Hand–arm vibration in cycling

Abstract

Numerous workers are exposed to vibrations which can turn out to be fatal for the health. Athletes can be included in this population, in particular cyclists who are exposed to vibration due to the irregularity of the road. This nuisance depends of the duration of exposure and the range of vibrations. While the worker is mostly directly excited by a vibrating system, the cyclist is indirectly subjected to it. He undergoes the vibrations of an excited sub-structure which is the bicycle. So the bicycle plays the role of a vibration filter or amplifier. In this paper we propose to (i) study the transmission of vibrations to the cyclist after excitation on a paving road, (ii) calculate the limit time of exposure to this type of excitation rate by the use of the standard ISO 5349 and the European directive 2002/44/EC, and (iii) compare the weighting curve of the standard with a vibrations transmissibility curve obtained between the collarbone and the stem. For this particular case of an excited sub-structure, a weighting curve is proposed by considering the first modal frequency of the bicycle.

Full text

Article
Hand–arm vibration in cycling
X Chiementin
1
, M Rigaut
1
, S Crequy
1,2
, F Bolaers
1
and
W Bertucci
2
Abstract
Numerous workers are exposed to vibrations which can turn out to be fatal for the health. Athletes can be included in
this population, in particular cyclists who are exposed to vibration due to the irregularity of the road. This nuisance
depends of the duration of exposure and the range of vibrations. While the worker is mostly directly excited by a
vibrating system, the cyclist is indirectly subjected to it. He undergoes the vibrations of an excited sub-structure which is
the bicycle. So the bicycle plays the role of a vibration filter or amplifier. In this paper we propose to (i) study the
transmission of vibrations to the cyclist after excitation on a paving road, (ii) calculate the limit time of exposure to this
type of excitation rate by the use of the standard ISO 5349 and the European directive 2002/44/EC, and (iii) compare the
weighting curve of the standard with a vibrations transmissibility curve obtained between the collarbone and the stem.
For this particular case of an excited sub-structure, a weighting curve is proposed by considering the first modal
frequency of the bicycle.
Keywords
Hand–arm vibrations, ISO5349 standard, cycling
Received: 18 November 2011; accepted: 26 July 2012
1. Introduction
Vibrations exposure of hand–arm system can be a risk
for workers’ health and safety. Numerous employees
are directly exposed to these vibrations which can be
the origin of repetitive strain injury (RSI), vascular
injuries and nervous injuries (Griffin, 1990; Leclercq,
2001). The causes and the consequences of these vibra-
tions are widely studied in the labor world.
The transmission of vibrations to the body depends
on its position, which means that the effects of vibra-
tions are complex. The exposure to vibrations provokes
movements and effects in the organism, which can
cause discomfort, negatively affect performance and
pose a risk to health and safety. This discomfort is
strongly linked to the power absorbed (Pradko et al.,
1965). In the case of a hand–arm system, there are four
types of disorders: vascular disorders, neurological dis-
orders, carpal tunnel syndrome and RSI (Bovenzi et al.,
1980; Friden, 2001; Pyykko et al., 1986; Taylor, 1988).
The standard ISO 5349-1 (EN-ISO-5349-1, 2001)
defines the procedure of measuring vibration for the
hand–arm system. As part of a hand–arm system, the
measure must be made by accelerometers on the
machine–hand interface. A frequency-weighted
acceleration is derived from this measure to give the
probability of damage. Despite this standard some stu-
dies were realized to measure the acceleration directly
on the desired joint to obtain unweighted accelerations.
Effectively, the standard does not consider other
important parameters such as the couplings forces
(Pekkarinen et al., 1989; Starck et al., 1990). The
advantage of such methods is to consider the absorp-
tion of vibrational energy according to the strength of
prehension and the size of the wrist (Aldien et al.,
2006). Some studies have reported that the use of
power consumption is a parameter more reliable than
the weighting in frequency recommended in the stand-
ard ISO 5349-1 (Burstro
¨m, 1990; Burstro
¨m and
1
GRESPI, Groupe de Recherche en Sciences Pour l’Inge
´nieur, Universite
´
de Reims Champagne-Ardenne, France
2
LISM, Laboratoire d’Inge
´nierie et Sciences des Mate
´riaux, Universite
´de
Reims Champagne-Ardenne, France
Corresponding author:
X Chiementin, GRESPI, Groupe de Recherche en Sciences Pour
l’Inge
´nieur, Universite
´de Reims Champagne-Ardenne, 51687 Reims
Cedex 2, France.
Email: xavier.chiementin@univ-reims.fr
Journal of Vibration and Control
19(16) 2551–2560
!The Author(s) 2012
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1077546312461024
jvc.sagepub.com
at BU Reims Champagne-Ardenne on August 19, 2015jvc.sagepub.comDownloaded from

