UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The UBC ring shear device 1980

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
UBC_1980_A7 B68.pdf
UBC_1980_A7 B68.pdf [ 4.74MB ]
Metadata
JSON: 1.0062554.json
JSON-LD: 1.0062554+ld.json
RDF/XML (Pretty): 1.0062554.xml
RDF/JSON: 1.0062554+rdf.json
Turtle: 1.0062554+rdf-turtle.txt
N-Triples: 1.0062554+rdf-ntriples.txt
Citation
1.0062554.ris

Full Text

THE UBC MNG SHEAR DEVICE BRUCE W. BOSDET B . A . S c , Un i vers i ty ,of B r i t i s h Columbji, 1975 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE i n THE FACULTY OF GRADUATE STUDIES Department of C i v i l Engineering The Univers i ty of B r i t i s h Columbia / i We accept this thes is as conforming / to the required standard / ' THE UNIVERSITY OF BRITISH COLUMBIA October, 1980 (c) Bruce W. Bosdet In presenting th i s thes is in pa r t i a l fu l f i lment of the requirements for an advanced degree at the Un ivers i ty of B r i t i s h Columbia, I agree that the L ibrary shal l make it f ree ly ava i lab le for reference and study. I further agree that permission for extensive copying of th i s thesis for scho lar ly purposes may be granted by the Head of my Department or by his representat ives. It is understood that copying or pub l i ca t ion o f th is thes is fo r f i nanc ia l gain sha l l not be allowed without my written permission. Department of C i v i l Engineering The Univers i ty of B r i t i s h Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date October 10, 1980 i i ABSTRACT The purpose of th is thesis has been to design and develop a p r ac t i c a l apparatus for determining the residual strength of c lay s o i l s . To provide background for the study, current knowledge regarding residual strength is reviewed, inc lud ing the fo l lowing po i nts. 1. Residual s trength, defined as the lowest drained strength a s o i l can e x h i b i t , is at ta ined at large shear displacements. 2. Residual condit ions resu l t when p a r t i c l e s located in shear bands with in the f a i l u r e zone become a l igned in the d i r e c t i o n of shear. 3. Residual s t rength, derived from i n t e r p a r t i c l e bonding, is inf luenced by c ry s ta l s t ructure and, in ac t i ve clay minera ls , by pore water chemistry. The r ing shear t e s t , performed by applying a tors iona l shear load to an annular shaped specimen, is p a r t i c u l a r l y su i ted for residual strength determinations because of unl imited un i -d i rec t iona1 s t r a i n c a p a b i l i t i e s . The UBC Ring Shear Device was designed to combine v e r s a t i l i t y with uncomplicated operat ion. Features of the design are as fol1ows. 1. Var iab le sample height up to 0.75 inches. 2. Smoothly var i ab le normal s t ress up to 200 psi de l i vered through an a i r p i s ton. 3. Smoothly var iab le rate of shear from 3-2 inches per year to 9 inches per hour. i i i *t. A n o n - t i l t i n g loading platen which reduces required machine tolerances and improves contro l o f sample losses during te s t i ng . 5. Automatic data a c q u i s i t i o n . 6. A simple method of sample placement. Residual strength determinations obtained with the UBC Ring Shear Device demonstrate i t s e f f i c i e n t and e f f e c t i v e operat ion. Min- imal superv is ion is required and test resu l ts are ea s i l y in terpreted . Mu1ti-reversa1 d i rec t shear tests for residual strength were undertaken for comparison with the r ing shear re su l t s , but no s a t i s f a c t o r y resu l t s were obtained due to excess pore pressures with in the test specimens. Recommendations fo r improvements to both r ing shear and d i r e c t shear devices are given. i v CONTENTS Page CHAPTER I INTRODUCTION 1 CHAPTER II RESIDUAL STRENGTH BEHAVIOUR OF CLAYS 2.1 INTRODUCTION ^ 2.2 SHEAR BEHAVIOUR OF CLAY 2.3 FAILURE PLANE STRUCTURE 7 2.4 CLEAVAGE 9 2.5 BONDING 11 2.6 STRESS DEPENDENT BEHAVIOUR 16 CHAPTER III RESIDUAL STRENGTH TESTING METHODS 20 3.1 INTRODUCTION 20 3.2 DIRECT SHEAR TEST 21 3.3 TRIAXIAL TEST 22 3.4 RING SHEAR TEST 23 CHAPTER IV THE UBC RING SHEAR DEVICE 26 4.1 INTRODUCTION 26 4.2 GENERAL DESCRIPTION 26 4.3 DETAILED DESCRIPTION 29 4.3-1 The Drive System 29 4.3.2 The Upper Ring Assembly 31 4.3.3 The Loading System 33 4.3.4 Measurement of Normal Loads 35 4.3.5 Measurement of Shear Forces 35 4.3.6 Displacement Measurements 37 4.4 SAMPLE PLACEMENT 37 4.5 DATA COLLECTION 40 V Page CHAPTER V TESTING PROGRAM 41 5.1 INTRODUCTION 41 5.2 RING SHEAR PERFORMANCE TESTS 42 5.3 RESIDUAL STRENGTH TESTS ON HANEY CLAY 48 5.3-1 So i l Descr ipt ion 48 5.3.2 Ring Shear Tests 49 TEST PROCEDURES AND RESULTS 49 DISCUSSION AND INTERPRETATION OF RESULTS 50 5.3.3 Direct Shear Tests 59 TEST PROCEDURES AND RESULTS 59 DISCUSSION AND INTERPRETATION OF RESULTS 59 5.3.** Review and Evaluation of Results 67 5.4 RESIDUAL STRENGTH TESTS ON HAT CREEK CLAY 69 5.^.1 So i l Descr ipt ion 69 5.4.2 Ring Shear Tests 70 TEST PROCEDURES AND RESULTS 70 DISCUSSION AND INTERPRETATION OF RESULTS 76 5.4.3 D irect Shear Tests 79 TEST PROCEDURES AND RESULTS 79 STRAIN CONTROLLED CONSOLIDATION TEST 84 DISCUSSION AND INTERPRETATION OF RESULTS 85 5.4.4 Review and Evaluation of Results 87 5.5 RESIDUAL STRENGTH TESTS ON IRANIAN CLAY 89 5.5.1 Soi l Descr ipt ion 89 5.5.2 Ring Shear Tests 90 5.6 SUMMARY OF TESTING PROGRAM 99 CHAPTER VI SUMMARY AND EVALUATION 100 6.1 RESIDUAL STRENGTH 100 6.2 UBC RING SHEAR DEVICE 101 6.3 TESTING PROGRAM 102 6.4 EVALUATION 103 REFERENCES 106 APPENDIX I: DERIVATION OF EQUATION FOR DETERMINING tan 0' 109 APPENDIX II: DIRECT SHEAR APPARATUS 110 v i TABLES Page Table 2.1 Cleavage and Residual F r i c t i o n Angle of Some Clay Minerals 10 Table 2.2 Bonding Along Cleavage Planes and Residual Strength 13 Table 5-1 Schedule of Ring Shear Tests 43 Table 5-2 Schedule of D irect Shear Tests 44 Table 5-3 Test Results : RS 2 and RS 4 on Haney Clay 54 Table 5-4 Test Results : RS 8 on Haney Clay 55 Table 5-5 Test Results : D irect Shear Tests on Haney Clay 62 Table 5.6 Test Results: Ring Shear Tests on Hat Creek Clay 74 Table 5-7 Test Results: Direct Shear Tests on Hat Creek Clay 8 3 Table 5.8 Test Results: Ring Shear Tests on Iranian Clay 97 v i i ILLUSTRATIONS Page Fi gure 2. 1 Shear Behaviour of Clays Under Fu l l y Drained Condi t ions 6 Fi gure 2. 2 Shear Zone Textures Strength State and Fabric at the Residual 8 Fi gure 2. 3 Residual Shear Strength versus Apparent E f f e c t i v e Stress for Sodium Montmori1 Ionite at Various Pore F lu id Salt (NaCl) Concentrations 15 Fi gure 2. 4 Relat ionship Between Shearing Resistance and Normal Stress for D i f fe rent Minerals 17 Fi gure 2. 5 Shearing Resistance Stress on the Shear 1/3 Power versus Normal E f f e c t i v e Plane Raised to the Minus 19 Fi gure 4. 1 The UBC Ring Shear Dev i ce 27 Fi gure 4. 2 Major Components of UBC Ring Shear Device 28 Fi gure 4. 3 Drive System 30 Fi gure 4. Upper Ring Assembly and Suspension System 32 Fi gure 4. 5 Loading System 34 Fi gure 4. 6 Sample Cutting Unit 38 Fi gure 5. ,1 Shearing Resistance vs. Displacement - RS 3 46 Fi gure 5-,2 Shearing Resistance vs. Displacement .- RS 2 51 Fi gure 5-• 3 Shearing Resistance vs. Displacement - RS 4 52 Fi gure 5. .4 Shearing Resistance vs. Displacement - RS 8 53 Fi gure 5. ,5 Mohr Rupture Diagram: Summary of Results of Residual Strength Tests on Haney Clay 56 Fi gure 5. .6 Shearing Resistance Displacement - DS 3 vs. Cumulative 60 v i i i ILLUSTRATIONS (Cont'd) Page Figure 5-7 Figure 5-8 Figure 5-9 Figure 5-10 Figure 5-11 Figure 5.12 Fi gure 5-13 Fi gure 5 • 1*» Fi gure 5.15 Figure 5-16 Figure 5-17 Figure 5.18 Figure 5.19 Figure 5-20 Figure 5-21 Figure 5-22 Shearing Resistance vs. Cumulative Displacement - DS 5 Determining the Influence of an Incl ined Fai1ure Surface Shear Plane at End of Test ing - DS 3 Shearing Resistance vs. Displacement - RS 5 Shearing Resistance vs. Displacement - RS 6 Shearing Resistance vs. Displacement - RS 7 Mohr Rupture Diagram: Summary of Results of Residual Strength Tests on Hat Creek Clay Relat ion Between Clay Content and Residual Shearing Resistance for Various So i l s Shearing Resistance vs. Cumulative Displacement - DS 1 Shearing Resistance vs. Cumulative Displacement - DS 2 Shearing Resistance vs. Cumulative Displacement - DS k Shearing Resistance vs. Displacement - RS 9 Shearing Resistance vs. Displacement - RS 10 Shearing Resistance vs. Displacement - RS 11 Total Normal and Shear Stress vs. Displacement - RS 10 Mohr Rupture Diagram: Summary of Results of Residual Strength Tests on Iranian Clay 61 64 66 71 72 73 75 78 80 81 82 91 92 93 95 98 i x ACKNOWLEDGMENTS Many people have contr ibuted to the completion of th i s p ro jec t . Fred Zurkirchen, a machinist at U.B.C., pa t i en t l y advised me in the p rac t i ca l aspects of the equipment design and constructed the r ing shear device. Golder Assoc iates , Vancouver, and Sydney H i l l i s , a consu l t ing engineer in Vancouver, provided in teres t ing so i l specimens for the tes t ing program. Friends and family provided the needed encouragement to complete the wr i t ten work when the tes t ing was done. However, th i s thes i s would never have been completed without the a id of my superv i sor , R.G. Campanella, Department of C i v i l Engineering, U.B.C., who suggested the project and provided support and guidance throughout. 1. CHAPTER I INTRODUCTION A s o i l subjected to shear displacement under drained condi - t ions w i l l normally exh ib i t a r i se in shearing res i s tance with increas ing displacement unt i l a maximum res i s tance, the peak s t rength, is reached. With continued displacement, the shearing res i s tance of the s o i l w i l l decrease un t i l a minimum res i s tance, unaffected by further displacements, is a t ta ined . The minimum res i s tance, termed residual s trength, is the lowest shearing res i s tance the so i l can exh ib i t under drained cond i t ions . This general re l a t i onsh ip between shear displacement and the drained shear strength of s o i l has long been recognized. In p a r t i c u l a r , peak strength and the behaviour of s o i l s at small s t ra in s has been studied in great d e t a i l . However, the f i e l d of res idual strength and s o i l behaviour at large s t ra ins has been large ly unexplored un t i l recent years. The r e l a t i v e i n a c t i v i t y in th is f i e l d can be a t t r i bu ted to p rac t i ca l d i f f i c u l t i e s in tes t ing to large s t ra ins and to an under- assessment of the importance of residual strength to the s t a b i l i t y of earth s t ruc tures . Ear ly studies into the res idual strength behaviour of s o i l s were ca r r i ed out in the 1930's. The i n a b i l i t y o f conventional shear te s t i ng techniques to model f i e l d condit ions by producing large un i - d i r e c t i o n a l shear s t ra ins led the ear ly researchers to devise large displacement rotat iona l shear devices of varying design. Several of these devices performed tors iona l shear tests on annular s o i l specimens 2. and, as such, were the f i r s t r ing shear devices. The i n i t i a l in teres t in res idual strength apparently stemmed from the recognit ion that residual s t rength, which could be as l i t t l e as 20 percent of peak strength, governs the s t a b i l i t y of prev ious ly f a i l e d s tructures (Hvorslev, 1 939 ) . This in teres t was shor t ly lost as a t tent ion turned to the study of peak strength behaviour and i t s broader app l i c a t i on to unfa i led s lopes. However, a t tent ion returned to residual strength in the mid -1960 ' s when res idual strength was c l e a r l y l inked to the i n s t a b i l i t y o f cer ta in s o i l s which had not previous ly f a i l e d . In the Fourth Rankine Lecture, Skempton (1964) presented evidence to show that the long-term s t a b i l i t y o f slopes in overcon- so l idated f i s sured clays is governed, at least in the lower bound, by residual strength through the mechanism of progress ive f a i l u r e . A progress ive f a i l u r e may occur i f the load on a small sect ion of a large so i l mass surpasses the peak s o i l s trength. As the peak strength is exceeded, the shearing res is tance of the s o i l f a l l s toward the res idual value and the loads previous ly ca r r i ed by the small sect ion of s o i l are t rans ferred to adjacent areas, causing the stresses in these areas to exceed the peak strength a l so . In th i s manner, a f a i l u r e may progress throughout a large s o i l mass without ever causing a large sca le move- ment. This is p a r t i c u l a r l y true in heavi ly overconsol idated c lays which genera l ly pass the peak strength at r e l a t i v e l y small s t r a i n s . Bjerrum (1967) has indicated that an overconsol idated c lay need not contain f i s su res for i t s long-term s t a b i l i t y to be governed by residual s t rength. 3- The app l i c a t i on of res idual strength to the s t a b i l i t y of un fa i l ed s lopes, as ou t l i ned by Skempton, sparked new research into the changes that occur in so i l during the t r a n s i t i o n from peak to residual strength and led to the development of new rotat iona l devices for determining the res idual strength of s o i l s . The UBC r ing shear apparatus is one of these new devices. The purpose of th i s thes i s project has been to design, construct and test a p r ac t i c a l apparatus for determining the res idual strength of s o i l . During the course of the study, ava i l ab l e l i t e r a t u r e was reviewed to develop an understanding of res idual strength behaviour and the h i s t o r i c a l development of the res idual strength apparatus. Chapter II describes residual strength behaviour and reviews current knowledge and theory regarding the causes of th i s behaviour. Chapter III b r i e f l y reviews and evaluates the commonly used methods for determining residual strength and traces the h i s t o r i c a l development of r ing shear devices. The design of the UBC Ring Shear Device is described in de ta i l in Chapter IV. Chapter V presents the resu l t s of the te s t ing program undertaken to evaluate the new r ing shear device and to explore te s t ing methods and procedures. The resu l t s of repeated d i r ec t shear t e s t s , conducted for purposes of comparison, are a l so presented and discussed. A summary of the work and an evaluat ion of the equipment is provided in Chapter VI, together with recommendations for improvements to the apparatus. CHAPTER I I RESIDUAL STRENGTH BEHAVIOUR OF CLAYS 2.1 INTRODUCTION Considerable advances have been made s ince the mid-1960's in our understanding of s o i l behaviour at large s t r a i n s . It has been shown that granular and clayey s o i l s exh ib i t s im i l a r macroscopic behaviour in shear, but that only the clayey so i l s , and to a lesser extent dense sands, exh ib i t the s i g n i f i c a n t drop from peak to res idual strength which resu l t s in important engineering app l i c a t i on s . For th i s reason, c lays are the focus of a t tent ion in the study of res idual s trength. This chapter provides a review of c lay behaviour in shear and the microscopic mechanisms which inf luence the residual strength of these so i1s . 2.2 SHEAR BEHAVIOUR OF CLAY A l l s o i l s subjected to shear displacement under drained condit ions w i l l eventual ly a r r i ve at a shear strength and water content that are functions of the s o i l composition and appl ied stresses on ly , regardless o f the i n i t i a l s tate o f the s o i l . The strength of the s o i l is then the lowest that can be achieved and is termed the ult imate or residual s t rength. Further shearing w i l l not a f f e c t the strength of the so i l once the residual condi t ion is e s tab l i shed . Because res idual strength is unaffected by the i n i t i a l s t ructure or s t ress h i s to ry of the s o i l s , i t is sometimes considered to be a fundamental s o i l para- meter, an unchanging c h a r a c t e r i s t i c which i d e n t i f i e s the s o i l . 5. The s t r e s s - s t r a i n and volume change c h a r a c t e r i s t i c s exh ib i ted by a c lay before reaching the res idual condi t ion are funct ions of the s t ress h i s tory of the s o i l . Heavi ly overconsol idated c lays have high peak strengths and genera l ly exh ib i t a large decrease from peak to res idual s trength. The reduction in strength is accompanied by an increase in void ra t i o and water content. Normally consol idated clays genera l ly have lower peak strengths than overconsol idated c lays and exh ib i t a smaller decrease from peak to residual s trength. This decrease in strength is accompanied by a reduction in void r a t i o and water content. This typ ica l s o i l behaviour is i l l u s t r a t e d in Figure 2 . 