International Symposium Ground Anchors

PROCEEDINGS INTERNATIONAL SYMPOSIUM GROUND ANCHORS
Limelette test field results
VOLUME 1
14 May 2008 Hotel Mtropole Brussels
BGGG - GBMS
BBRI
Proceedings of the International Symposium Ground Anchors Limelette test field results 14 May 2008, Brussels, Belgium Volume 1
TABLE OF CONTENTS
PREFACE INTRODUCTORY ADDRESS ORGANISATION WORKING GROUP GROUND ANCHORS AKNOWLEDGEMENTS CONTRIBUTIONS General Framework and Viewpoint with regard to Ground Anchors, Soil Nails and Tension Piles Prof. J. Maertens (Jan Maertens bvba & K.U. Leuven) Ground Anchors : Overview of Types, Installation Methods and Recent Trends Ir. F. De Cock (Geo.be) Full Scale Load Test Program in Limelette : Overview of the Test Campaign, Set up & General Results Ir. N. Huybrechts & Ir. O. Tomboy, (BBRI), Prof. J. Maertens (Jan Maertens bvba & K.U. Leuven) & Prof. A. Holeyman (UCL) Excavation of the Anchors : Measurements and Observations Ir. O. Tomboy & Ir. N. Huybrechts (BBRI) Integrated Analysis of the Load Test Results & Suggestions for a Harmonised Anchor Design and Test Methodology in Belgium in the Eurocode 7 Framework Ir. N. Huybrechts, Ir. M. De Vos & Ir. O. Tomboy (BBRI) & J. Maertens (Jan Maertens bvba & K.U. Leuven) Practical Experience of TUC RAIL with Ground Anchors and Micro-Piles Ir. W. Maekelberg, Ir. Q. Bollens, Ir. J. Verstraeten, Ir. F. Theys, E. De Clercq (TUC RAIL) & Prof. J. Maertens (Jan Maertens & Partners bvba & K.U. Leuven) Experience with Ground Anchors of the Federal Public Buildings Service Ir. Ph. Debacker (Federal Public Buildings Agency) Experience with Ground Anchors of the Flemish Ministry Ir. I. Marin (Ministry of Flemish Community) Design Guidelines for Non-driven Tension Piles Underneath under Water Concrete Slabs Ir. A.C. Vriend, reporter C-152 & Accon (NL) Experience in France with Ground Anchors Prof. J.P. Magnan (LCPC, F) Application of Ground Anchors, Nails and Tension Piles in Europe and Current Status of the EN 1537 Ground Anchors Dr.-Ing. W.R. Linder Chairman of CEN TC 288 Execution of Special Geotechnical Works & Brckner Grundbau (DE) Dr. Caesar Merrifield, Convener of CEN TC 288 WG 14 Anchors & Coffey Geotechnics (UK)
PREFACE
Ground anchors are commonly used in Belgium in different building and civil engineering applications (anchorage of retaining walls, submerged structures, quay walls, stabilisation of slopes, ). Unfortunately, up to now no Belgian geotechnical standards, which deal with such elements, exist. As a consequence, the project specifications of different owners integrate generally, without coherence, different design and test methodologies coming from other countries, mostly from French and German standards. In addition a lot of new anchoring techniques, for which the current geotechnical codes are in general not yet adapted, appeared the last decade on the Belgian market. Particular situations are those where not all the anchors are tested and/or pre-stressed. Finally, within the framework of the European standardisation, which is fully developing, the construction sector experienced a strong need for a better understanding of the applied anchoring techniques and appropriate design methods, in particular for the establishment of the Belgian national annex of the Eurocode 7. All the above-mentioned aspects encouraged the Belgian Building Research Institute (BBRI - CSTC - WTCB) to undertake the establishment of a uniform guidance for the execution, the design and the testing of anchoring systems in Belgium. In order to address this issue the BBRI initiated in 2004 a national research project on ground anchors. The project is guided by an inter-professional Working Group under supervision of the project partners K.U. Leuven (Prof. J. Maertens) and UCL (Prof. A. Holeyman). Financial support for the research project has been obtained from the Belgian Federal Public Service Economy and the Belgian Normalisation Institute. Backbone of the research project is the extended real scale load test campaign on approximately 50 ground anchors performed at the proof station of the BBRI in Limelette. During this symposium the various aspects of this extended load test campaign in Limelette (B) will be reviewed : soil investigation, anchor installation methods, load testing and interpretation, as well as observations related to the excavated anchors. Suggestions for an approach to design and test ground anchors in Belgium following the Eurocode 7 principles will be put forward. Moreover attention will be paid to the experience with ground anchors of three main organisations on the Belgian building market: Tuc Rail, the Federal Public Buildings Service, and the Ministry of the Flemish community. Finally three contributions from neighbouring countries (NL, F & DE) will deal with developments in anchoring techniques and design approaches in their country and/or in Europe. It is the hope of the organizers that the results of the Limelette load test campaign and the discussions held during this international symposium might be of value in the development of European geotechnical standardization. The organization committee
INTRODUCTORY ADDRESS
Research has historically been performed in Belgium by common efforts of independent bodies and individual contractors. It is only recently that the BBRI introduced the concept of more global research projects starting a few years ago with the research on screw piles, and the development of new guidelines for the design of piled foundations in the frame of the NAD of the EC7. In this context, the creation of the Belgian Federation of Foundations Contractors (ABEF) has filled in a gap as corresponding partner to the BBRI and other Federal bodies, in order to globalize the reflections and interests of the Foundation Industry, as a whole. This can be considered as a major achievement and a clear progress towards the past. Until recently, as stated by Nol Huybrechts, Belgium was characterized by the absence of codes or guidelines. This situation, which the contractors relatively easily could deal with, was judged, with some reason, dangerous and difficult by administrations and consulting offices. It was also difficultly compatible with the introduction of the European codes and norms. The BBRI has launched several projects in order to remediate this situation, inscribing its efforts in the global perspective of the Eurocodes. ABEF naturally contributed to consolidate the Contractors views, adding some entrepreneurial perspective to these efforts. ABEF was created in 1998 by 6 founding members, gathering the biggest contractors in the field of Deep Foundations. It has welcomed in the meantime 6 new members, and currently represents 80 % of the Belgian Deep Foundations market. Establishing a solid base for sound competition and reliable and safe execution of deep foundations is our global objective. Contributing to this research is part of this global objective. Other concrete steps over the last years are: o Work out a specific education program for workmen of the different members together with VDAB/FOREM structured around 3 modules corresponding to 3 levels of education. o Develop common Working Conditions in order to improve the contractual and technical quality of the relationship with the clients/general contractors. o Actively participate in the different working groups and research programs TIS National Annex of the Eurocode 7 (piles) Research Program on Screw Piles Research Program on retaining walls o Participate in the activities of the European Federation of Foundation Contractors and other international Working Groups TC288: Bored piles and Slurry Walls Contract & Qualification Working Group of the EFFC As President of ABEF, I am particularly pleased to welcome you all, to this seminar. What will be presented today is the result of years of efforts, installation and testing which the Deep Foundation Industry has contributed to both financially and intellectually. We hope you will benefit of this investment. Enjoy your stay in Brussels. ir Maurice Bottiau ABEF President BGGG/GBMS Vice-President Franki Foundations Group Belgium Group Commercial Director
ORGANIZATION
SYMPOSIUM ORGANISING COMMITTEE
Ir. G. Breyne, Chairman of the Technical Committee Structural Work of the Belgian Building Research Institute (BBRI) & SERCK Prof. A. Holeyman Catholic University of Louvain (UCL) Ir. N. Huybrechts, Belgian Building Research Institute (BBRI) Prof. C. Bauduin Besix & University of Brussels (VUB) Ir. M. Bottiau Chairman of ABEF (Belgian Member Society of EFFC) & Franki Geotechnics B Prof. J. Maertens Jan Maertens bvba & Catholic University of Leuven (K.U.Leuven) Ir. G. Simon Ministry of Equipment and Transport of the Walloon Region, Geotechnical Direction Ir. O. Tomboy Belgian Building Research Institute (BBRI)
WORKING GROUP GROUND ANCHORS

Ir. C. Bauduin (Besix & University of Brussels VUB) E. Dupont (ABEF & FUNDEX) M. Bottiau (ABEF & FRANKI GEOTECHNICS B) N. Charue (TRACTEBEL DEVELOPMENT ENGINEERING) B. Cloet (VOTQUENNE) W. Cromheecke (DENYS) F. De Cock (GEO.BE) M. De Vos, N. Huybrechts, C. Legrand, O. Tomboy, V. Whenham (BBRI) Ph. Debacker (Federal Public Buildings Agency) H. Gille (BACHY) D. Grabo (EURODRILL BELGIUM) K. Haelterman (Ministry of the Flemish Region, MOW, Geotechnical Division) E. Heirwegh (HEVI) A. Holeyman (UCL Unit Gnie Civil et Environnement) W. Hoppenbrouwers (SECO) W. Maekelberg (TUC RAIL) J. Maertens (KU.Leuven & JAN MAERTENS) I. Marin (Ministry of the Flemish Region, MOW, Steel Structural Division) J. Market, W. Smet (SMET F&C) M. Meersman, L. Smet, T. Smet (CVR ) M. Roovers (FONDEDILE BELGIUM) G. Simon, Ph. Welter (Ministry of the Walloon Region, MET, Geotechnical Division) C. Trve (CFE) G. Wellens (FONTEC) K. Van der Eecken, J. Wyckman (DYWIDAG-SYSTEMS INTERNATIONAL) S. Vandemeulebroecke (PLANET ENGINEERING) T. Vanlangenhoven (OLIVIER) J.-L. Vieslet (PROFFUND) H. Wanzeele (FREYSSINET BELGIUM)
ACKNOWLEDGEMENTS
The BBRI staff, responsible for the research program ground anchors, wishes to thank: for the financial support of the research project on Ground Anchors: - The direction of the Belgian Building Research Institute - The Belgian Federal Public Service Economy - The Belgian Standardization organization NBN - The sponsors of the test program in Limelette, in particular: ABEF (financial contribution) BACHY (financial contribution & installation on own cost) CVR (installation on own cost1) DENYS (installation on own cost) DSI (strands, Dywidrill hollow bars and hydraulic jacks) EURODRILL (financial contribution) ISCHEBECK (titan hollow bars) FONDEDILE (financial contribution) FONTEC (financial contribution) FREYSSINET (strands and hydraulic jacks) OLIVIER (financial contribution) PROFFUND (financial contribution) SMET F& C (financial contribution)
for the project elaboration: - The project partners K.U.Leuven (Prof. J. Maertens) and UCL (Prof. A. Holeyman) for their expertise during the research program - The Working Group Ground Anchors (practical elaboration of the project) - Mr. A. Bernard, Mr. R. Bonsangue and Mr. C. Verbeke from the Belgian Building Research Institute (for their technical support and contributions, and their hard and fine work during the whole project).
for the development of the inclined CPT device and the performance of the inclined CPT: - Ministry of Flanders, MOW, Geotechnics Division
for the organization of the Symposium Ground Anchors Limelette test field results: - The Symposium Organizing Committee
CONTRIBUTIONS
BBRI
BGGG GBMS
International Symposium 14 May 2008 Ground Anchors
General Framework and Viewpoint with regard to Ground Anchors, Soil Nails and Tension Piles
Prof. Ir Jan Maertens Jan Maertens & Partners bvba and KU Leuven
BBRI & BGGG-GBMS
Ground Anchors 14.05.2008
1.
INTRODUCTION
When starting the BBRI research program on ground anchors there was a lot of discussion within the Working Group concerning the types of anchors to be tested. This was mainly due to the fact that there is no clear definition and/or classification of ground anchors. So the difference between ground anchors, soil nails and tension piles is not always clear. The existing European Standards for ground anchors, micropiles and soil nailing contain a lot of valuable information. However a lot of cases exist which are not covered by the existing execution standards and/or by Eurocode 7.
2.
EUROPEAN STANDARDS EXECUTION OF SPECIAL GEOTECHNICAL WORKS:

Chapter 1 : Scope An anchor consists of an anchor head, a free anchor length and a fixed anchor length which is bonded to the ground by grout. Chapter 3 : Terms, definitions and symbols: Anchor : an installation capable of transmitting an applied tensile load to a load bearing stratum.
2.1. EN 1537 1999 : ground anchors.
Figure 1: sketch of a ground anchor details of anchor head and head protection omitted.
Chapter 9.7. acceptance test: Each working anchor shall be subjected to an acceptance test. The objectives . Comment : According to EN 1537 1999 a ground anchor has an anchor head, a free anchor length and a fixed anchor length and has to be subjected to an acceptance test.
General framework and viewpoint with regard to ground anchors, soil nails and tension piles
J. Maertens 14.05.2008 p. 3/9
BBRI & BGGG-GBMS
2.2. EN 14199 2003 : Micropiles

Chapter 1 : Scope 1.2. Micropiles are structural members to transfer actions to the ground and may contain bearing elements to transfer loads or to limit deformations. . Chapter 3 : Terms and definitions 3.1. Micropile: piles which have a small diameter (smaller than 300 mm outer diameter for bored piles and smaller than 150mm for displacement piles) and can be installed with small rigs. Chapter 9 : Micropile testing 9.3.2.3.3. unless otherwise specified for micropiles working in tension at least one static load test should be performed for the first 25 micropiles and 1 for each next 25 micropiles.
2.3. prEN 14490 : Soil nailing

Chapter 1: Scope 1.1 Soil nailing is a construction technique used to enhance / maintain the stability of a soil mass by installation of reinforcing elements (soil nails). Typical examples of soil nailing are given in Annex A.
Chapter 3 : Terms and definitions 3.15. soil nail: reinforcement element installed into the ground , usually at a sub-horizontal angle that mobilises resistance with the soil along its entire length. Chapter 9 : Supervision, testing and monitoring 9.3.2.1. The frequency and procedures for soil nail load testing should be based on a consideration of the consequence of failure, as defined in EN 1990 , EN 1997 and EN 1990. 9.3.2.2. Table 1 describes the principal types of soil nail load tests, their purpose, when they are required and actions to be taken in the event of an noncompliant test result. Annex A gives guidance on test procedures, acceptance
General framework and viewpoint with regard to ground anchors, soil nails and tension piles J. Maertens 14.05.2008 p. 4/9
BBRI & BGGG-GBMS
criteria and the equipment to be used for soil nail load tests. Table 2 suggests the frequency of soil nail load tests based on the category of geotechnical structure.
J. Maertens 14.05.2008 p. 5/9
BBRI & BGGG-GBMS
3.
GROUND ANCHORS IN BELGIAN DAILY PRACTICE

In Belgium the following anchor types and installation methods are used actually: a) Anchors with tendon elements of high strength steel (= mostly strands): - Drilling method : double or single rod system; - Injection: - double rod system : under pressure; - single rod system : under pressure or gravitary. - Testing: almost always , cfr. comment - temporary anchors = 1,3 x service load (SL) : - permanent anchors = 1,5 x SL. - Prestressing : always till 0,8 1,0 x SL. Comment : With tendon elements of high tension steel prestressing is always necessary to limit the deformations . Testing is not always possible. In some cases it is not possible to fully test ground anchors installed on Berlin walls with laggings of prefabricated concrete and installed in very soft upper layers. b) Anchors with tendon elements of low strength steel (= GEWI bars or self drilling rods) and with a free length: - Drilling method : double or single rod system. - Injection : - double rod system : under pressure - single rod system : under pressure or gravitary - sometimes post grouting through self drilling rods (after 1 or 2 hours). - Testing : - not always. - Prestressing: - not always. Comment : With tendon elements of low strength steel it is not necessary to prestress the ground anchors when certain deformations can be allowed, f.i. for retaining walls along streets, . c) Anchors with tendon elements of low strength steel and without a free length ( = mostly self drilling rods): - Drilling method : single rod system; - Injection : under pressure or gravitary; - Testing : not possible
J. Maertens 14.05.2008 p. 6/9
BBRI & BGGG-GBMS
- Prestressing : not possible. Comment : This anchor types should be considered as tension piles (= micropiles working in tension) or soil nails. The major problem with ground anchors in Belgium is that actually: - there is no clear definition of a ground anchor; - the design is done for all types of anchors with almost the same unit shaft resistances and with the same factors of safety. So in most cases the execution method and the number of tests is not taken into account in the design.
4.
PROPOSAL
In order to cover all types of ground anchors which are actually installed in Belgium the following classification is proposed : - prestressed ground anchors; - passive ground anchors; - tension (micro)piles; - soil nails. a) Prestressed ground anchors : - prestressed ground anchors may have a tendon element of high strength or low strength steel. - they always have a free length and a fixed length; - the fixed length is installed behind the so called active wedge and in this way that the necessary factor of safety is available for the overall stability; - they are always tested. When testing is not possible due to an excessive deformation of the retaining wall or the reaction system a higher value of the safety factor has to be introduced. - they are always prestressed by means of a hydraulic jack. b) Passive ground anchors: - passive ground anchors always have a tendon element of low strength steel; - they always have a free length and a fixed length; the fixed length is installed behind the so called active wedge and in this way that the necessary factor of safety is available for the overall stability; - the number of tests to be performed has to be clearly specified in the tender documents and/or in the method statement; - it has to be demonstrated that the displacement of the anchor head is smaller than the allowable displacement of the retaining structure. c) Tension Piles: - tension (micro)piles may have a tendon element of high strength or of low strength steel; - they have only a fixed length;
4.1. Classification
J. Maertens 14.05.2008 p. 7/9
BBRI & BGGG-GBMS
- the length of the piles is determined in this way that the necessary factor of safety is available for the overall stability, cfr. figure R 66-1 from EAU Recommendations.
- the number of tests to be performed has to be clearly specified in the tender documents and/or method statement. Tests on working piles can only be performed for vertical piles. For inclined piles a free length of min. 2 meters has to be provided ( = especially installed piles); - it has to be demonstrated that the displacement of the pile head is smaller than the allowable displacement of the retaining structure. d) Soil nails: - soil nails have always a tendon element of low strength steel; - they have only a fixed length; - soil nails are used as soil reinforcement and not as anchors.
4.2. Design
The ultimate skin friction over the fixed length is determined taking into account the applied drilling and injection technique. The factors of safety to be introduced are determined based on the type and number of executed tests : - preliminary tests - tests on sacrificial anchors - acceptance tests on working anchors.
4.3. Control tests

Following types of tests can be considered. a) Prestressed ground anchors: - preliminary tests on sacrificial anchors - suitability tests on sacrificial anchors
J. Maertens 14.05.2008 p. 8/9
BBRI & BGGG-GBMS
- suitability tests on working anchors?? (= not foreseen in EC7) - acceptance tests on all working anchors. b) Passive ground anchors: - preliminary tests on sacrificial anchors - suitability tests on sacrificial anchors - suitability tests on working anchors?? (= not foreseen in EC7) - acceptance tests on working anchors. c) Tension piles: - preliminary tests on sacrificial piles - suitability tests on sacrificial anchors or on selected working anchors (provided with a free length for inclined anchors) - acceptance tests on selected working anchors (provided with a free length for inclined anchors). d) Soil nails: - preliminary tests on sacrificial nails - suitability tests on sacrificial nails.
5.
CONCLUSIONS
In Belgium there is actually a lot of confusion concerning the design of ground anchors. This is mainly due to the fact that almost always the same design method is used regardless the installation method and the type and the number of tests performed. The information given in the available European Standards is not precise enough to allow a correct design of all the types of anchors that are actually installed. In order to obtain a better agreement with the actual daily practice it is necessary to extend EN 1537 with passive ground anchors and to define in EN 1997 (= EC 7) the safety factors which have to be taken in to account when acceptance tests are not performed on all working anchors. Further on it is proposed to elaborate a technical document dealing with the use of tension (micro)piles as anchoring elements.
J. Maertens 14.05.2008 p. 9/9
BBRI
BGGG GBMS
Ground Anchors : Overview of Types, Installation Methods and Recent Trends
Ir. Flor De Cock Geotechnical Expert Office Geo.be
BBRI & BGGG-GBMS
1.
INTRODUCTION
Together with the large variety in applications, ground conditions, specific project demands, local experience, etc. a wide panoply of ground anchor types has been developed over the last 40-50 years. The developments have meanly been focussing on: improving the soil-anchor interaction, allowing for higher capacities or shorter anchor lengths simplifying, improving or accelerating the installation process assuring faster or longer activity of the anchor. The scope of the execution code EN 1537:1999 is limited to anchors consisting of an anchor head, a free anchor length and a fixed anchor length which is bonded to the ground by grout, whereby ground encompasses both soil and rock. Our overview of anchor types and their installation methods is both broader and narrower, since: also mechanical non grouted anchors and expander bodies are considered rock anchor types are only included as far as their components and installation process is similar to their use in soil. Are not considered : deadman anchors, specific rock bolts.
2.
APPLICATIONS OF GROUND ANCHORS

The application of ground anchors has been dealt with in the contribution of Prof. J. Maertens to this seminar. A demonstrative presentation is given in Figure 1 (Samwoo). In Belgium as worldwide, the most common use of anchors is likely to be : Slightly inclined temporary or permanent tiebacks for retaining walls; with working loads most often in the order of 250-500 kN for temporary use (e.g. for construction pits) and of 500-1000 kN for permanent anchors (e.g. for quay-walls) Vertical anchoring to safeguard underground constructions from buoyancy.
Figure 1 : applications of ground anchors (Samwoo)
Ground anchors: overview of types, installation Methods and recent trends ir. Flor De Cock 14.05.2008
p. 3/23
BBRI & BGGG-GBMS
3.
TERMS AND DEFINITIONS

From EN 1537:1999 one maintains the following definitions and terms (Figure 2) :
Figure 2 : sketch of a (grouted) ground anchor (EN 1537:1999)
1. 2. 3. 4. 5. 6. 7. 8. 9. 10.
Anchorage point at jack during stressing Anchorage point at anchor head in service Bearing plate (or anchor plate) Load transfer black Structural element to be anchored Soil-rock Borehole Debonding sleeve Tendon Grout body (if any)
The tendon bond length Ltb (which is the bond length to be considered in the design) corresponds to the part of the tendon that is bonded directly to the grout and may be equal to or less than the fixed anchor length depending on location of the end point of the debonding sleeve.
4.
GROUND ANCHOR TYPES CLASSIFICATION

The result of our search for existing ground anchor classifications that were sufficiently covering most of the ground anchor types in general and the presentday Belgian market in particular, was quite poor. So we tried to establish ourselves a proposal for anchor classification. But what should be the most adequate basis for such a classification : The similarity in soil-mechanical behaviour and thus also to some extend the similarity of the appropriate design method and parameters? The composition (shape, components, materials, ) of the anchor?
p. 4/23
BBRI & BGGG-GBMS
The installation method? The market position and situation?
We also considered the different aspects that may distinguish the anchors : The shape and configuration of the anchor : e.g. cylindrical, with bulb(s), with anchor plate or helix, with one or several tendon elements (multi stage), The nature of the tendon or anchoring body; e.g. steel strands, bars, wires, tubes, plates, ; glass fibre, The bonding material : cement-grout, mortar, resin, The installation method : drilling tools and auxiliaries, one phase or 2-phase installation, The grouting method (if any) The terms of use : temporary or permanent, recoverable, The method(s) of corrosion protection Prestressed or not The method of anchor-soil interaction. The proposed classification is given in Figure 3, which starts from the latter aspect, related to the fixation method of the anchor body to the ground. In parallel, the grouting method (which is an essential influencing factor for the anchor capacity) as well as the term of use, are incorporated in the classification scheme.
METHOD OF FIXATION TO THE GROUND
By Friction
Tensile type anchor e.g. strand anchor threadbar anchor hollow bar anchor TMD anchor Compressive type anchor e.g. Duplex anchor e.g. plate anchor helix anchor expander body e.g. Grouted screw anchor
By ground pressure
Combined
GROUTING METHODS
* primary gravity grouting * primary pressure grouting * primary jetgrout pressure * secondary pressure grouting - global post-grouting - selective post-grouting Primary low pressure grouting during screwing-in of the anchor
(IGU-BE)
(IGU-FR) (IRS-FR)
Normally no grouting (except for formation of the expander body)
TERM OF USE
Temporary
* remaining in the ground * recoverable * partially removable (free length) * entirely removable (free length and bond length) * destructable
Permanent
* remaining in the ground
Figure 3 : ground anchor classification proposal
The various anchor-soil interaction systems are scheduled in Figure 4 : Friction anchors own their capacity from the bonding stress (shear stress) at the tendon/grout or the grout/ground interface. In tensile type anchors, the mobilisation of the bonding stress starts at the proximal end of the fixed
p. 5/23
BBRI & BGGG-GBMS
length; the distribution along the bonding length depends on the soil shearcharacteristics (shear stress as a function of displacement) and the anchor tendon elasticity. When this shear stress has reached the maximum shear strength and drops to a residual value, progressive debonding occurs as anchor load increases and the bonding is more and more transferred towards the distal end of the fixed length. To reduce this detrimental effect of the progressive debonding, the concept of multiple anchors may be used (Barley, 1997). This system involves the installation of a multiple of unit anchors into a single borehole. Each unit anchor has its own individual tendon, its own unit fixed length of borehole, and is loaded with its own unit stressing jack. The loading of all the unit anchors is carried out simultaneously by a multiple synchronised jacks which ensures that the load in all unit anchors is always identical. Compressive type anchors also use the shear stress at the grout/ground interface, but contrary to the tensile type anchors, the total anchor load is transferred to the very end of the anchor by a central steel tendon which is fixed to the end cap and the outer steel compression tube. As the elasticity of this tube and the surrounding grout mantle under compression is smaller than the elasticity of the steel tendons in tensile type anchors, the distribution of the bonding stress initiated at the anchor end is more uniform than with tensile type anchors and so the progressive failure effect is less pronounced. Plate or helix anchors meanly consist of a steel tendon which transfers the anchor load to a steel end blade. The anchor load generates compressive ground pressures at the blade/ground interface which tends to develop a slip surface in the ground. These anchors also may be called end bearing.
Tensile type anchor Single stage
Tensile type anchor Multi stage
Figure 4a and 4b : anchor-ground interaction for tensile type anchors
p. 6/23
BBRI & BGGG-GBMS
Compressive type anchor
End bearing type anchor
Figure 4c and 4d : Anchor-ground interaction for compressive and end bearing type anchors
5.
GROUND ANCHOR COMPOSITIONS INSTALLATION HEADLINES
AND
5.1. Friction anchor of the tensile type

Friction anchors of the tensile type are without any doubt the most popular anchor type worldwide. In its most traditional and generalised way, the installation uses flush-drilling and is done in 4 phases (Figure 5) 1. Phase 1 : DRILLING : rotational drilling of temporary outer casing and/or inner rod, using water-flushing to destruct the ground and to evacuate the cuttings 2. Phase 2 : PRIMARY GROUTING : after reaching the required drilling depth, one starts to inject grout (cement-grout) instead of water, and filling up the drill hole (or the temporary outer casing, if any) from bottom to top with grout while extracting the drill rod 3. Phase 3 : ANCHOR BODY INSTALLATION : the anchor body, including or not post-injection devices, is then lowered in the drill hole or in the casing, and the casing is extracted while additional cement-grout is pumped (under low or medium pressure) as needed 4. Phase 4 : (POSSIBLE) SECONDARY GROUTING (Post-grouting) : next day, a secondary pressure grouting is executed using the installed postinjection devices (grout tube, tube--manchettes TAM, )
p. 7/23
BBRI & BGGG-GBMS
Figure 5: Typical execution phases of flush-drilled friction anchors
Complementary to the above mentioned general installation process, the following details and remarks need to be added. 1. THE DRILLING Instead of water, also drilling-muds or water-bentonite may be used. In e.g. rock or hard clays, percussion or rotary-percussion drilling with air pressure is common. Alternatively, the very-high-pressure grouting method (Jetgrouting) (Figure 6) is used for drilling and grouting. After drilling of the rods using water- or groutflush (phase 1), the rods are slowly rotated and extracted, while cement-grout is injected under very high pressure through the injection nozzles nearby the drill bit (phase 2). As such, a cemented grout-body with diameters of 0.3-0.4 m or more is formed, in which the anchor body is installed (phase 3).
p. 8/23
BBRI & BGGG-GBMS
Figure 6: installation of friction anchors using the very high pressure technique (Jetgrouting)
Drilling tools can be : outer casing and inside drill rod, each with a specific recovered drill bit casing with slightly enlarged recovered drill bit drill rod with enlarged sacrificial drill bit. The different methods, their advantages and inconveniences are further discussed in 6. 2. THE PRIMARY GROUTING The primary grouting may be a low pressure grouting (+/- gravity filling or < 5 bars) or a medium pressure (> 5 bars) grouting. The latter case is indicated in the classification table and hereafter (see ) as the IGU-BE type (Injection Globale Unique Belgium). 3. THE ANCHOR BODY : tendon, sheathing, tubes, grout The anchor tendon for friction type anchors can be: (Figure 7) steel strands, typically 2 to 7 seven-wire strands, steel grade 1570/1770 or 1670/1860 N/mm, yield load of about 220 to 250 kN per strand
p. 9/23
BBRI & BGGG-GBMS
(Figure 8) threadbars, diameter in the range of 30 to 60 mm, steel grades from
500/550 N/mm up to 1080/1230 N/mm and so with yield loads in the range of 400 to 1.500 kN (occasionally) also steel wires or tubes, or glass fiber
Figure 7 : strand anchors composition, assembled strand tendon with post-grouting tube
Figure 8 : threadbar anchors composition, installed and prestressed anchors
p. 10/23
BBRI & BGGG-GBMS
For specific needs and demands of the client, the anchor tendon can be made partially or entirely removable, e.g. : For threadbars : by using a specific coupler between free length and bond length or at the anchor foot For strands : by incorporating a weaker breaking point between free length and bond length (Figure 9)
Figure 9 : partially removable strand anchor
Over the free anchor length, the tendon strands or threadbar are surrounded by a smooth sheathing, often in PVC, to allow for a friction-free extension of the free tendon length. Finally, over the free anchor length as well as the bond length, the tendon and its surrounding first grout mantle, a plastic or steel sheathing for corrosion protection may be incorporated. 4. THE SECONDARY GROUTING (Post-grouting) The secondary grouting is started when the primary grout has achieved its initial set, in general after 16-20 hours. It may be performed in 1 or 2 stages. The device for this secondary grouting can consist of a small diameter perforated grout tube (diameter 20 mm), a central plastic TAM with the strands distributed around this tube, a steel TAM with the strands or threadbar inside. Depending on whether the grouting is global (that means that the entire device is pressure-grouted over its full length) or selective (when the injection valves or manchettes are pressurised individually using a double packer injection tube), one uses the notations : IGU-FR : Injection Globale Uniforme IRS-FR : Injection Rptitive Slective.
p. 11/23
BBRI & BGGG-GBMS
5.2. Friction anchor of the tensile type self boring system

In particular for temporary anchors with low and medium working loads (150500 kN), the so-called self boring anchor system is gaining increased use. In Belgium as in Europe, the anchor components are mainly distributed by the manufacturers DSI or Ischebeck. The system takes benefit of the triple use of hollow steel threadbars provided with a sacrificial drill bit. In fact, the bar elements are used : as drill rod as injection tube for the water-, air- or grout-flush and as final anchor tendon. As such, the phases 1, 2 and 3 as mentioned above in 5.1 for the traditional installation, are combined in 1 single process step. The continuous thread with slow pitch can be cut on site to the required segments and easily extended by screwed couplers. Sheathing assures for the free anchor length (which can be recovered). (Figures 10a and 10b).
Figure 10a and 10 b: schematical drawing of self boring anchors (DSI and Ischebeck)
The outer diameter of the bars typically goes from about 30 mm up to 70 mm. With steel yield strengths of about 500 to 600 N/mm, the yield load ranges from about 200 kN up to 1200 kN. The oversized drill bit type is chosen as a function of the ground conditions. The main advantages of the system are : the high productivity
p. 12/23
BBRI & BGGG-GBMS
the flexibility and simplicity to adjust and assemble the anchor length the ability to work with small drill rigs in confined conditions
The drilling method, the type, flow and injection pressure of the flushing material and the penetration speed (including sometimes a moving up and down of the bars) shall be chosen adequately in order to stabilize the bore hole and to minimize soil disturbance and relaxation. After drilling the required anchor length, cement grout with low W/C ratio (in the order of 0.5) is pumped through the hollow bar and the drill bit, while the drill hammer continues rotating. It is claimed that the effect of this grouting in combination with the rotating and vibrating rods, is similar to a pressure grouting as in the conventional anchor types.
Photos 2a to 2f : components and installation of self boring anchors
p. 13/23
BBRI & BGGG-GBMS
5.3. Friction anchor of the compressive type (Duplex anchor)

The duplex anchor body is composed of two parts: the compression tube and the tie-rod (Figure 11). Thanks to the injected slurry, the compression element is bound to the ground over an exactly defined length. The transmission of the tensile forces at the extremity of the anchor is assured by the fact that the tie-rod stays freely extensible along its whole length and is only fixed with a screw connection at the very end of the compression element. In this way, the compression element is exclusively incited into compression and the tie-rod into tension. For temporary duplex anchors, the tie-rod can be easily recovered by simple unscrewing. In order to make the compression element destructible, the steel compression tube will not be made in one piece, but of a volley of elements joined together by point welding only. This fragmented compression tube will perfectly resist the compression during the post-tensioning, but will easily break under flexion and future foundation works or earthworks.
Figure 11 : composition of compressive type anchor
5.4. Ground pressure (end bearing) anchors

