Contribution of Steel Casing to Single Shaft Foundation Structural Resistance (2018)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

5 1.1. Introduction The use of drilled shafts for bridge foundations is a well-established technology. Drilled Shafts: Construction Procedures and LRFD Design Methods (Brown et al. 2010), published by the Federal Highway Administration (FHWA), provides a comprehensive overview of current knowledge on this topic, covering geotechnical engineering, construction methods, testing, costs, and struc- tural design issues related to the use of this foundation system. This extensive information is considered well-known, and information from it is not repeated here. The key point of interest here is that casings are often used in drilled shafts. Although this is done primarily for support of the drilled shaft excavation, these casings, which are typically steel tubes (joined by welding if necessary), may be considered a structural element of the completed drilled shaft—effectively creating a composite structural member. Many structural engineers do not rely on this composite strength. This is reflected in Drilled Shafts: Construction Procedures and LRFD Design Methods (Brown et al. 2010), which recommends designing the shaft as a rein- forced concrete beam-column (reinforcement cages are typically present in the shaft). AASHTO suggests that the steel casing may be considered in computing structural strength, provided the casing is at least 1/8 inch and that some of the casing thickness is discounted to account for possible future corrosion (when applicable). To complement this, some state departments of transportation (DOTs) developed details to ensure that the steel casing acts compositely with the concrete such that they can count on it in designing the shafts. Others have not encouraged such an approach. In essence, questions remain on how to achieve this composite action, which hampers the broader adoption of this approach. The perceived substantial financial savings that may occur when treating the shaft as a composite member provide a significant incentive to answer these questions. While it is conservative to only consider the reinforced concrete section of a drilled shaft in design when calculating its strength, it is realistic to expect that, in some conditions, full com- posite action (by engaging the casing) is possible. However, to be able to rely on such composite action, it must be determined under which conditions the full composite action can be achieved. For example, this may depend on the condition of the casing surface (clean versus coated by bentonite) before it is filled with concrete. 1.2. Objectives and Scope of Work The objective of this research was to investigate how to account for the contribution of steel casing to the structural resistance of a single shaft foundation supporting a column and to propose revisions to AASHTO LRFD Bridge Design Specifications (2012) and AASHTO C h a p t e r 1 Background

6 Contribution of Steel Casing to Single Shaft Foundation Structural resistance Guide Specifications for LRFD Seismic Bridge Design (2014) based on the findings of this study. The research considered flexural and shear effects under axial and lateral loading for strength and extreme event limit states for a reinforced concrete-filled steel tube (RCFST) single enlarged shaft. Fifteen specific tasks in five phases were completed toward this objective. The completed tasks are as follows: • Phase I: Planning – Task 1: Conduct a critical review of relevant specifications, technical literature, and owner and industry experiences. – Task 2: Propose analytical and testing programs to investigate concrete confinement and composite action. – Task 3: Prepare Interim Report No. 1. • Phase II: Analytical Program – Task 4: Execute the approved work plan for the analytical program. – Task 5: Identify proposed areas of AASHTO LRFD Bridge Design Specifications and AASHTO Guide Specifications for LRFD Seismic Bridge Design that may require modification. – Task 6: Finalize the testing program to validate the findings of the analytical program. – Task 7: Prepare Interim Report No. 2. • Phase III: Testing Program and Analytical Program Validation – Task 8: Execute the testing program according to the approved work plan. – Task 9: Finalize the analytical program validation based on testing results. – Task 10: Prepare Interim Report No. 3. • Phase IV: Proposed Modifications to AASHTO – Task 11: Based on the analytical and testing investigations, develop specification and commentary language for proposed changes to AASHTO LRFD Bridge Design Specifica- tions and AASHTO Guide Specifications for LRFD Seismic Bridge Design supported with examples. – Task 12: Quantify the potential economic impact of the proposed revisions. – Task 13: Prepare Interim Report No. 4. • Phase V: Final Products – Task 14: Update proposed modifications to AASHTO LRFD Bridge Design Specifications and AASHTO Guide Specifications for LRFD Seismic Bridge Design after consideration of review comments and prepare ballot items for AASHTO Highway Subcommittee on Bridges and Structures consideration. – Task 15: Prepare a final report that documents the entire research effort. 1.3. Review of Current Practice on the Design of Concrete-Filled Steel Tube (CFST) Drilled Shafts Design manuals related to design of drilled shafts, published online by 50 state DOTs, were reviewed at the beginning of this project in 2013, along with Drilled Shafts: Construction Proce- dures and LRFD Design Methods, published by FHWA (Brown et al. 2010), and AASHTO LRFD Bridge Design Specifications (2012) and AASHTO Guide Specifications for LRFD Seismic Bridge Design (2011), published by the American Association of State Highway and Transportation Officials (AASHTO). Parameters compared are (1) method of connection of oversized drilled or Type II drilled shafts with columns; (2) minimum embedment length of column cage required in drilled shafts; (3) contribution of casing in calculation of section structural strength; (4) mini- mum thickness of casing, and; (5) minimum diameter and concrete cover required for drilled