1 . As with peak strength behaviour, the residual strength of a c lay is described in terms of a f r i c t i o n angle, 0̂ ., and a cohesion in tercept , c^., as fo l lows: T = c 1 +G V | tan 0' L<" r r = c 1 + ((P-u) tan 0' r r where ' 7 \ . = residual shear s tress CT" = tota l normal s t ress on the shear plane u = pore water pressure CT1 = (T*- u = e f f e c t i v e normal s tress For most natural s o i l s , the res idual strength cohesion intercept is c lose or equal to zero, and the f r i c t i o n angle is less than the peak f r i c t i o n angle. The residual strength behaviour is commonly described by the shearing res i s tance r a t i o , TVcr1 = tan 0' ( for c ' = 0) FIGURE 2.1 Shear Behaviour of Clays Under Fu l l y Drained Conditions (from Mi tche l1, 1976) 7. 2.3 FAILURE PLANE STRUCTURE The physical process by which a f a i l u r e plane develops in c lay subjected to shear displacement has been studied by Morgenstern and Tchalenko (1967)- Their microscopic observations of kao l in c lay i n d i - cate that c lay deforms in simple shear up to the displacement at peak strength. However, beyond the peak strength, the so i l gradual ly develops a complex system of shear s t ructures and p a r t i c l e o r i en ta t i on s which lead to the establishment of a shear zone. At large displacements, as the clay approaches the residual cond i t i on , the shear zone cons is ts of two thin shear bands, in which the microscopic plate-shaped clay p a r t i c l e s are or iented in the d i r e c t i o n of shear displacement, enc los ing a region of compression textures, as shown in Figure 2.2. The compression textures resu l t from the p a r a l l e l alignment of c lay p a r t i c l e s in the regions between shear bands such that the p a r t i c l e s are or iented with the basal c rys ta l planes approximately perpendicular to the major p r inc ipa l s t re s s . The dominant mechanism of deformation at large s t ra in s is by s l i p between the basal cleavage planes of adjacent p a r a l l e l a l igned p a r t i c l e s within the two thin shear bands. A l l deformations are accounted for by basal plane s l i p or p a r t i c l e ro ta t i on . Shear f a i l u r e through p a r t i c l e s does not appear to occur. The study by Morgenstern and Tchalenko a lso showed that the p a r t i c l e s must be in v i r t u a l l y perfect p a r a l l e l alignment before the res idual strength is obta ined. Small local p a r t i c l e d i so r ien ta t i ons s i g n i f i c a n t l y increase the measured f r i c t i o n angle. Tests ca r r i ed out during the study ind icate that the i n i t i a l o r i en t a t i on of samples composed of p a r t i c l e s with a strong preferred p a r a l l e l alignment 500 ji™ ~ 7 Shear Zone Textures and Fabric at the Residual Strength State (from M i t c h e l l , 1 9 7 6 ; a f t e r Tchalenko, 1968 ) (a) Structures formed in d i rec t shear test (b) P r inc ipa l displacement shears in d i rec t shear test S.j, i n i t i a l f a b r i c a t t i tude (fmajor p r inc ipa l stress d i rect ions at peak and ^ ^ residual strengths, respect ive ly P, thrust shears R, Riedel shears hatchings, p a r t i c l e o r ien ta t i on in compression texture white areas, p a r t i c l e s in i n i t i a l f ab r i c a t t i tude 9- does not a f f e c t the displacement required to obtain the res idual cond i t i on . However, p re - cu t t i ng the shear plane p r i o r to shearing great ly reduces the required displacement, presumably by improving c lay p a r t i c l e alignment. 2.4 CLEAVAGE The residual strength of c lay s o i l s depends, in par t , on the c ry s ta l cleavage mode. Although, as discussed in the previous s ec t i on , s l i p along the basal cleavage planes is the dominant mechanism of deformation at large s t r a i n s , some c lay minerals have no easy s l i p planes and exh ib i t higher residual strengths. For example, a t t a p u l g i t e , a f ibrous needle-shaped c lay mineral which shears p r e f e r e n t i a l l y along a stepped cleavage plane, has a very high residual angle of f r i c t i o n (see Table 2.1 ). The high shear strength of a t t apu l g i t e may a l so be re lated to the intermeshing of p a r t i c l e s which occurs even at large s t ra in s (Chattopadhyay, 1972, p. 234). K a o l i n i t e , i11 i te and mont- m o r i l l o n i t e , with a strongly preferred basal c leavage, have low residual strengths. With the poss ib le exception of montmori1lonite, s l i p does not appear to occur on cleavage planes with in the mineral c r y s t a l s , but between cleavage planes of adjacent p a r t i c l e s (Chattopadhyay, 1972, p. 61). TABLE 2.1 Cleavage and Residual F r i c t i o n Angle of Some Clay Minerals Mineral Cleavage Mode Bonding Along Cleavage Planes Residual F r i c t i o n Angle 0' r (degrees) Montmori1loni te Easy (001) basal cleavage Secondary valence plus exchangeable ion l ink 8.5 Kaoli n i te Easy (001) basal cleavage Secondary valence plus hydrogen bond i ng 11 Attapulg i te Sta i rcase 1i ke (110) cleavage Si - 0 - Si weak 1i nk 30 ( a f ter Chattopadhyay, 1972) 11. 2.5 BONDING Based on studies o f s o i l creep as a rate process, M i t c h e l l , Singh and Campanella (1969) developed a hypothesis concerning the re l a t i onsh ip between bonding, e f f e c t i v e stress and strength of s o i l . Chattopadhyay (1972) reached subs tan t i a l l y s i m i l a r conclus ions based on his study of the inf luence of pore water chemistry on the residual strength of c l ay s . M i t che l l e t al (1969) postu late that the only s i g n i f i c a n t mechanism by which e f f e c t i v e normal and shear s tresses are transmitted through a s o i l mass is by intergranular contact. The in tergranu lar contact is e s s e n t i a l l y s o l i d - t o - s o l i d , although adsorbed water layers may sometimes behave as part of the c ry s ta l s t ruc tu re . The shearing res is tance of the s o i l is derived from inter-atomic and inter -molecu lar bonding across the contact zone between adjacent p a r t i c l e s . D irect physical evidence of s o l i d - t o - s o l i d i n t e r p a r t i c l e contacts in c lays has been obtained in the form of scanning e lec t ron microscope photographs of c lay p a r t i c l e s (Matsui, Ito, M i tche l l and Abe, 1980). Microphotographs of kao l in c lay p a r t i c l e s obtained a f t e r shearing show occasional scratches on p a r t i c l e surfaces which are interpreted as tracks formed by i n t e r p a r t i c l e f r i c t i o n . No such tracks were observed on the p a r t i c l e s p r i o r to shearing. For fur ther evidence of s o l i d - t o - s o l i d contacts in clays Matsui et al (1980) point to the work of Koerner, Lord and McCabe ( 1 9 7 7 ) . who recorded the acoust i c emissions of cohesive s o i l s during shear. It is suggested that the acoust i c emissions occur as a resu l t of s o l i d - t o - s o l i d i n t e r p a r t i c l e contacts and that such emissions should not be expected i f the contacts occurred between " s o f t " adsorbed water layers . 12. The shear strength of c lay is governed by the strength and concentrat ion of bonds in the plane of deformation (Chattopadhyay, 1972). As shown by Morgenstern and Tchalenko (1967), beyond the peak strength displacement, deformation occurs by s l i p between mineral cleavage planes of adjacent p a r a l l e l o r iented mineral c r y s t a l s in the two th in shear bands. The s l ippage involves the continuous rupture and formation of bonds in the i n t e r p a r t i c l e contact areas. Between minerals with the same cleavage mode, the mineral with the lesser bond energy per unit o f contact area w i l l have the lower residual s t rength. The general re l a t i onsh ip between bonding and residual strength is i l l u s t r a t e d by the data tabulated in Table 2.2. For normally consol idated c l ay s , the number of bonds formed across an i n t e r p a r t i c l e contact zone appears to be proport ional to the e f f e c t i v e normal force at the contact (Mitchel l et a l , 1969)- This e f f e c t may be expla ined by the Terzaghi-Bowden and Tabor adhesion theory of f r i c t i o n which states that the i n t e r p a r t i c l e contact area is proport ional to the e f f e c t i v e normal f o rce . An increase in e f f e c t i v e normal force resu l t s in a proport ionate increase in the s i ze of the contact zone, the number of bonds per unit of contact area remaining constant. However, in the absence of shear deformation, reducing the conso l ida t ion pressure appl ied to a clay does not resu l t in a proport ionate reduction in the number of bonds, thus accounting for the higher shear strength of overconsol idated c lays . Water does not appear to a f f e c t the strength of bonds, but does a l t e r the e f f e c t i v e ( p a r t i c l e - t o - p a r t i c l e ) s t ress and, there fore , the number of bonds which form. TABLE 2.2 Bonding Along Cleavage Planes and Residual Strength Mi neral Quartz A t tapu lg i te Mi ca Kaoli n i te 111i te Montmori1 Ion i te Ta lc Graph i te MoS„ Mode of Cleavage No d e f i n i t e cleavage Along (110) plane Good basal (001) Basal (001) Basal (001) Bonding Along Cleavage Planes S i-0-S i , weak Secondary valence (0.5 to 5 kcal/mole) + K- l inkages Secondary valence (0.5 to 5 kcal/mole) + H bonds Secondary valence (0.5 to 5 kcal/mole) + K l inkages Exce l lent basal (001) Secondary valence (0.5 to 5 kcal/mole) + exchangeable i on 1i nkages Basal (001) Basal (001) Basal (001) Secondary valence (0.5 to 5 ka1/mo1e) van der Waa11s Weak in ter layer Residual F r i c t i o n Angle, 0'r 3 5 degrees 30 degrees 17 to 2k degrees 12 degrees 10.2 degrees k to 10 degrees 6 degrees 3 to 6 degrees 2 degrees P a r t i c l e Shape Bui ky Fibrous and needle-shaped Sheet Platy Platy P l a t y - f i lmy Platy Sheet Sheet From M i t c h e l l , 1976; as adapted from Chattopadhyay, 1972. 14. The true e f f e c t i v e normal force by which contacts are made and bonds are formed is composed of ex te rna l l y app l ied forces ( inc lud ing the se l f -weight of s o i l ) and/or i n te rna l l y generated physico-chemical forces . In genera l , the physico-chemical forces are r e l a t i v e l y sma l l , the true e f f e c t i v e s tress is v i r t u a l l y equal to the apparent e f f e c t i v e s t re s s , and the strength of c lay s o i l s appears purely f r i c t i o n a l in character . However, in very ac t i ve c lay minerals, such as sodium montmori1lonite, large physico-chemical forces carry part of the normal s t res s . As the pore water chemistry great ly inf luences the magnitude of the physico-chemical forces in ac t i ve c l ay s , changes in pore water chemistry cause changes in the apparent residual angle of f r i c t i o n , as i l l u s t r a t e d in Figure 2.3. Studies by Chattopadhyay (1972, p. 326) ind icate that, i f the physico-chemical forces are taken into account, the behaviour of sodium montmori1 Ionite is purely f r i c t i o n a l , s im i l a r to the behaviour of less ac t i ve c l ay s . He concludes that every mineral has an unchanging true, or i n t r i n s i c , residual angle of f r i c t i o n . E l e c t r o l y t i c concentra- t ion may inf luence water content and physico-chemical s t res ses , thereby a l t e r i n g the true e f f e c t i v e s t re s s , but i t has no e f f e c t on the strength generating mechanism or the true res idual angle of f r i c t i o n . In p r ac t i c a l app l i c a t i on s , however, the apparent res idual angle of f r i c t i o n measured is the macroscopic or mass s o i l property, no allowance being made for the inf luence of physico-chemical fo rces . A de ta i l ed theoret i ca l treatment of th i s microscopic model of shear strength in c lay s o i l s is provided by Matsui et al ( 1 9 8 0 ) , inc luding cons iderat ion of the i n t r i n s i c or true angle of f r i c t i o n and 15- FIGURE 2.3 Residual Shear Strength Versus Apparent E f f e c t i v e Stress for Sodium Montmori1lonite at Various Pore F lu id Salt (NaCl) Concentrations (a f ter Chattopadhyay, 1972) 16. the e f fec t s of physico-chemical fo rces , d i l a tency , cementation, suct ion and stress h i s to ry . 2.6 STRESS DEPENDENT BEHAVIOUR Studies have shown that, for many c l ay s , the apparent res idual f r i c t i o n angle is greater at normal stresses less than 30 psi than at higher s t resses , as shown in Figure 2.k. As recorded by Mi tche l l (1976, p.315), two poss ib le explanations for th is curvature of the res idual strength envelope in the low normal s tress region are as fo l lows. 1. The work required to a t t a i n a per fect p a r t i c l e alignment in the f a i l u r e plane under low normal stresses exceeds the work required to shear the so i l without such alignment. Therefore, the p a r t i c l e s are less than p e r f e c t l y a l igned under low normal stresses and the shear strength is correspondingly increased. Under higher normal s t resses , less work is required i f the p a r t i c l e s are p e r f e c t l y a l igned and the lower res idual strength re su l t s . 2. Under low normal s tresses the p a r t i c l e contact areas behave e l a s t i c a l l y . By e l a s t i c junct ion theory, i t can be shown that the real contact area increases p ropor t i ona l l y with ((P) •' at the e l a s t i c contact. The shearing re s i s tance , /> -1/3 tan fb'r = <-/(T~ ' , must therefore vary as (<T̂ ) . At higher normal stresses the p a r t i c l e contact areas behave p l a s t i c a l l y and the s o l i d - t o - s o l i d contact areas are d i r e c t l y proport ional 17. 0 1 2 3 4 5 6 7 8 Normal Stress o'„ (kg/cm2) FIGURE 2.4 Relat ionship Between Shearing Resistance and Normal Stress for D i f fe rent Minerals (From M i t c h e l l , 1976; a f t e r Kenny, 1967) 18. to the e f f e c t i v e normal f o rce . This concept is supported by -1 /3 the p lot of tan 0' versus <?") for various c lays shown r r n in Figure 2.5. M i tche l l suggests that e i t he r of the preceding hypotheses is p l au s ib le and that fur ther research w i l l be required to f u l l y determine the cause of the stress dependent behaviour of f r i c t i o n a l re s i s tance . 19. FIGURE 2.5 Shearing Resistance Versus Normal E f f e c t i v e Stress on the Shear Plane Raised to the Minus 1/3 Power (From Mi t che l1 , 1976) 20. CHAPTER I I I RESIDUAL STRENGTH TESTING METHODS 3.1 INTRODUCTION The major methods which have been used for res idual strength determination are the d i r ec t shear, t r i a x i a l and ring shear te s t s . The d i r ec t shear and t r i a x i a l apparatus, o r i g i n a l l y developed for the study of peak strength behaviour, are small displacement devices that have been adapted for use in determining residual s t rength. The r ing shear apparatus was designed s p e c i f i c a l l y for residual strength te s t i ng . Many of the features des i rab le in a res idual strength device are a lso genera l ly required for any drained shear apparatus. For e f f i c i e n c y , the so i l specimens should be simple to prepare and the test device should be easy to operate with a minimum of superv i s ion. The device should be v e r s a t i l e , o f f e r i n g a wide se lec t i on of appl ied s t r a in rates and normal loads. To f a c i l i t a t e pore pressure d i s s i p a t i o n in low permeable s o i l s ( f u l l y drained cond i t i on s ) , the a b i l i t y to test th in samples at very low s t r a i n rates would be necessary. Uniform stress and s t r a in condit ions along the shear plane are des i rab le for r e l i a b l e in terpre ta t ion of r e su l t s . The unique design requirement of a res idual strength device is the large shear displacement necessary to ensure that the shear plane is f u l l y developed and the res idual condi t ion has been achieved. The shear displacement should be un i -d i rect iona1 to model f i e l d cond i - t ions and to minimize disturbances on the shear plane which might increase the shearing res i s tance of the s o i l . 21 . 3 . 