Effect of vibration exposure duration on discomfort.
University of Southampton, Institute of Sound and Vibration Research,
The comfort of a seated person exposed to vibration is known to depend on the magnitude, frequency content,
and direction of the excitation. A review of the literature showed that very little is known about the effects of the
duration of exposure to vibration on comfort.
This thesis investigates the effects of body support, frequency, waveform, and direction of excitation on the
Subjective Discomfort Time-Dependency (SDTD) during vibration so as to improve understanding of the
mechanisms involved (e.g. the biodynamic responses of the body and muscle activity) and elaborate a model
predicting how discomfort evolves with exposure duration. To achieve these objectives, a new method of
measuring the discomfort time-dependency was developed and tested.
The Subjective Discomfort Time-Dependency has been investigated in 27 experimental sessions, each with
twelve subjects seated on a conventional car seat. In each session, subjects were exposed to one stimulus.
The new developed method requires the subjects to adjust the magnitude of the vibration in order to keep
constant their discomfort. The SDTD was obtained by measuring the platform acceleration over the exposure
duration. At specific time-intervals, subjects were also asked to indicate the locations of their discomfort and
provide discomfort ratings for these locations. Results showed that the amount of vibration to achieve a
constant level of discomfort decreased over time (mainly during the first 15 minutes of exposure). This implies
that the sensitivity of vibration increases with duration. Fore-and-aft excitations generated a greater SDTD for
most stimuli. For 1-Hz lateral sinusoidal motion, the sensitivity of vibration increased at a greater rate with a
harness than without. Stimuli at 1 Hz produced SDTD that were less dependent on the duration of exposure
than stimuli at higher frequencies. The waveforms of the vibration had little effect on the SDTD. The discomfort
rating showed that prolonged exposure to vibration produced discomfort mainly at the neck.
Because discomfort was mainly felt at the neck and that the SDTD depended on the frequency, it was
hypothesised that the type of neck muscle activity produced during exposure to vibration depends on the
frequency. Neck muscle activity was measured with 12 seated subjects during 10 minutes of fore-and-aft
sinusoidal vibration. The r.m.s. magnitudes of the raw EMG and of the phasic and tonic components of the
EMG were calculated (it was assumed that phasic muscle activity arose from the periodic vibration whereas
the tonic muscle activity was needed to respond to a static load). Results showed that the frequency of
vibration had no effect on the EMG r.m.s values but affected the phasic and tonic components of the EMG.
Phasic activity was greatest at 1 Hz and decreased as the frequency increased. Tonic activity showed the
opposite tendency. As for the SDTD studies, the frequency of excitation seems to have an effect on the phasic
and tonic components of the neck muscle activity.
Phasic and tonic neck muscle activities represent different types of head motions. Because the content of
phasic and tonic activities of the EMG signal seems to be linked with the effects of vibration exposure duration
on discomfort, it was hypothesised that predicting the head motions may help estimating the comfort timedependency.
A three degree-of-freedom lumped parameter model was developed to predict floor-to-head
transmissibility. The model was then calibrated to estimate the head motions using the floor-to-head, seat-tohead,
and seat transmissibility measured with 12 subjects, exposed to fore-and-aft sinusoidal, narrow-band
random, and broad-band random vibration. Results showed that the model can estimate the head motions
around the frequencies of resonances (mode shapes), but requires improvement to be accurate at the other
frequencies. The estimated mode shapes showed three types of head motions: at 1.4 Hz the head and neck
moved in phase; at 3.5 Hz, there was a resonance of the backrest and the head and neck moved in phase, but
with a greater head motion than neck motion; and at 6.9 Hz the head and neck moved out of phase.
The subjective, physiological, and biodynamic studies suggest that the SDTD increases when the neck
muscles attempt to control head motions by producing greater tonic, and less phasic, activity. The lumped
parameter model identified through the mode shapes three types of head motions corresponding to different
comfort time-dependencies. It was hypothesized that the phase and modulus of the seat-to-head
transmissibility may indicate the amount of phasic and tonic activity produced. Through neck muscle activity, a
model predicting seat-to-head transmissibility may also predict the time-dependency of discomfort.
This thesis proposes a new method for determining the time-dependency of discomfort caused by whole-body
vibration. Discomfort time-dependencies have been shown to depend on the frequency of vibration, direction
of excitation, and body support. Mechanisms responsible for the discomfort time-dependency have been proposed.
Actions (login required)