Background 7 shaft. In this report, AASHTO LRFD Bridge Design Specifications is referred to as AASHTO BDS, AASHTO Guide Specifications for LRFD Seismic Bridge Design is referred to as AASHTO SGS, and Drilled Shafts: Construction Procedures and LRFD Design Methods is referred to as FHWA (Brown et al. 2010). DOTs continuously update their design manuals and changes may have since occurred in some of the documents reviewed. When information was not found to be available online, or in cases of confusing informa- tion, design engineers working for the respective states were contacted by email. Combining the information available in online design documents and gathered through email, a list of design and detailing requirements for each DOT was prepared. Results for each of the main parameters described above have been tabulated and compared, to better understand current practices in the structural design of encased drilled shafts. Internal documents and undocumented practices not available in public domain may exist, but are beyond the scope of this study and not covered in this literature review. Overall, most states typically follow the design requirements in the above AASHTO or FHWA publications, sometimes with minor differences in practice, but 11 states have sig- nificantly different state-specific provisions related to the above structural designing aspects. Table A.1 in Appendix A provides the tabulated information for each of the five key design parameters listed above for those 11 states, compared to the reference provisions of AASHTO BDS (2012), AASHTO SGS (2011), and FHWA (Brown et al. 2010) presented at the top of the table. Footnotes to the table provide clarifications for state-specific practice and details. A sum- mary of those findings is presented in Section 1.3.1. Section A.2 of Appendix A lists the state- specific requirements of the 37 DOTs that primarily follow the standard documents published by AASHTO and FHWA; it also lists the source of the information collected for each state and highlights differences (if any) in specified values for parameters related to the structural design and detailing of drilled shafts. Most of these differences relate to minimum cover or diameter requirements, for drilled shafts with or without permanent casings. A brief summary of the comparison for all the DOT guidelines related to the above five structural design aspects of drilled shafts is provided in Section 1.3.1. Kentucky, New Jersey, and Wisconsin DOTs are still referring to the previous edition of Drilled Shafts: Construction Procedures and Design Methods (O’Neil and Reese 1999) in their design manuals. Maine DOT is still referring to the first edition of the FHWA publication series on drilled shafts (O’Neil and Reese 1988) in their design manual. Two DOTs, from New Mexico and Oklahoma, were not found to have any specified requirement for the design of drilled shafts in their publicly available design guidelines and have not emailed back any information in this regard in response to inquiries. Section 1.3.2 describes the concept of Type I and Type II drilled shafts, together with a description of some DOT requirements for minimum increases in diameter from column to shaft specified to develop plastic hinge at the top of shaft (including some DOT’s specified thick- ness increases that are provided without explicitly stating whether plastic hinge will form at top of the shaft or not). To support the findings presented in Sections 1.3.1 and 1.3.2, Section B.1 of Appendix B pro- vides typical drilled shaft details as given in the design manuals of the DOTs that included such information (namely, California, Washington, Illinois, Kentucky, Kansas, Montana, Massachu- setts, Nevada, Oregon, Pennsylvania, South Carolina, and Texas). States that did not provide such standard details either explicitly or implicitly referred to those provided in the FHWA and AASHTO publications (except for the two states mentioned previously that were not found to have any specified requirements for the design of drilled shafts).

