classified it as clay or silty/sandy clay. Due to the close proximity of the four borings to each other, it did not seem likely that the soil composition would vary appreciably between each boring, so it was assumed that the cohesive soils were clay. As the majority of the varved soil at the middle of the adjacent building is clay, the whole layer between the fill and decomposed rock was considered to be clay in the analysis. Mohr-Coulomb Soil Model — The simple linear elasto-plastic Mohr-Coulomb (MC) soil model was used in the numerical analyses with five input parameters: Young’s modulus (E), Poisson’s ratio (v), cohesion (c), friction angle (ϕ) and dilatancy angle (ψ). The limitation of this soil model is that it cannot capture the small strain stiffness that more advanced soil models can, (e.g., hardening soil model, MIT-S1, etc.) but the latter require many more input parameters that are obtained from project specic lab testing. For this project, only SPT N-values and visual classication were available in determining engineering properties. Due to the lack of data, a simplied MC model was selected, and input parameters determined using empirical correlations and engineering judgement. Details about estimating the soil properties from SPT N-values can be found in Maniscalco et al. (2018). Since the correlations used in this study depend solely on SPT N-values, it is important that the field N-values were corrected to 60% hammer efficiency, as most of the correlations in literature use this normalization to eliminate variation in data. The majority of correlations available for coarse-grained soils using SPT N-values (e.g., to determine ϕ and E) are for sands; thus, assumptions have to be made when using correlations to estimate the properties of the fill and decomposed rock layers. For clay, the correlations between SPT N-values and S are reasonable, while u correlations with E are much more approximate and most references provide ranges of values based on qualitative consistency of clay. Modelling of Secant Wall and Internal Bracing — The secant pile wall was modeled using plate elements with equivalent elastic bending stiffness of the composite wall elements consisting of the 78 • DEEP FOUNDATIONS • JULY/AUG 2019 Fill Soil Model N Values (blow/ft) Average N60 (blow/ft) Drainage Type 60 Soil Unit Weight, γ (pcf) Drained Elastic Modulus, E' (ksf) Poisson's Ratio, v' Cohesion, c' (ksf) Friction Angle, ϕ' (deg) Undrained Shear Strength, Su (ksf) Dilatancy Angle, ψ (deg) MC 4 to 20 10 Drained 120 400 0.30 0.0 32 - 0.0 Clay MC 4 to 23 13 Undrained B 115 158+9.2z 0.40 - - 0.88+0.06z - Soil property input parameters for PLAXIS model Secant Wall Structure Type Axial Stiffness, EA (kip/ft) Bending Stiff., EI (kip-ft2/ft) Width, d (ft) Poisson's Ratio, v' Modulus of Elasticity, E (ksf) Unit Weight, γ (pcf) Area, A (ft2) Moment of Inertia, I (in4) Spacing, Lspacing (ft) Plate 2,289,000 1,667,000 2.956 0.15 - - - - - steel core beam in secondary piles and concrete “lagging” in between. The overall moment of inertia (I) was calculated for the wall per unit length based on the geometry of the secant wall, while a composite E accounted for the concrete and steel beam by taking a weighted average of each material’s contribution to the stiffness. The three levels of internal corner bracing consisted of wide flange steel walers that spanned the full length of the wall. The layout was similar for each bracing level but the member sizing varied per level. The bracing was modeled in PLAXIS using elastic anchor elements that capture the average bracing stiffness per level of bracing and per unit length of wall. The axial stiffness of the bracing per unit length of wall was calculated by deter- mining the average equivalent perpen- Strut: Strut: Levels 1&2 Level 3 8,227 - - - - - - - 15.92 Structural property input parameters for PLAXIS model dicular strut stiffness of each bracing level and then by dividing that stiffness by the strut spacing. Modelling of Adjacent Building — Limited information on the adjacent 25- story building and its foundations was available. From test pits excavated adjacent to the building, it was clear that the building basement was founded on H- piles. Based on the geotechnical report, the piles were assumed to be HP10x57. In NYC, H-piles are typically used as end bearing piles, so they were assumed to be bearing on rock. To incorporate the effects of the adjacent building’s basement and its pile foundations on the west wall, a distance of 13 ft (4 m) of soil behind the wall was deactivated in PLAXIS and supported by a rigid box consisting of plate elements 11,566 - - - - - - - 15.92 8 ea. Square Pile Group F.E. Anchor F.E. Anchor Embedded Beam Row - - - - 155,900 136 25 120.3 14 Decomposed Rock Rock MC 9 to REF 30 Drained 140 1555+132.5z 0.30 0.0 38 - 0.0 Linear Elastic - - Drained 176 238,217 0.20 - - - -
