Bridge Superstructure Tolerance to Total and Differential Foundation Movements (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.

NCHRP Project 12-103 1 NCHRP Project 12-103 Bridge Superstructure Tolerance to Total and Differential Foundation Movements Final Report 1 Introduction 1.1 Motivation and Research Objectives Article 10.5.2.2 in AASHTO LRFD provides guidance for tolerable movements and movement criteria for highway bridges. The commentary, Article C10.5.2.2, states the following: “….angular distortions between adjacent foundations greater than 0.008 rad. in simple spans and 0.004 rad. in continuous spans should not be permitted in settlement criteria……” The referenced article and its commentary provide other information, but the above criteria are the only quantifiable “deemed to satisfy” type of guidance related to the effect of foundation movements on bridge structures. Taken literally, these criteria indicate that for a span of 100 ft, a LD support movement of 9.6 in. and 4.8 in. for simple span and continuous span, respectively, would be tolerable. These values are in addition to any uniform settlement over the span. Designating such large values as tolerable settlements run contrary to the sensibilities of bridge designers. Therefore, while current practice can vary significantly from state to state, it is not uncommon for bridge designers to be conservative and use arbitrarily small values of angular distortions. Depending on the specific case, this excessive conservatism can lead to a significant increase in cost (e.g. through use of deep foundations instead of more cost-effective shallow foundations) with no clear evidence that it produces a commensurate improvement in long-term bridge performance. The aforementioned issues related to the application of current angular distortion criteria are further compounded by the fact that the application of the SE load factor for support movements is either not used at all or not used consistently by bridge designers. To address these issues, Samtani and Kulicki (2016) developed step-by-step flow chart based processes for rational incorporation of foundation deformations in the AASHTO LRFD bridge design process. Their work also performed calibration of the SE load factor to incorporate uncertainty in predicted foundation movements and demonstrated the

NCHRP Project 12-103 2 incorporation of the aforementioned AASHTO LRFD criteria for angular distortion through a series of examples for two-, four- and five-span bridges for a wide range of foundation settlements. Ballot items for incorporation of the work by Samtani and Kulicki (2016) in Sections 3 and 10 of AASHTO LRFD have already been prepared by others. While the work by Samtani and Kulicki (2016) is valuable from the viewpoint of providing rational guidance to bridge designers for incorporation of foundation movements into the bridge design process its scope did not include an evaluation of the angular distortion criteria itself in the context of current AASHTO LRFD. The primary objective of this research study is to concentrate on the evaluation of the current “deemed to satisfy” angular distortion criteria and refine them to meet the needs of the modern bridge design requirements as per current AASHTO LRFD. Thus, this research study does not overlap with the work by Samtani and Kulicki (2016) but is complementary in the sense that once this research study is completed then any refined angular distortion or other similar criteria can be easily incorporated into the overall framework developed by Samtani and Kulicki (2016) that is currently being considered by AASHTO for inclusion in AASHTO LRFD. The purpose of this research was to develop a comprehensive understanding of the levels of support movements that bridges may tolerate before exceeding strength or service limit states. This research was devised to satisfy the following objectives: 1) Develop analytical procedures to objectively determine the acceptable levels of bridge foundation movements based upon superstructure tolerance considering AASHTO LRFD strength and service limit states (Phase II). 2) Propose revisions to the AASHTO LRFD Bridge Design Specifications that provide rational guidance for foundation movement limits that shall include vertical and rotational movements (Phase III and IV). Phase II focused on satisfying Objective 1 through (a) estimating tolerable support movements for common bridge types, (b) identifying the parameters that influence the level of tolerable support movement, and (c) developing simplified expressions for estimating tolerable support movement. In addition, this research examined the current AASHTO LRFD criteria, which was developed under previous specifications (i.e. AASHTO Standard Specifications), in light of the results obtained through (a). For the estimation of tolerable support movements, this research examined the following four types of support movements:

