1Assignment 3: (Due: 5:00 pm, 4th November 2016, Online submission) Length of report (not including cover page): 5 to 10 pages 1. AVAILABLE DATA The site investigations for the project were carried out and details are contained in the provided information. The project is a 20 storeys building with commercial ground floor suspended over two levels of basement car parks. The diameter of secant piled wall is 500mm and spacing between the centres of the piles is 800mm. The toe of wall is embedded into soil layer at a depth 11m below the ground level. The wall was designed for construction of two basements. The maximum excavation depth was set to be 8.2m and one level of anchor was adopted to construct the basements. 2. TASKS You are required to assess and interpret the information. Make necessary assumptions if needed. You must indicate the assumptions in your report. The following specific tasks must be completed: a. Based on Figure E1, determine the Soil Profile along Section D-D. Use this to develop a general Soil Profile for your calculation. b. Based on the Bore-logs, in each soil layer, determine (in table):  Unit Weight  Relative Density (if relevant)  Friction Angle (if relevant)  Undrained shear strength, Su (if relevant)  Dilation Angle (if relevant)  Young's Modulus, E  Poisson's Ratio, .  Coefficient of Earth Pressure at Rest, Ko.  Active Pressure Coefficient, Ka. (if relevant)  Passive Pressure Coefficient, Kp. (if relevant) c. Identify and summarise the required soil parameters. Table 1. Soil Parameters (examples) Stratum  (kN/m3) su (kPa) ` (deg) c` (kPa) E (MPa) Soil Layer A 18 - 28 - - Soil Layer B - - - - - : - - - - - Soil Layer n - - - - - d. Determine the Total and Effective vertical stresses (along Section D-D) profile (see Figure E2). Also, determine the Total and Effective horizontal stresses (along Section D-D) profile (see Figure E2). e. Along Section D-D, plot the distributions of active pressure behind the wall and passive pressure in front of the wall (refer to Chapter 6 of Craig's Soil Mechanics). f. Comment on any engineering implications that may impact the design decisions of the retaining structure (basement of the building).Figure E1: Borehole Location Plan (approximately) 0 200 360 450 Total vertical stress, v (kN/m2) Depth 0 200 400 600 Total Horizontal Stress (kN/m2) 40 30 20 10 0 Depth (m) Total Horizontal Stress B C D Total Horizontal Stress (kN/m2) Figure E2: Example of Vertical Stress Profile Note: Refer to Chapter 3 of "Craig's Soil Mechanics".N S W E D D` GA3(A) GA3 BH1 GA1(A) GA1 GA4 BH2 GA2(A) GA2 0 1.5 Scale 1:250 3 4.5BoreholesEXPLANATION OF NOTES, ABBREVIATIONS & TERMS USED ON BOREHOLE AND TEST PIT REPORTS DRILLING/EXCAVATION METHOD AS* Auger Screwing RD Rotary blade or drag bit NQ Diamond Core - 47 mm AD* Auger Drilling RT Rotary Tricone bit NMLC Diamond Core - 52 mm *V V-Bit RAB Rotary Air Blast HQ Diamond Core - 63 mm *T TC-Bit, e.g. ADT RC Reverse Circulation HMLC Diamond Core – 63mm HA Hand Auger PT Push Tube BH Tractor Mounted Backhoe ADH Hollow Auger CT Cable Tool Rig EX Tracked Hydraulic Excavator DTC Diatube Coring JET Jetting EE Existing Excavation WB Washbore or Bailer NDD Non-destructive digging HAND Excavated by Hand Methods PENETRATION/EXCAVATION RESISTANCE L Low resistance. Rapid penetration possible with little effort from the equipment used. M Medium resistance. Excavation/possible at an acceptable rate with moderate effort from the equipment used. H High resistance to penetration/excavation. Further penetration is possible at a slow rate and requires significant effort from the equipment. R Refusal or Practical Refusal. No further progress possible without the risk of damage or unacceptable wear to the digging implement or machine. These assessments are subjective and are dependent on many factors including the equipment power, weight, condition of excavation or drilling tools, and the experience of the operator. WATER Water level at date shown Partial water loss Water inflow Complete water loss GROUNDWATER NOT OBSERVED The observation of groundwater, whether present or not, was not possible due to drilling water, surface seepage or cave in of the borehole/test pit. GROUNDWATER NOT ENCOUNTERED The borehole/test pit was dry soon after excavation. However, groundwater could be present in less permeable strata. Inflow may have been observed had the borehole/test pit been left open for a longer period. SAMPLING AND TESTING SPT 4,7,11 N=18 30/80mm RW HW HB Standard Penetration Test to AS1289.6.3.1-2004 4,7,11 = Blows per 150mm. N = Blows per 300mm penetration following 150mm seating Where practical refusal occurs, the blows and penetration for that interval are reported Penetration occurred under the rod weight only Penetration occurred under the hammer and rod weight only Hammer double bouncing on anvil DS Disturbed sample BDS Bulk disturbed sample G Gas Sample W Water Sample FP Field permeability test over section noted FV Field vane shear test expressed as uncorrected shear strength (sv = peak value, sr = residual value) PID Photoionisation Detector reading in ppm PM Pressuremeter test over section noted PP Pocket penetrometer test expressed as instrument reading in kPa U63 Thin walled tube sample - number indicates nominal sample diameter in millimetres WPT Water pressure tests DCP Dynamic cone penetration test CPT Static cone penetration test CPTu Static cone penetration test with pore pressure (u) measurement Ranking of Visually Observable Contamination and Odour (for specific soil contamination assessment projects) R = 0 R = 1 R = 2 R = 3 No visible evidence of contamination Slight evidence of visible contamination Visible contamination Significant visible contamination R = A R = B R = C R = D No non-natural odours identified Slight non-natural odours identified Moderate non-natural odours identified Strong non-natural odours identified ROCK CORE RECOVERY