5.4.1 Plate anchors
This system involves the installation by ramming a temporary drill rod and a metal plate in the ground to the required depth. A wire rope or a steel threadbar (GEWI) is eccentrically coupled to the anchor plate. After driving to the required depth, the tensioning of the rope or the threadbar causes the eccentrically loaded plate to rotate and to be positioned perpendicular to the bore and the tensile direction. The plate surface typically ranges from about 100 cm up to 1.000 cm (exceptionally 2.500 cm), what corresponds to an equivalent diameter of about
p. 14/23
BBRI & BGGG-GBMS
120 mm to 350 mm. The structural strength of the anchor lies in the order of 100 kN up to 400 kN.
Photos 3 : plate anchors (Manta RayTM supplier JLD International)
5.4.2 Helix anchors

Helix anchors or helical anchors consist of a steel rod, provided with one or more steel screw blades of equal or increasing diameter. They are installed without any use of water-flush or grouting, by screwing mechanically with a lead-guided drill motor or occasionally with a hand-held power-drive head. The helix diameter ranges between about 150 and 350 mm. The structural strength of the anchor system goes from about 250 kN through 800 kN.
p. 15/23
BBRI & BGGG-GBMS
Photos 4 : helix anchors (ChanceTM supplier Sub-Surface Technologies)
5.4.3 Expander body anchors

The expander body anchor comprises a package of folded sheet (e.g. steel plating) which is installed in the ground by ramming or in a predrilled bore hole. After installation, the end body is expanded by grouting with concrete or mortar into a cylindrical or spherical anchoring body. In the Soilex system, strands or a threadbar are inserted through the drill pipes up to the very end of the expander body, and connected to this body by adhesion or by an end screw. Diameters of the expanded body commonly ranges from about 400 mm through 800 mm and allows to achieve capacities of 300-800 kN in dense sands.
Photos 5 : Expander body anchors (Soilex)- 5a Permanent anchor 5b Temporary anchor
p. 16/23
BBRI & BGGG-GBMS
6.
DETAILS ON THE INSTALLATION METHODS BELGIAN PRACTICE

As explained in 5.1, friction type anchors are commonly drilled using a casing. This casing prevents the drill hole from collapse, but is also essential to allow for a primary grouting under pressure. The installation of the casing is possible in different ways : - with flushing of a double system, using both outer casing and inside drill rod - with flushing of a single system, using only a casing and a slightly oversized bit - with flushing of a single system, using only a drill rod and an oversized bit - by ramming (often dry) of a casing with a sacrificial end bit 1. Double drill system (casing+rod) (also called double rod tubes/tiges) (Phase 1) The outer casing provided with a spherical drill bit and the drill rod with a drill head are simultaneously inserted by rotational flush-drilling. The diameter of the spherical drill bit is only a few mm larger than the outer diameter casing (e.g. 140mm for a 133mm casing). The diameter of the drill head on the drill rod is chosen such as to allow a free movement of the rods inside the casings. The spoil and soil cuttings are evacuated to surface by the annular space in between the casing and the rods. This allows to minimise the bore hole to the diameter of the casing-bit, and so also to minimise the ground relaxation and disturbance. (Phase 2) After reaching final depth, the bore hole is cleaned by intensive injection of clean water throughout the rods. Next, the same rods are used to inject the grout, while the rods are systematically recovered. The casings so far are kept in place. (Phase 3) The anchor body is inserted in the casing, after which the casing is extracted while additional cement-grout is pumped. 2. Single drill system with casing (also sometimes mistakenly referred to as single rod (Phase 1) In the single drill system, only the casings, provided with a spherical recovered drill bit or with a full-surface sacrificial drill head, are inserted by rotational flush-drilling. Both drill bit and drill head are only slightly larger than the casing. The soil cuttings have now to be evacuated by the outside space between casing and ground. Generally, the required flow rate and pressure are higher than with the double drill system. This may lead to an excessive ground transport; beside
6.1. Drilling methods and tools for friction type anchors
p. 17/23
BBRI & BGGG-GBMS
the advantage of having as such an increased anchor diameter, the enlarged bore hole may collapse and result in soil disturbance and subsidence. (Phase 2) After reaching final depth, the bore hole is cleaned and filled with grout by means of a flexible that is lowered in the casing. (Phase 3) The further procedure is similar to the double drill system. Compared to the double drill system, the installation procedure is easier and faster, and therefor also more economical. However, it should not be used for drilling underneath existing structures or in collapsible soils. The grout consumption is likely to be higher than in the double system, but the grout pressure is generally smaller. 3. Single drill system with drill rod (Phase 1) This drilling method differs from the former by the use of a drill rod with an oversized sacrificial drill head (e.g. drill head of 150mm for rods 90mm). One often uses a stabilising fluid, such as bentonite or cementbentonite, to prevent collapse of the bore hole. The ground spoil is evacuated by the stabilising fluid at low pressures and low flow rate. Consequently, there is no increased bore hole diameter, nor any risk for instability of the bore hole. (Phase 2 and 3) After replacement of the bentonite with cement-grout (gravity filling) and extraction of the drill rods, the anchor body with post-grouting devices or a steel TAM is inserted. (Phase 4) As the single rod system does not allow for a primary pressure grouting, a post-grouting is required. One of the methods, referred to as the IRSmethod, is described hereunder. 4. Rammed casing The casing is provided with a sacrificial end bit which is soil- and watertight. The casing is driven by high frequency percussion drilling. At final depth, the casing is internally cleaned with water, the drill bit is disconnected, after which the further procedure as mentioned in point 3. is followed. Basically, the soil is not removed but laterally displaced during the driving. This is beneficial for the grouting effect and the anchor capacity. But because of inconveniences, such as frequent material brake by hard driving or early driving stop, the method has lost interest. It only remains frequently and successfully used in hard clays and soft rocks (e.g. marl, shale, chalk) by using percussion drilling and air-flush to evacuate the cuttings.
p. 18/23
BBRI & BGGG-GBMS
6.2. Installation procedure of the IGU-BE anchor

First of all we must clarify that the Belgian IGU system (IGU-BE) is different from the IGU system as described in the French Recommandation Tirants dAncrage TA 95 . The French IGU system is the abbreviation for : Injection Globale et Unique = Global and Unique Injection The Belgian IGU-BE system has to be explained as a Selective and Unique Injection (Injection Slective et Unique) because the injection along the anchor bulb is executed stepwise as the casing is pulled back. Description of the different phases : 1. Installation of the drilling unit and beginning of the drilling operation (Photo 5.1). 2. Installing of the first inner drilling rod into the casing. The length of the rod and casing elements is generally 2 m (Photo 5.2). 3. Drilling of the first rod and casing (Photo 5.3). 4. Adding new casing with inside rod (Photo 5.4). 5. Continuation of the drilling operation with the double system casing/rod. The cuttings are flushed through the annular space between inner tubes and casings (Photo 5.5). 6. Washing out of the casings by moving the inner tubes up and down (Photo 5.6). 7. Filling up the casings with a cement grout (generally with a W/C factor of 0.5) from the bottom of the casing/drill hole to the top (Photo 5.7). 8. Removing of the inner rods (Photo 5.8). 9. Installation of the strand tendons or in other cases threadbars (Photo 5.9). Stepwise grout injection over the anchor bond length and pulling back of the casings. The usual grout pressure is between 5 and 15 bars (occasionally up to 30 bars). The injection is stopped when ground bursting occurs or when an injection volume of 50 to 100 l has been reached (Photo 5.10). 10. Pulling back of last casings over the free length, further filling up with cement grout and end of the execution (Photos 5.11 and 12).
p. 19/23
BBRI & BGGG-GBMS
p. 20/23
BBRI & BGGG-GBMS
Photos 5-1 to 5-12 : installation of the IGU-BE grout anchor
6.3. Installation procedure for the IRS-FR anchor

The primary pressure grouting as explained before in the IGU-BE procedure has some limitations : the grouting is a one-shot (Unique) only, the grouting pressure may be limited by e.g. the presence of soft layers and no further grouting is possible after recovering of the casings. These inconveniences do not exist with the IRS-FR-method (Injection Rptitive Selective). The method is applicable in combination with all 4 drilling methods prescribed in 6.1. The key factor consists in the drilling of a stable bore, which allows the insertion of a grouting device (TAM). As explained before, the stability can be guaranteed by a casing or a stabilising drill fluid. The grouting device consists of : - Over the bond length : a (steel) grouting tube in screw-coupled segments and closed with an end cap (Photo 6-1 and 6-2). Photo 6-3 shows details of the grouting openings in the tube, protected by a rubber manchet (generally 1 per m) and steel conical rings - Over the free length : a PVC-tube in glue-connected segments, connected to the grouting tube by a purpose made coupler. (Photo 6-4))
p. 21/23
BBRI & BGGG-GBMS
Photos 6-1 to 6-4 : Steel TAM for post-grouting in IRS-FR anchor type
The whole is inserted in the bore hole, filled with fresh cement grout. The casings (if any) are extracted. One allows now for the grout to set, what generally takes about 12 hours. After that, the post-grouting operation can start. Therefore, a single packer (in the case of a global post-grouting IGU-FR) or a double packer (in the case of the IRS-FR) is used. Photos 6-5 and 6-6 show the packer devices in un-inflated and inflated situation. The un-inflated double packer is inserted in the grouting tube and positioned in front of a manchette. The 2 rubber packers are pneumatically inflated by means and cement-grout is pumped in the closed space in between both packers; this fresh grout escapes through the manchette openings, breaks the initial grout mantle and permeates in the surrounding ground. This operation is repeated at every single manchette.
Photos 6-5 and 6-6 : packer for post-grouting in IRS-FR anchor type
p. 22/23
BBRI & BGGG-GBMS
6.4. Des-activation of anchors

6.4.1. Bar anchors
When the post tensioning is limited to low forces, the bar anchor can be desactivated with a torque wrench. When the post tensioning is too high, the bar anchor has first to be tensioned again to allow to unscrew the nut of the anchor. The overlength has to be long enough (min. 0,8 m). In both cases the nut must be greased and protected sufficiently to avoid corrosion and to allow the unscrewing operation. When the above procedures are not working, the bars can be cut with a cutting torch. Attention has to be paid for security reasons. Nobody may stand in front of the anchor head.
6.4.2. Strand anchors

The anchor has to be post tensioned again to allow removing the anchor wedges from the anchor plate. Again the anchor head and the over length of the strands (min. 0,8 m) had to be protected sufficiently to avoid corrosion and damaging of the strands and wedges. When the above procedure is not working, the strands can be cut with a cutting torch taking the same security measures as mentioned before.
6.4.3. Filling up and sealing of the anchor reservations

The part of the anchor into the retaining structure has to be removed with a cutting torch before filling up and sealing of the anchor reservations. Corrosion of the anchor extremities can cause water leakages through the retaining structure.
7.
REFERENCES
Barley, A.D. The single bore multiple anchor system. ICE seminar. London 1997. EN 1537:1999 Execution of special geotechnical work Ground anchors. CEN Samwoo. Company website www.swanchor.com.
p. 23/23
BBRI
BGGG GBMS
Full Scale Load Test Program in Limelette: Overview of the Test Campaign, Set up & General Results
Ir. Noel Huybrechts & Ir. Olivier Tomboy Belgian Building Research Institute (BBRI) Geotechnical & Structural Division
Prof. Ir. Jan Maertens Jan Maertens bvba & Catholic University of Leuven (KUL)
Prof. Dr. Ir. Alain Holeyman Catholic University of Louvain (UCL)
BBRI & BGGG-GBMS
1.
INTRODUCTION
The Belgian Building Research institute (BBRI) initiated in 2004 a research program on ground anchors (BBRI 2004-2008). This research program is subsidized by the Belgian Federal Public Service Economy and the Belgian standardization institute NBN. The main objective of the project is the establishment of a uniform guidance for the execution, the design and the testing of anchoring systems in Belgium. Such guidance should be complementary to the content of European standards addressing anchors. The research program was elaborated under supervision of the project partners K.U.Leuven (prof. J. Maertens) and UCL (prof. A. Holeyman) and in collaboration with the inter-professional BBRI Working Group Ground anchors, composed of all relevant parties, the anchorage contractors in particular. Within the framework of the project a major real scale load test campaign has been organized on the Limelette test field. At this occasion, different types of ground anchors were installed in different soil layers encountered in Limelette (quaternary loam, heterogeneous clayey sand and tertiary Bruxellian sand) and load tested. This contribution gives a general overview of the test campaign, the set up and its results. Where relevant, reference is made to the other contributions to this symposium.
2.
TEST CAMPAIGN SET UP

It was proposed by the project team to set up an extended load test campaign on the terrains of the proof station of the BBRI in Limelette. The site of Limelette has already been used for several real scale test campaigns in the past and is for that reason very well documented. The most extensive soil investigation campaign, comprising in-situ (CPT, PMT, SPT, DMT, ) and laboratory testing was performed on the screw piles test site in 2001, at a distance of 150 m from the anchor test site. The paper dealing with this extended soil investigation program (Van Alboom & Whenham, 2003) has been added in Annex B of Volume 2 of the proceedings of this symposium. Moreover the Limelette test field offers the following advantages: geological layers that are frequently encountered in Belgium occur, e.g. Quaternary loam and Tertiary Ledian/Bruxellian sand, the site offers time and space to perform comparative scientific load testing, which is almost never possible on real foundation sites, due to the fact that groundwater is found at large depth, it offers the possibility to excavate and inspect foundation elements easily. Figure 1 gives some typical CPT that have been performed on the ground anchor test site in Limelette.
Full scale load test program Limelette N. Huybrechts et al. 14.05.2008
p. 3/18
BBRI & BGGG-GBMS
Limelette - Test site ground anchors CPT E10-E11-E34 -21 qc (MPa) 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 0 -1 -2 -3 -4 -5 -6 -7 -8 -9 -10 -11 -12 -13 -14 -15 -16 -17 -18 -19 -20 -21 -22 -23 -24 -25 -26 -27 -28 -29 -30 -31 -32 -33 -34
Quaternary loam (silt)
Heterogeneous clayey sand Bruxellian tertiary sand
Figure 1 : Some typical CPT results on the ground anchor site in Limelette
Relative Depth (R+xxm)
22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 Friction Ratio Rf (%)
CPT E11 CPT E10 CPT E34 CPT 21

p. 4/18
BBRI & BGGG-GBMS
The following layers can be identified on Figure 1 : R - 1.50m R - 6.50 m Quaternary loam (silt) R - 6.50 m R- 9.50 m Heterogeneous clayey sand containing silex stones (only in the upper part of this layer). This layer appear in lenses as was only present at a limited part of the test site R - 9.50 m Tertiary Bruxellian sand layer; at larger depths sand stone layers occur All three layers were found interesting to install and test ground anchors but it was however decided to operate in two phases. In phase 1 a preliminary test campaign on 5 classical IGU ground anchors (Injection Globale et Unique) was performed in 2005. The aim of these preliminary tests was: - to have a first estimate of the ultimate anchor capacity in the loam (silt) layer, for which the grout body is situated at limited depth ( 4m below the soil surface) , and to verify the total stability of the ground mass above the anchor during the tests, - to have a first estimate of the anchor capacity in the dense Bruxellian sand layer, - to evaluate the feasibility of the instrumentation principles for strand anchors that were worked out in the laboratory; - to evaluate the usefulness of a supplementary extended load tests campaign on different ground anchor types at the Limelette test site. Based on the results of the preliminary tests and the discussions in the Working Group Ground Anchors, phase 2 of the test program (extended test campaign) was established: 49 more anchors of the following anchor types, significant for the techniques applied on the Belgian market, were installed and load tested (44) in 2006: - strand anchors for which the drilling is performed with casing and inner tubes (IGU or 2T) in Tertiary Bruxellian sand and in a heterogeneous clayey sand layer - strand anchors for which the drilling is performed with casing and lost oversized point (1T) in the Tertiary Bruxellian sand layer - strand anchors of the previous type but with provision to perform a two stage post-injection (1T+inj) in Tertiary Bruxellian sand and in the heterogeneous clayey sand layer - self boring hollow bar anchors from the Dywidrill type in the Quaternary loam layer and in Tertiary Bruxellian sand - self boring hollow bar anchors from the Ischebeck Titan type in Tertiary Bruxellian sand and in the heterogeneous clayey sand layer. In Figure 2, the position of the different test anchors is given. With regard to Figure 2 it should be remarked that: All the cone penetration tests were performed with an electrical E1 cone. It concerns also a series of inclined CPT. With regard to the results of these CPT, reference is made to Volume 2 of the proceedings of this symposium. The heterogeneous clayey sand layer (see figure 1) was only significantly present in the zone E5-6 to E12-13.
p. 5/18
BBRI & BGGG-GBMS
Reference is made to phase 3 of the test campaign : it concerns 8 more ground anchors: 5 self boring hollow bar anchors performed in sand with intensive percussive drilling and 3 anchors of the IRS type (Injection Rpititive et Slective) installed in the loam (silt) layer. These anchors were recently installed (March and April 2008) and will be load tested in May/June 2008. The results of the supplementary tests in phase 3 are not reported in the proceedings of this symposium. A supplementary Addendum dealing with the phase 3 load tests will be published later.
VE15 VE14 E14 VE13 E37 E13 E12 VE12 E11 VE11 E36
E16-17
E15
E16
E17
E18
E19 E20 E21 E22 E23
-1 E15
E35
3 E1 -1
2E1
E17-18
E18-19
E19-2 0
E2 1-2 2
VE10
E10 E34
E2 0 -2 1
E1 5 4- 1
E24
VE9
E9
E1
E9 10
E2 2
E1 112
E25
-2 3
13
0-1
E2
24 3E2 425
6 is 5-2 -27b E2 E26 7 -2 8 bis E26 E27-2 8 E27-2 E28-29 bis E28-29
E26
VE8
E8
E27
E33
E8-9
VE7
E7
E7-8
E6-7
E28
E29
E32 E6 VE6 E5
E5-6
E29-30
E30
E30-3 1
E4-5
E31
E34
VE5 E4 VE4 E3 VE3
-3 E2
L3 E2 VE2 E1 VE1 L4
L2 LCM2 LCM1
L1
SCM3 SCM2 S1
SCM1
L0
VE17 8 9 VE16 7 10 VE18
S2
VERTICAL CPTE
6
INCLINED CPTE
5
DIRECTION OF INCLINED CPTE

INCLINED ANCHORS (PHASE 1)
4 3 2
INCLINED ANCHORS (PHASE 2) VERTICAL ANCHOR (PHASE 2) INCLINED ANCHORS (PHASE 3) VERTICAL ANCHOR (PHASE 3)
Figure 2 : Ground anchor test site Limelette : position of the ground anchors phase 1 (2005); Phase 2 (2006-2007); Phase 3 (2008) and position of the inclined and vertical CPTE
With the extensive load test program on anchors in Limelette it has been envisaged to obtain more information with regard to the several aspects and parameters that are summarised below:
p. 6/18
BBRI & BGGG-GBMS
Ultimate anchor capacity of the different anchorage systems in loam (silt), heterogeneous clayey sand and tertiary Bruxellian sand. Influence of the installation method on the anchor capacity. Influence of the inclination on the anchor capacity. Influence of the drill diameter on the anchor capacity. Friction losses in the free length (Lfree) of the anchor, and in particular the effectiveness of the free length (Lfree) of self boring hollow bar anchors. A comparison of the (French) maintained load test procedure TM3 with the (German) cyclic test procedure TM1 (see EN1537 and Pr EN ISO 22477-5). The (non-) lineair increase of anchor capacity with Lfixed, and the way the friction resistance is mobilised over Lfixed Performance of the double corrosion protection system(s). The influence of the absence of spacers between the strands. Relation between anchor dimensions, installation methodology, and anchor capacity.
3.
ANCHOR INSTALLATION
Different types of ground anchors, significant for the systems applied on the Belgian market, were installed on the Limelette test site. The different systems are briefly illustrated in the Figures 3 to 5. For a detailed report of the installation, and the observations and monitoring data during installation, reference is made to Volume 2 of the proceedings of this symposium. For a more general overview of ground anchors types, reference is made to the contribution of F. De Cock (2008) to this volume.
Figure 3 : Installation of IGU anchors drilling with casing water flushing via inner tubes stepwise grout injection procedure of Lfixed
p. 7/18
BBRI & BGGG-GBMS
Figure 4 : Installation of 1T anchors drilling with casing only at its end provided with an oversized lost point flushing with water
Figure 5 : Installation of SA anchors drilling with hollow bar at its end provided with an oversized drill bit flushing with water- cement mix
p. 8/18
BBRI & BGGG-GBMS
With regard to the anchor installation in Limelette, it should however be remarked that: The technique called IGU (Injection Globale et Unique) as applied in Belgium differs from the IGU technique in France, where a global injection is performed starting from a tube manchettes (TAM), cfr. AFNOR (1992) and De Cock (2008). Some of the IGU anchors have on purpose been gravity filled with grout; these anchors are symbolized with 2T-grav. For the IGU and 1T anchors in Limelette the tendon existed out of strands of pre-stressed steel. Almost all these tendons have been instrumented with strain gauges (cfr. Volume 2). Some of the 1T anchors were executed with reduced dimensions of the casing and a lost oversized point; these anchors are symbolized with 1Tred. Some other 1T anchors were provided with a 2-stage post-grouting system and are symbolized with 1T+inj. For the SA anchors, different drill bit diameters have been applied. In order to realize a free length, plastic tubing was provided between the coupling sleeves. For the SA anchors, two variants have been executed: one variant without post-grouting (SA-xx-Dy) and the other variant with post grouting (SA-xxIs). As these anchors were installed with limited (SA-xx-Dy) or no (SA-xxIs) percussion during drilling, it has recently been decided to install in Limelette some supplementary SA anchors with intensive percussive drilling (phase 3). The SA anchors were rinsed after installation. In this way it was possible to install a retrievable extensometer system, developed by BBRI, in the hollow bars upon testing. According to the Belgian practice for temporary anchors, the tendons were not provided with centralizers. Some of the strand anchors were even on purpose installed without spacers between the strands. As mentioned before, recently three supplementary IRS anchors have been installed in the loam (silt) layer (phase 3). Via this installation technique the tendons have been fixed to ground by means of 2 post injection stages via a TAM (tube manchettes)
4.
ANCHOR LOAD TESTS

The 5 preliminary anchors of phase 1 have been installed on 30 and 31 May 2005, and load tested between June 27th and July 5th 2005. The 44 anchors of phase 2 have been installed between 3 May and 2 June 2006, and the load tests have been performed in the period between June 27th and December 19th 2006. Two test methods were applied, it concerns - a maintained load test procedure (Test Method 3 or TM3) - a cyclic test procedure (Test Method 1 or TM1) The applied test procedures have been based on the requirements in the informative annex of EN 1537:1999 Execution of special geotechnical works Ground anchors and on the specifications of the anchor test standard PrEN ISO 22477-5. Both procedures are illustrated in Figures 6 and 7.
p. 9/18
BBRI & BGGG-GBMS
Load procedure : TM3 110 100 90 80 load [% Pp] 70 60 50 40 30 20 10 0 0 50 100 150 200 250 300 Time [min] 350 400 450 500 550 600
# step
% Pp Period of 0 60 60 60 60 60 observation [min](*) (*) Reduced period of 30 to 45 min has been generally applied for Steps 1 & 2
Datum load 10
Step 1 20
Load level [% Pp] Step Step 2 3 40 50
Step 4 60
Step 5 70
Step 6 80 60
Step 7 90 60
Step 8 100 60
Figure 6 : General test scheme Test Method 3 of the prEN ISO 22477-5: MLT-procedure
Load procedure : TM1 110 100 90 80 load [% Pp] 70 60 50 40 30 20 10 0 0 20 40 60 80 100 120 Time [min]
Load level [% Pp] Cycle 1 10 Minimum period of observation for adopted TM1 [min]
140
160
180
200
220
240
Cycle Cycle Cycle Cycle Cycle Cycle Cycle 2 3 4 5 6 7 8 10 10 10 10 10 10 10 1 20 40 50 60 70 80 90 1 20 40 50 60 70 80 90 100 15 (60 or 180)(*) 20 40 50 60 70 80 90 1 10 10 10 10 10 10 10 10 1 (*) Cycle 8 -Extended period of observation for creep-displacement monitoring at Pp (60 min in noncohesive soil; 180 min in cohesive soil)
Figure 7 : General test scheme Test Method 1 based on prEN ISO 22477-5: cyclic procedure
p. 10/18
BBRI & BGGG-GBMS
For the maintained load test procedure TM3 it was the aim to obtain anchor failure in 8 load steps, starting from a datum load Pa = Min[50kN;10%Pp], with Pp the maximum estimated test load. In reality anchor failures have been obtained after 5 to 15 load steps. This corresponds with 70 to 150 % of the estimated maximum load. In general the first two load steps have been reduced to 30 or 45 minutes For the cyclic load tests, some slight changes have been integrated in the test procedures TM1 of the PrEN ISO 22477-5 (6 cycles until maximum load). The magnitude of the load step Q was determined, in order to obtain anchor failure after 8 cycles. The duration of the load steps has been based on the evaluation of the creep (). Hereby the load was maintained until was constant. For the cyclic tested anchors in Limelette, anchor failure was in general obtained after 7 to 9 cycles. Two tests have on purpose been performed with a very low number of cycles (5) and a very high number of cycles (14), in order to evaluate the possible effect of the number of cycles on the ultimate anchor load. Figure 8 gives some illustrations from the load test set up for the inclined and the vertical anchors.
Figure 8 : Illustration of the anchor load test setup
The load was applied by means of hollow hydraulic jacks and transmitted to the reaction wall (inclined anchors) or bearing plates on the ground surface (vertical anchors). The force was regulated by means of a high precision PLC-controlled 700 bar hydraulic central of the BBRI. In this way it is possible to ensure a
p. 11/18
BBRI & BGGG-GBMS
continue regulation of the applied load in a very accurate way (0.1% of the maximum jack capacity). Measurements of the hydraulic jack pressure (digital manometer on hydraulic central), the applied force (dynamometer), the anchor head displacement (displacement transducers), and the deformation measurements of the tendon (strain gauges or extensometer) were automatically and continuously recorded (each 10 seconds). The displacement of the reference system was regularly controlled by means of optical measurements. The results of the load test on each individual anchor is reported in Volume 2 of the proceedings of this symposium. Figures 9 to 16 illustrate the reported test results on one anchor. With regard to the execution of the tests it should be remarked that: Only with load cells with a sufficient maximum load capacity with regard to the maximum test load and with a height of at least 2 times the load cell diameter, reliable load measurements were obtained for inclined anchor testing. For some inclined anchors, flat load cells were used, resulting in unstable measurements. The load Pcorrelation expressed in the figures below, is based on a correlation analysis between the load cell measurements Pload cell and the hydraulic pressure readings, and represents in our opinon the most reliable assessment of the applied load. Most of the anchors failed at the grout-ground interface. However for some anchors, when loaded up to the maximum test load Pmax, determined by the steel limit, geomechanical failure did not yet occur. Some of those anchors were submitted to a cyclic loading between datum load Pa and Pmax. In the most cases a limited number of cycles was needed to obtain geomechanical failure of these anchors. In one case mechanical rupture of some steel wires in the strands occurred under an applied load of 80% of the characteristic load capacity of the tendon. With regard to the creep curve : the creep values have for each load step been deduced from the anchor head displacement measurements (s) with time on a logarithmic scale (see Figure 15) and in general according to the definitions in prEN ISO 22477-5. It has been found that it is not always evident to obtain a nice and smooth creep curve, even for load tests applied under laboratory conditions as it was the case in Limelette. The creep values are very sensible to various factors, the precision with which the applied load can be kept constant on first instance. To anticipate on this, it is necessary to perform continuous measurements of load, hydraulic pressure and anchor head displacement in order to detect, explain and possibly correct irregularities in the creep curve. Finally it should be remarked that for the cyclic test method TM 1 in general creep curves of better quality were obtained.
p. 12/18
BBRI & BGGG-GBMS
Figure 9 : Identification table of anchor E13-14
Limelette - Test site ground anchors qc (MPa) 20 24
0 0 -1 -2 -3 -4 -5 -6 -7 -8 -9 -10 -11 -12 -13 -14 -15 -16 -17 -18
12
16
28
32
36
40
44
CPT E13 CPT E14 CPT E36
Ext 5 Ext 6
Figure 10 : Position of anchor E13-14 with regard to nearby CPT, position of instrumentation
Ext 1 Ext 2 Ext 4 Ext 12 Ext 11
p. 13/18
BBRI & BGGG-GBMS
BBRI Research 'Ground Anchors' - Site Limelette Anchor E13-14 (SA - sand) 1600 1400 Load P (kN) - Pressure (bar) 1200 1000 800 600 400 200 0 0:00 P_load cell (kN) P_correlation (kN) Pressure (bar) s (mm) 160 140 120 100 80 60 40 20 0 10:00 11:00 12:00 Anchor head displacement s (mm)
1:00
2:00
3:00
4:00
5:00
6:00
7:00
8:00
9:00
Time (hours:minutes)
Figure 11 : Time data (each 10 sec.) of Pload cell, hydraulic pressure, and the anchor head displacement s(mm)
BBRI Research 'Ground Anchors' - Site Limelette Anchor E13-14 (SA - sand)
3000 Deformation measurements Lfixed (strain) 2700 2400 2100 1800 1500 1200 900 600 300 0 0:00 EXT 11 EXT 12 EXT 4 EXT 2 EXT 1 EXT 6 EXT 5 Theor.
1:00
2:00
3:00
4:00
5:00
6:00
7:00
8:00
9:00
10:00
11:00
12:00
Time (hours:minutes)
Figure 12 : Time data (each 10 sec.) of the tendon deformation measurements
p. 14/18
BBRI & BGGG-GBMS
BBRI Research 'Ground Anchors' - Site Limelette: Anchor E13-14 (SA-sand) Load Pcorrelation (kN) 0 0 20 40 60 (6') 80 100 120 140 200 400 600 800 1000 1200 1400 1600
Anchor head displacement s (mm)
(a) (b)
(a) sel;Lfree,tot (b) sel;Lfree,tot+1/2Lfixed
After 6' in step 8 (918 kN), s accelerated. The anchor has been unloaded
Figure 13 : Pcorrelation(kN) s (mm) curve (data each 10 seconds)

BBRI Research 'Ground Anchors' - Site Limelette: Anchor E13-14 (SA-sand) Load Pcorrelation (kN) 0 0 20 40 60 80 100 120 140 (6') 200 400 600 800 1000 1200 1400 1600
Anchor head displacement s (mm)
(a) (b)
(a) sel;Lfree,tot (b) sel;Lfree,tot+1/2Lfixed
After 6' in step 8 (918 kN), s accelerated. The anchor has been unloaded
Figure 14 : Pcorrelation (kN) s (mm) (step data)
p. 15/18
BBRI & BGGG-GBMS
BBRI Research Ground Anchors - Site Limelette Anchor: E13-14 Time (logarithmic scale) 1 0 step 1 : 178 kN step 2 : 353 kN step 3 : 469 kN 20 step 4 : 560 kN step 5 : 650 kN step 6 : 740 kN step 7 : 831 kN Anchor head displacement (mm) 40 step 8 : 918 kN 10 100
60
80
100
120
Figure 15 : log t s diagram (time t on a logarithmic scale)
p. 16/18
BBRI & BGGG-GBMS
BBRI Research 'Ground Anchors' - Site Limelette: Anchor E13-14 (SA-sand) 5 5' --> 60' 30' --> 60' 4 (6') 3 (mm) Pmax
0 0 200 400 600 800 Load Pcorrelation (kN) 1000 1200 1400 1600
Figure 16 : Creep curve Pcorrelated - , values deduced from log t s diagram
5.
CONCLUSIONS
This contribution has given a general overview of the set up of a real scale load test campaign on different ground anchor types in Limelette. It concerns phase 1 (5 preliminary test anchors) and phase 2 (extended test campaign on 44 anchors) in particular. It has been shown that an overall quality control for the project planning, the anchor installation, and the load testing itself has been assured in order to obtain test results of high quality. For more details reference is made to Volume 2 of the proceedings of this symposium. Furthermore 29 of the tested anchors have been excavated. A summary of the observations and measurements of the real anchor dimensions is given in the contribution of Tomboy et al. (2008) in this Volume. Volume 2 contains all detailed measurements. The interpretation of the test results, taking into account a.o. the observations on the excavated anchors, is addressed in the contribution of Huybrechts et al (2008) to this symposium. Finally, a 3rd phase of the test campaign in Limelette has been activated. 8 more ground anchors have recently been installed (March and April 2008) and will be load tested in May/June 2008. It concerns 5 self boring hollow bar anchors performed in sand with intensive percussive drilling and 3 anchors of the IRS type installed in the loam (silt) layer. A supplementary Addendum to these Volumes, dealing with the phase 3 load tests, will be published later.
p. 17/18
BBRI & BGGG-GBMS
6.
REFERENCES
AFNOR 1992, Fondations profondes pour le btiment, NFP 11-212, DTU 13.2 CEN, 1999. EN1537 - Execution of special geotechnical works Ground anchors CEN, 2004. Pr EN ISO 22477-5 Geotechnical investigation and testing Testing of geotechnical structures Part 5: Testing of anchorages De Cock, F. 2008. Ground Anchors : overview of types, installation methods and recent trends, Proceedings of the international symposium on ground anchors, May 14th 2008, Brussels Huybrechts, N., De Vos, M., Tomboy, O. & Maertens, J. 2008. Integrated analysis of the anchor load test results in Limelette and suggestions for a harmonized anchor design and test methodology in Belgium in a EC7 framework, Proceedings of the international symposium on ground anchors, May 14th 2008, Brussels Ministre de lEquipement, du Logement, et des Transports, 1993. Fascicule 62-V, Rgles techniques de conception et de calcul des fondations des ouvrages de gnie civil, cahier des clauses techniques gnrales applicables aux marchs publics se travaux (France) Tomboy, O. & Huybrechts, N. 2008. Excavation of the ground Anchors: measurements and observations, Proceedings of the international symposium on ground anchors, May 14th 2008, Brussels Van Alboom, G. & Whenham, V. 2003. Soil investigation campaign at Limelette (Belgium):Results, Proceedings of the 2nd Symposium on screw piles, May 7th 2003, Brussels PrEN ISO 22477-5, Geotechnical investigation and testing Testing of geotechnical structures Part 5: Testing of anchorages EN1537, Execution of special geotechnical works Ground anchors
p. 18/18
BBRI
BGGG GBMS
Excavation of the Anchors Measurements and Observations
Ir. Olivier Tomboy & Ir. Noel Huybrechts Belgian Building Research Institute (BBRI) Geotechnical & Structural Division
BBRI & BGGG-GBMS
1.
INTRODUCTION
Within the framework of the BBRI Research program on ground anchors and more specifically at the location of the extended test campaign in Limelette it was decided to excavate a number of the tested anchors. The main reason for this decision was: - to measure the real dimensions and shape of the different anchor types, - to determine the real position (inclination) of the anchors, - to observe the effects of post-grouting operations on the shape and the dimension of the anchor, - to look at the surface roughness, - to observe fissuring patterns. All this information has been analysed in detail and an overview is given in this contribution. For more details about the results of the investigation, reference is made to Volume 2 of the proceedings of this symposium. The further objective of the excavation is to link this detailed information with: - the anchor installation procedures and the observed installation parameters, - the nominal values of anchor material and/or drill tools applied on the site, - the theoretical imposed position of the anchors, - the results of the anchor load tests. This integrated analysis is summarized in the contribution of Huybrechts et al. (2008a) to this volume,
2.
DATA COLLECTED ON THE LIMELETTE SITE

