Conference PaperPDF Available

Ebony vs. Rosewood: experimental investigation about the influence of the fingerboard on the sound of a solid body electric guitar

  • Institut Supérieur de l'Electronique et du Numérique, Lille, France
Ebony vs. rosewood: experimental investigation about the influence of the
fingerboard on the sound of a solid body electric guitar
Arthur Pat´
e, Jean-Lo¨
ıc Le Carrou, Benoˆ
ıt Fabre
LAM / Institut Jean Le Rond d’Alembert
UPMC Univ Paris 06, CNRS UMR 7190
Beyond electronics, lutherie also has something to do with
the sound of the solid body electric guitar. The basis of
its sound is indeed the conversion of the string vibration to
an electrical signal. The string vibration is altered by cou-
pling with the guitar at the neck. Electric guitar lutherie
being a huge topic, this paper focuses on the influence of
the fingerboard on the string vibration. An experimen-
tal study is carried out on two guitars whose only inten-
tional difference is the fingerboard wood: ebony or rose-
wood. The well-known ”dead spot” phenomenon is ob-
served, where a frequency coincidence of string and struc-
ture at the coupling point leads to an abnormal damping
of the note. Striking is the different behaviour of each fin-
gerboard wood about dead spots: affected notes, as well as
how much they are affected, differ with the wood.
Physical studies about the solid body (without soundbox)
electric guitar have been mainly focused on electronics,
whether it is on the string transduction by the pickup [1,2],
the effects and processing chain [3] or the amplifier [4],
often with the purpose of doing numerical synthesis. The
characteristics of the pickup (transducing the velocity of
the string into an electrical signal), effect pedals (trans-
forming this signal with endless possibilities), amplifier
(far away from high-fidelity), loudspeakers (reproducing
and distorting the final sound) are of course of significant
importance. But lutherie is at least partially responsible
for the sound. The vibration of the string is altered by the
coupling to a moving structure (the guitar) at its ends. The
structure may vibrate and exchange energy with the string,
like it is the case for e.g. the classical guitar [5] or the harp-
sichord [6].
The coupling of a string to a structure is described in [7].
The admittance of the structure at the coupling point causes
the frequencies and dampings of the coupled-string partials
to differ from those in the uncoupled case (string with two
rigid ends). This admittance at the coupling point is known
as the ”driving-point admittance”. It is defined by the ra-
tio in the frequency domain between the velocity V(ω)of
Copyright: c
2013 Arthur Pat´
e et al. This is an open-access article distributed
under the terms of the Creative Commons Attribution 3.0 Unported License, which
permits unrestricted use, distribution, and reproduction in any medium, provided
the original author and source are credited.
the structure at the coupling-point and the force F(ω)ap-
plied on the same point. Driving-point admittance can be
obtained by classical measurements on relevant coupling
points between the string and the structure, typically on the
neck [8,9]. The real part of this driving-point admittance is
called the driving-point conductance. It provides addition-
nal damping to the string [7]. Measurements in [9] qualita-
tively link a measured high conductance value at a specific
frequency with the fast decay of the note at the same fre-
quency, when fretting point and measurement point are the
same. Notes affected by an abnormally big damping are
known as ”dead spots”. Damping inhomogeneity among
notes is known to be disturbing for the players.
The vibrational behaviour of the structure, seen by the
string as end conditions, is influenced by many parame-
ters. Electric guitars can differ in many things [10]: shape
of the body and headstock, wood used for body, finger-
board or neck, bridge type, nut material, size and material
of the frets, neck profile. . . Each of these lutherie parts
changes the vibrational behaviour of the structure and then
may alter the sound.
Fleischer and Zwicker [8] studied a Gibson Les Paul and
aFender Stratocaster, which have been the two reference
models in the electric guitar industry [10]. Differences in
modal behaviour are found and are attributed to the sym-
metry of the headstock. However, these two guitars dif-
fer not only in the headstock shape, but also in the wood
species, the body shape, . . .
In order to draw conclusions about the influence of a
lutherie parameter, this parameter should be the only vary-
ing one. This paper is part of a broader project [11, 12]
aiming at studying the influence of each lutherie parameter
taken separately. Here the spotlight is on the study of the
fingerboard wood on the sound. An experimental investi-
gation of ebony and rosewood fingerboards is presented.
These are two out of three (the other one being maple) typ-
ical woods used for solid body fingerboards.
Section 2 gives a simple model of string-structure cou-
pling and its consequences on string frequency and dam-
ping. Section 3 describes the experimental protocol, and
quantitatively checks the model of section 2: the string
damping value can be predicted from the conductance value.
Section 4 discusses the change in sound induced by the
change in fingerboard wood.
A simple model of a stiff lossy string connected at one
end to a moving body is proposed. The moving-end string
model is derived as small perturbations of the stiff string
model simply-supported at its two ends. The theory has al-
ready been detailed by [7] and it is briefly described here.
Let xbe the axis of the string at rest position and yits mo-
tion normal to the fingerboard plane. The string is simply-
supported at x= 0 and x=L. It is stretched with tension
T. The string is also characterised by its mass per unit
length ρL, second moment of area Iand Young’s Modulus
E. The dispersion relation is:
where c=qT
ρLis the wave velocity, κ=qT
EI is the
stiffness term and k0
Lis the quantized (nis a posi-
tive integer) wavenumber for simply-supported end condi-
The string loses energy through three damping mecha-
nisms: visco-elasticity, thermo-elasticity and air damping.
