Test Bench for Analysis of Harmful Vibrations Induced

to Wheelchair Users

A. Ababou, N. Ababou, T. Morsi and L. Boukhechem

Laboratory of Instrumentation, University of Science and Technology Houari Boumediene,

BP32 El Alia BabEzzouar 16111 Algiers, Algeria

Keywords: Harmful Vibrations, Whole-Body Vibration, Wheelchair, MEMS, Accelerometers, Signal Processing.

Abstract: In this paper, a test bench for analysis of harmful vibrations that can be transmitted by manual wheelchair ti

its user is presented. The vibration generating device developed in the laboratory and sensors positioning

are described. Vibration measurements were carried out using four tri-axis MEMS accelerometers and a

position sensor. Experimental data were noisy, so they were first filtered before acceleration amplitude can

be assessed using two methods that we propose: a so-called ‘cyclogram method’ and ‘DFT principal peak

signal magnitude’ on the other hand. The second one is faster and results showed that the prototype

developed in the laboratory can provide harmful vibrations in 2-10Hz frequency range and can be used to

check the vibration transmissibility of wheelchair.

1 INTRODUCTION

According to the World Health Organization more

than one billion people in the world live with some

form of disability, of whom nearly 200 million

experience considerable difficulties in functioning.

In the years ahead, disability will be an even greater

concern because its prevalence is on the rise (World

Health Org. and World Bank, 2011) The results of

studies conducted by Van Sickle et al. and Wolf et

al. (Van Sickle et al., 2001; Wolf et al., 2005) have

shown that a wheelchair user is subjected to

vibrations that often exceed the limits specified by

the International Standard Organization ISO

2631-

1standard. Directive 2002/44/EC of the European

Parliament and of the Council dated June 25 2002

lays down minimum requirements for the protection

of workers from the risks arising from vibrations.

Manufacturers of machines and employers should

make an adjustment regarding risks related to

exposure to whole-body vibration. The directive

lays down the exposure action to 0.5m/s² and the

exposure limit value to 1.15m/s² (European

Commission Directive, 2002). These limits have

been set to indicate the level of vibration that results

in productivity loss of workers and it seems to be

analogous to wheelchair users by reducing their

activity levels (Pearlman et al., 2013). Whole-body

vibrations (WBV), acting via the buttocks, the back

and the feet of a sitting person may cause chronic

spinal cord injuries (Johanning, 2011). Rather than

measuring vibration exposure on a road, Maeda et

al. (Maeda et al., 2003) used a vibrating table to

vibrate on subjects sitting in a wheelchair. Subjects

were subjected to vertical vibration inducing more

discomfort, and reported the neck as being the place

where localized pain resulting from vibrations.

Other studies have been focused on how the

selection of the cushion and backrest can affect

WBV absorption by the body and examined the

influence of different surfaces on the sidewalk

vibration exposure of wheelchair users (Garcia-

Mendez et al., 2012; Qiu and Griffin; 2012; Wolf et

al., 2007).

Frequencies associated with the effects of WBV

on health, activities and comforts vary between

0.5Hz and 100Hz (ISO, 1997). According to ISO

2631-1 standard on human vibration, individuals in

a seated position are at risk of injury due to whole-

body vibrations when exposed for long periods of

time. Exposure to vibrations in the range of 4-10Hz

frequency may cause pain in the chest and abdomen.

Back pain occurs frequently with vibrations in the

frequency range of 8-12Hz, and vibration in the

frequency in 10-20Hz range can cause headache,

eyestrain, and irritation in the intestines (Kittusamy

and Buchholz, 2004; Bovenzi, 2010; Johanning,

147

Ababou A., Ababou N., Morsi T. and Boukhechem L..

Test Bench for Analysis of Harmful Vibrations Induced to Wheelchair Users .

DOI: 10.5220/0004806001470153

In Proceedings of the International Conference on Biomedical Electronics and Devices (BIODEVICES-2014), pages 147-153

ISBN: 978-989-758-013-0

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

2011).

Human exposure to vibration may be measured

using accelerometers. However, weighting filters are

required to correlate the physical vibration

measurements to the human’s response to vibration.

ISO 2631 standard describes suitable weighting

filters, but does not explain how to implement them

for digitally recorded acceleration data. Rimell and

Mansfield described in their paper the

implementation of the weighting filters used in three

different standards as digital IIR (Infinite Impulse

Response) filters and provided all the necessary

formulae to directly calculate the filter coefficients

for any sampling frequency (Rimell and Mansfield,

2007).

