NDTCE’09, Non-Destructive Testing in Civil Engineering Nantes, France, June 30th – July 3rd, 2009

Testing of Compressive and Bending Strength of Concrete and Monitoring Acoustic Emission Parameters Dunja MIKULIĆ1, Bojan MILOVANOVIĆ2, Ivan GABRIJEL3 1

Faculty of Civil Engineering, University of Zagreb, Zagreb, Croatia Faculty of Civil Engineering, University of Zagreb, Zagreb, Croatia, [email protected] 3 Faculty of Civil Engineering, University of Zagreb, Zagreb, Croatia 2

Abstract The mechanism of compressive failure is still in progress to be clarified; this has been tried with acquisition and analysis of acoustic emission parameters. The processing of the AE data is often not trivial and because of that, simple tests on the concrete samples were made. This paper deals with the relations between load and acoustic emission activities in concrete under different loading conditions. The compressive strength loading test was done with decreasing the friction between the specimen and the loading plates by means of rubber backing and by that decreasing the shear stresses. During testing, the acoustic emission activity was acquired. The three-point bending strength testing was also done. Cyclic loading was applied in both uniaxial compression strength testing and three-point bending strength testing in order to observe the Kaiser effect and Felicity ratio. Presented paper includes the differences in reading the acoustic emission signal and its behavior during the loading, just before the fracture and at the very time the concrete is fractured.

Résumé Le mécanisme de rupture en compression n’a pas encore été tout-à-fait éclairci. Quelques efforts en ce sens ont été faits avec la définition et l’analyse des paramètres d’émission acoustique (AE). Souvent le traitement des données AE ne peut pas être considéré comme banal et, pour cette raison, des essais simples on été faits sur les échantillons de béton. L’auteur analyse les relations entre la charge et les activités d’émission acoustique en béton sous conditions de chargement différentes. L’essai de résistance à la compression a été fait en réduisant le frottement entre l’échantillon et la plaque de chargement à l’aide d’une pièce de caoutchouc, et en réduisant les contraintes de cisaillement. Au cours de l’essai, les émissions acoustiques ont été mesurées. La résistance à la flexion trois points a également été déterminée. La charge cyclique a été appliquée au cours de l’essai de résistance à la compression simple et aussi dans l’essai en flexion trois points et cela afin de respecter l’effet Kaiser et le rapport Felicity. Les différences dans la lecture du signal d’émission acoustique, et son comportement au cours du chargement, juste avant la fracture et à l’instant même de la fracturation du béton, sont également présentés.

Keywords Concrete, compressive strength, bending strength, Acoustic emission, Kaiser Effect 1

Introduction

Acoustic emission (AE) waves are elastic waves due to dislocation motions in a solid such as cracking [1]. It is common experience that the failure of a concrete specimen under load is accompanied by a considerable amount of audible noise. In certain circumstances, some audible noise is generated even before ultimate failure occurs. Sub-audible sounds can be

NDTCE’09, Non-Destructive Testing in Civil Engineering Nantes, France, June 30th – July 3rd, 2009

detected at stress levels of perhaps 50% of the ultimate strength. With the sophisticated equipment available today, sound can be detected at much lower loads, in some cases below 10% of the ultimate strength [2]. These sounds, both audible and sub-audible, are referred to as acoustic emission and they are defined as the class of phenomena whereby transient elastic waves are generated by the rapid release of energy from localized sources within a material [3, 7]. Acoustic emissions, which occur in most materials, are caused by irreversible changes, such as dislocation movement, twinning, phase transformations, crack initiation, and propagation, debonding between continuous and dispersed phases in composite materials. In this paper compressive strength loading test was done on concrete cubes and three-point bending strength on prisms. There was no reinforcement or fibers in the samples and therefore only cracking processes should cause acoustic emission. When compression strength is tested acoustic emission could be generated due to friction between the specimen and the loading plates and cracking of concrete. In order to remove the acoustic emission events caused by friction, there was rubber backing installed between the specimen and the loading plates. Cyclic loading was applied in both uniaxial compression strength testing and three-point bending strength testing in order to observe the Kaiser effect, the phenomena when acoustic emission is found not to occur in concrete that has been unloaded until the previously applied maximum stress has been exceeded on reloading. Felicity effect as the appearance of significant acoustic emission at a load level below the previous maximum applied level was observed also. The goal of this experimental study was to gain experience in Acoustic Emission analysis and to understand its behavior in different failure mechanisms. 2