As it was practically and budgetary not possible to excavate all the tested anchors (50), a zone containing at least one anchor representative for each installation technique applied on the Limelette site has been selected. This zone is illustrated in Figure 1, and contains 13 vertical and 16 inclined anchors. Details about the excavated anchor (types and nominal dimensions) are given in the Table 1. The symbols used to indicate the different anchor types as well as details about the different installation procedures are given in Huybrechts et al. (2008b). With regard to the nominal dimensions, it should be mentioned that they correspond with the dimensions of the boring tools (casing diameter, drill bit diameter, and oversized lost point diameter)
2.1. Selected anchors on the Limelette site
Excavation of the anchors O. Tomboy & N. Huybrechts 14.05.2008
p. 3/26
BBRI & BGGG-GBMS
VE15 VE14 E14 VE13 E37 E13 E12 VE12

E16-17
E15
E16
E17
E18
E19 E20 E21 E22 E23
E36
E11 VE11 E35

E1 14 32 E1
E 15 -16
E17-18
E18-19
E19-2
VE10
E2 1-2 2
E10 E34
E1 1-
E2 0 -2 1
5 4- 1 E1
E24
0
3 -1
E25
VE9
E9
E1 0-1 1
E910
12
E26
25 4E2 6 5-2 2 E -27 E26
VE8
E8
E27
E33
E 89
E7 VE7
E7-8
E6-7
E28
8
E27-2
E28-29
E29-30
E29
E32 E6 VE6 E5 VE5

E5-6
E30
E30-3 1
E4-5
E31
4 E3-
E4 VE4 E3 VE3
E2
-3
L3 E2 VE2 E1 VE1 L4
L2 LCM2 LCM1
L1
SCM3 SCM2 S1 S2
SCM1
L0
CPTE
VERTICAL ANCHOR
Figure 1: Selected zone for the excavation Table 1: Excavated anchors: anchor type, nominal dimensions INCLINED ANCHORS VERTICAL ANCHORS
Anchor n
E02-03 E03-04 E04-05 E05-06 E06-07 E07-08 E08-09 E09-10 E10-11 E11-12 E12-13 E13-14 E14-15 E15-16 E16-17 E17-18
Anchor type
IGU SA-175-Is IGU IGU SA-175-Is IGU SA-130-Is IGU 1T + inj. 1T + inj. IGU SA-130-Is 1T red. IGU IGU 1T
nom [mm]
140 175 140 140 175 140 130 140 180 180 140 130 150 140 140 180
Anchor n
VE3 VE4 VE5 VE6 VE7 VE8 VE9 VE10 VE11 VE12 VE13 VE14 VE15
Anchor type
2T-grav. IGU 1T SA-150-Dy SA-150-Is 1T SA-150-Dy 2T-grav. IGU IGU IGU 1T + inj. 1T + inj.
nom [mm]
140 140 180 150 150 180 150 140 140 140 140 180 180
p. 4/26
BBRI & BGGG-GBMS
2.2. Execution of the excavation

The excavation was realised in the period between 4 and 19 June 2007. A total soil volume of about 10 000m has been extracted and stored at short distance. Figure 2 shows the finalised excavation.
Figure 2: Finalised excavation test site ground anchors Limelette
Particular remarks to be made with regard to the excavation works are the following: - the excavation is realised with a crane of CAT 330C type (Figure 3), - permanent standby of BBRI staff was provided in order to guide the crane operator, and to avoid damage of the ground anchors (Figure 3), - the soil in the direct neighbourhood of the anchors was manually removed, - observations detected during the excavation works were noted by the site staff. Due to the limited bending stiffness of the vertical anchors and due to the effect of load testing on the anchor grout, a lot of material in the free length of the vertical anchors was lost during the excavation works. In order to limit the loss of information with regard to Lfree of the vertical anchors, it was decided to perform as much as possible observations and measurements during the excavation works and to cut them off at regular levels. This implied that excavation process was a very delicate operation.
p. 5/26
BBRI & BGGG-GBMS
Figure 3: Overview of the excavation works
2.3. Methodology and presentation of the results

Thanks to the permanent standby of BBRI staff during the excavation works, observations detected have been immediately noted. Moreover, complete description of the excavated anchors have been realised after the end of the excavation (few days later). Besides the observations, a systematic methodology has been adopted to analyse each excavated anchors: a photographic report with a picture each 50 cm has been created; a description of the soil encountered around the anchor and a description of the fissures observed have been made, including general comments; the dimension (diameter) of the excavated anchors has been measured. For the inclined anchors the diameter Dmeas has been measured in one plane. For the vertical anchors Dmeas has been deduced from perimeter measurements. Measurements were made each 20 cm or less in case of irregular shape; the real position of the inclined anchor has been measured, either with a total station (for all the excavated anchors) or with a pressure sensor (only for hollow bar anchors) and has been compared to the theoretical one For each excavated anchor a detailed data fiche has been established. This datafiche contains all the above-mentioned information. An example is given in Figures 4 to 6. Each data fiche is incorporated in the Annex F of the Volume 2 of the proceedings of the symposium.
p. 6/26
BBRI & BGGG-GBMS
BBRI Research Ground Anchors Site Limelette. Anchor E13-14 ( SA-130-Is Sand)
0 0.5 m
0.5 1.0 m
1.0 1.5 m
1.5 2.0 m
2.0 2.5 m
2.5 3.0 m
3.0 3.5 m
3.5 4.0 m
4.0 4.5 m
4.5 5.0 m
5.0 5.5 m
5.5 6.0 m
L Fixed
L Free
p. 7/26
BBRI & BGGG-GBMS
6.0 6.5 m
6.5 7.0 m
7.0 7.5 m
7.5 8.0 m
8.0 8.5 m
8.5 9.0 m
9.0 9.5 m
9.5 10.0 m
Figure 4: Example of detailed report on excavated anchors: photographic report
p. 8/26
BBRI & BGGG-GBMS
Figure 5: Example of detailed report on excavated anchors: description of the encountered soil, fissuring patterns and general remarks with regard to the observations
p. 9/26
BBRI & BGGG-GBMS
BBRI Research 'Ground anchors' Anchor E13-14 (Sand SA-130-Is)
Limelette - Test site ground anchors
14 -5.8 -6.3 -6.8 -7.3 -7.8 -8.3 -8.8 -9.3

free-fixed
13
D meas. D nom
12
16
qc (MPa) 20 24 28 32 36 40
44
12
11
10
CPT E13 CPT E14 CPT E36 total station pressure sensor CSTC pressure sensor VLAAO
5 -9.8 -10.3 -10.8 -11.3 -11.8
Length from bottom (m)
Relative depth (R+xxm)
-1 -200
-100
100
-12.3 200
0 -1 -2 -3 -4 -5 -6 -7 -8 -9 -10 -11 -12 -13 -14 -15 -16 -17 -18
Equivalent radius (mm)
Figure 6: Example of detailed report on excavated anchors: measured versus nominal anchor diameter (left) - measured versus theoretical anchor position (right)
p. 10/26
BBRI & BGGG-GBMS
2.4. General observation and measurements

2.4.1. Grout degradation/fissuring
In general, most of the anchors were visually undamaged just after excavation. However, (sudden) cracks arised just after excavation, and progressive degradation was observed mainly for inclined anchors with time due to a.o. thermal effects, residual stresses, grout shrinkage due to contact with air, . For the vertical anchors, generalised cracks in de grout along the 2 first meters beneath soil surface have been observed (Figure 7). The cracks can be attributed to the static load test itself. Due to that generalised degradation little information on anchor diameter is available close to the ground surface. Such degradation was by far more important for the vertical self boring hollow bar anchors (SA) for which small information is consequently available in Lfree. For strands anchors (IGU, 1T, and 1T+inj) transversal (micro-)fissures at regular distance (15 20 cm) are remarked in several zones of Lfixed. In general the transition zone Lfixed/Lfree is characterized by transversal and longitudinal grout cracks, but large differences between individual anchors have been reported. For all types of anchors with strand tendons, it has been observed that for several anchors at the physical end of the tendon a (unreinforced) block of grout is completely torn off. For SA systems no fissuring is in general observed in Lfixed. This is probably due to the fact that the hollow bars exist out of construction steel that, compared with prestressed steel, has a considerably lower yield strength. Consequently the hollow bar tendons are subjected to smaller steel deformation during anchor loading. Another reason might be the presence of the thread (with small pitch) on the outside of the hollow bar, which might lead to a micro-fissuring pattern with very small interdistance. In the transition zone Lfixed-Lfree it appeared that the grout was highly fragmented for the vertical SA while only some longitudinal fissuring is noticed for the inclined SA.
Figure 7: Overview of the grout degradation of vertical anchors near the ground surface (left) Typical fissuring patterns in Lfixed for strand anchors (right)
p. 11/26
BBRI & BGGG-GBMS
2.4.2. Grout aspect

From a general point of view, the grout aspect is found to be dependent on both the installation technique and the encountered soil: In the tertiary sand (Lfixed) - For the IGU systems, the surface shows traces due to the withdrawal procedure of the casing (Figure 8). A regular shape has also been observed for the 1T and SA anchors (Figure 8). Conversely, the 1T-red presented a shape by far more irregular than 1T. For the SA finally, no effect (enlargement) due to the post-injection operation was observed at the bottom of the anchors. The effects of post-grouting 1T+inj can be clearly observed; a secondary grout mantle is present around the anchor body up to 1 m beneath and above the position of the manchettes, the grout surface is irregular and rough, and grout vanes on the surface can be observed (Figure 8). In the heterogeneous layer At the position of the intermediate heterogeneous clayey sand with gravel, ground was difficult to remove from the grout, and gravel is mixed to the outer grout. An irregular shape is observed mainly for SA and 1T. As observed in the sand layer, the effects of post-grouting 1T+inj can be clearly noticed (secondary grout mantle, irregular and rough grout surface, and grout vanes). In the (weaker) silt layer - The erosion induced during drilling involved local large diameters, in particular for 1T and 1T+inj. Unfortunately, little information is available due to (a) the limited anchors excavated length in this layer (inclined anchors) or (b) the high degradation of the grout in this layer (vertical anchors) Finally it should be mentioned that in some zones of the SA anchors free length Lfree some soil inclusions have been observed (Figure 9). This is probably due to the less intensive flushing in Lfree during drilling.
Figure 8. Grout surface for IGU system (left), SA system (centre), and 1T + inj. System (right)
p. 12/26
BBRI & BGGG-GBMS
Figure 9. Example of local soil inclusions in Lfree for SA
2.4.3. Grout cover

Due to the absence of centralisers, it is not surprising that the grout cover of the strands appeared to be locally very limited. Measurements on inclined anchors revealed that for IGU and 1T systems the strands are always positioned at the bottom of the cross-section of the grout body. Same observations and measurements have been made on self boring hollow bar anchors. Beneath the tendon, the grout cover is generally inferior to 20 mm, locally even less than 10 mm (Figure 10). For vertical anchors, the grout cover of the tendon appeared also to be locally very limited (less than 10 mm Figure 10).
Figure 10. Example of decentred tendons for inclined and vertical anchors
2.4.4. Anchor inclination

For all the IGU anchors and for most of the 1T (+inj) anchors the real anchor position deviates little from the theoretical anchor position. For some 1T anchors however, a strong local deviation of the inclination is noticed in the heterogeneous layer (when the anchor passes through zone containing silex stones). The SA tend to deviate more from the theoretical anchor position. In addition some SA present a local deviation in their alignment when they enter in the dense sand layer. Anyway, with regard to the inclination of the anchors and the real anchor position that has been measured, it can be concluded that all anchors satisfy largely the limits that are put forward in the EN 1537 which specifies that
p. 13/26
BBRI & BGGG-GBMS
overall borehole deviation tolerance should be limited to 1/30 of the anchor length.
2.4.5. Anchor diameter

Figures 11 to 18 give an overview of the measured versus nominal diameter for the different types of excavated anchors. In the same figures the average value of the ratio between Dmeas and Dnom along the fixed length Lfixed is given as well. In general, the diameter of the IGU & 2T-grav type anchors is relatively constant and coincides well with the nominal diameter, except in sand where local thickenings manifest, probably due to the drilling procedure, e.g. the more intensive rinsing (moving up & down inner tubes) before a new 2-m length casing element is added. Those thickenings imply a Dmeas which is somewhat higher than Dnom. For the 1T anchors, the diameter remains relatively constant with respect to the depth and a somewhat higher diameter (10%) than the nominal one is observed in the sand layer. However, the average Dmeas of the 1T-red anchor in sand, for which the same drilling procedure as that one for 1T has been applied but with a reduced casing diameter of 133 mm (instead of 152 mm) and a lost drill point diameter of 150 mm (instead of 180 mm), is significantly higher (40%) than Dnom. From the measurements and observations, it can be concluded that the real anchor diameter and the obtained anchor shape depends strongly on the flow rate of the drill fluid and in particular the ratio between flow rate and drilling diameter. It is obvious that during drilling those parameters determine in a direct way the (impact) velocity of the drill fluid on the surrounding soil, and are a determining factor for the amount of soil that is eroded. For the 1T+inj system that was provided with two post-grouting tubes, the effect of the post-grouting operations has clearly been observed. The secondary grout mantel increases significantly the anchor diameter with regard to Dnom. Along the zone of Lfixed where the manchettes were present, one can observe an increase of the diameter with 16 % with regard to Dmeas observed for the 1T system (in sand and in heterogeneous layer) It is remarkably that for inclined anchors the largest effect of the post-injection can be observed 0.5 to 1 m beneath the lowest injection point (manchette). After investigation it has been found that the end of the post-injection tube was still closed. Similarly to the 1T system, the diameter of SA is found to be regular in Lfixed and coincides in general well with the nominal diameter. For SA, no significant increase with regard to the drill bit diameter has been observed. Based on the argumentation here above, it can be concluded that the flow rate (and/or the pressure) of the drill fluid, cement grout in this case, was probably too low in order to realise an increase of the real anchor diameter with regard to Dnom. Out of the measurements no significant difference was found between the Dywidrill system installed with percussion (occasionally) and the Ischebeck system installed without percussion.
p. 14/26
BBRI & BGGG-GBMS
IGU 25 - SAND Dnom = 140 mm (=casing diameter)

BBRI Research 'Ground anchors' Anchor E04-05 (Sand IGU-140) 14 14 BBRI Research 'Ground anchors' Anchor E15-16 (Sand IGU-140)
BBRI Research 'Ground anchors' Anchor E02-03 (Sand IGU-140)
BBRI Research 'Ground anchors' Anchor E16-17 (Sand IGU-140) 14
14 13
D meas. D nom
13
D meas. D nom
-5.8 13
-5.8
D meas. D nom
-5.8
13 12 11 10 9 8 7 6 5
D meas. D nom
-5.8 -6.3 -6.8 -7.3 -7.8 -8.3 -8.8 -9.3

free-fixed
12 11 10 9 -7.8 -8.3 -8.8 -9.3

free-fixed
-6.3 12 -6.8 11 10 9 -7.8 -8.3 -8.8

free-fixed
-6.3 12 -6.8 -7.3
-6.3
11 -7.3
-6.8
10
-7.3
9 8 7 6 -9.3 -9.8 4 3 -10.8 2 1 0 -11.3 -11.8 -10.8 -11.3 -11.8 5 -9.8 4 -10.3 3 2 1 0 5 6 7 8
-7.8
-8.3
-8.8
-9.3
-9.8
-9.8 4
-10.3
free-fixed
-10.3
-10.3 3 2 1 0 -10.8 -11.3 -11.8
-10.8
-11.3
-11.8
-1 -200 -100 Equivalent radius (mm) 0 100
-100
100
-12.3 200
-1 -200
-12.3 200
-1 -200
-100
100
-12.3 200 Equivalent radius (mm)
-1 -200
-100
100
Dmeas,Lfixed = 1.07*Dnom
Figure 11. Dmeas versus Dnom for the inclined IGU anchors in sand (inclination 25: with regard to the horizontal)
p. 15/26
BBRI & BGGG-GBMS
IGU 25 - HETEROGENEOUS Dnom = 140 mm (=casing diameter)

BBRI Research 'Ground anchors' Anchor E07-08 (Heterog. IGU-140)
BBRI Research 'Ground anchors' Anchor E09-10 (Heterog. IGU-140) 14
BBRI Research 'Ground anchors' Anchor E05-06 (Heterog. IGU-140)

14 13 12 11 10 9 -7.8 -8.3 -8.8 -9.3
free-fixed
BBRI Research 'Ground anchors' Anchor E12-13 (Heterog. IGU-140) 14
14
D meas. D nom
13
-6.3
12 -6.3
D meas. D nom
-5.8
13
-5.8 13 12 11 10 9 8 7 6 5
-9.8 4 3
D meas. D nom
-5.8
D meas. D nom
-5.8 -6.3 -6.8 -7.3 -7.8 -8.3 -8.8 -9.3

free-fixed
12
-6.8
11 10 9 -7.8 -8.3 -8.8 -9.3
free-fixed
-6.3
-6.8 -7.3
11
-7.3
-6.8
10
-7.3
9
8 7 6 5 -9.8 4 -10.3 3 2 1 0
-1 -200 5 6 7 8
-7.8
-8.3
-8.8
-9.3
free-fixed
Relative depth (R+xxm) -9.8 4

-10.3
-9.8
-10.3
-10.8
2 1 0
-10.3 3
-10.8 -11.3 -11.8
2
-11.3 -11.8
-10.8
2 1 0
-10.8 -11.3 -11.8
-11.3
-11.8
-1 -200
-100 Equivalent radius (mm) 0 100
-100
100
-12.3 200
-1 -200
-12.3 200
-100
100
-1 -200
-100
100
Dmeas,Lfixed = 1.11* Dnom
Figure 12. Dmeas versus Dnom for the inclined IGU anchors in the heterogeneous clayey sand layer (inclination 25: with regard to the horizontal)
p. 16/26
BBRI & BGGG-GBMS
IGU Vertical Dnom = 140 mm (=casing diameter)

BBRI Research 'Ground anchors' Anchor VE11 (Sand IGU-140)
BBRI Research 'Ground anchors' Anchor VE12 (Sand IGU-140) 2
D meas. D nom
top
BBRI Research 'Ground anchors' Anchor VE04 (Sand IGU - 140)

2 1
top
BBRI Research 'Ground anchors' Anchor VE13 (Sand IGU-140) 2 1 0

D meas. D nom
top
2
D meas. D nom
1
0
0
soil surface
soil surface
D meas. D nom
0
-1
-1 -2 -3 -4 -5 -6 -7 -8 -9 -10 -11 -12
bottom
bottom free-fixed
top
-1
-2
-2 -3 -4 -5 -6 -7 -8 -9 -10 -11 -12 -13 -14
-1
soil surface
-2
-3 -4 -5 -6 -7 -8
free-fixed
soil surface
-3
-4
-5
-6
-7
-9 -10 -11 -12 -13 -14 -15 -200 Equivalent radius (mm) -100 0 100 200
-14 -200 -100 0 100 Equivalent radius (mm) -13
-8
-9
free-fixed
-10
-11
free-fixed
-12
bottom
-13
bottom
-14
-200
-100
100
200
200
-200
-100
100 Equivalent radius (mm)
200
Figure 13. Dmeas versus Dnom for the vertical IGU anchors, Lfixed mainly in sand
p. 17/26
BBRI & BGGG-GBMS
2T-grav.Vertical - Dnom=140 mm(= casing diameter)

BBRI Research 'Ground anchors' Anchor VE03 (Sand IGU - 140) 2 1
top
BBRI Research 'Ground anchors' Anchor VE10 (Sand IGU-140) 2 1 0

D meas. D nom
D meas. D nom
0
top
-1
soil surface
-1 -2
soil surface
-2 -3 -4 -5 -6 -7 -8 -9
free-fixed
-3 -4 -5 -6 -7 -8 -9 -10 -11 -12

bottom bottom free-fixed
-10 -11 -12 -13 -14 -200 Equivalent radius (mm) -100 0 100 200 -14 -200 -100 -13
200
Figure 14. Dmeas versus Dnom for the vertical 2T-grav anchors, Lfixzd mainly in sand
p. 18/26
BBRI & BGGG-GBMS
1T 25 - Sand Dnom = 180 mm

BBRI Research 'Ground anchors' Anchor E10-11 (Het. 1T+inj.-180) 14
1T-red 25 - Sand Dnom = 150 mm 1T+inj 25 - Heterogen. Dnom = 180 mm

BBRI Research 'Ground anchors' Anchor E11-12 (Het. 1T+inj.-180) 14
D meas.
1T+inj 25 - Heterogen. Dnom = 180 mm
BBRI Research 'Ground anchors' Anchor E17-18 (Sand 1T-180)

D meas.
14
-4.8
13
D nom
D meas. D nom
-5,8
-5,8 -6,3 -6,8 -7,3
-5.3 12 -6.3
BBRI Research 'Ground anchors' Anchor E14-15 (Sand 1Tred.150) 14 D meas. -5.8 D nom 13
13
13
-6,3
D nom
12
-5.8 11 -6.8
12
12 11 10 9 8 7 6
11
11 10 9 -7,8 -8,3 -8,8 -9,3

free-fixed
-6,8 -7,3
-6.3 10 9 -7.8 -8.3 -8.8 -9.3

free-fixed
10
-7.3
9 8
-6.8
-7,8 -8,3 -8,8 -9,3 5

free-fixed
-7.3
8 7 6 5 4
7 6 5 -9.8 4 -10.3 3
-7.8 7
-8.3
free-fixed

-9,8
-8.8
-9,8 4
-10,3
-9.3
3
-10.8
-10,3 3
-10,8 -11,3 -11,8
-9.8 2
2 1 0
2 1 0
-10,8 -11,3 -11,8
1 0 -1 -200 -100 Equivalent radius (mm) 0 100 -12.3 200 -11.8
-10.3 1 -11.3
-10.8
-1 -200
-100
100
-11.3 200
-1 -200
-100
100
-12,3 200 Equivalent radius (mm)
-1 -200
-100
100
-12,3 200 Equivalent radius (mm)
Figure 15. Dmeas versus Dnom for the inclined 1T and 1T-red anchors in sand and the inclined 1T+inj anchors in the heterogeneous clayey sand layer (for all anchors : inclination equals 25 with regard to the horizontal and Dnom corresponds with the diameter of the lost drill point)
p. 19/26
BBRI & BGGG-GBMS
1T vertical Dnom = 180 mm

BBRI Research 'Ground anchors' Anchor VE08 (Sand 1T-180) 2 2 BBRI Research 'Ground anchors' Anchor VE14 (Sand 1T+inj. - 180)
1T- vertical Dnom = 180 mm

BBRI Research 'Ground anchors' Anchor VE15 (Sand 1T+inj. - 180) 2
D meas.
1T+inj vertical Dnom = 180 mm
1T+inj vertical Dnom = 180 mm
BBRI Research 'Ground anchors' Anchor VE05 (Sand 1T - 180)

D meas.
top
2 1
D nom D meas. D nom
D meas. D nom
top
1 0
D nom
0 -1
soil surface
top
0 -1
top
-1
-1 -2 -3 -4 -5 -6 -7
soil surface
-2 -3 -4 -5 -6 -7 -8
free-fixed
soil surface
-2 -3 -4 -5 -6 -7 -8 -9 -10 -11
bottom free-fixed
soil surface
-2
-3
-4
-5
-6
-7
-8 -9 -10 -11 -12 -13 -14 -200 Equivalent radius (mm) -100 0 100 200 -13 -14 -200 -100 0 -12
free-fixed
-8 -9 -10 -11
bottom
-9
free-fixed
-10
-11
-12
bottom
-12 -13 -14

100 Equivalent radius (mm) 200
bottom
-13
-14
-200
-100
100
200
-200
-100
200
Figure 16. Dmeas versus Dnom for the vertical 1T and 1T+inj anchors with Lfixed mainly in sand (Dnom corresponds with the diameter of the lost drill point)
p. 20/26
BBRI & BGGG-GBMS
SA-175-Is 25 - Sand Dnom = 175 mm

BBRI Research 'Ground anchors' Anchor E13-14 (Sand SA-130-Is) 14 13 12 -6.3
D meas. D nom
SA-130-Is 25 - Sand Dnom = 130 mm
SA-175-Is 25 - Heterog. Dnom = 175 mm
SA-130-Is 25 - Heterog. Dnom = 130 mm
BBRI Research 'Ground anchors' Anchor E03-04 (Sand SA-175-Is)
14
-5.8
13
12 11 10 9 -7.8 -8.3 -8.8 -9.3
free-fixed
D meas. D nom
-5.8
-6.3 12 11 10 9 8 7 6 5 -9.8 4
-10.3 3
BBRI Research 'Ground anchors' Anchor E06-07 (Heterog. SA-175Is) 14 D meas. -5.8 D nom 13
BBRI Research 'Ground anchors' Anchor E08-09 (Heterog. SA-130Is) 14 D meas. -5.8 D nom 13 -6.3 -6.8 -7.3 -7.8 -8.3 -8.8 -9.3
free-fixed
12
11 10 9 -7.8 -8.3 -8.8 -9.3
free-fixed
-6.3
-6.8
11
-7.3
-6.8
-7.3
-6.8
10
-7.3
9
8
-7.8
8 7 6 5
8
7 6 5 -9.8 4
-8.3
-8.8
-9.3
free-fixed
Relative depth (R+xxm) -9.8 4 -10.3 3 -10.8 -11.3 -11.8 2 1 0 -10.8 -11.3 -11.8
-9.8
4
3
-10.8
-10.3
2
-10.3
3
2 1 0
2
1 0 -1 -200 -100 Equivalent radius (mm) 0 100 -12.3 200 -11.8 -11.3
-10.8
-11.3
-11.8
-1 -200
-100
100
-12.3 200
-1 -200
-100
100
-1 -200
-100
100
Figure 17. Dmeas versus Dnom for the inclined self boring hollow bar anchors SA-xx-Is in sand and in the heterogeneous clayey sand layer (for all anchors : inclination equals 25 with regard to the horizontal and Dnom=xx corresponds with the diameter of the drill bit)
p. 21/26
BBRI & BGGG-GBMS
SA-150-Dy vertical Dnom = 150 mm

BBRI Research 'Ground anchors' Anchor VE07 (Sand SA-150-Is)
BBRI Research 'Ground anchors' Anchor VE09 (Sand SA-150-Is) 2 1 0 -1
top soil surface
SA-150-Is vertical Dnom = 150 mm

2 1 0 -1
D meas. D nom
D meas. D nom
SA-150-Dy vertical Dnom = 150 mm
BBRI Research 'Ground anchors' Anchor VE06 (Sand SA-150-Dy)

D meas. D nom
-1
top soil surface
-2
-2 -3 -4 -5 -6 -7 -8 -9 -10 -11
free-fixed
top soil surface
-2 -3 -4 -5 -6 -7 -8 -9 -10 -11
free-fixed
-3
-4
-5
-6
-7
free-fixed
-9
-10
-11
bottom
-12
-12 -13 -14
bottom
-8
-12 -13 -14
bottom
-13
-14 0 100 200
-200
-100
-200
-100
100
200
-200
-100
200
Figure 18. Dmeas versus Dnom for the vertical self boring hollow bar anchors SA-150-Dy and SA-150-Is, Lfixed mainly in sand (Dnom corresponds with the diameter of the drill bit)
p. 22/26
BBRI & BGGG-GBMS
3.
DATA COLLECTED ON A WORK SITE IN KNOKKE

During excavation works for the realisation of an underground parking on a site in Knokke (B), anchors formerly installed during the construction of the neighbouring basement were encountered. It concerned self boring hollow bar anchors from the Ischebeck type, installed in dense quaternary sand (dune sand) by the company CVR. Bars of the type TITAN 30/16 and drill bit with a diameter of 90 mm (type HV375) were used This excavation offered the opportunity to gather further information concerning the real dimension of these anchor types. The excavation took place on 4th March 2008, and 2 self boring hollow bar anchor were dug out. Figure 19 gives an overview of the work site and illustrates a CPT which provides an idea about the cone resistance of the sand layer in which the anchors have been installed.
2 10 20 30 qc [MPa]
3.1. Introduction
5-
10-
15-
Figure 19: Overview of the work site in Knokke CPT results
3.2. Methodology & Measurements results