Let ξ0
nbe a damping coefficient taking into account those
three damping mechanisms. It is frequency-dependent be-
cause it depends on the partial number n. The damping
nis generally added as the imaginary part of the complex
angular frequency, so that the dispersion relation becomes:
Yet the string is not simply-supported at its ends. One end
is connected to the bridge and the other end is connected
at the neck to a fret or to the nut. What [8, 9] showed was
checked: most of the time the motion of the end connected
to the bridge is small compared to the motion of the end
connected to the neck. The bridge end (at x= 0) is still
assumed to be rigid whereas the other end (at x=L) is
connected at the neck to the admittance of the moving gui-
tar. The moving end at x=Lonly causes small perturba-
tion δkn1to the wavenumber k0
n, so that the corrected
wavenumbers kn=k0
n+δknare used.
At x=L, the string’s admittance is defined as the ratio
between its velocity and the force being applied on it:
Ystring (L, ωn) =
∂t (L, t)
∂x (L, t)=jtan(knL)
where Zc=ρLTis the characteristic impedance of the
string. At x=L, the string and the structure are connected
and must have the same admittance. Letting Y(L, ω)be
the admittance of the structure at the connection point, one
has :
Y(L, ω) = Ystring (L, ω)(4)
Remembering that tan(k0
nL)=01and assuming that
ZcY(L, ωn)12, equation 3 leads to the expression of
nis the the simply-supported end solution for the wavenumber
2The impedance of the structure is much greater than the characteristic
impedance of the string, resulting in a reflection of travelling waves in the
string at the connection point
LjY(L, ωn)Zc
with which equation 1 becomes:
L1 + n2π2E I
2T L220
njY(L, ωn)Zc
ωnare the complex angular frequencies of a stiff lossy
string having a moving end. Modal frequency is defined
as :
2L1 + n2π2
Im(Y(L, ωn))
and modal damping as:
2Re(Y(L, ωn)) (8)
The imaginary part of the body admittance implies a shift
in the simply-supported string frequencies, affecting the
inharmonicity [13]. Nevertheless, measured admittance
imaginary parts on the tested guitars never lead to a fre-
quency shift larger than 1Hz. For this reason this paper
only discusses the influence of the real part of the admit-
tance, the conductance.
The main effect of string-structure coupling is the damping
due to the conductance. The experimental study identifies
the conductance terms C(ω) = Re(Y(ω)) at the points
where the strings couple to the structure 3, that is on the
3.1 The two guitars of the study
This experimental study is intended to determine what dif-
fers in the sound when changing the fingerboard wood.
The fingerboard should therefore be the only varying lu-
therie parameter. In order to fulfil this recommendation,
a collaboration with instrument-makers was developped.
Two guitars were made by luthiers from Itemm 4, a french
leading lutherie training-center. The two guitars follow the
specifications of the Gibson Les Paul Junior DC, a ver-
sion of one of the two most important solid body electric
guitars in history [10]: original shape, quartersawn ma-
hogany for body, neck and head, set-in neck, same equip-
ment (bridge, bone nut, P-90 pickup). The only inten-
tional difference between the two guitars is the fingerboard
wood. One guitar has an ebony fingerboard (E) and the
other one a rosewood fingerboard (R). It should be kept
in mind that other parameters may differ between the two
guitars, mainly because of the wood variability and the
handmade process. For schedule reasons, the guitars could
not be measured prior to the gluing of the fingerboard.
3Measurements of this section are made at the connected end of the
string, so the dependence in Lof Y(L, ω)is no longer specified.
impact point
woodpiece bridge
Figure 1. Setup for driving-point conductance measure-
ment along the 5th string’s axis, at a particular fret. An
accelerometer is put on the one side of the fret. The ham-
mer strikes at the other side of the fret. A very light piece
of wood moves aside the strings and allows the accelerom-
eter to stay between the two strings.
Samples of the woods used for the fingerboards were pro-
vided. Ebony density ρE=1180kg.m1and rosewood
density ρR=751kg.m1are simply measured. Longi-
tudinal Young’s moduli EE= 3.02 1010Pa and ER=
2.30 1010Pa are identified with simple bending test. The
two fingerboard woods have different characteristics: fit-
ting the neck with one or another fingerboard wood may
then change the vibrational behaviour of the instrument.
3.2 Experimental setup
The experimental setup is sketched in figure 1. The con-
ductance is measured at every potential coupling point be-
tween string and structure, that is at every fret-string cross-
ing on the neck. As in [9], only the conductance normal
to the plane of the fingerboard is studied. Only the cou-
pling of the string polarisation in this direction is studied
in this paper. As usual, force F(ω)is applied with an im-
pact hammer equipped with a force sensor, and velocity
V(ω)is measured with an accelerometer. Impact and mea-
surement points must be as close as possible in order to ob-
tain actual driving-point conductance. The modal domain
(where peaks and modes are well identified) is from 20Hz
to 700Hz. The useful impact bandwidth is from 20Hz to
2000Hz. It is decided to consider the coupling of only the
fundamental frequency with the structure, so that n= 1 in
all equations of the section 2. The guitar is laid on elastic
straps supported by a frame. Resonant frequencies of the
system {frame-straps}is below the resonant frequencies of
the guitars, so that this setup provides a good approxima-
tion for free boundary conditions. Modeling clay is put on
the pegs and on the screw of the truss rod to prevent them
from vibrating. Paper is used to avoid string vibrations,
which are unwanted here for the study of the guitar only.
