Sherif S. AbdelSalam

Enhanced Load-Transfer Analysis for Friction Piles Using a Modified Borehole Shear Test

This study discusses the development and use of a modified borehole shear test (mBST) to improve the prediction of the load-displacement and load-distribution responses for axially loaded friction piles in cohesive soils using the load-transfer analysis method (i.e., t–z analysis). Unlike available approaches that rely on empirical or semi-empirical correlations to generate the shear stress displacement at the soil–pile interface (i.e., t–z curves), the mBST enables direct field measurement of the t–z curves at the soil–pile interface. As part of this study, three full-scale vertical static load tests (SLTs) were conducted on instrumented steel H piles. The t–z analysis was carried out utilizing the TZPILE software with measured t–z curves via the mBST (i.e., the TZ-mBST model). When comparing results of the analysis with the measured responses for the three test piles, it was found that: (1) the TZ-mBST provides proper prediction for the initial part of the measured load-displacement response from SLT results, with a difference not exceeding 10 %; (2) the TZ-mBST analysis provides acceptable prediction of the pile capacity; (3) the TZ-mBST analysis matches the load distribution along the pile length with a maximum difference of 8.3 %; and (4) the analysis with directly measured t–z curves using the mBST provide improved predictions of the pile response when compared to the empirical CPT-based analysis.

Reliability and 3D Modeling of Flexible Walls with EPS Inclusion

AbstractLengthy retaining walls are typically required in several geotechnical applications. Compacted soil is the classic backfill behind such walls, yet soil has a high unit weight that imposes large lateral loads and may lead to outsized wall dimensions. Geofoam or foam made of expanded polystyrene (EPS) can substitute the soil backfill due to its light weight and high compressibility and can reduce the lateral static loads imposed on walls. In this paper, EPS mechanical properties were measured in addition to the interface friction between this geomaterial and other materials, such as sand and concrete. A reliability analysis was conducted to develop the partial resistance factors required for the main EPS properties. Using a three-dimensional (3D) numerical analysis, the reliable EPS properties were used in the hardening soil constitutive model to simulate flexible walls with EPS inclusions of various thicknesses. Analysis outcomes were verified against measurements from a physical prototype and were...

Load and Resistance Factor Design Calibration for Bridge Pile Foundations: Investigation of Design and Construction Practices in Iowa County, Iowa, Jurisdictions

In response to the documented advantage of the load and resistance factor design (LRFD) approach over the working stress design approach, FHWA issued a policy memorandum on June 28, 2000, requiring all new bridges initiated after October 1, 2007, to be designed according to the LRFD approach. This paper presents current design and construction practices for bridge pile foundations that were established through a survey of Iowa county engineers as well as consulting firms that design piles for the county offices. Conducted during the second and third quarters of 2008 as part of an ongoing research study centered on the LRFD approach for pile foundations in Iowa, this study highlights the benefits associated with a thorough description of regional deep foundation design and construction practices on successive LRFD resistance factor calibrations. The specific topics targeted in the study were current foundation practice; the extent of timber pile usage; pile analysis and design; and pile drivability, pile design verification, and quality control. Since this is the first study to have been conducted at the county jurisdictional level on the topic of pile foundation design after the FHWA mandate, the current status of the implementation of the LRFD approach for bridge foundation design was examined. The responses obtained from 44 Iowa county engineers and eight consulting firms revealed that (a) regional variation in pile foundation practice cannot be captured via a state-level investigation, inhibiting the performance of effective regional LRFD calibrations, and (b) the degree of implementation of the LRFD approach at the county level is about 84%.

