dfi journal - deep foundations institute “simplified procedure” for evaluating...
TRANSCRIPT
Vol. 7, No. 1 August 2013
DFI JOURNALThe Journal of the Deep Foundations Institute
PAPERS:Commentary on the Selection, Design and Specifi cation of Ground Improvement for Mitigation of Earthquake-Induced Liquefaction – Ground Improvement Committee of DFI [3]
Liquefaction Mitigation Synthesis Report prepared for the Ground Improvement Committee of the DFI – Timothy C. Siegel [13]
Grouted Micropiles for Foundation Remediation in Expansive Soil (8th Michael W. O’Neill Lecture) – John D. Nelson, Kuo-Chieh Chao, Daniel D. Overton,
Zachary P. Fox, Jesse S. Dunham-Friel [32]
Relationship between Installation Torque and Axial Capacities of Helical Piles in Cohesive Soils – Mohammed Sakr [44]
Ultimate Lateral Resistance of Piles in Cohesive Soil – Lassad Hazzar, Mourad Karray, Mounir Bouassida, Mahmoud N. Hussien [59]
TECHNICAL NOTE: Direct Solution of the Brinch-Hansen 90% Pile Ultimate Failure Load – Don W. Dotson [69]
Deep Foundations Institute is the Industry Association of Individuals and Organizations Dedicated to Quality and Economy in the Design and Construction of Deep Foundations.
DFI JOURNAL Vol. 7 No. 1 August 2013 [1]
From the Editors and Publisher 2013 DFI Board of TrusteesPresident:Robert B. BittnerBittner-Shen ConsultingEngineers, Inc.Portland, OR USAVice President:Patrick BerminghamBermingham Foundation SolutionsHamilton, ON CanadaSecretary:Matthew JanesIsherwood AssociatesBurnaby, BC CanadaTreasurer:John R. WolosickHayward Baker Inc. Alpharetta, GA USAImmediate Past President:James A. MorrisonKiewit Infrastructure Engineers Omaha, NE USAOther Trustees:David BorgerSkyline Steel LLCParsippany, NJ USAMaurice BottiauFranki Foundations BelgiumSaintes, BelgiumDan BrownDan Brown and Associates, PLLCSequatchie, TN USAGianfranco Di CiccoGDConsulting LLCLake Worth, FL USARudolph P FrizziLangan Engineering &Environmental Services Elmwood Park, NJ USABernard H. HertleinGEI Consultants Inc. Libertyville, IL USAJames O. JohnsonCondon-Johnson & Associates, Inc.Oakland, CA USADouglas KellerRichard Goettle, Inc.Cincinnati, OH USASamuel J. KosaMonotube Pile CorporationCanton, OH USAKirk A. McIntoshAMEC Environment & Infrastructure, Inc.Jacksonville, FL USARaymond J. PolettoMueser Rutledge Consulting EngineersNew York, NY USAArturo L. Ressi di CerviaKiewit Infrastructure GroupWoodcliff Lake, NJ USAMichael H. WysockeyThatcher Engineering Corp.Chicago, IL USA
Journal PublisherManuel A. Fine, B.A.Sc, P.Eng
Journal EditorsAli Porbaha, Ph.D., P.E. Central Valley Flood Protection Board Sacramento, CA, USADan A. Brown, Ph.D. Dan Brown and Associates, Sequatchie, TN, USAZia Zafir, Ph.D., P.E. Kleinfelder Sacramento, CA, USA
Associate EditorsLance A. Roberts, Ph.D., P.E.RESPEC Consulting & ServicesRapid City, SD USAThomas Weaver, Ph.D., P.E.Nuclear Regulatory CommissionRockville, MD USA
Published By Deep Foundations Institute
Copyright © 2013 Deep Foundations Institute. AII rights reserved. Written permission must be obtained from DFI to reprint journal contents, in whole or in part.
ContactDFI Headquarters326 Lafayette AvenueHawthorne, NJ 07506staff@dfi .orgwww.dfi .org
DFI, its directors and offi cers, and journal editors assume no responsibility for the statements expressed by the journal’s authors. International Standard Serial Number (ISSN): 1937-5247
Mission/Scope The Journal of the Deep Foundations Institute publishes practice-oriented, high quality papers related to the broad area of “Deep Foundations Engineering”. Papers are welcome on topics of interest to the geo-professional community related to, all systems designed and constructed for the support of heavy structures and excavations, but not limited to, different piling systems, drilled shafts, ground improvement geosystems, soil nailing and anchors. Authors are also encouraged to submit papers on new and emerging topics related to innovative construction technologies, marine foundations, innovative retaining systems, cutoff wall systems, and seismic retrofit. Case histories, state of the practice reviews, and innovative applications are particularly welcomed and encouraged.
DFI JOURNAL
The DFI Journal has been encouraging the Technical Committees of the DFI to produce committee authored papers describing state-of-the-art subjects within the realm of their committee’s special interests. We are pleased to include in this edition the first such paper, “Commentary on the Selection, Design and Specification of Ground Improvement for Mitigation of Earthquake-Induced Liquefaction, authored by the Ground Improvement Committee. A companion paper which reports" the results of a DFI Committee Project Fund program follows, in the form of a report authored by Timothy C. Siegel entitled "Liquefaction Mitigation Synthesis Report". We are hopeful that will set a precedent, to be followed by other committee authored papers.
This edition includes a paper authored by John D. Nelson et al, “Grouted Micropiles for Foundation Remediation in Expansive Soil”, which was the subject of the 8th Michael W. O’Neill Memorial Lecture. There is also another paper from Mohammed Sakr, who has been a prolific contributor to the DFI Journal on the subject of Helical Piles, which discusses the relationship between installation torque and axial capacity of helical piles in cohesive soils.
We wish to thank our dedicated reviewers for their attention to detail and service to the authors, the DFI Journal publication, and the industry in providing input to enhance quality of the papers. We are pleased to report that new names are popping up as well as repeat services by some very responsive reviewers.
We continue to encourage submission of case history papers in particular. We are also open to publishing another themed edition and again request that any Technical Committee desiring to have their topic as the focus of a future themed edition contact the Publisher.
Other comments, suggestions, and submissions are welcome and may be submitted via the DFI website at www.dfi.org.
DFI JOURNAL Vol. 7 No. 1 August 2013 [3]
Commentary on the Selection, Design and Specifi cation of Ground Improvement for Mitigation of Earthquake-Induced LiquefactionBy the Ground Improvement Committee of The Deep Foundations Institute
ABSTRACTThe evaluation of earthquake-induced liquefaction has become a routine part of geotechnical engineering design. For a given project, if an analysis identifies a potential for liquefaction and the consequences of liquefaction are deemed unacceptable, then some form of hazard mitigation is required. Mitigation efforts may consist of removing the liquefiable soils, bypassing the liquefiable soils with deep foundations, structurally accommodating the deformations or strength loss caused by liquefaction, or preventing the onset of liquefaction through ground improvement. The fundamental ground improvement mechanisms for liquefaction mitigation include densification, drainage, and reinforcement. When evaluating, recommending and specifying various ground improvement methods for liquefaction mitigation, practitioners should understand the fundamental mechanics involved and applicability and limitations of the various methods. The DFI Ground Improvement Committee offers a review of the fundamental mechanics and commentary on the applicability and limitations of each method to provide clarity and guidance on the issues related to ground improvement for liquefaction mitigation.
INTRODUCTIONLiquefaction and its effect on engineered structures was recognized as an earthquake hazard in the 1960s after the widespread liquefaction-induced damage caused by the 1964 Niigata (Japan) and Alaskan earthquakes (Seed and Lee, 1966; Seed and Idriss, 1967). Independently and concurrently, Whitman (1971) and Seed and Idriss (1971) proposed a “simplified procedure” for evaluating earthquake-induced (cyclic) liquefaction. The simplified procedure evaluates the potential for liquefaction based on the relationship between earthquake-generated cyclic shear stresses and empirically-based liquefaction resistance as a function of field testing (e.g., Standard Penetration Test N-values, Cone Penetration Test tip resistance, etc.). Alternative methods for the evaluation of cyclic liquefaction have been proposed by others (e.g., Arulmoli et al, 1985; Poulos et al, 1985; Kayen and Mitchell, 1997; Andrus and Stokoe, 2000) but the simplified procedure remains the most commonly used liquefaction evaluation methodology. Although liquefaction analysis and the design of liquefaction mitigation have been part of engineering practice in the western United States for at least 40 years and guidelines are in place (CGS, 2008; Martin and Lew, 1999), only with the widespread adoption
of the International Building Code (IBC) in 2000 did earthquake hazards become an important design consideration in the central and eastern U.S.
As a result of the increased seismic demand presented in the IBC, many sites throughout the U.S. are classified as liquefiable under the design earthquake parameters. In the case of the IBC, engineers are instructed to “address” liquefaction. Options for addressing liquefaction include the following: 1) move the project to a different site that is not liquefiable, 2) design the structure to withstand liquefied conditions, or 3) use ground improvement to reduce the risk of liquefaction to an acceptable level. Because moving the project or designing for the consequences of liquefaction is often technically or financially unfeasible, liquefaction mitigation by ground improvement is frequently a preferred option.
The membership of the Deep Foundations Institute (DFI) includes government agency engineers, private consultants, and contractors; all of which have significant roles in the design, construction, and evaluation of ground improvement methods for the mitigation of liquefaction. As such, the DFI has a vested interest in examining the state-of-practice for the benefit of its members and their clients. This paper describes the fundamental
[4] DFI JOURNAL Vol. 7 No. 1 August 2013
mechanics of the most common mitigation methods, and provides brief commentary on the state-of-the-practice.
LIQUEFACTION EVALUATION Workshops were held in 1996 and 1998 by the National Center for Earthquake Engineering and Research (NCEER) in an effort to develop a consensus on the evaluation of liquefaction potential. These workshops led to the publication of a summary report (Youd et al., 2001) which, for a short time, served as a consensus document on the evaluation of liquefaction triggering.
Numerous modifications and additions to the simplified procedure have been proposed in the last decade resulting in a diminished consensus on liquefaction evaluation procedures. Additionally, following the 1999 Kocaeli earthquake and the 1999 Chi-Chi earthquake, various researchers expanded the range of potentially liquefiable materials to include some fine-grained soils that would previously not have been considered in a liquefaction evaluation. Various screening procedures have been proposed (e.g., the “Chinese” Criteria summarized in Youd et al, 2001; Bray and Sancio, 2006; Boulanger and Idriss, 2006; Idriss and Boulanger, 2008). The “Chinese” Criteria has been generally disregarded as a valid design method; however a general consensus on the alternate methods has not been achieved. This paper offers no guidance on the evaluation and screening procedures, but as subsequently presented the selected evaluation procedures must be clearly identified and communicated in the project documents related to liquefaction mitigation.
FUNDAMENTAL MECHANISMS OF GROUND IMPROVEMENT FOR LIQUEFACTION MITIGATIONA survey of the available ground improvement liquefaction mitigation techniques by the National Research Council (NRC, 1985) determined that three fundamental mechanisms are usually involved: 1) densification, 2) drainage, and/or, 3) reinforcement. These three methods are discussed in the following sections:
Densification. For sands below the groundwater table, the resistance to liquefaction is largely a function of relative density (Seed and Lee,
1966). It rationally follows that a substantial number of liquefaction mitigation techniques (e.g., vibro-compaction, vibro-stone columns, dynamic compaction, compaction grouting, etc.) are intended to sufficiently densify the soil so that liquefaction will not occur, or its consequences may be controlled, during the design earthquake. When compared to other methods, densification is attractive because improvement can be verified using the properties of the improved soil (e.g. post-improvement Standard Penetration Test N-values or Cone Penetration Test tip resistances). Baez, 1995 developed design densification models that allow an estimation of approximate improvement levels when using vibro-stone columns. Design and construction considerations of densification include (but are not limited to) the following:
• Fines Content: As the fines content of a granular soil increases, the effectiveness of all densification methods will decrease. Figs 1 and 2 illustrate this trend. Additionally, whether the fines are plastic or non-plastic and/or silt-sized or clay-sized is also important. Even a small clay fraction may limit the ability of a soil to be effectively densified (Mackiewicz & Camp, 2007). Therefore, densification methods may not be able to mitigate liquefaction in silty and clayey soils. However, it has been possible, in some cases, to increase densification of silty sands and silts when wick drains are pre-installed in combination with vibro-stone columns (Luehring et al., 2001; Seed et al., 2003, ). Micaceous sands may also present a challenge to densify. This is because the mica portion
sand silt claygravel
[FIG. 1] - Gradation curves that lie to the left of the transition zone are more easily densifi able. Soil gradation
curves within the transition zone require additional engineering judgment and test programs.
DFI JOURNAL Vol. 7 No. 1 August 2013 [5]
is typically retained on the No 200 sieve and is therefore not included in the fines content. However, the flat sheet particles are not easily re-arranged into a denser configuration via vibratory and cavity expansion ground treatments.
• The densification process (via vibratory energy or undrained cavity expansion) elevates pore pressures and will temporarily destroy ageing-related bonding, cementation, micro-structure, etc. As a result, penetration testing (e.g., SPT N-value or cone tip resistance) performed soon after densification may not be representative of the true degree of improvement, and the post-improvement penetration resistances may be expected to increase over time (Mitchell and Solymar, 1984; Schmertmann, 1987; Mesri et al., 1990; Charlie et al., 1992). In practice, a minimum 7 day “rest” period is often necessary to evaluate densification effects. For many projects, it is not feasible to delay the project in order to confirm the effectiveness of the ground improvement. Consequently, the analysis of the field data may consider the effects of time on penetration resistance using published relationships (Joshi, et al., 1995; Leon et al., 2006).
• There will be variation in the degree of soil improvement between the point of application of the vibroflot, point of impact,
compaction grout, etc. Degan (1997) reports a 20% variation in CPT tip resistance over a lateral distance of 20 inches (500 mm). While it would be ideal that the design computations and post-improvement testing include consideration of the lateral variation of the improvement, it is conservative to perform the post-improvement testing at the maximum distance between adjacent application points.
Drainage. By definition, cyclic liquefaction is the state of essentially zero effective stress that results when the ratio of excess pore pressure to the initial vertical effective stress (also called the pore water pressure ratio) is essentially 1. Liquefaction can be mitigated in sands if the development of high excess pore water pressure can be prevented using drains. Seed and Booker (1977) published design charts for vertical gravel drains based on the soil properties, the liquefaction susceptibility, and the earthquake conditions. More recently, Pestana et al. (1997) developed the finite element computer program FEQDrain to assist in the design of prefabricated drains consisting of a corrugated perforated plastic pipe with a geosynthetic covering (known commercially as EQ drains). As illustrated in Fig. 3, the compressibility of sand increases dramatically once the pore pressure ratio exceeds 0.6. Therefore, the objective of the design of an earthquake drain liquefaction mitigation program is to determine the spacing such that the pore pressure ratio is maintained below 0.6 to minimize deformation.
[FIG. 2] – Relationship of compactibility to CPT friction ratio (after Massarsch 1991) A larger friction ratio is
typically indicative of a higher fi nes content.
[FIG. 3] - Relationship of compressibility to peak pore pressure ratio (Seed and Booker 1977).
0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
16.00
0.00 0.20 0.40 0.60 0.80 1.00
Peak Pore Pressure Ratio
Nor
mal
ized
Coe
ffic
ient
of
Vol
um
etri
c C
ompr
essi
bilit
y
Dr = 30%
Dr = 40%
Dr = 50%
Dr = 60%
Dr = 70%
Dr = 80%
Dr = 90%
[6] DFI JOURNAL Vol. 7 No. 1 August 2013
Design and construction considerations for the use of enhanced drainage as a liquefaction mitigation method include (but are not limited to) the following:
• Aggregate drains have been used with success in Japan. These drains are constructed using aggregate gradations that consider filter requirements for the soils in which they are being installed. Additionally, the drains are installed using low-energy methods that do not cause crushing of the aggregate or mixing of the aggregate and surrounding soil.
• Conventional stone column or aggregate pier construction in the US may not create an element that is capable of effectively functioning as a drain for purposes of liquefaction mitigation (Green, 2012). Even with the typical highly permeable aggregate that such columns or piers use, mixing/infiltration of the surrounding soils into the stone and/or crushing of the aggregate during the compaction process results in an in situ matrix with measured soil intrusions of about 20% by weight and permeability values on the order of 1E-02 cm/sec (Baez and Martin, 1995). If stone columns or aggregate piers are intended to enhance drainage, consideration must be given to the gradation and hardness of the backfill stone, the potential for mixing during the installation process, and the process used to construct the columns.
• The spacing of drains is dependent on the permeability of the soil that is to be mitigated. The spacing will become impractical for silts or sands with significant fines content. Oftentimes, soils that are most appropriate for mitigation through drainage are also appropriate for densification (high permeability sands).
• Although drains can successfully mitigate liquefaction and the associated substantial loss of soil strength, the potential for volumetric compression may remain after drain installation and consideration must be given to allowable deformations. Large shaking table test research in Japan (Iai, 1988) has demonstrated that the volume of water drained during the seismic event is approximately equal to the amount of settlement observed at the surface of the drain treated ground. This suggests that
the drains effectively facilitate, rather than prevent volume change, which is not a desired consequence for an effective liquefaction countermeasure designed to reduce seismic settlements.
• When using drainage as the sole liquefaction countermeasure (i.e., no densification or reinforcement), the designer is also cautioned to take into account the variability of the brief high seismic pulses from large earthquakes and their effect on a temporary clogging of the drain which may render it ineffective. Seed et al., 2003 refer to a drainage countermeasure as a “brittle” solution which may only be effective if it promotes the rapid pore pressure dissipation during the few critical seconds of the earthquake.
• The effectiveness of an earthquake drain installation cannot be verified through field testing. Therefore, designers must rely on the analytical design method.
Reinforcement. Liquefaction mitigation by shear reinforcement relies on the installation of stiffer elements within a soil mass to reduce the cyclic shear stress applied to liquefiable soils. Soil reinforcement options include: full soil treatment (via permeation grouting, jet grouting, or mass soil mixing), cellular or panel reinforcement (using jet grouting, soil mixing, or slurry wall systems), or individual column elements (using jet grout columns, mechanically mixed columns, stone columns, aggregate piers, grout columns, etc.). Post-earthquake observations suggest that reinforcement also reduces the earthquake-related settlement by providing an improved axial stiffness.
Baez (1995) presented a design methodology based on fundamental principles of strain compatibility between reinforcing elements and soil, and force equilibrium to calculate the reduction in cyclic shear stress on a soil mass as a function of the soil shear modulus, reinforcement shear modulus, and the amount of treatment. This methodology has been used in practice to design reinforcement-based ground improvement programs for liquefaction mitigation. However, numerous researchers have evaluated the Baez procedure and concluded that, for columnar reinforcement, it significantly overestimates the effectiveness of reinforcement in terms of shear stress reduction (Goughnour and Pestana, 1998;
DFI JOURNAL Vol. 7 No. 1 August 2013 [7]
Martin and Olgun, 2006, Olgun and Martin, 2008, Rayamajhi et al., 2012, Nguyen et al., 2012, Boulanger, 2012). More specifically, the research findings indicate that there is a lack of strain compatibility between the soil and the reinforcement element (stone column, aggregate pier, soilcrete column, etc.) and significant benefits of the stiffer element in terms of shear reinforcement are not realized. In contrast, research shows that wall panels arranged to form a cellular pattern maintain shear strain compatibility between the soil and the reinforcement and contribute significantly to shear stress reduction. In summary, if shear stress reduction is the design objective, as might be the case for a soil that is not easily densified or drained, current research and models indicate that wall panels or a cellular pattern of reinforcement can be effective, whereas discrete columns are not. Cellular and panel reinforcement geometries have been widely and successfully implemented in Japan and an increasing use of these geometries for non-densifiable soils will likely occur in the U.S.
Numerous researchers have evaluated the performance of improved sites after they have been subjected to earthquakes and sites with ground improvement have out-performed (i.e., settled less, suffered less foundation damage, etc.) similar nearby sites without ground improvement (Iai et al, 1994; Mitchell et al, 1995; Yasuda et al, 1996; Mitchell and Wentz, 1998; Mitchell et al, 2000; Hausler, 2002; Martin and Olgun, 2006). In particular, some sites with column reinforcement (e.g., stone columns, soilcrete columns, jet grout columns) out-performed sites without any improvement indicating that even if the column reinforcement does not provide a shear stress reduction benefit as initially assumed (i.e., does not prevent the onset of liquefaction), it may still be effective in limiting the consequences of liquefaction. Not all of the possible mechanisms for the improved performance are fully understood but a likely component is a reduction in vertical deformation due to the increased axial stiffness provided by the elements (Martin and Olgun, 2006). Additionally, the increase in the effective lateral stress that is produced by some improvement methods (e.g., aggregate piers, stone columns) may reduce shear strains during shaking, thereby reducing the potential or
extent of liquefaction. Confinement pressures and the engagement of discrete columns via caps and mats that connect the columns are another possible contributing factor to improved performance, as compared to free field conditions. Current analytical models have not evaluated such conditions, but experimental centrifuge tests (Adalier et al. 2003) show improved liquefaction consequence results for discrete columns subjected to building pressures and confinement. Most discrete column applications include a building or structure slab or mat atop the discrete columns.
With respect to the mechanism of reinforcement for liquefaction mitigation, a number of issues remain for consideration:
• The shear stress reduction potential for individual reinforcing elements (e.g., stone columns, aggregate piers, grout columns, soilcrete columns) appears to be very small. The stress reduction potential decreases as the diameter of the element decreases and the efficiency of the system decreases as the modulus of the element increases (Boulanger, 2012) but it does increase as the confining stress increases (Green et al 2008). Discrete column designs based on the Baez (1995) concept may be significantly unconservative.
• Panel or cellular reinforcement can effectively reduce the shear stresses within a soil mass and the recent work by Nguyen et al. (2012) provides a design methodology that is applicable for all sites. This methodology generally matches the Baez (1995) methodology when panel coverage is in excess of about 25%.
• Individual reinforcing elements may be effective in reducing vertical displacements following ground shaking. A rational analytical approach for the use of individual elements to reduce seismic settlement (while not eliminating liquefaction potential) has not been developed.
• Much like drainage for mitigation, the effectiveness of reinforcement cannot be field verified post-treatment. Engineers must rely on theoretical analysis for their design, the methods for which have not reached an industry-wide consensus.
[8] DFI JOURNAL Vol. 7 No. 1 August 2013
SPECIFICATION CONSIDERATIONSDevelopment of appropriate specifications for liquefaction mitigation is not a trivial matter and is an area of practice that needs improvement. The means and methods for ground improvement are extremely diverse and, as a result, ground improvement programs are frequently contracted using a performance or design-build specification. When compared to a detailed design-bid-build approach, these contracting methods offer many benefits to both owners and specialty geo-constructors, but the following items should be addressed when developing the specifications:
• The design earthquake criteria should be thoroughly described in the project specifications. If the specialty geo-constructor is allowed to develop the design criteria, the owner’s representative should be of sufficient sophistication to confirm that the methodology used in developing the design criteria is consistent with the state-of-practice.
• Because there is no consensus on liquefaction evaluation and screening procedures and different methods will yield different results, the acceptable evaluation and screening method(s) should be specifically defined in the project specifications. The definition should include the required procedures for evaluating the efficacy of the ground improvement. If post-improvement in situ testing is required, the interpretation and correction (e.g., corrections to N-values for energy, rod length, overburden stress, etc.) procedures should be specified.
• A “seismic” or post-earthquake settlement tolerance is frequently specified. The selection of the settlement tolerance should reflect the performance objectives (i.e., collapse prevention in accordance with the International Building Code or a more stringent serviceability requirement). With respect to collapse prevention, post-earthquake reconnaissance routinely shows that structures tolerate very large liquefaction-induced settlement (e.g., 0.1 to 1 m or 0.33 to 3.3 ft) without collapse. Examples from the 2010-2011 Christchurch Earthquakes are shown in Figs. 4 and 5. With respect to serviceability requirements, it should be recognized that deformations
associated with liquefaction can only be crudely estimated. The expectation of a guaranteed maximum settlement with little or no tolerance is unrealistic.
• As noted above, ground improvement programs that rely on densification are particularly attractive since the effectiveness of the densification can be evaluated using post-improvement in situ testing. For such programs, a post-improvement criteria (e.g., cone penetrometer tip resistance, Standard Penetration Test N-value, dilatometer horizontal stress index, etc.) is frequently specified. The criteria should reflect the values needed in accordance with the specified evaluation procedure and design hazard. Additionally, it must be recognized that some soils (e.g., silty sands, slightly
[FIG. 4] Six story structure that experienced 0.26m of liquefaction-induced differential settlement (after
Cubrinovski and McCahon, 2011)
[FIG. 5] Two story structure that experienced 0.1 to 0.25m of liquefaction-induced settlement (after
Cubrinovski and McCahon, 2011)
DFI JOURNAL Vol. 7 No. 1 August 2013 [9]
clayey sands, silts) will not be able to be densified and the post-improvement criteria must make allowances for such strata. The criteria should also reflect the time-dependency of the post-improvement test results.
• Use of Cone Penetration Test (CPT) is widely accepted as an economical way to evaluate improvement based on tip resistance. A boundary for what may be considered non-liquefiable, and therefore not in need of improvement, is typically the use of the calculated parameter, Ic (soil behavior type index). However, the designer must recognize that this index was originally defined for non-disturbed and often normally consolidated conditions. Ground treatments with vibratory energy and cavity displacement countermeasures work by remolding and reconstituting the soil structure. Therefore, pre-treatment Ic soil type definitions do not necessarily match post treatment Ic calculations (Baez, 2005). A calibration of this parameter may need to be taken into account for the proper interpretation of post treatment CPT results.
• For soils that cannot be densified and/or for ground improvement methods that cannot be evaluated via post-improvement testing, the efficacy of the program must be based on construction observation and the fundamental mechanics and empirical observations. The tools available to researchers and practitioners have advanced significantly, but continuing research has illustrated limitations on using past practices (such as the lack of strain compatibility). Liquefaction mitigation solutions should be based on sound soil mechanics, particularly when designing mitigation programs that are not field verifiable.
CONCLUSIONS AND RECOMMENDATIONS This document presented a brief overview of the three mechanisms - densification, drainage and reinforcement - currently used for liquefaction mitigation within the geotechnical construction industry. The summaries provided describe the basic mechanics and potential concerns related to each method. Significant concerns include the following:
• When used in appropriate soils, densification allows for improvement verification, unlike drainage and reinforcement. However, densification is only applicable in cohesionless soils with less than 20% fines (and a significantly lower clay content). In cases where wick drains have been pre-installed, densification may be possible for soils with fines content up to 65% and small clay fractions. Note that successful cases using wick drains and stone columns generally require area replacement ratios in excess of 20-30% as well as wick drains that are close to the densification point. Engineers are encouraged to consider the soil characteristics and drainage properties when writing specifications that require post-treatment verification of densification.
• Because ground improvement methods that apply drainage and/or reinforcement are not amenable to post-treatment verification, the analysis used to design these types of ground improvement methods must be based on fundamental mechanical principles and empirical observations.
