Rock Mechanics in Civil and Environmental Engineering – Zhao, Labiouse, Dudt & Mathier (eds) © 2010 Taylor & Francis Group, London, ISBN 978-0-415-58654-2 A new laboratory test to evaluate the problem of clogging in mechanical tunnel driving with EPB-shields M. Feinendegen & M. Ziegler Geotechnical Engineering, RWTH Aachen University, Aachen, Germany G. Spagnoli & T. Fernández-Steeger Engineering Geology and Hydrogeology, RWTH Aachen University, Aachen, Germany H. Stanjek Clay and Interface Mineralogy, RWTH Aachen University, Aachen, Germany ABSTRACT: During mechanical tunnel driving in fine grained soil or rock the excavated material often sticks to the cutting tools or conveying system, which may cause great difficulties in its excavation and transport. In the joint research project INPROTUNNEL of RWTH Aachen University together with industrial partners this problem is faced on different scales particularly for the method of Earth Pressure Balanced (EPB) shield tunnelling. Main topic of this paper is the development of a new laboratory test to detect the adhesion/clogging propensity of a rock or soil already in the preliminary phase of a project and to quantify these as far as possible. Furthermore, a first draft of a new classification scheme for the clogging potential is presented. 1 INTRODUCTION Mechanical tunnel driving with Tunnel Boring Machines (TBM) is a world-wide popular method within tunnelling, whereby the limits of its application (diameter, length, overburden, water pressure, subsoil, etc.) are being constantly pushed ahead. Frequently bedrock zones with strongly changing strength properties have to be crossed. During the excavation and transport of the material the mechanical wear often causes a loss of strength which can even lead to a complete disintegration of the composite structure. In many cases and particularly in combination with water inflow, the excavated material sticks to the cutting tools or conveying system. This may cause great difficulties in its excavation, transport and re-use or dumping: High energy demand, blocking or breakdown of excavation tools (Fig. 1), clogging of screw or band conveyors, problems in stability during the reuse caused by lower shear resistance of the (possibly conditioned) excavation material, etc. An important factor for the performance of a tunnel construction project is the detailed knowledge of the expected geological and geotechnical conditions since the choice of suitable construction methods (face support, cutting tools, material transport, supporting and lining, etc.) depends on the resulting effects on the construction processes. Here, the problem of adhesion/clogging of excavated material to the surfaces of cutting and transportation equipment in particular is of key importance. Figure 1. Blocked roller bit. The adhesion of clays or clayey soft rocks in mechanical tunnel driving has already been investigated in several research projects (Jancsecz 1991, Wilms 1995, Thewes 1999, Burbaum 2009); nevertheless, no generally accepted (standardized) test currently exists to determine the clogging potential from a practical (tunnel) construction point of view. Main topic of this paper is the development of a standard method to detect the changing geo-technical properties and the resulting adhesion/clogging propensity of a rock or soil already in the preliminary phase of a tunnelling project and to quantify these -as far as 429 Figure 3. Laboratory tests: ball test, blade test. Figure 2. Load types for the adhesion. possible- by means of a newly developed laboratory test. 2 PRINCIPLES OF ADHESION AND CLOGGING Figure 4. Cone pull-out test: proctor pot and cone drill. Decisive factors for the occurrence of adhesion and/or clogging are the availability of water as well as swellable clay minerals, while the magnitude of adhesion changes depending on the consistency of the soil. For a characterization of the relevant mechanisms the following three criteria may be defined (Fig. 2): 1. Load type (shear – pressure/shear – tension), 2. Direction of loading (normal – tangential), 3. Ratio adhesion force – soil resistance (depending in particular on plasticity and consistency). Especially in the complex geometric surrounding of a TBM with highly different mechanical wear of the excavated material, a combination of these is relevant for the amount of soil adhering to a steel tool surface. Adherence does only then occur when there are adhesion forces acting, though a high bond stress does not