Lundstro
¨m, 1998). Griffin (1990) used the laser vibrom-
eter to determine the accelerations. However, this tool
is limited to the extent in one direction and the vibra-
tion of the body cannot be subjected to a uniaxial
vibration. Studies have brought on the transmissibility
of vibration in the hand–arm system showing the res-
onances (Aatola, 1989; Cherian et al., 1996; Pyykko
et al., 1976; Reynolds and Angevine, 1977).
The sports world is also widely exposed to vibra-
tions, however the standard is not used. In particular,
cyclists are regularly exposed to vibrations which are
generated by the road profile. Akuthota et al. (2005)
highlights the risk of carpal tunnel syndrome nerve on
long cycle trips. Braithwaite (1992) proposes a pos-
ition of hands slowing down the appearance of the
pinching of the carpal nerve in cycling, and a regular
change of positions. Haloua et al. (1987) is interested
in the compression of the ulnar nerve at the racing
cyclists. Professionals can accumulate 6 hours of train-
ing on the bike per day. Some ‘traditional’ races which
borrow paved areas significantly excite cyclist runners.
Loss of feeling in the hands implies a decrease in the
grip strength and results in decreased local vascular-
ization (Farkilla and Pyykko, 1979). Carpal tunnel
syndrome can also be a result of vibration and the
degeneration of the lunate bone of the hand
(Radwin et al., 1997). Studies were interested in redu-
cing the effect of these vibrations (Greenslade and
Larsson, 1997; Griffin, 1998). What is the risk taken
by the riders in facing these vibrations? After presen-
tation of the standard for the prevention of vibrations
for the worker, this paper is interested in: (i) quantify-
ing the transfer of the vibrations emitted by the profile
of the road on the cyclist; (ii) determining the critical
time of exposure through the standard; and (iii) com-
paring the weighting curve given by the standard with
that obtained by measuring the exit of the hand–
arm system. From this comparison, a weighting
curve based on the bike modal frequency will be
proposed.
2. Theory and the ISO5349 standard
The EN ISO 5349-1 standard (EN-ISO-5349-1, 2001)
expresses the measurement procedure and methods for
assessing human exposure to hand-transmitted vibra-
tion. The measurement should be made closer to the
hand grip on the vibrating system, and translated into
an effective amplitude a
hv
, which is the square root of
the sum of the square frequency-weighted accelerations
for each axis a
hw,i
:
ahv ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
a2
hw,xþa2
hw,yþa2
hw,z
qð1Þ
The weighting of the acceleration is related to the
fact that the risk of damage is not equal to all frequen-
cies and a frequency weighting is used to represent the
probability of damage due to different frequencies. This
weighting is defined as a total frequency weighting
function H(s)
HðsÞ¼HbðsÞHwðsÞð2Þ
which is the product of a band limiting filter H
b
(s)
HbðsÞ¼ s242f2
2
ðs2þ2f1s=Q1þ42f2
1Þðs2þ2f2s=Q1þ42f2
2Þ
with s¼j2f
ð3Þ
and a frequency weighting filter H
w
(s)
HwðsÞ¼ ðsþ2f3Þ2Kf2
4
ðs2þ2f4s=Q2þ42f2
4Þf3
ð4Þ
The values f
i
refer to the resonance frequencies
(f
1
¼6.310, f
2
¼1258.9, f
3
¼15.915, f
4
¼15.915). The
values Q
i
refer to the selectivities of the poles
(Q
1
¼0.71, Q
2
¼0.64) and Kis the gain (K¼1). The
vibration exposure is quantified by a measure with the
reference time equal to 8 hours: A(8). This is
calculated based on the amplitude and the duration of
exposure
Að8Þ¼ahv ffiffiffiffiffiffiffiffiffiffiffi
T=T0
pð5Þ
where Trepresents the daily duration (h) of exposure to
vibration and T
0
is the reference duration of 8 hours.
This standard is used by the European directive
2002/44/EC (Directive, 2002/44/EC, 2002), and makes
employers responsible for ensuring that the risks of
whole-body vibration are eliminated or minimized.
The directive defines thresholds of exposure from
which the employer has to control (2.5 m/s
2
for the
hand–arm system) and thresholds exposure limits to
which workers must not be subjected (5 m/s
2
for the
hand–arm system).
3. Experimental tests
3.1. The cycle
All of the study has been realized on a racing bicycle
used for competition with a carbon frame. The charac-
teristics are listed in Table 1.
2552 Journal of Vibration and Control 19(16)
at BU Reims Champagne-Ardenne on August 19, 2015jvc.sagepub.comDownloaded from

