OPTIMIZATION OF RANDOM RUBBLE MASONRY RETAINING WALL DESIGN A. N. Santhajeewa (118633P) Degree of Master of Engineering in Structural Engineering Department of Civil Engineering University of Moratuwa Sri Lanka April 2015 OPTIMIZATION OF RANDOM RUBBLE MASONRY RETAINING WALL DESIGN Assurappulige Nalaka Santhajeewa (118633P) Dissertation submitted in partial fulfillment of the requirements for the degree Master of Engineering in Structural Engineering Design Department of Civil Engineering University of Moratuwa Sri Lanka April 2015 i DECLARATION I declare that this is my own work and this dissertation does not incorporate without acknowledgement any material previously submitted for a Degree or Diploma in any other University or institute of higher learning and to the best of my knowledge and belief it does not contain any material previously published or written by another person except where the acknowledgement is made in the text. Also, I hereby grant to University of Moratuwa the non-exclusive right to reproduce and distribute my dissertation, in whole or in part in print, electronic or other medium. I retain the right to use this content in whole or part in future works (such as articles or books). Date: 8th of April 2015 A.N. Santhajeewa The above candidate has carried out research for the Masters Dissertation under my supervision. Date: 8th of April 2015 Dr. Mrs. D, Nanayakkara ii ABSTRACT The conventional earth retaining structures built using Random Rubble Masonry (RRM) are designed as gravity retaining structures where weight of the structure is used for its stability. In Sri Lanka, RRM retaining walls is the most common type of retaining structure for low retaining heights .However; in general, engineers are reluctant to adopt RRM for retaining heights more than 3m high, due to comparatively large sections obtained as the result of conventional design practice. More optimal and creative solutions could be obtained even for low retaining heights, if design material properties of RRM are known. In this study, use of flexural strength of RRM and adopting a Reinforced Concrete (RC) Tie- back at the top of the retaining wall to optimize the conventional design was explored. The experimental investigation was carried out to find out the flexural, compressive and shear strength of RRM. Further, bond strength between Reinforced Concrete (RC) and RRM was investigated. These tests results have been used to ascertain the adoptability of suggested optimizations. From the experimental study, it was concluded that magnitude of material strengths of RRM are sufficient for considerable optimization by taking into account the effect of flexural strength of RRM and adopting a Tie- back. The width of the base of wall section reduction for 3m high retaining wall was 28% as the result of the optimization. Keywords: Random Rubble Masonry, Retaining walls, Optimization, Tie- back, Flexural Strength. iii ACKNOWLEDGEMENT There are many individuals who deserve acknowledgement for their contribution towards successful completion of this research. First, I would like to express my gratitude to my supervisor, Dr. Mrs. D. Nanayakkara for her valuable advice, guidance and assistance throughout the entire period of study. I am much grateful for sharing her vast knowledge and expertise on the field of Masonry. Secondly, my sincere acknowledgement is towards my employer, Central Engineering Consultancy Bureau for granting me the sponsorship for following the course and other assistances provided for my research works. I am much grateful to the Head of the Department of Civil Engineering, the Course Coordinator of Master of Structural Engineering, the staff of the Department of Civil Engineering and the staff of the structural laboratory for their valuable guidance and corporation related to all experimental works. The assistance rendered by undergraduate students Buddhi, Maithri and Suresh for experimental works also gratefully acknowledged. My very special thanks go to my dear wife Mahesha for her continuous encouragement, assistance and patience during the entire period. My research would never be successful without her tremendous support. Lastly, there are many friends and colleagues who have not been personally mentioned here that I am much indebted to their contribution at various stages of the research to make it successful. iv TABLE OF CONTENTS Declaration of the Candidate & Supervisors i Abstract ii Acknowledgements iii Table of Contents iv List of Figures viii List of Tables x List of Abbreviations xi List of Appendices xii 1. Introduction 1 1.1 General 1 1.2 Need for Research 2 1.3 Optimization of Gravity Retaining wall 3 1.3.1 Effect of Tie – back 3 1.3.2 Effect of Flexural Strength 3 1.4 Objectives of the Research Study 4 1.5 Methodology 4 1.6 Outline of the Dissertation 5 2. Literature Review 6 2.1 Introduction 6 2.2 Design of Retaining Walls 6 2.2.1 Limit State Design Method 6 2.2.2 Conventional Design Method 7 2.2.2.1 Stability Analysis in Conventional method 7 2.2.2.1.1 Stability Analysis in Craig, R.F [10] 8 2.2.2.1.2 Stability Analysis by Liu,C and Evett, J.B[11] 10 2.3 Adopting Tie back Effect in Gravity Retaining Wall Design 12 v 2.3.1 Innovative Earth Retaining System Adopted for the Proposed Printing Complex at Mawaramandiya 12 2.3.2 Wall Foundations of Proposed Block no.10- Mahinda Rajapaksha Vidyalaya, Homagama 14 2.4 British Standards relevant to Random Rubble Masonry Design 15 2.5 Standard Construction Practices of Random Rubble Masonry in Sri Lanka 16 2.5.1 Type of Stones 16 2.5.2 Sizes of Stones 16 2.5.3 Dressing of Stones 18 2.5.4 Mortar 18 2.5.5 Mortar Joints 19 2.5.6 Laying 20 2.5.7 Curing 20 2.6 Previous Experimental Investigations on Material properties of Random Rubble Masonry 20 2.6.1 Compressive Strength 20 2.6.2 Shear Strength 21 2.7 Experimental Investigations on Masonry- Concrete Interface 23 3. Different Approaches used for Design of Random Rubble Masonry 25 3.1 Introduction 25 3.2 Case 1 - Design of RRM Retaining wall using Conventional Method 26 3.3 Case 2 - Retaining Wall assuming RRM will not fail due to flexure 31 3.4 Case 3 – Design of Retaining Wall with the Tie back effect 35 3.5 Summary of the Results obtained from Three Case Studies 41 4. Experimental Study 43 vi 4.1 General 43 4.1.1 Preparation of Test Specimens 44 4.2 Experimental Set-up 45 4.2.1 Testing for Flexural Strength of RRM 45 4.2.2 Testing for Shear Strength of RRM 47 4.2.3 Testing for Shear Strength at Concrete- RRM Interface 49 4.2.4 Testing for Compressive Strength of RRM 51 5. Analysis of Test Results 52 5.1 Flexural Strength 52 5.1.1 Experimental Results 52 5.1.2 Evaluation of Results 56 5.1.3 Comparison of Test Results with Brick/ Block Masonry Flexural Strengths 58 5.2 Shear Strength 59 5.2.1 Experimental Results 59 5.2.2 Evaluation of Results 61 5.2.3 Comparison of Results with previous research findings 64 5.3 Shear Strength at Concrete- RRM interface 64 5.3.1 Experimental Results 64 5.3.2 Evaluation of Results 67 5.3.3 Comparison of Results with previous research findings on Shear strength at Concrete- Masonry interface 68 5.4 Compressive Strength 69 5.4.1 Experimental Results 69 5.4.2 Evaluation of Results 69 5.4.2.1 Compressive Strengths of each Sample 69 5.4.2.2 Mean Compressive Strength 70 5.4.2.3 Characteristic Compressive Strength 71 vii 5.4.3 Comparison of Results with previous research findings on Compressive Strength of RRM 71 5.5 Summery of Test Results obtained by the Experimental Study 72 6. Conclusions and Recommendations 73 6.1 Use of Experimental Results for the improvements of RRM Retaining wall Design 73 6.1.1 Flexural Strength 74 6.1.2 Effect of Tie back 74 6.2 Suggestions for Future Works 75 Reference List 77 Appendix A 79 Appendix B 81 Appendix C 82 Appendix D 83 Appendix E 84 viii LIST OF FIGURES Page Figure 1.1 RRM retaining wall in front of Nuwara Eliya Post Office 2 Figure 1.2 Application of Tie-back for RRM retaining walls 3 Figure 2.1 Loads and base reactions of retaining walls 8 Figure 2.2 Modified RRM retaining wall system adopted 13 Figure 2.3 Construction of Tie back arrangement at the site 13 Figure 2.4 Wall foundations at the rear side of the class block 14 Figure 2.5 Typical bond patterns and Specifications for Bushing, amount of Chips and through stone 17 Figure 2.6 Types of Stones used in RRM 18 Figure 2.7 Types of Joints used in RRM 19 Figure 2.8 Triplet setup by Milosevic J. [17] 22 Figure 2.9 Relationship between Normal stress and Shear stress of samples 23 Figure 2.10 (a): Testing Setup 24 (b): Shear deformation while applying the load 24 Figure 3.1 Loadings acting on Retaining wall for Case 1 26 Figure 3.2 Loadings acting on Retaining wall for Case 2 31 Figure 3.3 Assumed Base Pressure Variation for Case 2 33 Figure 3.4 Loadings acting on Retaining wall for Case 3 36 Figure 3.5 Possible Shear Failure Planes of RRM for Case 3 40 Figure 4.1 (a): Preparing samples for Flexural Strength Test 44 (b): Pre-Compressed Specimens 45 Figure 4.2 (a): The Plane of Bending is Vertical 45 (b): The Plane of Bending is Horizontal 45 Figure 4.3 The test set up for Specimens bent about Vertical Axis (Plane of Bending is horizontal) 46 Figure 4.4 The test set up for Specimens when the plane of bending is vertical 46 Figure 4.5 Set up for Triplet tests as in [3] 47 ix Page Figure 4.6 Set up adopted for Shear Strength Test 48 Figure 4.7 Test set up for Shear Strength Test 49 Figure 4.8 Set up for investigating Shear Strength at Concrete-RRM Interface Test 50 Figure 4.9 Set up for Compressive Strength Test 51 Figure 5.1 Failure patterns of Specimens (When the Plane of Bending is horizontal) (a): Specimen 1 53 (b): Specimen 2 53 (c): Specimen 3 53 Figure 5.2 Failure Patterns of Specimens (When the Plane of Bending is Vertical) (a): Specimen 4 54 (b): Specimen 5 55 (c): Specimen 6 55 Figure 5.3 (a): Failure Patterns of Specimens – Specimen 1 59 (b): Failure Patterns of Specimens – Specimen 2 59 (c): Failure Patterns of Specimens – Specimen 3 60 (d): Failure Patterns of Specimens – Specimen 4 60 (e): Failure Patterns of Specimens – Specimen 5 60 (f): Failure Patterns of Specimens – Specimen 6 60 Figure 5.4 Variation of individual Shear Strength values with the Pre- Compressive Stresses 62 Figure 5.5 Failure Pattern of Specimens for Shear Test (a): Specimen 1 65 (b): Specimen 2 65 (c): Specimen 3 66 (d): Specimen 4 66 (e): Specimen 