LOCAL VIBRATIONS-MECHANICAL IMPEDANCE OF THE

HUMAN HAND’S GLABROUS SKIN

RONNIE LUKDSTR&
Department of Emironmental Medicine. University of Umea. S-901 87 Umea. Sweden

,\bstmct-The mcuhanical point impedance has been studied in ten diRrent areas of the glabrous skin oithe
human hand on three male and three female subjects within the frequency range of 20-10000 Hz.
For all tested areas the impedance decreased with increasing frequency down to a minimum talue.
corresponding to rhe natural frequency of the skin. After that. the mechanical impedance was dIrectI
proportional to the frequency. The highest natural frequency, about 200 Hz, was measured in the distal arm
of the finger and the lowest, about 80 Hz, in the proximal areas of the palm (thrnar). Small differences in
internal damping were also showed to exist.
A great amount of handheld tools used in industry have their maximum vibrational levels within the
natural frequency range of the skin.
In order to avoid adverse effects the skin’s mechanical properties should therefore carefully be taken into
consideration al designing vibrating tools.

ISTRODCCTIO> vibration (Franke. 1951; Reynolds and Soedel. 1972;

Taylor, 1974; Hixon, 1976; Mishoe and Suggs. 1977;
In order to understand or even predict how man is Olesen and Randall, 1979). The lower the impedance,
influenced by local vibrations, basic knowledge is the easier it will be to make the whole system vibrate.
needed as regards a number of different physical and Most mechanical systems have a frequency or some
biological factors. Man‘s response to vibrafioH is frequencies where the impedance is particularly low.
decided. among other things, by its frequency. in- At these so called resonance or natural frequencies,
tensity. durability, direction. contact point and the very little energy is usually needed to maintain the
pressing force of the operator himself when using vibration. The amplitude of the vibration can even be
the tool. The correlation of the characteristics of the greater inside the sys!em than the source itself.
vibration source and the generated injuries has also Man’s hand-arm system has also proved to have
proved to be a very complex problem. Furthermore, resonance frequencies. These are individual to a large
the problem has been accentuated by large biological extent and depend on several Factors. like vibration
variations mainly as regards vibration sensibility of the direction, squeeze force around a handle. muscle
individual. (Verillo. 1970: Lidstriim. 1976; Lundstrijm tension, arm’s position and others. Working with tools
and Landstram, 198 1). whose operation frequency is equal to that of the
From a physical point of view the hand and the arm hand-arm system naturally increases the risk of harm-
can be regarded as a mechanical system, consisting ofa ful influence. Such situations would be avoided if
large number of more or less homogeneous separate man’s mechanical impedance would be taken into
masses of various size, nature and design, e.g. different consideration more than it is at the present.
bone structures. muscle groups and sensory bodies, The purpose of this investigation has been to
held together by different kinds of springy and viscous describe mechanical properties of the inner side of the
tissues. Since man’s hand-arm system consists of a hand in a few different situations. The skin is mostly
great number of such separate elements plus the fact the direct contact point between the hand and the
that there is a great variation as to the individual, it is vibrating surface. This kind of knowledge is thus of
naturally extremely difficult to construct a practical great importance to increase the knowledge of how the
model. Such studiescan be found in literature, though, hand responds to vibration and for the biological
where more or less sophisticated hand-arm models effects that have proved to appear, for instance ‘white
have been constructed. Despite the fact that these fingers’ and decreased tactile sensibility of the fingers.
models are simplified, a comparatively good idea can Furthermore the results can be of great importance at
be got of how the hand-arm system behaves when introducing vibration isolating materials or for other
exposed to vibrations (Reynolds and Soedel, 1972; damping activities.
Mishoe and Suggs. 1977; Bystriim P[ al.. 1982).
When describing vibration response of mechanical
DEFlSlTlOS OF IIECHASICAL POlST I\!PEDASCE
systems. the mechanical impedance serves as a good
measure of how apt the system is to follow an applied The mechanical impedance, z, can be described as a
mechanical structure’s resistance to being caused to
vibrate according to an applied vibration. The dynamic

