Dot 63530 DS1-1

IDAHO TRANSPORTATION DEPARTMENT
RESEARCH REPORT
A Precast Pier System for Accelerated

Bridge Construction (ABC) in Idaho
RP 281
By
Mustafa Mashal, Arya Ebrahimpour, Mahesh Acharya, Jared Cantrell, Corey

Marshall, Ali Shokrgozar
Idaho State University
Prepared for
Idaho Transportation Department
ITD Research Program, Contracting Services
Highways Construction and Operations
August 2021
Disclaimer
This document is disseminated under the sponsorship of the Idaho Transportation Department and the
United States Department of Transportation in the interest of information exchange. The State of Idaho
and the United States Government assume no liability of its contents or use thereof.
The contents of this report reflect the view of the authors, who are responsible for the facts and
accuracy of the data presented. The contents do not necessarily reflect the official policies of the Idaho
Transportation Department or the United States Department of Transportation.
The State of Idaho and the United States Government do not endorse products or manufacturers.
Trademarks or manufacturers’ names appear only because they are considered essential to the object of
this document.
This report does not constitute a standard, specification, or regulation.
A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 2

Technical Report Documentation Page
1. Report No. 2. Government Accession No. 3. Recipient’s Catalog No.
FHWA-ID-21-281
4. Title and Subtitle 5. Report Date
A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho August 2021
6. Performing Organization Code
7. Author(s) 8. Performing Organization Report No.

Mustafa Mashal (https://orcid.org/0000-0003-4654-0531), Arya Ebrahimpour,
Mahesh Acharya (https://orcid.org/0000-0002-3009-7099), Jared Cantrell,
Corey Marshall, and Ali Shokrgozar
9. Performing Organization Name and Address 10. Work Unit No. (TRAIS)
Idaho State University
Department of Civil and Environmental Engineering 11. Contract or Grant No.
921 S. 8th Ave., Mail Stop 8060 ISU-19-01
Pocatello, ID 83209-8060
12. Sponsoring Agency Name and Address 13. Type of Report and Period Covered
Idaho Transportation Department (SPR) Final Report
Highways Construction and Operations, Contracting Services, Research 01/01/2019 – 06/01/2021
Program 14. Sponsoring Agency Code
PO Box 7129 RP-281
Boise, ID 83707-7129
15. Supplementary Notes
Project performed in cooperation with the Idaho Transportation Department and Federal Highway Administration.
16. Abstract
Accelerated Bridge Construction (ABC) in seismic regions is currently a research focal point. ABC is bridge construction that
uses innovative planning, design, materials, and construction methods in a safe and cost-effective manner to reduce the onsite
construction time that occurs when building new bridges or replacing and rehabilitating existing bridges. The research at Idaho
State University included large scale experimental and analytical investigation of a new precast pier system for applications in
seismic zones. The precast pier system uses hollow structural tubes filled with concrete in plastic hinge locations. Extensive
testing on precast cantilever and bent piers was conducted to investigate seismic performance under quasi-static cyclic loads.
Results were compared against benchmark cast-in-place specimens. Testing showed superior performance of the precast pier
compared to cast-in-place. Limited analytical and some parametric studies are carried out to assist the designer in
computational work. Furthermore, flexural analysis, interaction diagrams, design and detailing considerations, construction
and assembly technology, limitations, inspection, and some proposed post-earthquake repair methodologies are presented.
17. Key Words 18. Distribution Statement
Accelerated Bridge Construction; Pipe connection; Large-scale Copies available from the ITD Research Program
testing, Pushover, Analytical, Parametric, Seismic
19. Security Classification (of this report) 20. Security Classification (of this page) 21. No. of Pages 22. Price
Unclassified Unclassified 205 None

Acknowledgments
The authors are thankful of assistance from Michael Johnson, P.E. (Project Manager) from the Idaho
Transportation Department, Ed Miltner, P.E. from the Federal Highway Administration (Idaho Advisor);
and Elmar Marx, P.E. (Senior Bridge Engineer) from the Alaska Department of Transportation for peer-
reviewing this report.
Committee
Each research project is overseen by a Technical Advisory Committee (TAC), which is led by an ITD project
sponsor and project manager. The TAC is responsible for monitoring project progress, reviewing
deliverables, ensuring that study objectives are met, and facilitating implementation of research
recommendations. ITD’s Research Program Manager appreciates the work of the following TAC members
in guiding this research study.
• Matt Farrar, P.E.
• Dan Gorley, P.E.
• Leonard Ruminski, P.E.
• Michael Johnson, P.E.
• Ned Parrish
• Ed Miltner, P.E.
• Craig Wielenga, P.E.

Table of Contents
Executive Summary..................................................................................................................................... 16
Summary ........................................................................................................................................... 16
Project Objective .................................................................................................................................. 16
Project Tasks ......................................................................................................................................... 16
Task 1. Literature review................................................................................................................... 16
Task 2. Experimental Investigation ................................................................................................... 17
Task 3. Numerical Investigation ........................................................................................................ 19
Task 4. Parametric Case Studies........................................................................................................ 20
Task 5. Design/Detailing Considerations and Recommendations .................................................... 20
Task 6. Discussions ............................................................................................................................ 21
Task 7. Final Project Presentation to ITD Bridge Section Staff.......................................................... 21
Task 8. Submission of the Final Report to ITD .................................................................................. 21
Report Overview................................................................................................................................... 21
1. Introduction.......................................................................................................................................... 23
Accelerated Bridge Construction.......................................................................................................... 23
Concept for a Precast Pier System in Seismic Regions ......................................................................... 25
2. Literature Review ................................................................................................................................. 30
Introduction.......................................................................................................................................... 30
Recent Developments .......................................................................................................................... 30
Conclusion ............................................................................................................................................ 39
3. Experimental Investigation................................................................................................................... 40
Cantilever Column: Cast-In-Place ......................................................................................................... 40
Introduction ...................................................................................................................................... 40
Prototype Structure .......................................................................................................................... 40
Testing Arrangement and Design Considerations............................................................................. 41
Construction...................................................................................................................................... 47
Material Properties ........................................................................................................................... 49
Test Setup and Instrumentation ....................................................................................................... 49
Testing Results .................................................................................................................................. 51

Summary ........................................................................................................................................... 60
Cantilever Column: Precast .................................................................................................................. 61
Introduction ...................................................................................................................................... 61
Overview ........................................................................................................................................... 61
Material Properties ........................................................................................................................... 62
Design of Precast Specimen .............................................................................................................. 62
Construction...................................................................................................................................... 71
Test Setup and Instrumentation ....................................................................................................... 73
Testing Results .................................................................................................................................. 73
Summary ........................................................................................................................................... 82
Cantilever Columns: Precast vs. Cast-In-Place ..................................................................................... 83
Bent Pier System: Cast-In-Place............................................................................................................ 89
Introduction ...................................................................................................................................... 89
Prototype Structure .......................................................................................................................... 89
Design of the CIP Specimen .............................................................................................................. 92
Construction...................................................................................................................................... 96
Testing Arrangement ...................................................................................................................... 101
Instrumentation .............................................................................................................................. 103
Loading Protocol ............................................................................................................................. 105
Material Properties ......................................................................................................................... 108
Testing Results ................................................................................................................................ 109
Summary ......................................................................................................................................... 120
Bent Pier System: Precast................................................................................................................... 121
Introduction .................................................................................................................................... 121
Overview ......................................................................................................................................... 121
Precast Bent System Design ............................................................................................................ 123
Construction.................................................................................................................................... 130
Testing Arrangement ...................................................................................................................... 134
Testing Results ................................................................................................................................ 135
Summary ......................................................................................................................................... 147
Bent Pier System: Precast vs. Cast-In-Place ....................................................................................... 148

4. Analytical Investigation ...................................................................................................................... 149
Cantilever Column .............................................................................................................................. 149
Cast-In-Place Pier ............................................................................................................................ 149
Precast Pier ..................................................................................................................................... 153
Bent Pier System ................................................................................................................................ 156
Cast-In-Place Bent ........................................................................................................................... 156
Precast Bent .................................................................................................................................... 157
5. Parametric Case Studies ..................................................................................................................... 158
Description of the Bridge.................................................................................................................... 158
Nonlinear Static Analysis .................................................................................................................... 159
Nonlinear Dynamic Analysis ............................................................................................................... 161
Cast-In-Place Bridge Response ........................................................................................................ 165
Precast Bridge Response ................................................................................................................. 167
Cast-In-Place Vs. Precast Bridge ......................................................................................................... 168
6. Design and Detailing Considerations.................................................................................................. 169
Introduction........................................................................................................................................ 169
Flexural Analysis ................................................................................................................................. 169
Interaction Diagrams .......................................................................................................................... 171
Embedment Length ............................................................................................................................ 172
Unbonded Length ............................................................................................................................... 175
Construction Technology and Assembly ............................................................................................ 175
Limitations .......................................................................................................................................... 178
Inspection ........................................................................................................................................... 178
Proposed Repair Methodologies ........................................................................................................ 179
Summary............................................................................................................................................. 183
7. Conclusions......................................................................................................................................... 184
Future Research Work ........................................................................................................................ 186
Cited Works............................................................................................................................................... 188
Appendices................................................................................................................................................ 190
Appendix A. [Design Calculations of the Column Specimen] ............................................................. 191
A.1 Cast-In-Place Column Calculations ........................................................................................... 191

A.2 Precast Column Calculations .................................................................................................... 192
Appendix B. [Design Calculations of the Bent Specimen] .................................................................. 193
B.1 Cast-In-Place Bent Calculations ................................................................................................ 193
B.2 Precast Bent Calculations.......................................................................................................... 194
Appendix C. [Grout Product Data Sheet]............................................................................................ 195
Appendix D. [Interaction Diagrams and Tables for the Precast Column] .......................................... 199
D.1 HSS Pipe Connection f’c = 4 ksi ................................................................................................. 199
D.2 HSS Pipe Connection f’c = 8 ksi ................................................................................................. 202
D.3 Cast-in-place Moment Capacities with f’c = 4 ksi...................................................................... 205

List of Tables
Table 1. Summary of Performance Factors for the Cast-In-Place and Precast Columns ............................ 19
Table 2. Summarized Comparison of CIP and Precast Bent........................................................................ 19
Table 3. Actual f’c values (psi) ..................................................................................................................... 49
Table 4. Summary of the Cast-In-Place Loading Protocol ........................................................................... 52
Table 5. Damage Observation from Testing of the CIP Column, Drift % .................................................... 56
Table 6. Summary of the Performance Points from Testing of the CIP Column......................................... 56
Table 7. Equivalent Viscous Damping Coefficients for Hysteretic Damping Component ........................... 59
Table 8. Actual f’c Values for the Precast Column (psi)............................................................................... 62
Table 9. Summary of Performance Values from Testing of the Precast Column ....................................... 75
Table 10. Damage Observation from Testing of the Precast Column, Drift (%) ......................................... 79
Table 11. Summary of the Performance Points from Testing of the Precast Column ................................ 79
Table 12. CIP Bent Test Day Compressive Strength .................................................................................. 108
Table 13. CIP Bent Test Day Split Tension ................................................................................................. 108
Table 14. CIP Bent Loading Protocol Summary......................................................................................... 111
Table 15. Precast Bent Loading Protocol Summary .................................................................................. 140
Table 16. Uniform Hazard Spectrum Data Points for Montpelier, Idaho ................................................. 163
Table 17. Earthquake Records form Different Regions ............................................................................ 165
Table 18. Maximum Base Shear and Displacement for CIP Bridge (Absolute Values Shown) ................. 167
Table 19. Maximum Base Shear and Displacement for Precast Bridge (Absolute Values Shown) ........... 168
Table 20. Common Corrosion Rates from WSDOT Bridge Design Manual ............................................... 170

List of Figures
Figure 1. Column-to-Footing Connection (Not to Scale)............................................................................. 26
Figure 2. Column-to-Cap Beam Connection and Cap Details (Not to Scale)............................................... 27
Figure 3. Thin-Walled Columns with Varying Thickness a) Uniform Column, b) Graded-Thickness, c)
Graded-Thickness Sections (courtesy of Al-Kaseasbeh and Mamaghani 2019) ......................................... 31
Figure 4. a) Column without Grouted Couplers, b) Column with Grouted Couplers, c) Cross Section
Designs (courtesy of Qu et al. 2015) ........................................................................................................... 32
Figure 5. Details for Bent Cap Pocket Connections (courtesy of Tazarv and Saiidi 2015a) ........................ 33
Figure 6. Precast Column for Testing Mechanical Bar Splices (courtesy of Tazarv and Saiidi 2015b) ........ 34
Figure 7. Hollow Precast Concrete Columns (courtesy of Kim et al. 2016) ................................................ 34
Figure 8. Grouted Splice Sleeve Connection Located in the Plastic Hinge of the Column (courtesy of Ameli
at el. 2016) .................................................................................................................................................. 35
Figure 9. Column to Footing Connection Alternatives: a) Socket Connection, b) Pocket Connection
(courtesy of Galvis and Correal 2017)......................................................................................................... 36
Figure 10. CIP and Precast Bent Column Details (courtesy of Mehraein and Saiidi 2016) ......................... 36
Figure 11. Low-Damage Seismic Design (courtesy of Mashal and Palermo 2019a) ................................... 37
Figure 12. Precast Bent with Member Socket and Grouted Duct Connections (courtesy of Mashal and
Palermo 2019b)........................................................................................................................................... 38
Figure 13. Pipe-Pin Connection Detail (courtesy of Zaghi and Saiidi 2010) ................................................ 39
Figure 14. Detail Views of a Typical Bridge in Idaho ................................................................................... 40
Figure 15. Scaling of a Typical Column, a) Full-Scale Typical Column, b) 0.25 Scale Typical Column ......... 41
Figure 16. Testing Arrangement ................................................................................................................. 42
Figure 17. ACI Testing Protocol ................................................................................................................... 42
Figure 18. Loading Protocol ........................................................................................................................ 44
Figure 19. Comparison of SAP-2000 and Response-2000 ........................................................................... 45
Figure 20. Cast-In-Place Footing Reinforcing Rebar Layout ........................................................................ 46
Figure 21. Cast-In-Place Test Specimen ...................................................................................................... 47
Figure 22. CIP Construction: a) Completed Rebar Cage, b) Footing Pour, c) Finished Footing, d) Footing
Curing, e) Column Formwork Setup, f) Finished Test Specimen................................................................. 48
Figure 23. Column Instrumentation............................................................................................................ 50
Figure 24. Typical Test Setup with Load Cells ............................................................................................. 51
Figure 25. Loading Protocol for the Cast-In-Place Column ......................................................................... 52
Figure 26. CIP Specimen at a) 0.45% drift, b) 2.67% drift, c) 5.32% drift, d) 8.87% drift, e) 10.64% drift, f)
Ruptured Bar ............................................................................................................................................... 53
Figure 27. Force vs. Displacement Plot for the Cast-In-Place Column........................................................ 54
Figure 28. Force vs. Drift Plot for the Cast-In-Place Column ...................................................................... 55
Figure 29. Backbone Curve for the Cast-In-Place Column .......................................................................... 55
Figure 30. Cast-In-Place Curvature Distribution ......................................................................................... 57
Figure 31. Dissipated Energy Plot for the CIP Column (1 kJ = 737.56 ft-lb) ................................................ 58
Figure 32. Corrected Area-Based Hysteretic Damping for the CIP Column ................................................ 60