TCR = Total Core Recovery (%) SCR = Solid Core Recovery (%) RQD = Rock Quality Designation (%) 100 runcoreofLength eredcovrecoreofLength   100 runcoreofLength ofLength eredcovrecorelcylindrica    100 runcoreofLength Axial mm100coreoflengths     Combinations of these basic symbols may be used to indicate mixed materials such as sandy clay. CLASSIFICATION AND INFERRED STRATIGRAPHY Soil and Rock is classified and described in Reports of Boreholes and Test Pits using the preferred method given in AS1726 – 1993, (Amdt1 – 1994 and Amdt2 – 1994), Appendix A. The material properties are assessed in the field by visual/tactile methods. Particle Size Plasticity Properties Major Division Sub Division Particle Size BOULDERS > 200 mm COBBLES 63 to 200 mm Coarse 20 to 63 mm GRAVEL Medium 6.0 to 20 mm Fine 2.0 to 6.0 mm Coarse 0.6 to 2.0 mm SAND Medium 0.2 to 0.6 mm Fine 0.075 to 0.2 mm SILT 0.002 to 0.075 mm CLAY < 0.002 mm 0 10 20 30 40 0 10 20 30 40 50 60 70 80 Liquid Limit (%) Plasticity Index (%) MOISTURE CONDITION AS1726 - 1993 Symbol Term Description D Dry Sands and gravels are free flowing. Clays & Silts may be brittle or friable and powdery. M Moist Soils are darker than in the dry condition & may feel cool. Sands and gravels tend to cohere. W Wet Soils exude free water. Sands and gravels tend to cohere. CONSISTENCY AND DENSITY AS1726 - 1993 Symbol Term Undrained Shear Strength Symbol Term Density Index % SPT "N" # VS Very Soft 0 to 12 kPa VL Very Loose Less than 15 0 to 4 S Soft 12 to 25 kPa L Loose 15 to 35 4 to 10 F Firm 25 to 50 kPa MD Medium Dense 35 to 65 10 to 30 St Stiff 50 to 100 kPa D Dense 65 to 85 30 to 50 VSt Very Stiff 100 to 200 kPa VD Very Dense Above 85 Above 50 H Hard Above 200 kPa In the absence of test results, consistency and density may be assessed from correlations with the observed behaviour of the material. # SPT correlations are not stated in AS1726 – 1993, and may be subject to corrections for overburden pressure and equipment type. FILL GRAVEL (GP or GW) SAND (SP or SW) SILT (ML or MH) CLAY (CL, CI or CH) ORGANIC SOILS (OL or OH or Pt) COBBLES or BOULDERS CL Low plasticity clay CL/ML Clay/Silt OL or ML - Low liquid limit silt CI Medium plasticity clay CH High plasticity clay OH or MH High liquid limit silt OL or ML Low liquid limit siltTERMS FOR ROCK MATERIAL STRENGTH & WEATHERING AND ABBREVIATIONS FOR DEFECT DESCRIPTIONS STRENGTH Symbol Term Point Load Index, Is(50) (MPa) Field Guide EL Extremely Low < 0.03 Easily remoulded by hand to a material with soil properties. VL Very Low 0.03 to 0.1 Material crumbles under firm blows with sharp end of pick; can be peeled with knife; too hard to cut a triaxial sample by hand. Pieces up to 30 mm can be broken by finger pressure. L Low 0.1 to 0.3 Easily scored with a knife; indentations 1 mm to 3 mm show in the specimen with firm blows of pick point; has dull sound under hammer. A piece of core 150 mm long by 50 mm diameter may be broken by hand. Sharp edges of core may be friable and break during handling. M Medium 0.3 to 1 Readily scored with a knife; a piece of core 150 mm long by 50 mm diameter can be broken by hand with difficulty. H High 1 to 3 A piece of core 150 mm long by 50 mm diameter cannot be broken by hand but can be broken with pick with a single firm blow; rock rings under hammer. VH Very High 3 to 10 Hand specimen breaks with pick after more than one blow; rock rings under hammer. EH Extremely High >10 Specimen requires many blows with geological pick to break through intact material; rock rings under hammer. ROCK STRENGTH TEST RESULTS u Point Load Strength Index, Is(50), Axial test (MPa) w Point Load Strength Index, Is(50), Diametral test (MPa) Relationship between Is(50) and UCS (unconfined compressive strength) will vary with rock type and strength, and should be determined on a site-specific basis. UCS is typically 10 to 30 x Is(50), but can be as low as 5. ROCK MATERIAL WEATHERING Symbol Term Field Guide RS Residual Soil Soil developed on extremely weathered rock; the mass structure and substance fabric are no longer evident; there is a large change in volume but the soil has not been significantly transported. EW Extremely Weathered Rock is weathered to such an extent that it has soil properties - i.e. it either disintegrates or can be remoulded, in water. HW DW MW Distinctly Weathered Rock strength usually changed by weathering. The rock may be highly discoloured, usually by iron staining. Porosity may be increased by leaching, or may be decreased due to deposition of weathering products in pores. In some environments it is convenient to subdivide into Highly Weathered and Moderately Weathered, with the degree of alteration typically less for MW. SW Slightly Weathered Rock is slightly discoloured but shows little or no change of strength relative to fresh rock. FR Fresh Rock shows no sign of decomposition or staining. ABBREVIATIONS FOR DEFECT TYPES AND DESCRIPTIONS Defect Type Coating or Infilling Roughness B Bedding parting Cn Clean Sl Slickensided X Foliation Sn Stain Sm Smooth C Contact Vr Veneer Ro Rough L Cleavage Ct Coating or Infill J Joint Planarity SS/SZ Sheared seam/zone (Fault) Pl Planar CS/CZ DS/DZ IS/IZ SV Crushed seam/zone (Fault) Decomposed seam/zone Infilled seam/zone Schistocity Vein Un St Undulating Stepped Vertical Boreholes – The dip (inclination from horizontal) of the defect is given. Inclined Boreholes – The inclination is measured as the acute angle to the core axis.Appendix A: Relevant Empirical Charts2.2.1 Walls and Supports Retaining walls in deep excavation can be divided in three main systems, which are cantilever wall system, propping system and tie-back system. A good designer is required to fully understand of these three systems in order to make a proper judgment of an appropriate retaining system to be adopted. The choice of support system depends on the soil type, depth of excavation, availability