UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Filtration performance of geotextiles in cyclic flow conditions : a laboratory study Hawley, Rashmi A. 2001

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2001-0211.pdf [ 14.27MB ]
Metadata
JSON: 831-1.0063728.json
JSON-LD: 831-1.0063728-ld.json
RDF/XML (Pretty): 831-1.0063728-rdf.xml
RDF/JSON: 831-1.0063728-rdf.json
Turtle: 831-1.0063728-turtle.txt
N-Triples: 831-1.0063728-rdf-ntriples.txt
Original Record: 831-1.0063728-source.json
Full Text
831-1.0063728-fulltext.txt
Citation
831-1.0063728.ris

Full Text

FILTRATION PERFORMANCE OF GEOTEXTILES IN CYCLIC FLOW CONDITIONS: A LABORATORY STUDY By R A S H M I A . H A W L E Y B . S c . E n g . (Civil Eng ineer ing) - Universi ty of N e w Brunswick , Freder ic ton , C a n a d a , 1998 A T H E S I S S U B M I T T E D I N P A R T I A L F U L F I L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R O F A P P L I E D S C I E N C E in T H E F A C U L T Y O F G R A D U A T E S T U D I E S D E P A R T M E N T O F C I V I L E N G I N E E R I N G W e accep t this thesis a s conforming to the required s tandard T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A M a y 2001 © R a s h m i Hawley , 2001 In presenting this thesis in partial fulfillment of the requirements for an advanced degree at the University of British Columbia, I agree that the library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Civil Engineering The University of British Columbia Vancouver, Canada R. Hawley Authorization ABSTRACT Geotextile filters are often used as a replacement for, or in combination with, traditional granular filters in many engineering works. Conventional design criteria, which are largely empirical, are generally sufficient for applications where flow is unidirectional and the soil is internally stable. However, for conditions including reversing flow regimes and potentially internally unstable soils, these criteria may not be adequate and performance tests may be necessary. The gradient ratio test is a performance test that assesses soil-geotextile compatibility under an applied hydraulic gradient. The gradient ratio device developed at UBC is a modified version of the ASTM apparatus, which allows the application of both unidirectional and reversing flow to soil-geotextile systems at varying hydraulic gradients and confining pressures. In this research work, three soils were tested in combination with seven geotextiles, using the modified gradient ratio device. Two of the geotextiles were nonwoven materials and 5 were woven, with AOS values ranging from 0.212 mm to 0.600 mm. The soils were a Fraser River sand, a copper mine-waste tailings and a Port Coquitlam silty sand. The mine-waste tailings and Port Coquitlam silty sand were recognized as potentially 'problematic' from a filtration standpoint. The soils had a relatively narrow range of D8s (from 0.330 mm to 0.215 mm), and a moderate range of coefficient of uniformity, C u (from 1.8 to 5.8). The tests therefore provided results for AOS/D8 5 values ranging from 0.6 to 2.8. The intent was to gain insight to (i) the influence of geotextile type (woven versus nonwoven), (ii) the influence of flow regime (unidirectional versus cyclic), and (iii) the validity of existing design guidance for the range of soil and geotextile combinations used in testing. Based on the very limited comparison of three geotextiles of the same opening size, it appears there is little difference in behaviour of these woven and nonwoven geotextiles. All tests were relatively stable, with insignificant quantities of soil passing through the geotextiles. Results for the Port Coquitlam silty sand, which yielded the most soil passing through the geotextile, showed a small difference in the grain size distribution of the passing soils. It appeared that more of the finer material passed through the woven geotextile, than the corresponding nonwoven. The influence of flow regime was studied from tests in unidirectional flow, and cyclic flow with and without confining stress. No significant influence of the frequency was found in testing for the flow reversal at frequencies of 0.02 Hz and 0.1 Hz. The Fraser River sand is stable in all tests and therefore, the influence of flow regime does not appear to be significant. The mine waste tailings are stable in all unidirectional flow, and generate a very subtle trend towards piping instability in cyclic flow as the AOS/D8 5 approaches 2.0. The G R A S T M and G R M O D values correspondingly are less than unity thus indicating the onset of piping. The Port Coquitlam silty sand behaved slightly differently R. H a w l e y II A b s t r a c t than the other two soils, in that it is stable in both unidirectional flow and confined cyclic flow, but experienced significant piping and collapse of the soil structure with unconfined cyclic flow. The soil yielded catastrophic piping during sample preparation when the AOS/D8 5 was 2.8. The results were used to evaluate the design criteria of CGS (1992) and Luettich et al. (1992) in unidirectional flow, and CGS (1992), Luettich et al. (1992) and Holtz et al. (1997) in cyclic flow. The CGS (1992) and Luettich et al. (1992) guidance were found to be slightly conservative for soil-geotextile filtration compatibility in unidirectional flow. For cyclic flow, all three criteria were again found to be reasonable, but somewhat overly conservative. R. Hawley Abstract TABLE OF CONTENTS Page ABSTRACT ii LIST OF TABLES vi LIST OF FIGURES vii ACKNOWLEDGMENTS ix 1.0 INTRODUCTION 1 1.1 Purpose of Study 2 1.2 Scope of Study 2 2.0 LITERATURE REVIEW 4 2.1 Basic Principles in Filter Design 4 2.1.1 Soil Retention 4 2.1.2 Permeability 6 2.1.3 Strength/Survivability and Durability 7 2.2 Current Design/Specification Guidance 7 2.2.1 Unidirectional Flow 9 2.2.2 Cyclic Flow 10 2.3 Filtration Testing 13 2.3.1 Unidirectional (Steady State) Flow 13 2.3.2 Cyclic (Dynamic) Flow 18 3.0 APPARATUS AND PROCEDURES 23 3.1 Modified Gradient Ratio Device 23 3.1.1 Apparatus 23 3.1.2 Data Acquisition System 26 3.1.3 X-ray Particle Analysis 26 3.2 Procedures 27 3.2.1 Sample Preparation 27 3.2.2 Test Set-Up 28 3.2.3 Multi-Stage Testing Procedure 29 3.2.4 Particle Size Analysis 31 R. Hawley i v Table of Contents 4.0 TEST MATERIALS 32 4.1 Geotextiles 32 4.2 Soils 34 5.0 TEST RESULTS 38 5.1 Head Losses in the Permeameter 38 5.2 Particles Passing Through the Geotextile 39 5.3 Pre-test and Post-test Gradations 41 5.4 Water Head Distributions 44 5.5 Permeability 46 5.6 Gradient Ratio 54 5.7 Repeatability 55 6.0 ANALYSIS OF TEST RESULTS 58 6.1 Influence of Geotextile Type 58 6.2 Influence of Flow Regime 59 6.3 Design Criteria 65 6.3.1 Unidirectional Flow 66 6.3.2 Cyclic Flow 68 7.0 IMPLICATIONS FOR PRACTICE 71 7.1 Design Guidance 71 7.2 Limitations of Testing 71 8.0 CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK 73 LIST OF REFERENCES 76 APPENDIX A - Internal Stability of Test Soils and Calculation of Design Criteria 80 APPENDIX B - Pre-test and Post-test Gradations 87 APPENDIX C - Water Head Distributions 108 APPENDIX D - Tabulation of Key Results 129 R. Hawley v Table of Contents LIST OF TABLES Page Table 2.1 Summary of Permeability Criteria (after Christopher and Fischer, 1992) 7 Table 2.2. Typical Hydraulic Gradients (after Giroud, 1996) 8 Table 2.3. Cyclic load and related period (after Mouw et. al., 1986) 8 Table 2.4. Existing geotextile retention criteria (modified after Hameiri, 2000) 9 Table 2.5. Filtration criteria that include cyclic flow conditions (after Hameiri, 2000) 10 Table 3.1. Multi-stage testing sequence 29 Table 3.2. Data collection set-up 30 Table 4.1. Physical Properties of Geotextiles Tested 32 Table 4.2. Test soil properties 35 Table 4.3. Test codes (and A O S / D 8 5 ) of soil-geotextile combinations 35 Table 5.1. Summary of mass of soil passing during each test stage (g/m2) 39 Table 5.2. Summary of gradations of mass of soil passing (where adequate sample) 43 Table 5.3a. Average permeabilities (x 10"3 cm/s) for unidirectional and post cyclic stages: 50 F R S tests Table 5.3b. Average permeabilities (x 10"3 cm/s) for unidirectional and post cyclic stages: 51 MWT tests Table 5.3c. Average permeabilities (x 10"3 cm/s) for unidirectional and post cyclic stages: 52 P C S tests Table 5.4. Summary of sample length after each stage of P C S tests (mm) 53 Table 5.5. Summary of gradient ratio test results 54 Table 5.6a. Test results: Test F160(a) 56 Table 5.6b. Test results: Test F160(b) 56 Table 6.1. Comparison of filter performance: nonwoven versus woven 58 Table 6.2. Design criteria used in this study 65 R. Hawley VI List of Tables LIST OF FIGURES Page Figure 2.1. Filter bridge formation (after Holtz et al., 1997) 5 Figure 2.2. Giroud, 1982 soil retention criteria for steady state flow conditions (after 12 Luettich et al., 1992) Figure 2.3 Heerton, 1982 soil retention criteria for dynamic flow conditions (after 14 Luettich etal., 1992) Figure 2.4. Gradient ratio permeameter set up diagram, ASTM D 5101 15 (after ASTM, 1996) Figure 2.5. UBC modified gradient ratio device (after Fannin et al., 1996) 17 Figure 2.6. Cyclic flow test: Italian device (after Tondello, 1998) 20 Figure 2.7. Cyclic flow test: Singapore bi-directional flow apparatus 21 (after Chew et al., 2000) Figure 3.1. UBC modified gradient ratio device (after Hameiri, 2000) 24 Figure 3.2. The permeameter and port locations 24 Figure 3.3. Modeling the reversing flow regime in the permeameter 25 (after Hameiri, 2000) Figure 3.4. Differential pressure transducer and manometer setup 26 (modified after Hameiri, 2000) Figure 4.1a. SEM photograph: Nonwoven geotextile, AOS = 0.212 mm (Mirafi 140N) 33 Figure 4.1b. SEM photograph: Woven geotextile, AOS = 0.212 mm (FW700) 33 Figure 4.1c. SEM photograph: Woven geotextile, AOS = 0.600 mm (HP570) 33 Figure 4.2. Soil gradations 34 Figure 4.3a. SEM photograph of FRS: subangular to subrounded particles 36 Figure 4.3b. SEM photograph of MWT: angular to subangular particles 37 Figure 4.3c. SEM photograph of PCS: subrounded particles 37 Figure 5.1. Flow Rate, Q, dependent response of total head loss, h 1 7 38 Figure 5.2. Preferential channeled type piping as observed during tests 40 P404 and P402. Figure 5.3a. Pre- and post-test gradations, example of Scenario 1 (a, b): 41 D 5 0 constant (test F402) Figure 5.3b. Pre- and post-test gradations, Scenario 2: D 5 0 increases (test M700) 42 Figure 5.3c. Pre- and post-test gradations, Scenario 3: D 5 0 decreases (test P404) 42 Figure 5.4a. Water Head Distributions: Test F402 (i17 = 2) 45 Figure 5.4b. Water Head Distributions: Test M570 (i17 = 4) 45 Figure 5.4c. Water Head Distributions: Test P402 (i17 = 5) 45 Figure 5.5a. Cumulative Head Loss Between Ports During Stage CYC50S: Test F402 47 Figure 5.5b. Cumulative Head Loss Between Ports During Stage CYC10S: Test F402 47 Figure 5.6a. Cumulative Head Loss Between Ports During Stage CYC50S: Test M570 48 R. Hawley VII List of Figures Figure 5.6b. Cumulative Head Loss Between Ports During Stage CYC 10S: Test M570 48 Figure 5.7a. Cumulative Head Loss Between Ports During Stage CYC50S: Test P402 49 Figure 5.7b. Cumulative Head Loss Between Ports During Stage CYC10S: Test P402 49 Figure 5.8a. Water head distribution of unidirectional stages: Test F160(a) 57 Figure 5.8b. Water head distribution of unidirectional stages: Test F160(b) 57 Figure 6.1a. Influence of flow regime: FRS tests, i 1 7 = 2 60 Figure 6.1b. Influence of flow regime: MWT tests, h7 = 4 61 Figure 6.1c. Influence of flow regime: PCS tests, i 1 7 = 5 61 Figure 6.2a. Unidirectional results: G R M O D 62 Figure 6.2b. Cyclic results: G R M O D 62 Figure 6.3a. Test P404: Failure pattern during stage CYC10N 64 Figure 6.3b. Test P402 Failure pattern during stage CYC10N 64 Figure 6.4a. Unidirectional results: CGS (1992) guidelines 67 Figure 6.4b. Unidirectional results: Luettich et al. (1992) guidelines 67 Figure 6.5a. Cyclic results: CGS (1992) guidelines 69 Figure 6.5b. Unidirectional results: Luettich et al. (1992) guidelines 69 Figure 6.5c. Cyclic results: Holtz et al. (1997) guidelines 70 R. Hawley VIM List of Figures ACKNOWLEDGMENTS I wish to express my sincere gratitude to my supervisor, Dr. R. J. Fannin, P.Eng., whose patience, support, enthusiasm and continual encouragement made the submission of this thesis possible. Also, for your development of a university-industry partnership in support of the research, I am deeply grateful. The National Science and Engineering Research Council of Canada provided core funding for this project. TC Mirafi Inc. manufactured the geotextiles and also provided industry funding. I would like to thank Mr. Harald Schrempp and Mr. Doug Smith of the UBC Civil Engineering workshop for their precision and expediency in making modifications to the gradient ratio device. Many thanks also to Mr. Scott Jackson and Mr. John Wong for their advice and help regarding the computer and electrical systems used in this project. Thanks also to Dr. Avikam Hameiri for your guidance and for answering my many questions. I would also like to thank Mr. Dal Scott of Highland Valley Copper and Ms. Karen Thompson of Klohn-Crippen Consultants Ltd. for providing the mine-waste tailings for testing. I also wish to express love and many thanks to my family in New Brunswick, who have always been a solid foundation supporting me through every part of my life. Finally, I am deeply thankful to my husband, Hugh, for his constant love and encouragement across the 6000 km that separated us for two years. R. Hawley i x Acknowledgements 1.0 INTRODUCTION Geotextile filters are often used as a replacement for, or in combination with, traditional granular filters in many engineering works including earth dams, coastal erosion protection, and waste containment facilities. They offer comparable performance, improved economy, consistent properties, and ease of placement (Christopher & Fischer, 1992). It is imperative, where the consequences of poor performance are critical, to provide proper design guidance for such filters. Conventional design criteria, which are largely empirical, are generally sufficient for routine applications. In such applications, flows are unidirectional and soils are internally stable. However, for more demanding conditions, such as reversing flow regimes and potentially internally unstable soils, these criteria may not be adequate and performance tests may be required (Fannin & Pishe, 2001). The underlying problem remains that drainage systems employing geotextile filters are based on empirical design guidance that is fashioned after traditional granular filters. Being a relatively new material, there are few long-term performance data on which to evaluate the applicability and success of these criteria for geotextiles. Consequently, there can be difficulty in designing with confidence and hence a reluctance to use these materials. This causes a reversion to the use of traditional granular filters, which themselves can be constructed with poor quality materials, depending largely on availability, and improper construction practices. Many factors influence the performance of geotextile filters, including (i) type of application, (ii) soil properties, (iii) filter properties, (iv) fluid properties, (v) hydraulic conditions, (vi) confining stresses and (vii) construction practices (Fannin & Pishe, 2001). Specification of a compatible soil-geotextile system is based on strength, soil retention and consideration of relative permeability. Retention criteria limit the amount of piping through the geotextile while promoting the development of a filter bridge above the geotextile (Lawson, 1982). Retention criteria also include mitigation of long term clogging of the geotextile pores. Relative permeability criteria, however, only require that the permeability of the composite layer of the geotextile, soil bridge and immediate upstream filter zone be compatible with that of the base soil. Performance tests, such as the gradient ratio test as first proposed by Calhoun (1972) and standardized by ASTM (1992), are used to assess directly the compatibility of soils and geotextiles under different imposed hydraulic gradients (Fannin et al., 1994a,b). There does exist a reasonably large body of test data to support the specification criteria for geotextiles in unidirectional flow. However, cyclic flow has not been extensively studied due to the very few laboratory test devices that simulate cyclic flow conditions. The gradient ratio device developed at UBC is a modified version of the ASTM apparatus, which allows the application of both unidirectional and reversing flow to soil-R. Hawley 1 Chapter 1 geotextile systems at varying hydraulic gradients and confining pressures. The objective of this research work is to test various soils that are recognized to be challenging in filtration applications, with geotextiles, using this modified gradient ratio device. 1.1 Purpose of Study The purpose of the work is to evaluate the performance of seven geotextiles in combination with three different soils, in severe flow conditions, using the UBC Modified gradient ratio device. The soils are identified, by the experience of industry leaders, as having the potential to be problematic and hence present a challenge to many geotextile filters. The geotextiles comprise both nonwoven and woven materials provided by TC Mirafi Inc. This study serves three main purposes, namely to perform gradient ratio tests using problematic soils to assess empirical design guidance and make recommendations to industry regarding these soils, to quantify the relative performance of woven and nonwoven geotextiles having similar apparent opening sizes (AOS) in a comparative study, and to determine the influence of flow regime on soil-geotextile filtration compatibility. Additionally, findings of the study will contribute to the data bank of gradient ratio test results using this apparatus, on woven and nonwoven geotextiles, with 'real' and challenging soils. 1.2 Scope of Study The thesis is based on an interpretation of gradient ratio test results for seven geotextiles tested against three soils, yielding a total of 21 tests. Repeatability tests were also conducted, and those results are presented. A multi-stage testing procedure allows the imposition of •either steady state (unidirectional) flow, or unsteady (cyclic) flow at a selected frequency of flow reversal (0.02 Hz or 0.1 Hz). The vertical confining stress (unconfined or 25 kPa) is also controlled during the test. The geotextiles have AOS values ranging from 0.212 mm to 0.600 mm. The soils have a D 8 5 in the range 0.330 mm to 0.215 mm, and a C u between 1.8 and 5.8. Interpretation of the results is based on measurements of sample height, head loss along the sample, flow rate, mass of soil passing through the geotextile, gradation of the soil passing through the geotextile and visual observations of piping or clogging behaviour. The analysis of results addresses not only the gradient ratio values, but also examines their changes associated with the mass of soil passing through the geotextile and the permeability values deduced from flow rates through the soil-geotextile system. The synthesis of results provides a basis on which to draw conclusions on the filtration compatibility of these problematic soils, in unidirectional and cyclic flow, and to determine the relative performance of nonwoven and woven geotextiles. R. Hawley 2 Chapter 1 Specifically, the intent therefore is to perform tests using the UBC Modified Gradient Ratio device to determine (i) the influence of geotextile type (woven versus nonwoven), (ii) the influence of flow regime (unidirectional versus cyclic), and (iii) the validity of existing design guidance for the range of soil and geotextile combinations used in testing. R. Hawley 3 Chapter 1 2.0 LITERATURE REVIEW Geotextiles are now routinely used in many filtration and drainage applications, as an alternative to granular filters. They have proven a reliable alternative to granular filters for many reasons, including improved economy, consistent properties and ease of placement (Christopher & Fischer, 1992). Potential cost advantages include an expedient construction, reduction of maintenance costs, the ability to use less or lower quality drainage aggregate, reduction of excavation volume by using smaller drains, and elimination of collector pipes (Holtz et al., 1997). Geotextiles were once placed as a simple 'filter cloth' with little or no regard for design and specification of compatible materials. However, there is a growing confidence in these materials in routine applications due to the many design criteria that now exist in the literature. These criteria, however innumerable for steady state or unidirectional flow, have not yet fully addressed severe conditions such as cyclic flow with very high gradients. This review describes the basic principles of filtration and geotextile filter design. Many of the available design criteria are presented for both steady state (unidirectional) and non-steady state (cyclic or dynamic) flow conditions. As filtration testing is a key factor in determining these criteria, the most common methods used are then briefly described. Specifically, the theory behind gradient ratio (GR) testing "is presented and the developments of GR testing for unidirectional and reversing flow are reviewed. The objective is to summarize relevant research and design guidance and thereby outline the factors contributing to the need for this study. 2.1 Basic Principles in Filter Design Filter design is predicated on one major principle: it must allow unimpeded flow of water across the system while providing adequate retention of soil particles and preventing their migration. This function must be maintained throughout the design life of the project, which means that the filter cannot become unacceptably blocked (i.e. blinded) or clogged. This principle leads to three major design criteria that must be satisfied: (i) soil retention, (ii) permeability and (iii) strength (or survivability) and durability. These are briefly described in the following sections. 