EFT_Stage3_Details

The test structure consisted of a rigid diaphragm (a rectangular steel frame filled with reinforced concrete) supported on four replaceable steel columns at its corners.  The shake table study was conducted at the University of Illinois at Urbana-Champaign.  The concrete mass weighed 10.1 kips to fit the table capacity, and the column spacing was 60´72 inches to fit the hole-pattern of the base plate of the table.  Four plates with tapped holes were welded on the steel frame of the diaphragm to provide connections for the columns.

In the shake table study, the columns were bolted to a ½-inch thick base plate, which was bolted to the diaphragm of the table using ½-inch diameter bolts.  In the EFT study, the columns were bolted to a ¾-inch thick base plate, which was anchored to the strong floor using 1-inch diameter threaded rods (36-inch long).  Same as the development stage study, the effective forces were applied to the structure by a 35-kip MTS 244.52 actuator controlled by a 90-gpm MTS 256.09 servovalve, which was in turn controlled by an MTS 407 analog controller.  The velocity feedback compensation schemes were implemented using a dSpace DS1102 DSP controller with a TI TMS320C31 floating-point digital signal processor with a 2 kHz sampling rate. (Click the pictures to see a movie for the tests)

Displacement potentiometers were attached to a stationary reference frame, two of which were placed at mass’s sides to capture any torsion of the specimen. Accelerometers were attached both to the mass’s lower side and to the table. A velocity transducer was attached between the diagonal braces and an I-beam fixed on the table.

The Linear Variable Differential Transformer (LVDT) housed within the actuator was used to determine the mass displacement. The structural velocity was monitored by a tachometer type velocity transducer placed on the side of the mass opposite the actuator and aligned with the center of the mass. The actuator force was measured by a load cell mounted on the actuator piston. The spool opening (i.e., spool position) of the servovalve was measured by an LVDT inside the main-stage servovalve house.

The measured damper forces indicates that the damper was not viscous (i.e., the damper force is not proportional to the velocity). Two viscous dampers were connected back to back such that symmetric (yet not viscous) damping (8%) was obtained.  Therefore, the structural response was difficult to predict even with elastic columns.