Section 3.3 experimentally checks the model of section
100 200 300 400 500 600 700 800 900 1000
Frequency [Hz]
Re(Y) [m.s−1.N−1]
f0(F4) = 349 Hz
Figure 2. Driving-point conductance at the 6th fret along
the 2nd string’s axis. Solid line is used for the guitar E
and dashed line for the guitar R. Gray line highlights the
fundamental frequency of the F4 played at this place.
3.3 Validation of the model
In order to validate the model of section 2, a simple check
is done. Figure 2 is an example (further discussed in sec-
tion 3.4) of measured driving-point conductance: here at
the 6th fret along the 2nd string for both guitars. The cor-
responding note is F4 with fundamental frequency fF4=
349Hz. This note is also recorded by picking the string
with a guitar pick, fretting the 6th fret with a capo and
taking the output signal of the guitar pickup. Figure 3
shows the temporal evolution of the fundamental of this
note. This temporal evolution is extracted from the recorded
signal. It is obtained by computing a short-time Fourier
transform of the signal and determining the envelope of
the bin centered on the fundamental frequency. The time
constant τ, defined as the time needed for the amplitude to
get divided by Euler’s number e, can be estimated from the
fundamental envelope signals for guitars E and R. The esti-
mation of the global damping terms ξE
F4= (2πfF4τE
2.8 104and ξR
F4= (2πfF4τR
F4)1= 1.4 103is straight-
forward. Superscripts ”R” and ”E” refer to guitar R and E
respectively. Subscript refers to the note.
Identical string sets provided the string for both guitars.
String damping ξ0
F4is then assumed to be the same for
both guitars. The magnetic pickup could be a cause of
string damping as well. Since the guitars are equipped with
pickups of the same model series, the magnetic damping is
assumed to be the same for both guitars. This magnetic
damping can be included in ξ0
F4. According to equation 8
with C(ω) = Re(Y(ω)), one expects :
Estimating the conductance values from measurements pre-
sented in figure 2 (CE(2πfF4)=1.9 103m.s1.N1and
CR(2πfF4)=3.2 102m.s1.N1) leads to:
F4= 1.1 103(10)
2πCR(2πfF4)CE(2πfF4)= 1.0 103
F4can be identified by subtracting the term Zc
from the experimental ξF4. The line with crosses in figure
Time [s]
Normalised amplitude
Measure (E)
Measure (R)
Model with rigid ends
Figure 3. Temporal evolution of the fundamental fre-
quency of the F4 (6th fret and 2nd string) played on both
guitars. Solid line is for guitar E, dashed line is for gui-
tar R, crosses show the computed decay of the same string
with rigid ends.
3 is the decay curve with identified ξ0
F4= 2.3 104.
The small difference between the two lines of equation
10 can be explained by our estimation of τ, and by the ac-
curacy of our measurements. Nevertheless, the two values
are quite close and the model of section 2 is validated.
3.4 Observation of dead spots
Figure 2 shows that at the fundamental frequency of the
note, the conductance takes a low value for the guitar E.
Thus the factor Re(Y)in equation 8 is small and so should
be the string damping due to coupling with support. This is
checked in figure 3 and in the calculations of section 3.3:
the experimental computed decay is close to the intrinsic
decay (i.e. the rigid-ends case).
On the other hand, figure 2 shows a high conductance
value for the guitar R. Figure 3 and calculations of section
3.3 confirm the ”abnormal” damping of the fundamental.
This damping is indeed higher for the guitar R, and the de-
cay curve exhibits two slopes instead of the single slope
decay for ”normal” cases.
These two phenomena are consistently checked on the
two guitars: a low conductance value leaves the note’s
decay unperturbed (live spot), a high conductance value
makes the decay of the note shorter (dead spot).
When looking at other notes, it is found that both finger-
boards exhibit dead spots. However, the note studied in
this section showed a difference between the two finger-
board woods: for the same note at the same place on the
neck, a guitar exhibited a live spot whereas the other ex-
hibited a dead spot. Section 4 deals with the differences in
sound that may appear between the two guitars.
The two fingerboard woods lead to the same dead/live spot
phenomenon. However, it does not break out the same way
depending on the instrument.
Figure 4. Guitar E: driving-point conductance values in
the frequency range [20Hz–1000Hz] for all frets along the
2nd string. White crosses spot the fundamental frequencies
of the notes played at each fret of the 2nd string. Unit of
conductance is m.s1.N1.
Figure 5. Rosewood-fingerboard guitar: driving-point
conductance values in the frequency range [20Hz–
1000Hz] for all frets along the 2nd string. White crosses
spot the fundamental frequencies of the notes played at
each fret of the 2nd string. Unit of conductance is
4.1 Dead spot location
Section 3.4 indicates that a difference between the two gui-
tars is the places where dead spots occur. Since the repre-
sentation of figure 2 is hard to handle if one wants to have
an overview of every fret of one string, figures 4 and 5 pro-
pose a synthetic view of the string-structure frequency co-
incidences, a ”dead spot map”. The frequency-dependant
driving-point conductance values at every fret along one
string are represented on the same plot. Conductance value
is transcribed as a continuous color coding from blue (very
low conductance) to red (conductance peak). For each
fret, the fundamental frequency of the note is plotted with
a white cross. Hence, whenever a white cross gets close
enough to a red spot, a dead spot is reached.