In this paper a test bench developed in our

laboratory is presented. It can produce harmful

vibrations that can be transmitted by a wheelchair to

its user on one hand, and can give quantitative

assessment of vibration magnitude on the other

hand. Vertical as well as rocking strokes can be

generated at frequencies in 2-10Hz range. As a first

step, the characterization of the vibrations generated

by the device is investigated.

2 BASIC VIBRATIONS

When a manual wheelchair user moves outdoor, one

can consider that he can be submitted to

multidirectional vibrations that can be a composition

of three basic vibrations, i.e. vertical (along z-axis),

lateral (along x-axis) and fore-and-aft (along y-axis)

vibrations as shown schematically in Fig.1. The

vertical vibration noted V.V. is generally induced by

the quality of the tires and the road conditions

(Pearlman et al., 2013). The lateral vibration noted

L.V. corresponding to a roll, may be due to veiled

wheels, to the passage of a single wheel above a

small drop, or uneven road surfaces (Cooper et al.,

2011). The fore-and-aft vibration noted F.V.

corresponding to a pitch, may be caused by

corrugation of the road, by crossing small steps, a

door sill or by certain sort of floor tiles or for all-

terrain wheelchairs riding (Burton et al., 2010;

Rispin and Wee, 2013).

Vertical, lateral and fore-and-aft vibrations in

this work are generated by a device that can be

configured so that only vertical vibration or rocking

vibration is selected. The magnitude of vibration

can be quantified either by its amplitude (mm), its

velocity (m/s) or its acceleration (m/s²).

Figure 1: Basic vibrations that can be induced on a manual

wheelchair: vertical vibration V.V, lateral vibration L.V

and fore-and-aft vibration F.V.

3 EXPERIMENTAL

3.1 Experimental Set-up

The test bench schematic shown in Fig. 2 consists of

vibrating table which comprises three blocks: a base

anchored to the ground and supporting four coil

springs, a movable plate having guide tubes and a

vibration generator controlled by a variable-speed

drive.

Figure 2: Schematic of the vibrating table.

The base consists of a framework based on four feet

on the ground and secured with 4 coil springs.

Theses springs allow the upper part of the test bench

to vibrate freely along the three axes so as to

produce vertical as well as lateral and fore-and-aft

movements. The vibration plate is composed of a

rectangular frame surface with ties for supporting a

motor with its guiding device, two cogwheels, two

pinions and a tensioner pulley. Each cogwheel shaft

sustains an eccentric weight which generates

BIODEVICES2014-InternationalConferenceonBiomedicalElectronicsandDevices

148

mechanical vibration at each rotation.

Figure 3 shows the details of the transmission of

the movement using a toothed belt (CC). In the case

(a), the motor M drives the two wheels R1 and R2 in

the same direction, the belt tension is provided by

the tensioner T. Resulting movement of eccentric

weights B and B' causes rocking motion left to right

and right to left. In the case (b), the pinion P is used

to invert the rotation of the toothed wheel R2

relative to that of the wheel R1. The resulting

movement of the eccentric weights B and B'

generates a vertical vibration.

Rotational speed of the induction motor M was

controlled by a Siemens Micromaster variable-speed

drive (VSD).

Rotational speed of the motor evolution versus

the VSD excitation frequency has been investigated

using a DT-2269 digital stroboscope from Digital

Instruments and a linear behaviour of rotational

speed has been observed.

Figure 3: Patterns of the transmission system of the

movement - (a) generation of rocking vibration due to the

resulting force-couple : left rocking (90°) and right

rocking (270°) - (b) generation of vertical vibration

resulting from centrifugal force action up (270°) and down

(90°) on the two eccentric weights. Positions of B and B’

when rotated by 90°, 180°, 270° and 360°.

3.2 Sensors Positioning

For vibration amplitude measurements four

MMA7260Q tri-axis MEMS accelerometers with

analog outputs have been used. A linear

potentiometric position sensor has also been placed

under the vibrating table to measure the amplitude

(in mm) of the generated mechanical vibrations.

Accelerometers static and dynamical calibrations

were performed before measurements have been

carried out. Experimental data of the position sensor

calibration curve have been adjusted by a line whose

slope was equal to 1.93V/mm with a correlation

coefficient equal to 0.99997.

The sensors were arranged as shown in Fig. 4.

The four accelerometers were placed in a horizontal

plane so that the x-axis pointed to the left of the

subject, the y axis and the z-axis pointed to forward

and downward respectively. Their sensitivity was set

to 200mV/g and they were also equipped with

analog anti-aliasing filters with cutoff frequency

equal to 40Hz.