3.2. Modal analysis
Before the tests under real conditions, a modal analysis
is realized. The bike is suspended by elastic bands to
approximate free–free boundary conditions. The sig-
nals are recorded by an acquisition system OROS
OR36 with eight inputs and two outputs. Two tri-axis
accelerometers (Bruel and Kjaer 4525 B) with a sensi-
tivity of 10 0.22 mV/g are positioned on the frame
and the stem. A shock hammer is used to reveal the
normal mode. An averaging of 15 transfer functions
obtained between the input (a striking on the hub of
the front wheel) and the output (accelerometer) is done.
We obtain three characteristic lines, 29–75–181–418 Hz
on the frequency span 0–1000 Hz (Figure 1).
3.3. In real conditions
A voluntary competitive (regional-level) cyclist realized
all of the tests on the bicycle presented in Section 3.1.
His anthropologic data are weight 67 kg and height
1.84 m. The signals are collected by an acquisition
system OROS OR36. The signals are collected with a
sampling frequency of 20,000 Hz during 5 s. This
system, of 5.2 kg weight, is put on the cyclist’s shoulder.
A battery of 12 V, 2.3 Ah feeds the system and allows it
operate autonomously for 50 min. To start the acqui-
sition a pulse push button is used. One of the terminals
of the switch is connected to an output of the analyser
and generates a continuous signal of 5 V while the other
terminal of the push button is connected to an input
device. In order to allow the pilot to have the desired
position on the bike a temporization of 1 second delays
the acquisition. Two tri-axis sensors 4525 B from Bruel
and Kjaer, sensitivity 100.2 mV/g on the three axes,
are used. The first is positioned on the stem with the
axis Xpositive in the direction of movement and the
axis Zcollinear with gravity (Figure 2(a)). The second
is positioned successively on the wrist, elbow and clav-
icle of the rider so that when the pilot’s arm is extended,
a line can be drawn through the wrist, elbow and col-
larbone, with emphasis on the bony parts of each
member. The rider realizes 10 tests for each position,
the results are averaged because the experienced vibra-
tion is not identical for each test. The sensor positions
on the wrist are slightly down compared with the bump
of the ulna, 10 mm on the Xaxis and 10 mm on the Y
axis. The attachment is made using an elastic bandage
(Figure 2(b)). These positions are chosen to obtain a
bone portion. Thus, we limit the vibration response of
the muscles.
The tests are performed on a paved street in Reims
(France). The paving stones are laid in staggered rows
and their average sizes in the direction of rolling are
12011 mm, with a gap of 253 mm. The tests are
realized at movement speed, from 5 km/h to 35 km/h,
with an increment of 5 km/h (Figure 3).
From the paving stones and the rolling speed, the
excitations frequencies f
th
, are determined by
fth ¼v
d=3:6ð6Þ
where dis the distance between the centers of two
paving stones in meters and vis the rolling speed in kilo-
meters per hour. The frequencies of engagement are
thus calculated for each of the speeds studies (Table 2).
4. Results and discussion
4.1. Distribution of the vibrations
4.1.1. Root mean square values. This section concerns the
root mean square (RMS) on the stem and on three
points of the cyclist members, wrist, elbow and collar-
bone (Figures 4 and 5).
On the frame, the more excited axes are the Zaxis
(vertical) and Xaxis (direction of movement) while the
excitations along the Yaxis are at a level almost two
times lower. These results are consistent because in the
commitment of the pavement, the bike receives a shock
in its sense of movement, but also a vertical shock due
to the height of the pavement. The RMS values for the
axes Xand Yincrease for the range 5 to 30 km/h, then
begin a decline between 30 and 35 km/h. From the
speed of 30 km/h the bike is less excited along
the axes Xand Y, which increases the comfort of the
rider because the vibrations transmitted to the rider are
lower.
On the wrist, the excitations on the Zaxis are the
most important, they evolve strongly between 15 and
20 km/h. The excitation evolution of the Xaxis fol-
lows those of the Zaxis until 20 km/h, then they
remain stable with values between 45 and 55 m/s
2
.
Table 1. Technic characteristics of the bike
Frame : Carbon monocoque 3K
Fork : Carbon The One
Transmission : Shimano Ultegra - Shimano 12 25, 10 speeds
Cranksets : Shimano Ultegra 50 34
Breaks : Shimano Ultegra
Wheels : Mavic Ksyrium Equipe
Tires : Michelin Lithion 700-23c, pressure 7 bars
Handlebar : 3T Pro
Stem : 3T arx Pro
Saddle: fiz
´i:k Pave
Weight: 8.10 kg
Chiementin et al. 2553
at BU Reims Champagne-Ardenne on August 19, 2015jvc.sagepub.comDownloaded from