of space, budget, its advantage and disadvantage (Ou 2006). The cantilever wall system is the most suitable at shallow excavations where the lateral earth pressure and water pressure is not a problem. However, when excavation goes deeper, a more reliable retaining wall system is required. A propping system or tieback system is suitable in this case. Walls and struts system is widely used in trench excavations. Struts are most vulnerable elements in the retaining system and are usually structural over-designed. The failure of a strut could meet serious consequences and might lead to progressive collapse of the excavation. Buckling failure tend to be sudden, which carries further risk. Moreover, the cost of propping system is usually small compared with the cost of retaining wall. As a result, ef7ficient design of the propping system is encouraged, while a major reduction in overall construction cost should not happen in this area. 2.2.2 Preloading of Support When struts (or tieback anchors) applied in the excavation as supports, preload is often exerted onto struts. Struts can be preloaded at the time of installation. Load is applied to the prop either by jacking between the strut and the waling or by using a jackable flying strut to push back the wall before inserting a strut. Goldberg et al. (1976) and O'Rourke (1981) emphasized that preloading struts is effective in minimizing wall movements. This observation was based primarily on information from case histories. There are two reasons why preloading is beneficial. One is that preloading braces removes slack from connections. Another reason is that the preloading braces reload the soil behind the wall, which also makes soil stiffer. Under normal condition, when the struts are placed at the early stage of excavation (shallow level), the preload is able to push the retaining wall out if the preload is not too small. As the excavation is carried out, if the struts are placed at deeper levels, with the earth pressure growing with depth, the preload of struts will not be capable of pushing the wall outward easily (Ou et al., 1998). Preloading a strut with a higher load than is needed to take up slack in the support system can result in a higher strut load than would otherwise be required. There is a little benefit in introducing additional load in this way (Clough and O'Rourke 1990). Another thing to pay attention with is temperature changes can increase or reduce the strut load and some cases has completely eliminated the effect of pre-loading.2.2.3 Support force determination Limit equilibrium method: This is a simplified method used to determine the support forces in the past. It is usually reliable in determining the support force for a singly propped wall (see Figure 2.1), but not easy to be applied to a multi-propped wall. Pressure envelope method: For multi-propped walls, as it is difficult to predict the pressure distribution acting on the back of the wall, empirical methods have been developed. Pressure envelope method is one of those empirical methods, in which envelopes of apparent earth pressure are established from load measurement in props. The most frequently envelopes are those of Terzaghi and Peck (1967), subsequently modified by peck (1969) as shown in Figure 2.2. Other available methods are those given by Jack (1971) and Henkel (1971), both of which are based largely on a simplified theoretical approach rather than case study data. There methods are less commonly used than Terzaghi and Peck's method because of the relative simplicity of the latter approach and the confidence given by its basis on actual case study data. Deformation methods: with the development of hardware and software of computers, more complex methods of analysis have been well developed and widely available to be utilized in practice. These methods are collectively known as "deformation methods", which can be sub-divides into groups as below:  Beam on springs  Beam on elastic continuum  Finite difference methods  Boundary element methods  Finite element methods 2.3 Previous Research Study in Deep Excavation The papers written by Peck (1969), Lambe (1970), Goldberg et al. (1976), O'Rourke (1981), and Clough and O'rourke (1990) made significant contributions to the geotechnical engineering profession's understanding of deep excavation. Since Peck's landmark paper, technologies in deep excavation has a significant progress owing to the quality of construction that can be achieved, the amount of field performance data available, and the sophistication of analysis that can be performed. 2.3.1 Peck (1969) Peck (1969) considered deep excavations with vertical sides require lateral support. Many important issues, like Lateral movements, ground settlement next to excavations, base failure by heave, method for reducing ground settlement next to excavations, and earth pressure diagram for deep excavation design, are discussed by Peck. The observations in Peck's paper are based on his personal experience and information frompublished case histories. Peck summarized information from case histories on ground settlements adjacent to excavations and showed that settlement next to deep excavations correlate to soil type. The author proposed three zones of settlement profiles based on soil conditions and workmanship in Figure 2.3. There are three major themes highlighted in Peck's discussion of deep excavations. One is the importance of soil type and properties on the performance of deep excavations. The second is the importance of the depth of excavation. The third is the importance of what Peck called "workmanship" in controlling movements. Workmanship includes factors such as prompt installation of support. Poor workmanship (for example, late or sloppy installation of supports) could easily cause larger movements.Figure 2.1: Limit equilibrium methods for a propped wall Figure 2.2: Pressure