• Although post-earthquake observations indicate that reinforcement can effectively mitigate the effects of liquefaction, consensus has not been reached for developing a state-of-the-practice design methodology for liquefaction mitigation using soil reinforcement. Furthermore, recent research indicates that columnar reinforcement is not as effective in reducing the soil shear stress as previously believed. This is resulting in an inconsistent and potentially unconservative range of designs for this method. Engineers and agencies must be conscious of this inconsistency when evaluating reinforcement proposals and designs and continue to rely on fundamental mechanics and the most current research findings.
The geotechnical engineering community will be well served by a continued focus on the mechanics, effectiveness, and limitations of all liquefaction mitigation methods. With each new earthquake, the engineering knowledge base expands, and the engineering practice will evolve.
[10] DFI JOURNAL Vol. 7 No. 1 August 2013
REFERENCES1. Adalier, K., Elgamal, A., Meneses, J.
and Baez, J.I. (2003) “Stone columns as liquefaction countermeasure in non-plastic silty soils”, Soil Dynamics and Earthquake Engineering 23(7), pp. 571-584.
2. Andrus, R. D., and Stokoe, K. H., II (2000). ‘‘Liquefaction resistance of soils from shear-wave velocity”, Journal of Geotechical and Geoenvironmental Engineering, ASCE, 126(11), pp. 1015–1025
3. Arulmoli, K., Arulanandan, K., and Seed, H.B. (1985) “New Method for Evaluating Liquefaction Potential”, Journal of Geotechnical Engineering, 111(1), pp. 95-114.
4. Baez, J.I. (1995) “A design model for the reduction of soil liquefaction by vibro-stone columns”, Ph.D. thesis, University of Southern California, Los Angeles, CA.
5. Baez, J.I., and Martin, G.R., (1995) “Permeability and Shear Wave Velocity of Vibro-Replacement Stone Columns”, Soil Improvement for Earthquake Hazard Mitigation, Edited by Roman D. Hryciw, ASCE. Geotechnical Special Publication No. 49. October, 1995.
6. Baez, J.I. (2005) “Liquefaction Mitigation of Fine Grained Soils”, 2005 US-Japan Workshop on Ground Improvement: New Applications and Challenging Soils for Ground Improvement Technologies, Kyoto, Japan, September 8-10, 2005.
7. Boulanger, R.W. (2012) “Shear reinforcement effects for liquefaction mitigation”, DFI Liquefaction Forum: Consequences and Mitigation, St. Louis, MO.
8. Boulanger, R.W. and Idriss, I.M. (2006). “Liquefaction susceptibility criteria for silts and clays”, Journal of Geotechnical and Geoenvironmental Enineering, 132(11), pp. 1413-1426.
9. Bray, J.D. and Sancio, R.B. (2006). “Assessment of the liquefaction susceptibility of fine-grained soils”, Journal of Geotechnical and Geoenvironmental Engineering, 132(9), pp. 1165-1177.
10. CGS, California Geological Survey (2008) “Guidelines for Evaluating and Mitigating Seismic Hazards in California”, Special Publication 117, Public Information Offices of the California Geological Survey.
11. Charlie, W.A., Rwebyogo, M.F.J. and Doehring, D.O. (1992). “Time-dependent cone penetration resistance due to blasting”, Journal of Geotechnical Engineering,118(8), pp. 1200-1215.
12. Cubrinovski, M. and McCahon, I. (2011). “Foundations on Deep Alluvial Soils”, Technical Report Prepared for the Canterbury Earthquakes Royal Commission, University of Canterbury, Christchurch, NZ
13. Degan, W.S. (1997) Vibroflotation Ground Improvement, Vibroflotation AG, Altendorf, 248 p.
14. Goughnour, R.R., and Pestana, J.M., (1998). “Mechanical behavior of stone columns under seismic loading”, Proceedings 2nd International Conference on Ground Improvement Techniques, Singapore.
15. Green, R.A. (2012) “Liquefaction risk mitigation by excess pore pressure dissipation through compacted gravel piles”, DFI Liquefaction Forum: Consequences and Mitigation, St. Louis, MO.
16. Green, R.A., Olgun, C.G., and Wissmann, K.J., (2008). “Shear stress redistribution as a mechanism to mitigate the risk of liquefaction”, Proceedings Geotechnical Earthquake Engineering and Soil Dynamics IV, ASCE GSP 181, Sacramento, CA.
17. Hausler, E.A., (2002). “Influence of ground improvement on settlement and liquefaction: a study based on field case history evidence and dynamic geotechnical centrifuge tests”, Ph.D. Dissertation, Department of Civil and Environmental Engineering, University of California, Berkeley.
18. Iai, S. (1988) “Large Scale Model Tests and Analyses of Gravel Drains”, Report of the Port and Harbour Research Institute Japan, Vol 127, No. 3.
19. Iai, S., Matsunaga, Y., Morita, T., Miyata, M., Sakurai, H., Oishi, H., Ogura, H., Ando, Y., Tanaka, Y., and Kato, M. (1994). “Effects of remedial measures against liquefaction at 1993 Kishiro-Oki Earthquake”, Proceedings 5th U.S.-Japan Workshop on Earthquake Resistant Design of Lifeline Facilities and Countermeasures Against Soil Liquefaction, NCEER-94-0026, Nov, pp. 135-152.
DFI JOURNAL Vol. 7 No. 1 August 2013 [11]
20. Idriss, I.M. and Boulanger, R.W. (2008) Soil Liquefaction During Earthquakes, Monograph MNO-12, Earthquake Engineering Research Institute.
21. Joshi, R.C., Achari, G., Shenbaga, R.K., and Wijeweera, H. (1995). “Effect of aging on the penetration resistance of sands”, Canandian Geotechnical Journal, Vol 32, pp. 767-782.
22. Kayen, R.E. and Mitchell, J.K. (1997) “Assessment of Liquefaction Potential During Earthquakes by Arias Intensity”, Journal of Geotechnical and Geoenvironmental Engineering, 123(12), pp. 1162-1174.
23. Leon, E., Gassman, S.L., and Talwani, P. (2006). “Accounting for soil aging when assessing liquefaction potential”, Journal of Geotechnical and Geoenvironmental Engineering, 132(3), pp. 363-377.
24. Luehring, R., Snorteland N., Stevens, M., and Mejia, L. (2001) “Liquefaction Mitigation of a Silty Dam Foundation Using Vibro-Stone Columns and Drainage Wicks: A case History at Salmon Lake Dam”, 21st USSD Annual Meeting and Lecture Proceedings, Denver, Colorado, July 30 – August 03, 2001.
25. Mackiewicz, S. M., and Camp, W. M. (2007). “Ground Modification: How Much Improvement?”, Proceedings Geo-Denver, ASCE GSP 172, Denver, CO.
26. Martin G.R., and Lew M. (Editors) (1999). “Recommended Procedures for Implementation of DMG Special Publication 117 – Guidelines for Analyzing and Mitigating Liquefaction in California”, Southern California Earthquake Center, University of Southern California, March.
27. Martin, J.R., II, and Olgun, C.G. (2006). “Liquefaction mitigation using jet-grout columns – 1999 Kocaeli earthquake case history”, Ground Modification and Seismic Mitigation, ASCE GSP 152, pp. 349-358.
28. Massarsch, K.R., (1991). “Deep Soil Compaction Using Vibratory Probes”, ASTM Symposium on Design, Construction, and Testing of Deep Foundation Improvement: Stone Columns and Related Techniques, Robert C. Bachus, Ed. ASTM Special Technical Publication, STP 1089, Philadelphia, pp. 297-319.
29. Mesri, G., Feng, T.W. and Benak, J.M. (1990). “Postdensification penetration resistance in clean sands”, Journal of Geotechnical Engineering, 116(7), pp. 1095-1115.
30. Mitchell, J.K., Baxter, C.D.P., and Munson, T.C. (1995). “Performance of improved ground during earthquakes”, Soil Improvement for Earthquake Hazard Mitigation, ASCE GSP No. 49, pp. 1-36.
31. Mitchell, J. K., Martin, J. R., Olgun, C. G., Emrem, C., Durgunoglu, H. T., Cetin, K. O., and Karadayilar, T. (2000). "Performance of Improved Ground and Earth Structures”, Earthquake Spectra, 16(Supplement "A"), pp. 191-225
32. Mitchell, J.K. and Solymar, Z.V. (1984). “Time-dependent strength gain in freshly deposited or densified sand”, Journal of Geotechnical Engineering, 110(11), pp. 1559-1576.
33. Mitchell, J.K., and Wentz, F.J., Jr., (1998). “Improved-ground performance during the earthquake”, The Loma Prieta, California, earthquake of October 17, 1989 - Liquefaction, Holzer, T.L., Ed., U. S. Geological Survey Professional Paper 1551-B, pp. B241-B272.
34. Moseley, M.P. and Kirsch, K. (2004) “Ground Improvement, 2nd Edition”, Spon Press, New York, NY.
35. National Research Council (1985) “Liquefaction of Soils During Earthquakes”, National Research Council, Committee on Earthquake Engineering, Washington, District of Columbia.
36. Nguyen, T.V., Rayamajhi, D., Boulanger, R.W., Ashford, S.A., Lu, J., Elgamal, A., and Shao, L. (2012) “Effect of DSM grids on shear stress distribution in liquefiable soil”, Proceedings GeoCongress 2012, State of the Art and Practice in Geotechnical Engineering, ASCE GSP 255, Oakland, CA, pp. 1948-1957.
37. Olgun, C.G. and Martin, J.R., II, (2008) “Numerical modeling of the seismic response of columnar reinforced ground”, Proceedings Geotechnical Earthquake Engineering and Soil Dynamics IV, ASCE GSP 181, Sacramento, CA.
[12] DFI JOURNAL Vol. 7 No. 1 August 2013
38. Pestana, J.M., Hunt, C.E. and Goughnour, R.R. (1997) “FEQDrain: A Finite Element Computer Program for the Analysis of the Earthquake Generation and Dissipation of Pore Water Pressure in Layered Sand Deposits with Vertical Drains”, Report No. UCB/EERC-97/15, University of California, Berkeley, 88 p.
39. Poulos, S.J., Castro, G., and France, J.W. (1985) “Liquefaction Evaluation Procedure”, Journal of Geotechnical Engineering, 111(6), pp. 772-792.
40. Rayamajhi, D., Nguyen, T.V., Ashford, S.A., Boulanger, R.W., Lu, J., Elgamal, A., and Shao, L. (2012) “Effect of discrete columns on shear stress distribution in liquefiable soil”, Proceedings GeoCongress 2012, State of the Art and Practice in Geotechnical Engineering, ASCE GSP 255, Oakland, CA, pp. 1918-1927.
41. Seed, H.B. and Lee, K.L. (1966) “Liquefaction of Saturated Sands During Cyclic Loading”, Journal Soil Mechanics and Foundation Division, ASCE, Vol. 92, No. SM6, pp. 105-134.
42. Seed, H.B. and Idriss, I.M. (1967) “Analysis of Soil Liquefaction: Niigata Earthquake”, Journal Soil Mechanics and Foundation Division, ASCE, Vol. 93, No. SM3, pp. 83-108.
43. Seed, H.B. and Idriss, I.M. (1971) “Simplified Procedure for Evaluating Soil Liquefaction Potential”, Journal Soil Mechanics and Foundation Division, ASCE, Vol. 97, No. SM9.
44. Seed, H.B. and Booker, J.R. (1977) “Stabilization of Potentially Liquefiable Sand Deposits Using Gravel Drains”, Journal Soil Mechanics and Foundation Division, ASCE, Vol. 103, No. GT7, pp. 757-768.
45. Seed, R.B., Cetin, K.O., Moss, R.E.S., Kammerer, A.M., Wu, J., Pestana, J.M., Riemer, M.F., Sancio, R.B., Bray, J.D., Kayen, R.E., and Faris, A. (2003) “Recent Advances in Soil Liquefaction Engineering: A Unified and Consistent Framework”, 26th Annual ASCE Los Angeles Geotechnical Spring Seminar, Keynote Presentation, H.M.S Queen Mary, Long Beach, California.
46. Schmertmann, J.H. (1987) Discussion on “Time-dependent strength gain in freshly deposited or densified sand by J.K. Mitchell and Z.V. Solymar", Journal of Geotechnical Engineering, pp. 117(9), pp. 171-176.
47. Whitman, R.V. (1971) “Resistance of Soil to Liquefaction and Settlement”, Japanese Society of Soil Mechanics and Foundation Engineering. Vol. 11, No. 4. December.
48. Yasuda, S., Ishihara, K., Harada, K. and Shinkawa, N. (1996). “Effect of improvement on ground subsidence due to liquefaction”, Soils and Foundations, JSSMFE, Special Issue, January, pp. 99-107.
49. Youd, T.L., Idriss, I.M., Andrus, R.D., Arango, I., Castro, G., Christian, J.T., Dobry, R., Finn, W.D.L., Harder, L.F., Jr., Hynes, M.E., Ishihara, K., Koester, J.P., Liao, S.S.C., Marcuson, W.F., III, Martin, G.R., Mitchell, J.K., Moriwaki, Y., Power, M.S., Robertson, P.K., Seed, R.B., Stokoe, K.H., II. (2001) “Liquefaction resistance of soils: Summary report from the 1996 NCEER and 1998 NCEER/NSF workshop on evaluation of liquefaction resistance of soils”, Journal of Geotechnical and Geoenvironmental Engineering, 127(10), pp. 817-833.
DFI JOURNAL Vol. 7 No. 1 August 2013 [13]
Liquefaction Mitigation Synthesis ReportPrepared for: The Ground Improvement Committee of the Deep Foundations Institute
By: Timothy C. Siegel, P.E., G.E., D.GE, Dan Brown and Associates PC, Knoxville; (865) 357-1715;[email protected]
PROLOGUE
This report presents the results of a synthesis on the design and analysis of ground improvement for
liquefaction mitigation. The synthesis included an industry survey concerning the practice of ground
improvement for liquefaction mitigation. Participation in the survey was solicited by advertisements
in several trade magazines and by e-mail for the DFI membership. The survey participants numbered
150. Their professional roles include consulting engineers, specialty contractors, design engineers,
government engineers, and academicians. They represent a variety of geographical areas including
North/Central/South America, United Kingdom, Middle East, Caribbean, Hawaii, Japan, India, Egypt,
France, Australia and New Zealand. Upon completion of the survey, several professionals in the fi eld
of liquefaction and ground improvement were interviewed for them to elaborate on the survey results.
The interviews are included in the Appendix of this report. Financial support for the project was
provided by DFI and Dan Brown and Associates PC.
The concept of the liquefaction mitigation synthesis was developed by DFI’s Ground Improvement
Committee in recognition that:
(a) The results of recent research and post-earthquake reconnaissance have challenged previously long-
held beliefs about liquefaction and associated mitigation techniques, and;
(b) The DFI membership and the engineering/construction industry are interested to know if and how
engineers and designers are subsequently adjusting their practice in consideration of recent research
and post-earthquake reconnaissance.
For more detailed information on recent research and post-earthquake reconnaissance, presentations
are available from the State-of-the-Art Forum: Liquefaction Consequences and Mitigation that
was held in St. Louis in 2012. A commentary of the state-of-practice in ground improvement for
liquefaction mitigation (prepared by DFI’s Ground Improvement Committee) is included in this
issue of the DFI Journal.
The author would like to thank the participants of the survey and especially Mr. Mike Jeffries, Dr. Les
Youd, and Dr. Ikuo Towhata for their willingness to share their expertise in interviews. The author also
acknowledges Mary Ellen Bruce of DFI, Billy Camp of S&ME, Inc., and Marty Taube of DGI Menard
(and Chair of DFI’s Ground Improvement Committee) for their signifi cant contributions.
INTRODUCTIONCyclic liquefaction is a phenomenon where high excess pore-water pressure develops in saturated soil as a result of cyclic loading (Seed and Lee, 1965 and 1966). When the ratio of pore-water pressure to the total vertical stress is essentially 1 (i.e., the state of zero effective stress), the soil is considered “liquefied” and loses a large portion of its shear resistance. At lower relative densities (less than approximately 70%), soils may contract resulting in large ground settlements. Soils with higher relative densities (greater than approximately 70%) are dilative, preventing a substantial loss of
shear strength and large ground settlements. Liquefaction and its impact on engineering structures came to the engineering forefront in the 1960s due, in part, to the widespread liquefaction-induced damage (primarily settlement, tilting and lateral displacement of buildings) that occurred as a result of the 1964 Niigata (Japan) and 1964 Good Friday Alaska earthquakes (Grantz, et al., 1964; Japan National Committee on Earthquake Engineering, 1965; Seed and Idriss, 1967). Liquefaction also caused severe damage in the 1959 Jaltipan (Mexico) earthquake (Marsal, 1961) and the 1960 Chilean earthquake (Duke
[14] DFI JOURNAL Vol. 7 No. 1 August 2013
and Leeds, 1963). In 1971, the “simplified procedure” (Seed and Idriss, 1971; Whitman, 1971) for anticipating liquefaction was developed based on the soil conditions and the design earthquake. Seed and Booker (1977) proposed the installation of columnar gravel drains in soil of high liquefaction potential to prevent the development of excessively high pore-water pressure. In this way, a routine framework was established. First, the liquefaction potential is evaluated using a rational method involving either field or laboratory testing. Second, if a potential for liquefaction is present and the effects of liquefaction are determined to present unacceptable risk to the performance of the structure, then ground improvement can be designed to mitigate the risk.
Over the past 40 years, the evaluation of liquefaction and the methods for mitigation design has continued to rapidly evolve. The evolution process is not without difficulties. Due, in part, to the use of the research community as the primary technical source serving practicing engineers and contractors, new insights are continuously being delivered. Practicing engineers and contractors are challenged to implement the conclusions of the most recent findings in a coherent and rational manner. However, considerable controversy and an absence of consensus exist regarding several aspects related to liquefaction including the subject of the efficacy of various mitigation techniques. The resulting negative impacts may include over-conservatism (and increased cost) by the designers and consultants, conflicts within the design team, and confusion among owners and their representatives.
The dichotomy between research and practice on the subject liquefaction is not new. In his technical note from 1979, Ralph Peck advised that engineers and those that depend on engineers would be well served to distinguish between research (or science) and practice:
In short, engineering science and engineering practice are not identical. Advances in science may temporarily appear to run counter to good practice. When this occurs, the implications should be evaluated carefully, but it should by no means be assumed that the latest scientific advancement is always the right direction. Science has its own ways of making progress, as evidence accumulates it corrects
its errors and improves its predictions. In the end, it is certain to improve practice as well. But science may temporarily mislead the unwary, and it should not intimidate either the experienced engineer or the overburdened regulatory agency.
Another way of saying this is that engineering practice should be careful not to assume that the most recent opinion on the subject of liquefaction must be correct to the extent that it automatically invalidates those that precede it.
LIQUEFACTION MITIGATION SYNTHESISAs primarily an effort to support the Deep Foundations Institute (DFI) membership, this synthesis attempts to help define the current state-of-practice in liquefaction mitigation by surveying practicing engineers and specialty contractors involved in the selection and implementation of ground improvement techniques for liquefaction. The survey was divided into four sections: (1) general practice, (2) liquefaction analysis, (3) mitigation design, and (4) verification. This report summarizes the results of the survey and presents the conclusions that may be made from the survey results.
General Practice
These questions are intended to provide a profile of the participant. Taken as a summary, this information will characterize the population involved in the survey. The questions and the answers (in terms of percentages) are:
1. What best describes your position in the liquefaction mitigation industry?
• Consulting Engineer (58%)
• Design Engineer (16%)
• Specialty Contractor (14%)
• Other (12%)
Commentary: The ground improvement industry, especially for liquefaction mitigation, is different from most other geotechnical/foundation designs. The design engineer for liquefaction mitigation is typically employed directly by the specialty contractor performing the installation. Because the owner and the owner’s consultants often do not have the technical expertise to prepare or review the liquefaction mitigation design, they provide
DFI JOURNAL Vol. 7 No. 1 August 2013 [15]
the performance requirements (which can be very stringent). Such an environment has its problems ranging from performance requirements that are unrealistic to significant differences between competing designs.
2. What area of the US are most of your
projects involving liquefaction
(check all that apply)?
• California (29%)
• Western US (ID, UT, NE, MT, WY) (6%)
• Pacifi c Northwest (OR, WA) (19%)
• Midwest (New Madrid) (10%)
• Southeast (Charleston) (10%)
• Other (26%)
Commentary: Earthquake and liquefaction
concerns have been an essential part of
engineering design in California and other parts
of the Western United States since the 1960s.
Because of the adoption of the International
Building Code and the associated increase in
seismic demand since the mid to late 1990s,
liquefaction analysis and mitigation have become
part of engineering design in parts of the eastern
United States.
3. How many projects involving liquefaction
analysis, mitigation design and/or
construction do you participate in over a one
year period?
• less than 5 (39%)
• between 5 and 10 (38%)
• between 10 and 40 (18%)
• over 40 (5%)
4. How standardized do you believe the state-
of-practice in liquefaction mitigation using
ground improvement techniques is?
• very consistent and uniform (3%)
• somewhat consistent and uniform (33%)
• somewhat non-uniform (46%)
• highly non-uniform (18%)
Commentary: As recently as the last few years,
research and post-earthquake reconnaissance
have provided results that contradict previously
held beliefs on liquefaction and the effectiveness
of some mitigation efforts. This is one reason
why it has been particularly diffi cult to establish
a more uniform and standardized state-of-
practice in ground improvement for liquefaction
mitigation and that it is not unexpected that over
half of the participants believe that state-of-
practice is “somewhat” or “highly non-uniform”.
5. How signifi cant have recent considerations
of liquefaction of fi ne-grained soil (i.e., >
30% passing No.200 sieve) been to your
projects?
• very signifi cant (24%)
• signifi cant (39.5%)
• marginally signifi cant (24.5%)
• not signifi cant (12%)
Commentary: Results of recent research
(Boulanger and Idriss, 2006; Bray and Sancio,
2006) and earthquake reconnaissance (Martin
and Olgun, 2008) support that fi ne-grained soils
can be susceptible to liquefaction. Engineering
practice has, in many areas, incorporated fi ne-
grained soils into the triggering analysis and
design of mitigation by adapting the procedures
developed for sands.
Liquefaction Analysis
These questions relate to liquefaction analysis in engineering practice. The questions and the answers (in terms of percentages) are:
Test 1 2 3 4 5
Standard penetration test (SPT) 58% 27% 8% 3% 4%
Cone penetration test (CPT) 32% 43% 16% 5% 3%
Shear-wave velocity test 3% 15% 43% 30% 9%
Cyclic lab test 1% 5% 11% 35% 49%
Liquefaction maps 6% 10% 22% 27% 35%
[16] DFI JOURNAL Vol. 7 No. 1 August 2013
6. How often are the following techniques used for site characterization associated with liquefaction analysis on your projects? 1 is most often…..5 is least often.
Commentary: The standard penetration test
(SPT) (Seed et al., 1983 and 1985) can involve
signifi cant error due to the variation in energy
delivered by the hammer during the test. In this
regard, the more controlled in situ tests – cone
penetration test (Robertson and Wride, 1998)
and shear-wave velocity test (Andrus and Stokoe,
2000) – provide more repeatable and reliable
results. However, there can be limitations
with any test. For example, Dr. Brady Cox of
the University of Texas at Austin showed that
calcareous sands experienced liquefaction during
the 2010 Haiti earthquake even though the
shear-wave velocity profi le would have indicated
otherwise (DFI Presentation, 2012).
7. What presumptive analytical maximum
depth do you consider in your liquefaction
analysis?
• 30 ft (9 m) or less (8%)
• between 30 ft and 40 ft (9 m and 12 m)
(4%)
• between 40 ft and 50 ft (12 m and 15 m)
(25%)
• between 50 ft and 75 ft (15 m and 23 m)
(25%)
• No presumptive analytical maximum
depth (38%)
Commentary: The recently published FHWA
reference manual entitled LRFD Seismic
Analysis and Design of Transportation
Geotechnical Features and Structural
Foundations (2011) recommends that
liquefaction be evaluated over the greatest of
the following depths: (a) at least 20 ft (6 m)
below the lowest expected foundation level for
deep foundations, or (b) 80 ft (24 m) below
the existing ground surface or lowest proposed
fi nished grade. It should be noted that the
geologic and hydrogeologic setting of the site
should also be part of the basis for determining
the required depth of analysis.
8. What water level do you use in the
liquefaction analysis for level sites?
• water level observed during fi eld
exploration (43%)
• an assumed elevated water level for
earthquake (48%)
• ground surface (9%)
Commentary: Selection of the water level
determines the minimum depth of potential
liquefaction and an emphasis on identifi cation
of the water level during site exploration is
well placed. A signifi cant amount of published
research exists (e.g., Okamura and Soga, 2006;
Hossain et al, 2013) that supports the conclusion
that partially saturated soils (even those soils near
saturation) have a signifi cantly greater resistance
to liquefaction than fully saturated soil. It is
understood that the water table can fl uctuate, but
trapped air is typically present for short term high
water table events. It may be overly conservative
to select a high water table for liquefaction
analysis, especially where the high water table is
a temporary condition.
9. What water level do you use in the
liquefaction analysis for slopes?
• water level observed during fi eld
exploration (40%)
• an assumed elevated water level for
earthquake (53%)
• ground surface (7%)
10. What is the minimum thickness of soil layer
that you consider to be signifi cant with
respect to liquefaction potential?
• no minimum thickness (31%)
• 6 to 12 inches (150 to 300 mm) (20%)
• 12 to 24 inches (300 to 600 mm) (23%)
• greater than 24 inches (600 mm) (26%)
Commentary: Whether or not to exclude thin
zones that may be categorized as liquefi able
can have a dramatic effect on code-related
design decisions. For example, the 2012
International Building Code requirements for steel
reinforcement for cast-in-place deep foundations
can be controlled by the location of “strata that
are liquefi able”. To illustrate an extreme but not
unrealistic scenario, a long reinforcing cage length
could be interpreted to be a code requirement based
on a few data point(s) within a very detailed CPT
sounding. Judgment should be applied as there are
typically other important considerations, such as
constructability and quality.
DFI JOURNAL Vol. 7 No. 1 August 2013 [17]
11. What screening criteria do you primarily
use to differentiate between “sand-like” and
“clay-like” cyclic behavior of soils?
• Chinese criteria (9%)
• Idriss and Boulanger (69%)
• Bray and Sancio (13%)
• Other (9%)
Commentary: The Chinese criteria are no longer
considered to provide a suitable indication of
“clay-like” cyclic behavior.
12. What reference in published literature do you
primarily use for estimating liquefaction-
induced settlement for sands?
• Tokimatsu and Seed, 1987 (45%)
• Ishihara and Yoshimine, 1990 (30%)
• Zhang, Robertson and Brachman,
2002 (16%)
• Other (9%)
Commentary: In his 2012 H. Bolton Seed
lecture in Oakland, California, Dr. Geoffrey
Martin presented laboratory test results
supporting that sands may experience
different degrees of liquefaction-induced
compression depending on their gradation,
shape, etc. Shamoto et al. (1996) showed that
the liquefaction-induced compression can be
uniquely related to the relative compression
defi ned by Δe/(ei – e
min).