5 66 (f): Specimen 6 66 x LIST OF TABLES Page Table 2.1 Summary of guidance on British and British European Standards relevant to Natural Stone 15 Table 2.2 Characteristic Compressive Strength of RRM for Mortar designation of 1:5 21 Table 2.3 Comparison of Characteristic Compressive Strength of RRM and Brick work for Mortar designations of 1:5 and 1:8 21 Table 3.1 Summary of results obtained from Three Case Studies 41 Table 5.1 Results of the test on Flexural Strength (When the Plane of Bending is horizontal) 52 Table 5.2 Results of Flexural Strength Test (When the Plane of Bending is Vertical) 54 Table 5.3 Flexural Strength of RRM specimens 56 Table 5.4 Characteristic Flexural Strength of RRM 57 Table 5.5 Flexural strength of Brick and Block Masonry as per BS 5628-1:1992 58 Table 5.6 Results of the Test on Shear Strength 59 Table 5.7 Shear Strength results for different Pre-Compressive Stresses 62 Table 5.8 Results of Shear Strength Test 65 Table 5.9 Results of test carried out for Shear Strength at Concrete- Masonry Interface 68 Table 5.10 Results of Compressive Strength Test 69 Table 5.11 Compressive Strength Results of each sample 70 Table 5.12 Characteristic Compressive Strength of RRM for Mortar designation of 1:5 71 Table 5.13 Summery of Strength Parameters of RRM 72 Table 6.1 Results obtained through different Design Approaches 73 Table 6.2 Extent of Optimization for 1-3m Retaining Heights 75 xi LIST OF ABBREVIATIONS Abbreviation Description RRM Random Rubble Masonry RC Reinforced Concrete BS British Standard ICTAD Institution of Construction Training & Development HM Hydraulic Mortar AM Air Lime Mortar ASTM American Society for Testing and Materials xii LIST OF APPENDICES Appendix Description Page Appendix – A Flexural Strength -Experimental Data and Results 79 Appendix – B Shear Strength -Experimental Data and Results 81 Appendix – C Shear Strength at Concrete Masonry Interface – Experimental Data and Results 82 Appendix – D Compressive Strength – Experimental Data and Results 83 Appendix – E Calibration Reports of Proving Rings 84 1 Chapter 01 INTRODUCTION 1.1 General Masonry built with rubble stones of random sizes and shapes are referred as Random Rubble Masonry (RRM). The stones used in RRM are only hammer dressed and they are bonded together with comparatively thick mortar joints. RRM is one of the oldest forms of masonry construction used all over the world. It was widely used due to its durability and availability of materials. However, nowadays, higher cost of stones, depletion of sources of materials due to excessive extraction to produce aggregates for construction works, difficulties associated with transporting and handling, identification of cheaper construction materials have limited its usage. Nevertheless, RRM has been commonly used in Sri Lankan construction industry, mainly for walls exposed to moisture environments and for walls where aesthetic considerations are governed. Earth retaining walls built using RRM have also been wildly used in Sri Lanka over the centuries, especially in housing construction and in infrastructure development projects. Still, the RRM retaining wall is the most common retaining wall type in Sri Lanka for low retaining heights. Figure 1.1 shows a retaining wall built along a property line. 2 Figure 1.1: RRM Retaining Wall in front of Nuwara Eliya Post Office 1.2 Need for Research In general, RRM retaining wall is designed as a Gravity retaining structure where the weight of the structure is used for its stability. Engineers do the design by proportioning the sections using “Middle-third rule” or “Flexural Formula”. Usually, the use of RRM for retaining heights over 3m is not economical. Even if stones are readily available and cheap in the locality, engineers are reluctant to use RRM for larger retaining heights due to large sections gained as a result of the traditional design practice. More optimal and creative solutions could be obtained even for low retaining heights, if design material properties of RRM are known. However design guild lines or strength parameters given in Codes of Practice, National Standards or Specifications relevant to RRM are not sufficient for engineers to deal with this material with confidence. Even the Sri Lankan engineers have not taken much of interest in doing research in this area compared to research carried out in respect of other construction materials. Hence study on RRM is worthwhile to enhance the effective use of RRM. 3 1.3 Optimization of Gravity Retaining Wall In this study, the following effects were taken into consideration to optimize traditional gravity retaining wall design. 1.3.1 Effect of Tie - back The Tie- back reduces the Bending moment developed in the retaining wall due to lateral loads. This can be achieved by introducing Reinforced Concrete (RC) beams which are laterally restrained, cast on top of retaining walls. Application of this arrangement for a house constructed on a sloping ground is illustrated in Figure 1.2. In this case, Retaining walls along grid 1 and 2 are supported by the Tie-back arrangement of RC beams. Figure 1.2: Application of Tie-back for RRM retaining walls 1.3.2 Effect of Flexural Strength In general, gravity retaining walls are proportioned so that no tension is developed at the Soil- Base interface, and hence no flexural stresses develop within the material as well. However, masonry possesses low flexural strength and hence, use of this flexural strength for optimizing traditional gravity wall was investigated. 4 1.4 Objectives of the Research Study The objective of this research was to investigate the following material properties of RRM experimentally in order to optimize the conventional RRM gravity Retaining wall design: a) Flexural Strength; b) Shear Strength; c) Compressive Strength and d) Bond between RC- RRM interface. 1.5 Methodology The main objective of this study is to optimize the traditional gravity retaining wall design by considering the effects of the tie-back and the flexural strength of RRM. To achieve the above goal, following Methodology was adopted:  A literature review on previous research wok carried out in the area of study;  Review the literature evolved for design guidance for RRM;  Gather information related to RRM retaining walls, including the most commonly used mortar mixes for RRM, Retaining wall heights & wall thicknesses and currently adopted design methods etc.;  Study test methods to investigate flexural strength, shear strength, compressive Strength and the bond strength at the concrete- masonry interface;  Carry out experimental investigation to evaluate flexural strength, shear strength, compressive strength and bond between Concrete – RRM interface;  Comparison of the results with previous research findings of similar nature; and  Optimizing the Retaining wall design, considering the effect of strength parameters found from the experimental study. 5 1.6 Outline of the Dissertation The second chapter of this dissertation deals with the literature review, which includes the study on retaining wall design approaches, previous attempts of optimization of RRM retaining walls, Standard guidance/construction practices of RRM and the previous research findings relevant to RRM. The third chapter summarizes the different approaches used for design of RRM retaining wall, which were carried out to assess the effect of tie- back and the flexural strength of RRM to conventional design approach. The fourth chapter provides all details of experimental investigation and the analysis of results is presented in Chapter 5. Further, it includes a comparison of obtained results with the previous research findings of similar nature. The dissertation concludes with Chapter 6, indicating the conclusions of the study and giving suggestions for further research work. 6 Chapter 02 LITERATURE REVIEW 2.1 Introduction The literature review was carried out to gather information from previous research studies in this area of study and to acquire the knowledge on widely adopted design approaches of RRM. This chapter summarizes the important and most relevant information gathered from the literature for this research study. 2.2 Design of Retaining Walls There are two widely accepted design approaches available for design of gravity retaining walls as follows: i) Limit state design method; and ii) Conventional design method. 2.2.1 Limit State Design Method The principle of Limit state design was introduced with the BS 8002- Code of Practice for Earth retaining structures, 1994[6]. The latest revision of the British Standard for it, which is Eurocode 7 [7] is also based on same design principle. According to the principles of limit state design, an earth-retaining structure must comply with both ultimate and serviceability limit states. Ultimate limit states are those involving the collapse or instability of the structure as a whole or the failure of one of its components. Serviceability limit states are those involving excessive deformation, leading to damage or loss of its function [8]. 7 In this approach, earth retaining structure is checked as to whether both these limit states are satisfied, after application of partial factors to actions and soil properties. Characteristic values of soil strength parameters are divided by an appropriate partial safety factor to obtain design values. The design values of actions, on the other hand, are obtained by multiplying characteristic values by an appropriate partial safety factor. 2.2.2 Conventional Design Method As its name implies conventional method is a much older approach which is based on code CP2- Earth retaining structures, 1951. This method deals with “Factor of safety” in terms of moments, sliding force and bearing capacity. The Factor of safety considered to take into account all the uncertainties in the load evaluation and material properties. With the introduction of BS 8002,1994 [6] CP2 had to be officially withdrawn; however, still its design approach is very popular among design engineers. It is a rather simple approach which has been successfully adopted for design of earth retaining structures [9]. Several popular geotechnical and soil mechanics text books published before the year 2000 and even after, have illustrated their design examples based on conventional method. This is one of the main reasons for its popularity. The Craig’s Soil Mechanics book which has been published seven times, in 1974, 1978, 1983,1987,1992,1997 and 2004, contains design examples of retaining wall design based only on conventional approach up to 1997edition [10]. In its 2004 version [8], design examples are explained based on both approaches. 