A similar methodology as that one adopted for the Limelette site was used : permanent standby of BBRI staff was provided in order to guide the crane operator, the soil in the direct neighbourhood of the anchors was manually removed, a photographic report with pictures each 50 cm has been created, soil and fissuring patterns descriptions have been made, the diameter of the grout has been measured. Dmeas.has been deduced from perimeter measurements. Measurements were made each 20 cm or less in case of irregular shape. The results of these observations and measurements are given in the Annex A of this contribution.
p. 23/26
BBRI & BGGG-GBMS
Figure 20 presents for both anchors the measured diameter as a function of the length from the anchor bottom. Due to the presence of the ground water level, it was not possible to excavate the last 50 cm of the anchors. Moreover, the stability of the surrounding slopes enforced the site staff to stop the excavation of the anchor A1 at 1.5 m from the bottom. Due to this, the effect of the postgrouting operation at the anchor bottom could not be observed. Figure 20 shows that the measured diameter of the anchor is in good agreement with the nominal diameter (90 mm). A ratio Dmeas,av/Dreal of about 1.1 is found.
180
A1 A2
measured anchor diameter [mm] 150
120
90
60
30
0 0 1 2 3 4 5 6 7 8 9 length from the anchor bottom [m]
Figure 20: Dmeas versus Dnom for the inclined self boring hollow bar anchors SA-90Is investigated on the work site of Knokke
4.
CONCLUSIONS WITH REGARD TO THE REAL ANCHOR DIAMETER & COMPARISON WITH LITERATURE DATA
Table 2 summarises for each anchor type an overview of the average ratios of Dmeas/Dnom along Lfixed, that were obtained from the observations in Limelette and Knokke From the values in Table 2 and the observations mentioned before the following conclusions can be drawn: For the IGU anchors it can in general be concluded that the measured diameter is somewhat higher than the nominal anchor diameter due to local thickenings induced during the drilling For the systems with lost oversized drill point (1T) or drill bit (SA), it can be concluded that the real anchor diameter Dmeas and the obtained anchor shape depend strongly on the drilling procedure, i.e. the flow rate of the drill fluid and/or the ratio between flow rate en drilling diameter in particular. The comparison between the diameter measured on 1T-red and 1T anchors shows
p. 24/26
BBRI & BGGG-GBMS
clearly this effect. For SA installed on the Limelette site, it can be stated that the flow rate (and/or the pressure) of the drill fluid, cement grout in this case, was probably too low in order to realise an increase of the real anchor diameter with regard to the drill bit diameter Dnom. For the 1T+inj system that was provided with two post-grouting tubes, the effect of the post-grouting operations has clearly been observed. Along the zone of Lfixed where the manchettes were present, a significant increase of the diameter up to 30% may be expected with regard to Dnom (in sand and heterog. layer)
Table 2 Overview of the Dmeas/Dnom ratios along Lfixed deduced from the investigated sites VERTICAL ANCHORS TYPE Dmeas Dnom [mm]. [mm]. 144 140 IGU 194 180 1T 180 1T+inj. 233 150 SA-150 156 INCLINED ANCHORS 154 140 IGU 201 180 1T 180 1T+inj. 216 150 1T+red. 209 90 SA-90(*) 98 130 SA-130 136 166 175 SA-175
(*)
Dmeas. / Dnom. [-] 1.03 (1.00 - 1.08) 1.08 (1.07 & 1.09) 1.29 (1.28 & 1.30) 1.04 (1.03 - 1.07) 1.10 1.12 1.20 1.40 1.09 1.05 0.95 (1.04 - 1.14) (1.15 - 1.24)
(0.97 & 1.12) (0.86 & 1.04)
values deduced from the work site in Knokke
These results can be compared with the values in Table 3 that contains data concerning empirical factors (EXP) that accounts for an increase of the anchor dimension in Lfixed with regard to the nominal diameter.
Table 3. Value of the coefficient commonly used in Belgium [TA 95 (1995) and EBA (2004)] Soil Gravelly sands Fine to coarse sands Silt Clay IRS (after TA 95) 1.5 1.6 1.4 1.5 1.4 1.6 1.8 - 2 IGU (after TA 95) 1.2 1.3 1.1 1.2 1.1 1.2 1.2 SA (after EBA) 2.0 1.5
When comparing the values in Tables 2 & 3, one can concluded that the values of EXP obtained from the observations in the framework of the BBRI research on ground anchors deviate from the literature data, in particular for the self boring hollow bar anchors. With regard to those self boring hollow bar anchors, supplementary tests (phase 3) are actually performed on the Limelette test field in order to assess the influence of the drilling procedure (intensive percussive drilling) on the real anchor diameter and on the anchor capacity. During this phase 3 of the test program, IRS anchors were installed in the loam layer and are actually load tested as well. In the coming months, some of the IRS and SA anchors installed in the loam will probably be excavated in order to assess their real dimensions. The results of these observations will be subject of an addendum to the Volumes of this symposium.
p. 25/26
BBRI & BGGG-GBMS
Finally, based on the observations made in the (clayey) sand layer on IGU systems and systems with lost point (1T) or drill bit (SA), it is proposed for the moment to introduce Dnom (diameter of the casing, lost point or drill bit) in the design.
5.
REFERENCES
Huybrechts, N., De Vos, M., Tomboy, O., and Maertens, J. 2008. Integrated analysis of the anchor load test results in Limelette and suggestions for a harmonized anchor design and test methodology in Belgium in a EC7 framework, Proceedings of the international symposium on ground anchors, May 14th 2008, Brussels. Huybrechts, N., Tomboy, O. Maertens, J. and Holeyman, A. 2008b. Full scale load test program in Limelette: overview of the test campaign, set-up & general results, Proceedings of the international symposium on ground anchors, May 14th 2008, Brussels. Recommandation TA 95 - Tirants dancrage 1995, Recommandations concernant la conception le calcul, lexcution et le contrle. Bustamante, M. & Doix, B. 1985. Une mthode pour le calcul des tirants et des micropieux injects, In Bull. liaison laboratoire des Ponts et Chausses, n 140, Nov.-Dec. page 75-92 EBA Zulassung2004. Verwendung von Verpresspfhlen System Ischebeck TITAN zur temporren Sicherung von Baugrubenwnden bei den Eisenbahnen des Bundes.
p. 26/26
BBRI
BGGG GBMS
Annex A
BBRI & BGGG-GBMS
A1
Type : Inclined 35 SA-90-Is 8.5 : : : : NA NA Sand 20-30
Installation method :
Anchor information
Lbehind retaining wall [m]: Instrumentation
Test information Soil information
Type Type qc,av ,global [MPa]
Annex A - p. 3/13
BBRI & BGGG-GBMS
BBRI Research Ground Anchors Site Knokke. Anchor A1 ( SA-90-Is Sand)
0 0.5 m
0.5 1.0 m
1.0 1.5 m
1.5 2.0 m
2.0 2.5 m
2.5 3.0 m
3.0 3.5 m
3.5 4.0 m
4.0 4.5 m
4.5 5.0 m
5.0 5.5 m
5.5 6.0 m
The lengths are given with respect to the retaining wall. The anchor bottom is situated at 8.5 m from the retaining wall.
Annex A - p. 4/13
BBRI & BGGG-GBMS
6.0 6.5 m
6.5 7.0 m
7.0 7.2 m
Annex A - p. 5/13
BBRI & BGGG-GBMS
Soil description
0 7.2 m Yellow/white coloured dune sand
Fissures description
Not fissured except at 2.0 m where a longitudinal fissure occurred.
General comments : - Visual aspect : relatively straight anchor with some local deviations (picture a) - Surface shows traces due to the drilling procedure. The traces disappear along the 2 last meters close to the bottom. - Due to practical reasons, the anchor bottom could not be excavated, no enlargement due to post-injection could consequently be observed - Local small enlargement at 3m (picture b)
Picture a
Picture b
Annex A - p. 6/13
BBRI & BGGG-GBMS
Ischebeck anchors : Knokke Anchor A1 ( 90mm)

Equivalent radius (mm) -100 0 -50 0 50 100
D meas. D nom
3 Length from wall (m)
assumed end
Annex A - p. 7/13
BBRI & BGGG-GBMS
Annex A - p. 8/13
BBRI & BGGG-GBMS
A2
Type Installation method : : Inclined 35 SA-90-Is 8.5 : : : : NA NA Sand 20-30
Anchor information
Lbehind retaining wall [m]: Instrumentation
Test information Soil information
Type Type qc,av ,global [MPa]
Annex A - p. 9/13
BBRI & BGGG-GBMS
BBRI Research Ground Anchors Site Knokke. Anchor A2 ( SA-90-Is Sand)
0 0.5 m
0.5 1.0 m
1.0 1.5 m
1.5 2.0 m
2.0 2.5 m
2.5 3.0 m
3.0 3.5 m
3.5 4.0 m
4.0 4.5 m
4.5 5.0 m
5.0 5.5 m
5.5 6.0 m
Annex A - p. 10/13
BBRI & BGGG-GBMS
6.0 6.5 m
6.5 7.0 m
7.0 7.2 m
Annex A - p. 11/13
BBRI & BGGG-GBMS
Soil description
0 8.0 m Yellow/white coloured dune sand
Fissures description
Not fissured
General comments : - Visual aspect : relatively straight anchor with some local deviations (picture a) - Surface shows traces due to the drilling procedure. The traces disappear along the 2 last meters close to the bottom - Due to practical reasons, the anchor bottom could not be excavated, no enlargement due to post-injection could consequently be observed
Picture a
Annex A - p. 12/13
BBRI & BGGG-GBMS
Ischebeck anchors : Knokke Anchor A2 ( 90mm)

Equivalent radius (mm) -100 0 -50 0 50 100
D meas. D nom
3 Length from wall (m)
assumed end
Annex A - p. 13/13
BBRI
BGGG GBMS
Integrated Analysis of the Anchor Load Test Results in Limelette & Suggestions for Harmonised Anchor Design and Test Methodology in Belgium in a EC7 Framework
Ir. Noel Huybrechts, Ir. Monika De Vos & Ir. Olivier Tomboy Belgian Building Research Institute (BBRI) Geotechnical & Structural Division
Prof. Ir. Jan Maertens Jan Maertens bvba & Catholic University of Leuven (KUL)
BBRI & BGGG-GBMS
1.
INTRODUCTION
This contribution summarizes the analysis of the load test results on different anchor types performed in the framework of phase 1 and phase 2 of the anchor test campaign in Limelette. In the contributions of Huybrechts & Maertens (2008) and Tomboy & Huybrechts (2008) to this Volume a general overview has been given of the anchor installation techniques, the test results and the observations and measurements on the excavated anchors. In this contribution it is the objective to summarize the methodology that has been applied to analyse the test data taking into account all available data and to formulate some general conclusions with regard to the Limelette anchor test campaign. For a detailed report of this integrated analysis, reference is made to Volume 2 of the proceedings of this symposium. Finally, some suggestions for a harmonised anchor design and test methodology in Belgium taking into account Eurocode 7 principles and anchor practice in Belgium are formulated.
2.
ANALYSIS OF THE TEST RESULTS

The ultimate anchor load Pu of the anchors has been deduced from the creep curves obtained during the anchor load tests. For all anchors Pu has, according to prEN ISO 22477-5, conventionally been determined as the load for which: the creep value =5 in the case where the maintained load test procedure (TM3) has been applied, the creep value =2 in the case where the cyclic load test procedure (TM1) has been applied. The results for the 49 test anchors are summarized in Table 1. It should be remarked that most of the anchors failed at the grout-ground interface during load testing. However, for some anchors this Pu value might be a slight underestimation of Pu as the mentioned conventional creep values were not yet reached at the maximum test load Pmax (due to the steel limit), or due to structural failure of the tendon (in one case). Furthermore it is worthwhile to mention that in some cases the difference between the values of Pu determined out of a the conventional creep values mentioned above, or determined as asymptotic value in the creep curve, (specified in different normative references), is significant. This has in particular been found for the cyclic anchor tests (TM1) and the load test in the more cohesive soil layers. Figure 1 gives an example. A detailed analysis can be found in Volume 2 of the proceedings of this symposium.
2.1. Determination of Pu
Integrated analysis & suggestions for NA-EC7 N. Huybrechts et al. 14.05.2008
p. 3/21
BBRI & BGGG-GBMS
Table 1 : Ultimate anchor load Pu of the anchors tested in Limelette Pu determined out of creep curve following prEN 22477-5 conventional criteria =5 (TM3) or =2 (TM1)
(*)Preliminary anchors
p. 4/21
BBRI & BGGG-GBMS
Inclined anchors of IGU type in Sand Layer

1600 1400 1200
(6') (60') (increasing) (increasing) (60') (60') (60') (60')
Load (kN)
1000
(increasing)
800 600 400 200

IGU-5m-TM3 IGU-6m-TM3 S1 S2 IGU-4m-TM3 E2-3 E26-27 IGU-5m-TM1 E4-5 E21-22 E30-31
Pmax Pu
0
E16-17 E28-29 E15-16 E27-28
Figure 1 : Comparison of ultimate anchor load Pu versus maximum applied test load (the value between brackets equals the time in minutes that Pmax has been maintained before the anchor failed when increasing is noted that means that the anchor failed during increasing the load for the next step if no value is given than Pmax corresponds with mechanical steel limit or Pu)
2.2. Interpretation of the instrumentation

The test anchors in Limelette were instrumented with strain sensors, with which the deformations of the anchors tendon were measured on several positions. For the strand anchors (IGU - 2Tgrav 1T 1T+inj) strain gauges were fixed on the strands in Lfixed. However, although more than 90% of the strain gauges survived anchor installation a lot of them were damaged during the load test it self so that only a limited amount of data was available for analysis. For the self boring hollow bar anchors an extensometer device, which could be installed in the hollow bars during testing, was developed by BBRI. In general high quality data were obtained with this measurement device. By multiplying the deformation measurements with a modulus of elasticity E and a section A the corresponding load can be obtained. The factor EA (rigidity) of the tendons was deduced from the measurement result by means of the Fellenius method (2001). Form this method a non-linear stress () -strain () behaviour, especially for the hollow bar tendons, was derived and applied (see Figure 2).
p. 5/21
BBRI & BGGG-GBMS
900 800 700 600 stress (MPa) 500 400 300 200 100 0 0 500 1000 1500 2000 Fellenius Ischebeck TITAN 73/45 (A=2260mm) Fellenius Dywidrill T76N (A=1900m) Fellenius strands (A=1057 mm) theoretical relationship (E=210 GPa) fy TITAN 73/45 fy T76N 2500
-6
3000
3500
4000
4500
strain (10 )
Figure 2 : Stress-strain behaviour of tendons deduced from Fellenius (2001) method
With this approach the following could be obtained for several anchors; the friction losses in the free length Lfree, which exist out of internal friction losses between tendon and plastic tubing and out of friction mobilised along the outer of the grout column in Lfree the load distribution in the anchors fixed length Lfixed the mobilisation curves of the unit shaft friction (qsi curves). For the anchors installed in the heterogeneous clayey sand and the tertiary sand layer, this analysis revealed average total friction losses in Lfree of 14% of the ultimate anchor load Pu for the strand anchors and 19% of Pu for the self boring hollow bar anchors. For the anchors installed in loam, the load losses in Lfree corresponding with the ultimate anchor load Pu have been estimated on 7% (based on limited information). For more details with regard to this analysis reference is made to Volume 2 of the proceedings of this symposium.
2.3. Determination of qsu global approach

The analysis in the previous point revealed that the values of the ultimate anchor loads (Pu), which are deduced from the test measurements and given in Table 1, should be corrected for friction losses, in order to quantify the load that has really been transmitted to Lfixed. By correcting Pu for friction losses in Lfree and taking into account the measured anchor dimensions from Tomboy & Huybrechts (2008), an estimate has been made of the real global ultimate unit shaft friction qsu along Lfixed for all the tested anchors. In Figures 3 (heterogeneous clayey sand and tertiary sand) and 4 (loam/silt) these qsu values are expressed as a function of the average cone resistance qc,av. along Lfixed of each individual anchor. On the figures 3 and 4 some other curves have been represented as well. It concerns some qs-qc curves that are used on a regular base for anchor design in
p. 6/21
BBRI & BGGG-GBMS
Belgium based on CPT data; i.e. curves from TA95 (CFMS, 1995) for IGU and IRS anchors and an empirical rule which states that qs=0.0033.qc with a maximum of 466 kPa.
BBRI project Ground Anchors - Result Limelette qs(qc) in Sand & Heterogeneous layer 1000 900 800 700 600 qsu (kPa) 500 400 300 200 100 0 0 4 8 12 16 20 qc,av (MPa) 24 28 32 36 40 qs = [Pu-W tot]/(..Dnom.Lfixed) - with real or average if not available - with real loss W tot or average if not available IGU-TA95 IRS-TA95 qc/30 IGU-sand-TM3 IGU-sand-TM1 IGU-Verticaal-TM3 IGU-Verticaal-TM1 IGU-Het.-TM3 IGU-Het.-TM1 1T-sand-TM3 1T-Dred.-sand-TM3 1T-Verticaal-TM3 1T+inj-Verticaal-TM3 1T+inj-Het.-TM3 SA-sand-TM3 SA-Verticaal.-TM3 SA-Het.-TM3 2T-grav -Verticaal
Figure 3 : qsu-qc,av for anchors tested in tertiary sand and clayey sand in Limelette only the real load on the fixed length is considered (by taking into account friction losses in Lfree and the real anchor diameter)
BBRI project Ground Anchors - Result Limelette qs(qc) in Silt layer 300
250 IGU-TA95 200 qsu (kPa) IRS-TA95 qc/30 150 IGU-silt-TM3 SA150-silt-TM3 100 qs = [Pu-W tot]/(..Dnom.Lfixed) - with = 1 for SA and IGU - with average loss W tot SA76mm-silt-TM3
50
0 0 2 4 qc,av (MPa) 6 8 10
Figure 4 : qsu-qc,av for anchors tested in silt (loam) in Limelette - only the real load on the fixed length is considered (by taking into account friction losses in Lfree - is assumed to be 1)
p. 7/21
BBRI & BGGG-GBMS
2.4. Relation between execution dimensions and test results
method,
real
anchor
In Volume 2 of the proceedings of this symposium, an attempt has been made to represent on some figures some important anchor installation data together with the measured anchor dimensions, the test method and the test results. Two examples are given in Figure 5 for a 1T+inj anchor and an IGU anchor.
Figure 5 : Dmeas versus Dnom for an inclined 1T+inj anchor in the heterogeneous clayey sand layer and an inclined IGU anchor in the tertiary sand layer - inclination equals 25 with regard to the horizontal and Dnom corresponds with the diameter of the lost drill point(1T+inj.) or diameter of the casing (IGU).
p. 8/21
BBRI & BGGG-GBMS
3.
CONCLUSIONS ANCHOR LOAD TEST RESULTS LIMELETTE

In the previous points a detailed analysis of the anchor load test results has been presented: the test results have been corrected for friction losses in the anchors free length and the real anchor diameters have been introduced in order to assess as precisely as possible the real global unit shaft friction qsu along Lfixed. When the results of this analysis are linked with the anchor execution parameters and the results of the inclined versus vertical CPT the conclusions summarized here below can be drawn. Anchor capacity of the different anchorage systems in sand and clayey sand From Figure 6, which is an adopted representation of the qsu-qc values from Figure 3, it can be deduced that, for the anchors of which Lfixed is installed in clayey sand with average cone resistances qc,av along Lfixed of 12 to 14 MPa and in tertiary Bruxellian sand with qc,av between 16 and 26 MPa, the following values of the ultimate global unit shaft friction qsu are obtained for anchors with Lfixed between 4 to 6 m (after correction for losses in the free length and taking into account Dreal) : For the IGU anchors : 0.015qc,av < qsu < 0.030qc,av For the 1T anchors : 0.013qc,av < qsu < 0.020qc,av For the SA anchors : 0.015qc,av < qsu < 0.020qc,av The gravity filled anchors 2T-grav show qsu value in the neighbourhood of the lower boundaries of 1T and SA anchors at 0.013qc,av to 0.015qc,av The 1T +inj. show qsu value in the neighbourhood of the higher boundary of the IGU anchors at 0.025qc,av to 0.030qc,av
BBRI project Ground Anchors - Result Limelette qs(qc) in Sand & Heterogeneous layer 1000 900 800 700 600 qsu (kPa) 500 400 300 200 100 0 0 4 8 12 16 20 qc,av. (MPa) 24 28 32 36 40 qs = [Pu-W tot]/(..Dnom.Lfixed) - with real or average if not available - with real loss W tot or average if not available Min.-qc/65 Max - qc/33 Min.-qc/75 Max - qc/50 Min.-qc/65 Max - qc/50 IGU-sand-TM3 IGU-sand-TM1 IGU-Verticaal-TM3 IGU-Verticaal-TM1 IGU-Het.-TM3 IGU-Het.-TM1 1T-sand-TM3 1T-Dred.-sand-TM3 1T-Verticaal-TM3 1T+inj-Verticaal-TM3 1T+inj-Het.-TM3 SA-sand-TM3 SA-Verticaal.-TM3 SA-Het.-TM3 2T-grav -Verticaal
Figure 6 : qsu qc curves of the anchors tested in clayey sand and sand - only the real load on the fixed length is considered (by taking into account friction losses in Lfree and the real anchor diameter)
p. 9/21
BBRI & BGGG-GBMS
On the same Figure 6 it can be observed that: the variation of the results is o very high for the IGU anchors, o lower for the 1T anchors, o considerably low for the SA anchor; the result of the gravity filled anchors 2T-grav are situated in the intervals of the 1T and SA anchors; the results of the IGU anchors that are situated in the area of the 1T and SA anchors are almost all vertical anchors. Out of these observations it might be concluded that the 1T and the SA anchors that have been performed at the Limelette test site can be considered as gravity filled anchors. This is not surprising when looking at the installation monitoring of these anchors (almost no pressure during injection). The fact that for the SA anchors the results can be situated in a narrower interval is probably due to the very regular form of these anchors (see Tomboy, 2008). Moreover the high variation obtained for the IGU anchors evidences the effect of the stepwise grout injection procedure, and the beneficial effect of pressure grouting on the obtained qs values. It is not surprising that the qsu values of vertical IGU anchors are situated in the zone of the gravity filled anchors, as it could in general be observed that the grout injection procedure was not very successful for the vertical anchors (probably due to the shorter lengths Lfixed and Lfree). Moreover, the shape of the vertical anchors seems in general to be some what more regular than the shape of the inclined anchors. The beneficial effect of pressure grouting and an irregular anchor form on the anchor capacity is also confirmed by the results obtained for the 1T+inj anchors. Anchor capacity of the different anchorage systems in silt (loam) For the anchors of which Lfixed is installed in loam, the real unit shaft friction has been determined based on limited information: only for one anchor the friction losses in the free length were determined, and based on observations for the sand anchors it has been assumed that Dreal = Dnom. This leads to the following results in loam with cone resistances between 3.4 and 4.5 MPa (see Figure 4). For the IGU anchors : For the SA-150-Dy : For the SA-76-Dy : qsu = 0.040 to 0.045 qc,av qsu = 0.030 to 0.040 qc,av qsu = 0.045 to 0.072 qc,av
Based on these results one can conclude that: The variation of the results obtained for the IGU anchors is rather low. This can probably be explained by the fact that the dimensions of the IGU anchors in the loam layer are regular (assumption based on observation in Lfree of excavated IGU anchors). Moreover it has been observed during installation that the grout pressures that could be realised in loam are comparable for the different anchors. With regard to this stepwise grout injection it was found that the realised pressures were not high (maximum 8 bar, mostly 5 bar) and that for each injection step grout leakage to the surface occurred.
p. 10/21
BBRI & BGGG-GBMS
For the SA-150-Dy anchors installed with a drill bit with diameter of 150 mm, the qsu values that are obtained are somewhat lower than those obtained for the IGU anchors. Out of the differences between the results of the IGU and the SA-150-Dy anchors in loam, and out of the observations during installation, one could conclude that the SA-150 anchors in loam can be considered as gravity filled anchors and that the grout injection procedure for the IGU anchors show some beneficial effect but less significant than in sand. For the SA-76-Dy anchors installed with a drill bit with diameter of 76 mm, the obtained qs values are 50% higher than the SA-150-Dy. As in general it is assumed that in cohesive layers the unit shaft friction is independent from the anchor diameter (cfr. Ostermayer & Barley, 2003), especially for gravity filled anchors, this could possibly indicate that the real diameter is considerably higher than the nominal drill bit diameter. Influence of the inclination on the anchor capacity (anchors in sand) On first instance, looking at the ultimate anchor load deduced from the test results (see Table 1), it was found that for all systems the capacity of the vertical anchors was significantly lower than the capacity of the inclined anchors. However, after corrections for friction losses in the free length and taking into account the measured anchor dimensions to determine the anchors real unit shaft friction, this difference becomes insignificant for the SA anchors and less significant for the 1T anchors. Only for the IGU anchors the vertical anchor capacity remains significantly lower than the inclined anchor capacity, but as explained before, the main reason for this is probably the less successful grout injection procedure compared to that of the inclined IGU anchors. Probably the less regular form and straightness of the inclined anchors might explain some of the remaining differences as well. Based on this argumentation it can be concluded that the inclination in itself has no significant influence on the anchor capacity in the Limelette sand layer. This was also confirmed by the comparison between cone resistances obtained with vertical and inclined CPT showing no significant differences (See Volume 2 of the proceedings of this symposium). With regard to the influence of the length on the bond stress, the bond stress evolution and/or the (non-) linear increase of anchor capacity with Lfixed Based on the obtained information about bond stress evolution along Lfixed as deduced from the load distribution measurements (see Volume 2) it can be concluded that: For SA anchors with Lfixed up to 6 m in heterogeneous clayey sand and tertiary sand tested according to the maintained load test procedure (TM3), the value of the unit shaft friction qsi in the different anchor zones in Lfixed continue to increase; no peak value and consecutively no residual value of qsi have been observed. For some of these anchors that did not fail under Pmax such behaviour was however observed when the anchor was submitted to subsequent Pa-Pmax cyclic loading.
p. 11/21
BBRI & BGGG-GBMS

For IGU and 1T anchors in heterogeneous clayey sand and tertiary sand only limited information was obtained from the measurements, but it is assumed that the same conclusions as for SA anchors can be drawn. For IGU anchors in loam (silt) with Lfixed= 5m, that were all tested according to the maintained load test procedure TM3, measurements on one preliminary tested strand anchor show that the qsi values of the first part of Lfixed evolutes to a peak value and drops back to a residual value. Although observed in a limited way for the more cohesive loam layer, it is assumed that when anchors are submitted to a maintained load test procedure (TM3), fixed anchor lengths up to 6m are too short to observe in the soil layers in Limelette a non-linear increase of anchor capacity with length. It also assumed that submitting anchors to cyclic testing favours a non-linear increase of anchor capacity with length; as design methods in the UK and Germany are calibrated with cyclic test methods, this is probably the reason that they have introduced a length dependent efficiency factor in their design (decreasing anchor capacity with length).
With regard to the influence of the test method on the anchor capacity The maintained load test procedure (Test Method 3 or TM3 according to the PrEN ISO 22477-5) has been applied as reference test method. However, some anchors have been tested according to the cyclic test procedure Test Method 1 (TM1) of PrEN ISO 22477-5. It concerns; 3 inclined IGU anchors in sand 2 vertical IGU anchors in sands 2 inclined IGU anchors in the heterogeneous clayey sand layer In Table 2 the results of the load tests in terms of an ultimate global unit shaft friction qsu (corrected for friction losses in Lfree and based on Dreal) and the ratio with regard to the average cone resistance is given for the anchors tested according to the TM1 method. In the same table the results of the reference tests on the similar anchor types tested according to TM3 are also given. Based on the results in Table 2 the following conclusions can be drawn: For the IGU anchors in sand the ultimate unit shaft friction qsu for anchors tested with TM1 is somewhat lower (~6%) than for the anchors tested with the TM3 method. Within the high variation obtained for all IGU anchors, such a difference can not be considered as significant. This seems not surprisingly for the Bruxellian sand layer and for limited fixed anchor lengths of 5 m. Within the anchors tested according to the TM1 method no influence of the number of cycles on the anchor capacity could be deduced. For the IGU anchors in the heterogeneous clayey sand layer the difference between qsu obtained from TM1 versus TM 3 method is higher, even up to 15%. The more cohesive character of this soil layer could possibly explain such a difference. However as mentioned before it should be emphasized that large variations are obtained for the IGU anchors. Furthermore, as highlighted before, the ultimate anchor load Pu that has been considered in this analysis has been determined according to the conventional creep criteria of the PrEN ISO 22477-5. Analysis has shown that the Pu value obtained in this way is, in some cases, lower than the ultimate anchor load
p. 12/21
BBRI & BGGG-GBMS
deduced from an asymptotic approach in the creep curve. This was especially the case for the anchors installed in the more cohesive layers and the anchors tested according to the cyclic TM1 method (example see Figure 1).
Table 2 : Comparison of qsu values obtained for IGU anchors tested with TM1 and TM3
With regard to the influence of the anchor diameter on the unit shaft friction For the SA anchors in sand, different anchor diameters have been realised (Dnom = 130 mm, 150 mm and 175 mm). Based on the measurements of the real anchor diameters where it was shown that Dmeas Dnom and the detailed analysis of the test results, represented in Figures 3 and 6 it can be concluded that the obtained ultimate unit shaft friction is independent from the anchor diameters in this diameter range. For the SA anchors installed in loam; anchors with Dnom=76 mm and 150 mm were installed. Assuming Dreal Dnom, large differences in qs values were obtained. However these anchors have not been excavated and the real anchor diameter could not be determined. Anyway, from literature (cfr. Ostermayer & Barley, 2003) it is assumed that in cohesive soil layers anchor diameter does not influence the qs value. The same reference mention however that for pressure grouted anchors realised in sand the anchor capacity in terms of kN/m is independent from the anchor diameter (in a range of 100 to 150 mm). This means that qsu decreases with anchor diameter. In the Limelette test campaign this parameter has not been investigated for the IGU anchors in sand. For the 1T anchors in sand this parameter was introduced but upon excavation it was shown that, although the drilling was performed with different diameters of the drill tools, that the real diameter were equal, so that no conclusions could be drawn on that point.
p. 13/21
BBRI & BGGG-GBMS
With regard to the effectiveness of the free length (Lfree) of self boring hollow bar anchors (SA) One of the parameters that has been investigated is the effectiveness of the free length of the self boring hollow bar anchors. This was questioned, especially due to the fact that the plastic tubing in Lfree is not continuously present over the complete length (only between the coupling sleeves) and because grout infiltration between tendon and plastic tubing is to be expected during installation. Based on the results of the load tests on self boring hollow bar anchors and especially based on the load distribution obtained from the extensometer measurements it can be concluded that: For loads up 20 to 30% of Pmax, a relative stiff load-displacement behaviour was measured. Together with the load distribution measurements it could be deduced that the load was almost completely transferred to the soil in the free length of the anchor. However from that moment it could be deduced from the measurements that the plastic tubing was detached from the surrounding grout and load was transferred to the fixed length. This means that the plastic tubing fulfils its role to guarantee a free anchor length. The only difference that could be observed compared with the strand anchors, for which the strands in the free length were greased, is that for the SA anchors the internal friction between tendon and plastic tubing is 5% higher. With regard to the influence of the absence of spacers between the strands. Some anchors have been installed without spacers between the strands. It concerns the IGU anchors E09-10 and E07-08 installed in the heterogeneous clayey sand layer. No influence on tendon-grout bond capacity nor on the anchor capacity (grout-ground bond) has been observed for these anchors.
4.
SUGGESTIONS FOR A HARMONISED DESIGN APPROACH ACCORING TO EC7

As in the past no geotechnical standards existed in Belgium, many owners have established their own technical specifications. This has lead to large variations in the applied rules for anchor testing and design in Belgium. Within the framework of the BBRI anchor research program (BBRI, 2004-2008) an analysis has been made of several anchor specification documents and anchor testing and design practices in Belgium. Taking into account the specifications in the Eurocode 7 and the Pr EN ISO 22477-5 a proposal for a uniform methodology for anchor testing in Belgium (terminology, test methods, aim of testing, ) has been elaborated. The proposal that has been discussed in the Working Group Ground anchors is presented in Table 3.
4.1. Test methodology
p. 14/21
BBRI & BGGG-GBMS
Table 3: Proposal for a uniform anchor testing methodology in Belgium

Investigation tests (NL : Basisproeven Fr: Essais pralable) Suitability tests (NL: Geschiktheidsproeven Fr: Essais de conformit) Acceptance tests (NL: Aanvaardingsproeven Fr: Essais de rception) active passive
Goal of the test
To confirm that a particular anchor design will be adequate in particular ground condition
To confirm that each executed anchor suits to support the load under serviceability conditions
On site or elsewhere
To establish the ultimate resistance (Ra and Rc) of an anchor at the grout-ground interface and to determine the characteristics of the anchorages in the working load range on site on another site with same stratigraphy reference tests (same stratigraphy and same anchorage technology) On site On site Proof anchors Before starting the work (contract determines if the contractor can start before acceptance of the results) Working anchors Afterwards
On proof anchors or working anchors
Proof anchors
Time of the test
Sufficiently long before starting the works
Compulsory or recommended + min. number
Tendering authority decides if investigation tests are compulsory (depends on availibility of references in the considered soil type and load range) If required : min 2 or 3
Owner decides if compulsory or not (depends on homogeneity, magnitude of the loading, cyclic loading or not, and number of tests if required If required : min 2 or 3
On each anchor
Owner decides if acceptance tests are compulsory or not, and the number if required (If required : number of tests) Pp = 1.15-1.25 * Pk (Pk : 1*Gk + 1*Qk) Pp = 1.15-1.25 * Pk (Pk : 1*Gk + 1*Qk)
Proof load(Pp)
Test method
To failure : Pp > Re (estimated resistance) (e.g. Pp=1.5 * Re) Untill =5 or 2 depending on Test Method TM1 & TM3 (ISO) (owner shall determine the test method)
1.5 *Pk load under serviceability conditions (no distinction between permanent and temporary anchors). Pk : 1*Gk + 1*Qk TM1 & TM3 (ISO) (owner shall determine the test method) Criteria: see TA 95 and DIN4125 If OK : approbation If NOK : disapproval or limited loading
Interpretation & consequences if pos/neg resultat
Rc and Ra s-p,s-t,-p qs => statistics => calculations
TM1 & TM3 (ISO) (owner shall determine the test method) Criteria: see TA 95 and DIN4125 If OK: approbation (SF proven) If NOK: reduced permissible anchor load or further tests
Pk : Characteristic value of the load Ra : ultimate load Rc : creep load Pp : proof load Proof anchor = sacrificial anchor
Gk : Characteristic value of the permanent load Qk : Characteristic value of the variable load Re : estimated resistance : creep factor
p. 15/21
BBRI & BGGG-GBMS
4.2. Design methodology

After analysis of the Belgian practice with regard to anchor testing and design methodology, 6 possible situations for anchor design, as summarised in Table 4, are proposed.
Table 4 : Possible situations deduced from Belgian Practice
(*) (**)
with comparable experience not applicable
Acceptance tests Investigation tests Suitability tests Active anchor 1 2 On M anchors on the job site on M anchors on an other similar site Or M reference tests on M anchors on an other similar site Or M reference tests / (*) / (*) / (*) / on N anchors on site / on N anchors on site / / on all anchors on all anchors on all anchors on all anchors on all anchors NA(**) Passive anchor on X anchors on X anchors on X anchors on X anchors on X anchors / Based on test results Based on test results Based on test results Based on calculations Based on calculations Based on calculations Design approach
4 5 6
Remark that the type and number of anchor tests is an important variable in this table. Furthermore a difference is made between active anchors and passive anchors. This anticipates on Belgian practice, in particular on the application of hollow bar anchors, existing out of steel with considerably lower yield strength compared to pre-stressed steel, and which are in certain conditions not always pre-stressed. For passive anchors, also a design situation (6) that is only based on a calculation rule without any testing at all is proposed.
4.2.1. Design based on test results

It is proposed to introduce different factors of safety, in order to take into account the type of tests and the number of tests performed
p. 16/21
BBRI & BGGG-GBMS
4.2.2. Design based on calculations