2 DIRECT SHEAR TEST The d i r ec t shear test is the most common method used for determining residual s t rength. L i t t l e or no mod i f i ca t ion of the equipment is required. Due to the short length of t rave l o f the shear box, large s t ra in s are accumulated e i the r by repeated reversa ls of the d i r e c t i o n of t ravel or by repeated u n i - d i r e c t i o n a l shears, repos i t ion ing the shear box at the end of each t r a v e l . Some researchers cut the shear plane p r i o r to shearing in an attempt to acce lerate development of the residual cond i t i on . The major advantages of the common shear box method app]ied to residual strength tes t ing would appear to be the ready a v a i l a b i l i t y of su i tab le equipment and the simple method of operat ion. Disadvantages include the inconvenience imposed by repeated repos i t ion ings or reversals of the shear box, the poss ib le inf luence on resu l t s of f l u c tua t i ng s t ress condit ions in the sample, and d i f f i c u l t i e s encountered in in terpret ing the test re su l t s . The stress f luc tuat ions occur in the sample during tes t ing due to the changing c ro s s - sec t iona l area of the shear plane and due to the a l t e rna t ing d i r e c t i o n of appl ied shear s tress when the d i r e c t i o n of t rave l is reversed. Considering that the compres- sion textures wi th in the shear zone are or iented r e l a t i v e to the major p r inc ipa l s t re s s , as described in Chapter II, and that a major change in the d i r e c t i o n of the major p r i nc ipa l s t ress occurs during a reversal of t r a v e l , shear box reversals would be expected to cause some phys ica l d i s rupt ion of the shear plane, poss ib ly delaying the development of residual cond i t ions . The d i f f i c u l t i e s in in terpre t ing test resu l t s appear to a r i s e from curvature or i n c l i n a t i o n of the shear p lane, as encountered 22. by Hermann and W o l f s k i l l (1966, p. 118) in d i r ec t shear tests on weak c lay sha les , or from premature termination of t e s t i ng . Based on a review of publ ished data, La Gatta (1970) suggests that termination of te s t i ng before the residual condi t ion has been adequately es tab l i shed is a common f a i l i n g of res idual strength studies employing the d i r ec t shear test method. Considerations which can resu l t in ear ly termination of te s t i ng include excessive sample loss through the loading frame gap or time l i m i t a t i o n s . Some of the problems encountered in d i r e c t shear te s t ing for residual strength during the course of th is thes is study are discussed in de ta i l in Chapter V. 3-3 TRIAXIAL TEST The t r i a x i a l test is more d i f f i c u l t to adapt for residual te s t ing than the d i rec t shear and, for th i s reason, has not been widely used. The major d i f f i c u l t y encountered in t r i a x i a l te s t ing for residual strength is the i n a b i l i t y to obtain large shear s t ra in s without severely d i s t o r t i n g the sample. Other disadvantages include the correct ions which must be made for the const ra in ing e f f e c t of the rubber membrane surrounding the sample, for the changing c ro s s - sec t iona l area of the f a i l u r e plane, and for the hor izonta l component of the ram load i f the loading cap is incapable of l a te ra l movement. In t r i a x i a l te s t ing for residual s t rength, because only l im i ted displacements can be achieved, methods must be employed to promote shear plane development at small s t r a i n s . Some success is claimed for t r i a x i a l tests on samples with pre-cut shear planes or 23. pre -ex i s t i n g f i s su res or iented at an angle of 45 + 0^/2 to the h o r i z o n t a l , where 0J . is the estimated angle o f residual f r i c t i o n (Chandler, 1966; Pet ley , 1966; Webb, 1969). Pet ley (1966) found that the displacement to the res idual cond i t ion could be s i g n i f i c a n t l y reduced by smoothing the pre-cut shear surfaces with a piece of glass drawn over the surfaces in the d i r e c t i o n o f shear. However, complicated test procedures and the uncerta inty of residual strength determinations obtained at small shear displacements make the t r i a x i a l test unsuited for normal p rac t i ca l use in th i s a p p l i c a t i o n . 3.4 RING SHEAR TEST The r ing shear test was developed s p e c i f i c a l l y for the purpose of determining residual s trength. The test is performed on an annular disc-shaped specimen with the normal load appl ied in the ax ia l d i r e c t i o n and the shear load appl ied tangentia 11y. The s o i l is sheared in a plane perpendicular to the axis by rotat ing the top of the specimen with respect to the base. The ring shear test has the f a c i l i t y for unl imited u n i - d i r e c - t iona l s t ra in s ( a l l s t ra ins are tangentia l in th i s to r s iona l test) with a constant so i l c ro s s - sec t i on . The most serious disadvantage of the te s t , other than the r e l a t i v e l y complicated const ruct ion o f the apparatus, would appear to be the va r i a t i on in s t r a i n rate across the shear plane which can a f f e c t pore pressure d i s s i p a t i o n . However, the v a r i a t i o n can be minimized by choosing appropriate sample dimensions and tests ind icate that res idual shear strength is i n sens i t i ve to small changes in s t r a i n rate (Townsend and G i l b e r t , 1974, p. 12). 2k. A var iant form of the r ing shear test is c a r r i ed out on s o l i d disc-shaped specimens. Despite theore t i ca l object ions to the large va r i a t i on in s t r a i n rate across the sample which ex i s t s in th i s type of t e s t , some reasonable resu l t s have been achieved using th i s method (La Gatta, 1970). Ring shear devices were i n i t i a l l y developed in the ear ly 1930's by a number of independent researchers inc lud ing Tiedemann, Gruner and H a e f e l i , Cooling and Smith, and Hvorslev (Hvorslev, 1939). The more successful devices incorporated upper and lower pa irs o f con f in ing rings pos i t ioned to cause f a i l u r e to occur wi th in the sample body away from the end platens. A l l of the ear ly devices were s t r e s s - c o n t r o l l e d . In 19^7, Hvorslev d i rected the development of a r ing shear machine that could be operated in e i ther a s t r e s s - c o n t r o l l e d or s t r a i n - c o n t r o l l e d mode (La Gatta, 1970, p. 10). With the exception of th i s work by Hvorslev, the evo lut ion of the r ing shear apparatus v i r t u a l l y ceased un t i l in teres t in residual strength was rekindled in the I960 's. Of several devices developed s ince the ear ly 1960's, the more s i g n i f i c a n t in terms of design advancement and innovation are the device developed at Harvard Univers i ty (La Gatta, 1970) and the device developed j o i n t l y by the Imperial College and the Norwegian Geotechnical Ins t i tute (Bishop et a l , 1971). The I.C.-N.G.I, machine tests th ick (3/k inch) annular specimens using s p l i t conf in ing r ings. Sample loss through squeezing between the conf in ing rings is regulated by a prec i se gap control mechanism. The Harvard machine can test annular or s o l i d disc-shaped specimens with e i t h e r s o l i d or s p l i t conf in ing r ings . The sample height is va r i ab le from 1 to 25 mm, but th in samples (2 to 3 mm 25. thick) are normally used. Both the Harvard and the I.C.-N.G.I, devices are s t r a i n - c o n t r o l l e d . Many of the ear l y devices are descr ibed in d e t a i l by Hvorslev (1939). The advantages and disadvantages o f most r ing shear devices constructed p r i o r to 1971 are reviewed by Bishop et al (1971). 26. CHAPTER IV THE UBC RING SHEAR DEVICE 4.1 INTRODUCTION The UBC r ing shear machine resu l t s from the author ' s thes i s project to design and develop a p r ac t i c a l residual strength apparatus. A f te r cons ider ing various options and a l t e r n a t i v e s , the r ing shear conf i gurat ion was accepted as the most advantageous approach to res idual strength determinat ion. The pre l iminary design was begun in May 1976, and construct ion was completed in January of the fo l lowing year. The UBC r ing shear device operates on the same bas ic p r i n c i p l e s as ear ly devices such as that designed by Hvorslev (1939) and more modern devices such as that developed j o i n t l y by the Norwegian Geotech- n ica l Ins t i tute and the Imperial Col lege (Bishop et a l , 1971). However, i t is hoped that the emphasis on operat ional s i m p l i c i t y and p r a c t i c a l i t y in the UBC design w i l l prove a useful contr ibut ion to the evo lut ion of residual strength test devices. 4.2 GENERAL DESCRIPTION A photograph of the UBC ring shear device is provided in Figure 4.1. A schematic sect ion of the device i den t i f y ing major components is presented in Figure 4.2. As shown in the sec t i on , the annular s o i l sample is contained at the sides between upper and lower pa irs of c o n f i n - ing rings and at the top and bottom between porous s t a in le s s s teel p latens. The lower conf in ing rings and porous platen are fastened to a turntable which is rotated at a chosen ra te . The upper conf in ing rings and porous  MAJOR COMPONENTS 1 Annular Soi1 Sample 2 Upper and lower outside conf in ing rings 3 Upper and lower ins ide conf in ing rings k Porous s ta in le s s steel platens 5 Turntable 6 Turntable base 7 Moment t rans fer arms 8 Moment-measuring force transducers 9 Confining r ing gap 10 A i r pi ston 11 Water Reservoi r FIGURE k.2 Major Components of UBC Ring Shear Device 29- platen are restra ined from rotat ing by moment-transfer arms which abut against two moment-measuring force transducers. The s o i l sample is thereby compelled to shear in the hor izonta l plane i n te r sec t i ng a small gap maintained between the upper and lower conf in ing r ings . The normal load, which is appl ied through the upper porous p la ten, is derived from and regulated by a i r pressure suppl ied to an a i r p i s ton . Sample d r a i n - age is provided through the porous platens which are connected to a water reservo i r encompassing the conf in ing r ings. The major machine parameters are as fo l lows: So i l Sample: Outside diameter 5-5 inches Inside diameter 3-5 inches I n i t i a l height Var iab le up to 0.75 inches Basal Drainage Path: Var iab le up to 0.25 inches Normal S t ress : Var iab le up to 200 psi with 100 psi house l i ne ai r pressure Rate of Shear (at center of shear p lane) : Smoothly va r i ab le -1 -4 1.5 x 10 to I.5 x 10 inches/minute or -3 -6 6.1 x 10 to 6.1 x 10 inches/minute with an opt ional gearbox 4.3 DETAILED DESCRIPTION 4.3.1 The Drive System Components of the turntable dr ive system, other than the motor and an opt iona l gearbox, are i d e n t i f i e d in a schematic sect ion of the UBC r ing shear device in Figure 4.3. The turntable is dr iven by a Motomatic E-550 MGHD d.c. servo motor with a heavy duty gearhead (manufactured by E l e c t r o - C r a f t Corporat ion, Hopkins, Minnesota). DRIVE SYSTEM 1 Turntable 2 Turntable alignment shaft 3 Rotary bearings provid ing l a tera l alignment h Thrust bearings providing v e r t i c a l support 5 Spur gear 6 Pinion gear 7 Chain dr ive to motor Other Components 8 Turntable base FIGURE 4.3 Drive System o 3 1 . With a master control un i t , the motor-gearhead combination provides a smoothly var i ab le output from 0.001 to I rpm. The motor-gearhead is connected by a chain dr ive to the pinion which d i r e c t l y dr ives the large spur gear o f the turntab le . When fur ther speed reductions are des i red , a small opt ional gearbox is used to reduce the output of the motor-gearhead. The turntable is supported v e r t i c a l l y by a s ing le large thrust bearing. Lateral alignment is provided along the hollow turntable alignment shaft by two rotary bearings set in the turntab le base. As previous ly mentioned, the dr ive system provides a smoothly var i ab le rate of shear at the center of the sample from 1.5 x 10 ' to -4 -3 -6 1.5 x 10 inches/minute, or from 6.1 x 10 to 6.1 x 10 inches/ minute with the opt ional gearbox. A performance test o f the dr ive system revealed smal l , gentle f luc tuat ions in turntable s t r a i n rate of -2.5%. As the period of tooth engagement equals the per iod of the s t r a i n rate f l u c t u a t i o n s , the behaviour is a t t r i bu ted to s l i g h t imperfections in the tooth design of the spur and pin ion gears. The magnitude of the f l u c t u a t i o n s , however, is considered too small to inf luence res idual strength measurements. 4.3.2 The Upper Ring Assembly A schematic sect ion of the UBC r ing shear device i den t i f y ing components o f the upper r ing assembly and suspension system is presented in Figure 4.4. The upper r ing assembly cons i s t s o f both the ins ide and outs ide upper conf in ing rings and the moment-transfer arms to which they are attached. The assembly is supported and a l igned by the UPPER RING ASSEMBLY 1 Outside upper conf in ing r ing 2 Inside upper conf in ing r ing 3 Confining ring support / moment t rans fer arms 4 Ro l ler bearings SUSPENSION SYSTEM 5 Center support shaft 6 Threaded cy l inder for adjust ing length of shaft and width of conf in ing r ing gap 7 Removable metal bar for c los ing conf in ing r ing gap 8 Loading bal1 9 Load c e l l to measure v e r t i c a l (s ide) f r i c t i o n on upper conf in ing rings 10 Bottom plate FIGURE k.h Upper Ring Assembly and Suspension System 33. center shaft . A threaded cy l i nder at the base of the center shaft is used to shorten or lengthen the sha f t , thereby c l o s ing or opening the gap between the upper and lower conf in ing r ings. During sample conso l idat ion the threaded cy l inder is t ightened against a removable metal bar, tens ioning the center shaft and f i rmly c l o s i ng the c o n f i n - ing r ing gap. This conf igurat ion is shown in Figure 4.4. During tes t ing the threaded cy l i nder is t ightened against the loading ba l l of the load c e l l res t ing on the bottom p late of the r ing shear machine unt i l the conf in ing r ing gap is jacked open the des ired amount. The gap width during tes t ing is t y p i c a l l y set at 0.002 to 0.004 inches. The load c e l l on the bottom p late measures s ide f r i c t i o n of the so i l sample and the upper porous platen on the upper conf in ing rings plus the weight of the upper r ing assembly. 4.3-3 The Loading System A schematic sect ion o f the UBC r ing shear device i den t i f y i n g components of the loading system is provided in Figure 4.5. The normal load is derived from a i r pressure suppl ied to a Bel lofram ro)1 ing- diaphragm a i r p i s ton. The piston shaf t , a l igned by a l i nea r ba l l bushing, t ransfers the load through a conical adapter, a loading cy l i nde r and the upper porous platen to the so i l sample. The loading cy l i nder is s l o t ted at 90 degree in terva l s to accommodate four support arms for the outs ide upper conf in ing r ing . Two of the conf in ing r ing support arms, shown in Figure 4.5, double as moment-transfer arms. The only contact between the support arms and the loading cy l i nde r is through two r o l l e r bearings which allow f r i c t i o n l e s s d i f f e r e n t i a l movement LOADING SYSTEM 1 A i r supply 2 Ro l l ing diaphragm a i r p iston 3 Piston shaft k Linear bal1 bushing 5 Conical adapter 6 Thrust bearing 7 Loading cy1inder 8 Loading plate - a s t i f f backing for porous piaten 9 Upper s ta in le s s s teel porous platen 10 Annular s o i l sample Other Components I I Ro l ler bearings 12 Outside upper conf in ing r ing 13 Confining ring support / moment t rans fer arms FIGURE k.S Loading System 35. between the components. Each component of the loading system from the a i r p iston to the upper porous platen is a l igned by the preceding component to produce a r i g i d , n o n - t i l t i n g un i t . The purpose of the carefu l alignment is to prevent t i l t i n g of the upper porous platen in the event of uneven s o i l compression. The n o n - t i l t i n g platen reduces the to lerance required between the upper platen and conf in ing rings to allow for smooth d i f f e r e n t i a l movement of components. If the sample compresses unevenly, the n o n - t i l t i n g platen w i l l cause a non-uniform stress d i s t r i b u t i o n in the sample. In such a case, the highly remolded s o i l in the f a i l u r e zone w i l l tend to flow p l a s t i c a l l y under the stress gradient unt i l near- uniform stress condit ions are re -e s tab l i shed . 