Figure 4: Sensors positioning on the vibrating table, the

manual wheelchair and the user: (1) position sensor; (2)

Acc-T; (3) Acc-F; (4) Acc-W; (5) Acc-H.

Acc-T, Acc-F, Acc-W and Acc-H are respectively

the accelerometers placed on the vibrating table,

footrest, frame of the manual wheelchair and on the

head (vertex) of seated subject. The accelerometer

Acc-T placed on the table gives the same signal as

the vibration provided by the vibration generator.

Accelerometers Acc-W and Acc-F respectively

placed on the wheelchair frame and footrest are used

to compare the signal supplied by the vibration

generator and the vibration transmitted to two

extreme points of the wheelchair in order to quantify

the vibration magnitude transmitted by the

wheelchair. The last accelerometer placed on the

vertex of the subject sitting in the wheelchair

provide information on the quality of vibration

transmitted to the point of the subject is seated

furthest both the wheelchair system for generating

vibration. The level of vibration transmitted to the

subject's head can then be assessed.

TestBenchforAnalysisofHarmfulVibrationsInducedtoWheelchairUsers

149

3.3 Acquisition and Data Processing

For acquisition and processing the analog data ssued

from the four tri-axis accelerometers and position

sensor, a 12-bit acquisition DAQCard-6062E from

National Instruments has been used and data

sampled at 1kHz. Measurements have been carried

out for duration equal to 10s on a healthy male

subject 1.75m tall and 75kg weight seated in a

manual wheelchair for eight different values of

vibration frequency: 2, 3.2, 4.4, 5.6, 8, 9, 10 and

10.6Hz. The selected vibration duration can be

considered as sufficiently short to avoid health

problems for the subject (ISO, 1997). Fig. 5 shows

an example of signal acceleration output along z-

axis at a vibration frequency equal to 4.4Hz

provided by the accelerometer Acc-W located on

wheelchair frame after analog filtering. For more

convenience, raw acceleration data are displayed

only for duration equal to 5s. The signal remains

noisy after analog anti-aliasing filtering.

Figure 5: Example of z-axis signal issued from

accelerometer (Acc-W) located on wheelchair frame.

The DFT of the vibration signal expressed in m/s²

and depicted in Fig. 5 is presented in Fig.6. One can

notice the presence of one principal peak at 4.4Hz

and several harmonics at higher frequencies. The

first principal peak has been observed for all the

eight vibration frequencies applied to the

wheelchair.

Because of lack in the DFT response of other

peaks with appreciable magnitude compared to the

first one, we did not consider in this work the

weighting filters usually used in literature to

correlate the physical vibration measurements to the

human’s response to vibration and described in

ISO2631 standard. A digital band-pass filter has

been applied to the vibration signal of Fig.5.

Selected low and high cutoff frequencies for the

filtered signal presented in Fig.7 have been set to

= 3Hz and f

= 7Hz respectively.

Figure 6: DFT response in semi-log plot of the raw

acceleration signal presented in Fig. 4.

Figure 7: Acceleration signal depicted in Figure 5 after

band-pass filtering (fcl= 3Hz ; fch= 7Hz).

4 RESULTS AND DISCUSSION

In order to assess quantitatively the vibrations

magnitude along x, y and z axis measured by all

accelerometric sensors, the frequency of the

variable-speed driver has been varied from 10Hz to

50Hz. These values correspond to vibrations

delivered by the test bench to the wheelchair-user

system with a frequency varying in 2-10.6Hz range.

The graphs shown in Fig. 8 depict for a column

from top to bottom an example of measured then

filtered 10.6Hz vibration magnitude at the vibrating

table (T), the wheelchair frame (WF), footrest (F)

and the user head (H) along one axis(x, y or z).

When considering a raw, from left to right, the

measured accelerations correspond respectively to

vertical acceleration along z-axis, fore-and-aft

acceleration along y-axis and lateral acceleration

along x-axis at one of the four selected locations.

BIODEVICES2014-InternationalConferenceonBiomedicalElectronicsandDevices

150

Figure 8: Example of vibration signals measured by the

four accelerometers. In this example, vibration frequency

f=10.6Hz.

Thus, for a given vibration frequency generated by

the test bench, a comparison between different

points of measure along the same axis (x, y or z) can

be achieved by considering one column (among

three columns) and between different accelerations

patterns at one location (vibrating table, wheelchair

frame, wheelchair footrest or user head) is obtained

by considering one raw (among four raws). One can

note on these graphs that acceleration amplitude

varies with both time and location. To assess the

amplitude of vibration measured by the

accelerometers for different values of vibration

frequency, we have represented cyclograms

corresponding to the evolution of the acceleration

value (in one direction and for one accelerometric

sensor) versus the position of vibrating table plate of

the test bench. The cyclogram associated with

vertical wheelchair acceleration at f=4.4Hz (See

Fig.5 and Fig.7) is presented in Fig. 9.