The excitations on the Yaxis are in low growth but
constant.
On the elbow the excitations between the Xand Y
axes are similar until 25 km/h, from this speed the exci-
tations on the Yaxis are stable whereas those on the X
axis increase slightly. The excitations on the Zaxes
have the same evolution as the Yaxis but with values
twice as large. After 25 km/h (46 Hz) the RMS values
on the Yand Zaxes decrease softly.
On the collarbone, for the range 5 to 25 km/h the
more excited axis is the Zaxis with a peak value at 20
km/h, from this speed the excitations fall on this axis.
This is the Yaxis which became the more excited for the
range 25 to 35 km/h.
On different body parts and the stem the Zaxis
remains the most severely excited, this is explained by
the nature of the shock, the paving attack creates a
shock along the Zaxis. The Yaxis is generally the
least excited except for the collarbone. The most excited
body part is the wrist, more particularly from 20 km/h.
The hand–arm system responds like a filter because the
vibrations amplitude decreases between the wrist, the
elbow and the collarbone. One can note that the shoul-
der is excited very little because the RMS values
between 10 and 35 km/h are almost constant.
4.1.2. Transmissibility function. After having quantified the
RMS at different parts of the cyclist and the cycle, we
Figure 1. Transfer function of the bike.
Figure 2. (a) Sensor position on the stem and (b) sensor position on the wrist.
2554 Journal of Vibration and Control 19(16)
at BU Reims Champagne-Ardenne on August 19, 2015jvc.sagepub.comDownloaded from

calculate the vibrational transfer T, that is the ratio
between the RMS value computed at a point of the
rider a
hc
and the RMS value calculated at the stem a
hw,i
:
HmeasuredðwÞ¼ ahc
ahw,i
ð7Þ
The transmissibility functions on the three axes
between the wrist and the stem show that the acceler-
ations on the Yand Zaxes are amplified (Figure 6).
The transmissibility ratios are higher than 1 from 20
km/h for the Xand Yaxes, and from 25 km/h for the
Zaxes. The maximum transmissibility ratio is reached
at a speed of 30 km/h (55.5 Hz). One can also note a
stabilization of the amplification from 30 km/h. The
accelerations on the elbow are amplified for the Y
axis from 20 km/h. A ratio of 1.38 is reached for 25
km/h (46 Hz). The accelerations on the Xaxis are lar-
gely reduced (from 90% for a speed of 5 km/h to 40%
for a speed higher than 20 km/h). The accelerations on
the Yaxis are reduced until a speed of 20 km/h and are
then stable. The transmissibility on the three axes stay
steady from 25 km/h. The collarbone reduces largely
the vibration in the three axes. The transmissibility
does not exceed 0.4 on the Xand Zaxes. The transmis-
sibility is higher on the Yaxes but stays below to 1. In a
global way the amplification of the vibration occurs so
from 20 km/h, a 38 Hz frequency of engagement. Note
that this frequency of 38 Hz corresponds to the
resonant frequency of the wrist (Chiementin et al.,
2011). For speeds lower than 20 km/h the accelerations
are widely reduced.
4.2. Exposure threshold
4.2.1. Time-scale and spectral analysis. As mentioned in
ISO 5349-2 (EN-ISO-5349-2, 2001), we record here
the spectral behavior of the study. The excitation fre-
quency of the hand–arm system is determined in order
to understand the comparison between the norm and
our results. The signals collected are largely unsteady
due to the varying distance between two centers of suc-
cessive blocks and the unstabilized velocities of the
bike. That is why an analysis time scale is realized.
A continuous wavelet transform of the Morlet mother
wavelet, denoted by 
a,b
(t), is used (Mallat, 2008). This
transformation defines, at every moment, the scale (fre-
quency) of the analyzed signal s(t)
Ca,b¼ZsðtÞa,b
tb
a