Figure 33. Plastic Stress Distribution .......................................................................................................... 64
Figure 34. Embedment Length with Embedded Ring (courtesy of WSDOT) .............................................. 66
Figure 35. Precast Column Detail ................................................................................................................ 68
Figure 36. Precast Column to Footing Assembly ........................................................................................ 70
Figure 37. Precast Column Construction: a) Footing Rebar Cage, b) Column Rebar Assembly, c) Finished
Footing, d) Finished Column, e) Grout Flowing out of Grout Vent, f) Finished Test Specimen.................. 72
Figure 38. Location of Strain Gauges on the HSS Member ......................................................................... 73
Figure 39. Pipe Failure During the Testing Procedure. ............................................................................... 74
Figure 40. Loading Protocol for the Precast Column .................................................................................. 74
Figure 41. Precast Specimen During Testing at: a) 0.45% Drift, b) 2.67% Drift, c) 5.32% Drift, d) 8.87%
Drift, e) 13.27% Drift, f) Failed HSS Member .............................................................................................. 76
Figure 42. Force vs. Displacement plot for the Precast Column ................................................................. 77
Figure 43. Force vs. Drift Plot for the Precast Column................................................................................ 78
Figure 44. Backbone Curve for the Precast Column ................................................................................... 78
Figure 45. Precast Curvature Distribution .................................................................................................. 80
Figure 46. Dissipated Energy Plot for the Precast Column (1 kJ = 737.56 ft-lb) ......................................... 81
Figure 47. Area-Based Hysteretic Damping for the Precast Column .......................................................... 82
Figure 48. Backbone Curve for the Cast-In-Place and Precast Columns ..................................................... 84
Figure 49. Cumulative Dissipated Energy for Both Columns (1 kJ = 737.56 ft-lb) ...................................... 85
Figure 50. Dissipated Energy for Each Cycle for Both Columns (1 kJ = 737.56 ft-lb) .................................. 86
Figure 51. Corrected Area-Based Hysteretic Damping for Both Columns .................................................. 87
Figure 52. Residual Drift for the Cast-In-Place and Precast Columns ......................................................... 88
Figure 53. Elevation and Top View of SH-36 Bridge Over Bear River ......................................................... 90
Figure 54. Bear River Bent System Elevation View ..................................................................................... 90
Figure 55. CIP Bent Specimen ..................................................................................................................... 92
Figure 56. CIP Pier Detail............................................................................................................................. 93
Figure 57. CIP Pier Cross-Section ................................................................................................................ 94
Figure 58. CIP Cap Beam ............................................................................................................................. 95
Figure 59. CIP Cap Beam Cross-Section ...................................................................................................... 95
Figure 60. CIP Footing ................................................................................................................................. 96
Figure 61. CIP Footing Construction: a) Column Cage, b) Place Column Cage, c) Tying Top Mat, d) Anchor
Sleeves, e) Footing Pour.............................................................................................................................. 97
Figure 62. CIP Footing Placement ............................................................................................................... 98
Figure 63. Completed CIP Pier .................................................................................................................... 98
Figure 64. Cap False-Work .......................................................................................................................... 99
Figure 65. CIP Cap Construction: a) Cap Cage, b) Cap Cage Placement, c) Cap Complete False-Work, Cage,
and Formwork ............................................................................................................................................. 99
Figure 66. Completed Cap Pour ................................................................................................................ 100
Figure 67. CIP Bent Prepared for Instrumentation ................................................................................... 100
Figure 68. Actuator Cap Anchors .............................................................................................................. 101
Figure 69. CIP Bent Testing Arrangement ................................................................................................. 102

Figure 70. CIP Instrumentation Layout: Front Elevation View.................................................................. 103
Figure 71. CIP Instrumentation Layout: Top View .................................................................................... 104
Figure 72. CIP Instrumentation Layout: Rear Elevation View ................................................................... 104
Figure 73. ACI Loading Protocol ................................................................................................................ 106
Figure 74. Displacement Capacity of a Pier in a Bent with Fixed-Fixed Supports (Caltrans 2013) ........... 107
Figure 75. CIP Bent Loading Protocol........................................................................................................ 108
Figure 76. CIP Bent Testing: a) Cycle 1: Interface Cracking, b) Cycle 3: Hairline Crack Development, c)
Cycle 4: 0.0157 in. (0.4 mm) Crack Development, d) Cycle 8: Spalling and 0.0984 in. (2.5 mm) Cracking, e)
Cycle 9: Base Connection Spalling, f) Cycle 13: Spiral Exposure ............................................................... 110
Figure 77. CIP Bent Testing: g) Cycle 14: Longitudinal Exposure, h) Cycle 15: Longitudinal Rebar Fracture
.................................................................................................................................................................. 111
Figure 78. CIP Bent Force-Displacement Hysteresis ................................................................................. 112
Figure 79. CIP Bent Force-Drift Hysteresis ................................................................................................ 113
Figure 80. CIP Bent Force-Drift Backbone Curve ...................................................................................... 113
Figure 81. South Column: Top (D1-S – D4-S) ............................................................................................ 114
Figure 82. South Column: Bottom (A1-S – A4-S)....................................................................................... 115
Figure 83. North Column: Top (D1-N – D4-N) ........................................................................................... 115
Figure 84. North Column: Bottom, (A1-N – A4-N) .................................................................................... 116
Figure 85. CIP Bent Dissipated Energy per Cycle and Cumulative ............................................................ 117
Figure 86. Moment Curvature Curve (Caltrans 2013)............................................................................... 117
Figure 87. Bilinear Approximation for CIP Bent ........................................................................................ 118
Figure 88. Residual Drift of CIP Bent ......................................................................................................... 119
Figure 89. Proposed Connection: Precast Footing Detail ......................................................................... 121
Figure 90. Proposed Connection: Precast Column Detail ......................................................................... 122
Figure 91. WSDOT Pipe Embedment Cross-Section.................................................................................. 126
Figure 92. Precast Bent Cross-Section ...................................................................................................... 128
Figure 93. Precast Bent Cap Detail ............................................................................................................ 129
Figure 94. Precast Bent Specimen ............................................................................................................ 130
Figure 95. Embedded HSS Pipe Prepared for Installation......................................................................... 131
Figure 96. Precast Bent Footing Elements ................................................................................................ 131
Figure 97. Precast Bent Column Prepared for Concrete........................................................................... 132
Figure 98. Cap Rebar Cage Placement ...................................................................................................... 133
Figure 99. Precast Bent Column Erection ................................................................................................. 133
Figure 100. Precast Bent Cap Erection...................................................................................................... 134
Figure 101. Precast Bent Testing: a) Cycle 3: Hairline Cracking, b) Cycle 5: Vertical Hairline Cracking .... 135
Figure 102. Precast Bent Testing: c) Cycle 6: Spalling Development, d) Cycle 6: Gap Opening 3/16-inch, e)
Cycle 9: Slab Spalling Bottom, f) Cycle 11: Slab Spalling Top, g) Cycle 12: Slab Spalling North, h) Cycle 14:
Gap Opening ½ inch .................................................................................................................................. 136
Figure 103. Precast Bent Failure: a) Cycle 18: HSS Pipe Exposure, b) Cycle 19: Perpendicular Face
Cracking, c) Cycle 24: Confinement Failure, d) Cycle 24: Failure .............................................................. 137

Figure 104. Precast Bent Top Connection Post Test: a) Bulging HSS Pipe, b) HSS and Grout Separation, c)
HSS Bulging, d) HSS Bulging and Separation ............................................................................................. 138
Figure 105. Precast Bent Bottom Connection Post Test: a) South Column, b) North Column ................. 139
Figure 106. Cycle 24 Peak State ................................................................................................................ 139
Figure 107. Precast Bent Force-Displacement Hysteresis ........................................................................ 141
Figure 108. Precast Bent Force-Drift Hysteresis ....................................................................................... 142
Figure 109. Precast Bent Force-Drift Backbone Curve.............................................................................. 142
Figure 110. South Column: Top (D1-S – D4-S) .......................................................................................... 143
Figure 111. South Column: Bottom (A1-S - A4-S) ..................................................................................... 144
Figure 112. North Column: Top (D1-N – D4-N) ......................................................................................... 144
Figure 113. North Column: Bottom (A1-N – A4-N) ................................................................................... 145
Figure 114. Precast Bent Dissipated Energy Per Cycle and Cumulative (1 kJ = 737.56 ft-lb) ................... 145
Figure 115. Bilinear Approximation for Precast Bent ............................................................................... 146
Figure 116. Residual Drift of Precast Bent ................................................................................................ 147
Figure 117. Details of CIP Pier Specimen .................................................................................................. 149
Figure 118. OpenSees Model of the Experimental CIP Column................................................................ 150
Figure 119. Schematic of the Fitted Bi-Linear Curve for the Footing Bond-Slip Moment Rotation ......... 151
Figure 120. Combined Strain Vs. Number of Half Cycles to Failure Plot (Based on Brown & Kunnath 2000
and Hawileh, et al. 2009) .......................................................................................................................... 152
Figure 121. Numerical and Experimental Hysteresis Force-Displacement for CIP Column...................... 153
Figure 122. Details of the Precast Pier Specimen ..................................................................................... 154
Figure 123. Detailed Sections of the Model for Precast Column: a) Zone 1, b) Zone 2, and c) Zone 3 .... 154
Figure 124. Numerical and Experimental Hysteresis Force-Drift for Precast Column .............................. 155
Figure 125. Numerical and Experimental Hysteresis Force-Displacement for CIP Bent Pier ................... 156
Figure 126. Numerical and Experimental Hysteresis Force-Displacement for Precast Bent Pier............. 157
Figure 127. Plan View of the SH-22 over I-15 Bridge at Dubois (NTS) ...................................................... 158
Figure 128. Elevation View of the SH-22 over I-15 Bridge at Dubois (NTS) .............................................. 159
Figure 129. Dubois Bridge Nonlinear Model with Node Numbers ........................................................... 160
Figure 130. Dubois Bridge Nonlinear Model with Element Numbers ...................................................... 160
Figure 131. Base Shear Vs. Displacement for both CIP and Precast Column............................................ 161
Figure 132. AASHTO Seismic Coefficient Design Spectrum Constructed with the Three-Point Method . 162
Figure 133. Design Spectrum .................................................................................................................... 164
Figure 134. Displacement vs. Time for Landers Earthquake .................................................................... 166
Figure 135. Pushover vs. Dynamic Response for Landers Earthquake for the Bridge with CIP Columns in
Direction 2................................................................................................................................................. 166
Figure 136. HSS Member Moment Capacity Interaction Diagram ........................................................... 172
Figure 137. Embedment Length Determination Flow Chart ..................................................................... 173
Figure 138. Embedment Length Comparison for HSS Round (in.) ............................................................ 174
Figure 139. Embedment Length Comparison for Pipe (in.) ...................................................................... 174
Figure 140. a) Receiving HSS Member, b) Footing Reinforcement, c) Finished Footing, d) Column
Reinforcement, and e) Welded Centering Fins ......................................................................................... 177

Figure 141. Repair Methods, a) Embedded Ring Repair, b) Shear Stud Repair, and c) Weld Bead Repair
.................................................................................................................................................................. 179
Figure 142. Finite Element Model for Repair Methods ............................................................................ 180
Figure 143. Moment vs. Drift for Repair Methods ................................................................................... 180
Figure 144. Pedestal Rotation for Repair Methods .................................................................................. 181
Figure 145. Repair Methods with the Highest Results, a) ER-2x4t, b) SS-x4, c) wb-x3 and wb-x3-1 ........ 181
Figure 146. Proposed Repair Method for the ITD Precast Pier: a) Embedded Ring, b) Shear Stud, c) Weld
Bead .......................................................................................................................................................... 182
Figure 147. Mission Bridge, BC, Canada ................................................................................................... 182
Figure 148. Proposed Concrete Jacketing Repair Method for Precast Column Connection .................... 183

List of Abbreviations and Acronyms
AASHTO The American Association of State Highway and Transportation Officials
ABC Accelerated Bridge Construction
ACI American Concrete Institute
ASTM American Society for Testing and Materials
CFST Concrete Filled Steel Tubes
HSS Hollow Structural Section
ITD Idaho Transportation Department
PBES Prefabricated Bridge Elements and Systems
PVC Polyvinyl Chloride
SLAB Structural LAB
SIBC Slide-in Bridge Construction
UHPC Ultra-High Performance Concrete
WSDOT Washington State Department of Transportation

Executive Summary
Summary
In this project, a new precast pier with emulative connections is proposed for ABC in seismic regions. The
precast pier offers ample tolerance, simple construction, and better seismic performance in addition to
other benefits that will be discussed later. It uses structural steel tubes (pipes) filled with concrete in the
plastic hinge zones. The pipes are placed at the end of the column during prefabrication. During assembly,
the column pipe is inserted inside another pipe with larger diameter that is placed in the cap beam or
footing. The gap between the pipes are filled with high-strength grout on-site to complete the connection.
In this type of connection, the concrete filled steel pipe provides shear/flexural resistance as well as
concrete confinement in the plastic hinge. Results and observations from large-scale testing showed that
the proposed precast pier connection outperforms cast-in-place. Based on the results from this research,
ITD has incorporated the proposed connection in design of an upcoming bridge in Idaho that is located in
a seismic zone.
Project Objective
The objective of this project is to investigate seismic performance of a proposed pipe connection by ITD
and comparison against cast-in-place using the current force-based seismic design philosophy in the
United States.
Project Tasks
Task 1. Literature review
This task includes a review of the technologies available/implemented for precast pier systems in the
context of ABC from ITD and other Departments of Transportation (DOTs) in the United States. This task
includes the following sub-tasks:
a) Literature review on ABC in moderate-to-high seismicity
b) Available technologies for seismic resistant connections between precast elements
c) Comparison of past studied connections from the perspectives of transportation, seismic

performance, constructability, etc.
d) Experimental testing techniques for ABC bridge substructure systems
e) Review of the connections that have been deployed in actual ABC substructure systems in Idaho
and elsewhere

A significant amount of research has been conducted to improve and expand the ABC scope and
knowledge in seismic regions. The connections discussed include thin-walled steel columns, high strength
precast concrete, pocket connections, socket connections, grouted ducts, grouted splice sleeves,
mechanical bar splices, hollow precast reinforced concrete, precast shell column, dampers, and pipe-pin
connections. Research for ABC in seismic regions continues to be performed to improve and/or simplify
the construction of bridges. The research presented in this report aims to present a simple alternative to
the current grouted duct, pocket, socket, and pipe-pin connections.
Task 2. Experimental Investigation
This task is the main focus of the project and includes large-scale testing of precast and cast-in-place pier
systems (two bents and two cantilever columns) under quasi-static cyclic loading. The columns are
octagonal section. This task includes the following sub-tasks:
a) Design of specimens in accordance with AASHTO/ITD Bridge Design Manuals for experimental
testing
b) Design of test-setup
c) Construction of the specimens
d) Materials characterization
e) Assembly of the specimens in the Structural LAB (SLAB) at Idaho State University
f) Instrumentation and data acquisition system for experimental testing
g) Experimental testing of the specimens up to failure point
h) Disposal/demolition of the tested specimens
i) Processing of testing results (e.g., force-displacement hysteresis, moment-rotation etc.)
The cast-in-place specimen is intended to be the benchmark for the precast solution. The cast-in-place
column is designed following the latest version of the AASHTO LRFD Bridge Design Specifications. The
footing and column are constructed using traditional methods and materials. The column is tested under
quasi-static loading until failure. The column behaved in a ductile manner. The column achieved its design
base shear of 25-kip. The maximum base shear obtained during the testing procedure is 37.8-kip. The
maximum displacement achieved is 7.7 in. which corresponded to a ductility of just below 7.5. Damage to
the column during the test is within the first 18 in. of the column which is the expected plastic hinge length
for a well detailed and confined section. The total energy dissipated during the testing procedure is 456
kJ. A hysteretic damping plot showed that the column had similar values to a Takeda-Thin model up to a
displacement ductility of slightly lower than 3.5. Above that, the column had higher values of hysteretic
damping compared to a Takeda-Thin model.