Modeling Load-Transfer Behavior of H-Piles Using Direct Shear and Penetration Test Results

The load transfer analysis (or t–z analysis) has long been used to predict the load-displacement response of axially loaded driven piles. However, the t–z curves along the pile length and q–z curve at the pile tip, required for the t–z analysis, are routinely obtained based on empirical correlations using field and/or laboratory soil tests. This study focuses on the use of a modified Direct Shear Laboratory Test (mDST) to directly quantify the t–z curves and the use of a penetration test, namely the Pile Tip Resistance test (PTR) to quantify the q–z curve, for partial-displacement piles. As part of this study, two instrumented steel H-piles driven in sandy soils were load tested and soil layers at the two sites were characterized using in situ and laboratory tests. A load transfer analysis was conducted utilizing the directly quantified t–z and q–z curves from the mDST and PTR, respectively, to calculate the response of the load tested piles. When compared to the measured load-displacement response and load distribution along pile length, the t–z analysis based on the mDST and PTR measurements showed very good agreement with the measured pile responses. Therefore, and despite the limited database availability at present, the proposed mDST–PTR based model is promising as it represents a simple and cost-effective means to accurately predict the load–displacement response of partial-displacement piles driven in cohesionless soils.

Reliability of Load-Transfer Approach in the Design of Large Diameter Bored Piles

Pursuing current global trend to practice a reliability-based pile design methodology, most codes have calibrated resistance factors accounting for the strength limit, while the serviceability limit is still being assessed following a deterministic approach. Recently, the Load and Resistance Factor Design (LRFD) parameters were regionally developed by AbdelSalam et al. (2015) for the strength limit utilizing the electronic database “EGYptian Pile Test”. This database contained results from more than 320 pile load test, most of them for large diameter bored piles. In this study, the database was upgraded to include load-transfer outcomes for all the available data using finite difference program Allpile v.6.5, which provided separate skin- and end-bearing behaviors. In addition, the total load-displacement acquired from the load-transfer analysis was employed to develop the LRFD resistance factors for groups of piles sorted by pile diameter, length, and soil conditions. These resistance factors were calculated based on limited total settlement of 1% and 2.5% of the pile diameter, which indirectly accounts for the serviceability limits. A comparison between the strength- and the serviceability-based resistance factors was conducted, and it was found that designs based on the serviceability is more efficient in case of large diameter bored piles.

3D Modeling of EPS Geofoam Buffers Behind Diaphragm Walls

In an attempt to reduce lateral earth pressures acting on diaphragm walls, decrease the dependency on anchors, and optimize the wall structural design, expanded polystyrene geofoam (EPS) was introduced as a compressible buffer between the wall and the retained soil. Based on verified outcomes from the literature, EPS buffers is an effective solution that can significantly reduce the static lateral earth pressure acting on flexible walls. In this paper, a 3D numerical model was developed for a small-size diaphragm wall with EPS buffer using the finite element (FE) program PLAXIS 3D. The constitutive properties utilized in the model were measured as part of the material characterization phase of this research project, and the model was intended to capture the short-term behavior of the retained soil using EPS buffers with various thicknesses. To verify the FE results, a physical instrumented prototype was assembled to mimic the modeled diaphragm wall with EPS. The comparison showed a decent agreement between the FE results and the prototype measurements. From the main outcomes, lateral pressure on diaphragm walls was significantly reduced by around 37% using a relatively thin EPS buffer.

Closure to “Pile Setup in Cohesive Soil. I: Experimental Investigation” by Kam W. Ng, Matthew Roling, Sherif S. AbdelSalam, Muhannad T. Suleiman, and Sri Sritharan