2.2.2.1. Stability Analysis in Conventional Method. In conventional method, stability analysis of a retaining walls deals with following failure modes: 8 i) Sliding; ii) Overturning; and iii) Bearing Capacity. Since code CP2 was published more than 60 years ago, it is now a rare document in design offices. Due to this reason, the procedures given in text books for the stability analysis, are widely used. The procedures of the stability analysis given in the following two text books were considered in this study. a) Craig,R.F., Craig’s Soil Mechanics, 6th Edition[10] b) Cheng Liu, Evett J.B, Soil and Foundations, 7th Edition[11] 2.2.2.1.1 Stability Analysis, Craig, R.F [10] Checking for sliding, overturning and bearing capacity are done using two basic equations. Figure 2.1 shows Loads and reactions acting on a retaining wall. Figure 2.1 : Loads and base reactions of retaining walls Source : Craig,R.F., [10] 9 Factor of safety (Fs) against sliding, ignoring the passive resistance, is given by, 𝐹𝑠 = 𝑅𝑣 𝑡𝑎𝑛𝛿 𝑅ℎ Where, 𝑅𝑣, 𝑅ℎ : Vertical and horizontal components of the resultant force R 𝛿 : angle of friction between the base and the underlying soil Usually, for factor of safety, a value of at least 1.5 is specified. Both overturning and bearing capacity checks are done with single equation (2.2) which can be derived from the flexural formula. The Flexural formula can be applied to the base of the retaining wall to find out the maximum and minimum base pressure as follows, 𝑝 = 𝑅𝑣 𝐴 ± 𝑀𝑦 𝐼 Where, Rv : Vertical components of the resultant force R p : Minimum or maximum base pressure A : Area of the base. (Since 1m length of wall is under consideration, this equals to base width of B) M : Moment about center line of the base 𝐼 : Second moment area about the base center line y : the distance to the edge of the base from base center line. The derivation can be made as follows, 𝑝 = 𝑅𝑣 𝐵 ± 𝑅𝑣 × 𝑒 × 𝐵/2 (1 × 𝐵3 12 ) ( 2.1 ) 10 = 𝑅𝑣 𝐵 ± 𝑅𝑣 × 𝑒 × 6 𝐵2 𝑝 = 𝑅𝑣 𝐵 (1 ± 6𝑒 𝐵 ) Where, 𝑝 : Minimum or maximum base pressure B : Base width 𝑒 : Eccentricity of base resultant If the maximum base pressure is less than the allowable bearing capacity of the underlying soil, check on the bearing capacity is considered to be satisfactory. If the minimum base pressure is greater than zero, overturning criterion is considered satisfactory. When base pressure is positive throughout the base, the whole base width is in contact with the underlying soil. Therefore, overturning is considered satisfactory without evaluating the factor of safety against overturning. 2.2.2.1.2 Stability Analysis by Liu,C and Evett, J.B[11] The three factors of safety with regard to stability analysis are given as follows: F.S (sliding) = F.S(overturning) = F.S (bearing capacity) = ( 2.2 ) Sliding Resistance Force Sliding Force Ultimate bearing capacity of soil Maximum base presussure Total righting moment about toe Total overturning moment about toe ( 2.4 ) ( 2.3 ) ( 2.5 ) 11 The minimum factors of safety for sufficient stability specified in this book are as follows, F.S (Sliding) = 1.5 (If the passive earth pressure of the soil at the toe in front of wall is neglected); F.S (Overturning) = 1.5 (for granular backfill soil); or = 2.0 (for cohesive backfill soil); and F.S (Bearing capacity) = 3.0. The ratio between ultimate bearing capacity and allowable bearing capacity varies from 2.5 to 3. Hence, when equation 2.5 is re-written for allowable bearing capacity instead of ultimate bearing capacity, the factor of safety against bearing capacity of 1 could be considered as satisfactory. The maximum base pressure is found from the flexural formula. Before applying this formula, it was checked to make sure that vertical reaction of the base is within the one- third of the base width. This is to ensure that base pressure is positive throughout the base width. The one- third concept can be explained using the equation 2.2 In order to have a positive contact throughout, 𝑝𝑚𝑖𝑛 = 𝑅𝑣 𝐵 (1 − 6𝑒 𝐵 ) > 0 Hence, (1 − 6𝑒 𝐵 ) > 0 𝑒 > 𝐵 6 12 2.3 Adopting Tie- back Effect in Gravity Retaining Wall Design Some applications where Tie- back effect has been utilized effectively are discussed in this section. The first application has been used for retaining wall design of a printing complex at Mawaramandiya [12]. The second application is an own experience of the author in which the Tie- back effect is adopted effectively. In both cases, proper lateral load transferring from wall to Tie back has been assumed by design engineers through their judgment. If there were proper design information, even more optimized designs would have been adopted. 2.3.1 Innovative Earth Retaining System Adopted at the Proposed Printing Complex at Mawaramandiya [12] The proposed land for this project was located in a sloping terrain, thus earth retaining both along the site boundary and at some intermediate locations had been adopted. The heights to be retained have varied from 1m to 7m. Concrete anchor blocks tied back by RC tie beams have been used to enhance the lateral stability of the gravity retaining walls. Figure 2.2 illustrates the details of modified RRM retaining wall system adopted and Figure 2.3 shows the Construction of Tie back arrangement at the site. 13 Figure 2.2 : Modified RRM retaining wall system adopted. Source : Fernando S et al. [12] Figure 2.3 : Construction of Tie back arrangement at the site Source : Fernando S et al. [12] 14 According to the cost comparison analysis, it was reported that about 15% of overall cost could be saved by adopting this system compared to conventional gravity retaining wall. 2.3.2 Wall Foundations of Proposed Block No.10- Mahinda Rajapaksha Vidyalaya, Homagama The proposed building was a four storied class room block built on a land with a steep slope. The height of the wall foundations at the rear side of the building had to be increased up to 3m and were designed as retaining walls to retain the soil to support ground slab as well. Figure 2.4 illustrates the wall foundations at the rear side of the class room block. Figure 2.4 : Wall Foundations at the Rear Side of the Class Room block Set of RC tie beams were cast on the retaining walls and wall foundations in order to reduce slenderness of columns. These tie beams have been advantageous in terms of improving the lateral stability of the retaining walls as well. 15 2.4 British Standards relevant to Random Rubble Masonry Design During the past fifty years, British Standard has been consistently limiting its design recommendation on natural stone and stone masonry. Prior to 4-5 decades, some of codes of practice entirely dealt with natural stone masonry. However when these were replaced with modern British and European standard, very limited guidance for natural stone masonry design was provided. Table 2.1 shows the different Code of Practices existed and currently operating in the field of stone masonry. Table 2.1: Summary of guidance on British and British European Standards relevant to Natural Stone Standard Title Status CP 121.201: 1951 Code of practice for masonry walls ashlared with natural or cast stone Withdrawn CP 121.202 :1951 Code of Practice for Masonry Rubble Walls Withdrawn BS 5390:1976 Code of practice for Stone masonry Withdrawn BS 5628-1:2005 Code of practice for the use of masonry. Part 1: Structural use of unreinforced masonry Withdrawn BS 5628-3:2005 Code of practice for the use of masonry. Part 3: Materials and components, design and workmanship Withdrawn BS EN 1996-2:2006 Eurocode 6. Design of masonry structures. Part 2: Design considerations, selection of materials and execution of masonry Current BS EN 771-6:2011 Specification for masonry units; Natural stone masonry units Current Source : 1) Urquhart[13] 2) http://shop.bsigroup.com 16 2.5 Standard Construction Practices of Random Rubble Masonry in Sri Lanka Currently, Random Rubble Masonry is the most widely used stone masonry construction method in Sri Lanka. The Cabook stone masonry and Coursed Rubble masonry (rubble masonry with approximately rectangular bond patterns) have some applications, but extremely small compared to RRM. However, limited references are available for construction specifications of Random Rubble Masonry specifically in Sri Lankan context. Though not comprehensively addressed, ICTAD specification for Building Works [14] is considered to be the main reference available for rubble masonry construction in Sri Lanka. Few research papers [5, 15] on Random Rubble Masonry also have contributed towards covering the deficiency. Few design guidelines specified by above references regarding selection, preparing and laying of stones are summarized below. 2.5.1 Type of Stones For Random Rubble Masonry, Granite, Charnockites and Gneiss are mainly used. They shall be hard, sound, free from decay, weathering and defects like cavities, cracks, flaws, sand holes, veins patched of soft or loose materials. Stones shall be angular as far as possible and stones with round surfaces shall not be used to avoid single point contact. ICTAD specifications[14] does not specify material properties of stones, but minimum compressive strength of 10 N/mm2 and water absorption less than 10% [5] are generally preferred. 2.5.2 Sizes of Stones Sizes of stones are specified basically considering practical aspects and the stability of the masonry construction. Length of stones is limited in longitudinal direction of the wall to allow for differential settlements (Refer Figure 2.5). Following criteria 17 have been established by ICTAD specifications [14] in terms of selecting suitable sizes of stones: i. Stone shall be small enough to lift manually; so limiting weight can be considered as 25kg; ii. Length of stone < 3 x Height of stone ; iii. Breadth of stone < 0.75x wall thickness (except for Through stones); iv. Breadth of stone > 150mm; and v. Height of stone < 300mm. In case of Hearting stones, i. Stone shall not pass through a circular ring of 150mm diameter; and ii. Length or Breadth of stone > 100mm. Figure 2.5 illustrates typical bond patterns and Figure 2.6 shows the types of stones used in RRM. Figure 2.5: Typical Bond Patterns and Specifications for Bushing, Amount of Chips and Through Stone. Source: ICTAD specification for Building works [14] 18 Figure 2.6: Types of Stones used in RRM Source: Chandrakeerthy [15] 2.5.3 Dressing of Stones Stones shall be hammer dressed to enable them to come into close proximity with the neighboring stone. The Bushing ( irregularities of the face of the wall) in the face shall not project more than 40mm on an exposed face and 10mm on a face which is to be plastered. 