DTK :: Bieberstein, A., Saucke, U. - THE KARLSRUHE CUT-OFF WALL TESTING UNIT (KTU) FOR MONITORING IN-SITU PERMEABILITY (German Dam Research ...)

Deutsches TalsperrenKomitee (DTK)
MITGLIED DER
INTERNATIONALEN TALSPERRENKOMMISSION (ICOLD)

URL dieser Seite

THE KARLSRUHE CUT-OFF WALL TESTING UNIT (KTU) FOR MONITORING IN-SITU PERMEABILITY

A. Bieberstein & U. Saucke
Division of Embankment Dams and Landfill Technology, Institute of Soil Mechanics and Rock Mechanics, University of Karlsruhe, Germany

ABSTRACT: The paper presents a testing unit for monitoring the hydraulic permeability of cut-off walls. The testing unit was developed, constructed and applied in initial tests in practice at the Institute of Soil and Rock Mechanics in Karlsruhe and different building sites. It can be implemented under conditions generally prevailing at cut-off walls with one and two phase methods and enables the permeability of cut-off walls to be measured by means of percolation at predetermined depths over any length within certain limits, both immediately after completion (and hardening) or at any time afterwards. All elements of the KTU are described as well as the formula for evaluating hydraulic permeability values. Further, the installation in a cut-off wall underneath a flood water dike is presented with first measurement results.

1 INTRODUCTION

It is a known fact that it is a complex undertaking to prove that cut-off walls are sufficiently impermeable. A number of factors play a role here, which make the tests and evaluations involved either more or less difficult or even unreliable -depending on the cut-off material, method of construction, geology of the site etc.:

The cut-off material may behave differently in the panel compared to laboratory test specimens, so that the latter may not necessarily be representative.
The overall permeability of the cut-off wall is generally greater than the permeability of the individual wall panels or even the cut-off material itself, because, for example, the joints and other imperfections in the panels have a negative influence.
On site, the wall acts together with a 'sealing base layer' (which has either been built on purpose or exists in the form of a natural sealing layer), the sealing effect or permeability of which influences the whole behaviour of the encapsulation in a manner difficult to estimate in advance.

These conditions are taken into account, when building actually takes place, by means of a quality assurance system with a graduated programme of tests and controls, ranging from measurements on specimens in the laboratory and quality controls of different kinds during construction to a pumping test on the finished object (Ratnam et al. 2001).
Unfortunately from time to time it is discovered that - despite all efforts of those involved during the planning and construction phase - the observations made at the "moment of truth", i.e. at the end of any construction work, are not as favourable as expected, or indeed are completely unsatisfactory.
Past experience has shown that a method for measuring in-situ permeability of cut-off wall panels would be useful, in as much as it would avoid the necessity of testing bore holes (which are undesirable for various reasons, and from a testing point of view technically unreliable). With this in mind, a testing unit was developed, constructed and applied in initial tests in practice at the Institute of Soil and Rock Mechanics in Karlsruhe, Germany, which can be implemented under conditions generally prevailing at cut-off walls with one and two phase methods. It enables the permeability of cut-off walls to be measured by means of percolation at predetermined depths (over any length within certain limits) immediately after completion (and hardening) or at any time afterwards.
The following is a report on this testing unit and the first results obtained under building site conditions. The unit was designed with a view to attend to two requirements: obtain essential information about the quality of cut-off wall panels under specific applications, since it should be adaptable to specific site conditions and be economical to acquire.

2 DESIGN AND APPLICATION

The basic intention, on which the design is based, is to insert a permanent testing control unit of a preselectable length, which has been prepared in a workshop and can then be let down to any depth in newly constructed cut-off panels on site, in the open panel. The unit should be mainly made up of components easily available on the market and should require as few special parts as possible and be simple to use on site.
In figure 1 a unit according to this conception is shown. The KTU presented has a testing length of 1.80 m; however, by linking several elements together the length can easily be extended.

Figure 1. Complete testing unit, 2 m in length in this case, with all the components.

A connecting hose exists between the body of KTU and a plastic chamber with tube connections. The hose located inside the cut-off wall protects the pipe for the water supply and the tubes for the installation phase. The pipe for the water supply leads to the water reservoir, which feeds the system during the permeability test.
The thin plastic tubes for the installation phase form the connections, which are required when installing the unit in preparation for the actual permeability test. By means of suitable parts inserted in the testing tube they serve to prevent the initially fluid cut-off wall material with its greater density from draining into the testing unit, in which there is only hydrostatic counterpressure from the water inside. Any drainage of this kind would lead to accelerated consolidation and compression as with a filter cake around the testing unit and would, of course, distort the measuring results.
Further, at the lower end of the KTU an iron ballast element is attached to equalize the resulting uplift depending on the density of the cut-off wall material and the volume of the KTU. The iron weight makes it possible to suspend the body of the KTU vertically in the open panel. A sketch of the testing unit after it has been installed and prepared for operation is shown in figure 2.

Figure 2. Sketch of the KTU after installation and prepared for operation.