envelope methods for multi-propped walls (Terzaghi and Peck 1967)Figure 2.3: Summary of settlements adjacent to open cuts in various soils, as function of distance from edge of excavation (Peck 1969) 2.3.2 Lambe (1970) Lambe's (1970) paper on braced excavation focused on design and analysis of deep excavations and their support systems. Lambe reviewed factors which influence the movement of soil due to excavation and the engineering of deep excavations. He included three case histories of excavations for the MBTA subway in Boston, and applied the state of the art in design and analysis to each of the three cases and then compared predictions to measured performance. The author concluded that the state of the art for the design and analysis of braced excavations was far from satisfactory, since support system loads and ground movements could not be predicted with confidence. Lambe also suggested that the finite element method, and experience shared through published case histories, were the two most promising ways for gaining an understanding of deep excavation performance. 2.3.3 Goldberg et al. (1976) Goldberg et al. (1976) published a report with three volumes on design recommendations, design considerations, and construction techniques for lateral support systems. This report is a comprehensive source of information on the state of practice in 1976. The writer estimates maximum horizontal wall movement, maximum ground settlements, and the shape of the settlement profile of the ground surface adjacent to excavations through the measurements and performance of 63 case histories.In the report, Goldberg et al. suggested quantifying the stiffness of the support system by dividing the bending stiffness of the wall by the maximum support spacing (h) raised to the fourth power (EI/h4). 2.3.4 O'Rourke (1981) O'Rourke (1981) examined ground movement caused by braced excavations and related construction activities. He pointed out the importance of site preparation activities on ground movements. He listed relocation and underpinning of utilities, dewatering, support wall construction, and deep foundation installation as a few of the site preparation activities that can cause ground movements. He also studied the relationship between the deflected shape of the excavation support wall and the ratio of horizontal to vertical movement of the ground surface by reviewing performance data from seven case histories. He concluded from his analysis that the ratio if horizontal to vertical movements of the ground surface is 1.6 for pure cantilever deformation and 0.6 for pure bulging deformation of the wall. O'Rourke also drew conclusions about the effects of brace stiffness, pre-stressing of braces, and timing of brace placement. He observed that the effective stiffness of braces could be as low as two percents of the ideal stiffness (AE/L) due to the effects of compression in connections and bending of braces. 2.3.5 Clough and O'Rourke (1990) Clough and O'Rourke (1990) studied the movement due to deep excavation by examining information from case histories and previous studies. They divided movements into two types. One is movement due to the excavation and support process, and the other is movement caused by auxiliary construction activities. They summarized movement information from case histories to aid in estimating maximum wall movements and settlement profile of the ground next to excavations. They concluded from their study that movements due to deep excavations could be predicted within reasonable bounds if the significant sources of movement are considered. Clough and O'Rourke also found that the ratio of the maximum settlement induced by the construction of diaphragm walls to the depth of the trench is 0.15% (Figure 2.4) and illustrated the effect of support system stiffness on wall displacements. (Figure 2.5) 2.4 Empirical Approach on Deep Excavation Project Empirical studies attempt to develop general relationships between observed ground movements and construction activities based on actual observations from a number of similar excavations. As the empirical studies developed, it gradually becomes an isolated analysis approach named observational method, which has been widely used in many deep excavation cases. According to Peck (1969a), the observational method provides a way of ensuring safety while achieving economy in terms of construction cost.Figure 2.4: Envelop of ground surface settlements induced by trench excavations (Clough and O'Rourke, 1990) Figure 2.5: Chart for estimating maximum lateral wall movements and ground surface settlements for support systems in clays (Clough and O'Rourke, 1990)Appendix B: Relevant Empirical Equations and Tables1 4.3.1 Standard penetration Test (SPT) Standard penetration test (SPT) is the most commonly used in-situ test in practice because it is simple, cost effective and can be applied to all types of soils. However, the use of SPT seems to over predict the values of the soil parameters compared to other tests. Energy Correction Skempton (1986) suggested a correction factor N60 based on the standard practices as the average energy ratio of the drop hammer on the drill rod is 55% to 60% of the theoretical free fall energy for SPT. Therefore, 60 percent hammer energy is defined as the correction of SPT N value. Normally, N60 values provide better design parameters when they correlate with strength parameters, bearing capacity, unit weights, liquefaction susceptibility and other properties. The conversion from N to N60 is as follows: 60 60.0 NCCCE N  RSBm Equation 4.1 Where: N60 = the SPT N value corrected for field procedures Em = the hammer efficiency (from Table 4.1) CB = the borehole diameter correction (from Table 4.2) CS = the sample correction (from Table 4.2) CR = the rod length correction (from Table 4.2) N = the measured SPT N value The SPT data is also adjusted using an overburden