akustik“ sind sicher Chladni, Fourier
und Helmholtz zu nennen.
• Ernst Florens Friedrich Chladni (1756-1827) veröf-
fentlichte im Jahre 1787 seine Forschungsergebnisse
in der Schrift "Entdeckungen über die Theorie des
Klanges". Er machte durch Anstreichen mit einem
Geigenbogen die Eigenschwingungsformen frei ge-
lagerter Platten sichtbar. Wie die Skizze zeigt, wur-
den die Platten mit Sand bestreut, der (bei Wahl der
jeweils geeigneten Lagerung und Anstreichstelle)
durch die erzeugten Vibrationen von den Schwin-
gungsbäuchen zu den Knotenlinien wandert:
Quelle: www.av-medien.net/html/katalog/ akustik_1203.htm
Im 20. Jahrhundert wurde diese Methode (mithilfe
eines elektronischen Funktionsgenerators) vielfach
im Bereich des Geigenbaues beim sog. free-plate-
tuning angewandt. Die nachfolgende Abbildung
zeigt die damit sichtbar gemachten Schwingungs-
form eines Geigenbodens:
© Meisteratelier für Geigenbau Martin Schleske 2004. [www.schleske.de]
Quelle: www.uni-erfurt.de/ kommunikationswissenschaft
• J.B. Fourier zeigte 1822 in seiner ,,Théorie analyti-
que de la chaleur``, daß sich ein Vorgang beliebiger
Form, der sich nach einer Periode T wiederholt (z.B.
die Schwingung einer gestrichenen Geigensaite)
vollständig aus einzelnen harmonischen Frequenz-
komponenten aufbauen läßt. Die Grund- und Ober-
schwingungen nennt man Fourierkomponenten.
Auf ihn geht die sog. Fouriertransformation zu-
rück - ein Algorithmus, anhand dessen Signale vom
Zeitbereich in den Frequenzbereich transformiert
werden können. Die Fourriertransformation läßt
sämtliche Frequenzkomponenten sichtbar werden,
aus denen zeitliche Vorgänge bestehen. Durch den
Einsatz des Computers erhielt in den 1970er Jahren
der 150 Jahre zuvor von Fourier entwickelte Algo-
rithmus erstmals „flächendeckende“ Anwendung.
Die sog. Fast Fourier Transformation (FFT) ist heute
vermutlich die am häufigsten angewandte Methode
der Ingenieurswissenschaften. Auch die computerba-
sierte Modalanalyse, durch die das Schwingungsver-
halten von Strukturen sichtbar gemacht werden
kann, basiert auf Fouriers Grundlagenwerk aus dem
Jahre 1822.
• Hermann Ludwig Ferdinand von Helmholtz (1821-
1894) veröffentlicht 1865 das berühmt gewordene
Werk „Lehre von der Tonempfindung“. Es markiert
einen Meilenstein in der Erkenntnis der Akustik des
Gehörs. Ihm gelingt es mithilfe unterschiedlich gro-
ßer Glaskugeln (s. Abb. oben), deren eine Öffnung
in den Gehörgang des Ohres gesteckt wird, eine erste
Form der Spektralanalyse des Schalles vorzuneh-
men. Der Luftresonanz-Mechanismus, der bei seinen
Experimenten zum Tragen kommt, wird nach ihm
als „Helmholtzresonanz“ bezeichnet. Auch die Gei-
ge weist durch ihre ff-Löcher eine Helmholtzreso-
nanz auf, die für die Abstrahlung des tieffrequenten
Grundtonbereiches verantwortlich ist.
Artikel „Zeitgeist und Geigenbau“ Seite 4
 Nach Helmholtz ist auch der sägezahnförmige
Schwingungsverlauf der gestrichenen Saite benannt.
• Lord Rayleigh's Buch „Theory of Sound“ (1877)
wird Standardwerk zum aktuellen akustischen Wis-
sen. Es enthält wesentliche Beiträge zur Berechnung
schwingender Strukturen, behandelt das Thema Ab-
strahlung, Beugung und Streuung von Schall.
ENTWICKLUNGEN IM 20. JAHRHUNDERT
Im Jahr 1907 erfindet Lee de Forest die
Verstärkerröhre. Die damit verbundene Mög-
lichkeit, Klang aufzuzeichnen, zu konservieren
und später wiederzugeben, wie auch die Analy-
semethoden der elektronischen Meßtechnik,
eröffnen der Akustik neue Horizonte.
Es entsteht das Forschungsgebiet der
sog. Psychoakustik, die den Zusammenhang
zwischen den physikalischen Eigenschaften
eines Schallsignals und den daraus resultieren-
den Hörempfindungen untersucht – der Zu-
sammenhang also zwischen physikalischen
Reizgrößen und psychischen Empfindungsgrö-
ßen.
Im Bereich der Instrumentenakustik
werden zahlreiche Grundlagenwerke verfasst,
die Aufschluß über die akustische Funktion der
Instrumente geben. Darunter das Standardwerk
„Die Physik der Geige“ [Stuttgart 1981] von
L.Cremer, oder die umfassenden Lehrbücher
„The Physics of Musical Instruments“ von
N.H.Fletcher und T.D.Rossing [New York
1991] und „Fundamentals of Musical A-
coustics“ von Arthur H.Benade [New York
1976].
Vor allem der Catgut Acoustical Socie-
ty, die über Jahrzehnte hinweg stark vom Enga-
gement Carleen M. Hutchins lebte, ist es zu
verdanken, dass sich in der zweiten Hälfte des
20. Jahrhunderts eine kleine, internationale
„Gemeinde“ aus ambitionierten Akustikfor-
schern, Instrumentenbauern und Geigenliebha-
bern zusammengefunden hat, die auf regelmä-
ßigen Internationalen Symposien für Musikali-
sche Akustik (ISMA) Forschungsergebnisse
© Meisteratelier für Geigenbau Martin Schleske 2004. [www.schleske.de]
austauscht. Es würde den Rahmen dieser Skiz-
zen sprengen, all die Ergebnisse im Bereich der
Geigenakustik des 20. Jahrhundert darstellen zu
wollen. Hier sei v.a. auf das halbjährlich er-
scheinende Journal der Catgut Acoustical So-
ciety verwiesen und auf die beiden Buchbände
„Reserach Papers in Violin Acoustics 1975-
1993“ [Acoustical Society of America, 1997].
Dort sind in über 120 Artikeln Forschungser-
gebnisse zu Themen wie Schallabstrahlung,
gestrichene Saite, Steg, Stimmstock, Platten-
schwingungen, Luftresonanzen, Holz, Lack und
Psychoakustik dargestellt.
Erwähnenswert sind sicher die praxisbe-
tonten Arbeiten von:
• Eric Jansson („Department of Speech, Music and
Hearing“ KTH Stockholm)
• Helmut A. Müller (Müller-BBM), der 1983 erstmals
die akustische Funktion der Geige mithilfe der Mo-
dalanalyse darstellte
• Jim Woodhouse (Universität Cambridge, UK), der
u.a. EDAX-Analysen von Grundierungen und Holz-
analysen an altitalienischen Geigen mithilfe der Ras-
terelektronen-Mikroskopie durchführte
Mit Blick auf das heutige „Klima“ im
Geigenbau erscheint mir bemerkenswert, dass
die Kluft zwischen universitärer Forschung und
praktischem Instrumentenbau kleiner zu wer-
den scheint. Mehr und mehr Geigenbauer öff-
nen sich den Analysemethoden der Akustik und
nehmen Forschungsergebnisse aufgeschlossen
zur Kenntnis. Die zunehmende Nähe zwischen
Akustikforschung und Geigenbau zeigt sich
aktuell etwa an der Fusion der Catgut Acousti-
cal Society mit der Violin Society of America.
Auch die jährlich stattfindenen Oberlin-
Workshops in den USA, bei denen Fragestel-
lungen der Akustikforschung, der Klangeinstel-
lung und der Werkstattmethoden einen gleich-
berechtigten Platz finden, dokumentieren, dass
der Geigenbau sich aus einem abgeschlossenen
„Werkstatt-Kämmerlein-Milieu“ zu befreien
beginnt.
Artikel „Zeitgeist und Geigenbau“ Seite 5
VISIONEN UND ENTWICKLUNGEN Ein wesentliches Merkmal von Intelligenz
ist Offenheit, d.h. die Fähigkeit, Neues aufzu-
Ich vermute, dass die Geigenbauwerk-
nehmen und in das eigene Werden und Schaf-
statt der Zukunft sehr viel selbstverständlicher
fen einzubeziehen. Intelligente, entwicklungs-
als bisher akustische Laborarbeit in ihr tägli-
fähige Systeme kommunizieren mit ihrer Um-
ches Arbeiten integrieren wird. Es werden neu-
welt, treten in Wechselwirkungen, um sie krea-
artige Messgeräte als „Diagnosewerkzeuge“ im
tiv für die eigene Entwicklung nutzbar zu ma-
Umgang mit Materialeigenschaften und Klang-
chen.
fragen zum Einsatz kommen.
Wenn wir heute das hundertjährige Jubi-
Neu entwickelte Streichinstrumente der Zu-
läum des Verbandes Deutscher Geigenbauer
kunft werden den akustischen Herausforderun-
(VDG) feiern, dann könnte dies auch Anlass
gen der heutigen großer Säle besser gewachsen
sein, uns zu fragen, wie nachfolgende Genera-
sein als bisherige Instrumente. Durch die Ent-
tionen in 100 Jahren wohl über uns heute den-
deckung des akustischen Potentials der Kohle-
ken werden, ob sie uns als borniert oder kreativ,
faser und einer konsequenten Kombination mit
als abgeschottet oder offen, als eingeschlafen
den attraktiven Eigenschaften des Holzes wer-
oder innovativ empfinden werden.
den Instrumente mit einer größeren dynami-
Wo treten wir heute in Wechselwirkungen
schen Bandbreite und Modulierbarkeit entste-
mit Entwicklungen links und rechts unseres
hen als dies je der Fall war. (Mit dem Material
Blickfeldes? Wo suchen wir jenes kreative
Holz wurde „das Ende der Fahnenstange“ in
Spannungsfeld aus Kunst und Wissenschaft?
der ersten Hälfte des 18. Jahrhunderts bereits
Wo werben wir heute im VDG als neue Mit-
erreicht. Die Materialeigenschaften der Spitzen-
glieder passionierte Laien, die freilich mit Blick
instrumente Stradivaris können wir mit reinem
auf die Geigenbaupraxis Anfänger, zugleich
Holz nicht übertreffen). Aber würde Stradivari
aber Spitzenvertreter ihres Faches sein können:
in seiner Innovationskraft heute leben, hätte er
Architekten, Physiker, Komponisten, Histori-
längst die faszinierenden akustischen Eigen-
ker, Bildhauer, Musiker, Hirnforscher?
schaften von Kohlefaserverbund entdeckt und
würde in eine neue Blütezeit des Geigenbaues Ein attraktiver Geigenbauverband wird
vordringen. gewiss einer sein, von dem Impulse ausgehen;
ein Verband, in dem Wechselwirkungen mit
SCHLUSS Vertretern und Entwicklungen aus Musik,
Kunst und Wissenschaft stattfinden; ein Ver-
Diese Skizzen sollen ausreichen, uns zu
band in dem die Spannung lebendig gehalten
vergegenwärtigen, in welch innovativem Milieu
wird zwischen Tradition und Innovation, zwi-
die Geige entstehen und sich entwickeln konn-
schen Kontinuität und Erneuerung.
te. Je tiefer man das akustische System der
Geige kennen lernt, desto mehr wird man stau-
Gabriel Weinreich, einer der großen Akustiker
nen über die Intelligenz dieses Systems und
unserer Zeit, Professor für Physik an der Uni-
man wird sich fragen, wie es zu dieser Intelli-
versity of Michigan stellte an das Ende seines
genz kam. Ich bin überzeugt, die genannten
Festvortrags Musical Acoustics in the Twentieth
Wechselwirkungen spielten dabei eine wesent-
Century angesichts des 75-jährigen Bestehens
liche Rolle. Vielleicht können sie eine Ermuti-
der Acoustical Society of America ein Wort,
gung sein, gerade angesichts unseres Jubiläums,
dem ich mich uneingeschränkt anschließen
konstruktiv nach der Kreativität und Offenheit
möchte: „Letztlich aber wohnt der Erforschung
unseres Verbandes zu fragen.