8 Contribution of Steel Casing to Single Shaft Foundation Structural resistance 1.3.1. Requirements for Specific Design Parameters This section summarizes the findings from a comparison of AASHTO, FHWA, and DOTs design requirements on five key structural design parameters for drilled shafts (used in Table A.1 of Appendix A, where data is presented for the 11 states having significant differences in require- ments from AASHTO and FHWA). More specifically: • Section 1.3.1.1 summarizes requirements for the connection of Type II (or oversized) drilled shaft with column without pile cap. • Section 1.3.1.2 summarizes the minimum requirement of embedment length of column reinforcement, together with a brief review of recent research suggesting that the develop- ment length requirements specified by Washington State DOT (WSDOT) are effective for seismic applications and less conservative than those of AASHTO SGS (2011) and Caltrans SDC (2010). • Section 1.3.1.3 summarizes DOT practices with respect to contribution of thickness of steel casing when calculating structural resistance of encased drilled shaft. • Section 1.3.1.4 summarizes minimum casing thickness requirements. • Section 1.3.1.5 summarizes requirements for the minimum diameter of drilled shaft. 1.3.1.1. Connection Between Type II or Oversized Drilled Shaft and Column When non-contact splice must be used at the top of drilled shafts, there are two methods for connecting an oversized or Type II drilled shaft with column (Brown et al. 2010): • A column cage is extended into the shaft for some specific development length. • A “splice cage” is used, and additional lap splices are provided into the column. None of the DOTs have mentioned other methods for connecting oversized shafts with piers or columns. 1.3.1.2. Minimum Embedment Length Required for Column Reinforcement Except for the Washington, Arizona, Oregon, Montana, Pennsylvania, Kentucky, and South Carolina DOTs, which have defined their own criteria, state DOTs follow the AASHTO BDS (2012) and AASHTO SGS (2011) for determination of minimum embedment length of column reinforcement in shafts. As a result, five different criteria exist for determining the minimum embedment length of column or splice cage reinforcement extended into drilled shafts. 1. The AASHTO SGS (2011) states: “Column longitudinal reinforcement should be extended into oversized shafts in a staggered manner with the minimum embedment lengths of Dc,max+ ld and Dc,max+ 2ld, where Dc,max is the larger cross-section dimension of the column and ld is the development length in tension of the column longitudinal reinforcement bars determined in accordance with Article 5.11.2.1 of AASHTO BDS (2012) using expected values of material properties.” 2. The AASHTO BDS (2012) Articles 5.11.2.1 and 5.11.2.2 provide the reference equations for calculating the development length ld, which is either taken as-is or multiplied by 1.25 for applications in AASHTO Seismic Zones 2, 3, and 4, per requirement of Article 5.10.11.4.3. AASHTO requires that this value be further multiplied by 1.0, 1.3, or 1.7 for Class A, B, and C splices, respectively, where these classes are defined based on the percentage of longitu- dinal steel, As, spliced within required lap length, and the ratio of the provided As over the required As. The commentary in AASHTO’s Fifth Edition (2012) states that “the development length of column longitudinal reinforcement in drilled shafts is from WSDOT-TRAC Report WA-RD 417.1, title Noncontact Lap Slices in Bridge Column-Shaft Connections.” This statement was not included in prior editions.

Background 9 3. The approach in Section 8.4.9 of South Carolina DOT Seismic Design Specifications for High- way Bridges (2008) specifies that “longitudinal column reinforcement shall be extended into oversized shafts in a staggered manner with the minimum embedment lengths of 2Dc,max and 3Dc,max where Dc,max is the largest cross-sectional dimension of the column.” 4. Montana and Pennsylvania DOTs have specified embedment length of splice cages into drilled shafts as 30 db, where db is the diameter of the longitudinal bars. Typical cross-section details of drilled shafts provided by Montana and Pennsylvania DOTs are provided in Appendix B. 5. The approach taken by Arizona, Oregon, Kentucky, and Washington State DOTs requires an increase in the minimum development length calculated with AASHTO BDS (2012) Article 5.11.2.1 in proportion to the difference between diameter of shaft and diameter of column. However, the requirements from these 3 DOTs vary as follows: – Section 1.1.5.5 of the Oregon DOT’s Bridge Design and Drafting Manual (2013) states: “The splice region is (1.7Ldb + a) rounded up to the nearest 3 inches.” Ldb is the basic devel- opment length per AASHTO BDS Article 5.11.2.1, and a is 0.5 times the difference in the shaft spiral diameter and column spiral diameter. Refer to Figure B.13 of Appendix B for further details. A similar approach is adopted by the Kentucky Transportation Cabinet. Refer to Figure B.9 of Appendix B for a typical cross-section of a drilled shaft. – The Arizona DOT’s Bridge Design Guidelines (2011), Appendix A-Example 2.2, states: “Where the distance between spliced rebar exceeds 6 inches, the development length must be increased to reflect the lack of a contact splice. This is done by assuming a 1:1 distribu- tion between bars resulting in increasing the lap length by the distance of separation. . . . For the column/shaft splice, all the reinforcing is spliced in the same location. Since there is less than twice the required reinforcing a Class C splice is required.” – WSDOT follows the recommendations from McLean and Smith (1997) for determin- ing minimum development. The DOT’s TRAC Report WA-RD 417.1 states that column longitudinal reinforcement in drilled shafts is typically straight. Embedment shall be a minimum length equal to lns = ls + s, which is to be multiplied by 1.25 in seismic zone where column plastic hinging is expected, where � lns = minimum length of the column longitudinal reinforcement extended in drilled shaft, � ls = the larger of 1.7 × lac or 1.7 × ld (for Class C lap splice), � lac = development length from the AASHTO SGS, Article 8.8.4 for the column longitu- dinal reinforcement, � ld = tension development length from AASHTO BDS, Article 5.11.2.1 for the column longitudinal reinforcement, and � s = difference between diameter of shaft and diameter of column. WSDOT also specifies that the modification factor in AASHTO BDS Article 5.11.2.1.3 that allows ld to be decreased by the ratio of (As-required)/(As-provided) shall not be used in seismic applications, because plastic hinging is expected and all longitudinal steel will be yielding. It is also significant that WSDOT states that the requirements of AASHTO SGS (2011) for develop- ment length of column bars extended into oversized pile shafts should not be used in the more severe seismic regions. The FHWA (Brown et al. 2010) design guidelines refer to Articles 5.11.2.1 and 5.11.2.2 of the AASHTO BDS (2012). Although the above requirements vary, they agree that the non-contact splice length provided should be based on standard splice length plus offset distance (as shown by McLean and Smith 1997). More recent research by Murcia-Delso (2013) established that the minimum required embedment length to develop the full capacity of a bar can be taken as ld + s + c, where ld is the tension development length given in AASHTO BDS, s is the spacing between the longitudinal bars extending from the column and the shaft, and c is the concrete cover at top of the shaft,