NCHRP Project 12-103 3 • LD movements – This type of support movement occurs when all bearing locations at a single support undergo the same amount of vertical translation. The tolerable level of this type of support movement may be limited by Strength and Service limit states or ride-ability criteria. • TD movements – This type of support movement occurs when a single support undergoes a vertical translation that varies linearly in the transverse direction. The tolerable level of this type of support movement may be limited by Strength and Service limit states or ride-ability criteria. • Horizontal movement – This type of support movement occurs when a support moves horizontally in the longitudinal direction of the superstructure. The tolerable level of this type of support movement is limited by the displacement capacity of the movement systems (joints and bearings). • Total (uniform) movements – This type of support movement occurs when all supports of a bridge undergo an equal vertical translation. The tolerable level of this type of support movement is limited by either clearance or ride-ability criteria. To accomplish the goals for Phase II, a large number of steel and PS concrete multi-girder bridge configurations and parameters were examined through a multivariate parametric study. This study employed an automated member sizing algorithm as well as automated three-dimensional (3D) FE model construction, simulation, and results extraction approaches to permit a large sample population (or “suite”) of bridges to be examined. Tolerable support movements were calculated for each sample and the results were compared to the current AASHTO LRFD guidance when applicable. The guidance was deemed conservative if the observed level of tolerable support movement was found to be greater than what is suggested by AASHTO LRFD. In contrast, the guidance was deemed unconservative if the observed level of tolerable support movement was found to be less than what is suggested by AASHTO LRFD. Objective 2 was satisfied in Phases III and IV of the project. The proposed revisions to AASHTO LRFD Bridge Design Specifications are presented in Section 10 of this report. These revisions are based upon the results presented herein. Expressions were developed for estimating the maximum tolerable LD support movement of steel and PS concrete multi-girder bridges. Expressions for TD support movements were not included in the proposed revisions as it was concluded, after discussion with the project panel, that TD support movements cannot easily be accounted for in the current state of practice for designing foundations.

NCHRP Project 12-103 4 1.2 Project Organization To satisfy the research objectives described above, the project activities were organized into four phases as shown in Figure 1-1. The specific objectives for each phase are provided in Table 1-1. Figure 1-1 - Tasks descriptions for each phase. Table 1-1 - Objectives of each phase. Phase Project Objectives I (a) Establish the state-of-the-knowledge related to LD and TD support movements

NCHRP Project 12-103 5 (b) Develop efficient and accurate modeling approaches capable of simulating the response of common bridge types to support movements (c) Examine the sensitivity of the initial set of parameters to responses generated from support movements (d) Revise the proposed work plan based on the results of (b) and (c) (e) Identify design provisions that may require revisions based on the results of this research program II (a) Develop estimates of tolerable support settlements for both strength and serviceability limit states for common bridge types (b) Identify the critical parameters that influence tolerable support movements (c) Develop simple expressions to estimate the level of tolerable support movement based on the identified influential parameters III (a) Develop recommendations and associated ballot items for the inclusion of Phase II results within the AASHTO LRFD Specifications (b) Identify the influence of the proposed recommendations on current practice IV Develop a comprehensive report documenting the entire research project 1.3 Research Summary As outlined in Table 1-1, Phase II focused on estimating tolerable support movements for common bridge types, identifying the parameters that influence the level of tolerable support movement, and developing simplified guidance or refining current guidance for tolerable support movement. Accomplishing these objectives requires a large number of bridge configurations and parameters (and combinations of these) to be examined. This was done through a multivariate parametric study. Task 2.1: Definition and Construction of 3D FE Bridge Models For the primary bridge types identified in Phase I (i.e. multi-girder steel and multi-girder PS concrete bridges) the project team automated both the member sizing (“design”) process as per the AASHTO LRFD specifications and the generation of 3D FE models (as defined through the examination of multiple model forms during Phase I). Based on a series of sensitivity studies conducted during Phase I, the suite of influential parameters shown in Table 1-2 were selected for inclusion within the parametric study. These parameters were divided into discrete and continuous parameters and sampled using a full factorial Design of Experiments (DoE) approach and Latin Hypercube Sampling (LHS) approach, respectively. A DoE experiment is an experiment that involves assigning discrete possible values or “levels” to parameters (e.g., span length) under consideration and evaluating all possible combinations of these “levels” across all parameters. Such an approach permits a thorough evaluation of the effect of