4.3 -^ Measurement of Normal Loads The normal load on the f a i l u r e plane is ca l cu la ted by subt rac t - ing the s ide f r i c t i o n on the upper conf in ing rings from the tota l appl ied normal f o r ce . The tota l app l ied normal force is a pre-determined func- t ion of the a i r pressure suppl ied to the a i r p iston as measured by a pressure transducer. The s ide f r i c t i o n is ca l cu la ted by subtract ing the weight of the upper r ing assembly and suspension system from the force on the load c e l l at the base of the center shaft . 4.3-5 Measurement of Shear Forces Shear forces developed across the f a i l u r e plane during te s t ing are t rans ferred through the upper conf in ing rings and loading platen to the moment-transfer arms. Shear forces t rans fer red to the upper 36- conf in ing rings are passed d i r e c t l y to the moment-transfer arms to which the rings are attached. Shear forces t rans fer red to the loading platen are passed through the loading cy l i nder to r o l l e r bearings attached to the moment-transfer arms. A thrust bearing between the conical adapter and the loading c y l i n d e r prevents the t rans fer o f moments to the loading p i s ton . A pa i r of force transducers measure the moment ca r r i ed by the moment t rans fer arms. During t e s t i n g , the loads on the moment-measuring force transducers are maintained in balance, equal in magnitude but opposite in d i r e c t i o n . In th is con f i gu ra t i on , the forces form a near -per fect force couple, reducing s ide- loads on the center shaft to a minimum. Minimizing the shaft s ide - load helps maintain alignments and reduces loads on bearings and bushings. The loads on the force transducers are maintained in balance by occasional minor adjustments to the screw- mounted loading t ips of the transducers. The adjustments do not i n t e r - rupt tes t ing or the c o l l e c t i o n of data. The res idual f r i c t i o n angle is ca l cu la ted from the moment and normal load on the shear plane, both measured q u a n t i t i e s , by the fo l lowing equat ion, derived in Appendix I: 3M (r, + r 2 ) tan 0 r = 2W (r f + r , r 2 + r\) where M = sum of moments on shear plane W = tota l normal load on shear plane T| ^ ins ide radius of s o i l specimen r„ = outs ide radius of s o i l specimen 37. A study by the I. C. - N.G.I, group (Bishop et a l , 1971) has shown that the above equation w i l l not be in ser ious e r ror for any reasonable d i s t r i b u t i o n of shear stresses in the sample. 4.3-6 Displacement Measurements Displacements are measured using l i nea r va r i ab le displacement transformers (LVDT'S). Angular displacements of the turntab le are converted to v e r t i c a l displacements using a spur gear and screw arrange- ment, the re su l t ing v e r t i c a l displacements being measured by an LVDT. Conso l idat ion, sample loss and the width of the conf in ing r ing gap are d i r e c t l y measured by a s ing le LVDT suspended between the shaft of the a i r p iston and the moment t rans fer arm. These v e r t i c a l displacements can b e I dent i f i ed by the character and timing of the motion. Sudden changes in v e r t i c a l displacement resu l t from adjustments made to the conf in ing r ing gap. Very gradual changes in v e r t i c a l displacement t y p i c a l l y occur from the gradual loss of s o i l through the conf in ing r ing gap. Ve r t i c a l displacements that are nei ther sudden nor very gradual, and which fo l low a change in conf in ing pressure, are the resu l t o f sample conso l idat ion or swe l l . 4.4 SAMPLE PLACEMENT Undisturbed samples are prepared by cut t ing a maximum 3A inch th ick s lab of s o i l and trimming i t to the approximate dimensions of the annular sample des i red . The f i n a l trimming is done with a cu t t ing unit that is pushed into the s o i l . The cut t ing un i t , shown in Figure 4.6, cons i s t s of the upper conf in ing r ings , moment-transfer arms, loading cy l i nder and upper porous p la ten. The conf in ing rings FIGURE k.6 Sample Cutt ing Unit This un i t , composed of the upper conf in ing r ings , moment- t rans fer arms and loading c y l i n d e r , is used for f i na l trimming of undisturbed samples or as a mold for remolded samples. The un i t , complete with s o i l sample, is then mounted into the r ing shear device for t e s t i ng . 39. have sharp t ips and behave well as cutters provided the sample has been c a r e f u l l y trimmed to prevent excess ive bu i ld -up of s o i l at the cu t t ing t i p s . A f te r excess s o i l is removed, the cu t t ing unit is mounted in the r ing shear machine and the conf in ing r ing gap is f i rm ly c lo sed . The so i l sample is then gently pushed into pos i t i on in the lower c o n f i n - ing rings by applying a small downward force through the a i r p i s ton loading system. Remolded samples are placed in a s im i l a r manner except that the cu t t ing unit is used as a mold into which the remolded material i s pushed or spread. With minor adjustments, such as p lac ing a spacer beneath the lower porous p la ten, the minimum drainage path from the shear plane to the lower platen can be shortened from a maximum of \/k inch to any p rac t i c a l minimum. Shortened drainage paths are preferred for c lays with low permeabi l i ty to induce a more rapid d i s s i pa t i on of pore water pressures in the shear plane region. For samples which contain sand s ized p a r t i c l e s , the shear plane to platen separation should be s u f f i c i e n t to prevent sand grains lodged against the porous platen from i n t e r f e r i n g with natural shear plane development. The to ta l sample height can a lso be var ied from the maximum height of 3/k inch to any p rac t i c a l minimum. Thick samples are usual ly pre fer red fo r c lays and clayey s i l t s with r e l a t i v e l y high permeab i l i t i e s because the greater sample mass allows continued te s t ing for periods of weeks or months despite minor sample losses by squeezing through the conf in ing r ing gap. Thick samples are a lso required when porous platens 40. with r ibs are used. Although tests ind icate that f r i c t i o n of the s o i l against the platens and conf in ing rings is normally s u f f i c i e n t to ensure shearing w i l l occur in the body of the sample, ribbed porous platens are used when s o i l s l ippage adjacent to the platens is a concern. The ribbed platens contain 16 equal ly-spaced r ibs which protrude 1/16 inch into the sample over i t s en t i r e width. 4.5 DATA COLLECTION The UBC r ing shear machine is designed for use with an automatic e l e c t r o n i c data c o l l e c t i o n system and can be ea s i l y adapted for computerized data reduct ion. Although deta i l ed superv is ion is not required, i t has proved prudent to check the various e l e c t r o n i c and mechanical systems every day or two to ensure smooth operat ion. CHAPTER V TESTING PROGRAM 5.1 INTRODUCTION A te s t ing program was adopted ear l y in th is thes i s study to provide a means of determining appropriate tes t ing techniques and procedures and a means of eva luat ing the UBC Ring Shear Device. The program began in Ju ly 1976 with a ser ies of 5 s ing le stage m u l t i - reversal d i rec t shear tests undertaken to provide data for comparison and poss ib le c o r r e l a t i o n with r ing shear test re su l t s . These t e s t s , which var ied from 0.5 to 3 months in durat ion, were completed in the course of one year. Due to the need for de ta i l ed test superv i s ion , the d i r ec t shear equipment was operated only part - t ime and cumulative shear displacements did not exceed 6 inches for any of the 5 t e s t s . A desc r ip t i on of the d i r ec t shear apparatus used in th i s study is provided in Appendix II. Ring shear tes t ing began in mid-January 1977 fo l lowing completion of the new device. A ser ies of 11 r ing shear tests were performed over an 8 month per iod. The ser ies was comprised of 2 tests fo r machine performance and 9 tests for residual s t rength, inc lud ing several mult i - s tage tests in which residual strength determinations were made at 2 or more normal loads. In view of the experimental nature of the apparatus, a l l test r e su l t s , whether successful or inconc lus ive , are presented and discussed in th is chapter. kl. Three s o i l s having widely varying propert ies were included in the test program. Two o f the s o i l s , Haney Clay and Hat Creek Clay, were tested in both the r ing shear and d i rec t shear apparatus. The th i rd s o i l , Iranian Clay, was tested in the r ing shear apparatus on ly. This chapter presents the resu l t s of res idual strength te s t ing on these three s o i l s . To provide an order ly presentat ion of data, tests performed on each type of s o i l are discussed in separate sect ions . In each case, the d iscuss ion of test resu l t s is preceded by a de ta i l ed descr ip t ion of the s o i l tes ted. Plots and tables o f test resu l t s are a l so presented. The ring shear and d i r ec t shear te s t ing programs are summarized in Tables 5.1 and 5-2, re spec t i ve l y . The tests in each program are numbered according to the chronolog ica l order of te s t ing . P r i o r to the presentat ion and d iscuss ion of res idual strength test r e s u l t s , there is a d i scuss ion of machine performance tests c a r r i ed out ear ly in the r ing shear te s t ing program. 5.2 RING SHEAR PERFORMANCE TESTS Two r ing shear t e s t s , RS 1 (Ring Shear Test 1) and RS 3, were conducted to invest igate machine performance only. RS 1, the i n i t i a l test fo l lowing construct ion of the new device, es tab l i shed that the mechanical systems were funct ion ing as planned. The test a l so provided an opportunity to invest igate d e t a i l s o f sample preparat ion and the p r a c t i c a l techniques of t e s t i ng . However, the e l e c t r o n i c measurement systems were not f u l l y i n s t a l l e d at the time of te s t ing and no useful s o i l strength data was recorded. TABLE 5-1 Duration (days) Test So i l Consoli dation Sheari ng RS 1 H.C. 3 1 RS 2 H.C. - 22 RS 3 H.C. - 8 RS 4 H.C. 14 7 RS 5 H.C.C. 1 5 RS 6 H.C.C. 1 20 RS 7 H.C.C. 2 7 RS 8 H.C. 1 28 RS 9 L C . 1 4 RS 10 I.C. - 27 RS 11 I.C. - 14 H.C. - Haney Clay H. C.C. - Hat Creek Clay I. C. - I rani an CIay Schedule of Ring Shear Tests End of Test 1/18/77 2/12/77 3/18/77 4/11/77 5/15/77 6/7/77 6/16/77 7/16/77 7/22/77 8/18/77 9/18/77 No. of Stages 1 5 1 1 1 10 1 Total Di splacement (i nches) 23-5 12.7 11.1 6.8 44.0 24.8 28.1 3.1 20.1 50.7 Comments Machine Performance Test Pre-cut f a i l u r e plane Machine Performance Test So i l marked with model- l ing c lay wedges Test ended due to leaking 0 - r ing seal Test ended due to high side f r i c t i o n on conf in ing rings Test ended due to high side f r i c t i o n Test ended due to f au l ty ai r regulator TABLE 5.2 Schedule of Direct Shear Tests Test Soi 1 Test Duration End of No. of (days) Test Stages Total Displacement (i nches) Comments DS 1 DS 2 H.C.C. H.C.C. 17 55 3/14/76 11/9/76 3.3 DS 3 H.C. 87 2/28/77 5.3 DS 4 H.C.C. 79 5/20/77 3.1 Remo1ded, s i eved and reconsol idated sample DS 5 H.C. 50 7/19/77 5.3 Pre-cut f a i l u r e plane H.C.C. - Hat Creek Clay H.C. - Haney Clay -e- jr- The s o i l used in th is f i r s t test was remolded Haney Clay. Following a short 3 day test per iod , the sample was removed a/id examined. As expected, a wel1-developed shear plane was located with in the body o f the sample as def ined by the conf in ing r ing gap. As a resu l t of the successful operat ion and performance of the equipment in RS1, res idual strength te s t ing was begun immediately the rea f te r . RS3 was an inves t i ga t ion to determine i f any sample s l ippage was occurr ing adjacent to the porous p latens. A sample of undisturbed Haney Clay marked by 2 th in s l i c e s of coloured o i l - b a s e modell ing c lay was tes ted. The i n i t i a l pos i t ion of the modell ing c lay wedges was c l e a r l y i d e n t i f i e d on both upper and lower outs ide conf in ing rings p r i o r to t e s t i n g . The so i l was sheared to a to ta l displacement of about 12.5 inches. The normal load was maintained at about 75 p s i , except for a 16 hour period during which the house l i ne a i r pressure supplying the loading p iston was l o s t . During that per iod the normal load dropped to about 2 psi for a displacement of 2.6 inches. The test resu l t s are p lot ted in Figure 5-1 . At the end of the test i t was observed that no s l ippage had occurred in the upper conf in ing r ings, but that a l im i ted s l ippage of about 0.5 inches had occurred in the lower r ings. This s l ippage is be l ieved to have resulted from the d i s t r i b u t i o n of e f f e c t i v e normal s tresses wi th in the sample fo l lowing the drop in app l ied normal load. Due to the low permeabi l i ty of Haney Clay, the sudden reduction in appl ied normal load created negative pore water pressures wi th in the body of the sample approximately equal in magnitude to the Shearing Resistance o o O — 1 U o C D cz m CO CO CO "D 3 " fD fD 00 O CU 3 " — • -1 fD 3 — • CU fD 3 - i 3 CQ CT — • fD CO o n: -h cu — • — • in CU 3 n> r t n o ~< cu fD C L 3 3 n> o n fD — - — • fD 3 — OJ r t < < 3 i/i , — ^ io 3 • — cu 3 n -t CT • — 7r — • * — ^ CU fD W -< C L X I CU — n r t fD 3 " m e z 3 fD r t C L CD 1 fD l/> 73 O S ^1 ^1 73 H CU O r+ r+ (D CU O ~ -h Z o co -! 3 " 3 fD CU CU —1 -\ -o ^— -t X fD in in o c 1 -t vn fD — . . . 3 — • \ . 3 un 3 in • • 3 N> —> Ver t i ca l Displacement T o o ( in.) 47. drop in normal s t re s s . Therefore, the e f f e c t i v e s t ress wi th in the body of the sample remained v i r t u a l l y unchanged, f a l l i n g only gradual ly as the excess negative pore pressures d i s s i pa ted . However, due to the high permeabi l i ty of the p latens, the negative pore pressures at the s o i l - to -p laten contact d i s s ipa ted almost in s tant ly re su l t i ng in a sudden drop in e f f e c t i v e normal s t re s s . Hence, the sample was compelled to shear or ' s l i p ' at the p l a t e n - t o - s o i l in ter face where the e f f e c t i v e normal s tress and the re lated shearing res i s tance were very low. The s l ippage lasted only unt i l the negative pore pressures in the body of the sample d i s s ipa ted s u f f i c i e n t l y to reverse the balance of shearing res is tance between the platen and the estab l i shed f a i l u r e plane with in the sample. Slippage natura l l y occurred in the lower conf in ing rings because the upper conf in ing rings are deeper with a correspondingly greater contact area and f r i c t i o n a l res i s tance to s l ippage. Soi l s l ippage adjacent to the porous platens is not considered a serious problem with the UBC Ring Shear Device. However, as mentioned in a previous sec t i on , porous platens f i t t e d with metal r ibs to prevent so i l s l ippage are ava i l ab le for use should there be any doubt. A p lot of the shearing res i s tance versus cumulative shear d i s - placement fo r RS3 is presented in Figure 5.1. The res idual shearing res i s tance of about 0.05, indicated on the p l o t , corresponds to a residual f r i c t i o n angle of 3 degrees. However, a systematic experimental e r r o r , described in the d iscuss ion of test resu l t s for Haney Clay, e f f e c t i v e l y increased the measured normal stress by a constant value. The values of residual shearing res is tance and f r i c t i o n angle, as corrected for th i s systematic e r r o r , are 0.18 and 10.1 degrees, re spec t i ve l y . 48. This low f r i c t i o n angle is apparently that of the modell ing c lay used to mark the pos i t i on o f the sample. An examination of the sample a f t e r removal from the r ing shear apparatus revealed the shear plane was completely coated with a th in trans lucent layer of modell ing c l ay . This resu l t may be i n d i c a t i v e of how thin seams of weak material can s i g n i f i c a n t l y a f f e c t the strength behaviour of more competent s o i l s . 