Figure 9: Example of cyclogram corresponding to

acceleration value versus table displacement.

From the different cyclograms obtained from

experimental measures, acceleration amplitudes

along z-axis (vertical vibration), y-axis (fore and aft

vibration) and x-axis (lateral vibration) have been

quantitatively assessed for the eight values of

vibration frequency delivered by the test bench at

the four selected locations.

Furthermore, for each temporal acceleration

signal, we considered the principal peak magnitude

of the DFT signal. Fig.10 shows the variation of the

acceleration amplitude obtained from cyclograms as

a function of the DFT magnitude, expressed in

arbitrary units due to scaling factor. It is worth

noticing that a correlation between these two

variables seems to be present. Experimental data

were fitted by a straight line with a slope found to be

equal to 4.83.10

-3

. Hence, acceleration amplitude

can easily be deduced from DFT principal peak

height with a relative uncertainty less than 20%.

According to ISO2631-1 and European

Commission Directive (European Commission

Directive, 2002) the vibration should not exceed the

weighted acceleration a

=0.8m/s² in the case of

long-term exposure and a daily exposure time of 8h.

Furthermore, ISO2631-1 recommends weighting the

frequencies of the measured vibration according to

the possible deleterious effect associated with each

frequency.

Frequency weightings are required for three

orthogonal directions (x -, y – and z -axes) at the

interfaces between the body and the vibration.

TestBenchforAnalysisofHarmfulVibrationsInducedtoWheelchairUsers

151

Figure 10: variation of acceleration amplitude obtained

from cyclograms versus DFT signal magnitude.

The weighted acceleration in the case of whole-body

vibration is expressed (Bovenzi, 2005) by the

relation (1)

)a()a4.1()a4.1(a 

(1)

Once the values of a

, a

and a

has been obtained

from cyclograms, the weighted acceleration has

been calculated for the vibrating table, the

wheelchair frame and footrest, and the user head for

the eight different values of frequency vibration.

Deducing acceleration amplitudes directly from

DFT vibration magnitude for a given frequency is

another method which is simpler than considering

the cyclogram method.

We used the DFT magnitude method and

equation (1) to obtain weighted accelerations for

different vibration frequencies. The results presented

in Fig. 11 are associated with the test bench in

rocking vibration configuration while those depicted

in Fig.12 are associated with the test bench in

vertical vibration configuration. Weighted

acceleration values are expressed in m/s².

For vibration frequencies lower than 5Hz, the

obtained weighted acceleration values are

sufficiently low to consider the vibration not

harmful when the test bench is in rocking vibration

configuration. When vibration frequency exceeds

5Hz, the weighted acceleration value associated with

the wheelchair increases drastically, reaching almost

10m/s² for f > 8Hz. In the second case, except for f

= 2Hz, all other values of weighted acceleration

show that the test bench generates harmful

vibrations for vibration frequencies effects.These

frequencies are principally inducing back pain and

backbone disorders. Hence, the device developed in

the laboratory can be used to check the quality of the

wheelchairs that are marketed in developing

countries from the vibration transmissibility point of

view.

Figure 11: weighted acceleration versus vibration

frequency with test bench rocking vibration (r.v)

configuration. F: wheelchair footrest; W:wheelchair

frame; T: vibrating table; H: user head.

Figure 12: weighted acceleration versus vibration

frequency with test bench in vertical vibration (v.v)

configuration. F: wheelchair footrest; W:wheelchair

frame; T: vibrating table; H: user head.

5 CONCLUSIONS

A test bench for analysis of harmful vibrations that

can be potentially induced to a manual wheelchair

user has been developed in the laboratory.

Vibrations generated by the device were measured

using MEMS accelerometers and a position sensor.

The experimental signals were noisy and were first

filtered before being processed. Two methods have

been used to assess vibration magnitude, the

cyclogram method and DFT principal peak

magnitude method. The obtained weighted

acceleration values showed that the device

developed in the laboratory provides harmful

BIODEVICES2014-InternationalConferenceonBiomedicalElectronicsandDevices

152

vibrations and can be used to check the vibration

transmissibility of manual wheelchair.

In further work, this test bench will be used to

investigate the vibration effect and the wheelchair

design (rigid frame, foldable, wheel camber) as well

as tires and cushion damping effects on the harmful

vibration magnitude.