10 Contribution of Steel Casing to Single Shaft Foundation Structural resistance based on experiments on specimens that exhibited satisfactory performance (i.e., “no anchor- age failure and damage in the shafts was limited to a local cone failure and splitting cracks near the column-shaft interface”) when designed following this rule. Their results from finite ele- ment analysis also support their experimental finding. Based on those findings, Murcia-Delso commented that the minimum criteria required for determining the non-contact development length specified in AASHTO SGS (2011) and Caltrans SDC (2010) was conservative. This sup- ported findings by Chang and Dameron (2009) who demonstrated, using finite element analy- ses, that the requirements of AASHTO SGS (2011) were conservative. 1.3.1.3. Contribution of Casing in Calculating Strength of Encased Drilled Shaft AASHTO BDS (2012) states that steel casing may be considered structurally effective in resisting axial loads and bending moments. FHWA (Brown et al. 2010) follows the AASHTO BDS (2012). As most state DOTs do not explicitly address the issues, it could be inferred that they do not disapprove of this practice when referring to AASHTO BDS (2012) while not explicitly providing state-specific criteria to the contrary. However, personal communications with DOT engineers revealed that some DOTs (e.g., Kansas and North Carolina DOTs) do not allow considering the contribution of casing to shaft resistance, while others (e.g., Kentucky and Virginia DOTs) advise to neglect the effect of permanent casing in calculating structural resistance. Nevada and South Carolina DOTs explicitly state that casing should not be consid- ered in determination of the structural resistance of the shaft, but “it should be considered for evaluating the seismic response of foundation.” 1.3.1.4. Minimum Thickness of Casing AASHTO BDS (2012), FHWA (Brown et al. 2010), and DOTs require that a minimum effec- tive thickness of casing be used for it to be considered structurally effective. The effective thick- ness is defined as the thickness of the steel pipe after considering an allowance for section loss due to the effects of corrosion. AASHTO BDS (2012) and Caltrans Bridge Design Specifications (2003) require a minimum thickness of 1⁄8 in. A few states were found to deviate from this mini- mum requirement, namely: • WSDOT and Kentucky DOT require 3⁄8 in. • Delaware DOT requires ¼ in. • Kansas mentions standard available casing thickness of 5⁄16 in. for 30–36 in. diameter shafts. • Missouri requires ½ in. for more than 6 ft. diameter shafts. FHWA (Brown et al. 2010) does not mention any minimum requirement explicitly. The required minimum thickness of casing can also be either governed by structural reinforcement or stability requirement or by strength required for driving. Minimum strength of casing required for driving is dependent upon site condition and driving method. More thickness is required for installation with vibratory or impact hammer. Typical thickness of casing is 2–2.5 in. when installed by rotary/oscillator method (Brown et al. 2010). 1.3.1.5. Minimum Diameter of Shaft Typical diameters of drilled shafts range from 3 ft to 12 ft (Brown et al. 2010). AASHTO BDS (2012) requires a minimum shaft diameter of 30 in. for visual inspection (which was also found to be explicitly specified by the Louisiana and South Carolina DOTs). A few states have different requirements, namely: • Kansas, Rhode Island, and Pennsylvania DOTs require 36 in. • Ohio DOT requires 42 in. • Florida and Georgia DOTs require 48 in. for non-redundant bridge drilled shafts.