137
138 RONNE LUNDSTR~W
force. K applied to the structure. divided by the
obtained velocity, I;, defines the mechanical impedance.
z = 7jjr;.
If force and velocity have been measured at the same
point, it is called ‘point impedance’ and describes the
resistant and absorbing properties of the structure.
The force and velocity vectors, F and r, can be
expressed mathematically by the following equations
7 = f expj(wl +#I
r = c expiWr
where cut = 2nft and 4 equals the phase angle between
force and velocity.
If the equations (2) and (3) are inserted in the
definition of impedance (I), the following is obtained:
The absolute size of the mechanical impedance is
thus dependant on not only the amplitudes of F and c
but also on their reciprocal phase relations. When F
and c are in phase with each other, the impedance
reaches its maximum value. Furthermore the impe-
dance can be divided into two components-one real
(Z,,,) and one imaginary (Z,,). Theis size can be
determined exactly when the phase relationships are
known.
A mechanical system, almost however complicated it
may be, can be decomposed into a more or less
numerous number of separate elements, consisting of
masses, springs and viscous dampers and constitute
together what is usually called a mass-spring system.
Figure 1 shows how the separate impedances are
variable with frequency and how they are generally
represented in symbols.
SPRING DAMPER MASS
iIy; [ 2,== ( I74
FREQUENCY
Fig. I. The figure illustrates the separate elements into which
the mechanical impedance can be decomposed: springs,
dampers and masses.The upper half of the figure shows the
separate elements’ symbols in mechanical models, and the
lower one indicates how the specific impedances are
frequency conditioned.
METHODS
Apparatus
Sine-wave signals for stimulation were delivered by
an increasing frequency from 20-10000 Hz from a
right angle towards a chosen part of the skin of the
hand’s inner side by means ofa mini vibrator, driven by
a power amplifier and a signal generator. A view of the
(1) mechanical stimulator set up is shown in Fig. 2. On the
vibrator an impedance head (Bruel & Kjaer 8001) was
mounted, by means of which could be registered the
dynamic force and at the same time the acceleration
that the skin was exposed to for the moment.
The signal generator also had a feed-back network
by means of which the vibration level could be held
(2) constant, independent of test frequency and static load.
(3) It is necessary that the generator’s feed-back network
gets continuous information of instantaneous vib-
ration amplitude out at the vibrator. The acceleration
signals from the impedance head were therefore con-
nected, after being amplified, to the generator’s feed-
back network. During the experimental series the
acceleration’s amplitude was kept on a constant level.
The amplified force signal, on the contrary, was
connected directly to a level recorder. The phase
relation of force and acceleration was registered at the
same time with a phasemeter and an oscilloscope. The
phasemeter was connected immediately after the first
two amplifiers for force and acceleration. These were
identical with similar adjustment and a similar phase
.shift of the signals consequently appeared.
The vibrator was set up in accordance with the
principle of a beam balance and it has therefore been
possible to superimpose all the test stimulations at a
constant static pressure both within and between each
test sequence. Two quite realistic loads, 0.5 and 1 N,
have been used in the experiments. The pressure
against the skin, caused by the static and dynamic load,
is also dependent on the size of the test surface. Two
cylinder-shaped test probes. made of stiff PVC plastic,
with plane circular contact surfaces of an area of 0.5
and I cm2 each have been used.
The vibration amplitude’s effect on skin impedance
has also been investigated by using different acceler-
ation levels, especially 120 and l-10 dB, (relative
1 pm s - 2, rms).
Experimental procedure
The subjects of the experiment, three men and three
women, ages 30-40yr, were asked to lie down on a
bunk and rest their right arms and hands on a specially
constructed test table. After that the position ofthe test
device was adjusted so that the surface of the chosen
test probe covered one of the ten test points defined in
Fig. 3. Having achieved the right position, the subjects
were asked to lie still and relaxed before and during the
frequency sweep. Each test period took about 4 min to
carry out. Another test point was then adjusted and a
new sweep was started. When all the ten test points
were gone over, one of the experimental variables were
changed.
Twenty-eight frequencies in the range of
20-10000 Hz were chosen at the treatment for further
analysis. Directly from the chart of the level recorder,
both force and phase angle values were transferred to a
 _. .

