Abstract:
In accordance with an embodiment of this invention, voltages are received from the pilot valve torque motor, the main servo valve and the actuator indicating the value of the torque motor current, spool position and actuator displacement. The received voltages are compared to a preselected threshold value. When one or more of the voltages exceed a threshold value, an attenuator control voltage is generated. An attenuator circuit is coupled between the reference driver signal generator and the DC servo amplifier. The attenuator control voltage, if present, is applied to the attenuator circuit and scales down the reference driver signal as a function of the error signal level thereby to limit the torque motor current and consequently the displacements of the main servo valve spool and actuator shaft.
Abstract:
The invention is a method of performing a plurality of vibratory seismic surveys simultaneously. A plurality of vibratory sources transmit signals into the earth. Each vibratory source successively transmits the same signal, except that an offset phase of the signal is selectively shifted for successive transmissions. The offset phase of the signal transmitted by each vibratory source is selected to enable the signal from each of the vibratory sources to be recovered by data processing.
Abstract:
A seismic vibrator for shaking the ground is driven by an input sweep signal. A control signal, which is a function of the output force level of the vibrator, modulates the sweep signal to limit the output force level to a value that will prevent ground decoupling. At the beginning of a sweep, an adaptive preset signal over-rides the control signal during a desired initial time portion of the sweep signal. The preset signal for a given sweep is derived from a sample of the control signal that existed during a specified portion of an immediately previous sweep.
Abstract:
A method is disclosed for improving the signal-to-noise ratio of low level seismic signals resulting from weak acoustic sources. The method has application to the summation of signals from swept-frequency sources. A reference model of the level of a valid seismic signal is built for each of a number of time windows during a seismic-data recording cycle. The model is selected by obtaining the average of the absolute magnitudes for each time window from a number of sweeps to form a set of averages. The median of each set is selected and is padded by a suitable coefficient to provide the reference model for each time window. Subsequently, a normal recording is made. Each data sample amplitude of the recording is compared with the reference model corresponding to the time window that includes the data sample. If the amplitude level of the data sample exceeds the reference model level, that sample is suppressed prior to summing.