In his 2013 Ralph B. Peck lecture in San Diego,
California, Dr. Jonathan Bray presented the
results of post-earthquake reconnaissance and
concluded that the procedures described above
are not applicable for building settlements.
While the procedures may be applicable to
free-fi eld conditions, they do not represent
the conditions within the zone of infl uence of
foundations. In general, these procedures are
expected to under-predict building settlement,
particularly for thinner liquefi able strata.
13. Do you estimate liquefaction-induced
settlement of liquefi able fi ne-grained soils
using the published charts for sands?
• Yes (45%)
• No (55%)
Commentary: Considering the recent
developments that show that fi ne-grained
soils are liquefi able, the absence of research
concerning liquefaction-induced compression
for these soils is understandable. Dr. Ed
Kavazanjian at the Arizona State University
stated that limited research suggests that
published literature for estimating liquefaction-
induced settlement for sands provides reasonable
results for non-plastic silts (DFI Seminar, 2012).
14. What approach do you primarily use for
estimating lateral spread in liquefi able sand?
• Empirical correlations (53%)
• Laboratory-based methods (13%)
• Newmark sliding block analysis (15%)
• Numerical modeling/analyses (10%)
• Other (9%)
15. What approach do you primarily use for
estimating lateral spread in liquefi able fi ne-
grained soil?
• Empirical correlations (47%)
• Laboratory-based methods (13%)
• Newmark sliding block analysis (18%)
• Numerical modeling/analyses (11%)
• Other (11%)
16. How much confi dence do you put in the
calculated lateral spread displacement?
• 0 to 10% (12%)
• 10 to 50% (64%)
• 50 to 90% (21%)
• greater than 90% (3%)
Commentary: Dr. Scott M. Olson of the
University of Illinois presented the results of
his research indicating that actual lateral spread
displacements are within one-half to 2 times
the predictions made based on the accepted
estimation approaches (DFI Seminar, 2012)
Mitigation Design
These questions relate to liquefaction mitigation
design in engineering practice. The questions
and the answers (in terms of percentages) are:
[18] DFI JOURNAL Vol. 7 No. 1 August 2013
17. How do you rank the following engineering
tools for liquefaction mitigation designs?
Engineering Tool
Very important
Less important
Least Important
theory/analysis/modeling
49% 33% 19%
local precedence
25% 34% 41%
published reconnaissance of earthquake damage
26% 33% 40%
Commentary: While it is recognized that analysis should be validated by field performance, this is problematic in the practice of earthquake design where the opportunities for first-hand observations are rare. The implementation of liquefaction mitigation techniques solely on precedence does not explicitly consider the variation in soil and seismic conditions; however, numerical modeling without calibration and validation can provide misleading results.
18. How often do you use the following fundamental approaches to mitigate liquefaction on your projects?
Fundamental Approach
Most often
Less oftenLeast often
densification 51% 35% 14%
reinforcement 40% 46% 14%
drainage 9% 19% 72%
19. How often do you use the following densification methods to mitigate liquefaction on your projects?
Densification method
Most often
Less often
Least often
vibrocompaction 66% 27% 8%
dynamic compaction
18% 41% 41%
compaction grouting 16% 32% 51%
20. What technical resources (literature,
software, etc.) do you primarily use in
designing against liquefaction using
densifi cation?
Commentary: There was a wide variety
of responses to this question. Specialty
contractor’s typically responded that their design
approaches were proprietary. For the remaining
respondents, the results to this question may be
broadly categorized as follows:
• The criteria are established by the consultant
but the design, implementation and
verifi cation are made the responsibilities of
the specialty contractor;
• Spreadsheets and commercial software, and;
• Numerical models (e.g., Plaxis and FLAC)
21. How often do you use the following
reinforcement methods to mitigate
liquefaction on your projects?
Reinforcement method
Most often
Often Less often
Least often
vibro-stone columns
55% 28% 11% 6.5%
rammed aggregate piers
19% 28% 17% 35%
grout columns 11% 20% 42% 26.5%
deep soil mixing cells
15% 24% 30% 32%
22. What technical resources (literature,
software, etc.) do you primarily use in
designing against liquefaction using
reinforcement?
Commentary: The results to this question may
be broadly categorized as follows:
• No analysis is typically performed but
rely on precedence and judgment with the
recognition that reinforcement may not fully
mitigate the liquefaction;
• Use of methodology proposed by Baez and
Martin (1993)
• Use of numerical models (e.g., Plaxis and
FLAC)
23. How often do you use the following drainage
methods to mitigate liquefaction on your
projects?
Drainagemethod
Most often
Less often
Least often
EQ drains 22% 20% 57%
gravel drains 41% 44% 15%
pre-fabricated vertical drains
37% 36% 28%
DFI JOURNAL Vol. 7 No. 1 August 2013 [19]
Commentary: Vertical gravel drains were
described by Seed and Booker (1972). However,
they are not widely used in the United States.
One reason is a concern about their effectiveness
and reliability. It is recognized that vertical
gravel drains need to reliably provide a high
ratio of permeability between the drain material
and the adjacent soil to prevent the buildup of
high excess pore water pressure. Dr. Russell
Green with Virginia Tech (DFI Seminar, 2012)
presented research results showing that high
degree of control during installation is required
to maintain an effective gradation in order to
achieve the target permeability.
From the comments provided by participants, it
may be concluded that the use of drains is rarely
relied upon as the primary or sole mechanism for
mitigating liquefaction in the U.S. The use of
EQ drains is focused on parts of the US, namely
Charleston, SC.
24. What technical resources (literature,
software, etc.) do you primarily use in
designing against liquefaction using
drainage?
Commentary: FEQDrain (Pestana et al., 1997)
was recognized as a technical resource for the
design of EQ drains. Gravel drains, which have
been more popular in Japan (Towhata, 2008),
can be designed by using the charts presented
by Onoue (1988).
25. What is the typical liquefaction-induced
settlement tolerance or design criteria used
on your foundation projects?
• No liquefaction as determined by a
required post-improvement SPT or
CPT resistances (23%)
• 1 inch (25 mm) (27%)
• 3 inches (76 mm) (21%)
• Greater than 3 inches (76 mm) (6%)
• No maximum settlement so long as there
is an adequate factor-of-safety against
bearing capacity failure (23%)
Commentary: Participants commented that the
type of structure and that whether the design
is to be determined based on life safety or
serviceability were important considerations
regarding selection of the tolerable settlement.
Assuming that most foundations can tolerate
about one inch of settlement with only cosmetic
damage, the results indicate that about half of
the participants of the participants typically
design for serviceability. Considering that
we often use the one inch as the tolerable
foundation settlement for non-seismic conditions
as well, it is very conservative to use either one
inch of liquefaction-induced settlement or no
liquefaction even for serviceability.
26. Is the typical liquefaction-induced settlement
tolerance or design criteria used on
your foundation projects reasonable and
achievable?
• Yes (82%)
• No (18%)
27. What is the typical lateral spread tolerance
used on your foundation projects?
• less than 1 foot (0.3 m) (55%)
• 1 to 3 ft (0.3 to 0.9 m) (35%)
• greater than 3 ft (0.9 m) (10%)
28. Is the typical lateral spread tolerance or
design criteria used on your foundation
projects reasonable and achievable?
• Yes (87%)
• No (13%)
29. What primary reference do you use in
estimating residual strength of liquefi ed
soils?
• Seed and Harder, 1990 (23%)
• Olson and Stark, 2002 (17%)
• Idriss and Boulanger, 2008 (50%)
• Other (10%)
30. Would you be in favor of performance-
based design where the tolerable ground
movements were more closely related to the
design of the structure?
• Yes (96%)
• No (4%)
[20] DFI JOURNAL Vol. 7 No. 1 August 2013
Verification
These questions relate to verifi cation of
liquefaction mitigation efforts in engineering
practice. The questions and the answers (in
terms of percentages) are:
31. For mitigation dependent on densifi cation,
what approximate percentage of your
projects includes post-improvement
verifi cation testing?
Commentary: The responses ranged from 0
to 100%. Approximately ½ of the participants
responded that 100% of their projects included
post-improvement verifi cation testing and the
majority of the remaining participants responded
with values that were between 25% and 50%.
32. How often do you use the following
techniques for post-improvement verifi cation
testing?
Verification test technique
Most often
Less often
Least often
Standard Penetration Test
40% 42% 18%
Cone Penetration Test 55% 38% 6%
Shear-Wave Velocity Test 5% 20% 76%
33. When evaluating the densifi cation by the
CPT, do you use the fi nes content estimated
from the pre-improvement or
post-improvement?
• Pre-Improvement (61%)
• Post-Improvement (39%)
Commentary: The fi nes content interpreted
from CPT data can change between the pre-
improvement testing and post-improvement
testing. This emphasizes that a good practice
is to validate CPT data with the fi nes content
determined from laboratory gradation testing
performed on samples collected in the fi eld.
34. For sites improved by densifi cation, does
your post-improvement liquefaction analysis
consider ageing effects?
• Yes (29%)
• No (71%)
Commentary: Research supports that dynamic
compaction, blasting and vibro-compaction can
temporarily destroy inter-particle structure and
bonds associated with aging. Therefore, the
cone tip resistance is expected to increase with
time after improvement using these techniques
(Mitchell and Solymar, 1984; Schmertmann,
1986; Mesri et al., 1990; Charlie et al., 1992).
Other densifi cation methods such as compaction
grouting, displacement piles, or compaction piles
may also have the same effect although it has
not been documented. Lunne et al. (1997) states
the recommended procedure is to perform fi eld
trials at the start of the project by performing
CPT at different time intervals after compaction
to evaluate the signifi cance of any time effect.
There are fi nancial drawbacks to such fi eld trials
including extending the construction schedule
and requiring a greater amount of CPT services.
35. For sites improved by densifi cation, does
your post-improvement liquefaction analysis
consider lateral stress relaxation?
• Yes (27%)
• No (73%)
Commentary: Mejia and Boulanger (1995)
performed SPT and CPT to evaluate the
effects of compaction grouting in silt and sand.
The study observed a large increase in the
penetration resistance one week after treatment.
A loss of approximately 30% of the average
increase was subsequently observed within the
following 18 months.
36. For sites improved by densifi cation, does
your post-improvement liquefaction analysis
consider lateral variation in the degree of
densifi cation?
• Yes (48%)
• No (52%)
Commentary: Degen (1998) reports that the
practice of testing at the mid-point between
three vibro-compaction improvement points
(assuming an equilateral triangular spacing)
introduces “a rather large additional factor of
safety into the design”. Field data suggest that
the CPT resistance is about 20% higher, only
500 mm (20 in) away from the midpoint.
DFI JOURNAL Vol. 7 No. 1 August 2013 [21]
37. How do you model the response of
liquefi ed soil when evaluating lateral loading
on a deep foundation?
• use a p-multiplier of 0.1 for loose sand
and 0.25 for dense sand (16%)
• use the equivalent fl uid pressure of the
liquefi ed sand (19%)
• use a p-y curve for soft clay based on
the residual strength (26%)
• use the Rollins et al. (2005) liquefi ed
sand p-y curves (27%)
• other (12%)
CONCLUSIONS
On the basis of the subject survey, the following
conclusions are presented:
1. The state-of-practice is perceived to be
“somewhat” to “highly” non-uniform by a
majority of the survey respondents. This
illustrates the need for continued efforts to
develop greater consensus within engineering
practice for many of the issues included in this
synthesis.
2. The SPT and CPT are the two primary tools
for evaluating the site conditions for the design
and verifi cation for ground improvement for
liquefaction mitigation.
3. A majority of the survey respondents use
an elevated ground water level or a ground
water level at the ground surface for their
liquefaction analysis. Such a practice is
expected to introduce conservatism because
unsaturated soil (even soil near saturation)
has a higher resistance to liquefaction than
saturated soil.
4. Almost one-third of the survey respondents
do not apply a minimum liquefi ed thickness
when performing liquefaction analysis. Such a
practice may introduce conservatism, especially
where liquefaction is isolated to one or a few
thin zones within the subsurface profi le.
5. A performance-based design where the design
criteria are determined based on the tolerance(s)
of the proposed structure, is overwhelmingly
preferred. While two levels of design
performance were recognized (serviceability
and life safety), it was not clear which level
provides the basis of most designs.
6. Densifi cation is the most implemented
primary mechanism for liquefaction mitigation
and is followed by reinforcement. Post-
improvement testing for densifi cation projects
may involve signifi cant judgment to consider
the effects of cementation associated with
aging, stress relaxation and lateral variation in
improvement.
7. While the owner’s consulting engineers
typically defi ne the densifi cation requirements,
it is the specialty contractor (and/or their
subconsultant) that is given the responsibilities
of design, implementation and verifi cation of
the means and methods.
8. The application of reinforcement for
liquefaction mitigation relies on precedence
and judgment, as well as, the results of
numerical modeling. Research is in progress
to better defi ne the effi cacy of reinforcement
and to develop simplifi ed design methods
(Nguyen et al., 2012; Rayamajhi et al., 2012.)
9. Drainage as the primary mechanism does not
appear to be widely implemented in the U.S.
Towhata (2008) reports that the installation
of drains for liquefaction mitigation has
experienced a decrease in Japan.
REFERENCES
1. Andrus, R.D. and Stokoe, K.H. II (2000)
“Liquefaction resistance of soils from shear-
wave velocity”, Journal of Geotechnical and Geoenvironmental Engineering, ASCE,
126(11), pp. 1015-1025.
2. Arulmoli, K., Arulanandan, K., and Seed,
H.B. (1985) “New method for evaluating
liquefaction potential”, Journal of Geotechnical Engineering, ASCE, 111
(1), pp. 95-114.
3. Baez, J.I. and Martin, G.R. (1993) “Advances
in the design of vibro systems for the
improvement of liquefaction resistance”, Proceedings of the Symposium on Ground Improvement, Vancouver Geotechnical
Society, B.C. Canada.
[22] DFI JOURNAL Vol. 7 No. 1 August 2013
4. Boulanger, R.W. and Idriss, I.M. (2006)
“Liquefaction susceptibility criteria for silts
and clays”, Journal of Geotechnical and Geoenvironmental Engineering, 132(11), pp.
1413-1426.
5. Bray, J.D. and Sancio, R.B. (2006)
“Assessment of the liquefaction
susceptibility of fi ne-grained soils”, Journal of Geotechnical and Geoenvironmental Engineering, 132(9), pp. 1165-1177.
6. Charlie, W.A., Rwebyogo, M.F.J. and
Doehring, D.O. (1992) “Time-dependent
cone penetration resistance due to blasting”
7. Cox, B.R. (2012) Liquefaction lessons
learned from recent post-earthquake
reconnaissance, DFI State-of-the-Art Forum:
Liquefaction Consequences and Mitigation,
St. Louis, MO.
8. Degen, W.S. (1998) Vibration Ground Improvement, Vibrofl otation AG,
Altendorf, 194 p.
9. Duke, C.M. and Leeds, D.J. (1963)
“Response of soils, foundations and earth
structures to the Chilean earthquake of
1960”, Bulletin Seismological Society of America, 63(2).
10. Grantz, A., Plafker, G. and Kacherdoorian,
R. (1964) “Alaska’s Good Friday
Earthquake, March 27, 1964”, Geologic Survey Circular 491, Department of the
Interior, Washington.
11. Hossain, M.A., Andrus, R.D. and Camp,
W.M. (2013) “Correcting liquefaction
resistance of unsaturated soil using wave
velocity”, Journal of Geotechnical and Geoenvironmental Engineering, in press.
12. Idriss, I.M. and Boulanger, R.W. (2008)
“Soil liquefaction during earthquakes”,
Monograph MNO-12, EERI.
13. Ishihara, K. and Yoshimine, M. (1992)
“Evaluation of settlements in sand deposits
following liquefaction during earthquakes",
Soils and Foundations, 32(1), pp. 173-188.
14. International Code Council (2012)
International Building Code, 690 p.
15. Japan National Committee on Earthquake
Engineering (1965) “Niigata Earthquake of
1964”, Proceedings, 3rd World Conference on Earthquake Engineering.
16. Kavazanjian, E. Jr. (2012) Evaluation and
mitigation of liquefaction impacts: an
overview, DFI State-of-the-Art Forum:
Liquefaction Consequences and Mitigation,
St. Louis, MO.
17. Kavazanjian, E. Jr., Wang, J-W., Martin,
G.R., Shamsabadi, A., Lam, P., Dickenson,
S.E., and Hung, C.J. (2011) LRFD Seismic Analysis and Design of Transportation Geotechnical Features and Structural Foundations, Report No. FHWA-NHI-11-032, 592 p.
18. Kayen, R.E. and Mitchell, J.K.
(1997) “Assessment of liquefaction
potential during earthquakes by Arias
Intensity”, Journal of Geotechnical and Geoenvironmental Engineering, 123(12),
pp. 1162-1174.
19. Lunne, T., Robertson, P.K. and Powell,
J.J.M (1997) Cone Penetration Testing in Geotechnical Practice, Spon Press, 312 p.
20. Marsal, R.J. (1961) “Behavior of a sandy
uniform soil during the Jaltipan Earthquake,
Mexico”, Proceedings, 5th International Conference on Soil Mechanics and Foundation Engineering, Paris.
21. Martin, J.R. II and Olgun, C.G. (2008)
“Soil improvement for damage mitigation
along Izmit Bay during the 1999 Kocaeli
earthquake”, Geotechnical Engineering for Disaster Mitigation and Rehabilitation.
22. Mejia, L.H. and Boulanger, R.W. (1995)
“A long term test of compaction grouting
for liquefaction mitigation”, Earthquake-Induced Movements and Seismic Remediation of Existing Foundations and Abutments, ASCE, GSP No.55, 94-109.
23. Mesri, G., Feng, T.W. and Benak, J.M.
(1990) “Post-densifi cation penetration
of clean sands”, Journal of Geotechnical Engineering, ASCE, 116(7), pp. 1095-1115.
DFI JOURNAL Vol. 7 No. 1 August 2013 [23]
24. Mitchell, J.K. and Solymar, Z.V. (1984)
“Time-dependent strength gain in freshly
deposited or densifi ed sand”, Journal of the Geotechnical Engineering, ASCE,
104(GT7), pp. 995-1012.
25. Nguyen, T.V., Rayamajhi, D., Boulanger,
R.W., Ashford, S.A., Lu, J., Elgamal, A. and
Shao, L. (2012) “Effect of DSM grids on
shear stress distribution in liquefi able soil”,
Proceedings, Geo-Institute GeoCongress,
Oakland, pp. 1948-1957.
26. Okamura, M. and Soga, Y. (2006) “Effects
of pore fl uid compressibility on liquefaction
resistance of partially saturated sand”, Soils and Foundations, 46(5), pp. 695-700.
27. Olson, S.M. (2012) Lateral spreading during
liquefaction, DFI State-of-the-Art Forum:
Liquefaction Consequences and Mitigation,
St. Louis, MO.
28. Olson, S.M. and Stark, T.D. (2002)
“Liquefi ed strength ratio from liquefaction
fl ow case histories”, Canadian Geotechnical Journal, 39, pp. 629-647.
29. Onoue, A. (1988) “Diagrams considering
well resistance for designing spacing ratio of
gravel drains”, Soils and Foundations, 28(3),
160-168.
30. Pestana, J.M., Hunt, C.E. and Goughnour,
R.R. (1997) “FEQDrain: A fi nite element
computer program for the analysis of the
earthquake generation and dissipation of
pore water pressure in layered sand deposits
with vertical drains”, Report No. UCB/EERC-97/15, Earthquake Engineering
Research Center, College of Engineering,
University of California at Berkeley.
31. Peck, R.B. (1979) “Liquefaction potential:
science versus practice”, Journal of Geotechnical Engineering Division, ASCE,
105(GT3), pp. 393-398.
32. Rayamajhi, D., Nguyen, T.V., Ashford,
S.A., Boulanger, R.W., Lu, J., Elgamal,
A. and Shao, L. (2012) “Effect of discrete
columns on shear stress distribution in
liquefi able soil”, Proceedings, Geo-Institute GeoCongress, Oakland, pp. 1948-1957.
33. Robertson, P.K. and Wride, C.E. (1998)
“Evaluating cyclic liquefaction potential
using the cone penetration test”, Canadian Geotechnical Journal, 35, 442-459.
34. Rollins, K.M., Gerber, T.M., Lane, J.D. and
Ashford, S.A. (2005) “Lateral resistance
of a full-scale pile group in liquefi ed
sand”, Journal of Geotechnical and Geoenvironmental Engineering, 131(1), pp.
115-125.
35. Schmertmann, J.H. (1986) “CPT/DMT
QC of ground modifi cation at a power
plant”, Proceedings of the ASCE Specialty Conference, In Situ ’86: Use of In Situ Tests
in Geotechnical Engineering, Blacksburg,
pp. 985-1001.
36. Seed, H.B. and Booker, J.R. (1977)
"Stabilization of Potentially Liquefi able Sand
Deposits Using Gravel Drains", Journal of the Soil Mechanics and Foundations Division, ASCE, Vol. 103, No. GT7, pp. 757-
768
37. Seed, R.B. and Harder, L.F., Jr. (1990)
“SPT-based analysis of cyclic pore pressure
generation and undrained residual strength”
in H.B. Seed Memorial Symposium, J.M.
Duncan, Editor, Bi-Tech Publishers Ltd,
Vancouver, Canada, Vol.2, pp. 351-376.
38. Seed, H.B. and Idriss, I.M. (1967) “Analysis
of soil liquefaction: Niigata Earthquake”,
Journal of the Soil Mechanics and Foundations Division, ASCE, 93(SM3),
pp. 83-108.
39. Seed, H.B. and Idriss, I.M. (1971)
“Simplifi ed procedure for evaluating soil
liquefaction potential”, Journal of Soil Mechanics and Foundations Division, ASCE,
97(SM9), pp. 1249-1273.
40. Seed, H.B., Idriss, I.M., and Arango, I.
(1983) “Evaluation of liquefaction potential
using fi eld performance data”, Journal of Geotechnical Engineering, 109(3), pp. 458-
482.
[24] DFI JOURNAL Vol. 7 No. 1 August 2013
41. Seed, H.B. and Lee, K.L. (1965) “Studies
of the liquefaction of sands under cyclic
loading conditions”, Report No. TE-65-5 to the State of California Department of Water Resources, University of California at
Berkeley, 46 p.
42. Seed, H.B. and Lee, K.L. (1966)
“Liquefaction of saturated sands during
cyclic loading”, Journal of the Soil Mechanics and Foundations Division, ASCE,
92(SM6), pp. 105-134.
43. Shamato, Y., Sato, M. and Zhang, J-M.
(1996) “Simplifi ed estimation of earthquake-
induced settlements in saturated sand
deposits”, Soils and Foundations, 36(1), pp.
39-50.
44. Tokimatsu, K. and Seed, H.B. (1987)
“Evaluation of settlements in sands due
to earthquake shaking”, Journal of the Geotechnical Engineering Division, ASCE,
113(8), pp. 861-878.
45. Towhata, I. (2008) Geotechnical Earthquake Engineering, Springer Series in Geomechanics and Geoengineering,
46. Whitman, R.V. (1971) “Resistance of soil
to liquefaction and settlement”, Soil and Foundations, 11(4).
47. Youd, T.L. (2011) “Evaluation and
mitigation of liquefaction hazard”,
Geostrata, ASCE, 15(5), pp. 53-54.
48. Youd, T.L., Hansen, C.M. and Bartlett, S.F.
(2002). “Revised multilinear regression
equations for prediction of lateral spread
displacement”, Journal of Geotechnical and Geoenvironmental Engineering, 128(12), pp.
1007-1017.
49. Zhang, G., Robertson, P.K. and Brachman,
R.W.I. (2002) “Estimating liquefaction-
induced ground settlements from CPT for
level ground", Canadian Geotechnical Journal, 39, pp. 1168-1180.
APPENDIX: INTERVIEWS
Interview with Mike Jefferies with Golder Associates who co-authored with Ken Been the book entitled “Soil Liquefaction: A critical state approach” (2006) from CRC Press.
(1) What are your reservations regarding the use of numerical modeling in the design of ground improvement for liquefaction mitigation? Please elaborate.
The biggest reservation is the generic
classifi cation “numerical modeling”, which
would seem to allow anybody to pick up a
standard package (Plaxis, Flac) with some
standard properties lifted from the user manual
to generate a result… “Garage In, Garbage
Out”.
In principle I am a great fan of numerical
modeling and think it will become the way
forward. But, when specifying “modeling”
it is also essential to add the following:
a) an appropriate stress strain model. you
can do a lot with standard Mohr Coulomb
provided that an appropriate dilation angle
is chosen and G = Gmax/3, but for many
ground improvement projects we will need
to go further and adopt a “good” stress
strain model that predicts how the proposed
improvement changes the soil stiffness and
strength. And the problem that you then run
into is that none of the standard numerical
codes used by engineers in practice have
such models as one of their “menu” choices.
So, you wind up with a decent numerical
model that actually requires a user defi ned
model where you code one of the “good”
models for yourself – predictably, rarely
done! Things will improve in the future
as the software developers pre-pack good
models into commercial codes, but even
then users will need to be aware of how
their, seemingly innocent, choice of model
impacts their results.
b) Relevant and reliable soil properties.
Too many times I have seen people
(both clients and fellow engineers) ask
for sophisticated fi nite element analysis
using SPT blowcounts as the basis of
soil properties; this is complete nonsense
and simply produces delusions about
adequacy. The very minimum is that
anything using numerical analysis must
have measured Gmax data as the starting
point for the analysis (at least we get the
DFI JOURNAL Vol. 7 No. 1 August 2013 [25]
elastic response right, and Gmax/3 is not a
bad approximation for the secant modulus
to mobilized strength for many soils at
usual FS). This is actually not a big $$$
requirement, as geophysical methods to
assess Gmax are cheap and easy to do in
situ and ‘bender elements’ are becoming
readily available in commercial testing labs.
It is a matter of appropriate understanding
and attitude. How often do you see the
analyst reporting calibration of their
model to the soil behavior they are
trying to capture?
c) Validation studies have been done.
Although one might like to think (and
hope) that numerical modeling would be
formally correct, in reality there are ways of
setting up the problems, sorting out initial
conditions, and dealing with the loading
conditions that all affect the results. As
well as modeling in 2D when the works
are 3D… All of which means that the
procedures used need to be validated
against ‘case histories’. This validation
of modeling procedures is actually a
requirement in the European standard EN
1997 (Eurocode 7), but not often done by
the consulting fi rms I know.
If you put (a) – (c) together, you wind up with
my fear that simply giving numerical methods
as an allowable/desirable design approach
will allow all sort of bogus work to be put in
front of clients as good engineering. Done
well, numerical methods are a brilliant
technology to help us in what we do. But, they
are done badly 95% of the time (in my view)
and it would be better to not do them at all in
this situation.
(2) What future improvements do you consider most important regarding the different technical aspects related to ground improvement for liquefaction mitigation - liquefaction analysis, design, verifi cation? Please elaborate.
