Abstract

The committee on nondestructive inspection of steel-reinforced concrete structures in the Federation of Construction Materials Industries, Japan, has recently published new standards on estimation of concrete properties by elastic-wave methods. Three methods for nondestructive evaluation (NDE) are specified. The one is a test method for concrete strength based on the velocity and wave properties transmitted in concrete. Another deals with an improved method for quantitative estimation of the depth of a surface crack. The other is a monitoring method for active cracks in concrete by acoustic emission (AE). Proposed standards are summarized and demonstrated with various experimental results.

1. Test Method for Compressive Strength of Concrete by Ultrasonic Waves

1.1Fundamentals
A dominant factor to govern the concrete strength is the strength of cement paste, which is empirically dependent on the water-to-cement ratio. Accordingly, it is proposed that the strength of the concrete is estimated from the relation between the strengths and the velocities of the ultrasonic waves in the reference concrete. In order to determine the strength of concrete to be tested, the effects of aggregate on the strength are compensated from such wave parameters as attenuation property and spectral characteristic. As for the mixture proportion in the reference concrete, the volume ratio of coarse aggregate is 75%, the volume ratio of fine aggregate to total aggregate is 50% and the maximum gravel size is 20 mm. In the test, the velocity of the ultrasonic wave is measured by the through-transmission method.

Concerning the sensor setting, sensors are arranged in the two ways of through-transmission and reflection. To measure the velocity and the spectral amplitudes, two sensors are set up in the through-transmission method. The attenuation coefficient of the back-scattered wave is measured by employing one sensor in the reflection method.

1.2 Measurement
The surface of concrete, to which the sensor is attached, shall be cleaned and smoothed. In the case that there exist surface coating, painting, and others which might prevent from transmitting the wave energy, these surface layers shall be ground off. For reinforced concrete members, rebars shall be located prior to the measurement, by applying such techniques as the ground-penetration radar and the electro-magnetic induction. The measurement of ultrasonic waves shall be performed at the location apart from rebars.

After calibrating the arrival time difference in the face-to-face measurement of the driving sensor and the receiver, the arrival time Dt of the received signal is measure by unit of msecond. In the measurement of the velocity, five cycles of burst-type waves with 100 kHz input frequency are driven. Taking into account the length of the measured thickness T (mm), the velocity Vo (m/sec) is determined from,

(1)

To detect a back-scattering wave, a pulse wave with frequency range from 200 kHz to 600 kHz is driven. The back-scattered wave is approximated by the exponential curve, and then the attenuation coefficient a is estimated as the unit of neper/msecond.

In the through-transmission test, a pulse wave with frequency range from 50 kHz to 300 kHz is driven. A received signal in time domain is truncated (gated) for the duration of 2 msec, and is transformed into frequency domain via FFT. Thus, the transfer function of the received signal is calculated, and then deconvoluted by the Fourier transform of the face-to-face signal. The spectral index of the transfer function St (%) is calculated from,

(2)

where St₀ is obtained, integrating from 0 Hz to the maximum frequency fo of the gated signal, and St₁ is obtained, integrating from 2fo/3 to fo as shown in Fig. 1.

Fig 1: Frequency range to be integrated.

The velocity shifts DV due to the volume ratio of aggregate, the volume ratio of fine aggregate to total aggregate, and the maximum size of aggregate are calculated based on the attenuation coefficient a, the volume ratio of aggregate Va (%) shall be estimated from an empirical relation,

(3)

Then, the variation of the velocity Rv from that of the reference concrete, of which the volume ratio is equal to 75%, is calculated from,

(4)

The first compensated velocity V₁ is estimated from the measured velocity Vo by,

(5)

In order to compensate the effect of total volume of aggregate on the spectral index St, a compensated value Stc is obtained from an empirical relation,

(6)

To compensate the effects of the volume ratio of fine aggregate to total aggregate and the maximum size of aggregate, the velocity shift DV is determined by the relations,

(7)

The second compensated velocity V_H is obtained from V₁ - DV. Eventually, measured velocity Vo is compensated, and the velocity V_H is determined. The compressive strength s is estimated from the relation of the reference concrete given by,

(8)

Some results estimated by this procedure are compared with the strengths obtained in the compression tests as shown in Fig. 2. Reasonable agreement clearly demonstrates the improvement of the proposed procedure. Thus, the procedure is effective as a practical method for determination of the strength from the ultrasonic-wave velocity.

Fig 2: Strengths obtained and estimated.

2. Method for Estimating Surface-Crack Depth in Concrete by Ultrasonic-Wave Velocity

The depth of surface cracks in concrete has been estimated by applying the velocity of ultrasonic wave[2]. The time-of-flight of a diffracted wave from the crack-tip is applied to estimate the depth. Taking into account the effects of reinforcing steel, the crack-opening width, and the crack orientation, the measuring method is improved in the standard.

2.1 Measurement
The arrival time shall be determined by either the direct observation of the first motion or the threshold detection. The resolution of time sampling shall be shorter than 1 msec. Concrete surface shall be cleaned and smoothened without sealing crack mouths. The distance between the driving and receiving sensors shall be longer than 150 mm. At least three distances shall be tested, because of the variation in the data. The maximum distance shall be determined, taking into account the attenuation property of concrete. Locations of rebars shall be estimated by referring to the specification, and/or measured by the electromagnetic radar and the electromagnetic induction as accurately as possible. It is noted that the opening of the crack mouth shall be measured by a crack gauge and so forth.

