Acoustic Emission Technique

The acoustic emission technique is a non-destructive technique which detects and locates damage at the moment of occurrence. Acoustic emissions (AE) are high frequency transient sound waves, which are emitted during local stress redistributions caused by structural changes, such as crack growth. The technique is proposed here as a monitoring technique for the detection of damage initiation and assessment of the rate of damage evolution during creep deformation.

When a set of several sensors is applied for the acoustic emission measurement, localisation of the

Rise

Rise

Event duration

Figure 5. Main features of an AE signal (burst type).

Event duration

Figure 5. Main features of an AE signal (burst type).

source of the emissions is possible by taking into consideration the geometrical arrangement of the sensors and the moment of arrival of a wave at each individual sensor. This technique is frequently used for detection of the location of damage within isotropic materials. As masonry is a highly heterogenic material, with a different propagation speed of the energy waves in different directions, localization of damage is very difficult and will therefore not be taken into account here.

An AE wave, detected by a sensor is called a "hit". In order to filter out the continuous low-amplitude background noise, a threshold is defined and only sound waves passing this amplitude-threshold are detected. Two or more hits, captured by different sensors and originating from the same event are simply referred to as one "event". The main features of an AE signal (burst type) are schematically indicated in Figure 5.

3.1 The use of acoustic emissions for damage detection

Examples of damage assessment by AE monitoring in brittle materials under mechanical loading can be found in literature (Grossi 1996, Eberhardt 1997, Colombo 2003). References describing the use of this technique on masonry on the other hand are to a much smaller degree available in literature, especially regarding the monitoring of damage evolution in masonry as a consequence of long-term creep deformation. Acoustic emission monitoring has been used on masonry arch bridges (Tomor & Melbourne 2007). A practical study of the creep phenomenon in masonry towers combined with acoustic emission monitoring was performed by Carpinteri (Carpinteri 2007).

The amount of hits, detected during a certain time interval, depends on various specific boundary conditions of the test setup, such as the threshold level, the quality of the coupling between the sensor and the test specimen, the density, coherence and speed of wave propagation in the material and the interference of surrounding test equipment to name a few. Therefore, the necessary precautions have to be taken in order to keep these boundary conditions as much constant as possible. Concerning this remark, it also follows that not the absolute amount of detected events, but rather the change in detection level or event detection rate is a determining factor for the assessment of the damage accumulation. The experimental research, discussed below, indicates that the event rate is related to the rate of damage increase within the masonry.

3.2 AE testing on masonry cores

To gain more insight in the described creep phenomena, short-term creep tests were carried out on masonry core samples in combination with acoustic emission (AE) detection. It concerns masonry cores, composed of brickwork and cement mortar, which were taken from an existing structure in order to test the material's compressive strength. As the drilled cores showed a very good coherence, also creep testing could be carried out together with AE detection in order to test the combined use of both deformation and AE measurements. The short-term creep tests were carried out based on the knowledge obtained during the previously discussed test program. From the compression tests, an average compressive strength (fc) of 6.58 MPa (± 1.15 MPa standard deviation) was found for the masonry cores. The loading scheme for the accelerated creep tests was set to start at a load of 20% of fc and then the load was increased in steps of 10% at the beginning of the test and 5% during the last load steps. The initial load of 20% was taken to be considerably low, to take into account the large scatter on the com-pressive strength of the samples. By setting a low initial load, each sample would have enough load increment steps during the test. Each loading step was kept constant for at least 20 minutes, longer time increments were used if the strain rate was not showing a constant or decreasing evolution after 20 minutes. By following this scheme, each test had a duration of approximately 6-8 hours.

Acoustic emissions, the stress-strain evolution and the evolution of the vertical strain in time could be followed online during the test. Therefore, following sensors were applied for each test (Fig. 6):

- Two acoustic emission sensors (range 250700 kHz, peak amplitude 375 kHz);

- Two LVDT's with a range of ± 1 mm to measure the horizontal strain (type Schaevitz LBB);

- Two LVDT's with a range of ± 2.5 mm to measure the vertical strain (type Schaevitz LBB).