© Meisteratelier für Geigenbau Martin Schleske 2004. [www.schleske.de] Artikel „Zeitgeist und Geigenbau“ Seite 6
der Musikinstrumente ein ganz besonderer Reiz
inne. Denn über Jahrhunderte (wenn nicht gele-
gentlich sogar Jahrtausende) hinweg wurde
durch „trial and error“ eine derart beeindru-
ckende Genialität im Bau jener Instrumente
erreicht. Es hat etwas Erfüllendes, dem ge-
heimnisvollen, einzigartigen Charakter von
Musikinstrumenten einen kleinen Baustein lo-
gischen Verständnisses zur Seite zu stellen und
gerade dadurch ehrfürchtig zu erkennen, wel-
che Werke der Mensch über Epochen hinweg
zu entwickeln in der Lage war - durch Intuition,
Geduld und die Gnade Gottes“.

[Zurück zur Themensuche des Handbuch Geigen-

akustik‘]
http://www.schleske.de/index.php?http://www.schle
ske.de/09extras/extras3handbuch.shtml

© Meisteratelier für Geigenbau Martin Schleske 2004. [www.schleske.de] Artikel „Zeitgeist und Geigenbau“ Seite 7
Martin Schleske: „Eigenschwingungen im Werdegang einer Geige“. Diplomarbeit an der Fachhoch-
schule München im Fachbereich „Technische Physik“ 1994.

