Geoscience Reference
In-Depth Information
One peculiarity of these tests is that the size L of the wave zone was much greater
than characteristic wavelength of the radiation. Even if the characteristic frequency
of the fracture-induced signals reaches the value of the order of 1 MHz, then L
300 m. Since the antennas were located at the shorter distances, only near-field
components can be measured in these experiments. The complexity of the problem
is that the electric field decreases inversely proportional to the distance cubed; that
is, more rapidly than that in the wave zone. On these conditions the antenna operates
as an opened capacity rather than the usual radiowave antenna. Moreover, the size of
antenna is much smaller than a typical wavelength so that one should calibrate the
antenna before performing a measurement in order to give a correct interpretation
of these data. One more problem is that the standard formulas for energy radiated
by the source cannot be applied in this area.
Warwick et al. (
1982
) have noted that the electromagnetic signals appear
synchronously with acoustic emission of the fractured sample. The frequencies of
about 50 kHz dominate in the recording of electric field and the magnetic field is
maximized in the range of 1-2 MHz. The similar experiments reported by Yamada
et al. (
1989
) have shown that the spectrum of electromagnetic signals caused by
failure had a maximum within 0.5-1 MHz. Brady and Rowell (
1986
) and Cress et al.
(
1987
) have found that the maximum of spectral density of the electric signals has
lower frequency and lies in the range 0.9-5 kHz. Ogawa et al. (
1985
) have studied
fracture of granite samples under uniaxial loading and bending moment of external
forces. In both cases wide-band electric fields (0.01-100 kHz) were detected.
Oscillations in the radio-band lasted about 15 ms have been observed during
compression of the basalt and granite samples (Martelli et al.
1989
). The fractured
samples were surrounded by seven inductance coils, which measure the signals
within a frequency band from 500 Hz to 830 kHz. Martelli et al. (
1989
)have
assumed that the observed effect could be resulted from the low-frequency vibra-
tions of plasma arising under the fracture.
Simultaneous recording of the acoustic and electric signals before and during
destruction of the samples has been made by Yamada et al. (
1989
). Cylindrical
granite samples with length of 62.5 mm and diameter of 25 mm were undergone
by uniaxial compression with slow strain rate (about 10
6
s
1
). In this case the
complete failure of the samples occurred in tens minutes. The acoustic emission
sensors placed in different points of the sample surface recorded the impulses due to
the process of microcracking. To measure electromagnetic impulses the inductance
coils were put on the cylindrical sample with small gap, and their resonance
frequency is changed from 80 kHz to 1.2 MHz. It was found that 15 events of
the electromagnetic emission fall on 211 events of acoustic emission in one of the
experiments and 31 events against 135 ones were detected in the next experiment.
However, these results could be different from those in the case of higher sensitivity
of the electromagnetic sensors.
Yamada et al. (
1989
) have found that the acoustic emission increases up to
the moment of complete failure of the sample whereas the electromagnetic one is
more intensive at the early stage of loading. It was supposed that the acoustic and
electromagnetic impulses are simultaneously emitted because they are controlled
Search WWH ::
Custom Search