The precast specimen is designed to match the capacity and ductility of the cast-in-place benchmark. The
precast footing and column are constructed incorporating HSS members. The Concrete Filled Steel Tube
(CFST) at the interface of the column and footing provides flexural and shear resistance as well as
confinement. The specimen is designed using the CFST equations provided in the WSDOT Bridge Design
Manual and the latest version of AASHTO LRFD Bridge Design Specifications. The contribution of the
unconfined concrete cover at the interface was ignored for the flexural design of the column. An
elastomeric pad is provided at the column and footing interface to prevent cracking during smaller drift
ratios and allow rocking of the column. The embedment length of the pipe is selected such to develop the
plastic capacity of the column pipe without any premature failure or pullout. An unbonded length of the
pipe is provided to distribute inelastic strain during larger drift ratios and to improve the low-cycle fatigue
performance of the pipe. The unbonded length region was designed to be the ductile link in the specimen.
The footing and other parts of the column are designed to be capacity protected regions. Testing results
show good performance of the precast column with enhanced energy dissipation. Damage to the column
during the test was observed to be within the first 18 in. of the column which thought to be the plastic
hinge region. The loss of the cover concrete occurred during the 4.42% drift ratio. This was later than what
was observed in the testing of the cast-in-place benchmark. The failure mechanism for the connection
started with an “Elephant-Leg” buckling of the pipe over the unbonded region, followed by the fracture
of the pipe on one side. There was a rapid decrease in the lateral capacity of the column when the buckling
increased. The maximum force obtained during the testing procedure was 41.2-kip which was
approximately 10% higher compared to the cast-in-place benchmark. The maximum displacement
achieved was 9.4 in. which is about 20% higher compared to the cast-in-place benchmark. The ultimate
displacement ductility is just below 6.5. The lower value of ductility compared to cast-in-place (7.5) was
the fact that the precast column had a higher yield point compared to cast-in-place. The total energy
dissipated during the testing procedure was 1,025 kJ. This was 2.25 times of the cast-in-place benchmark.
The hysteretic damping plot showed that the precast column, overall, had similar values of hysteretic
damping of Takeda-Fat and Ramberg-Osgood models. This is thought to be due to enhanced confinement
(e.g., Concrete Filled Steel Tube) and detailing considerations. The precast column reached a hysteretic
damping of 20% at the failure point. Table 1 presents the summary of performance factors obtained from
the experimental investigation of the cast-in-place and precast cantilever columns. Similarly, Table 2
presents the summary of performance factors obtained from the experimental investigation of the cast-
in-place and precast bent. The overstrength factor in the table are from experimental studies and can be
seen that they are comparable; however, having slightly higher overstrength factor for precast than CIP is
because CFST provide better confinement as well as producing more strain hardening. Further research
can be carried out to obtain the appropriate overstrength factors for the precast pipe connection when a
performance based, or displacement based designed is utilized.
Most importantly, the CIP has slightly higher displacement ductility at failure point because of the smaller
yield compared to precast. In reality, precast system is more ductile and provides better ductility. So, the
measure should be the energy dissipation (which is 6.21 times higher than CIP) and not the ductility.
Overall, the precast system provided great elastic and plastic deformation capacity.

Table 1. Summary of Performance Factors for the Cast-In-Place and Precast Columns
Performance Factors Cast-in-place column Precast column
Yield Displacement 0.9 in. (1.2% Drift Ratio) 2.6 in. (3.3% Drift Ratio)
Base Shear at Yield 33 kip 37.5 kip
Maximum Curvature at (7.99% Drift Ratio) 0.0038 radian 0.0036 radian
Ultimate Displacement Ductility 7.4 6.3
Ultimate Displacement Capacity 7.7 in. (9.9% Drift Ratio) 9.6 in. (12.3% Drift Ratio)
Ultimate Base Shear 37.8 kip 41.2 kip
Total Energy dissipated 456 kJ 1025 kJ
Table 2. Summarized Comparison of CIP and Precast Bent
CIP Precast Precast (% Based)
Max Force 66-kip 71.4-kip 108%

Max Displacement 4.14 in. 7.66 in. 185%
(4.94%) (9.15%)
Moment Capacity 460-kip-ft. 498-kip-ft. 108%
Initial Stiffness 56.7-kip/in. 38.7-kip/in. 68%
Initial Yield 0.5% 1.13% 226%
(0.42 in.) (0.95 in.)
Global Yield (Bilinear approximation) 0.7% 1.49% 209%
(0.596 in.) (1.246 in.)
Correlating Moment Capacity at Global Yield 392-kip-ft. 433-kip-ft. 110%
Energy Dissipation 342 kJ (252,246 ft-lb) 2,125 kJ (1,567,320 ft-lb) 621%
*[466 kJ (343,704 ft-lb)] *[136%]
Overstrength Factor 1.76 2.18 124%
Displacement Ductility (Ultimate Base Shear) 3.69 2.03 55%
Displacement Ductility (Failure Point) 7.48 6.02 80%
Note: *Precast Cumulative Dissipated Energy at Failure of CIP Bent
Task 3. Numerical Investigation
This task supplements the experimental investigation and includes analytical modeling of the precast and
cast-in-place piers in a software package. Appropriate analytical models are developed for the proposed
type of connection. This task includes the following sub tasks:
a) Finite Element Analysis of the proposed connection
b) Simplified analytical modeling of the proposed connection

c) Calibration/comparison of analytical models against experimental data
d) Hysteresis response under cyclic loads
The analytical modeling in this report is aimed to provide a practical tool for bridge engineers when
considering new connection details. Analytical models were created for the CIP column and precast
column with proposed pipe connection using the Open System for Earthquake Engineering Simulation
(OpenSees) software. To predict the experimental results, low-cycle fatigue data were included in the
OpenSees model. Results show good agreement between the analytical and experimental data. Similarly,
analytical models were created for the CIP bent pier and precast bent pier with proposed pipe connection
using OpenSees software. The results showed good agreement between the analytical and experimental
data. It was difficult to capture shear deformation and Bauschinger effects for the precast bent pier after
6% drift ratio.
Task 4. Parametric Case Studies
This task is limited in scope and includes the following sub tasks:
a) Building analytical models of an actual bridge in Idaho that is located in a high seismic region. The
models represent both cast-in-place and the proposed precast connection
b) Conduct nonlinear static (pushover) and nonlinear dynamic (time-history analysis) on the models
and comparing the global seismic response of the bridge (e.g., formation of plastic hinges, force-
displacement response etc.)
c) Comparing capacity versus demand for the proposed precast connection and cast-in-place
construction
d) Summarizing findings from global seismic analysis of the bridge structures
Results from nonlinear static and dynamic analyses showed that the bridge with the precast pier closely
emulates the cast-in-place bridge behavior. Both models had similar strength, ductility, and stiffness. Since
the location of the Dubois bridge was assumed to be in Montpelier, nonlinear dynamic analysis showed
that both cast-in-place and precast bridge models would reach their yield capacity under the four
earthquake records. This is to be expected since the design seismic accelerations for Montpelier are larger
than those of Dubois. In summary, existing nonlinear analysis techniques for cast-in-place bridges can be
used to perform a system level analysis of a bridge with the proposed ITD connection.
Task 5. Design/Detailing Considerations and Recommendations
This includes the following sub tasks:
a) Recommendations for design of bridges with the proposed connection type in seismic regions.
b) Recommendations for precasting and construction on-site

c) Proposing assembly sequence for precast elements on-site
d) Limitations on implementation of the proposed connection
e) To ease the designers in choosing the correct size of the pipe to match cast-in-place strength, a
series of interaction curves similar to those shown in Figure C6.12.2.3.3-2 of the AASHTO Specs
are developed based on existing information on Concrete-Filled Steel Tubes (CFSTs).
f) Potential post-earthquake repair methods
The flexural analysis of the pier can be conducted similar to CFSTs. Some design considerations and
recommendations for construction/assembly are proposed. A series of interaction diagrams were created
using the WSDOT Bridge Design Manual. The interaction diagrams show the moment capacity increasing
as the HSS member diameter and wall thickness increase. The interaction diagrams do not account for an
axial load but can be used as an effective tool in sizing the HSS/pipe to match a certain cast-in-place
capacity. Some limitations of the precast column were presented with respect to the capacity of the
connection, precasting technology, and the weight of the precast element. Several potential repair
methods using an embedded ring, welded beads, and concrete jacketing were discussed. The proposed
repair methods should be experimentally tested to verify their effectiveness in restoring stiffness and
strength to the precast pier. Some post-earthquake inspection and indication of damage to the HSS/pipe
was also discussed.
Task 6. Discussions
Based on existing literature and results/observations from experimental testing of the proposed
connection, some discussions are provided for the following:
a) Transportation, constructability, and inspection
b) Post-earthquake inspection methodologies and damage signs
c) Recommendations for future research work
Task 7. Final Project Presentation to ITD Bridge Section Staff
Task 8. Submission of the Final Report to ITD
Report Overview
This report is divided into seven chapters.
1) Chapter 1 presents introduction to ABC and the challenges associated with ABC in seismic regions.
This chapter discusses concepts for a precast pier system for ABC in seismic regions.
2) Chapter 2 presents literature review on ABC in seismic regions.

3) Chapter 3 presents the design, detailing, construction, and results from experimental testing of
large-scale precast and cast-in-place specimens.
4) Chapter 4 presents the details from the analytical investigation.
5) Chapter 5 presents the details from the parametric case studies.
6) Chapter 6 describes flexural analysis, design and detailing considerations, construction

technology, limitations, inspection, and some proposed post-earthquake repair methods for
bridges incorporating the ITD precast pier system.
7) Chapter 7 presents conclusions and recommendations for future research work.

1. Introduction
Accelerated Bridge Construction

For the past several years, Accelerated Bridge Construction (ABC) has been gaining popularity in the
United States and other parts of the world. The Federal Highway Administration defines ABC as “bridge
construction that uses innovative planning, design, materials, and construction methods in a safe and
cost-effective manner to reduce the onsite construction time that occurs when building new bridges or
replacing and rehabilitating existing bridges” (U.S. Department of Transportation/Federal Highway
Administration 2019). While ABC offers great advantages, applications of ABC in seismic regions have been
limited due to concerns about the seismic performance of the connection between precast elements. Past
earthquakes have shown vulnerability of the connection between precast elements due to limited
ductility and strength.
In 2017, ASCE published an infrastructure report card for the nation. The report states that there are
614,387 bridges in the United States. Approximately 4 in 10 of the bridges are 50 years old or older. 56,007
of the nation’s bridges were classified as structurally deficient in the year 2016, with approximately 188
million trips across a structurally deficient bridge each day. A structurally deficient bridge requires a
significant amount of maintenance and rehabilitation or needs to be replaced. Overall, the number of
structurally deficient bridges is decreasing, but at the same time the average age of the bridges in the
nation is rising, 43 years old, and approaching the end of their designed service lives. The ASCE 2017
infrastructure report card states that “The most recent estimate puts the nation’s backlog of bridge
rehabilitation needs at $123 billion” (ASCE 2017).
ASCE also published an infrastructure report for the state of Idaho in 2018. When the report was
published, there were 4,492 bridges in Idaho. Out of the total number of bridges, 393 were considered
structurally deficient, 45 of which were located on the state highway system, and 280 on local highway
systems. 837 of the bridges located along the state highway system had reached or exceeded their
designed 50-year life span. According to the ASCE report, the number of bridges on the state highway
system that are at or exceeding their designed 50-year life span will increase to 911 in 2021 (ASCE 2018).
To improve the next infrastructure report card, the bridges in the United States need to be either repaired
or replaced quickly and efficiently. One of the ways the bridges can be replaced quickly and efficiently is
by incorporating ABC methods. Fortunately, the state and highway officials are making a difference, but
more techniques need to be developed to hasten the process. In addition, bridges located in seismic zones
must be designed using ABC connections that will offer adequate performance during an earthquake.
There are several advantages to using ABC methods, which are evident in most ABC projects. One of the
advantages is limiting the duration of disruptions to traffic during construction because of the increased
construction speed. Another advantage is improved public perception. The public’s perception is
improved because they see the construction progressing faster than usual. Faster project delivery is also

an advantage to ABC. Due to the speed of the construction, the negative economic impact on local
businesses is reduced which is counted as another advantage. The use of precast concrete is an advantage
because formwork can be reused which reduces costs. Other noteworthy advantages include the
following: better quality control of the materials being used, lower machinery and equipment costs,
higher durability of the bridge elements, a reduced weight of the bridge structure, increased level of
safety, and less environmental impacts to the surrounding area (U.S. Department of
Transportation/Federal Highway Administration 2019). The major advantages of ABC are listed below:
1. Limited disruption to traffic during construction, especially in populated areas
2. Fast project delivery
3. Cost savings related to the use of formwork
4. More accuracy in bridge elements due to their prefabrication
5. Better quality control of the materials used in the bridge elements
6. Lower machinery and equipment costs
7. Higher durability of the bridge elements
8. Reduced weight of the bridge structure
9. Higher level of safety
10. Less environmental impacts
There have been plenty of examples for application of ABC in zones of low seismicity, however, application
of ABC in regions with moderate-to-high seismicity has been limited. This is due to uncertainty about the
seismic performance of the connections between the precast elements. Lessons learned from the past
earthquakes have shown specific vulnerability of precast connections in zones of high seismicity.
There are disadvantages associated with using ABC and precast concrete. One of the disadvantages is the
difficulty of maintaining tight tolerances. As projects develop with precast concrete, occasionally the
precast elements do not fit or align as intended. If the concrete members are not aligning, then a new
concrete element will need to be constructed and precious time and money are wasted. Misalignment
issues become more common with longer/taller precast elements. Another disadvantage is encountered
when the concrete elements have bars extruding from the concrete. In a case where bars are extruding
from the concrete, the bars need to be protected from being damaged during transportation and
installation. Another disadvantage is the cost of each concrete member. The process involved with precast
concrete elements requires a specific skillset to prepare, pour, transport, and install the member, which
is not the same as the traditional skillsets required for cast-in-place concrete. One more disadvantage
related to the overall cost occurs during the transportation of the precast elements. Concrete is heavy
and as a result, the cost associated with the transportation process is high.