2.5.4 Mortar ICTAD specifications [14] doesn’t carry special recommendation for Random Rubble Masonry and it specifies general mortar designations (1:5, 1:6 and 1:8 cement: sand proportions) for RRM as well. However general practice is to adopt 1:5 mortar designation as most of local applications of RRM are in contact with water. 19 2.5.5 Mortar joints Different joint finishes can be created depending upon the desired aesthetic requirement. Most commonly used joints in RRM are illustrated in Figure 2.7. Mortar Joints in RRM shall be within 6mm to 20mm. When joints are more than 20mm, they should be well packed with chip stones to limit the joint thickness within the limits. When plastering or pointing is not required to be done, the joints shall be struck flush and finished at the time of laying. Otherwise, joints shall be racked to a minimum depth of 20mm by a racking tool during the progress of work, when the mortar is still green. Figure 2.7: Types of Joints used in RRM Source: Chandrakeerthy [15] 20 2.5.6 Laying All stones shall be clean, free of dust and shall be wetted prior to use. Every stone shall be carefully fitted to the adjacent stones, so as to form neat and close joints. A sufficient number of Through stones (see Figure 2.5) shall be adapted to bond adjacent stones together. At least one through stone shall be built into the wall at the intervals of 1.8m horizontally and 0.6m vertically. Such stones shall be at least 150mm square at the face and run through the thickness of the walls up to 600mm. In case of walls exceeding 600mm in thickness, more than one stone may be used to run through the full thickness of the wall with overlaps of not less than 150mm. Where through stones of suitable lengths are not available, concrete block of grade 15 shall be used. 2.5.7 Curing Masonry work shall be kept constantly moist on all faces for a minimum period of 7 days. 2.6 Previous Experimental Investigations on Material properties of Random Rubble Masonry 2.6.1 Compressive Strength Chandrakeerthy[5,15] has investigated the compressive strength of RRM for 1:5 and 1:8 mortar designations. For both cases 300mm thick 600mmx600mm wall panels built with Chanockites stones, have been tested. All test panels have been built according to general guidelines (sizing, dressing, laying, mortar thicknesses, curing etc.) specified in ICTAD specifications [14]. Testing of panels (age of testing, load application etc.) have been carried out in accordance with BS 5628-1 [16]. Test results reported in these research publications are tabulated in Table 2.2 and Table 2.3. 21 Table 2.2: Characteristic Compressive Strength of RRM for Mortar Designation of 1:5 Mortar Designation Mortar Mix (cement:sand) Compressive Strength of Stones (N/mm2) (iii) 1:5 20 30 40 50 60 80 100 1.07 1.60 1.84 2.08 2.31 2.31 2.31 Source: Chandrakeerty [5] Table 2.3: Comparison of Characteristic Compressive Strength of RRM and Brickwork for Mortar Designations of 1:5 and 1:8 Mortar Designation (Cement: Sand) Characteristic Compressive Strength (N/mm2) Brick Work Random Rubble Masonry Flush Jointed Brick Compressive Strength : 2.0N/mm2 Brick Compressive Strength: 2.8N/mm2 Compressive Strength of Stone: 40N/mm2 Compressive Strength of Stone: 60N/mm2 and above iii (1:5) 1.00 1.40 1.84 2.31 iv (1:8) 0.88 1.23 0.68 0.86 Source: Chandrakeerty [15] 2.6.2 Shear Strength An experimental study was carried out by Milosevic et al. [17] to investigate the shear strength of rubble masonry. In this study, rubble triplets having 400mm X 600mm X 400mm dimensions were built with three stone layers. These layers were intentionally formed to allow for the middle stone layer to slide between the top and bottom layers, when it was subjected to a shear load as shown in Figure 2.8, while the top and bottom layers were latterly restrained. 22 Figure 2.8: Triplet setup by Milosevic et al. [17] In this test setup, failure surfaces were imposed to be parallel with the stone layers. Hence, this test doesn’t represent the actual shear failure of Random Rubble Masonry of which well-defined failure surfaces cannot be expected. Figure 2.9 shows the relationship between normal stress and shear stress of samples. Following abbreviations have been used in this figure: HM : Hydraulic Mortar; and AM : Air lime Mortar. A- Pre compression Load B- Concrete Slab C- Shear Load D- Lateral restrains to top & bottom layers 23 Figure 2.9 : Relationship between Normal stress and Shear stress of samples Source : Milosevic et al. [17] According to their analysis, Initial shear strength (shear strength when normal stress equal to zero) of masonry samples built with hydraulic mortar (HM) can be considered as 0.2 Mpa. However, it was not reported any information on mortar designation, strength of stones or type of stones which have been used for the investigation. 2.7 Experimental Investigation on Masonry- Concrete Interface Shear behavior at the Masonry – Concrete interface relevant to Sri Lankan brick masonry has been investigated by Premadasa et al. [4]. In this investigation, tests were carried out using a series of Brick- Concrete block couplets bonded with mortar. The brick-concrete couplets were tested in accordance with ASTM C 952-02 [18]. 24 For preparation of testing specimens, a wire cut brick of standard size (215mm x 105mm x 65mm) and a concrete block of the same size cast with Grade 25 concrete have been used. Testing has been carried out for 1:5, 1:6, 1:8 mortar designations and 10mm, 15mm mortar thickness. Test specimens have been cured for 28 days prior to testing. During testing, concrete portion of the specimen has been fixed and shearing load was applied to the Brick through a hydraulic jack. This is illustrated in Figure 2.10. Figure 2.10 : (a) - Test Setup ; (b) – Shear deformation while applying the load Source : Premadasa et al. [4] According to the results, average shear strength at the interface for 10mm mortar joint with 1:5 mortar designations was 0.2 N/mm2. For other mortar designations and for increased mortar thickness lesser strength values have been observed. Figure 2.10 (a) Figure 2.10 (b) 25 Chapter 03 DIFFERENT APPROACHES USED FOR DESIGN OF RANDOM RUBBLE MASONRY 3.1 Introduction In order to study the effects of Tie back and flexural strength of masonry on the design of RRM gravity retaining wall, a 3m high RRM retaining wall was designed considering following 3 cases: 1. Design of Conventional Retaining wall using Conventional Method; 2. Design of Retaining wall assuming RRM will not fail due to flexure; and 3. Design of Retaining wall with the Tie back effect. Following Design Information were assumed;  Density of the backfill : 17 kN/m3  Density of the RRM : 22 kN/m3  Surcharge pressure : 5 kN/m2  Characteristic values of the shear strength parameters for the backfill,  C = 0 , ∅ = 30°  Angle of friction between the base and the foundation soil , δ = 30°  Allowable bearing capacity of underneath soil is 250 kN/m2 and water table is below the base of Retaining wall. 26 3.2 Case 1 - Design of RRM Retaining Wall Using Conventional Method The loads act on the retaining wall for Case 1 is shown in the Figure 3.1. Figure 3.1: Loads acting on Retaining wall for Case 1 In this case, a value of 2.1m was assumed for base width. The weight of soil above the retaining wall contributes to the stability of the wall. In order to simplify the analysis, the retaining wall can be treated as a rectangular one with a base width of B. The average density of the wall (γwall ) is considered as 20kN/m3 Applying Rankin theory for evaluating lateral soil pressure, the active pressure coefficient, 𝐾𝑎 = 1 − 𝑠𝑖𝑛∅ 1 + 𝑠𝑖𝑛∅ 𝐾𝑎 = 1 − 𝑠𝑖𝑛30° 1 + 𝑠𝑖𝑛30° 𝐾𝑎 = 0.33 27 Lateral soil pressure at the bottom, 𝑃2 = 𝐾𝑎 × 𝐻 × 𝛾𝑠𝑜𝑖𝑙 = 0.33 × 3.45 × 17 = 19.3 𝑘𝑁/𝑚2 Lateral pressure due to surcharge, 𝑃1 = 𝐾𝑎 × 5 = 0.33 × 5 = 1.7 𝑘𝑁/𝑚2 The resultant forces due to lateral pressures can be calculated as, 𝐹1 = 𝑃1 × 3.45 = 5.9 𝑘𝑁/𝑚 𝐹2 = 𝑃2 × 3.45/2 = 33.3 𝑘𝑁/𝑚 The weight of simplified retaining wall, 𝑊 = 𝐵 × 𝐻 × 𝛾𝑤𝑎𝑙𝑙 = 2.1 × 3.45 × 20 = 144.9 𝑘𝑁/𝑚 Considering vertical and horizontal equilibrium, 𝑊 = 𝑅𝑣 = 144.9 𝑘𝑁/𝑚 ( 3.1) 28 𝑅ℎ = 𝐹1 + 𝐹2 = 33.3 + 5.9 = 39.2 𝑘𝑁/𝑚 By taking moments about the toe of the wall (O), 𝑅𝑣 × 𝑥 = 𝑊 × 𝐵 2 − 𝐹1 × 1.725 − 𝐹2𝑥1.15 Substituting the results of eqation 3.1 and values of F, 144.9 × 𝑥 = 144.9 × 2.1 2 − 5.9 × 1.725 − 33.3𝑥1.15 𝑥 = 0.71𝑚 Eccentricity of Rv can be wrtten as, 𝑒 = 𝐵 2 − 𝑥 = 2.1 2 − 0.71 = 0.34 m Stability analysis was done based on Craig, R.F [10]. Check for the minimum base presure 𝑝𝑚𝑖𝑛 = 𝑅𝑣 𝐵 (1 − 6𝑒 𝐵 ) = 144.9 2.1 (1 − 6 × 0.34 2.1 ) = 2.0 𝑘𝑁/𝑚2 > 0, hence check for minimum pressure is satisfactory. 29 i) Check for the Maximum Base Presure 𝑝𝑚𝑎𝑥 = 𝑅𝑣 𝐵 (1 + 6𝑒 𝐵 ) = 144.9 2.1 (1 + 6 × 0.34 2.1 ) = 136.0 𝑘𝑁/𝑚2 < 250𝑘𝑁/𝑚2, hence check for maximum pressure is satisfactory. ii) Check for Sliding Factor of safety against sliding, 𝐹𝑠 = 𝑅𝑣 𝑡𝑎𝑛𝛿 𝑅ℎ = 144.9 𝑡𝑎𝑛30° 39.2 = 83.6 39.2 = 2.1 > 1.5 Hence check for sliding is satisfactory. These three checks covered the stability analysis as described in Craig R.F[10] Check on minimum bearing pressure ensures the overturning criterion of the retaining wall. This is an indirect approach to check the overturning stability and the factor of safety against overturning has to be calculated in order to find out the actual status of the overturning stability. By using the method specified by Cheng Liu [11], Factor of Safety against Overturning can be computed as follows; 30 𝐹. 𝑆 (𝑂𝑣𝑒𝑟𝑡𝑢𝑟𝑛𝑖𝑛𝑔) = Total righting moment about toe Total overturning moment about toe = 𝑊 × 𝐵/2 𝐹1 × 1.725 + 𝐹2 × 1.15 = 144.9 × 1.05 33.7 × 1.725 + 5.9 × 1.15 𝐹. 