| dc.description.abstract | A dry-point ultrasonic instrument can be used to detect the degree of
deterioration in concrete. Most data shows that wave velocity histograms
are often a bimodal distribution. The first peak will appear at a velocity of
3,600 m/s to 4,500 m/s, with the second peak at about 2,000 m/s to 2,600
m/s. Literatures, however, haven’t investigated how wave velocity
histograms appear or what their micro-mechanisms are.
In order to explore the bimodal distribution of wave velocity
histograms, a series of physical model tests were carried out in this paper.
Through experimental observation and theoretical analysis, a wave
velocity simulation model was constructed. In the physical models test, a
series of acrylic and concrete model with different depths of fracture
created by saw blade. Apparent wave velocities of various locations were
measured by moving the position of the probe. The measurement results
showed that when the probe was far from the cracks, the wave velocity
value was slightly lower than the theoretical value. When the probe was
near the crack, however, the wave velocity value dropped sharply and
was much lower than the theoretical value. The rapidly decreasing area
continued to expand in conjunction with the increasing depth of the
iv
cracks. The reason for this was that when the probe was too close to the
cracks, the wave velocity was unusually low or unable to be measured.
After regularization of the measured acrylic and concrete wave velocities,
the two were found to be quite consistent. This showed that the
phenomenon of wave velocity decline and material were unrelated. When
the probe was too close to the crack, there was an area of abnormally low
velocity measurement, which this paper defines as the “blind zone.” The
normalized wave velocity was about 0.48 ~ 0.49. In this paper, the
distance of the probe was 15 cm. When the crack depth was within 4.5
cm, the zone was about the same as the depth of the crack. When it was
more than 6 cm, they were all in the blind zone.
When the probe was close to the crack, the wave was harder to
transmit and therefore may have caused the wave path to increase. The
distance difference (ΔL) of the different measurement points may be
obtained from the results of the physical model test. The physical model
of the test results showed that they were close, which appears to indicate
that it was not related to the material, but rather the probe position and
fracture distance.
The distance difference from the physical model test may take into
v
account the blind effect, but the in-situ width of the crack was much
smaller than the artificial width. Furthermore, if the artificial cracks were
not completely separated and ΔL was used, there is the possibility of
excessive consideration of the effects of the blind zone. The measured
data from the wave velocity showed that the probe across the cracks was
the main cause of the decrease in wave velocity, but there was no
significant decrease in the wave velocity of the cross-cracks. A possible
reason was that the cracks were shallow or healing. The crack density
obtained from the current survey should also be reduced. In order to
consider the reduction of the blind zone effect and fracture density, this
paper introduces the dead zone reduction factor (α) and fracture density
reduction factor (β).
In order to simulate the grid method to measure the wave velocity of
concrete, the data for intact concrete wave velocity, crack density, and
crack depth can be obtained from a local survey. The distance from the
dry-point ultrasonic probe was fixed to 15 cm and the crack density was
multiplied by the crack density reduction factor (β). Using the Poisson
distribution equation, the wave path of the probe can be obtained
randomly. If the probe is in the blind zone, the total path length is the sum
vi
of the wave propagation path and the distance difference multiplied by
the blind zone reduction factor (α). If the probe is in a non-blind zone, the
total path is the wave path. The total path length divided by the probe
distance (15 cm) and multiplied by the complete material wave velocity
gives the apparent wave velocity. Repeat this procedure, and the apparent
wave velocity histogram and cumulative probability mass function may
be obtained. The α and β reduction factors proposed in this paper have
their physical meaning, but there is still no specific or effective
measurement method. According to the current wave velocity
measurement data, the α and β reduction factors will be found. From the
simulation results constructed in this paper and the wave velocity
accumulation curve of the current measurement, numerical solutions
showed a good agreement with the experiments. The model proposed in
this paper can reasonably simulate the bimodal phenomenon of the wave
velocity histogram and explain its mechanism.
Keywords：P-wave velocity、Surface crack | en_US |