When considering the seismic performance of an ABC structure, other challenges arise. Some of the
challenges are: inelastic deformations that develop during cyclic behavior, large concentrated forces that
develop where key elements join together, and a rapid loss of strength may be evident due to localized
buckling (Marsh 2018).
Precast connections are classified in two different ways, emulative and non-emulative. An emulative
connection mimics the seismic performance of a traditional cast-in-place connection while a non-
emulative connection improves the performance of the connection. The connections that are emulative
may incorporate bar couplers, grouted ducts, pocket connections, and member socket connections while
the non-emulative connections are segmental post-tensioned rocking piers and hybrid or dissipative
controlled rocking connections (Mashal et al. 2016). Emulative connections are usually more desirable
than a non-emulative connection because of the confidence with regards to the durability and seismic
performance of cast-in-place connections (Marsh et al. 2011).
Concept for a Precast Pier System in Seismic Regions

The Idaho Transportation Department (ITD) has proposed a new precast pier system for ABC in seismic
regions. The connection is somewhat similar in appearance to the pipe-pin connection investigated by
Zaghi and Saiidi (2009). However, it would be pipe-fixed connection designed to carry full moments
expected to develop at the top and bottom of the column. In the type of connection proposed in this
research project, the diameter and thickness of the steel pipe protruding from the column into the cap
would be selected to resist all design moments and shears at the column-to-cap beam interface. The
connection should have similar moment capacity as that provided by the column longitudinal rebars.
Figure 1 and Figure 2 presents sketches and details of the proposed connection and precast cap beam.

Figure 1. Column-to-Footing Connection (Not to Scale)

Embeded Steel Tube High Strength Precast Shell High Early Strength
in the Precast Cap In-Situ Grout In-Situ Concrete
Embeded Steel Tube

in the Precast Column
Figure 2. Column-to-Cap Beam Connection and Cap Details (Not to Scale)

The proposed connection incorporates hollow structural sections (HSS) made from steel located within
the plastic hinge of the column and is referred to as a pipe connection. HSS is installed at the end of the
precast column. Approximately half of the HSS is extruding out of the precast column and have centering
fins welded to the outside edge of the pipe. In the proposed pier system, the column main reinforcement
would be terminated at the top and the bottom with head terminators to improve the rebar development
length. In this case, precast column reinforcement would not protrude into the cap, eliminating the risk
of its damage during transportation and a cumbersome need of alignment with the grouted ducts or
grouted couplers. Instead, the column’s steel pipe would be easily “fed” into the larger diameter steel
pipe (sleeve) embedded within the precast hollow shell pier cap. Centering fins will be pre-welded to the
column pipe at the precast plant and will be sized to provide relatively snug fit inside the footing and the
cap pipes, eliminating the need for cumbersome cap alignment and additional external bracing for either
column or the cap after their erection. During the assembly on-site, these centering fins would be used to
fine-tune the cap final position and to secure the cap on top of column in a similar fashion as a Christmas
tree stand without the need of external bracing. The centering fins are not important structurally, but
rather are only for inserting the column into the footing or cap beam.
The advantage of the proposed system is that the only fine-tuning required is the correct setting of the
top of footing elevation and the correct positioning of the footing pipe. Once this is done, the rest of the
components will be automatically aligned in correct positions, like “Lego blocks”. Grout will be pumped
into the gap between the two pipes to secure the connection after the column is erected. For the column-
to-cap beam connection, the next step would be to fill the voids within the hollow cap shell using High-
Early Strength (HES) concrete. This would complete the cap installation. It is possible to use expansion
joint filler or elastomeric ring placed around the column’s pipe in order to isolate the cap and column
concrete around its perimeter and to allow for some level of deformations without crushing the concrete.
The column steel pipe may also be wrapped with expansion joint filler at the column-to-cap connection
in order to avoid stress concentrations and to allow for a small local yielding/gap opening without crushing
of concrete. Similar connection detail may also be incorporated in the column-to-footing connection as
shown in Figure 2. In this case, the concrete filled steel pipe would protrude from the bottom of column
into the pocket within the footing and then the pocket would be filled with HES grout. Similar expansion
joint fillers would be used around the pipe in this connection as well, as to allow for certain amount of
deformations without crushing of concrete.
The main advantage of the proposed connection lies in its simplicity, ample installation tolerance, ease of
erection, use of hollow/partially hollow/solid piers caps (based on project specific preference), and
potentially superior performance at reduced costs. The proposed connection does not require any
proprietary components and materials, making it more competitive and easier to construct. The pier
would be analyzed as if it were built by conventional methods. The size and thickness of concrete filled
connecting pipes would then be selected to closely match the column nominal moment resistance.
Experimental and analytical investigations need to determine the level of ductility of such concrete filled
pipe in comparison with cast-in-place reinforced concrete column and to adjust, as required, assuring the
capacity protected design of the cap beam and the footing.

The column-to-cap beam and column-to-footing connection allow some flexibility (gap opening in the
connections) during smaller earthquakes, thus reducing the chances of hairline cracking in the column
and yielding of the rebars/pipes. Self-centering is provided by gravity loads in this case.
The research project stems from the list of disadvantages associated with some existing ABC connections
such as grouted ducts or grouted couplers. Many of the current ABC methods that require couplers or
grouted ducts have difficulties connecting the concrete elements due to alignment issues. The proposed
connection offers ample tolerance compared to grouted ducts or grouted couplers.

2. Literature Review
Introduction
Several research projects have been devoted to the expansion of ABC methodologies. Some of the
institutions that have contributed to ABC research in seismic regions include the University of Nevada-
Reno, Texas A&M University, University of Canterbury in New Zealand, Idaho State University, and
Washington State University. This chapter will briefly discuss a variety of related literature which includes
thin-walled steel columns, high-strength precast concrete, pocket connections, socket connections,
grouted ducts, grouted splice sleeves, mechanical bar splices, hollow precast reinforced concrete, precast
shell column, dissipative controlled rocking, and pipe-pin connections. Most of these connections are
emulative cast-in-place. Emulative cast-in-place connections aim to achieve a similar seismic performance
to cast-in-place (e.g., formation of plastic hinges). Non-emulative cast-in-place connections such as
dissipative controlled rocking aim for a better and low damage seismic performance compared to
traditional cast-in-place (e.g., self-centering after the earthquake, replaceable energy dissipaters etc.).
Each type of connection has its advantages and disadvantages. Therefore, it is difficult to identify and
select a single ABC connection that works for all projects and types of seismicity.
Recent Developments
Al-Kaseasbeh and Mamaghani (2019) investigated the hysteretic behavior of circular thin-walled steel
columns. The control test column had a continuous wall thickness while the five experimental columns
had gradient wall thicknesses with volumes and outside diameters equaling that of the control specimen.
The columns are analyzed with a constant axial load and a bidirectional cyclic horizontal loading protocol.
The analysis of the members is completed using a finite-element model that allows the material and
geometric properties to vary along the length of the pipe, as shown in Figure 3. The gradient columns
displayed a significant increase in ultimate strength, ductility, and post-buckling behavior compared to
the controlled column with a uniform thickness.

Figure 3. Thin-Walled Columns with Varying Thickness a) Uniform Column, b) Graded-Thickness, c)
Graded-Thickness Sections (courtesy of Al-Kaseasbeh and Mamaghani 2019)
Ou et al. (2015) performed six cyclic tests on large-scale columns. These large-scale tests are part of
Taiwan’s new reinforced concrete research effort to develop standards for high strength reinforced
concrete structures to incorporate into high-rise building construction projects. The purpose of the
research is to investigate the seismic performance of high strength concrete in precast reinforced
columns. Each of the six columns is constructed from high-strength concrete, high strength longitudinal
bars, and transverse reinforcing bars. Two different construction methodologies are incorporated into the
design of the columns. The two methodologies are grouted coupler splices for the longitudinal
reinforcement in the plastic hinge zones and butt-welded splices for the transverse reinforcement, shown
in Figure 4. The effect of using grouted coupler splices and butt-welded splices in the plastic hinge zone is
compared to a cast-in-place column using conventional construction methods. After analyzing the results,
it appears that the performance of the precast grouted coupler splices is equivalent to the conventional
cast-in-place column. The results also showed the precast columns with butt-welded splices had a lower
ultimate drift capacity compared to the conventional hooked transverse reinforcement. The reduction in
ultimate drift capacity is a result of the longitudinal reinforcement buckling sooner than the traditional
transverse hook methodology.

Figure 4. a) Column without Grouted Couplers, b) Column with Grouted Couplers, c) Cross Section
Designs (courtesy of Qu et al. 2015)
Tazarv and Saiidi (2015a) investigated the use of pocket connections in high seismic regions. Although the
AASHTO Scan 11-02 had studies regarding the seismic performance of these pocket connections,
additional research was required to develop practical and reliable pocket connections. As a result of the
research project, pocket connections are found to be a useful way to join precast columns and pier caps
together. Through their literature review research, it was discovered that if the bent cap is properly
designed, the effects of pockets, with regards to the seismic performance of the cap, are negligible and
the connection simulated a cast-in-place column connection. Based on the lessons learned from the
research, five details for precast pocket connections are shown in Figure 5. Some of the previous precast
cap beam models that were constructed with a pocket connection yielded. The failure of the pocket
connections is a result of inadequate design. The use of precast pocket Alt-5 connections, shown in Figure
5, reduced the onsite construction time by 75%. The other pocket connections reduced the onsite
construction time by 42%.

Figure 5. Details for Bent Cap Pocket Connections (courtesy of Tazarv and Saiidi 2015a)
Tazarv and Saiidi (2015b) also investigated mechanical bar splices located in the plastic hinge. One of the
precast columns is shown in Figure 6. Their research revealed that the performance of a coupler is
dependent upon the loading rate and the manufacturer of the coupler. It is also concluded that careful
placement of large couplers in the column is crucial to obtaining the ultimate capacity of the coupler. In
most cases, the seismic performance of smaller couplers is determined to be adequate. Finally, it is
estimated that incorporating the mechanical bar splices at both ends of a precast column can reduce the
construction time by nearly 60%.

Figure 6. Precast Column for Testing Mechanical Bar Splices (courtesy of Tazarv and Saiidi 2015b)
Kim et al. (2016) developed two types of cast-in-place concrete-filled hollow precast concrete columns to
reduce the weight of large precast concrete columns and to increase the structural integrity of the cap
beam to column joints. The two proposed types of hollow precast concrete columns are shown in Figure
7. Cyclic loading tests were performed on the two columns and a conventional reinforced concrete column
to measure the seismic resistance of the columns. The results from the test procedure showed that the
stiffness and the maximum strength of the experimental test columns are comparable to the controlled
conventional reinforced column. However, the displacement ductility of the experimental columns is
lower. The energy dissipated from the experimental columns is slightly lower than the controlled
reinforced column. The experiment also concluded that the hollow core sections reduced the weight of
the precast elements by 62% and 51%.
Figure 7. Hollow Precast Concrete Columns (courtesy of Kim et al. 2016)
Ameli et al. (2016) performed tests on seismic column-to-footing connections using grouted splice sleeves.
The location of the connection is in either the column or footing, depending on the test specimen. The

grouted splice sleeves are selected to be tested because they have good construction tolerances and offer
a bond-related load transfer mechanism. Figure 8 shows a typical grouted connection located in the plastic
hinge of the column. To test the connections, three half-scale precast cantilever column test specimens
and one cast-in-place cantilever control test specimen were tested under quasi-static cyclic loads. The
results from the test demonstrated that the columns with grouted splice sleeves located in the plastic
hinge of the column had a lower displacement ductility capacity compared to the conventional cast-in-
place column. The results improved for the columns with grouted splice sleeves located in the footing. To
expand the test results, the reinforcing bars inside of the footing were intentionally de-bonded a length
equal to eight times the diameter of the reinforcing bars. Displacement ductility is any performance of
the connection past the yielding point without a significant decrease in the structure, or column, lateral
capacity.
Figure 8. Grouted Splice Sleeve Connection Located in the Plastic Hinge of the Column (courtesy of
Ameli at el. 2016)
Galvis and Correal (2017) investigated the characterization of the seismic behavior of a column foundation
connection for Accelerated Bridge Construction. This research indicated that grouted duct connections
have the potential to be improved. Several challenges arise during the design and construction of grouted
ducts, such as construction tolerances and large diameter bars. An alternative to the grouted duct
connections is a pocket connection. The pocket connection investigated is created from a corrugated steel
tube and non-shrinkage grout that bonds all of the longitudinal reinforcing bars together. The benefit of
using a pocket connection is large construction tolerance. A disadvantage to using a pocket connection is
a large amount of connecting materials and the foundation’s reinforcement details are complex. Another
alternative to grouted ducts is a socket connection. A socket connection allows the column to be installed
without any additional connecting members between the column and footing. To improve the bond
strength of the column and footing, the pre-cast column has a roughened surface where it is in contact
with the footing. A benefit of using the proposed socket connection is there are not any reinforcing bars
extruding out of the base which makes transporting and handling the precast elements easier. A
disadvantage of using the socket connection is the same as the pocket connection discussed prior (e.g.,
placing of footing rebars). A drawing of both the pocket and socket connections are shown in Figure 9.