“None of the above”; the biggest problem I
see is intellectual dishonesty/incompetence in
academia where we have largely outsourced
all development – possibly a surprising view,
but it is detailed in my book! In reality,
liquefaction is not diffi cult to analyze, but
what is acceptable has been hijacked by one
faction (lets us be charitable and call them the
‘engineering geology view’) and they have the
support of the regulators. As engineers, we
can do much better than present practice but it
simply is not allowed.
On the bright side, I think the survey results
show a super-majority consensus on post-
treatment validation, and it would be easy to
pull an ASTM standard together on this issue.
The only real issue seems to be the degree of
post-treatment aging we include and how that is
assessed.
(3) What practices (either technical or non-technical) by those involved - consultant, designer, specialty contractor, owner - do you consider problematic to the consistency within the practice? Please elaborate.
a) Continued use of the SPT. The test, and
its correction factors, are so variable that
even a single organization is challenged
in producing a uniform standard between
projects if they base their work on the SPT.
b) Uncertainty and/or lack of understanding
of how “fi nes content” affect liquefaction
susceptibility, leading to wildly inconsistent
engineering in anything other than clean
sand. And it is not the contractors or
owners who are the problem; academia is at
a loss and consultants seemingly take little
interest in challenging them.
c) Lack of best-practice guides. I’ve just been
part of a consensus guide to compaction
grouting, and really similar guides are
needed for dynamic compaction and vibro-
densifi cation (in its various forms).
(4) Are you satisfi ed with the current approach for considering liquefaction-induced settlement and lateral spread?Do you consider these estimates as relative measures of liquefaction severity or actual/accurate values? Please elaborate.
No! None of the approaches to settlement
are properly based on soil behavior (post
[26] DFI JOURNAL Vol. 7 No. 1 August 2013
liquefaction settlement is a consolidation
problem requiring Cc as a basic input
parameter; any method without Cc as a soil
property is bogus mechanics). Similar
comments follow about lateral spreads with the
exception of Newmark’s method.
(5) Do you have any other comments that haven't been covered by the survey or this interview?
Could I suggest the compaction grout guide as
a prototype of what is needed across other areas
of the ground improvement industry?
Interview with Dr. Les Youd, Professor Emeritus of Civil Engineering at Brigham Young University.
(1) Do you have any reservations regarding the use of numerical modeling in the design of ground improvement for liquefaction mitigation? Please elaborate.
I have major reservations on this issue.
Numerical modeling is only one of several
tools that should be applied by designers.
Over reliance on numerical modeling can lead
to nonsensical results because of imperfect
or inadequate models. With the present state
of practice, it is generally impossible to
construct accurate models of (1) subsurface soil
stratigraphy, (2) lateral and vertical variances
in stratigraphy, (3) soil properties, (4) variances
in soil properties in space and time, and (5)
imposed seismic loads in both space and time.
Numerical analyses are generally useful to
gain a rough perspective of expected results
and to perform parametric analyses to estimate
how the results might change with variations
in stratigraphy, soil properties, loading
assumptions, etc., but not as the sole basis
for design.
A more important tool, in my opinion,
is assessment of case histories of past
performance compiled from post-earthquake
investigations and successful or unsuccessful
similar projects. Empirical procedures
are generally applied in present practice;
these procedures were primarily developed
from analysis of case histories, and thus are
grounded on observed actual performance.
Although past performance and empirical
techniques may not allow exact duplication of
site conditions and constraints for each project,
they usually provide realistic estimates as a
guide to the design engineer.
The third tool that should be applied is
engineering judgment. Expert guidance from
those with past earthquake experience and
from analysis and design should be sought
after to assure that sound engineering judgment
is applied on all critical ground modifi cation
projects. Conversely, design of critical ground
modifi cation projects should not be entrusted
to inexperienced engineers, although they may
have attained expert computer skills but with
little practical experience.
(2) What future improvements do you consider most important regarding the different technical aspects related to ground improvement for liquefaction mitigation - liquefaction analysis, design, verifi cation? Please elaborate.
I believe that much money has been wasted in
the past on mitigation to prevent liquefaction
from occurring, when the occurrence of
liquefaction would not lead to signifi cant
damage. To avoid such waste, education
of the profession is needed to increase
understanding of the following key points
(extracted from a short paper I prepared for
the ASCE publication Geo-Strata (Youd,
September-October 2011, p. 53-54)). This
paper was written to increase this needed
understanding. “In analyzing liquefaction
the following fundamental questions should
be asked and appropriate answers and actions
determined: 1. Will liquefaction occur? If the
answer to this question is “no,” mitigation is
obviously not required. If the answer is “yes,”
the analysis proceeds to the second question: 2.
Will liquefaction lead to potentially damaging
ground deformations, displacements or ground
failure? If the answer is “no,” liquefaction will
not cause signifi cant damage and the hazard
can be safely accepted without mitigation. If
the answer is “yes,” the analysis proceeds to the
third question: 3. What mitigative measures are
DFI JOURNAL Vol. 7 No. 1 August 2013 [27]
required to reduce the hazard to an acceptable
risk?” Only when this level of understanding
has been gained should the analysis proceed to
design of mitigation measures.
Verifi cation of the effectiveness of ground
modifi cations has been a major issue for
several projects I have encountered. Additional
research, discussion and consensus building is
required to improve verifi cation procedures for
use in engineering practice.
(3) What practices (either technical or non-technical) by those involved - consultant, designer, specialty contractor, owner - do you consider problematic to the consistency within the practice? Please elaborate.
Procedures for liquefaction hazard evaluation
and mitigation relies heavily upon empirical
procedures, which are generally based on
analyses of collected case histories and
performance assessments. Development
of empirical procedures usually occurs
through research and analyses by individual
investigators or teams of investigators; these
investigators or teams do not always (or seldom)
agree; thus development of empirical procedures
tends to be a messy and often chaotic process;
disagreements and disputes are common and
to be expected. Most of this chaos occurs at
the researcher and consultant levels. Because
of this chaos, practitioners and designers often
are confused or uncertain as to which expert
they should rely on or which procedure they
should follow. With time the chaos usually
calms as procedures are vetted or tested and
consensus builds. Sometimes professional
societies or other professional groups can
speed the process through workshops or expert
panels to develop consensus guidelines for
engineering practice. Such was the case with
the NCEER/NSF workshop I chaired in 1996
on evaluation of liquefaction resistance, which
developed consensus guidelines that calmed the
atmosphere for triggering evaluations for about
10 years. Groups such as DFI may assist by
organizing or supporting workshops or panels
of this type to build consensus and reduce chaos
and develop improved and more consistent
usage within the profession.
(4) Are you satisfi ed with the current approach for considering liquefaction-induced settlement and lateral spread? Do you consider these estimates as relative measures of liquefaction severity or actual/accurate values? Please elaborate.
a) Lateral Spread: As an author of the
empirical MLR procedure, one of the more
widely used procedures for evaluating
lateral spread displacement, I feel that the
MLR procedure is a valid procedure if
applied within the limitations specifi ed by
the authors (Youd et al 2002 from Journ
of Geotech and Geoenviron Engr, v. 128,
no 12, p. 1007-1017). This procedure
provides mean predicted values that are
demonstrated to be accurate within a factor
of plus or minus two if applied within the
specifi ed limits. Extension beyond the
stated limits leads to greater uncertainty
of results. Because the MLR procedure is
empirical, it is not valid for all conditions
that may be encountered. In some instances
extrapolation using numerical procedures
may allow reasonable, but still uncertain,
results for a wider range of site conditions.
For example, inclusion of a deep foundation
for a bridge in a sediment cross-section to
be analyzed creates a condition beyond that
in the empirical database. The additional
infl uence of this bridge foundation could
be analyzed through numerical procedures.
An accuracy of plus or minus 2 may seem
too uncertain for engineering applications,
however, such uncertainties are common in
other geotechnical engineering calculations.
For example, similar uncertainty is
inherent in calculations of bearing capacity
and foundations settlement under static
conditions.
One of the greater sources of erroneous
results that I have encountered reviewing the
work of others using the MLR procedure is
insuffi cient geotechnical information. Often
analyses are made on the basis of one or
a few boreholes or soundings exacerbated
by an improper assumption that penetrated
soil layers are laterally homogeneous
[28] DFI JOURNAL Vol. 7 No. 1 August 2013
and continuous across and beyond site
boundaries. If critical layers thin or pinch
out or important facies changes occur
within the layer, inaccurate to nonsensical
predicted displacements may be calculated.
Thus, I feel that adequately accurate tools
are available for calculation of lateral
spread displacements for many applications.
However, MLR and other empirical
procedures need to be verifi ed and updated
as additional earthquakes occur and new
case histories are developed.
b) Ground Settlement: The same general
limitations apply to empirical procedures
for calculation of liquefaction-induced
ground settlement as for lateral spread.
However, the limitations for ground
settlement do not seem to be as well
defined as for lateral spread. For example,
most empirical settlement procedures
appear to be based on relatively clean sand
conditions. Limits on the procedure with
respect to silt and gravel contents do not
seem to be clearly defined. Thus, the
user must evaluate soil conditions at a site
in question, compare those conditions
against those implicit in the development
of the empirical procedure, and then
make an unspecified adjustment for
incompatible soil conditions. Such
adjustments increase the uncertainty of
the calculated settlements.
(5) Do you have any other comments that haven't been covered by the survey or this interview?
From reading the survey text and these
questions, diffi culties faced by design engineers
appear to stem from one or more of the
following issues:
(1) lack of adequate communication between
researchers, expert consultants, analysts, and
designers;
(2) confusion within these same groups with
respect to which procedures should be
recommended for application in practice,
and;
(3) lack of adequate research and verifi cation
of procedures to answer many fundamental
questions.
DFI and other professional organizations
could play a major role in fostering
communication, supporting studies to
develop and verify procedures, education of
professionals at all levels, assisting profession
to identify unresolved issues, and assisting
in the development of support for research,
workshops and other means to resolve
important issues.
Interview with Ikuo Towhata, Professor of Geotechnical Engineering at the University of Tokyo and author of the “Geotechnical Earthquake Engineering” from Springer Series.
(1) Do you have any reservations regarding the use of numerical modeling in the design of ground improvement for liquefaction mitigation? Please elaborate.
Everybody points out the shortcomings of the
use of numerical analysis in design. Those
shortcomings are caused by the complex stress-
strain-dilative behavior of soils, heterogeneous
subsoil conditions that cannot be fully
recorded by current soil investigations and
many others. Although those critical attitudes
are understandable, I feel that some critics
use those shortcomings as an excuse for not
challenging advanced (numerical) studies.
Recent desires towards seismic performance
design require approaches that can calculate
residual deformation of structures in place of
the conventional factor of safety. I would ask a
question whether or not the traditional
non-numerical approaches are more reliable and
more useful than the numerical approaches. The
traditional approaches often rely on empirical
knowledge and their use is certainly limited
within the range of available knowledge. It is
risky to apply them to totally new soil and load
conditions. Moreover, the traditional methods
cannot be applied to the behavior of complicated
underground structures that are subject to
liquefaction of soils around. In this regard, we
should not discriminate numerical methods.
They should be considered to be tools which
DFI JOURNAL Vol. 7 No. 1 August 2013 [29]
give us indices that help us assess the seismic
performance of structures to be designed.
Do not misunderstand that I am trying to
favor numerical methods. Good numerical
methods have to be associated with elaborate
but costly fi eld / laboratory investigations. I
feel that many current projects do not allocate
reasonable budget to investigations, resulting
in unexpected diffi culty during the later
construction stages. A small investigation
budget results in a great loss of money and time
during construction.
The attitudes of numerical people are also a
problem. They do not go to the site. They prefer
to stay in a comfortable offi ce and 100% trust
documented data. They do not imagine that the
reality is more complicated than information
in paper.
Because my most interested fi eld of study is
the assessment of liquefaction-induced large
displacement, I should make one more point
about numerical approaches. To my knowledge,
the constitutive models that are employed
in major computer codes were developed in
1970s and 80s when nobody cared that the
liquefaction-induced large deformation of
subsoil. Also, even today laboratory devices
cannot reproduce such large shear deformation
as 30%, 50%, or more. Laboratory tests after
the onset of liquefaction are not possible
because of segregation of water and sand grains
within a tested specimen. Therefore, those
constitutive models are not fully supported by
laboratory test data after onset of liquefaction
and development of large shear deformation.
In summary, I would propose to use both
simple traditional approaches and numerical
approaches and compare them prior to drawing
the fi nal conclusion.
(2) What future improvements do you consider most important regarding the different technical aspects related to ground improvement for liquefaction mitigation - liquefaction analysis, design, verifi cation? Please elaborate.
After the M=9 gigantic earthquake in 2011,
I encountered a very diffi cult problem of soil
improvement. It was how to improve soil
(reduction of liquefaction vulnerability) under
existing houses with relatively low fi nancial
burden to house owners.
The current solution is two-fold. For a frequent
design earthquake (return period being about 50
years), public and private funds are combined.
Liquefaction vulnerability is mitigated by either
constructing underground rigid walls under
streets and house-lot borders to constrain cyclic
shear deformation of soil, or pumping ground
water to lower the ground water level and to
create an unliquefi able soil crust. The former
has a limitation that the spacing of walls cannot
be very small because of existing houses at the
surface. The latter has a problem of possibly
triggering consolidation settlement in the
underlying thick soft clay.
Note further that house owners have to spend
their own money on soil improvement, if
they wish to do it, against a stronger design
earthquake with a return period of hundreds
of years. This is diffi cult and costly because
the ground surface is occupied by a house
and compaction or other traditional soil
improvement is not possible.
Soil improvement under existing structures
is further important for big industries as
well. This is because the intensity of design
earthquake tends to be increased and existing
old structures cannot satisfy the safety
requirement under newer regulations.
I suppose that the following kinds of soil
improvement are feasible under existing
structures; installation of drainage, compaction
grouting by good technicians, injection
of colloidal silica, and construction of
underground walls around the foundation
of houses. Noteworthy is that some
improvement methods cannot prevent the
onset of liquefaction but reduce the residual
deformation. Thus, methodology is necessary
to assess the residual deformation of subsoil
and surface structures (possibly by a numerical
method) and also to determine the allowable
extent of deformation.
[30] DFI JOURNAL Vol. 7 No. 1 August 2013
Gravel drains became less popular after the
1995 Kobe earthquake. This is because the
intensity of design earthquake was increased
and design calculation could not prove that
gravel drains under the stronger design
earthquake can still maintain the development
of excess pore water pressure less than 50%
of the full liquefaction. However, it is possible
that the columns of gravel drain maintain
some rigidity during a strong earthquake and
reinforce the stability and integrity of subsoil.
Further study is needed in this direction.
(3) What practices (either technical or non-technical) by those involved - consultant, designer, specialty contractor, owner - do you consider problematic to the consistency within the practice? Please elaborate.
Owners try not to spend suffi cient money
on subsoil investigation. Hence all the input
data for analysis have to be determined by
SPT-N only. Although liquefaction of fi ne-
grained soil is important, plasticity index is
hardly measured. Owners should understand
that they should allocate more money on
soil investigation so that they can avoid
unnecessary big expenditures on construction
and unnecessary liquefaction damage during
future earthquakes.
Some consultants do not want to visit sites.
They prefer to stay in the offi ce and analyze
the supplied borehole data. For them, the
data on paper is the reality and they do not
want to experience the reality on site. One
reason for this situation is found in the
owners who do not pay suffi cient money for
fi eld activities. To accurately interpret bore
hole data, it is important for consultants to
have good knowledge of local soils and local
geological history as well as history of human
action on soils such as land reclamation and
consolidation settlement. Hence, it is possible
that a local experienced consultant is better
than an international famous consultant.
(4) Are you satisfi ed with the current approach for considering liquefaction-induced settlement and lateral spread? Do you consider these estimates as relative
measures of liquefaction severity or actual/accurate values? Please elaborate.
My attitude towards numerical methods was
described already in (1). I do not think that
numerical methods are worse than simplifi ed
and traditional factor-of-safety approach. All
the methods give us an index to assess the
performance of a designed structure.
Numerical methods can assess the lateral
displacement of liquefi ed subsoil with an
error of +- 50%. This is not too bad. The error
of non-numerical methods is most probably
similar.
Some people state that the Newmark rigid
block analogy is better than other methods
to assess the liquefaction-induced lateral
displacement. I would say that the use of the
Newmark method in liquefaction problem
is beyond the applicability of this method,
because Newmark intended to calculate the
movement of a "rigid block" subjected to
seismic action. Liquefi ed soil is never a rigid
block.
(5) Do you have any other comments that haven't been covered by the survey or this interview?
There are several methods of subsoil
investigation. In publications, I often encounter
such an article in which the author insists that
his device is able to determine all the required
soil data accurately. In reality, this is diffi cult.
Every method has good points and bad points.
I suggest that we should combine different
methods and get the best subsoil data. It is good
to interpolate a big spacing among SPT by
means of less expensive CPT or other device.
It is often stated that engineering judgment is
extremely important and that less experienced
engineers should not be trusted. Then it
becomes important how to produce the next
generation of experienced engineers. If a
young engineer is not trusted, he will never
become an experienced one. Moreover, the
engineering judgment is a kind of empiricism.
During the medieval time, technology relied
totally on empiricism. There was no systematic
education. Hence, the power of technology was
DFI JOURNAL Vol. 7 No. 1 August 2013 [31]
poor. This situation was changed drastically
during the time of modern technology and
education because the problems were analyzed,
interpreted, understood, and solved. Long
patience behind a “meister” is not necessary
any more. It should be borne in mind that too
much emphasis on empiricism will push things
back to the medieval time.
[32] DFI JOURNAL Vol. 7 No. 1 August 2013
Grouted Micropiles for Foundation Remediation in Expansive Soil (8th Michael W. O’Neill Lecture)John D. Nelson, Ph.D., P.E., D.GE., CEO and Principal Geotechnical Engineer, Engineering Analytics, Inc., and Professor Emeritus, Colorado State University, Fort Collins, Colorado, USA: 970-488-3111, [email protected]
Kuo-Chieh Chao, Ph.D., P.E., Vice President and Senior Geotechnical Engineer, Engineering Analytics, Inc., Fort Collins, Colorado, USA; 970-488-3111, [email protected]
Daniel D. Overton, M.S., P.E., President and Principal Geotechnical Engineering, Engineering Analytics, Inc., Fort Collins, Colorado, USA
Zachary P. Fox, M.S., Geotechnical Engineer, Engineering Analytics, Inc., Fort Collins, Colorado, USA
Jesse S. Dunham-Friel, M.S., P.E., Geotechnical Engineer, Engineering Analytics, Inc., Fort Collins, Colorado, USA
ABSTRACTFoundation underpinning is a common component of remediation schemes for distressed foundations on expansive soils. For many applications in expansive soil, micropiles have distinct advantages over other techniques. This paper will concentrate on the design and construction of micropiles in expansive soil. It discusses the nature of building distress and the relationship between foundation movement and soil heave. It presents methods for determining the factors that are required for the design of micropiles. Such factors include calculation of expected free-field heave, depth of soil wetting, and prediction of pier movement. A finite element program developed by the authors and others to determine pier heave and internal forces is presented. The input parameters that are required for pier analysis are discussed, and the nature of the output and the sensitivity of the results to the output are described. A case example illustrates the advantages of micropiles over other methods.
INTRODUCTIONHeave of expansive soils is a common cause of differential movement of building foundations resulting in structural distress. For foundations constructed on soils consisting of highly expansive clay, underpinning of the foundation is the most reliable method of remediation. Recently, micropiles have found increasing use for underpinning, particularly in the Front Range of Colorado because of the reliability of the method and its ease of installation. The drilling equipment for micropiles is easily attached to existing foundations utilizing the weight of the structure for reaction. This makes the use of micropiles advantageous in places such as crawlspaces, garages, basements, and other confined areas.
Appropriate design of the micropiles involves careful site investigation, calculation of anticipated free-field heave, and then analysis of the required micropile length. The successful
performance of the micropiles also involves careful attention to detail during construction.
The following sections present examples of the nature of distress caused by heave of expansive soil and typical foundation types that have been underpinned. They outline the geotechnical engineering parameters that are necessary for design and present methods for analysis of the micropiles. The input parameters required and methods of determination of these parameters are discussed. Important aspects of the construction are also discussed. A case example is used to demonstrate the advantages of micropiles in terms of ease of installation and reliability.
FOUNDATION TYPES AND NATURE OF DISTRESS When highly expansive soils are encountered on a given site, the most reliable foundation type is a deep foundation consisting of drilled
DFI JOURNAL Vol. 7 No. 1 August 2013 [33]
pier and grade beam systems. Drilled pier and grade beam foundations isolate the structure from the expansive soils by creating a void space beneath the superstructure such that only the shaft of the drilled pier is in contact with the problematic soil. As will be discussed in greater detail later, uplift forces acting on the upper portion of a pier due to soil heave in the active zone are resisted by the embedment or anchorage zone below. Distress in pier and grade beam foundations caused by expansive soils is typically the result of differential pier heave and manifests itself through cracking of the pier and/or grade beam causing distortion of the superstructure above. Fig. 1 shows a grade beam that experienced diagonal cracking due to pier heave. Fig. 2 shows a diagonal crack in a 30 inch (762 mm) diameter drilled pier near the intersection with the grade beam. In this case lateral forces were also imposed on the pier due to soil heave.
[FIG. 1] - Grade beam crack due to pier heave
[FIG. 2] - Diagonal crack in a 30 inch drilled pier
Associated with the pier and grade beam foun-dation is a structural floor to isolate the floor slab from the soil. Slab-on-grade basement floors can experience large amounts of heave that can also be transmitted to the superstruc-ture above. Fig. 3 shows a scenario where sig-nificant slab heave has necessitated the cutting of the interior wall studs in the basement of a residence to avoid lifting the first floor. Fig. 4 shows a “center lift” condition in a basement slab-on-grade.
[FIG. 3] - Wall studs sawed off due to slab heave
[FIG. 4] - Differential heave of basement slab
Regardless of the foundation type, distress as-sociated with expansive soils typically results in significantly increased maintenance and repair costs throughout the life cycle of the structure. Additionally, differential movements result in racked doors and windows which in addition to inconvenience, may result in loss of emer-gency egress. To remediate such distress and losses of functionality, foundations are often underpinned with structural elements such as micropiles.
[34] DFI JOURNAL Vol. 7 No. 1 August 2013
DESIGN OF MICROPILES IN EXPANSIVE SOILSMicropiles have been used to underpin foundations since the early 1950s and they are increasingly being used for underpinning foundations experiencing heave due to expansive soils. Despite their increasing usage, there is a lack of published literature regarding micropile design, installation, or performance in expansive soils. The following offers a method for analysis of the behavior of micropiles installed in expansive soils.
Heave Prediction
Free-field heave is defined as the amount of heave the ground surface will experience without any applied surface load. The distribution of heave with depth is the primary data on which pier heave is calculated. Therefore, a review of free-field heave calculations is presented below.
Various heave prediction methods have been developed based on results of one-dimensional oedometer tests (Fredlund et al. 1980; U.S. Army Corps of Engineers, 1983; Nelson and Miller, 1992; Fredlund and Rahardjo, 1993; Fredlund et al. 2012). These methods utilize the net mechanical stress, σ’ = (σ – u
a), and
the matric suction, h = (ua - u
w) as the stress
state variables. In these variables, σ is the total stress and u
a and u
w are the pore air and pore
water pressures. The soil heave takes place as the suction is decreased. These methods are commonly referred to as “oedometer” methods. The oedometer methods all use the same basic equation for calculation of heave. The equation for heave of a soil layer of thickness, ∆z
i,
subjected to an applied stress, ∆σ’v, is
DESIGN OF MICROPILES IN EXPANSIVE SOILS
icv
vvoiHi zC Δ+Δ= ⋅ '
''
logσ
σσρ [1]
and the heave of the entire soil column is,
DESIGN OF MICROPILES IN EXPANSIVE SOILS
=
=n
i 1iρρ [2]
where: ρ = free-field heave; ∆zi = thickness of
each soil layer; σ’vo
= overburden stress; ∆σ’v =
applied stress; σ’cv
= constant-volume swelling pressure, and C
H = constitutive parameter.
The parameter CH defines the amount by which
a soil sample will swell when it becomes wetted. The method presented here is characterized
by the manner in which the CH parameter is
determined. It considers both the change in suction due to wetting and the applied stress that is acting on the soil when it is wetted.
The determination of CH is depicted in Fig. 5
which is a three dimensional plot of the stress paths followed during the inundation and heave of a soil. In a conventional consolidation-swell oedometer test, a sample of soil is consolidated under an inundation stress, labelled as σ’
i in
Fig. 5. The initial state of the soil under the inundation stress, σ’
i, is represented by the
point labeled K. At that point the soil suction is equal to some value labelled as h
c1. The initial
percent swell, εs%
, at point K (and H) is equal to zero. When the sample is inundated, the suction is reduced to h
o and the soil swells along the
path KB. The projection of that stress path on the plane defined by the axes for ε
s% and log σ’
is the line GB. The sample is then loaded back to its original height along the path BA. The value of stress corresponding to point A is the “consolidation-swell swelling pressure”, σ’
cs.
[FIG. 5] - Stress Paths for Soil Expansion
In a conventional constant-volume oedometer test, the sample is constrained from swelling during inundation and the stress required to prevent swell is determined. The initial point for this test is also point K but because it is constrained from swelling it develops a confining stress as the suction decreases to h
o and the stress path would be along a line
such as KE. The value of stress corresponding to point E is the “constant volume swelling pressure”, σ’
cv. Due to hysteretic effects, the
value of σ’cv
is generally less than that of σ’cs.
The reason for this is somewhat intuitive in that it should be easier to prevent water molecules
DFI JOURNAL Vol. 7 No. 1 August 2013 [35]
from entering into the soil lattice than to force the water out once it has entered into the soil.
For an element of soil at some depth in the ground, the initial stress conditions could be at some point such as J. When that sample is inundated it will swell along a stress path such as JD. Point D will fall between points B and E. Our experience and data has shown that the line BDE is close to being a straight line (Justo et al. 1984; Reichler, 1997; Nelson et al. 2006; Fredlund et al. 2012). Thus, the slope of the line BDE defines the constitutive relationship between the percent swell, ε
s%, that a soil will
undergo, and the applied stress when it is wetted. The slope of that line is C
H.
It is important to note that, as shown in Fig. 5, the line BDE which defines C
H represents the
expansion, or heave, that will occur due to suction changes under different values of applied stress. Thus, it is a constitutive relationship that incorporates both of the independent stress state variables, σ’ and (u
a –
uw), for use in computing heave.
Fig. 6 shows the projection of the stress paths shown in Fig. 5 onto the ε
s% and log σ’ plane.
The results of both consolidation-swell test and constant-volume test are shown as the lines GBA and GFE, respectively.
The heave index, CH, is the slope of the line BDE
in Fig. 6 and is equal to:
[FIG. 5] Stress Paths for Soil Expansion
s%CH 'cv100 log 'i
ε=
×
[3]
where εs%
is the percent swell corresponding to σ’
i expressed as a percent, and σ’
i is the vertical
stress at which the sample is inundated.