Zusammenfassung
Es wurde untersucht, welchen Einfluss die Bearbeitung der Geigenplatten auf die Eigenschwingun-
gen der Geige hat. Dazu wurde nach dem Modell einer ‚Antonio Stradivari‘, anno 1715 eine Geige
angefertigt. In vierzehn verschiedenen Stadien der Plattenbearbeitung (lokale Verringerungen der
Plattendicken) wurden die Eigenfrequenzen und Eigenformen sowohl der freien Platten als auch des
in jedem Stadium zusammengefügten Korpus gemessen.
Als Messmethode wurde ein „Analoges Abtasten“ entwickelt. Dabei wurde die Struktur über Körper-
schall sinusförmig und bei kontinuierlicher Ortsverschiebung angeregt. Über einen ortsfesten Be-
schleunigungsaufnehmer wurde die Schwingungsantwort gemessen.
Ein Vergleich der Eigenfrequenzen im Endstadium des Werdeganges mit denen im Anfangsstadium
zeigt: Der Betrag der über die ersten sieben Moden gemittelten Frequenzverschiebung je Mode
betrug für die freie Decke 1.8 Halbtöne, für den freien Boden 5.6 Halbtöne, für den Korpus jedoch
lediglich 1.0 Halbtöne.
Aus den gemessenen Eigenfrequenzverschiebungen geht hervor, dass die Lage der Eigenfrequen-
zen des Instrumentes nur zu einem geringen Ausmaß durch die Ausarbeitung der Platten zu beein-
flussen war. Ein einfacher Zusammenhang zwischen einer Frequenzabstimmung der freien Platten
und den resultierenden Eigenfrequenzen des Korpus war nicht erkennbar. Dies lässt die Bedeutung
eines „Plate-Tunings“ der freien Platten fragwürdig erscheinen.
Anders verhielt sich der Einfluss der Werdegangstadien auf die Eigenformen: Die Änderungen der
Eigenformen, die durch die Plattenbearbeitungen erzielt wurden, waren beim zusammengefügten
Instrument deutlich größer bei den freien Platten.
 -1-

Klangfarben der Erkenntnis

Festvortrag anlässlich der 25. Jahrestagung der Kanzlerinnen und Kanzler der Hochschulen
für angewandte Wissenschaft der Bundesrepublik Deutschland, 2.-4. September 2009, Pina-
kothek der Moderne, München.

Martin Schleske. Geigenbaumeister & Dipl.-Physik-Ing. (FH)

Als ich vor 15 Jahren meinen Abschluss an der Hochschule für Angewandte Wissenschaften
in München gemacht habe, war ich vielleicht eine Ausnahme, denn ich habe nicht studiert, um
anschließend als Physiker oder Ingenieur zu arbeiten, sondern um das zu tun, was ich davon
auch schon tat – allerdings anders und besser als zuvor: nämlich Konzertgeigen zu bauen.

Ich hatte längst meine Geigenbaulehre in Mittelwald absolviert. Aber ich habe gemerkt, dass
ich die Akustik der Geige nicht verstehe und dass ein Physikstudium nötig ist, um die Dinge
besser zu verstehen. Nach dem Physikstudium und Jahren der Praxis habe ich dann die noch
wichtigere Prüfung gemacht: Die Meisterprüfung.

Das führt mich zu meinem ersten Gedanken: Verstehen und Beherrschen. Mit dem Verstehen
ist das ja so eine Sache. Dass man eine Prüfung bestanden hat, heißt ja nicht, dass man die
Sache auch verstanden hat. Das war für mich die Haupterkenntnis des Studierens. Man betritt
ein Universum der Wissensgebiete und kann nur einen winzig kleinen Teil davon wirklich
verstehen.
Wenn man anfängt, sich damit wirklich zu beschäftigen, merkt man: Das allermeiste kann
man nur zur Kenntnis nehmen; Manches hat man verstanden; Und nur von wenigen Dingen –
es sind die Dinge, mit denen man arbeitet, die einen faszinieren, und in die man mit Herzblut
reingeht – nur von wenigen Dingen kann man vielleicht einmal sagen, dass man sie be-
herrscht. Es ist das, wo man sich als Mensch nicht außen vorlässt, sondern eine Leidenschaft
entwickelt, die Dinge zu erforschen.
 -2-

Ein solcher Bereich, der nicht nur das Denken, sondern auch das eigene Herzblut und Emp-
finden betrifft, war für mich von Anfang an der Klang.

Das führt mich zu einem zweiten Gedanken, an dem ich etwas Grundsätzliches deutlich ma-
chen möchte: Akustik und Klang. So wie man Klang und Akustik unterscheiden muss, bewege
ich mich in meinem Arbeitsalltag zwischen dem Atelier und dem Labor. Es sind zwei Welten.
In den vergangenen Jahren habe ich – unterstützt u.a. durch Mittel der Innovationsförderung
des Bayerischen Wirtschaftsministeriums – einige Entwicklungsarbeit geleistet – vor allem im
Bereich der akustischen Materialmodifikationen. Da sind mittlerweile fünf Patente entstan-
den, aber es liegt noch viel Entwicklungsarbeit vor mir.

Das Wesentliche ist die Kombination zwischen Labor und Atelier. Diese Kombination hat mit
dem eigentümlichen Verhältnis zwischen Akustik und Klang zu tun, auf das ich gleich zu
sprechen kommen möchte.

Man muss wissen und begreifen, dass Klang und Akustik zwei unterschiedliche Welten sind.
Sie haben unterschiedliche Sprachen und unterschiedliche Werkzeuge der Erkenntnis. Bei der
Akustik reden wir von physikalischen Reizgrößen. Sie sind messbar und man kann mit ihnen
rechnen. Beim Klang aber geht es um Empfindungsgrößen. Sie sind erfahrbar und man
braucht gute Ohren. Wenn Musiker mir in meinem Atelier auf ihrem Instrument vorspielen –
und zeigen, wo das Instrument Probleme hat – dann besteht mein Kopf nur noch aus zwei
riesig großen Ohren. Nur noch Hören! Sonst nichts. Zuhören, hinhören!

Akustik ist messbar, Klang ist erfahrbar. Akustik gibt es auch ohne den Menschen, Klang aber
entsteht in den Empfindungen des Menschen. Auch wenn der Klang seine physikalische Ur-
sache in der Akustik hat, ist Klang doch mehr als bloße Physik. Akustik und Klang sind darin
wie das menschliche Gehirn – auf der einen Seite – und das menschliche Bewusstsein ander-
seits.

Diese Spannung zwischen Akustik und Klang ist wie eine Allegorie für das Geheimnis, dass
der Mensch nicht nur aus sichtbarer Materie besteht, sondern auf eine rätselhafte Weise Be-
wusstsein hat – und dass er diese Tatsache auch reflektiert!
Das ruft in uns die staunende Frage wach: „Wie kommt der Geist in die Materie?“ – so das
gleichnamige Buch des amerikanischen Philosophen Colin McGinn (München 2005). Unser
Gehirn ist nicht nur denkendes Fleisch, sondern wirkender und selbstbewusster Geist. Der
Klang einer Geige hat nicht nur eine physikalisch-akustische Gestalt, sondern auch einen wir-
kenden, ästhetischen Gehalt.

Unser Bewusstsein ist dem denkenden Gehirn anvertraut wie der Klang dem schwingenden
Holz. Ist es nicht ein faszinierendes Geheimnis, dass bewusstseinslose physikalische Teilchen
und bewusstseinslose chemische Prozesse bewussten Geist entstehen lassen können. Wer dar-
über nicht staunt, hat darüber nie wirklich nachgedacht.