Figure 9. Column to Footing Connection Alternatives: a) Socket Connection, b) Pocket Connection
(courtesy of Galvis and Correal 2017)
Mehraein and Saiidi (2016) examined the seismic performance of bridge column-pile-shaft pin
connections. The experimental test specimens for this project are two bent assemblies. The cap beam is
precast, which is connected to columns via a pocket connection. One of the columns in each of the bents
is cast-in-place, while the other is constructed using a precast shell. The reinforcement for each of the
columns is shown in Figure 10. The shake table testing results confirmed that the proposed precast shell
design satisfies the safety and performance requirements outlined in the code.
Figure 10. CIP and Precast Bent Column Details (courtesy of Mehraein and Saiidi 2016)
Mashal and Palermo (2019a) performed experimental work investigating a non-emulative low-damage
seismic design for ABC. The project intends to minimize the amount of damage imposed on precast bridge
elements during an earthquake. The experiment utilized dissipative controlled rocking connections
between the precast column and the footing or cap beam as shown in Figure 11. The use of the dissipative
controlled rocking connections keeps the traditional plastic hinges located at the column to footing and
column to cap beam interfaces from forming. The connection utilizes an unbonded post-tensioned tendon

located inside of the precast column to encourage the column to centering itself to its original position
after a seismic event. The dissipaters located on the exterior of the column are used to dissipate the
seismic kinetic energy. A steel jacket armor is used to increase the confinement of the concrete and reduce
any damage that may occur during a seismic event. After several cycles of quasi-static testing of the
connection, there was no damage or measurable post-seismic displacement. As a result of the
experiment, the design was used in the Wigram-Magdala Link Bridge located in Christchurch, New
Zealand. The bridge continues to stand today after enduring through a 7.8 magnitude earthquake in 2016,
just months after it was built. Although, the epicenter of the earthquake was far away from the bridge,
the bridge is expected to perform well during future seismic events.
Figure 11. Low-Damage Seismic Design (courtesy of Mashal and Palermo 2019a)
Mashal and Palermo (2019b) also performed experimental work investigating the use of two different
types of connections in a single precast column. The two connections are a member socket connection at
the base connecting the columns to the footings and a grouted duct connection at the top connecting the
columns to the cap beam. A half-scale test specimen is constructed with the intent to simulate a cast-in-
place connection. Quasi-static cyclic loading is used during testing. The results from the experiment
showed similar amounts of energy being dissipated in the plastic hinges of the columns. Several large
cracks developed where the member socket connections are located and only a few large cracks at the
grouted duct locations. The member socket connections exhibited a larger amount of strength
deterioration than the grouted duct connections which is a result of the starter bars intentionally being
de-bonded a length of 100 mm during the construction. The results indicate that including an unbonded
length of the starter bars at the connection can reduce the amount of spalling and strength degradation
in the plastic hinges. Figure 12 depicts an accurate representation of the tested column.

Figure 12. Precast Bent with Member Socket and Grouted Duct Connections (courtesy of Mashal and
Palermo 2019b)
Zaghi and Saiidi (2010) experimented with the use of a pipe-pin connection. A pipe-pin connection is
essentially a concrete-filled steel tube embedded at the end of a column. More details of the connection
are shown in Figure 13. The observations and conclusions from this research are numerous. An
observation for the experimentation is that the column can rotate significantly without altering the
performance of the connection. The mode of failure that is assumed when designing the test specimens
is a shear failure. However, if the pipe thickness is large the connection can fail in shear or result in bearing
failure of the concrete surrounding the pipe. One of the conclusions from the experiment is that the
bearing strength of concrete against the pipe is two to six times the axial compressive strength of the
concrete. After the experiment had concluded, the pipe-pin connection was disassembled, and the pipes
were straight, intact, and damage-free; while the steel exterior can receive small dents located where the
pipe came in contact with it. The finite element analysis used to model the pipe-pin connections returned
results that are within 5% of the actual experimental data.

Figure 13. Pipe-Pin Connection Detail (courtesy of Zaghi and Saiidi 2010)
Conclusion
A significant amount of research has been conducted to improve and expand the ABC scope and
knowledge in seismic regions. The connections discussed include thin-walled steel columns, high strength
precast concrete, pocket connections, socket connections, grouted ducts, grouted splice sleeves,
mechanical bar splices, hollow precast reinforced concrete, precast shell column, dampers, and pipe-pin
connections. Research for ABC in seismic regions continues to be performed to improve and/or simplify
the construction of bridges. The research presented in this report aims to present a simple alternative to
the current grouted duct, pocket, socket, and pipe-pin connections.

3. Experimental Investigation
Cantilever Column: Cast-In-Place
Introduction
This section presents the design, construction, and testing of a cast-in-place cantilever column. The
purpose of a cast-in-place cantilever column is to set a benchmark to compare the experimental precast
cantilever column. The design discussed in this chapter follows the 2017 AASHTO LRFD Design
Specifications (AASHTO 2017). This chapter also presents a prototype structure and the testing
arrangement for both the cast-in-place and precast columns.
Prototype Structure
Details of a typical bridge located in Idaho, US-95 over US-20/26, are shown in Figure 14 and are used to
determine reasonable dimensions for the prototype structure. The prototype structure can be assumed
to be built in South-East Idaho which is the most seismically active region in the state. To obtain the height
and diameter of the prototype structure, the typical bridge details are scaled by a factor of approximately
0.25. The 0.25 scale is used to accommodate for the height restraints within the Idaho State University
Structural LAB (SLAB). Applying the approximately 0.25 scale to the typical drawings reduces the height
from 40 ft-10 in. to 10 ft-2.5 in. and the diameter from 6 ft-0.75 in. to 1 ft-6.2 in., as shown in Figure 15.
Figure 14. Detail Views of a Typical Bridge in Idaho

Figure 15. Scaling of a Typical Column, a) Full-Scale Typical Column, b) 0.25 Scale Typical Column
After the typical drawings are appropriately scaled, the dimensions of the prototype column and footing
are determined. The prototype structure will have an overall height of 10 ft-4 in. The footing size is 4 ft. x
4 ft. x 3 ft. (L x W x H). The logic behind the proposed dimensions is a result of the SLAB’s strong floor
sleeve spacing and to ensure the footing will not fail before the column. The column is octagonal, having
a diameter of 18 in., and a height of 7 ft-4 in. measured from the top of the footing up. Using a footing
height of 3 ft., the distance from the top of the footing to the center of the actuator is 6.5 ft.
Testing Arrangement and Design Considerations
The 0.25 (1/4th) scale cantilever columns are tested via a uniaxial lateral load. Using a uniaxial load to test
the column represents a seismic event pushing and pulling the piers in the transverse direction. In this
type of loading, the abutments are assumed to resist the loads in the longitudinal direction of the bridge.
A 50-kip axial load is applied to the column throughout the entire testing procedure. The axial load is
expected to vary slightly and is kept as close to a constant value as possible (deviation of less than
approximately 5%). The deviation of the axial load is inconsequential to the testing results. This axial force
50 𝑘𝑘𝑘𝑘𝑘𝑘𝑘𝑘
corresponds to about 5% of the ultimate axial capacity of the column (axial ratio in % = = 5%) where
𝐴𝐴𝑔𝑔 𝑓𝑓𝑐𝑐′
Ag is the gross cross-sectional area of the column and fc’ is the compressive strength of the concrete. The
footing is assumed to be rigidly fixed to the ground (e.g., no soil-structure interaction). The appropriate
loads are shown in Figure 16.

Figure 16. Testing Arrangement
The loading protocol established for the experiment is obtained from the American Concrete Institute
(ACI) (ACI Committee 374 2013) which is a quasi-static cyclic loading protocol. The loading protocol begins
at 0.5 of the yield displacements then proceeds to 1, 2, 3, 4 times of the yield displacement until failure,
as shown in Figure 17.
Figure 17. ACI Testing Protocol

The testing procedure is stopped when the column lateral capacity is reduced to 50% of the ultimate
lateral force. The yield displacement is calculated using the following equation from Priestley et al. (2007):
2
𝜑𝜑𝑦𝑦 �𝐻𝐻 + 𝐿𝐿𝑠𝑠𝑠𝑠 �
∆𝑦𝑦 =
3
Where:
𝜑𝜑𝑦𝑦 = Yield curvature

H = Height of the column, inches
Lsp = Strain penetration length, inches
The yield curvature is calculated using the equation shown below:
2.25�𝐹𝐹𝑦𝑦𝑦𝑦 �
𝜑𝜑𝑦𝑦 =
𝐸𝐸𝑠𝑠 𝐷𝐷
Where:
Fye = Effective yield strength of the steel, i.e., 110% of yield strength of the steel, ksi
Es = Modulus of elasticity of the longitudinal rebars, ksi
D = Column diameter, inches
The strain penetration is calculated using the following equation:
𝐿𝐿𝑠𝑠𝑠𝑠 = 0.15�𝐹𝐹𝑦𝑦𝑦𝑦 �𝑑𝑑𝑏𝑏𝑏𝑏
Where:
dbl = Diameter of the longitudinal rebars, inches

Fye = Effective yield strength of the steel, i.e., 110% of yield strength of the steel, ksi
Yield displacement equation becomes:
0.000284 ∗ (78 + 7.7425)2

∆𝑦𝑦 = = 0.692 𝑖𝑖𝑖𝑖.
3
Figure 18 shows a graphical representation of the loading protocol used during the testing procedure. The
loading protocol from ACI is primarily developed for buildings. There is no load protocol for bridges at the
moment because it depends on seismicity. So, the loading protocol that is used, is the minimum required
from ACI. It is common in some places, e.g., Alaska uses 3-cycles; however, the idea is to account for
strength degradation from the test. This can be achieved from the loading cycle as in Figure 18. The
loading rate of the lateral actuator is 1 mm/sec with appropriate pauses to allow adequate time to observe
and record any changes in the column during testing.

Figure 18. Loading Protocol
Once the dimensions of the prototype structure have been established, the reinforcing steel is
appropriately selected. The design of the reinforcing steel located in the column and footing is in
accordance with the 9th edition of AASHTO LRFD Bridge Design Specifications (AASHTO 2020).
The column base shear is selected so it does not exceed the capacity of the equipment in the SLAB. The
base shear force has been selected to be 25-kip. Using this base shear, the longitudinal reinforcing has
been determined to be twelve #6 rebars equally spaced in a circular pattern. For confinement purposes,
a spiral having a pitch of 1.5 in. is used along the entire length of the column except where the footing
reinforcing crosses the longitudinal bars. Where the footing reinforcement crosses the longitudinal bars,
a mechanical splice is used on each side of the reinforcing bars. The minimum cover requirements of the
spiral are 1.5 in. Guidelines on rebar hooks, bends, and development lengths are conservatively
considered with all the rebar in the column.
The footing has reinforcing bars in each direction on both the top and bottom using #6 rebars. The rebar
will have a spacing of roughly 3.5 in. between parallel bars and a minimum cover of 2 in. Guidelines on
rebar hooks, bends, and development lengths are conservatively considered with all the rebar in the
footing.
The moment capacity of the proposed cantilever column has been determined for a base shear of 25-kip.
In the calculations, an 18 in. diameter column having 1.5 in. of cover, twelve #6 rebars, with #3 spirals are
used. The compression strength of concrete, f’c, is taken to be 4 ksi. The rebar properties are 60 ksi for the
yield strength and 29,000 ksi for the modulus of elasticity. Using Response-2000 and SAP-2000 the yield
moment capacity of the round column is calculated to be approximately 165 kip-ft without an axial load.

Adding the 50-kip vertical axial force the moment capacity of the column increases to 187 kip-ft. A
comparison of the two programs, before the axial load, is shown in Figure 19. Using a moment arm of 6.5
ft., the shear demand from the moment capacity is approximately 25-kip which is the targeted base shear.
Referring to Figure 19, the reason why Response-2000 obtained a lower curvature is that the program
does not have the capability to introduce spiral stirrups. Response-2000 thinks of the shear reinforcing as
hoops compared to spirals which provide more confinement and ductility. Response-2000 does not have
a confined model and is not typically used for seismic. In terms of capacity, the values from both programs
are comparable. It should be noted that SAP-2000 uses a fiber model to compute the moment-curvature.
A fiber model is a model with the capacity to perform an analysis with nonlinear behavior distributed
across the cross-section of the element in the model.
Figure 19. Comparison of SAP-2000 and Response-2000
The parameters used to calculate the moment capacity of the footing are f’c equal to 4 ksi, ten #6 rebars
in both the top and bottom reinforcing layers with 2 in. of cover, rebar yield strength of 60 ksi, and a
modulus of elasticity of 29,000 ksi. Using these values, the moment capacity is calculated to be 1,000.2
and 998.5 kip-ft using SAP2000 and Response2000, respectively. An image depicting the footing is shown
in Figure 20.

Figure 20. Cast-In-Place Footing Reinforcing Rebar Layout
In summary, the column is octagonal with a diameter of 18 in. The longitudinal reinforcement is comprised
of twelve #6 rebars equally spaced and a #3 spiral having a pitch of 1.5 in. to confine the concrete. The
calculated moment capacity of the column has been determined to be 165 kip-ft.
The footing is 4 ft. x 4 ft. x 3 ft. with ten #6 rebars reinforcing the top and bottom in each direction while
maintaining a spacing of approximately 3.5 in. between the bars and 2 in. of cover. The moment capacity
of the footing has been calculated to be roughly 1,000 kip-ft. Comparing the column moment capacity to
the footing moment capacity, the column should reach its ultimate strength before the footing is yielded.
Figure 21 provides a visual representation of the test specimen.

Figure 21. Cast-In-Place Test Specimen
Construction
Now that the design criteria have been determined, the construction of the cast-in-place cantilever
column is next. To begin, the formwork is built. After the formwork is complete, rebar is cut to the required
lengths and bent appropriately. Next, the rebar is tied together, and the formwork is secured around the
tied rebar (Figure 22a). Once the rebar and formwork are set, the concrete for the column footing is
poured via Pocatello Ready Pour and ISU civil engineering students as shown in (Figure 22b). The footing
is allowed to cure for three days before the formwork is removed. After the formwork is removed the
concrete is covered with burlap and plastic and wetted daily to allow the concrete to continue curing in
the most ideal conditions (Figure 22d). Once the footing reached seven days of curing, the footing is
moved to the structural laboratory, the column formwork is assembled, column cap reinforcement is tied
in place, then the column concrete is poured (Figure 22e). The column is allowed to cure for three days,
then the formwork is removed. The column is then covered with burlap and plastic to ensure ideal curing
conditions are present for 28 days. After the 28 days, the column is uncovered, painted, and instrumented
in preparation to be tested. Images of the construction progress are shown in Figure 22.

a) b) c)
d) e) f)
Figure 22. CIP Construction: a) Completed Rebar Cage, b) Footing Pour, c) Finished Footing, d) Footing
Curing, e) Column Formwork Setup, f) Finished Test Specimen

Material Properties
The 28-day compressive strength of concrete, f’c, is designed to be 4,000 psi. The yield strength of the
longitudinal and transverse reinforcement of the rebar is 60,000 psi with a modulus of elasticity of 29×106
psi. The actual compressive strength of the concrete on the test day is summarized in Table 3.
Table 3. Actual f’c values (psi)
Test 1 Test 2 Test 3 Test 4 Average

Footing 3,980 4,740 4,450 5,330 4,630
Column 5,280 4,680 4,970 4,460 4,850
Test Setup and Instrumentation
A Campbell Scientific data acquisition system is used to collect important data points from the instruments
such as load cells, strain gauges, and linear potentiometers to analyze data after the testing procedure. A
total of 39 instruments are used and are discussed in the following paragraphs and summarized in Figure
23. The testing arrangement is shown in Figure 24.
The lateral and vertical loads are measured using load cells. The lateral load cell is mounted between the
hydraulic ram and the end of the actuator. The vertical load cell is mounted between a hydraulic jack and
a steel beam which is resisting any vertical movement of the hydraulic jack.
Lateral displacement at the top of the column is measured by string potentiometers shown as “IPC”,
“OPC”, and “ASP” in Figure 23 and representing in-plane, out-of-plane, and actuator displacements,
respectfully. The spring potentiometers are mounted to a steel structure that is independent of the testing
arrangement. Each of the instruments is connected to the column at the centerline of the actuator.
The footing displacement is monitored using horizontally and vertically mounted spring potentiometers
shown as “FHI”, “FHO”, “FVN” and “FVS” in Figure 23 and representing footing horizontal in-plane, footing
horizontal out-of-plane, footing vertical on the north end, and footing vertical on the south end of the
footing, respectfully. The spring potentiometers are also mounted independently of the testing
arrangement. Spring potentiometers are positioned on the footing to measure any sliding or rocking of
the footing that may occur. Data collected from these instruments are used to correct the lateral
displacement at the top of the column.
The column deformation is obtained by using rod potentiometers, these are shown as “A”, “B”, and “C”
symbols in Figure 23. The rod potentiometers are attached to the face and back of the column in three
distinct zones: Zone A measured from the top of the footing to 18 in., Zone B measured from 18 in. to 36
in., and Zone C measured from 36 in. to 54 in. This deformation data is used to calculate the curvature of
the section up the height of the column.