NCHRP Project 12-103 6 each parameter on the chosen response variable (e.g., tolerable settlement, moment, shear, etc.), as well as the effects of interactions between various parameters on the response variable. The LHS is a stratified method of sampling random numbers that attempts to distribute samples evenly over the sample space (i.e. practical upper and lower bounds for each parameter). Thus, the use of this hybrid (DoE and LHS) approach makes it virtually impossible for any interactions of factors to be missed as all possible combinations of parameters are accounted for and thereby a high level of confidence in the results is attained. The work done for Task 2.1 is further detailed in Section 3 of this report. The work plan and key action items for Phase II are shown in Figure 1-2. Task 2.2: Estimation of Maximum Tolerable Support Movements Using samples of parameters noted in Table 1-2, the suites of notional bridges were defined using the automated member-sizing software. FE models of these bridges were then constructed and analyzed for the dead load, superimposed dead load, and live loads defined by the limit states shown in Table 1-3. In addition to these demands, each model was analyzed for LD and TD support movement at each support (see Section 4.1 for detailed definitions of LD and TD support movement). Using these results the tolerable support movements for each limit state shown in Table 1-3 were computed. To ensure the results are independent of the specific sample size selected, a convergence criterion was defined and implemented to ensure that addition samples would not influence the results generated. The work done for Task 2.2 is further detailed in Section 4 of this report.

NCHRP Project 12-103 7 Figure 1-2 - Flow chart of key action items for Phase II of this project. Automate and validate LRFD design for primary bridge types (T2.1.2) Generate two independent samples of the primary bridge parameters (T2.1.1) Generate automated designs for each sample set (T2.1.2) Generate additional parameter sets for sample (T2.1.1) Update FE modeling software to accommodate primary bridge types (T2.1.3) Generate FE models for each design within each sample set (T2.1.3) Simulate live load, dead load, and support movement demands and extract key responses (T2.2) Did the independent samples converge? (T2.1.1) Evaluate each Limit State and compute tolerable displacements (T2.2) No Yes Identification of key parameters and trends (T2.3) Spot checking of additional bridge types (T2.4) Reporting (T2.5) Data Analysis Preliminary Activities Data Generation

NCHRP Project 12-103 8 Table 1-2 - Parameters investigated in this project. Parameter Bounds Notes Di sc re te P ar am et er s Bridge Type Steel and PS Concrete Multi-girder These bridge types were selected as they represent the most commonly designed and built structures Bridge Continuity Simple, 2-Span Continuous, 3-Span Continuous These levels of superstructure continuity allow for no continuity, continuity at one end of each span, and continuity at both ends of a span, respectively Co nt in uo us P ar am et er s Span Length 40 ft to 160 ft Typical span lengths for multi-girder bridges. Girder Spacing 5 ft to 12 ft Typical bounds of girder spacing for multi-girder bridges. Skew 0⁰ to 60⁰ Larger skew angles typically require advanced analytical methods. Bridge Width 36 ft to 72 ft Approximately 2 to 4 lanes. This nominal bridge width will be adjusted based on girder spacing. Span-to-Depth Ratio 20 to 30 These represent typical bounds on girder depth and govern the relationship between girder strength and stiffness.