A possible approach to integrate the design situations, dealing with a calculation rule, in a design methodology according to the Eurocode 7 principles is suggested in Figure 7. This approach is mainly inspired on the Belgian design methodology for axially loaded piles according to EC7 that was recently published (BBRI, 2008). In this design methodology the friction resistance is deduced from the cone resistance measurements in CPT. It is suggested that the different factors determining overall safety would be dependent on: the available soil investigation (correlation factors i) the number of suitability test, (model factor Rd) the number of acceptance tests (safety factor a)
DESIGN APPROACH
Ra , d a . EXP .
J CPT results calculation rule : a, p, EXP J values Ra,calc,i
Rd i a
* p
qc
1 1 .
Rd = f(anchor system , #Suit. tests)
J values Ra,calib,i = Ra,calc,i / Rd a3 and a4 = f(#J)
characteristic value Ra,k = min (Ra,calib,mean / a3 ; Ra,calib,min / a4)
a =f(#Accept. tests) design value Ra,d = Ra,k / a Figure 7 : Proposal for an approach for the determination of the pull-out resistance
p. 17/21
BBRI & BGGG-GBMS
a) Proposal calculation rule for ultimate unit shaft friction In the formula represented in Figure 7 the ultimate unit shaft friction is qc represented by a . * .
It is proposed to start from a shaft friction calculation rule as published in qc (BBRI, 2008). The basic friction equals in that case * and p* is the empirical
factor that provides the ratio between the shaft friction and the cone resistance depending on the soil type pile (independent from pile or anchor type) For sand this basic curve is represented in Figure 8. On the same figure the experimental data from the tests in Limelette (Clayey sand and tertiary sand) are added, as well as some calculation rules that are frequently applied in Belgium. The basic friction calculated in this way needs to be multiplied by an empirical installation factor a that accounts for the installation procedure of the anchor and ground type. Based on the information obtained in Limelette and based on the contribution of De Cock (2008) in this Volume the following classes might be defined for grout type anchors: CAT 1 : Gravity filled anchors CAT 2 : Anchorage systems filled with primary pressure grouting CAT 3 : Anchorages with a global post inject grouting system CAT 4 : Anchorage with post grouting performed repetitively and selectively injection pressure (IRS) With regard to the installation factor it is obvious that these relate to each other as follows: 1 a,CAT1 < a,CAT2 < a,CAT3 < a,CAT4 Specific values are not yet put forward at this stage as this is still under discussion in the Working Group Ground Anchors. Anyway, based on the test results in Limelette it can be concluded that : The SA and 1T anchors can be situated in the CAT 1 (gravity filled anchors). The IGU anchors, of which Belgian practice differs from the French practice, are in general situated in CAT 2, although several of them, especially the anchors where the pressure grouting was not successful, are situated in CAT 1. The T1+inj system, with simple serial post-injection system can be situated in CAT 3. One system can belong to a different category, depending on execution parameters, pressure grouting in particular. This is shown by the wide spread in the results of the IGU anchors. Also for the 1T anchors, it is not excluded that, depending on drilling procedure, anchor lengths and soil type, CAT 2 anchors can be realised. With regard to the post-grouting of the SA anchors: in Limelette this post-grouting procedure was not successful although other references proved the contrary (cfr. Maekelberg et al, 2008). In phase 3 of the test campaign in Limelette, more tests on SA anchors, which have been post-grouted, will be performed soon. Actually it is anyhow not clear how to verify the effect from such post grouting (monitoring?) and how to quantify the effect of the post grouting on the anchor capacity.
p. 18/21
BBRI & BGGG-GBMS
Previous points underline the importance of the monitoring of the execution parameters, pressure grouting parameters in particular. The challenge however, is to elaborate criteria that determine in which category the anchors belong for a specific site. Actually, no conclusions could be drawn with regard to an efficiency factor dealing with a decreasing unit shaft friction with increasing anchor length.
BBRI project Ground Anchors - Result Limelette qs(qc) in Sand & Heterogeneous layer Min.-qc/65 Max - qc/33 Min.-qc/75 qs = [Pu-W tot]/(..Dnom.Lfixed) Max - qc/50 Min.-qc/65 - with real or average if not available Max - qc/50 - with real loss W tot or average if not available IGU-sand-TM3 IGU-sand-TM1 IGU-Verticaal-TM3 IGU-Verticaal-TM1 IGU-Het.-TM3 IGU-Het.-TM1 1T-sand-TM3 1T-Dred.-sand-TM3 1T-Verticaal-TM3 1T+inj-Verticaal-TM3 1T+inj-Het.-TM3 SA-sand-TM3 SA-Verticaal.-TM3 SA-Het.-TM3 2T-grav -Verticaal TA 95 - 1.15*IGU TA 95 - 1.15*IRS qc/30*1.3 NA EC7 - pile EBA Ischebeck (1.5*150kPa) 0 4 8 12 16 20 24 28 32 36 Ostermayer 40 (Lf=4-6m / m.d. to d.) qc,av. (MPa)
1000 900 800 700 600 qsu (kPa) 500 400 300 200 100 0
Figure 8 : Comparison of experimental data Limelette anchor test campaign and main rules found in the literature; for the experimental data from Limelette only the real load on the fixed length is considered (by taking into account friction losses in Lfree and the real anchor diameter)
b) Emperical factor EXP In the formula represented in Figure 7, an empirical factor EXP that accounts for an increase of the diameter of the fixed length due to the installation procedure is introduced. Based on the observations up to now it is proposed to introduce the nominal diameter of the anchor in the design methodology (EXP = 1). The nominal diameter corresponds in this case with the dimensions of the drilling tools: diameter of casing (IGU), lost oversized point (1T), diameter drill bit (SA). Hereby it should be remarked that: For the self boring hollow bar anchors (SA), some discussions still exist and some extra tests are still ongoing in order to verify in which way drilling procedure might influence the anchor diameter. For smaller diameter drill bits, some indications exist that an enlargement of the diameter is obtained, in our opinion due to the effect mentioned in the next point. In phase three of the test campaign in Limelette, the possibility exists to excavate some of the SA anchors performed with a small drill bit (76 mm) in the loam (silt) layer.
p. 19/21
BBRI & BGGG-GBMS
For the systems with flushing around casing or hollow bar (1T, SA) it has been observed that the ratio (flow rate drill fluid)/(diameter drill tool) influences the anchor dimensions, but for the moment it seems very difficult to quantify this effect in practice. A multiplication factor on the diameter (EXP > 1) may be applied if postgrouting operations are performed, or if it can be proved by the contractor. Based on the observations on the 1T+inj anchors in Limelette, it seems reasonable to propose for each successful injection stage with the simple serial injection system a factor EXP 1.10 over the anchor area where the injection points (manchettes) are present. For the IRS anchors a more detailed study is necessary. As mentioned before, in phase three of the anchor test campaign in Limelette, IRS anchors have been installed in the loam (silt) layer and will probably be excavated soon. These observations and other experimental data with regard to IRS anchors, a.o. validated in the TA-95, will be used to determine EXP factors for IRS anchors.
5.
CONCLUSIONS
In this contribution an overview has been given of the methodology to analyse the results of the anchor load tests (phase 1 and phase 2) on the Limelette test field. The ultimate skin friction qsu along Lfixed has been deduced taking into account friction losses in the anchors free length and the real anchor diameters. The results of this analysis have been linked with the anchor execution parameters and the results of the inclined versus vertical CPT in order to draw the general conclusions. Furthermore the principles of a uniform test and design approach for anchors in Belgium according to Eurocode 7 have been explained. The principles set out in this contribution take into account the Belgian anchorage practice on the one hand. On the other hand it is inspired on the methodology for pile foundations that was recently published in the Belgian recommendations for the design of axially loaded piles according to EC7 (BBRI, 2008). Herewith it is aimed to obtain a coherent application of the EC7 in Belgium, and to link the safety factors with aspects as quality assurance during execution and the number of tests on anchors that are performed. Together with other available test data, a.o. from the third anchor test phase that is going on for the moment in Limelette, this information will be transmitted to the Belgian Commission responsible for the establishment of the national annex (and background documents) of the Eurocode 7. It is the role of this commission to work out these principles and to propose values for the different factors.
p. 20/21
BBRI & BGGG-GBMS
6.
REFERENCES
BBRI, 2004-2006 & 2006-2008. Ground Anchors Establishment of a standardized design method for ground anchor taking into account execution methodology. Research program subsidized by the Belgian Federal Public Service Economy and the Belgian standardization institute, Conventions CC CCN-119 & CC CCN-169 BBRI, 2008. Richtlijnen voor de toepassing van Eurocode 7 in Belgi. Deel 1 : het grondmechanisch ontwerp in uiterste grenstoestand van axiaal op druk belaste funderingspalen / Directives pour lapplication de lEurocode 7 en Belgique. Partie 1 : Dimensionnement gotechnique ltat limite ultime de pieux sous charge axiale de compression. Document available in Dutch and French on www.tis-sft.wtcb.be and www.bggg-gbms.be (English version will be available by mid-2008) CEN, 2004. Pr EN ISO 22477-5 Geotechnical investigation and testing Testing of geotechnical structures Part 5: Testing of anchorages CFMS, 1995. Tirants dancrages Recommandations T.A. 95 concernant la conception, le calcul, lexcution et le contrle, Editions Eyrolles, Paris De Cock, F. 2008. Ground Anchors : overview of types, installation methods and recent trends, Proceedings of the international symposium on ground anchors, May 14th 2008, Brussels Fellenius, B.H. 2001 From strain measurement to load in an instrumented pile, Geotechnical News, March 2001. Maekelberg, W., Bollens, Q., Verstraeten, J., Theys, F., De Clercq, E. & Maertens, J., 2008. Practical Experience of TUC RAIL with Ground Anchors and Micro-Piles, Proceedings of the international symposium on ground anchors, May 14th 2008, Brussels Ostermayer, H. & Barley, T. 2003. Ground anchors : paragraph 2.5 of the Geotechnical Engineering Handboek Vol. 2 procedures, edited by U. Smoltczyk and published by Ernst & Sohn, 2003, Berlin. Tomboy, O. & Huybrechts, N. 2008. Excavation of the ground Anchors: measurements and observations, Proceedings of the international symposium on ground anchors, May 14th 2008, Brussels
p. 21/21
BBRI
BGGG GBMS
Practical Experience of TUC RAIL with Ground Anchors and Micro-Piles
Ir Maekelberg Wim Ir Bollens Quentin Ir Verstraelen Jan Ir Theys Frank Ir De Clercq Eric TUC RAIL N.V., Belgium Design Department, Unit Soil and Water
Prof. Jan Maertens Jan Maertens bvba & Catholic University of Leuven (KUL)
BBRI & BGGG-GBMS
1.
INTRODUCTION
The latest years, the most important jobsites of TUC RAIL are the realizations of the high speed railway in Belgium and the regional express network around Brussels. The high-speed railway, presently constructed, passes through Belgium and links Paris-Brussels-Lige-Kln and Paris-Brussels-Antwerp-Amsterdam. At this moment, the first jobsites of the regional express network around Brussels are being started. For these projects, new constructions are often realized next to the existing tracks in service, so special care must be given to the stability of those tracks. Soil anchors and micro-piles are often used to limit the deformations of the retraining walls and to stabilize embankments of pore quality. Furthermore, new constructions often have to be realized within limited work space, which has an impact on the installation methods used to install the foundations. As micropiles can be installed with little rigs to great depths, these techniques are often used in those situations. Soil anchors and micro-piles also have an important implication on the stability of the construction, so different kinds of in situ-tests are necessary to verify the calculated bearing capacity and the integrity of those elements. The results of these tests as well as the applied design methods and some practical experiences, are discussed in this article.
2.
THE USE AND INSTALLATION OF GROUND ANCHORS AND MICRO-PILES FOR RAILWAY INFRASTRUCTURES
Ground anchors and micro-piles are used for various purposes and applications. The choice between both is mainly determined by their specific application or the amount of working space available for installation. Drilling techniques for the installation of both are similar. However, their configuration and purpose are different: Ground anchors are tension only and consist of an anchor head which fixes the anchor to the structure, an unbonded, or free section, which causes the anchorage force to shift to deeper soil parts, near the bonded, or fixed section. Distinction is made between active and passive ground anchors. Active ground anchors are always pre-tensioned. The ground anchor is fixed to the structure with a certain pre-tensioning force, which causes an elastic elongation of the ground anchor to occur. At this equilibrium condition, the reinforced structure will no longer be subjected to these displacements, which also causes less settlement to occur behind the structure. This in term is a big advantage for existing railway infrastructures. Passive ground anchors are not pre-tensioned. These are used when it is not desirable to transfer a load to the tied-back structure or when pre-tensioning is not necessary to limit deformation of the structure.
2.1. Definitions
Practical experience of TUC RAIL with ground anchors and micro-pilesMaekelberg et al.
p. 3/42
BBRI & BGGG-GBMS
Micro-piles are structural members to transfer actions, both tension and compression, to the ground. Micro-piles have a small diameter (smaller than 300 mm outer diameter) and can be installed with small rigs. They are fully grouted and the anchorage force is developed along its full length. Since the build-up of the anchorage force occurs immediately behind the wall, there is no use in applying a pre-tensioning force. This is referred to as passive anchorage. Soil nails fall in this category.
2.2. Drilling of ground anchors and micro-piles

The following drilling techniques can be used: 1. Flushed rotary drilling with singular casing, 2. Flushed rotary drilling with dual casing, 3. Drilling with self-drilling rods. These different techniques are described in F. De Cock 2008 [1]. TUC RAIL only accepts techniques 2 and 3 to be used on its sites. The main reason for this is that most drillings have to be carried out in the railway embankment, which is usually of intermediate quality. The embankments exist mainly of loose, loamy sands or a mixture of ballasting materials and incineration residues. For tied-back walls next to or near existing structures which are sensitive to settlements, such as active railways, it is necessary to limit the possibility of borehole instability and consequently, settlements. In the case of flushed rotary drilling with a dual casing, the outer casing ensures the borehole stability. When drilling with self-boring rods, the drilling cement fluid 1 itself stabilizes the borehole. During the drilling process, a cake is formed on the borehole sides, on which the drilling cement fluid exerts a stabilizing excess pressure which keeps the borehole open. An added advantage of drilling with a cement fluid is that any voids or loosely packed soils in the drilled soils are filled. A W/C-ratio of 1 is often used as drilling cement fluid. However, the W/C-ratio depends on the nature of the drilled soil layers and depth of the drilling. The lightest cement fluid allowed on sites of TUC RAIL, has a W/C-ratio of 2, and is used for drilling through homogeneous clay. The fine grains present in clay soils mix with the drilling cement fluid, fulfilling a similar function as the already present cement in stabilizing the borehole. With increasing depth, it becomes harder and harder to circulate the drilling cement fluid (and clear soil particles) to the surface, which limits the use of this technique to a depth of about 30 m. Both the flushed rotary drilling with dual casing as the drilling with self-boring rods can be performed with rigs with a width of 3 m, length of 7 m and a boom height that varies with the working conditions from 2 to 7 m (see figures 1a and 1b). When the working area is limited, the use of even smaller rigs may become necessary. Often used smaller rigs have a surface area of 1,5 m by 3 m. The smallest rig has a width of only 1,0 m and a length of 2,5 m. Photos 1a and 1b show two confined working conditions and the use of small rigs.
1
Drilling cement fluid = mixture of cement and water with a given water/cement-ratio (W/Cratio)
p. 4/42
BBRI & BGGG-GBMS
When access to the site is difficult, the boom can also be fixed to an excavator, as shown in photo 2. This gives the ability to work at greater heights and at further distance from the working platform. In some cases, the ground anchors and micro-piles are installed by hand, as shown in photo 3. The boom is fixed to a small and mobile structure. However ground anchors and micro-piles installed with smaller rigs are limited in borehole diameter, depth and bearing capacity.
Figure 1a: Rig for installation of ground anchors and micro-piles Cross section
Figure 1b: Rig for installation of ground anchors and micro-piles Plan view
p. 5/42
BBRI & BGGG-GBMS
Photo 1a: Installation of micro-piles Rig width 1.5m
Photo 1b: Installation of ground anchors Rig width 1.0m
p. 6/42
BBRI & BGGG-GBMS
Photo 2: Installation of nails Boom fixed on excavator
Photo 3: Installation of nails Manual device
2.3. Reinforcement and injection of ground anchors and micro-piles

After drilling of the borehole, the drilling cement fluid is always replaced by a injected cement fluid with a W/C-ratio of 0,6 for temporary ground anchors, and 0,5 for permanent ground anchors. Reinforcement depends on the drilling technique or function of the ground anchor and micro-pile.
p. 7/42
BBRI & BGGG-GBMS
Ground anchors drilled with the dual casing technique can be reinforced with soft steel rods, S460N, or strands. When strands are used, the free length of each strand should be sealed within a PVC tubing filled with anti-corrosive grease. Ground anchors installed with self-boring rods are always reinforced with a hollow rod in soft steel, S460N. The hollow rods have a continuous thread along their full length. Independent of drilling technique, the free length is realized with a smooth PVC piping across the soft steel reinforcement. For temporary ground anchors, these smooth pipes can be placed along the reinforcement rods, between the couplers. The inner diameter of the pipes is a little smaller than the outer diameter of the couplers, and the length of the pipe is a little longer than the theoretical distance between the couplers. This ensures a tight seal between the couplers and pipe. For permanent ground anchors, the smooth pipe needs to be placed continuously along the reinforcement rod and couplers to avoid long-term corrosion around both the rods and couplers. The inner annulus of the PVC pipe needs to be filled with cement fluid or any other corrosion protective material. For permanent ground anchors, the use of strands is not allowed since the use of a corrosion protection agent on these strands can lead to an improper installation and these types of reinforcements tend to fail in a more brittle way and are more susceptible to corrosion, certainly in railway applications. The strands work as separate elements, and with corrosion of the strands and failing of one strand, the force is distributed among the remaining strands, which are also subjected to corrosion. This can lead to failure of the ground anchor as a whole. Micro-piles are always reinforced with rods of soft steel, with a continuous thread along their full length. For permanent micro-piles, a corrugated PVC pipe is placed along the head of the pile in the fresh cement fluid. The length of this pipe amounts to 2 m, of which 1,5 m is placed within the fresh cement fluid of the pile. After execution of the pile, a lowering of the grouted top level can occur, this has to be replenished within 24h. During replenishment of the micropile, the inner part of the corrugated PVC pipe needs to be filled to a level of 0,5 m above the top of the pile. Both insure that settlements will not cause the hole to collapse and that the reinforcement rod is grouted along its full length. For inclined micro-piles, the corrugated PVC pipe aids in protecting the reinforcement at the pile cut-off level (as shown in figure 2). When micro-piles are anchored in foundation footings, the corrugated PVC pipe is cut off at about 10 cm above the bottom level of the footing (see figure 3). This section is anchored to the foundation footing and presents a corrosion protection against infiltrating water that is situated beneath the foundation. For micro-piles which are attached to a wale, the PVC pipe is removed till the outer limit of the wale (see figure 2). It is most important to guarantee the grout coverage of the reinforcements. For this purpose, spacers are used for both permanent and temporary ground anchors or micro-piles. These are placed every 3 m and must guarantee coverage of at least 15 mm at the couplers. Minimal coverage at the reinforcement rods or strands is 25 mm. For permanent ground anchors, the reinforcement is limited to soft steel to which an excess thickness is added to account for corrosion. This excess thickness amounts to 2 mm.
p. 8/42
BBRI & BGGG-GBMS
Figure 2: Anti-corrosion protection for ground anchor Ribbed tube
Figure 3: Anti-corrosion protection for micro-pile Ribbed tube
After drilling the borehole for the ground anchor or micro-pile with a dual casing, relaxation of the surrounding soil can occur. This is compensated for by an injection of cement fluid every 2 m while pulling the casing (IGU, Injection Globale et Unique sous faible pression, Global and unique injection under low
p. 9/42
BBRI & BGGG-GBMS
pressure pi) (0,5pl<pi<pl 2 and pi>1,0 MPa) [2]. Alternatively, IRS injection (Injection Rptitive et Slective sous pression leve, repeatedly and selective injection under high pressure) (pi>pl and pi<4 MPa) [2] uses specialized equipment to inject through injection tubes at selected injection sleeves (manchettes), one by one. This has so far been used only once at sites of TUC RAIL. Sometimes, separate injection pipes (min of 2) are added to the reinforcement with injection openings every 0,5 m, which cant be individually selected for injection. This enables a final injection with a minimum of two phases. With this type of installation, the injection opening with the least resistance will open first, and others may not open at all. So it can thus not be guaranteed that the injection is carried out across the entire length of the ground anchor. Since injection is carried out with a minimum of 2 phases, this technique can only be regarded as an enhanced IGU injection. In the case of self-drilling rods, relaxation of the soil around the hole is compensated for by performing a post-injection through the inner annulus of the reinforcement rods. This injection is done with a cement fluid with a W/C-ratio of 0,6 for temporary anchors and 0,5 for permanent anchors. The waiting time between installing the ground anchor or micro-pile and the post-injection depends mainly on soil types and length of the ground anchor or micro-pile, and varies between a minimum of 30 min and a maximum of 3 h. With every post-injection, it is attempted to inject around 5 l/m of anchor at a moderate injection rate. The maximum injection pressure is 60 bars. Surface breakthrough must be avoided during post- injection.
2.4. Applications
Ground anchors and micro-piles are used for various purposes and different structures, both as temporarily and permanent reinforcements and foundations. Temporary applications consist mainly of: Tied-back walls (see figure 4). The primary functions of the ground anchor are to ensure the stability of the construction and to limit displacements of the wall as well as the accompanying settlement behind the wall. This is an important issue next to active railways, Temporary foundations of a building to allow excavations close to the building (see figure 5). This is often the case with soil remediation, Soil nailing for retaining construction pits (see photo 4). Foundations of bridges and overpasses where limited working space is available for the installation of piles, Deepening of foundations and walls (see figure 6), Tying-back of walls (see figure 7), Stabilizing existing railway embankments (see figure 4).

2
Permanent applications are mainly:
Pi = injection pressure; pl = Mnard limit pressure (pressiometer)
p. 10/42
BBRI & BGGG-GBMS
Figure 4: Anchoring of retaining wall next to active railway
Figure 5: Temporary foundations of a building
Photo 4: Construction pit with soil nailing
p. 11/42
BBRI & BGGG-GBMS
Figure 6: Deepening of wall foundations
Figure 7: Tied-back retaining wall with secans-piles
p. 12/42
BBRI & BGGG-GBMS
2.5. Mechanical ground anchors/micro-piles

In some cases, the ground anchors must be able to accommodate a tensile force immediately after installation. This can be necessary to avoid long hardening times of classic cement grouted ground anchors, and thereby increase efficiency and speed of installation of the entire structure. Two types of mechanical ground anchors are used: screwed and plate dowels (see photo 5a and 5b). This type of foundation needs larger displacements to mobilize a same amount of bearing capacity. This is inherent to the concept. At present, this type of reinforcement has only been used in specific cases. This type of ground anchor is not discussed further in this article.
Photo 5a: Screwed dowels
Photo 5b: Plate dowels
3.
DESIGN OF GROUND ANCHORS AND MICROPILES (TUC RAIL)

When designing ground anchors or micro-piles, both practical and design considerations are taken into account.
3.1. Practical considerations to the design

Practical considerations are mainly the accessibility of the construction site and available surface area for installing the ground anchors and micro-piles. Based on local conditions, the appropriate equipment must be used while optimizing construction efficiency. This must take into account the capability of the equipment with respect to the drilling diameter and attainable depth. In some
p. 13/42
BBRI & BGGG-GBMS
cases, the use of different types of rigs may become necessary to attain a feasible design. E.g. the use of smaller rig may be required to install the first rows of nails in a deep cut, hereby creating a working platform for bigger rigs to carry out the remaining part of the installation. These first rows may be installed with manual equipment, to a limited depth (see photo 3 and figure 8a and b). After installation of these first rows, a working platform can be created on which a bigger rig can install longer nails between the shorter, manually installed, nails, insuring stability of the complete deep cut (see figure 8c). These secondary nails can be installed with a boom fixed to an excavator, which enables the machine to reach greater heights. Progressive cuts may make use of more regular rigs, increasing construction efficiency (see figure 8d).
Figure 8a: Nailed Wall 1st phase: manual installation
Figure 8b: Nailed Wall 1st phase: manual installation - Detail
p. 14/42
BBRI & BGGG-GBMS
Figure 8c: Nailed wall 2nd phase: nails executed with boom fixed on excavator
Figure 8d: Nailed wall other phases : nailed installed with normalrigs
3.2. Calculation of ground anchors and micro-piles

The bearing capacity of a ground anchor or micro-pile is calculated based on the expected loads and combination of loads. A distinction is made between permanent loads and variable loads. Permanent loads consist of the weight of the structure (e.g. the weight of the bridge and abutments), soil loads that are transferred to the construction by friction along the contact, as well as any pre-tensioning load applied. Variable loads are loads linked to traffic loads or possible thermal loads. For railway constructions, the most conservative loads as caused by UCI convoys are taken into account, including accompanying braking and acceleration loads. For the case of ground anchors, the variable and permanent loads are added to obtain a maximum load in service limit state (SLS). This maximum load is compared to the resistance of the ground anchor to failure, for which a factor of safety of 1,7 (temporary ground anchor) to 2 (permanent ground anchor) is required (see [2], annex 3, pp 139-150).
p. 15/42
BBRI & BGGG-GBMS
To minimize the risk of failure of the structure, the design of retaining walls has to ensure that at least 2 ground anchors are placed on each wale. For permanent ground anchors, the maximum load is limited to about 350 kN per ground anchor. This way, the structure is anchored by several ground anchors, which minimizes the effect of failure of one of the ground anchors. The minimum length of the fixed section of the ground anchor is 4 m. To determine the free length of a ground anchor, the safety against deep sliding needs to be determined. The ground anchor is always anchored behind the slip plane with the lowest safety. In the case of a single row of ground anchors, the KRANZ method can be used to perform the stability check (see [2], annex 1, pp 117-128). However, the minimal (practical) free length amounts to 3 m. With micro-piles, the different permanent and variable loads are summed to obtain a maximum load for each loading condition. This is done by: Fmax = PL FPL + ML FML With Fmax = maximum load of each loading condition (kN) = weighing factor on permanent loads = permanent load taken into consideration (kN) = weighing factor on variable loads = variable loads based on the conservative position of UCI convoys and wind (kN) (1)
PL
FPL
ML
FML
Table 1 summarizes the weighing factors for each loading condition (see [3], Chapter 5, pp 22-25). Table 1: Summary of different loading conditions
PL
ML
1.35 1.00
1.00
1,45 1,00 0,00
FULSfundamental FSLScharacteristic
FSLSpermanent
RULSfundamental RSLScharacteristic
RSLSpermanent
For each load condition these maximum loads are compared to the minimal required resistance of the micro-pile, as indicated in table 1. In each loading condition, the resistance of the micro-pile requires a specific factor of safety against failure of the micro-pile. Rug Rus R ; ) R = u = min( (3) S Sg Ss
With Ru Rug = failure of the ground anchor or micro-pile (kN) = geotechnical failure resistance of ground anchor or micro-pile (kN)
p. 16/42
BBRI & BGGG-GBMS
Sg Rus Ss
= specific factor of safety against Rug (see table 3) = structural failure resistance of the ground anchor or micro-pile (kN) = specific factor of safety against Rus (see table 4)
The minimum failure resistance Ru of the micro-pile is determined as the minimum of the structural failure resistance Rus (failure of the pile itself) and the geotechnical failure resistance Rug (failure between pile and soil) (see [4], informative annex D), as can be deduced from formula (3). Ru = min( Rug , Rus )
(4) failure resistance according to
The
Rug is calculated Recommandations Tirants dancrage 1995 [2] as follows: Rug = Dd Li q sui
i
geotechnical
(5)
With
Dd
= drill bit diameter expansion factor = diameter of drill bit (m)
Dd = diameter of drilled hole (m)

Li q sui
= fixed length of the ground anchor or micro-pile in the considered soil (m) = unit friction resistance of the ground anchor or micro-pile in the considered soil (kN/m2)
The unit friction resistance q sui is always based on soil investigation. In Belgium, soil investigation consists mostly of CPT tests. In certain cases, also pressiometer test results are available. TUC RAILs view on soil investigation was already discussed in W. Maekelberg 2003 [5]. Based on the soil identification obtained through drillings, and the cone resistance from CPT tests, the unit friction resistance is determined by abacuses for each soil type [2]. The cone resistance used to determine the unit friction resistances is an average value for each distinct soil layer. Based on experience from pull-out tests on ground anchors and micro-piles in similar soil conditions, these values can be modified. The value to be used depends on soil type and technical installation. Table 2 presents an overview of the value as described by TA95 [2].
p. 17/42
BBRI & BGGG-GBMS
Table 2: Overview of values in common soils

IGU (0,5pl<pi<pl and pi>1,0 MPa) value value (TA 95) (TUC RAIL) 1,3 1,4 1,3 1,1 1,2 1,2 1,1 1,2 1,2 1,2 1,1 1,1 1,2 1,1 1,1 1,05
Gravel Sand Loam Clay Marl & Chalk Rock
Experience with Belgian soils shows that the q su -values as reported by Bustamante are low in the case of sandy soils [2]. Therefore, when calculating ground anchors and micro-piles, an upper limit for the value is used. For the clayey soils, the q su -values as reported by Bustamante are realistic. However, taking into the account that most Belgian clays are rather stiff in nature, the specified value is rarely obtained. Therefore, a smaller value is used. For vertical tension piles, the geotechnical failure resistance
Rug is the
minimum of the geotechnical failure resistance as discussed above and the active weight around the tension pile. Therefore a check of the active weight around the tension pile must be performed (see figure 9). This is detailed in [2], annex 2, pp 129-138.
Figure 9: Active weight around tension piles
The specific factors of safety against Rug are summarized in table 3.
p. 18/42
BBRI & BGGG-GBMS
Table 3: Specific factors of safety against Rug for each load condition Sg RULSfundamental RSLScharacteristic RSLSpermanent
1,40 1,54 = 1,40 1,10 1,96 = 1,40 1,40
For temporary conditions, the safety factor for the SLS loading condition is reduced to 1,7 instead of 1,96. The structural failure resistance Rus of a ground anchor or micro-pile is determined by the steel section of the reinforcement. The structural failure resistance is calculated as follows:
Rus = f y As
(6)
With
fy As
= elastic limit of steel (N/mm2) = net steel section, after deduction of excess thickness to account for corrosion (mm2)
The specific factors of safety against Rus for each loading condition are summarized in table 4.
Table 4: Specific factors of safety against Rus for each load condition Ss RULSfundamental RSLScaracteristque RSLSpermanent
1,15 1,25 1,54
For permanent ground anchors, only soft steel reinforcement is used. The steel section incorporates an excess thickness of generally 2 mm to account for corrosion. Up to 4 mm of excess thickness is used for permanent pre-tensioned ground anchors.
4.
FIELD TESTS MICRO-PILES
ON
GROUND ANCHORS AND
4.1. Definitions
The following different types of tests can be performed (see [2], chapter 5, pp 69-75 and chapter 6, pp 83-103) and [4]:
Investigation test (according to EN 1537): these tests are performed in statistically sufficient numbers in the vicinity of the construction site, in similar soils as the final installed ground anchors or micro-piles. These tests are to be carried out well before start of the construction and until failure.
p. 19/42
BBRI & BGGG-GBMS
Based on interpretation of the test results, a characteristic q su - value can be determined which can be used to design the ground anchors and micro-piles.
Suitability tests (according to EN 1537): these tests are carried out at the start of the construction, just before start of the further installation, to verify the assumed q su -value and to validate the used installation method. For each encountered soil type, at least 2 tests are performed till failure. Control tests: these tests are carried out on installed ground anchors or micro-piles to verify the correct installation. The maximum load during these tests is however limited. One test is carried out for each 40 instalments, with a minimum of 2 per job site. Acceptance test (according to EN 1537): these tests are performed on all active ground anchors, just before pre-tensioning them.
Since investigation tests require a separate early tendering, these tests are rarely carried out by TUC RAIL. The design of ground anchors and micro-piles always makes use of the recommendations in TA 95, as explained above. Suitability tests are used to verify the assumptions made in the design calculations, and are thus usually carried out at the start of each construction. For jobsites where only micro-piles are used, the suitability test is the only test performed. When ground anchors are included, control and acceptance tests are carried out along with the suitability tests.
4.2. Suitability and control tests on ground anchors and micro-piles

All tests are carried out on ground anchors and micro-piles, which are installed with the same technique, and at the same angle as planned for the final construction. To eliminate the friction in the top soft layers, a smooth PVC pipe is inserted across the reinforcement (rod and couplers) along these soft layers. Alternatively, the friction in these soft layers is taken into account when determining the applied load. After sufficient curing (minimum 8 days), the loading test is carried out on the ground anchors and micro-piles to a load of twice the calculated service load for suitability tests, and 1,30 time the service load for control tests. The execution of these tests is always confided to a certified laboratory. Since the calculation of micro-piles always neglects the bearing capacity at the tip, tensile test are performed for both ground anchors and micro-piles. The reaction device to perform the tests must always be able to withstand the applied load, and must be designed as such. The test set up is shown in figures 10a and b and photo 6. It needs to be noted that while performing a load test up to 500 kN, independent of the presence of a free length, a circle of at least 1 m radially around the ground anchor or pile, must be free of any support of the reaction device or measuring point. For maximum loads larger than 500 kN, this free zone extends to 2 m around the ground anchor or pile.
p. 20/42
BBRI & BGGG-GBMS
Figure 10a: Reaction device for pull out test Cross section
Figure 10b: Reaction device for pull out test plan view
p. 21/42
BBRI & BGGG-GBMS
Photo 6: Reaction device for pull out test micro-pile
Figure 11 shows the test set up for an inclined ground anchor. The load is applied with a hydraulic piston which is operated manually with a feeding pump. The flow rate of the pump allows precise control of the applied pressure, which needs to remain constant during each loading step. Since the difference between consecutive load steps may be significant in some cases, the use of electric pumps in parallel with the manual pump is allowed. In any case, the pumps must avoid any abrupt changes in pressure to occur and certainly any temporary overloads.
Figure 11: Reaction device for pull out test ground anchor
p. 22/42
BBRI & BGGG-GBMS
The measurement of the applied load must be accurate within 2 % of the theoretically applied load. The system (jack and measuring device) must be calibrated every 6 months in an independent certified laboratory and is always accompanied by a calibration certificate. The test procedure is detailed further in the following references [6] and [7]. For ground anchors, also the French standard NF P94-153 [8] is used. The loading stages, as described in the above references, are adapted as follows:
At the start of the test, a load Q0 is applied to secure the piston to the reaction device with a certain pressure. At load Q0, a reference measurement of the displacement is taken. The maximum load Qmax is applied in 10 equal stages, each lasting for 60 min (investigation tests) or 30 min (suitability tests). The duration of stages 0,1 . Qmax and 0,3 . Qmax is limited to 1 min, with measurement of the displacement. When failure does not occur at the maximum load of Qmax, two extra loads are applied (1,1 . Qmax and 1,2 . Qmax), as far as the stress in the reinforcement remain below 90% of the yield limit of the steel. After the final loading stage, the ground anchor or pile is unloaded in steps of 0,2 . Qmax each having a duration of 5 min. Displacements are measured at 0, 1, 2 and 5 min. The final stage, Q0, is maintained for 10 min with an extra displacement measurement at 10 min.