correction that compensates for the effects of the effective stress. Deep tests in a uniform soil deposit will have higher N values than in the shallow tests of the same soil. So the overburden correction adjusts the measured value to the corrected value as below (N1)60 = CN  N60 Equation 4.2 where: CN = the correction factor for overburden pressure. (N1)60= the N60 value corrected to a reference stress of one atmosphere2 Table 4.1: Hammer efficiency factors for SPT correction (Clayton, 1990) Country Hammer Type Hammer Release Mechanism Hammer Efficiency Em Argentina Donut Cathead 0.45 Brazil Pin weight Hand dropped 0.72 China Automatic Trip 0.60 Donut Hand dropped 0.55 Donut Cathead 0.50 Colombia Donut Cathead 0.50 Japan Donut Tombi trigger 0.78-0.85 Donut Cathead 2 turns + special release 0.65-0.67 UK Automatic Trip 0.73 US Safety 2 turns on cathead 0.55-0.60 Donut 2 turns on cathead 0.45 Venezuela Donut Cathead 0.43 Table 4.2 Borehole, Sample, and Rod Correction Factors (Skempton, 1986) Factor Equipment Variables Value Borehole diameter factor, CB 65-115 mm 1.0 150 mm 1.05 200 mm 1.15 Sampling method factor, CS Standard sampler 1.00 Sampler without liner 1.20 Rod length factor, CR 3- 4 m 0.75 4-6 m 0.85 6-10 m 0.95 >10 m 1.00 The value of CN suggested by Skempton (1986) is as follows: aw N p C /1 2    Equation 4.3 where:  w : the effective overburden pressure at the depth of testing pa : the pressure in atmosphere Another suggestion from was Liao and Whitman (1985) based on laboratory testing for CN and is given as:3 5.0 C N      pw a     Equation 4.4 A comparison was made between two formulae by Kulhawy and Mayne (1990). Basically, the methods give similar corrections for v0  5.0 pa . Therefore, the suggestion from Skempton is used to calculate the SPT N60 in this thesis. Major soil parameters estimation from SPT (i) The friction angle ´: Many researchers provided recommendations correlating the effect friction angle (´) of granular material with SPT N-value (e.g. Meyerholf, 1956, DeMello, 1971, Peck et al. 1974, Schmertmann, 1975, Hatanaka and Uchida, 1996). So far, there were no reliable correlations between in-situ tests and the effective value of friction angle for cohesive soils, Therefore, the total stress analysis will be applied in modelling the cohesive soils and the values of friction angles can be assumed to be zero. Peck et al. (1974) and Meyerholf (1956) recommended a range of friction angles depended on the N value. Details are presented in Table 4.3. In 1971, DeMello published a chart of empirical Correlation between N60 and  for uncemented sands (see Figure 4.1). Three years later, Peck et al. (1974) gave the correction between N value and ′ in the form of a curve (see Figure 4.2). Terzaghi's bearing capacity factors, Nq and Nr have also been included on the same plot. Schmertmann (1975) established a correlation with SPT N-value and overburden pressure as following equations, which has a good agreement with DeMello's empirical correlation chart: 34.0 1 3.202.12 tan'      v a p N   Equation 4.5 Table 4.3: SPT N-value versus the friction angle ′ N value (blows/300 mm) Relative Density Friction angle (degrees) Peck et al. (1974) Meyerhof (1956) < 4 Very Loose < 29 < 30 4 ~ 10 Loose 29 ~ 30 30 ~ 35 10 ~ 30 Medium 30 ~ 36 35 ~ 40 30 ~ 50 Dense 36 ~ 41 40 ~ 45 > 50 Very Dense > 41 > 454 Figure 4.1: Correlation between N60 and for uncemented sands (DeMello, 1971) Hatanaka and Uchida (1996) collected high quality " undisturbed" freezing samples and provided a correlation of the blow-count measured in-situ with the friction angle evaluated in the laboratory. These results have been adjusted from the Japanese 78% efficiency to an equivalent 60% value (designated N60) and normalized to a stress-level of one atmosphere, designated (N1)60, and related to the triaxial-measured value of ′. The equation suggested by Hatanaka and Uchida (1996) was:   N 601 20)(4.15'  Equation 4.6 where the energy-corrected and stress-normalized N-value is obtained from: 5.0 60 601 / )(         vo atm N N  Equation 4.7 The parameter atm =1 bar = 100 kPa = reference stress equal to one atmosphere.5 Figure 4.2: Relationships between SPT N-value and ′, Nq and Nr (Peck et al., 1974) (ii) Undrained shear strength, Su It is required to know the undrained shear strength (Su, also cu) for stability analysis of retaining structures in deep excavation. Only cohesive soils have undrained shear strength while the shear strength of cohesionless soils comes to zero. Su is normally determined by means of the laboratory and field vane shear tests (VST), unconsolidated undrained (UU) compression test. In addition, for saturated fine-grained soils undrained shear strength can be obtained by taking the half of unconfined compressive strength by the unconfined compression (UC) test: Su = qu /2 Equation 4.86 As shown in Table 4.4, Tschebotarioff (1973), Parcher and Means (1968) and Terzaghi and Peck (1967) suggest approximate undrained shear strength for fine-grained soils based on the SPT-N value and consistency. Table 4.4: Relation between SPT N-value and Su for fine-grained soil accordance with consistency N-value Consistency Undrained shear strength Su (kpa) Tschebotarioff (1973) Parcher and Means (1968) Terzaghi and Peck (1967) < 2 Very Soft 15 < 12 < 12.5 2 ~ 4 Soft 15 ~ 30 12 ~ 25 12.5 ~ 25 4 ~ 8 Medium 30 ~ 60 25 ~ 50 25 ~ 50 8 ~ 15 Stiff 60 ~ 120 50 ~ 100 50 ~ 100 15 ~ 30 Very Stiff 120 100 ~ 200 100 ~ 200 > 30 Hard > 225 > 200 > 200 Stroud (1974) has proposed the relation between SPT-N value and undrained shear strength in accordance with plastic index and corrections: Su = f1 N Equation 4.9 f1 is the ratio of Su to the SPT-N value. f1 = f (Ip). It decreases with increasing plasticity index (Ip). The Value of f1 depends on the plasticity index of the clay (see Figure 4.3). In Stroud's point of view, Su varies approximately between 4 and 