𝑆 (𝑂𝑣𝑒𝑟𝑡𝑢𝑟𝑛𝑖𝑛𝑔) = 3.1 > 1.5 This shows that the factor of safety is well beyond the minimum requirement and base width could have been further reduced while keeping the required minimum Factor of Safety of 1.5. However, this will lead to a negative bearing pressure, which means tension between base and soil. In reality, soil cannot furnish any tensile resistance. Hence equations derived based on flexural formula are no longer valid to compute the soil pressure at this stage. Further, when bearing pressure becomes negative, flexural stresses develop within the retaining wall material itself. It can be assumed that both take place at the same time since same equation is used to compute the bearing pressure based on the flexural formula which is applicable to stress distribution of retaining wall material at the base level as well. In case of masonry retaining walls, development of tension or flexure is generally avoided. These are the reasons for keeping the minimum bearing pressure greater than zero even with a higher factor of safety against overturning. 31 Though it is assumed that masonry does not possess any flexural strength, indeed, it has some flexural strength. In a retaining wall, if it is allowed to develop small flexural stress which is less than the flexural strength, while keeping the required factor of Safety against overturning, retaining wall section can be optimized. In this case, flexural formula cannot be used to compute the soil pressure distribution. This is illustrated in the next design example. 3.3 Case 2 - Retaining Wall Assuming RRM will not Fail Due to Flexure Figure 3.2: Loads acting on Retaining wall for Case 2 In this case try for B=1.5m Lateral soil pressures and 𝑅ℎ are the same as in the previous case. Hence, 𝑃1 = 1.7 𝑘𝑁/𝑚2 𝑃2 = 19.3 𝑘𝑁/𝑚2 𝐹1 = 5.9 𝑘𝑁/𝑚 𝐹2 = 33.3𝑘𝑁/𝑚 𝑅ℎ = 39.2 𝑘𝑁/𝑚 32 The weight of retaining wall, 𝑊 = 𝐵 × 𝐻 × 𝛾𝑤𝑎𝑙𝑙 = 1.5 × 3.45 × 20 = 103.5𝑘𝑁/𝑚 Considering vertical and horizontal equilibrium, 𝑊 = 𝑅𝑣 = 103.5 𝑘𝑁/𝑚 By taking moment about toe of the wall (O), 𝑅𝑣 × 𝑥 = 𝑊 × 𝐵 2 − 𝐹1 × 1.725 − 𝐹2𝑥1.15 Subtituding results of Eqation 3.2 and values of F1 anf F2 103.5 × 𝑥 = 103.5 × 1.5 2 − 5.9 × 1.725 − 33.3 × 1.15 𝑥 = 0.28𝑚 Eccentricity of 𝑅𝑣 can be wrtten as, 𝑒 = 𝐵 2 − 𝑥 = 1.5 2 − 0.28 = 0.47𝑚 𝐵 6 = 0.25𝑚 < 𝑒 , Henec, base reaction doesn’t lie within one-third of the base width. ( 3.2) 33 i) Check for Maximum Base Presure Base reaction doesn’t lie within the one- third of the base width; hence base pressure will not be positive throughout the base. Figure 3.3: Assumed Base Pressure Variation for Case 2 The flexural formula is not applicable in this situation. Instead, soil pressure can be calculated using basic equations of statics. Since 𝑅𝑣 represents the resultant of base pressure, 𝑄 = 𝑞 × 𝑑 2 = 𝑅𝑣 = 103.5 𝑘𝑁/𝑚 𝑥 = 𝑑 3 = 0.28 𝑑 = 0.84𝑚, subsituding d in the equation 3.3, 𝑞 × 0.84 2 = 103.5 𝑞 = 246 𝑘𝑁/𝑚2 < 250 𝑘𝑁/𝑚2 , hence check on maximum base pressure is satisfactory. ( 3.3) 34 ii) Check for Sliding Factor of safety against sliding, 𝐹𝑠 = 𝑅𝑣 𝑡𝑎𝑛𝛿 𝑅ℎ = 103.5 𝑡𝑎𝑛30° 38.9 = 59.7 39.2 = 1.52 > 1.5 Hence check for sliding is satisfactory. iii) Check for Overturning 𝐹. 𝑆 (𝑂𝑣𝑒𝑟𝑡𝑢𝑟𝑛𝑖𝑛𝑔) = Total righting moment about toe Total overturning moment about toe = 𝑊 × 𝐵/2 𝐹1 × 1.725 + 𝐹2 × 1.15 = 103.5 × 0.75 33.3 × 1.15 + 5.9 × 1.725 𝐹. 𝑆 (𝑂𝑣𝑒𝑟𝑡𝑢𝑟𝑛𝑖𝑛𝑔) = 1.60 > 1.5 Hence check for overturning is satisfactory. Since all three checks are satisfactory, the stability of retaining wall can be considered as satisfactory. However, in this case, flexural stresses develop at the heel of the wall and it was assumed that flexural strength of masonry is capable of resisting it. 35 The magnitude of flexural stress developed can be estimated by using the flexural formula. The same equation used to compute the minimum base pressure in Chapter 2 (eq. 2.2), can be used to evaluate the magnitude of the flexural stress developed in the base. If the Flexural Stress at the heel is t, 𝑡 = 𝑅𝑣 𝐵 (1 − 6𝑒 𝐵 ) = 103.5 1.5 (1 − 6 × 0.47 1.5 ) = −60.7 𝑘𝑁/𝑚 2 = −0.061 𝑁/𝑚𝑚2 If RRM can resist 0.061 𝑁/𝑚𝑚2 flexural stress, retaining wall design is satisfactory with the proposed base width of 1.5m. 3.4 Case 3 – Design of Retaining Wall with the Tie-Back Effect In this case, the effect of a Tie back on the design of a retaining wall was investigated. The Tie back effect is expected to get through the 225mm x 225mm Reinforced Concrete (RC) beam cast on the retaining wall. This RC tie beam has to be laterally restrained at certain intervals through tie beams cast perpendicular to the retaining walls or through cross walls. 36 Figure 3.4: Loads acting on Retaining wall for Case 3 Try for B=1.5m In this case, lateral movement of the retained soil is restrained than previous two cases due to the tie back at the top of the wall. In order to mobilize the active pressure on the wall, the lateral deformation of the soil mass has to reach a certain level. In general, for cantilevered retaining walls, it is assumed that this minimum deformation is reached [8]. If the lateral deformation of soil is zero, the corresponding lateral pressure is called at –rest earth pressure and it is greater than the active earth pressure. In this case, lateral deformation is not zero since lateral deformation is restrained by a RC tie beam which has no significant lateral stiffness compared to the Retaining walls. Hence, assuming a lateral pressure in between at- rest and the active pressure is more appropriate. 37 At rest soil pressure coefficient, 𝐾0 = 1 − 𝑠𝑖𝑛∅ = 1 − 𝑠𝑖𝑛30° = 0.5 Active pressure coefficient, 𝐾𝑎 = 0.33 Pressure coefficient K= 0.4 is assumed which is in between 𝐾𝑜 and 𝐾𝑎. Lateral soil pressure at the bottom, 𝑃2 = 𝐾 × 𝐻 × 𝛾𝑠𝑜𝑖𝑙 = 0.4 × 3.45 × 17 = 23.5 𝑘𝑁/𝑚2 Lateral pressure due to surcharge, 𝑃1 = 𝐾 × 5 = 0.4 × 5 = 2.0 𝑘𝑁/𝑚2 The resultant forces due to lateral pressure can be calculated as, 𝐹1 = 𝑃1 × 3.45 = 6.9 𝑘𝑁/𝑚 𝐹2 = 𝑃2 × 3.45/2 = 40.5 𝑘𝑁/𝑚 𝑅𝑣 and 𝑊 are the same as the previous case (neglecting the weight of the tie beam). Hence, 𝑊 = 𝑅𝑣 = 103.5 𝑘𝑁/𝑚 38 Considering horizontal equilibrium, 𝑅ℎ = 𝐹1 + 𝐹2 − 𝑇 𝑅ℎ = 47.4 − 𝑇 By taking moment about toe of the wall (O), 𝑅𝑣 × 𝑥 = 𝑊 × 𝐵 2 − 𝐹1 × 1.725 − 𝐹2𝑥1.15 + 𝑇 × 𝐻 Substituting values for W, B, and F, 103.5 × 𝑥 = 103.5 × 1.5 2 − 6.9𝑥1.725 − 40.5 × 1.15 + 𝑇 × 𝐻 103.5 × 𝑥 = 19.1 + 𝑇 × 3.45 There are three unknowns and only two equations were available. Therefore, for diffent values of T, these two equations were resolved until get acceptable values for other parameters. As a first trial T= 10kN was assumed. Then 𝑥 = 0.52𝑚 and 𝑅ℎ = 37.4 𝑘𝑁/𝑚 Eccentricity of Rv can be wrtten as, 𝑒 = 𝐵 2 − 𝑥 = 1.5 2 − 0.52 = 0.23𝑚 (< 𝐵 6 = 0.25𝑚) , hence, base pressure will be positive throughout. Stability checks can be done as follows. ( 3.4) ( 3.5) 39 i) Check for Minimum Base Pressure 𝑝𝑚𝑖𝑛 = 𝑅𝑣 𝐵 (1 − 6𝑒 𝐵 ) = 103.5 1.5 (1 − 6 × 0.23 1.5 ) = 5.5 𝑘𝑁/𝑚2 > 0, hence check for minimum pressure is satisfactory. ii) Check for Maximum Base Pressure 𝑝𝑚𝑎𝑥 = 𝑅𝑣 𝐵 (1 + 6𝑒 𝐵 ) = 103.5 1.5 (1 + 6 × 0.23 1.5 ) = 132.5 𝑘𝑁/𝑚2 < 250𝑘𝑁/𝑚2, hence check for maximum pressure is satisfactory. iii) Check for Sliding Factor of safety against sliding, 𝐹𝑠 = 𝑅𝑣 𝑡𝑎𝑛𝛿 𝑅ℎ = 103.5 𝑡𝑎𝑛30° 37.4 = 59.7 37.4 = 1.6 > 1.5 40 iv) Check for Overturning 𝐹. 𝑆 (𝑂𝑣𝑒𝑟𝑡𝑢𝑟𝑛𝑖𝑛𝑔) = Total righting moment about toe Total overturning moment about toe = 𝑊 × 𝐵 2 + 𝑇 × 3.45 𝐹1 × 1.725 + 𝐹2 × 1.15 = 103.5 × 0.75 + 10 × 3.45 40.5 × 1.15 + 6.9 × 1.725 𝐹. 𝑆 (𝑂𝑣𝑒𝑟𝑡𝑢𝑟𝑛𝑖𝑛𝑔) = 1.92 > 1.5 Hence check for overturning is satisfactory. Hence check for sliding is satisfactory and T = 10 kN. In this case overall stability can be maintained without allowing flexural stresses to develop within the masonry. However, the capability of transferring tie back force to the RC tie beam through the masonry- Concrete interface and through the narrowest masonry section at the top of the retaining wall, as shown in the Figure 3.5, need to be investigated. Figure 3.5: Possible Shear Failure Planes of RRM for Case 3 41 The expected shear stress at the two shear planes are computed as follows, Shear stress at Concrete- Masonry interface, 10 𝑘𝑁/𝑚 225 × 1000𝑚𝑚2 = 0.044 𝑁/𝑚𝑚2 Shear stress at Masonry- Masonry interface (shear stress of RRM masonry), 10 𝑘𝑁/𝑚 350 × 1000𝑚𝑚2 = 0.028𝑁/𝑚𝑚2 3.5 Summary of the Results Obtained from Three Cases The results obtained from three cases are summarized in Table 3.1. Table 3.1: Summary of Results Obtained from Three Cases Parameter C a se 1 -C o n v en ti o n al R et ai n in g w al l C a se 2 - R et ai n in g w al l as su m in g R R M w il l n o t fa il d u e to fl ex u re C a se 3 - R et ai n in g w al l w it h a T ie b ac k Base Width /m 2.1 1.5 1.5 Max. Bearing Pressure / kNm-2 136.0 246 132.5 Min. Bearing Pressure / kNm-2 2.0 0 5.5 Flexural stress at the heel/ Nmm-2 - 0.061 - Shear stress at Concrete - Masonry interface/ Nmm-2 - - 0.044 Shear stress at the narrowest masonry section/ Nmm-2 Negligible Negligible 0.028 Factor of Safety - Overturning 3.1 1.60 1.92 Factor of Safety - sliding 2.1 1.52 1.6 Percentage of base width reduction / % - 28 28 42 This study shows that with the effect of flexural strength of RRM and tie back, the conventional retaining wall design can be optimized to a considerable level. According to analysis, approximately, one-third of the base width can be reduced if those effects were taken into consideration. However, the flexural effect (Case2) can be considered for ground conditions with higher bearing capacities only. In relation to conventional gravity wall, the tie- back effect further reduces the maximum bearing pressures and the difference between maximum to minimum soil pressures. The combination of both effects would be able to adopt for further optimized solutions. 43 Chapter 04 EXPERIMENTAL STUDY 4.1 General This chapter summarizes the details of the experimental investigation carried out in the research study. The main objectives of the experimental study are: a) To determine the flexural strength of RRM; b) To determine the shear strength of RRM; c) To determine the compressive strength of RRM; and d) To determine the shear strength at the concrete- masonry interface. To determine above strength parameters of RRM, tests were carried out in accordance with the following codes of practice.  Flexural Strength : BS EN 1052-2: 1999, Methods of test for masonry - Part 2: Determination of flexural strength [2].  