[FIG. 6] - Determination of Heave Index, CH
The value of CH can be determined from the
results of a consolidation-swell test and a constant volume test using identical samples of the same soil. However, in practice it is virtually impossible to obtain two identical samples from the field. Therefore, it is convenient to utilize a relationship between σ’
cs and σ’
cv so that C
H
can be determined from a single consolidation-swell test. On the basis of data that has been assembled from a number of different sources, it was found that experimental data corresponded well to Equations 4a and 4b. The authors have found that for use in the Front Range area of Colorado a value of λ
l of
0.6 is reasonable when Equation 4a is used or a value of λ
a of 0.3 is reasonable when
Equation 4b is used. However, for application of these equations to other soil types, it would be prudent to perform tests to determine an appropriate value for the soil being considered (Nelson et al. 2012a).
[FIG. 6] Determination of Heave Index, CH
)(logloglog '
''i
'cv
i
csl σ
σλσ +=
[4a]
[FIG. 6] Determination of Heave Index, CH
)( '''i
'cv icsa σσλσ −+=
[4b]
Zone of Soil Contributing to Heave
The depth of soil that is contributing to heave at a particular point in time depends on two factors. These are the depth to which water contents in the soil have increased since the time of construction, and the expansion potential of the various soil strata. As water migrates through a soil profile different strata become wetted, some of which may have more swell potential than others. Consequently, the zone of soil that is contributing to heave varies with time.
The amount of heave that will occur at a particular time depends on the manner in which the groundwater migrates in the soil and the expansion potential of the soil at depth. Movement of the soil surface will begin almost immediately after construction, whereas some time will be required for the soil at deeper depths to become wetted. Thus, the surface of the soil will begin to heave almost immediately, but movement of piers will be delayed, sometimes by several years.
The term “active zone” has been in common usage in the field of expansive soils. However, the usage of that term has taken different
[36] DFI JOURNAL Vol. 7 No. 1 August 2013
meanings at different times and in different places. Therefore, for purposes of clarity and consistency, the following definitions have been put forth (Nelson et al. 2001).
Active Zone, ZA, is that zone of soil that is
contributing to heave due to soil expansion at a particular point in time.
Zone of Seasonal Moisture Fluctuation, Zs, is
that zone of soil in which water contents change seasonally due to climate changes.
Zone of Wetting, Zw, is that zone in which
water contents have increased beyond the pre-construction conditions.
Depth of Potential Heave, Zp, is the depth to
which the overburden vertical stress equals or exceeds the swelling pressure of the soil. This represents the maximum depth of Active Zone that could occur.
Design Active Zone, ZAD
, is the zone of soil that is expected to become wetted during the design life of the structure. It may be less than the depth of potential heave if the entire depth of potential heave is not expected to become wetted. If water migration analyses are not performed and if the depth of potential heave is of reasonable value for design, it is prudent to assume the depth of the design active zone is equal to the depth of potential heave.
Construction of buildings and pavements in arid regions typically results in a reduction of evapotranspiration from the soil. Additionally, the introduction of irrigation typically exceeds the evapotranspiration of the vegetation. These factors as well as others result in the development of a wetting front that progresses downward in the soil. Above the wetting front, the soil may be saturated or unsaturated. The difference in soil suction between the wetter and drier zones will result in downward flow of water, and the wetting front will continue to move downward until an impermeable boundary or a water table is reached (McWhorter and Nelson, 1979). Once a low permeability boundary is reached by the wetting front, a perched water table will be formed. Full wetting of the soil profile would be expected to occur if the soil above the wetting front is saturated and the wetting front advances to below the depth of potential heave. If full wetting is not expected to occur, analyses should be conducted to determine the water content profile at the end of the design life.
Pier Heave Calculations
The simplest method used to design piers in expansive soil is termed the “Rigid Pier” method. This method assumes that the pier will not move and determines a required pier length by equating the negative, or downward, skin friction below the depth of the design active zone, plus the dead load, to the uplift pressures exerted on the pier by the swelling soil. Chen (1965), O’Neill (1988), and Nelson and Miller (1992) present methods for rigid pier analysis in expansive soil. Rigid pier design generally produces conservative pier lengths for a light structure founded on a deep deposit of highly expansive soil. The rigid pier design works well if the stratum of expansive soil is not thick and is underlain by a stable non-expansive stratum. However, in a deep deposit of expansive soil, the design rigid pier length is generally not practical for a light structure.
In reality almost all structures are able to tolerate some amount of pier heave. The amount of tolerable heave to be used for design depends on the nature of the structure. Methods of analysis of pier heave were developed by Poulos and Davis (1980) and were adapted by Nelson and Miller (1992) to develop design charts for calculating pier heave. This method is termed the “Elastic Pier” method. The elastic pier method calculates the pier heave assuming the pier is a stiff inclusion in an elastic half space. The elastic pier method presented in Nelson and Miller (1992) was developed for piers with uniform properties installed in a uniform soil profile. Additionally, the elastic pier analysis formulation breaks down when the length to diameter ratio becomes too great. Micropiles typically have non-uniform properties with depth, are often installed in non-uniform soil profiles, and have large length to diameter ratios. Therefore, the elastic pier method is not well suited for their analysis.
Finite element approaches to pier analyses provide versatility to consider such details as non-uniform soil or pier interface properties with depth and large length to diameter ratios. Nelson et al. (2012b) presents one such finite element based numerical analysis approach termed APEX for Analysis of Piers in EXpansive Soils. This finite element based approach is discussed below.
DFI JOURNAL Vol. 7 No. 1 August 2013 [37]
APEX Formulation
The APEX formulation is discussed in detail in Nelson et al. (2012b) and is briefly summarized below. Swell is assumed to be isotropic and it is simulated using conventional analyses of thermal strains in solids. The constitutive equations are as follows:
( )1rr rr zz isoE θθε σ ν σ σ ε= − + + [5]
( )1zz rr isoEθθ θθε σ ν σ σ ε= − + + [6]
( )1zz zz rr isoE θθε σ ν σ σ ε= − + + [7]
where: εiso
= isotropic swelling strain; and εrr,
εθθ, εzz = components of stress and strain in
cylindrical coordinates. The pier-soil interface is accounted for with a mixed boundary condition. The mixed boundary condition is shown in Fig. 7 and can be written as follows:
( )t p tF k H U= − [8]
where: Ft = nodal force tangent to pier; H
p =
pier heave; Ut = nodal displacement tangent to
pier; and k = parameter used to adjust shear stress, which serve a purpose similar to a spring constant.
Fig. 8 depicts the manner in which APEX calculates pier heave. The pier is modeled as a rigid body connected to an elastic, expansive medium by springs with a spring constant k. Fig. 8a illustrates the conditions before swell takes place when there are no uplift forces on the pier. Fig. 8b illustrates the conditions after swelling takes place but before any pier heave, when the shear forces exerted on the pier result in an upward force on the pier. At this point the pier is not in equilibrium, and the pier is then allowed to move up until forces are balanced. Fig. 8c illustrates the condition after forces are balanced and the pier is in equilibrium.
The APEX formulation allows for movement between the pier and the expansive soil mass by either slip between the pier-to-soil interface or failure of the soil adjacent to the pier. The slip and soil failure mechanisms are calculated at each iteration to evaluate which condition governs.
[FIG. 7] - Boundary Conditions: (a) Soil Boundary, (b) Mixed Boundary (after Nelson et al. 2012b)
[FIG. 8] - Schematic of pier and soil interface: (a) initial-no force on pier, (b) soil heave-upward force on pier, (c)
pier heave-resultant force on pier is zero (after Nelson et al. 2012b)
Input Parameters for APEX
The soil and heave profiles are the primary input parameters used in the APEX analysis. Detailed and accurate characterization of the soil profile to the full depth to which the soil will influence the behavior of the piers is a critical element of pier design. If the depth of exploration is too shallow, or if an insufficient number of samples are collected and tested, variations in the soil profile will not be detected. Fig. 9 shows two examples of soil profiles.
Fig. 9a illustrates a simplified soil profile where one relatively uniform expansive soil such as clay or claystone is encountered to the full depth of exploration. In this case the incremental heave is high at the surface and decreases exponentially with depth to the depth of potential heave. Relatively uniform soil
[38] DFI JOURNAL Vol. 7 No. 1 August 2013
profiles such as that shown in Fig. 9a are rarely encountered in the field. Instead it is typical to encounter multiple soil layers with varying expansion potential such as in the profile shown in Fig. 9b. Fig. 9b illustrates a complex soil profile which is typical of many expansive soil sites in the Front Range Area of Colorado. The soil profile shown in Fig. 9b has multiple layers with varying expansion potential. Accurate analysis of pier heave constructed in complex soil profiles such as that requires a detailed analysis which can account for the variations in heave with depth. Incremental free-field heave computed for such profiles is the most important input parameter in the APEX analysis. The free-field heave profile can be determined by predicting heave versus depth as discussed in the above sections.
The primary elastic input properties used in APEX analysis are the Young’s modulus (E), Poisson’s ratio (ν), and coefficient of lateral stress (K
o). The sensitivity of the analysis to
those parameters is discussed in detail in Nelson et al. (2012b).
Example calculations performed by Nelson et al. (2012b) have demonstrated that changes to the interface friction parameters, α, do not substantially affect the calculated pier heave but do have a significant impact on the tensile force in the pier when the frictional interface is uniform with depth. However, if
the upper portion of the micropiles installed in expansive soils is sleeved with PVC, the frictional properties for each part require accurate determination of the value for α with depth along the micropile. The APEX analysis developed by the authors and others allows for α to be changed with depth in order to allow for accurate representation of the frictional properties at all locations along the micropile.
The frictional interfaces that typically occur during the construction of micropiles in expansive soils are soil to grout, grout to PVC, and PVC to soil as is discussed in Schaut et al. (2011). The values of α presented in the literature for a concrete to soil interface generally range from 0.1 to 0.25 (Chen, 1988; O’Neill, 1988; Sorochan, 1991; Nelson and Miller, 1992). However, field test results presented by Benvenga (2005) indicate that α generally ranges from about 0.4 to 0.6 and can be as high as 0.9. Schaut et al. (2011) completed testing on the soil to grout interface as well as the grout to PVC and PVC to soil interfaces using typical micropile construction materials and claystone soil from the Front Range Area of Colorado. The results of this research indicate that the value of α depends on the method of testing, whether the soil is remolded and what the water content of the soil is during testing. It was shown that PVC casing reduces the frictional resistance along the cased section of the micropile.
[FIG. 9] - Examples of soil and heave profi les: (a) uniform expansive soil and (b) complex soil profi le
DFI JOURNAL Vol. 7 No. 1 August 2013 [39]
Output Results from APEXTypical APEX input and output for a soil profile is presented on Fig. 10. Fig. 10a shows the heave profile input into the APEX program. Fig. 10b shows the distribution of slip along the pier. This figure indicates that for this case, slip was the failure mechanism along the entire length of the pier and therefore soil failure was not experienced. Fig. 10c shows the distribution of shear stress along the pier with positive shear stresses in the anchorage zone and negative shear stresses in the uplift zone. Fig. 10d shows the axial force in the pier with the maximum value occuring at the interface between the uplift and anchorage zones.
[FIG. 10] - Typical output from APEX Program: (a) cumulative heave used as input, (b) variation of slip along pier, (c) shear stress distribution along pier, (d) axial force distribution (after Nelson et al. 2012a)
portion of the micropile is cased with a PVC sleeve while the bottom portion has grout in direct contact with the soil. Depending on the method of construction and the fit between the PVC casing and the drilled hole, grout can flow up in the annulus between the PVC and the side of the hole as shown on Fig. 12. An all thread bar is typically used to reinforce the micropile and provide a means for attachment to the foundation.
[FIG. 12] - Schematic of typical micropile in expansive soil (after Schaut et al. 2011)
Pier Design Chart
An example of a pier design chart that was derived using the results of APEX analyses is shown in Fig. 11. Design charts of this nature can be developed using APEX and can be used to design micropiles. E
A is the modulus of
elasticity of the soil in bars.
[FIG. 11] - Pier heave versus pier length - nonlinear free-fi eld heave (after Nelson et al. 2012a)
MICROPILE TYPES AND TYPICAL CONSTRUCTION IN EXPANSIVE SOIL Micropiles have been classified into different types based on construction technique (FHWA, 2005 and AASHTO, 2012). A typical micropile installed in expansive soil has a configuration similar to that shown in Fig. 12. The upper
[40] DFI JOURNAL Vol. 7 No. 1 August 2013
Micropiles used in expansive soils typically consist of drilling a 4 to 6 inch (102 to 152 mm) hole using a hydraulic drill rig that mounts to the foundation as shown in Fig. 13. After drilling, casing made of rigid PVC pipe or other material is placed in the open hole in order to reduce the friction between the micropile and the surrounding expansive soil. Fig. 14 shows PVC casing placed in the drilled hole prior to grout placement. Micropiles are typically tremie grouted from the bottom of the hole which often allows the grout to flow up the inside of the PVC casing as well as into the annulus between the side of the hole and casing as discussed in Schaut et al. (2011). Soil swelling or worn cutting teeth on the auger bit may restrict flow of grout into the annulus between the soil and the casing. Fig. 15 shows a micropile during tremie grouting. After curing of the grout, the micropile is then connected to the bracket.
[FIG. 13] - Micropile drill rig bolted to grade beam foundation
[FIG. 14] - Micropile prior to grouting with annulus around the outside of casing
[FIG. 15] - Micropile during grouting. Note that the grout is tremied into the hole
CASE EXAMPLEAn interesting case example regarding the use of grouted micropiles in the remediation of distressed structures is the case of a single family home originally constructed on spread footings in Loveland, Colorado during the summer of 1995. After the original homeowners reported evidence of structural distress, local geotechnical and structural engineering firms were hired to investigate potential causes. Results of their investigations indicated that differential movement of the expansive soils beneath the residence had resulted in the basement and garage slabs being up to 7.0 in (178 mm) out of level, respectively. The respective reports recommended underpinning of the residence and a number of alternative underpinning methods were proposed including helical piers, push-pins, straight shaft piers and micropiles. A combination of steel push pins and helical piers were ultimately used to underpin the residence during the winter of 2001. The push pins and helical piers were recommended to be installed to a minimum depth of 30 ft (9 m). In 2010, continuing structural distress was observed by the homeowners and measured by performing inverted joist level surveys of the basement and garage. Figs. 16 and 17 illustrate some of the observed distress to the residence in 2010 after the initial underpinning.
DFI JOURNAL Vol. 7 No. 1 August 2013 [41]
[FIG. 16] - Diagonal Brick Cracking
[FIG. 17] - Diagonal Drywall Cracking
The lack of as-built information regarding the installation of the steel push-pins and the nature of the distress caused suspicion that the push-pins may not have been installed to the depths specified. To investigate, a geophysical survey was conducted by Zonge International, Inc. using a magnetic difference meter and a conductivity meter. In order to conduct the survey, a micropile drill rig was used to drill 4-in (102 mm) diameter holes to a depth of 35 ft (10.7 m) adjacent to three existing push-pins and one helical pier. The boreholes were cased with PVC pipe and the geophysical meters were inserted into the boreholes allowing data collection along the entire length of the boreholes. Fig. 18 shows the magnetometer/conductivity probe with the PVC-cased borehole in the background. Results of the geophysical survey clearly showed that the depth of the push-pin piers and helical piers ranged from 9 to 21 ft (2.7 to 6.4 m), significantly less than the depth specified. Due to the very hard state of the claystone beneath the residence it was
not possible to install the push-pins and helical piers to the depths specified. The micropile drill rig used in this investigation was secured to the foundation in order to drill in the hard claystone. The fact that the exploratory holes for the geophysical testing were drilled to a depth of 35 ft (10.7 m) using micropile technology shows that grouted micropiles could have been drilled and installed to the depths specified in the underpinning plans. In contrast, push pins and helical piers had not been able to be successfully installed to the depths specified, thereby significantly reducing their ability to resist uplift caused by heave of the expansive soils. This case study demonstrates the advantage of micropiles as compared to other underpinning options for use in hard expansive soil conditions.
[FIG. 18] - Magnetic Gradiometer, Testing Apparatus and Cased Hole
CONCLUSIONS The authors offer the following conclusions regarding the use of grouted micropiles for underpinning of foundations on expansive soils.
• Grouted micropiles began to find substantial use in the United States as far back as the 1970s. Since that time they have been used for a wide variety of applications including use as a structural element to underpin foundations constructed on expansive soils.
• The design of grouted micropiles in expansive soils is complex due to the use of low friction casing, large length to diameter ratios and typically complex soil and wetting profiles. The use of finite element based solutions can be used to model pier heave and tensile force if the free field
[42] DFI JOURNAL Vol. 7 No. 1 August 2013
heave and other input parameters are determined accurately.
• The depth and degree of wetting must be accurately determined for use in a formulation such as APEX to predict pier heave and tensile force in micropiles installed in expansive soil.
• Micropiles have distinct advantages as compared to alternative methods for underpinning of foundations on expansive soils. These advantages include ease of construction, ability to use the foundation as a reaction block on which to secure drilling equipment, ability to be installed in confined spaces and ability to be advanced to a specified design depth in stiff expansive soil.
REFERENCES1. American Association of State Highway and
Transportation Officials (AASHTO), 2012. Design specifications customary U.S. units. Publication Code LRFDUS-6.
2. Benvenga, M. M., 2005. "Pier-soil adhesion factor for drilled shaft piers in expansive soil", Master’s Thesis, Colorado State University, Fort Collins, Colorado.
3. Chen, F. H., 1965. "The use of piers to prevent the uplift of lightly loaded structures founded on expansive soils", Proceedings of the International Conference on Expansive Soils, College Station, Texas.
4. Chen, F. H., 1988. Foundations on expansive soils. Elsevier. New York, NY.
5. Federal Highway Administration (FHWA), 2005. Micropile design and construction reference manual. Publication No. NHI-05-039, NHI Course No. 132078, U.S. Dept. of Transportation. December.
6. Fredlund, D. G. and Rahardjo, H., 1993. Soil mechanics for unsaturated soils. John Wiley & Sons, New York, NY.
7. Fredlund, D. G., Hasan, J. U., and Filson, H., 1980. "The prediction of total heave", Proceedings 4th International Conference on Expansive Soils, Denver, Colorado, June 16-18, pp. 1-11.
8. Fredlund, D. G., Rahardjo, H., and Fredlund, M. D., 2012. Unsaturated soil mechanics in engineering practice. John Wiley & Sons, Hoboken, NJ.
9. Justo, J. L., Saura, J., Rodriguez, J. E., Delgado, A., and Jaramillo, A., 1984. "A finite element method to design and calculate pier foundations in expansive-collapsing soils", Proceedings of the 5th International Conference on Expansive Soils, Adelaide, Australia, pp. 199-123.
10. McWhorter, D. B. and Nelson, J. D., 1979. "Unsaturated flow beneath tailings impoundments", Journal. Geotechnical and Engineering Division, ASCE, November, Vol. 105(GT11), pp. 1317-1334.
11. Nelson, J. D. and Miller, D. J., 1992. Expansive soils: problems and practice in foundation and pavement design. John Wiley and Sons, New York, NY.
12. Nelson, J. D., Chao, K. C., Overton, D. D., and Schaut, R. W., 2012a. "Calculation of heave of deep pier foundations", Geotechnical Engineering Journal of the Southeast Asian Geotechnical Society and Association of Geotechnical Societies in Southeast Asia, Vol. 43, No. 1, pp. 12-25.
13. Nelson, J. D., Overton, D. D., and Durkee, D. B., 2001. "Depth of wetting and the active zone", Proceedings of the Geo-Institute Shallow Foundation and Soil Properties Committee Sessions at the ASCE Civil Engineering Conference 2001, Houston, Texas, October 10-13, in “Expansive Clay Soils and Vegetative Influence on Shallow Foundations” ed. C. Vipulanandan, M.B. Addison, and M. Hansen, GSP115, pp. 95-109.
14. Nelson, J. D., Reichler, D. K., and Cumbers, J. M., 2006. "Parameters for heave prediction by oedometer tests", Proceedings of the 4th International Conference on Unsaturated Soils. Carefree, Arizona. April, GSP 147, pp. 951-961.
15. Nelson, J. D., Thompson, E. G., Schaut, R. W., Chao, K. C., Overton, D. D., and Dunham-Friel, J. S., 2012b. "Design considerations for piers in expansive soils", Journal of Geotechnical and Geoenvironmental Engineering, ASCE, Vol. 138, No. 8, pp. 945-956.
DFI JOURNAL Vol. 7 No. 1 August 2013 [43]
16. O’Neill, M. W., 1988. "Special topics in foundations", Proceedings of the Geotechnical Engineering Division National Convention, ASCE, Nashville, Tennessee, pp. 1-22.
17. Poulos, H. G. and Davis, E. H., 1980. Pile foundation analysis and design. John Wiley, New York, NY.
18. Reichler, D. K., 1997. "Investigation of variation in swelling pressure values for an expansive soil", Master’s Thesis, Colorado State University, Fort Collins, Colorado.
19. Schaut, R. W., Nelson, J. D., Overton, D. D., Carraro, J. A. H., and Fox, Z. P., 2011. "Interface testing for the design of micropiles in expansive soils", Proceedings of the 36th Annual Conference on Deep Foundations, Boston, MA, Oct 18-21, Deep Foundations Institute.
20. Sorochan, E. A., 1991. Construction of buildings on expansive soils. A.A. Balkema Publishers, Brookfield, VT.
21. U.S. Army Corps of Engineers, 1983. Technical manual TM 5-818-7, foundations in expansive soils. Washington, DC., September 1.
[44] DFI JOURNAL Vol. 7 No. 1 August 2013
Relationship between Installation Torque and Axial Capacities of Helical Piles in Cohesive SoilsMohammed Sakr, PhD., P.Eng, Manager, Geotechnical Engineering, Vertex, Sherwood Park, AB Canada; Ph: (780) 920-0652; [email protected]
ABSTRACTThe empirical relationship between torque measured during installation and pile capacity has gained a wide acceptance in the helical pile industry in the last few decades. This paper presents a theoretical model developed to estimate torsional resistance of cohesive soils to helical pile installation. The proposed torsional resistance model was then used to establish torque factor, K
t, a factor that is widely
used in the industry. The results of the study indicated that the Kt factor is a function of the load path
(i.e. tension or compression), pile geometry and soil properties. The assessed Kt factors in tension
and compression were validated by the results of 74 installation records and full-scale helical pile load tests. A parametric study was also performed to qualitatively assess the relative importance of different parameters that affect the K
t factors.
INTRODUCTIONEmpirical relationship between measured torque during installation and helical pile capacity is widely used in the industry in North America especially for small-size helical piles. The empirical relationship can be expressed as (Hoyt and Clemence, 1989; Canadian Foundation Engineering Manual (CFEM) 2006, Perko 2009).
Qt = K
tT [1]
where
Qt = ultimate capacity of screw pile;
Kt = empirical factor; and
T = average installation torque.
Torque-load correlation factors, Kt, were
statistically established based on a large database, and the method has been used successfully in the installation of thousands of piles and anchors over the past two decades, as indicated by Hoyt et al. (1995). However, the obtained K
t values did not differentiate between
compression and tension loading. For example, Hoyt and Clemence proposed a K
t factor equal
to 9.8 m-1 (32.2 ft-1) for 89 mm (3.5 in) diameter round shaft helical piles regardless of load path (i.e. compression or tension). The K
t
values published in literature and adapted in the Canadian Engineering Foundation Manual (CFEM; 2006) were mainly based on the results of pullout (tension) tests for small diameter piles. Therefore, the available K
t values may be
not suitable for estimating the axial capacities
of helical piles under compressive loads or for large diameter helical piles.
Perko (2001) proposed a correlation between installation torque and pile capacity based on an energy model similar to that model for driven piles. However, the main limitation to the energy model is that it requires numerous parameters, some of which are not easily measurable during pile installation.
Perko (2009) proposed another empirical relationship between K
t and effective
shaft diameter (deff
) based on exponential regression analysis of over 300 load tests in both compression and tension. The empirical equation can be expressed as:
92.0eff
kt dK
λ= [2]
where
kλ = fitting factor equal to 1433 mm0.92/m
deff
= shaft diameter for round shaft, mm.
It should be noted that in Eqn. 2, torque factor is inversely proportional to pile shaft diameter. However, the main limitations to Eqn. 2 are the lack of explanation of the physical meaning of fitting factor and the estimated K
t values in
compression and tension were identical.
Despite the effort over decades to empirically correlate between installation torque and pile capacity, a comprehensive relationship has not been attainable. Moreover, some practitioners
DFI JOURNAL Vol. 7 No. 1 August 2013 [45]
have misgivings about determining the capacity of helical piles using only torque measurements, without taking into consideration geotechnical parameters (Cannon, 2000). There are a number of factors that affect installation torque such as pile configuration, soil conditions, operator skill level, and accuracy of measurements. Pile configuration such as shaft size and shape, number of helices, diameter of helix, and pitch size are some parameters that affect torque measurements. The presence of cobbles or boulders during installation results in a sharp rise in torque values, which is not necessarily an indication of better soil conditions. The use of empirical torque correlations is viewed with reservation by some engineers, who see the dependency of the procedure adopted by the installer on the results (Beim and Luna, 2012). Installation procedures such as applying down-pressure (crowd) on the pile, use of predrilling process and speed of rotary head are other factors that impact torque measurements. Methods of measuring torque using either a differential hydraulic pressure measured using mechanical devices or using an electronic load cell attached to the pile head may also impact the torque measurements. The frequency of calibrating torque measurement devices is another parameter that affects the quality and reliability of torque measurements.
The main objective of the paper is to propose a comprehensive theoretical model to estimate torque factors for helical piles installed into cohesive soils. The proposed torque factors can be used to assess torque requirements for selecting the suitable equipment for installation. They can be also used as a quality control measure for production piles to accept or reject installed piles. Moreover, the proposed torque factors can be used to approximately assess axial pile capacities in tension and compression. Other objectives of the study are to evaluate the effect of different parameters on torque factors, K
t and to assess their relative importance.
THEORETICAL MODELHelical piles are typically installed through the use of mechanical torque applied at the pile head with a rotary hydraulic head. Fig. 1 shows a typical installation of helical piles. Torque measured during pile installation is a function of numerous factors that includes pile configuration, soil conditions, method of
[FIG. 1] Typical Helical Pile Installation
installation, operator experience, and accuracy of measurements. Pile configuration such as shaft diameter, shape of pile shaft, number of helices, diameter of the helices, and pitch are measurable values and can be included in the theoretical torque model. Soil conditions and groundwater level have a considerable effect on pile installation and can be reasonably evaluated at the geotechnical investigation stage. However, there are other non-measurable parameters that affect measured torque values such as installation procedure, method of torque measurements, and accuracy of torque measuring devices. Methods of installation are highly dependent on the manufacturer’s installation specifications, availability of equipment, operators’ experience, and speed of installation. The accuracy of measuring torque devices depends on the method used for torque reading, such as differential pressures or using strain gauges at pile head. The reliability of torque measurements depends on the frequency of equipment calibration. These factors are difficult to quantify and do not affect the torsional resistance of soils to pile installation, but they adversely affect the measurements of torque values. Therefore, installation procedures and quality of measurements will not be considered for the development of a theoretical torque model.