Mensch! Du bist aus „Lehm vom Acker“ (Hebräisch: Adama) geschaffen und bist dir doch
dessen bewusst, dass du mehr als ein Materieklumpen bist. Adam, der Mensch! Du bringst
Kunst und Wissenschaft hervor; empfindest Liebe und Hoffnung; leidest unter deiner Be-
grenztheit und deiner Endlichkeit; bist fähig eine Symphonie zu komponieren und schaffst aus
Steuergeldern das Orchester, das sie spielt. Du empfindest Angst und Glück, bist fähig zur
 -3-

Sünde wie zur Treue. Du fragst dich, wer du bist und was du sollst; du fragst dich, was du
giltst und stellst dich selbst in Frage; du sendest deinen Geist zur Erforschung in die Welt und
bleibst doch – wie sehr du dich beim Denken auch beobachten magst – dir selbst ein Rätsel.
Und welch eine Sehnsucht und Vollmacht entstehen, wenn erst der Glaube hinzukommt. Je-
nes geheimnisvolle Feuer des menschlichen Bewusstseins!

So sehe ich die Beziehung zwischen Akustik und Klang wie ein Gleichnis für die Beziehung
zwischen Gehirn und selbstbewusstem Geist. Akustik ist Physik, aber Klang hat mit Empfin-
dung – also mit Bewusstsein – zu tun.

Vom Klang des Menschseins also zurück zum Klang der Geige. Eine Geige zu bauen, bedeu-
tet, dem Musiker eine Stimme zu geben. Das ist weit mehr als ein bloß physikalischer Vor-
gang. Man greift damit an das Innere des Menschen. Wer einen wirklich großen Musiker mit
seinem Instrument beobachtet -- Ich habe das Glück, dass das in meinem Atelier immer wie-
der der Fall ist -- , der wird erleben, dass das Instrument wie ein Teil seines Körpers ist. Es ist
wie angewachsen. Es ist ein Teil seines Ichs.
Wenn Musiker mit Problemen ihres Instrumentes in meine Werkstatt kommen, dann habe ich
regelmäßig das Gefühl, es ist wie die Begegnung mit einem körperbehinderten Menschen.
Wenn ein Musiker sagt, er kommt in den Ton der Geige nicht mehr rein; in den hohen Lagen
macht das Instrument „zu“, es klingt matt, die Ansprache des Klanges hat seine Unmittelbar-
keit verloren…, dann ist es gleichbedeutend als würde er sagen, sein Arm sei lahm, seine Fin-
ger würden schmerzen. Das Instrument ist wie ein Teil seines Körpers. Als Geigenbauer gibt
man dem Musiker seine Stimme.

Einer der schönsten Rückmeldungen, die ich vor kurzem über eine meiner Geigen bekommen
habe, die ich für eine Geigerin aus Freiburg gebaut habe, war, dass sie mir schrieb: „Die
Klänge dieser Geige sind wie Medizin, die einen nach einem harten Arbeitstag wieder leben-
dig werden lässt.“ Der Klang ist tatsächlich wie die Stimme der Seele.

Psychoakustik. Die vergangenen Jahre habe ich (basierende auf den Austausch mit der Psy-
choakustikgruppe der Universität in Cambridge) gemeinsam mit einem guten Freund (Prof.
Mores, der an der HAW in Hamburg tätig ist) ein psychoakustisches Werkzeug entwickelt.
 -4-

© Martin Schleske. Meisteratelier für Geigenbau 2009

Ich möchte es kurz demonstrieren. Es geht dabei um die psychoakustische Auswertung des
gehörten Schalls. Die meisten Attribute des Psychoakustik betreffen leider nur die Lärmbe-
kämpfung – also die Fragen, wie man das Störende, Hässliche und Schädliche vermeiden
kann; sehr viel weniger Grundlagenforschung wurde bislang zu Fragestellungen des Wohl-
klanges betrieben, also der Frage, was gesund, schön, hilfreich und wohltuend ist.
(Zwei Doktoranten der HAW in Hamburg schreiben zur Zeit ihre Dissertationen über The-
menstellungen meiner Werkstatt. Es ist wunderbar, wenn Andere die Zeit haben, für einen
weiter nachzudenken).

Für jeden Geigenbauer ist ein großes Vorbild die menschliche Gesangsstimme. Seit Jahren ist
das für mich besonders Maria Callas (die ich leider nie live erleben konnte). Es gibt bei ihren
Solo-Arien immer wieder einzelne Töne, wo man vor lauter Intensität und Schönheit Gänse-
haut bekommt. Ich wollte wissen, was bei diesen Tönen auf der Basilarmembran unseres In-
nenohres eigentlich passiert. (Sie wissen: In dieser in der nur erbsengroßen Cochlea ist diese
hochempfindliche Membran mit ihren je ca. 3.500 Haarzellen, die durch Wanderwellen in der
Flüssigkeit der Gehörschnecke angeregt werden und über den Hörnerv ins Gehirn feuern). Die
Erregungsmuster der Basilarmembran sind natürlich etwas ganz anderes – und viel näher an
der tatsächlichen Empfindung – als bloße Messungen im akustischen Schallfeld.

Es zeigt sich, dass tatsächlich bei den besonders stark wirksamen Tönen auch ein besonders
deutlich erkennbares – sogar visuell sehr ansprechendes! – Erregungsmuster entsteht. Es
scheint so als habe unser Gehirn sehr viel Freude an der Mustererkennung.
Maria Callas hat es geschafft, bei diesen Tönen ihre Resonanzräume so abzustimmen, dass
sich innerhalb der Vibratoperionden (etwa 7-mal / Sekunde) die geradzahligen und ungerad-
zahligen Obertöne in ihrem Feuern ständig abwechseln. So entsteht eine faszinierende Klang-
farbenmodulation.
 -5-

Der gute Klang beginnt mit dem guten Holz. Einer von 10.000 Stämmen ist ein sog. Sänger-
stamm. Ich habe es beim Schlagen immer wieder erlebt. Wo die gewöhnlichen Baumstämme
nur ein dumpfes Geräusch von sich geben, haben diese Sänger beim Fällen einen glockenarti-
gen Klang. Aus ihnen kann man gute Geigen bauen.

Die wichtigste Klangeigenschaft – man könnte auch sagen, die klangliche Handschrift die ich
in meinen Instrumenten suche – ist für mich die Modulierbarkeit des Tones. Man muss den
Klang unter dem Bogen kneten können wir Ton in den Händen des Töpfers. Da muss ein Wi-
derstand entstehen. Bei einer guten Geige ist das Spielgefühl (vor allem auf den tiefen Saiten)
als würde man durch frisch gefallenen Schnee laufen – dieses Knirschen und das komprimie-
rende Gefühl unter den Füßen. Modulierbarkeit heißt: Der Ton lässt sich in unterschiedliche
Richtungen färben, von einer sanften Wärme bis zu einer aggressiv beißenden Attacke.

Gute Musiker wissen sehr genau, was sie suchen. Und sie sind in ihrem Spielen erstaunlich
konsistent. Das wird von manchen Forschern, die auf den Musikakustikkonferenzen Vorträge
halten, aber spürbar nur mit ihrer eigenen Forschung, nicht aber mit Profimusikern zu tun
hatten, oft unterschätzt.