Experimental Details

2.1 AE Testing Apparatus For the measurements, following equipment was used: a resonant type piezo-electric sensor with an operating frequency range from 35 kHz to 100 kHz and a peak sensitivity of 75 V/(m/s), a preamplifier with a fixed gain of 40 dB and a µDISPTM AE main system. In order to eliminate mechanical and electro-magnetic disturbances, a high-pass filter with a cut-off frequency of 20 kHz, and a low-pass filter with a cut-off frequency of 200 kHz were used. The threshold level was set to 35dB, slightly above the previously measured background noise. The AE signals were captured with the AE WIN software and further analyzed with Noesis software for advanced analysis of acoustic emission data. 2.2 Compressive Strength Compressive strength testing was conducted in four different ways: (1) according to HRN EN 12390-3:2001; (2) cyclic loading; (3) with perforated rubber backing between the specimen and the loading plates; (4) with imperforated rubber backing between the specimen and the loading plates. Compressive strength of the samples and loading steps during cyclic loading are given in table 1. Table 1.

Properties of hardened concrete

LOADING TYPE

ACCORDING TO HRN EN 12390-3

CYCLIC LOADING STEPS

PERFORATED RUBBER

IMPERFORATED RUBBER

LOADING STRESS (MPa)

61.89

21.12 50.70 66.35

22.87

25.17

NDTCE’09, Non-Destructive Testing in Civil Engineering Nantes, France, June 30th – July 3rd, 2009

During the compressive strength testing, the crosshead velocity was kept constant at 0.25 MPa/s. Cyclic loading was performed in such manner that the loading steps would be approximately 1/3 of concrete compressive strength. Applied load, crosshead displacement and AE events were monitored during the testing of every specimen. It has to be emphasized that with use of the rubber backing between the specimen and the loading plates, shear stress in the sample was decreased. By applying this step, the failure mechanism of concrete specimens was changed. When compression strength is tested according to HRN EN 12390-3, AE could be generated by friction and cracking of concrete, the failure of concrete is increasing gradually with constant increase of damage degree [5]. When the rubber backing was used during the compressive tests the specimens were failing suddenly with unstable spreading of cracks. The result was that acoustic emission could be generated only from cracking of concrete. During the cyclic loading of concrete cubes appearance of Kaiser effect and Felicity ratio were monitored. Two samples of concrete cubes were tested for each configuration of compressive strength. AE parameters were gathered with four piezo-electric sensors when compressive strength was tested, one sensor on each side of the concrete cube. When bending strength was tested only two sensors were used to gather the AE parameters. Because of the large variance in absolute values of the AE parameters, only the results of one representative sample were chosen to be presented. Although the differences in absolute values of the AE parameters are significant for each of the sensors, the shapes of the graphs that are plotted by the AE parameters remain very similar with respect to time, for all used sensors. 2.2.1 Compressive Strength According to HRN EN 12390-3

Figure 1. Applied load and hits rate vs. time

Figure 2. Applied load & amplitude vs. time

From the figures 1 and 2 it can be seen that there is a lot of acoustic activity during the entire duration of compressive strength testing and fracture of specimens cannot be defined as brittle. The corresponding AE parameters, the hits rate and the maximum amplitude follow the increase of the applied load in a different way. The hits rate is changing rapidly at the beginning of the test, which is explained with the adjustment of the concrete specimen under the loading plates and resulting friction cracks. Afterwards the hit rate is lower just before the compressive strength is reached, when the hit rate increases rapidly (Fig. 1). The absolute acoustic emission amplitude rate (Fig. 2) is high at the beginning of the test, decreases continuously toward the end and increases rapidly just before the compressive strength is reached. 2.2.2 Cyclic Loading The results of the cyclic loading of the concrete cubes are presented in figures 3 and 4. The hits rate is changing rapidly from the beginning of the test, after unloading the specimen there

NDTCE’09, Non-Destructive Testing in Civil Engineering Nantes, France, June 30th – July 3rd, 2009

is additional activity only after exceeding previous loading. When the loading was below 70% of ultimate load, Kaiser effect can clearly be seen, but when previous loading is greater than 70 % of the ultimate loading, Kaiser effect is not evident and Felicity ratio can be calculated (Fig. 3). The absolute acoustic emission amplitude rate is high during loading; this is explained with adjustment of the sample under loading plates and friction between them.

Figure 3. Applied load and hits rate vs. time

Figure 4. Applied load & amplitude vs. time

2.2.3 Continuous loading with the use of perforated rubber backing From figures 5 and 6 it can be seen that there is a little acoustic activity during the beginning stage of the test especially in comparison with the compressive strength testing without the use of rubber backing. The hits rate and the absolute acoustic emission amplitude rate are changing from approximately the middle of the test, and are rising significantly when the ultimate load is achieved.

