Research Team:
Dr. André R. Barbosa
Civil & Construction Engineering, OSU
Arijit Sinha
Wood Science & Engineering, OSU
Christopher Higgins
Rajendra Soti
This testing project focuses on structural testing to generate data that can be used to support the performance-based design of the framework project to be built in Portland, Oregon, namely in support of facilitating the design of rocking CLT walls. Specifically, this phase of testing would be developed using panels manufactured by DR Johnson. The testing being addressed in this scope of work includes CLT crushing tests and CLT wall panel tests
Two sets of experiments were conducted to evaluate the structural performance of CLT wall panels The first set of experiments included in-plane shear tests of two CLT wall panel specimens designated as “wall panel 1” and “wall panel 2”. The second set of experiments included compression testing of five cross-laminated panels tested in compression with a buckling restraint to characterize the post-peak behavior of the panel. The CLT panels were manufactured by DR Johnson Lumber Co., Oregon, using Douglas-fir Larch species group lumber. The panels for the two sets of experiments were fabricated to the required finished dimensions prior to shipping to Oregon State University’s Structural Engineering Research Library and Wood Structural Engineering Laboratory, respectively. The test set-up and instrumentation plan was peer-reviewed and approved prior to commencement of testing as the testing performed was a subset of testing performed for the Framework Project.
Cyclic, quasi-static, loading in accordance with the abbreviated basic loading history in CUREE Publication No. W-02 was performed, except that trailing, cycles were taken equal to the primary cycle (rather than 75% of it). The CUREE “delta” of 1.0 in. was set and maintained constant for all specimens for comparison purposes. Axial load was maintained constant at 135 kips. Following the CUREE protocol, the specimens were subjected to cyclic loading with increasing displacement amplitudes at a constant displacement rate of 0.01 in/sec until the target (CUREE “delta”) displacement was achieved, at which point the displacement rate was increased to 0.025in/sec until the end of the test. Prior to testing, the moisture content was measured and recorded. The average measured moisture content for wall panel 1at the time of the test was 14.1% while the recorded average moisture content for wall panel 2 was 14.6%.
A monotonic load was applied to the CLT panel using a steel plate between the head of the CLT panel and the actuator. The test was controlled by a constant rate of displacement of the hydraulic actuator. The rate of displacement of the actuator was 0.05” per minute until 20% of the post-peak load was achieved. Then the rate was increased to 0.15” per minute until 80% post-peak load. As received the average moisture content of the panels was 8% as measured by a pin-type moisture meter. One surface of the panels was painted black and speckled with white paint for strain measurements using digital image correlation techniques.
Four locations are reported for the east face of the panel, and two locations for the west face of the panel. It can be seen that the tangent stiffness across the instrumental panel zones is similar across all locations, except the north panel zones which show higher stiffness compared to other panel zones. The higher stiffness in north panel zones could be attributed to the presence of relatively strong substrate laminations in the north end of this wall panel. Some discontinuities are visible in the results. These correspond to instances when the instrumentation stopped recording useful data, either due to loss of support of the instrumentation or due to the local failure of the substrate laminations in shear at the specific locations where the sensors were installed.
A summary of results for the crushing tests is presented in Table 1. The mean peak load for the CLT panels was 473 kips with a coefficient of variation (cov) of 4.9%. The bulk modulus of elasticity (MOE) estimated over 48″ was 1.29 million psi. This calculation does not exclude the perpendicular laminations and hence it is reported as bulk MOE. The cover for bulk modulus is 3.1%. Table 1 also presents the load at which the first onset of yielding is observed, the onset of delamination of the lamellas occurred. Corresponding displacement values are also reported in Table 1.
These 13 observations were made after a visual assessment of pictures and videos recorded during testing, DIC analysis, and corroborating those with testing notes. Additionally, visual estimates were checked against DIC data. The DIC tends to provide a lower bound for the onset of failure stresses and the corresponding strain than human observation. The definition of the onset of failure that DIC uses is user-dependent and if altered will provide a different result. Since there are no certain criteria for the onset of failure, visual estimates are presented and relied upon. Results used to develop data in Table 1 can be provided in an electronic format upon request and approval from DR Johnson Lumber due to the large size of the files.