Background 11 • Texas DOT requires 30 in. for specific girder bridges. However, 24 in. drilled shafts are com- monly used for a concrete slab span bridge. • Missouri DOT requires a length-to-diameter ratio in the range of 3 ≤ L/D ≤ 30. Typically, states also require that the diameter of drilled shafts be at least 6 in. greater than the column diameter. FHWA (Brown et al. 2010) states that the reinforcing cage shall be con- centric with the drilled shaft excavation, within a tolerance of 1-½ in. Typical required concrete cover for drilled shaft, which for column reinforcement is approximately 3 in., is specified to be approximately 6 in. This difference in cover thickness can be used to facilitate alignment of the cages (Brown et al. 2010). 1.3.2. Type I and Type II Shafts 1.3.2.1. Definition Drilled shafts are often described as Type I and Type II in design guides or specifications for seismic areas (e.g., Caltrans SDC 2010). In Type I designs, a longitudinal cage is extended from the shaft into the column without splices near the ground line. Constructability issues include the need to work with a cage extend- ing over a great length above the top of the shaft and large cranes to handle the taller cage. Casing is sometimes extended above grade. If the casing is fully extended over the height of the column, to create a pile bent, composite behavior is possible and many existing research results support the AASHTO equations already available to design the column/shaft as a composite member, even in absence of internal reinforcement. Many DOTs prefer to use reinforced concrete columns and terminate the casing at ground level, for various reasons, including concerns of corrosion of the casing above ground level. Examples of Type I details are shown in Figure 1.1. In Type I designs, maximum bending moment of the column is typi- cally reached below ground level. Figure 1.1. Schematics of Caltrans single column Type I and Type II shafts (Caltrans 2010).

12 Contribution of Steel Casing to Single Shaft Foundation Structural resistance In Type II designs, the drilled shaft is sized to have a greater moment resistance than the column it supports to ensure that plastic hinging develops in the column above the shaft. One advantage of this approach is that it allows inspection of the plastic hinge following an earth- quake (which would not be easily achievable for Type I designs where hinging is below ground level). An example of Type II detail is shown in Figure 1.1; a more detailed example is shown in Figure 1.2. In Type II designs, the column reinforcement is often extended into the shaft for the length necessary to develop the strength of both the column and the shaft reinforcement. 1.3.2.2. Minimum Diameter of Shaft Only two state DOT design guidelines (WSDOT and Caltrans) specifically mention requiring a minimum increase of 24 in. in shaft diameter with respect to column diameter when develop- ing a Type II behavior in order to achieve plastic hinging at the top of the shaft—although it is not implied that this is the only requirement to be met to achieve Type II behavior (Caltrans SDC 2010). Incidentally, some DOTs (Nevada, Kansas, Indiana, Montana, Pennsylvania, and South Carolina) provide typical shaft detail for oversized shafts without specifying a required mini- mum increase in shaft diameter required for developing plastic hinging at the top of the shaft. The Oregon DOT Bridge Design and Drafting Manual (2013), Section 1.1.5.5, requires that for shafts with a diameter less than 6 ft, maximum column size must be 1 ft smaller than shaft diam- eter, and for shafts with diameters greater than 6 ft, the required difference in diameter increases to 2 ft. Similarly, Nevada DOT’s NDOT Structures Manual (2008) states that the diameter of a drilled shaft shall be at least 1.5 ft greater than the largest dimension of supported single column. FHWA (Brown et al. 2010) also specifies that Type II drilled shafts shall have at least an 18 in. greater diameter than column. However, AASHTO BDS (2012) and AASHTO SGS (2011) do not have such minimum requirements for shaft diameter with respect to column diameter for developing plastic hinging at the top of shaft. The above requirements are generally applicable to drilled shafts, with or without casing. 1.3.3. Standard Details A number of states have provided standard details for drilled shafts. They are included in Appendix B for DOTs of the following states: • Illinois • Massachusetts • Indiana • Kansas • Kentucky • Montana • Nevada • Oregon • Pennsylvania • South Carolina • Texas • Washington 1.3.4. Review of the Other Countries’ Design Codes Requirements The design requirements of Concrete Structures Standard of New Zealand, Canadian Highway Bridge Design Code of Canada, and Design of Composite Steel and Concrete Structures–Part 2: General Rules and Rules for Bridges of Europe for design of the drilled shafts were also reviewed. These requirements are provided in Appendix B.

Figure 1.2. Washington State DOT Type II shaft construction details (WSDOT Bridge Design Manual 2012).