dt where a,bðtÞ¼ex2=2cosð5tÞ
ð8Þ
The study was conducted on the scale 500–2000 corres-
ponding to a frequency range of 20–81 Hz
f¼fc=ð2aÞð9Þ
Figure 3. (a) Tests on a paved street in Reims and (b) data acquisition systems OROS.
Table 2. Characteristic values
Speed (km/h) f
th
f
exp
Error A(8) (m/s
2
)T
limit
(min)
5 9.3 – – 3.30 138
10 18.5 – – 7.14 29
15 27.8 – – 9.46 17
20 37.0 35 1.45 13.03 9
25 46.3 45 1.42 12.9 9
30 55.6 53 2.39 14.14 7.5
35 64.8 65 0.15 14.11 7.5
Chiementin et al. 2555
at BU Reims Champagne-Ardenne on August 19, 2015jvc.sagepub.comDownloaded from

Figure 4. Values of a
hv,i
for (a) stem and (b) wrist.
Figure 5. Values of a
hv,i
for (a) elbow and (b) collarbone.
2556 Journal of Vibration and Control 19(16)
at BU Reims Champagne-Ardenne on August 19, 2015jvc.sagepub.comDownloaded from

where f
c
is the central frequency, ais the scale and brep-
resents the time. The analysis of Figure 7 for a speed of
20 km/h confirms the appearance of nonstationary
measurements, however, we can distinguish windows
of the time where the signal is stationary. These time
slots allow us to determine the experimental frequencies
of commitment of the bike and are listed in Table 2.
The comparison of theoretical and experimental fre-
quencies shows, that there is a good fit for speeds
greater than or equal to 20 km/h. Under that speed,
the frequency of 25 Hz is predominant corresponding
to the first modal frequency of the bike (see
Section 3.2).
4.2.2. Measurement of A(8). This section evaluates
the severity of cycling activity on cobblestones
through the variable A(8) given by ISO 5349-1 (EN-
ISO-5349-1, 2001), in which the thresholds are fixed
by the European directive 2002/44/EC (Directive,
2002).
For each speed, the effective frequency-weighted
acceleration is computed. The values A(8) are estimated
for a duration of 1 hour and are listed in Table 2. The
threshold of surveillance fixed by the European direct-
ive is 2.5 m/s
2
. This value is exceeded from the first
speed of engagement. The threshold of risk, fixed to
5 m/s
2
, is exceeded between the speeds of 5 and
10 km/h. We thus note that the cyclist is even widely
excited for low speeds.
From the threshold of risk reduction of 5 m/s
2
, the
time exposure limit, T
limit
, is calculated. For the speed
range of tests, the maximum exposure time is 137 min-
utes at 5 km/h and 7 min for a speed of 35 km/h. The
calculated values of A(8) are listed in Table 2. For
information, many cycle races offer portions of pave-
ment. We can name the most famous of the ‘classics’
Paris–Roubaix, at the speed of 35 km/h, the time of
7 min corresponds to a distance of 4 km, that is the
first four paved sectors while the race presents 27
sectors.
Figure 6. Transfer rate of vibration.
Chiementin et al. 2557
at BU Reims Champagne-Ardenne on August 19, 2015jvc.sagepub.comDownloaded from

