An advanced process technolgoy needs proper control by reliable and - as much as possible - objective measurements (1). One of such methods will be presented and discussed in the following. It concerns the continuous monitoring of the setting and hardening of concrete. These properties are very relevant for the slip-forming process and also for concretes which contain long-time retarders.
Fig. 1 Set-up with ultrasonic generator and receiver. Dimensions of the container: 300 mm x 300 mm x 80 mm
At the top and the bottom two small aluminium plates were placed in contact with the fresh concrete. US transducers (Geotron Elektronik, UPG-D and UEAE, resp.) touch the aluminium plates. A signal is generated by a USG (Geotron Elektronik) and sent to the upper transducer which triggers the data acquisition system. The transducers are broad-band transducers. The signal is preamplified, digitized and recorded by a transient recorder with a sampling rate of 100 ns and a 12 bit resolution. This test set-up has proven successfull after a precursor failed due to the coupling of the transducers to the container walls. Due to this coupling, waves travelled through the walls and interfered with the waves travelling through the concrete. This new test set-up is used for the measurement of the compressional wave propagation velocity in an automated manner up to every 6 minutes.
When the frequency was to be analysed a signal has been generated by a 4 mm steel ball which was shot from a mechanical gun to the upper aluminium plate. The receiver is a piezo transducer of the UEAE series. This method was superior to the US generator since a larger frequency range could be received and evaluated.
| Table 1. Concrete composition | |||||||||||
| Component l | Mass (kg/m3) | Cement CEM I 32.5 R | 320.0 | Aggregates 0/2 mm t | 635.3
| 2/8 mm | 610.7 | 8/16 mm | 531.9 | Water | 176.0
| |
The resulting water-cement ratio was 0.55. The concrete strength class is C20/25 (i.e. 25 MPa cube compressive strength at 28 days) and the consistence is KP (which means "plastic" with a spread table value of 350-410 mm). The second concrete which is the same as the first one contained a commercial retarder (Pozzutec 50 G). The amount was 8, 16 and 25 mg per kilogramm of cement, resp. In a third mix a super retarder (Delvo Stop 10 G) was used with a dosage of 11, 22 and 33 ml per kilogramm of cement, resp. The water-cement ratio has been varied between 0.40 and 0.55 while the remaining composition was kept alike. The consistence has been controlled by a small amount of plasticizer. Finally the portland cement has been replaced by a furnace slag cement CEM III/A. The mixes tested are summarized in Table 2.
| Table 2. Series of mixes tested | |||||||||||
| Series number | Important parameter | 0 | Reference mix acc. to Table 1 | 1 | Commercial retarder Pozzutec 50 G | 2 | Super retarder Delvo Stop 10 G
| 3 | Water-cement ratio 0.40, 0.45, 0.50
| 4 | Blast furnace slag cement, CEM III/A 32,5 R
| |
Fig. 2 Velocity of the compressional wave vs. age of concrete
The low value at the beginning is smaller than that of water (1430m/s). L'Hermite (2) has measured 150 to 300 m/s and explained this phenomena by a spring-mass model of many degrees of freedom. Biot's theory of the behaviour of masses in a fluid has also been applied and it was found that this theory can describe this phenomena (10). After about 9 hours the velocity increases only slowly. A plot of the results of tests with four water-cement ratios is shown in Fig. 3.
Fig. 3 Velocity of the compressional wave as function of age and w/c ratio
The lines start at about 300 m/s and develop continuously to values between 2500 and 3500 m/s after 8 hours. The lower the water-cement ratio the faster the velocity increases. This feature is well known from measurements of the compressive strength and the elastic modulus of young concrete. If a certain wave velocity is reached the concrete is not workable any more. Van der Winden (3) considered a velocity of 1000 to 1500 m/s as the end of workability. Regarding Fig. 3 the concrete with w/c = 0.40 would reach this after about 2.5 hours and the concrete with w/c = 0.55 at about 4.5 hours. The dosage of a commercial retarder can be applied such that the duration of workability can be adjusted to the need of the construction site. Fig. 4 shows that the increase of velocity depends strongly on the dosage of retarder.
Fig. 4 Velocity of the compressional wave as function of age and dosage of retarder
When 25 mg per kilogramm of cement are used the velocity after 30 h stays at a value which is only about two third of the value reached without or with little retarder. The super retarder is aimed to stop hydration for many hours or even several days. When a concrete cannot be placed on a certain day it would be possible to store it to a later time and to use it when needed. Fig. 5 shows that the super retarder has a great effect already at a dosage of 11 ml per kilogramm of cement.
