SUMMARY

The spatial distribution of material characteristics in a large concrete specimen is subjected to differences due to a variety of influences. Results of destructive and non-destructive tests are presented in this paper. One major aim of this study is to investigate the spatial variation of the Young's Modulus within a large specimen and its comparability to specimen of the same concrete mix but different in size and subjected to different weathering. The concrete properties found by non-destructive investigations have a well defined relationship to destructively achieved results. The curing of concrete elements and the weathering they are exposed to have great influence on the ultrasonic pulse velocity. This fact has to be paid attention to even in apparently simple ultrasound measurements in situ like the determination of thickness. A critical assessment of the applied ultrasonic speed of sound is needed.

1 INTRODUCTION

For an accurate examination of a building it is necessary to consider any of its components. Economic reasons force to restrict the investigation to single components. In case of concrete structures this is tough because the components are hardly produced under constant boundary conditions. Even if components are made of the same concrete charge their technological characteristics will differ. Furthermore the material characteristics inside a component are subjected to variations.

The main objectives of this research is to investigate the influence of the spatial variation of concrete characteristics and their development due to their geometry and the weathering conditions they are exposed to on the speed of sound.

1.1 Concrete
Concrete is a conglomerate consisting mainly of water, cement and aggregates. Under certain circumstances admixtures and additives are added to the mix. One adventageous feature of concrete is that it can be produced according to its particular application. Correspondingly there are a lot of different kinds of concrete. For the investigation light weight concrete, normal concrete and high strength concrete had been regarded.

The Young's modulus indicates the resistivity of a material against deformation caused by loading. Concrete behaves nonlinear and therefore this resistivity against deformation depends on the load level. The static Young's modulus is measured as specified by DIN 1048-5 by applying a load level of 1/3 of the compressive strength of the specimen. The "load level" caused by ultrasound waves is some decades lower. The corresponding modulus is defined as "dynamic Young's modulus", i.e. the slope of the s /e curve close to the origin.

Hydration products responsible for the hardening of cement can only be built up as long as there is enough supply of humidity over a longer period of time. Normally the added water is sufficient for this process. But its amount can be reduced by the loss of the humidity at the surface. The extent of loss depends on the ability of the concrete to reserve the water and the temperature of the fresh concrete, but primarily on the weathering conditions, that means on the temperature, on the relative humidity of the air, on the insolation and last but not least on the geometry of the element. The curing of the concrete elements is of great importance for the development of their characteristics. Dependant on the cement used a lack of curing has negative effect on the degree of hydration and with this on the compressive strength of the specimen. It is stated by [1] that the relative humidity in the center of a cylinder specimen with a diameter of d = 20 mm reaches the critical concrete humidity of 80 %, at which the hydration process is interrupted, after 5 days, while at a cylinder with a diameter of d = 150 mm this value occurs not until after about 150 days.

1.2 Ultrasound wave velocity
Both longitudinal and shear waves can propagate in a solid. In an isotropic, homogeneous, elastic solid, the velocity of the shear wave is given by

where E is the Young's modulus, n the Poisson's ration and r the mass density of the material. Liquids and gases are unable to support shear waves, and thus V_S in these materials is zero.

2 EXPERIMENTAL TESTS

2.1 Test specimen
The tests were performed at light weight concrete, normal concrete and high strength concrete specimens. The description of the specimen compositions is shown in Table 1.

For all 3 types of concrete a cube 1000 mm x 1000 x mm x 1000 mm (Figure 1) and a cuboid 1000 mm x 1000 mm x 500 mm was produced. (Figure 2) The curing was performed according to the specifications of DIN 1045.

Fig 1: Cube 1000mm x 1000mm x 1000mm.

Fig 2: Partition of the cuboid 1000 mm x 500 mm x 500 mm.

The cuboid was cut in two cubes. One of them was used for the investigation of the spatial distribution of the mass density. From the other 9 cores with a diameter of 100 mm and a length of 200 mm were drilled as shown in Figure 2. From the remaining part a slab 500 mm x 500 mm x 300 mm was cut out. For one year the cores were stored under controlled conditions at 20°C and a relative humidity of 50% and the cubes and the slabs were exposed to natural environment.

The static Young's modulus and the s-wave velocity were measured at the cores. On the slab and the cube just the ultrasound velocity measurements were performed. The dynamic Young's modulus was calculated for all of them.

2.2 Ultrasonic measurements
The propagation of ultrasound in the concrete specimen is influenced by a variety of technological parameters (aggregates, porosity, etc.). Due to the inhomogeneity of concrete the signal-to-noise ratio of the ultrasonic investigation is quite low. But due to the performance of the ultrasonic velocity measurements in through transmission technique the obtained data were not significantly influenced by the noise.

2.3. Standard tests
Mass density and static Young's modulus of the specimen were measured as specified by DIN 1048.

3 EXPERIMENTAL RESULTS

The deviations of the non-destructively measured apparent velocities are caused by the spatial variation of the concrete characteristics within the specimen.

The relevant concrete characteristics for the evaluation of the ultrasound velocity are primarily the Young's modulus and the mass density.

3.1 Spatial variation of the static Young's modulus within the specimen
For all three types of concrete the results are discussed below.

Figure 3a-c shows the spatial variation of the Young's modulus inside the high strength concrete (HSC), normal concrete (NC) and light weight concrete (LWC), respectively. The vertical direction of the specimen is identified as Top-Middle-Bottom with respect to the direction of the concreting. (see also Figure 3) The nine drilled cores are illustrated as circles comprising the according values as a percentage of the value in the center of the specimen. In this case they are the measured static Young's moduli.