Figures 4 and 5 can be used to quantify the number of
dead spots of a guitar. Here on the 2nd string, the rosewood-
fingerboard guitar has one dead spot (6th fret) and none at
any other place. ”Corresponding” dead spot for the gui-
tar E is moved to the 7th and 8th frets. A first remark can
be done, when looking at the ”dead spot maps” for all six
strings (five are not showed in this paper): the number of
Figure 6. Driving-point conductance at the 1st fret along
the 4th string’s axis. Solid line is used for the guitar E and
dashed line for guitar R. Gray line highlights the funda-
mental frequency of the D#3 played at this place.
dead spots is roughly the same between the two guitars, but
their locations often slightly (one or two frets) differ.
A second remark is that the conductance peaks (for ex-
ample figures 4 and 5 around 400Hz and between 600Hz
and 700Hz) seem to be higher for the guitar R than for the
guitar E. This is the purpose of section 4.2.
4.2 Dead spot dangerousness
Another kind of difference between the two fingerboard
woods is the amplitude of conductance peaks, that is the
potentially high damping of the note. That is what we
call the ”dangerousness” of a dead spot. Most of the mea-
surements show that higher values of driving-point con-
ductance are reached for the rosewood-fingerboard guitar.
Figure 6 illustrates this tendency. It shows the measure-
ment at 1st fret along the 4th string for both guitars. The
conductance at the frequency of the note (D#3,155Hz) is
high for both guitars. Estimation of experimental damping
coefficient ξas in section 3.3 leads to ξE
D#3 = 1.0 103
and ξR
D#3 = 2.4 103in this case. ξ0
D#3 is estimated as
in section 3.3: ξ0
D#3 = 3.0 104. For both guitars, ξD#3
is much higher than ξ0
D#3: this clearly reveals a common
dead spot. However, the dead spot is more pronounced for
the guitar R than for the guitar E. Guitar R’s conductance
peak is closer to the frequency of the note than guitar E’s
one, and guitar R also has a higher peak guitar E’s one.
Whatever the tuning (determining the frequencies of the
notes) is, the string-structure coupling still occurs because
the neck conductance still takes non-zero values. In order
to characterise this conductance amplitude difference ten-
dency between guitars E and R in a more tuning-indepen-
dent way, a mean conductance value is computed. For each
measurement (each fret) along a string, the mean of the
conductance is computed in the frequency range [20Hz–
2000Hz]. Figure 7 presents these computed mean conduc-
tance values as a function of the place on the neck along
the 2nd string. For every fret the guitar R clearly stands
out from the guitar E with systematically higher mean con-
ductance. This would mean that whatever the tuning is, the
rosewood-fingerboard guitar is likely to grasp more vibrat-
ing energy from the string.
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
x 10
Fret number
Mean Re(Y) [m.s1.N1]
Figure 7. Mean value of conductance in the fre-
quency range [20Hz–2000Hz] as a function of fret num-
ber/measurement place along the 2nd string. Solid line is
used for the guitar E and dashed line for the guitar R.
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22
Fret number
Figure 8. Computed time constants τfor each note of the
2nd string. A bandpass filter ([20Hz-2000Hz]) was applied
to each note. Solid line is used for the guitar E and dashed
line for the guitar R.
In figure 7 it can be seen that above 15th fret, the two
curves become closer and the mean conductance tends to
become smaller. This is because 15th fret and upper frets
are close to the neck-body junction, an area where the neck
motion is smaller.
These mean conductances can be linked to the computed
time constants τfor each note along the 2nd string showed
in figure 8. This computation is slightly different from sec-
tion 3.3: a bandpass-filter (20Hz to 2000Hz) is applied to
the pickup signal and τis computed from this filtered sig-
nal. Guitar E almost always has a higher time constant.
The smaller damping (for every partial in the frequency
range [20Hz-2000Hz]) due to the smaller mean conduc-
tance for guitar E results in a higher time constant τ.
The mean conductances and the time constants are com-
puted along the five other strings. The tendency is con-
firmed: the guitar E almost always has a lower mean con-
ductance value and a higher time constant. Rosewood might
then perturb the string more than ebony.
Previous results on the influence of the structure on the
vibration of the string have been confirmed. Because the
vibrational behaviour of the electric guitar is highly depen-
dent on the lutherie parts, which are numerous, it was de-
cided to focus on the influence of a single lutherie parame-
ter: the most prominent difference between the two guitars
of the study was the wood of the fingerboard (ebony or
rosewood). Comparative study of sound and driving-point
conductance on these two guitars indicate that the wood of
the fingerboard may have an influence upon the:
dead spot location: the spatial and frequency coin-
cidence of string and guitar resonances happens at
different places depending on the fingerboard wood
dead spot dangerousness: when this coincidence
happens, the string damping may be bigger for rose-
wood-fingerboard guitar
Experimental investigation about dead spots and the re-
lated discussion are naturally not only valid for the fun-
damental of the string but also for partials. As equation 8
shows, each string partial may couple with higher structure
modes. Hence the timbre is affected by the fingerboard.