5.3 RESIDUAL STRENGTH TESTS ON HANEY CLAY 5-3-1 Soi1 Descri pt ion Haney Clay is a dark-grey, s en s i t i ve marine-deposited s i1 ty c lay . The undisturbed c lay has a f l o ccu l a ted s t ructure be l ieved to resu l t from the in teract ion of the so i l p a r t i c l e s with the s a l t so lut ions of i t s marine depos i t ional environment. However, th i s f l occu l a ted s t ructure is metastable because the clay has been sub- sequently u p l i f t e d above sea level and leached of s a l t by fresh water i n f i l t r a t i o n (Vaid and Campanella, 1977)- Due to the reduced s a l t concentrat ion, the clay p a r t i c l e s take up a more stable dispersed s t ructure upon remolding. The change from a f l occu la ted to a dispersed s t ructure is accompanied by a s i g n i f i c a n t loss in s o i l s t rength, as evidenced by the s e n s i t i v i t y of Haney Clay (Campanella and Gupta, 1969). For de ta i l ed information regarding the s t ructure and behaviour of s e n s i t i v e c l ay s , the reader is referred to Mi tche l l and Houston (1969) and Houston and Mitchel l (1969). *»9. The s o i l was block sampled from a deposit near the town o f Haney, B r i t i s h Columbia, some 20 miles east o f Vancouver. Due to the uniformity and a v a i l a b i l i t y of the mate r i a l , Haney Clay has been studied in numerous experimental works conducted at the Un ivers i ty o f B r i t i s h Columbia over a period in excess of 15 years. Typica l c h a r a c t e r i s t i c s and propert ies for Haney Clay are as fol lows (Byrne and Aok i , 1969; Bosdet, Lum and Negussey, 1976): S p e c i f i c g rav i ty 2.8 L iqu id l im i t 44% P l a s t i c i t y index 18% Natural water content 41% Percent f i n e r than 2 microns 46 Unconfined compressive strength 1550 psf S e n s i t i v i t y 12 Maximum past pressure 7000 psf -8 Permeabi l i ty 2 x 10 cm/s 5.3.2. Ring Shear Tests TEST PROCEDURES AND RESULTS Three ring shear tests for the purpose of determining res idual strength were performed on Haney Clay: RS 2, RS 4, and RS 8. In the case of RS 2, the shear plane was cut p r i o r to shearing. RS 2 and RS 8 were mult i - s tage tests in which residual strength determinations were obtained at 4 and 10 d i f f e r e n t normal loads, re spec t i ve l y . RS 4 was terminated prematurely fo l lowing the f i r s t stage of loading when a fau l ty o - r i ng seal caused water to leak from the reservo i r into the center column of the r ing shear apparatus. 50. The test data is presented on p lots o f shearing res i s tance and v e r t i c a l displacement versus cumulative shear displacement in Figures 5.2, 5-3 and 5.4. Ve r t i c a l displacement data was not recorded during RS 2, hence, th is p lot is omitted in Figure 5-2. The tabulated resu l t s of each stage of te s t ing are presented in Tables 5.3 and 5-4. The residual strength resu l t s are a l so summarized on a Mohr p lot in Figure 5-5- DISCUSSION AND INTERPRETATION OF RESULTS The peak strength behaviour of Haney Clay during r ing shear te s t ing was markedly d i f f e r e n t in RS 2 than in RS 4 and RS 8, as shown in Figures 5-2 to 5-4. RS 2 reached a peak shearing res i s tance of less than 0.5 at a displacement of 0.03 inches, corresponding to about 4 percent s t r a i n at the center of the sample. In cont ras t , RS 4 and RS 8 reached a peak shearing res i s tance of about 0.54 at a displacement and s t r a i n of about 0.4 inches and 50 percent, re spec t i ve ly . The reduced peak strength and peak strength displacement in RS 2 is a t t r i b u t e d to the e f f e c t of p re - cu t t i ng the shear plane before t e s t i n g . Beyond the peak strength displacement, the measured or apparent shear strength dropped temporari ly below the f i n a l recorded res idual strength in RS 2 and RS 8, and to a lesser degree in RS 4. Because the residual strength is the lowest drained strength of a s o i l , i t is evident that the s o i l was not f u l l y drained during th i s period and that excess pore pressures were in f luenc ing shear strength measurements. The development of excess pore pressures during the i n i t i a l stages of te s t ing is be l ieved to be re lated to the s e n s i t i v i t y of Haney Clay. During i n i t i a l shear ing, the metastable f l o ccu l a ted s t ructure Total Normal Pressure (1bs. /sq. in. ) T 1 T 8 10 12 Shear Displacement ( in.) 125 165 CO (0 ^ 0.4 -|- 0) <?0.3 H 56 S 0.2 20 22 i i r 24 26 — i 1 I 1 1 1 r- 28 30 32 3 4 Shear Displacement ( in.) 36 38 40 FIGURE 5.2 Shearing Resistance vs. Displacement - RS 2 Specimen: Undisturbed Haney Clay with Pre-cut Shear Plane Total Normal Pressure (1bs./sq. in.) 75 Rate of Shear (x 10 in./min.) 25 220 2 0.5i ro 2 0.4- U l ™ 0.3- S 0.2-1 JO to o.H t / (Tn Vert. Disp. 14 1 1 i r r- 6 8 10 Shear Displacement ( in.) 12 FIGURE 5.3 Shearing Resistance vs. Displacement - RS 4 Specimen: Undisturbed Haney Clay CO 7 3 m 7 3 CO C O Shearing Resistance o o o Shearing Resistance o o o o ro cr<1 o o CO -a zr fD fD CO S 3 o CU zr COT ~\ fD 3 — • 01 fD 3 -1 3 I Q 7 3 fD D is cr U l X I U J 3 Q - U l 01 C D — • r t o U l QJ fD r t 3 3 C O fD — - i fD 3 cr r t fD < C u — ^ U> • — I O - rr. 3 01 C 3 • 3 — • - — ' fD U l - < X I O QJ ce m e n t ay ce m e n t U J u i ui •XJ U J OV U J o O O N M U J ro U J O I O col rV 1- ON' OO zr oo f D . 0) -1 o Ul X I sr.° o fD 3 fD 3 r t — ro 3 1 I o O C O U ) o' I - r — O U l 1̂ 1—. S 3 tp ON I U l U J _ 3 U J ON ro ON cr —• o Ver t i ca l Displacement ( in.) o o • • • • U J ro —» o Ver t i c a l Displacement ( in.) TABLE 5.3 Test Results: RS 2 and RS 4 on Haney Clay Test Stage Normal Stress Shearing Resistance 0 ' r Shearing Rate Cumulative Displacement -4 RS 2 1 26 psi i n s u f f i c i e n t displacement- 2.9x10 in/min 2.38 inches 2 51 " .370 20.3° 5.6x10" 4 " 18.53 o -4 3 76 n .376 20.6 5.6x10 " 19.33 o -4 4 125 " .400 21.8 5-6x10 1 1 21.02 " 5 165 " .413 22.7° 5.6x10 _ l 4 1 1 23-51 Average 0'r = 21.3° RS 4 1 75 " .379 20.8° 2.2x10~ 3 1 1 11.08 " * i n s u f f i c i e n t displacement to achieve residual condit ion at this load. By : l i nea r regression analys i s for RS 2 and 4, 0 ' r = 23 -8°, C r = -4.4 psi VJ1 -E- TABLE 5 . 4 Test Results: RS 8 on Haney Clay Normal Stress Shearing Resistance 0 ' r Shearing Rate Cumulative Displacement 3 1 . 5 psi 26 123 91 66 4 0 32 130 124 98 . 4 4 6 . 4 3 8 . 4 3 5 . 4 4 5 . 4 4 0 . 4 29 . 4 3 2 .451 . 4 4 2 . 4 4 7 2 4 . 0 4 ° 1 . 2 x 1 0 ~ 3 in/min 23.65° 23-51° 23-99° 23-75° 23.22° 23.36° 24.28° 23.85° 5.7x10 6.2x10 6 . 2 x 1 0 6 . 2 x 1 0 6 . 2 x 1 0 6 . 2 x 1 0 6 . 2 x 1 0 6 . 2 x 1 0 24.08 6 . 2 x 1 0 9 . 8 1 inches 1 1 . 5 9 1 6 . 0 7 1 7 . 9 8 2 0 . 0 7 2 1 . 7 3 2 2 . 6 0 2 4 . 2 9 2 6 . 0 6 2 8 . 1 1 Average 0 ' r = 2 3 - 8 By l i nea r regression analys i s for RS 8 , 0 ' r = 2 4 . 1 , C r = - 0 . 7 psi Normal Stress (1bs./sq. in.) FIGURE 5.5 Mohr Rupture Diagram: Summary of Results of Residual Strength Tests on Haney Clay It is considered that the results of RS 2 and k, i f corrected for a c a l i b r a t i o n e r r o r , would be coinc ident with the results of RS 8. ON 57. of the sen s i t i ve c lay is broken down wi th in the shear plane by remolding and is replaced by a dispersed s t ructure with greater p a r a l l e l o r i en t a t i on of c lay p a r t i c l e s . Under constant volume cond i t ions , the new dispersed s t ructure is less capable of supporting loads and part of the normal s t ress must be t rans fer red to the pore water, c reat ing excess pore pressures (Mitchel1 and Houston, 1969). However, once the s t ruc tura l change is complete with in the shear plane, the natural d i s s i pa t i on of excess pore pressures re -es tab l i shes drained condit ions throughout the sample. Excess pore pressures can a lso be generated a f t e r the shear plane is f u l l y es tab l i shed i f the sample is sheared too rapid ly to allow f u l l d i s s i pa t i on of pore pressures. This e f f e c t was observed in RS 3 where, at a displacement of about 12 inches (see Figure 5-4), the rate of shear was increased 6 f o ld producing a rapid drop in measured shear strength due to the bui ld-up of pore pressures. A subsequent reduction in the rate of shear brought about quick d i s s i - pation of the excess pressures and return to equ i l i b r i um. Following establishment of the shear plane and d i s s i pa t i on of excess pore pressures, the shearing res i s tance recorded in RS 8 became v i r t u a l l y constant regardless of normal load, as shown in Figure 5.4. This is the an t i c ipa ted resu l t based on previous exper ience, as described in Chapter 2. However, the resu l t s of RS 2, tabulated in Table 5-3> record an increase in apparent shearing res i s tance with normal s t re s s . The s ing le residual strength determination made in RS k is in c lose agreement with one of the determinations made in RS 2. 58. The true re l a t i onsh ip between the resu l t s of these r ing shear tests is best i l l u s t r a t e d by the Mohr p l o t , Figure 5-5- As shown in the f i gu re , the resu l t s of RS 8 ind icate that Haney Clay has a residual angle of f r i c t i o n of 23.8 degrees and no residual cohesion. The resu l ts of RS 2 and RS h exh ib i t the same angle of f r i c t i o n , but the data is sh i f ted to ind icate a negative cohesion in tercept . The s h i f t in cohesion intercept is evidence of a constant and systematic experimental e r ro r . The e r ro r ended fo l lowing r e c a l i b r a t i o n of the transducers at the end of RS k, i nd i ca t ing that the source of the problem was a z e r o - s h i f t e r ror in transducer c a l i b r a t i o n . When corrected for th i s e r r o r , the resu l t s of RS 2 and RS k would be co inc ident with the resu l t s of RS 8, i nd i ca t ing that Haney Clay has a residual angle of f r i c t i o n of 23.8 degrees, with no residual cohesion. 59. 5.3-3 Di rec t Shear Tests TEST PROCEDURES AND RESULTS Two s i n g l e stage d i r e c t shear t es t s f o r the purpose o f de termin ing res idua l s t reng th were performed on Haney C l a y : DS 3 and DS 5, c a r r i e d out at a p p l i e d normal s t r e s s e s o f 34 and 87 p s i , r e s - p e c t i v e l y . Both samples were p laced as und is turbed specimens, but the shear p lane was cut p r i o r to t e s t i n g in DS 5- The t e s t s c o n s i s t e d o f a s e r i e s o f a l t e r n a t i n g forward and reverse t r ave rses o f the shear box, the length o f each t r a v e r s e not u s u a l l y exceeding 0.3 i nches . The shear s t r e s s versus d isp lacement behaviour o f the s o i l in the i n d i v i d u a l t r ave rses c o n s i s t e d o f v a r i a t i o n s o f the t y p i c a l shear behaviour o f a normal ly c o n s o l i d a t e d c l a y , i l l u s t r a t e d in F igure 2.1. The tes t r e s u l t s have been summarized in p l o t s o f the u l t ima te shear ing r e s i s t a n c e determined in each t r a v e r s e of the shear box versus cumula t ive shear d i sp lacement , presented in F igures 5-6 and 5-7. V e r t i c a l d isp lacement data i s a l s o shown on these f i g u r e s . The res idua l s t reng th r e s u l t s as i n t e r p r e t e d from t h i s data are presented on the Mohr p l o t , F igure 5-5- Tabulated r e s u l t s are presented in Table 5-5- DISCUSSION AND INTERPRETATION OF RESULTS The r e s u l t s o f d i r e c t shear t e s t s on Haney Clay i n d i c a t e that there i s a reduced d isplacement to peak s t reng th when the f a i l u r e p lane is cut p r i o r to s h e a r i n g , a l s o demonstrated in r i ng shear t e s t s . DS 3, c a r r i e d out on und is turbed s o i l , and DS 5, c a r r i e d out on a p re -cu t specimen, a t t a i n e d peak s t reng th at cumula t ive d isp lacements o f 0.6 and 0.1 i nches , r e s p e c t i v e l y . Shear Displacement ( inches) FIGURE 5.6 Shear ing Res is tance v s . Cumulat ive Displacement Specimen: Haney Clay To ta l Normal Pressure = 33-9 psir Typ ica l Rate o f Shear = 1 0 x 10 i n . / m i n . * U l t ima te va lue in each recorded t rave rse DS 3 ON o ' I 1 1 1 1 1 2 3 4 5 Shear Displacement (inches) FIGURE 5.7 Shearing Resistance vs. Cumulative Displacement - DS 5 Specimen: Haney Clay Total Normal Pressure: 87 -3 p s i Typica l Rate of Shear: 7 x 10 in./min. - Ultimate value of each recorded traverse TABLE 5-5 Test Results: Direct Shear Tests on Haney Clay T e s t Normal Stress Shearing Resistance 0'r apparent Shearing Rate Cumulative Displacement DS 3 33.85 psi 5-F .350 F 19.3° 1 x ]0~k in/min 5-3 inches R .412 R 22.4° 1 x 10 _ Z | DS 5 87.3 11 .310 17.2° 7 x 10~5 " 5.3 " F - value obtained in forward shear d i rec t i on R - value obtained in reverse shear d i rec t i on ON 63. DS 3 exh ib i ted a gradual f a l l in shearing res i s tance beyond peak strength displacement u n t i l , at a displacement of about 2.5 inches, the shearing res is tances measured in the forward and reverse d i rec t i ons d iverged, with the higher res i s tance being reg i s tered in the reverse d i r e c t i o n . This dichotomy in measured res i s tance was maintained through- out the remaining test per iod. As ind icated by the lack of divergence in the i n i t i a l test period and confirmed by subsequent examination, the divergence in res i s tance was not caused by an er ror in the c a l i b r a - t ion of the proving r ing used to measure shear fo rce . The d iverg ing res is tance phenomenon, a l so observed in tests on Hat Creek Clay, is apparently the resu l t of an i nc l i ned f a i l u r e sur face. An inc l i ned f a i l u r e surface is mechanically permiss ib le in d i rec t shear tes t ing because the freedom of movement of the upper load- ing frame includes v e r t i c a l and, to a lesser degree, rotat iona l motion. To analyze the e f f e c t of an inc l ined plane of f a i l u r e , the v e r t i c a l and hor izonta l appl ied loads have been broken down into components act ing p a r a l l e l and perpendicular to the i nc l i ned f a i l u r e surface in Figure 5.8. The resolved normal and shear forces as re lated by the f r i c t i o n angle can be descr ibed by two equations, one fo r each d i r ec t i on of t r a v e l , in two unknowns. The two unknowns, which can be determined from these equations are the angle of shear plane i n c l i n a - t ion and the angle of f r i c t i o n of the s o i l . The angle of f r i c t i o n is the residual angle provided drained residual condit ions have been es tab l i shed with in the shear plane. The equations re su l t ing from the preceding ana lys i s are presented in Figure 5.8. Toward the end of t e s t i n g , DS 3 exh ib i ted a gradual decrease in the angle of f r i c t i o n measured in the forward d i r e c t i o n and a gradual eh. Vsin0 Vcos0 H sin0 \" H xcos0 0 -9- V » 1 H 2 N S i n c l i n a t i o n of shear plane angle of f r i c t i o n for s o i l appl ied v e r t i c a l force appl ied hor izonta l force in forward d i r e c t i o n appl ied hor izonta l force when shear d i r ec t i on is reversed resolved forces normal to f a i l u r e plane resolved forces p a r a l l e l to f a i l u r e plane In terms of forces , the equation of s o i l strength is as fo l lows. S = N tan -6- (1) Resolve the appl ied forces shown above and subs t i tu te in equation (l) to ob ta in : (h^cos 0 - Vsin0) = (Vcos0 + H 1sin0) tan-8- (2) Reverse the d i r ec t i on of appl ied hor izonta l fo rce , resolve and subs t i tu te in equation (1) to ob ta in : (H 2cos0 + Vsin0) = (Vcos0 - h^s i n0) tan -9- (3) Combine equations (2) and (3), c o l l e c t and s imp l i f y terms, to determine the i n c l i n a t i o n of the shear plane. V(H 1 - H j tan 20 = ^ (h) V + H l H 2 The so i l f r i c t i o n angle is obtained by enter ing the determined angle of i n c l i n a t i o n in equations (2) or (3). FIGURE 5 - 8 Analys is of Direct Shear Results Influenced by an Incl ined Fa i l u re Surface 65- increase in the f r i c t i o n angle measured in the reverse d i r e c t i o n . The f i na l recorded values were about 19-3 and 22.4 degrees in the forward and reverse d i r e c t i o n s , re spec t i ve l y . By applying the ana lys i s described above, i t is ca l cu la ted that the true angle of f r i c t i o n at the end of tes t ing was 20.8 degrees, the d iverg ing recorded values re su l t ing from a shear plane i n c l i n a t i o n of 1.5 degrees. At the end of DS 3, the sample was removed and photographed. As seen from a view of the upper loading frame in Figure 5-9, the shear plane is well e s tab l i shed , h ighly po l i shed, and s1 i ckens ided. The shear plane is not f l a t but rounded, having protruded into the lower loading frame. The shape of the f a i l u r e plane may resu l t from the gradual loss of remolded s o i l through the gap between the loading frames. Lost s o i l is replaced by fresh material from the upper loading frame on ly , re su l t ing in a gradual eros ion of s o i l from the lower loading frame. The shear plane develops a rounded shape due to edge constra ints which prevent uniform so i l e ros ion. Herrman and W o l f s k i l l (1966, p. 120) encountered s im i l a r shaped shear planes in repeated d i rec t shear tests on weak c l ay - sha le s , noting that the apparent shearing res i s tance was increased by the i n c l i n a t i o n of the f a i l u r e sur face. However, as measurements were made in only one traverse d i r e c t i o n , they did not record the corresponding dec l ine in apparent shearing res i s tance in the opposite d i r e c t i o n of t r a v e l . Beyond peak strength displacement in DS 5, the shearing res i s tance f e l l to a temporary plateau ear ly in t e s t i n g , followed by a fur ther drop to a f i n a l p lateau. At the end of t e s t i n g , DS 5 exhib i ted an angle of f r i c t i o n of about 17.2 degrees with some minor sca t ter of data. FIGURE 5.9 Shear Plane at End of Tes t ing - DS 3 View of upper loading frame. Note the h ighly po l i s hed , s 1 i ckens ided sur face. The curvature of the shear plane may a l t e r the apparent shearing res i s tance of the s o i l . ON 67. 