Fig. 2. View showing the mechanical skin stimulation set up. For information concerning associated
electronics, see text.

139
 Local vibrations-mechanical impedance of the human hand’s glabrous skin
Fig. 3. The skin’s mechanical impedance has been tested on
ten different points on the right hand’s inside. The iest points
have got a numerical code of I-IO in the text according 10 the
figure.
computer which on this basis and by means of the
constant acceleration amplitude determined the mech-
anical impedance of the skin. In these calculations was
also a correction for the additional force produced by
the impedance head itself, due to the mass below the
force gauge.
RESULTS
How the skin impedance had been influenced by the
different experimental conditions was studied by com-
paring measurement results from various variable
combinations. An increase of the acceleration leads to
an increase of force of the corresponding size, provided
that the mechanical properties of the skin are linear
within the amplitude range in question. According to
the definition, the size of the impedance will not be
influenced. Only minor impedance variations, caused
by the amplitudes 120 and 140dB could be noticed
though, with an exception for the very lowest frequen-
cies (< 100 Hz). Figure 4 shows the relation between
-5-1 ,
I I 1
120 130 140 150
ACCELERATIONLEVEL. dB
Fig. 4. The relation.between skin impedance and stimulation
amplitude at the fr qucncies of SO( l ). 100 ( n ). 200 ( A ) and
400 (*) Hz. The resuI t refers to test point 8 for one of the male
subjects, Pro& area: 1 cm’ and static load: 1 N. For more
information see text.
I41
impedance and stimulation amplitude for one individ-
ual at the frequencies of 50, 100, ZOOand 400 Hz of test
point 8. The two highest frequencies, 200 and 400 Hz,
have given good linearity within the studied amplitude
range of 120-150dB. It can furthermore be noticed
that the impedance decreases with the amplitude for
the other two frequencies, especially at 50 Hz, which
indicates that the impedance in this case is not linear.
The larger contact surface proved to cause a some-
what higher force component. of about l-3 dB, which
really means that the impedance value had risen
correspondingly, whereas the position of the reson-
ance frequency did not seem to be much influenced.
A doubling of the static load caused a slight increase
of the impedance’s force component. On the contrary,
there was an increase of the resonance frequency of
about 1040 Hz, probably because of the fact that the
skin was precompressed to a somewhat larger extent.
Furthermore, no significant differencies between the
men and women studied could be stated to exist.
In the light of the comparatively small variations
that have been obtained. depending on the different
experimental variables, Fig. 5 shows the skin
impedance’s magnitude and phase of each test point as
an average value in which the results of all the variable
combinations are included. As a measure of how the
impedance has varied, standard deviations around the
average values have also been drawn.
It is true of all the test points that at the start the
impedance decreases with the frequency down to a
minimum level, after which it increases. The frequency
at the minimum impedance defines the resonance
frequency of the skin,/,. At frequencies lower thanj,,
the skin behaves mechanically like a spring element. At
frequencies higher thanf,, on the contrary, the skin
loads the impedance head like a mass element. The
magnitude of the impedance at f, describes the skin’s
viscous and damping properties.
On the basis of the average graphs in Fig. 5 it can be
stated that skin shows some differences as regards
mechanical properties, depending on what test point
has been studied. It is first of all the positions of the
resonance frequency of the different parts of the skin
that have varied. The highest resonance frequencies,
about 200 Hz, have been recorded at the two distal test
points of the finger (1 and 2). and the lowest, about
80 Hz, for test point 10 in thenar. Low resonance
frequencies, about 100 and 1IO Hz respectively, have
been measured on hypothenar (point 8) and in the
central part of the palm (point 9). The skin’s properties
at the points 4,5,6 and 7 on the distal part of the palm
proved to be identical on the whole. The average
resonance frequency for these points varied within the
range of 160-185 Hz.
The phase angle graphs for all the test points looked
fairly similar (Fig. 5). Exactly at and close to/,, force
and velocity were in phase on the whole. Abovef, the
phase angle increased continuously with the frequency
up to about 90” (+ 15’). Usually the phase angles were
very close to 90” over 1 kHz. Correspondingly, the
 RONSIE LUNDSTR~M

