1/D
t
(mm/MPa)
MO
E (Pa
)
OSL
LVL
Figure 3.14. Simulated MOE versus 1/D
t
for LVL and OSL at 10% compaction
63
3.6.4 Numerical Modeling of OSB with Surface VTC Strands
To study the effect of VTC strands on the mechanical properties of OSB panels,
the surface layers were replaced by VTC strands. The VTC strands were placed on both
surface layers using three differences cases. The first case had 20% by weight of VTC
strands placed on the top and bottom surface layers (10% each surface). The second case
had 40% by weight of VTC strands placed on the top and bottom surface layers. In the
third case, instead of weight percent, there was 50% by volume of VTC strands for the
two face layers (about 75% by weight of VTC). The first two cases mimic an
experimental study to Rathi (2009). The third case, which is discussed first, replaces all
surface strands with VTC strands to compare to the previous results on commercial OSB.
3.6.4.1 50% VTC by Volume
In this case, 50% VTC strands by volume were placed on the two face layers
(25% on each face). The VTC strands in this case had the same thickness and length as
the normal strands. Compared to the case in Figure 3.10, only the face (surface) layer
strands were changed to VTC strands. The core remained as unmodified strands. Figure
3.15 gives the MPM calculations for axial modulus for these OSB panels as a function of
mat compaction and glue-line stiffness. Like OSB with unmodified strands, the modulus
decreased as the interfacial compliance (1/
D
t
) increased. The MOE values are much
higher compared to the results without VTC (see Figure 3.10) because the surface strands
provide most of the stiffness and these strands were replaced by VTC strands. A panel
with 1% coverage would have an axial modulus about 25% lower than one with 100%
coverage. The result shows the influence of interfacial stiffness was higher when using
VTC strands. Comparing the case without VTC (Figure 3.10) to this case (Figure 3.15),
there is at least 93% increase in modulus at the same levels of compaction. In Figure
3.15, all the data in this plot are obtained from average of at least five runs. The error bars
on one curve show the standard deviation of the results. There was little change in
standard deviation for different values of 1/D
t
.
64
5400
6400
7400
8400
9400
10400
11400
0.00
0.01
0.02
0.03
0.04
0.05
Do'stlaringiz bilan baham: |