Shear Strength : BS EN 1052-3: 2002 , Methods of test for masonry- Part 3: Determination of initial shear strength [3]  Compressive Strength : BS EN 1052-1: 1999, Methods of test for masonry- Part 1: Determination of compressive strength [1]. All these codes of practice deal with masonry constructed with regular size units which are laid in a specified bonding pattern. Standard test methods to determine the above strength properties of RRM are not specified in any of these codes of practice. Therefore, this investigation had to be based on these codes due to unavailability of any other standard specification. The investigation on shear strength at Concrete- RRM interface was based on the experimental approach adopted by Premadasa et al. [4] to determine the shear behaviour at the Brick – Concrete interface. 44 4.1.1 Preparation of Test Specimens  All RRM panels were made to 300mm thick which is the minimum practically possible thickness to be constructed.  9″x6″ size of rubble were used since this is the most commonly used size for constructing retaining walls  1:5 (cement: sand) mortar mix was used for all panels as it is the widely used mortar mix for RRM.  Thickness of Mortar joints were maintained within 6-20mm and flush joints were used.  Curing: All specimens were closely covered with polythene sheets and maintained them undisturbed until testing.  Specimens for Flexure and shear tests were pre-compressed by placing concrete test cubes on the top surface of the specimen, which is equivalent to 3.6 × 10−3 𝑁/𝑚𝑚2 pre- compression. Figure 4.1:(a) shows the preparation of specimens and the pre-compressed specimens are shown in Figure 4.1:(b) Figure 4.1: (a) – Preparing samples for Flexural Strength Test Figure 4.1: (b) – Pre-compressed Specimens 45 4.2 Experimental Set-up 4.2.1 Testing for Flexural Strength of RRM Two types of test specimens were constructed for the following two cases; 1) When the plane of bending is vertical - (see Figure 4.2(a)) 2) When the plane of bending is horizontal - (see Figure 4.2(b)) Three samples were tested under four point loading for each type of test. Loads were applied to samples through two inner bearings while two outer bearings were used to support them laterally. Figure 4.2 (a): The Plane of Bending is Vertical Figure 4.2 (b): The Plane of Bending is Horizontal Outer Bearings Outer Bearings Inner Bearings 46 A gradually increasing load was applied using a hydraulic jack (Manufacturer: EVERPAC) on inner bearing via Proving ring (Manufacturer : MARUTO). Proving ring with 100 kN (10 Tons) load capacity had been used for all 6 specimens. The test set-ups are shown in Figure 4.3 and Figure 4.4. Figure 4.3: The Test set up When the Plane of bending is horizontal Figure 4.4: The test set up for Specimens when the Plane of Bending is Vertical A – Hydraulic jack B- Proving ring C- Inner bearings D- Outer bearings B C A D B- Proving ring C- Inner bearings D- Outer bearings D C B 47 The dimensions of the samples to the nearest 5mm and maximum dial gauge reading for all cases were recorded. Appendix A includes all data relevant to results of the experiment. Only planes of failure occurred within the inner bearings were identified as satisfactory failure modes as this was the region when pure bending was expected. 4.2.2 Testing for Shear Strength of RRM BS EN 1052-3: 2002 [3] specifies triplet test for investigating shear strength of masonry units which are laid in a specific bond pattern. This test provides two straight failure planes which are parallel to the masonry units. Figure 4.5 indicates the set-up for triplet test. Figure 4.5: Set up for Triplet tests as in [3] Source: BS EN 1052-3: 2002 [3] 48 In the case of RRM, this type of straight failure surfaces cannot be expected. Further, the shear failure of a retaining wall provides single plane shear failure. Hence, testing set-up shown in the Figure 4.6 was adopted. In this set-up, supporting plates and loading plates were positioned so that only single failure plane is obtained. Figure 4.6: Set up adopted for Shear Strength Test The pre- compression was applied with a hydraulic jack through a proving ring of 100kN capacity. The shear load was applied with the hydraulic jack through the proving ring of 29 kN (3 Tons) capacity. Six samples of 600mm × 300mm ×450mm were tested. Three were tested for zero pre- compression and the rest were tested with 4.5 kN, 6.0 kN, 9.0 kN pre- compression forces. These forces were selected as they reflect the brick wall weights of typical heights, built on the retaining walls. The pre- compression of 6kN is equivalent to the compression due to the weight of 10′ high 9″ thick brick wall. However, during the application of shear loading, pre-compression load increased 49 due to the slight inclination of the specimen. The pre-compression load at the time of failure was recorded as the pre- compression load. Figure 4.7 shows the typical loading arrangement for shear strength test. Dimensions of the samples to the nearest 5mm and maximum dial gauge readings for all cases were recorded and included in Appendix B. Figure 4.7: Test set up for Shear Strength Test 4.2.3 Testing for Shear Strength at Concrete- RRM Interface. Six numbers of test specimens were prepared by casting a 225mm × 225mm Concrete beam on 300mm thick 450mm × 450mm RRM samples. When preparing specimens, concrete mix was directly poured on the top surface of the RRM samples. The concrete mix of 1: 1.5: 3 (19mm) was used. Figure 4.8 illustrates Set up for investigating shear strength at Concrete-RRM interface test A – Screw Jack B- Hydraulic Jack C- Loading Plate D- Supporting Plates D D C B A 50 Figure 4.8: Set up for investigating Shear Strength at Concrete-RRM Interface Test The shear load was applied with the hydraulic jack through the proving ring of 29kN (3 Tons) capacity. Dimensions of the RRM samples and concrete tie beam to the nearest 5mm and maximum dial gauge readings for all cases were recorded and included in Appendix C. Figure 4.9 illustrates the set up for Shear strength at RC- RRM interface test. Figure 4.9: The set up for Shear strength at RC- RRM interface test A – Hydraulic Jack B- Proving Ring C- Loading Beam D- Supporting Plates C B A D D 51 4.2.4 Testing for Compressive Strength of RRM Three numbers of 300 thick 600mm×600mm RRM samples were used for the investigation. The sizes of samples are in conformity with the BS EN 1052-1: 1998 and also are similar to those tested by Chandrakeerthy[5,15]. The specimens were loaded by means of a 200 Ton-Amsler compression testing machine. Steel and timber platens were provided to the top surface of the sample in order to have a uniform pressure distribution over the surface. The steady loading rate of 0.15N/(mm2.min) was maintained for all samples. The test set-up for Compressive Strength of RRM is shown in Figure 4.10. Figure 4.10: Set up for Compressive Strength Test Dimensions of the RRM samples to the nearest 5mm and maximum dial gauge readings for all cases were recorded. Refer Appendix D for the results. A – Compression Testing Machine B – Loading Platens B A 52 Chapter 05 ANALYSIS OF TEST RESULTS 5.1 Flexural Strength 5.1.1 Experimental Results a) When the Plane of Bending is horizontal The Dimensions of the test panels and test results when the plane of bending is horizontal are given in the Table 5.1. Table 5.1: Results of the test on Flexural Strength (When the Plane of Bending is horizontal) Specimen Dimensions/ (mm) Load /(N) Comments on Failure Pattern Average Width Average Height Average Thickness 1 1378 1000 300 19,078 Not satisfactory, Failure at an outer bearing 2 1400 1000 300 37,278 Satisfactory 3 1366 1000 300 39,874 Satisfactory The failure patterns of the test specimens are shown in Figures 5.1(a) – (c). 53 Figure 5.1 (a) – Specimen 01 Figure 5.1 (c) – Specimen 03 Figure 5.1: Failure patterns of Specimens (When the Plane of Bending is horizontal) Failure at out-bearing Figure 5.1 (b) – Specimen 02 Figure 5.1 (c) – Specimen 03 54 b) When the Plane of Bending is vertical The Dimensions of the test panels and test results when the plane of bending is vertical are given in the Table 5.2. Table 5.2: Results of Flexural Strength Test (When the Plane of Bending is Vertical) The failure patterns of the test specimens 4, 5 and 6 are shown in Figures 5.2(a) – (c). Figure 5.2 (a) – Specimen 4 Specimen Dimensions/ (mm) Load/ (N) Comment on Failure Pattern Average Width Average Height Average Thickness 4 560 1400 300 5182 Satisfactory 5 565 1390 300 5182 Satisfactory 6 560 1380 300 5182 Satisfactory 55 Figure 5.2 (b) – Specimen 5 Figure 5.2 (c) – Specimen 6 Figure 5.2: Failure Patterns of Specimens (When the Plane of Bending is Vertical) 56 5.1.2 Evaluation of Results The flexural strength of each specimen can be calculated by using following formula [2]. 𝑓𝑥𝑖 = 3𝐹𝑖,𝑚𝑎𝑥(𝑙1 − 𝑙2) 2𝑏𝑡𝑢2 𝑁/𝑚𝑚2 Where, 𝑓𝑥𝑖 - Flexural strength of an individual masonry specimen, (N/mm 2) 𝐹𝑖,𝑚𝑎𝑥 - Maximum load, (N ) 𝑙1 - Spacing of outer bearings, (mm) 𝑙2 - Spacing of inner bearings, (mm) 𝑏 - Width of the section in the plane of bending, (mm) 𝑡𝑢 - Width of the masonry unit, (mm) The Flexural Strength of Specimen 2, 𝑓𝑥𝑖 = 3 × 37278 × (1200 − 450) 2 × 1000 × 3002 𝑁/𝑚𝑚2 = 0.47 𝑁/𝑚𝑚2 Similarly, flexural strengths of all specimens were evaluated and tabulated in Table 5.3. The result of specimen 1 is discarded due to unsatisfactory failure mode. Table 5.3: Flexural Strength of RRM Specimens Plane of Bending Specimen No 𝑓𝑥𝑖 (N/mm2) Horizontal 1 - 2 0.47 3 0.50 Vertical 4 0.12 5 0.11 6 0.12 57 The characteristic flexural strength of masonry units can be calculated by using the following formula [2]. 𝑓𝑥𝑘 = 𝑓𝑚𝑒𝑎𝑛 1.5 Where, 𝑓𝑥𝑘 - Characteristic flexural strength of masonry,(N/mm 2) 𝑓𝑚𝑒𝑎𝑛 - Mean flexural strength of masonry specimens, (N/mm 2) This formula is specified for five specimens, however due to non-availability of results of five specimens for each Plane of Bending case and the deviation of the results is very small, the Mean Flexural Strength is calculated using the available test results. When the Plane of Bending is horizontal, the Characteristic Flexural Strength of RRM is: 𝑓𝑥𝑘 = (0.47 + 0.5)/2 1.5 = 𝟎. 𝟑𝟐 𝑵/𝒎𝒎𝟐 When the Plane of Bending is vertical, the characteristic flexural strength of RRM is: 𝑓𝑥𝑘 = (0.12 + 0.11 + 0.12)/3 1.5 = 𝟎. 