14 Contribution of Steel Casing to Single Shaft Foundation Structural resistance For expediency in this report, the Canadian Highway Bridge Design Code is shortened to Canadian Code, the Design of Composite Steel and Concrete Structures–Part 2: General Rules and Rules for Bridges is shortened to Eurocode, and the Concrete Structures Standard of New Zealand is termed the New Zealand Code. 1.4. Review of Design and Detailing Requirements for CFSTs Some American and international codes that include design provisions for CFST were reviewed. Parameters compared here are those relevant to the objective of developing the composite strength of encased drilled shafts. This review includes comparison of: (1) the tube diameter- to-thickness ratio limits specified to ensure development of plastic capacity; (2) the effective stiffness, EIeff, of CFST used to evaluate deflections, deformations, and buckling capacity; and (3) the strength of CFST. Reviewed American codes are Building Code Requirements for Structural Concrete (ACI 318-11); Specifications for Structural Steel Buildings (AISC 360-10); Seismic Provisions for Structural Steel Buildings (AISC 341-10); Drilled Shafts: Construction Procedures and LRFD Design Methods (Brown et al. 2010); AASHTO LRFD Bridge Design Specifications (2012); and AASHTO Guide Specifications for LRFD Seismic Bridge Design (2011). International codes compared are Canadian Highway Bridge Design Code (Canadian Stan- dards Association 2006); Specification for Design and Construction of Concrete-Filled Steel Tubular Structures, from the China Association for Engineering Construction Standardization (CECS 28:90 1991); Eurocode 4: Design of Composite Steel and Concrete Structures: Part 2: Gen- eral Rules and Rules for Bridges (British Standards Institution 2005); Design and Construction of Concrete-Filled Steel Tube Column System in Japan (Morino and Tsuda 2003); and Concrete Structures Standard of New Zealand (Standards New Zealand 2006). Since Standard for Struc- tural Calculation of Steel Reinforced Concrete Structures, 5th Ed. (in Japanese) by Architectural Institute of Japan (AIJ 2001) is not available in English, work from Morino and Tsuda’s 2003 paper, Design and Construction of Concrete-Filled Steel Tube Column System in Japan, has been used as a reference in this report. For expediency in this report, Building Code Requirements for Structural Concrete, American Concrete Institute is referred to as ACI; Specifications for Structural Steel Buildings is referred to as AISC 360; Seismic Provisions for Structural Steel Buildings is referred to as AISC 341; Specification for Design and Construction of Concrete-Filled Steel Tubular Structures from the China Association for Engineering Construction Standardization is referred to as Chinese Code; and Design and Construction of Concrete-Filled Steel Tube Column System in Japan is referred to as Japanese Code. The inelastic deformation capacity, compressive stiffness, global buckling resistance, and load capacity of the CFSTs are significantly greater than the corresponding steel tube without the concrete fill. The behavior of steel tubes depends on the stiffness and strength of the tube itself, and the presence of concrete fill helps control premature local buckling. Over the last three decades, a significant amount of research has been done worldwide to develop an understanding of the behavior of CFST. Hajjar et al. (2013) have developed a publicly available database that provides a summary of all known experimental, computational, and analytical research on com- posite members, connections, and frames (see the wiki “Steel-Concrete Composite Structural Systems” available at http://www.northeastern.edu/compositesystems). In this database, each previously conducted study is summarized, with focus on experimental setup and properties, analytical and computational methods, and key results from the work. The reader is referred to this database for more in-depth literature review related to behavior of CFST.

Background 15 Section 1.4.1 describes the minimum requirement of tube diameter (D) to thickness (t) ratio required by American and international codes. A minimum value of D/t is required to prevent premature local buckling of the steel tube and to allow developing the plastic moment capacity of CFST. Section 1.4.2 describes the method for calculating effective stiffness EIeff mentioned in Ameri- can and international codes. An effective stiffness is needed to calculate buckling capacity and to determine deflections of CFST. To support the findings of Sections 1.4.1 and 1.4.2, details of the design provisions along with code references are mentioned in Appendix C. 1.4.1. Requirement of Maximum Permitted D/t Ratio The design requirements presented in this section are those specified by various codes to ensure local stability of the steel tube used in CFST. By extension, they should apply to the design of encased drilled shafts if the intent is to develop their full composite strength. Hollow tubular members are susceptible to buckling both inward and outward. However, in the presence of a concrete core, steel tubes typically only buckle outward. This is a higher mode of buckling, which explains why the presence of the concrete core delays the onset of local buck- ling. After repeated cycles in inelastic buckling, the typical failure mode of CFST occurs by fracture of the steel tube at the location of local buckling (at the points of highest inelastic strains on the buckled wave). This local buckling and failure upon repeated cyclic loading therefore depends on the D/t ratio, where D is outer diameter of steel tube and t is thickness of steel tube. The start of local buckling is delayed when members have smaller D/t ratios; members with large D/t ratios can develop local buckling before the member yields, and limits on this ratio are typically prescribed in seismic design when this behavior is not desirable. Ductility and hysteretic energy dissipation of a CFST beam-column member also depends on the D/t ratio, with lower D/t ratios correspond- ing to greater ductility and hysteretic energy dissipation capacities. All codes control local stability of CFST by restricting the maximum permitted D/t ratio. For seismic applications, these limits are typically more stringent, as the ability to develop full plastic moment capacity directly depends on the D/t ratio since members with low D/t ratio are able to sustain larger strains and delay the onset of local buckling further after first yielding of steel. A summary of limits specified by different codes is presented in Table 1.1, together with the cor- responding limit values for two different values of the yield strength, Fy. FHWA (Brown et al. 2010) refers to AASHTO BDS (2012) in this regard. To document the tabulated values, formulas given in various codes (with respective code references) are provided in Appendix C. In the equations, • D = Diameter of encased shaft/composite column. • t = Thickness of casing. • E = Modulus of elasticity of steel. • Fy = Steel yield stress. For CFST members expected to develop plastic hinging during an earthquake, more strin- gent limits are typically applicable. For example, the AISC 341 high ductile limit is intended to ensure the development of large and stable plastic rotations, whereas the moderate ductile limit is for other flexural members only expected to undergo moderate plastic rotation (on the order of 0.02rad. or less) or for members that, although being part of a lateral load resisting system, are not acting as the primary energy dissipating elements. By analogy, casing in a Type I shaft design that would be expected to develop large plastic hinging in the CFST shaft could be