[46] DFI JOURNAL Vol. 7 No. 1 August 2013
The theoretical model assumes that the exerted torque during helical pile installation into cohesive soils is resisted by torsional shear along the pile shaft and torsional shear along the helices, as shown in Fig. 2. The main assumptions that are considered for the development of the proposed torque model include the following:
1. Down pressure force (crowd) applied on the pile during installation is neglected.
2. Torsional shear along the pile shaft is equal to axial unit shaft friction.
3. The soil layer is assumed to be a homogenous layer that extends to infinite depth.
4. Resisting torque during pile installation is independent of the speed of the robust hydraulic head.
5. The pile is advanced into the soil at a constant penetration rate equal to the pitch, and soil disturbance is minimal.
6. Helices are a true spiral shape, and their projected area is equal to the size of a disk with a diameter equal to the helix diameter.
Therefore, the exerted torque during pile installation may be given by the following expression:
∑+=N
his TTT1
[3]
where:
Ts = torsional moment acting on pile shaft,
(kN.m)
Thi = torsional moment acting on helix i, (kN.m)
N = number of helices
In the present model, the torsional resisting moment of the pile shaft is a function of the shaft resistance and can be given by the following equation:
2dQ
T ss =
[4]
where:
d = shaft diameter, (m) for round shaft piles and equivalent diameter for square shaft piles.
Assuming a homogenous soil layer, the shaft friction resistance of the helical pile may be given by:
ss LfdQ π= [5]
[FIG. 2] Torsional Moments during Pile Installation
where:
L = embedded pile length, (m)
fs = unit shaft friction, (kPa).
Therefore, the torsional resistance of pile shaft to installation can be given by:
2
2s
sLfd
Tπ
= [6]
Eqn. 6 assumes that the torsional stress on the pile-soil interface reaches a limiting value equal to the pile-soil unit shaft friction (Basile, 2010). The unit shaft friction for piles installed in cohesive soils can be given from the following equation (CFEM 2006):
urs Cf α= [7]
where:
α = adhesion factor, (m)
Cur
= remoulded undrained shear strength, (kPa).
Adhesion factor, α is given by:
u
a
Cp26.0
21.0 +=α ≤ 1 [8]
where:
pa = atmospheric pressure, (101 kPa).
Therefore Eqn. [6] can be rewritten as:
Ts=πd 2Lα
jC
uj
2 [9]
DFI JOURNAL Vol. 7 No. 1 August 2013 [47]
The torsional resisting moment on a helix during pile installation, as indicated in Fig. 3, is analogous to a giant vane shear test with a height equal to the pitch of helix, p, and diameter equal to the diameter of helix, D. Therefore, the torque required to shear the soil surrounding the helix can be estimated as the torque required to shear a cylinder of soil with height equal to the pitch of helix, p, and diameter equal to the diameter of the helix, D (Fig. 3). The torque required to separate the cylinder of soil is proportional to the undrained shear strength of the clay (C
u). The model
assumes that installation is consistent with minimal soil disturbance and that the pile is advanced into the soil at a constant rate equal to pitch per full revolution of drive head.
The torque required to separate a helix i is the sum of the torsional moment at the upper surface of the helix, T
ti; torsional moment due
to the resistance of the separated cylinder, Tci;
and the moment at the bottom of helix, Tbi; as
shown on the following equation:
bicitihi TTTT ++= [10]
Torque due to shearing the cylinder of soil around a helix i can be expressed as:
uiiii
uici CpD
daD
CT22
2
π== ∫ [11]
where
Cui = undrained shear strength at helix level i, (kPa).
D = helix diameter, m.
Torque due to shearing resistance at the top of the soil cylinder can be expressed as:
utiii
D
d utiti CdD
rdaCTi
12)( 33
2
2
−== ∫ π [12]
whereC
uti = undrained shear strength at top of helix
level i, (kPa).
Di = helix i diameter, m.
Torque due to shearing resistance at the bottom of the soil cylinder can be expressed as:
ubiii
D
d ubibi CdD
rdaCTi
12)( 33
2
2
−== ∫ π [13]
where
Cubi
= undrained shear strength at bottom surface of helix i, (kPa).
[FIG. 3] Torsional Resistance Model of a Helix during Pile Installation
Therefore, resisting moment acting on helix i can be expressed as:
2/)(12
)( 233
uiiiubiutiii
i CpDCCdD
T ππ ++−
= [14]
Assuming that soil around the helix i is homogenous, (i.e. C
ui = C
uti = C
ubi), Equation [14]
can be rewritten as:
⎟⎟⎠
⎞⎜⎜⎝
⎛ −+=
6)(
2
332iiii
uiidDpD
CT π [15]
It should be noted that, since the installation of a helical pile typically requires a relatively large number of revolutions to install the pile to final depth, remoulded undrained shear strength values are suggested for use in Equation [15]. However, for the bottom helix, the intact undrained shear strength values (peak values) are suggested for use. For the bottom helix, the resisting torsional moment can be expressed as:
⎟⎟⎠
⎞⎜⎜⎝
⎛ −++=
12)(
122
31
31
311
21
11dDDpD
CT uπ [16]
[48] DFI JOURNAL Vol. 7 No. 1 August 2013
MODEL VERIFICATIONIn order to verify the proposed theoretical torque model, the measured torque values versus depth for a helical pile with double helices, ST72, reported by Sakr (2012b) are presented in Fig, 4. Pile ST72 had a shaft diameter of 406 mm (16 in), with two helices, 813 mm (32 in) in diameter spaced at 1.63 m (5.35 ft) (i.e. two times helix diameter). The pitch for both helices was 152 mm (6 in). The side friction from Cone Test Penetration Test (CPT) sounding at the test location is also presented in Fig. 4. The main advantage of a CPT test is that it provides a near-continuous soil profile and, therefore, the data is of great importance for verification of the proposed torque model.
Soil properties at the test site location, as interpreted from CPT data and summarized in Table 1, consisted of surficial sandy silt, to a depth of about 1.9 m (6.2 ft) below existing grade, over a stiff to very stiff silty clay layer, to a depth of about 13.7 m (45 ft) , underlain by a hard silty clay layer that extended to the end of sounding at 16.4 m (53.8 ft). The estimated undrained shear strength, Cu, for the stiff and very stiff silt clay were 80 kPa and 115 kPa (11.6 psi and 16.7 psi), respectively. The estimated undrained shear strength of the lower hard silty clay layer was 180 kPa (26 psi). Based on the results of the CPT sounding, soil properties for each soil layer were relatively consistent with the exception of few peaks, such as at a depth of about 14.6 m (48 ft), where an abrupt increase in sleeve friction was observed.
The measured torque values during pile ST72 installation are also presented in Fig. 4. The torque values were measured using differential pressures displayed on mechanical gauges. The estimated torque values using equations [3], [9]; [15] and [16] at different depths for pile ST72 are also presented in Fig. 4. The following observations were made based on comparing between measured torque values at pile head and estimated torsional resistance of pile ST72:
1. Measured torque at the pile head increased considerably as the upper helix advanced into the ground. The estimated torsional resistance of the soil followed a similar trend to the measured values. As expected when the bottom helix travels through different soil layers, the torque value shows an abrupt change to reflect the properties of the soil layer at the bottom helix level.
2. When both helices travel through the same soil layer, torsional resistance increases
0 100 200 300 400 500 600 700
0
2
4
6
8
10
12
14
16
18
50.00 100.00 150.00 200.00 250.00 300.00 350.00
Cone side shear, kPa
Dep
th, m
Torque, kN-m
ST72 Torque estimated CPT Sleeve Frictiion
Dense sand φd 33°, γ=19 kN/m3
Avg Torque 88 kN-m
Hard silty clay, cu 180kPa, γ=19 kN/m3
Avg Torque 296 kN-m
Very stiff silty clay,cu 115kPa, γ=18 kN/m3
Avg Torque 233 kN-m
Stiff silty clay, cu 80kPa, γ=18 kN/m3
Avg Torque 160 kN-m6.9
13.7
10.8
1.9
[FIG. 4] Comparison between Measured Torque and Estimated Torsional Resistance of Pile ST72 during Installation
[TABLE 1] Summary of Soil Properties
Depth
mSoil description
Total unit weight, kN/m3
Undrained Shear Strength, kPa
Frictional resistance angle,
φ (o)
0 – 1.9 Sand, compact 18.5 - 33
1.9 – 6.9 Glacial Till, stiff 18 80 0
6.9 – 13.7 Glacial Till, very stiff 18 115 0
13.7 – 16.4 Glacial Till, very stiff to hard 19 180 0
DFI JOURNAL Vol. 7 No. 1 August 2013 [49]
linearly due to the increase of resisting moment of the shaft.
3. Torque measured up to a depth of about 8 m (26.2 ft) was considerably higher than the estimated torque.
4. The slope of the measured and estimated torque between depths of about 9 m and 13 m (29.5 ft and 42.7 ft) was similar.
5. In general, the measured and estimated installation torque values both agreed reasonably.
6. The estimated torsional resistance at the end of pile installation agreed reasonably with the measured values.
7. Spikes were observed from the CPT data (sleeve friction) indicating that the soil layers were not truly homogenous.
The comparison between measured and estimated torque values at different levels indicated that both estimated and measured values generally followed similar trend and agreed reasonably. It should be noted that remoulded undrained shear strength values were used for the torque estimate. The estimated lower torque values up to depth of 9 m (29.5 ft) could be as a result of using remoulded shear strength as opposed to intact undrained shear strength of native materials.
COMPARISON BETWEEN MEASURED AND ESTIMATED TORQUE VALUESAs indicated in Eqns. [3]; [9] and [15], the estimated torsional soil resistance values at the end of installation are a function of shaft size (i.e. either diameter or width of square shaft helical piles); helix diameter; pitch; and number of helices. The estimated torsional resistance is also a function of soil strength parameters, including undrained shear strength, adhesion and soil sensitivity. Moreover, as indicated on the model assumption, it is assumed that the pile is advanced into the soil at a constant rate equivalent to the pitch. Therefore, operator experience and pile installation consistency affect the measured torque values during pile installation. Auguring effect, where the pile is advanced at a smaller rate less than pitch size, may cause additional soil disturbance and reduce the measured torque during installation. Sensitivity of the torque reader and accuracy and frequency of calibrating the
device are other factors that affect the quality of measurements.
In order to compare between measured and estimated torque values, a total of 74 installation tests were considered in the present study. Pile configuration and a summary of soil parameters are presented elsewhere (Zhang 1999; Tappenden 2007; Livneh and El Nagger 2008; Cerato and Victor 2009; Sakr 2008, 2011, 2012a and 2012b; Beim and Luna 2012). The selected installation cases were selected to satisfy the following conditions:
1. Pile configuration, including shaft sizes and helix diameters, cover a wide range. For example, shaft sizes varied between square shafts 44 mm (1.73 in) in width to round shafts with diameters up to 508 mm (20 in). Helix diameters varied between 203 mm (8 in) and 1016 mm (40 in). The number of helices varied between 1 and 4. Pitch was either 76 mm or 152 mm (3 in or 6 in).
2. Selected piles were also installed by different operators so that the data reflect the variability in installation procedures for different contractors.
3. Installation technique included standard installation and use of predrilled pilot holes.
4. Soils considered for the study varied between soft clays to clay shale with undrained shear strength, C
u, that varied
between 5 kPa and 400 kPa (0.73 psi and 58 psi) (very soft to very hard clay materials).
The measured and estimated torque values for different piles were compared in Fig. 5. The prediction ratio, which is the ratio between estimated and measured torque values, varied between 0.78 and 1.29. The data were also linearly fitted with a standard deviation of 0.96. Therefore, the proposed torque model reasonably estimated installation torque for helical piles considered in the study.
RELATIONSHIP BETWEEN AXIAL TENSILE CAPACITY AND INSTALLATION TORQUEThe empirical torque-capacity relationship, expressed in Eqn. [1], assumes a proportional relationship between measured torque at the end of installation and axial pile capacity by torque factor. To separate different torque factors (i.e. compression versus tension); torque factor in tension, K
t, is defined as the ratio
[50] DFI JOURNAL Vol. 7 No. 1 August 2013
between axial tensile capacity of the pile and the measured torque at the end of installation. Therefore, K
t, can be expressed as:
thtshst
t KKTQ
TQ
TQ
K +=+== [17]
Eqn. [17] suggests that the torque factor in tension, K
t, can be uncoupled into two
components including torque factor due to shaft, K
ts, and torque factor due to helices, K
th.
For predicting the ultimate uplift resistance of a helix i installed into cohesive soils considering the individual helix capacity method (i.e. neglecting the interaction factors between different helices), the following expression may be used (Das and Seeley, 1975):
D + NC = D + Q =Q 11ui
N
111
N
1hih hHuiHihH AAA γγ ′′ ∑∑ [18]
9)(2.1 ≤=DDN h
U [19]
where:
γ’ = Average effective unit weight of soil above the top helix, (kN/m3);
Qhi = ultimate helix i resistance, (kN);
Cui = undrained shear strength of soil layer at
helix i, kPa
Dhi = depth to helix i, (m);
Di = diameter of helix i, (m);
Nui = Dimensionless uplift bearing capacity
factor for helix i;
0
40
80
120
160
200
240
280
320
360
0 40 80 120 160 200 240 280 320 360
Est
imat
ed T
orqu
e (k
N.m
)
Measured Torque (kN.m)
CompressionTension
Equity Line
[FIG. 5] Comparison between Measured and Estimated Torque Values
AHi
= net surface area of bearing helix (helix area – shaft area), m2; and
N is the number of helices.
For round shaft piles, the net surface area of
helix i, )4
(22 dD
A hiHi
−=π where d = shaft
diameter, (m). Therefore the torque factor due to shaft, K
ts can be estimated as:
⎟⎟⎠
⎞⎜⎜⎝
⎛⎟⎟⎠
⎞⎜⎜⎝
⎛ −+++⎟⎟
⎠
⎞⎜⎜⎝
⎛ −++
==
∑ = 12)(
1226)(
22
131
31
311
21
12
332 dDDpDC
dDpDC
QdT
QK
uN
iiiii
uis
sts
π [20]
It should be noted that neglecting the interaction between different helices is a reasonable assumption for most cases where the spacing between helices is equal to or greater than three times the helix diameter. However, for the cases where spacing between helices is less reduction of helix capacities due to interaction should be considered. Therefore, the torque factor due to helices can be expressed as following:
( )
⎟⎟⎠
⎞⎜⎜⎝
⎛⎟⎟⎠
⎞⎜⎜⎝
⎛ −+++⎟
⎟⎠
⎞⎜⎜⎝
⎛ −++
+==
∑
∑
=
=
12)(
1226)(
22
31
31
311
21
12
332
11
/
dDDpDC
dDpDC
dQ
DNCA
TQ
K
uN
iiiii
uis
N
nhuiuiHi
hth
ππ
γ
[21]
RELATIONSHIP BETWEEN AXIAL COMPRESSIVE CAPACITY AND INSTALLATION TORQUESimilar to torque-factor in tension, torque factor in compression K
c, defined as the ratio between
the axial compressive capacity and measured torque at the completion of pile installation can be expressed as the sum of torque factor due to the shaft, K
cs and torque factor due to helices, K
ch.
For simplicity, the frictional resistance of pile shaft in tension may be assumed similar to that value in compression. Hence, Eqn. [20] can be used to estimate torque factor due to shaft resistance, K
cs.
For the case of compressive loading for helical piles founded in cohesive soils, the axial compressive resistance of a helix i can be estimated as follows:
Qchi
= AHi
Cui N
ci [22]
where
Nci = dimensionless bearing capacity factor for
helix i;
(Nc = 9 for helices smaller than 0.5m; N
c = 7 for
helices between 0.5 m (20 in); and Nc = 6 for
helixes larger than 1 m (40 in) in diameter)
DFI JOURNAL Vol. 7 No. 1 August 2013 [51]
Tables 2 and 3. The measured torque factors were based on the results of full-scale axial compressive and tension tests reported in literature. A total of twenty-one axial tensile load tests and fifty-three axial compression tests were used to assess the tension and compression torque factors (K
t and K
c). The
theoretical torque factors in tension and compression were estimated using Eqns. [20], [21] and [25].
The comparison between measured and estimated torque factors in tension and compression are also presented in Figs. 6 and 7. It can be seen that there is a reasonable agreement between measured and estimated torque factors both in tension and compression. However measured K
c values for square
shaft piles were generally lower than the estimated values. Possible reason for that is that for square shaft piles, soil disturbance is considerably higher compared to round shaft piles due to the rotation of the square shaft
For round shaft piles, the ultimate compressive resistance for helix i can be expressed as:
ciuihichi NCdDQ )(4
22 −=π
[23]
For round shaft piles, the ultimate compressive resistance of the bottom helix 1 can be expressed as:
11211 4 cubhch NCDQ π
= [24]
The torque factor in compression for helices can be expressed as:
⎟⎟⎠
⎞⎜⎜⎝
⎛⎟⎟⎠
⎞⎜⎜⎝
⎛ −+++⎟⎟
⎠
⎞⎜⎜⎝
⎛ −++
⎟⎠
⎞⎜⎝
⎛−+
==
∑
∑
=
=
12)(
1226)(
22
)(41
31
31
311
21
12
332
2
2211
21
dDDpDC
dDpDC
dQ
NCdDNCD
TQ
K
uN
iiiii
uis
N
nciuihicubh
chch
π
[25]
COMPARISON BETWEEN THEORETICAL AND MEASURED TORQUE FACTORSMeasured and estimated torque factors in tension and compression are presented in
[TABLE 2] Comparison between Measured and Estimated Torque Factors in Tension, Kt.
Pile ID Soil TypeShaft
Diameter
m
No of helices
Theoretical K
t
m-1
Measured K
t
m-1
Reference Notes
T1 Silty Clay 0.089 1 18.2 12.7 Sakr (2011)
ST62 Clay Till 0.406 2 4.8 4.8 Sakr (2012a) predrilled holeST72 0.406 2 5.8 6.2
ST14 Clay Shale 0.406 1 4.4 5 Sakr (2012a) predrilled holeST5 0.324 2 6.3 5.7
TL Clay Till 0.219 3 8.4 9.9 Zhang (1999)TS 0.219 3 9.3 7.6
Tprod 0.219 2 9.8 9.2
T7 Clay Till 0.273 1 8.8 9.8 Tappenden (2007)T8 Clay Till 0.273 2 9.3 10.9
7 Clay Till 0.0445 3 25.6 21.3Livneh and El Naggar
(2008)8 0.0445 3 26 26.5
16 0.0445 3 25.5 23.6
20 0.0445 3 24.7 29.2
10 0.0445 3 23.9 24.7
T1 Clay 0.219 3 7.7 7 Sakr (2008)
T2 Clay Till 0.273 2 7.7 10.7
T1 0.273 1 8.5 9.9
Clay Till 0.508 1 4.9 2.6 Sakr (2012b)
Clay Till 0.508 2 3.8 1.9
Clay Till 0.508 2 4.4 5.2
[52] DFI JOURNAL Vol. 7 No. 1 August 2013
[TABLE 3] Comparison between Measured and Estimated Torque Factors in Compression
Pile ID Soil TypeShaft
Diameter
m
No of helices
Theoretical K
t
m-1
Measured K
t
m-1
Reference Notes
Silty Clay 0.089 1 19.2 19.8 Sakr (2011)
ST61 Clay Till 0.406 2 6.1 5.5 Sakr (2012a) predrilled holeST15 Clay Shale 0.508 1 5.2 7.1
ST14 Clay Shale 0.406 1 6.3 6.8
ST7 Clay Shale 0.406 1 7.7 7.3
Clay Till 0.219 3 14.4 9.2 Zhang (1999)0.219 3 10.9 7.6
0.219 2 10.9 10.8
C7 Clay 0.178 1 9.5 11.8 Tappenden (2007)C8 Clay 0.219 1 9 9.9
C9 Clay 0.178 2 9.7 12.8
C10 Clay Shale 0.24 2 9.8 14.1
C11 Clay Till 0.273 1 12.1 8.9
C12 Clay Till 0.273 2 10.1 8
C15 Clay 0.14 3 13.7 13.7
C16 Clay 0.114 2 15.4 18.1
C17 Clay 0.114 1 15.9 21.1
2 Clay Till 0.0445 3 28.6 42.6 Livneh and El Naggar
(2008)
Square shaft
6 0.0445 3 25.5 37.8
Clay 0.324 1 6.3 5.6 Sakr (2012b)
Clay 0.324 1 6.3 5.3
Clay 0.324 1 6.3 6
Clay 0.324 3 7.8 7.1
Clay 0.324 2 7.7 7.9
Clay 0.324 2 8.3 9.3
Clay Till 0.324 4 6.5 6.1
Clay Till 0.324 3 6.3 6.1
Clay Till 0.324 2 7.6 7.8
Clay Till 0.324 2 7.3 6.8
C1 Clay 0.324 3 5.9 7.1 Sakr (2008)
C2 0.324 4 6.9 7.6
C2 Clay Till 0.273 2 8 14.1 Sakr (2012b)
C1 0.273 1 8.1 12.1
0.324 1 6.7 5.5
0.406 1 5.4 5.6
0.508 1 4.4 4.2
0.508 1 4.6 4.4
0.508 2 5 5.3
0.508 3 5.5 6
0.508 2 6.2 6
DFI JOURNAL Vol. 7 No. 1 August 2013 [53]
and causing void around pile shaft during installation. The mean prediction ratio for K
c is 1.06 with a coefficient of variation of
17.8%, while the mean prediction ratio for Kt
is 0.98 with a coefficient of variation of 17.2%. Generally, torque factors in tension were more predictable than torque factors in compression. This can be explained by the fact that during helical pile installation, torque measured at the end of installation can be considered to be representing average soil conditions within pile embedment depth. However, for the case of torque factor in compression, there are no torque measurements within the soil layer immediately below the bottom helix (which considerably affects the performance of helical
piles in compression), and therefore torque factor in compression may not be accurate.
It can be seen from Tables 2 and 3 that torque factors in tension were generally lower than torque factors in compression. For example, pile ST61 and ST62 with similar configurations, measured torque factors in tension and compression were 4.8 and 5.5, respectively. It can be also seen from Tables 2 and 3 that torque factors for square shaft piles were considerably higher than those for cylindrical shafts. Discrepancies between measured and estimated torque factors were also higher for square shaft piles.
Pile ID Soil TypeShaft
Diameter
m
No of helices
Theoretical K
t
m-1
Measured K
t
m-1
Reference Notes
0.406 1 5.3 4.6
0.508 2 4.4 4.6
0.508 2 4.6 5.7
HP5 Varved Clay 0.073 3 26.7 54.1 Beim and Luna (2012)HP10 0.073 3 26.7 46.6
HP15 0.073 3 26.7 44.3
HP4 0.073 3 32.1 33.3
HP7 0.073 3 32.1 31.9
HP9 0.073 3 32.1 40.4
HP12 0.073 3 32.1 43.3
HP14 0.073 3 32.1 37.7
[TABLE 3] Comparison between Measured and Estimated Torque Factors in Compression (continued)
y = 0.9955xR² = 0.9234
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
0 5 10 15 20 25 30 35 40
Est
imat
ed T
orqu
e Fa
ctor
, Kt(m
-1)
Measured Torque Factor, Kt (m-1)
Measured and Estimated TorqueEquity LineLinear (Measured and Estimated Torque)
[FIG. 6] Comparison between Measured and Estimated Torque Factors in Tension
[FIG. 7] Comparison between Measured and Estimated Torque Factors in Compression
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
45.00
50.00
55.00
0 5 10 15 20 25 30 35 40 45 50 55
Est
imate
d T
orq
ue F
act
or,
Kc
(m-1
)
Measured Torque factor, Kc (m-1)
Measured and Estimated Torque Equity Line
[54] DFI JOURNAL Vol. 7 No. 1 August 2013
PARAMETERS INFLUENCING TORQUE FACTORSParameters affecting torque factors in cohesive soils can be grouped into five main groups:
1. soil properties such as undrained shear strength, adhesion and sensitivity;
2. pile configuration (i.e. shaft diameter, helix diameter, pitch, number of helices and embedment depth);
3. installation procedure;
4. reliability of torque measurements; and
5. loading path (i.e. tension or compression).
Parameters 1 to 4 are discussed in more detail in the following sections, while the last parameter has been discussed earlier.
Soil Properties
In order to evaluate the effect of undrained shear strength on torque factors, a hypothetical pile configuration was assumed, consisting of a round shaft pile, 0.324 m (12.75 in) in diameter with a single-helix diameter of 0.762 m (30 in) in diameter. Estimated torque factors in tension, K
t versus embedment depth ratio (i.e.
embedment depth divided by helix diameter) for cohesive soils with undrained shear strength are presented in Fig. 8. A homogeneous soil layer with undrained shear strengths that varied between 25 kPa (3.64 psi) ( soft clay) and 400 kPa (58 psi) (very hard clay) was assumed. It can be seen from Figure 8 that at shallow embedment depths, up to depth of about 4.5 D, the softer soils showed lower K
t factors
compared to harder soils. At deeper embedment depths, the harder clay soils showed relatively
high Kt factors. However, the increase in
torque factor Kt due to increasing undrained
shear strength was relatively insignificant. For example, the torque factor for a pile installed into very hard clay (C
u = 400 kPa or 58 psi) at
an embedment depth of about 8 helix diameters were about 7 m-1 (23 ft-1) compared to 6 m-1 (19.7 ft-1) for a pile with a similar configuration installed in soft clay (C
u = 25 kPa or 3.64 psi)
Adhesion around a pile shaft is typically proportional to the undrained shear strength, as indicated in Eqn. [7]. The adhesion factor expressed in Eqn. [8] is inversely proportional to the undrained shear strength of soils. For soft clay, the adhesion around a pile shaft is equivalent to the undrained shear strength value.
Sensitivity of soils is another factor that affects the torsional resistance of soils to installation and torque factors. For sensitive clays where the ratio between undisturbed shear strength and remolded shear strength is high, the effect of soil disturbance on pile installation is expected to be high. Therefore, the predicted torsional soil resistance to installation using undisturbed undrained shear strength is likely to be in error.
Pile Configurations
To evaluate the effect of increasing shaft diameter on torque factor in tension, K
t, Eqns.
[20] and [21] were used to estimate torque factors for helical pile with a single-helix, 0.4 m (1.3 ft) in diameter, installed into cohesive soil with undrained shear strength C
u = 50 kPa
(7.25 psi). The shaft diameters were 89 mm (3.5 in), 178 mm (7 in) and 273 mm (10.75 in) respectively. The torque factors in tension, K
t,
versus embedment depth ratios are presented in Fig. 9. It can be seen from Fig. 9 that, as expected, increasing the shaft diameter resulted in considerably reducing torque factors. For example, at an embedment depth of 7.5 D, the torque factors for piles with shaft diameters of 89 mm, 178 mm and 273 mm (3.5 in, 7.0 in and 10.75 in) were 17 m-1, 12 m-1 and 8 m-1 (55.8 ft-1, 39.4 ft-1,and 26.2 ft-1) respectively.