The depth of the surface crack shall be determined after the Tc-To method specified in the RILEM NDT1. By measuring Tc and To at more than three locations along the surface crack, the depth d is estimated from,

(9)

where L is the distance between the driving and the receiving sensors after the calibration. Tc is the time-of-flight, traveling via the crack tip, and To is the time-of-flight on the non-cracked surface.

2.2 Estimation of related effects with errors
In the case that a rebar crosses a crack, the effective distance ds is calculated from,

(10)

where Lc is the cover-thickness and Ls is the horizontal distance from the rebar to the crack. If the depth d determined by eq. 10 is shorter than ds, the measured depth is free from the effect of rebars. This is because the depth is underestimated due the effect of rebars if the crack depth is nucleated as deeper than the cover-thickness as seen in Fig. 3.

Fig 3: Relations between crack depths and horizontal distances from enforcement.

In the case that a crack-mouth opening (i.e. the angle of crack opening to crack depth) is so small that the cracked surfaces may contact together and the estimated depths become smaller than the actual depths. Consequently, the effective depth dw associated with the

crack-mouth opening angle is obtained from an empirical relation,

(11)

where w is the crack-mouth opening. From empirical data, it is found that the crack depth d is accurately determined if the d > dw.

In the case that the depth decreases with the increase in the measuring locations, the crack is probably inclined. This case could give information safe enough, because the depths estimated are mostly deeper than the actual depth. The error of estimated depth, d, can be within 15% provided that the following relation is hold,

(12)

where L is the distance between the driving and receiving sensors

In the case that the crack is saturated with water, the depth might be underestimated. Thus, it is noted that the presence of water inside the crack shall be taken into careful consideration.

3. Monitoring Method for Active Cracks in Concrete by Acoustic Emission

Concrete structures are known to be no longer maintenance-free. Consequently, a variety of inspection techniques are under development and going to be established. Because acoustic emission (AE) technique is promising to estimate active cracks in concrete structures, a monitoring method is developed. For in-situ inspection and monitoring, estimation of active cracks is of significant important for damage evaluation[3].

3.1 Monitoring
AE sensors shall be sensitive enough to detect AE signals generated in the target structure, taking acoustic coupling into consideration. Sensitivity calibration of AE sensors shall be performed by employing the standard source or an equivalent piezoelectric sensor. AE sensor shall be robust enough against temperature changes, moisture conditions and mechanical vibrations in the environments.

AE parameters obtained by a monitoring system are AE count, AE hit, AE event, peak amplitude, AE energy, rise time, duration time, arrival time differences in AE sensor array and so forth. AE sensors are attached at proper locations to cover the target area.

In advance to AE measurement, the noise level shall be estimated. Then, counteract against external noises, wind, rain, sunshine and so forth shall be conducted to decrease the noise level as low as possible. In the case that the noises have similar frequency contents and amplitudes to AE signals, or sources of the noises are unknown, characteristics of the noises shall be estimated prior to the measurement. Then, separation of AE signals from the noises shall be made by the use of filters after determining the proper frequency range.

3.2 Estimation and criterion
AE test shall be conducted under loads which must not cause any damages on functions and performance of the structure during detection and location of active cracks. Based on the spatial area to be covered, AE sensors of proper frequency characteristics shall be selected. In advance to the test, attenuation properties of the target structure shall be estimated, by employing the standard source or the equivalent. Then, a sensor array shall be determined so as to keep the equivalent sensitivities in all the sensors.

AE test shall be carried out routinely or temporally under the following loads:

1. Service load lower than the serviceable limit
2. Incremental load lower than the serviceable limit
3. Live and repeated loads in service

Taking the following analyses into consideration, active cracks in the structure are estimated. The results are applied to evaluate the structural integrity and the damaged degree of the concrete structure. Time history of AE parameters shall be investigated. Time variations of AE parameters is applicable to estimate cracking conditions in the structure. In most cases, abrupt increase of AE activity is associated with damage accumulation. Spatially, concentration of AE cluster results in nucleation of visible cracks. To classify active cracks, AE parameters of the rise time and the peak amplitude in Fig. 4 are applied to calculate RA value, and the averaged frequency Fa of each waveform is calculated as,

where the definitions of AE ringdown-count and the duration time shall be referred to Fig. 4.

Fig 4: AE Waveforms and pareters.

From these two parameters, classification of cracks is performed. This classification shall be based on the moving average of more than 50 events. On the basis of the criterion, cracks due to AE events are reasonably classified. One example is shown in Fig. 5. Because tensile cracks are closely related with nucleation of cracks and shear cracks occur mostly on the existing crack-surface, estimation of damage could be performed readily.

Fig 5: Cracks classified by two AE parameters.

4. Concluding Remarks

In the Federation of Construction Materials Industries, Japan, three codes are standardized for nondestructive evaluation on concrete properties by elastic-wave methods. Concerning the test method for compressive strength, reasonable agreement with the strengths obtained in the compression tests confirms that the procedure is improved as a practical method. In order to estimate the depth of the surface crack, the effects of reinforcing steel, the crack-opening width, and the inclined crack are clarified for the practical method. On the basis of the proposed criterion, cracks due to AE events are reasonably classified. Thus, estimation of damage could be performed readily.

References

1. In-Place Method for Determination of Strength of Concrete, ACI 228.1R-89, 1989.
2. Recommendations for Nondestructive Methods of Test for Concrete, BS 4408, Part 5, 1974.
3. M. Ohtsu, M. Uchida, T. Okamoto and S. Yuyama,"Damade Assessment of Reinforced Concrete Beams qualified by AE," ACI Structural Journal, Vol. 99, No. 4, 411-417, 2002.