NCHRP Project 12-103 9 Table 1-3 - Limit states investigated in this project. Limit State Description Strength I Flexure and shear limitations under basic load combinations related to normal vehicle use of the bridge Service II Flexure limitations intended to control yielding of steel structures under basic load combinations related to normal vehicle use of the bridge Service I & III Compression/tension limitations under the load combination for analysis of longitudinal compression/tension in PS concrete superstructures Task 2.3: Identification of Influential Parameters Once convergence was achieved the results were plotted and interpreted to understand and explain the underlying mechanisms. In addition, controlling limit states and members (interior or exterior girders) were identified. Both classical and exploratory data analysis techniques were employed to identify the parameters that influence superstructure tolerance to LD and TD support movements. Once identified, the influential parameters were used to develop a set of simplified expressions, modifications to the current AASHTO LRFD guidance. These expressions are provided with the proposed revisions in Section 10 of this report. The work done for Task 2.3 is detailed in Sections 5 and 6 of this report. Task 2.4: Spot Checking of Secondary Bridge Types In addition to the primary bridge types of multi-girder steel and PS concrete, limited parametric analyses were performed on a secondary group of structures to “spot check” if the findings of the primary study were applicable to other, less common, superstructure types. The goal of this task was to identify any differences between the primary (steel and PS concrete multi-girder) and secondary bridge types in either observed trends or levels of tolerable support movement by examining the governing combinations of bridge configuration parameters. The bridge types that were incorporated in this study (referred herein as “secondary” bridge types) include:

NCHRP Project 12-103 10 • Closed Steel Box Girders Bridges • Open Steel Box Girder Bridges (Tub Girder Bridges) • Cast-in-Place Concrete Multi-Cell Box Girder Bridges The work done for Task 2.4 is further detailed in Section 7 of this report. Functional Limitations on Support Movement Although not explicitly a part of the originally proposed tasks (shown in Figure 1-1), the Research Team investigated ride quality and clearance requirements, and developed a proposed approach to translating these performance requirements into generalized support movement criteria. This work is detailed in Section 8. Framework for Estimation of Tolerable Support Movement Finally, the approach used throughout this research to estimate tolerable support movements was summarized in a general form. It is anticipated that for unique bridges (beyond the scope of this project) as well as in cases where designers require more precision than provided by “deemed to satisfy” criteria, this refined analysis procedure would be employed. This work is detailed in Section 9. 1.4 Report Outline The outline of this report follows the work plan for Phase II detailed in Figure 1-2. A summary of each section is given. • Section 2: Current AASHTO Guidance Related to Support Movement presents a summary of the current AASHTO LRFD guidance and its origin. • Section 3: Definition and Construction of 3D FE Bridge Suites (T2.1) presents the processes used to define the sample populations of steel and PS concrete multi-girder bridges and build the 3D FE bridge suites for each sample population. • Section 4: Estimation of Maximum Tolerable Support Movements (T2.2) outlines the procedure developed and executed for determining superstructure tolerance to support movement. • Section 5: Steel Bridge Results (T2.2 and T2.3) presents the results of the computed tolerable support movements as well as an interpretation of the trends and influential parameters observed for steel multi-girder bridges.

NCHRP Project 12-103 11 • Section 6: Pre-Stressed Concrete Bridge Results (T2.2 and T2.3) presents the results of the computed tolerable support movements as well as an interpretation of the trends and influential parameters observed for PS concrete multi-girder bridges. • Section 7: Secondary Bridge Types (T2.4) presents the comparison of the spot check bridges to the findings of the primary study. • Section 8: Functional Limitations on Support Movements discusses limitations on tolerable support movements based on ride quality and clearance criteria. • Section 9: Refined Approach for Estimating Maximum Tolerable Support Movement presents an approach to estimating tolerable support movement using FE models in cases that fall outside the applicable ranges of the simple expressions identified herein. • Section 10: Conclusions and Recommendations summarizes the findings of this research and details the proposed revisions to AASHTO LRFD Bridge Design Specifications, which include expressions for estimating maximum tolerable support movement Appendices A through F include supplementary data related to each task of this research.