2.1.1 Soil Retention Generally, retention criteria for geotextile filters are based on traditional granular filter design whereby a characteristic geotextile opening size (On) is specified to be less than a characteristic grain size of the soil to be retained (Dn). This criterion is established such that some of the finer particles may pass (or wash through) without disturbing the integrity of the base soil, while R. Hawley 4 Chapter 2 still promoting the development of a filter bridge This bridge is formed with the coarser particles of soil filtering the smaller particles next to the geotextile, which will serve to retain the base material (Figure 2.1). Figure 2.1. Filter bridge formation (after Holtz et al., 1997) Lawson (1982) states that the geotextile only serves as a catalyst for forming the filter bridge after which point, it is the bridge that maintains the function of the system. The geotextile, however, must be chosen properly in order for optimal formation of the bridge or self-filtering layer. This is also reinforced by Giroud (1996), who observes that the migration of all soil particles need not be prevented for proper retention of the base soil, but rather, the soil behind the filter must remain 'stable'. This means, that after the initial washout of fines, a steady state is achieved whereby the soil bridge is intact and flow is unimpeded. This phenomenon is examined and a 'washout model' is presented by Lafleur et al. (1989) and Lafleur (1998). Lafleur et al. (1989) have shown that for broadly graded cohesionless soils, a particle migration of 2.5 kg/m2 is a threshold above which instability can occur, based on test soils with Cu values greater than 8 and D 8 5 values from 1.8 to 19. This, however, may be too large a soil mass migrating into a downstream drain (Lafleur, 1999). Bhatia and Huang (1995) found that anything below approximately 3 kg/m2 was relatively insignificant. Lawson (1998), on the other hand, suggested the use of rate of soil passing through the geotextile as an index for piping since, with time, continuous piping could lead to loss of serviceability and potential collapse. For internally unstable soils, however, Kenney and Lau (1985) recommend a maximum opening size for retention of soils as well as a minimum opening size in order to prevent clogging near the base/filter interface caused by internal migration of fines within the soil skeleton. This recommendation is based on work with granular filters. R. Hawley 5 Chapter 2 Given the intent of specifying a geotextile with openings small enough to retain the base soil and promote the formation of a filter bridge, the geotextile should not be so tight as to yield clogging or blinding of the geotextile. Piping has been known to manifest itself quickly whereas clogging may occur quickly or gradually over the long term. As Holtz et al. (1997) have stated, it is the life of the structure that must be considered. A retention criterion ensures that the geotextile provides adequate flow during the design life of the structure. So, even if some openings become plugged, the flow rate will be maintained at an operable level. This criterion sometimes requires the use of long-term filtration tests to ensure compatibility between site-specific soils and the geotextiles used. For example, for clogging resistance, a Gradient Ratio value of 3 is recommended as an upper bound (US Army Corps of Engineers, 1977). Also as additional qualifiers, Christopher & Holtz (1985) recommend that a 30% porosity minimum be maintained for nonwoven geotextiles, and Calhoun (1972) recommends a percent open area of 4% for woven geotextiles in order to reduce the risk of long term clogging or blinding of the geotextile. This is especially important for silty soils where clogging is a definite possibility. These recommendations on clogging resistance have been subsequently incorporated in more recent documentations, including the Canadian Foundation Engineering Manual (CGS, 1992) and Holtz et al. (1997). The existing design guidance for both unidirectional and cyclic flow regimes is further discussed in section 2.2. 2.1.2 Permeability This criterion in based on the premise that the permeability of the geotextile (kg) must be larger or equal to that of the soil against which it is placed (ks) in order to provide unimpeded flow through the soil-geotextile system and therefore avoid any build up of excess pore water pressures in that soil. This criterion often conflicts with the retention criterion in that the openings must be small enough to retain the soil, yet large enough to allow free passage of water. Permeability is generally specified directly using the relationship, kg > C ks, where the constant C depends on the severity of the flow conditions. Carroll (1983), Holtz et al. (1997) and the CGS (1992) recommend that C should be equal to 10 for critical applications and for severe soils and hydraulic conditions. Critical applications are defined by a high risk of loss of life and/or structural damage due to drain failure, very high repair costs versus installation costs of drain and no evidence of drain clogging before potential catastrophic failure. Severe conditions are defined by a gap-graded, pipable or dispersible soil to be drained, a high hydraulic gradient and/or dynamic, cyclic, or pulsating flow conditions. A summary of the permeability criteria is presented in Table 2.1 R. Hawley 6 Chapter 2 Table. 2.1. Summary of Permeability criteria (after Christopher and Fischer, 1992) Source Criterion Remarks Calhoun (1972), Schober & Teindl (1979), Wates (1980), Carroll (1983), Haliburton et al. (1982), Christophers Holtz(1985), USFHWA (1985), CGS (1992) kg > ks Steady state flow Noncritical applications Nonsevere soil conditions Carroll (1983), Christopher & Holtz(1985), USFHWA (1985), CGS (1992) kg > 10 ks Critical applications Severe soil or hydraulic conditions Giroud (1982) kf > 0.1 ks No factor of safety French Committee of Geotextiles and Geomembranes (1986) Based on permittivity, with 4>> 103"5 ks Critical 105 ks Less critical 104 ks Clean sand 103 ks Koerner (1990) fallow > FS x ^req'd Factor of safety, FS, based on application and soil conditions 2.1.3 Strength/Survivability and Durability The issue of survivability, see AASHTO (1998), is addressed to ensure the geotextile is sufficiently durable to withstand stresses from installation and placement. These guidelines are given in the form of minimum average roll values of grab strength, burst strength, seam strength, puncture strength and trapezoidal tear strength for high and moderate survivability. However, since this is a not a hydraulic issue, it will not be expanded upon in this review. Durability, or endurance, is also a prominent issue as it pertains to the long-term integrity of the material. As stated by Holtz et al. (1995), geotextiles have been shown to be basically inert materials for most environments and applications. However, certain applications may expose the geotextile to chemical or biological activity that could drastically influence its filtration properties or durability. Again, this issue is beyond the scope of this paper and is not further discussed here. 2.2 Current Design / Specification Guidance Many factors must be considered when selecting a geotextile for compatibility in a soil-geotextile filtration application. These comprise four categories (Bhatia et al., 1991): (a) soil properties, (b) geotextile properties, (c) hydraulic conditions, and (d) external conditions. Relevant soil conditions include particle size gradation, relative density, particle shape and permeability. The R. Hawley 7 Chapter 2 geotextile properties of concern are opening size, porosity, mass per unit area and permittivity. Hydraulic conditions include gradients, flow directions and frequencies, while external conditions include type and function of earth structure, type of loading and nature of adjacent soils. Designing for proper hydraulic conditions requires the knowledge or estimation of representative gradient values and in the case of cyclic flows, the characteristic frequency of flow reversal. Typical values for these latter two conditions, namely hydraulic gradients and frequency, are provided in Tables 2.2 and 2.3 respectively. Table 2.2. Typical Hydraulic Gradients (after Giroud, 1996) Drainage Application Typical Hydraulic Gradient Standard dewatering trench 1.0 Vertical wall drain 1.5 Pavement edge drain 1.0 Landfill leachate collection/detection removal system 1.5 Landfill leachate collection removal system 1.5 Landfill closure surface water collection removal system 1.5 Dam toe drains 2.0 Dam clay cores 3.0to>10 Inland channel protection 1.0 Shoreline protection 10 Liquid impoundment with clay liners >10 Note: Critical applications may require designing for higher gradients than those given. Table 2.3. Cyclic load and related period (after Mouw et. al., 1986) Phenomenon Frequency (Hz) Remark Storm surges, 1Q-6 to be considered stationary for geotextiles Tidal waves Seiches (waves resulting from ^Q-3 atmospheric or seismic disturbances), Translation waves, Swell transition area Wind waves 0.1 Ship waves 0.5 cyclic Turbulence Dynamic impact 100 R. Hawley 8 Chapter 2 Soil properties, geotextile properties, and hydraulic conditions have been the subject of various studies from which many geotextile design criteria have been established. The following sections present these criteria divided into those that pertain to unidirectional flow (section 2.2.1) and those that pertain to cyclic flow (section 2.2.2). 2.2.1 Unidirectional Flow A significant body of test data exists to describe soil/geotextile interaction for unidirectional (steady state) flow, and validate design criteria for soil retention (Faure and Mylnarek, 1998). Some of the more common criteria using the filtration opening size (FOS) of the geotextile, as well as the Giroud (1982) criterion using O g 5, are shown in Table 2.4. Table 2.4. Existing geotextile retention criteria (modified after Hameiri, 2000) Source Criterion Remarks FCGG (1986) FOS/D8 5< 0.38-1.25 FOS/D15>4 - Dependent on soil type, compaction, hydraulic and application conditions - For soils from which fines can easily be put in suspension CGS (1992) FOS/D8 5 < 1.5 FOS/D8 5 < 3 - Uniform soils - Broadly graded soils OMT (1992) FOS/D8 5 < 1 and FOS > 0.5D85 or FOS > 0.040 mm UBC (Fannin et al., 1994a) FOS/D8 5 < 1.5 FOS/D5 0 < 1.8 FOS/D8 5 < 0.2 FOS/D50 < 2 FOS/D 5 0| <2.5 FOS/D1 5 < 4 1<Cu<2 3<Cu<7, and where D50,i is the mean particle size of the soil fraction smaller than the FOS of the geotextile Lafleur (1998, 1999) FOS/D, < 1 1 < FOS/D3 0 < 5 - For internally stable soils, where the indicative grain size diameter (Dj) is a function of Cu and the gradation shape profile - For internally unstable soils Giroud (1982) 095/D5o<(9-18)/Cu - Dependent on soil C u and density, assumes fines in soil migrate for large C u R. Hawley 9 Chapter 2 In addition to the design criteria presented above, the Giroud (1982) criteria are presented in the form of a flow chart leading step by step to the dimensioning of geotextiles under steady state flow (see Figure 2.2). 2.2.2 Cyclic Flow Cyclic flow conditions may occur in highway applications due to traffic loading or in coastal applications due to wind, ship and tidal activity. It is for these types of situations that the geotextile filter must be designed properly and according to applicable hydraulic conditions. As stated by Holtz et al. (1997), if the geotextile is not properly weighted down and in intimate contact with the soil to be protected, and the dynamic, cyclic or pulsating loading conditions produce high localized hydraulic gradients, the soil particles can move behind the geotextile. Thus, the use of 095/D85 = 1 (unidirectional criterion) is not conservative, because the bridging network will likely not develop and the geotextile will be required to retain even-finer particles. In contrast to the unidirectional criteria provided in section 2.2.1, there are few design criteria for cyclic flow conditions that have been validated by test data. Hameiri (2000) has provided a summary of filtration criteria that include cyclic flow conditions, together with comments on the validity of these criteria. Inspection of Table 2.5 reveals a considerable variation in the empirical relationships used for design of geotextiles subject to cyclic loading (Fannin & Hameiri, 1999). Table 2.5. Filtration criteria that include cyclic flow conditions (after Hameiri, 2000) 0 „ ., . Characteristic _ , Source Criterion Remarks pore size The grain diameter that must be ._ retained is set equal to the _ . . - For nonwoven Schoberand opening size. This diameter is ^ ' ^ ' H " ? 'J*! Sand ^ n , T .„ M r i 7 n , , ? . . , ., , with sand (acc. 0.01 mm < D 5 0 < 0.3 mm Teindl (1979) dependent on the intensity of , _ . , }n-.c. . A e ~ r-fl ,j«, n ui * t to Ogink, 1975) and1.5<C„<5 flow and the allowable amount of a ' u soil loss. Lawson(1982) D 5 0 >. O 9 0 > D 1 5 Dry sieving with glass beads - Have been used in design and limited number of erosion control structures and appears to validate the criterion C*D 8 5 > 0 9 5 C = a function of grain size u . , FCGG(1986) distribution, soil density, Hydro-dynamic hydraulic flow, and geotextile sieving function R. Hawley 10 Chapter 2 U.S. D.O.T FHWA (1995) 0.5*D85 > Og Dry sieving - A revised Christopher with glass and Holtz (1985) criterion beads (see Holtz et al., 1997) lngold(1985) O90,w/D5o < 1 Very similar to the wet sieving using sand, developed by Heerton (1982) - Semi-empirical Heerton & Wittmann (1985) Cyclic Loading: D 5 0 > Ogo.w* * based on Heerton (1982) criterion for dynamic flow Static load conditions: 5 < Cu O 9 0 < 10D50 and O90 < D9o Cy< 5 O 9 0 < 2.5D50 and O90 < D90 Fine grained cohesive material: 10D50 > O9 0,w and D 9 0 ^ O9 0,w and 0.1 mm > O9 0,w Wet sieving with sand - Based on field investigations. - In the case of dispersive soil it may be necessary to employ a more stringent design (Heerton & Wittmann, 1985). - In case of silt it can be very hard to meet. - These criteria formed the base to Luettich et al. (1992) criteria, which in turn is the source to Koerner (1998) criteria. Christopher & Holtz (1985) < 50% passing #200 D 1 5 > 0 9 5 (if the soil can move beneath the Dry sieving fabric) or 0.5D85 > O 5 0 with glass >50% passing #200 beads 0.5D85 > O 5 0 - For dynamic, pulsating and cyclic flow PIANC (1990) 0.7D90 >O90 > 0.05 mm for 5>C y D 9 0 > O 9 0 > 0.05 mm for Cu>5 Suspected Wet sieving. - Proposed as a general guideline for the design of flexible revetments CGS (1992) < 50% passing #200 095(FOS)<D15 >50% passing #200 take lesser of the following criteria: 095(FOS) < 0.5D15 095(FOS) < 0.3 mm. Hydrodynamic sieving - Should not be used for pulsing loads such as in highways and railways where the flow are small and cyclic loads are large Klein Breteler (1994) Published preliminary design criteria for geometrically open geotextile constructions on sand and cohesive soils for geotextiles with 0.1 mm < O 9 0 < 0.3 mm - Reported results shows scattering (Berensen & Smith, 1996) - Not clear whether this criteria can be used for cross section cyclic flow Mlynarek et al. (1999) 0 9 5 < A*D, or 0 9 5 < B mm and 0 9 5 > C*D) or 0 9 5 > D mm A, B, C and D are determined according to the nature of the retained soil and the severity of the hydraulic conditions Hydrodynamic sieving - Semi - empirical R. Hawley 11 Chapter 2 o V O W LU > W a: LU co Q CN CD CC CD _C O Q) _ l s CD 4^ ro to c o o o CD -*—1 ro to ro CD ro Q) O o c CD o CO CN CO a> T 3 o 1 CN csi CD 1_ _, LU 5 W LU o o a: _ LL CL H R. Hawley 12 Chapter 2 In the same fashion as the Giroud (1982) criteria, the Heerton (1982) criteria are presented in a flow chart as shown in figure 2.3. Notably, these criteria distinguish between severe dynamic conditions (severe wave attack) and mild dynamic conditions (mild water currents) by consideration of flow type. Severe conditions are given by high turbulent flow, wave attack or pumping phenomenon. Conversely, mild conditions are given by laminar flow including the change of flow direction (Heerton, 1982). 2.3 Filtration Testing Performance testing is not only necessary to assess the behaviour of the soil-geotextile system in critical applications, but is necessary to provide a basis on which to develop and validate any of the design criteria reported in section 2.3. Currently, the suite of test procedures for geosynthetic drainage products is derived from three primary sources: the American Society for Testing and Materials (ASTM), Geosynthetics Research Institute (GRI), and the US Army Corps of Engineering (COE). International standards include the International Organization of Standardization (ISO), Canadian General Standards Board (CGSB), British Standards (BS) and West German Institute of Normalization (DIN), as reported in Boschuk & Zhou (1992). From all of the available standards, the tests applicable to filtration and specifically geotextile filtration are relatively limited. Bertram (1940), who performed permeability tests on graded granular filters and recognized the need to used de-aired water in such tests, has provided the fundamental basis for performance testing of filter materials. To this day, however, there does not exist general acceptance of a single test that best evaluates soil-geotextile compatibility and performance. It is not the intent of this review to describe the details of these tests. Rather, a brief mention is provided together with references from which the reader may obtain more information. 2.3.1 Unidirectional (Steady State) Flow For unidirectional flow conditions, Koerner and Ko (1982) and Siva and Bhatia (1993) have used the Long Term Flow (LTF) test along with others. This test is intended to simulate soil-geotextile interaction, but does not reproduce field stress conditions. It does allow the collection of soil passing through the geotextile, but does not provide water head distributions and has a very small sample size. The Hydraulic Conductivity Ratio (HCR, ASTM D5567) test is intended to better simulate field conditions as it is performed in a triaxial cell, thereby allowing application of effective stresses to the sample (Luettich and Williams, 1989, Williams and Abouzakhm, 1989). However, it is R. Hawley 13 Chapter 2 i < < co CO < LU LU 2 LU CO o wu. CO LU LU i CO X : => r- I LU O V cc X Q V o A CC X Q >-fi o < X \— LU CC o < o v O CO o h-CO 0. A CL ^ < s ^  CO | _ < a: ^1 O Q < Al O Q LU Q CD >-_J cc o m v O E IT) o d v IT) _ CO > CO < LU CC CD 8 _ i LU — X E E E E ro _l m o v HAN WE d S h- cf A 7 3 LU CD o Q CC S < o o CD E E CM cn CD ro "S o '•4—; Q) =3 cu co o T5 c o o o "E ro a T3 i £ ro cu o '.4—' c cu -I—I cu o CO CN CO cn o •c CU cu X cri CN cu i ZJ g> LL R. Hawley 14 Chapter 2 best suited to soils with relatively low permeability, typically less than 5 x 10"2 cm/s. The Fine Fraction Filtration (F3) test, as researched by Hoover (1982), Legge (1990), Montero and Overmann (1990), Sansome and Koerner (1992), operates on the evaluation of different fractions of the upstream soil under the least desirable installation conditions. This test can be performed rapidly and its interpretation is relatively simple. It is appropriate for situations where intimate contact of the upstream soil and the geotextile cannot be assured. The Gradient Ratio (GR) test, initiated by Calhoun (1972) of the US Army Corps of Engineers, was adopted as a standard test method used to assess soil-geotextile compatibility in unidirectional flow conditions by ASTM in 1990. According to ASTM D5101, measurements of hydraulic head are taken and specified port locations to establish the water head distribution and hence, the variation of hydraulic gradient across the soil-geotextile system. Figure 2.4 shows a schematic diagram of the permeameter used in testing. MAKE FLOW RATE READINGS FROM THIS OUTFLOW PORT MANOMETERS PERMEAMETER CONSTANT HEAD DEVICES Figure 2.4. Gradient ratio permeameter set up diagram, ASTM D 5101 (after ASTM, 1996) R. Hawley 15 Chapter 2 With reference to Figure 2.4, the gradient ratio is defined as the ratio of the hydraulic gradients measured in the soil-geotextile section, isg, to that of the soil itself, i s . It is defined by: G R A S T M = isg/is = ise/ias = (Ah56/^56)/(Ah35/^ 35) [Eqn. 2.1] Where Ah 5 6 = differential head between ports 5 and 6 (soil-geotextile) Ah 3 5 = differential head between ports 3 and 5 (soil) £ 5 6 = vertical distance between ports 5 and 6 ^ 3 5 = vertical distance between ports 3 and 5 If GR = 1, the distribution of water head is linear, and therefore the soil geotextile system is unchanged by flow through the system; in other words, it is stable. If GR < 1, then some particles of soil have migrated through the geotextile, thereby leaving a less permeable zone close to the geotextile. In contrast, if GR > 1, a less permeable zone has developed close to the geotextile due to clogging and/or blinding of the filter. Haliburton & Wood (1982) reported that a GR of 3 is the limit above which clogging will likely occur in the field. Additional work in this area by Dierickx (1986), Scott (1980) and others has contributed to the improvement of the GR test and to the body of available test data. However, the recommendation of GR < 3, despite being cited by several agencies (FHWA, 1985, CGS, 1992), is founded on few test data and is therefore under careful examination by others. Fisher et al. (1999) conducted a systematic study of the influence of procedural variables on the ASTM gradient ratio test. They demonstrated that the methods used to prepare samples and the use of the optional procedures as recommended by ASTM, can cause large variations in gradient ratio test results. They concluded that microfilters due to recirculated water can lead to unstable permeability values during testing and this effect is dependent on the quality of water used. They also confirmed earlier work of Fannin et al. (1994a) that a chlorine algaecide reduces biological growth and consequently improves the repeatability of tests. With regard to saturation, they recommend that the permeameter must always be filled slowly with C 0 2 treatment as per ASTM guidelines to ensure saturation. Conversely, presoaking the geotextiles had no observable effect on filtration behaviour. Also, compaction of the sample produces lower permeabilities that stabilize earlier than with the loose samples recommended by ASTM. Finally, early disturbances produce marked changes in the gradient ratio values and therefore extreme care in avoiding disturbance during testing, especially in the initial stages, must be exercised. R. Hawley 16 Chapter 2 Fannin et al. (1994a, 1994b, 1996) have contributed to the development of the G R device and the interpretation of its test results in a number of ways. A series of modifications to the existing ASTM device have led to the design of the UBC modified gradient ratio device (see Figure 2.5). Fannin et al. (1994a) first contributed by the addition of an energy dissipator to the water inlet to prevent localized disturbances of the top of the soil specimen at high flow rates. Also, the use of a commercial liquid bleach is recommended as an algaecide to mitigate the development of any internal clogging. Water pluviation for uniformly graded soils, as opposed to back saturation with C 0 2 is also used to reconstitute a completely saturated, replicable homogeneous sample. In the case of broadly graded soils, a slurry deposition technique is used. Finally, three additional ports (2, 4, 6) were included to better define the variation of head loss across the sample. With these additional ports, top blinding was more easily distinguishable (port 2) as was the behaviour close to the geotextile (port 6). The latter serves to provide an enhanced index, isg, and hence a modified gradient ratio, G R M O D , that is a more sensitive indicator of soil-geotextile performance than G R A S T M (Fannin et al., 1994b). a - soil sample b - geotextile c - perforated rigid top plate d - piston e - cell top f - cell base g - collection trough h - LVDT i - ports 1 to 7 j - water inlet k - perforated rigid base plate I - permeameter cell wall m - reservoir bath n - constant head overflow tube P - vertical load Q - inlet flow rate Figure 2.5. UBC modified gradient ratio device (after Fannin et al., 1996) R. Hawley 17 Chapter 2 Subsequent modifications by Fannin et al. ( 1 9 9 6 ) introduced the ability to apply a target normal stress through a top loading plate, however, it was found that no significant variation in gradient ratio with normal stress was observed. Also, the flow regime was controlled either by a constant head or by the imposition of a constant volumetric rate. A flow pump enabled the application of either unidirectional or cyclic flow conditions. Finally, the mass of soil passing through the geotextile was collected thus providing a better ability to monitor piping of soil during sample preparation and testing. With reference to Figure 2 . 