3 EVALUATION OF TEST RESULTS

When the testing unit, which has been completely preconstructed in the workshop, has been inserted into the cut-off wall element to be tested and the hardening process of the cut-off wall material has been completed, the aid device (special packer technique) to prevent the formation of filter cakes is released. After the unit has been connected to the water supply, the apparatus is ready for use.
The system dimensions, which are essential for the evaluation, are given in figure 2. Since generally longer testing stretches of at least a few meters in length will be selected, it is sufficient to interpret the flow out of the testing unit through the wall area to be tested as a horizontal two-dimensional problem and thus to ignore the three-dimensional upward and downward discharges (cf. Brauns & Charles de Oliveira 1987).
In this way, one is on the safe side regarding the evaluation of the permeability of the wall, and the related equation to determine the permeability of the wall from two readings of the water level in the water reservoir at different times is:

(1)
where r_F and h_F are radius and length of testing stretch; h₀ and h_t are the water heads above ground water at the beginning and the end of the period of observation (see figure 2); (t_t - t₀) is the duration of the period of observation; d_w is the thickness of the cut-off wall.
The testing unit is designed on the basis of filter pipes with a 2" diameter. A water barrel with a burette can be used for the water reservoir.
Suitable measuring periods are, of course, determined by the selected length of the test stretch and the actual permeability of the tested area of wall on the one hand, and on the other, by the chosen size of the water supply and the accuracy of the reading of the water level, i.e. the volume of water consumed. It could be useful to switch between the water barrel and the burette by means of a three-way valve, so that in addition to carrying out a longer infiltration test using the water in the barrel, also a short term reading is made possible by operating the burette.
When planning the application of the testing unit, the question arises as to the quantity of the water supply, taking into account the geometrical system requirements and the expected permeability of the wall. The equation for the fall velocity of the water level _(vs) in the water supply with diameter (2r_R) to be selected is given as an orientation aid:

(2)
The fall velocity at the beginning of the test serves mainly as an orientation aid, i.e. at t = 0 and h = h₀. For this condition the last factor in the equation equals 1 and the initial fall velocity becomes

(3)
(This equation of course applies basically for each condition h during the test, when the momentary h is used as h₀).

4 APPLICATION INSIDE A CUT-OFF WALL UNDERNEATH A FLOOD WATER DIKE

For the first time the KTU was installed in a cut-off wall for the encapsulation of a quantity of hazardous waste (cf. Brauns et al. 2000). The successful operation of the KTU encouraged the authors to put the KTU in a cut-off wall below a flood water dike close to a large river (see figure 3).

Figure 3. Cross section of the dike with subsoil and cut-off wall (detail: cf. figure 5).

In addition to the initial application a method had to be developed to lead the connecting hose with the supply tubes through the body of the dike, in particular through the clay core which has to act as a seal. When doing this there is also an opportunity to determine the permeability of the cut-off wall beyond the ground water table with regard to the behaviour depending on whether it is loaded under flood water conditions. In particular, the influence of the desiccation process of the cut-off wall material with regard to the hydraulic permeability can be investigated.
The cross section of the dike is presented in figure 3. The dike with the clay core and an access road located on the landside is based on a clay layer with a thickness of about 2.0 m deposited by fluvial processes, overlying a gravel and sand layer at least 10 m thick. Underneath a tertiary layer with a low hydraulic permeability adjoins this.
With this picture in mind it is obvious that the installation process has to be separated into different steps. The first step consists of hanging the KTU into the fresh slurry of the cut-off wall (cf. figure 4).

Figure 4. KTU just before installation into the fresh slurry of the cut-off wall.

After the hardening process of the fresh slurry, work takes place to build the body of the dike in several layers. In order to protect the connecting hose is led inside a plastic pipe through the clay core up to the crest of the dike (cf. figure 5). The pipe ends in a manhole consisting of concrete elements at the top of the dike.

Figure 5. Cross section of the dike with cut-off wall, clay core and manhole for test equipment (cf. figure 3, detail).

The water reservoir and the supply tubes are placed inside the manhole, which can be covered with an iron lid (cf. figure 6). As a result of this construction all the measurement elements are well protected on the one hand, whilst on the other hand, everything is easily accessible, so that long time measurements can be carried out as well.

Figure 6. Manhole, located in the crest of the dike, with water barrel and supply tubes.

Since the installation process of the KTU on this site, two measurements have been carried out with two different KTUs. One KTU is located below the ground water table, and one above it. The evaluated hydraulic permeability values are presented in figure 7 in accordance with equation (1). The first reading was taken after 3 months, the second after one year. The hydraulic permeability values of both units are satisfactory, since the values are less than required according to the quality assurance system. Also the process of decreasing permeability is as expected. Further readings, which will be taken in coming years, will show the behaviour of the hydraulic permeability depending on the position related to the ground water level.

Figure 7. Example of a measuring result, evaluated as a wall permeability coefficient from a one phase cut-off wall.

5 CONCLUDING REMARKS

In this article, a testing technique for the determination of the permeability of individual cut-off wall panels in situ is described. Furthermore, the installation of the unit inside a cut-off wall underneath a flood water dike and some successful measurements are presented.
The unit is designed for practical use and for this reason is composed mainly of commonly available parts, in addition to a few specially made elements. It is intended that the unit be assembled ready for use in a workshop so that it only needs to be suspended in the newly made open cut-off wall on site. The length of the test stretch can be more or less selected as required and prefabricated (in meter steps). Special measures are necessary to prevent the formation of filter cakes around the testing unit cover during the period of installation in the cut-off wall material, which has not yet hardened. It was seen that it is possible to solve this problem.
The testing unit remains in the cut-off wall to be monitored and is thus completely constructed out of non-degradable and otherwise durable materials. As a result, once a KTU has been installed, it is available for reuse at the testing site for a long period of time.

REFERENCES
BRAUNS, J. & CHARLES DE OLIVEIRA, L.M. 1987. Abströmung aus Prüflöchern in Dichtwänden. Geotechnik 1987, H. 1.

BRAUNS, J., BIEBERSTEIN, A., REITH, H. 2000. Das Karlsruher Dichtwand-Prüfgerät zur in-situ-Prüfung der Durchlässigkeit von Schlitzwänden. Bautechnik 77, H. 9.

RATNAM, S. et al. 2001. An in situ permeability measurement technique for cut-off walls using the Cambridge self boring pressuremter. Proceedings of the 15. International Conference on Soil Mechanics and Geotechnical Engineering, Istanbul 2001, Volume 1.