always lead to extensive clogging. Actually, in the cone pull-out tests that were performed (see 3.2) the highest tensile forces were measured for a consistency of Ic = 0.85, even though the amount of material sticking to the test device was quite small. This is most probably due to the fact, that the resistance (cohesion and tensile strength) of the stiff soil is even higher than the bond strength between clay and steel, which causes a failure at the surface of the cone. In a soft soil the resisting inner forces are usually smaller than the bond strength. The resulting failure within the soil can cause sometimes extensive clogging problems. However, if the soil water content exceeds a critical value (e.g. for a pasty consistency), the surplus of free water will have the effect of a lubricating film which again considerably reduces the adherence of clay to the steel surfaces. 3 LABORATORY TESTING purpose mainly modified direct shear tests as well as separation tests, typically with steel pistons, have been carried out (Schlick 1989, Beretitsch 1992, Thewes 1999, Zimnik 2000, Burbaum 2009). However, one precondition for an exact measurement of adhesion forces is, that there is no adherence of soil to the testing device. Particularly for piston pull tests this cannot be ensured. Furthermore, separation tests do not account for the influence of the soil parameters on the adherence. Clogging does only then occur, when the resisting forces within the soil matrix are smaller than the bond stress between clay and steel surface. 3.1 Developed test layouts For a better identification and quantification of the above mentioned effects, different classification test setups (Fig. 3) were designed and a number of test series were performed. Since the results of these first experiments were not satisfactory, a new test layout was developed. The equipment and the test procedure for the so called “cone pull-out test” are shown in Figure 4 and Figure 5. The sample material is compacted in a standard proctor device, a steel cone is inserted into a pre-drilled cone shaped cavity and loaded for 10 minutes with the magnitude of the applied load between 2.3 kN/m2 and 50 kN/m2 depending on the consistency. The load is then taken off and the specimen is placed in a test stand where the cone is pulled out with a velocity of 5 mm/min. The tensile forces and the displacements are recorded. 3.2 In the relevant literature up to now most authors defined the stickiness of different fine-grained soils by a determination of the adhesion forces. For this Results from cone pull-out tests Six different clays with varying mineralogy (illite, kaolinite, smectite, etc.) were tested in a number of test series with different cones (variable inclination: 10◦ , 430 Figure 5. Cone pull-out test: application of load and pull test stand. Figure 7. Test results for different consistencies. Figure 8. Normalization. Figure 6. Test results for different cone inclinations. 31◦ , 45◦ , 58◦ , 72.6◦ ) and soil consistency (Ic = 0.20, 0.40, 0.55, 0.70, 0.85). Some exemplary results are illustrated in the following. It should be mentioned, that all curves normally represent the mean values of four tests. Only when the deviation is too large, the respective data are neglected. In general, the scatter over all test series was quite small with 79% of all results showing a deviation of less than 15% from the mean value. Figure 6 shows the progress of the vertical tensile stresses for the so called “clay 3” tested with the different cone inclinations at a consistency of Ic = 0.70. It can be seen, that with the “nearly flat” cone 0 (10◦ ) tensile forces can only be measured for displacements less than 3 mm, while with the “steep” cone 4 (72.6◦ ) they are acting in a quite large range up to 11mm. After several comparative tests, only cone 3 (58◦ ) was used furthermore, since it provided the most characteristic results for all analysed soils. In Figure 7 the respective results for different consistencies tested with cone 3, are shown. Here the stiff material (Ic = 0.85) shows quite high tensile stresses at very short ways whereas for the softer material the maximum decreases with tensile forces still acting over large displacement ways. For a better comparison of the different behaviours the tensile stress-displacement curves are then normalized by dividing all stress data by the maximum stress Figure 9. Adhering soil for Ic = 0.20 and Ic = 0.4 (0◦ = viewing direction). value and dividing all