Figure 5. Applied load and hits rate vs. time

Figure 6. Applied load & amplitude vs. time

2.2.4 Continuous loading with the use of imperforated rubber backing Similarly to tests with perforated rubber, it can be seen that there is a little acoustic activity during the beginning stage of test (Fig. 7 and 8). In both cases this is explained with the absence of friction between the loading plates and the sample. The hits rate and the maximum amplitude follow the increase of the applied load in a similar way; they are changing rapidly from approximately the middle of the test, and are rising significantly when the ultimate load is achieved.

NDTCE’09, Non-Destructive Testing in Civil Engineering Nantes, France, June 30th – July 3rd, 2009

Figure 7. Applied load and hits rate vs. time

Figure 8. Applied load & amplitude vs. time

2.3 Bending Strength Bending strength testing was conducted in two different ways: according to HRN EN 12390-5:2001 and cyclic loading. Beam specimens were tested with a span length of 30 cm at the crosshead velocity of 0.02 MPa/s. Cyclic loading was performed in such manner that the loading steps would be approximately 1/3 of concrete bending strength. Cyclic loading was performed to monitor the appearance of Kaiser effect. The average bending strength of the samples was 5.35 MPa. 2.3.1 Bending Test of Concrete Prisms The results of the bending test of concrete prisms are presented in Figures 9 and 10.

Figure 9. Applied load and hits rate vs. time

Figure 10. Applied load & amplitude vs. time

The AE parameters, the hits rate and the maximum amplitude follow the increase of the applied load in a similar way. The hits rate remains rather negligible during the test and rapidly increases just before the bending strength is reached (Fig. 9). The maximum amplitude changes during the tests with the increase of applied load, and reaches the extreme amplitude at breaking point (Fig. 10). 2.3.2 Cyclic Bending Test of Concrete Prisms Cyclic loading results confirm the existence of the Kaiser effect. Unlike the cyclic loading during compression strength testing there was no significant AE at load level below the previous maximum applied level, (Figures 11 and 12).

NDTCE’09, Non-Destructive Testing in Civil Engineering Nantes, France, June 30th – July 3rd, 2009

Figure 11. Applied load and hits rate vs. time Figure 12. Applied load & amplitude vs. time 3

Discussion and Conclusions

The experiments described above were carried out with the primary objective to gain experience with AE testing equipment and techniques. While the main goal was reached, new questions were raised during the course of the study as well. The results of the investigation were summarized by the following concluding statements: The acoustic emission methods can be used as a reliable method for detection of cracking in concrete. Maximum amplitude varies during the test, but it seems that the maximum value is reached when a major crack occurs in the concrete. During the unloading, the AE activities disappear. The absolute values of the AE parameters as given by the individual AE sensors are difficult to be compared directly. Sensor characteristics, quality of the contact between each sensor and the concrete surface, and local concrete porosity significantly influence acoustic acquisition. It can be noticed that there is a difference on the type of the loading conditions. Uniaxial loading results with a weak acoustic activity before the failure of the sample, respectively to the triaxial compressive strength testing.

Acknowledgments This research was performed within scientific project “From Nano- to Macro-structure of Concrete”, 082-0822161-2990, funded by Croatian Ministry of education, science and sport.

References 1. Grosse, C.U.; Ohtsu, M. - Acoustic Emission Testing, Basics for research-Applications in Civil Engineering, Springer – Verlag Berlin Heidelberg, 2008. 2. Mindess, S. - Handbook on Nondestructive Testing of Concrete, Chapter 16 - Acoustic Emission Methods - University of British Columbia, CRC Press LLC, 2004. 3. ASTM E 1316-02a, Standard Terminology for Nondestructive Examinations, Section B: Acoustic Emission, Annual Book of ASTM Standards, ASTM, West Conshohocken, 2003. 4. Šajna, A.; Kovač, J.; Stipanović, I.; Mikulić, D.;: Determination of bond and flexural strenght of reinforced concrete by acoustic emission., Proceedings, 2006-NDE-Conference on Civil Engineer, St. Louis, Missouri, SAD, 2006 5. Ukrainczyk, V., Beton – Struktura, Svojstva, Tehnologija, Alcor, Zagreb, 1994. 6. Šajna, A.; Kovač, J. Bajt Ž. : Acoustic emission monitoring of cracking in reinforced concrete specimens, 2nd International Symposium on Advances in Concrete through Science and Engineering, 11-13 September 2006, Quebec City, Canada 7. Šajna, A.; Bremec, T.: Use of the acoustic technique on reinforced concrete bending loaded specimens; Proceedings of the First International RILEM Symposium SACoMaTiS 2008, Rilem Publications S.A.R.L., 2008. pg. 211-221.