FREQUENCY,Hz

Fig. 5. The average value for skin’s mechanical impedance and phase relationship between the force
transmitted to the skin, f, and the resulting velocity, r, for all the ten test points. The average values (the
central lines/groups) have been obtained after 30 measurements per analysed frequency. The standard
deviations around the average values are indicated with a lower and an upper graph. The impedance is
denoted in dB rel. I Nsm- ‘. For more information see text.

phase angle grew negative at decreasing frequency in system can be described mathematically, and starting
relation tof,. The standard deviations of the average off from the impedance graphs of Fig. 5, it has been
values as regards both impedance and phase graphs possible to estimate the skin’s mechanical constants
showed that there was little scattering apart from the which is shown in Table 1.
area around the resonance frequencies. z = T/a
A mechanical model, constructed on the basis of z= z,+z,+z,
obtained impedance and phase angle data shows that z = c +j(wm -k/o)
human skin can roughly be regarded as a mechanical z = [CZ+(wm-k/o#]t’*.
parallel system, consisting of a spring, a damper and a
mass. Figure 6 illustrates how the impedance of such a There proved to be little differences as to the viscous
 Local vibrations-mechanical impedance of the human hand’s glabrous skin I43

F The springy component is due to the fact that the

tissues are compressed and extended respectively
within their elastic range without consuming vibration
energy. At wide amplitudes though. the elastic range of
m the tissue can be exceeded, with more or less large
lesions as a consequence. At the tissue’s resonance
frequency of course this risk is at its largest. as there
can even be an amplification of the vibration
amplitude.
SKIN
The absorption ofvibration energy takes place in the
Fig. 6. Mechanical model of the skin. with appertaining
viscous parts of the skin tissues. The mechanical energy
mathematical formulae. constructed on the basis of impe- is there transformed into heat by means of inner
dance and phase angle data gained here. friction. The damping coefficient of the different test
points proved to be roughly the same, which implies
that the skin’s energy absorbing properties are about
Table I. Approximate values for the skin’s mechanical the same as well.
material constants, calculated from the impedance graphs in Skin’s ability to absorb energy per unit of time, P.,_.
Fig. 5. on the basis of formulae given in Fig. 6.
can be estimated from the impedance and phase angle
k c
data shown above. It is able to show that the absorp-
Testpoint (IrN m-l) (Ns m-‘1 tion is proportional to velocity and force transmitted
to the skin and that they are interrelated as regards
I 5.6 ?.I 3.0 phase relation in accordance with the following for-
2 4.2 I.0 2.4
3 2.3
mula (Johnson, 1975; Lidstram, 1976).
I.0 2.2
4 3.6 2.1 2.5
Pebr= ~Fcos(4,.,)(Nms-0.
5 3.1 2.2 2.3
6 3.2 2.8 2.5 If velocity and force are in phase with each other,
1 2.8 2.2 2.2
there is maximal energy absorption, whereas it is at its
8 I.5 2.4 2.9
9 2.0 2.4 2.2 4.0
lowest point at a phase shift of k 90’. On the basis of
IO 1.5 1.9 2.8 5.0 Fig. 5 it can be stated that the largest amount of energy
is absorbed at an exposure to vibration within the
frequency range of W-200 Hz. This conclusion can be
drawn from the phase angle graphs presented. At high
properties of the skin between the various test points or low frequencies, the phase angle coefficient.
with the exception of points 2 and 3. whose viscosity COS(&~,~),approaches to zero. Very little energy, thus.
coefficient was about half as large. Nor did there prove will be absorbed by the skin at these frequencies. The
to be any marked difference between the testpoints, skin impedance varies both as regards magnitude and
looking upon the skin as a mass substance. The shape phase in different parts of the hand, which implies, as a
of the impedance graphs of test points 9 and 10 consequence, that the energy absorption will vary as