Teflon sheets were placed between the metal plates of the press and the specimen to immobilize fric-tional forces. The test was carried out stress-controlled in order to be able to keep a constant stress level, even if deformation of the specimen would occur.

Figure 6. Accelerated creep tests (a) and cyclic accelerated creep tests (b) on masonry core, test setup with AE sensors and LVDT's.

Two different types of accelerated creep tests were performed:

- Standard tests with small stress increases, alternated with periods during which the stress level is kept constant, as explained above. They will be called accelerated creep tests (ACT);

- Creep tests similar to the previous ones, the difference being that before every stress increase, an unloading cycle is included. This way, it is easier to capture the evolution of the elastic parameters as damage increases and distinguish between the immediate elastic deformation and time-dependent deformations. These tests will be called cyclic accelerated creep tests (CACT).

For the ACT (Fig. 6a), the vertically placed LVDT's were positioned in a cylindrical metal frame in which the test specimen was placed. Consequently, also the deformation of the Teflon sheet and the compression of small irregularities on the specimen's surface were measured by the strain gauge. During the CACT (Fig. 6b), higher specimens were used, so that the vertical LVDT's could be placed on the specimen itself. The horizontal strain gauges were always positioned in the metal frame (Fig. 6a-b).

The stress-strain relation for a typical ACT and a CACT test are presented in Figures 7-8. Horizontal strains are indicated as negative, vertical strains as positive.

Figure 8a very clearly shows the unloading and reloading cycles, from which the decrease ofthe elastic modulus can be determined. From this graph, the elastic modulus does not appear to decrease substantially when damage increases as the slope of the reloading curves remains rather constant.

The evolution of the horizontal and vertical strains in time for both tests is indicated in figures 7b-8b. Both curves show an elastic deformation at stress increase, a small primary creep phase at the beginning of each

Figure 7a. Stress-strain curve for an accelerated creep test on masonry cores.

Figure 7b. Strain evolution in time for an accelerated creep test on masonry cores.

load step and a secondary phase at the end of the load step. As the time increment was rather short, a clearly pronounced secondary phase with constant strain rate was not always obtained for every step. At some load steps, the strain rate is still decreasing at the end of the step, which indicates that there is still primary creep present. All tests showed a stress-dependent strain rate, as was also detected during the accelerated creep tests on the masonry wallets discussed above.

Figure 7b shows a too large vertical strain increase at the initial load step as a consequence of the compression of the Teflon sheet when the first load is applied. All specimens reached the tertiary creep phase very fast during the last load step as a consequence of the short time intervals taken. Tertiary creep would probably also have been reached during the second last load step if the time increment could have been longer. Due to the time restriction (the test had to be concluded within one day) this was not possible.

Figure 9. Acoustic emission measurement during accelerated creep test (ACT), events indicated in time by differential bars and cumulative (above). Load increment steps during ACT (below).

Figure 8a. Stress-strain curve for a cyclic accelerated creep test on masonry cores.

Figure 8b. Strain evolution in time for a cyclic accelerated creep test on masonry cores.

Figure 9. Acoustic emission measurement during accelerated creep test (ACT), events indicated in time by differential bars and cumulative (above). Load increment steps during ACT (below).

Figure 8b. Strain evolution in time for a cyclic accelerated creep test on masonry cores.

Table 1. Dimensions of the test specimen and resulting compressive stress during short-term creep tests.

Test

diameter

height

fc

Type of

specimen

(mm)

(mm)

(MPa)

test

Core 1

Was this article helpful?

0 0
Renewable Energy Eco Friendly

Renewable Energy Eco Friendly

Renewable energy is energy that is generated from sunlight, rain, tides, geothermal heat and wind. These sources are naturally and constantly replenished, which is why they are deemed as renewable.

Get My Free Ebook


Post a comment