Das Thema meines Vortrags war „Klangfarben der Erkenntnis“. Was meine ich damit? Vie-
les ist längst angeklungen. Geigenbau bedeutet für mich Forschen und Gestalten. Man muss
dazu Forscher, Handwerker Musiker und Künstler gleichermaßen sein. So habe ich in den
letzten Jahren vier Wege der Erkenntnis entdeckt, die alle ihre Bedeutung haben. Es ist wie
vier unterschiedliche Klangfarben, die sich zu einem guten Gesamtklang ergänzen sollen:

(1) Ratio. Das rationale Denken folgt den Gesetzen der Natur. Um besser herauszufinden,
wie die Dinge funktionieren, habe ich Physik studiert und nutze in meinem Labor die
Erkenntnisse und Methoden der akustischen Forschung. So kann ich die Schwingun-
gen der Strukturen, die ich erschaffe, besser verstehen. Hier begreift man Zusammen-
hänge. Es ist erfüllend, Dinge zu verstehen und in einer treffsicheren Theorie darzu-
stellen. Tódor Kármán sagte einmal zu Recht: „Nichts ist praktischer als eine gute
Theorie“, denn hier kommt es tatsächlich zu einer Erkenntnis der Ordnung.

(2) Empirie. Das empirische Erforschen beruht auf Experimenten. Viele Fragen sind zu
komplex, als dass man sie in geschlossenen, „ordentlichen“ Modellen erfassen könnte.
Die Gesetze, die wir erkannt haben, reichen nicht aus. Wir sind noch zu dumm. Da
führen gezielte Experimente und wachsames Beobachten weiter. So habe ich es etwa
mit der Erforschung des Geigenlackes gemacht, als ich in meiner kleinen, alchemie-
ähnlichen Lackküche hunderte Lackrezepturen gekocht und in ihrer klanglichen Aus-
wirkung auf Probehölzern gemessen habe. (Als Messergebnisse entstehen Punktewol-
ken von Daten, in die man Trendlinien legt). Da begreift man die Dinge, ohne das Ge-
setz zu verstehen.

(3) Intuition. Das intuitive Handeln schöpft aus der Erfahrung. Wenn man eine Sache
durch intensivste Beschäftigung und Leidenschaft nicht nur zur Kenntnis genommen
hat, sondern sie beherrscht, dann schöpft man aus einem unbewussten Erfahrungs-
schatz, der weit größer ist als das, was man bewusst begriffen hat. Man tut das Richti-
 -6-

ge und weiß nicht warum. So geht es mir vor allem in der Ausarbeitung der Wöl-
bungsform. Ich habe zu wenig Wissen und habe viel zu wenig experimentiert und
dennoch gelingt die Geige. Es ist beglückend zu merken, dass man viel mehr begriffen
hat als man weiß. Wir öffnen uns diesem Weg, indem wir beginnen, unserm Leben zu
vertrauen. Das sollten wir tun.

(4) Inspiration. Das inspirative Erkennen folgt der Gnade. Hier geschehen Dinge, die sich
nicht aus der bisherigen Erfahrung oder Intuition, geschweige denn dem Wissen, ab-
leiten lassen. Es sind rettende Momente, Offenbarungen mitten im Alltag, beglücken-
de Führungen und Fügungen. Auch dies ist ein Weg, das Richtige zu tun. Dieser vierte
Weg ersetzt nicht die drei anderen Wege, die alle ihre Bedeutung, ihre Art, und ihre
Verheißung haben. Denn in ihnen allen leben und erleben wir, was es bedeutet,
Mensch zu sein.

Man darf die Wege nicht gegeneinander ausspielen, wie man ja auch – gleichnishaft gespro-
chen – die Funktionen des eigenen Körpers nicht gegeneinander ausspielen würde. Müsste
man hier eine Zuordnung treffen, so würde ich sagen: Das rationale Denken ist mein „Kopf“;
die Empirie „die Hände“; die Intuition „der Bauch“; die Inspiration „das „Herz“. Es leuch-
tet sofort ein: Das Herz ist nicht wichtiger als der Kopf, und die Hände sind nicht wichtiger
als der Bauch. Wir sollen ganze Menschen sein!
Was uns treibt, sind unsere Nöte und unsere Fragen. Wir können unseren Fragen die vier We-
ge der Erkenntnis weisen. An ihnen werden wir wachsen und werden die Welt, für die wir
geschaffen sind, erforschen, ergründen (und durch unser Kunst und Liebe) auch gestalten.

Vielen Dank.

Martin Schleske, Pinakothek der Moderne, München den 2. September 2009

„Den Akustischen Geheimnissen der Geige auf der Spur“, Verband Deutscher Geigenbauer (Jahres-
hauptversammlung), Mittenwald 2000.