The sound differences that may be induced by the change
of fingerboard wood can then have consequences in:
instrument-making: the guitar maker could attempt
to change the resonance coincidences: for example
fingerboard thickness, shape (the so-called ”slim”
and ”slapboard” fingerboards by Fender) or sawing
angle are parameters changing the modes of the struc-
ture. Hence the instrument-maker can reduce the
differences between the woods or on the contrary in-
crease them.
playing: the same note can be played at different
places on the neck. Depending on the location and
dangerousness of the dead spots, the player may be
forced to avoid certain places on the neck and to con-
form his playing to the guitars’ sound.
tuning: actually, the frequency coincidence between
the string and the structure depends on the tuning of
the string. In order to avoid a too strong coupling,
the guitar player can slightly change the tuning of the
strings. This could be an explanation to the fact that
some guitar players say that a guitar sounds better
with a special tuning (e.g. all the strings a whole-
tone lower) than with the standard tuning A-440Hz.
A perceptual study involving the two guitars of this paper
has been carried out. The analysis is in progress and is
expected to tell us to what extent the differences found here
are perceptible for the guitar player.
The authors pay a special tribute to the luthiers Vincent
Charrier, Lo¨
ıc Keranforn, Lisa Marchand, Bela Pari, Alex-
andre Paul and Julien Simon and warmly thank as well
Yann-David Esmans, Fred Pons and Pierre Terrien from
Itemm. The efficient mediation of Vincent Doutaut made
the collaboration with instrument-makers possible.
The authors also thank R´
emi Blandin for his helpful con-
tribution to the measurements on the guitars, and Benoˆ
Navarret for fruitful discussion.
[1] D. Queen, “From boom boxes to Beatles, Baez and
Boccherini, the electric guitar at the crossroads,” in
Proceedings of the 31st Convention of the AES, 1966.
[2] N. G. Horton and T. R. Moore, “Modeling the mag-
netic pickup of an electric guitar,American Journal of
Physics, vol. 77 (2), pp. 144–150, 2009.
[3] O. Kr ¨
oning, K. Dempwolf, and U. Z ¨
olzer, “Analysis
and simulation of an analog guitar compressor,” in Pro-
ceedings of DAFx11, 2011.
[4] I. Cohen and T. H´
elie, “Real-time simulation of a guitar
power amplifier,” in Proceedings of DAFx10, 2010.
[5] C. Lambourg and A. Chaigne, “Measurements and
modeling of the admittance matrix at the bridge in gui-
tars,” in Proceedings of SMAC 93, 1993, pp. 448–452.
[6] N. H. Fletcher, “Analysis of the design and perfor-
mance of harpsichords,” Acustica, vol. 37, pp. 139–
147, 1977.
[7] A. Hirschberg, J. Kergomard, and G. Weinreich, Me-
chanics of musical instruments. Wien–New York:
Springer Verlag, 1995.
[8] H. Fleischer and T. Zwicker, “Mechanical vibrations
of electric guitars,” Acta Acustica united with Acustica,
vol. 84, 1998.
[9] H. Fleischer and T. Zwicker, “Investigating dead spots
of electric guitars,” Acta Acustica united with Acustica,
vol. 85, pp. 128–135, 1999.
[10] A. Pat´
e, B. Navarret, R. Dumoulin, J.-L. Le Carrou,
B. Fabre, and V. Doutaut, “About the electric guitar: a
cross-disciplinary context for an acoustical study,” in
Proceedings of Acoustics 2012, Nantes, France, 2012,
pp. 2717–2722.
[11] J.-L. Le Carrou, J. Frelat, A. Mancel, and B. Navar-
ret, “Guitare ´
electrique : quel r ˆ
ole pour les ´
ements de
lutherie ?” in 10`
eme Congr`
es Franc¸ais d’Acoustique,
[12] A. Pat´
e, J.-L. Le Carrou, B. Navarret, D. Dubois, and
B. Fabre, “A vibro-acoustical and perceptive study of
the neck-to-body junction of a solid-body electric gui-
tar,” in Proceedings of Acoustics 2012, Nantes, France,
2012, pp. 2711–2716.
[13] C. E. Gough, “Acoustical studies of stringed instru-
ments using string resonances,” in Proceedings of
SMAC 83, 1983, pp. 19–45.
... Cependant, c'est bien le mouvement de la corde qui fournit le signalà l'entrée de la chaîneélectroacoustique : la corde est donc au moins partiellement responsable du son de l'instrument. Etant connectéeà une structure vibrante, elle voit son comportement dynamique perturbé par couplage mécanique [3,4,5]. Chaqueélément de lutherie (bois et forme du manche, du corps, type de frette. . ...
... Les guitares du test perceptif ontégalement fait l'objet d'uneétude mécanique [5]. Celle-ci montre que les guitares a touche en palissandre présentent des conductances sur le manche plusélevées : elles ont donc davantage tendanceà altérer la vibration des cordes. ...
... Rosewood is also used for tailpieces and pegs of violins (Hiziroglu 2016). A comparative study of sound on two guitars using ebony and rosewood indicated that rosewood as the fingerboard was superior (Paté et al. 2013), accentuating its demand. The difference among guitars having the back in rosewood (D. latifolia) as defined by guitar makers is mainly a longer and more sustained sound (Bucur 2016). ...