4.3. Norm comparison/transmissibility
This section compares the weighting curve provided by
the standard and the curve of transmissibility between
the collarbone and the stem. Figure 8 shows that we
obtain a good correlation for a speed of 25 km/h and
higher. However, under this speed the errors are
widely significant, we obtain an average relative error
of 28.7% and a maximal relative error of 87.1%
between the weighting curve of the standard and our
measures. This speed limit corresponds to the first
resonance frequency of the cycle. Thus, in the case
of a sub-structure excited (the cycle), we propose a
frequency weighting H
w,proposed
(s) which has the same
mathematical formulation as Equation (4), but whose
characteristics are: f
3
¼29.0, f
4
¼29.0, Q
2
¼0.64,
K¼0.54. Here f
3
,f
4
correspond to the first resonant
frequency of the cycle, Kis the maximum weighting
factor experimental and Q
2
stays identical as on the
standard. These remarks consider the standard and the
dynamic of the sub-structure excited (Figure 8). So the
mean relative error is 7.0% and a maximum relative
Figure 7. Absolute value of C
a,b
for a speed of 20 km/h.
Figure 8. Transmissibility function from the ISO 5349-1 standard, from experimental values and proposed model.
2558 Journal of Vibration and Control 19(16)
at BU Reims Champagne-Ardenne on August 19, 2015jvc.sagepub.comDownloaded from

error of 13.6% between the proposed weighting curve
and our measures.
5. Conclusion
This paper is concerned with the propagation of vibra-
tions in the hand–arm system of the cyclist, especially
when discussing the ISO5349-1 standard. These vibra-
tions generated by the road profile can be harmful to
the health of the athlete, and mastery can be transpose
into performance.
Atfirst,itisshownthatforaunidirectional excitation
on the pavement, the vibrations felt by the cyclist are
expressed in all three axes, and especially along the
normal axis of movement. It is also shown that the vibra-
tions are amplified from a speed of 20 km/h that is 38 Hz,
which corresponds to the resonance of the wrist. The radial
axis of movement is the most amplified. This excitation is
similar to a wavering and oscillation type deformation.
Second, this paper applies the ISO 5349-1 standard
and the European directive to estimate the dose of
vibrations, A(8), and the time of maximal exposure.
This time, relatively weak, is estimated at 7 min for a
35 km/h speed.
Finally, this paper shows that the standard over-
states the dose of vibration before the first modal fre-
quency of the sub-structure. Thus, for a sub-structure
excited a frequency weighting filter based on the first
modal frequency of the cycle is proposed.
Further tests are planned as part of an agreement with
a secondary school (Lycee Arago) in Reims. This school
educates young top-level sportsmen and sportswomen.
Moreover to quantify the effects of the stoutness, subjects
of different levels will be added to the study.
This study precedes a model of the hand–arm system
to prevent risk of exposure to vibrations. The design of
a cycle also appears suitable for reducing the dose to
meet the standard of health.
Funding
This research received no specific grant from any funding
agency in the public, commercial, or not-for-profit sectors.
References
Aatola S (1989) Transmission of vibration to the wrist and
comparison of frequency response function estimators.
Journal of Sound and Vibration 131: 497–507.
Akuthota V, Plastaras C, Tobey KLJ, Press J and Garvan C
(2005) The effect of long-distance bicycling on ulnar and
median nerves: An electrophysiologic evaluation of cyclist
palsy. American Journal of Sports Medicine 33: 1224–1230.
Aldien Y, Marcotte P, Rakheja S and Boileau P-E (2006)
Influence of hand forces and handle size on power absorp-
tion of the human hand–arm exposed toZ–Haxis vibra-
tion. Journal of Sound and Vibration 290: 1015–1039.
Bovenzi M, Petronio L and DiMarino F (1980)
Epidemiological survey of shipyard workers exposed to
hand–arm vibration. International Archives of
Occupational and Environmental Health 46: 251–266.
Braithwaite I (1992) Bilateral median nerve palsy in a cyclist.
British Journal of Sports Medicine 26: 27–28.
Burstro
¨m L (1990) Absorption of Vibration Energy in the
Human Hand and Arm. Ph.D. thesis. Sweden: Lulea
University of Technology.
Burstro
¨m L and Lundstro
¨m R (1998) Absorption of vibration
energy in the human hand–arm exposed to sinusoidal
vibration. International Archives of Occupational and
Environmental Health 61: 213–216.
Cherian T, Rakheja S and Bhat R (1996) Ananalytical inves-
tigation of an energy flow divider to attenuate hand-trans-
mitted vibration. International Journal of Industrial
Ergonomics 17: 455–467.
Chiementin X, Bertucci W, Crequy S, Bolaers F and Estocq P
(2011) Vibration of the upper limbs in cycling. Computer
Methods in Biomechanics and Biomedical Engineering 14:
19–20.
Directive 2002/44/EC (2002) Directive of the european parlia-
ment and of the council of 25 June 2002 on the minimum
health and safety requirements regarding the exposure of
workers to the risks arising from physical agents (vibra-
tion), Sixteenth individual Directive within the meaning of
Article 16(1) of Directive 89/391/EEC.
EN-ISO-5349-1 (2001) Mechanical vibration, Measurement
and evaluation of human exposure to hand-transmitted
vibration, Part 1: General requirements.
EN-ISO-5349-2 (2001) Mechanical vibration, Measurement
and evaluation of human exposure to hand-transmitted
vibration, Part 2: Practical guidance for measurement at
the workplace.
Farkilla M and Pyykko I (1979) Blood flow in the contralat-
eral hand during vibration and hand grip contractions of
lumberjacks. Scandinavian Journal Work Environment and
Health 5: 368–374.
Friden J (2001) Vibration damage to the hand: clinical pres-
entation, prognosis and length and severity of vibration
required. Journal of Hand Surgery 26: 471–474.
Greenslade E and Larsson T (1997) Reducing vibration
exposure from hand-held grinding,sanding and polishing
powertools by improvement in equipment and industrial
processes. Safety Science 25: 143–152.
Griffin M (1990) Handbook of Human Vibration. London:
Academic Press.
Griffin M (1998) Evaluating the effectiveness of gloves in
reducing hazards of hand-transmitted vibration.
Occupational and Environment Medicine 5: 340–348.
Haloua J, Collin J and Coudeyre L (1987) Paralysis of the ulnar
nerve in cyclists. Annales de Chirurgie de la Main 6: 282–287.
Leclercq C (2001) La main traumatique du sportif. Paris:
Masson Sport.
Mallat S (2008) A Wavelet Tour of Signal Processing, 3rd edn.
New York: Academic Press.
Pekkarinen J, Starck J and Pyyko I (1989) High-speed digital
method to measure impulsive hand–arm vibration. In:
Proceedings of the 3rd International Symposium of the
ISSA. Vienne.
Chiementin et al. 2559
at BU Reims Champagne-Ardenne on August 19, 2015jvc.sagepub.comDownloaded from

