INSTRUMENTATION CORRECTIONS TO WAVE VELOCITY DATA

The detection of the arrival of seismic stress waves for a field measurement of wave velocity requires that the incoming signal be recorded and displayed by an instrument such as a seismograph. Filters, introduced into recording circuits to limit noise or to prevent aliasing in digital recording, can have significant effects on the observed arrival times and on the computation of wave velocity. Such filtration is often not taken into account in processing the data, but the example that follows indicates how the effects can be evaluated and the data corrected by simple techniques. The properties of electronic filters are, of course, well known to electrical engineers, and the following comments are concerned with an application in which their importance has not always been appreciated by those trained in other disciplines. Two basic conclusions can be drawn from this work: 1. The characteristics of instrumentation can have important effects on the results of field geophysical tests. In particular, the effects of filtration should not be ignored. 2. Regression analysis of travel time versus distance is effective in determining timing error and correcting for it. There is a widespread belief that, if one utilizes a hammer with reversible polarity, shear wave arrivals in cross-hole seismic tests can be identified unambiguously because the direction of first motion in the shear wave reverses when the direction of the impact is reversed. The direction of first motion in the P-wave is assumed not to reverse. Unfortunately, shear wave sources such as the Bison hammer also produce P-waves of opposite polarity when the direction of impact changes. The reason that P-waves reverse polarity is that a downward excitation of the hammer causes initial compression below the level of the hammer and initial tension above it. An upward excitation reverses this pattern so the P-waves must reverse polarity. A theoretical argument is that in a linearly elastic material, the superposition of equal and opposite upward and downward excitations gives a case of no net excitation. If the P-waves do not reverse polarity, they will not cancel, and there will be energy propagated with no input. Non-reversing P-waves may be caused by several factors, including late arrivals of reflected or refracted waves and secondary dilatational effects near the source. However, this misconception probably arises because in the down-hole test, shear wave energy is often created by striking a horizontal beam or plate on the surface in opposite horizontal direction. This unreversed vertical load causes an unreversing P-wave, which can be observed in field data. (Author)