𝟎𝟖 𝑵/𝒎𝒎𝟐 The characteristic flexural strength values obtained for each case are tabulated in Table 5.4. Table 5.4: Characteristic Flexural Strength of RRM Plane of Bending Characteristic Flexural strength (N/mm2) Horizontal 0.32 Vertical 0.08 58 5.1.3 Comparison of Test Results with Flexural Strength of Brick/ Block Masonry. Table 5.5 indicates the flexural strength values given in BS 5628-1:1992. Table 5.5: Flexural strength of Brick and Block Masonry as per BS 5628-1:1992 [16] The mortar mix of 1:5 (cement: sand) corresponds to mortar designation (iii) (highlighted area). The Table 5.4 shows that characteristic flexural strength result of RRM (when the Plane of Bending is horizontal) which is comparable with those of block masonry. The test result of RRM (when the Plane of Bending is vertical) is very much smaller when compared with the brick and block masonry. 59 5.2 Shear Strength 5.2.1 Experimental Results The Dimensions of the test panels and results for the Test on Shear Strength are given in the Table 5.6. Table 5.6: Results of the Test on Shear Strength The failure patterns of the test specimens are shown in Figures 5.3(a) – (f). Specimen Dimensions/ (mm) Load (N) Pre- Compression Load/ (N) Failure Pattern Average Width Average Height Average Thick. 1 455 600 300 8092 4834 Satisfactory 2 455 600 300 13841 9527 Satisfactory 3 450 600 300 4640 12133 unsatisfactory 4 450 600 305 5273 Satisfactory 5 450 605 310 8954 Satisfactory 6 450 600 310 3374 Satisfactory Figure 5.3(a) – Specimen 01 Figure 5.3(b) – Specimen 02 60 Figure 5.3: Failure Patterns of Specimens 1 – 6 Figure 5.3 (c) – Specimen 03 Figure 5.3 (d) – Specimen 04 Figure 5.3 (e) – Specimen 05 Figure 5.3 (f) – Specimen 06 61 5.2.2 Evaluation of Results The shear strength and the pre- compressive stress on a sample can be calculated by using following formulae [3]. 𝑓𝑣𝑜𝑖 = 𝐹𝑖,𝑚𝑎𝑥 𝐴𝑖 𝑁/𝑚𝑚2 𝑓𝑝𝑖 = 𝐹𝑝𝑖 𝐴𝑖 𝑁/𝑚𝑚2 Where, 𝑓𝑣𝑜𝑖 - Shear strength of an individual sample, (N/mm2) 𝑓𝑝𝑖 - Pre-compressive stress of an individual sample, (N/mm2) 𝐹𝑖,𝑚𝑎𝑥 - Maximum shear force, (N ) 𝐹𝑝𝑖 - Pre-compressive force, (N ) 𝐴𝑖 - Cross sectional area of a sample parallel to the shear load, (mm) The shear strength of specimen 1, 𝑓𝑣𝑜𝑖 = 8092 455 × 300 𝑁/𝑚𝑚2 𝑓𝑣𝑜𝑖 = 0.059 𝑁/𝑚𝑚 2 The pre-compressive stress on specimen 1, 𝑓𝑝𝑖 = 𝐹𝑝𝑖 𝐴𝑖 𝑁/𝑚𝑚2 = 4834 455 × 300 𝑁/𝑚𝑚2 = 0.035 𝑁/𝑚𝑚2 62 Similarly, shear strength results for all specimens are calculated and presented in Table 5.7 Table 5.7: Shear Strength results for different Pre-Compressive Stresses Specimen No Shear Strength (𝑓𝑣𝑜𝑖) Nmm-2 Pre-compressive Stress (𝑓𝑝𝑖) Nmm-2 1 0.059 0.035 2 0.101 0.070 3 0.034 0.090 4 0.038 0.000 5 0.064 0.000 6 0.024 0.000 Figure 5.4 shows graphically the variation of Shear Strength with Pre-Compressive Strength. Figure 5.4: Variation of Individual Shear Strength values with the Pre-Compressive Stresses 63 When plotting the variation of shear strength with pre-compressive stresses, the results correspond with specimen 3 (which is highlighted) was discarded since its failure mode was not satisfactory. As specified in BS EN 1052-3:2002 [3], shear strength parameters can be evaluated as follows, Referring to Figure 5.4, Initial shear strength, 𝑓𝑣𝑜 = 0.042 N/mm 2 (Shear strength when pre-compression is zero) Angle of Internal Friction, 𝛼 = tan−1 [ (0.1 − 0.042) 0.08 ] = tan−1[0.725] = 35.9° The Characteristic Initial Shear Strength, 𝑓𝑣𝑜𝑘 = 0.8𝑓𝑣𝑜 = 0.8 × 0.042 = 𝟎. 𝟎𝟑𝟑 𝑵/𝒎𝒎𝟐 The Characteristic Angle of Internal Friction, 𝛼𝑘 = tan −1[0.8𝑡𝑎𝑛𝛼] = tan−1[0.8 × 0.725] = 𝟑𝟎. 𝟏° 64 The Characteristic frictional coefficient, 𝜇𝑘 = tan 𝛼𝑘 = tan 30.1 = 𝟎. 𝟓𝟕 5.2.3 Comparison of Results with previous research findings. The investigation carried out by J.Milosevic et al. [17] reported the Characteristic Shear Strength parameters of RRM based on the triplet test. According to Milosevic’s et al. [17] findings, Characteristic initial shear strength of RRM = 0.16 N/mm2 Characteristic frictional coefficient ( tan 𝛼) = 0.98 Compared to above values, the results obtained for local RRM in this research study are extremely low. The samples used for Milosevic’s investigation were built in three stone layers and the middle layer was pushed through other two layers. Hence, this test doesn’t make a fully representation of the actual shear strength of Random Rubble Masonry of which well-defined failure surfaces cannot be expected. There can be differences of strength parameters due to dissimilar mortar designations and the difference of workmanship. However, the mortar designation which had been used for making samples was not mentioned in the research paper. 5.3 Shear Strength at Concrete- RRM interface. 5.3.1 Experimental Results Experimental results for all 6 specimens are summarized and shown the Table 5.8. Failure modes of all samples are illustrated subsequently in Figures 5.5(a)-(f). 65 Table 5.8: Results of Shear Strength Test Specimen Dimensions of Tie beam/ (mm) Dimensions of Masonry Panel/ (mm) Load /(N) Failure Surface Length Height Width Width Height Thick. 1 450 225 225 450 600 300 3201 Masonry- Masonry 2 450 225 225 450 600 300 5675 Masonry- Masonry 3 450 225 225 450 600 300 12117 Concrete- Masonry 4 430 225 225 430 600 300 6366 Concrete- Masonry & Masonry- Masonry 5 450 225 225 450 600 300 14128 Concrete- Masonry 6 450 225 225 450 600 300 5503 Masonry- Masonry Figure 5.5 (a) – Specimen 1 Figure 5.5 (b) – Specimen 2 66 Figure 5.5: Failure Pattern of Specimens Used for Shear Test, (a) - 1, (b) -2, (c) -3, (d) - 4, (e) -5, (f)-6 Figure 5.5 (c) – Specimen 3 Figure 5.5 (d) – Specimen 4 Figure 5.5 (e) – Specimen 5 Figure 5.5 (f) – Specimen 6 67 5.3.2 Evaluation of Results In some cases, failure has not been occurred at the Concrete- Masonry interface as expected. Instead, it has occurred at Masonry- Masonry interface. Only samples failed at Concrete- Masonry interface can be used for evaluating the shear strength at the concrete-masonry interface. Indeed, the other results with Masonry- Masonry interface failure was due to the shear failure of RRM. The results of the sample (sample 4) of which failure occurred across both failure surfaces were discarded. The shear strength at the Concrete- Masonry interface, 𝑓𝑣𝑜𝑡𝑖 = 𝐹𝑖,𝑚𝑎𝑥 𝐴𝑡𝑖 𝑁/𝑚𝑚2 Where, 𝑓𝑣𝑜𝑡𝑖 - Shear strength at the Con.- Masonry of an individual sample, (N/mm2) 𝐹𝑖,𝑚𝑎𝑥 - Maximum shear force, (N ) 𝐴𝑡𝑖 - Cross sectional area of the tie beam parallel to the shear load, (mm) The shear strength at Concrete- Masonry interface of specimen 3, 𝑓𝑣𝑜𝑡𝑖 = 12117 225 × 450 𝑁/𝑚𝑚2 = 0.120 𝑁/𝑚𝑚2 The shear strength of RRM can be evaluated by using the same equation as in 5.2.2. The shear strength of specimen 2, 𝑓𝑣𝑜𝑖 = 𝐹𝑖,𝑚𝑎𝑥 𝐴𝑖 𝑁/𝑚𝑚2 = 5675 450 × 300 𝑁/𝑚𝑚2 = 0.042 𝑁/𝑚𝑚2 68 Similarly, shear strength results for all specimens are calculated and presented in Table 5.9. Table 5.9: Results of test carried out for Shear Strength at Concrete- Masonry Interface Specimen Shear strength at Con.- Masonry (𝑓𝑣𝑜𝑡𝑖) /Nmm -2 Shear strength of RRM (𝑓𝑣𝑜𝑖) /Nmm -2 1 - 0.024 2 - 0.042 3 0.120 - 4 - - 5 0.140 - 6 - 0.041 The Average shear strength at Concrete- Masonry interface is 0.13 N/mm2 As evaluated in section 5.2.2, Characteristic shear strength at Concrete- Masonry interface, 𝑓𝑣𝑜𝑡𝑘 = 0.8 × 𝑓𝑣𝑜𝑡 𝑁/𝑚𝑚 2 = 0.8 × 0.13 𝑁/𝑚𝑚2 = 0.104 𝑁/𝑚𝑚2 5.3.3 Comparison of Results with previous research findings on Shear strength at Concrete- Masonry interface According to the findings of Premadasa et al.[4] average shear strength at the interface of 10mm mortar joint with 1:5 mortar designations was 0.2 N/mm2. For other mortar designations and for increased mortar thickness lesser strength values of shear strength have been observed. 69 In this study, the average shear strength at the interface of Concrete- RRM was 0.13N/mm2 which is slightly lower than the Brick- Concrete interface. 5.4 Compressive Strength. 5.4.1 Experimental Results The Dimensions of the test panels and results for the Test on Compressive Strength are given in the Table 5.10. Table 5.10: Results of Compressive Strength Test Specimen Dimensions/ (mm) Load/ (N) Average Width Average Height Average Thickness 1 595 600 300 147150 2 570 600 300 128511 3 600 600 300 209934 5.4.2 Evaluation of Results 5.4.2.1 Compressive Strengths of each Sample The Compressive strength of each sample can be calculated by using following equation [1]. 𝑓𝑖 = 𝐹𝑖,𝑚𝑎𝑥 𝐴𝑖 𝑁/𝑚𝑚2 Where, 𝑓𝑖 - Compressive strength of an individual sample, (N/mm2) 𝐹𝑖,𝑚𝑎𝑥 - Maximum load on a sample, (N) 70 𝐴𝑖 - Loaded cross sectional area of an individual sample, (mm) The compressive strength of specimen 1, 𝑓𝑖 = 147150 595 × 300 𝑁/𝑚𝑚2 = 0.82 𝑁/𝑚𝑚2 Similarly, compressive strength of other specimens were calculated and tabulated in Table 5.11. Table 5.11: Compressive Strength Results of Each Sample 5.4.2.2 Mean Compressive Strength Mean compressive strength, 𝑓 = (𝑓1 + 𝑓2 + 𝑓3) 3 𝑁/𝑚𝑚2 = (0.82 + 0.75 + 1.17) 3 𝑁/𝑚𝑚2 = 0.91 𝑁/𝑚𝑚2 Specimen Compressive strength (𝑓𝑖) /Nmm-2 1 0.82 2 0.75 3 1.17 71 5.4.2.3 Characteristic Compressive Strength Characteristic Compressive strength as specified in [1], 𝑓𝑘 = 𝑓 1.2 𝑜𝑟 𝑓𝑖,𝑚𝑖𝑛 = 0.76 𝑁/𝑚𝑚2or 0.75𝑁/𝑚𝑚2, whichever is the smaller. 𝑓𝑘 = 0.75 𝑁/𝑚𝑚 2 5.4.3 Comparison of Results with previous research findings on Compressive Strength of RRM Chandrakeerthy[5,15] has investigated on the compressive strength of RRM for Sri Lankan rubble stones using 300mm thick 600mm X 600mm panels. Test results reported in the study are given in Table 5.12. Table 5.12: Characteristic Compressive Strength of RRM for Mortar designation of 1:5 Mortar Designation Mortar Mix (cement: sand) Compressive Strength of Stones (N/mm2) 20 30 40 50 60 80 100 (iii) 1:5 1.07 1.60 1.84 2.08 2.31 2.31 2.31 Source: Chandrakeerty [5] It is reported that, tests have been carried out for various strength grades of stones. Characteristic compressive strength of RRM with stones of 20 N/mm2 compressive strength was slightly greater than the results obtained in this research study. Whichever is the smaller 72 5.5 Summery of Test Results obtained by the Experimental Study Table 5.13 gives the summery of Test results obtained by this research study. Table 5.13: Summery of Strength Parameters of RRM Strength Parameter Characteristic Strength / (N/mm2) Flexural Strength ( When plane of bending is horizontal) 0.32 Flexural Strength (When plane of bending is vertical) 0.08 Shear Strength 0.033 Shear Strength at Concrete- RRM interface. 