In order to evaluate the effect of varying helix diameters on the torque factors in tension, K
t, the torque factors were estimated for a
pile with a shaft diameter of 406 mm (16 in) equipped with a single helix installed into clay material with undrained shear strength of 50 kPa (7.25 psi). Helix diameters of 0.763 m
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
0 2 4 6 8 10 12 14 16 18 20
Estim
ated
Tor
que
Fact
or, K
t(m
-1)
Embedment Depth Ratio (H/D)
Cu = 50 kPa100 kPa200 kPa400 kPa25 kPa
[FIG, 8] Effect of Undrained Shear Strength on Torque Factors in Tension, Kt.
DFI JOURNAL Vol. 7 No. 1 August 2013 [55]
(30 in) and 1.2 m (48 in) were considered for the comparison, and the results are presented in Fig. 10. As seen in Fig. 10, increasing helix diameter resulted in slightly increasing K
t.
Fig. 11 shows piles with helix and shaft diameter ratios of 2, 2.5 and 3. As seen in Fig. 11, increasing the ratio between D/d resulted in considerably increasing torque factors. For example, K
t for piles with D/d ratios
of 2, 2.5 and 3 at embedment depth of 7.5 D were about 6.5, 7.5 and 9.0 respectively.
Torque factors in tension Kt for piles with a
shaft diameter of 0.324 m (12.75 in), single helix and pitch of either 76 mm or 152 mm (3 in or 6 in) are presented in Fig. 12. As seen in Fig. 12, the pitch had a minor effect on the torque factors. However, it should be noted that the pitch for piles with small shaft diameters may have more pronounced effects (Sakr 2012b).
As indicated in Figs. 8 to 12, the torque factors increased with increasing embedment depths for relatively short piles (i.e. piles with embedment depth ratios up to about 8), beyond which the torque factors were relatively independent of embedment depth.
Torque factors in tension Kt for piles with
single, double or triple helices are presented in Fig. 13. The torque factors were estimated using a pile with shaft diameter of 0.324 m (12.75 in) and helix diameter of 0.763 m (30 in), installed into clay with undrained shear strength of 50 kPa (7.25 psi). The assumed spacing between different helices is 3D, and pitch is 152 mm (6 in). As seen in Fig. 13, K
t at lower embedment
depths is inversely proportional to number of
[FIG. 10] Effect of Varying Helix Diameters on Torque Factors in Tension, Kt.
[FIG. 12] Effect of Varying Pitch Size on Torque Factors in Tension, Kt.
[FIG. 11] Effect of Varying Helix to Shaft Diameter Ratio on Torque factors in Tension, Kt.
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
0 2 4 6 8 10 12 14 16 18 20
Estim
ated
Tor
que
Fact
or, K
t(m
-1)
Embedment Depth Ratio (H/D)
Undrained Shear Strength Cu = 50 kPaShaft Dia = 406 mm
Helix Dia 762 mm
Helix Dia 1200 mm
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
0 2 4 6 8 10 12 14 16 18 20
Estim
ated
Tor
que
Fact
or, K
t(m
-1)
Embedment Depth Ratio (H/D)
Undrained Shear Strength = 50 kPaD/d = 3D/d = 2D/d = 2.5
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
0 2 4 6 8 10 12 14 16 18 20
Est
imat
ed T
orqu
e F
acto
r, K
t(m
-1)
Embedment Depth Ratio (H/D)
Cu = 50 kPaShaft Dia = 324 mm
Single Helix 763 mm in Dia
pitch = 76 mm
pitch = 152 mm
[FIG. 9] Effect of Shaft Size on Torque Factors in Tension, Kt.
0.00
2.00
4.00
6.00
8.00
10.00
12.00
14.00
16.00
18.00
20.00
22.00
24.00
0 2 4 6 8 10 12 14 16 18 20
Est
imat
ed T
orqu
e Fa
ctor
, Kt(m
-1)
Embedment Depth Ratio (H/D)
Undrained Shear Strength, Cu = 50 kPaHelix Dia. = 0.4 m
168 mm shaft89 mm shaft273 mm shaft
[56] DFI JOURNAL Vol. 7 No. 1 August 2013
helices. However, at higher embedment depths, K
t increased with increasing the number of
helices.
The method of installation is one of the major factors that affects the quality of torque values used in practice. In general, methods of installations that cause more soil disturbance negatively impact the torque measurements and reduce the reliability of torque data. For example, the presence of cobbles or boulders during installation results in a sharp rise in torque values, which is not necessarily an indication of stronger soil conditions. Installation procedures such as applying down pressure, predrilling, or advancing the pile at a smaller rate than the pitch are other factors that impact torque measurements. An auguring effect (or spinning), where pile rotation is continued and little or no advancement into ground (usually occurs when pile hit hard soil layer or spinning on rock), is likely to considerably reduce the torque requirement during installation and cause significant soil disturbance.
Reliability of Torque Readings
Methods of measuring torque during pile installation are mainly either using a mechanical gauge that measures the differential pressure across the gear motor, or using an electronic torque transducer that consists of a series of strain gauges attached to the drive head. Theoretical torque using differential pressure may be estimated as:
( )24πη×××Δ
=PGRCIDPT [26]
where
PΔ = Differential pressure across the motor, (psi);
CID = Cubic inch displacement of the hydraulic motor;
PGR = Planetary gear ratio; and η = Combined motor and planetary gear efficiency.
It should be mentioned that, when using torque measuring method based on differential pressures, the hydraulic gear motor torque versus differential pressure curves may not reflect the manufacturer’s stated performance data. Equipment and hydraulic line size may also affect the torque versus differential pressure curve for the same motor. Deardorff (2011) advocated that installation speed and flow rate at the lower end of differential pressure curve may also affect the torque versus differential pressure curve for the same motor and same equipment line setup. Therefore, the use of torque measurement based on differential pressures may not be accurate. Electronic torque transducers may provide more reliable means of measuring torque during pile installation. Frequency of calibrating torque measurement devices is another factor that affects the reliability of torque measurements.
Moreover, a clear definition of the average torque value at the end of installation is required. For example, Hoyt and Clemence (1989) averaged the installation torque over the final distance of penetration equal to three times the largest helix diameter. Some contractors specify the use of average torque over the last 0.3 m (1 ft), while others specify that torque should be averaged over the last 1 m (3.3 ft) of installation.
CONCLUSIONSThis paper presents a theoretical model for predicting torsional resistance to helical pile installation into cohesive soils. The developed model was validated by comparing the estimated torque to the measured values at different embedment depths for a case reported in the literature. The measured and estimated
[FIG. 13] Effect of Increasing Number of Helices on Torque Factors in Tension, Kt.
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
9.00
10.00
0 2 4 6 8 10 12 14 16 18 20
Est
imat
ed T
orqu
e Fa
ctor
, Kt(m
-1)
Embedment Depth Ratio (H/D)
Single helix2 helices3 helices
DFI JOURNAL Vol. 7 No. 1 August 2013 [57]
torque values at the end of pile installation from total of seventy four (74) field installations reported in the literature were also compared. Both measured and estimated torque values agreed reasonably. The developed torque model was then used to assess torque factors. The following general conclusions may be drawn:
1. Torque required to install piles into cohesive soils can be reasonably estimated using the theoretical model developed in this paper.
2. Based on the proposed torque model, theoretical torque factors (K
t and K
c) that
represent the ratio between ultimate capacity and installation torque were proposed.
3. Parameters required to assess theoretical torque factors include pile geometry and soil strength parameter such as undrained shear strength. The required soil parameters to assess torque factors are standard parameters and therefore it is relatively easy to obtain for the project site.
4. Pile geometry, including shaft shape, ratio of helix to shaft diameter, and embedment depth has a considerable effect on the torque factors. Other factors that influence torque factors include pitch and number of helices.
5. Torque factors in compression and tension for piles with similar configurations installed into similar cohesive soils are different. It was found that torque factors in compression are generally higher than those factors in tension.
6. Torque measurements are also influenced by other factors such as the method of installation, operator experience, and accuracy of the measurement device. Therefore, in absence of precise installation procedures and quality torque device measurements, installation torque readings should be used with caution and may only be used to qualitatively assess installation.
REFERENCES1. Basile, F. 2010. "Torsional response of
pile groups", Proceedings 11th DFI & EFFC International Conference on Geotechnical Challenges in Urban Regeneration, London, UK.
2. Beim, J., and Luna, S.C. 2012. "Results of dynamic and static load tests on helical piles in the varved clay of Massachusetts", DFI Journal, Deep Foundations Institute, In print.
3. Cannon, J.G. 2000. "The application of high strain dynamic pile testing to screwed steel piles", In Proceedings of 6th International Conference On the Application of Stress Wave Theory to Piles, Sussumu Niyama and Jorge Beim ed., Sao Paulo, Brazil, pp. 393-398.
4. Cerato, A.B., and Victor, R. 2009. "Effects of long-term loading on fluctuating water table on helical anchor performance for small wind tower foundations", Journal of Performance of Constructed Facilities, ASCE 23(4): pp. 251-261.
5. CFEM. 2006. Canadian Foundation Engineering Manual. 4th Edition. Canadian Geotechnical Society, Technical Committee on Foundations, BiTech Publishers Ltd., Richmond, BC.
6. Das, B.M. and Seeley G. R. 1975. "Breakout resistance of horizontal anchors", Journal of Geotechnical Engineering Division, ASCE, 101(9): pp. 999–1003.
7. Deardorff, D. 2011. "A comparison of gear motor performance curves for helical pile installation", Seminar presented at DFI Helical Foundations and Tiebacks Specialty Seminar, Deep Foundations Institute, March 17, 2011, Dallas, Texas, USA.
8. Hoyt, R.M., and Clemence, S.P. 1989. "Uplift capacity of helical anchors in soil", Proceedings of the 12th International Conference on Soil Mechanics and Foundation Engineering, Rio de Janerio, Brazil, Vol. 2, pp. 1019-1022.
9. Hoyt, R., Seider, G., Reese, L. C., and Wang, S. T. 1995. "Buckling of helical anchors used for underpinning: Foundation upgrading and repair for infrastructure improvement", Edited by William F. K. and Thaney, J. M. Geotechnical Special Publication No. 50, ASCE, pp. 89-108.
10. Livneh, B., and El Naggar, M.H. 2008. "Axial testing and numerical modeling of square shaft helical piles under compressive and tensile loading", Canadian Geotechnical Journal, 45: pp. 1142–1155.
[58] DFI JOURNAL Vol. 7 No. 1 August 2013
11. Perko, H.A. 2001. "Energy method for predicting the installation torque of helical foundations and anchors", New Technologies and Design Developments in Deep Foundations, Reston, VA, ASCE, pp. 342:352.
12. Perko, H. A. 2009. Helical Piles: A Practical guide to design and installation. John Wiley & Sons. New York, N.Y.
13. Sakr, M. 2008. "Helical piles for power transmission lines: Case study in Northern Manitoba, Canada", Ninth International Conference on Permafrost, Extended Abstracts, NICOP 2008, Fairbanks, Alaska, USA, pp. 261:262.
14. Sakr, M. 2011. Helical piles - "An effective foundation system for solar plants", 64th Canadian Geotechnical Conference and Pan-AM CGS, Toronto, Ontario, 2-6 October 2011, Toronto.
15. Sakr, M. 2012a. "Installation and performance characteristics of high capacity helical piles in cohesive soils", DFI Journal, 6(1): 41-57, July 2012.
16. Sakr, M. 2012b. "Torque prediction of helical piles in cohesive soils", The 65th Canadian Geotechnical Conference (CGC); at the Fairmont Hotel, Winnipeg, Manitoba, Canada from September 30 to October 3, 2012.
17. Tappenden, K.M. 2007. "Predicting the axial capacity of screw piles installed in Western Canadian soils", MSc. Thesis, The University of Alberta, Edmonton, Alberta, Canada.
18. Zhang, D.J.W. 1999. "Predicting capacity of helical screw piles in Alberta soils", MSc. Thesis, The University of Alberta, Edmonton, Alberta, Canada.
DFI JOURNAL Vol. 7 No. 1 August 2013 [59]
Ultimate Lateral Resistance of Piles in Cohesive SoilLassaad Hazzar, University of Sherbrook, Canada; (819) 446 5100; [email protected]
Mourad Karray, University of Sherbrook, Canada
Mounir Bouassida, University of Tunis El Manar, Tunis, Tunisia
Mahmoud N. Hussien, University of Sherbrook, Canada
ABSTRACT The ultimate lateral resistance of piles in cohesive soil is studied using the well-known finite difference code, FLAC2D. The Modified Cam Clay (MCC) constitutive relation is adopted in the analyses to model the cohesive soil behavior, whereas the structural pile model with three degree of freedoms, available in FLAC2D library, is adopted to model the piles. The reliability of Broms's method, still used in the current design practice of piles under lateral loads, is verified. Comparisons between the ultimate lateral resistances of piles and those deduced from the graphs proposed by Broms (1964) are presented in graphs. Different factors thought to affect the lateral resistance of piles in cohesive soil, not adequately considered in Broms's method, such as clay stiffness, pile length, pile diameter and axial load are parametrically studied. A special concern is devoted to elucidate the effects of over-consolidation ratio (OCR) on the ultimate lateral resistance of piles in cohesive soil.
INTRODUCTIONPile foundations have been used extensively for supporting both axial and lateral loads for a variety of structures including heavy buildings, transmission lines, power stations, and highway structures. In some cases, the lateral loads may be relatively light and there is no need to account for them in pile design; however, in other cases, lateral loads govern the design of piles. A key element in the design of pile foundations under lateral loads is the determination of the ultimate lateral resistance that can be exerted by the soil against the pile (Murff and Hamilton, 1993). For example, the ultimate lateral resistance is required for calculating the p-y curves, which have been used extensively in recent years in piles design.
Several methods have been published for predicting the ultimate lateral resistance of pile in cohesive soils (Brinch Hansen, 1961; Broms, 1964; Poulos and Davis, 1980; Fleming et al.,1992; Reese and Van Impe, 2001). However, these methods often produce significantly different predictions of the ultimate resistance. This makes it difficult for engineers to effectively select the appropriate method when designing laterally loaded piles in cohesive soils.
Because the problem of determining the ultimate resistance of a laterally loaded pile is a three dimensional (3D) and nonlinear
problem, finding a rigorous solution is very unlikely. Thus existing solutions for the ultimate lateral resistance of the pile are either of a semi-empirical nature or employ approximate analysis which often involves many simplifications (Jamiolkowski and Garassino, 1977). These approximations may account for the significantly different ultimate resistance values obtained from the different methods. This makes it difficult for practicing engineers to effectively select the appropriate method when designing laterally loaded piles in cohesive soils. In this paper an assessment of the most important method, Broms's method, still used in the current design practice of piles under lateral loads, is done. A two-dimensional (2D) finite difference code, FLAC2D (Version 6, Manual [2008]) is used to this end. The Modified Cam Clay (MCC) constitutive relation is adopted in the analyses to model the cohesive soil behaviour, whereas the structural pile model with three degrees of freedom, available in FLAC2D elements library, is adopted to model the piles. Different factors thought to affect the ultimate lateral resistance of piles in cohesive soil such as clay stiffness, pile length, pile diameter and axial load are parametrically studied. A special concern is devoted to elucidate the effects of over-consolidation ratio (OCR) on the ultimate lateral resistance of piles in cohesive soil. The investigations were carried out for single piles in a type of clay which has
[60] DFI JOURNAL Vol. 7 No. 1 August 2013
been used by several studies conducted at the University of Sherbrook, Quebec, Canada.
The existing methods of predicting the ultimate lateral resistance of pile foundations under lateral loads are first reviewed, to be followed by the main part of the study with respect to the effects of clay stiffness, pile length, pile diameter, axial loads and OCR on the lateral ultimate resistance of pile foundations. The primary findings from this study are summarized as conclusions.
EXISTING METHODS OF PREDICTION OF ULTIMATE LATERAL RESISTANCEThe existing methods used to estimate the lateral resistance of vertical piles can be divided into two main categories: methods of ultimate lateral resistance and methods of acceptable deflection at a given working lateral load. First investigation by Terzaghi (1955) consisted in the use of variable passive earth coefficients for modeling the lateral reaction of soil as a function of its internal angle of friction. Adopting the method proposed by Brinch Hansen (1961), the pile is assumed to rotate with respect to its centre of rotation, the ultimate lateral load is then estimated and the shearing force and bending moment diagrams are drawn. Broms (1964) presented a method to determine the ultimate lateral load in cohesive and cohesionless soils. Kasch (1977), stated that using Rankine’s passive states will result in very conservative solutions. Reese (1977) developed a
computer program that is widely used to predict the performance of piles subjected to lateral loading. This program solves a differential equation derived on the assumption that the pile is linearly elastic and that the soil reaction may be represented as a line load. In recent years, extensive research and developments have been undertaken to predict theoretically the behavior of laterally loaded piles in clayey soils (Poulos and Davis, 1980; Brown and Shie, 1991; Fleming et al.,1992; Liang, 1998; Reese and Van Impe, 2001).
Broms's method is still used in the current design practice of piles under lateral loads to calculate the lateral bearing capacity of piles because of its simplicity. This method will be briefly reviewed in the next paragraph.
Broms’ method (1964), proposed for the prediction of lateral resistance of vertical piles, is similar to that developed by Brinch Hansen without consideration of c-φ’ soil parameters. In fact, Broms’s method is based on earth pressure for calculation of lateral resistance of vertical piles, but quite simple assumptions are made for the distribution of ultimate soil resistance over the length of the pile. These methods study two types of piles, a short-rigid and long-flexible, embedded in mono layered half space. Broms (1964) elaborated charts for determination of the ultimate lateral load for each class as illustrated in Figs. 1(a) and 1(b) respectively (FHWA, 1997). Figs. 1(a) and 1(b) show also that the ultimate lateral resistance of the piles is affected by pile head conditions.
[Fig. 1] Ultimate Lateral Load of Piles in Cohesive Soils; (a) Short Pile, (b) Long Pile (Broms, 1964)
DFI JOURNAL Vol. 7 No. 1 August 2013 [61]
The ultimate lateral resistance of a fixed head pile is higher than that of free-head conditions for both cases of short and long piles. In this method, the load-deflection relationships of laterally loaded piles driven into cohesive soils are similar to the stress-strain relationships as obtained from consolidated-undrained tests (Broms, 1964). In fact, Broms method calls for some comments that be discussed later. Broms’ contribution does not consider the effect of axial loading on lateral bearing capacity of piles.
In this study, the finite differences method is implemented to examine how the lateral load capacity of the pile is influenced by varying the length of the pile, its diameter and by considering the vertical component of load as well. The numerical modeling also aims to verify whether the OCR for clayey soil has a significant effect when determining the lateral capacity at failure of the loaded pile. Note that the OCR has not been taken into consideration in several previous investigations made about the ultimate lateral capacity of piles. Elsewhere, as for Broms’ method, the vertical load component was not considered in prior analyses, this study aims to clarify how the behavior of laterally loaded piles will be affected when subjected to additional vertical load.
Therefore, the main objective is to draw design charts making possible the design of laterally loaded piles in cohesive soils and to compare the numerical predictions by FLAC2D (Fast Lagrangian Analyses of Continua in 2D) program with the Broms’ solution.
STUDIED MODEL
Finite-Difference code
In this study the finite-difference code FLAC2D (Fast Lagrangian Analysis of Continua) is used to model the behavior of single piles embedded in a clay layer in non-symmetric loading plane strain condition. FLAC2D is a commercial finite difference program that adopts an explicit numerical scheme which solves the dynamic equations of motion (even for static problems) in conjunction with an incremental constitutive law over a small time step, at discrete points in space. This method is particularly well adapted for analyzing nonlinear behavior of soils.
Geometry of the model
Fig. 2 shows the general layout and meshing of the finite differences model. Side boundary displacements were fixed in the horizontal direction, while those at the bottom boundary were fixed in both the horizontal and vertical directions. The pile is modeled by means of a structural pile model, available in the FLAC2D library, with three degrees of freedom: horizontal and vertical displacements and a rotation with respect to perpendicular axis of the plan in Fig. 2. As shown in Fig. 2, the pile toe is anchored in a separate stratum (rock). The finite difference analyses were performed in two stages. In the first stage (self-weight analysis), the in-situ stresses were initialized in the soil due to the self weight of the soil. Properties of the pile were set to be zero during this stage of analysis. During the second stage
[Fig. 2] Numerical plane strain model
[62] DFI JOURNAL Vol. 7 No. 1 August 2013
of analysis (lateral load analysis), the actual properties of soil and pile were assigned. The applied loading was simulated by the application of a lateral load at the top of the pile. The modeling of the pile installation process is rather complicated, so that pile is assumed to be in a stress-free state at the beginning of the analysis, and the effect of the pile installation is ignored.
Soil properties
The soil was a soft post-glacial clay of marine origin, sampled from the site of Grande Baleine River (Demers, 1980). Two specimens (COE-01 and COE-02) have been tested to identify the geotechnical characteristics of this clay.
These samples were extracted from three (3) holes using a sampler developed by the University of Sherbrook (Quebec-Canada). The tool provides specimens of clay having a diameter ranging between 250 and 270 mm (9.8 and 10.6 in) and about 350 mm (14 in) in height. In the laboratory, the samples were cut into slices of 110 to 120 mm (4.3 4.7 in) in height, surrounded by paraffin and stored in a room where the humidity hovers around 97% and the temperature is maintained at about 14 °C (57 °F) . Laboratory tests such as Consolidation oedometer tests and triaxial tests CU (isotropically consolidated triaxial tests and sheared under undrained conditions) were carried out on the soil samples. In situ tests such as the Swedish cone were also carried out to measure the undrained shear strength of the clay. The obtained value of the undrained shear strength of the clay was found to vary between 40 and 80 kPa (5.8 and 11.6 psi). Table1 summarizes the recorded geotechnical characteristics of tested clays.
The Modifi ed Cam-Clay Model (CCM) (Roscoe
and Burland, 1968) was adopted as quite
appropriate, particularly for materials whose
behavior is infl uenced by volume variation.
In fact, the CCM may be used to represent
materials when the infl uence of volume change on
bulk property and resistance up to failure should
be taken into consideration, as for soft clays.
The CCM may give softening behavior
for particular stress paths. Without special
regularization techniques, this softening behavior
may lead to mesh dependency and numerical
instability. The use of the CCM in practical
applications is not recommended.
The CCM is expressed in terms of three variables:
the mean effective pressure, p; the deviator
stress, q; and the specifi c volume, v. In the FLAC
implementation of this model, principal stresses σ
1, σ
2, σ
3 are used, the out-of-plane stress, σ
zz,
being recognized as one of these. (By convention,
traction and dilation are positive.)
The generalized stress components p and q may
be expressed in terms of principal stresses, as
follows:
( )1 2 3
2
1p = − σ + σ + σ
2 21 2 2 3 3 1
31q ( ) ( ) ( )2
= σ −σ + σ −σ + σ −σ [1]
(Note that 2q 3J= , where J2 is the second
invariant of the effective stress deviator tensor).
The incremental strain variables associated with
p and q are the volumetric strain increment, Δe,
and distortional strain increment, Δeq , and we
have
1 2 3
2 2 2q 1 2 2 3 1 3
e e e e
2e ( e e ) ( e e ) ( e e )3
Δ = Δ + Δ + Δ
Δ = Δ − Δ + Δ − Δ + Δ − Δ [2]
Where Δej, j = 1, 3 are principal strain
increments. The principal strain increments may
be divided into elastic and plastic parts so that
pei i ie e e i 1,3Δ = Δ + Δ = [3]
The specifi c volume, υ, is defi ned as:
[TABLE 1] Geotechnical properties of clay studied
Test no Plasticity
index, Ip (%)
Initial void
ratio, e0 (-)
Effective
stress, σ0’
(kPa)
Pre-con-
solidation
pressure, σP’
(kPa)
Compression
index, CC (-)
Swelling
coeffi cient
CS (-)
Total unit
weight, γ
(kN/m³)
Undrained
shear
strength, cu
(kPa)
COE-01 11.7 1.59 40.7 105 0.90 0.08 16.7 16.0-39.0
COE-02 7.0 1.57 41.0 112 0.88 0.06 16.7 43.0- 62.0
DFI JOURNAL Vol. 7 No. 1 August 2013 [63]
s
VV
υ = [4]
Where Vs is the volume of solid particles
(assumed incompressible), contained in a volume,
V, of soil. The incremental relation between
volumetric strain, e, and specifi c volume has the
form
e ΔυΔ =υ
[5]
Starting with an initial specifi c volume, υ0,
we may thus write, for small volumetric strain
increments,( )0 1 eυ = υ + [6]
Where e is the current accumulated volumetric
strain.The incremental expression of Hooke’s law
in principal axes may be expressed in the form
)( )(( )
e e e1 1 1 2 2 3
e e e2 1 2 2 2 3
e e e3 3 3 2 2 3
e e e
e e e
e e e
Δσ =α Δ + α Δ + Δ
Δσ =α Δ + α Δ + Δ
Δσ =α Δ + α Δ + Δ
[7]
Where: α1 = K + 4G/3; and α
2 = K − 2G/3.
In this study, eight material parameters were
required to specify the soil model, including
either the elastic bulk modulus “K” or elastic
shear modulus “G”, mass density “ρ”, Poisson’s
ratio “µ”, slope of the normal consolidation
line “λ”, slope of the elastic swelling line“κ”,
frictional constant “M”, pressure of reference “p1”
and the specifi c volume at pressure of reference,
p1, on the normal consolidation line “υλ”.
Fig. 3 presents the oedometer curve in the
semi-logarithmic plot (υ, ln p) where p, is the
effective vertical pressure and, υ, the specifi c
volume of specimen.
The material properties adopted in the analyses
for soft, medium and hard clay are presented
in Table 2. In this table, the coeffi cient of earth
pressure (K0) is defi ned as the ratio of effective
horizontal stresses (σh) to applied effective
vertical stresses (σv) at zero stress strain
(Donath, 1981):
h 0 vKσ = ⋅σ [8]
Alpan (1967) indicated that K0 is a function of
over consolidation ratio (OCR), defi ned as the
[Fig. 3] Oedometer curves of tested clays (Demers, 1980)
[64] DFI JOURNAL Vol. 7 No. 1 August 2013
ratio of initial pre-consolidation pressure to the
in situ overburden effective stress, and in over
consolidated clays:
n
0(NC)K
0(OCR) 0(NC)K K OCR= ⋅
0.15 0.233log (Ip)= +
Ip / 281n 0.54 10−= ⋅
[9]
[10]
[11]
Pile properties
The pile is modeled as a structure element
made up of concrete material characterised by a
Poisson’s ratio of 0.2, a unit mass of 2500 kg/m3
(4200 lb/ yd3) and Young’s modulus equal to
25 GPa (3.6x106 psi).
The length D and the diameter b of pile are
variable in order to investigate their infl uences
on the lateral bearing capacity of the pile.
The ultimate lateral load of the pile, Qu, is
represented by the dimensionless factor defi ned
by “Qu/cu b2” for which the infl uence of several
parameters will be studied.