5 , G R A S T M is calculated as: G R A S T M = isg/is = i57/i35 = (A IW^^AIW^ 35) [Eqn. 2 . 2 ] The G R M O D is defined by: G R M O D = isg/is = Whs = (Ah6 7/47)/(Ah3 5/^35) [Eqn. 2 . 3 ] 2.3.2 Cyclic (Dynamic) Flow The methods presented in section 2 .2 .1 are some of those available to test soil-geotextile filtration performance in unidirectional flow. For cyclic flow, two devices have been used to represent these more extreme conditions. The Dynamic Filtration (DF) test is a very severe and accelerated test to be used to assess filter compatibility under dynamic conditions (Narejo & Koerner, 1 9 9 2 ) . The DF test is basically a flow rate (or permittivity) test that is conducted after the application cycles of dynamic pulsing. This relatively new method represents the worst hydraulic conditions that a geotextile filter can probably sustain and the authors are uncertain as to its ultimate applicability. The G R device and similar variations have also been modified to accommodate cyclic flow, where the choice is either to use flow-control or head-control. Fannin et al. ( 1 9 9 6 ) , in Vancouver, Cazzuffi et al. ( 1 9 9 6 ) , in Italy and Chew et al. ( 2 0 0 0 ) , in Singapore have all developed flow-control systems of gradient ratio testing. In Italy, Cazzuffi et al. ( 1 9 9 9 ) and Tondello ( 1 9 9 8 ) impose a cyclic flow perpendicular to the interface, a cyclic flow parallel to the interface, and vary the boundary conditions of effective stress and contact geometry between the geotextile and filter material (i.e. cover layer). The test apparatus is based on flow-control, using a pump with 'pushing units' dependent on the amount of 'discharge' required. Periods ranging from 2 to 1 0 sec, but a maximum of 2 0 sec, are possible. The maximum volume R. Hawley 18 Chapter 2 allowable for each loading cycle is approximately 12 litres for the large unit and 0.5 units for smaller unit (discharge type). A deformable or non-rigid cylinder is used to enable the application of a known and uniform effective stress to the interface. The non-rigid wall moves with the soil (virtually no side friction) and therefore very little shear stress is mobilized. This allows the base soil to transfer the vertical stress to the interface. An effective stress up to 150 kPa is possible in this system. Three small pore pressure transducers are inserted into the specimen, and any soil passing through the geotextile is collected. Cobbles represent a cover layer (as in embankment protection rip rap), placed on a rigid grid. The geotextile is placed on top of the cobbles and the base soil placed using either pluviation (uniformly-graded soil) or aerial deposition in thin strata (well-graded soil). Tondello (1998) claims that the latter method is successful in avoiding segregation of broadly graded soils. The system is back-saturated using de-aired water and the application of cycles of flow to remove the air bubbles. Figure 2.6 shows this a schematic diagram of this apparatus. To determine the number of cycles required to properly simulate long-term behaviour of the system, a battery of preliminary tests is performed with different soils, geotextiles, gradients and periods. In those tests, it was observed that after 1000/1200 loading cycles, the average interface gradient becomes constant with time. Therefore the number of cycles is set to 1500. The results are analyzed by determining the ratios of gradients i i n t to i r e f in a similar manner to the ASTM gradient ratio procedures. In this case, i i n t represents the gradient across the soil-geotextile interface comprising the geotextile and 10 cm of the base soil, while i r e f represents the gradient across 18.6 cm of the base soil. The latter gradient is measured across the middle section of the soil sample and is used as a reference gradient assumed to be unaffected by interface action. Results of this study (Tondello, 1998) show that, in almost all cases, stable interfaces can become unstable with an increase in gradient, or a decrease in effective stress. This was characterized by a proposed erosion limit state (i.e. piping failure) envelope. Tests on unstable soils at low effective stresses showed that lack of confinement results in internal migration of the fines to the top of the sample, producing unreliable test results. Findings from the Cazzuffi et al. (1996) and Tondello (1998) studies for reversing flow can be expected to guide our understanding for unidirectional flow. The apparatus described by Chew, S.H. et al. (2000) is a 'bi-directional' apparatus very similar to that developed by Cazzuffi et al. (1996). It is also a flow-controlled system that consists of three parts: a steel sample chamber, a two-way waved generator, and a water reservoir and washout collector. A pneumatic loading device is attached to the sample chamber to provide a constant overburden load during the test. The sample setup is also the same as that of Cazzuffi et al. (1996). It comprises a supporting steel grid, aggregate layer simulating protective stone layer (rip rap), geotextile filter and subsoil layer to be protected. A two-way wave generator used to apply cyclic bi-directional wave load R. Hawley 19 Chapter 2 of varying amplitude and period. The period during this particular study was 2 - 1 5 sec and a total vertical pressure of 110 kPa was applied. Fines that wash out are collected in a water reservoir and washout collector. Three pore pressure transducers are used: one just below geotextile sample, a second located 60 mm above the first, and a third 215 mm above the second. The gradient ratio is given by the ratio of i i n t to isoi|, which represent the gradients at the interface and of the subsoil respectively. At the interface, i i n t is defined as the gradient between transducers one and two in the subsoil, isoi| is the gradient between transducers two and three. In total, 1000-1500 cycles were found adequate to achieve dynamic equilibrium therefore, 1750 cycles were imposed. The apparatus is shown schematically in Figure 2.7. Figure 2.6. Cyclic flow test: Italian device (after Tondello, 1998) R. Hawley 20 Chapter 2 .— Steel Sample Chamber Loading Device Two-way Wave Generator i N 7 / Water Reservoir and Washout Collector Load Cell LVDT . Subsoil . Geotextile Filter Stone Layer Steel Grid Figure 2.7. Cyclic flow test: Singapore bi-directional flow apparatus (after Chew et al., 2000) The method of flow-control as described above is one of the two alternatives for modeling filtration applications in the laboratory, the second being the method of head-control. Head-control offers a direct comparison with field conditions in a manner that flow-control does not. Often, in the field, it is the head distribution that is known and the volumetric flow rate is a predicted quantity. The UBC Modified Gradient Ratio device was initially conceived as a flow-control device (Fannin et al., 1996), but was later modified to include the option of head-control (Fannin & Hameiri, 1999; Hameiri, 2000). The desire was to simulate those common situations where head difference or hydraulic gradient is the governing factor in filtration applications. The pump-controlled system, as discussed in section 2.2.1, is here replaced with a series of constant head tanks and a valve that allows a reversal of flow direction at any frequency and duration (see Figure 3.1). The operation is based on three tanks: an inlet tank (I), an outlet tank (O) and a tank that serves both purposes (l-O) depending on the direction of flow. This apparatus is automated using differential pressure transducers and a computerized system for process control as well as data acquisition. The automated data collection allows for readings during R. Hawley 21 Chapter 2 the crit ical p h a s e after f low reversa l , wh ich is be l ieved key to unders tand ing the potential for deve lopmen t of stabil ity (Fann in & Hamei r i , 1999). A s this dev ice w a s u s e d in this s tudy, further detai ls on the test ing appara tus are reported in Chap te r 3. Hamei r i (2000) per formed an ex tens ive ser ies of tests, us ing g lass b e a d s (uniform, broadly g raded and gap g raded) and nonwoven geotext i les in order to c o m m i s s i o n the dev ice , eva lua te exist ing des ign criteria and invest igate filtration p h e n o m e n a under both static and d y n a m i c condi t ions. G l a s s b e a d s w e r e se lec ted for that invest igat ion s ince they are essent ia l ly spher i ca l . H is observa t ions led to m a n y recommenda t i ons for future work including a mult i -stage test s e q u e n c e w h e r e b y unidirect ional s tages are used after the cyc l i c s tages to ' charac ter ize s a m p l e homogene i ty ' , and addit ional ly to conduc t tests on w o v e n geotext i les to enab le a direct c o m p a r i s o n with n o n w o v e n geotext i les. T h e s e recommenda t i ons , and the need to ex tend the s tudy to ' real ' and prob lemat ic so i ls , led to the object ives of this current study. R. Hawley 22 Chapter 2 3.0 APPARATUS AND PROCEDURES 3.1 Modified Gradient Ratio Device The apparatus used in this study is the Modified Gradient Ratio device that was designed at UBC (Hameiri, 2000). It is a modified version of the ASTM device in that it allows for the application of unidirectional or cyclic loading, imposition of a normal stress to simulate in-situ confining pressures, collection of particles passing through the geotextile, and has additional measurements of water head along the sample length for a more detailed analysis of soil/geotextile compatibility. In addition, the system is completely automated using a process control and data acquisition system, through a personal computer and applicable software. The equipment and set-up are described briefly in the following sections. However, for more information, the reader is referred to Fannin et al. (1996), Fannin & Hameiri (1999) and Hameiri (2000) for details on the development and design. 3.1.1. Apparatus Figure 3.1 shows a schematic diagram of the UBC modified gradient ratio device. A rigid-wall permeameter holds the soil and geotextile specimens and is made of 8 mm thick Plexiglas, which facilitates visual observation during testing. It accommodates a specimen of 102 mm diameter and a length of approximately 125 mm. The base and top plates are made of anodized aluminum and the loading piston is placed on top of the soil through which the normal stress is applied. Underneath, is a collection trough, which collects the soil that passes through the geotextile sample. The collection trough is made up of upper and lower section, where the upper section is made of a Plexiglas funnel with an internal slope of 45° that directs the particles passing through the geotextile into the lower section. The lower section is made of a flexible silicon tube with a diameter of 19.1 mm to facilitate the acquisition of discrete samples at any time during the test. Measurements of water head are taken at port locations as shown in Figure 3.2. Port 1 is located on the top plate to establish the water head at the top of the sample. Ports 2, 3, 5 and 6 are located 101 mm, 75 mm, 25 mm and 8 mm above the geotextile. Port 7, which establishes the water head below the sample, is located on the upper part of the collection trough. With reference to Figure 3.1, G R A S T M is calculated as: G R A S T M = isg/is = kil'hs = (Ah57/£57)l(Ah35/£35) [Eqn. 3.1] and G R M O D is calculated as: G R M O D = isg/is = 'Uhs = (Ah 6 7/^ 6 7)/(Ah 3 5/^35) [Eqn. 3.2] R. Hawley 23 Chapter 3 Axial load L V D T Permeameter Geotextile Collection trough Flow measurement Pressure transducer Figure 3.1. UBC modified gradient ratio device (after Hameiri, 2000) A x i a l load Top cover plate Y R i g i d wal l permeameter Geotext i le f l i p p e r co l lec t ion t rough , - B o t t o m plate Figure 3.2. The permeameter and port locations R. Hawley 2 4 Chapter 3 The hydraulic system was designed to model the behaviour of a soil element, where the geotextile is placed on a revetment face and is subject to steady or alternating wave action while the opposite side is subject to relatively stable hydraulic head (Hameiri, 2000). This is illustrated schematically in Figure 3.3. The hydraulic supply system comprises three constant head tanks to impose the hydraulic gradient that is controlled by H and L (see figure 3.1). De-aired water is supplied by a peristaltic pump to the top (I) tank that overflows into the middle (l-O) tank during unidirectional flow, which is subsequently driven by gradient, H/L, through the sample into the bottom (0) tank to be recirculated. During cyclic flow, a solenoid valve (Valve 1) is used to switch the direction of flow in the upward direction by driving the water from the top (I) tank to the middle (l-O) reservoir. The valve is switched at predetermined frequencies depending on imposed conditions. A collection trough captures the soil particles passing through the geotextile during the test for further analysis. Figure 3.3. Modeling the reversing flow regime in the permeameter (after Hameiri, 2000) Also shown in Figure 3.1 is the flow measurement system. It uses the overflow from the O tank during downward flow to establish the volumetric flow rate, Q, during the test. Valve 2 is used to route the water to the flow measurement tube at discrete intervals. A differential pressure transducer is used to record the volume of water passing through the sample over predetermined time intervals. Also, a linear variable differential transformer (LVDT) is used to measure any change in sample R. Hawley 2 5 Chapter 3 height during the test by monitoring the displacement of the top loading plate. Again, for details on apparatus design and technical specs of all hardware, refer to Hameiri (2000). 3.1.2 Data Acquisition System Differential pressure transducers (DPT) are used to measure directly the differential water heads between ports. Each port is also connected to a manometer, which is used only to calibrate the DPTs to allow for automated data collection. Figure 3.4a shows the set-up of the transducers to the manometers as designed by Hameiri (2000). However, during this study, the setup in Figure 3.4b was used for data collection. The data acquisition system comprises a DAS board, a desktop computer, a signal conditioning unit and a data acquisition program (LabTech Notebook). The DAS board is a multifunctional board with 12-bit high speed Analog to Digital converters, digital counters and digital Input/Output. All data are written directly to output files and are stored on the PC. Depending on the phase and its frequency of flow reversal, the frequency of readings is varied (see section 3.2.3). 3.1.3. X-ray Particle Size Analysis A Sedigraph 5100 X-ray particle size analyzer from Micromeritics Instrument Corporation was used to obtain gradations of the soil passing through the geotextile. The system comprises a particle size analyzer, a PC compatible computer and a printer. The machine operates on Stokes' Law and can therefore be used in-lieu of the traditional hydrometer method of particle size R. Hawley 26 Chapter 3 analysis for fine-grained soils. This machine offers many advantages over the hydrometer method, including excellent repeatability, speed and very small sample requirements. In addition, continuous gradations are provided and data are available directly in digital format. The mass of soil required for analysis is approximately 2 to 5 grams, depending on the specific gravity of the soil. While it is claimed to provide gradations for soils up to a maximum diameter of 300 pm, experience in this study suggests the analysis of soils with a maximum particle size greater than 150 pm is challenging. To obtain a full grain size distribution for those soils including a maximum particle size exceeding 150 pm, the soils were sieved and split between those particles passing the no. 200 sieve and those particles retained on the no. 200 sieve. Those particles passing were analyzed using the Sedigraph machine, while those retained were analyzed using dry sieving. The results of both analysis techniques were then combined to produce the complete grain size distribution. 3.2 Procedures The test procedure consists of preparing the geotextile and soil samples, setting up the apparatus, running the multi-stage test, and finally analyzing the soil, if any, that passed through the geotextile. The following sections describe these procedures in detail. The procedure followed is adapted from Hameiri (2000), with a few modifications for this specific study. 3.2.1 Sample Preparation The geotextile is first cut into a 109 mm diameter circle. This is slightly larger than the inside diameter of the permeameter to ensure a proper seal and no possibility of preferential flow paths along the edge of the sample. Following this, the sample is placed in a bath of de-aired water and squeezed manually until there is no visual observation of air bubbles. Then the geotextile is left to soak in the bath overnight to further ensure saturation. Two methods of sample reconstitution are used in this study. Water pluviation is used for one soil that is relatively uniform, whereas a slurry deposition technique is used for the more broadly graded soils (see section 4.2). Water pluviation is a technique that has been found by many researchers, including Lee & Seed (1967), Finn et al. (1971), Chaney & Mullis (1978) and Vaid & Negussey (1986) to replicate a saturated, homogeneous sample. This technique simulates the deposition of sand through water found in many natural environments and mechanically placed hydraulic fills (Kuerbis & Vaid, 1988). Slurry deposition, on the other hand, is more appropriate for broadly graded soils as it minimizes the propensity for particle segregation and inhomogeneity within the sample. R. Hawley 27 Chapter 3 The soil samples are first prepared by measuring a known dry mass of the soil (approximately 1800 g) into 500 mL flasks. Then, water is added and the resultant soil slurry is boiled to remove any of the entrapped air. The saturated soil is then allowed to cool to room temperature, which in the laboratory is consistently 23 - 24°C. 3.2.2 Test Set-Up The next step is to set up the apparatus by assembling the frame and the permeameter, connecting the ports to the manometers and transducers, pluviating or depositing (as appropriate) the soil in a slurry form into the permeameter, and finally attaching the top cover plate and the hydraulic supply system (see Figures 3.1 and 3.2). A more detailed description is given in the following paragraphs. The collection trough is filled with de-aired water and a perforated plate and wire mesh are placed on the frame. The geotextile sample is then placed on top of the wire mesh. This is done as quickly as possible to prevent any desaturation of the geotextile. The permeameter is immediately placed on top of the geotextile and frame and clamped down using three wing nuts. The permeameter is then filled with de-aired water. The reader is referred to Fannin et al. (1994) for a complete description of the test set up. The next step is to connect the ports to the manometers and transducers in sequence ensuring air-free connections. The purpose of the manometers here are only to provide a small amount of flow, which facilitates the air-free connection and allows the system to come into hydrostatic equilibrium under atmospheric pressure. It is important to note that the manometers are not used for any measurements of water head following the calibration of the transducers. The transducers are used for all water head measurements. Once the ports are connected, the soil samples are pluviated or deposited into the permeameter until a height of approximately 120 mm is reached. The transparent permeameter allows for visual observation of potential segregation that may occur in more broadly graded soils. During the pluviation process, the side of the manometer is tapped with a rubber hammer so as to provide some compaction as according to Fisher et al. (1999): compaction of the sample produces higher densities and therefore lower permeabilities. The technique allows any targeted initial density to be obtained. Once the soil is placed, a siphon is used to level the top surface to approximately 100 mm height. This action also eliminates some of the finer particles that might otherwise settle on top following pluviation. Thereafter, the top cover plate assembly, including loading plate and LVDT, is attached. In addition, the top port (No. 1) is connected to the corresponding manometer and transducer, while R. Hawley 28 Chapter 3 the inlet-outlet tank is connected to the top cover plate assembly, again, ensuring air-free connections. After these connections, the solenoid valve (valve 1, see Figure 3.1) is connected to the permeameter by running water out of the outlet tank to ensure an air free connection. Once all connections are made, the test is ready to begin. 