7. It is taken nearly 4–5 for medium plastic clay, 6 –7 or higher for plasticity index less than 20 and 4.2 for plasticity index more than 30. Figure 4.3: Correlation between SPT N-value and undrained shear strength Su for overconsolidated clays (Stroud, 1974)7 (iii) Poisson's ratio  Poisson's ratio is a function of stresses but can be assume to be a constant. Little information is available for the relationship of Poisson's ratio with SPT N. However, different authors had established certain range of value for: Poulos and Davis (1980) suggested that the following ranges for values of : Overconsolidated stiff clays:  = 0.1~0.2 Medium clays:  = 0.2~0.35 Soft normally consolidated clays: = 0.35~0.45 Besides, Kulhawy and Mayne (1990) also gave: For clay:  = 0.2-0.4 For dense sand:  = 0.3-0.4 For loose sand:  = 0.1-0.3 (iv) Young's modulus E The Young's modulus is the basic property in the elastic model and the Mohr-Coulomb model. However, the value of the Young's modulus depends on many factors such as testing types, confining pressures and loading rates. There is only one type of Young's modulus in cohesionless soil which is drained Young's modulus E′, meanwhile, there are two types of Young's modulus in cohesive soils which are drained and undrained Young's modulus E′ and Eu respectively. For cohesive soils, Kulhawy and Mayne (1990) recommended the undrained modulus as: For soft clay: Eu = 1.5-4 (Mpa) For stiff clay: Eu = 8-20 (Mpa) A number of authors (Ohya et al. 1982; Tsuchiya and Toyooka, 1982; Leach and Thompson, 1979; Webb, 1970) established different relations for various soils and a summary is presented in Figure 4.4.8 Figure 4.4: Relationship of equivalent undrained Young's Modulus values and SPT N-values For cohesionless soils, Poulos and Davis (1980) gave the range of drained Young's modulus for sands as: For loose sand: E′ = 10-20 Mpa For medium sand: E′ = 20-50 Mpa For dense sand: E′ = 50-100 Mpa For fine sands: E′ = 0.5 N60 MPa For clean sands: E′ = N60 MPa In this study, correlations by other authors (D'Appolonia and D'Appolonia, 1970; Denver, 1982; Schutze and Menzenbach, 1961; Webb, 1970; Ohya et al., 1982; Tsuchiya and Toyooka, 1982; Komomik 1974; Chrisltoulas and Pachakis, 1987; Yamashita et al., 1987) are also collected and summarised in Figure 4.5. Correlation Method 1. Clays (Ohya et al.., 1982) Lateral load Lester 2. Mudstone (Tsuchiya and Toyooka, 1982) Pressure meter 3. Mudstone (Leach and Thompson, 1979) Pile tests 4. Alluvial clays ( Tsuchiya and Toyooka, 1982) Pressuremeter 5. Glacial clays (Tsuchiya and Toyooka, 1982) Pressuremeter 6. Clays and sands (Webb, 1970) -----9 Figure 4.5: Correlation between equivalent drained Young's Modulus values with SPT N-values Correlation Method 1. NC Sand (D'Appolonia and D'Appolonia, 1970) ----- 2. OC sand (D'Appolonia and D'Appolonia, 1970) Driven piles 3. Dry sand ( Denver, 1982) Screwplate/ pressuremeter 4. Sand (Schutze and Menzenbach, 1961) ----- 5. Saturated sand (Webb, 1970) ----- 6. Sands (Ohya et al., 1982) Lateral load tester 7. Alluvial sands (Tsuchiya and Toyooka, 1982) Presuremeter 8. Glacial sands (Tsuchiya and Toyooka, 1982) Presuremeter 9. Sand (Komomik 1974) Driven piles 10. Sand (Chrisltoulas and Pachakis, 1987) Driven piles 11. Sands (Yamashita et al., 1987) Driven piles10 4.3.2 Cone Penetration Testing (CPT) The cone penetration test (CPT) is similar to the standard penetration test (SPT). The distinct difference is that, a steel cone is pushed into the soil in CPT test while a thick-walled sampler is driven into the soil in SPT test. The main types of cone penetration devices include mechanical cone, mechanical-friction cone, electric cone and piezocone. The study here of cone penetration test is particularly focus on Electric Cone Penetration Test (CPT) and Piezocone Test (CPTU), known as Cone Penetration Test with pore water measurement. The advantages of CPT test are continuous measurements of resistance to penetration of the cone, and resistance of a surface sleeve. From those measurements, parameters of cone resistance qc, Sleeve friction fs can be obtain. In the piezocone penetrometer, pore measure is measured typically at one, two or three locations as show in Figure 4.6. These pore pressure are known as: on the cone (u1), behind the cone (u2) and behind the friction sleeve (u3). Figure 4.6: Representation of cone penetrometer. The empirical correction approaches available for interpretation of undrained shear strength Su from CPT/CPTU results can be group under three main categories as follows: (i). Su estimation using total cone resistance. The estimation of Su from CPT using cone resistance is made from the following equations:   k voc u N q S   Equation 4.10 Where: Nk is an empirical cone factor and vo is the total on situ vertical stress. The cone factor, Nk, is an important parameter that enables Su to be estimated form measurements of cone tip resistance, qc, from the CPT. The cone factor is generally assumed to be constant for a particular clay layer. Kjekstad et al. (1978) showed that for overconsolidated clays, with Su from triaxial compression tests as the reference strength, an average value of Nk was 17. Lunne and11 Kleven (1981) showed that for normally consolidated marine clays with field shear vane test as the reference test, Nk varies between 11 and 19 with an average value of 15. A modification and improvement of the above approach, using CPTU data, is to employ cone resistance corrected for pore pressure effects, qt, instead of measured cone resistance qc. The cone factor is expressed as:   u vot kt S q N   Equation 4.11 Aas et al. (1986) presented corrections between cone factor Nkt and plasticity index (see Figure 4.7). The values Nkt of varied between 8 and 16 for plasticity index (Ip) varing between 3% and 50%. The Nkt increases with increasing plasticity. La Rochelle et al. (1988) showed that using Su