Elongation of the rebar within the column is monitored by strain gauges. The strain gauges are attached
to the longitudinal rebar at the column to footing interface before pouring the concrete.
Figure 23. Column Instrumentation

Gravity Load (50
Servo-valve actuator kip or 4% f’c Ag)
Reaction Frame
Specimen
Load Cell
High-strength rods
simulating gravity loads
Strong Floor Footing Anchors
Figure 24. Typical Test Setup with Load Cells
Testing Results
On the first cycle, hairline cracks appeared on the column within the first 0 – 18 in. from the base of the
column where the plastic hinge is expected to form. During the second cycle, hairline cracks spread into
the 18 – 36 in. of the column. By the third cycle, hairline cracks spread above the 36 in. mark and other
lower cracks continued to spread and get larger. On the fourth cycle, a crack approximately 1mm wide is
seen at the column-footing interface. During the fifth and sixth cycles, cracking continued to develop with
the base crack opening up to 4mm and the concrete began to spall. For the eighth, ninth, tenth, and
eleventh cycles; all of the cracks continued to develop and the column concrete near the footing had all
spalled off. During the twelfth cycle, there was a loud pop followed by a significant drop in lateral force.
The drop-in force was not significant enough to discontinue the test. The thirteenth cycle resulted in two
additional loud pops which are followed by a significant drop in lateral force. The drop in lateral force was
significant enough to terminate the testing procedure. The loud pops are a result of a longitudinal rebar
breaking and the locations where they occurred are shown in Figure 25. Figure 26 shows images of the
lower half of the column during the testing procedure.
Throughout the testing procedure, the programmed displacement is not the actual displacement of the
column. As a result of this, there is a cycle missing between cycles eight and nine to achieve the desired
displacement. One of the reasons why the programed displacement values are not the same as the actual
values is because during the testing procedure the reaction frame is deflecting. Another reason is that the
footing is sliding a little throughout the testing procedure. The sliding of the footing is documented via

the instruments placed on the footing before the testing procedure. Using the collected data, the sliding
of the footing is removed from the column displacement to obtain the column displacement relative to
the footing. A summary of the loading protocol is shown in Table 4.
Figure 25. Loading Protocol for the Cast-In-Place Column
Table 4. Summary of the Cast-In-Place Loading Protocol
Cycle Programmed Displacement (in.) Programmed Drift (%) Actual Displacement (in.) Actual Drift (%)
1 0.20 0.26 0.16 0.20
2 0.35 0.45 0.26 0.33
3 0.69 0.88 0.46 0.59
4 1.38 1.77 0.86 1.10
5 2.08 2.67 1.50 1.93
6 2.77 3.55 2.15 2.76
7 3.46 4.44 2.84 3.64
8 4.15 5.32 3.50 4.49
9 5.54 7.10 4.86 6.23
10 6.23 7.99 5.54 7.10
11 6.92 8.87 6.23 7.99
12 7.61 9.76 6.94 8.90
13 8.30 10.64 7.71 9.89

a) b) c)
d) e) f)
Figure 26. CIP Specimen at a) 0.45% drift, b) 2.67% drift, c) 5.32% drift, d) 8.87% drift, e) 10.64% drift,
f) Ruptured Bar
After the testing procedure has concluded the data is analyzed. The maximum displacement of the column
during the testing procedure is 7.7 in. The maximum load applied to the column during the testing

procedure is 37.8 kip which corresponds to a 245.7 kip-ft moment capacity. Plots showing the Force vs.
Displacement and Drift during the test are shown in Figure 27 and Figure 28, respectively. These plots
show the yield displacement of the column and the corresponding base shear. The column reached its
design flexural capacity at 25-kip and behaved in a ductile manner. Rebar ruptures can be identified on
the plots as a sudden reduction in capacity in the first quadrant. The backbone curve is also shown in
Figure 29. Table 5 and Table 6 show significant values from the cast-in-place column testing procedure.
Figure 27. Force vs. Displacement Plot for the Cast-In-Place Column

Figure 28. Force vs. Drift Plot for the Cast-In-Place Column
Figure 29. Backbone Curve for the Cast-In-Place Column

Table 5. Damage Observation from Testing of the CIP Column, Drift %
Cracking Spalling 1st Bar Rupture 2nd Bar Rupture 3rd Bar Rupture End of Testing
0.2 % 4.49% 8.9% 9.89% 9.89% 9.89%
Table 6. Summary of the Performance Points from Testing of the CIP Column
Yielding Yielding Yielding Yielding Ultimate Ultimate Ultimate

Predicted Predicted Experimental Experimental Strength Strength Strength
Drift (%) Base Shear Drift (%) Base Shear Experimental Experimental Experimental
(kips) (kips) Drift (%) Base Shear Ductility*
(kips)
0.88 25.58 1.15 35.1 9.89 37.8 7.4
Note: *Displacement ductility (µ) was 7.4 at the end of testing
The distribution of curvature along the height of the first 54 in. of the column is shown in Figure 30. The
column is expected to yield and fail within the plastic hinge of the column. The plastic hinge is calculated
by the following equation from Priestley et al. (2007):
𝐿𝐿𝑝𝑝 = 0.08𝐻𝐻 + 𝑙𝑙𝑠𝑠𝑠𝑠
Where:
H = Height of the column, inches

Lsp = Strain penetration length, inches
Strain penetration can be calculated using the following equation from Priestley et al. (2007):
𝑙𝑙𝑠𝑠𝑠𝑠 = 0.15𝐹𝐹𝑦𝑦𝑦𝑦 𝑑𝑑𝑏𝑏𝑏𝑏
Where:
Fye = 1.1 times the longitudinal bar yield strength, ksi

dbl = Diameter of the longitudinal bar, inches
Using a yield strength of 60 ksi and a bar diameter of 0.75 in, the strain penetration becomes 7.4 in. Using
a column height of 78 in. the plastic hinge can be calculated to be 13.7 in. Figure 30 shows that yielding
occurred in the bottom 18 in. of the column. From 18 in. to 36 in. the column approached the yield point
but never reached it. Because the region from 18 in. to 54 in. never reached the yield point, this region
essentially remained elastic throughout the testing procedure.

Figure 30. Cast-In-Place Curvature Distribution
The dissipated energy plot is shown in Figure 31. The dissipated energy is obtained by finding the area of
each loop in the force versus displacement plot. The area is obtained using a MATLAB program. Each of
the loops represents the total energy dissipated in a single cycle. Using input units of newtons and meters,
the output energy units are of joules. For each drift cycle, the first loop dissipated more energy than the
second because the column becomes weaker with each push-pull cycle. The largest difference in
dissipated energy occurred during the 12th and 13th cycles which corresponds to the longitudinal rebars
rupturing. The total dissipated energy during the testing procedure is 336,328 ft-lb (456 kJ).

Figure 31. Dissipated Energy Plot for the CIP Column (1 kJ = 737.56 ft-lb)
The area-based hysteretic damping is calculated using the following equation from Chopra (2017):
𝐴𝐴ℎ
𝜉𝜉𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎−𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 =
2𝜋𝜋𝐹𝐹𝑚𝑚 𝛿𝛿𝑚𝑚
Where:
Ah = Area of a loop in the force vs. displacement plot, N-m

Fm = Maximum force in the loop, N
δm = Maximum displacement in the loop, mm
To compare the hysteretic damping values to a theoretical model, a correction factor should be applied
to convert the values to an equivalent time-history-calibrated hysteretic damping. The equation used to
convert the area-based damping values to the time-history-calibrated hysteretic damping values is shown
below from Priestley et al. (2007):
𝜉𝜉ℎ𝑦𝑦𝑦𝑦𝑦𝑦 = 𝜉𝜉𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎−𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 [−0.018𝜉𝜉𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎−𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 + (0.0875𝜇𝜇 + 0.723)]
Where:
µ = Displacement ductility
ζarea-based = Area based hysteretic damping

Displacement ductility is used to characterize the seismic response of a structure (Vielma and Mulder
2017) and is calculated using the following equation from Priestley et al. (2007):
𝐻𝐻
𝜇𝜇 = 0.01(%𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷𝐷)
∆𝑦𝑦
Where:
H = Height of the column, m
Δy = Yield displacement, m
A corrected area-based hysteretic damping plot is shown in Figure 32. As a comparison, the Takeda Fat,
Takeda Thin, Elastic-Perfectly Plastic (EPP), and Ramberg-Osgood (RO) models have also shown in Figure
32. The Takeda Fat model represents a ductile reinforced concrete frame structure. The Takeda Thin
model represents a ductile reinforced concrete wall or column structure. The Elastic-Perfectly Plastic (EPP)
model represents an elastic model and the Ramberg-Osgood (RO) model represents a ductile steel
structure (Priestley et al. 2007). The equivalent viscous damping coefficients used are summarized in Table
7 which are used in the following equation to obtain the hysteretic damping models from Priestley et al.
(2007):
1 1
𝜉𝜉ℎ𝑦𝑦𝑦𝑦𝑦𝑦 = 𝑎𝑎 �1 − � �1 + �
𝜇𝜇 𝑏𝑏 (𝑇𝑇𝑒𝑒 + 𝑐𝑐)𝑑𝑑
Where:
ζhyst = Hysteretic damping

Te = Effective period representing one column, 1 sec
Table 7. Equivalent Viscous Damping Coefficients for Hysteretic Damping Component
Model a b c d
EPP 0.224 0.336 -0.002 0.250
Takeda Thin 0.215 0.642 0.824 6.444
Takeda Fat 0.305 0.492 0.790 4.463
Ramberg-Osgood 0.289 0.622 0.856 6.460
Figure 32 shows the cast-in-place column having a relatively linear trend between the displacement
ductility and the hysteretic damping. The hysteretic damping is approximately equal to the Takeda Thin
model until a displacement ductility of 3.5. The cast-in-place column hysteretic damping curved achieved
a value of 24.7% when the testing procedure is terminated.

Figure 32. Corrected Area-Based Hysteretic Damping for the CIP Column
Summary
The cast-in-place specimen is intended to be the benchmark for the precast solution. The cast-in-place
column is designed following the latest version of the AASHTO LRFD Bridge Design Specifications. The
footing and column are constructed using traditional methods and materials. The column is tested under
quasi-static loading until failure. The column behaved in a ductile manner. The column achieved its design
base shear of 25-kip. The maximum base shear obtained during the testing procedure is 37.8-kip. The
maximum displacement achieved is 7.7 in. which corresponded to a ductility of just below 7.5. Damage to
the column during the test is within the first 18 in. of the column which is the expected plastic hinge length
for a well detailed and confined section. The total energy dissipated during the testing procedure is 456
kJ. A hysteretic damping plot showed that the column had similar values to a Takeda-Thin model up to a
displacement ductility of slightly lower than 3.5. Above that, the column had higher values of hysteretic
damping compared to a Takeda-Thin model.

Cantilever Column: Precast
Introduction
This section presents the design, construction, and testing of a precast cantilever column. The precast
cantilever column is intended to have the same capacity as the cast-in-place column in terms of resisting
lateral force and drift capacity. Using precast elements will increase the rate of construction and reduce
overall construction costs. One of the foreseeable issues is with regards to the proper alignment of the
hollow structural section (HSS) made from steel that is incorporated into the design of the column. Proper
alignment is crucial because small errors can have a significant impact during the assembly of the precast
members. The design discussed in this chapter closely follows the 2019 WSDOT Bridge Design Manual
(WSDOT 2019). The overall structural dimensions, testing arrangement, and instrumentation are identical
to the cast-in-place column.
Overview
As shown in Figure 1 of Chapter 1: Introduction, the pipe connection incorporates two HSS members
which is a representation of the pipe connection assembly. One of the HSS members is installed at the
end of the precast column and the other is embedded into the footing. Approximately half of the column
HSS is extending out of the precast column and with centering fins welded to the outside edge of the pipe
and provides flexural capacity, shear capacity, and confinement for the concrete at the plastic hinge zone.
The centering fins are for inserting the column into the footing and are not structurally important. Headed
rebars are used to develop the strength of the rebar during a seismic event. Inserting the precast column
into the footing is made possible by the footing HSS member. The footing HSS member will remain hollow
until the column is inserted into it. An unbonded length on the column HSS member is provided to reduce
the stresses in the HSS member and emulate a cast-in-place behavior.
An elastomeric bearing pad is placed between the column and footing to eliminate any undesired
concentrated loading between the two elements and allows the column to rock during smaller
earthquakes to prevent cracking. After the protruding HSS column member has been inserted into the
footing HSS member, the remaining voids between the two elements are filled using a non-shrinkage
grout via PVC pipes strategically placed to allow the grout to fill all the desired voids without compromising
the integrity of the footing or column before the grouting procedure. The grout is poured or pumped into
the grout inlet pipe while the grout vent provides an escape for air that would otherwise be trapped. The
gap between the two HSS members is approximately 1 in. to 2 in. to ensure the grout performs properly.
Any of the rebars in the footing that are impeded by the HSS member in the footing are terminated using
a welding bar coupler or hooks. The column and footing design that is not impeded by the HSS members
follows the AASHTO LRFD Bridge Design Specifications (AASHTO 2017).

