- Torsion test was evaluated as a method to determine the shear strength
parallel to grain of full-size structural composite lumber (SCL). Experimental
results of eighty-five rectangular (44 x 140 mm²) specimens were verified by finite
element analysis. Laminated strand lumber (LSL), laminated veneer lumber
(LVL) and parallel strand lumber (PSL) were tested using torsion. For
comparison, ASTM shear block tests were performed in joist (LR-plane) as well
as in plank (LT-plane) orientation with a shear plane of 2600 mm².
In torsion, a different failure behavior was observed for the three wood
composites. LSL failed abruptly, in a brittle manner, on the center of the short
side along the strands (LT-plane). LVL and PSL, on the contrary, failed gradually,
in a more ductile manner, mostly on the center of the long side across the
strands or veneers (LR-plane). LVL and especially PSL displayed a distinct
decrease in torsional rigidity with increasing torque, while LSL showed a more
linear torque-twist relationship. LSL was able to bear a much higher torque than
LVL and PSL. Checks, like lathe checks (rotary peeling), internal checks (drying), or not properly glued veneer or strand interfaces, as well as void areas, possibly
decrease the shear resistance of LVL and PSL.
Due to the different failure behavior of the wood composites, orthotropic
theory was applied to evaluate shear stresses in torsion. The ratio of shear
moduli in the two longitudinal planes (GLT and GLR) had a significant influence on
the orthotropic shear stresses. The following mean shear stresses were defined
as shear strength in torsion, depending on the failure plane: LSL in plank
orientation tLR = 6.43 MPa; LVL in joist orientation tLT = 7.96 MPa; PSL in joist
orientation tLT = 6.82 MPa. The shear strength of LSL in joist orientation could
not be determined, since the shear plane coincided for all specimens with the LTplane.
However, it must definitely be higher than the concomitant shear stress of
tLT = 12.7 MPa.
Shear strength of SCL was found to be much higher in joist orientation,
where failure occurs across the veneers or strands, compared to plank
orientation ('tLT > tRL) for both test methods, torsion and ASTM block. ASTM
shear blocks of joist oriented LSL failed rather in compression parallel to grain
(crushing), or a combined failure mode of crushing and shear, than in pure shear.
In joist orientation, LSL has much higher shear strength (tTL) than LVL and PSL,
while in plank orientation all three composites show similar shear strengths (tRL).
In torsion, LVL showed a slightly higher shear strength than PSL, whereas the
ASTM block test indicated the reverse. Shear strength of SCL based on torsion is
much lower than shear strength based on shear block (tT/tASTM < 1), as opposed
to what has been observed earlier for solid-sawn lumber (SSL).
Shear strength based on tests of small ASTM shear blocks is not
representative of the shear strength of full-size specimens. Torsion test is
recommended as standard test method to determine the pure shear strength of
full-size SCL, since it is the only test method that imposes a state of pure shear
stress on the specimen. This will complement the way other stresses are
determined, i.e., pure tensile, compressive and bending strength.