Distant ~otions from a Building Vibration Test By

Horizontal ground motion generated by vibration tests of the nine-story Millikan Library Building on the Caltech campus was recorded on 1he surface of the ground in the Pasadena area at distances up to 3 miles from the building. Later it was learned that the vertical component of the motion also was recorded by the seismograph on Mt. Wilson, 6.7 miles from the Library and 4,800 ft higher in elevation. The magnitude of the acceleration varied from 2.04 X 10-2g at the excitation level on the ninth floor of the building to 3.2 X 10-~g at Mt. Wilson. Simple calculations show that multistory buildings are particularly well-suited for inducing large dynamic forces in the ground with relatively small equipment. INTRODUCTION During the vibration tests of the nine-story Millikan Library Building at the California Institute of Technology (Jennings and Kuroiwa, 1968, Kuroiwa, 1967), buildinginduced motion was clearly recorded off campus at the Caltech Seismological Laboratory. The building is founded on approximately 900 ft of alluvium overlying bedrock. After an additional test confirmed that the building vibration was indeed the source of the motion, an experiment was conducted to measure the surface motion in the Pasadena area. It was found that the motion could be measured easily by portable seismographs at distances up to 3 miles from the building site. After the test was completed, it was learned that the vertical component of the motion was recorded by the seismograph atop Mt. Wilson, 6.7 miles distant from the Library and 4,800 ft higher in elevation. INSTRUMENTATION AND TEST PROCEDURES The building was excited into resonance in the N-S direction by two of the vibration generators developed at Caltech for the State Division of Architecture under the sponsorship of the Earthquake Engineering Research Institute. Details of the vibration generators, now the property of the University of California, and descriptions of their use in the Millikan Library tests are available in the literature (Hudson, 1962; Jennings and Kuroiwa, 1968; Kuroiwa, 1967). The accelerometer-recorder system used to measure the building motion and the motion on the nearby soil also has been described elsewhere (Kuroiwa, 1967). The surface motion in the area not immediately adjacent to the Library was measured by six of the portable seismographic stations owned and developed by the Seismological Laboratory of the California Institute of Technology. These instruments use Ranger-type seismometers as the sensing element and record on film. Their characteristics have been described in a paper by F. E. Lehner and F. Press (1966). Beginning at approximately 7 p.m., December 15 1966, the building was excited into resonance, using full weights in the eccentric mass vibrators. The fundamental mode in the N-S direction (1.92 cycles/sec) was chosen as this enabled the greatest amount of energy to be fed into the building by the two 11⁄2-hp vibrators. As a result of this choice, all reported measurements with the exception of that from Mr. Wilson, are for N-S motion. The purpose of this first part of the test was to calibrate the seismo2037 2038 BULLETIN OF THE SEISMOLOGICAL SOCIETY OF AMERICA graphic instruments as accelerometers at the chosen frequency by recording motion at the same point by both the seismometers and the aeeelerometers. The aeeelerometerrecorder system then was calibrated in a routine manner on a tilt table (Knroiwa, 1967) to complete this portion of the experiment. The calibration point, shown in Figure 1, was chosen as far from the building as it was possible to obtain usable records with the aecelerometers. Later examination of the film records from the seismometers showed that the calibration was not completely successful in that the motion appeared to be too strong for the seismometers, although they were attenuated fully. The motion at the site was very nearly sinusoidal according to the aeeelerometers, but this was not the case on the