Full-text available
Key message Fortified forest management strategies, coupled with in situ and ex situ conservation strategies, are needed for long-term conservation of rare, valuably important Dalbergia latifolia. Abstract Indian rosewood (Dalbergia latifolia Roxb.; Fabaceae) is a highly valued and commercially important tree species. In its global natural habitats, D. latifolia is sparsely distributed. Apart from its uses in furniture, plywood, veneer and carved wood products, it is globally known for its use in the guitar industry. Possessing unique tonal properties and beautiful wood finishing, it is considered an iconic material in the manufacture of guitars. High demand for the wood, coupled with its growing rarity, has resulted in excessive harvest and an illegal trade. In parallel, habitat destruction has led to a rapid decline in its natural populations globally. Today, most of the best trees have already been extracted, and rosewood forests are under severe threat of depletion. Conservationists worldwide are concerned about the fate of rosewood mainly because it takes many decades to grow to a commercially viable size and centuries to reach full maturity. With the increasing rarity of timber, exponentially growing demand for wood, and rampant illegal trade, rosewood prices have soared in the global market. It has thus been listed in Appendix II of CITES from 2016 and categorised as ‘Vulnerable’ by the IUCN. This review provides a comprehensive set of information about the taxonomy, distribution, morphology, pollination and seed biology, regeneration, insect pests and diseases, and wood quality of Indian rosewood. It also discusses trade, alternative species, and conservation options. The natural habitats of D. latifolia need to be more stringently monitored so that appropriate management interventions can be adopted to conserve the genetic resources of this valuable timber species. Since the regeneration status of D. latifolia in forest areas is still unknown, long-term monitoring and an understanding of the factors influencing stand establishment would ensure that this tree species is preserved in natural forests.
... As this study has shown, by using a material, i.e., ash wood with greater mechanical stiffness (Table 2), we achieve higher modal frequencies of the body of the electric guitar as well as of the whole instrument (Table 3). In this way, it also influences the characteristics, i.e., the modal frequency of the guitar neck, which is otherwise considered to be more important for the acoustic behavior of an electric guitar compared to the body of this instrument [3,13,27]. ...
Full-text available
Research show that the vibrations of the strings and the radiated sound of the solid body electric guitar depend on the vibrational behavior of its structure in addition to the extended electronic chain. In this regard, most studies focused on the vibro-mechanical properties of the neck of the electric guitar and neglected the coupling of the vibrating strings with the neck and the solid body of the instrument. Therefore, the aim of the study was to understand how the material properties of the solid body could affect the stiffness and vibration damping of the whole instrument when comparing ash (Fraxinus excelsior L.) and walnut (Juglans regia L.) wood. In the electric guitar with identical components, higher modal frequencies were confirmed in the structure of the instrument when the solid body was made of the stiffer ash wood. The use of ash wood for the solid body of the instrument due to coupling effect resulted in a beneficial reduction in the vibration damping of the neck of the guitar. The positive effect of the low damping of the solid body of the electric guitar made of ash wood was also confirmed in the vibration of the open strings. In the specific case of free-free vibration mode, the decay time was longer for higher harmonics of the E2, A2 and D3 strings.
... Numerical simulations involving all guitar elements can be conducted aiming to investigate possible alterations in tone derived from wood distinctions. In this context, Paté et al. (2013) studied alterations in sound and damping due to the conductance involved in stringstructure coupling for two electric guitars having ebony and rosewood fingerboards, and it was verified that whatever the tuning is, it is likely the rosewood fingerboard grasps more energy of vibration from the string. Sproßmann et al. (2017) affirm that a high hardness value indicates good abrasion resistance on the fingerboard surface due to strings scratching while guitar playing. ...
Full-text available
In face of scarcity in the supply of non-traditional Brazilian woods properly treated for use in high quality musical instruments, pieces of Amazonian wood species muiracatiara (Astronium lecointei) and maçaranduba (Manilkara huberi) purchased in the common internal Brazilian timber market were examined. These species were pre-selected for use in fingerboards of acoustic and electric guitars due to similar properties with ebony (Diospyros crassiflora). Variabilities of elastic modulus parallel to grain and density were investigated inside wooden pieces. In addition, referred parameters were used in calculation of speed of sound. Statistical tests were performed in order to compare both species and revealed inequality for variances of dynamic elastic modulus (Ed) and speed of sound, but equality for density. Equality of means was also examined via unequal variance t-test. Despite color differences, lower variability of M. huberi led to the indication of this species as likely capable to substitute satisfactorily ebony in fingerboards manufacturing.
... However, corresponding studies on the fretboard are still insufficient and unsystematic [14]. Paté et al. [15,16] studied two guitars that only differ in the fretboard material (ebony or Indian rosewood). The well-known "dead spot" phenomenon was observed, where a frequency resonance of the string and the main structure at the coupling point lead to an abnormal damping of the note. ...