ACKNOWLEDGEMENTS

This work was supported in part by MESRS

ministry under grant J0200220100018. The authors

wish to thank Mr H. Zerouali for his participation in

the prototype developing.

REFERENCES

Bovenzi, M., 2005, ‘Health effects of mechanical

vibration’, G. Ital. Med. Lav. Erg., vol.27, n°1, pp.58-

64.

Bovenzi, M., 2010, ‘A longitudinal study of low back pain

and daily vibration exposure in professional drivers’,

Industrial Health, vol.48, pp.584-595.

Burton, M., Fuss, F. K., and Subic A., 2010, ‘Sports

wheelchair technologies’. Sports Technology vol.3,

n°3, pp. 154–167.

Cooper, R. A., Teodorski, E. E., Sporner M. L., and

Collins, D. M., 2011, ‘Manual wheelchair propulsion

over cross-sloped surfaces: a literature review’.

Assistive Technology, vol.23, pp.42–51.

European Commission Directive ECD 2002/44/EC of the

European Parliament and of the Council, 2002. ‘on

the minimum health and safety requirements regarding

exposure of workers to the risks arising from physical

agents (vibration) ’, Official Journal of the European

Communities vol. L177, pp.13-19.

Garcia-Mendez, Y., Pearlman, J. L., Cooper, R. A.,

Boninger, M. L., 2012, ‘Dynamic stiffness and

transmissibility of commercially available wheelchair

cushions using a laboratory test method’, Journal of

Rehabilitation Research & Development, vol.49, n°1,

pp. 7-22.

International Organization for Standardization, 1997,

‘Mechanical vibration and shock Evaluation of human

exposure to whole-body vibration Part 1: General

requirements (ISO 2631-1)’.

Johanning, E., 2011, ‘Diagnosis of whole-body vibration

related health problems in occupational medicine’,

Journal of Low Frequency Noise, Vibration and

Active Control, vol.30, n°3, pp.207-220.

Kittusamy, N. K. and Buchholz, B., 2004, ‘Whole-body

vibration and postural stress among operators of

construction equipment: a literature review’, Journal

of Safety Research, vol.35, n°3, pp.255-261.

Maeda, S., Futatsuka, M., Yonesaki, J., Ikeda, M., 2003,

‘Relationship between questionnaire qurvey results of

vibration complaints of wheelchair users and vibration

transmissibility of manual wheelchair’, Environmental

Health and Preventative Medicine, vol.8, pp. 82-89.

Pearlman, J., Cooper, R., Duvall, J., Livingston, R., 2013,

‘Pedestrian pathway characteristics and their

implications on wheelchair users’, Assistive

Technology, vol.25, pp.230-239.

Rimell, A. N. and Mansfield, N. J., 2007, ‘Design of

digital filters for frequency weightings required for

risk assessments of workers exposed to vibration’,

Industrial Health, vol.45, n°4, pp.512–519.

Rispin, K., and Wee, J., 2013, ‘A paired outcomes study

comparing two pediatric wheelchairs for low resource

settings; the Regency pediatric wheelchair and a

similarly sized wheelchair made in Kenya’. Assistive

Technology, DOI: 10.1080/10400435. 2013.83784

Qiu, Y., Griffin, M. J., 2012, ‘Biodynamic response of the

seated human body to single-axis and dual-axis

vibration: effect of backrest and non-linearity’,

Industrial Health, vol.50, n°1, pp.37–51.

VanSickle, D. P., Cooper, R. A., Boninger, M. L.,

DiGiovine, C. P., 2001. ‘Analysis of vibrations

induced during wheelchair propulsion’, Journal of

Rehabilitation Research & Development, vol.38, n°4,

pp. 409-421.

Wolf, E. J., Pearlman, J., Cooper, R. A., Fitzgerald, S. G.,

Kelleher, A., Collins, D. M., Boninger, M. L., Cooper,

R., 2005, ‘Vibration exposure of individuals using

wheelchairs over sidewalk surfaces’. Disability &

Rehabilitation, vol.27, n°23, pp. 1443-1449.

Wolf, E., Cooper, R. A., Pearlman, J., Fitzgerald, S. G.,

Kelleher, A., 2007, ‘Longitudinal assessment of

vibrations during manual and power wheelchair

driving over select sidewalk surfaces’, Journal of

Rehabilitation Research & Development, vol.44, n°4,

pp. 573-580.

World Health Organization, World Bank, 2011, World

report on disability. Available from

http://whqlibdoc.who.int/hq/2011/WHO_NMH_VIP_

11.01_eng.pdf.

TestBenchforAnalysisofHarmfulVibrationsInducedtoWheelchairUsers

153