The unloading stages are as follows:

For interpretation of the results, one can refer to [2] and [6]. When the maximum applied load approaches or equals the ultimate resistance, the failure resistance can be determined. The service limit resistance of the ground anchor or micro-pile corresponds to:
RSLS 0.90 Rc for temporary ground anchors and micro-piles RSLS 0.80 Rc for permanent ground anchors and micro-piles
with = yield of the ground anchor or pile, as determined by the test (kN) When the maximum applied load is not close enough to the ultimate resistance of the ground anchor or micro-pile, the yield of the ground anchor or micropile can not be deduced. Then the service limit resistance corresponds to the highest load at which the elongation e between times t+5min and t+60min(30min) meets the following requirements: e min(1,25mm;10 4 L free ) for temporary applications,
Rc
e min(1,00mm;10 4 L free ) for permanent applications.
with
L free
= the free length of the groundanchor (m)
p. 23/42
BBRI & BGGG-GBMS
4.3. Additional tests on ground anchors and micro-piles

4.3.1. Validation of the free length ([4], 9)
In this type of test, the ground anchor is subjected to a minimum of 3 loadingreloading cycles between 0 and the tensioning force. This allows drawing up the force-displacement curve. The apparent free length, as determined from the force-displacement curve, subtracted with the tensioning length, as the distance between the fixation point at the tierod (nut) and the clamp within the tensioning device, is compared to the specified free length as indicated on the construction drawings. Acceptance criteria are as follows:
The apparent length minus tensioning length should not exceed the sum of the specified free length and half of the specified tensioning length, as indicated on the construction drawings. Alternatively, they should not exceed 1,1 time the specified free length. The apparent length minus tensioning length may not be smaller than 80% of the specified free length.
At the start of construction, the first 5 ground anchors are tested. Afterwards, one out of 20 ground anchors is tested. At each of these tests, the loss in tensioning force at the anchor nut is determined, to take this into account when tensioning the tierods.
4.3.2. Acceptance tests on ground anchors

Before tensioning the ground anchors, the necessary tests (suitability tests, control tests, check of the free length, check of the tensioning losses) need to be performed and approved. Tensioning can only begin at least 8 days after installation (just as well as the tests). Before tensioning the ground anchors to 80% of their service load, all ground anchors are subjected to an acceptance test at which the ground anchors are tensioned to a force of 1,3 times the service load. During these fixation tests, the elongation of the ground anchor is measured during 15 minutes. The measured displacements between the measurements at 3 and 15 minutes, need to be smaller than 1 mm. If this doesnt occur, the further procedure is described in [2].
4.3.3. Measurement of time dependency of stress

For some important constructions, certain control measurements are taken at the ground anchors. During the first year after tensioning the tierods measurements are taken every 3 months. During the second year after tensioning, a measurement is performed every 6 months. Finally, during a 5 year period, annual measurements are performed. In a first instance, ground anchors that were subjected to a control test are fitted with an analogue measuring gauge (see photo 7). However, other ground anchors can be equipped as well.
p. 24/42
BBRI & BGGG-GBMS
Photo78: Analogue measuring gauge
5.
CASE STUDY 1: SOIL NAIL REINFORCED WALL, ANTWERP-BERCHEM

In order to construct the HSL-connection between Brussels and Amsterdam, as well as to increase the capacity of the Antwerp railway station, a tunnel beneath the central station was constructed between 1998 and 2007 (see figure 12). The access ramp is situated between the Berchem and Antwerp stations, and is situated beneath the existing railway embankment.
5.1. Introduction
Figure 12: Tunnel under Antwerp Cross section
The first construction phase (1998 2001) consists of the realisation of the tunnel side walls and roof. This was done in two sub phases. As separation
p. 25/42
BBRI & BGGG-GBMS
between both work areas, a soil nail reinforced wall is situated across the entire length between the work area and an active railway between Berchem and Antwerp station (see figure 13). The soil nails were installed with self-drilling rods. During the construction, several suitability tests were carried out, and some installed nails were excavated and measured.
Figure 13: Tunnel under Antwerp - Sub phase 1 Cross section
5.2. Suitability tests

5.2.1. Installation of nails
The nails were drilled, as described above, with the exception that:

The drilling cement fluid (W/C=1) was not replaced with an injected cement fluid (W/C=0,6) at the end of the installation. Post-injection was carried out with a W/C = 1.
The nails were installed at a 20 angle and were anchored in the fill between the supporting walls of the railway embankment (see figure 13). The hollow reinforcement rods have an outer diameter of 30 mm and inner diameter of 16 mm. The drill bit had a diameter of 75 mm. The length of the nails was limited due to the distance to the supporting walls. To check the pull-out failure of the nails, a total of 8 test-nails were installed and subjected to suitability testing. These test-nails are, just as the final nails, anchored in the fill between the supporting walls. This embankment consists mainly of heterogeneous fill and sand, with an average cone resistance of about 2 MPa (see figure 14).
p. 26/42
BBRI & BGGG-GBMS
Figure 14: Tunnel under Antwerp - Soil characteristics
The test-nails have a free length, obtained with a PVC pipe along the first 2 m of the tie rod. Every test nail had a fixed length of 3 m. Both the free and fixed lengths are minimal lengths to avoid side effects of the embankment walls (see figure 13). The test procedure used was described in paragraph 4. The maximum load was applied in 8 or 10 steps, each lasting 30 min. A maximum load with a value of 2 times RSLS , or thus the theoretical failure load, was proposed.
5.2.2. Interpretation of results

Table 5 summarizes the RSLS -values of the 8 test-nails, with their accompanying displacements. The indicated unit friction was determined by assuming that the full tensile force was equally distributed along the fixed section of the nail (3 m) and with a borehole diameter of 0,15 m, as suggested by the manufacturer of the nails.
Table 5: RSLS -values for test nails RSLS -values
Load Force per meter Unit friction Displacement (kN) (kN/m) (kN/m) (mm) V17-1 64,0 21,33 45,3 3,30 V19-1 80,8 26,94 57,2 4,60 V22-1 94,4 31,47 66,8 6,70 V22-2 123,2 41,06 87,12 9,30 V23-1 128,4 42,80 90,8 6,80 V23-2 56,0 18,67 39,6 0,72 V23-3 134,4 44,80 95,1 7,50 V26 1 103,2 34,4 73,0 4,70
These results are presented graphically in figure 15 and annex A. When testing nails V17-1 and V23-2 failure occurred prematurely. A check of the injection form showed that injection pressure of nail V17-1 was only 5 bars. Post-injection of nail V23-2 was carried out immediately after installation, so no pressure could be built up at the tip of the nail, which also leads to a lack of compaction of the surrounding soil. This explains the lower service resistance of these nails. Both nails also have a lower stiffness as compared to the other nails (see figure 15). Based on the load-displacement curves for the other nails (see figure 15), it can be concluded that the behavior of the nails is reproducible for loads smaller than
p. 27/42
BBRI & BGGG-GBMS
the service load. Only when a nail approaches its failure load, the behavior starts to become more diverse. It can also be noted that elongation of the nail quickly rises above the elongation of the free length + of the fixed length. This means that the force in the nail is rather quickly transferred towards the tip of the nail. Both observations and the failure of the nails clearly shows the importance of the post-injection to attain the specified ultimate failure resistance, especially when the nails are placed in loose sands.
Figure 15: Tunnel under Antwerp - Test results Force displacement diagram
5.3. Observations after excavation of some nails

In a second phase, the active railway was shifted towards the already finished part of the tunnel roof. At this stage, the soil nail reinforced wall was broken up (see figure 16). This demolition was used to carefully excavate a number of nails.
5.3.1. Excavated nails and applied measurements

A zone of 4 nails was excavated (see photo 8a and b) partially by hand, and partially with a small excavator, to minimize damaging the nail. All exposed nails had a length of 6 m. The excavation showed that the upper part of the nails was drilled through hard heterogeneous fill material. The bottom part against the supporting walls was embedded in moderate homogeneous sands. After exposure of the nails, the periphery was measured at several distances from the bottom level. With these measurements, an average diameter for the nail was calculated. Based on the outer diameter of the reinforcement rod (30 mm), the grout coverage could be deduced. The diameter of the nails as a function of the distance from the nail top is shown in figure 17a and b.
p. 28/42
BBRI & BGGG-GBMS
Figure 16: Tunnel under Antwerp - Sub phase 2 Cross section
Photo 8a: Tunnel under Antwerp - Excavation of nails
p. 29/42
BBRI & BGGG-GBMS
Photo 8b: Tunnel under Antwerp - Excavation of nails
Figure 17a: Tunnel under Antwerp - Excavation of nails Measured Diameters of nails
p. 30/42
BBRI & BGGG-GBMS
Figure 17b: Tunnel under Antwerp - Excavation of nails Shell extent of nails
5.3.2. Observations based on measurements

The photos of the exposed nails are shown in annex B. Based on the observations during the excavation and the measurements of the diameter, the nail length can be divided in 3 sections:
A first part of the nail was difficult to expose without damaging the grout. Measurements of the diameter were not possible. However, during excavation, it could be noticed that this part of the nail was significantly smaller than the rest of the nail (see figure B1). In the second section, the diameter remains fairly constant (see figure B2). The third section corresponds to the grout plug formed with the final injection. This plug was very irregular in shape with long branches for 3 exposed nails. One nail showed little difference between section 2 and 3. Figures B3 to B6 show the different nail tops.

Table 6 summarizes the measured values and soil types for each of the 3 sections.
Table 6: Average nail diameters
Distance to nail top Soil Consistency Diameter Average diameter Average grout coverage Section 1 0 2,00m Heterogeneous soil Dense 30 40mm Section 2 2,00 5,50m Sand Moderate 70 130mm 85mm 27mm Section 3 5,50 6,00m Sand Moderate 110 260mm -
Table 5 and Figures 17a and b clearly show that the specified diameters were nowhere attained in this soil (i.e. 150 mm for a drill bit of 75 mm). With the final injection, a grout plug was clearly formed in 3 out of 4 cases. The diameter of these plugs varies from 160 to 260 mm. The size of these plugs can probably be linked to the consistency of the surrounding soils.
p. 31/42
BBRI & BGGG-GBMS
During the excavation, it could also be noted that the grout in section 2 showed an inner void between the outer limit and reinforcement (see figure B7 and B8). Cutting of the nail showed the presence of pure water under pressure, even though the nails were not drilled through a groundwater table. These observations can be explained by the fact that a W/C-ratio of 1 is too high to obtain a solid grout body. The high W/Cratio causes the cement fluid to become unstable, and as a result, hardening of the grout causes a decantation of cement particles. The water filling the void in the first phase, has evacuated through the outer shell of the grout. The cement shell present at the outer limit of the void, is caused by the cake formation during drilling of the nail. At the inside of the rebar, a similar decantation occurred, while the water remained present until cutting of the nail. To avoid such voids, the use of a W/C-ratio < 0,70 was advised, as noted in reference [11]. For loose soils, a W/C-ratio of 0,5 is advised for both the drilling cement fluid as injected cement fluid.
5.4. Conclusion case study 1

In order to asses the tensile strength of the nails, a number of suitability tests were carried out. The results of these tests lead to the conclusion that the installation details for the realization of the nails are of great importance to the final tensile resistance. The load displacement curves show that the tensile force is quickly passed on towards the tip of the nail, which could be a consequence of the grout plug formed with the post-injection. When exposing the nails, the diameter of the nails turned out to be relatively small compared to the grout plug obtained with the final injection, and also, an expansion factor of 2 for the drill bit is overoptimistic. The form of the grout plug obtained at the tip of the nail could be linked to the moderate consistency of the sands near the tip. Surrounding the reinforcement rods, a hole was systematically present, which was caused by the use of a non stable cement fluid (W/C=1). To avoid such voids, a W/C-ratio < 0,5 to 0,7 is advised. Exposure of the nails allowed a clear observation of the processes which are at work during installation and testing of the nails. The hereby made observations led to conclusions regarding installation and design of such type of soil nails. The tender documents were adapted, including more severe restrictions on the W/C-ratio of the grout.
p. 32/42
BBRI & BGGG-GBMS
6.
CASE STUDY 2 : THE USE AND ROLE OF GROUND ANCHORS AND MICROPILES IN THE VIADUCT STRUCTURES CROSSING THE COMPRESSIBLE VALLEYS OTTIGNIES-SOUTH
The Belgian National Railways Society (SNCB-NMBS) has undertaken the huge project of developing a Regional Express Network around Brussels (see figure 18). This project aims to progressively open up Brussels by providing more efficient public transportation: the train traffic on the main railways entering Brussels will be intensified and made faster. In order to meet these objectives, the main railways, basically consisting of two tracks, need to be broadened to four tracks. One of these railways connects Brussels to Louvain-la-Neuve. This railway, like most of the Belgian railway network, was constructed between 1850 and 1900, and crosses the hilly area South-East of Brussels. The crossing of the alluvial valleys was carried out at that time by means of high embankments, which were manually compacted. The widening of these fills is very difficult for many reasons:

6.1. Introduction
technical reasons: risks of settlement of the fills, instability of the embankment slopes social reasons: many houses or buildings are built close to the embankments, limiting the possibility of widening without numerous expropriations financial: the land value in those regions: the project has to stay, as much as possible, in the limits of the territory already owned by SNCB-NMBS.
An innovative solution had to be developed making use of the available space between the top and the base of the embankment. The chosen design is a viaduct structure, founded on micro-piles and leaning against the existing embankments. Anchors and micro-piles play an important role in this design. The main principles of this design are described here.
6.2. Geological context

Figure 19a and b represents a typical CPT profile executed from the railway platform, through the embankment, in an alluvial valley area. It consists of :
One more or less thick layer of filling material constituted of loamy sands or a mixture of ballasting materials and incineration residues. This layer is heterogeneous, poorly compacted and potentially prone to settlement under high loads One layer of alluvia, poorly or not consolidated Deep resisting layers : sands, clays or rocky materials

p. 33/42
BBRI & BGGG-GBMS
Figure 18: Plan of regional network project around Brussels
Embankment
Alluvium
Weathered rock
Figure 19a: CPT-test through existing railway embankment in the vicinity of compressive valley
p. 34/42
BBRI & BGGG-GBMS
Alluvium
Weathered rock
Figure 19b: CPT-test at the basis of the existing railway embankment in the vicinity of compressive valley
6.3. Design
The viaduct structure on this main railway, at Ottignies south, is conceived as a tunnel in open air, and consists of 3,4 meters long prefabricated elements of viaduct, founded on four micro-piles drilled through the embankment and the alluvial layers and anchored in the rocky, sandy or clayey layers under the alluvial deposit. The basic principle of this structure is represented on the cross section in figure 20. The structure is progressively built back ahead, working in consecutive phases. The elements of this structure are necessary for the stability of the structure, during construction or in service state. Those elements and their role are described hereafter:
Berliner retaining wall: In order to be able to build a working platform, which will become the future foundation of the railway platform, two Berlin type retaining walls have to be built (see photo 9):
o The wall A, realised in the slope of the embankment, aims to create the working platform for the execution of the micropiles of the viaduct. o The wall B, realised along the existing tracks, aims to retain the current embankment during earthworks.
In order to realise those walls, HEA soldier piles are placed through the embankment. The distance between the HEA profiles is 1,7 m for the wall A, 1,33 m for the Wall B. The sheeting of those walls consists of concrete flagstone. (see photo 9).
p. 35/42
BBRI & BGGG-GBMS
Photo 9: Retaining walls A and B of the viaduct
Figure 20: Typical cross section of viaduct
p. 36/42
BBRI & BGGG-GBMS
Helical screw anchors / nails C: During construction, as the excavators have to ride on the freshly placed platform that immediately can be loaded, wall A has to be temporarily tied-back in the embankment. To this end, helical anchors are used: these are actually rods screwed into the soil (see photo 5a). This technique was chosen because no heardening time of the cement has to be taken into account and because of its high speed and efficiency of implementation. It also has the possibility to be unscrewed and reused. Nevertheless, as no experience with this technique exists in Belgium, especially in railway application, many trials of installation in a railway embankment and several pull-out tests have been performed in order to validate the use of helical nails in this specific context. Self-drilling nails D: Inclined micro-piles are bored in the railway embankment, beneath the existing tracks. These nails aim to stabilize wall B and the embankment during and after construction, and help to transfer the lateral loads of the viaduct to the embankment. Self-drilling grouted nails E and F: as the embankment in its present state has a theoretical safety factor against gliding of about 1,00, micro-piles E and F are executed in order to increase this safety factor to a value of 1,15 after construction of the viaduct. Self-drilling micro-piles G and H: those micro-piles are the real foundations of the viaduct elements. They transfer the loads to the resisting layers below the alluvial deposits.
The viaduct elements may then be placed upon these foundations.
6.4. Feasibility trials

In its tender documents, TUC RAIL has decided to allow the use of self-drilling rods which do not necessarily have a continuous thread along their entire length, for the vertical micro-piles used as deep foundations of the bridges and for the G and H micro-piles of the viaduct sections. However, as the smooth tubes are often used with drill bits that doesnt have been tested before, the tender documents required the realisation of preliminary feasibility micro-piles. The feasibility micro-piles have to be excavated after hardening over a depth of min. 5m, in order to determine:

the optimal drilling parameters (density and quality of the grout, drilling parameters,), the adequate material, fitting with the various soils to be drilled through (drill bit, centralizers,), in order to obtain the necessary pile diameter and thus the grout thickness needed for the protection of the rod against corrosion.
Those trials have been executed in similar soil conditions than the ones of the building site. As it was impossible to implement them through the railway embankment, they were executed close to the base of the fill. The results of those trials are summarized hereunder.
p. 37/42
BBRI & BGGG-GBMS
6.4.1. Drilling parameters and grout conditions

The tests permitted the determination of the optimal parameters in order to ensure an efficient drilling in the soft to medium soft soil layers. Those are:

Drilling advancement speed : 48 m/h (80 cm/min) Rotational speed : 120 rpm
The optimal drilling cement fluid has a W/C-ratio of 2. The fine grains present in alluvial clayey soils mix with the drilling cement fluid, fulfilling in stabilizing the borehole. In these kinds of soils this ratio gives a good compromise between fluidity, helping the polluted grouts to be evacuated from the boring hole, and density, necessary to stabilize the borehole, especially in the railway fill. The injection cement fluid must have a W/C of 0,5, which permits to ensure a good replacement of the drilling grout and to make a good coating of the rod. It is also necessary to regularly move the rods up and down several times in order to clean the boring hole and so ensure the pile diameter and avoid soil inclusions into the pile.
6.4.2. Drill bits and spacers

The vertical micro-piles have to be drilled through several different soils, of very different hardness. The drill bits have to be adapted in consequence. In case of anchorage into clayey or sandy layers, a drill bit with a cross-shape, equipped with thin lateral injection channels, gave entire satisfaction (see photo 10) In case of anchorage into rocky layers (shales) this drill bit has been tested but was unacceptable. The drilling through the abrasive rock eroded the drill bit (see photo 11). The drilling speed rapidly decreased and the drilling quickly became impossible. Moreover, the diameter of the bit was reduced and the pile diameter was not ensured anymore. A special designed and reinforced drill bit had to be used. Nevertheless, in order to drill through the thick embankment and alluvial layers without clogging the bit, thin metal sheets were perpendicularly welded on the bit. Those sheets are able to shear the soft soil, and then break with the contact of the rock (see photo 12). The spacers are necessary to ensure the correct position of the rod in the borehole, and to create a correct pile diameter. Nevertheless the feasibility tests revealed that a spacer which is free to move around the rod can cause some problems:
During the drilling, the spacer moves around the rod and creates kinds of channels along the borehole surface. By these channels the grout finds an easy way to break through, towards the soil surface. This leads to a final helical-shaped pile, with irregular diameter (see photo 13). A spacer in an inclined oppressed final position may become an obstacle for the injection grout, leading to uncoated rod areas, just below the spacer (se photo 14). Under the spaces some soil inclusions occurred.
p. 38/42
BBRI & BGGG-GBMS
In order to prevent these problems, TUC RAIL required to weld the spacer on to the rod couplers (see photo 15) and to rotate with a rotational speed of min 120 rpm, or 1,5 rotations per cm. In this way the helical-shaped pile can be avoid and the total replacement of the drilling mud, cement fluid mixed with soil particles, by the injection cement fluid can be more guaranteed.
Photo 10: Cross-shaped drill-bits
Photo 11: Cross-shaped drill-bit eroded by the drilling through hard rocks (on the right) compared with unused drill bit
p. 39/42
BBRI & BGGG-GBMS
Photo 12: Drill bit especially designed for the rock drilling, and furnished with metal sheets to shear the soft soil layers
Photo 13: Helical-shaped micro-pile, with irregular diameter
p. 40/42
BBRI & BGGG-GBMS
Photo 14: Large uncoated area below the centralizer
Photo 15: Centralizers welded on the rod sleeves
7.
CONCLUSION
An overview of the practical experiences of TUC RAIL with ground anchors and micro-piles has been given. In the vicinity of railways, different design aspects have to be considered. A part from the calculation of the ground anchors and micro-piles, other design considerations referring to the feasability of the installations as well as the different execution details for long term durability of the ground anchors and micro-piles are taken into account.
p. 41/42
BBRI & BGGG-GBMS
Given the importance of the ground anchors and micro-piles for the stability of the constructions, different tests are done to guarantee the bearing capacity of the ground anchors or micro-piles. In two case studies, TUC RAILs point of view is presented. In some cases, excavation of the ground anchors or micropiles is done, which leads to a greater understanding of the importance of certain execution details.
8.
REFERENCES
[1] Ground Anchors: Overview of types, installation methods and recent trends, F. De Cock, Proceedings of International Symposium on Ground Anchors, Brussels, May 2008 [2] Tirants dAncrage, Recommandations T.A. 95, Recommandations concernant la conception, le calcul, lexcution et le contrle, Comit Franais de la Mcanique des Sols et des Travaux de Fondations, Editions Eyrolles, ISBN 2-212-01813-4, 1995 [3] Rgles Techniques de Conception et de Calcul des Fondations des Ouvrages de Gnie Civil, Fascicule N 62 Titre V, Ministre de lquipement, du logement et des transports, Circulaire n 93-66 du 20 dcembre 1993, [4] Execution of Special Geotechnical Work Ground Anchors, NBN-EN 1537, maart 2000, [5] De grondmechanische aspecten bij de aanleg van de HSL in Belgi, W. Maekelberg e.a., Proceedings of Conference on Geotechnical Aspects for Important Projects in Belgium KVIV september 2003 [6] Essai statique de pieu isol sous charge axiale, L.C.P.C. (Laboratoire Central des Ponts et Chausses-France), februari 1989 [7] Subcommittee on Field and Laboratory Testing - Int. Soc. Soil Mechanics Foundation Engineering - Axial Pile Loading Test part 1: static Loadinggeotechnical Testing - June 85. [8] Essai statique de tirants dancrage, NF P 94-153, AFNOR, Dcembre 1993 [9] Praktische ervaringen met nagels met zelfborende stangen voor de HSLwerken te Berchem, W. Maekelberg e.a., Proceedings of Innovative conference, 11nd Edition, KVIV november 2001 [10] Projet National Clouterre ; Recommandations CLOUTERRE, pour la conception, le calcul, lexcution et le contrle des soutnements raliss par clouage des sols, Presses de lcole nationale des Ponts et Chausses, 1991. [11] Verankerungen und vernagelungen im grundbau; Wichter, L en Meiniger, W. ; Bauingenieur-Praxis ; Ernst & Sohn 2000
p. 42/42
BBRI
BGGG GBMS
Annex A
BBRI & BGGG-GBMS
1.
RESULTS OF PULL OUT TESTS
Figure A.1: Measured values of displacements versus loads
Practical experience of TUC RAIL with ground anchors and micro-piles Maekelberg et al.
Annex A - p. 3/5
BBRI & BGGG-GBMS
Figure 17a: Tunnel under Antwerp - Excavation of nails Measured Diameters of nails
Annex A - p. 4/5
BBRI & BGGG-GBMS
Figure 17b: Tunnel under Antwerp - Excavation of nails Shell extent of nails
Annex A - p. 5/5
BBRI
BGGG GBMS
Annex B
BBRI & BGGG-GBMS
1.
DETAILED PHOTOS OF EXCAVATED NAILS
Figure B.1: Measured diameters 1st part of nails
Figure B.2: Measured diameters 2nd part of nails
p. 3/6
BBRI & BGGG-GBMS
Figure B.3: Measured diameters 3rd part of nail 1 - Nail end
p. 4/6
BBRI & BGGG-GBMS
p. 5/6
BBRI & BGGG-GBMS
Figure B.7: Measured diameters cavity in nail shell
Figure B.8: Measured diameters cavity in nail shell
p. 6/6
BBRI
BGGG GBMS
Experience with Ground Anchors of the Federal Public Buildings Service
Ir. Philippe Debacker Rgie des Btiments
BBRI & BGGG-GBMS
The Buildings Agency is an organism of the federal state, which is responsible for the federal public buildings. On the one hand, the agency makes studies and builds every year a number of new buildings, most of the time in an urban environment with all the technical complications that result from this situation. On the other hand, the agency also renovates or rebuilds every year a number of old already existing buildings. Among the new buildings there are, for instance, the Belgian prisons and courthouses, the main Brussels museums, office buildings, buildings for the European Community, all of the European schools, the Belgian buildings for the international exhibitions in Sevilla, Hannover, Nagoya, Zaragoza and soon in Shanghai. Among the old buildings, renovation works have been carried out at the Saint Michael's cathedral, the arcades of the Jubilee Park, the Palace of Congresses at the Mont des Arts and even the Lion's Mound on the Hill of Waterloo which has been consolidated by a whole series of ground anchors. For all of these works, a good knowledge of the structures is necessary, regardless if it concerns reinforced concrete, pre-stressed concrete, steel or wooden structures, stonework and last but not least geotechnics. We attach great importance to the correct dimensions of everything that touches the ground, like the shallow or deep foundations, the supporting works or the stability of the slopes, because in the field of geotechnics, every mistake in the design is unforgivable and the reinforcements that have to be realized afterwards always require very high costs. As you might have guessed when mentioning our activities concerning the buildings in urban environment, it regularly happens that we have to open an excavation so that we can place one or more lower ground floors. A first technique (Figure 1) that we use occasionally is that of successive excavations applying the stross method after having concreted the floors, one after the other on the ground temporarily kept in place. This technique has the great advantage not to be extended under the adjacent constructions with the whole problem of unknown obstacles and above all of the authorisations that need to be obtained. When we last applied this technique of stross working, we would have had to negotiate with a dozen of neighbours. Moreover, this technique guarantees the blockage on each level of the horizontal movements of the retaining walls. Because this realization process doesnt imply the use of ground anchors, we shall not give any further description. This leaves us then with the technique that consist of looking in the ground external to the excavation for the reaction necessary to maintain the retaining wall.
Experience with ground anchors of the Federal Public Buildings Service Ph. Debacker
p. 3/15
BBRI & BGGG-GBMS
When the floor adjacent to the wall (Figure 2) supports a pavement, a road or even an average building built on a raft foundation, we are not against the use of nails combined with a normal wall or even with the technique of guniting combined with successive excavations. The item of the dimensions of these nails is nevertheless not relevant for this symposium for two main reasons : we cant distinguish a free length nor a fixed length a prestressing force is not applicable to these nails
And this leads us finally to the ground anchors that are the subject of this study day. I will give you a description of the way we usually proceed. During the study that precedes the contracting (Figure 3), we choose the levels of the different anchoring beds, as a rule 1m above the future floors. The calculation of the wall, which we wont develop further here, takes into account the different phases of realization : successive excavations with the corresponding positions of the anchorages. This way, we obtain for every anchoring bed a horizontal linear force, expressed in kN/running m (SLS value). The total quantity of kN thus obtained and expressed in fixed quantities or even in presumed quantities is the only item of the contract that allows us to pay the ground anchors. We simply ask the tenderer to commit himself to a unit price : a certain amount in euros per horizontal kN. From here, the subcontractor indicated by the awarding Contractor is in charge of the further dimensioning : On the basis of this horizontal linear force, this subcontractor chooses: - the tonnage(s) of his ground anchors - their angle inclination - their free length and their fixed length. This inclination is expressed by a cosine that can highly penalize the capacity of the anchors. If the geologic configuration allows it, the subcontractor may choose the shorter ground anchors, but more inclined with shorter fixed lengths in the good layer, but this may be to the detriment of the horizontal capacity of each anchor and thus of the total number of anchors. This whole reasoning is freely carried out by the subcontractor who has to present to the approval of the Buildings Agency, his realization plans mentioning, for every anchor, its capacity, its inclination, its free length, its fixed length and its position on the wall. He therefore uses the results of the in situ soil tests, executed in sufficient numbers, preceding the tender. These tests are added to the tender file. These in situ tests are essentially CPT tests. For buildings designed for the European Communities, we add frequently PMT tests to simplify the price offerings of non Belgian contractors.
p. 4/15
BBRI & BGGG-GBMS
We think that its better to multiply the in situ tests, rather than to take samples so-called non reworked accompanied by tedious and expensive laboratory tests. The dimensioning of the fixed length is justified by graphs of the Bustamente type but nevertheless the subcontractor remains fully responsible. On the other hand, the dimensions of the free length (Figure 4) is justified by the application of the Krantz method or by every other similar method. This whole technical procedure linked to the anchors and the corresponding contractual aspect are the subject of a standard text for Special Specifications (Figure 5). This text is integrated in the Standard Specifications n 904 of the Buildings Agency in the same way as all the other articles concerning the structuring elements of a construction. I hereby seize the opportunity to mention that all the texts of the 904, which are regularly updated, are available for free on the website of the Buildings Agency. Under the general title Prestressed ground anchors there is, in French and in Dutch, a text for the temporary ground anchors, the definitive ground anchors and for the possible preliminary tests. These articles of the Special Specifications find their origins in old documents of the Ministry of public Works that have been entirely revised and adapted to the current practices. The French document TA 95 is the main basis for their drafting. A particular attention is paid to the tightening of the ground anchors. I will give you a brief survey of these Special Specifications (Figure 6) : If TS is the contractual Service Load of the ground anchor (SLS value), the tightening is carried out in successive stages until reaching 125% of the Service Load TS. These stages correspond successively to 12.5%, 25%, 50%, 75%, 100% and finally 125% of TS. At each stage, the load T is measured in the ground anchor together with the corresponding lengthening L of the extremity of the reinforcement of the ground anchor. The stopping time at each stage is limited to the time which is strictly necessary for taking the measurements. These couples of values ( L, T) are put into a graph. This graph is filled up with 2 straight lines passing through the origin. The equation of this two lines is T = AE L L AE represents the steel of the ground anchor. L is equal to the free length for the first straight line and equal to the free length increased by half of the fixed length for the second straight line. All the successive couples ( L, T) must be located between those 2 straight lines. The broken line that links these points must constitute a regularly increasing line without any abrupt angular deviations. In this way, it has been shown that the free and fixed lengths are truly correct and that the tightening of the anchor bulb is progressively carried out.
p. 5/15
BBRI & BGGG-GBMS
The last stage at 125% of TS is maintained during 15 minutes. The ground anchor is accepted if the lengthening L caused by the settling and measured between the time t + 3 minutes and t + 15 minutes is inferior or equal to 1,5mm. This procedure is generally applied on our yards. It allows us to have a simple but efficient control of each ground anchor. This has also allowed us to do without the Acceptance Tests that were formerly executed on a certain number of ground anchors picked by hazard after the realization of all the ground anchors. The last problem to deal with is that of the temporary ground anchors and the definitive ground anchors. As a rule, we set aside the definitive ground anchors because you can ask yourself what the importance is of an anti-corrosion guarantee of 10 or even 20 years for a public building designed to last for at least a hundred years. It is always possible to let the structure of the ground floors take over the ground pressures which were initially taken over by the ground anchors. When theres a great unbalance in the pressures resulting from a highly inclined site, the use of buttresses is the only reasonable solution (Figure 7). In this connection, I would like to mention an anecdote that happened to us twice during the last decennium. In the Ministry of External Relations in Brussels and later on in the Courthouse of Lige, we had actually put in place buttresses in order to guarantee the definitive stability of the high part of the constructions. Nevertheless, when we were considering to remove the temporary ground anchors like stipulated in the specifications, certain members of the study team wondered what the utility was of such an action. Why not leave them in place, knowing that, in addition, the commissioning authority and the subcontractor will not be suffering any financial looses doing so. On the contrary, for the same price, the client obtained at once a belt and straps to take over the ground pressures. At that time, some people pointed out the danger of leaving under pressure the ground anchors which are insufficiently protected against corrosion. These ground anchors, they said, could break abruptly during the following years and the heads of the ground anchors could then act like real cannon balls passing through the open spaces in the lower ground floors. Some people even pretended they had knowledge of such experiences. This seems at least unlikely because the corrosion acts slowly on the steel section of the ground anchors. After a noticeable reduction of this section, the steel weakens and lengthens, the prestressed strain disappears progressively and all danger is taken away. If someone in this room has really had knowledge or experienced an accident of this kind, I would ask him or her to come forward during the Coffee Break so that we can revise our point of view.
p. 6/15
BBRI & BGGG-GBMS
Finally, its only after a lot of discussions and hesitations that we definitively kept this temporary ground anchors under pressure. A last reflection to finish this presentation. Here is a picture of the head of a ground anchor, taken on a building site (Figure 8). How can we justify the lack of horizontal stiffeners between the heads of the ground anchors while the retaining wall executed in contiguous piles doesnt allow the flexion in the horizontal plane? Lets take into account that there is not any upper connection cross beam put in place before the beginning of the excavations. Under the bearing plate, the pressure on the concrete has to be close to its maximum allowable value. The vertical component of the load in the ground anchor that equals TS.sin , seems only to be equilibrated by the concrete acting under the lateral section of the steel plate. The next picture (Figure 9) is another illustration of a construction that, in our opinion, contains a lot of risks. But all those things are kept in place and thats for the better; this is the charm of geotechnics.
p. 7/15
Figure 1- Stress method
BBRI & BGGG-GBMS
Figure 2- Soil nailing p. 9/15
BBRI & BGGG-GBMS
Figure 3- Application of Ground Anchors p. 10/15
BBRI & BGGG-GBMS
Figure 4- Krantz method p. 11/15
BBRI & BGGG-GBMS
Figure 5- Standard text of the Buildings Agency Special Specifications (n 904) p. 12/15
BBRI & BGGG-GBMS
Figure 6- Tightening of the anchors according to the Special Specifications
p. 13/15
BBRI & BGGG-GBMS
Figure 7 p. 14/15
Figure 8 Anchor Head
Figure 9 Anchor Head
p. 15/15
BBRI
BGGG GBMS
Experience with Ground Anchors of the Flemish Community
Ir. Inge Marin Ministry of Flemish Community
Application of ground anchors in civil constructions of the Flemish Government
Ir. Inge Marin Regional Government of Flanders, Dept. of Mobility and Public Works
Contents
Some of our projects Most used type of ground anchor for quay walls Corrosion protection of permanent ground anchors Ground anchor design Tests on ground anchors Some test cases
Some of our projects

Deurganckdock Embankment renovation New lock in Evergem Projects given in concession by the Flemish government Port of Genk
Deurganckdock
New lock in Evergem
New lock in Evergem
Most used type of ground anchor for quay walls

Task: Construction design life = 100 years Permanent ground anchors Pre-stressed (deformations) Cost (in comparison with pre-stressed bars) Handling on the construction site Strand anchors
Corrosion protection of permanent ground anchors

Task: Construction design life = 100 years Visual inspection after construction = impossible Industrial sites Corrosion protection! Built-in redundancy Demand for a solid and continuous corrosion protection (< NBN EN 1537:2000) General rule: each part is double protected + attention to transition zones
Example of double corrosion protection anchor head

Example of double corrosion protection free length

Example of double corrosion protection bond length


Exceptions on the general rule Test: crack width in grout is limited sheathing & cement grout = 2 protection layers E.g. deadman anchors visual inspection Temporary anchors
Ground anchor design

Principle: Failure of one random anchor may not lead to the failure of the entire construction.