16 Contribution of Steel Casing to Single Shaft Foundation Structural resistance considered both highly or moderately ductile (depending on their expected plastic hinge rota- tions), while a Type II shaft could only be considered moderately ductile. The AASHTO SGS (2011) limits, as well as the Japanese and New Zealand ones, are intended to be used for seismic design, and the AASHTO BDS limit shown in Table 1.1 is for development of full plastic moment, but no special designation is provided for the other codes in that table. There is significant variability in the tabulated limit values, even for seismic applications, with values varying between 44 and 102 for Grade 50 steel. The limit provided by the AASHTO SGS commentary (2012) is 81 (the value of 0.044E/Fy provided in Table 7.4.2-1: Limiting Width– Thickness Ratios for “Round HSS in axial compression or flexure” is substantially more con- servative, and not applicable to CFST). CFST having higher D/t values develop more pinched hysteretic loops due to earlier and more severe development of local buckling. Code and Designation Related Code Provision* Limit for Fy = 50ksi Limit for Fy = 36ksi AISC 360 (non-seismic) Compact/ Noncompact 52 73 AISC 341 (seismic) Highly Ductile 44 61 Moderately Ductile 87 121 ACI (non-seismic) 68.1 80.3 AASHTO BDS Full Plastic Moment Capacity 48.2 56.8 AASHTO SGS (Commentary only) Ductile Elements 81.2 112 Eurocode (non-seismic) (Fy is in MPa) 61.3 85.2 Canadian CISC/CSA-S6 (Fy is in MPa) 81.2 112.8 Japanese Code (Fy is in MPa) 102.2 142 Chinese Code (Fy is in MPa) 70.2 82.7 New Zealand Code 68.1 80.3 * Unless specified otherwise, measurements are in U.S. units. Table 1.1. D/t ratio comparison.

Background 17 1.4.2. Effective Flexural Stiffness The effective stiffness of a CFST member is used for calculating its global buckling capacity, its period (used in dynamic analysis of its response), its deflections for compliance with possible drift and serviceability limits. In this section, equations for calculating the effective stiffness of CFST mentioned by different codes are reviewed. Provisions mentioned in codes for calculat- ing the effective flexural stiffness of CFST should, by extension, be applicable to encased drilled shafts that are intended to develop their full composite action. Again, significant variability exists in the value of this design parameter when comparing the equations specified by different codes. The general equation for representing effective stiffness in many codes is of a form equivalent to (1.1)EI E I CE Ieff s s c c= + where E is the modulus of elasticity of the respective materials, I is the moment of inertia, the subscripts “eff” refers to effective concrete section, s refers to steel, c refers to concrete, and C is a coefficient calibrated to represent the contribution of concrete to the total stiffness and having different values in different codes. For comparison, in AISC 360 (2010), C is expressed as C3, and is equal to: 0.6 2 0.9 (1.2)3C A A A s c s = + +     ≤ where A is the area and where the value of C in the Eurocode is 0.6. The commentary to the AASHTO SGS (2011) does provide the following two equations: 2.5 (1.3)EI E I E I s s c c = + 0.88 0.352 (1.4)EI E I A nA E Is s c s s s= +   ≥ where n = modular ratio. The first equation is a modified form of the equation given in Article 5.7.4.3 of AASHTO LRFD Specifications (2012), and the second equation is a modified form of the one given in Article 6.9.5.1 of AASHTO LRFD Specifications (2012). The effect of the variability of axial load on the effective stiffness value is not included in the equations provided in American codes. Canadian and Japanese codes do not specify equations for calculating of the effective flexure stiffness of CFST. The following equation is provided in the WSDOT (2012) design specifications: (1.5)EI E I C E Ieff s s c c= + ′ 0.15 2 0.9 (1.6)C P P A A A s s c ′= + ° + +   < 0.95 (1.7)°= ′ +P f A F Ac c y s where P is the applied axial load and P° is the compressive strength of the CFST section. This equation was proposed by Roeder et al. (2010) based on a comparison of results from 50 tests

