Biomedical Engineering Reference
In-Depth Information
Fig. 4.14
a Loading device for creep characterization of elastomeric foams, b plate compression
test
Fig. 4.15 Shaded contour plots of direct stress of a the plate creep test model at time 10,800 s,
b and c the indenter creep test model with a cylindrical punch (section cut view) at time 10,800 s
(b HR-foam model and c VE-foam model); arrows indicate loading direction
compression test, multi-objective optimization was performed, taking both creep
tests into account (see Sect. 3.4.2 ) .
Creep-Data: Creep testing was chosen along with stress relaxation tests since
the body support material was loaded with a constant body weight, due to gravity,
during body-support interaction (cf. outlines regarding creep behaviour in Sect.
3.2.2.2 ).
In the parameter optimization process, material data was fitted using a Prony
series expansion, see ( 3.330 ) and ( 3.332 ). Since the accuracy of this approach is
sensitive to the number of decades of time data (the information in the first second
of recording is as valuable as the information from 1 to 10 s), emphasis was laid on
proper data collection early in the experiments. Possible non-linear viscoelastic
effects of the examined foam materials were handled by capturing the time-
dependent foam material behavior in the operating range, corresponding to the
load case of body weight buttock loading on a soft foam device. Specifically,
dimensions of the test specimen were chosen, such that indenter contact stress
approximately corresponded to the stress to be expected in a buttock-support
interaction simulation. Sample height and contact area, as well as indenter weight,
were chosen accordingly.
Within the first minute indenter displacement values were recorded with a
sample rate of 50 Hz. During the remaining experimental creep time of 10,800 s
the sample rate was 20 Hz.
Search WWH ::




Custom Search