3.2.3 Multi-Stage Test Procedure The test is run in stages: one hydrostatic (HYD) stage to gauge initial conditions, four unidirectional (UNI) stages and three cyclic (CYC) stages giving a total of eight stages. The imposed system gradient in all tests, /17, is approximately four to represent the worst condition given the physical geometric limitations of the equipment. Also, as Bertram (1940) argued, using large gradients may compensate for the short time scale of laboratory experiments over the design life of a filter. The variables in each stage are the applied normal stress, the frequency of flow reversal and the duration or the limiting condition of the stage. The limiting condition is reached once the water head measurements are observed to become stable. The applied normal stress is either zero (unconfined) or 25 kPa. If the soil is unconfined and subjected to a gradient of 4, quick conditions may manifest and lead to a loosening of the soil and an increase in permeability. The confining stress of 25 kPa was chosen to prevent quick conditions from occurring, and conversely, it was removed in the expectations that quick conditions will occur. The frequency of flow reversal is either zero (unidirectional), 0.02 Hz, or 0.1 Hz. Table 3.1 shows the testing program, sequence of each stage and the test variables. The duration of each stage was selected to provide sufficient opportunity for the sample to yield a characteristic response as evident from the measurements of water head distribution. Table 3.1. Multi-stage testing sequence Stage 1 HYD Stage 2 UNI1 Stage 3 CYC50S Stage 4 UNI2 Stage 5 CYC10S Stage 6 UNI3 Stage 7 CYC10N Stage 8 UNI4 Type Hydro-static Unidir-ectional Cyclic Unidir-ectional Cyclic Unidir-ectional Cyclic Unidir-ectional Normal Stress (kPa) 0 0 25 25 25 25 0 0 Frequen-cy (Hz) 0 0 0.02 (T= 50 sec) 0 0.1 (T=10 sec) 0 0.1 (T=10 sec) 0 Duration/ Limiting Condition 5 minutes or until stable 90 minutes or until stable 1080 cycles (15 hours) 30 minutes or until stable 260 cycles (43.3 min) 30 minutes or until stable 260 cycles (43.3 min) 30 minutes or until stable R. Hawley 29 Chapter 3 The data collection is also varied depending on the frequency of flow reversal. There are three different data collection set-ups used in this study as shown in Table 3.2. Table 3.2. Data collection set-up STAGE Sub-Stage (1-4) Frequency, f (Hz) Duration, d (sec) Elapsed Time, t (sec) Applicable Stages (1-8) 1 1 10 10 HYDROSTATIC 2 3 0.1 0.0033 50 1740 60 1800 HYD 4 0.00167 3600 5400 1 1 10 10 UNIDIRECTIONAL 2 0.1 50 60 UNI1, UNI2, 3 0.0033 1740 1800 UNI3, UNI4 4 0.00167 3600 5400 CYCLIC (0.02 Hz) 1 2 3 1 0.02 0.001 522.5 2500 54,000 522.5 3022.5 57,022.5 CYC50S 4 1 577.5 57,600 1 2 104.5 104.5 CYCLIC 2 0.2 400 504.5 CYC10S (0.1 Hz) 3 0.005 2000 2504.5 CYC10N 4 2 95.5 2600 The hydrostatic stage is used to define the initial condition, and confirm system saturation. A duration of five minutes was found to be more than adequate, as defined by water head readings that are constant with time. The first unidirectional stage (Stage 2) establishes the homogeneity of the sample, permeability, and the gradient ratio results under a hydraulic gradient of approximately 4 and zero normal stress. This stage is run for 90 minutes or until the readings are stable. Stage three is the first cyclic stage with a frequency of 0.02 Hz. One cycle takes 50 seconds, therefore, the direction of flow is switched every 25 seconds. A 25 kPa normal stress is applied prior to this stage, which runs for 1080 cycles, or 15 hours. Stage four is a unidirectional stage that is run for 30 minutes or until stable under the same normal stress of 25 kPa. This stage is used to characterize the post cyclic response immediately following the influence of the 0.02 Hz cyclic stage. Stage five is the second cyclic stage, which is run at the higher frequency of 0.1 Hz. One complete cycle takes 10 seconds, therefore, the direction of flow is switched every five seconds. This stage is run for 260 cycles, or 43.3 minutes. Stage six is another unidirectional stage, which again is run for the same purpose of characterizing the sample immediately after cyclic flow. Stage seven is the third R. Hawley 30 Chapter 3 and last cyclic stage. It is run at the same frequency and duration as its precursor (0.1 Hz for 260 cycles), however, the normal stress is set at zero to yield unconfined conditions. The final stage is another unidirectional stage with zero normal stress that is used to determine effect of the preceding cyclic stage and the post-test condition of the sample. The three cyclic stages (Stages 3, 5 and 7) are selected to impose conditions that are progressively more severe with each subsequent stage. Therefore, if a sample fails in the first cyclic stage with a normal stress, it can be assured to fail in the next higher frequency stage (with normal stress). If it fails in the second cyclic stage (Stage 5), then it can be assured to fail once the sample is unconfined in the final cyclic stage (Stage 7) and the test need not be continued. The unidirectional stages simply 'punctuate' these cyclic stages in order to establish the post-disturbance conditions. 3.2.4 Particle Size Analysis At any interval during the test, the lower collection trough (Figure 3.1) is clamped at discrete intervals to separate and collect the soil passing through the geotextile. Upon completion of the test, the soil is removed and its mass is determined. It is then taken to the X-ray particle analyzer (Sedigraph 5100 as described in Section 3.1.3) to determine the grain size analysis of the soil particles passing through the geotextile. This information provides the basis of evaluating soil retention criteria for filter design for that specific geotextile/soil combination. R. Hawley 31 Chapter 3 4.0 TEST MATERIALS Various combinations of soils and geotextiles were tested in order to assess their compatibility in filtration and drainage applications. Three soils were tested against seven geotextiles. The soils and geotextiles used are described in the following sections. 4.1 Geotextiles The geotextiles consisted of two needle-punched nonwoven and five woven materials. All are made of polypropylene fibers. They were tested at the request of the manufacturer, TC Mirafi Inc. The Apparent Opening Size (AOS) ranges from 0.212 mm to 0.600 mm. These and other physical properties are reported in Table 4.1, where it is noted that geotextiles of the same AOS values can have very different permittivity values. For example, the geotextiles with codes 140, 160 and 700 all have an AOS of 0.212 mm, but their permittivity values are 1.310, 1.192 and 0.511 sec"1 respectively. Notably, the smallest value corresponds to the woven geotextile (700). Table 4.1. Physical Properties of Geotextiles Tested (sources:a IFAI, 1999; b TC Mirafi, personal correspondence;c laboratory measured) Perm- Punc- T r a P e z o i d Grab OJ Mass / AOS a .... ., b ,-, ,-, * a Tearing Tensile/ ~ . ... o. , A O T M ittivity Perm- Flow ture 0 , . / fa _. ,. a Geotextile £ Unit ASTM y b b Strength Elongation (code) Area 0 D4751 e a b ' " t y K a t e *_J .J! ASTM ASTM U 4 4 y i U 4 H > " D4533 D4632 NW/ . . 2 N , v , -u , . > (gal/ „ ... (kN) (kN) >A/ (g/m ) (mm) (sec ) (cm/s) . , _ 2 X (kN) _,.v^ W v a ' v ' v ' v y min/ft ) v ' CMD x MD CMD x MD Mirafi HON (140) NW 287 0.212 1.310 0.290 97 0.31 0.22x0.22 0.53x0.53 Mirafi 160N (160) NW 185 0.212 1.192 0.134 88 0.42 0.27x0.27 0.71x0.71 Mirafi Filterweave 700 (700) W 218 0.212 0.511 0.021 38 0.60 0.45x0.27 1.65x1.11 Mirafi Filterweave 500 (500) W 225 0.300 0.769 0.049 57 0.60 0.47x0.62 1.22x1.74 Mirafi Filterweave 404 (404) W 282 0.425 0.881 0.080 65 0.67 0.67x0.89 1.78x1.40 Mirafi Filterweave 402 (402) W 304 0.425 2.003 0.194 148 0.47 0.51x0.33 1.62x0.89 Geolon HP570 (570) W 453 0.600 0.366 0.061 27 0.87 0.80x0.80 2.12x1.95 R. Hawley 32 Chapter 4 0 se 140 N 2mm se140N 700um at Figure 4.1a. S E M photograph: Nonwoven geotextile, A O S = 0.212 mm (Mirafi 140N) se FW 700 2mm se FW 700 700um Figure 4.1b. S E M photograph: Woven geotextile, A O S = 0.212 mm (FW700) se HP 570 5mm Figure 4.1c. S E M photograph: Woven geotextile, A O S = 0.600 mm (HP570) R. Hawley 33 Chapter 4 In Figure 4.1a, the random pattern of opening sizes is evident due to the needle-punching process in geotextile Mirafi MON. The characteristic opening size (AOS) of this geotextile is 0.212 mm: inspection of the image confirms it is representative of the 0 9 5 of the material (the opening size through which 95% of the particles during dry sieving passes through the geotextile). Figures 4.1b and 4.1c show two different types of woven geotextile with different weaves and opening sizes. The former, Mirafi Filterweave 700, has a single weave pattern and an AOS of 0.212 mm as in Figure 4.1a. Figure 4.1c shows the Geolon HP570 geotextile with an AOS of 0.600 mm. This is a different type of weave altogether, which can also influence its hydraulic and mechanical behaviour. Notably, this particular geotextile has the largest opening size and the largest strength values (see Table 4.1). 4.2 Soils Three soils were used with varying gradations and particle shapes. The gradations are shown in Figure 4.2. £ 60 i l 50 fr =* t= \ i I i i i : : : : i • FRS (Cu = 1.8) \ \\ MWT (Cu =3.3) PCS (Cu = 5.8) \ \ v X \\ \ \ \ \ • \ • - • -4 • = ^ ^ 4 -< > — X 10.000 1.000 0.100 0.010 0.001 Diameter (mm) Figure 4.2. Soil gradations The Fraser River Sand (FRS), is a naturally occurring river deposit (alluvial soil). It is a subrounded fine sand with little silt: the grain size distribution is uniform with a coefficient of uniformity (Cu) of 1.8 and a D 8 5 of 0.330 mm. The Mine Waste Tailings material (MWT), from the Highland Valley copper mine located in the Interior of British Columbia, is an angular to subangular deposit with a C u of 3.3 and a similar D 8 5 of 0.290 mm. The Port Coquitlam Silty sand (PCS), river-deposited material, is a silty sand having a C u of 5.8, and a significantly smaller D 8 5 of 0.215 mm. Two of the soils, MWT and R. Hawley 3 4 Chapter 4 PCS, were identified through discussion with geotechnical consultants as being potentially 'problematic' from a filtration standpoint and, as such, are compared against the FRS. A summary of soil gradation properties is given in Table 4.2. Also, the grain shapes are discussed in the following sections. Given the D 8 5 values of these soils and the geotextile AOS ranges as reported in section 4.1, the AOS/D 8 5 ratio for the geotextile-soil combinations range from 0.6 to 1.8 for the FRS, 0.7 to 2.1 for the MWT and 1.0 to 2.8 for the PCS. This information along with the test codes used for reporting purposes is summarized in Table 4.3. Table 4.2 Test soil properties Soil Code Soil Description D 8 5 D 6 0 D 5 0 D 3 0 D15 D-io C u G s (mm) (mm) (mm) (mm) (mm) (mm) (D6 0/D o) Uniformly FRS Graded Fine Sand, trace silt 0.330 0.280 0.260 0.220 0.170 0.155 1.8 2.50 Uniformly Graded Fine Sand, some silt MWT 0.290 0.200 0.178 0.126 0.081 0.060 3.3 2.50 PCS Broadly Graded Silty Sand 0.215 0.185 0.178 0.126 0.074 0.032 5.8 2.75 Table 4.3. Test codes (and AOS/D 8 5) of soil-geotextile combinations Geotextile Mirafi Soil 140N Mirafi 160N Mirafi Filterweave 700 Mirafi Mirafi Filterweave Filterweave 500 404 Mirafi Filterweave 402 Geolon 570 FRS F140 (0.6) F160 (0.6) F700 (0.6) F500 (0.9) F404 (1.3) F402 (1.3) F570 (1.8) MWT M140 (0.7) M160 (0.7) M700 (0.7) M500 (1.0) M404 (1.5) M402 (1.5) M570 (2.1) PCS P140 (1.0) P160 (1.0) P700 (1.0) P500 (1.4) P404 (2.0) P402 (2.0) P570 (2.8) R. Hawley 3 5 Chapter 4 In order to ascertain the internal stability of these soils, the Kenney and Lau (1985, 1986) method was applied and all soils were found to be internally stable. Appendix A contains gradations with this method applied to all three soil gradations. Some Scanning Electron Microscope (SEM) images of the soils are shown in Figure 4.3 in order to visually observe the difference in grain shape. Figure 4.3a shows the Fraser River Sand (FRS). Inspection shows the particles have nearly plane sides with unpolished surfaces but have rounded to well-rounded corners and edges. Therefore, according to the ASTM D2488, criteria for describing angularity of coarse-grained particles, the FRS soil is identified as being subangular to subrounded. The mine waste tailings (MWT) are shown in Figure 4.3b and are classified as angular to subangular. Angular particles have sharp edges and relatively plane sides with unpolished surfaces, while subangular particles are similar to the angular description but have rounded edges. The Port Coquitlam (Figure 4.3c) is classified as subrounded according to ASTM D2488, where subrounded is again defined as having nearly plane sides and well-rounded corners and edges. se F R S 1mm Figure 4.3a. SEM photograph of FRS: subangular to subrounded particles R. Hawley 36 Chapter 4 se MW11 mm Figure 4.3b. SEM photograph of MWT: angular to subangular particles se PCS 1mm Figure 4.3c. SEM photograph of PCS: subrounded particles R. Hawley 37 Chapter 4 5.0 TEST RESULTS Given the objectives of the study, namely to compare the performance of nonwoven and woven geotextiles, investigate the influence of unidirectional versus cyclic flow and assess the results against existing design guidance, the results are reported in such a manner as to facilitate these objectives. Firstly, the mass of soil passing through the geotextiles for all tests is summarized and presented. Based on these results, selected pre-test and post-test soil gradations as well as water head distributions are reported. The significant permeability values are summarized and reported, as are the values of both GRASTM and GRM0D. Two identical tests were performed in order to demonstrate the repeatability of the test procedure. Those results are also presented. A complete set of results for the pre- and post-test gradations is provided in Appendix B, together with the water head distributions in Appendix C. A tabulation of key results for each test is given in Appendix D. 5.1 Head Losses in the Permeameter Observations of water head distribution across each of the three soils used in testing have shown that as the flow rate, Q, changed, so did the total head loss across the sample (h17). This change in h17 caused the targeted system gradient (i17) also to vary, depending on the permeability of the soil. Hameiri (2000) measured this effect with a previous study, and found that as the flow rate increased (through soils of higher permeability), h17 decreased. Results of both studies are presented in Figure 5.1. The findings of this study confirm the observation that an increase in flow rate (associated with a higher permeability) yields a lower total head loss across the sample, h17. Consequently, for tests conducted with a constant head difference (H, see Figure 3.1), the actual system hydraulic gradient, i ]7 (= h17/L), is different for each soil. It is lower for soils of greater permeability. The phenomenon is attributed to flow-induced head losses in the permeameter. 0) o c tu •a ro a x ro o »0 Hz (this study) 0^.02 Hz A 0.1 Hz Q0 H Z (Hameiri, 2000) <>0.02 Hz n0.2 Hz 2 3 4 Flow Rate, Q (cm 3 / sec) Figure 5.1. Flow Rate, Q, dependent response of total head loss, h17 R. Hawley 38 Chapter 5 5.2 Particles Passing Through the Geotextile The mass of the particles passing through the geotextile during sample preparation and each stage of a test and each stage of the test was collected. These results are summarized in Table 5.1. Table 5.1. Summary of mass of soil passing during each test stage (g/m2) Test Sample UNI1 CYC50S UNI2 CYC10S UNI3 CYC10N UNI4 a (kPa) Preparation 0 25 25 25 25 0 0 f(Hz) 0 0.02 0 0.1 0 0.1 0 F140 0 0 0 0 0 0 0 0 F160 0 0 0 0 0 0 0 0 F700 0 0 0 0 0 0 0 0 F500 0 0 0 0 0 0 0 0 F404 0 0 0 0 0 0 0 0 F402 83 0 6 0 0 0 0 0 F570 213 0 40 0 0 0 0 0 M140 0 0 26 0 0 0 0 0 M160 0 0 13 7 0 0 0 0 M700 154 0 4 0 0 0 2 0 M500 163 0 60 0 0 0 0 0 M404 374 21 313 9 115 4 128 16 M402 214 23 162 0 17 0 53 0 M570 721 37 1246 0 0 0 88 0 P140 635 0 34 0 0 0 39 0 P160 33 0 56 0 0 0 0 0 P700 754 15 49 0 0 0 28 92 P500 732 65 55 0 0 0 51 0 P404 1270 75 239 0 165 0 4953 240 P402 1614 106 95 0 0 0 4349 217 P570 17756 NP NP NP NP NP NP NP Note: NP = test not performed due to continuous piping during sample preparation These results show that, for all tests on the Fraser River sand (FRS), no significant quantity of soil passed through the geotextiles. This observation recognizes the exception of a negligible amount passing through the two geotextiles with larger opening sizes, tests F402 and F570. Here, negligible is defined in terms of engineering consequences where a loss of a significant quantity of fines leads to catastrophic piping failure. Lafleur et al. (1989) set a value of 2500g/m2 as a boundary for initiation of this piping failure. Therefore, significant is defined as being a mass of soil passing greater than 2500 g/m2. The FRS sand tests were therefore relatively stable in all cases. R. Hawley 39 Chapter 5 With the Mine Waste Tailings (MWT), results show a negligible to moderate amount of soil passing, with the exception of the cyclic stage (0.02 Hz) for the test M570. Given the Lafleur et al. (1989) threshold of 2500 g/m2 for 'significant' piping, these results are relatively stable for all tests with some, although insignificant amount of piping in the latter test (M570). This test, however, soon stabilized after the first cyclic stage (CYC50S) as no further significant piping was observed. It can be noted that no additional soil passed during stages UNI2, UNI3, UNI4, or CYC10S, while a negligible amount passed during the unconfined cyclic stage, CYC10N. The Port Coquitlam silty sand (PCS) showed insignificant piping occurring with those geotextiles having smaller opening sizes (tests P140, P160, P700 and P500), see Table 5.1. Tests P404 and P402 (AOS of 0.425 mm) showed a different behaviour in that minor quantities of soil passed during the first unidirectional and cyclic stages, but significant piping took place in the later unconfined, 0.1-Hz frequency cyclic stage (CYC10N). Again, 'significant' piping is defined as greater than 2500 g/m2 as stated by Lafleur et al. (1989). Notably, for these two tests, the unidirectional stages, UNI2 and UNI3 yielded zero soil passing, whereas following the instability caused by piping during CYC10N, the UNI4 stage did show some additional movement to have occurred. The mass of soil passing during stage CYC10N in tests P404 and P402 represents 2.9 % and 2.3 % of the original samples respectively. Visual observation through the permeameter in these two tests revealed some minor loss of fines that was followed by a sudden collapse of the sample structure and large amounts of soil passing through the geotextile. It did not occur as a gradual loss of fines near the geotextile that left the remainder of the sample relatively intact. The piping action was actually preferentially located on one side of the sample. The collapse occurred in the middle third of the sample leaving a horizontal gap and a vertical channel that continued to draw soil particles preferentially down the channel through the geotextile (shown schematically in Figure 5.2). The P570 test could not be prepared, as the soil particles continuously fell through the geotextile during pluviation. p p o Figure 5.2. Preferential channeled type piping as observed during tests P404 and P402 R. Hawley 40 Chapter 5 5.3 Pre and Post Test Gradations The soil gradations before and after testing provide valuable information on the fraction of soil passing through the geotextile during the test. All pre- and post-test gradations as well as the gradation of the passing soil (where available) are provided in Appendix B. Figures 5.3a, 5.3b and 5.3c represent three scenarios of pre- and post-test gradations observed in this study. Figure 5.3a shows Scenario 1, where the plots lie essentially on top of one another. In this scenario, two explanations are valid. In the first case (Scenario 1a), the mass of soil passing is equal to zero and the sample remains unchanged. In the second case (Scenario 1b), the mass of soil passing is not zero and there is a loss of a portion of the entire size distribution. Figure 5.3b, Scenario 2, represents gradations for a sample that loses all fines below a certain size. This necessarily results in an increase in D 5 0 after testing due to the loss of its finer fraction. Figure 5.3c shows a Scenario 3, where D 5 0 decreases likely due to a loss of fines (< 75 um) and some of the matrix (75 to greater than 200 urn). Table 5.2 shows the C u, D85, D 1 5 and D 5 0 values for the mass of soil passing for those tests in which a significant amount of soil passed (during sample preparation and permeation) to facilitate using the Sedigraph X-ray analyzer to determine the grain size distributions. It was found by experience with the X-ray analyzer that any less than 3 g (400 g/m2) would not provide an adequate sample for the machine. An exception was test M570, where the particles were too coarse to analyze in the Sedigraph machine and therefore a sieve-shaker was used. Notably, in tests M404 and M402, the particles were too coarse for the Sedigraph machine and there was also insufficient sample for a sieve analysis. <D C C O u 0-100 90 80 70 60 50 40 30 20 10 0 10.000 • FRS (Cu = 1.8) - K - Post Test Gradation \ : I M —4£* •-• 1.000 0.100 Diameter (mm) 0.010 0.001 Figure 5.3a. Pre- and post-test gradations, example of Scenario 1 (a, b): D 5 0 constant (test F402) R. Hawley 41 Chapter 5 Figure 5.3b. Pre- and post-test gradations, Scenario 2: D 5 0 increases (test M700) Based on the scenarios described and the mass of soil passing data as reported in Table 5.2, the pre- and post-test gradations can be categorized. All the Fraser River sand tests had zero mass of soil passing during permeation and therefore behaved as in Scenario 1a (Figure 5.2a). The same scenario applies to tests M140 and M160. Tests M700 and M500 behaved as in Scenario 2 with a larger post-test D5 0. Tests M404, M402 and M570 show pre- and post-test gradations like Scenario 1b, however, it is noted that in M404 and M402, some additional fines are lost thereby causing a R. Hawley 42 Chapter 5 slight increase in D 5 0. In test M570, the soil passing through the geotextile basically represented the entire soil sample. In fact, it was the larger fraction of soil of the parent sample that appeared to have migrated through the geotextile. All Port Coquitlam silty sand tests behaved as in Scenario 3, where the D 5 0 decreases due to a loss in fines and some of the soil matrix. In tests P140 to P402, where there was adequate sample for analysis, the mass of soil passing represented a slightly finer fraction of the parent soil. However, the gradation of the soil passing during the sample preparation of test P570 is virtually identical to the parent soil. Notably, the post-test gradations where a significant mass of soil did pass through the geotextile are average distributions and do not represent the spatial variations with depth, but the overall change in particle size distributions. Table 5.2. Summary of gradations of mass of soil passing (where adequate sample) Test Mass passing Mass passing C u D 8 5 D 5 0 Di 5 (sample preparation) g/m2 (during permeation) g/m2 (mm) (mm) (mm) Parent Soil FRS 1.8 0.330 0.260 0.170 F140 0 0 - - - -F160 0 0 - - - -F700 0 0 - - - -F500 0 0 . - - - -F404 0 0 - - - -F402 83 6 - - - -F570 213 40 - - - -Parent Soil MWT 3.3 0.290 0.178 0.081 M140 0 26 - - - -M160 0 20 - - - -M700 154 6 - - - -M500 163 60 - - - -M404 374 606 - - - -M402 214 255 - - - -M570 721 1371 2.9 0.310 0.230 0.110 Parent Soil PCS 5.8 0.215 0.178 0.074 P140 635 73 4.4 0.067 0.033 0.012 P160 33 56 - - - -P700 754 184 10.5 0.076 0.033 0.007 P500 732 171 8.8 0.086 0.040 0.006 P404 1270 5672 6.1 0.119 0.056 0.011 P402 1614 4767 14.9 0.090 0.022 0.005 P570 17756 NP 5.9 0.210 0.180 0.076 Note: NP = test not performed due to continuous piping during sample preparation R. Hawley 43 Chapter 5 5.4 Water Head Distributions The resolution of measurement using the differential pressure transducers was determined to be ± 0.5 mm of water for a range of 0 to 60 cm of differential water head. A primary objective of the study is to characterize the response of tests that exhibited a significant quantity of piping versus those that did not. Additionally, the response to frequency of flow reversal during the cyclic stages for those tests where piping occurred is of interest. These issues are now examined with reference to the variation of water head distribution with time. For purposes of demonstration, three tests are selected that illustrate a characteristic response (F402, M570 and P402). Recall, as shown in Table 5.1, test F402 was stable throughout testing. Test M570 showed a moderate, although insignificant (Lafleur et al., 1989), amount of soil passing during the 0.02 Hz cyclic stage (CYC50S). In contrast, test P402 piped significantly during the unconfined 0.1 Hz cyclic stage (CYC10N), with additional moderate quantities passing during the UNI1 and CYC10S stages. In the interests of brevity, the water head distributions for stages of unidirectional flow in all tests are reported in Appendix C. Figures 5.4a, 5.4b and 5.4c show the respective water head distributions during the unidirectional stages (UNI1, UNI2, UNI3, UNI4) that punctuate the cyclic stages. Three different responses are represented in the plots. In Figure 5.4a (Test F402), stability is apparent in that the head distributions lie on top of one another throughout the entire testing sequence. The nearly linear shape implies a homogeneous sample. In Figure 5.4b (test M570), where some 1250 g/m2 passed during the CYC50S stage (see Table 5.1), the water head distribution for stage UNI2 has changed from that observed initially. Thereafter it remains constant. The final distribution is as expected with a gradient ratio (GR) value less than unity (both ASTM and MOD) that is associated with a loss of soil through the geotextile: the GR values are provided in section 5.5. Figure 5.4c shows test P402 where some material was observed to pass through the geotextile the first unidirectional stage (106 g/m2) as well as during the first cyclic stage (95 g/m2). In this case, this soil passing is indicative of a short-duration wash through of material associated with establishing stability as opposed to the piping phenomenon indicative of instability. This is reflected in the slightly smaller head losses observed adjacent to the geotextile in the UNI2 and UNI3 stages. After the final and faster cyclic stage with no confining stress (UNI4), the significant piping is shown by the extremely steep water head distribution in the lower third of the sample. Notably, this plot does not follow the classic shape of a 'piping' scenario as in Figure 5.4b. The response is attributed to the manner in which the PCS soil piped and subsequently failed, a point that is further discussed in Chapter 6. R. Hawley 44 Chapter 5 14 ? 