indicates though, that more than one mass substance is well. On the basis of data of this kind, the flow of
involved. In the table therefore, two values are given. energy into the hand under various vibration circum-
The lower one refers to the mass in question at the stances can be estimated.
highest frequencies. A somewhat heavier mass sub- Comparing the mechanical properties that the skin
stance is reflected around the resonance frequency. proves to have had to the perception thresholds earlier
Large variations, on the contrary, would be es- found by others, there prove to be evident similarities.
tablished for the springy properties of the different The perception threshold for vibrations is frequency
parts of the skin. The stiffness, k, was greatest at the dependent and is typically U-formed (Verillo, 1970;
two distal points of the finger (1 and 2). whereas the test Lundstriim and Landstriim, 1981; Johansson et al.,
points 8 and 10 in the palm had the lowest. 1982; Johannsson and Vallbo. 1982). If perception is
expressed in terms of vibration velocity man’s most
DISCUSSION sensitive frequency range is between 100-300 Hz, i.e.
within the range where the skin’s mechanical reson-
The outcome of this investigation shows that it is ance frequencies are to be found.
comparatively eas# to make the skin vibrate within the A large part of the handheld tools used today have
frequencies of SCrXIO Hz. This depends partly on what an operation frequency within or close to the
part of the skin is being studied. At the very resonance frequency range of 100-200 Hz. Naturally enough this
frequency and near it, the energy contents of the is unfortunate, considering the mechanical properties
vibration need not be very great to cause a strong of the skin. As is shown earlier, the skin is fairly easily
mechanical load on skin and underlying tissues. The caused to vibrate with large mechanical loads as a
load consists above all of two components, a springy consequence, while the energy absorption is con-
one and a viscously damping one. siderable. Depending on the low perception threshold,
l-l-! RONNIE LUNDSTR~M
furthermore. a large number of sense-organs are
activated, whose task is to inform the central nervous
system of the mechanical state of the skin. Tactile
information from the hand’s skin forms the basis of
our ability to identify things and assess surface struc-
tures, and our experience of pressure, movement and
touch. If the tactile information flow is disturbed, there
appears a deterioration of the hand’s accuracy of
motion, with increased risk of accidents as a con-
sequence. (Verillo, 1970; Bjerker et al., 1972).
‘White fingers’ is the symptom of an injury most
often associated within the field of working environ-
ment with vibrations. This injury manifests itself in
attacks ofcontractions of the peripheral blood vessels,
particularly in the fingers (Taylor, 1974; Lidstriim,
1976; Lundstriim and Landstriim 1981). The reason
for this special symptom has not yet been made clear
but is considered to be due to functional disturbances
of peripheral nerves and blood vessels. The operating
frequency of handheld tools commonly used, mainly
coincide with the skin’s resonance frequencies.
Therefore, it cannot be excluded that symptoms of
vibration injuries, such as ‘vibration induced white
fingers’, may be explained by the great amount of
vibration energy absorbed during resonance exitation.
An existing connection between the operating
frequency of the handheld tools and the occurrence of
‘white fingers’ has been pointed out earlier, among
others by Hyvlrinen et al. (1973). In their investigation
it was established that a machine’s operating frequency
of about 125 Hz resulted in a higher frequency of
injures. At the designation of new machines and tools,
it should therefore be a matter of utmost necessity to
search for technical solutions where this critical
frequency range could be avoided as far as possible.
According to the latest proposal from the
International Standard Organization (ISO/DIS 5349,
1982). measurements and hygienic estimations are to