Fig 3: Spatial variation of the static Young's modulus within the specimen.

As expected due to the hydraulic pressure the static Young's modulus increases vertically from the top to the bottom of the specimen at an extend of about 9% at the high strength concrete and at the normal concrete. Horizontally the Young's modulus remains relatively constant. The single values vary just at a rate of 2% at the high strength concrete and at about 4% at the normal concrete, respectively. (Figure 3a & b)

The results at the light weight concrete don't show a distinct trend in any direction. The average variation of the values vertically and is about 3% and horizontally about 4%. (Figure 3c) This is reasonable because at light weight concrete the static Young's modulus is essentially influenced by the properties of the aggregates. The surface of light aggregates is craggier than that of normal aggregates. Thus during the hydration process the reaction products of the cement grow into the slots and provide a better mechanical connection between the cement matrix and the light aggregates at the transition zone.[4]

The weakest part of the system switches from the transition zone into the light weight aggregate itself. In [3] it is shown that the light aggregates are distributed relative constantly in the specimen and cause therefore relatively constant Young's modulus.

3.2 Mass density
The second important parameter influencing the velocity inside a specimen is the mass density. The investigation has shown that the mass density increases in the direction of the concreting from the top to the bottom of the specimen at high strength concrete at an extend of about 2.4%, at normal concrete at about 1% and at light weight concrete at about 4.5%. Horizontally the mass density scatters at a rate of maximal 1%.

3.3 Velocity measurement
For the non-destructive ultrasonic evaluation of the thickness of an element when just one side is accessible the knowledge of its velocity is essential. One way to determine the appropriate velocity is to drill a hole at a chosen area of the element, measure the thickness and correlate it to the time of flight of the ultrasound at the same area .

At our investigation the velocity is measured at the center of the specimen of well known thickness and set to 100% in figure 4a-c. It can be seen, that an application of this velocity on other areas of the specimen evokes a maximum deviation of about 1% at all three types of concrete.

Fig 4: Spatial variation of the non-destructively measured thickness of the specimen.

The deviation is in the range of the measuring precision of the ultrasonic measuring device. DIN 18201, 18202 and 18203 specify the tolerance thresholds for the construction site. Accordingly for elements with external dimensions less than 3m an irregularity of up to ± 12 mm is acceptable.

It can be seen clearly, that the spatial variation of the static Young's modulus and the mass density within the specimen has no significant influence on the results of the thickness measurements of LWC, NC and HSC elements as long as the drilled cores are regarded in the described way.

3.4 Comparison of static and dynamic Young's modulus
The comparison of the static and dynamic Young's modulus shows that the dynamic Young's modulus is slightly higher than the static one. Their difference decreases with increasing static Young's modulus.(Figure 5) This is in good agreement with the statements in [1].

Fig 5: Comparison of static and dynamic Young's modulus of high strength concrete (HSC), normal concrete (NC) and light weight concrete (LWC).

With the actual s-wave velocity in the center of the specimen the appropriate dynamic Young's modulus for the slab and the cube can be evaluated. (Fig. 6) The dynamic Young's modulus increases from the core to the slab for all three types of concrete. That means, elements exposed to the weathering reach higher rate of hydration resulting from the more humid storage condition. [1]

Fig 6: Dynamic Young's modulus at the core, slab and cube at LWC, NC and HSC.

The further increase of the dynamic Young's modulus from the slab to the cube at all three types of concrete indicates higher rates of hydration but beneath the storage condition in this case it is also a consequence of the larger volume. The larger the volume of an element is, the slower it dries out and the cement grains get more time to hydrate as described in chapter 1.

It is the question whether or not the obtained velocity can be applied to specimen of different shape and exposed to different weather conditions.

To demonstrate the effect the calculation of the thickness of the slab and the cube for all three types of concrete using the pulse velocity measured at the cores is show in table 2.

Column 2 contains the velocity obtained at the cores. The thickness of the specimen and the thickness calculated based on the time of flight measured and the velocity measured at the cores are displayed in column 3 and 4. The remarkable deviation is listed in column 5.

The deviation increases from the core to the slab at a rate of about 11% an NC, of about 7% at LWC and 3% at HSC, respectively. From the core to the cube the increase is even more significant. At NC it is about 18%, at LWC about 13% and at HSC about 6%.

The investigations indicate in which extend the velocity of plane elements is influenced by their thickness and the weathering conditions they are exposed to. Both characteristics trace back to the degree of hydration the elements have reached. The higher the degree of hydration is the higher is the actual velocity of sound within the specimen.

4 CONCLUSIONS

1. The spatial variation of the static Young's modulus and the mass density inside the specimen has no evident influence on deviation of the thickness measurements.
2. The spatial variation of the static Young's modulus is higher than that of the dynamic Young's modulus.
3. The dynamic Young's modulus governing the speed of sound waves of a plane element increases with its thickness and the humidity during its storage period.
4. The transferability of the velocity measured on elements with different shapes and exposed to different weathering conditions is problematic for the differing degree of hydration causes differences in Young's modulus and therefore in the speed of sound.

5 REFERENCES

1. Grübl/Weigler/Karl: Beton, Ernst&Sohn, 2001
2. Krautkrämer: Werkstoffprüfung mit Ultraschall, Springer Verlag, 1986
3. G. Astyrakakis: Ultraschallausbreitung im Beton - Variation des lokalen Elastizitätsmoduls, Studienarbeit, 2002
4. K. Klemt: The bearing and deformation behaviour of concrete subjected to uniaxial compressive short-term loading, Dissertation, VDI Verlag, 2001