1 2 o s 10 o C3 E o - -UNI1 - B - UNI2 _x-UNI3 -#-UNI4 // • -10 15 Head (cm) 20 25 Figure 5.4a. Water Head Distributions: Test F402 (i17 = 2) 14 E 1 2 o> 10 o ID o E o 0) u c <0 *-» u> D UNI1 - UNI2 - « - UNI3 UNI4 y 10 20 30 Head (cm) 40 50 Figure 5.4b. Water Head Distributions: Test M570 (i17 = 4) E 14 12 10 8 6 4 2 0 UNI1 • -UNI2 _ K _ UN 13 - # - UNI4 7// -<> " 10 20 30 Head (cm) 40 50 60 Figure 5.4c. Water Head Distributions: Test P402 (i17 = 5) R. Hawley 45 Chapter 5 In cyclic flow, the Fraser River sand (Figure 5.5) shows the distribution of water heads with time for the 0.02 Hz and 0.1 Hz stages respectively (both confined at 25 kPa). Flow reversal is controlled by the constant head tanks (see Figure 3.1), and occurs with change in direction of head loss. A positive head loss results in downward flow and, conversely a negative head loss results in upward flow (see Figure 3.3). The two distributions are very similar, implying frequency is not a dominant issue. The soil is coarse enough to allow for the pore pressures to stabilize to its steady state value. At the higher frequency (Figure 5.5b) this takes a little longer for Ah 6 7 (curve Dh67), defined in Eqh. 3.2. With the Mine Waste Tailings (Figure 5.6b), there is a similar response in that the Ah 6 7 also exhibits a slight time lag and the head difference just attains a constant value as the direction of flow is reversed. With the Port Coquitlam silty sand (Figure 5.7), the same behaviour is present, but perhaps to a greater extent due to the finer grain size distribution of this soil. The phenomenon of a time lag to stabilize the water head distribution in cyclic flow has been observed in previous studies (Hameiri, 2000). Due to this time lag, it can be said that the values are not actually in 'real time' and therefore values of gradient ratio used for analysis and interpretation of results will be based on the observations of head distribution in the stages of unidirectional flow (UNI2, UNI3, and UNI4) immediately following the imposed stages of cyclic flow (CYC50S, CYC10S, CYC10N). It is noted that flexibility of differential transducers can also cause a time lag when the flow is reversing. Also, in Figure 5.7a, Ah 1 7 and Ah 3 7 appear to be equal during upward flow (negative head losses) during stage CYC50S. This phenomenon was also observed during a similar test P404. 5.5 Permeability Based on the measurements of water head and volume of water as measured by the flow measurement system, the coefficients of permeability are reported with a resolution of ±0.4 x 10"3 cm/s. The coefficients of permeability were calculated using Darcy's law. The measurements of head loss along the sample (Ah67, Ah 5 6, Ah 3 5 and Ah13) allow the corresponding permeabilities to be deduced knowing the volumetric flow rate (Q) during any point of downward flow. Specifically, the values of interest are k35, ks7 and k@7, which describe the permeability of the soil sample, the soil-geotextile zone (ASTM procedure) and the soil-geotextile zone (UBC Modified procedure) respectively. Additionally, k17 is of interest as it represents the average permeability of the sample. These values are summarized in Tables 5.3a, 5.3b and 5.3c, for the FRS, MWT and PCS tests respectively. It is noted in the PCS series tests, a small variation in local gradients did exist, however, these small variations did not appear to significantly affect the permeability values in table 5.3c or overall sample behaviour. R. Hawley 4 6 Chapter 5 CO CO o _ l •a ro o X 30 20 _ 10 E o -10 -20 -30 Dh17 Dh37 Dh57 . Dh67 1043 1044 1045 1046 Time (min) 1047 f s ( I i L _ > • L I * L L • L • r If-— in I [ j I I ! 1 f 1048 Figure 5.5a. Cumulative Head Loss Between Ports During Stage CYC50S: Test F402 Dh17 Dh37 , Dh57 Dh67 E o CO CO o _ l •o ro CD X 30 20 10 -10 -20 -30 ! T r * fl I I I i •• > • T l . \ H j L i l T J ' ' ' , J " T i I I 1 r L d I I r • jl i 1 ******** 7 ! 1138 1138.2 1138.4 1138.6 Time (min) 1138.8 1139 Figure 5.5b. Cumulative Head Loss Between Ports During Stage CYC 10S: Test F402 R. Hawley 47 Chapter 5 925 926 927 928 Time (min) 929 930 Figure 5.6a. Cumulative Head Loss Between Ports During Stage CYC50S: Test M570 80 60 40 £ 20 cn S o T J CO | -20 -40 -60 -80 Dh17 - -Dh37 ^-Dh57 Dh67 1044 • J ! % I! \\ h -• | j y • > .* r' I fl • 1 • * ri | : 5 n • * 1044.2 1044.4 1044.6 Time (min) 1044.8 1045 Figure 5.6b. Cumulative Head Loss Between Ports During Stage CYC10S: Test M570 R. Hawley 4 8 Chapter 5 80 60 40 ? 20 o CO S o 1 -20 x -40 -60 -80 *» 927 ^ D h 1 7 - Dh37 k- Dh57 _ _Dh67 v M ./ 928 929 930 Time (min) * I f 11—m * 931 932 Figure 5.7a. Cumulative Head Loss Between Ports During Stage CYC50S: Test P402 Dh17 -Dh37 Dh57 _ . _ Dh67 1081.5 1081.7 1081.9 1082.1 Time (min) 1082.3 1082.5 Figure 5.7b. Cumulative Head Loss Between Ports During Stage CYC 10S: Test P402 R. Hawley 4 9 Chapter 5 Since in general, the first unidirectional stage (UNI1) did not induce any significant loss of soil (see Table 5.1), the value of k35 during this stage will be used as a characteristic or 'baseline' value for comparative purposes in subsequent stages. In effect, it is taken to represent the 'initial' soil permeability. Using this rationale, the average initial permeability of the FRS is approximately 25 x 10"3 cm/s (for a range of 21.9 to 27.4 x 10~3 cm/s). The average initial permeability of the MWT is approximately 1.5 x 10~3 cm/s (range of 1.0 to 2.3 x 10"3 cm/s) and that of the PCS is approximately 0.1 x 10"3 cm/s (for a range of 0.04 to 0.12 x 10"3 cm/s). In all cases, the narrow range of k35 values measured in the UNI1 stage is indicative of the homogeneity of the reconstituted soil samples. Table 5.3a. Average permeabilities (x 10"3 cm/s) for unidirectional and post-cyclic stages: FRS tests Stage UNI1 UNI2 UNI3 UNI4 Normal Stress (kPa) 0 25 25 0 Frequency (Hz) 0 0 0 0 Elapsed Time: (min) 81 1075 1161 1254 Test F140 21.8 23.1 23.1 23.5 k35 22.0 24.6 24.8 24.9 k57 26.9 25.2 24.6 25.5 k67 28.8 28.5 29.3 29.7 F160 k^ 20.6 22.2 22.3 22.7 k35 22.2 25.7 26.0 26.2 k57 21.2 20.2 19.9 20.4 k6? 23.8 24.6 25.3 25.6 F700 k^ 21.6 23.7 23.7 24.2 k35 21.9 24.6 24.7 24.9 k57 26.6 27.7 27.3 28.6 k67 27.6 28.5 29.4 29.9 F500 k^ 24.1 24.4 24.3 24.0 k35 26.5 26.8 26.7 26.6 k57 23.0 23.5 23.7 23.3 k67 19.3 19.1 20.3 19.9 F404 k^ 25.1 26.6 26.5 26.5 k35 25.5 27.7 27.7 27.8 k57 27.5 29.0 28.7 28.8 k67 23.4 27.9 28.1 28.2 F402 22.7 25.5 25.5 25.3 k35 28.2 30.8 30.6 30.4 k57 21.0 22.6 22.6 22.6 k67 18.9 22.6 23.2 23.6 F570 ki7 25.1 28.2 28.3 28.5 k3s 27.4 30.7 30.9 30.9 k57 29.1 32.4 32.2 32.6 k67 26.9 34.7 34.7 35.1 R. Hawley 50 Chapter 5 Thereafter, in each test Table 5.3a shows a nearly constant value of permeability during the respective stages of unidirectional flow. This is indicative of the relative compatibility of the materials (see Table 5.1). Virtually no soil passed through the geotextile, and therefore no significant change occurred in permeability. The greatest change was in test F570 where some, albeit negligible, loss of particles took place during stage CYC50S. In this case, the permeability of the soil adjacent to the geotextile (k67) increased by approximately 30%, from 26.9 to 34.7 x 10"3 cm/s, and remained relatively constant thereafter. Table 5.3b. Average permeabilities (x 10"3 cm/s) for unidirectional and post-cyclic stages: MWT tests Stage UNI1 UNI2 UNI3 UNI4 Normal Stress (kPa) 0 25 25 0 Frequency (Hz) 0 0 0 0 Elapsed Time: (min) 81 1075 1161 1254 Test M140 kn 1.4 2.5 2.5 2.5 k35 1.5 2.4 2.4 2.4 k57 1.4 2.7 2.6 2.6 k67 1.3 3.2 3.3 3.3 M160 k-|7 1.0 2.1 2.0 1.9 k35 1.0 2.1 2.0 2.0 k57 0.7 1.4 1.2 1.2 k67 0.9 1.6 1.5 1.6 M700 k^ 1.6 2.6 2.4 2.3 k35 1.9 2.6 2.4 2.3 k57 1.6 2.7 2.6 2.5 k67 0.9 2.1 2.1 2.2 M500 k^ 1.2 2.3 1.8 1.9 k35 1.0 2.4 1.9 1.9 k57 1.2 2.0 1.5 1.5 k67 0.9 2.6 2.5 2.6 M404 k17 1.4 2.2 2.0 2.0 k35 1.5 1.8 1.6 1.5 k57 1.6 4.0 3.9 4.5 k67 1.5 10.6 9.3 11.0 M402 ku 1.4 2.4 2.3 2.2 k3s 1.4 2.2 2.1 2.2 k57 2.0 3.8 3.7 3.4 k67 2.5 4.7 4.6 5.1 M570 k^ 2.1 3.2 3.2 3.0 k35 2.3 2.9 2.8 2.7 k57 2.2 8.3 8.0 7.2 k67 3.1 11.1 10.4 12.7 R. Hawley 51 Chapter 5 Upon inspection of Table 5.3b, it is evident that the changes in permeability of the MWT are more pronounced than with the FRS. In tests M140 and M160, the permeability of the soil-geotextile zone, k6 7, behaved as the rest of the sample. In the remainder of the tests (M700, M500, M404, M402 and M570) where a small but insignificant mass of soil was lost, the change in k 6 7 is greater than the rest of the sample. In all cases, the change took place during the CYC50S stage and appears as an increase in permeability for the UNI2 values, and is essentially constant throughout the remainder of the test. Tests M404 and M570 show the most significant increase in k6 7. Typically, the sensitivity of the k 6 7 value to change is much greater than that of the k 5 7 value (as per ASTM). Table 5.3c. Average permeabilities (x 10"3 cm/s) for unidirectional and post-cyclic stages: PCS tests Stage UNI1 UNI2 UNI3 UNI4 Normal Stress (kPa) 0 25 25 0 Frequency (Hz) 0 0 0 0 Elapsed Time: (min) 81 1075 1161 1254 Test P140 k-i7 0.12 0.06 0.08 0.06 k35 0.12 0.05 0.08 0.06 k 5 7 0.09 0.05 0.07 0.06 k67 0.13 0.07 0.10 0.07 P160 kl7 0.04 0.05 0.08 0.06 k35 0.04 0.04 0.06 0.04 k57 0.03 0.04 0.06 0.04 k67 0.04 0.05 0.08 0.06 P700 0.05 0.03 0.06 0.07 k3s 0.06 0.03 0.07 0.07 k57 0.03 0.02 0.04 0.06 k67 0.04 0.02 0.04 0.03 P500 ki7 0.06 0.09 0.06 0.05 k35 0.06 0.08 0.06 0.05 k57 0.04 0.06 0.05 0.03 k67 0.04 0.05 0.03 0.03 P404 ki7 0.04 0.03 0.04 0.13 k3s 0.04 0.03 0.04 0.14 k 5 7 0.03 0.02 0.03 0.08 k67 0.03 0.02 0.04 0.03 P402 ki7 0.05 0.05 0.06 0.18 k35 0.04 0.06 0.07 0.10 k57 0.03 0.03 0.03 0.71 k67 0.03 0.06 0.06 0.26 P570 k-i7 NP NP NP NP k35 NP NP NP NP k57 NP NP NP NP k67 NP NP NP NP Note: NP = not performed due to continuous piping during sample preparation. R. Hawley 52 Chapter 5 In the P C S test series (Table 5.3c), the P140 test showed an early decrease in permeability between stages UNI1 and UNI2. As shown in Table 5.4, this and other samples experienced an early decrease in sample length, which remained constant throughout the remainder of the test. Tests P160, P700 and P500 showed essentially no change in permeability. The small variations can be attributed to the insignificant quantities of soil passing (see Table 5.1) and the consolidation during the C Y C 5 0 S stage (see Table 5.4). P404 and P402 experience an increase in permeability, together with a significant loss of soil during stage C Y C 1 0 N (see Table 5.1). Test P570 piped continuously during sample preparation and thus was not performed. All samples, except P160, gave a sufficient mass of soil passing through the geotextile to enable collection and subsequent particle size analysis as shown in Table 5.2. In the F R S tests and the MWT tests, observations of sample length indicated no change during testing. In contrast, with the P C S , there was a notable change in sample length (see Table 5.4). Note that the strain values for tests P404 and P402 represent the change in length until stage C Y C 1 0 N , at which point the severe piping prevents any further comparison with the other tests. Due to localized piping experienced in these tests, as noted in section 5.1, it was not possible to quantify the behaviour after this point. In the absence of any significant loss of soil particles, the axial strain observed in tests P140, P160, P700 and P500 are attributed to consolidation induced during the C Y C 5 0 S stage. Table 5.4. Summary of sample length after each stage of P C S tests (mm) Test a (kPa) f(Hz) Sample Prep-aration UNI1 0 0 C Y C 5 0 S 25 0.02 UNI2 25 0 C Y C 1 0 S 25 0.1 UNI3 25 0 C Y C 1 0 N 0 0.1 UNI4 0 0 Axial Strain (%) P140 90 89 87 87 87 87 87 87 3.3 P160 104 101 98 98 98 98 98 97 6.7 P700 104 101 100 100 100 100 97 97 6.7 P500 100 98 93 93 93 93 92 92 8.0 P404 95 92 90 90 89 89 n/a n/a 6.3 P402 110 107 104 104 104 104 n/a n/a 5.5 P570 n/a n/a n/a n/a n/a n/a n/a n/a n/a Note: n/a = not available due to excessive piping R. Hawley 53 Chapter 5 5.6 Gradient Ratio It is of interest to charac te r i ze the filtration compatibi l i ty in three s ta tes : (1) unidirect ional f low, (2) cyc l i c f low with su rcharge , with interest in f low f requency in f luences (if any) , and (3) cyc l i c f low with no su rcharge . There fore , the gradient ratio resul ts are repor ted for s t ages U N I 1 , UNI3 and UNI4, thus represent ing the three states respect ive ly (see Tab le 5.5). T h e empi r ica l des ign criteria for cyc l ic f low usual ly carry the qualif ier that the geotext i le be held down and in good contact with the so i l . In this regard, it cou ld be a rgued that they apply to the C Y C 5 0 S and C Y C 1 0 S s tages (cycl ic f low with surcharge) . T h e s e results are ref lected in the UNI3 da ta , w h e r e a s the C Y C 1 0 N (cycl ic f low with no surcharge) results are ref lected in the UNI4 da ta . T h e A O S / D 8 5 and A O S / D 5 0 va lues are most common l y used in des ign criteria for unidirect ional f low, therefore, they are a lso inc luded for re ference. Simi lar ly, the A O S / D 1 5 and A O S / D 5 0 va lues are reported for cyc l i c f low condi t ions. T h e G R A S T M va lues are def ined by E q n . 2.2 and the G R M O D va lues by E q n . 2 .3 . Tab le 5.5. S u m m a r y of gradient ratio test resul ts Unidirectional Flow Post Cyclic Flow UNI1 UNI3 UNI4 Tes t A O S / D 8 5 A O S / D 5 0 A O S / • i s GRASTM GRMOD GRASTM GRMOD GRASTM GRMOD F 1 4 0 0.6 0.8 1.2 0.8 0.8 1.0 0.8 1.0 0.8 F 1 6 0 0.6 0.8 1.2 1.0 1.0 1.3 1.0 1.3 1.0 F 7 0 0 0.6 0.8 1.2 0.8 0.8 0.9 0.8 0.9 0.8 F 5 0 0 0.9 1.2 1.8 1.2 1.4 1.1 1.3 1.1 1.3 F 4 0 4 1.3 1.6 2.5 0.9 1.1 1.0 1.0 1.0 1.0 F 4 0 2 1.3 1.6 2.5 1.3 1.5 1.4 1.3 1.3 1.3 F 5 7 0 1.8 2.3 3.5 0.9 1.0 1.0 0.9 0.9 0.9 M 1 4 0 0.7 1.2 2.6 1.1 1.2 0.9 0.8 0.9 0.7 M 1 6 0 0.7 1.2 2.6 1.4 1.1 1.6 1.3 1.6 1.2 M 7 0 0 0.7 1.2 2.6 1.2 2.0 0.9 1.1 0.9 1.0 M 5 0 0 1.0 1.7 3.7 0.9 1.1 1.2 0.8 1.2 0.7 ' M 4 0 4 1.5 2.4 5.2 0.9 0.9 0.4 0.2 0.3 0.1 M 4 0 2 1.5 2.4 5.2 0.7 0.6 0.6 0.5 0.6 0.4 M 5 7 0 2.1 3.4 7.4 1.0 0.7 0.3 0.3 0.4 0.2 P 1 4 0 1.0 1.2 2.9 1.2 0.9 1.0 0.8 1.0 0.8 P 1 6 0 1.0 1.2 2.9 1.3 0.9 1.0 0.7 1.0 0.7 P 7 0 0 1.0 1.2 2.9 1.6 1.2 1.5 1.7 1.2 2.0 P 5 0 0 1.4 1.7 4.1 1.3 1.5 1.4 1.9 1.4 1.9 P 4 0 4 2.0 2.4 5.7 1.4 1.2 1.3 1.1 1.7 4.7 P 4 0 2 2.0 2.4 5.7 1.4 1.3 2.2 1.2 0.1 0.4 P 5 7 0 2.8 3.4 8.1 N P N P N P N P N P N P Note : N P = not per formed R. Hawley 54 Chapter 5 Visual inspection of Table 5.5 allows for comparison of G R A S T M and G R M O D as an index of soil-geotextile compatibility. In general, when G R = 1, it was found that G R A S T M was essentially equal to G R M O D - In most cases when G R » 1, G R M O D was greater than G R A S T M - Also, when G R « 1, then G R M O D was less than G R A S T M - Hence, G R M O D is considered a more sensitive index of both change and compatibility. The F R S test results show that in unidirectional flow (UNI1), G R values are approximately equal to unity and the tests are therefore stable. In the case of cyclic flow with surcharge (UNI3), the G R values again are approximately equal to one and the tests are stable. In cyclic flow with no surcharge (UNI4), the same trend is apparent and tests are stable. This is confirmed by the essentially zero mass of soil passing data in Table 5.1. The MWT results show that in unidirectional flow (UN11), the G R values are close to unity and the tests are again stable. In cyclic flow with surcharge (UNI3), the G R values decrease with increasing A O S / D n ratios. Therefore, it appears that there is an increasing instability with geotextile A O S . In cyclic flow with no surcharge (UNI4), the same trend exists as in UNI3 and there is little difference in G R values. Therefore, the removal of the surcharge appears to have little effect on the response of the MWT tests under cyclic flow. The data in Table 5.1 confirm these results as most of the loss of soil occurred during stage C Y C 5 0 S . The P C S test results indicate that in unidirectional flow (UNI1), the tests are stable with G R values close to unity. In cyclic flow with surcharge (UNI3), the tests are relatively stable, with most G R values greater than one. Data in Table 5.1, however, show that there is some soil loss in the C Y C 5 0 S stage. In cyclic flow with no surcharge (UNI4), there appears to be a variable response with no strong pattern like that in the MWT tests. In Table 5.1, soil does pass during the C Y C 1 0 N stage; therefore, it appears that with the P C S tests, removal of surcharge does significantly affect performance especially with the larger A O S / D n values. 5.7 Repeatabi l i ty In order to demonstrate that the test results are repeatable and therefore reliable with respect to sample preparation and quality of data, two separate tests with identical conditions were performed. The results of these tests, the F160, are summarized in Tables 5.6a and 5.6b. Figures 5.7a and 5.7b show the water head distributions of both tests F160(a) and F160(b) respectively. It is evident from the results shown in Tables 5.4a and 5.4b as well as Figures 5.7a and 5.7b, that the gradient ratio test is repeatable and the tests are therefore reliable. It is noted, however, that the R. Hawley 55 Chapter 5 permeability values reported during the cyclic stages are measured only during downward flow. As mentioned previously, the time lag phenomenon that occurs during these transient conditions do not allow for reliable determination of water heads in real time and therefore these measurements (during stages CYC50S, CYC10S and CYC10N) are mainly for comparison purposes for assessment of repeatability. Table 5.6a. Test results: TestF160(a) Stage El. Time (min) Ah 6 7 (cm) Ah 5 7 (cm) Ah 3 7 (cm) Ah 1 7 (cm) Ah 3 5 (cm) Gradient Ratio Permeability (x 10"3 cm/sec), k17 Distance from Geotextile, cm - 0.8 2.5 7.5 9.9 - 5.0 ASTM MOD -UNI1 81 1.6 5.5 15.9 22.3 2.3 10.5 1.0 0.9 2.1 CYC50S 941 1.5 2.4 11.7 17,7 1.8 6.0 0.8 1.6 2.9 UNI2 993 1.5 5.9 15.2 21.3 2.2 9.3 1.3 1.0 2.2 CYCIOs 1052 1.6 6.0 15.2 21.0 2.1 9.2 1.3 1.1 3.0 UNI3 1076 1.5 6.0 15.3 21.3 2.2 9.2 1.3 1.0 2.2 CYC10N 1137 1.6 6.0 15.2 21.1 2.1 9.2 1.3 1.1 3.0 UNI4 1160 1.5 5.9 15.1 21.0 2.1 9.2 1.3 1.0 2.3 Table 5.6b. Test results: Test F160(b) Stage El. Time (min) Ah 6 7 (cm) Ah 5 7 (cm) Ah 3 7 (cm) Ah 1 7 (cm) Ah 3 5 (cm) Gradient Ratio Permeability (x 10"3 cm/sec), k17 Distance from Geotextile, cm - 0.8 2.5 7.5 10.0 - 5.0 ASTM MOD -UNI1 81 1.6 5.5 16.0 23.1 2.3 10.5 1.0 1.0 2.1 CYC50S 941 1.6 2.3 12.2 19.5 2.0 7.3 0.6 1.4 2.8 UNI2 993 1.6 5.9 16.2 22.3 2.2 10.3 1.1 0.9 2.2 CYCIOs 1052 1.6 5.9 15.5 24.1 2.4 9.6 1.2 1.0 3.0 UNI3 1076 1.5 6.0 15.3 23.2 2.3 9.3 1.3 1.0 2.1 CYC10N 1137 1.6 6.1 15.5 22.2 2.2 9.4 1.3 1.1 3.0 UNI4 1160 1.5 5.9 15.2 22.1 2.2 9.3 1.3 1.0 2.2 R. H a w l e y 56 C h a p t e r 5 14 12 H 10 8 O o 6 C J UNI1 - UNI2 -x-UNI3 - -.-UNI4 M y 10 15 Head (cm) 20 25 Figure 5.8a. Water head distribution of unidirectional stages: Test F160(a) 14 12 a 10 0> o C5 E o re 4 UNI1 . UNI2 -x-UNI3 - _#_UNI4 •*/ A u y y 10 15 Head (cm) 20 25 Figure 5.8b. Water head distribution of unidirectional stages: Test F160(b) R. Hawley 57 Chapter 5 6.0 ANALYSIS O F T E S T R E S U L T S The test results presented in Chapter 5 are used to address two issues of filtration compatibility: the influence of geotextile type and the influence of flow regime. Additionally, the results are compared with selected design criteria for both unidirectional and cyclic flow conditions, in order to examine their success in characterizing the behaviour of the soil-geotextile combinations used in testing. 6.1 Influence of Geotextile Type Three of the geotextiles, one woven and two nonwoven, have the same AOS value of 0.212 mm. Therefore the results of nine tests are examined, comprising the three geotextiles and three soils (FRS, MWT, PCS). The results are summarized for the stages of initial unidirectional flow (UNI1) and post-cyclic unidirectional flow (UNI4) in Table 6.1. The mass of soil passing is the cumulative quantity throughout the tests (excluding sample preparation). Table 6.1. Comparison of filter performance: nonwoven versus woven Unidirectional Flow (UNI1) Post Cyclic Flow (UNI4) Test (Geo-textile AOS/ D 8 5 AOS/ D 5 0 AOS/ D 1 5 Soil Passing, Gradient Ratio So/7 Passing, Gradient Ratio Type) mp mp g/m2 ASTM MOD g/m2 ASTM MOD F140 (NW) 0.6 0.8 1.2 0 0.8 0.8 0 1.0 0.8 F160 (NW) 0.6 0.8 1.2 0 1.0 1.0 0 1.3 1.0 F700 (W) 0.6 0.8 1.2 0 0.8 0.8 0 0.9 0.8 M140 (NW) 0.7 1.2 2.6 0 1.1 1.2 26 0.9 0.7 M160 (NW) 0.7 1.2 2.6 0 1.4 1.1 20 1.6 1.2 M700 (W) 0.7 1.2 2.6 0 1.2 2.0 6 0.9 1.0 P140 (NW) 1.0 1.2 2.9 0 1.2 0.9 73 1.0 0.8 P160 (NW) 1.0 1.2 2.9 0 1.3 0.9 56 1.0 0.7 P700 (W) 1.0 1.2 2.9 15 1.6 1.2 184 1.2 2.0 The FRS test series, where AOS/D8 5 and AOS/D5 0 are 0.6 and 0.8 respectively, yields a similar behaviour regardless of geotextile type. No soil was lost during either unidirectional or cyclic flow. In the MWT test series, M140, M160 and M700 also resulted in no loss of soil during unidirectional flow, R. Hawley 58 Chapter 6 and subsequent cyclic loading yielded an insignificant amount of wash through (26, 20 and 6 g/m2 respectively). The PCS test series, which had a larger ratio of AOS/D8 5, reveals a slight difference between the nonwoven and woven geotextiles. The nonwoven tests (P140 and P160) showed no soil passing in the UNI1 stage and a small amount passing by the end of the test (UNI4). In the woven test (P700), the quantity of soil lost was moderately greater (15 g/m2 and 184 g/m2) than that in the two companion tests. Grain size curves of the soil passing are available for tests P140 and P700 (see Appendix B). Inspection shows the maximum size and size distribution for P140 (D9 5 ] P = 0.095 mm, D 5 p = 0.007 mm) to be smaller than that for P700 (D95,p = 0.140 mm, D5,p = 0.002 mm). The results of this comparison, therefore, indicate that these particular woven and nonwoven geotextiles behave similarly in that the filtration compatibility is equally satisfied. In all tests, the quantity of soil passing is not considered significant, based on the criterion of Lafleur et al. (1989). There is a trend with the results that suggests increasing washout of soil with greater AOS/D 8 5 ratio, which is to be expected (see section 6.3). The limited data on the size distribution of these soils indicate the woven geotextile has potential to pass more soil than the nonwoven of equal AOS, and that both the maximum particle size and the size distribution of the passing soils are larger. 