be executed starting off from a ‘broad-banded
frequency weighted value’. The weighting shall be done
within the range of 6.3-1250 Hz with a filter, whose
characteristics, expressed in terms of vibration vel-
ocity, decrease at the start by 12 dB octave-’ from 6.3
to 16 Hz. From 16 to 1250 Hz the velocity signals must
not be influenced by the filter. So the hand-arm system
is considered to be equally sensible within this latter
frequency range. The skin, being the direct contact
surface between hand and tool has, according to this
investigation, resonance frequencies that are in this
frequency range. Before setting up future standards,
the mechanical properties of the hand-arm system
should therefore be taken into careful consideration.
rfckno~lecfgement-I wish to thank Professor Axe1 Ahlmark
and Dr. Ludwik Liszka for valuable comments on the
manuscript. The financial support by the Swedish Work
Environmental Fund (Project no. 781156) and the technical
assistance by Asta Lindmark are greatly acknowledged.
REFERENCES
Bystr6m. B.-O., Nilsson. A. and Olsson, E. (1982)
Development of artificial hands for use in chain saw
vibration measurement. J. Sound I’ibrul. 82, I I I-I 17.
Bjerker. N.. Kylin, B. and Lidstriim. L-M. (1972) Changes in
vibratory threshold after exposure to powerful vibration.
Ergonomics. 15, 399-406.
Franke. E. K. (195 1) Mechanical impedance of the surface of
the human body. J. appl. Physiol. 3, 582-90.
Hixon, E. L. (1976) Mechanical impedance. Shock and
Vibration Hand-book (Edited by Harris. C. M. and Crede.
C. E.), 2nd ed. McGraw-Hill, New York.
Hyvgrinen. J.. Pyykkii, I. and Sundberg. S. (I 973) Vibration
frequencies and amplitudes in the aetiology of traumatic
vasospastic disease. Lancet 14, 79 I-794.
ISO/DIS 5349 (1982) Guide for the measurement and the
assessmentof human exposure to vibration transmitted to
the hand. Revised Gent September 1982.
Johansson. R. S.. Landstriim, U. and Lundstriim. R. (1982)
Responses of mechanoreceptive aff’erent units in the
glabrous skin of the human hand to sinusoidal skin
displacements. Brain Res. 244, 17-25.
Johansson, R. S. and Vallbo, A. B. (1983) Tactile sensory
coding in the glabrous skin of the human hand. Trends in
Neuroscience (in press).
Johnson, 1. (1975) A method for the measurement of energy
absorption in the human hand using various types of hand
tools. Aeronautical Research Institute of Sweden, FF.4
Memo 97, l-20.
Lidstriim, L-M. (1976) Vibration injury among rock drillers.
chiselers and grinders: some views on the relationship
between the quantity of energy absorbed and the risk of
injury. Vibration and Work: Proceedings of the Finnish-
Soviet-Scandinavian Vibration symposium, Helsinki. IO-I 3
March, 1975. (Edited by Korhonen, O.)pp. 81-88. Institute
of Occupational Health, Helsinki.
LundstrBm, R. and Landstriim, U. (1981) Biological effects
of vibrations, Swedish National Board of Occupational
Safety and Health, Education report Vol. 81, No. 5. pp. I -
74.
Mishoe. J. W. and Suggs, C. W. (1977) Hand-arm
vibration-part II. vibrational responses of the human
hand. J. Sound Vibrat. 53, 545-558.
Olesen, H. P. and Randall, R. B. (1979) A guide to
mechanical impedance and structural response techiques.
Briiel & Kjaer Application Note 17-l 79.
Reynolds, D. D. and Soedel, W. (1972) Dynamic responseof
the hand-arm system to a sinusoidal input. J. Sound
Vibrar. 21, 339-353.
Reynolds, D. D. and Keith, R. H. (1977) Hand-arm
vibration. part I: analytic model of the vibration response
characteristics of the hand. 1. Sound Vibrar. 51. 237-253.
Reynolds, D. D. and Angevine, E. N (1977) Hand-arm
vibration. part II: vibration transmission characteristics of
the hand and arm. J. Sound Vibrar. 51, 255-265.
Suggs, C. W. and Abrams, Jr. C. F. (1971) Mechanical
impedance techniques for evaluating the dynamic charac-
teristics of biological materials. J. agric. Engng Res. 16,
307-315.
Taylor. W. (Ed.) (1974) The Vibration Syndrome. Academic
Press, London, New York.
Verillo, R. T. (1970) Subjective magnitude functions for
vibrotaction. IEEE Trans. Man-Machine Sysr. MMS-11,
19-24.