5.3.4 Review and Evaluat ion of Results The resu l t s of r ing shear tests on three samples of Haney Clay ind icate that th i s s o i l has a residual angle of f r i c t i o n of 2k degrees. The uniformity of the data obtained over a wide range of normal loads provides a strong degree of confidence in th i s determination of res idual f r i c t i o n angle. A z e r o - s h i f t c a l i b r a t i o n er ror during RS 2 and RS k produced data which, whi le f a l s e l y ind ica t ing a negative cohesion in tercept , is nevertheless in f u l l agreement with the determined angle of res idual f r i c t i o n . The resu l t s of the d i rec t shear tests c a r r i ed out on Haney Clay, DS 3 and DS 5, ind icate that Haney Clay has a residual f r i c t i o n angle of 21 and 17 degrees, re spec t i ve l y . A divergence in shearing res i s tance recorded in the forward and reverse traverse shear d i rec t ions observed in DS 5, is bel ieved to resu l t from the i n c l i n a t i o n of the shear sur face. As the shear plane i n c l i n a t i o n is smal l , in the order of I to 2 degrees, the v e r t i c a l motion caused by the i n c l i n a t i o n is masked by the downward movement due to sample losses which, in th is ser ies of t e s t s , is 1.5 to 3.0 times greater. The d iverg ing res is tance phenomena does not occur in r ing shear te s t ing because the conf in ing rings are not capable of v e r t i c a l movement or rotat iona l motion about a hor izonta l ax i s . These constra ints make shear plane i n c l i n a t i o n mechanically impermiss ible. For a va r ie ty of reasons the degree of confidence in the residual strength determinations made using the d i r ec t shear device is qu i te l im i ted . Factors in f luenc ing th i s judgement inc lude: I. S i g n i f i c a n t l y reduced tota l displacements in r e l a t i on to the large displacements obtained in the r ing shear apparatus. 68. 2. Moderately scattered data. 3- L imited number of res idual strength determinations (two) having c o n f l i c t i n g re su l t s . The strong degree of confidence in the resu l t s of r ing shear tests and the uniform nature of Haney Clay would suggest that the true res idual f r i c t i o n angle is 2k degrees, and that the values obtained in d i r ec t shear tests are too low. Barr ing errors in measurement, the source of the lowered values must be excess pore pressures wi th in the so i1 samples. The most probable source of the excess pore pressures would appear to be ingest ion of water into the shear plane during t e s t i n g . At the end of each t raverse, part of the sample is extended into the re servo i r . This mater i a l , exposed to water and carry ing no normal load, is f ree to swe l l . As shear displacement continues, the swelled so i l is drawn back into the f a i l u r e plane and again subjected to normal loading, generating excess pore pressures wi th in the swelled s o i l during the reconso l idat ion per iod. Repeated exposure of the ends of the f a i l - ure plane to water might be expected to fo s ter a gradual bu i ld -up of excess pore pressure in the sample. Other factors which might inf luence the pore pressures in the sample include the rate of shear, the rate of s o i l loss from the loading frame gap, and the length of pause between traverses - a longer pause a l lowing greater d i s s i pa t i on of pore pressures and greater swel l ing of exposed s o i l s . However, the author has found no apparent c o r r e l a t i o n in the ava i l ab le data between these parameters and the measured shear strength. 69. 5-4 RESIDUAL STRENGTH TESTS ON HAT CREEK CLAY 5.4.1 So i l Descr ipt ion Hat Creek Clay is a heav i ly overconso l idated, b r i t t l e , h ighly p l a s t i c , dark blue c lay . The s o i l was core sampled from extensive c lay deposits associated with coal measures in the Hat Creek Va l ley of B r i t i s h Columbia. A study performed at the Un ivers i ty of Western Ontario shows that the s o i l contains a high percentage of sodium montmori1 Ionite with var i ab le amounts o f carbonate, quartz and fe ldspar (Quigley, 1976). The same study presented the fo l lowing typ ica l prop- e r t i e s for the c lay : L iquid 1imit 109 to 178% P l a s t i c 1imit 34 to 39% Natural water content 35% Percent carbonate I to 68 The samples tested at UBC, contain ing a r e l a t i v e l y high percen- tage of s i l t and sand s ized p a r t i c l e s , had the fo l lowing c h a r a c t e r i s t i c s : L iqu id 1imi t 84 to 119% P l a s t i c 1imi t 27 to 28% P l a s t i c i t y index 68 to 92% Natural water content 29% Percent greater than 2 microns 40 to 50% Based on the resu l t s of a s t r a i n - c o n t r o l l e d conso l ida t ion t e s t , Hat Creek Clay has a permeabi l i ty in the order of 10 ' ° cm per second. 70. 5.4.2 Ring Shear Tests TEST PROCEDURES AND RESULTS Three s ing le stage r ing shear tests were performed on Hat Creek Clay: RS 5, RS 6 and RS 7> c a r r i ed out at appl ied normal s tresses of 26.5, 60 and 27 p s i , r e spec t i ve l y . The test data is presented in p lots of shearing res i s tance versus displacement in Figures 5.10 to 5-12. The f i n a l r e su l t s , tabulated in Table 5.6, are p lo t ted on a Mohr diagram in Figure 5-13- To f a c i l i t a t e sample placement, the hard and b r i t t l e clay was crushed by mortar and pest le before being placed for t e s t i ng . The s o i l sample for RS 5 was crushed and placed at the natural water content. For RS 6, the so i l was crushed at the natural water content, but l i g h t l y wetted as i t was p laced. The s o i l for RS 7 was crushed, wetted and remolded unt i l i t exh ib i ted a p l a s t i c consistency before being placed. To reduce the time required for pore pressure d i s s i pa t i on at the shear plane, the shortest drainage path to the f a i l u r e plane was reduced to 1/8 inch for th is test ser ies by p lac ing a spacer beneath the lower porous p la ten. Due to the large sand f r a c t i o n in the s o i l , further reductions in the drainage path would have increased the p robab i l i t y that sand grains lodged against the platen might protrude through the f a i l u r e plane, d i s rupt ing f a i l u r e plane develop- ment and increas ing the apparent shearing res i s tance of the s o i l . During tes t ing of Hat Creek Clay, sand p a r t i c l e s from the s o i l sample had a tendency to become wedged between the conf in ing r ings , hampering adjustment of the gap. The p a r t i c l e s were dis lodged Shearing Resistance o o o o o to UJ .e- Ul ON -L u i . 3 CO rr (D 0) O r t ru o -! 3 01 o- ro • in un ONI o I Ul I o t o o U l o o Ve r t i c a l Displacement ( in.) o c z 73 co oo T3 3" CD CD O CD — -1 3 — CD 3 3 IQ 73 CD 1/1 OJ - . rt Ul rt- O Ol -1 3 CD O CD CD 7T < O OJ -< o 01 n CD 3 CD 3 70 OO Shearing Resistance Shearing Resistance o o T V o o -p- un J L_ I co UJ o U J N> U J OJ <^ o CD 3 CD 3 rt- ^ U J — CO 3 O -C- - f O N o O N o O i un u i O CD T3 U0 P> - „ ^ o o ON-I co-I CD -I o 0) o CD 3 CD 3 rt- — K> 3 O N oo O U J U l O N o o o U l O N o Si T D 1) 1 0 I U l V e r t i c a l Displacement ( in. ) Ve r t i ca l Displacement ( in.) •ZL Shearing Resistance Shearing Resistance o o o o o o o oo .e- u i f-o oo .t- u i ro —» o t-o —* o Ver t i ca l Displacement ( in.) Ve r t i c a l Displacement ( in.) 'IL TABLE 5.6 Test Results : Ring Shear Tests on Hat Creek Clay Test Stage Normal Stress Shearing Resistance 0'r Shearing Rate Cumulative Displacement R S 5 1 26.5 psi .kkS 2k.0° 5.6x10~Z' in/min 6.84 inches R S 6 1 60 " .382 20.9° 1.9x10_Zt " 43.98 " RS 7 1 27 1 1 .461 24.8° 2.9x10 - 3 " 30.33 -c- cr in Xi 1/1 Q) 1_ fD 20 i 10 J 10 ^ 2 4 . 8 ° DS 4 DS 1 RS 6 —1 1 1 1 1~~ 20 30 40 50 60 Normal Stress (1bs./sq. in.) DS 2 9.4° "~r~ 70 FIGURE 5.13 Mohr Rupture Diagram: Summary of Results of Residual Strength Tests on Hat Creek Clay 76. by clamping the rings shut whi le shearing the sample. The problem was e l iminated by maintaining a very narrow conf in ing r ing gap. Due to the swel l ing nature of Hat Creek Clay, material lost through the conf in ing r ing gap increased in volume producing the appear- ance of excess ive sample losses. However, the actual rate of sample loss as measured by the change in sample thickness was minor. DISCUSSION AND INTERPRETATION OF RESULTS At the end of the f i r s t r ing shear test on Hat Creek Clay, RS S, the apparent f r i c t i o n angle (neglect ing poss ib le excess pore pressures) was 2k degrees. As shown in Figure 5-10, the shearing res i s tance was s t a b i l i z i n g when the test was ended such that s i g n i f i c a n t changes in shearing res i s tance would not an t i c ipa ted i f te s t ing had been extended to any reasonable displacement and durat ion. Considering that the accepted values of res idual f r i c t i o n angle for montmori1 Ionite clay ranges from k to 10 degrees (see Table 2.2), the shearing res i s tance encountered in tes t ing was higher than expected. S imi lar resu l t s were obtained during RS 6 and RS 7 which, at the end of t e s t i n g , exh ib i ted apparent res idual f r i c t i o n angles of 20.9 and 24.8 degrees, re spec t i ve l y . The high values of s o i l shearing res i s tance are a t t r i bu ted to shear plane d i s rupt ions caused by sand p a r t i c l e s in the s o i l . The d i s rupt ions to shear plane development may have been i n t e n s i f i e d by the shortened drainage path should the 1/8 inch separat ion between porous platen and shear plane have been i n s u f f i c i e n t to prevent i n t e r - act ion between sand p a r t i c l e s lodged against the platen and the shear 7 7 . plane. The inf luence of sand and s i l t content on the shearing r e s i s t - ance of several d i f f e r e n t c lays is i l l u s t r a t e d in Figure 5 -1^ . As shown in the f i g u r e , the curves for montmori1 Ionite c lay ind icate that Hat Creek Clay, having a sand and s i l t f r a c t i o n of kO to 50 percent, would be expected to exh ib i t a shearing res i s tance s i m i l a r to that o f the pure c lay . However, the continual loss o f f ines through the con- f i n i n g r ing gap during te s t i ng may have concentrated the coarser p a r t i c l e s in the shear plane, increas ing the shearing res i s tance s i gni f i cant ly . The samples appear to have been we l l -d ra ined at the end of t e s t i n g , showing l i t t l e or no va r i a t i on in measured shearing res i s tance with increas ing displacement or with changes in rate of shear. Further evidence that the samples were we l l -dra ined is provided by the s i m i l a r i t y of resu l t s obtained from RS 5, where the s o i l was placed at i t s natural water content, and RS 7, where the s o i l was wetted and remolded p r i o r to placement. Very high values of s ide f r i c t i o n ( f r i c t i o n a l forces of the loading platen and the so i l against the upper conf in ing rings) were en- countered in RS 5- In previous te s t s , the measured s ide f r i c t i o n had been in the order of 10 to 20 percent of the tota l appl ied load. In RS 5 , the s ide f r i c t i o n began at a very low value but s tead i l y increased during the te s t . When the side f r i c t i o n exceeded 30 percent of the appl ied load, the test was terminated to prevent poss ib le damage to the equipment. The high values of s ide f r i c t i o n are a t t r i bu ted to corros ion welts on the upper conf in ing r ings, but could a l so resu l t from misalignment of machine components during assembly or from granular s o i l p a r t i c l e s wedged between the conf in ing rings and the upper loading p la ten. 78. Residual Strength Envelope A. a f t e r Skempton, 1964 B e s t - f i t Relat ions f o r : B. Viennese Clay a f t e r Borowicka, 1965 C. Na - Montmori11 on i te ~~20 ' 40 ' 60 ' 80 ' ^00 Clay Fract ion (percent < 2 microns) FIGURE 5-14 Relat ion Between Clay Content and Residual Shearing Resistance for Various So i l s 79. 5.4.3 D i rect Shear Tests TEST PROCEDURES AND RESULTS Three s ing le stage d i r ec t shear tests for the purpose of determining res idual strength were performed on Hat Creek Clay: DS 1, ! DS 2, and DS 4, c a r r i ed out at appl ied normal s tresses o f 52, 78.5 and 44.5 p s i , re spec t i ve l y . To prepare the sample for DS 1, the s o i l was broken and ground with mortar and pe s t l e , and subsequently placed in l i f t s in the d i r ec t shear box. A f te r each l i f t was p laced, the loading block was l a i d on top and hammered f i rmly to compact the so i l The sample was made as deep as poss ib le to allow for l a te r conso l idat ion and loss o f sample by squeezing through the loading frame gap. The sample for DS 2 was placed in a s i m i l a r manner, except that each l i f t of s o i l was spr ink led with d i s t i l l e d water to moisten the sample. The sample for DS 4 was wetted, remolded, and s ieved to remove the medium to coarse sand f r a c t i o n . The remaining s o i l was reconsol idated under s t r a i n - c o n t r o l l e d cond i t ions , trimmed and placed in the shear box. The test resu l t s have been summarized in p lots of the ult imate shearing res i s tance determined in each traverse of the shear box versus cumulative displacement, Figures 5.15 to 5-17- V e r t i c a l displacement data is a l so shown on these f i gu res . The res idual strength resu l t s interpreted from the data are presented on a Mohr p lo t in Figure 5-13- Tabulated resu l t s are presented on Table 5-7- o cz 73 m un un —i -H CO CO -< O T3 3" CO T J r i - fD fD 3" CZ — ' al O o> fD —J o — ' —• Ol r i - OJ 3 —• ~i —. ' (D 3 3 o 3 IO O OJ 73 -! • - —• r t OJ 3 73 ui fD r t OJ fD T J fD — ' 3: Ul — ' < OJ —• Ol OI O r t Ul o — ' - h -1 r t fD C fD o 01 3 CD OO </l 3 fD 3" Ul CD n 3 —• fD c fD fD r t 0) -1 7T CD < —^ fD .. O Ul —• 01 • 3 O o ZT U l U l -< o 3" o c fD ~\ • 3 Ul fD X oo c O — ' o —1 T J Ol —\ o Ul r t O- 1 — —• fD U l < ta- fD r t 3 C3 -1 • —• OJ Ul < 3 T J fD — • — - I Ul • O fD fD 3 fD 3 CT CO U n J Shearing Resistance* o o o M UO -C" I I L_ i o i o 1 o I o N> —> O O O un 3 3 fD fD O - C L O - I OJ -i O - (D fD QJ OJ - i -1 -1 -1 fD fD o o o o 3 3 V e r t i c a l Displacement (inches) '08 1 1 — i T r 1 2 3 5 Shear Displacement (inches) FIGURE 5.16 Shearing Resistance vs. Cumulative Displacement - DS 2 Specimen: Hat Creek Clay Total Normal Pressure: 78.4 psi.,- co Typica l Rate of Shear: 25 x 10 in./min. 7* * Ult imate value in each recorded traverse Shear Displacement (inches) FIGURE 5-17 Shearing Resistance vs. Cumulative Displacement - DS h Specimen: Hat Creek Clay Total Normal Pressure: hh.k p s i Typica l Rate of Shear: k x 10 in. /min. » Ultimate value in each recorded traverse TABLE 5.7 Test Results: Direct Shear Tests on Hat Creek Clay Test Normal Stress Shearing Res i s tance- 0' apparent--* Shearing Rate Cumulative Displacement o -4 DS 1 51.8 psi F .166 3-h 5.0x10 in/min 3-28 inches DS 2 78.4 " F .198 11.2° 2.5x10 _ Z t " S.hk DS k hh.k " S .290 16.2° 3-7x10" 5 " 3.10 F - value obtained in forward shear d i r ec t i on S - 'average' value of scattered results Tests ended prematurely due to cumulative sample losses, Residual condi t ions not achieved. 