Zusammenfassung
„In seinem Vortrag wird Martin Schleske nach einer „klingenden“ Einführung in grundsätzliche As-
pekte der Geigenakustik einige seiner aktuellen Forschungsergebnisse präsentieren. Am Beispiel
verschiedener, typischer Referenzinstrumente, darunter die“ Schreiber“–Stradivari 1712 (ex. Wie-
niawski, ex Zukerman) und einer Montagnana 1729, werden akustische Besonderheiten des gege-
benen Schwingungsverhaltens und der Schallabstahlung erkennbar gemacht. Es wird der Frage
nachgegangen, inwieweit der spezifische Klangcharakter objektiven akustischen Kriterien gegen-
über gestellt werden kann. Ferner wird versucht, Zusammenhänge zwischen Akustik und jeweili-
gem konstruktiven Konzept der Instrumente aufzuzeigen. Die Instrumente werden anhand von
Konstruktionsdiagrammen, Akustikanalysen und Klangaufnahmen präsentiert. Zur Anwendung für
den Geigenbau werden Grenzen und Möglichkeiten der gezielten Anfertigung von Klangkopien dis-
kutiert. Einige der dazu geeigneten Methoden werden vorgestellt.“
 Martin Schleske | Workshop for Violin Making
Georg-Hellmair-Platz 382, D-86899 Landsberg am Lech, Germany
www.schleske.de
 Martin Schleske | Workshop for Violin Making
Georg-Hellmair-Platz 382, D-86899 Landsberg am Lech, Germany
www.schleske.de
 Martin Schleske | Workshop for Violin Making
Georg-Hellmair-Platz 382, D-86899 Landsberg am Lech, Germany
www.schleske.de
 Martin Schleske | Workshop for Violin Making
Georg-Hellmair-Platz 382, D-86899 Landsberg am Lech, Germany
www.schleske.de
 Martin Schleske | Workshop for Violin Making
Georg-Hellmair-Platz 382, D-86899 Landsberg am Lech, Germany
www.schleske.de
 Martin Schleske | Workshop for Violin Making
Georg-Hellmair-Platz 382, D-86899 Landsberg am Lech, Germany
www.schleske.de
 Martin Schleske | Workshop for Violin Making
Georg-Hellmair-Platz 382, D-86899 Landsberg am Lech, Germany
www.schleske.de
 Martin Schleske | Workshop for Violin Making
Georg-Hellmair-Platz 382, D-86899 Landsberg am Lech, Germany
www.schleske.de
 Martin Schleske | Workshop for Violin Making
Georg-Hellmair-Platz 382, D-86899 Landsberg am Lech, Germany
www.schleske.de
 Martin Schleske | Workshop for Violin Making
Georg-Hellmair-Platz 382, D-86899 Landsberg am Lech, Germany
www.schleske.de
 Martin Schleske | Workshop for Violin Making
Georg-Hellmair-Platz 382, D-86899 Landsberg am Lech, Germany
www.schleske.de
 Martin Schleske | Workshop for Violin Making
Georg-Hellmair-Platz 382, D-86899 Landsberg am Lech, Germany
www.schleske.de
 Martin Schleske | Workshop for Violin Making
Georg-Hellmair-Platz 382, D-86899 Landsberg am Lech, Germany
www.schleske.de
 Martin Schleske | Workshop for Violin Making
Georg-Hellmair-Platz 382, D-86899 Landsberg am Lech, Germany
www.schleske.de
 Martin Schleske | Workshop for Violin Making
Georg-Hellmair-Platz 382, D-86899 Landsberg am Lech, Germany
www.schleske.de
 Martin Schleske | Workshop for Violin Making
Georg-Hellmair-Platz 382, D-86899 Landsberg am Lech, Germany
www.schleske.de
 Martin Schleske | Workshop for Violin Making
Georg-Hellmair-Platz 382, D-86899 Landsberg am Lech, Germany
www.schleske.de
Proceedings of the Stockholm Music Acoustics Conference, August 6-9, 2003 (SMAC 03), Stockholm,
Sweden
PSYCHOACOUSTIC INVESTIGATIONS ON THE POSSIBILITY OF AURALLY
IDENTICAL VIOLINS
Paul Geissler1), Otto Martner1), Carsten Zerbs1), Martin Schleske2)
1)
Müller-BBM GmbH, Planegg near Munich, Germany
2)
Meisteratelier für Geigenbau, Munich, Germany
PGeissler@MuellerBBM.de
ABSTRACT
“Is it possible to make sound copies of violins?” To answer this
question one needs to know the ingredients which make two
violins judged by musicians to be aurally identical. Using a
questionnaire among 54 musicians verbal attributes related to
sound quality characteristics of violins were collected and
ranked. Subjective listening tests were carried out to check
which verbal attributes are suitable and do evoke the same
meaning in different persons. Useful attributes for the ranking
could be found using statistical measures for consistency and
concordance. Additionally, it was found that the measurement
uncertainty of frequency response measurements today is smaller
than the analysis resolution of the human ear in regard to
frequency and amplitude in the frequency range important for
the hearing sensation and for room acoustic situations typical for
a violin maker’s workshop. The results give strong indications
that the psychoacoustics characteristics of the human ear and the
vibroacoustic characteristics of the violin body in combination
with the attributes judged to be important for violin sound
quality will in principle allow for the idea of making sound
copies of a given violin.
1. INTRODUCTION
1.1. What are sound copies?
Professional musicians sometimes use the sound quality
characteristics of a given instrument as a reference and asks the
violin maker “Can you make me a violin that sounds like this
instrument?”. To be even more precise: if the customer holds,
say one of the famous old Italian violins in his hand, the question
would be “Can you make me a new violin that sounds like this
masterpiece in my hand?”
The focus is on the verb “sound”. A violin maker learns
how to make geometrical copies of masterpieces. This is cer-
tainly useful to get a violin body to start with. Of course, due to
the differences in the wood material the geometrical copies do in
general not sound like the masterpieces.
But if a violin maker is able to build with up-to-date wood
material a new violin whose modal characteristics, the eigenfre-
quencies, damping factors and mode shapes, are identical to the
modal characteristics of the reference violin with its old wood
material, it should in principle be possible. If the modal identity
can be attained, one could expect that the copy also has the same
radiation characteristics as the reference violin and hence - for
SMAC-1
the same listener in the same room - the same hearing impression
for both instruments. This concept of modal copies or “tonal
copies” has recently been tackled by Schleske [1].
1.2. Can the procedure be simplified?
How accurate must these modal copies be built? Are differences
between the modal characteristics of the copy and the reference
violin allowed? Taking into account the finite resolution of the
human hearing perception and the always present statistical
fluctuations in the reverberant sound field of a room one could
imagine that the copying procedure can eventually be simplified,
if only the differences between copy and reference violin are
smaller than the audible differences.
It should also be noted that the question which is formu-
lated here for violins is a fundamental question in sound quality
investigations of other products and services as well. The ability
to evaluate if two physically not 100% identical products do
evoke the same hearing impression for potential customers is the
basis for product sound design. Telephone and broadcasting are
two branches where simplification procedures based on the
human hearing perception have been investigated and applied
successfully.
1.3. Acknowledgements
The results of this paper are based on work carried out within the
research project „Material investigations and advanced methods
of production and quality control of bowed instruments“
(VIOLIN) funded by the European community under contract
BRST-CT98-5465. The project was structured into different
working groups. The „Meisteratelier für Geigenbau Martin
Schleske“ and the company „Müller-BBM“ formed the working
group „sound“.
The team members of the working group “sound” thank
Helmut A. Müller for help, comments and fruitful discussions.
2. QUALITY ATTRIBUTES
2.1. Questionnaire
The musician’s evaluation of a violin is based on a number of
perceived sensations and attributes. The sound received at his
ears is one of the most important criteria but there are also
additional criteria such as playability, reaction of the instrument
and others. The interests of the musician certainly play a role,
Proceedings of the Stockholm Music Acoustics Conference, August 6-9, 2003 (SMAC 03), Stockholm,
Sweden
too: Beginners tend to weigh attributes such as playability and
reaction much higher than professional musicians. The profes-
sional musician usually is much more interested in the sound
because he is able to fully control the instrument at hand.
As a first step psychoacoustic investigations were carried
out to find out relevant attributes used by musicians for the
evaluation of the sound of a violin.
A questionnaire asking for relevant acoustic attributes of
violins and the attribute’s ranking was designed. The question-
naire was mailed to about 100 musicians in Germany. 54 musi-
cians (26 female, 28 male) sent answers. The musicians were
asked: “Please list attributes you would choose to evaluate a
violin in respect to sound and playability” and “Please rank the
above attributes according to the importance that you personally
give them”.
For analysis the more than 70 different attributes (or attrib-
ute pairs) were grouped empirically into different general fea-
tures. For example the attributes bright/dark (hell/dunkel),
warm/cold (warm/kalt) or round/soft (rund/weich) were grouped
to the general feature timbre (Klangfarbe), while attributes such
as easy/difficult (leicht/schwer) or direct (direkt) were grouped
to the general feature reaction (Ansprache). A weight of 1 was
assigned to the attribute ranked highest, the second rank was
given a weight of ½ and so on.
2.2. Results of the questionnaire analysis
The final result of the analysis, together with some interpretation
concerning the correlated physical response quantity, is shown in
table 1.
feature % physical response quantities correlated
timbre 38 spectral distribution of frequency response
reaction 19 coupling between player and violin
sustain 13 coupling between violin and room
balance 11 balance of spectral distribution
loudness 9 energy of frequency response
modulation 7 dynamic range of frequency response
others 3 color, optics and other attributes
Table 1: Features, physical meaning and percentile contribution.
The results show that timbre with a contribution of 38% to-
gether with the 11% contribution of balance of timbre are judged
by musicians to be the most important criteria for sound quality.
Loudness contributes only 9%. Of nearly equal importance is
modulation with 7% contribution: a feature that stands for the
variability of the timbre and a large dynamical range between
forte and piano playing, with the piano passages still markedly
above the “operational noise floor” given by the mere bow/string
interaction. It is believed that all these features are closely re-
lated to the structural dynamic characteristics of the violin itself.
These features make a total of 66% of the sound quality features.
Sustain (13%) is judged to be important as well. It is as-
sumed that this feature is related to the coupling between the
violin and the room acoustical situation. One could also argue,
however, that sustain is for a given room a feature linked with
the power of the violin itself. In this case nearly 80% of the
features refer to sound quality characteristics of the violin.
SMAC-2
A very important feature is reaction which is not a pure lis-
tening criterion but takes the feedback of the instrument on the
player’s action into account. However, this feature is in parts
also related to the modal characteristics of the violin. “Other”
features (3%) were mostly optical attributes.
The results show that the majority of the selected quality
attributes refer to variables linked with frequency response func-
tions. The most interesting frequency response function here
relates the sound pressure at the listener’s ear to the force at the
bridge in bowing direction.
It is important to recognize that there are today psycho-
acoustic models that allow us to calculate from the measured or
calculated sound pressure as a physical quantity psychoacoustic
quantities that relate to the various aspects of the hearing per-
ception. For example, for the evaluation of the loudness one has
the models of Moore [2] or Zwicker [3], for the evaluation of
timbre one has models of Benedini [4] and v. Bismarck [5],[6].
Knowing that we can calculate and evaluate the attributes
related to the relevant hearing perception once we know the
sound pressure at the ear, it is useful to investigate into the scal-
ing of psychoacoustical criteria and into consequences for the
determination of frequency response functions during the violin
making.
3. LISTENING TEST
3.1. Setup
Based on the results of the questionnaire a listening test was
developed and carried out. Because the main interest here was
methodological, the listening test was carried out with only 3
female and 7 male test persons, aged between 30 and 45. It was
executed as a complete paired comparison test.
For the listening test the attributes bright (hell), nasal (na-
sal), pleasant (angenehm), reaction (Ansprache), balanced
(ausgeglichen), colorful (farbenreich) and passionate (leiden-
schaftlich) were selected because they were typical for the dif-
ferent features of the questionnaire analysis.
For each of the selected attributes a characteristic sound ex-
ample was recorded. Sound recordings of 8 different violins
played by 2 different professional musicians were made. The
binaural recordings were made with an artificial head system.
They encompassed short pieces from violin compositions, scales
and tones. Using an audio editor short sequences were compiled
from these recordings. A time period of 5 seconds duration of
this sound was prepared for all eight violins (played by the same
musician).
The consistency of the decisions of the listeners was
checked. The results showed the difficulties to assess the attrib-
utes selected. It was further analysed how the test persons ranked
the instruments for all the attributes of interest. As a measure of
the concordance between the judgements of different test per-
sons the concordance coefficient [11] was determined. It was
found that the concordance coefficient in the ranking was rela-
tively high for the attributes pleasant and reaction, medium for
bright and balanced and relatively low for nasal, colorful and
passionate.
Not all values were found to be significant on a level of at
least 5%, i. e. with an hypothesis error probability α < 5%.
Table 2 shows the concordance coefficient for all test persons
and for those with consistent answers.
Proceedings of the Stockholm Music Acoustics Conference, August 6-9, 2003 (SMAC 03), Stockholm,
Sweden
attributes Wa αa (%) Wc αc (%) bright
4
bright 0.29 0.5 0.45 3,5
3
no. 1
nasal 0.18 8.5 0.13 > 40 3
passionate nasal
-5 -3 2 no. 2
pleasant 0.68 10 0.65 10
2
no. 3
reaction 0.65 5⋅10-5 0.70 10-3 1
1 no. 4
balanced 0.17 20 0.45 3 0
no. 5
colourful 0.21 4.5 0.27 > 40
colourful pleasant
no. 6
passionate 0.26 1.5 0.72 20
no. 7