displacement data by the corresponding value. The results for the above mentioned tests on clay 3 with cone 3 at different consistencies are shown in Figure 8. It can be seen, that the areas under these normalized curves are quite different in size and shape. These functions are then integrated with the result being a dimensionless number, which is then defined as the “clogging potential”. Additionally, after each test the mass of adhering soil (Fig. 9) is determined by weighing. It is referred to as “adherence” in the following. When plotting the clogging potential, which was derived from the tensile (=bond) stresses, over the consistencies and comparing it to the measured adherence, 431 of ranges with high, medium and low clogging potential especially with respect to EPB shield tunnelling, there is still a strong need for additional tests. For these experiments, soil samples from current tunnelling projects where clogging is expected or already observed will be examined. Furthermore, within the course of the INPROTUNNEL-project the new laboratory test will help to evaluate new concepts for manipulation methods to reduce adhesion and/or clogging in the practice of EPB tunnelling. Finally their testing in especially designed large-scale tests is planned. Figure 10. Clogging potential and adherence. ACKNOWLEDGMENTS The authors would like to thank the industrial partners Herrenknecht AG, Ed. Züblin AG, Marti Tunnelbau AG and Condat Lubrifiants for their valuable contributions as well as the BMBF/DFG “Geotechnologien” program (pub. no. 1313) for the financial support, which made this research possible. REFERENCES Figure 11. Draft of a classification scheme. a good correlation can be observed from the first tests series (Fig. 10). Both the clogging potential as well as the adherence show relatively high values in a soft to stiff consistency and a decrease towards the “wet” and the “dry” side. This corresponds quite well with the experiences of Weh et al. (2009a, b, c), who carried out extensive analyses of EPB tunnel drives with clogging problems. 3.3 Classification scheme The results obtained so far are a good basis for the derivation of a classification scheme to quantify the clogging potential of different types of fine-grained soil or rock. A first draft based on the results of the experiments carried out so far as well as on the evaluation and interpretation of project data is shown in Figure 11. In the future it may also be possible to evaluate the clogging potential only by determining the adherence with a simplified test procedure avoiding to perform the quite complex and time-consuming stress and displacement measurements. 4 FUTURE PROSPECTS Beretitsch, S. 1992. Kräftespiel im System SchneidwerkzeugBoden. Institut für Maschinenwesen im Baubetrieb, Nr. 41. Universität Fredericiana, Karlsruhe. Burbaum, U. 2009. Adhäsion bindiger Böden an Werkstoffoberflächen von Tunnelvortriebsmaschinen. Institut für Angewandte Geowissenschaften. Technische Universität Darmstadt. Jancsecz, S. 1991. Definition geotechnischer Parameter für den Einsatz von Schildvortriebsmaschinen mit suspensionsgestützter Ortsbrust. In Neue Chancen aus europäischen Impulsen. STUVA-Tagung 1991, Düsseldorf. Düsseldorf: Alba-Fachverlag. Schlick, G. 1989. Adhäsion im Boden – Werkzeug – System. Institut für Maschinenwesen im Bauwesen, Nr. 39. Universität Fridericiana, Karlsruhe. Thewes, M. 1999. Adhäsion von Tonböden beim Tunnelvortrieb mit Flüssigkeitsschilden. Institut für Bodenmechanik und Grundbau, Nr. 21. Gesamthochschule Wuppertal. Weh, M. et al. 2009a. Verklebungen bei EPB-Vortrieben in wechselndem Baugrund: Eintrittsbedingungen und Gegenmaßnahmen. In Tunnel – Räume für zukunftssichere Mobilität. STUVA-Tagung 2009, Hamburg. Gütersloh: Bauverlag. Weh, M. et al. 2009b. Maschinenvortrieb in verklebungsanfälligem Baugrund, Teil 1. Tunnel 28(1): 25–36. Weh, M. et al. 2009c. Maschinenvortrieb in verklebungsanfälligem Baugrund, Teil 2. Tunnel 28(2): 18–28. Wilms, J. 1995. Zum Einfluß der Eigenschaften des Stützmediums auf das Verschleißverhalten eines Erddruckschildes. Fachgebiet Baubetrieb und Bauwirtschaft, Nr. 12. Universität-Gesamthochschule Essen. Zimnik, A.R. et al. 2000. The adherence of clay to steel surfaces. In GeoEng 2000: An International Conference on Geotechnical and Geological Engineering. Melbourne, Australia., Lancaster, Basel: Technomic Publ.. For a verification of the test procedure and a practical scaling of the classification scheme, i.e. the definition 432