6.2 Influence of Flow Regime Four flow regimes are sequentially imposed in testing: unidirectional flow, cyclic flow with surcharge (0.02 Hz and 0.1 Hz) and cyclic flow with no surcharge (0.1 Hz). Comparison of the results allows an assessment of the influence of flow regime on soil-geotextile compatibility. The variation of mass of soil passing with AOS/D8 5, for each of these stages is illustrated in Figures 6.1a, 6.1b and 6.1c for the FRS, MWT and PCS soils respectively. The values of modified gradient ratio are reported in Figure 6.2 from the corresponding unidirectional stages (UNI1 to UNI4). For the Fraser River Sand (Figure 6.1a), the mass of soil passing for all AOS/D 8 5 combinations is negligible. The GR values (Figure 6.2) show no change between the unidirectional (UNI1) and post-cyclic responses (UNI2, UNI3, UNI4). In conjunction with the permeability data (see Table 5.3a), the relatively consistent values of k67 indicate that neither piping nor clogging were prominent issues. Filtration compatibility exists in all stages of testing for the Fraser River Sand. Results for the Mine Waste Tailings are given in Figure 6.1b. The response is stable in all stages except the 0.02 Hz stage (CYC50S). As the AOS/D 8 5 increases, so does the mass of soil passing through the geotextile. Notably, the gradation of the mass of soil passing is similar to that of the R. Hawley 59 Chapter 6 parent soil, therefore, it is the soil structure that passed through the geotextile. This loss of material does not continue in the subsequent unidirectional or cyclic stage with no surcharge. This result is supported by reduction in gradient ratio (see Figure 6.2b) for the UNI2 stage, and an increase in permeability that is consistent throughout the remainder of the test (Table 5.3b, test M570). It appears counter intuitive that in this test, soil passed during the CYC50S stage but not the CYC 10S or CYC 10N stages. Upon inspection of permeability data (see Appendix D), it appears that the overall sample permeability (k17) was higher during stage CYC50S by a factor of 3 as compared to stages CYC10S and CYC10N. The permeability (k 6 7) was higher by a factor of 5. Perhaps the slower frequency of flow reversal (0.02 Hz) allowed enough time for the matrix material to migrate out of the sample during downward flow and hence, increase the permeability of the soil-geotextile zone and the sample overall. During the faster frequency stages (0.1 Hz), there was perhaps insufficient time for the matrix soil to get carried out of the geotextile before the flow direction was reversed. The Port Coquitlam Silty Sand behaves differently (Figure 6.1c). A relatively insignificant amount of soil passed in all stages but CYC10N, during which the cyclic flow was unconfined at 0.1 Hz, and a significant piping failure took place. The response appears consistent, since it is observed in two tests (P404 and P402, see Table 5.1), each with the same AOS/D 8 5 ratio. The quantity of soil passing was the greatest observed of all tests performed. In contrast to other tests, the onset was relatively sudden once the AOS/D 8 5 exceeded a value of 1.4. The corresponding modified GR values for these two tests do not show a consistent behaviour (see Table 5.5 and Figure 6.2b). In test P402, the value of 0.4 indicates piping (post-cyclic). However, test P404 gave a value of 4.7, which theoretically, indicates a tendency towards clogging behaviour. CN E oi a E c CO CO rc Q_ O co CO co ro 2 5600 4800 4000 3200 2400 1600 « 800 0 | - H - U N I 1 ^ 4 ^ U N I 2 _ # _ U N I 3 O U N I 4 • — U 0.0 0.5 1.5 2.0 1.0 AOS/D85 Figure 6.1a. Influence of flow regime: FRS tests, i 1 7 = 2 R. Hawley 60 Chapter 6 5600 E 4800 Q. E 4000 C cn cn 3200 ro D. o 2400 — o cn 1600 cn ro 800 ro o H T| j UNI1 H_±_UNI2 - 0 - U N I 3 | | U N I 4 4 A ft 0.0 0.5 1.0 1.5 AOS/D85 2.0 2.5 Figure6.1b. Influence of flow regime: MWT tests, i 1 7 = 4 £ CL E cn c cn cn ra CL O CO cn cn ro ro +-> o 5600 4800 4000 3200 2400 1600 800 0 0.0 0.5 1- A L JNI1 JNI2 JNI3 C / ) • L / ) JNI4 / / / / / / t M — i i 1.0 1.5 AOS/D85 2.0 2.5 Figure 6.1c. Influence of flow regime: PCS tests, i 1 7 = 5 R. Hawley 61 Chapter 6 CXI I Q. E ro c CO CO ro 0_ O CO CO to ro 7000 6000 5000 4000 3000 2000 * 1000 4, FRS-UNI1 A MWT-UNI1 0 PCS-UNI1 — C b g g i n g Criteria: GR< 3 - - - Piping/Clogging Threshold: G R = 1 -La f leur Piping Limit: < 2500 g/m2 • I i • — — — — - - — — — - - — • — — I ! GR MOD CN E ro Q. E ro C CO CO ro a o CO CO CO CO o I-9000 8000 7000 6000 5000 4000 3000 2000 1000 Figure 6.2a. Unidirectional results: G R M O D ^ FRS-UNI2 A M W T - U N I 2 # P C S - U N I 2 0 F R S - U N I 3 A MWT-UNI3 o P C S - U N I 3 O FRS-UNI4 A MWT-UNI4 O P C S - U N I 4 - - . Piping/Clogging Threshold: G R = 1 _ — L a f l e u r Piping Limit: 2500 g/m2 — C l o g g i n g Criteria: GR< 3 o o -— — — — — — — -— — — A 1 2 3 GRMOD Figure 6.2b. Cyclic results: G R M O D R. Hawley 62 Chapter 6 The two tests appear to have a minor anomaly (based on the GR values, which in turn are based on measurement of head loss). Yet, both tests yield very similar quantities of passing soil for each stage (see Table 5.1 and Figure 6.1c). From this point of view, they seem quite 'reproducible', given the same AOS values. Upon comparison of the pre- and post-test gradations as well as the gradation of the soil passing (see Table 5.2), it appears that P402 passes a finer soil (D50,p = 0.022 mm) compared with test P404 (D50,p = 0.055 mm). This explains the GR < 1 for test P402, where k67 increased significantly. The GR > 1 in test P404 occurs due to a dramatic increase of k35 while k67 remained virtually unchanged. Both tests are similar with respect to the overall sample permeabilities (k17), the quantity of soil passing and the flow rate measurements. The difference lies in the subtlety of the locations of head measurement since both the gradient ratio and permeability values are determined based on head measurements. As described in section 5.1, a horizontal gap and vertical channel type failure was observed in both tests, where the channel occurred preferentially to one side. It is reasonable to speculate that in test P404, the horizontal gap occurs between ports 3 and 5, thus resulting directly in an increase in permeability k35 of a factor of approximately 3.5. It also seems reasonable that the vertical channel did not occur on the side where port 6 is located, which incidentally is located on the opposite side of the permeameter to ports 3 and 5. This would result in the somewhat misleading measurement of a head values, and hence permeability value (k67) that remains unchanged. This failure pattern is shown schematically in Figure 6.3a. It also seems reasonable that in test P402, the horizontal gap is located slightly higher (note that the sample height is also greater than in test P404), leading to a smaller increase in permeability (k35) by a factor of 1.3. In contrast to test P404, the vertical channel may have developed where port 6 is located and hence results in the measurement of a dramatic increase in head and hence permeability (k67) by a factor of approximately 4.3. This is shown schematically in Figure 6.3b. Therefore, despite the overall reproducibility of the two tests with respect to mass of soil passing and volumetric flow rate, the head measurements are perhaps not capturing the true behaviour of the sample thereby resulting in misleading GR and permeability values. R. Hawley 63 Chapter 6 14 12 E 10 Port 6 k17 = 1.3 x 10 cm/s Q = 0.05 cm3/s Port 3 Port 5 Figure 6.3a. Test P404: Failure pattern during stage CYC1 ON 14 12 ^ k v = 1.8 x 10'' cm/s Q = 0.06 cm3/s 20 30 Head (cm) 40 50 Port 3 Port 5 Figure 6.3b Test P402 Failure pattern during stage CYC1 ON R. Hawley 64 Chapter 6 6.3 Design Criteria The results are used to assess existing design criteria, namely the CGS (1992), Luettich et al. (1992), and Holtz et al. (1997) which are representative of those most widely used in practice. Table 6.2 shows the design criteria as they apply to the soils in this study, for both unidirectional and cyclic flow. The CGS (1992) criteria for unidirectional flow, based on the C u of the soil, are extracted from previous Christopher & Holtz (1985) work and are the same as the Holtz et al. (1997) guidance for steady state flow conditions. The Luettich et al. (1992) criteria are obtained using Figure 2.2, which originate from the Giroud (1982) soil retention criteria for steady state flow. The criteria are calculated based on the D10, D 6 0 and D 3 0 values (see Table 4.2) of each soil and the assumptions that the application favours retention, the soil is stable and is medium dense. For cyclic flow, the CGS (1992) guidance states that soil with < 50% passing the No. 200 sieve be designed simply with the criterion O g 5 < D15. The Luettich et al. (1992) criterion is determined based on Figure 2.3, which originates from the work of Heerton (1982). The criterion is based on the D 1 0 of the soils and assumes severe wave attack. The Holtz et al. (1997) guidance considers that if the geotextile is not properly weighted down and in intimate contact with the soil to be protected, or if dynamic, cyclic, or pulsating loading conditions produce high localized hydraulic gradients, then soil particles can move behind the geotextile. Thus, the use of B = AOS/Dn = 1 is not conservative, because the bridging network will not develop and the geotextile will be required to retain even finer particles. When retention is the primary criteria, B should be reduced to 0.5. The calculations for all criteria used in this study are provided in Appendix A. Figures 6.4 and 6.5 summarize the mass of soil passing against the appropriate AOS/Dn value, for unidirectional and cyclic flow respectively. Table 6.2. Design criteria used in this study Soil Cu Unidirectional Flow Cyclic Flow CGS Luettich et al. CGS Luettich et al. Holtz et al. (1992) (1992) (1992) (1992) (1997) FRS 1.8 AOS < 1.0 D 8 5 AOS < 1.9 D 5 0 AOS < 1.0 D 1 5 AOS < 1.0 D 5 0 AOS < 0.5 D 8 5 MWT 3.3 AOS < 1.7 D 8 5 AOS < 2.4 D 5 0 AOS < 1.0 D 1 5 AOS < 1.0 D 5 0 AOS < 0.5 D 8 5 PCS 5.8 AOS < 1.4 D 8 5 AOS < 2.2 D5o AOS < 1.0 D 1 5 AOS < 1.0 D 5 0 AOS < 0.5 D 8 5 The results are assessed on two bases: (i) test performance and (ii) conformance to specification. Test performance, or filtration compatibility, is based on the mass of soil passing where perfect compatibility includes the necessary action of a little 'wash-through' the geotextile in order to set-up a bridging zone immediately upstream of the geotextile. The intersection of the criterion and the mass of soil passing provides an indication of the overall conformance to design guidance. R. Hawley 65 Chapter 6 6.3.1 Unidirectional Flow For unidirectional flow (UNI1) results, the mass of soil passing is plotted against the appropriate AOS/Dn value for the CGS (1992) guidelines and Luettich et al. (1992) guidelines in Figures 6.4a and 6.4b respectively. The Fraser River sand has been shown to have a perfect filtration compatibility with all of the geotextiles used in testing. From Figure 6.4a, four of these tests conformed to the CGS criteria and three did not, namely those where the AOS/D8 5 was 1.3 and 1.8 (see Table 4.3). The absence of soil passing suggests that the CGS guidance may be slightly conservative. The Luettich et al. (1992) criteria, see Figure 6.4b, show all combinations but one (AOS/D50 of 2.3) to be in conformance. Therefore, the Luettich et al. (1992) guidance may also be slightly conservative. In the case of the mine waste tailings (Figure 6.4a), 6 results conform to CGS criteria, and one does not (AOS/D85 of 2.1). The intersection of the CGS criterion with the data occurs at a mass of soil passing of approximately 30 g/m2. Figure 6.4b shows that all tests conform to Luettich et al. (1992) criterion, except one test (AOS/D50 of 3.4), which is associated with approximately 20 g/m2 of soil passing. Therefore, the criteria again appear to be slightly conservative. The Port Coquitlam silty sand yielded 4 tests in conformance with CGS (1992) criteria and 3 that did not, including one not shown that resulted in catastrophic piping during the sample preparation stage (test P570: AOS/D 8 5 = 2.8 and AOS/D5 0 = 3.4). From Figure 6.4a, the intersection of the criterion with the data occurs at approximately 50 g/m2 of soil passing. It therefore appears the criterion may be slightly conservative. Similarly, Figure 6.4b shows all tests except the same three as above conformed to Luettich et al. (1992) specifications. The intersection of the criterion and experimental data occurs at approximately 80 g/m2 of soil passing. Therefore, the criterion appears very suitable for characterizing the filtration compatibility of the Port Coquitlam silty sand under unidirectional flow. All 'best fit' lines in Figures 6.4a and 6.4b are established by inspection. R. Hawley 66 Chapter 6 F R S M W T P C S C G S (1992) Criterion: F R S — C G S (1992) Criteria: MWT - - - C G S (1992) Criteria: P C S 4.0 Figure 6.4a. Unidirectional results: CGS (1992) guidelines 120 Figure 6.4b. Unidirectional results: Luettich et al. (1992) guidelines R. Hawley 67 Chapter 6 6.3.2 Cyclic Flow For three stages of cyclic flow (CYC50S, CYC 10S and CYC10N), the mass of soil passing is plotted against the appropriate AOS/Dn value for the CGS (1992), Luettich et al. (1992), and Holtz et al. (1997) guidelines in Figures 6.5a, 6.5b and 6.5c respectively. Most design guidance in cyclic flow emphasizes the need for 'contact' (confining stress) between the geotextile and the soil. Holtz et al. (1997) includes this as rational for a more conservative design guidance for cyclic flow as noted earlier in section 6.3. The cyclic flow regime is divided into three stages, two stages with surcharge, CYC50S and CYC10S and cyclic flow with no surcharge (CYC10N). The stages CYC50S and CYC 10S are grouped together as being 'confined' and CYC10N is 'unconfined'. Based on these test results, a confined envelope and an unconfined envelope are shown in Figures 6.5a, 6.5b, and 6.5c to allow for interpretation of the data. Figure 6.5a shows that all confined data exceed the CGS (1992) criterion. The confined envelope suggests that AOS/D1 5 < 5 for a very conservative limit of mp < 250 g/m2. Figure 6.5b shows that most data, except three FRS tests with an AOS/D5 0 of 0.8, exceed the Luettich et al. (1992) criterion. Again, for mp < 250 g/m2, the confined data suggest AOS/D5 0 < 2. Similar to the CGS (1992) criterion, Figure 6.5c shows that all data exceed the Holtz et al. (1997) criterion and for mp < 250 g/m2, the data suggest AOS/D8 5 < 1. Figures 6.5a, 6.5b and 6.5c show the proposed 'unconfined envelope' for this series of tests. However, interpretation of this envelope is difficult since the PCS is the only test series that yielded significant quantities of soil passing, and only one test (P500) falls between the AOS/Dn lower and upper limits of the other tests. It is noted, however, that none of the three design guidances claim to address unconfined conditions aside from the mention of ensuring adequate weight and providing intimate contact between the soil and geotextile. R. Hawley 68 Chapter 6 6000 E 5000 • F R S 4 M W T C Y C 5 0 S • P C S O F R S A M W T C Y C 1 0 S 0 P C S O F R S A M W T C Y C 1 0 N 0 P C S C G S (1992) Cr i ter ion 0 / / o U N C O N F I N E D E N V E L O P E C O N F I N E D E N V E L O P E 0.0 1.0 2.0 3.0 4.0 5.0 A O S / D -a R 6.0 7.0 8.0 Figure 6.5a. Cyclic results: CGS (1992) guidelines 6000 U N C O N F I N E D E N V E L O P E ' o F R S 4 M W T C Y C 5 0 S © P C S F R S A M W T C Y C 1 0 S o P C S o F R S A M W T C Y C 1 0 N O P C S • Leut t ich et a l . (1992) Cr i ter ion 0.0 0.5 C O N F I N E D E N V E L O P E 1.0 1.5 2.0 2.5 A 0 S / D 5 n 3.0 3.5 4.0 Figure 6.4b. Cyclic results: Luettich et al. (1992) guidelines R. Hawley 69 Chapter 6 6000 J 5000 ED a. £ 4000 in S. 3000 o CO 10 w ro 2000 « 1000 o 0.0 • F R S fl MWT • PCS 0 FRS A MWT O PCS 0 FRS A MWT 0 PCS •Holtz et al. (1997) Criterion o U N C O N F I N E D E N V E L O P E C O N F I N E D E N V E L O P E 0.5 1.0 1.5 A O S / D „ c fl 4 2.0 2.5 Figure 6.5c. Cyclic results: Holtz et al . (1997) guidelines R. Hawley 70 Chapter 6 7.0 IMPLICATIONS FOR PRACTICE The implications of this study for the application of geotextiles in geotechnical filters lie in assessing the results against existing design guidance. Additionally, the results contribute to a very limited database of existing studies. The limitations of testing must also be addressed in this section, as it is important that the practitioner understand the applicability of these results. 7.1 Design Guidance Results from testing were assessed against several commonly-used design guidance for unidirectional and cyclic flow: the C G S (1992), Holtz et al. (1997), and Luettich et al. (1992). The guidance is assessed using three soils, all of which are internally stable, which represent a wide range of gradations (Cu = 1.8, 3.3, 5.8) and AOS/D8 5 values (0.6 to 2.8). In unidirectional flow, the CGS (1992) criteria are acceptable in most cases, but since the quantities of soil passing are so small, they appear somewhat conservative. Luettich et al. (1992) guidance is also acceptable and, similar to CGS (1992) appears somewhat conservative given comparably small quantities of soil passing. In cyclic flow (with surcharge), all guidance appear very conservative, being associated with negligible to very small amounts soil passing. Therefore, they are acceptable as the results support their use. There may be reason to permit an increase to the AOS/Dn ratio, and still expect an adequate compatibility. From this study, there is no clear evidence to assess whether it is more advantageous to select D15, D5 0, or D 8 5 for design purposes. Cyclic flow (with no surcharge) results in a greater passage of soil that appears to develop very significantly over small increments of AOS/Dn. This is immediately apparent from the much steeper unconfined envelope in comparison to the confined envelope (see Figure 6.5). The results confirm this scenario to be problematic in design. Generally, the guidance is suitable for both nonwoven and woven geotextiles, recognizing the subtle variation in gradation of the soil that passes. Additionally, the frequency of cyclic flow was not found to be influential for the range examined in this study (0.02 Hz to 0.1 Hz). 7.2 Limitations of Testing Cazzuffi et al. (1999) and Tondello (1998) have published test results using a similar gradient ratio apparatus (see section 2.2.2). They were able to directly apply a known effective stress R. Hawley 71 Chapter 7 to the interface by using a deformable cylinder and a non-rigid wall at the interface. Therefore, they were able to vary the effective stress systematically and assess its influence on soil-geotextile compatibility. They applied effective stresses ranging from 0 to 150 kPa. At zero effective stress, the soil reached instability much more quickly and at effective stresses above 100 kPa, regardless of applied gradient, the soil was stable. However, it is noted that only two soils were tested in their study. These findings are similar to the relatively limited data presented in this study, where the lower the 'applied stress', the greater the tendency toward instability in cyclic flow. With respect to frequency effects in cyclic flow, Tondello (1998) and Cazzuffi et al. (1999) state that they were able to vary the frequency between 0.02 Hz and 0.5 Hz; however, they did not quantify the frequency effects on soil-geotextile compatibility. In general, in this study, it appeared that for one soil (MWT) the slower frequency stage (0.02 Hz) was more disturbing than the subsequent higher frequency stages (0.1 Hz). However, the results clearly indicate that for the PCS soil, the higher frequency is more detrimental to soil-geotextile compatibility. The relative stability of most tests does not facilitate the thorough quantification of frequency effects on soil-geotextile compatibility. Another consideration in this study is one of the duration of unidirectional flow (UNI1) , where it may influence the quantity of soil passing. Lafleur et al. (1989) and Lafleur (1999) use a mass of 2 5 0 0 kg/m2 as the limit for undesirable piping, given that enough time is allowed for the phenomenon to develop. The limit was proposed based on tests with durations of 3000 to 8 0 0 0 minutes. This is a potential concern when assessing the significance of the mass of soil passing through the geotextile in this current study, since the test stages are much shorter (UNI1 is 90 minute duration). However, the results have shown either a wash-out phenomenon of very small quantities of soil or catastrophic piping, therefore, this consideration is not viewed as a limitation. Finally, the variation of applied system gradient (i17) between the three soils is another potential limitation recognized in this study. However, due to the very small range of gradients, this is not believed to be a significant limitation. R. Hawley 72 Chapter 7 8.0 CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK The modified gradient ratio device was used in filtration testing of 21 soil-geotextile combinations, comprising 7 geotextiles and 3 soils. Two of the geotextiles were nonwoven materials and 5 were woven, with AOS values ranging from 0.212 mm to 0.600 mm. The soils included two broader gradations that were identified by industry as being potentially 'problematic' from a filtration standpoint. They each had a relatively narrow range of D85 (from 0.330 mm to 0.215 mm), and a wide range of coefficient of uniformity with C u (from 1.8 to 5.8). The tests therefore provided results for AOS/D 8 5 values ranging from 0.6 to 2.8. Water pluviation was used to prepare saturated, homogeneous samples for the more uniform soil, while a slurry deposition technique was used to prepare similar samples of the two soils with broader gradations. A multi-stage testing procedure was used in which unidirectional and cyclic flow conditions were imposed. The test variables included frequency of flow reversal (0.02 Hz, 0.1 Hz), and confining stress (25 kPa, zero). The hydraulic gradient applied to the sample varied with soil type, in the range of 2 to 5. The results describe the relative performance of nonwoven and woven geotextiles in different hydraulic conditions, and allow for an evaluation of criteria for unidirectional and cyclic flow as they exist in commonly used design guidance. The influence of geotextile type was examined from results for 2 nonwoven geotextiles and a woven geotextile, each with the same AOS of 0.212 mm. Based on this very limited comparison, it appears there is little difference in behaviour of these woven and nonwoven geotextiles. All tests were relatively stable, with < 200 g/m2 of material passing through the geotextiles. Results for the Port Coquitlam silty sand, which yielded the most soil passing through the geotextile, showed a small difference in the grain size distribution of the passing soils. It appeared that more of the finer material passed through the woven geotextile, than the corresponding nonwoven. The influence of flow regime was studied from tests in unidirectional flow, and cyclic flow with and without confining stress. No significant influence of the frequency was found in testing for the flow reversal at a frequency of 0.02 Hz and 0.1 Hz. The Fraser River sand is stable in all tests and therefore, the influence of flow regime does not appear to be an issue. The mine waste tailings are stable in unidirectional flow, and generate a very subtle trend towards a piping potential in cyclic flow as the AOS/D8 5 approaches a value of 2.0. The Port Coquitlam silty sand behaved slightly differently than the other two soils, in that it is stable in both unidirectional flow and confined cyclic flow, but experienced significant piping and collapse of R. Hawley 73 Chapter 8 the soil structure with unconfined cyclic flow. The soil yielded catastrophic piping during sample preparation when the AOS/D8 5 was 2.8. The results were used to evaluate the design criteria of CGS (1992) and Luettich et al. (1992) in unidirectional flow, and CGS (1992), Luettich et al. (1992) and Holtz et al. (1997) in cyclic flow. The CGS (1992) and Luettich et al. (1992) guidance were found to be slightly conservative for soil-geotextile filtration compatibility in unidirectional flow. For cyclic flow, all three criteria were again found to be reasonable, but overly conservative. Based on these conclusions, the following recommendations are made for additional studies: 1. Due to the flow-related head losses in the system, the gradient across the sample, i 1 7 varied. Tests should be performed such that i 1 7 remains constant so as to ensure identical severity of flow for all soils. 