Material Properties
For the proposed pipe connection, WSDOT design manual (WSDOT 2019) was referred. The WSDOT design
manual (WSDOT 2019) specifies materials to be used in the design of the concrete-steel tube members.
First, the concrete should be class 5000P which is 5,000 psi concrete. The second material specification is
to use ASTM A 709 GR 50 steel. ASTM A 709 GR 50 steel has a specified tensile strength of 65,000 psi and
a yield strength of 50,000 psi (Chapel Steel 2018).
In this study, the 28-day compressive strength of concrete (f’c) is 4,000 psi. Through trial and error, the
HSS member selected to obtain a moment capacity approximately equal to the cast-in-place member is
HSS8.625x0.625. The properties of an HSS8.625x0.625 member is 42,000 psi yield strength (Fy), 58,000 psi
ultimate strength (Fu), and a modulus of elasticity (E) of 29×106 psi. The dimensions of the HSS member
are 8.625 in. for the outside diameter and a pipe thickness of 0.581 in. The actual compressive strength
of the concrete and grout on the test day is summarized in Table 8.
Table 8. Actual f’c Values for the Precast Column (psi).
Element Test 1 Test 2 Test 3 Average

Footing/Column 3,800 4,180 3,930 3,970
Grout 5,410 5,420 5,370 5,400
Design of Precast Specimen
The Concrete Filled Steel Tube (CFST) at the interface of the column and footing (e.g., where the CFST in
unbonded) is the ductile link. This section is designed to provide flexural and shear capacities as well as
confinement for the plastic hinge. The footing and the section of the column above the unbonded region
are designed to remain elastic. For a simplified approach, the contribution of the unconfined cover
concrete towards the flexural resistance of the CFST section is ignored. The size of the CFST for the ductile
link was selected to closely match the flexural capacity of the cast-in-place specimen (e.g., similar base
shear).
The flexural design procedure for CFST has several resistance factors to use for a variety of situations. The
resistance factor used in this design is equal to one, which is used for an extreme event limit state. To
ensure the pipe is not subject to local buckling before developing the strength of the pipe, the WSDOT
(2019) advises the use of the following equation for members subject to plastic forces:
𝐷𝐷 𝐸𝐸
≤ 0.15
𝑡𝑡 𝐹𝐹𝑦𝑦
Where:
D = Outside diameter, inches

t = Wall thickness of HSS member, inches

For the selected HSS member the D/t is equal to 14.85 and 0.15E/Fy is equal to 103.57. In common
practice, both D and t should be adjusted for corrosion but due to the short life of the testing timeline,
this is considered irrelevant. Most importantly, the equation from WSDOT (2019) was adopted in this
research for CFST to be used in columns and not piles. The nominal compressive strength of the concrete
filled HSS member is determined by using the following equation:
𝑃𝑃0 = 0.95𝑓𝑓𝑐𝑐′ 𝐴𝐴𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 + 𝐹𝐹𝑦𝑦,𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝐴𝐴𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠
Where:
f’c = Compressive strength of a concrete, ksi

Aconcrete = Area of concrete, in2
Asteel = Area of steel, in2
Fy,steel = Yield strength of the steel, ksi
For the selected HSS member P0 has been calculated to be 949 kips. The equation used to determine the
nominal moment capacity is shown below:
𝑐𝑐 3 𝑟𝑟𝑚𝑚2
𝑀𝑀𝑛𝑛 (𝑦𝑦) = �𝑐𝑐�𝑟𝑟𝑖𝑖2 − 𝑦𝑦 2 � − � ∗ 0.95𝑓𝑓𝑐𝑐′ + 4𝑐𝑐𝑡𝑡 𝐹𝐹𝑦𝑦
3 𝑟𝑟𝑖𝑖
Where:
c = One half cord length of the compressive block

ri = Mean radius of internal reinforcement, in
rm = Mean radius of HSS member, in
y = Vertical distance of neutral axis from center of HSS in plastic stress distribution method, in
f’c = Compressive strength of the concrete
t = thickness of HSS member
c is calculated using:
𝑐𝑐 = 𝑟𝑟𝑖𝑖 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐
Where:
ri = Mean radius to the inside of the steel tube

θ is calculated using:
𝑦𝑦
𝜃𝜃 = 𝑠𝑠𝑠𝑠𝑠𝑠−1 � �
𝑟𝑟𝑚𝑚
Where:
y = Distance from the centroid of the specimen to the neutral axis during a seismic event, in
rm = Radius to the center of the steel tube, in
Because the neutral axis is expected to be equal to the centroid the variable y is taken equal to zero. Once
y is determined, the variables θ and c are calculated to be 0° and 3.73 in., respectfully. The nominal
moment capacity is calculated then to be 1842 kip-in. or 142.5 kip-ft.

The variable y can be used if the centroid and neutral axis are not equal. If this is the case, the column is
not perfectly vertical. Determining the variable y is difficult if the column is not vertical and is more of a
guess. Figure 33 shows the plastic stress distribution of the steel and concrete when the variable y is not
equal to zero. Material forces can be obtained by solving the following equations simultaneously.
� 𝐹𝐹𝑥𝑥 = 𝑃𝑃 + 𝑇𝑇𝑠𝑠 − 𝐶𝐶𝑐𝑐 − 𝐶𝐶𝑠𝑠 = 0
(𝐶𝐶𝑐𝑐 + 𝐶𝐶𝑠𝑠 )(𝑟𝑟 + 𝑦𝑦) 𝑇𝑇𝑠𝑠 (𝑟𝑟 − 𝑦𝑦)

� 𝑀𝑀0 = − =0
2 2
𝐴𝐴𝑠𝑠 = 𝐴𝐴𝑠𝑠𝑠𝑠 + 𝐴𝐴𝑠𝑠𝑠𝑠
𝑇𝑇𝑠𝑠 = 𝐴𝐴𝑠𝑠𝑠𝑠 𝐹𝐹𝑦𝑦
𝐶𝐶𝑐𝑐 = 0.95𝐴𝐴𝑐𝑐𝑐𝑐 𝑓𝑓𝑐𝑐′

𝐶𝐶𝑠𝑠 = 𝐴𝐴𝑠𝑠𝑠𝑠 𝐹𝐹𝑦𝑦
Where:
P = Axial load, kips

Ts = Resultant force of the steel in tension, kips
Cc = Resultant compressive concrete forces, kips
Cs = =Resultant compressive steel forces, kips
Ast = Area of steel in tension, in2
Acc = Area of concrete in compression, in2
Asc = Area of steel in compression, in2
r = Radius of the pipe measured to the outside of the pipe, in
f’c = Compressive strength of the concrete, ksi
Fy = Yield strength of the steel, ksi
Figure 33. Plastic Stress Distribution
To calculate the base shear, the following equations were used:

∅𝑉𝑉𝑛𝑛 = ∅(𝑉𝑉𝑐𝑐 + 𝑉𝑉𝑠𝑠 )
Where:

∅ = Capacity reduction factor, 0.75
Vn = Net shear, kips
Vs = Shear in the HHS steel member, kips
Vc = Shear in the concrete, kips
Vc is calculated using the following equation from McCormac (2014):
𝑉𝑉𝑐𝑐 = 2𝜆𝜆�𝑓𝑓𝑐𝑐′ 𝑏𝑏𝑤𝑤 𝑑𝑑

Where:
𝜆𝜆 = Modification factor, 1.0

f’c = Compressive strength of concrete, ksi
bwd = Area of concrete
Using appropriate values, the equation becomes:
𝑉𝑉𝑐𝑐 = 2(1)√4000(43.74) = 5,532 𝑙𝑙𝑙𝑙

Vs is calculated using the following equation from AISC (2011):
𝑉𝑉𝑠𝑠 = 0.6𝐹𝐹𝑦𝑦 𝐴𝐴𝑔𝑔 ⁄2
Where:
Fy = Yield strength of steel, ksi

Ag = Gross cross-section area of the member, in2
Using the appropriate values, the equation becomes:
0.6(42,000)(14.68)
𝑉𝑉𝑠𝑠 = = 184,968 𝑙𝑙𝑙𝑙
2
And the net shear equation becomes:
∅𝑉𝑉𝑛𝑛 = 0.75(5,532 + 184,968) = 142,875 𝑙𝑙𝑙𝑙
Two methodologies are considered to determine the embedment of the pipe into the column and footing.
The methodologies used are proposed by WSDOT and Edward P. Wasserman (Wasserman and Walker
1996). The WSDOT equation includes an annular ring, as shown in Figure 34, and is used to ensure full
plastic behavior of the concrete-filled steel tube is as follows:
𝐷𝐷02 5.27𝐷𝐷𝐷𝐷𝐹𝐹𝑢𝑢 𝐷𝐷0

𝑙𝑙𝑒𝑒 ≥ � + −
4 �𝑓𝑓𝑐𝑐′ 2
Where:
le = Calculated embedment length, in

D0 = Outside diameter of the annular ring, in
D = Diameter of the embedded pipe, in
t = Wall thickness of the embedded pipe, in
Fu = Ultimate strength of the embedded pipe, ksi

f’c = Compression strength of the concrete, ksi
Figure 34. Embedment Length with Embedded Ring (courtesy of WSDOT)
Eliminating the annular ring from the equation, e.g., D0 = 0, the equation becomes:
5.27𝐷𝐷𝐷𝐷𝐹𝐹𝑢𝑢
𝑙𝑙𝑒𝑒 ≥ �
�𝑓𝑓𝑐𝑐′
Incorporating the appropriate values into the equation an embedment length of 27.7 in. is adequate. For
the embedment length proposed by Wasserman, the following equations are used:
2𝐹𝐹𝑦𝑦 𝑍𝑍
𝑙𝑙𝑒𝑒 =
�700𝑓𝑓𝑐𝑐′ 𝑏𝑏
𝑑𝑑√𝜋𝜋
𝑏𝑏 =
2
Where:
Fy = Yield strength of the embedded pipe, ksi

Z = Plastic section modulus of the embedded pipe, in3
d = Outside diameter of the embedded pipe, in
Incorporating these values into the equations an embedment length of 23.7 in. As a result of these two
equations an embedment length of 2 ft-3 in. is used as the HSS embedment length.
The column longitudinal reinforcing is comprised of twelve #6 headed rebars equally spaced in a circular
pattern. For confinement purposes, a spiral having a pitch of 1.5 in. is used along the entire length of the

column with a mechanical splice located at each end of the spiral. The spiral should maintain 1.5 in. of
cover.
A portion of the HSS member just inside the concrete column is left unbonded. The unbonded length is a
way to force the HSS over a certain length above the footing and encourage a rocking type (gap opening)
motion at the base of the column during smaller movements (e.g., small earthquakes) without yielding of
the CFST. Also, during larger movements/earthquakes, the unbonded length distributes the inelastic strain
over a longer length of the CFST which enhances the low-cycle fatigue performance and ultimate
displacement ductility of the connection. Leaving unbonded length is optional, if there are concerns about
durability of the HSS, the unbonded length can be avoided.
In absence of experimental data on debonding of CFSTs, the unbonded length of the HSS member is
determined using PRESSS Design Handbook (Pampanin 2010). All of the equations from the PRESSS Design
Handbook are in metric units. To begin the elongation of the mild steel from the cast-in-place column
testing procedure is determined to be roughly 0.88 in. (22.4 mm). The strain penetration is then calculated
using the following equation:
Where:
Fye = 1.1 times the yield strength, ksi

dbl = Pipe thickness, in
Using a value of 60 ksi (413.7 MPa) for Fy and 0.75 in. (19.05 mm) for dbl, the strain penetration is
calculated to be 7.43 in. (188.7 mm). Strain in the steel is then calculated using the following equation:
∆
𝜀𝜀𝑠𝑠 =
𝑙𝑙𝑢𝑢𝑢𝑢 + 2𝑙𝑙𝑠𝑠𝑠𝑠
Where:
Δ = Elongation of the rebars, in

lub = Unbonded length, in
lsp = Strain penetration length, in
The elongation of the rebars during the cast-in-place testing procedure was determined to be 0.81 in.,
from the data collected at an 8.87% drift. Using a value of 0 for the unbonded length and 0.81 in. (22.4
mm) for the elongation of the rebars, the strain is calculated to be 5.90% for the cast-in-place column.
Using Δ = 0.81 in. (22.4 mm), Fy = 42 ksi (289.6 MPa), dbl = 0.581 in. (14.76 mm) or the pipe thickness, and
an unbonded length of 3 in. (76.2 mm), the strain in the HSS member is determined to be 7.75%. As a
result of this calculation, the unbonded length is determined to be 3 in. It should be noted that this
procedure is a rough estimate and more experimental and analytical data are required to quantify the
exact unbonded length of the HSS. Another proposed method of estimating the unbonded length is that
the unbonded length would equal 30-35% diameter of the HSS. For states such as Idaho where corrosion

risk is not such high as Pacific northwest, California, or areas around the coast; there should not be a major
concern regarding the debonding steel. However, for the areas where moisture ingrains or corrosion is
likely to occur are factors, then the unbonded length can be avoided for better durability.
Figure 35 shows the details of the precast column. The reason behind using the volumetric reinforcement
ratio of about 2.4%, i.e., #3 spiral rebars although CFST gets shear to simulate the CIP as close as possible
for the sake of good comparison with CIP until further research is available on what should be the
volumetric reinforcement ratio for the system. The tapered fins that are used in pipes are not structural,
they can be simple fillet welds because they are intended for the alignment and are not performing any
structural function.
Figure 35. Precast Column Detail
After the precast testing procedure, the gap opening for the precast column is measured to be 0.55 in.
(13.97 mm) on cycle 10 or 5.96 in. (7.64% drift) after which the instruments were removed from the

column. Using the values previously discussed, the actual strain in the HSS member is determined to be
5.29%.
The elastomeric bearing pad is designed according to the AASHTO LRFD Bridge Design Specifications
(AASHTO 2017). The elastomeric bearing pad is designed for a column with a diameter of 18 in., 50-kips
axial load, and has a 10.5 in. diameter hole through the center of the bearing pad. Using these parameters
and shear modulus of 0.13 ksi, a plain elastomeric bearing pad with a thickness of 0.5 in. is determined to
be sufficient. Further information can be found in AASHTO M251 (Standard Specification for Plain and
Laminated Elastomeric Bridge Bearings). For this research project, under consultation of ITD a ½ in. of this
material was deemed to be suitable.
The footing for the precast column was designed to remain elastic throughout testing. The footing design
is similar to the cast-in-place column. The changes that are made to the footing are to accommodate for
assembling the column to the footing. An HSS10.75x0.5 having a length of 28 in. is installed into the top
of the footing such that the top of the HSS member is level with the top of the footing. All of the footing
reinforcing rebars that are affected by the HSS member are adjusted by incorporating a 90° bend with an
appropriate amount of development length to accommodate this change. Guidelines on rebar hooks,
bends, and development lengths are conservatively considered with all the rebar in the footing. An image
depicting the precast column assembly to the footing is shown in Figure 36.

Figure 36. Precast Column to Footing Assembly
In summary, the precast column is octagonal with a diameter of 18 in. The longitudinal reinforcement is
comprised of twelve #6 headed rebars equally spaced with a #3 spiral having a pitch of 1.5 in. to confine
the concrete. The HSS8.625x0.625 has a development length of 2 ft-3 in. into the column and footing with
3 in. of un-bonded length at the footing to column interface. The calculated moment capacity of the
concrete-filled HSS member is 142.5 kip-ft This corresponds to a base shear of 21.9-kips for a cantilever
height of 6.5ft.
The footing is 4 ft. x 4 ft. x 3 ft. with ten #6 rebars reinforcing the top and bottom in each direction while
maintaining a spacing of approximately 3.5 in. between the bars and 2 in. of cover. An HSS10.5x0.5
member is embedded into the footing to receive the precast column. The moment capacity of the footing
is expected to be roughly 1,000 kip-ft. Comparing the column’s plastic hinge moment capacity to the
footing moment capacity, the CFST at the interface of the column and footing is designed to be the ductile
link with inelastic deformation.