from triaxial compression tests as the reference strength, Nkt varied from 8 to 29. Another research in 1988 by Powell and Quarterman shows that Nkt varied from 10 to 20 depending on plasticity index (Ip), also based on the triaxial compression tests. Figure 4.7: Relations between Computed cone factor Nkt and Plasticity index Ip (Aas et al. 1986) (ii). Su estimation using effective cone resistance. According to the study from Senneset et al. (1982), the expression of undrained stress Su can be found as follows:   ke t ke e u N uq qN S   2 Equation 4.12 Where, qe is the effective cone resistance, defined as the difference between the measured cone resistance and pore pressure. It is measured immediately behind the cone (u2).12 Senneset et al. (1982) indicated that the value of Nkt is equal to 93. Lunne et al. (1985) showed that varied between 1 and 13. It appears to correlate with the pore pressure parameter Bq. Karlsrud et al. (1996) used triaxial compression tests on high quality block samples to obtain Su values. Their resulting Nke and Bq plots results in a rather narrow band. (see Figure 4.8) Figure 4.8: Relationship between Cone factor Nke and Pore pressure parameter Bq (Karlsrud et al. 1996) (iii). Su estimation using excess pore pressure. A number of authors proposed the relationships between excess pore pressureu and Su. These relationships have the form of u u N u S    (   uuu 02 ) Equation 4.13 Lunne et al. (1985) found Nu to correlate well with Bq and to vary from 4 to 10 for North Clay by taking triaxial compression test strength as the reference strength. La Rochelle et al. (1988) found Nu varied between 7 and 9 for three Canadian Clays by using uncorrected field shear vane strength as the reference strength. Karlsrud et al. (1996) obtained Nu values varied between 6 and 8 with no clear dependency on Bq by using Su values from triaxial compression test on block samples (see Figure 4.9).13 Figure 4.9: Relationship between Cone factor Nu and Pore pressure parameter Bq (Karlsrud et al. 1996) The undrained Young's Modulus, Eu, is usually made using correlations with the undrained shear strength, Su, in the form: u   SnE u Equation 4.14 where n is a constant that depends on shear stress level, overconsolidation ratio, clay sensitivity and other factors (Ladd et al.1977). Because soil behaviour is non-linear, the choice of relevant shear stress level is very important. Figure 4.10a presents data for normally consolidated soils from Ladd et al.(1977) that shows the variation of the Eu/Su with stress level for seven different cohesive soils (15< Ip<75). Figure 4.10b shows the variation of Eu/Su with overconsolidation ratio (OCR) at two shear stress levels for the same soil types shown in Figure 4.10a. The drained Young's modulus, E, in sand mainly depends on relative density, overconsolidation ratio and current stress level. Figure 4.11 presents a chart to estimate the secant Young's modulus (E′s) for an average axial strain of 0.1% for a range of stress histories and ageing. A review of calibration chamber test results was made by Robertson and Campanella (1983) to compare measured cone resistance (qc) to measured peak secant friction angle (′). The resulting comparison is shown in Figure 4.12. The correlation for uncemented moderately incompressible, predominately silica sands proposed by Robertson and Campanella (1983) is shown in Figure 4.13, where qc increases linearly with  vo  for constant ′.14 Figure 4.10: Stiffness ratio, Eu/Su, as function of Ip (Ladd et al.1977)15 Figure 4.11: Evaluation of drained Young's modulus for silica sands (Baldi et al.1989) Figure 4.12: Relationship between bearing capacity number and friction angle from large calibration chamber tests (Robertson & Campanella 1983)16 Figure 4.13: Relationship among v , qc and ′ (Robertson & Campanella 1983) 4.3.3 Vane Shear Test The field Vane Shear Test (VST) is a means of determining the in-situ undrained shear strength of soft to medium stiff clays and silts. This consists of a cruciform vane on a shaft (Figure 4.14). The vane is inserted into the clay soil and a measured increasing torque is applied to the shear until the soil fails as indicated by a constant or dropping torque by shearing on a circumscribing cylindrical surface. During rotation, the torque (T) is measured and the maximum torque (Tmax) is used to calculate the undrained shear based on the vane geometry. Prior to calculation of undrained shear strength Su, the torque associated with rod friction (Trod) must be subtracted from the measured torque (Tnet=Tmax-Trod). Best practice involves using a sheath (or a slip coupling) to eliminate the rod friction, and thus Tnet would equal Tmax. The test is carried out rapidly. If suv is the undrained shear strength in the vertical direction, and suh is the undrained shear strength in the horizontal direction, then the maximum torque is     net uv suh D T D Hs 2 3 2  Equation 4.15 where H is the vane height and D is the vane diameter.17 Figure 4.14: Representation of the shear vane For isotropic soil:   sss uuhuv Equation 4.16 Whence, net su D T D H      32 2  Equation 4.17 The values of undrained shear strength from field vane tests are likely to be higher than can be mobilized in practice (Bjerrum 1972). This is attributed to a combination of anisotropy of the soil and the fast rate of shearing involved with the vane shear test. Because of these factors, Bjerrum (1972) proposed that, for field vane shear tests performed on the saturated normally consolidated clays, the undrained shear strength su be reduced according to the plasticity index of the clay. Figure 4.14 shows the updated Bjerrum (1972)'s field vane shear test correction factor by Ladd et al (1977), where the in situ undrained shear strength determined from the field vane test times the correction factor determined from Figure 4.15.18 Figure 4.15: Updated Bjerrum (1972)'s