Fig. 5 Velocity of the compressional wave as function of age and dosage of super retarder
With 33 ml, the end of workability can be delayed until about 3 days. Finally, the effect of the cement type on the wave velocity is shown in Fig. 6.
Fig. 6 Velocity of compressional wave in concrete with CEM III and CEM I as function of age
There is almost no difference until 8 h. After that time the CEM I concrete develops the wave velocity faster than CEM III does.
4.2 Energy transmission
Some of the energy of the input signal is dispersed into the concrete and not picked up by the receiver, another part is converted to heat. That part which is transported directly from the input to the output transmitter can be measured by evaluating the amplitude spectrum of all frequencies. The more elastic the material the larger the transmitted energy, the more viscous the less.
The US signals which have been used to produce the results of chapter 4.1 were not strong enough to transmit a measured energy up to about an age of 6 h for the reference concrete. However, the mix had set already and was not workable anymore. This means that the energy transmission from the US signals could not be used as a characterizing property.
Therefore, the impact by a steel ball was evaluated. This impact generates more energy and a broad frequency spectrum. Unfortunately, only some of the results have been processed yet. Fig. 7 shows the energy which has been transmitted through 80 mm concrete.
Fig. 7 Relative transmitted energy from a steel ball impact
The measured energy is given relative with respect to the maximum value measured when this maximum is adjusted to the growing stiffnes of the concrete. Fig. 7 shows that the energy increases at the age of 3 h for w/c = 0.40 and at 6.0 h for w/c = 0.55. The other concretes are between these limits. These ages compare very well with the end of workability as discussed in chapter 4.1.
This result may be pure chance and another geometry of the testing specimen may had lead to another answer. The systematic investigation on the relation between energy transmission and end of workability is still to be performed.
4.3 Frequency spectrum
When a steel ball hits a surface the resulting signal depends on mass diameter, and velocity of the steel ball, on the hardness of the impacting bodies and on the properties of the impacted material. When all parameters are constant with exeption of the impacted material this will determine the impact signal. The transmitted signal is additionally a function of the propagation distance. The Fresh concrete is more viscous than hardened concrete which makes that the signal has a low frequency in the beginning and a higher one later. Fig. 8 shows an example of the waves in the reference concrete from 22 minutes to almost 26 hours.
Fig. 8 Example of the waves received by the UEAE transmitter at increasing concrete age (Hours : minutes)
It can be clearly seen that the waves change considerably and that the frequencies increase with setting and hardening of the concrete. To make this phenomena more obvious the signals of Fig. 8 have been converted to the frequency domain by Fast Fourier Transform (FFT). Fig. 9a shows in the upper part the waterfall plot of the frequencies as function of age and, in the Fig. 9b, the contour plot.
| Fig. 9 Frequencies as function of age | |
 Fig. 9a) Waterfall plot (Amplitudes at early ages are enlarged) |
 Fig. 9b) Contour plot |
This is not satisfying yet. What would be needed is the prediction of setting and hardening behaviour from an accelerated test or from the measurements in the first hour after mixing. The compressional wave velocity at the age of 0.5 h is different for different water-cement ratios (Fig. 3) and dosages of retarding agents (Fig. 4). The restricted number of tests do not allow, however, to generalize this result yet, i.e. other mixes should be tested. Simultaneously, Biot's theory has to be extended to concrete and validated by testing. Both activities are on the way.
The test set-up should be analyzed again with respect to a smaller specimen, the discontinuous coupling of the receiver and the use of a single receiver in order to test several specimens subsequently with one testing device. This would be useful for a continuous production control in the ready mix plant and at the construction site as acceptance test if necessary.
 Prof. Dr.-Ing. H.W. REINHARDT , born in 1939, studied Civil Engineering and made
a thesis on photoelastic investigation of three-dimensional transient thermal stresses. Today he is Professor for Civil engineering materials at Stuttgart Univerity and Managing director of Otto-Graf-Institute (FMPA - Material Testing Research Institute) , Stuttgart, Germany.
email: reinhardt@po.uni-stuttgart.de | 
Dipl.-Geophys. C.U. GROSSE born in 1961, studied Geophysics at the University of Karlsruhe. Today he procceed on his thesis
(Deadline for July 1996) and is a research member at the Institute of Construction Materials, University of Stuttgart, Germany. email: christian.grosse@po.uni-stuttgart.de Author's Home Page
|