Specified in tender Working load Breaking load Lock-off load Anchor type (strands / bar) Spacing anchors Minimum steel section Inclination Start level grout body (IRS)

Specified by the contractor Bond length and diameter Mostly: IGU / IRS Injection pressure, other working parameters
Tests on pre-stressed ground anchors

Investigation tests: Not for known systems Suitability tests: In specific cases Not systematically in each project Tested until failure
Acceptance tests: Strand anchors: for each anchor in the construction
Tests on pre-stressed ground anchors

According to DIN 4125
Minimum 5 loading and unloading cycles + determination of creep displacement rate ks = (s2-s1)/log(t2/t1) more suitable for quay walls with alternating loading Test period suitability test in non-cohesive / cohesive soil
According to informative annex E of NBN EN1537:2000

tests less obvious not fully described
Conclusion: The German method is most suitable for most of our projects.
Some test cases

Suitability tests: Ground anchors in chalk bridge in Kanne Ground anchors for quay walls at the Leie in Kortrijk
Ground anchors in chalk bridge in Kanne

4 test anchors IGU-anchors Working load = 215 kN Bond length = 10 m Suitability test according to DIN 4125. Extra load levels: 2x, 3x, 3.5x, 4x working load
Ground anchors in chalk bridge in Kanne

Kanne IGU according to DIN 4125: A nchor 1
90 80 Displacement (mm) 70 60 50 40 30 20 10 0 0 200 400
Low creep
No failure at 860 kN ks < 2 mm

600 800 1000 Test load (kN )
Ground anchors for quay walls at the Leie in Kortrijk

2 test campaigns: 3 IGU and 6 IRS Working load = 700 kN Breaking load = 1400 kN Bond length: Anchor 1 IGU = 12 m Anchor 2 IGU = 14.5 m Anchor 3 IGU = 15 m Anchor 3 and 4 IRS = 15 m Anchor 5 IRS = 8 m IRS Anchors 4 and 5: Extra load levels until 2.25x working load

Kortrijk IGU according to DIN 4125: A nchor 1
100 90 Displacement (mm) 80 70 60 50 40 30 20 10 0 0 100 200 300 400 Test load (kN ) 500 600 700 800
Failure at 700 kN

250 200 150 100 50 0 0 200 400 600 800 Test load (kN ) 1000 1200 1400 1600
High creep
Displacement (mm)
Failure at 1400 kN

250 200 150 100 50 0 0 200 400 600 800 Test load (kN ) 1000 1200 1400 1600
High creep
Displacement (mm)
No failure at 1400 kN

Kortrijk IRS according to DIN 4125: A nchor 3
120 100 Displacement (mm) 80 60 40 20 0 0 200 400 600 Test load (kN ) 800 1000 1200
Low creep

Kortrijk IRS according to T.A .95: A nchor 4
0.25 0.2 0.15 0.1 0.05 0 0 0.2 0.4 T / TP 0.6 0.8 1
alfa

Kortrijk IRS according to DIN 4125: Anchor 5
180 160 Displacement (mm) 140 120 100 80 60 40 20 0 0 200 400 600 800 1000 1200 1400 1600 1800
Test load (kN)
BBRI
BGGG GBMS
Design Guidelines for Non-driven Tension Piles Underneath under Water Concrete Slabs
- Developments in The Netherlands regarding design, testing and quality control of vertical ground anchors -
Ir. Ad Vriend Accon Adviesbureau voor funderingstechnieken bv Dordrecht, The Netherlands
BBRI & BGGG-GBMS
ABSTRACT
In The Netherlands there is no specific guideline available for the design and testing of vertical ground anchors. In daily practice design engineers often make use of different standards and guidelines related to inclined ground anchors and micro piles, but in many cases these documents do not cover all specific design issues essential for vertical ground anchors or could even lead to wrong conclusions. This paper describes the highlights of ongoing studies by the Dutch CUR committee C152 which is responsible for writing a guideline dedicated to the design, testing and quality control of vertical ground anchors when applied to support under water concrete slabs against uplift by water pressures. It must be emphasized that the studies are still ongoing and consequently values as presented in this paper are to be interpreted as indicative only as they are still to be determined by the committee for final use.
INTRODUCTION
During the last 10 to 15 years in The Netherlands there has been a strong and successful increase of applying vertical ground anchors and micro piles as vertical tensile elements preventing the uplift of under water concrete slabs in building pits. Most of these tensile elements are well known and commonly practiced in the application as inclined anchors to support vertical retaining structures. Most of the experience in both design and execution rests with specialist contractors that install these ground anchors. In daily practice this is not a real problem as these contractors are responsible for their own products and after installation, all of these anchors are subjected to either suitability tests and/or acceptance tests to prove their capacity. In case of vertical anchors however this is quite different as in most projects only a small percentage of these anchors can be tested as it is too expensive and too time consuming to test them all. How to be sure that these vertical elements do meet their requirements regarding the bearing capacity and how can that bearing capacity be calculated anyway? Another difference between inclined ground anchors and vertical anchors is that the anchors supporting retaining structures can be pre-stressed after being tested in order to reduce the expected elongation to an acceptable level. Therefore in most projects the so called axial stiffness of the applied anchors does not play an important role and if so, counter measures can be taken by choosing a specified level of pre-stressing. In case of vertical ground anchors however the axial stiffness plays quite an important role due to its relative strong influence on the behaviour of and development of tensile stresses in the under water concrete slab that is being supported by these anchors. At the same time most of these relative slender anchors and micro piles are quite sensible for elongation due to the economical drive not to put too much steel into the ground. Unfortunately in most situations pre-stressing of vertical tensile elements is not a realistic option, so calculation of the rising of the pile head that can be expected when tensile force is increasing from zero to its maximum level has become a critical aspect in the
Design guidelines for non-driven tension piles underneath under water concrete slabs - Developments in The Netherlands A. Vriend 14.05.2008 p. 3/15
BBRI & BGGG-GBMS
design. The design of vertical tensile elements and under water concrete is following quite an interactive procedure. At the moment there is no specific code or guideline that can be referred to for the design of these piles and because of the essential differences between inclined and vertical ground anchors as describe above, there is a strong demand both by independent design engineers and specialist contractors to develop dedicated recommendations in which design aspects of both bearing capacity and axial stiffness are described as well as how to deal with in situ testing and quality control. The Dutch CUR Committee C152 is now working on these recommendations and in this paper the most important issues regarding design and testing are highlighted.
3.
BEARING CAPACITY OF VERTICAL GROUND ANCHORS

In general Dutch geotechnical design rules for piles both under compression and tension are based on the use of cpts where the bearing capacity is calculated by multiplying the cpt values with dedicated and prescribed factors to determine the friction and toe resistance. This method is nowadays also commonly used for ground anchors, but the values of these (friction) factors are not prescribed. This leads to a situation in which geotechnical design engineers are using different values and not surprisingly higher values are commonly proposed by specialist contractors and lower values by design engineers; more than once this has lead to intensive discussions between the two parties. Basically the bearing capacity of a vertical ground anchor and micro pile is calculated by using the following equation as presented in the Dutch recommendations CUR 2001-4 Design rules for tension piles [1]:
Fr ; tension ; max; d = sf 1 f 2tqc ; red ; ddz
0 L
(1)
where: Fr;tension;max;d = factored design value of the tension bearing capacity s = diameter of pile shaft or anchor (in case of ground anchors or micro piles commonly in the range of approximately 150 to 300 mm) = factor describing the possible positive influence caused by f1 increased density of the soil layer due to pile installation (in case of soil displacement piles f1>1.0; in case of anchors f1=1.0) f2 = factor describing the influence of reduced effective stresses due to the so called group effect, where piles are positioned with relative short ctc distances and more or less pulling onto the same sand particles (depending on ctc distances f2 varies between 1.0 to 0.5 or even less) t = friction factor, different per type of pile or anchor (see below) qc;red;d = design value for the cone resistance, in which aspects as reduced effective stresses due to excavation, reduced capacity due to time depended variation of axial forces as well as a partial material factor for piles or anchors under tension are taken into account
BBRI & BGGG-GBMS
= bond length of the pile or anchor
It must be noted that for slender anchors an additional factor describing the effectiveness of anchors with a relative long bond length should be taken into account in the equation as presented above. At the moment in the Dutch design practice there is no commonly used approach for this aspect, and will be assessed within the C152 committee. The determination of values for the friction factor t for tension bearing capacity of vertical ground anchors and micro piles as most commonly used in The Netherlands is still in the process of intensive study within the C152 committee, where numerous available load testing results are assessed. The following table presents ranges of indicative values that are generally used in daily practice (final design values are still to be determined by the C152 Committee).
Table 1. Overview of friction values for different types of ground anchors (indicative only)
Type of ground anchor Traditional bored ground anchor with double casing (overburden flush drilling) Bored ground anchors using one single casing (percussion flush drilling with lost bit) Self boring anchors Screwed ground anchors
Friction factor t (1), (2) 1.5 2.0 % 2.0 2.5% 1.0 1.5% 1.0 1.5%
(1) It must be noted that values could be well less when no (over)pressure is being built up during grout injection and formation of the fixed bond length; under good circumstances higher values can also been found. (2) Applicable for qc values of max 15 20 MPa. It is essential to realize that the above presented types of anchors all have in common that they are quite sensitive for the method of pile installation. The man who is operating the anchor rig is the one who can make or brake the end product; so the bearing capacity strongly depends on the experience and craftsmanship of the operator. Of course the human factor is not the only factor that determines the results, as the specific soil conditions on the site also play a very important role. When still in the process of making the first design it is therefore recommended to take a sufficiently safe design value for the friction factor t;d and we are inclined to propose to use more or less the average of the above presented values, depending on the specific type of anchor that is being used in the design. Once a project is approaching the start of the actual pile installation then we strongly are in favour of steering in the direction of executing in situ pile tests in order to determine the actual project specific design value. The proposed design value can then be verified and, in case of testing up to failure, even be optimized. For larger projects pile testing should always be a part of the process, but when having a smaller project one could from an economic point of view choose not to test but to use a relative safe design value in combination with a somewhat higher partial safety factor. The values of these partial material factors will depend on the testing regime that is chosen on the job site; at the moment the C152 Committee is studying the values of these factors and will try to follow the proposals as specified in the Eurocode.
BBRI & BGGG-GBMS
4.
AXIAL STIFFNESS OF VERTICAL GROUND ANCHORS

As described in the introduction, the axial stiffness of vertical ground anchors plays an important role in the design of the under water concrete slab. Different then when using tension piles with a larger shaft diameter, in case of applying an relative slender anchor element the axial stiffness could strongly effect the bending moments and tensile stresses in the under water concrete floor. This makes it important to develop a reliable design rule for calculating this axial stiffness. The C152 committee is still studying how to calculate the axial stiffness, but as per own experience of the author the following design approach proved to be a workable method in several large projects. The axial stiffness can be calculated by using the following equation:
krep = Frep lpilehead
(2)
where: krep = representative value of the axial stiffness Frep = representative value of the tension force from the under water concrete acting on the pile pile head = increase in rise of the anchor head caused by Frep This approach is valid only for tension forces slowly increasing form 0 to the unfactored Frep, corresponding with the loading of the tension piles during the process of pumping the water out of the building pit. The rise of the anchor head is the cumulative result of the following three contributing risings: lpilehead = lelastic + lanchorbody + lswelling (3)
where: lelastic = elastic elongation of the anchor lanchorbody = rising of the anchor along soil particles during mobilisation of friction lswelling = rising of lower soil layer(s) below toe level of the anchors due to swelling The contribution of the elastic elongation can easily be calculated by using the following equation:
lelastic = FrepLeff /( EA) anchor where: = representative value of the tension force acting on the anchor head Frep
(4)
BBRI & BGGG-GBMS
Leff
= effective length, determined by the level of the anchor plate in the under water concrete and the level of the so called fictive anchor point which lies at approximately 1/2th 2/3rd of the bond length below the top level of the anchor body
The contribution by the rising of the anchor along the soil particles during mobilisation of the friction is basically estimated at approximately 1-1.5 mm when the anchor body is successfully formed by high pressure grouting, to approximately 3-5 mm when the anchor body is formed under hydrostatic grout pressures. This includes possible creep which contribution is supposed to be more or less of minor importance when compared with all other cumulative contributions in vertical rising, but it must be noted that the aspect of neglecting the creep is still being studied at. The contribution by the swelling of soil layer below toe level of the anchors is more complicated to calculate, and should be dealt with by an experienced geotechnical engineer. In the situation of only sand layers below toe level then this contribution is often neglected, but when having a wide excavation and having a clay layer below toe level then it can give quite a substantial or even governing contribution. It must be noted however that swelling of a deeper soil layer(s) is not only influencing the rise of the anchors but also of the surrounding retaining walls of the building pit so in the end not the total swelling has to be taken into account but only a differential part. Within the C152 committee the study on the contributions in the rising of the anchor head is ongoing and promising results are expected. Complicated though is the translation form a single anchor to a group of anchors that are influencing each other.
5.
TESTING OF THE VERTICAL GROUND ANCHORS AND QUALITY CONTROL

Establishing a realistic framework and requirements for testing of vertical ground anchors is one of the most important issues within the C152 committee. It is obvious that testing of piles has great advantages especially when executed well before the start of actual piling works. Testing strongly increases the reliability and acceptance of vertical ground anchors as a product, especially when it is combined with a strict registration of parameters representative for the piling process enabling a useable tool for in situ quality control.
5.1.
Investigation tests
As described in the previous chapter, testing of the vertical ground anchors is essential for determination of the correct design value for the friction factor t. It is favourable to perform pile testing well in advance before starting the execution of the actual piling project, and to bring these test piles to failure. The results of the proven ultimate capacity can then be used to optimize the design of the
BBRI & BGGG-GBMS
vertical ground anchors. One disadvantage however is the fact that in most cases only a relative short fixed bond length can be used in order to be able to bring the test pile to failure, so it can be questioned if the results are sufficiently representative when used for longer fixed bond lengths of the actual production piles.
5.2.
Suitability tests
When there is less time and/or money available, pile testing could also be done on actual production piles without bringing them to failure. The test load should be such though that it can be proven by back calculation that at least the design value of t was mobilised. Disadvantage of performing suitability tests on actual production piles is that in most cases the subjected piles must be modified by applying a high capacity tendon to accommodate the relative high test load. This implies that before the start of the piling works it must be known which piles will be tested and it can then be questioned whether these piles are still representative.
5.3.
Acceptance tests
In order to verify the in situ axial stiffness of the production piles acceptance tests can be applied. The test piles are subjected to a load of at least the factored design value of the tensile force Fs;tension;d which is in the order of approximately 1.25 to 1.3 times Frep. Because these piles are designed as group piles, but can only be tested as single piles, the actual ultimate bearing capacity of such a production pile will be at least a factor 2 to 3 times higher than Frep, and from such a short load test no conclusions can be drawn about the bearing capacity of that tested pile, unless something has gone dramatically wrong during pile installation. Another difficult aspect that is still studied is the translation of the results of a single pile to the group of piles, as it is supposed that a pile in a group is having a somewhat reduced effective axial stiffness. The C152 committee is now in the process of evaluating and discussing several testing regimes, but in short the following tests and objectives can be summarized as presented in the following table.
5.4
Overview of proposed testing in according with EN-1537

Table 2. Overview of proposed anchor load testing in accordance with EN-1537 Type of load test Bearing capacity Axial stiffness Main objective of load test Investigation test (1) + optimisation of friction factor t;ult;d Suitability test (2) + verification of friction factor t;d Acceptance test (3) + verification of axial stiffness krep
(1) Test from existing ground level. Loading up to failure. (2) Test from existing ground level. Loading including compensation of all reduction and material factors. (3) Test in excavated building pit before pouring the under water concrete. Loading of production pile to factored design value of tension force.
Testing of vertical ground anchors before the start of the anchor works is not only important to determine a safe design value for the friction factor, but it also
BBRI & BGGG-GBMS
enables comparison of all items that are relevant for pile installation (grout injection pressures, volume of injected grout, w/c-factor, type of cement, diameter drilling bit, diameter bore casing, etc) as registered during installation of the initial test piles with the actual production piles. By using this information and following this procedure the quality control can take place during and immediately after installation of the production piles. In case of doubt over certain piles, these piles can additionally be subjected to an acceptance test to verify the load displacement behaviour.
6.
CASE STUDY
In the west of The Netherlands as part of a large building pit, more than 700 vertical ground anchors had to be installed in order to prevent uplift of an under water concrete slab. The vertical ground anchors (Gewi-piles) were installed from existing ground level using a single casing and percussion flush drilling with a lost bit. After lowering the tendon into the casing, the fixed bond lengths were formed by grout injection of at least 5-10 bar and gradual extraction of the casing at the moment of rapid increase of the torque on the casing which is slowly turning around during this process. The relevant specification of the vertical ground anchors and test piles are presented in figure 1.
Test piles I, II and III 0.5 m+ 0.5 m2.75 m3.25 mTest piles A0, A1, A2 and A3 1 2 3 1182 kN 1326 kN 1170 kN Pile type Fs;tension;d Gewi bar 63.5 mm FeB 555/700 63.5 mm FeB 555/700 63.5 mm FeB 555/700 Diameter anchor s = 200 mm s = 200 mm s = 200 mm Design value friction factor t = 2.0% t = 2.0% t = 2.0%
Levels in [m] to NAP 63.5 mm SAS 670/800 75 mm St. 835/1030 Pile type 1 Pile type 2 Pile type 3
10.4 m11.6 msand clay 16.0 m17.5 mLa = 5.0 m s = 200 mm 22.5 m23.0 m24.0 msand (dense) (water head 0.50 m-) 13.5 m-
sand (very dense) La = 9.0 m s = 200 mm
33.0 m-
33.0 m34.75 m36.5 m-
Figure 1. Typical cross section
6.1.
First series of investigation tests (test piles I, II and III)

For the first series of investigation tests it was chosen to install three piles with 5.0 m fixed bond length in the dense sand layer between 16.0 m- and 23.0 mNAP. It was expected that the test pile could be loaded to failure, which was calculated to occur at a test load of at least 1540 kN (based on the average qc values between 17.5 m- and 22.5 m- NAP, where qc values were limited to 20 MPa and using an optimized friction factor of t=2.5%). Actually, based on comparable load tests in dense sand, failure was expected at approximately 1600
BBRI & BGGG-GBMS
to 1800 kN. By using a Gewi+ bar 63.5 mm with steel grade SAS 670/800 as tendon the maximum test load could be increased up to 1910 kN.
During the process of pile installation, which took place from a working level of 0.50 m+ NAP and ground water level at 0.50 m- NAP, no problems or deviations were observed and grouting pressures of 8 to 15 bar were registered. The loading procedure as applied for testing piles I, II and III is presented in the following table.
Table 3. Loading procedure test piles I, II and III Load Test load Load held steps constant Fi 100 kN 20% 308 kN 20 min 40% 616 kN 20 min 60% 924 kN 30 min 80% 1232 kN 45 min 100% 1540 kN 60 min Fi 100 kN 30 min 100% 1540 kN 30 min Increased test load until failure occurs 110% 1694 kN 30 min 120% 1848 kN 30 min 124% 1910 kN 30 min
Figure 2. Representative CPT
Photo 1. Testing of piles
BBRI & BGGG-GBMS
Contrary to the expectations, all three test piles failed in a much too early stage, at respectively 977 kN, 1355 kN and 1450 kN. The last load steps during which the test piles were showing a stable behaviour (creep < 2.0 mm) were at 924 kN, and 1232 kN. This disappointing result turned out to be caused by the characteristics of the sand particles in the sand layer between 16.0 m- and 23.0 m- NAP in which the fixed bond lengths were installed. After receiving a representative borehole description and by doing some further geological investigation it was concluded that this sand layer was described as Dune sand consisting of fine to very fine and rounded sand particles. These specific characteristics could not be derived form the original CPT, and it became clear that the bore hole data and geological information was needed to determine the characteristics of this particularly sand layer. The assumed cause for this low bearing capacity is hidden in the combination of the installation process and the specific soil conditions. During the process of pile installation a relative thin zone around the anchor is effected. At first, during boring of the anchor, this zone is slightly loosened and then by forming the fixed bond length by grout injection with sufficient grouting pressures the sand particles are compacted and voids between the sand particles will partially be filled with cement.
zone effected by pile installation
anchor body
sand surrounding anchor
Figure 3. Effected zone around anchor body
In case of coarse and angled sand particles this zone is then brought back to the original situation or even better, resulting in a high capacity for transferring shear stresses to the surrounding sands. However, when having fine and rounded sand particles this behaviour is different as this zone cannot be brought back to the original situation due to the lower porosity and the lack of angled particles. The fine rounded sand particles are not able to mobilise the same amount of friction between the particles.
6.2
Second series of investigation tests (test piles A1, A2 and A3)

The disappointing results of the first three investigation tests have lead to the decision to install a second series of three test piles, but then with the fixed length in the deeper and very dense sand layer below 23.0 m- NAP. In order to have relative long fixed lengths with toe levels as close as possible to the actual Gewipiles that were designed for the building pit, much effort was put into finding special high capacity steel bars (strands were not favourable due to their relative high strain that would certainly influence the results). The higher the capacity of the tension bar the longer the fixed length could be, but on the other hand the fixed length should not be too long because of the need to load these additional test piles up to failure. This was not an easy task, but the piling contractor was lucky in finding bars 75 with steel grade St. 835/1030. Based on the maximum acceptable test load of 3320 kN that could be applied on these bars 75 mm, it
BBRI & BGGG-GBMS
was decided to install the 2nd series of test piles with 9.0 meter long fixed lengths. Failure of these test piles was anticipated not to occur before approximately 2900 to 3100 kN. In this way fixed lengths could be created between 24.0 m- and 33.0 m- NAP which was quite close to the designed toe levels of the actual Gewipiles in the building pit (see figure 1) and thus as representative as possible. During the installation of test piles A2 and A3, which now took place at a somewhat deeper level of 2.75 m- NAP with freatic ground water table at 3.25 mNAP (and with a water head in the deeper sand layer below the clay layer at 0.50 m- NAP) the piling contractor encountered new problems as (due to grout welling) it turned out to be difficult to build up the essential grout injection pressure when forming the fixed length from 33.0 m- up to 24.0 m- NAP in the coarse sand. To overcome this problem the piling contractor did not continue with further installation of test pile A1, but made some additional trial piles first to the depth of 33.0 m- NAP. After some successful adjustments the pile installation procedure was improved leading to an increased and acceptable grout injection (over)pressures of 4 to 5 bars. The piling contractor then continued with the successful installation of test pile A1 and an additional 4th test pile A0. Testing of these 4 piles of the 2nd series went similar to the first series but then up to a maximum test load of 3300 kN. Results were more or less in accordance with expectations: A2 and A3 failed too early at respectively 1910 kN and 2640 kN, where the other two piles A1 and A0 that were installed using the improved installation procedure showed good stability even when loaded up to 3300 kN without any indication of nearby failure.
Test pile A1 (grouting pressures > 5 bar) Test pile A2 (no grouting pressures)
Figure 4. Load displacement curves of test piles A1 and A2
After completion of this 2nd series of investigations tests again additional piles were installed and tested, giving confirmation of the good results of previous test piles A1 and A0.
BBRI & BGGG-GBMS
6.3.
Summary of results investigation tests

Table 4. Overview test results
Test piles I II III A2 A3 A0 A1 Fixed bond length 5.0 5.0 5.0 9.0 9.0 9.0 9.0 17.5 m- to 22.5 m17.5 m- to 22.5 m17.5 m- to 22.5 m24.0 m- to 33.0 m24.0 m- to 33.0 m24.0 m- to 33.0 m24.0 m- to 33.0 mExpected failure load 1600 - 1800 kN 1600 - 1800 kN 1600 - 1800 kN 2900 - 3100 kN 2900 - 3100 kN 2900 - 3100 kN 2900 - 3100 kN Failure load 977 kN 1355 kN 1450 kN 1910 kN 2640 kN > 3300 kN > 3300 kN Last stable load step 924 kN 1232 kN 1232 kN 1800 kN 2400 kN 3300 kN 3300 kN Effective shear stress (2), (3) 280 kN/m2 230 kN/m2 260 kN/m2 310 kN/m2 290 kN/m2 560 kN/m2 540 kN/m2 Comments fine rounded sand fine rounded sand fine rounded sand no grout pressure no grout pressure grout pressure 4-6 bar grout pressure 4-6 bar
(1) Diameter pile shaft in calculation s=200 mm (drill bit =180 mm). (2) Design values effective shear stress based on last stable load step and after taking into account estimated friction along free anchor length (respectively: 40 kN, 500 kN, 400 kN, kN, 730 kN, 100 kN and 220 kN). (3) Design of Gewi-piles in building pit based on d = t qc;limited = 2,0%x15,0MPa= 300 kN/m2.
Despite the somewhat disappointing results of the bearing capacity in the layer with fine and rounded dune sands, the underlying very dense layer with coarse sands provided more than sufficient compensation. Ultimately the proven bearing capacity was higher than required so the designed Gewi-piles could safely be installed, provided a good quality control for maintaining the adjusted and improved installation process. Recently we received information about another project in the same region where Gewi-piles were installed by a different piling contractor but in similar soil layers, and remarkably the same conclusions were drawn on the significant difference in bearing capacity in the fine and rounded dune sands when compared with the underlying coarse sands. This confirms the influence of the specific characteristics of the sand particles.
6.4
Acceptance tests on production Gewi-piles after wet excavation of the building pit
After completion of the excavation of the building pit in total 33 Gewi-piles have been subjected to an acceptance test in order to verify the axial stiffness of these piles. The piles were tested from a floating pontoon using an extension bar that was coupled to the pile head just above excavation level. For design of the under water concrete the representative value of the axial stiffness was estimated in the range between 45 50 MN/m which was quite well confirmed by the acceptance tests; most results showed even higher values.
BBRI & BGGG-GBMS
Photo 2. Acceptance test from floating pontoon
Photo 3. Building pit after completion
7.
SUMMARY AND CONCLUSIONS

- Though the Dutch design guidelines for non-driven tension piles underneath under water concrete slabs is still in the process of development, it is expected to be welcomed by both the independent design engineers and specialist piling contractors as it will provide a relative straight forward and clear procedure for design, testing and quality control when dealing with vertical ground anchors. - In case of large piling projects the use of the new guideline will lead to the increase of investigation tests or suitability tests prior to the actual piling works. This will have the benefit to be able to confirm the bearing capacity of the vertical ground anchors or even to optimize the pile design. - As all types of vertical ground anchors have in common to be sensitive for the experience and craftsmanship of the piling contractor and the operator in special, another important objective is to have better control over the installation process in general. - As shown in the case study, testing also makes it possible to discover unusual soil conditions, which has the great advantage of being able to deal with this before the start of the piling project and not when execution is already ongoing or even finished. - It can be questioned if the use of CPTs alone is sufficient for making a reliable design, as it does not provide essential information about the specific and essential characteristics of the sand grains. Borehole data and geological knowledge can provide valuable additional information, but it will remain indicative only. - Eventually, testing will be the best option to optimize the pile installation process for the specific local soil conditions and to determine the final pile design. - By comparing the basic pile installation parameters as registered during testing with these parameters during the execution of the actual piling works, it will be possible to detect potential defects in an early stage so that actions can be taken immediately if necessary. - In case of smaller projects, testing is often leading to unrealistic additional costs, and more conservative design with the use of higher partial safety factors can be an economical acceptable alternative.
BBRI & BGGG-GBMS
- By using the procedures as proposed in the recommendations it is expected that quality and overall safety level of vertical ground anchors will increase, leading to further successful and reliable application in future projects.
8.
REFERENCES
[1] CUR 2001-4; Design rules for tension piles Civieltechnisch Centrum Uitvoering Research en Regelgeving (CUR); juni 2001 [2] NEN-EN 1537; Execution of special geotechnical work ground anchors Nederlands Normalisatie-instituut; December 1999 [3] NEN 6743-1; Calculation method for bearing capacity of pile foundation compression piles Nederlands Normalisatie-instituut; November 2006 [4] NEN 6745-2; Load test on foundation piles static axial loading in tension Nederlands Normalisatie-instituut; December 2005 [5] Proceedings of the international symposium on anchors in theory and practice Salzburg, Austria; 9-10 October 1995 [6] Proceedings of the international conference Ground anchorages and anchored structures Institution of Civil Engineers, London, UK; 20-21 March 1997
BBRI
BGGG GBMS
Experience in France with Ground Anchors
Prof. Jean-Pierre Magnan LCPC
BBRI
BGGG GBMS
INTERNATIONAL SYMPOSIUM GROUND ANCHORS Limelette test field results 14 May 2008 Hotel Mtropole Brussels
Experience in France with Ground Anchors

J.P. Magnan, LCPC (France)
The French practice is described in

Ground Anchors
Recommendations for conception, calculation, execution and testing
RECOMMENDATIONS T.A.95 Previous editions : 1972, 1976, 1986 in French 1989 Balkema in English Prepared by 37 experienced engineers (Cambefort, Graux, Lebelle, Dupeuble, Clment, Dupeuble, Logeais, under the convenorship of P. Habib) 180 pages www.geotechnique.org
The recommendations T.A. 95 cover

1. 2. 3. 4. 5. 6. 7. 8.
A1 A2 A3
Introduction Definitions Material specifications Protection against corrosion Execution Testing Periodical control of anchor load Recommendations to the client
Appendix : Stability of retaining structures Appendix : Verification of the global stability of vertical anchors under slabs Appendix : A guide to the preliminary design of anchors
A look into (close) future

Recommendations T.A. 95 Introduction Definitions Material specifications Protection against corrosion Execution Testing Periodical control of anchor load Recommendations to the client
A1 A2 A3 Appendix : Stability of retaining structures Appendix : Verification of the global stability of vertical anchors under slabs Appendix : A guide to the preliminary design of anchors
EN 1537 Execution of anchors
EN 22477-5 Testing of anchorages
EN 1997-1 Eurocode 7 Part 1 NF P 94-282 Geotechnical design Retaining structures Embedded walls and anchorage EN 22477-5 Testing of anchorages Procedure 3
French Standard on Static Anchor testing NF P94183
What is French practice (experience)?