2. The influence of geotextile type was studied here for three geotextiles with the same AOS of 212 um. The limited data suggest a similar response. It is recommended that more coupled tests be performed with nonwoven and woven geotextiles at different AOS values, to obtain further data on this issue. 3. Also, with respect to cyclic flow, it is recommended that longer stage durations be imposed, especially for stages of a higher frequency. It is important to know whether a piping action would have established had there been more time. A longer duration of cyclic flow might allow for more soil to pass, and therefore the mass of soil passing over time during the stage could be better resolved. 4. The results of this study show that there appears to be no significant influence of frequency on filtration behaviour, where 0.02 Hz and 0.1 Hz were imposed. However, in other studies, the influence of frequency has been found to be significant. For example, Chew et al. (2000) used frequencies ranging from 0.07 Hz to 0.5 Hz and found the washout of fines to be highly dependent on this frequency. Tondello (1998) used frequencies ranging from 0.1 Hz to 0.5 Hz, however, the frequency dependence is not clearly quantified. On this basis, it is recommended that the frequency of flow reversal be systematically varied over a much broader range in order to better determine and quantify its effect. 5. For greater certainty in design practice, there is a need to establish an acceptable rationale for filtration compatibility concerning the use of the gradient ratio values or the measurement of the R. Hawley 74 Chapter 8 mass of soil passing, mp. Lafleur et al. (1989) suggest a limit of 2500 g/m2 for mass for soil passing, however, it must be determined whether this value is appropriate. This will assist to 'unify' the empirical design criteria for unidirectional and cyclic flow. In any case, it is recommended that the mass of soil passing should be measured in all studies as it provides valuable and sometimes critical information on filter behaviour that the gradient ratio values alone cannot provide. R. Hawley 75 Chapter 8 LIST OF REFERENCES Akram, A.H., and Gabr, M.A. (1997) Filtration of Fly Ash Using Nonwoven Geotextiles: Effect of Sample Preparation Technique and Testing Method, ASTM Geotechnical Testing Journal, GTODJ, Vol. 20, No. 3, September, pp. 263 - 271. ASTM. (1992) Standard Test Method for Measuring the Soil-Geotextile System Clogging Potential by the Gradient Ratio (D5101), in 1992 Annual Book of ASTM Standards, sect. 4, vol. 04.08. ASTM, Philadelphia, PA, pp.1090-1196. ASTM. (1996) Standard Test Method for Measuring the Soil-Geotextile Clogging Potential by the Gradient Ratio (D5101-96), in the Annual Book of ASTM standards, Vol. 04.09, ASTM Philadelpia. Bertram, G.E. (1940) An Experimental Investigation of Protective Filters, Soil Mechanics Series No. 7, Graduate School of Engineering, Harvard University, Cambridge, MA. Bhatia, S.K. and Huang, Q. (1995) Geotextile Filters for Internally Stable/Unstable Soils, Geosynthetics International, Vol. 2, No. 2, pp. 537-565. Bhatia, S.K., Mlynarek, J., Rollin, A.L., and Lafleur, J. (1991) Effect of Pores Structure of Nonwoven Geotextiles on Their Clogging Behavior, Proceedings, Geosythetics '91, Atlanta, GA, February 26-28, Vol. 2, pp. 629-642. Boschuk, J, Jr. & Zhou, Y. (1992) Existing Test Methods for Design of Geosynthetics for Drainage Systems, Geotextiles and Geomembranes, Elsevier, England, No. 11, pp. 461 - 478. Calhoun, C C , Jr. (1972) Development of Design Criteria and Acceptance Specifications for Plastic Filter Cloths, Technical Report, S-72-7, US Army Engineer Waterways Experiment Station, Vicksburg, MS, June, 83 pp. Canadian Geotechnical Society. (1992) Canadian Foundation Engineering Manual, 3 r d ed., Canadian Geotechnical Society, Richmond, BC, pp. 447-451. Carroll, R.G. (1983) Geotextile Filter Criteria, Transportation Research Record, No. 916, pp. 46-53. Cazzuffi, D.A,, Mazzucato, A., Moraci, N., and Tondello, M. (1996) A New Test Apparatus For The study of Geotextile Behavior As Filters In Unsteady Flow Conditions, Proceedings of Geofilters'96 Conference, Montreal, Quebec, Canada, May 29-31, pp.183-191. Cazzuffi, D.A., Mazzucato, A., Moraci, N., & Tondello, M. (1999) A new test apparatus for the study of geotextiles behaviour as filters in unsteady flow conditions: relevance and use, Geotextiles and Geomembranes, Elsevier, England, Vol. 17, No. 5-6, pp. 313 - 329. Chaney, R. and Mullis, P.J. (1978) Geotechnical Testing Journal, ASTM, Vol. 1, No. 2, pp. 107-108. Chang, D. T.-T., Hsieh, C , Chen, S.Y., Chen, Y.Q. (2000) Review Clogging Behaviour by the Modified Gradient Ratio Test Device with Implanted Piezometers, Testing and Performance of Geosynthetics in Subsurface Drainage ASTM STP 1390, J.B. Goddard, L.D. Suits, and J.S. Baldwin, Eds., ASTM. West Conshochocken, PA. Chew, S.H., Zhao, Z.K., Karunaratne, G.P., Tan, S.A, Delmas, Ph., and Loke, K.H. (2000) Revetment Geotextile Filter Subjected to Cyclic Wave Loading, Advances in Transportation and Geoenvironmental Systems Using Geosynthetics, Proceedings of Sessions of Geo-Denver 2000, August 5-8, Denver, CO, pp. 162 - 175. R. Hawley 76 References Christopher, B.R. and Fischer, G.R. (1992) Geotextile Filtration Principles, Practices and Problems, Geotextiles and Geomembranes, Elsevier, England, No. 11, pp. 337-353. Craig, R.F. (1997) Soil Mechanics, 6 t h Edition. E & FN Spon, London, UK. DeBerardino, S.J. (1992) Drainage Principles and the Use of Geosynthetics, Geotextiles and Geomembranes, Elsevier, England, No. 11, pp. 449-459 Dierickx, W. (1986) Model research on geotextile blocking and clogging in hydraulic engineering, Proceedings of the Third International Conference on Geotextiles, Vienna, IGS, pp. 775-777. Fannin, R.J. and Hameiri, A. (1999) A Gradient Ratio Device for Compatibility Testing in Cyclic Flow, Proceedings, Geosynthetics '99, April 28-30, Boston, MA, pp. 1033 - 1042. Fannin, R.J. and Pishe, R. (2001) Testing and specifications for geotextile filters in cyclic flow applications, Proceedings, Geosynthetics 2001, February 12-14, Portland, OR, pp. 423-435. Fannin, R.J., Vaid, Y.P. and Shi, Y. (1994a) A Critical Evaluation of the Gradient Ratio Test, ASTM Geotechnical Testing Journal, GTJODJ, Vol. 17, No. 1, March, pp. 35-42. Fannin, R.J., Vaid, Y.P. and Shi, Y.C (1994b) Filtration testing of nonwoven geotextiles, Canadian Geotechnical Journal, No. 31, pp. 555 - 563. Fannin, R.J., Vaid, Y.P., Palmeira, E.M. and Shi, Y.C. (1996) A Modified Gradient Ratio Device, Recent Developments in Geotextile Filters and Prefabricated Drainage Geocomposites, ASTM STP 1281, S.K Bhatia and L.D. Suits, Eds., ASTM, Philadelphia, pp. 100 - 112. Faure, Y. and Mylnarek, J. (1998) Geotextile Filter Hydraulic Requirements, Geotechnical Fabrics Report, Vol. 16, No. 4, May, pp. 30-33. Finn, W.D.L., Pickering, D.J. and Bransby, P.L. (1971) Sand liquefaction in triaxial and simple shear tests, Journal of the Soil Mechanics and Foundations Division, American Society of Civil Engineers, Vol. 97, No. SM4, pp. 639-659. Fischer, G.R., Mare, A.D. and Holtz, R.D. (1999) Influence of Procedural Variables on the Gradient Ratio Test, ASTM Geotechnical Testing Journal, GTJODJ, Vol. 22, March, pp. 22 - 31. Gabr, M.A., Akram, M.H., and Zayed, A.M. (1998) Field versus laboratory filtration performance of a nonwoven geotextile with fly ash, Technical Note, Geotextiles and Geomembranes, Elsevier, England, No. 16, pp. 247 - 255. Giroud, J.P. (1982) Filter Criteria for Geotextiles, Second International Conference on Geotextiles, Las Vegas, USA, August 1-6, Vol. 1, pp. 103 - 108. Giroud, J.P. (1996) Granular Filters And Geotextiles Filters, Proceedings of Geofilters'96 Conference, Montreal, Quebec, Canada, May 39-31, pp. 565-680. Hameiri, A. (2000) Soil Geotextile Filtration Behavior Under Dynamic Conditions of Vibration and Cyclic Flow. PhD Thesis, University of British Columbia, Vancouver. Heerton, G. (1982) Dimensioning the Filtration Properties of Geotextiles Considering Long-Term Conditions, Proceedings, Second International Conference on Geotextiles, Las Vegas, USA, August 1-6, Vol. 1, pp. 115-120. Holtz, R.D, Christopher, B.R., and Berg, R.R. (1997) Geosynthetic Engineering, BiTech Publishers, Richmond, BC, pp.29-68. R. Hawley 77 References Holtz, R.D., Christopher, B.R., and Berg, R.R. (1995) Geosynthetic Design & Construction Guidelines, Participant Notebook, Publication No. FHWA-HI-95-xxx, Federal Highway Administration, McLean, Virginia. Hoover, T.P. (1982) Laboratory testing of geotextile fabric filters, Proceedings of the Second International Conference on Geotextiles, Las Vegas, Vol. Ill, Industrial Fabrics Association International, St. Paul, MN, pp. 839-843. Industrial Fabrics Association International, IFAI (1999) Geotechnical Fabrics Report: Specifier's Guide 2000, J. Swedberg (Ed.), Vol. 17, No. 9. Kenney, T.C. and Lau, D. (1985) Internal stability of granular filters, Canadian Geotechnical Journal, No. 22, pp. 215-225. Koerner, R.M, and Ko, F.K. (1982) Laboratory studies on long-term drainage capability of geotextiles, Proceedings of the Second International Conference on Geotextiles, Las Vegas, NV, Vol. I, pp. 91-95. Kuerbis, R.H., and Vaid, Y.P. (1988) Sand Sample Preparation - the Slurry Deposition Method, Soils and Foundations, Vol. 28, No. 4, pp. 107-118. Lafleur J., Mlynarek J., and Rollin A. L. (1989) Filtration of Broadly Graded Cohesionless Soils, Journal of Geotechnical Engineering, ASCE, Vol. 115, No. 12, pp. 1747-1768. Lafleur, J. (1984) Filter testing of broadly graded cohesionless tills, Canadian Geotechnical Journal, No. 21, pp. 634-643. Lafleur, J. (1998) Particles Washout Associated with the Retention of Broadly Graded Soils by Geotextiles, Proceedings, Sixth International Conference on Geosynthetics, Atlanta, GA, March 25-29, Vol. 2, pp. 1001 - 1004. Lafleur, J. (1999) Selection of geotextiles to filter broadly graded cohesionless soils, Geotextiles and Geomembranes, Vol. 17, No. 5-6, pp. 299-312. Lawson, C. R. (1998) Retention Criteria and Geotextile-Filter Performance, Geotechnical Fabrics Report, Vol. 16, No. 2, pp. 26-29. Lawson, CR. (1982) Geotextile Requirements for Erosion Control Structures, Proceedings of the International Symposium on Recent Developments in Ground Improvement Techniques, AIT, Bankok, Nov. 29 - Dec. 3, pp. 177-192. Lawson, CR. (1992) Geotextile Revetment Filters, Geotextiles and Geomembranes, Elsevier, England, No. 11, pp. 431 -448. Lee, K.L. and Seed, H.B. (1967) Journal of the Soil Mechanics and Foundations Division, American Society of Civil Engineers, Vol. 93, No. SM1, pp. 47-70. Legge, K.R. (1990) A new approach to geotextile selection, Proceedings of the Fourth Internatioinal Conference on Geotextiles, The Hague, The Netherlands, pp. 269-272. Luettich, S.M., and Williams, N.D. (1989) Design of vertical drains using the hydraulic conductivity ratio analysis, Proceedings of Geosynthetics '89, San Diego, USA. Luettich, S.M., Giroud, J.P., and Bachus, R.C (1992) Geotextile Filter Design Guide, Geotextiles and Geomembranes, Elsevier, England, No. 11, pp. 355 - 370. R. Hawley 78 References Mannsbart, G. and Christopher, B.R. (1997) Long-Term Performance of Nonwoven Geotextile Filters in Five Coastal and Bank Protection Projects, Geotextiles and Geomembranes, Elsevier, England, No. 15, pp. 207-221 Mlynarek, J. (1998) Hydraulic behavior of geotextile filters in the field, Geotechnical Fabrics Report, Oct/Nov., Vol. 16, No. 8, pp. 30 - 35. Mlynarek, J., and Lombard, G. (1997) Significance of Percent Open Area (POA) in the Design of Woven Geotextile Filters, Proceedings, Geosynthetics '97, Long Beach, CA, March 11-13, Vol. 2, pp. 1093-1106. Molenkamp, F., Calle, E.O.F., Heusdens, J.J., and Koenders, M.A. (1979) Cyclic filter tests in a triaxial cell, Proceedings, 7th European Conference on Soil Mechanics and Foundation Engineering, Brighton, England, September, pp. 97-101. Montero, C M . and Overmann, L.K. (1990) Geotextile Filtration Performance Test, Proceedings of the Fourth International Conference on Geotextiles, Geomembranes and related products, Netherlands, Vol. 1, pp. 318. Mouw, K. A., Nederlof, K. D. C , Stuip, J., and Veldhuijzen, V. Z. R. (1986) Geotextiles in Shore and Bed Protection Works, Proceedings of the Third International Conference on Geotextiles, Vienna, Austria, April 7-11, Vol. 2, pp.349-354. Narejo, D.B. and Koerner, R.M. (1992) A Dynamic Filtration Test for Geotextile Filters, Technical Note, Geotextiles and Geomembranes, Elsevier, England, No. 11, pp. 395-400. Sansome, L.J. and Koerner, R.M. (1992) Fine fraction filtration test to assess geotextile filter performance, Geotextiles and Geomembranes, Vol. 11, pp. 371-393. Scott, J.D. (1980) The Filtration-Permeability Test, Proceedings of the First Canadian Symposium on Geotextiles, Canadian Geotechnical Society, Rexdale, ON, pp. 176-186. Shi, Y.C. (1993) Filtration behaviour of nonwoven geotextiles in the gradient ratio test. MASc Thesis, University of British Columbia, Vancouver. Siva, U. and Bhatia, S.K. (1993) Filtration Performance of Geotextiles with Fine-Grained Soils, Proceedings of Geosynthetics' 93 Conference, Vancouver, Canada, March 30 - April 1. Vol. 1, pp. 483-499. Tondello, M. (1998) Geotextile filters in unsteady flow conditions, Rivista Italina di Geotecnica, Associazione Geotecnica Italiana, Via Badini, Bologna, October-December, No. 4,, pp. 18 - 29. United States Army Corps of Engineers. (1977) Plastic Filter Cloth, Civil Works Construction Guide Specification No. CE-02215, Office Chief of Engineers, Washington, D.C., 16 p. Vaid, Y.P. and Negussey, D. (1988) Preparation of Reconstituted Sand Specimens, Advanced Triaxial Testing of Soil and Rock, ASTM STP 977, R.T. Donaghe, R.C. Chaney, and M.L. Silver, Eds., ASTM, Philadelphia, pp. 405 - 417. Williams, N.D. and Abrusakhm, M.A. (1989) Evaluation of Geotextile/soil Filtration Characteristics Using the Hydraulic Conductivity Ratio Analysis, Journal of Geotextiles and Geomembranes, Vol. 8, No. 1, pp. 1-26. R. Hawley 79 References APPENDIX A Internal Stability of Test Soils Calculation of Design Criteria R. Hawley 8 0 Appendix A INTERNAL STABILITY OF TEST SOILS FRASER RIVER SAND: 25.0 20.0 10.0 2F = 5.0 F = 2.5 0.0 < i * K • i | J • \ « X X • N >-« »-« >-©-« • « »- f > 10.000 1.000 4D = 0.4 D = 0.1 0.010 0.001 Diameter (mm) H » F .-. FRS is INTERNALLY S T A B L E R. Hawley 81 Appendix A MINE WASTE TAILINGS: an n H > F .-. MWT is INTERNALLY S T A B L E R. Hawley 82 Appendix A H » F .-. PCS is INTERNALLY S T A B L E R. Hawley 83 Appendix A CALCULATION OF DESIGN CRITERIA Test soil properties Soil Code r? 0 ' ' . ,. Description ^85 D60 D 5 0 D 3 0 Pis Dio C u G s (mm) (mm) (mm) (mm) (mm) (mm) (D60/D10) Uniformly F R S Graded Rounded Fine Sand 0.330 0.280 0.260 0.220 0.170 0.155 1.8 2.50 Uniformly MWT G r a d e d Anqular Fine 0.290 0.200 0.178 0.126 0.081 0.060 3.3 2.50 Sand Broadly PCS Graded Rounded 0.215 0.185 0.178 0.126 0.074 0.032 5.8 2.75 Sandy Silt (1) UNIDIRECTIONAL FLOW: (a) Design Guidance: CGS (1992): AOS < B D 8 5 FRS: Cu = 1.8 Cu<2 .-. B = 1 => AOS < 1.0 D 8 5 MWT: Cu = 3.3 2 < Cu < 4 .-. B = 0.5 Cu = 1.7 => AOS < 1.7 D 8 5 PCS: Cu = 5.8 4 < Cu < 8 .-. B = 8/Cu = 1.4 => AOS < 1.4 D 8 5 (b) Design Guidance: Luettich et al. (1992): Figure 2.2 FRS: D 1 0 = 0.155 mm .-. less than 10% fines & less than 90% gravel Application favours retention Stable soil Cu = D 6 0/D 3 0 = 0.280/0.220 = 1.27 < 3 .-. uniformly graded Medium dense 0 9 5 <1.5CuD 5 0 = 1.5 (1.27)D5o = 1.9D5o => AOS < 1.9 D 5 0 R. Hawley 84 Appendix A MWT: D 1 0 = 0.060 mm .-. less than 20% clay & more than 10% fines Non-plastic Application favours retention Stable soil Cu = D 6 0/D 3 0 = 0.200/0.126 = 1.59 < 3 .-. uniformly graded Medium dense 0 9 5 < 1.5 C u D 5 0 = 1.5 (1.59) D 5 0 = 2.4 D 5 0 => AOS < 2.4 D 5 0 PCS: D 1 0 = 0.032 mm .-. less than 20% clay & more than 10% fines Non-plastic Application favours retention Stable soil Cu = D6o/D3o = 0.185/0.126 = 1.47 < 3 .'. uniformly graded Medium dense 0 9 5 <1.5CuD 5 0 = 1.5 (1.47)D5o = 2.4D5o => AOS < 2.2 D 5 0 (I) CYCLIC FLOW: (a) Design Guidance: CGS (1992): FRS: Soil with < 50% passing No. 200 sieve => AOS < 1.0 D15 MWT: Soil with < 50% passing No. 200 sieve => AOS < 1.0 D 1 5 PCS: Soil with < 50% passing No. 200 sieve => AOS < 1.0 D 1 5 (b) Design Guidance: Luettich et al. (1992): Figure 2.3 FRS: D 1 0 = 0.155 mm .-. less than 50% fines & less than 90% gravel Severe wave attack => AOS < 1.0 D50 MWT: D 1 0 = 0.060 mm .-. less than 30% clay & more than 50% fines Non-Plastic Severe wave attack => AOS < 1.0 D50 MWT: D 1 0 = 0.032 mm .-. less than 30% clay & more than 50% fines Non-Plastic Severe wave attack => AOS < 1.0 D 5 0 R. Hawley 85 Appendix A (c) Des ign G u i d a n c e : Hol tz et a l . (1997) D y n a m i c F low condi t ions: If the geotext i le is not proper ly we ighted down and in intimate contact with the soi l to be protected, or if dynamic , cyc l ic or pulsat ing loading condi t ions p roduce high loca l ized hydraul ic gradients, then the soil part ic les c a n m o v e beh ind the geotext i le. T h u s , the use of B = 1 is not conserva t ive , b e c a u s e the br idging network will not deve lop and the geotext i le will be required to retain even f iner par t ic les. W h e n retention is the pr imary cri teria, B shou ld be reduced to 0.5, or: => A O S < 0.5 D 8 5 R. Hawley 86 Appendix A APPENDIX B Pre-test and Post-test Gradations R. Hawley 87 Appendix B Pre-Testing and Post Testing Gradations TestF140 100.0 . . , 10.000 1.000 0.100 0.010 0.001 Diameter (mm) R. Hawley 88 Appendix B Pre-Testing and Post Testing Gradations Test F160 FRS (Cu = 1.8) • Post Test Gradation 10.000 1.000 0.100 Diameter (mm) 0.010 0.001 R. Hawley 89 Appendix B Pre-Testing and Post Testing Test F700 Gradations 100.0 90.0 80.0 70.0 ~ 60.0 cu LI 50.0 c U S. 40.0 30.0 20.0 10.0". 0.0 10.000 • FRS (Cu = 1.8) Post Test Gradation 1.000 0.100 Diameter (mm) 0.010 0.001 R. Hawley 90 Appendix B 100.0 90.0 80.0 70.0 ~ 60.0 CD i l 50.0 c a> o i_ cu 40.0 30.0 20.0 10.0 0.0 10.000 Pre-Testing and Post Testing Gradations Test F500 Si FRS (Cu = 1.8) Post Test Gradation 1.000 0.100 0.010 Diameter (mm) 0.001 R. Hawley 91 Appendix B Pre-Testing and Post Testing Gradations Test F404 « FRS (Cu = 1.8) Post Test Gradation 10 0.1 Diameter (mm) 0.01 0.001 R. Hawley 92 Appendix B 100.0 90.0 80.0 70.0 ~ 60.0 i l 50.0 c a> o I 40.0 30.0 20.0 10.0 0.0 10.000 Pre-Testing and Post Testing Gradations Test F402 I FRS (Cu = 1.8) Post Test Gradation 1.000 0.100 Diameter (mm) 0.010 0.001 R. Hawley 93 Appendix B 100.0 90.0 80.0 70.0 ~ 60.0 CD il 50.0 .