field vane shear test correction factor by Ladd et al. (1977)4.3 Standard Penetration Test (SPT) Results 4.3.1 Introduction The standard penetration tests (SPT) was performed in the two projects because it can advance through hard stratum compared to the CPT. According to the borehole reports supplied by the special geotechnical company, the SPT tests followed the same procedure at both projects: A standard 50mm diameter thick-walled split tube sampler is driven up to 450mm into the ground from the bottom of the borehole by a 63.5kg mass hammer with 760mm freefall. The blows required to penetrate each 150mm (or part of) are recorded. Where the full 450mm penetration is achieved the total blows over the final 300mm are recorded as the "N" value for the test. 4.3.2 Corrected SPT Test Results Schmertmann and Palacios (1979) showed that measured SPT blow counts are inversely proportional to the energy ratio for blow counts less than 50. Seed et al. (1985) and Skempton (1986) subsequently proposed that measured blow counts should be corrected to the value that would have been recorded if a standard amount of energy had been transmitted through the rods. A standard value of 60% of the hammer potential energy has been adopted because it is the historical average measured for most drill rigs and operators. The energy corrected SPT blow count (N60) SPT is calculated as follows: 60 60.0 NCCCE N  RSBm Eq. 4.1 Where: N60 = SPT N value corrected for field procedures Em = hammer efficiency CB = Borehole diameter correction CS = sample correction CR = rod length correction N = measured SPT N value The SPT data also may be adjusted using an overburden correction that compensates for the effects of the effective stress. Deep test in a uniform soil deposit will have higher N values that shallow tests in the same soil. So the overburden correction adjusts the measured N values to (N1)60 (Liao and Whitman, 1985): ' 100 )( 601 60 60 z N kPa NCNN   Eq. 4.2 Where, (N1)60 = N60 value corrected to a reference stress of one atmosphere ; CN = correction factor for overburden pressure. σz' = Vertical effective stress at the test location.4.4 Soil Parameters 4.4.1 Unit Weight, γunsat, γsat Bowles (1996) presented empirical values for unit weight of granular soil based on SPT at about 6m depth and normally consolidated as given in Table 3-4 below. 4.4.2. Relative Density, Dr The relative density of sands may also be estimated from N60 ( see Jamiolkowski et al, 1988, Skempton 1986 ) as 60 ) 5.0 60 Dr  (100 N Eq. 4.3Where, N60 = Penetration resistance normalized to an effective energy delivered to the drill rod at 60 percent of theoretical free-fall energy, blows/300 mm. Where Dr > 35 percent, N60 should be multiplied by 0.92 for coarse sands and 1.08 for fine sands. 4.4.3. Friction Angle,  SPT test can be one of the methods to estimate the friction angle of cohesionless soil. Table 3-4 summarised the research on correlation of friction angle to SPT N value. 4.4.4. Dilation Angle, ψ In order to shear sand, the grains must physically ride over each other. This requires the sand to expand in the direction perpendicular to the shear. This expansion is known as dilation. When the soil is loose, the shearing process will actually cause contraction rather than dilation, as the sand particles readily bed in to a denser structure. Sands can display behaviour between these two extremes depending on the particular relative density. Apart from heavily over-consolidated layers, clay soils tend to show little dilation (≈0). Rock dilation also tends to be zero. The dilation of sand depends on both the density and on the friction angle. For quartz sands, the order of magnitude is    300 . For  - value of less than 30°, the angle of dilation is almost zero. A small negative value for the angle of dilation is only realistic for extremely loose sands. Table 3-8 shows the typical values for dilation angles. 4.4.5 Young's Modulus, EFor over-consolidated (compacted) cohesionless soil the E values are approximately proportional to the corrected SPT N value according to the equation: Young's modulus (kN/m2) = F x SPT N value Eq. 4.4 where F is in the range 2000 to 6000 for retaining walls in sands and gravels. 4.4.6 Poisson's Ratio, υ The selection of a Poisson's ratio is particularly simple when the elastic model or MohrCoulomb model is used for gravity loading. h v K    0 Eq. 4.5 As both models will give the well-know ratio of  )1(      h v Eq. 4.6 For one-dimensional compression it is easy to select a Poisson's ratio that gives a realistic value of K0. Hence, υ is evaluated by matching K0. In many cases one will obtain υ values in the range of 0.3 to 0.4. In general, such values can also be used for loading conditions other than one-dimensional compression. For unloading condition, however, it is more common to use values in the range between 0.15 and 0.25. Bowles (1993), Das (1994), 4.4.7 The Coefficient of Earth Pressure at Rest, Ko Schmidt, (1966), Mayne & Kulhawy, (1982), Hayat (1992) and Michalowski (2005) researched on this parameter intensively. There are some empirical relationships to obtain K0 as given below: For perfect elastic materials (Mohr-Coloumb Model) '1 ' 0    K  Eq.4.7 Where, υ' = Poisson's ratio. Normally consolidated loose sand (Jaky, 1944) K 0  sin1  Eq. 4.8 Dense compacted sand (Sherif et al, 1984)   5.5)sin1( 1 min 0   K  x compact Eq. 4.9 Where, ρcompact = actual compacted dry density;  min = minimum dry density (loosest state) of the sand. Normally consolidated clays (Brooker & Lreland, 1965) K 0  sin95.0  Eq. 4.10 Over-consolidated sand and clays (Mayne and Kulwahy, 1982) sin OC  NC )(0)(0 OCRKK Eq. 4.11 4.4.8 Active Pressure Coefficient Ka and Passive Pressure Coefficient Kp According to the Mohr- Coloumb theory, the active pressure coefficient Ka and passive pressure coefficient Kp are given by   sin1 sin1    K a ,   sin1 sin1    K p Eq.4.12