Factors affecting the dynamic response of the body and the vibration transmitted through seats

Abstract

The vibration transmitted through a seat is influenced by the dynamics of the seat and the
dynamics of the occupant. The principal objective of this thesis is to understand how the dynamics
of the body and factors affecting the dynamics of the body influence the vibration transmitted
through seats. Previous studies have shown that the apparent mass of the body and seat
transmissibility are affected by the seating environment (e.g. vibration input spectra, backrest,
hands position, foot position) and variability between people (i.e. physical characteristics), but
these effects have not previously been systematically explored for realistic seating conditions.
The apparent masses of 12 subjects were measured during exposure to random vertical vibration
(from 0.125 to 40 Hz) to investigate the effects of the seat backrest, the footrest and steering
wheel, and input spectra. In a rigid seat with no backrest, there were resonances in the apparent
mass of the body around 5 and 10 Hz (with 1.0 ms-2 r.m.s broadband vibration). In the same seat
with a rigid backrest, the median resonance frequency in the apparent mass increased from 5.47 to
6.35 Hz as the backrest was reclined to 30 degrees in 5 degrees increments; with a 100-mm foam
backrest, the median resonance frequency decreased from 5.18 to 4.49 Hz as the backrest was
reclined to 30 degrees. When subjects held a steering wheel, the mass supported on the seat
surface decreased and there was an additional resonance at 4 Hz in the apparent mass. Moving
the steering wheel away from the body reduced the apparent mass at resonance and increased the
apparent mass around the 4 Hz resonance. As the feet moved forward, the mass supported on the
seat surface increased, indicating that the backrest and footrest supported a lesser proportion of
the subject weight. Applying force (0, 50, 100, 150, 200 N) to either the steering wheel or the
footrest reduced the apparent mass at resonance and decreased the mass supported on the seat
surface. Narrowband inputs at ½-octave intervals (from 1 to 16 Hz) presented at five magnitudes
(0.25, 0.4, 0.63, 1.0 and 1.6 ms-2 r.m.s.) showed that the extent of nonlinearity previously observed
with broadband vibration was frequency-dependent: the magnitude of vibration at frequencies less
than 4 Hz had the greatest effect on the apparent mass at resonance, while vibration at
frequencies less than 8 Hz had the greatest effect on the resonance frequency.
A simple lumped parameter model was used to demonstrate that changes in the apparent mass
with backrest contact, backrest inclination, hand position, foot position and vibration magnitude
could be closely represented by changing the parameters in the model. Trends in model
parameters, the damping ratios, and the damped natural frequencies were identified as a function
of the model variables.
A study was designed to determine how the physical characteristics of 80 seated adults (41 males
and 39 females aged 18 to 65) affected their apparent mass and the transmission of vibration
through a seat. Multiple regression models showed that while the strongest predictor of the vertical
apparent mass at 0.6 Hz, at resonance, and at 12 Hz was bodyweight, weight was not strongly
associated with seat transmissibility. A lumped parameter seat-person model was used to show
that the dynamic stiffness of the seat increased with increased loading so as to compensate for
increases in apparent mass associated with increased sitting weight. As age increased from 18 to
65 years, the apparent mass resonance frequency increased by up to 1.7 Hz. This change was
greater than the 0.9-Hz increase in resonance frequency between sitting without a backrest and
sitting with a backrest reclined to 15° and greater than the 1.0-Hz reduction in resonance frequency
when the magnitude of vibration increased from 0.5 to 1.5 ms-2 r.m.s. Subject age was much the
strongest predictor of the seat transmissibility resonance frequency and the transmissibility at
resonance. The model was used to show that changes in the seat transmissibility with age could be
predicted from changes in the apparent mass with age.

More information

Identifiers

ORCID for Martin G.R. Toward: orcid.org/0000-0001-7334-4355

ORCID for M.J. Griffin: orcid.org/0000-0003-0743-9502

Catalogue record

Export record