Pradko F, Lee R and Greene J (1965) Human Vibration-
response Theory. Paper No. 65-WA/HUF-19. American
Society of Mechanical Engineering.
Pyykko I, Farkkila M, Toivanen J, Korhonen O and
Hyvarinen J (1976) Transmission of vibration in the
hand–arm system with special reference to changes in com-
pression and acceleration. Scandinavian Journal of Work,
Environment and Health 2: 87–95.
Pyykko I, Korhonen O, Farkkila M, Starck J, Aatola S and
Jantti V (1986) Vibration syndrome among finnish forest
workers, a follow-up from 1972 to 1983. Scandinavian
Journal of Work, Environment and Health 12: 307–312.
Radwin R, Armstrong T and Chaffin D (1997) Power hand
tool vibration effects on grip exertions. Ergonomics 30:
833–855.
Reynolds D and Angevine E (1977) Hand-arm vibration. Part
II: vibration transmission characteristics of the hand–arm.
Journal of Sound and Vibration 51: 255–265.
Starck J, Pekkarinen J and Chun L (1990) Transmission of
vibration from tool handle to the wrist and to the head.
The Kurume Medical journal 37: 1–11.
Taylor W (1988) Biological effects of the hand–arm vibration
syndrome: historical perspectives and current research.
Journal of Acoustics Society of America 83: 415–422.
2560 Journal of Vibration and Control 19(16)
at BU Reims Champagne-Ardenne on August 19, 2015jvc.sagepub.comDownloaded from