0.104 Compressive Strength 0.75 73 Chapter 06 CONCLUSIONS AND RECOMMENDATIONS 6.1 Use of Experimental Results for the improvements of RRM Retaining wall Design In chapter 3, three different design approaches were considered. The results obtained for those three cases are summarized in Table 6.1. Table 6.1: Results obtained through different Design Approaches 74 6.1.1 Flexural Strength. According to the above summery, if RRM is able to withstand 0.061N/mm2 flexural stress, the base width of the conventional design can be reduced by 28%. However, the maximum bearing pressure is increased up to 246 kN/m2 The results indicated that, characteristic flexural strength of RRM (when the plane of bending is vertical) is 0.08 N/mm2. Hence there is a possibility for an optimization of RRM retaining wall design considering the effect of flexural strength of RRM, However it depends on the bearing capacity of the soil. 6.1.2 Effect of Tie back If the concrete- masonry interface can transfer 0.044 N/mm2 shear stress and RRM can withstand 0.028 N/mm2 shear stress, the base width selected for the conventional design could be reduced up to 28%. According to the experimental results, characteristic shear strengths at the Concrete- Masonry interface and of RRM are 0.104 N/mm2, 0.033 N/mm2 respectively. Hence reducing the base width up to 28% with aid of the Tie back effect is possible. 6.2 Adopting Results of the Study for Other Retaining Heights The extent of optimization achievable for some other retaining heights is illustrated in the Table 6.3. The same design information which are listed in chapter 3, are assumed for all retaining heights. 75 Table 6.2: Extent of Optimization for 1-3m Retaining Heights Soil Retaining Height Parameter C a se 1 -C o n v en ti o n al R et ai n in g w al l C a se 2 - R et ai n in g w al l as su m in g R R M w il l n o t fa il d u e to f le x u re C a se 3 - R et ai n in g w al l w it h a T ie b ac k 1 Base Width /m 1 0.75 0.6 Percentage of base width reduction / % - 25 40 2 Base Width /m 1.55 1.15 1 Percentage of base width reduction / % - 26 35 3 Base Width /m 2.1 1.5 1.5 Percentage of base width reduction / % - 28 28 From the results, it can be concluded that base width of the retaining wall can be optimized considerably considering the effect of tie-back and allowing flexural stress, for retaining heights bellow 3m. 6.3 Suggestions for Future Works 1) Carry out investigation for other mortar designations. In this investigation, all samples were made with 1:5 (cement: sand) mortar mix. ICTAD specification [14] specifies mortar mixes of 1:5, 1:6, 1:7 and 1:8 for RRM. 1:5 mortar mix was selected for this investigation, as it is the most commonly used mortar mix for outdoor and below ground level applications [5]. However there are some instances where 1:6 mortar mix is also adopted for this type of applications. Hence, it is suggested to carryout investigation by using 1:6 mortar mix as well. 76 2) Investigation of the effect of strength of Rubble to the strength parameters of RRM. It is a known fact that not only the strength of mortar, but also the strength of the masonry unit affects the strength parameters of Masonry. However, in this investigation, the compressive strength of rubble samples was not investigated. Therefore it is recommended to investigate the influence of strength of different rubble stones on different strength parameters of RRM. 3) To increase the number of test specimens During the experimental investigation, there were several instances where failure patterns of the specimen were not acceptable. Hence results of those specimens had to be discarded. Also there can be damages during handling of specimens. Therefore it is necessary to test more samples to get more appropriate characteristic strength values. 77 REFERENCE LIST [1] BS EN 1052- 1:1999, “Methods of test for masonry -Part 1: Determination of compressive strength", British Standards Institution, London. [2] BS EN 1052- 2:1999, “Methods of test for masonry -Part 2: Determination of flexural strength", British Standards Institution, London [3] BS EN 1052- 3:2000, “Methods of test for masonry -Part 3: Determination of shear strength ", British Standards Institution, London [4] P.K.S. Premadasa, S.K. Liyanage, H.D.A.R. Ariyarathna, S.A.D. Induprabha and K.G.S. Dilrukshi, “Experimental approach to investigate concrete- Masonry interface”, Civil engineering research for industry, University of Moratuwa, 2011 [5] S.R.D.S. Chandrakeerthy, "A study of Random Rubble Masonry Construction in Sri Lanka", Annual Transactions of IESL, 1998 [6] BS 8002:1994, Code of Practice for Earth Retaining Structures, British Standards Institution, London. [7] BS EN 1997-1:2004, Eurocode 7: Geotechnical design – Part 1: General Rules, European Committee for Standardization, Brussels. [8] R.F.Craig, Craig’s Soil Mechanics, 7th edition, London, UK, Spoon Press, 2004, pp. 161-226. [9] S.O.Franklin and A.L.Olopade, “A Note on the Analysis and Design of Cantilever Retaining walls using the CECP2 and BS8002 Methods”, IJRRAS, vol. II , Issue I, April 2012 [10] R.F.Craig, Craig’s Soil Mechanics, 6th edition, London, UK, Spoon Press, 1997, [11] C. Liu and J.B.Evett, Soils and Foundations, 7th edition, New Jersey, Prentice Hall, 2008, [12] S. Fernando S and V. Jayasena, “Innovative earth retaining system adopted at the proposed printing complex for Aithken Spence Ltd. at Mawaramandiya”, Civil engineering research for industry, University of Moratuwa, 2011. 78 [13 ] D. Urquhart, Natural stone masonry in modern Scottish construction; A guide for designers and constructors, Scottish Stone Liaison Group, Dunfermline, pp.119- 125, 2008 [14] Sri Lanka. Institute for Construction Training and Development (ICTAD), Specifications for Building works- Volume 1, Colombo, Ministry of Housing and Plantation Infrastructure, Jul-2004 [15] S.R.D.S. Chandrakeerthy, "Random Rubble Stone Masonry with Architectural features”, Annual Transactions of IESL, 1999 [16] BS 5628- 1:1994, Code of Practice for Structural Use of Masonry-Part I:Un reinforced Masonry, British Standards Institution, London. [17] J. Milosevic, A.S.Gago, M.Lopes and R. Bento, " Experimental tests on Rubble Masonry specimens- Diagonal Compression, Triplet and Compression Tests", Technical University of Lisbon, Portugal [18] ASTM C 952-02:2003, Standard Test Method for Bond Strength of Mortar to Masonry units, ASTM international, USA. [19] W.H.Mosley, J.H.Bungey and R.Hulse, Reinforced Concrete Design, 5th edition, UK, Mac press, 2001 [20] M.D. Bolton, “Geotechnical design of retaining walls”, The Structural Engineer, vol. 74, no. 21, pp.365-369, Nov 5 1996. [21] T.N.W. Akroyd, “Earth-retaining structures: Introduction to the Code of Practice”, The Structural Engineer, vol. 74, no. 21, pp.360-364, Nov 5 1996. [22] M.D. Bolton, Codes standards and design guides in retaining structures, Thomas Telford, 1993 [23] C. Jayasinghe and K.M.C. Konthesinghe, “Effect of Tie beams on Lateral Strength of Masonry”, Annual Transaction of IESL,2008. 79 Appendix A: Flexural Strength -Experimental Data and Results A.1 Flexural Strength When Plane of Bending is Horizontal A.1.1 Experimental Data Date of making samples : 12-July-2013 Date of Testing : 2-September-2013 Distance between Outer bearings : 1200mm Distance between inner bearings : 550mm A.1.2 Experimental Results Table A.1: Results of flexural strength test (When plane of bending is horizontal) Specimen Dimensions/ (mm) Dial gauge reading Load* /(N) Comments on failure pattern Average Width Average Height Average Thickness 1 1378 1000 300 110 19,078 Not satisfactory, Failure at an outer bearing 2 1400 1000 300 215 37,278 Satisfactory 3 1366 1000 300 230 39,874 Satisfactory * The Dial gauge reading was converted into load in N, referring to the calibration Table of the Proving ring (Appendix- E) A.2 Flexural Strength When Plane of Bending is Vertical A.2.1 Experimental Data Date of making samples: 16-Jul-2013 Date of Testing : 10-Oct-2013 Distance between Outer bearings : 1200mm Distance between inner bearings : 550mm 80 A.2.2 Experimental Results Table A.2: Results of flexural strength test (When plane of bending is vertical) * The Dial gauge reading was converted into load in N, referring to the calibration Table of the Proving ring (Appendix- E) Specimen Dimensions/ (mm) Dial gauge reading Load* / (N) Comment on failure pattern Average Width Average Height Average Thickness 4 560 1400 300 30 5182 No clear crack pattern 5 565 1390 300 30 5182 No clear crack pattern 6 560 1380 300 30 5182 No clear crack pattern 81 Appendix B: Shear Strength -Experimental Data and Results B.1 Experimental Data Date of making samples: 31-Jul-2013 to 2-Aug-2013 Date of Testing : 24-Oct-2013 B.2 Experimental Results Table B.1: Results of Shear Strength Test Specimen Dimensions/ (mm) Dial gauge reading Load* /(N) Pre- Compression Failure pattern Average Width Average Height Average Thick. Dial gauge reading Load (N) 1 455 600 300 140 8092 28 4834 Satisfactory 2 455 600 300 240 13841 55 9527 Satisfactory 3 450 600 300 80 4640 70 12133 unsatisfactory 4 450 600 305 91 5273 Satisfactory 5 450 605 310 155 8954 Satisfactory 6 450 600 310 58 3374 Satisfactory * The Dial gauge reading was converted into load in N, referring to the calibration Table of the Proving ring (Appendix- E) 82 Appendix C: Shear Strength at Concrete Masonry Interface - Experimental Data and Results C.1 Experimental Data Date of making samples : 27-July-2013 Date of casing of RC beam : 30- July-2013 Date of Testing : 29-Oct-2013 C.2 Experimental Results Table C.1: Results of Shear strength at concrete masonry interface Specimen Dimensions of Tie beam/ (mm) Dimensions of Masonry panel/ (mm) Dial gauge reading Load * /(N) Failure surface Length Height Width Width Height Thick. 1 450 225 225 450 600 300 55 3201 Masonry- Masonry 2 450 225 225 450 600 300 98 5675 Masonry- Masonry 3 450 225 225 450 600 300 210 12117 Concrete- Masonry 4 430 225 225 430 600 300 110 6366 Concrete- Masonry & Masonry- Masonry 5 450 225 225 450 600 300 245 14128 Concrete- Masonry 6 450 225 225 450 600 300 95 5503 Masonry- Masonry * The Dial gauge reading was converted into load in N, referring to the calibration Table of the Proving ring (Appendix- E) 83 Appendix D: Compressive Strength -Experimental Data and Results D.1 Experimental Data Date of casting: 17-July-2013 Date of Testing: 24-Oct-2013 D.2 Experimental Results Table D.1: Results of Compressive strength test Specimen Dimensions/ (mm) Dial gauge reading /(Ton) Load*/ (N) Average Width Average Height Average Thickness 1 595 600 300 15 147150 2 570 600 300 13.1 128511 3 600 600 300 21.4 209934 * The Dial gauge reading was converted into load in N, referring to the calibration Table of the Proving ring (Appendix- E)