*—' c tu o £ 40.0 30.0 20.0 10.0 0.0 10.000 Pre-Testing and Post Testing Gradations Test F570 FRS (Cu = 1.8) . Post Test Gradation 1.000 0.100 Diameter (mm) 0.010 0.001 R. Hawley 94 Appendix B Pre-Testing and Post Testing Gradations TestM140 MWT ( C u = 3.3) -ji—Post Test Gradation 10.000 1.000 0.100 Diameter (mm) 0.010 0.001 R. Hawley 95 Appendix B Pre-Testing and Post Testing Gradations TestM160 - # - M W T ( C u = 3.3) — P o s t Test Gradation T 10.000 1.000 0.100 Diameter (mm) 0.010 0.001 R. Hawley 96 Appendix B Pre-Testing and Post Testing Gradations Test M700 R. Hawley 97 Appendix B Pre-Testing and Post Testing Gradations Test M500 R. Hawley 98 Appendix B Pre-Testing and Post Testing Gradations Test M404 10.000 1.000 0.100 Diameter (mm) 0.010 0.001 R. Hawley 99 Appendix B 100.0 90.0 80.0 70.0 ~ 60.0 tu c c CD O 50.0 S. 40.0 30.0 20.0 10.0 0.0 10.000 Pre-Testing and Post Testing Gradations Test M402 1.000 -»-- MWT (Cu = 3.3) —x—Post Test Gradation 0.100 Diameter (mm) 0.010 0.001 R. Hawley 100 Appendix B 100.0 10.000 Pre-Testing and Post Testing Gradations Test M570 MWT (Cu = 3.3) Post Test Gradation Soil Passing 1.000 0.100 Diameter (mm) 0.010 0.001 R. Hawley 101 Appendix B Pre-Testing and Post Testing Gradations Test P140 R. Hawley 102 Appendix B Pre-Testing and Post Testing Gradations Test P160 R. Hawley 103 Appendix B Pre-Testing and Post Testing Gradations Test P700 R. Hawley 104 Appendix B Pre-Testing and Post Testing Gradations Test P500 10.000 1.000 0.100 0.010 0.001 Diameter (mm) R. Hawley 105 Appendix B Pre-Testing and Post Testing Gradations Test P404 10.000 1.000 0.100 0.010 0.001 Diameter (mm) R. Hawley 106 Appendix B Pre-Testing and Post Testing Gradations Test P402 R. Hawley 107 Appendix B APPENDIX C Water Head Distributions R. Hawley 1 0 8 Appendix C Water Head Distribution Test F140 10 15 Head (cm) 20 25 R. Hawley 109 Appendix C Water Head Distribution TestF160 10 ? o s 8 o o> O E o -t 6 2 0 UNI1 - » - U N I 2 H K - U N I 3 ^ U N I 4 0 5 10 15 20 25 Head (cm) R. Hawley 110 Appendix C Water Head Distribution Test F700 R. Hawley 111 Appendix Water Head Distribution Test F500 12 10 O E o * 6 2 UNI1 -B-UNI2 _*_UNI3 _t_UNI4 J 0 5 10 15 20 25 Head (cm) R. Hawley 112 Appendix C Water Head Distribution Test F404 R. Hawley 113 Appendix C Water Head Distribution Test F402 R. Hawley 114 Appendix C Water Head Distribution Test F570 R. Hawley 115 Appendix C Water Head Distribution Test M140 R. Hawley 116 Appendix C Water Head Distr ibut ion Test M160 R. Hawley 117 Appendix Water Head Distribution Test M700 R. Hawley 118 Appendix C Water Head Distribution Test M500 Head (cm) R. H a w l e y 119 A p p e n d i x C Water Head Distribution Test M404 R. Hawley 120 Appendix C Water Head Distribution Test M402 R. Hawley 121 Appendix Water Head Distribution Test M570 R. Hawley 122 Appendix C Water Head Distribution Test P140 R. Hawley 123 Appendix C Water Head Distribution Test P160 10 ? _o x 8 <D * ^ o CD o E o - 6 2 0 UNI1 _ » _ U N I 2 _ * _ U N I 3 — I — UNI4 0 10 20 30 40 50 60 Head (cm) R. Hawley 124 Appendix C Water Head Distribution Test P700 14 0 10 20 30 40 50 Head (cm) R. Hawley 125 Appendix C Water Head Distribution Test P500 10 20 30 Head (cm) 40 50 R. Hawley 126 Appendix C Water Head Distribution Test P404 R. Hawley 127 Appendix C Water Head Distribution Test P402 R. Hawley 128 Appendix C APPENDIX D Tabulat ion of Key Resul ts R. Hawley 1 2 g Appendix D FRASER RIVER SAND (FRS) TESTS: Sample: F140 Stage El. Time Dh67 Dh57 Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) (cm) (cm) (cm) Gradient (cm) Distance - 0.8 2.5 7.5 9.9 i17 5.0 ASTM MOD UNI1 81.000 1.3 4.3 14.8 21.0 2.1 10.5 0.8 0.8 CYC50S 941.497 1.3 0.9 10.6 16.5 1.7 5.9 0.3 1.4 UNI2 993.705 1.3 4.8 14.5 20.5 2.1 9.7 1.0 0.9 CYC10S 1052.163 1.4 4.8 14.4 20.2 2.0 9.6 1.0 0.9 UNI3 1076.583 1.3 4.9 14.6 20.5 2.1 9.7 1.0 0.8 CYC10N 1137.373 1.4 4.8 14.5 20.3 2.1 9.7 1.0 0.9 UNI4 1160.643 1.3 4.7 14.4 20.2 2.0 9.6 1.0 0.8 Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec cm/sec cm/sec cm/sec cm/sec Q(avg) k(avg) k67 k35 k57 UNI1 81.000 3.777 0.022 1,603 0.029 2.097 0.022 1.717 0.027 CYC50S 941.497 4.046 0.030 1.626 0.030 1.183 0.042 0.371 0.133 UNI2 993.705 3.911 0.023 1.678 0.029 1.942 0.025 1.901 0.025 CYC10S 1052.163 5.248 0.031 1.699 0.038 1.926 0.033 1.911 0.034 UNI3 1076.583 3.921 0.023 1.635 0.029 1.937 0.025 1.948 0.025 CYC10N 1137.373 4.933 0.029 1.693 0.036 1.934 0.031 1.929 0.031 UNI4 1160.643 3.925 0.024 1.618 0.030 1.928 0.025 1.886 0.025 Sample: F160 Stage El. Time Dh67 Dh57 Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) (cm) (cm) (cm) Gradient (cm) Distance - 0.8 2.5 7.5 10.0 i17 5.0 ASTM MOD UNI1 81.000 1.6 5.5 16.0 23.1 2.3 10.5 1.0 1.0 CYC50S 941.497 1.6 2.3 12.2 19.5 2.0 7.3 0.6 1.4 UNI2 993.705 1.6 5.9 16.2 22.3 2.2 10.3 1.1 0.9 CYC10S 1052.163 1.6 5.9 15.5 24.1 2.4 9.6 1.2 1.0 UNI3 1076.583 1.5 6.0 15.3 23.2 2.3 9.3 1.3 1.0 CYC10N 1137.373 1.6 6.1 15.5 22.2 2.2 9.4 1.3 1.1 UNI4 1160.643 1.5 5.9 15.2 22.1 2.2 9.3 1.3 1.0 Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec cm/sec cm/sec cm/sec cm/sec - Q(avg) k(avg) k67 k35 k57 UNI1 81.000 3.900 0.021 2.025 0.024 2.100 0.023 2.200 0.022 CYC50S 941.497 4.500 0.028 1.975 0.028 1.460 0.038 0.920 0.060 UNI2 993.705 3.900 0.021 1.938 0.025 2.060 0.023 2.360 0.020 CYC10S 1052.163 5.500 0.028 1.950 0.035 1.922 0.035 2.356 0.029 UNI3 1076.583 4.000 0.021 1.875 0.026 1.860 0.026 2.400 0.020 CYC10N 1137.373 5.400 0.030 1.988 0.033 1.888 0.035 2.424 0.027 UNI4 1160.643 4.000 0.022 1.900 0.026 1.856 0.026 2.368 0.021 R. Hawley 130 Appendix D Sample: F700 Stage El. Time Dh67 Dh57 Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) (cm) (cm) (cm) Gradient (cm) Distance - 0.8 2.5 7.5 9.9 i17 5.0 ASTM MOD UNI1 81.000 1.3 4.4 15.0 21.3 2.1 10.6 0.8 0.8 CYC50S 941.497 1.3 0.5 10.3 16.1 1.6 5.9 0.2 1.4 UNI2 993.705 1.3 4.3 14.1 20.0 2.0 9.7 0.9 0.9 CYC10S 1052.163 1.4 4.3 14.0 19.7 2.0 9.6 0.9 0.9 UNI3 1076.583 1.3 4.4 14.1 20.0 2.0 9.7 0.9 0.8 CYC10N 1137.373 1.3 4.3 14.0 19.8 2.0 9.7 0.9 0.9 UNI4 1160.643 1.3 4.2 13.8 19.7 2.0 9.6 0.9 0.8 . Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec cm/sec cm/sec cm/sec cm/sec - Q(avg) k(avg) k67 k35 k57 UNI1 81.000 3.790 0.022 1.681 0.028 2.121 0.022 1.746 0.027 CYC50S 941.497 4.110 0.031 1.632 0.031 1.175 0.043 0.211 0.238 UNI2 993.705 3.911 0.024 1.679 0.029 1.947 0.025 1.728 0.028 CYC10S 1052.163 5.468 0.034 1.695 0.039 1.930 0.035 1.727 0.039 UNI3 1076.583 3.921 0.024 1.630 0.029 1.940 0.025 1.758 0.027 CYC10N 1137.373 4.933 0.030 1.682 0.036 1.936 0.031 1.727 0.035 UNI4 1160.643 3.925 0.024 1.606 0.030 1.929 0.025 1.679 0.029 Sample: F500 Stage El. Time Dh67 Dh57 Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) (cm) (cm) (cm) Gradient (cm) Distance - 0.8 2.5 7.5 9.5 i17 5.0 ASTM MOD UNI1 81.000 2.1 5.5 14.9 19.8 2.1 9.5 1.2 1.4 CYC50S 2355.300 1.7 2.1 11.0 16.1 1.7 8.9 0.5 1.2 UNI2 1107.668 2.1 5.3 14.6 19.4 2.0 9.3 1.1 1.4 CYC10S 2410.534 2.0 5.1 14.0 19.1 2.0 8.9 1.1 1.4 UNI3 1225.081 2.0 5.3 14.7 19.6 2.1 9.4 1.1 1.3 CYC10N 2455.298 2.0 5.0 14.0 19.0 2.0 9.0 1.1 1.4 UNI4 1299.032 2.0 5.4 14.9 20.0 2.1 9.5 1.1 1.3 Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec cm/sec cm/sec cm/sec cm/sec - Q(avg) k(avg) k67 k35 k57 UNI1 81.000 4.093 0.024 2.591 0.019 1.891 0.026 2.180 0.023 CYC50S 2355.300 4.130 0.030 2.141 0.024 1.789 0.028 0.823 0.061 UNI2 1107.668 4.080 0.024 2.613 0.019 1.863 0.027 2.126 0.023 CYC10S 2410.534 4.161 0.025 2.561 0.020 1.779 0.029 2.043 0.025 UNI3 1225.081 4.090 0.024 2.461 0.020 1.876 0.027 2.115 0.024 CYC10N 2455.298 4.173 0.026 2.528 0.020 1.799 0.028 2.010 0.025 UN14 1299.032 4.129 0.024 2.539 0.020 1.901 0.027 2.173 0.023 R. Hawley 131 Appendix D Sample: F404 Stage El. Time Dh67 Dh57 Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) (cm) (cm) (cm) Gradient (cm) Distance - 0.8 2.5 7.5 9.5 i17 5.0 ASTM MOD UNI1 81.000 1.7 4.5 14.3 18.9 2.0 9.8 0.9 1.1 CYC50S 1067.749 1.5 1.8 11.1 15.8 1.7 9.3 0.4 1.0 UNI2 1102.410 1.5 4.5 13.8 18.5 1.9 9.3 1.0 1.0 CYC10S 1161.356 1.5 4.5 13.8 18.4 1.9 9.3 1.0 1.0 UNI3 1183.448 1.5 4.5 13.8 18.5 2.0 9.3 1.0 1.0 CYC10N 1234.151 1.5 4.4 13.7 18.3 1.9 9.3 1.0 1.0 UNI4 1256.783 1.5 4.5 13.8 18.6 2.0 9.3 1.0 1.0 Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec cm/sec cm/sec cm/sec cm/sec - Q(avg) k(avg) k67 k35 k57 UNI1 81.000 4.079 0.025 2.132 0.023 1.960 0.025 1.814 0.028 CYC50S 1067.749 ' 4.138 0.030 1.838 0.028 1.868 0.027 0.721 0.070 UNI2 1102.410 4.229 0.027 1.853 0.028 1.866 0.028 1.787 0.029 CYC10S 1161.356 4.164 0.026 1.908 0.027 1.865 0.027 1.782 0.029 UNI3 1183.448 4.221 0.026 1.841 0.028 1.864 0.028 1.798 0.029 CYC10N 1234.151 4.182 0.027 1.887 0.027 1.856 0.028 1.764 0.029 UNI4 1256.783 4.228 0.026 1.835 0.028 1.860 0.028 1.799 0.029 Sample: F402 Stage El. Time Dh67 Dh57 Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) (cm) (cm) (cm) Gradient (cm) Distance - 0.8 2.5 7.5 9.6 i17 5.0 ASTM MOD UNI1 81.000 2.0 5.8 14.4 20.5 2.1 8.6 1.3 1.5 CYC50S 1044.834 1.8 2.2 10.6 15.8 1.6 8.4 0.5 1.3 UNI2 1078.839 1.8 5.7 14.1 19.4 2.0 8.4 1.4 1.4 CYC10S 1139.534 2.2 5.5 14.0 19.1 2.0 8.4 1.3 1.6 UNI3 1167.600 1.8 5.7 14.1 19.4 2.0 8.4 1.4 1.3 CYC10N 1222.067 2.3 5.6 14.0 19.2 2.0 8.5 1.3 1.7 UNI4 1254.046 1.7 5.7 14.1 19.5 2.0 8.4 1.3 1.3 Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec cm/sec cm/sec cm/sec cm/sec - Q(avg) k(avg) k67 k35 k57 UNI1 81.000 3.952 0.023 2.555 0.019 1.717 0.028 2.306 0.021 CYC50S 1044.834 4.077 0.030 2.188 0.023 1.674 0.030 0.884 0.056 UNI2 1078.839 4.216 0.026 2.285 0.023 1.675 0.031 2.282 0.023 CYC10S 1139.534 4.138 0.025 2.720 0.019 1.684 0.030 2.214 0.023 UNI3 1167.600 4.209 0.026 2.219 0.023 1.684 0.031 2.275 0.023 CYC10N 1222.067 4.121 0.025 2.934 0.017 1.693 0.030 2.231 0.023 UNI4 1254.046 4.191 0.025 2.170 0.024 1.687 0.030 2.273 0.023 R. Hawley 132 Appendix D Sample: F570 Stage El. Time Dh67 Dh57 Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) (cm) (cm) (cm) Gradient (cm) Distance - 0.8 2.5 7.5 9.6 i17 5.0 ASTM MOD UNI1 81.000 1.5 4.3 13.4 19.1 2.0 9.1 0.9 1.0 CYC50S 1039.802 1.2 1.6 10.2 15.4 1.6 8.6 0.4 0.9 UNI2 1074.502 1.2 4.1 12.7 18.0 1.9 8.6 0.9 0.9 CYC10S 1134.530 1.3 4.0 12.6 17.7 1.8 8.5 0.9 1.0 UNI3 1161.295 1.2 4.1 12.7 18.0 1.9 8.6 1.0 0.9 CYC10N 1217.161 1.4 4.0 12.6 17.7 1.8 8.6 0.9 1.0 UNI4 1253.538 1.2 4.1 12.7 17.9 1.9 8.6 0.9 0.9 Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec Q(avg) cm/sec k(avg) cm/sec k67 cm/sec k35 cm/sec k57 UNI1 81.000 4.081 0.025 1.859 0.027 1.822 0.027 1.716 0.029 CYC50S 1039.802 4.214 0.032 1.549 0.033 1.715 0.030 0.633 0.081 UNI2 1074.502 4.321 0.028 1.526 0.035 . 1.723 0.031 1.631 0.032 CYC10S 1134.530 4.239 0.028 1.672 0.031 1.709 0.030 1.606 0.032 UNI3 1161.295 4.333 0.028 1.527 0.035 1.717 0.031 1.647 0.032 CYC10N 1217.161 4.230 0.028 1.751 0.030 1.715 0.030 1.606 0.032 UNI4 1253.538 4.334 0.028 1.510 0.035 1.718 0.031 1.628 0.033 R. Hawley 133 Appendix D MINE WASTE TAILINGS (MWT) TESTS: Sample: M140 Stage Distance El. Time (minutes) Dh67 (cm) 0.8 Dh57 (cm) 2.5 Dh37 (cm) 7.5 Dh17 (cm) 8.5 System Gradient i17 Dh35 (cm) 5.0 Gradient ASTM Ratio MOD UNI1 CYC50S UNI2 CYC10S UNI3 CYC10N UNI4 81.0 1043.3 1083.5 1143.1 1169.2 1219.5 1241.4 4.5 3.1 3.1 5.9 3.1 6.2 3.0 13.2 4.1 11.6 11.6 11.7 11.3 11.5 36.9 29.6 36.9 37.7 37.2 37.5 36.8 43.4 34.5 42.0 42.0 42.2 41.6 41.8 5.1 4.1 4.9 4.9 5.0 4.9 4.9 23.7 25.5 25.3 26.0 25.4 26.1 25.3 1.1 0.3 0.9 0.9 0.9 0.9 0.9 1.2 0.7 0.8 1.4 0.8 1.5 0.7 Stage El. Time (minutes) Flow Rate cm3/sec Q(avg) Permeability cm/sec k(avg) i67 Permeability cm/sec k67 i35 Permeability cm/sec k35 i57 Permeability cm/sec k57 UNI1 CYC50S UNI2 CYC10S UNI3 CYC10N UNI4 81.0 1043.3 1083.5 1143.1 1169.2 1219.5 1241.4 0.587 0.856 1.007 1.540 1.014 1.595 0.990 0.001 0.003 0.002 0.004 0.002 0.004 0.002 5.612 3.813 3.823 7.324 3.815 7.769 3.715 0.001 0.003 0.003 0.003 0.003 0.003 0.003 4.744 5.100 5.063 5.208 5.081 5.223 5.053 0.002 0.002 0.002 0.004 0.002 0.004 0.002 5.263 1.649 4.625 4.651 4.698 4.536 4.611 0.001 0.006 0.003 0.004 0.003 0.004 0.003 Sample: M160 c Stage Distance El. Time (minutes) Dh67 (cm) 0.8 Dh57 (cm) 2.5 Dh37 (cm) 7.5 Dh17 (cm) 9.6 System Gradient i17 Dh35 (cm) 5.0 Gradient ASTM Ratio MOD UNI1 CYC50S UNI2 CYC10S UNI3 CYC10N UNI4 81.0 1048.2 1079.6 1141.6 1166.0 1220.7 1247.2 4.2 4.9 4.9 13.3 4.7 14.0 4.4 16.1 6.6 17.6 17.8 18.0 17.7 18.0 39.8 29.2 39.9 39.9 40.3 39.7 40.2 44.4 32.3 43.2 43.1 43.4 43.1 43.4 4.6 3.4 4.5 4.5 4.5 4.5 4.5 23.6 22.5 22.4 22.0 22.3 22.0 22.2 1.4 0.6 1.6 1.6 1.6 1.6 1.6 1.1 1.4 1.4 3.8 1.3 4.0 1.2 Stage El. Time (minutes) Flow Rate cm3/sec Q(avg) Permeability cm/sec k(avg) i67 Permeability cm/sec k67 i35 Permeability cm/sec k35 i57 Permeability cm/sec k57 UNI1 CYC50S UN!2 CYC10S UNI3 CYC10N UNI4 81.0 1048.2 1079.6 1141.6 1166.0 1220.7 1247.2 0.381 0.681 0.783 1.256 0.721 1.293 0.711 0.001 0.002 0.002 0.003 0.002 0.004 0.002 5.229 6.106 6.140 16.680 5.819 17.515 5.481 0.001 0.001 0.002 0.001 0.002 0.001 0.002 4.730 4.508 4.479 4.405 4.456 4.399 4.431 0.001 0.002 0.002 0.003 0.002 0.004 0.002 6.445 2.647 7.021 7.139 7.220 7.076 7.211 0.001 0.003 0.001 0.002 0.001 0.002 0.001 R. Hawley 134 Appendix D Sample: M700 Stage El. Time Dh67 Dh57 Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) (cm) (cm) (cm) Gradient (cm) Distance - 0.8 2.5 7.5 9.6 i17 5.0 ASTM MOD UNI1 81.0 6.3 11.6 30.8 43.7 4.6 19.2 1.2 2.0 CYC50S 939.8 4.3 6.7 28.9 38.8 4.0 10.0 1.4 2.7 UNI2 963.1 4.4 10.6 32.6 42.8 4.5 22.0 1.0 1.3 CYC10S 1023.7 7.5 10.3 33.8 42.9 4.5 23.5 0.9 2.0 UNI3 1049.2 4.1 10.5 33.3 43.2 4.5 22.8 0.9 1.1 CYC10N 1104.9 10.0 10.1 33.9 43.0 4.5 23.9 0.8 2.6 UNI4 1135.1 3.8 10.4 33.4 43.2 4.5 23.0 0.9 1.0 Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec cm/sec cm/sec cm/sec cm/sec - Q(avg) k(avg) k67 k35 k57 UNI1 81.0 0.601 0.002 7.835 0.001 3.849 0.002 4.630 0.002 CYC50S 939.8 0.823 0.002 5.320 0.002 1.997 0.005 2.696 0.004 UNI2 963.1 0.945 0.003 5.518 0.002 4.400 0.003 4.235 0.003 CYC10S 1023.7 1.385 0.004 9.342 0.002 4.694 0.004 4.136 0.004 UNI3 1049.2 0.890 0.002 5.069 0.002 4.563 0.002 4.205 0.003 CYC10N 1104.9 1.397 0.004 12.441 0.001 4.772 0.004 4.028 0.004 UNI4 1135.1 0.852 0.002 4.735 0.002 4.605 0.002 4.153 0.003 Sample: M500 Stage El. Time Dh67 Dh57 Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) (cm) (cm) (cm) Gradient (cm) Distance - 0.8 2.5 7.5 9.2 i17 5.0 ASTM MOD UNI1 81.0 4.2 10.6 35.1 37.4 4.1 24.5 0.9 1.1 CYC50S 928.5 3.0 -5.9 -10.4 -11.3 -1.2 -4.5 2.6 -4.1 UNI2 985.4 3.2 13.2 35.2 41.6 4.5 22.0 1.2 0.9 CYC10S 1050.9 3.2 14.1 36.7 42.1 4.6 22.6 1.2 0.9 UNI3 1080.2 2.8 14.1 36.8 42.2 4.6 22.7 1.2 0.8 CYC10N 1133.5 3.4 14.0 36.6 42.1 4.6 22.5 1.2 0.9 UNI4 1158.0 2.7 14.1 36.8 42.2 4.6 22.7 1.2 0.7 Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec cm/sec cm/sec cm/sec cm/sec - Q(avg) k(avg) k67 k35 k57 UNI1 81.0 0.401 0.001 5.214 0.001 4.907 0.001 4.241 0.001 CYC50S 928.5 0.690 -0.007 3.731 0.002 -0.901 -0.009 -2.347 -0.004 UNI2 985.4 0.861 0.002 4.034 0.003 4.408 0.002 5.268 0.002 CYC10S 1050.9 1.224 0.003 4.058 0.004 4.526 0.003 5.623 0.003 UNI3 1080.2 0.691 0.002 3.442 0.002 4.548 0.002 5.624 0.002 CYC10N 1133.5 1.233 0.003 4.257 0.004 4.506 0.003 5.610 0.003 UNI4 1158.0 0.708 0.002 3.393 0.003 4.535 0.002 5.632 0.002 R. Hawley 135 Appendix D Sample: M404 Stage El. Time Dh67 Dh57 Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) (cm) (cm) (cm) Gradient (cm) Distance - 0.8 2.5 7.5 8.9 i17 5.0 ASTM MOD UNI1 0.0 3.5 ' 10.5 33.9 44.1 5.0 23.4 0.9 0.9 CYC50S 904.0 0.9 3.6 33.4 39.5 4.4 6.1 1.2 0.9 UNI2 958.4 0.8 6.6 36.1 42.6 4.8 29.6 0.4 0.2 CYC10S 1017.4 17.4 7.3 38.3 43.8 4.9 31.0 0.5 3.5 UNI3 1043.7 0.8 6.2 36.9 42.8 4.8 30.8 0.4 0.2 CYC10N 1097.2 21.5 6.8 38.9 44.1 5.0 32.1 0.4 4.2 UNI4 1260.7 0.7 5.5 37.5 43.1 4.8 32.1 0.3 0.1 Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec cm/sec cm/sec cm/sec cm/sec - Q(avg) k(avg) k67 k35 k57 UNI1 0.0 0.556 0.001 4.421 0.002 4.675 0.001 4.190 0.002 CYC50S 904.0 0.846 0.002 1.078 0.010 1.211 0.009 1.446 0.007 UNI2 958.4 0.854 0.002 0.989 0.011 5.910 0.002 2.624 0.004 CYC10S 1017.4 1.836 0.005 21.692 0.001 6.204 0.004 2.924 0.008 UNI3 1043.7 0.782 0.002 1.030 0.009 6.150 0.002 2.460 0.004 CYC10N 1097.2 1.811 0.004 26.850 0.001 6.420 0.003 2.701 0.008 UNI4 1260.7 0.803 0.002 0.892 0.011 6.414 0.002 2.184 0.005 Sample: M402 Stage El. Time Dh67 Dh57 Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) • (cm) (cm) (cm) Gradient (cm) Distance - 0.8 2.5 7.5 8.3 i17 5.0 ASTM MOD UNI1 81.0 2.4 8.9 34.5 44.1 5.3 25.6 0.7 0.6 CYC50S 933.4 2.2 3.2 32.1 37.9 4.6 5.7 1.1 2.4 UNI2 960.2 2.1 8.2 36.3 • 42.7 5.1 28.1 0.6 0.5 CYC10S 1019.2 8.1 7.9 36.8 42.6 5.1 28.9 0.5 1.7 UNI3 1041.5 2.1 8.2 36.3 42.9 5.2 28.1 0.6 0.5 CYC10N 1103.1 5.2 8.4 35.6 42.6 5.1 27.2 0.6 1.2 UNI4 1125.2 1.8 8.5 34.8 43.1 5.2 26.3 0.6 . 0.4 Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec cm/sec cm/sec cm/sec cm/sec Distance - Q(avg) k(avg) k67 k35 k57 UNI1 81.0 0.591 0.001 2.952 0.002 5.128 0.001 3.564 0.002 CYC50S 933.4 0.847 0.002 2.713 0.004 1.149 0.009 1.277 0.008 UNI2 960.2 1.014 0.002 2.646 0.005 5.619 0.002 3.264 0.004 CYC10S 1019.2 1.443 ' 0.003 10.087 0.002 5.781 0.003 3.169 0.006 UNI3 1041.5 0.984 0.002 2.600 0.005 5.615 0.002 3.278 0.004 CYC10N 1103.1 1.710 0.004 6.528 0.003 5.440 0.004 3.340 0.006 UNI4 1125.2 0.947 0.002 2.267 0.005 5.253 0.002 3.397 0.003 R. Hawley 136 Appendix D Sample: M570 Stage El. Time Dh67 Dh57. Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) (cm) (cm) (cm) Gradient (cm) Distance - 0.8 2.5 7.5 10.8 i17 5.0 ASTM MOD UNI1 81.0 2.2 9.6 28.3 43.7 4.0 18.7 1.0 0.7 CYC50S 930.3 0.9 3.5 26.6 42.1 3.9 15.5 0.4 0.4 UNI2 985.5 0.9 3.8 26.1 42.3 3.9 22.3 0.3 0.3 CYC10S 1045.6 1.9 3.9 27.1 42.6 3.9 23.2 0.3 0.5 UNI3 1068.9 1.0 3.9 26.3 42.5 3.9 22.4 0.3 0.3 CYC10N 1128.9 1.9 4.2 27.1 42.5 3.9 22.9 0.4 0.5 UNI4 1162.3 0.8 4.2 26.5 42.5 3.9 22.3 0.4 0.2 Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec Q(avg) cm/sec k(avg) cm/sec k67 cm/sec k35 cm/sec k57 UNI1 81.0 0.694 0.002 2.768 0.003 3.749 0.002 3.826 0.002 CYC50S 930.3 0.976 0.003 1.169 0.010 3.099 0.004 1.383 0.009 UNI2 985.5 1.041 0.003 1.151 0.011 4.458 0.003 1.533 0.008 CYC10S 1045.6 0.382 0.001 2.434 0.002 4.647 0.001 1.566 0.003 UNI3 1068.9 1.024 0.003 1.201 0.010 4.485 0.003 1.569 0.008 CYC10N 1128.9 0.392 0.001 2.434 0.002 4.587 0.001 1.684 0.003 UN14 1162.3 0.974 0.003 0.939 0.013 4.468 0.003 1.667 0.007 R. Hawley 137 Appendix D PORT COQUITLAM SILTY SAND (PCS) TESTS: Sample: P140 Stage El. Time Dh67 Dh57 Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) (cm) (cm) (cm) Gradient (cm) Distance - 0.8 2.5 7.5 9.0 i17 5.0 ASTM MOD UNI1 81.0 3.5 15.6 40.9 44.8 5.0 25.4 1.2 0.9 CYC50S 961.3 3.6 4.8 32.4 35.1 3.9 27.5 0.4 0.8 UNI2 1014.5 3.4 14.2 41.7 44.8 5.0 27.5 1.0 0.8 CYC10S 1082.8 18.2 26.4 44.0 45.1 5.0 17.5 3.0 6.5 UNI3 1105.8 3.5 14.4 41.9 44.9 5.0 27.5 1.0 0.8 CYC10N 1160.4 16.5 25.1 44.1 45.1 5.0 19.0 2.6 5.4 UNI4 1195.3 3.5 14.1 43.1 44.9 5.0 28.9 1.0 0.8 Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec Q(avg) cm/sec k(avg) cm/sec k67 cm/sec k35 cm/sec k57 UNI1 81.0 0.048 1.18E-04 4.410 1.33E-04 5.075 1.16E-04 6.227 9.45E-05 CYC50S 961.3 0.152 4.78E-04 4.537 4.11E-04 5.504 3.39E-04 1.940 9.61 E-04 UNI2 1014.5 0.023 5.65E-05 4.277 6.58E-05 5.494 5.12E-05 5.685 4.95E-05 CYC10S 1082.8 0.457 1.12E-03 22.793 2.45E-04 3.509 1.59E-03 10.575 5.29E-04 UNI3 1105.8 0.034 8.38E-05 4.314 9.70E-05 5.501 7.61 E-05 5.757 7.27E-05 CYC10N 1160.4 0.688 1.68E-03 20.661 4.07E-04 3.800 2.22E-03 10.033 8.39E-04 UNI4 1195.3 0.026 6.42E-05 4.343 7.37E-05 5.783 5.54E-05 5.658 5.66E-05 Sample: P160 Stage El. Time Dh67 Dh57 Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) (cm) (cm) (cm) Gradient (cm) Distance - 0.8 2.5 7.5 10.4 i17 5.0 ASTM MOD UNI1 81.0 3.9 17.4 45.0 49.9 4.8 27.6 1.3 0.9 CYC50S 906.4 19.9 23.5 44.5 50.0 4.8 21.0 2.2 5.9 UNI2 943.7 3.8 16.7 46.8 50.3 4.8 30.1 1.1 0.8 CYC10S 1008.9 34.2 35.4 46.0 49.1 4.7 10.5 6.7 20.3 UNI3 1031.4 3.7 . 16.1 46.9 48.9 4.7 30.8 1.0 0.7 CYC10N 1098.0 37.3 35.9 44.9 47.6 4.6 9.0 8.0 25.8 UNI4 1125.7 3.4 15.3 45.5 47.3 4.5 30.2 1.0 0.7 Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec Q(avg) cm/sec k(avg) cm/sec k67 cm/sec k35 cm/sec k57 UNI1 81.0 0.017 4.41 E-05 4.836 4.37E-05 5.529 3.82E-05 6.942 3.05E-05 CYC50S 906.4 0.133 3.39E-04 24.934 6.54E-05 4.201 3.88E-04 9.381 1.74E-04 UNI2 943.7 0.020 5.09E-05 4.775 5.15E-05 6.024 4.09E-05 6.665. 3.69E-05 CYC10S 1008.9 0.560 1.45E-03 42.693 1.61 E-04 2.106 3.26E-03 14.178 4.84E-04 UNI3 1031.4 0.030 7.72E-05 4.569 7.95E-05 6.168 5.89E-05 6.431 5.65E-05 CYC10N 1098.0 0.616 1.65E-03 46.588 1.62E-04 1.803 4.18E-03 14.359 5.25E-04 UNI4 1125.7 0.021 5.75E-05 4.223 6.20E-05 6.032 4.34E-05 6.139 4.26E-05 R. Hawley 138 Appendix D Sample: P700 Stage El. Time Dh67 Dh57 Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) . (cm) (cm) (cm) Gradient (cm) Distance - 0.8 2.5 7.5 10.2 i17 5.0 ASTM MOD UNI1 81.0 3.7 15.1 33.7 46.3 4.5 18.6 1.6 1.2 CYC50S 934.5 10.6 12.7 30.5 41.2 4.0 17.7 1.4 3.7 UNI2 971.9 4.5 13.8 32.9 45.0 4.4 19.1 1.4 1.5 CYC10S 1043.7 31.2 33.7 39.6 44.6 4.4 5.9 ..11.4 32.9 UNI3 1066.3 5.1 13.8 32.3 44.0 4.3 18.5 1.5 1.7 CYC10N 1120.0 19.0 30.0 37.2 43.5 4.3 7.2 8.4 16.5 UNI4 1154.8 7.2 13.5 35.3 42.9 4.2 21.9 1.2 2.0 Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec Q(avg) cm/sec k(avg) cm/sec k67 cm/sec k35 cm/sec k57 UNI1 81.0 0.017 4.52E-05 4.615 4.45E-05 3.718 5.52E-05 6.042 3.40E-05 CYC50S 934.5 0.120 3.63E-04 13.195 1.11E-04 3.548 4.13E-04 5.095 2.87E-04 UNI2 971.9 0.010 2.80E-05 5.609 2.20E-05 3.824 3.23E-05 5.527 2.24E-05 CYC10S 1043.7 0.451 1.26E-03 38.965 1.42E-04 1.183 4.67E-03 13.471 4.1 OE-04 UNI3 1066.3 0.020 5.73E-05 6.399 3.86E-05 3.694 6.69E-05 5.535 4.46E-05 CYC10N 1120.0 0.855 2.45E-03 23.691 4.42E-04 1.432 7.31 E-03 12.015 8.71 E-04 UNI4 1154.8 0.024 7.03E-05 8.939 3.31 E-05 4.374 6.77E-05 5.381 5.50E-05 Sample: P500 Stage El. Time Dh67 Dh57 Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) (cm) (cm) (cm) Gradient (cm) Distance - 0.8 2.5 7.5 10.0 i17 5.0 ASTM MOD UNI1 81.0 5.4 14.6 37.2 45.6 4.6 22.6 1.3 1.5 CYC50S 930.9 18.7 19.4 39.6 46.5 4.7 20.1 1.9 5.8 UNI2 963.5 6.8 15.7 38.6 45.6 4.6 23.0 1.4 1.8 CYC10S 1025.6 30.4 29.8 40.3 45.7 4.6 10.5 5.7 18.1 UNI3 1048.1 7.0 15.9 38.6 45.6 4.6 22.7 1.4 1.9 CYC10N 1102.9 30.0 28.4 39.0 45.0 4.5 10.6 5.3 17.6 UNI4 1129.6 7.0 15.9 38.3 45.5 4.6 22.4 1.4 1.9 Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec cm/sec cm/sec cm/sec cm/sec - Q(avg) k(avg) k67 k35 k57 UNI1 81.0 0.021 5.64E-05 6.770 3.80E-05 4.516 5.69E-05 5.837 4.41 E-05 CYC50S 930.9 0.179 4.70E-04 23.358 9.36E-05 4.023 5.43E-04 7.777 2.81 E-04 UNI2 963.5 0.032 8.53E-05 8.486 4.58E-05 4.592 8.47E-05 6.270 6.20E-05 CYC10S 1025.6 0.480 1.28E-03 38.030 1.54E-04 2.104 2.79E-03 11.909 4.93E-04 UNI3 1048.1 0.023 6.30E-05 8.757 3.28E-05 4.543 6.33E-05 6.366 4.51 E-05 CYC10N 1102.9 0.523 1.42E-03 37.469 1.71 E-04 2.125 3.01 E-03 11.362 5.63E-04 UNI4 1129.6 0.018 4.84E-05 8.711 2.53E-05 4.489 4.90E-05 6.347 3.47E-05 R. Hawley 139 Appendix D Sample: P404 Stage El. Time Dh67 Dh57 Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) (cm) (cm) (cm) Gradient (cm) Distance - . 0.8 2.5 7.5 9.3 i17 5.0 ASTM MOD UNI1 81.0 4.6 16.1 39.7 46.9 5.0 23.6 1.4 1.2 CYC50S 932.0 12.4 13.5 36.9 43.6 4.7 23.4 1.2 3.3 UNI2 1013.3 4.3 15.3 38.4 45.3 4.9 23.1 1.3 1.2 CYC10S 1082.6 32.5 27.1 39.1 45.7 4.9 12.0 4.5 17.0 UNI3 1110.8 4.2 15.5 38.8 45.6 4.9 23.3 1.3 1.1 CYC10N 1165.8 28.3 27.2 33.0 36.9 4.0 5.7 9.5 30.8 UNI4 1209.3 18.0 20.1 43.9 48.6 5.2 23.9 1.7 4.7 Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec cm/sec cm/sec cm/sec cm/sec - Q(avg) k(avg) k67 k35 k57 UNI1 81.0 0.018 4.39E-05 5.733 3.86E-05 4.719 4.69E-05 6.443 3.44E-05 CYC50S 932.0 0.199 5.19E-04 15.555 1.56E-04 4.670 5.21 E-04 5.406 4.50E-04 UNI2 1013.3 0.014 3.48E-05 5.385 3.15E-05 4.619 3.67E-05 6.112 2.77E-05 CYC10S 1082.6 0.466 1.16E-03 40.615 1.40E-04 2.390 2.38E-03 10.852 5.25E-04 UNI3 1110.8 0.016 3.93E-05 5.260 3.66E-05 4.651 4.14E-05 6.203 3.11 E-05 CYC10N 1165.8 0.500 1.54E-03 35.364 1.73E-04 1.148 5.33E-03 . 10.895 5.62E-04 UNI4 1209.3 0.055 1.28E-04 22.557 2.96E-05 4.776 1.40E-04 8.023 8.34E-05 Sample: P402 Stage El. Time Dh67 Dh57 Dh37 Dh17 System Dh35 Gradient Ratio (minutes) (cm) (cm) (cm) (cm) Gradient (cm) Distance - 0.8 2.5 7.5 10.9 i17 5.0 ASTM MOD UNI1 81.0 4.8 16.0 39.4 46.4 4.3 23.3 1.4 1.3 CYC50S 932.0 12.6 15.6 38.3 45.9 4.2 22.7 1.4 3.5 UNI2 1013.3 3.4 20.0 40.2 50.6 4.6 20.2 2.0 1.1 CYC10S 1082.6 13.5 20.4 41.6 45.9 4.2 21.2 1.9 4.0 UNI3 1110.8 3.5 20.0 38.6 48.6 4.5 18.6 2.2 1.2 CYC1ON 1165.8 16.6 19.1 43.3 43.3 4.0 24.3 1.6 4.3 UNI4 1209.3 2.4 2.7 42.3 46.2 4.2 39.6 0.1 0.4 Stage El. Time Flow Rate Permeability i67 Permeability i35 Permeability i57 Permeability (minutes) cm3/sec Q(avg) cm/sec k(avg) cm/sec k67 cm/sec k35 cm/sec k57 UNI1 81.0 0.016 4.54E-05 6.021 3.21 E-05 4.665 4.15E-05 6.414 3.02E-05 CYC50S 932.0 0.121 3.52E-04 15.714 9.44E-05 4.533 3.27E-04 6.243 2.38E-04 UNI2 1013.3 0.020 5.31 E-05 4.277 5.76E-05 4.045 6.09E-05 7.990 3.08E-05 CYC10S 1082.6 0.463 1.34E-03 16.907 3.35E-04 4.244 1.33E-03 8.160 6.94E-04 UNI3 1110.8 0.023 6.23E-05 4.372 6.36E-05 3.720 7.47E-05 8.018 3.47E-05 CYC10N 1165.8 1.058 3.26E-03 20.756 6.24E-04 4.850 2.67E-03 7.624 1.70E-03 UNI4 1209.3 0.063 1.82E-04 2.970 2.60E-04 7.919 9.74E-05 1.090 7.08E-04 R. Hawley 140 Appendix D Sample: Stage Distance El . Time (minutes) P 5 7 0 Dh67 (cm) Dh57 (cm) Dh37 (cm) Dh17 (cm) System Gradient Dh35 (cm) Gradient A S T M Ratio M O D UNI1 C Y C 5 0 S UNI2 C Y C 1 0 S UNI3 C Y C 1 0 N UNI4 Continuous Piping During Sample Preparation, therefore no test performed Stage E l . Time Flow Rate Permeability i67 Permeability (minutes) cm3/sec cm/sec cm/sec - Q(avg) k(avg) k67 i35 Permeability cm/sec k35 i57 Permeability cm/sec k57 UNI1 C Y C 5 0 S UNI2 C Y C 1 0 S UNI3 C Y C 1 0 N UNI4 Continuous Piping During Sample Preparation, therefore no test performed R. Hawley 141 Appendix D 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0063728/manifest

Comment

Related Items