00 00 84. The shearing res i s tance versus displacement p lot for DS 1, Figure 5-15, does not show data for a rapid pre-shear ing period ca r r i ed out p r i o r to regular t e s t i ng . This sample was pre-sheared for a period of 5 consecutive test days during which the measured shear angle remained constant at 31 degrees. A f te r a three day pause in te s t ing to al low for equa l i za t i on of pore pressures, the measured shear angle f e l l to 21.7 degrees. Subsequent test resu l t s are summarized in Figure 5-15. STRAIN CONTROLLED CONSOLIDATION TEST As previous ly d iscussed, s o i l l a te r tested in DS 4 was remolded, s ieved to remove the medium to coarse sand f r a c t i o n , and reconsol idated in a s t r a in con t ro l l ed tes t . During reconso l i da t i on , free drainage was permitted at the top of the sample on ly , while pore pressure was monitored at the sample base. The basal pore pressures remained in the order of 90 to 100 percent of the appl ied s t ress throughout the test due to the extremely low permeabi l i ty of the s o i l . The appl ied s t r a i n rate was 2.4 x 10 ^ inches per minute, the lowest rate of which the test apparatus was capable. Although the assumptions used to analyze the resu l t s of the s t r a i n con t ro l l ed conso l idat ion test require that the basal pore pressure does not great ly exceed 10 percent of the appl ied stress (Byrne and Aok i , 1 969 ) , the resu l t s were nevertheless s u f f i c i e n t to demonstrate that the permeabi l i ty of Hat Creek Clay must be in the order of 10 ^ cm per second or less . This value is in agreement with that estimated by Quigley (1976). 85. DISCUSSION AND INTERPRETATION OF RESULTS As described above, DS 1 exh ib i ted high shear res i s tance over a 5 day rapid displacement preshearing period not recorded on the data p lo t s . It is l i k e l y that the o r i g i n a l dry s tate of the s o i l combined with i t s low permeabi l i ty resu l ted in the i n i t i a l te s t ing being done on e s s e n t i a l l y dry s o i l . The high shear angle measured during preshearing (31 degrees) is not unreasonable for a crushed dry c l ay . The subsequent dec l ine in the measured shearing re s i s tance , shown in Figure 5-15, would ind icate that water has reached the shear plane. The low permeabi l i ty of the s o i l appears to preclude the p o s s i b i l i t y of drainage from the porous base so ear ly in t e s t i n g . However, water may have been ingested into the shear plane as the ends of the s o i l sample were a l t e r n a t e l y thrust into the water reservo i r and drawn back into the shear plane during t e s t i n g . This mechanism, previous ly deta i l ed in the review of test resu l t s on Haney Clay, could expla in the drop in shear strength of the s o i l over a r e l a t i v e l y short period of time. At the completion of DS 1, the apparent angle of f r i c t i o n was about 9-5 degrees and decreasing. In DS 2, the apparent angle of f r i c t i o n remained approximately constant at 15 degrees throughout most of the test per iod , as shown in Figure 5-16. However, in the las t few reversals the shear angle dec l ined to about 11 degrees. In DS k, performed on reconsol idated Hat Creek Clay, the shearing res i s tance f e l l gradual ly at the s t a r t of the test before l e v e l l i n g out at an angle of f r i c t i o n of about 16 degrees. In each case, tes t ing was ended when cumulative sample loss made fur ther te s t ing imprac t i ca l . 86. As encountered in d i r ec t shear tests on Haney Clay, the tests on Hat Creek Clay general ly exh ib i ted greater shear res i s tance when tested in one d i r e c t i o n than in the other . This divergence of recorded res i s tance is a t t r i bu ted to an i n c l i n a t i o n of the f a i l u r e surface. By applying the ana lys i s summarized in Figure 5.8, i t is ca l cu la ted that the shear planes in DS 1, DS 2 and DS A had i n c l i n - at ions of about 2.2, 1.1 and 0.9 degrees, re spec t i ve l y , near the end of t e s t i n g . A l l of the d i r ec t shear tests on Hat Creek Clay exh ib i ted a genera l ly constant shearing res i s tance in the ear ly stages of t e s t i n g . In DS 1, the shearing res i s tance remained v i r t u a l l y constant throughout a 5 day rapid displacement preshearing period at a value corresponding to an apparent angle of f r i c t i o n of 31 degrees. In DS 2 and DS k, a constant value of 15 to 16 degrees was a t ta ined a f t e r a displacement of about 1.5 inches. The shearing res i s tance plateau may resu l t from shear plane d i s rupt ions caused by the high percentage of sand and s i l t in the s o i l . The removal of the coarse sand f r a c t i on in DS k, leaving the f ine sand and s i l t in the sample, appears to have had l i t t l e e f f e c t on th i s behaviour. From subsequent reductions in shearing res i s tance recorded in DS 1 and DS 2, i t would appear that the in ter ference to shear plane development suggested by the res i s tance plateau can be overcome by increased displacements. As prev ious ly d i scussed, the higher shearing res i s tance plateau recorded in DS 1 is a t t r i bu ted to the low i n i t i a l water content of the s o i l sample. 87. The resu l t s o f d i r ec t shear tests on Hat Creek Clay f a i l to adequately def ine the residual strength of the s o i l . Residual condit ions were not at ta ined in DS 1 and 2, as indicated by the decreasing shearing res i s tance at the end of te s t ing . Comparing the resu l t s of DS 4 (Figure 5.17) to DS 2 (Figure 5-16), i t appears that fur ther decreases in shearing res is tance would be expected in DS 4 i f the test had continued and that th i s test a l so was ended prematurely. Considering the low permeabi l i ty o f the s o i l , the length of the drainage path and the poss ib le e f f ec t s of water ingested into the shear plane during shearing, i t is probable that the so i l was not f u l l y drained during d i r ec t shear tests on Hat Creek Clay. 5.4.4 Review and Evaluation of Results The resu l t s of ring shear tests on Hat Creek Clay ind icate that th is s o i l has a residual angle of f r i c t i o n in the order of 21 to 25 degrees. The r ing shear samples appear to have been wel1-drained, showing l i t t l e or no va r i a t i on in measured shearing res i s tance with increased displacement or with changes in the rate of shear. However, the maintenance of a very narrow conf in ing r ing gap, which allows the clay f ines to escape but reta ins the sand f r a c t i o n , may have resul ted in a concentrat ion of the coarse f r a c t i o n in the shear plane, increas ing the shearing res i s tance of the s o i l . The d i r ec t shear tests on Hat Creek Clay were ended due to cumulative sample loss before residual condit ions could be e s tab l i shed . The rate of sample loss was very high, genera l ly in the order of 0.1 inch loss ( ve r t i c a l displacement) per inch of t r a v e l , some 15 to 60 times greater than sample losses incurred in r ing shear te s t s . It would appear that the d i rec t shear tests on Hat Creek Clay were not f u l l y drained. Considering that 10 to 20 percent of the shear plane is exposed to water at each end of the d i r ec t shear traverse and that montmori1lonit ic c lays have a low permeabi l i ty combined with a high capacity to swell when exposed to water, the potent ia l for the development of excess pore pressures by ingest ion of water into the shear plane is much greater with Hat Creek Clay than with most other s o i l s . The sand content and low permeabi l i ty of Hat Creek Clay pose formidable problems in tes t ing for residual s trength. To ensure that the v i r t u a l l y impermeable s o i l is dra ined, i t is necessary to reduce the thickness of the sample and/or the rate of displacement. As any s i g n i f i c a n t reduction in the rate of shear great ly increases the time required to achieve the large displacements needed in res idual strength te s t i ng , reduced sample thickness is the preferred opt ion. However, attempts to test thin samples of c lay contain ing sand may be f u t i l e i f sand grains lodged against the porous platens protrude through or otherwise d i srupt the shear plane. Tests may a l s o be i n f l uenced by s e l e c t i v e l oss o f f i n e s from the sample which would a l t e r the s o i l composi t ion and inc rease the measured shear ing res i s t ance . 5-5 RESIDUAL STRENGTH TESTS ON IRANIAN CLAY 5.5-1 Soi1 Descr i p t i on I ran ian Clay i s a homogeneous, l i g h t brown, c layey s i l t . The s o i l , which was chunk sampled from a Middle Eastern deser t a r e a , i s hard and b r i t t l e and has a very low na tu ra l water con ten t . Typ i ca l c h a r a c t e r i s t i c s and p r o p e r t i e s f o r I ran ian Clay are as f o l l ows (Lum and Negussey, 1977): S p e c i f i c g r a v i t y 2.78 L i q u i d 1imi t kl% P l a s t i c i t y index 18% 90. Percent s i l t 70 to 75 Percent c lay 18 to 30 Iranian Clay has a strong tendency to f l o c c u l a t e such that a suspension of the s o i l w i l l s e t t l e wi th in minutes unless mixed with a de f l occu l an t . This tendency to f l o c c u l a t e may have s i g n i f i c a n t l y reduced the apparent c lay content, tabulated above, as determined by hydrometer ana ly s i s . The agency which provided the s o i l has ind icated that the c lay port ion of the s o i l may be large ly composed of a t t a p u l g i t e . 5-5-2 Ring Shear Tests Three r ing shear tests were performed on Iranian Clay: RS 9, RS 10 and RS 11. Shearing res i s tance versus displacement p lots for these tests are presented in Figures 5-18, 5-19 and 5.20. The three samples were prepared by adding water and remolding the s o i l p r i o r to pi acement. The i n i t i a l test o f the s e r i e s , RS 9, exh ib i ted very high values of s ide f r i c t i o n on the wal ls o f the upper con f in ing r ings. Short ly a f t e r the s t a r t of t e s t i n g , the tota l load appl ied through the a i r p iston was lowered in an attempt to reduce the s ide f r i c t i o n . However, s ide f r i c t i o n remained high, in the order of 37 percent of the tota l appl ied load, and the te s t ing was ended to prevent poss ib le damage to the equipment. RS 10 was marred by f luc tuat ions in appl ied normal pressure which were caused by a f au l t in the a i r pressure regulator c o n t r o l l i n g the loading p i s ton . Due to the low permeabi l i ty of the s o i l and the short period of the pressure o s c i l l a t i o n s , undrained condit ions ex i s ted 0.7 S 0.6 c (0 ? 0.5 in <U * 0.4 01 c 0.3 <u oo 0.2 0.1 J 125 1 50 1 Rate of Shear (x 10 in./min.) 57 1 ' P - O - - */<Tn ^ ^.Vert. Di sp. i i i i r i i 1 1 -1 T " I I i 6 8 10 Shear Displacement ( in.) 12 c 0J E 0) o ro CL in \~ 0.0 r- ro o 0J h-0.1 14 FIGURE 5.18 Shearing Resistance vs. Displacement - RS 9 Specimen: Iranian Clay Total Normal Pressure (1bs./sq. in.) 60 1 30 Rate of Shear (x 10 in./min.) . 53 1 **3 1 i — i 1 1— 1 1 1 1 1 1— i i 1 1 1 ' I r 2 k 6 8 10 12 14 16 18 20 Shear Displacement ( in.) FIGURE 5-19 Shearing Resistance vs. Displacement - Specimen: Remolded Iranian Clay • RS 10 73 m o co co T J 3 - CD CD O Q> — -1 3 — CD 3 3 tCI 73 CD — U l -I — QJ U l 3 r t - . cu QJ 3 3 O CD o — < QJ U l U l T3 QJ n CD 3 CD 3 73 co Shearing Resistance o o o o O J -tr- on O N _L I 1 L Shearing Resistance o o o o oo o' oo oo CD m ui •o CU o CD 3 CD -C- 3 CO on o on on -e- on ON 31 T J co o CO zr CD QJ —' ~ J o ui I on 3 -o -i CD Ul in c -1 CD C O 3 - CD CO- QJ -I ui •a QJ o CD 3 CD 3 r t ON T r o co ro o C O o 0 ( on 3 3 ^ O O o o Ver t i c a l Displacement ( in.) Ve r t i c a l Displacement ( in.) •£6 94. in the shear plane region of the sample such that f l uc tua t ions in the e f f e c t i v e normal s tress were smal l . This is demonstrated by the constancy of the measured shear strength which is a funct ion of the e f f e c t i v e normal s t re s s . A p lot of tota l appl ied normal s tress and measured shear s tress versus displacement is provided in Figure 5.21. In r ing shear t e s t i n g , the test procedures are designed to create drained condit ions in the sample such that the to ta l and e f f e c t i v e normal stresses are equal. This permits c a l c u l a t i o n of shearing res i s tance, the r a t i o of the e f f e c t i v e shear s tress to the e f f e c t i v e normal s t re s s , without monitoring pore pressure. When undrained condit ions e x i s t , the e f f e c t i v e normal s tress is unknown and the parameter ca l cu la ted as shearing res i s tance is only the r a t i o of the shear s tress to the tota l normal s t re s s . This is the parameter shown on the shearing res i s tance p lot for RS 10, Figure 5-19- Despite the pressure o s c i l l a t i o n s , the e f f e c t i v e normal s tress wi th in the sample can be estimated by the average tota l normal s t res s . Using th i s approach, the resu l ts of RS 10 ind icate that Iranian Clay has a res idual angle of f r i c t i o n of about 29 degrees, measured at an average appl ied normal stress of about 62.5 p s i . The f i n a l test of the ser ies on Iranian Clay, RS 11, was hampered by a f a i l u r e in the automatic data recording system. Data was acquired by manual operat ion of the equipment, reducing the quantity of data ava i l ab le for study. However, the resu l t s of th is t e s t , p lo t ted in Figure 5.20, are cons i s tent , i nd i ca t ing that Iranian U> -O in i n OJ 70 H 60 A so A ro LQ c ro 30 H fj 20 10 H ro O / / Normal Stress Shear Stress 9 . 0 1 0 . 0 Shear Displacement (inches) 11.0 FIGURE 5 .21 Total Normal and Shear Stress vs. Displacement - RS 10 The o s c i l l a t i o n s in tota l appl ied normal stress occurred as a resu l t o f a f au l ty a i r pressure regulator. The constancy in shear s tress ind icates that, despite tota l normal stress f l u c tua t i on s , the e f f e c t stress on the shear plane remained nearly constant due to undrained condit ions within the so i l sample. 96. Clay has a residual f r i c t i o n angle of about 27.8 degrees. RS 11 was conducted at a normal s t ress of 75 p s i . The resu l ts of the ring shear tests on Iranian Clay, tabulated in Table 5.8, are shown on a Mohr diagram in Figure 5-22. These resu l t s are in c lose agreement with the accepted residual f r i c t i o n angle of a t t apu l g i t e , 30 degrees, as tabulated in Tables 2. I and 2.2. TABLE 5.8 Test Results : Ring Shear Tests on Iranian Clay Test Stage Normal Stress Shearing Resistance 0'r Shearing Rate Cumulative Displacement - 4 RS 9 1 125 psi Test incomplete 5-7x10 in/min 1.40 inches - 4 2 60 " Test incomplete 5-7x10 " 3-13 " _/, RS 10 1 60 " Scatter due to 4.3x10 11 1 4 . 4 0 " load f luc tuat ions ( - average .555) (29°) - 4 2 30 " Test incomplete 4.3x10 " 20.12 " RS 11 1 75 " -529 27-9° 1-8x10 _ 3 " 50.68 RS 1Q/ RS 27.9 29° —r 80 10 —r 20 30 4o 50 Normal Stress ( l b s . / sq . in . ) 60 T 70 FIGURE 5.22 Mohr Rupture Diagram: Summary of Results of Residual Strength Tests on Iranian Clay 99. 5.6 SUMMARY OF TESTING PROGRAM Ring shear tests were conducted on three c lay s o i l s having widely contras t ing p roper t ie s . Mu l t i - rever sa l d i r ec t shear tests were a l so conducted on two of the c l ays . Po s i t i ve determinations of res idual strength were obtained in r ing shear tests on Haney Clay and Iranian Clay. The resu l t s obtained in r ing shear tests on Hat Creek Clay are less d e f i n i t i v e due to the compl icat ing e f f ec t s o f sand and swel l ing c lay with in the soi1 compos i t ion. High values of s ide f r i c t i o n were encountered in some r ing shear te s t s . The problem is be l ieved to resu l t from corros ion welts ra ised on the conf in ing rings through chemical act ion of the s o i l pore f l u i d . Problems re l a t i n g to improper c a l i b r a t i o n of transducers and malfunctions of the data a cqu i s i t i on unit were a lso encountered. Despite such cons iderat ions , the ring shear device proved simple to operate and e f f i c i e n t at producing cons i s tent , ea s i l y interpreted re su l t s . No po s i t i ve determinations of res idual strength-were obtained in d i r ec t shear te s t ing . A l l d i r ec t shear resu l ts were lower than the values obtained in the ring shear dev ice, i nd i ca t ing that po s i t i ve excess pore pressure ex i s ted in the d i r ec t shear samples. In most tests the apparent shearing res is tance d i f f e r e d in forward and reverse shear d i r e c t i o n s , a va r i a t i on a t t r i bu ted to i n c l i n a t i o n of the shear plane. Inconvenient procedures and the protracted duration of the tests added to these d i f f i c u l t i e s . 