Full-text available
Abstract Fretboards of string instruments are usually made of rare woods that commonly have a high density, strength, and hardness; further, they are wear resistant, uniform in texture, and feature an elegant color. To reduce the consumption of scarce timber resources, especially of endangered tropical hardwood species, suitable replacement materials should be identified. The substitute can be either common tree species having similar characteristics, or fast-growing plantation wood that has undergone modifications to match the performance of precious woods. This study compares the anatomical structure, physical features, mechanical properties, and surface color of three precious woods traditionally used in fretboards (ebony, Indian rosewood, and African blackwood) against maple, which is used for the backboard, ribs, and necks of string instruments. Based on the data, a set of performance evaluation indices for selecting alternative materials for fretboards is proposed. In specific, the replacement wood should be a diffuse-porous tropical hardwood with few vessels and a smaller diameter, thick fibrous walls, and a cell wall rate of more than 50%. In terms of physical properties, it should have low swelling coefficients for moisture and water absorption, and dimensional stability. The replacement should also display hardness values greater than 9.0 kN in the cross-section and greater than 6.0 kN in the tangential and radial sections. Further, it should have a high modulus of rupture (> 149 MPa) and elasticity (> 14.08 GPa), good impact bending strength, and good wear resistance (80–150 mg/100 r). To satisfy the traditional aesthetics, the wood surface color should be black, dark brown, or dark purple-brown, with colorimetric parameters in the range of 0.0
... Some lutherie parameters that are said (byl uthiers, players)t oh avea ni nfluence on the sound of the electric guitar were investigated in previous mechanical studies: the wood of the body [3], the neck-to-body junction [4] or the wood of the fingerboard [7]. One of the final goals of mechanical studies of musical instruments is to find the physical parameters that are relevant to the musicians. ...
... Just as in the case of acoustic instruments (see [1] or [2] for example), the electric guitar string couples with the structure of the guitar. This coupling has been found to be well described with the knowledge of the vibratory behaviour of the structure [3,4,5,6]. Vibratory measurements on musical instruments have almost always been performed on whole and finished instruments. One of the side results of such measurements is that nominally "identical" instruments can present notable differences in their vibratory behaviours [7,8]. ...
Full-text available
The objective of the study was to investigate the effects of wood on the sound generated by electric string instruments. The practical aspect was to select such material properties that would allow the producers of this particular instruments, to find the appropriate kind of wood that can be used in building neck and body, in order to create instruments of specific sound. The research has been divided into two variant forms: marking and analyzing psychoacoustics parameters which characterize the sound timbre (sharpness, roughness and specific loudness) and selected acoustic sound parameters (RMS amplitude, maximum sound pressure level, equivalent continuous A-weighted sound level) with the analysis of the sound envelope. Both research forms were conducted to simplify the complexity of design a model replaced the actual guitar, simultaneously limiting changing factors, such as those related with the instrument construction. Each model consists of a plank of selected wood species with constant dimensions, a tuner, a bridge, and an attached string and a record equipment. The research was conducted on different types of wood, which included: maple, ash, alder, cedar, mahogany, basswood and additionally thermo modified wood (alder). On every single wood sample used for the research were marked such properties as: macro structural parameters of the wood, density and dynamic modulus of elasticity, specific modulus of elasticity, damping of sound radiation, acoustic resistance and index of merit. Due to given results and their analysis, it became possible to state that the specific type of wood used in guitar productions has a significant influence on its final signal. Especially on its max sound pressure level and sound envelope. Higher modulus of elasticity of the wood in longitudinal direction gives higher sustain and max peak level. These results indicate that the selection of the wood, for the purpose of guitar production, should be mostly depended on the modulus of elasticity in longitudinal direction. This parameter is highly correlated with the basic acoustic sound parameters, such as: peak signal level, equivalent sound or forming of the signal and its sustain. On the other hand, any type of wood did not show any significant influence on psychoacoustics parameters (sharpness, roughness, and specific loudness). Statistical analysis of the given psychoacoustics parameters for the signals recorded with electromagnetic pickup and microphone, also did not show any relevant impact of the wood selection and its properties on those parameters. It has been maintained that the crucial factors are the way the sound is recorded and the frequency of the string. It has also been shown that the changes of the wood acoustic features due to its thermal modification, did not have any significant impact on the psychoacoustics parameters of the sound.
Full-text available
For the purpose of making of a solid body of an electric guitar, the acoustic and mechanical properties of walnut (Juglans regia L.) and ash wood (Fraxinus excelsior L.) were researched. The acoustic properties were determined in a flexural vibration response of laboratory conditioned wood elements of 430 × 186 × 42.8 mm used for making of a solid body of an electric guitar. The velocity of shear and compression ultrasonic waves was additionally determined in parallel small oriented samples of 80 × 40 × 40 mm. The research confirmed better mechanical properties of ash wood, that is, the larger modulus of elasticity and shear moduli in all anatomical directions and planes. The acoustic quality of ash wood was better only in the basic vibration mode. Walnut was, on the other hand, lighter and more homogenous and had lower acoustic-and mechanical anisotropy. Additionally, reduced damping of walnut at higher vibration modes is assumed to have a positive impact on the vibration response of future modeled and built solid bodies of electric guitars. When choosing walnut wood, better energy transfer is expected at a similar string playing frequency and a structure resonance of the electric guitar.
Full-text available
The vibrations of strings are influenced by their end supports. As a result of non-rigid supports, energy can flow from the strings to the body of an instrument causing the string signal to decay faster than in the case of rigid supports. In electric stringed instruments featuring a neck and a fretboard such as guitars or basses, this mechanism can evoke effects of practical relevance at particular locations on the fretboard which players denote "dead spots". For a precise understanding of the causes of this phenomenon the vibrations of the bodies of electric gui-tars were measured. In addition, the energy transfer via the end supports of the strings was assessed using a straightforward experimental procedure. Emphasis was put on the measurement of the mechanical conductance under realistic playing conditions in situ. Experiments on electric guitars revealed that the conductance at the bridge is generally smaller than at the neck. As a rule, the neck conductance proves to be smaller in the fretboard plane than perpendicular to the fretboard. The out-of-plane neck conductance is suggested as a relevant measure for characterizing the end supports of the strings in evaluations of electric guitars and basses, in particular the phenomenon of dead spots.
