Seismic Joint Testing Program Completion
by Paul A. Bradford, P.E., PhD, Development Engineer, Watson Bowman Acme – BASF
Seismic Joint Testing In January of 2009 Watson Bowman Acme completed an 11 year program of seismic joint testing (Figs. 16 and 17) at the University of California at Berkeley, Richmond Fieldhouse Station. The project included three phases: proof of concept, value engineering, and performance limits. The joint tested, the XCel multi-movement modular expansion joint, is designed to respond elastically in all six degrees of freedom at high velocities. Its primary use is on long span structures where it is critical to accommodate vehicular traffic immediately after an extreme event. It has been implemented in other applications were its multi-movement capabilities are needed for everyday service, e.g. structures with severe skews.
The proof of concept stage, started in 1998, included identifying a system that would remain undamaged during an earthquake. Initial testing showed that standard systems modified with additional movement capacities incurred significant damage, rendering the joint inoperable. Gaps were so large that traffic would not have been able to pass over the joint. The joint was retrofitted with a special equidistance mechanism, along with bearings capable of high speed multi-rotational movement, and after several design iterations the joint worked well.
The value engineering phase included simplification and standardization of the system, with emphasis placed on using standard materials and near standard components. An outgrowth of this phase was the XR bearing, a wear resistant polyurethane bearing that has since been used to retrofit problematic installations with bearing rotation/wear issues.
The final phase of testing included pushing the joint to the maximum capacity of the test equipment, exceeding parameters of all expansion joints tested to date. In all, hundreds of tests were run and thousands of man-hours were spent during the joint’s evolution to produce the XCel design.
Much was learned during the course of the project, including;
• Joints respond much differently at high speeds than at slow speeds.
• Friction forces are predictable, and friction components can be implemented such that the expansion joint can act as force limiter to the adjoining structure segments.
• Testing components is necessary but not sufficient to establish system viability.
• Test failure modes sometimes came from unexpected sources.
• Consideration should be given to real world conditions, i.e. less than ideal sliding surfaces, installation misalignments, etc.
• Fabrication tolerances, alignments, and quality control measures are more important for high speed applications than for slow speed applications.
• Spring based equidistance systems used alongside sliding friction components do not perform well due to their slow response time.
• PTFE bearings, when designed correctly, are exceptionally durable. Metal on metal surfaces tended to bind and “screech” at high speeds.
• If it is critical that the joint perform correctly on a project with a high level of reliability, a section of the project joint should be tested. This is because at high speeds small variations in fabrication processes, materials, etc. can sometimes lead to large changes in system performance.
The test program showed that it is possible to modify standard modular joints to accommodate large displacements at high speeds. Peak velocities approached 50 inches per second, reaching the limits of the hydraulic pump system. Peak total displacements were 34 and 20 inches in the longitudinal and transverse directions respectively. Test excitations included harmonic motions at various displacements and frequencies, as well as seismic simulations of real time earthquakes.
Segmental bridges tend to be longer span structures, often assigned an Importance Category of “Critical”. They are typically designed for extreme events such that the structure responds with minimal damage, and must allow traffic to pass. The XCel provides a means to accomplish this design objective at the expansion joint locations.