100. CHAPTER VI SUMMARY AND EVALUATION RESIDUAL STRENGTH Residual strength is the lowest drained strength of a s o i l , a t ta ined at large shear displacements and unaffected by i n i t i a l s o i l s t ructure or stress h i s to ry . Many clayey s o i l s , p a r t i c u l a r l y i f overconso1idated, exh ib i t a res idual strength that is s i g n i f i c a n t l y lower than the peak value. Residual strength has important engineering app l i ca t ions in determining s t a b i l i t y of prev ious ly f a i l e d s lopes, where res idual condit ions current ly e x i s t , and in eva luat ing the long-term s t a b i l i t y of slopes in overconsol idated c l ay s , in which residual condit ions may develop through the mechanism of progress ive f a i l u r e . Clay deforms in simple shear p r i o r to reaching peak s trength. Beyond peak strength, shear displacement occurs by s l i p along the cleavage planes of adjacent p a r t i c l e s with in shear bands located in the f a i l u r e zone. Residual condit ions develop as the p a r t i c l e s in the shear bands a t t a i n p a r a l l e l alignment. The preferred cleavage mode of s o i l inf luences i t s residual s trength. 101. 5. Shear strength in c lays o r i g ina tes from inter -atomic bonding across s o l i d - t o - s o l i d contacts between s o i l p a r t i c l e s . Var iat ions in the residual strength of d i f f e r e n t minerals occurs due to var ia t ions in the strength and concentrat ion of bonding. 6. The residual strength of ac t i ve c lay minerals can be a l t e red by the inf luence of pore water chemistry on physico-chemical i n t e r p a r t i c l e forces of a t t r a c t i o n and repu l s ion. 6.2 UBC RING SHEAR DEVICE The ob jec t i ve of th is thes is project was to design and develop a p r ac t i c a l residual strength apparatus. Having considered a l t e r n a t i v e s , the r ing shear apparatus was se lected as the most p r a c t i c a l , e f f e c t i v e and adaptable tes t ing device yet devised for th i s a p p l i c a t i o n . Although the d i rec t shear and t r i a x i a l tests can be modified for res idual strength purposes, the i n a b i l i t y to obtain large uni -d i rect iona1 displacements and complicated or inconvenient test procedures severely l im i t the potent ia l value of these methods. The major features of the UBC Ring Shear Device are as fo l lows: 1. The device is capable of unl imited displacements at smoothly var i ab le rates o f shear from 3-2 inches/year to 9 inches/hour. 2. The appl ied normal s tress is smoothly var i ab le up to 200 l b . / s q . i n . Higher stresses can be obtained by increas ing the house l i ne a i r pressure suppl ied to the loading p i s ton . 3. The s o i l sample, with inner and outer rad i i o f 1.75 and 2.75 inches, re spec t i ve l y , has a var i ab le thickness up to a maximum of 0.75 inches. 102. 4. The length of the basal drainage path is va r i ab le up to a maximum of 0.25 inches. The top drainage path var ies in length with sample th ickness. 5. Sample placement is made simple and e f f i c i e n t by using the upper conf in ing rings as a mold for remolded samples and a cu t t ing unit for undisturbed samples. 6. The conf in ing r ing gap may be set at a very narrow width without i n te r f e r i n g with shear t e s t i ng . Typica l set t ings range from 0.002 to 0.004 inches. 7. Data a cqu i s i t i on units c o l l e c t and record data automat ica l ly in a form that is ea s i l y adaptable to computerized data reduct i on. 8. Only minimal superv is ion is required during t e s t i n g , a b r i e f equipment check every day or two normally being s u f f i c i e n t toensure smooth operat ion. 6.3 TESTING PROGRAM A ser ies of r ing shear tests were undertaken on three var ied c lay s o i l s as a means of eva luat ing the new dev ice. Repeatable and cons is tent resu l ts were obtained in r ing shear tests on Haney Clay and Iranian Clay, i nd i ca t ing that the residual angles of f r i c t i o n for these s o i l s are 24 and 28 to 29 degrees, re spec t i ve ly . Var iab le re su l t s , ranging from 21 to 25 degrees, were obtained in r ing shear tests on Hat Creek Clay. The va r i a t i on is a t t r i bu ted to d i s rupt ion of the shear plane by sand grains in the s o i l . This va r i a t i on may have been aggravated 103. by a gradual concentrat ion of sand-s ized p a r t i c l e s in the shear plane as f i n e r p a r t i c l e s were s e l e c t i v e l y lost through the conf in ing r ing gap. No determination of res idual strength was obtained in the d i r ec t shear apparatus. The resu l ts were incons is tent and lower than the res idual values as determined in the r ing shear dev ice, i nd i ca t ing that the tests were not f u l l y drained and that po s i t i ve pore pressures ex i s ted in the shear zone. 6.4 EVALUATION The UBC Ring Shear Device provides a simple and p r a c t i c a l method for determining residual s trength. Specimen preparat ion is quick and convenient, whi le operat ion of the device requires minimal superv i s ion . This v e r s a t i l e apparatus accomodates a wide range of sample heights , appl ied normal loads and rates of displacement. Test resu l t s are e a s i l y understood and in terpreted . The device funct ions e f f i c i e n t l y , averaging 13.5 test days per res idual strength determination during the test program. Increased e f f i c i e n c y is an t i c ipa ted as f a m i l i a r i t y is gained with tes t ing procedures and equipment. The success of the UBC Ring Shear Device in determining res idual strength is a t t r i bu ted in large measure to the natural advantages inherent in i t s mechanical con f i gura t ion . This conf igurat ion permits unl imited un id i rec t i ona l shear displacements to be appl ied to a s o i l sample of constant c ros s - sec t iona l area under near-uniform stress cond i t ions . These features provide an exce l lent laboratory model of the i n s i t u f i e l d cond i t i on . Also cont r ibut ing to the success of the device through 104. enhanced p rac t i c a l operat ion are the p rec i se l y con t ro l l ed conf in ing r ing gap and the automated tes t ing and monitoring features . While the UBC Ring Shear Device provides an e f f i c i e n f and e f f e c t i v e method for determining residual s t rength, minor modi f icat ions w i l l improve th i s c a p a b i l i t y . Problems encountered during r ing shear tes t ing include malfunctions of the data a c q u i s i t i o n unit and excess ive s ide f r i c t i o n on the upper conf in ing r ings. To avoid fur ther malfunc- t ions , the data a c q u i s i t i o n unit has been replaced with a new s e n s i t i v e and r e l i a b l e un i t . The excess ive values of s ide f r i c t i o n are be l ieved to resu l t from corros ion welts on the surface of the conf in ing r ings. It is recommended that th i s problem be e l iminated by rep lac ing the e x i s t i n g aluminum rings with co r ro s i on - re s i s t an t rings constructed of s t a i n l e s s s teel or anodized aluminum. The d i r ec t shear device proved to be an inconvenient and inadequate means of determining residual strength. The present and potent ia l performance of the d i r ec t shear device in res idual strength app l i ca t ions is l imi ted by a mechanical conf igurat ion which lacks the natural advantages of the r ing shear device. As a r e su l t , shear displacements must be reversed or repeated to accumulate large d i sp l ace - ments. The stress condit ions in the so i l sample, which s h i f t gradual ly during displacement, change abrupt ly with each reversal of shear d i r e c t i o n . These frequent d i s rupt ions in appl ied s t ress and s t r a i n produce data that is commonly sparse, contrad ic tory or d i f f i c u l t to i n te rpre t . However, with in these l im i ta t i ons s i g n i f i c a n t improvement in the performance of the d i rec t shear device may be poss ib le by 105. r e s t r i c t i n g the motion of the upper loading frame to a s ing le hor izonta l path. As present ly constructed, some v e r t i c a l , l a te ra l and rota t iona l motions of the upper loading frame occur during t e s t i n g . Mod i f i ca t ion of the device to prevent such motions, which could be accomplished by mounting the upper loading frame on a ca r r i age , would provide p rec i se control of the loading frame gap, reduce sample losses and permit te s t ing of th in samples. The shortened drainage path of th in samples would improve sample drainage, provid ing greater confidence in the test r e su l t s , and would probably reduce the duration of t e s t i n g . Re s t r i c t i n g the motion of the upper loading frame would a l so i n h i b i t shear plane i n c l i n a t i o n which hampered many of the d i r ec t shear tests c a r r i ed out during th i s study. Due to the extended duration of t e s t i n g , fur ther modi f icat ions to f u l l y automate the d i r e c t shear device would be e s sent i a l fo r con- venient and p rac t i ca l operat ion. Modi f icat ion of the d i rec t shear apparatus would improve i t s c a p a b i l i t y for determining residual s t rength. However, data would remain sparse and cumbersome to in terpret due to the incremental approach to cumulative displacements employed in th i s te s t , and the durat ion of tes t ing would remain extended as a resu l t of shear plane d i s rupt ions at each reversal of shear d i r e c t i o n . Therefore, the r ing shear device w i l l continue to provide the most p r a c t i c a l , convenient, e f f i c i e n t and e f f e c t i v e method for determining residual s t rength. 106. REFERENCES Bishop, A.W., G.E. Green, V.K. Garga, A. Andresen and J.D. Brown, 1971- A new ring shear apparatus and i t s app l i c a t i on to the measurement of residual s t rength, Geotechnique, V. 21, No. k, 273-328. Bjerrum, L., 1967. Progress ive f a i l u r e in slopes of overconso l idated p l a s t i c c lay and c lay sha les , Jour. So i l Mech. and Fdn. Div. , A .S .C .E . , V. 93, SM 5, 3-^9. Borowicka, H., 1965- The inf luence of c o l l o i d a l content on the shear strength of c l ay , Proceedings, 6th I n t ' l Conference on Soi l Mechanics and Foundation Engineering, Montreal, V. 1, 175-178. Bosdet, B., K. Lum and D. Negussey, 1976. One-dimensional conso l idat ion of Haney Clay, unpublished report on tests performed at the Univers i ty o f B r i t i s h Columbia, Vancouver. Byrne, P.M. and Y. Aok i , 1969- The S t ra in Contro l led Consol idat ion Test , So i l Mech. Series No. 9, Dept. of C i v i l Eng., Un ivers i ty of B r i t i s h Columbia, Vancouver. Campanella, R.G. and R.C. Gupta, 1969. E f fec t of Structure on Shearing Resistance of a Sens i t i ve Clay, So i l Mech. Series No. G~, Dept. of Ci vi 1 Eng. , Uni vers i ty of B r i t i s h Columbia, Vancouver. Chandler, R.J . , 1966. The measurement of residual strength in t r i a x i a l compression, Geotechnique, V. 16, No. 3, 181-186. Chattopadhyay, P.K., 1972. Residual Shear Strength of Some Pure Clay Minera ls , Ph.d. Thes i s , Un ivers i ty of A lbe r ta , Edmonton. Houston, Wi l l i am N. and James K. M i t c h e l l , 1969• Property i n t e r r e l a t i o n - ships in s en s i t i ve c l a y s , Jour. So i l Mech. and Fdn. D iv . , A .S .C.E . , V. 95, No. SM4, Hvorslev, M.J . , 1939- Tors ion shear tests and the i r place in the determination of the shearing res i s tance of s o i l s , Proc. Am. Soc. Test. Mater., V. 39, 999"1022. Herrmann, H.G. and L.A. W o l f s k i l l , 1966. Engineering Propert ies of Nuclear Craters ; Residual Strength of Weak Shales, U.S. Army Engineer Waterways Experiment S ta t i on , Technical Report, 3~699, Report 5- 107. Kenney,T.C., 1967- The Influence of mineral composition on the res idual strength of natural s o i l s , Proceedings, Geotechnical Conference, Os lo, V. 1, 123-129. Koerner, Robert M., Arthur E. Lord, J r . , and W. Martin McCabe, 1977- Acoust ic emission behavior of cohesive s o i l s , Jour. Geotechnical Eng. D iv . , A .S .C .E . , V. 103, No. GT8, 837-850. La Gatta, Daniel P., 1970. Residual Strength of Clays and Clay-Shales by Rotation Shear Tes t s , Harvard So i l Mechanics Ser ies No. .86, Harvard Un iver s i t y . Lum, K. and D. Negussey, 1977. Unpublished resu l t s of tes t ing at the Un ivers i ty of B r i t i s h Columbia, Vancouver. Matsui, Tamotsu, Tomio Ito, James K. M i t c h e l l , and Nobuharu Abe, 1980. Microscopic study of shear mechanisms in s o i l s , Jour. Geotechnical Eng. Div. , A .S .C .E . , V.106, No. GT2, 137-152. M i t c h e l l , James K., 1976. Fundamentals of So i l Behaviour, John Wiley & Sons, 422 p. M i t c h e l l , James K., Awtar Singh, and Richard G. Campanella, 1969• Bonding, e f f e c t i v e stresses, and strength of s o i l s , Jour. Soi l Mech. and Fdn. D iv. , A .S .C .E . , V. 95, No. SM5. M i t c h e l l , James K. and Wi l l iam N. Houston, 1969- Causes of c lay s e n s i t i v i t y , Jour. So i l Mech. and Fdn. D iv. , A .S .C .E . , V, 95, No. SM3. Morgenstern, N.R. and J.S . Tchalenko, 1967- Microscopic s t ructures in kao l in subjected to d i r ec t shear, Geotechnique, V. 17, No. 4, 309-328. Pet ley , D .J . , 1966. The Shear Strength of So i l s at Large S t r a in s , Ph.D. Thes i s , Un ivers i ty of London. Quigley, R.M., 1976. Unpublished reports to Golder Brawner and Associates Ltd. Skempton, A.W., 1964. Long-term s t a b i l i t y of c lay s lopes, Geotechnique, V. 14, No. 2, 77-101. Tchalenko, J . S . , 1968. The Evolut ion of Kink-bands and the Development of Compression Textures in Sheared Clays, Tectonophysics, V. 6, 159-174. 108. Townsend, Frank C. and Paul A. G i l b e r t , 1972*. Engineering Propert ies o f Clay Shales; Residual Shear Strength and C l a s s i f i - cat ion Indexes of Clay Shales, U.S. Army Engineer Waterways Experiment S t a t i on , Technical Report, S-71-6, Report 2. Vaid, Y.P. and R.G. Campanella, 1977- Time-Dependent Behaviour of an Undisturbed Clay, Soi l Mechanics Ser ies No. 33, Dept. o f C i v i l Eng., Un ivers i ty of B r i t i s h Columbia, Vancouver. Webb, D.L., 1969. Residual strength in conventional t r i a x i a l t e s t s , Proceedings, 7th I n t ' l . Conf. Soi l Mech., Mexico, V. 1, 433-^1- 109. APPENDIX I DERIVATION OF EQUATION FOR DETERMINING tan 0' Let = uniform shear s t ress over shear plane (T* = uniform normal s tress tan 0' = t/G- r^ = ins ide radius of s o i l sample r^ = outs ide radius o f s o i l sample M = sum of moments over shear plane as measured by force transducers W = tota l normal load on shear plane 'T" x Sample Area x Moment Arm 2 « r t a n 0 ' ) (2Tfrdr) (r) 1 • J i . = 2tt<Ttan0' \ 2 r 2 d r J r r 3 - r 3 = 2 t t ( T t a n 0 l - 2 — = — - Therefore: t a n 0 ' 3M " - • ? ) , 2 2, 3M x ( r 2 - r 1 } 2 ( r 3 - r 3 ) X 0 - ^ ( r 2 _ 2) 2 2 3M r 2 " r i 2W 3 3 r - r 2 A 1 3M(rl + r 2) 2W(r^ + r i r 2 + r|) 110. APPENDIX II DIRECT SHEAR APPARATUS The d i r e c t shear apparatus used in th i s study is a standard manual dr ive unit with a 2 by 2 inch shear box that has been f i t t e d with a va r i ab le speed motor d r i ve . The length of the shortest drainage path, from the porous base to the shear plane, is 3/8 inches. I n i t i a l l y the device had no se l f - s topp ing mechanism and was operated only during hours when an attendant was a v a i l a b l e . The attendant was required to observe and record load and displacement data, to maintain and adjust the gap between the upper and lower loading frames of the shear box, and to reverse the d i r e c t i o n of t rave l when necessary. During Direct Shear Test 3 (DS3), the apparatus was f i t t e d with a micro-switch system to stop the motor dr ive at the des i red end of t r a v e l . The se l f - s topp ing mechanism enabled a more continuous operat ion of the equipment to obtain large displacements in shorter time, but an attendant was s t i l l required to observe and record data, adjust the gap, and reverse the d i r e c t i o n of t r a v e l . More recent modi f icat ions have f u l l y automated the dr ive system and data c o l l e c t i o n , although per iod i c manual adjustments of the loading frame gap are s t i l l required.

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
China 18 0
France 16 0
Japan 4 0
United States 3 1
Canada 3 0
Germany 1 3
City Views Downloads
Unknown 18 4
Beijing 18 0
Tokyo 4 0
Edmonton 3 0
Ashburn 2 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items