Conference Paper
Full-text available
The string motion of the solid body electric guitar is captured by an electromagnetic transducer sending an electrical signal to an amplification system, providing the sound to be perceived. Transducer and amplification have been so far well investigated, but the vibrational aspect of the instrument in connection with lutherie has been rarely considered. The aim of the present study is to analyse mechanically and perceptually the own influence of a single construction parameter. Three guitars, whose single difference is the neck-to-body junction, were made specially for this study. The neck can be either screwed or glued to the body, or have a neck-thru construction. The guitars have been played by professionnal guitarists along with semi-directed interviews. The judgments on the guitars are actually very varied and the guitarists have a lot of judging criteria, including criteria in assumed relationship with time-frequency aspect of the sound. We concentrate on the confrontation of time-frequency representation of the notes of the guitars in relation with driving-point conductance. In general when the driving-point neck-conductance measured at the string/guitar coupling point is high at the frequency of the note, an unusual damping of the fundamental frequency is visible on the spectrogram. We nevertheless find a non-neglectible number of exceptions to that.
Full-text available
A long decay of the string vibrations of an electric guitar ("sustain") is considered as a quality attribute. In practice, there are particular locations on the fretboard where for one of the strings the sustain is shorter than at adjacent frets. The player calls this irregularity a "dead spot". It originates from the fact that the string causes the neck of the guitar to vibrate. As a consequence, energy flows from the string to the neck which results in a faster decay.Three structurally different electric guitars (symmetric and asymmetric heads; neck screwed and glued to the body, respectively) served as measuring objects. In a first step, the decay times of the string signals were measured. In a second step, a technique was applied which allowed for in-situ measurements of the mechanical point conductance on the neck of guitars. The experiments revealed a clear inverse relation between the decay time of the string vibrations and the magnitude of the neck conductance. A local high neck conductance indicates a dead spot. In conclusion, the driving-point conductance, measured on the neck perpendicular to the fretboard, promises to be a key parameter for the diagnosis of dead spots.
Le son rayonné par une guitaré electrique provient essentiellement dusys emé electro-acoustique amplifi-cateur du mouvement de la corde. Pou etudier les différences sonores entre instruments, le etudes se sont légitimement focalisées sur les différents « microphones »captant le mouvement de la corde. Cependant, le comportement dynamique de la structure de la guitare est important du fait du couplage de la corde au chevalet et aux frettes du manche. Le eléments de lutherie (manche, corps, chevalet, ...) jouent donc, potentiellement, un rôle dans la vibration de la corde. L'objectif de ce travail est de qualifier l'influence du corps de la guitare. Dans ce but, uné etude expérimentale est menée sur trois instruments dont le matériau du corps est différent (médium, acajou et frêne).Ap es l'analyse modale de chaqué elément et des guitares montées, le comportement dynamique des instruments est analysé. De plus, uné etude détaillée des mobilités aux endroits de couplage avec les cordes est entreprise pour qualifier l'influence du comportement dynamique de l'instrument sur la vibration de la corde. Enfin, une modélisation préliminairé eléments-finis est mise en place dans le but de pouvoir quantifier, dans le futur, le rôle de eléments de lutherie dans la vibration des cordes de la guitare.
The digital modeling of guitar effect units requires a high phys-ical similarity between the model and the analog reference. The famous MXR DynaComp is used to sustain the guitar sound. In this work its complex circuit is analyzed and simulated by using state-space representations. The equations for the calculation of important parameters within the circuit are derived in detail and a mathematical description of the operational transconductance am-plifier is given. In addition the digital model is compared to the original unit.
The magnetic pickup of an electric guitar uses electromagnetic induction to convert the motion of a ferromagnetic guitar string to an electrical signal. Although the magnetic pickup is often cited as an everyday application of Faraday's law, few sources mention the distortion that the pickup generates when converting the motion of a string to an electric signal, and fewer analyze and explain this distortion. We model the magnet and ferromagnetic wire as surfaces with magnetic charge and construct an intuitive model that accurately predicts the output of a magnetic guitar pickup. This model can be understood by undergraduate students and provides an excellent learning tool due to its straightforward mathematics and intuitive algorithm. Experiments show that it predicts the change in a magnetic field due to the presence of a ferromagnetic wire with a high degree of accuracy.
The mechanical and acoustic behaviour of a plucked string coupled to a soundboard is investigated in relation to the principles underlying harpsichord design. Radiated sound energy, acoustic spectrum and sound decay time are all considered in relation to string length, wire gauge and soundboard properties, and simple scaling rules giving satisfactory musical balance are derived. These ab initio design principles are related to a typical harpsichord and found to be in general accord with building practice.The measured musical properties, which show a nearly constant sound pressure level over the whole compass, increased harmonic development in the bass, and a decay time varying closely as the inverse 0.4 power of fundamental frequency, are close to what is expected from the analysis.