The authors utilized the results of full-scale pile load tests from five fully instrumented small displacement steel H-piles embedded in cohesive soils to illustrate that shaft frictional resistance, rather than end-bearing resistance, is the predominant contributor to pile setup. They also showed that pile setup can be accurately estimated for practical purposes from the soil characteristics determined from consolidation, undrained shear strength, and the standard penetration test (SPT). As noted by the authors, determining pile setup from soil characteristics significantly minimizes the use of static and pile dynamic testing, which are costly and impractical to undertake. The authors also suggested that increased resistance because of pile setup could lead to more cost-effective designs using shorter and a smaller number of piles than would be the case if pile setup was neglected. The subject of pile setup is not new, and over the years there have been a number of empirical methods proposed by several researchers to estimate pile setup. One of the earlier works on pile setup relating the radial consolidation effect on pile setup was by Soderberg (1962). Vesic (1977) used Fig. 13 in his report on the “Design of pile foundations” to show the results of available field data and the theoretical prediction of the effect of time after driving on the bearing capacity for both small and large displacement friction piles in clay. For small displacement piles (H-piles, which is the main focus of the authors’ paper), Vesic (1977) showed that almost all of the maximum pile capacity is achieved within 2 weeks after the end of initial driving of the piles (EOID). However, Vesic (1977) did not mention which methods (static or dynamic)were used to determine themaximumpile capacities and the time (how long after pile installation) when those tests were undertaken. This brings up an important aspect with respect to the reference or benchmark that was used to relate the pile setup in the authors’ paper, because as shown by Komurka et al. (2003), pile setup likely occurs in phases, with the first phase occurring during the nonlinear dissipation of porewater pressure, followed by a linear dissipation, and finally, by aging. It was, however, assumed by the authors that most of the setup was caused by dissipation of porewater pressure. As shown in Figs. 4(a and b) of the authors’ paper, the horizontal coefficient of consolidation decreases with an increase in undrained shear strength and an increase in SPT blow counts. The decrease in the values of the horizontal coefficient of consolidation as the soil becomes stiffer is likely true for undisturbed soils. However, this may not be always be the trend for a soil that is disturbed by a penetrating pile, which tends to create fractures in the soil matrix, and therefore, results in an increase in the coefficient of consolidation of the soil closest to the pile surface. In the absence of static or dynamic pile load tests, the geotechnical resistance of piles is generally obtained from conventional static analysis, whereby effective or total stress methods are used to determine the ultimate resistance, which is then modified by a factor of safety (FOS) in the case of working stress design methods or resistance factors smaller than unity in the case of LRFD design approach. The static formulas do not have any provisions for including pile setup, and as such, these resistances are based on the assumptions made by the designer in determining the most appropriate method and soil parameters to use in determining resistances caused by shaft and end-bearing resistances. An evaluation of the ultimate resistance by static analysis for the ISU 5 pile was made using the total stress approach with a pile adhesion of 75 kPa determined from the Federal Highway Administration (FHWA) National Highway Institute (NHI)-05-042 (FHWA 2006) for an undrained shear strength of 116 kPa. The shaft resistance was determined to be 1,275 kN, and the toe resistance was determined to be approximately 71 kN for a total ultimate capacity of 1,340 kN using the box area of the pile for calculating the shaft friction and the bearing capacity factor of 9 for the end-bearing and cross-sectional area of the box section at the pile toe. The assumption made here is that the pile is fully plugged. This calculated ultimate resistance is larger than that obtained from the dynamic and static load tests andwould have been used to obtain the ultimate limit state (ULS) resistance for structural design based on the LRFD approach. This brings to the fore the premise stated in the “Introduction” section of the authors’ paper—that pile setup integrated in the design processwould result inmore cost-effective designs. However, this would only be true if the pile setup generates a much greater resistance than obtained from the static analysis, which so far has been the customary recommendation made in typical geotechnical reports for the structural engineer. From the authors’ paper, the increase in resistance was evaluated based on dynamic testing at intervals of time after EOID, and the increases in resistance were based on static load testing or wave equation analysis of piles (WEAP) or Case Pile Wave Analysis Program (CAPWAP)-derived resistances. Unless a reference and/ or benchmark is established from which the increase in resistances can be monitored without undertaking the dynamic tests, it is doubtful that the increased resistance caused by setup can be readily integrated in design. This is because projects are generally designed with resistance values that are initially provided rather than resistance values that are anticipated in the long term. However, it is generally recognized that any increase to be obtained with time is an additional resistance that is welcome. This latter aspect is important in evaluating foundations for reuse or in increasing the loads on existing foundations in time. With the trend of monitoring foundation performance by using fiber optic technology in piles, and thereby having the opportunity to monitor resistance with time, the concept of pile setup can be used to advantage in foundation reuse. Although Zeevaert (1973) did not investigate pile set up per se, his investigation undertook tests on undisturbed and remolded samples in 1952 to better understand the behavior of positive skin friction piles used in design. The results of this study, performed on clays in Mexico City, used the coefficient of consolidation on both undisturbed and remolded samples to show that that the effective angle of friction, and therefore, soil strength increased with time. This was aided by using the classical Terzaghi time factor relationship, alongwith the thickness of the disturbed zone around the pile to determine the increase in the time factor with time. In a worked example, Zeevaert (1973) also showed that the calculated capacities