Limitations of 2D analysis
A pile foundation subjected to lateral loads is
a class of problem that incorporates pile-soil
interaction in 3D. In this paper, the soil-pile
interaction in the direction perpendicular to the
loading direction is not accounted for in the used
simple 2D fi nite difference formulation. This
simplifi cation leads to overestimation of the
lateral displacement of the pile compared to
the actual behavior encountered in the fi eld.
Prediction of lateral resistance of pile
The adopted modeling of a beam element
subjected to the lateral action/reaction of soil
is derived from the well-known equilibrium
equation of beams:
4
4d yEI p(x) 0dx
+ = [12]
Fig. 4 details how the horizontal resistance of
soil p(x) can be determined by adopting the
spring equation:
(x) k(x) y= ⋅ p [13]
k(x) : modulus of the horizontal reaction of soil
(kN/m²);
y : horizontal displacement of the pile at depth x
(m);
E : young’s modulus of the pile (kPa);
I : moment of inertia of the cross section at
x (m4);
x: current depth along the length of pile.
[Fig. 4] Model of soil reaction by elastic springs
Introducing the bending moment, M (kN.m) and
the shear force, V (kN) at depth x within a current
cross section of the pile, the equilibrium equation
provides relationships between the bending
moment and shear force, and, then as illustrated in
Soil rigidity ρ (kg/m3) G (MPa) K (MPa) µ (-) λ (-) κ (-) M (-) p1 (kPa) υλ (-) K
0 (-)
Soft clay
cu = 16.0 kPa
1670 4.80 12.48 0.33 0.262 0.065 0.77 1 5.3 0.63
medium clay
cu = 39.0 kPa
1670 11.70 30.42 0.33 0.262 0.065 0.77 1 5.3 0.63
Stiff clay
cu = 64.0 kPa
1670 19.20 49.92 0.33 0.257 0.064 0.77 1 5.25 0.58
[TABLE 2] Parameters according to CCM
DFI JOURNAL Vol. 7 No. 1 August 2013 [65] [Fig. 6] Behaviour of pile under lateral load
line for p=p1, is given by:
( ) ( )ln 2λΓ = υ − λ − κ × [15]
The numerical analysis has been conducted by
adopting zero free vertical distance from the
head of pile to the soil surface (ec=0), and
varied ratio D/b, D is the embedment of pile in
the clay layer. The comparison between
numerical predictions and Broms’ results are
presented in Fig. 7. For this case, it can be seen
that Broms’ assumption greatly overestimates
the ultimate lateral resistance of pile in purely
cohesive clays that was assumed equal to 9bcu,
but numerical predictions show that the soil
will collapse much earlier.
EFFECTS OF VERTICAL LOAD AND PILE DIAMETER ON ITS LATERAL RESISTANCE
The infl uence on pile diameter has been also
investigated. Fig. 8 shows that the variation of
pile diameter, especially when D/b is less than
14, does not signifi cantly affect the normalized
ultimate lateral bearing capacity of the pile.
The infl uence of vertical load on the ultimate
lateral bearing capacity is studied. At this stage,
fi rst, the ultimate vertical bearing capacity is
obtained, and then by introducing a factor of
safety equal to 3 the allowable vertical load is
deduced. The ultimate lateral bearing capacity
of the pile is fi nally determined. Results of
Fig. 9 show that the ultimate lateral resistance
will decrease when the vertical load component
increases. Therefore special care should be
accorded when it comes to the prediction of the
ultimate lateral resistance of a pile.
Fig. 6, the lateral resistance of soil is derived from
Eqn. (9). The complete solution is obtained once
the horizontal defl ection of pile is determined.
Therefore, we concluded that the lateral soil
reaction p (x) can be determined as follows
(see Fig. 5):
[Fig. 5] Shear and lateral load
For a pile of length D = 8.0 m (26 ft) and
diameter b = 0.8m (2.6 ft), Fig. 6 displays the
diagrams for profi les of pile behavior under a
lateral load equal at 250 kN (28.1 ton).
EFFECT OF SOIL STIFFNESS ON LATERAL RESISTANCE
The undrained shear strength has been varied in
order to study the effect of the stiffness of clayey
soils on the ultimate lateral resistance of the pile.
For capped plasticity model, like the Modifi ed
Cam Clay here investigated, the undrained shear
strength, cu, is uniquely related to the specifi c
volume, υ, by the equation [14]:
1u
Mpc exp
2Γ − υ=
λ [14]
Where the specifi c volume, Γ, at the critical state
[66] DFI JOURNAL Vol. 7 No. 1 August 2013
[Fig. 7] Effect of soil stiffness on lateral load capacity compared with Broms method
[Fig. 8] Effect of diameter on lateral bearing of capacity pile
[Fig. 9] Effect of vertical load, ec/b = 0
DFI JOURNAL Vol. 7 No. 1 August 2013 [67]
[Fig. 10] The ultimate lateral capacity vs OCR
EFFECT OF OVER-CONSOLIDATION RATIO
The infl uence of OCR on the ultimate lateral
load capacity of a single pile is presented in
this section. The same soil proprieties presented
in Table 2 were adopted with changing the
OCR values from 1 to 10 (K0 values vary
between 0.4 to 1.23).
Fig. 10 shows the effect of OCR on the ultimate
lateral capacity of the pile. When the OCR
increases from 2 to 10, the increase in the
ultimate lateral load is about 20%. The
increase of the OCR values is accompanied
by an increase in the K0 values. This increase
in the K0 values is due to the increase of
the horizontal lateral stress of the soil, the
confi ning pressure, compared to the applied
vertical stresses. The increase of the confi ning
pressure is the major deriving factor causing
the increase in the ultimate resistance of a pile
to lateral loads. Thus, it is concluded that the
role of pre-consolidation pressure cannot be
neglected in the pile design.
CONCLUSIONS
The ultimate resistance of pile foundations
embedded in cohesive soil has been studied
in this paper through a series of 2D fi nite
differences analyses. FLAC2D was employed
to this end. The Modifi ed Cam Clay (MCC)
constitutive relation was adopted in the
analyses to model the cohesive soil behavior;
whereas the structural pile model with three
degrees of freedom is adopted to model the
piles. The reliability of the well-known Broms'
method is discussed in this paper. Different
factors thought to affect the lateral resistance
of piles in cohesive soil, not adequately
considered in Broms' method, such as clay
stiffness, pile length, pile diameter and axial
load were parametrically studied.
Of the fi ndings of this study, the following
conclusions can be drawn:
- The ultimate lateral capacities of pile
foundations obtained from the current
fi nite differences analyses were found to
be smaller than that obtained by Broms'
graphs.
- The pile diameter, not considered in
Broms's method, seems to have a signifi cant
effect on the ultimate lateral capacities
- For low values of soil stiffness, Broms'
method overestimates the ultimate lateral
resistances of piles. As the soil stiffness is
increased, the ultimate lateral resistances
obtained from the current analyses
approach that obtained using Broms'
graphs.
- The axial load increases ultimate bearing
capacity and special care to choose the
ultimate bearing capacity of pile should be
taken.
- The CCM is a suitable model to describe
sensitive clays, and it is necessary to
[68] DFI JOURNAL Vol. 7 No. 1 August 2013
take care of the value of OCR or pre-
consolidation pressure in the design of piles
embedded in cohesive soils.
REFERENCES
1. Alpan, I. "The Empirical Evaluation of
the Coeffi cient K0 and K
or", Soils and
Foundations, Vol.7, No.1, 1967, pp. 31-40.
2. Brinch Hansen, J. "The Ultimate Resistance
of Rigid Piles against Transversal Forces",
Geoteknish Institute Bulletin No.12, Danish
Geotechnical Institute, Copenhagen,
Denmark, 1961, pp. 5-9.
3. Britto, A. M., and Gunn, M.J. Critical
State Soil Mechanics via Finite Elements.
Chichester U.K.: Ellis Horwood Ltd, 1987.
4. Broms, B.B. "Lateral Resistance of Piles in
Cohesive Soils", Journal of Soil Mechanics Foundation Division, Vol. 90(2), 1964, pp.
27-64.
5. Brown, D.A., and Shie, C.F. "Evaluation
of the Relative Infl uence of Major
Parameters for Laterally Loaded Piles in
three Dimensional Finite Element Models",
Civil Engineering Department, Harbert
Engineering Center. Auburn University,
Alabama, 1991.
6. Demers, B. "Résistance Cyclique d’une
Argile Extra-Sensible", Thesis M.Sc., University of Sherbrook, Quebec, Canada,
1980.
7. Donath, A.D. Untersuchungen Veber den
Erddruck auf Stuetz waende. Zeitschrift
Fuer Bauwesen, 1981.
8. Federal Highway Administration.
Design and Construction of Driven Pile
Foundations. Workshop Manual – Vol. I,
Publication no13, Washington, D.C. 1997.
9. Fleming, W.G.K., Weltman, A.J., Randolph,
M.F. and Elson, W.K. Piling Engineering.
Surrey University Press, London, 1992.
10. Itasca Consulting Group. FLAC: Fast
Lagrangian Analysis of Continua User’s and
Theory Manuals, Version 6.0, Minneapolis,
USA, 2008.
11. Jamiolkowski, M. and Garassino, A. "Soil
Modulus for Laterally Loaded Piles",
Proceedings, 9th International Conference, Soil Mechanics Foundation Engineering. Tokyo, 1977, pp. 87-92.
12. Kasch, V.R. "Lateral Load test of Drilled
Shaft in Clay", Research report 211-1. Texas Transportation Institute, Texas A&M
University, 1977.
13. Liang, R. "Development and
Implementation of New Driven Piles
Technology", The Ohio Department of
Transportation and the US Department
of Transportation, Federal Highway
Administration, 1998.
14. Murff, J.D. and Hamilton, J.M. "P-Ultimate
for Undrained Analysis of Laterally
Loaded Piles", Journal of Geotechnical Engineering, Vol. 119(1), 1993, pp. 91-107.
15. Poulos, H.G. and Davis, E.H. Pile
Foundation Analysis and Design. Wiley,
New York, 1980.
16. Reese, L.C. "Laterally Loaded Piles:
Program Documentation", Journal of Geotechnical Engineering Division, ASCE.
Vol. 103(GT4), 1977, pp. 287-305.
17. Reese, L.C. and Van Impe, W.F. Single Piles
and Pile Groups Under Lateral Loading. A.
A. Balkema, Rotterdam, 2001.
18. Roscoe, K.H. and Burland, J.B. On the
Generalized Stress-Strain Behavior of ‘Wet
Clay’. Engineering Plasticity, Cambridge
University Press, New York, 1968, pp. 535-
609.
19. Terzaghi, K. "Evaluation of Coeffi cients of
Subgrade Reaction", Géotechnique. Vol.
5(4), 1955, pp. 297-236",
DFI JOURNAL Vol. 7 No. 1 August 2013 [69]
TECHNICAL NOTEDirect Solution of the Brinch-Hansen 90% Pile Ultimate Failure LoadDon W. Dotson, PhD., PE, DGE, Chief Designer, Geo-Structural Design Group, AMEC Environment and Infrastructure, Adj. Prof., Dept. of Civil Engineering, Tennessee State University, Nashville, TN, USA’ Ph: (615) 333-0630; [email protected]
ABSTRACTIn 1962, Kondner prepared several papers dealing with hyperbolic stress-strain response of cohesive soils. The following year, Brinch Hansen proposed 80% and 90% failure criteria for stress-strain behavior of cohesive soils. Fellenius was instrumental in popularizing these failure criteria for pile load tests and offered a direct solution equation for the failure load according to the 80% failure criterion. Since the 90% Criterion has been incorporated into the International Building Code, equations for the direct solution of the failure load at the 90% Criterion would be useful to practicing engineers. This Technical Note supplies the derivation methodology for the 80% Criterion and provides expressions to determine the load and deflection at failure for the 90% Criterion.
INTRODUCTIONRobert Kondner (1962a, 1962b) proposed that the stress-strain behavior of cohesive soils in triaxial testing could be reasonably approximated by a two-constant rectangular hyperbola (Eqn. 1) which could be algebraically transformed into a linear relationship with determinable slope and intercept (Eqn. 2):
[1]
[2]
where: σ = stress ϵ = strain
a,b = constants
In a discussion of Kondner (1963), Hansen (1963) reported equations similar to Kondner’s, especially Eqn. 3.
[3]
(Note: Hansen’s original formulation contained an erroneous minus sign which was later corrected in Kondner, 1964.) The linear transformation of Eqn. 3 is shown in Eqn. 4.
[4]
Hansen (1963) observed that when this latter form gave a good approximation to the test data, it could be used to provide a general, simple failure criterion in which failure is
represented by the stress for which the strain is equal to four times the strain at a 20% smaller stress. This became known as the Brinch Hansen 80% Failure Criterion.
Hansen (1963) compared this 80% Failure Criterion to a definition he previously proposed (source not cited) in which he defined the stress at failure as equal to two times the strain at a 10% smaller stress (i.e., the 90% Failure Criterion). Hansen (1965) further noted that these hyperbolic curves “seem to apply, not only to direct shear tests, but to almost any test, in which shear stresses play a dominant role, for example triaxial tests, plate loading tests, pile loading tests.”
Fellenius (1975) compared various failure criteria for pile load testing. Fellenius (1980) gave more detailed examples of a number of these criteria, including the 80% Criterion along with algebraic expressions used to calculate the load and deflection at failure.
Fig. 1 is a digitization of a load-movement curve for a driven concrete pile load test using the Constant Rate of Penetration method (Fellenius, 1980). Fig. 2 is a plot of 15 of the data points prior to pile plunging using the transformed ordinate axis along with a “best fit” line. (Note: any consistent set of units could be used. Fellenius multiplied the ordinate by 103 for presentation. That same format has been preserved here.)
[70] DFI JOURNAL Vol. 7 No. 1 August 2013
[FIG. 1] - Pile Test Results (after Fellenius 1980)
[FIG. 2] - Brinch Hansen Transformation (after Fellenius 1980)
The equation of the line that can be fitted to the data in Fig. 2 is:
[5]
where C1 is the slope and C
2 is the y-intercept of
the fitted line. The load P can be determined at any point as:
[6]
Fellenius (1980) gives the following equation for the 80% failure load:
[7]
and the corresponding deflection:
[8]
The derivation of these expressions for the 80% Criterion were not provided, nor were similar expressions provided for the 90% Criterion. Since the 90% Criterion has been incorporated into the International Building Code (2000), and not the 80% Criterion, equations for the direct solution of the failure load at the 90% Criterion would be useful to the practicing engineer. This Technical Note supplies the derivation methodology for the 80% Criterion and provides expressions to determine the load and deflection at failure for the 90% Criterion. In practice, the results obtained from either direct solution method should be validated by comparison to the load-movement curve.
DERIVATIONFor both the 80% and 90% Failure Criteria, a plot of the pile test data is prepared in which pile head deflection (∆) is plotted along the abscissa and the square root of the deflection divided by the test load (P) is plotted along the ordinate (Fig. 2). A “best fit” linear approximation is fitted to the data and the slope and intercept are determined. The portion of the curve with the highest loads is of principal interest, thus points at lower loads may be omitted since they tend to skew the “best fit” line. Failure for the 80% Criterion corresponds to the values of P
u
and ∆ that satisfy Eqn. 6 and Eqn. 9 as a system of linear equations (equating P with P
u).
[9]
Eqn. 7 is the solution to this system of equations.
Similarly, failure for the 90% Criterion corresponds to the values of P
u and ∆ that
satisfy Eqn. 6 and Eqn. 10.
[10]
After solving and rearranging (see Appendix), an approximate solution can be written as:
[11]
with deflection [12]
As Hansen (1963) noted, the 80% and 90% failure criteria give approximately the same failure load, although the IBC code specifies the 90% Criterion.
(inches)
DFI JOURNAL Vol. 7 No. 1 August 2013 [71]
Using Eqn. 11 and the best-fit coefficients from Fellenius, the calculated 90% Criterion failure load is 209 tons (190 tonnes), which compares favorably with the 205 tons (186 tonnes) reported (Fellenius, 1980).
ACKNOWLEDGEMENTSI would like to thank my colleague Siphay Douangvilay for his review and comments along with the anonymous reviewers.
REFERENCES1. Fellenius, B. H. (1975). “Test loading of piles
and new proof testing procedure”, Journal of Geotechnical Engineering Division pp. 101(9), 855-869.
2. Fellenius, B. H. (1980). “The analysis of results from routine pile load tests”, Ground Engineering. 13(6), pp. 19-31.
3. Hansen, J. Brinch. (1963). “Discussion: Hyperbolic stress-strain response: cohesive soils”, Journal of Soil Mechanics, Foundations Division, 89(4), pp. 241-242.
4. Hansen, J. Brinch. (1965). “Some stress-strain relationships for soils”, The Danish Geotechnical Institute, Bulletin, No. 19, pp. 231-234.
5. International Building Code. (2000). International Code Council, Inc. Falls Church, VA 22041-3401.
6. Kondner, R. L. (1962a). “Hyperbolic Stress-Strain Relation in Direct Shear”, Technical Report, Civil Engineering Department, Northwestern University.
7. Kondner, R. L. (1962b). “Friction pile groups in cohesive soil”, Journal of Soil Mechanics, Foundations Division, 88(3), pp. 117-149.
8. Kondner, R. L. (1963). “Hyperbolic stress-strain response: cohesive soils”, Journal of Soil Mechanics, Foundations Division, 89(1), pp. 115-143.
9. Kondner, R. L. (1964). “Hyperbolic stress-strain response: cohesive soils – Closure”, Journal of Soil Mechanics, Foundations Division, 90(1), pp. 121-126.
APPENDIXSolve the following system of equations:
[A1]
[A2]
Solve Eqn. A1 for Pu
[A3]
Solve Eqn. A2 for Pu
[A4]
Set Eqn. A3 and A4 equal to each other:
[A5]
Find a common denominator
[A6]
Multiply the numerator on the left-hand side of the equal sign of Eqn. A6
[A7]
Multiply the numerator on the right-hand side of the equal sign of Eqn. A6
[A8]
Substitute the results of Eqn. A7 and Eqn. A8 into Eqn. A6 and rewrite
[A9]
Subtract the right-hand side of the equal sign from the left-hand side of Eqn. A9
[A10]
Factor √Δ out of the numerator
[A11]
Remove √Δ by dividing both sides of the equation by √Δ:
[A12]
Multiply both sides of Eqn. A12 by 9:
[A13]
[72] DFI JOURNAL Vol. 7 No. 1 August 2013
To avoid trivial or undefined solutions, the constraints on Eqn. A13 are that∆ ≠ 0, (C
2 + C
1∆) ≠ 0, and (2C
2 + C
1∆) ≠ 0.
Therefore, the numerator of Eqn. A13 must be = 0 for a valid solution. Set the numerator = 0 and solve for ∆:
[A14]
Rearrange terms
[A15]
Factor Eqn. A15 [A16]
Solve for ∆
[A17]
After evaluation, the approximate solution to Eqn. A17 is:
[A18]
Substituting Eqn. A18 into Eqn. A3 gives
[A19]
Simplifying
[A20]
If desired, a more precise solution can be obtained by substituting Eqn. A17 into Eqn. A3. In the Fellenius (1980) example, the difference between approximate solution Eqn. A20 and a more precise solution determined by incorporating Eqn. A17 into Eqn. A3, is 0.004 tons (0.0036 tonnes).
DFI JOURNAL Vol. 7 No. 1 August 2013 [73]
Underwriters: Gold
CANADA & WESTERN U.S.CHAMPION EQUIPMENT SALES, LLC 562.634.8180
EASTERN & CENTRAL U.S.AMERICAN EQUIPMENT & FABRICATING CORP401.438.2626
Your Solutions Provider.
www.soilmec.com
[74] DFI JOURNAL Vol. 7 No. 1 August 2013
DFI Journal Paper Review Process
The peer review process for documents considered for publication in the DFI Journal is still evolving. The following is a description of the current process, however, the publication is still in its infancy and the review process is still in a state of flux. DFI reserves the right to alter the procedures as necessary.
Paper SubmittalPapers may be submitted at any time. Authors wishing to submit their papers for consideration of publication in the DFI Journal are invited to access www.dfi-journal.org. The website will ask for a login or, for new submitters, will ask for creation of an account. Once logged in the author must upload a full paper in MS Word format as well as any ancillary files such as figures, photos and other graphics which are included in the paper. The paper is then converted to a PDF file which the author must approve before the paper will be released to the publisher and journal editors for viewing. The journal editors preliminarily review the paper for relevancy to the Journal mission.
Paper Review The journal editors assign those papers deemed to be worthy of consideration for Journal publication to the appropriate editorial board member, which currently consists of DFI technical committee chairmen and other industry leaders, so that appropriate reviewers for the paper topic can be obtained. Reviewers are chosen based on their knowledge, areas of expertise, and qualifications to act as a reviewer on the particular subject matter of the paper in question. At least three reviewers will be assigned to each paper.
After the reviewers are selected, they are provided with instructions and a password for entry into the website where they can view the paper PDF and submit their evaluation. The criteria on which they base their review fall under two areas: technical content and quality of paper presentation. The criteria for technical content include relevancy, originality, appropriate references to support statements, significance of results and exclusion of personal opinion and commercialism. The criteria for paper presentation include quality of figures, quality of English language, paper organization and completeness. The reviewers enter their evaluation by responding to a number of questions rating the paper as well as entry of comments to authors. They are also required to make a recommendation to the journal editors of: accept as is; accept with mandatory changes; or reject. The author is advised by automatic email of the posting of reviews and he/she can access the reviews and respond and/or modify the paper to satisfy comments by the reviewers. A second round review can then take place if necessary, ultimately leading to second round reviewer recommendations. The publisher and editors, acting as a final review committee, make the decision, based on the reviewers’ recommendations, as to acceptance of the paper for publication in the next issue of the journal or in a subsequent issue.
Throughout the process, automatic emails are sent out to reviewers when papers are ready for their review and to the authors to keep them aware of the progress of their paper.
Paper Finalization Upon acceptance, the final paper submission by the author and all graphic files are downloaded by the publisher for processing and formatting for publication. The publisher is provided with proofs by the production house and these are edited to ensure acceptable layout, the absence of typos, clarity of figures, etc. In most cases the author(s) are provided with a final PDF for their review and approval.
DFI JOURNAL Vol. 7 No. 1 August 2013 [75]
DFI Journal Call for Papers
The Deep Foundations Institute compiles and publishes a Journal of practical and technically rigorous papers on a bi-annual schedule. The DFI Journal is distributed to ~3,000 DFI members plus non-member subscribers.
The DFI Journal content is subject to quality technical review, and must meet a standard in quality on practical subjects dealing with case studies, deep foundations history, design, construction, testing, innovations and research in the field.
Each journal consists of at least five documents collected from technical papers that are invited or selected from papers submitted by international industry members based on this call. Papers presented at the DFI Annual Conference and Specialty Seminars may be included if expanded to the Journal standard and review process.
The editors are herein sending out a call for original papers for consideration of inclusion in the upcoming journals. Full draft papers up to 15 pages in length are to be submitted to: http://www.dfijournal.org for review. Authors will be required to create a login account and will be notified via email on the status of their submission.
Papers are solicited on the following topics:
• Case studies involving foundation systems with technical data support• Historical evolution of deep foundations• Relationship between use of design, construction and equipment• Quality control, quality assurance and non-destructive testing• Innovation in all aspects of deep foundations and earth retention• Practice-oriented research
The Publisher and the Journal Editorial Board will review submitted papers for acceptability for publication in the current or future issues of the Journal, subject to full peer reviews as described on the preceding page entitled "DFI Journal Paper Review Process". Authors of papers accepted for publication will be required to sign a copyright licence agreement.
Deep Foundations Institute was incorporated in 1976 in the State of New Jersey as a non-profit educational activity. DFI is a technical association of firms and individuals in the field of designing and constructing deep foundations and excavations. DFI covers the gamut of deep foundation construction and earth retention systems.
Although the bulk of the membership is in North America, the Institute is worldwide.
DFI’s strengths are:
• Communication of information concerning the state-of-the-art and state of the practice of deep foundation technologies
• Offering networking opportunities for our members
• Offering opportunities for members to improve the industry through publications produced by volunteer committees
• Offering educational conferences, seminars and workshops in the industry
The core strength of DFI is the broad spectrum of its membership. All disciplines participate on an equal footing, be they contractors, engineers, owners, academicians, equipment manufacturers and distributors or materials manufacturers and suppliers. All types of foundation systems are represented, whether installed by driving, drilling or other means. This diversity and openness without bias provides a forum for the free exchange of knowledge and a platform for the development of new technology and opportunity.
DFI is:
• An international network of heavy construction professionals dedicated to quality and economy in foundation design and construction
• A forum open to all construction professionals across disciplines and borders.
• A technological association devoted to gathering, storing and disseminating practical information
• A resource for identifying and locating the specialists and sources of expertise.
• An initiator and participant in research
Deep Foundations Institute Sustaining Members are Corporate Members of DFI who have voluntarily granted funding to the Institute for expanded support of the Industry. The fund is managed by the DFI Educational Trust.
DFI Sustaining MembersAECOM USA INC.AMEC - ENVIRONMENT & INFRASTRUCTUREAMERICAN EQUIPMENT & FABRICATING CORP.ANDERSON DRILLINGAPE/J&MBAUER FOUNDATION CORP.BAUER - PILECO INC.BEN C. GERWICK INC.BERKEL & COMPANY CONTRACTORS INC.BRASFOND FUNDAÇÕES ESPECIAIS S/ABRAYMAN CONSTRUCTION CORPORATIONCAJUN DEEP FOUNDATIONS LLCCASE FOUNDATION COMPANYCIPORT S.A.DEAN CONSTRUCTION CO. LTD.DEWITT CONSTRUCTION INC.FOUNDATION CONSTRUCTORS INC.FOUNDATION SUPPORTWORKS INC.FOUNDATION TECHNOLOGIES INC.GEOKON INC.GOETTLEHAYWARD BAKER INC.HJ FOUNDATION COMPANYKELLER FOUNDATIONS LTD.KLEINFELDERL.G. BARCUS & SONS INC.LANGAN ENGINEERING AND ENVIRONMENTAL SERVICESMCKINNEY DRILLING COMPANYMENARDMORETRENCHMUESER RUTLEDGE CONSULTING ENGINEERSNICHOLSON CONSTRUCTION COMPANYO.C.I. DIVISION / GLOBAL DRILLING SUPPLIERS INC.PND ENGINEERS INC.SAS STRESSTEEL INC.SCHNABEL FOUNDATION COMPANYTEI ROCK DRILLS INC.THATCHER ENGINEERING CORPORATIONURBAN FOUNDATION/ENGINEERING LLCWILLIAM F. LOFTUS ASSOCIATES FOUNDATION ENGINEERSWURSTER ENGINEERING & CONSTRUCTION INC.
DFI JOURNALThe Journal of the Deep Foundations Institute
Deep Foundations Institute326 Lafayette AvenueHawthorne, New Jersey 07506 USATel: 973-423-4030Fax: 973-423-4031www.dfi .org
International Standard Serial Number (ISSN): 1937-5247