Table 2: Values for the concordance coefficient Wa for all test no. 8
persons and Wc for consistently answering test persons (6 or less balanced reaction
circular triples, with the exception for passionate: 9 or less
circular triples) and values for the hypothesis error probability
Figure 2: BTL-scaling values of consistently answering test
αa and αa, respectively [11].
persons for violin no.1 to 8.
The results show that the attributes pleasant and reaction
The results are to be interpreted from a methodological per-
are rated approximately equal among the test persons. For these
spective. They show that one can find attributes like pleasant,
attributes the understanding of what pleasant or (good) reaction
reaction, bright and balanced that do evoke approximately the
means was equal among the test persons. On the other hand for
same meaning within different persons. They are obviously not
colorful or nasal the value is lower: obviously the understanding
only important but can additionally be used for scaling and
of these attributes is different among the different test persons or
ranking of the sound quality of different violins, whereas attrib-
the test persons had no clear image about these attributes.
utes like passionate, colorful or nasal evoke a different under-
For the scaling of the ranking evaluation the BTL-method
standing in different persons. They can not be properly scaled
[12] was chosen.
properly and should therefore be omitted for the ranking of
The results for the BTL-scale based upon the consistent de-
violins.
cisions is shown in the following table 3 and figure 2.
4. FREQUENCY RESPONSE FUNCTIONS
passionate

The above results show that the majority of features is implicitly

balanced
pleasant

reaction

colorful

described by the frequency response characteristics of a violin,

bright
violin

nasal

measured e. g. as the frequency response function of the sound

pressure at a specific point in the room (e. g. at the listener’s ear)
related to the force input at the bridge in bowing direction.
no. 1 0.67 0.31 2.96 2.96 0.90 0.55 0.27
Conceptually such a frequency response function can be
no. 2 0 0 2.75 3.48 2.02 1.49 0.55 subdivided into the structural frequency response functions of
no. 3 1.81 0.31 1.98 0.98 2.81 1.42 0 the acceleration or velocity of various points on the violin body
in normal direction related to the force input at the bridge, and
no. 4 2.04 0.52 0 1.89 1.59 0.03 0.55 the vibroacoustic frequency response functions of the sound
no. 5 2.02 0.96 1.17 2.38 2.24 1.14 0.55 pressure at the specific point in the room related to the
accelerations or velocities at the violin body.
no. 6 1.12 0.69 0.69 1.76 1.24 1.17 1.24 How accurate must the frequency response characteristics
no. 7 2.34 0.14 0.69 0.44 1.34 1.38 0.69 of the copy meet those of the reference? It is known that the
sound pressure frequency response level of a room, when it is
no. 8 2.14 1.48 2.62 0 0 0 1.65 excited by a sinusoidal sweep, shows irregularities due to the
overlap of the modes and phases. Cremer and Müller [10]
Table 3: BTL – scaling values, calculated from the cumulated showed, based on results from Kuttruf and Thiele [8] and
preference frequency of the consistently answering test persons. Schröder [9] that the number of level maxima nR in a given fre-
quency band ∆f can be written as
According to table 3 violin no. 1 is not particularly bright T
(0.67) but sounds very pleasant (2.96) and has a good reaction nR = ∆f (1)
(2.96). On the other hand violin no. 7 is ranked to be very bright 6.7
(2.34) but ranked low for pleasant (0.69) and reaction (0.44). with T being the reverberation time of the room.
Table 3 also shows that the violins in the test have no significant From this result on can estimate the number of level
difference in the ranking concerning the attribute nasal. maxima in the 1/12-octave centered around a halftone fH, nRH, to
be approximately nRH ≅ 0.009⋅TfH.
In a concert hall with a typical reverberation time of 2 s this
results approximately in 4 level maxima for the band around the