Given types of passive or prestressed anchors Specified types of accepted materials Proven ways of insuring protection against corrosion A design method based on investigation tests on each type of anchor in each type of soil The execution of suitability tests at the construction site The execution of acceptance tests on each anchor A preliminary design procedure based on Mnard pressuremeter tests Anchor test procedures based on incremental loading and interpretation in terms of anchor pull out resistance, anchor critical resistance and a limited anchor head displacement
Given types of anchors and material specifications
Specifications covered by EN 1537. Steel tendons, grouted by cement or resins. Free length and grouted length: equal protection. Protection against corrosion: depends on the aggressiveness of the ground, include metal or plastic sheaths and ducts, cement or resin grouts
Maintained anchor test procedure

Procedure 3 in EN 1537 (and prEN 22477-5)
100 90 60 min 80 60 min 70 60 min 60 60 min 50 60 min 40 30 20 10 0 0 50 100 150 200 250 300 60 min 2 min 2 min 2 min
Load [% PP]
Proof loads Pp differ.

60 min
60 min
8 x 1h Acceptance 10+15+5+5+1min Suitability 6 x 1h

350 400
Investigation
Time [min]
450 500
5 min
Proof loads and design parameters

For investigation tests Pp 1,25 Po or Pp Ra but Pp 0,9 Ptk Po = estimated service load : Ra resistance (failure) The tendon section may be adapted to avoir steel failure For suitability tests Pp a Pd (EC7, modified) Pd = design load For acceptance tests Pp 1,25 Pserv and Pp 0,9 Ptk Pp Pd (EC7, modified)
for checking creep
Basis for design

The design load must be calculated first, from the equilibrium of the retained structure. The design load on an anchor must be less than the structural design resistance of the tendon (see EN 1993-5) and than the design pull out resistance of the anchor (see NF P 14 282): Pd Ra,d/Rd = Ra,k / Rd a with Ra,k = min {Ra,m,mean/a1; Ra,m,mean/a2). If Ra,k comes from an investigation test, then Rd =1. If Ra,k comes from any calculation model, then Rd 1,4. For approach 2: a =1,1 for temporary and permanent anchors. for n = a1 a2 1 1,4 1,4 2 1,3 1,2 3 1,2 1,05 4 1,1 1 5 1 1
In addition, the service load is checked against the creep load obtained from tests.
Preliminary design
The pull out resistance Ra = qssDLs.
IRS IGU (Selective repeated injection) (Unique global injection) s qs s qs Gravel 1,8 Sandy gravel 1,6 - 1,8 0,1pLM + 0,05 Gravelly sand 1,5 - 1,6 Coarse, medium, fine, silty sands 1,4 - 1,5 0,1pLM + 0,05 Silt Clay 1,4 - 1,6 1,8 - 2,0 0,08pLM + 0,10 1,3 - 1,4 1,2 - 1,4 1,2 - 1,3 1,1 - 1,2 1,1 - 1,2 1,2
0,1pLM
0,1 pLM 0,06pLM+0,04
Marl, marly-limestone and weathered or fragmented chalk 1,1 - 1,2 1,8 0,07pLM + 0,13 Weathered or fragmented rock 1,2
0,05pLM+0,1
0,12pLM+ 0,08
1,1
0,1pLM+ 0,0
Conclusion
Our intention in France is to adopt design approach 2 for the design of retaining structures and anchorages and test procedure 3 for testing anchorages. In case design approach 3 is used, a model factor on the resistance has to be introduced.
BBRI
BGGG GBMS
Application of Ground Anchors, Nails and Tension Piles in Europe and Current Status of EN 1537 - Ground Anchors
Dr.-Ing. Wolf-R. Linder Brckner Grundbau GmbH, Essen, Germany Chairman CEN/TC 288
Dr Caesar M. Merrifield Coffey Geotechnics Limited, Manchester, United Kingdom Chairman CEN/TC 288/WG 13
BBRI & BGGG-GBMS
1.
INTRODUCTION
1.1. Programme of European Geotechnical Standards

Work on the Eurocodes started in 1979 with the objective of developing European design standards using the principle of partial safety factors and aiming at harmonizing design approaches in Europe for the replacement of existing national standards which were commonly using global safety factor approaches. Work was intensified after release by the European Commission of the Construction Products Directive (CPD) in 1988 which also necessitated European standardization of materials, building products, construction methods and testing. For coordination and supervision of European standardization the European Standards Committee (CEN) was established which commissioned the Technical Committee CEN/TC 250 to continue work on the Eurocodes. The provisional time schedule aimed at their publication and introduction already in 1993 and 1996, respectively. For geotechnical design and construction, the original programme foresaw four different standard packages to be dealt with by the Sub Committee SC 7: EC 7-1: Geotechnical design, EC 7-2: Laboratory testing, EC 7-3: Field testing and EC 7-4: Geotechnical construction.
Taking account of the great range of the subjects and realizing that only a comprehensive system of standards could be fully functional, the workload was later split between CEN/TC 250 SC7 and two new TCs: CEN/TC 250 SC7: EN 1997-1: Geotechnical design - general rules and EN 1997-2: Geotechnical design - ground investigation and testing; CEN/TC 288: CEN/TC 341: Geotechnical construction standards (Execution of special geotechnical work), established 1991 and Ground investigation and testing, established 2000.
Scope of CEN/TC341 working groups (WG) is the development of standards for: identification and classification of soil and rock, drilling and sampling methods and groundwater measurement (WG 1), cone penetration tests (WG 2), dynamic probing (WG 3), testing of geotechnical structures (WG 4) and borehole expansion tests (WG 5).
Whilst the work of the other WGs is running smoothly, that of WG 4 was initially sometimes controversial because of the different nature of subjects addressed in the standard. Testing of piles is an issue not directly connected to the execution of piles: Testing of prestressed ground anchors is however an integral part of the execution and an execution standard for anchors remains incomplete unless it contains respective provisions or is complemented by a testing standard available simultaneously.
Application of Ground Anchors and Status of EN 1537 Linder/Merrifield 14.05.2008
p. 3/17
BBRI & BGGG-GBMS
A summary of the most important geotechnical design, execution and testing standards is given in Table 1. Table 1: Special foundation engineering Standards: status
Committee Standard EN 1536 Bored Piles EN 1537 Ground Anchors EN 1538 Diaphragm Walls EN 12063 Sheet Piles EN 12699 Displacement Piles EN 12715 Grouting EN 12716 Jet Grouting EN 14199 Micro-Piles EN 14475 Reinforced Fills EN 14490 Soil Nailing EN 14679 Deep Mixing EN 15237 Vertical Drains EN14731 Deep Vibration EN 22477-1 Piles: Static axial compression testing EN 22477-5 Testing of anchors EN 22477-6 Testing of nails EN 1997-1 Geotechn. Design: General rules EN 1997-2 Ground investigation and testing Start Enquiry 1995 1996 1996 1997 1998 1998 1998 2002 2002 2002 2007 2003 2005 2003 2006 2005 ? Formal EN Vote Ratific. 1997 1997 1997 1998 2000 2000 2001 2004 2005 2009 2005 2006 2005 2009 2009 ? 2004 2005 2000 1999 1999 1999 2000 2000 2001 2005 2006 2010 2005 2007 2005 2010 2010 Notes confirmed: 2002 Systematic Review until 2010 confirmed: 2005 Systematic Review until 2011 confirmed: 2002 Systematic Review until 2010 confirmed: 2005 next Systematic Review: 2010 confirmed: 2005 next Systematic Review: 2010 confirmed: 2005 next Systematic Review: 2010 confirmed: 2006 next Systematic Review: 2011 first Systematic Review: 2011 First Systematic Review: 2011 final draft for Formal Voting being prepared first Systematic Review: 2010 first Systematic Review: 2012 first Systematic Review: 2010 final draft for Formal Voting being prepared final draft for Formal Voting being prepared work interrupted first Systematic Review: 2010
1992 1992 1992 1993 1994 1994 1994 1996 1997 1997 1999 1999 1999 2001 2001 2001 <1990
CEN TC288
CEN TC 341 WG 4
CEN TC 250 SC 7
2000
2005
2006
2007
first Systematic Review: 2012
Years in italics: anticipated or target dates
1.2 Drafting and Presentation of European Standards

European standards are mostly drawn up in English and are published in the three official languages of CEN, namely English, French and German, which are to be identical, especially regarding format, formulations and degrees of obligation of individual provisions.
p. 4/17
BBRI & BGGG-GBMS
CEN has prescribed strict rules for all parts of its standards including the presentation, figures, tables and the wording. The rules are contained in Internal Directives (CEN CENELEC (2006) Internal Regulations Part 3) and are identical for all CEN and ISO standards. The Standards distinguish between Requirements, Recommendations, Permissions, Possibilities or just Notes and respective wordings are prescribed. It must be kept in mind that common understanding and use of the language can deviate from the standardized formulations. This fact is of particular importance for the application of European Standards in different countries. Similar strict and formal requirements including procedures and time schedules are in place for the application for new works items for standardization, for the formation of Technical Committees, the drafting process and the presentation of draft standards for Enquiry, Formal Voting and Standard Review procedures. Member Bodies (MB) of CEN are the national standards institutions, e.g. AFNOR, IBN, NEN, BSI, DIN etc. Decisions regarding the acceptance of a new Work Item (WI), on acceptance of a draft to become a EN-Standard through Formal Voting (FV) or for the Systematic Review are taken by double majority-voting of MBs, i.e. requiring the single majority plus the 71 % weighted majority of MBs voting. The latter reflects to some extend the population of individual countries but is much in favour of smaller countries.
1.3 National Application Documents (NAD)

Legal and statutory requirements and conditions are different throughout Europe as are technical traditions, experience, education and the role of the parties involved in construction. Apart from relatively simple cases (like pure material standards), national differences will remain for some time. European standards should therefore rather define generally acceptable principles and requirements as common denominators than requirements that might not be acceptable throughout. They should leave room for national deviations as necessary, however establishing conditions for such permissions. Examples are the permission of reduced material testing requirements provided appropriate quality control systems are nationally specified (EN 1536) or different testing procedures for ground anchors (EN 1537 and EN 22477-5). The individual member countries are to check and provide for each individual standard and how it is matched with the respective national conditions. Normally a short national foreword or national appendix (NA) should be sufficient. More complicated issues can however require National Application Documents (NAD) for making a European standard operational.
1.4 General status of TC 288 Standards

Most standards on the work items defined by CEN/TC 250/SC7, TC 288 and TC 341/WG4 are completed and published both Europe-wide and nationally (see Table 1). The creation phase is basically completed. During a consolidation phase, the standards should gradually be improved and be harmonized within the family of geotechnical standards and with other reference standards. The periodic Systematic Review procedures are to be used for this purpose and main objectives should be as follows:
p. 5/17
BBRI & BGGG-GBMS
Gaining experience and making the standards operational with a minimum of NADs, Amending or making corrections as necessary, Condensing of standards (i.e. shortening), harmonisation of structure, terminology, definitions and providing similar layout, depth and appearance, Calibrating different language versions, Harmonisation with design, testing (e.g. anchors) and material standards (e.g. concrete, precast piles).
2.
GROUND ANCHORS NAILS TENSION PILES
2.1 Basic Differences

For the introduction of tensile forces into the ground, prestressed ground anchors are used as actively loaded elements. Non-prestressed anchors, nails or tension micro piles are passive elements on which the tensile forces develop as the result of earth and water pressures and are typically connected with larger displacements. The main parts of anchors, nails and tension micro piles and their respective principle load transfer are defined in Figures 1 and 2 (after EN 1537 and EN 14199). 4
Figure 1: Elements of a pre-stressed ground anchor (from EN 1537) and load transfer
p. 6/17
BBRI & BGGG-GBMS
Figure 2: Elements of a permanent soil nail (from EN 14490) and load transfer; tension micro piles act similar to nails
2.2 Context of Design, Execution and Testing Standards

The different types of ground anchors, nails and tension micro piles and their coverage in existing and planned European standards are listed in Table 2. The context of the standards within the European standard system is shown in Figure 3 illustrating the importance of the integrated approach and harmonization during the drafting and review processes. Table 2: Anchors, nails and tension micro piles covered by European standards
Type of Anchorage European Standard Execution (CEN/TC 288) Prestressed grouted anchor Grouted anchor Non Deadman anchor prestressed Screw anchor anchors Rock bolt Expander anchor Nails Tension micro piles n.a. = not applicable EN 1537 n.a. n.a. n.a. n.a. EN 14490 EN 14199 Testing (CEN/TC 341) EN 22477-5 n.a. n.a. n.a. n.a. n.a. EN 22477-6 (planned) EN 22477-2 (planned) EN 1997-1 Cl. 9.7 EN 1997-1 Cl. 7.6.3 & 10.2 EN 1997-1 Cl. 8 & 9.7 Design (CEN/TC 288)
p. 7/17
BBRI & BGGG-GBMS
EN 1997-2: Ground invest. and testing National Application Document (NAD) EN 1997-1: Geotechnical Design, Cl. 8 National Application Document (NAD)
Harmonisation
EN 1537: Ground Anchors National Application Document EN 22477-5: Testing of Achors National Application Document
EN XXX: Steel, National Application Document EN YYY: Grout, National Application Document
Figure 3: Context of execution, design and testing standards for ground anchors
3.
SCOPE OF EN1537 GROUND ANCHORS
3.1 Anchors included in the standard.

The standard EN1537 currently applies to the installation, testing, durability and monitoring of ground anchors where the load capacity of each anchor is tested. The standard also considers matters which should be taken into account into the design of anchors as part of a larger structural system. An annex (Annex D) which is an informative annex, provides more specific recommendations for the design of anchor capacity. Essentially the standard was developed in order to provide guidance on the execution" or the installation, testing and subsequent monitoring and maintenance of the anchor elements. At the time of development it was generally felt that the provision of design guidelines in EC7, Chapter 8, did not fulfil the needs of those who would be using the European standard initially. The scope of the anchors covered in the standard is quite explicit. It is also widely believed that the scope should be better defined, only including anchors which are posttensioned and which have a free length. This certainly excludes alternative anchor solutions such as tension piles, screw anchors, mechanical anchors, soil nails, expander anchors or deadman anchors.
p. 8/17
BBRI & BGGG-GBMS
3.2 Elements of anchor recognised by the Standard

To reinforce the definition of the anchor the standard recognises that it consists of an anchor head, a free anchor length and a fixed anchor length which is bonded to the ground by grout. It should be noted, however, that the standard is indiscriminate in its definition of the term "ground" which is taken to encompass both soil and rock. It is not in the scope of this paper to discuss the roles of the fixed and free lengths in the overall behaviour of ground anchors other than to say that the structural characteristics of an anchor as defined in the standard suggest that the mechanisms of load transfer and serviceability behaviour differ materially from other types of anchors which are not represented in the standard. The standard highlights the importance of the anchor head in that this is the primary load transfer element to the anchored system. The behaviour of the anchor as a load transfer element is dependent upon two main characteristics of the anchor head. These are; a) the ability of the head to maintain a load and, if necessary, to allow the anchor to be restressed at some time in its working life and b) the ability of the anchor head to resist wakening due to material degradation through corrosion. Figure 4 (taken from EN 1537, as is Figure 1) shows schematically the elements of an anchor as defined in the standard. This figure includes more detail than that which is required to define an anchor in terms of the standard.
Key
1 2 3 4 5 Anchorage point at jack during stressing Anchorage point at anchor head in service Bearing plate Load transfer block Structural element 6 7 8 9 10 Soil/rock Borehole Debonding sleeve Tendon Grout body in fixed length
Figure 4: Schematic representation of an anchor (EN 1537: 1999)
p. 9/17
BBRI & BGGG-GBMS
3.3 Temporary and Permanent Anchors

The standard state explicitly that it covers both temporary and permanent anchors. It goes on to define temporary anchors as those which have a design life of less than two years. A number of National Standards (e.g. DIN4125) accommodate anchors which may be defined as temporary anchors with extended design lives. This may include anchors which are required to be in service for up to five years. The standard is rigorous in this definition for reasons of safety, both to that of the integrity of the anchors and consequently the overall structures and also of third parties using and maintaining the structures in question. Opinion is divided on the necessity for this rigid definition and it has been suggested that a third definition of design life should be promoted that of extended temporary anchor. This, in essence would provide a mechanism to provide design and execution guidance on a broader range of anchor uses. With the development of more and more complex deep basement solutions to high rise buildings and the extension of time required for the temporary works to be effective, the application of extended temporary anchors to these solutions may well result in faster, cheaper and less disruptive construction solutions to be applied in innovative designs.
4. 4.1
PHILOSOPHY OF ANCHOR EXECUTION Integrated nature of design, installation, testing and monitoring
The successful implementation of a ground anchor solution relies on an integrated approach to the design, installation, testing and eventual in-service condition monitoring of the system. Anchor solutions are unique in that each and every working anchor is subjected to an acceptance test to ensure that it conforms to the design criteria. The elegance of approach which requires validation of material behaviour, design conformity and eventual serviceability checks, results in good engineered solutions. Incorporated within EN 1537 is an approach, regarded as industry standard across Europe and, indeed, internationally, which has been tested successfully over a number of years. Notwithstanding modest differences in the approach to design across Europe, the adoption of the limit state approach in EN 1537, based on the behaviour of the tendon, is rational and results in designs which have been shown to be not dissimilar to those obtained using national standards.
4.2 Design basis for anchors

The basis of design was highlighted by Merrifield et al (1997) and has seen limited use in a number of countries in Europe for the last ten years. Based on the limit state design approach, the recommendations in EN 1537 were complied with a view to simplifying the understanding of the design process. The ultimate limit state design demonstrates equilibrium of the design values of actions acting on a structure and the design resistance of the structure in every possible design situation. The serviceability limit state, however, is associated with proving that deformations and displacements of a structure using characteristic values of actions and material properties are acceptable. Inevitably specialist anchor contractors have continued to evaluate the capacity of their own an-
p. 10/17
BBRI & BGGG-GBMS
chor fixed length designs using the most appropriate equilibrium analyses. (Merrifield et al, 1997). The recommendations in the design are predicated on the fact that the behaviour of the high tensile steel tendon within an anchor is the most about which is known at the time of design, and therefore this is the element upon which the design is based. EN 1537 requires that design and verification of anchored structures at serviceability limit state should be executed by checking appropriate design situations using characteristic values of actions, ground properties and geometrical data. This is given in Annex D of EN 1537. The standard then refers the designer to clause 2.4.6 .of ENV 1997-1:1994 which addresses the limits of allowable displacement and deformation of the structure and the ground adjacent to anchored structures. The tolerance of supported structures to displacement and distortion should also be considered. Likewise EN 1537 requires verification that the design value of the effect of stabilising actions is at least or greater than the design value of the effect of the destabilising actions on the anchored structure. For example, when considering a limit state of rupture or excessive deformation of a section, anchor or connection, the following should be verified: Ed Rd where Ed Rd is the design value of the effect of actions, such as anchor force; is the corresponding design resistance, associating all structural properties with the respective design values.
This simplified process is then followed when considering either a limit state of rupture or excessive deformation of a section (i.e. tendon) safe in the assumption that Rd is considerably lower than either the characteristic internal or external anchor resistance (i.e. the tendon tensile strength Rik or the grout/ ground bond strength, Rak , respectively). In fact EN 1537 requires that;Rd = where Rk
R = 1.35.
Furthermore the anchor lock-off load P0 should be no greater than 0.60Ptk, the characteristic load capacity of the tendon. By following these guidelines, the designer ensures that the anchor, as a stressed element providing load transfer between the structure and the surrounding soil, should never be loaded to a condition which would compromise the behaviour of the anchor.
p. 11/17
BBRI & BGGG-GBMS
4.3 Harmonisation within different European approval systems

The harmonisation of the National Standards into a single European Standard has been undertaken in accordance with CEN/CENELEC (2006). These regulations are designed to allow the constitution of working groups comprising national experts, whilst maintaining their understanding and awareness of their own national positions, they have the right to act in a personal capacity in the working group. They, therefore, provide expert knowledge and experience in compiling a well-balanced and considered standard which may be used throughout Europe without prejudice. The harmonisation process for EN 1537 and its subsequent revisions necessitate interaction with other TC 288 standards and Eurocodes to ensure harmonisation of terminology, symbols and, in some cases, basis of execution and design. The harmonisation process should also be sensitive to the current legal systems prevailing in each sovereign state within the CEN area. In some countries the design and installation of anchors is subject to a greater approval application regime by regulatory authorities than in others. The standard should be cognisant of the differences in the regulatory and legal systems across Europe.
4.4 Principles of testing

4.4.1 Stress testing
Anchors are unique in the context of ground improvement elements, in that each working anchor should be subjected to a load acceptance test. EN 1537, in fact, requires each working anchor to be subjected to an acceptance test. This is the case in the vast majority of European and international standards. EN 1537, Clause 9, Testing Supervision and Monitoring, sets out three classes of stress testing; Investigation tests Suitability tests Acceptance tests.
Each category of test above has a different set of objectives. The first two may be regarded as subdivisions of the general category of assessment tests as defined in EN 1997-2. The investigation tests have been designed to allow the designer to establish the load resistance of the proposed anchors in relation tot the ground and the materials used in the anchor itself. These tests may also be used to evaluate the competence of the proposed contractor. Essentially these tests are undertaken to failure or the equivalent thereof. The suitability tests were designed to allow a greater understanding of the creep or load loss characteristics at proof and lock-off load levels and to provide an indication of the apparent free length of the anchor. The objectives of the acceptance tests on each and every working anchor are to measure the performance under working load conditions thus providing the designer and contrac-
p. 12/17
BBRI & BGGG-GBMS
tor with the assurance that both the design and installation processes have resulted in an anchor which is fit for purpose in accordance with the standard. The current standard does not prescribe a load or stress test method to be applied for any of the test categories above, but provides examples of three methods which are deemed to be acceptable. The inclusion of these methods is a consequence of the recognition that each is favoured traditionally in different parts of Europe. In the interests of one of the principles of harmonisation that no current practice, which is deemed to be acceptable, is to be discriminated against in the harmonisation process they were incorporated into the standard as an informative annex.
4.4.2 Durability testing

EN 1537, Clause 6, Materials and Products, highlights the importance of material compatibility and whole life durability. The clause considers each component in turn drawing the designers attention to the hazards associated with the loss of durability during the working life of the anchor. The necessity to protect the high tensile steel anchor tendon against corrosion is strongly reinforced, requiring stringent standards of protection to be applied. Whilst the principles of corrosion protection are the same for all parts of the anchor, the standard recognises that different detailed treatments are required for the bond length, the free length and the head. Case studies have shown that corrosion at the head and in particular where the tendon is restrained, is particularly severe and very often the area which suffers the most degradation due to corrosion. As in the case of stress testing, the practice of designing and determining adequate anchor durability varies across Europe. For this reason the minimum corrosion protection requirement for a permanent anchor tendon or tendon has been articulated as a single continuous layer of corrosion preventive material which does not degrade during the design life of the anchor. This requirement is to be either verified by an appropriate test, or the anchor is to be provided with a sacrificial layer of corrosion protection, hence ensuring the integrity of a single layer insitu. The standard provides details in informative annexes of systems tests to determine the adequacy of corrosion protection. These are i) the electrical resistance test of anchors insitu and under stress and ii) a laboratory based gun barrel systems test, again to check for any protection damage under stress. Discussion at the recent international conference on ground anchorages and anchored structures in service (2007) (to be published) underlined the difficulty the industry was having in the implementation of the electrical resistance method in the determination of acceptable corrosion protection. The reliability of the results was called into question, given the dependence of the system on climatic conditions.
4.5 Monitoring and maintenance

The routine monitoring and maintenance of working anchors, especially for those over 30 years old that may have been designed with corrosion protection considered inferior or inadequate by todays standards, is a matter of great concern. Routine programmes of inspection and monitoring, where satisfactory condition and service performance can
p. 13/17
BBRI & BGGG-GBMS
be confirmed, can extend the service life of anchored structures, such as post-tensioned dams, dry docks, bridges, sports stadia and multi-storey buildings etc which represent key elements of a countrys infrastructure. (Littlejohn and Mothersille, 2007). EN 1537 provides a modicum of guidance on monitoring and the level of records that should be retained. Very little is said on the necessity for regular maintenance. This may be seen as a disadvantage to the use of an anchored solution. Records show, however, that even very large structures such as dams, high buildings and steep slopes which, in the event of failure could cause dramatic loss of life and assets, are monitored and maintained on a regular basis as part of the whole life asset management cycle. Many examples of long term monitoring and maintenance and the current practices in both European and other countries, such as South Africa, Australia and the USA are recorded in the proceedings of the recent international conference on ground anchorages and anchored structures in service (2007).
5.
EXAMPLES OF FAILURES DUE TO LACK OF ANCHOR DURABILITY
The following plates illustrate the consequences of insufficient monitoring or anchor elements which were inadequately corrosion protected in the context of the standard today, courtesy of Dr Devon Mothersille, Geoserve Global Ltd.
Plate 1 Severely corroded anchor head showing deterioration of bearing plate and barrel loss due to strand slippage after 37 years service
Plate 2: Heavily corroded anchor having sustained tendon failure after 28 years in service in a marine environment in the UK
p. 14/17
BBRI & BGGG-GBMS
Plate 3: Anchor head after removal of cap showing corroded barrels and remnant grease after 30 years in service
There are a number of examples of anchors in service which have failed as a consequence of corrosion. Many of these cases could have been avoided if a coherent and reliable monitoring and maintenance regime had been established at the time of installation. Construction records are also useful in the forensic analysis of future anchor behaviour. These include grout pressures, volume grout take and stressing records.
6.
REVISION OF EN 1537 EXECUTION OF SPECIAL GEOTECHNICAL WORK GROUND ANCHORS
6.1 Basis of revision

The revision of EN 1537 is being undertaken by CEN/TC288/WG14 under the authority of TC288. The working group will draft the document in accordance with the work specifications, guidelines and time schedule provided by TC288 and in accordance with CEN drafting rules. As an individual expert, each member of the working group maintains his/her understanding and awareness of their own national position by being in contact with the national delegation to TC288 and with related standardization activities within the National Standards Body of their home country. The working group is currently assessing the comments received through the national enquiries and resolving issues associated with these comments. In each case the comments are being considered and incorporated within the draft revision, where appropriate. CEN recognises the importance of maintaining the right balance between the different interests when appointing experts to ensure that the breadth of technical and user expertise is obtained and to ensure that no interest group has a dominating position. This is the case with respect to CEN/TC288/WG14.
p. 15/17
BBRI & BGGG-GBMS
6.2 Separation of testing and design

The current revision of EN 1537 involves the redistribution of two elements which formed part of the original standard into other standards. These are;a) the methods of anchor stress testing in Annex E. These are to be incorporated into EN ISO 22477-5 currently being drafted by CEN/TC341/WG3 b) anchor design which will be incorporated into EC7. Appropriate reference to these matters will be made in the revised EN 1537 but the substantive text on these matters will reside in the standards into which these topics will be placed. General agreement exists between the working groups drafting EN 1537, EN ISO 22477-5 and the maintenance group responsible for EC7, that the transfer and subsequent issues of compatibility will be managed with appropriate care.
6.3 New developments

Whilst the revision of the current EN 1537 requires analysis of incompatibility across Europe, issues raised by National Bodies and the general improvement of text and understanding raised in the deliberations of the working group, there is the belief that the standard should address developments in anchor technology which have manifested themselves since the initial draft. Developments which should certainly be considered are the emergence of the single bore multiple anchor and the various designs of removable anchor. Developments are being made in remote condition monitoring or non-destructive integrity testing, both in Europe and other countries such as Japan (Ivanovi et al, 2007; Tanaka et al, 2007). The revised draft should also be so constructed to allow such developments to be incorporated if and when they are validated and considered sufficiently reliable for constant use on site.
6.4 Current status of EN 1537 revision

An inaugural meeting of CEN/TC288/WG14 was held in Manchester, UK on 14th and 15th April, 2008. The working group meets again in Munich on 19th and 20th June, 2008. The programme of revision is established and the working group is confident that the work will be finished in time for the TC288 meeting in June, 2009, at which time a decision to start the enquiry will be made.
p. 16/17
BBRI & BGGG-GBMS
7.
REFERENCES
CEN CENELEC, (2006) Internal Regulations Part 3: Rules for the structure and drafting of CEN/CENELEC Publications, Brussels. Ivanovi A., Neilson R., Starkey A., and Rodger A. (2007) Common anchorage issues addressed by numerical modelling. pp209-219, Proc. International Conference on Ground Anchorages and Anchored Structures in Service, ICE, London. Littlejohn, S & Mothersille, D (2007) Maintenance testing and service behaviour monitoring of permanent ground anchorages. Keynote Presentation, Proc. International Conference on Ground Anchorages and Anchored Structures in Service, ICE, London. Merrifield C., Barley A., Von Matt U. (1997) The execution of ground anchor works: The European Standard prEN1537. pp492-502, Proc. International Conference on Ground Anchorages and Anchored Structures, ICE, London. Tanaka K., Okada H. and Izakura M. (2007) The development of the method of detecting the anchor tensile force to impact elastic wave. pp238-249, Proc. International Conference on Ground Anchorages and Anchored Structures in Service, ICE, London. Whitworth, M and Parrish, S (2007) Long term monitoring and routine maintenance of ground anchorages at Devonport Royal Dockyard, Plymouth UK pp63-71, Proc. International Conference on Ground Anchorages and Anchored Structures in Service, ICE, London.
p. 17/17

International Symposium Ground Anchors

Enviado por

Dados do documento

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

International Symposium Ground Anchors

Enviado por

Direitos autorais:

Formatos disponíveis

PROCEEDINGS INTERNATIONAL SYMPOSIUM GROUND ANCHORS

Limelette test field results

WORKING GROUP GROUND ANCHORS

International Symposium 14 May 2008 Ground Anchors

BBRI & BGGG-GBMS

Ground Anchors 14.05.2008

EUROPEAN STANDARDS EXECUTION OF SPECIAL GEOTECHNICAL WORKS:

2.1. EN 1537 1999 : ground anchors.

J. Maertens 14.05.2008 p. 3/9

BBRI & BGGG-GBMS

Ground Anchors 14.05.2008

2.2. EN 14199 2003 : Micropiles

2.3. prEN 14490 : Soil nailing

BBRI & BGGG-GBMS

Ground Anchors 14.05.2008

J. Maertens 14.05.2008 p. 5/9

BBRI & BGGG-GBMS

Ground Anchors 14.05.2008

GROUND ANCHORS IN BELGIAN DAILY PRACTICE

J. Maertens 14.05.2008 p. 6/9

BBRI & BGGG-GBMS

Ground Anchors 14.05.2008

J. Maertens 14.05.2008 p. 7/9

BBRI & BGGG-GBMS

Ground Anchors 14.05.2008

4.3. Control tests

J. Maertens 14.05.2008 p. 8/9

BBRI & BGGG-GBMS

Ground Anchors 14.05.2008

J. Maertens 14.05.2008 p. 9/9

International Symposium 14 May 2008 Ground Anchors

Ground Anchors : Overview of Types, Installation Methods and Recent Trends

Ir. Flor De Cock Geotechnical Expert Office Geo.be

BBRI & BGGG-GBMS

Ground Anchors 14.05.2008

APPLICATIONS OF GROUND ANCHORS

Figure 1 : applications of ground anchors (Samwoo)

BBRI & BGGG-GBMS

Ground Anchors 14.05.2008

TERMS AND DEFINITIONS

Figure 2 : sketch of a (grouted) ground anchor (EN 1537:1999)

GROUND ANCHOR TYPES CLASSIFICATION

BBRI & BGGG-GBMS

Ground Anchors 14.05.2008

The installation method? The market position and situation?

Normally no grouting (except for formation of the expander body)

Figure 3 : ground anchor classification proposal

BBRI & BGGG-GBMS

Ground Anchors 14.05.2008

Tensile type anchor Single stage

Tensile type anchor Multi stage

Figure 4a and 4b : anchor-ground interaction for tensile type anchors

BBRI & BGGG-GBMS

Ground Anchors 14.05.2008

Compressive type anchor

End bearing type anchor

GROUND ANCHOR COMPOSITIONS INSTALLATION HEADLINES

5.1. Friction anchor of the tensile type

BBRI & BGGG-GBMS

Ground Anchors 14.05.2008

Figure 5: Typical execution phases of flush-drilled friction anchors

BBRI & BGGG-GBMS

Ground Anchors 14.05.2008

BBRI & BGGG-GBMS