18 Contribution of Steel Casing to Single Shaft Foundation Structural resistance on circular CFST and comparing various code provisions. Results showed that of all stiffness values calculated in accordance with American code provisions, the ACI provisions provided the best estimates of effective stiffness for flexural members, but that for members subjected to combined bending and compression it underestimated the stiffness. The values obtained by the AISC provisions overestimated stiffness of both CFST beams and beam-columns, but less so with increases in compressive load. The equation calibrated by Roeder et al. (2010) acknowledges the effect of axial load on bend- ing stiffness, as the additional concrete in compression under large axial compressive loads can also contribute to the column stiffness. Figure 1.3 compares the measured stiffness with those predicted by the Roeder et al. (2010) equation. The most recent version of WSDOT BDM (2016) as well as AISC 360 (2016) provide effective stiffness equations for reinforced concrete-filled steel tubes (RCFST). This will be presented in Section 3.2.7. 1.4.3. Strength of the CFST Shafts This section describes the difference in approaches of calculating flexural strength of CFST given in different codes. In Appendix C, the approach adopted by different national and inter- national codes for calculating the strength CFST is mentioned. In American codes, two approaches are found for calculating the strength of CFST members. One is a plastic stress distribution method (PSDM) and the other is a strain compatibility method. AISC allows use of either the PSDM or strain compatibility method for calculating the strength of CFST members. ACI uses a general strain compatibility method, essentially treat- ing the CFST member as a concrete column reinforced with a structural steel tube in addition to internal reinforcement. The AASHTO SGS (2011) and the Canadian code have adopted a plastic stress method proposed by Bruneau and Marson (2004) similar to the AISC PSDM. The AASHTO provisions only deal with CFST without internal reinforcement. Small variations exist between the various applications of each method. For example, there are three major differences between the ACI and AISC approaches: (1) AISC approximates the stress in the uniform block to be 0.95f ′c for CFSTs while ACI approximates the uniform strength as 0.85f ′c, (2) AISC uses the entire height of the stress block while ACI recommends that it be reduced by a factor β (β < 0.85 depending on concrete strength), and (3) ACI limits the axial strength to 80% of the squash load with an additional factor to account for eccentricity where AISC treats the instability explicitly. (Moon et al. 2013) Figure 1.3. Comparison of the measured stiffness to the stiffness predicted by the Roeder et al. (2010) proposed model.

Background 19 The coefficient of 0.95 is larger than the 0.85 used for the Whitney stress block calculation in order to take advantage of enhanced confinement of concrete due to presence of steel tubes. Figure 1.4 compares the AISC and ACI models. For comparison, the plastic stress distribution used in the Canadian code and in the AASHTO SGS (2011) uses the entire height of the stress block with full values of Fy and f ′c; corresponding strength equations and stress diagrams used are shown in Appendix C (and Figure 1.4a if it was modified to show f ′c instead of 0.95f ′c). Eurocode 4 also uses a plastic stress model similar to the AISC approach. However, in Euro- code 4, concrete stresses of 1.0 f ′c are used instead of 0.95 f ′c . The Japanese Code uses allowable stress design and is based on the elastic analysis of the structures. In Japan, such CFST sections are known as steel reinforced concrete (SRC). 1.5. Review of Finite Element Modeling Methods of Reinforced Concrete Shafts Some of the existing finite element modeling methods of reinforced concrete shafts and RCFSTs in available finite element software packages especially Abaqus (Simulia 2012) and LS-Dyna (LSTC 2013) were studied in this section. The details of this study are presented in Appendix D. In this research project, LS-Dyna software was used for the finite element modeling of reinforced concrete shafts and later for the modeling of the RCFSTs. Comparisons of the analyses results of reinforced concrete shafts modeled in Abaqus and LS-Dyna are provided in the subsequent chapter. Figure 1.4. Comparison of AISC and ACI models for predicting resistance of CFST. a) Plastic Stress Distribution Method (AISC). b) AISC Strain Compatibility Method. c) ACI Method (Roeder et al. 2010).