SMAC-3
Proceedings of the Stockholm Music Acoustics Conference, August 6-9, 2003 (SMAC 03), Stockholm,
Sweden
lowest tone of a violin (G3, fH ≈ 200 Hz) and 115 level maxima
for the band around the highest tone, five octaves above the
lowest tone, (G8, fH ≈ 6400 Hz). For a reverberation time of 0.5 s
– which is typical for a small living room or a violin maker’s
workshop – one would have about 1 level maximum per halftone
at 200 Hz and about 29 maxima per halftone at 6400 Hz.
A violin has typically only 1 mode, the so-called Helm-
holtz-resonance, in the frequency range 200 Hz to 400 Hz. The
number of level maxima per halftone for a violin, nVH, in this
frequency range can be estimated from the above result to be
approximately nVH ≅ 0.0003⋅fH. For 400 Hz this number is 0.12.
It is considerably smaller than the number of level maxima for a
room. Therefore it is important to match the Helmholtz-reso-
nance of the copy exactly to the Helmholtz-resonance of the
reference violin.
For frequencies between 400 Hz and 800 Hz one usually
observes 2 to 4 modes in frequency response curves of sound
radiated from violins. For 4 modes the number of level maxima
per halftone would be nVH ≅ 0.0006⋅fH. For 800 Hz one would
calculate the number of level maxima per halftone to be roughly
0.5, which is still smaller than the equivalent number for rooms.
The exact matching of eigenfrequencies in the copy is important
in this frequency range, too.
The exact matching of eigenfrequencies becomes less im-
portant, however, when nVH ≥ 1, i. e. when there is at least one
mode in the violin body per halftone. Estimates based on the
modal density of a 5 mm flat plate indicate that this happens at
about 1500 Hz. Woodhouse reports similar results based on
estimates of modal overlap factors [13], and observations from
mobility measurements show that the transition frequency where
nRH ≥ 1 can be even lower. Above approximately 1500 Hz the
exact matching of single modes is not very important as long as
the overall modal density and damping are preserved.
The limits in the resolution of human hearing can be stated
by the so-called just-noticeable sound changes. Data for just-
noticeable sound changes vary to a certain extent between differ-
ent authors. In [7] it is stated that the just-noticeable relative
frequency difference for pure tones between the 500 Hz and
2000 Hz octave presented to a listener for 500 ms or longer is
roughly 0.002 x frequency, i. e. about 2 Hz around 1000 Hz. For
octaves below 500 Hz and above 2000 Hz these values are lar-
ger. Also, for a shorter duration of the tone presentation – which
is likely when one evaluates the sound quality of different vio-
lins - these differences increase and are for a duration of 10 ms
approximately 10 times higher than the values stated above, i. e.
20 Hz around 1000 Hz. Concerning the sound level it is stated
that the just-noticeable level difference for white noise is about
0.5 dB for levels above 25 dB.
With an optimized measurement procedure, using a ham-
mer pendulum system, one is able to measure frequency re-
sponse functions, including the radiated sound pressure, up to 3
kHz with an uncertainty of 0.5 dB (and from 3 kHz to 10 kHz
with an uncertainty of 2 dB). The frequency resolution is typi-
cally dependent upon the measurement time but can be selected
so that the frequency resolution is smaller than the above men-
tioned just-noticeable changes.
5. CONCLUSIONS
It was shown that there are sound quality features of violins that
a relevant for musicians and do evoke similar perceptions in
SMAC-4
different listeners. These features can be scaled and used for
consistent sound quality ranking of violins. The features can be
analyzed from measured frequency response functions using
psychoacoustic models. For frequencies below 1500 Hz the
modal characteristics of the “tonal copy” and the reference violin
must be identical or very similar. For higher frequencies the
average modal density and damping must be preserved. The
uncertainty of the frequency response measurement is smaller
than the just-noticeable changes of our hearing perception. Thus
a violin maker can in principle measure the modal characteristics
of the copy for the final instrument and during the working
progress and evaluate if the sound quality is already “near
enough” – i. e. aurally indistinguishable - to the reference violin.
It is very clear that these results do not make violin making
easier, because it needs intuition, professional experience and
excellent workmanship to select the correct starting geometry
and to continuously control and tune the modal characteristic of
the violin.
6. REFERENCES
[1] Schleske, M., “Empirical tools in contemporary violin
making: Part I. Analysis of design, materials, varnish and
normal modes”, J. Catgut Acoust. Soc., vol. 4, no. 5 (Series
II), p. 50-64, 2002.
[2] Moore, B. C. J., Glasberg, B. R., Baer, T., “A model for the
prediction of thresholds, loudness and partial loudness”, J.
Audio Eng. Soc. 45, 1997.
[3] Zwicker, E., “A program for calculating loudness according
to DIN 45631 (ISO 532 B)”, J. Acoust. Soc. Japan (E), 12,
p. 39-42, 1991.
[4] Benedini, K., “Ein Funktionsmodell zur Beschreibung von
Klangfarbenunterschieden”, Biol. Cybernetics 34, p. 111-
117, 1979.
[5] v. Bismarck, G., “Timbre of steady sound. A factorial
investigation of its verbal attributes”, Acustica 30, p. 146-
159, 1974.
[6] v. Bismarck, G., “Sharpness as an attribute of the timbre of
steady sounds”, Acustica 30, p. 159-172, 1974.
[7] Zwicker, E., Fastl, H., Psychoacoustics - Facts and Models,
2nd revised edition, Berlin, 1999.
[8] Kuttruff, H., Thiele, R, “Über die Frequenzabhängigkeit des
Schalldrucks in Räumen”, Acustica 4, p. 614, 1954.
[9] Schröder, M.; “Die statistischen Parameter der Frequenz-
kurven von großen Räumen”, Acustica 4, p. 594, 1954.
[10] Cremer, L., Müller, H. A., Die wissenschaftlichen Grundla-
gen der Raumakustik, Band II, Teil 4, Wellentheoretische
Raumakustik, 2. Auflage, Stuttgart, 1976.
[11] Bortz, J., Lienert, G. A., Boehnke, K., Verteilungsfreie
Verfahren in der Biostatistik, Berlin, 2. Auflage, 2000.
[12] Bradley, R. A., Terry, M. E., “Rank analysis of incomplete
block design. I. The method of paired comparison.” Bio-
metrika, Vol. 39, p. 324-345, 1952.
[13] Woodhouse, J., “Body vibration of the violin – what can a
maker expect to control?”, J. Catgut Acoust. Soc., vol. 4,
no. 5 (Series II), p. 43-49, 2002.