Construction
For the construction of the precast specimen, the formwork from the cast-in-place test specimen is
modified slightly to accommodate for the HSS members and is reused. Rebar is cut to the required lengths
and bent appropriately. Next, the rebar is tied together, and the formwork is secured around the tied
rebar (Figure 37a). Before installing the HSS members, the exposed surfaces are sandblasted to remove
paint and roughen the surface to obtain a better bond with the concrete. The precast column is poured
horizontally because of the restrictions over dropping height for concrete as well as difficulty in keeping
the column standing upright with a pipe protruding from the end (Figure 37b). Once the HSS members,
rebar, and formwork are set, the concrete for the column and footing is poured (Figure 37c-d). Both the
footing and column are allowed to cure for three days before the formwork is removed. After the
formwork is removed the concrete is covered with burlap and plastic and wetted daily to allow the
concrete to continue curing in the most ideal conditions. Once the concrete reached seven days of curing,
the concrete footing is moved to the Structural LAB (SLAB). After another seven days, the column is also
moved to the structural laboratory and grouted into place, painted, and instrumented in preparation to
be tested. Images of the construction progress are shown in Figure 37. Before the column assembly, the
exposed surfaces of the HSS member embed in the footing is roughened and lightly wetted in preparation
for the grouting process. The grout used to fill the void between the two HSS elements is SikaGrout®-328.
After mixing the grout per the grout mixing instructions, the mixture is promptly pumped into a Polyvinyl
Chloride (PVC) pipe that extended from the top of the footing grout inlet pipe. After a few moments grout
is observed to be pouring out of both of the grout outlet pipes (Figure 37). The grout is allowed to flow
freely through the grout vent ducts for approximately 30 seconds to remove any air pockets that may be
trapped. Once the grout had flowed for approximately 30 seconds the PVC pipe valves are closed and the
grout is allowed to cure for 7 days before the testing protocol is initiated. Figure 37f shows the completed
precast specimen.

a) b) c)
d) e) f)
Figure 37. Precast Column Construction: a) Footing Rebar Cage, b) Column Rebar Assembly, c) Finished
Footing, d) Finished Column, e) Grout Flowing out of Grout Vent, f) Finished Test Specimen

Test Setup and Instrumentation
The testing protocol and arrangement for the precast column is similar to that of the cast-in-place column.
The instrumentation of the precast column is similar to the cast-in-place; however, the precast column
had a total of four strain gauges, two on each side, located on the column HSS member at the extreme
push and pull locations. On each extreme location of the HSS member, a strain gauge is placed on the
exposed surface about one inch from the surface of the concrete and the other was placed approximately
one inch inside of the concrete surface as shown in Figure 38. Arranging the strain gauges in this manner
makes so each of them is one inch from the column to footing interface.
Figure 38. Location of Strain Gauges on the HSS Member
Testing Results
After testing began, for the first two cycles, there are no visible hairline cracks. On the third cycle, hairline
cracks are visible below the first 0 – 36 in. of the column which continued to develop into the fourth cycle.
During the fifth cycle, concrete spalling began to occur on the north face of the column and by the seventh
cycle, spalling occurred on the south face as well. Cracking and spalling continued to develop until the 11th
cycle. On the 11th cycle, the unbonded length of the pipe is visible on both the north and south faces of
the column. On the 13th cycle, an “elephant leg” behavior is observed on the exposed HSS member. During
the 15th cycle, approximately 10 kips of lateral force is lost as a result of the pipe beginning to degrade.
On the 16th cycle, another significant drop in lateral force is observed due to a fracture of the pipe on one
side and the testing procedure was terminated due to safety reasons. The fracture is located very close to
the interface, the buckling of the pipe is obvious on the other side as shown in Figure 39.

Figure 39. Pipe Failure During the Testing Procedure.
Figure 40 shows the loading protocol used during the testing procedure. Similar to the cast-in-place
column, the programmed displacement is not the actual displacement due to the actuator frame
deflecting and the footing sliding. The actual deflection of the column for the push and pull cycles are
within 0.05 in. of each other. To keep the same loading protocol as the cast-in-place there is a missing
cycle between cycles eight and nine as shown in Figure 40. The same procedure is followed to correct the
column displacement data which is summarized in Table 9. Figure 41 shows images of the lower half of
the column during the testing procedure.
Figure 40. Loading Protocol for the Precast Column

Table 9. Summary of Performance Values from Testing of the Precast Column
Cycle Programmed Displacement (in.) Programmed Drift (%) Actual Pull Displacement (in.) Actual Pull Drift (%)
1 0.20 0.26 0.14 0.18
2 0.35 0.45 0.23 0.29
3 0.69 0.88 0.47 0.60
4 1.38 1.77 0.92 1.19
5 2.08 2.67 1.53 1.96
6 2.77 3.55 2.15 2.75
7 3.46 4.44 2.82 3.62
8 4.15 5.32 3.36 4.30
9 5.54 7.10 4.87 6.24
10 6.23 7.99 5.37 6.89
11 6.92 8.87 6.09 7.80
12 7.61 9.76 6.78 8.69
13 8.30 10.64 7.45 9.55
14 9.00 11.54 8.14 10.44
15 9.69 12.42 8.87 11.37
16 10.38 13.31 9.57 12.27

Figure 41. Precast Specimen During Testing at: a) 0.45% Drift, b) 2.67% Drift, c) 5.32% Drift, d) 8.87%
Drift, e) 13.27% Drift, f) Failed HSS Member
After the testing procedure has concluded the data is analyzed. The maximum displacement of the column
during the testing procedure is 9.57 in. which corresponds to a 12.27% drift ratio. The maximum load
applied to the column during the testing procedure is 41.2 kip which corresponds to a 267.8 kip-ft moment
capacity. The column showed a stable response with significant energy dissipation. When the column
reached the peak load, a reduction in the lateral load is observed as a result of the column engaging the
cover concrete. The lateral load continues to slowly decrease with each cycle because the HSS pipe is
experiencing low-cyclic fatigue. As the testing procedure progresses a significant drop in lateral force is

observed which is a result of the pipe buckling or “elephant leg” behavior. On the next cycle the lateral
force approaches the same lateral force then decreases as the pipe fails. Plots showing the Force vs.
Displacement and Drift during the test are shown in Figure 42 and Figure 43, respectively. These plots
show the yield displacement of the column and the design base shear of the column which was reached
before the pipe failed. The backbone curve is also shown in Figure 44. The initial kink in the backbone
curve is a decompression point or point at which a gap began to open as a result of the column rocking.
Table 10 and Table 11 show significant values from the precast column testing procedure.
Figure 42. Force vs. Displacement plot for the Precast Column

Figure 43. Force vs. Drift Plot for the Precast Column
Figure 44. Backbone Curve for the Precast Column

Table 10. Damage Observation from Testing of the Precast Column, Drift (%)
Cracking Spalling Significant Elephant-Leg Buckling Fracture of the Pipe End of Testing
0.6 % 4.42 % 8.7 % 10.5 % 12.27 %
Table 11. Summary of the Performance Points from Testing of the Precast Column
Yielding Yielding Yielding Yielding Ultimate Ultimate Ultimate

Predicted Predicted Experimental Experimental Strength Strength Strength
Drift (%) Base Shear Drift (%) Base Shear Experimental Experimental Experimental
(kips) (kips) Drift (%) Base Shear Ductility*
(kips)
0.88 21.9 1.7 27.2 12.27 41.2 6.3
Note: *Displacement ductility (μ) was 6.3 at the end of testing
The distribution of curvature along the height of the first 54 in. of the column is shown in Figure 45. Like
the cast-in-place column, the column is expected to yield and fail within the plastic hinge of the column.
The plastic hinge for the precast column can be calculated using the same equation as the cast-in-place
column from Priestley et al. (2007):
𝐿𝐿𝑝𝑝 = 0.08𝐻𝐻 + 𝑙𝑙𝑠𝑠𝑠𝑠
Where:
H = Column height, in
Using the strain penetration discussed earlier in this chapter and column height of 78 in. the plastic hinge
is calculated to be 13.07 in. Result showed that yielding occurred in the bottom 18 in. of the column. From
18 in. to 36 in. the column approached the yield point but never reached it. Because the region from 18
in. to 54 in. never reached the yield point, this region essentially remains elastic throughout the testing
procedure.

Figure 45. Precast Curvature Distribution
The dissipated energy plot is shown in Figure 46. The dissipated energy is obtained using the same
procedure as the cast-in-place column. The largest difference in dissipated energy occurred during the
15th and 16th cycles which corresponds to the HSS member located in the column buckling and fracture.
The total dissipated energy during the testing procedure is 756,001.2 ft-lb (1025 kJ).

Figure 46. Dissipated Energy Plot for the Precast Column (1 kJ = 737.56 ft-lb)
An area-based hysteretic damping plot is shown in Figure 47. The area-based hysteretic damping is
calculated using the same procedure as the cast-in-place column. The plot shows the precast column
having a relatively parabolic trend for the majority of the testing procedure. The column had higher
hysteretic damping at the beginning of the testing procedure because of the elastomeric damping pad
(e.g., contact damping) allowing the column to move without significantly increasing the load at the top
of the column. The precast column hysteretic damping is just below the Takeda Fat model after the
capability of the elastomeric bearing pad is exceeded, then approximately equal to the Takeda Fat model
from a displacement ductility of 4 to 5.5 after which the column hysteretic damping approached the
Ramberg-Osgood model until the testing procedure was ended.

Figure 47. Area-Based Hysteretic Damping for the Precast Column
Summary
The precast specimen is designed to match the capacity and ductility of the cast-in-place benchmark. The
precast footing and column are constructed incorporating HSS members. The Concrete Filled Steel Tube
(CFST) at the interface of the column and footing provides flexural and shear resistance as well as
confinement. The specimen is designed using the CFST equations provided in the WSDOT Bridge Design
Manual and the latest version of AASHTO LRFD Bridge Design Specifications. The contribution of the
unconfined concrete cover at the interface was ignored for the flexural design of the column. An
elastomeric pad is provided at the column and footing interface to prevent cracking during smaller drift
ratios and allow rocking of the column. The embedment length of the pipe is selected such to develop the
plastic capacity of the column pipe without any premature failure or pullout. An unbonded length of the
pipe is provided to distribute inelastic strain during larger drift ratios and to improve the low-cycle fatigue
performance of the pipe. The unbonded length region was designed to be the ductile link in the specimen.
The footing and other parts of the column are designed to be capacity protected regions. Testing results
show good performance of the precast column with enhanced energy dissipation. Damage to the column
during the test was observed to be within the first 18 in. of the column which thought to be the plastic
hinge region. The loss of the cover concrete occurred during the 4.42% drift ratio. This was later than what
was observed in the testing of the cast-in-place benchmark. The failure mechanism for the connection
started with an “Elephant-Leg” buckling of the pipe over the unbonded region, followed by the fracture

of the pipe on one side. There was a rapid decrease in the lateral capacity of the column when the buckling
increased. The maximum force obtained during the testing procedure was 41.2-kip which was
approximately 10% higher compared to the cast-in-place benchmark. The maximum displacement
achieved was 9.4 in. which is about 20% higher compared to the cast-in-place benchmark. The ultimate
displacement ductility is just below 6.5. The lower value of ductility compared to cast-in-place (7.5) was
the fact that the precast column had a higher yield point compared to cast-in-place. The total energy
dissipated during the testing procedure was 1,025 kJ. This was 2.25 times of the cast-in-place benchmark.
The hysteretic damping plot showed that the precast column, overall, had similar values of hysteretic
damping of Takeda-Fat and Ramberg-Osgood models. This is thought to be due to enhanced confinement
(e.g., Concrete Filled Steel Tube) and detailing considerations. The precast column reached a hysteretic
damping of 20% at the failure point.
Cantilever Columns: Precast vs. Cast-In-Place

Both columns are designed from the same prototype structure and as a result, are designed and built with
the same overall dimensions. The columns had a height of 7 ft-4 in. and a diameter of 18 in. The designed
moment capacity of the cast-in-place column is 165 kip-ft, after the AASHTO LRFD Bridge Design
Specifications (AASHTO 2017), and the moment capacity of the precast column pipe connection is 143 kip-
ft, after the WSDOT LRFD Bridge Design Manual (WSDOT 2019).
The testing setup and procedures are identical for each of the columns except the precast column endured
through more cycles to reach its failure point. The cast-in-place column reached 50% degradation of the
highest observed lateral force on the 13th cycle, which corresponds to 7.7 in. of displacement and 9.9%
drift after the displacement of the footing is removed from the data. The precast column reached 50%
degradation of the highest observed lateral force on the 16th cycle, which corresponds to 9.57 in. of
displacement and 12.3% drift after the displacement of the footing is removed from the data. The loading
protocol for the cast-in-place and precast columns is shown in Figure 25 and Figure 40 respectively. Figure
26 and Figure 41 show images of the cast-in-place and precast columns, respectively during the testing
procedure at approximately the same displacements.
The cracking drift ratio for the cast-in-place and precast columns are 0.2% and 0.6%, respectively. The
precast column exhibited flexibility during smaller drift ratios to prevent cracking to the column.
The maximum displacement of the cast-in-place column during the testing procedure is 7.7 in. which
corresponds to a 9.9% drift. The maximum load applied to the cast-in-place column during the testing
procedure is 37.8 kip which corresponds to a 245.7 kip-ft moment capacity. The maximum displacement
of the precast column during the testing procedure is 9.6 in. which corresponds to a 12.3% drift. This is
20% higher compared to the cast-in-place benchmark. The maximum load applied to the precast column
during the testing procedure is 41.2 kip which corresponds to a 267.8 kip-ft moment capacity. This is 9%
higher compared to the cast-in-place benchmark. The Force vs. Displacement plots for the cast-in-place
and precast testing procedures are shown in Figure 27 and Figure 42, respectfully. The Force vs. Drift plots
are shown in Figure 28 and Figure 43 for the cast-in-place and precast testing procedures, respectively.

Looking at each of the hysteresis plots, it is obvious that the area of the loops for the precast column is
larger compared to the cast-in-place column. In each of the plots, the enclosed area values are larger for
the precast column than the cast-in-place column. It can be observed that the cast-in-place column
yielded sooner compared to the precast column. The precast column had a yield drift which was 30%
higher compared to the cast-in-place column.
Figure 48. Backbone Curve for the Cast-In-Place and Precast Columns
The backbone curve for both testing procedures is shown in Figure 48. It can be observed that the precast
column has a lower stiffness compared to the cast-in-place column. The cast-in-place column had a
consistent stiffness of 36.7 kips/in. The precast column begins with a stiffness of 42.4 kips/in., transitions
to 16.44 kip/in., then has a stiffness of 8.5 kips/in. before achieving the maximum lateral load. Comparing
the values, the precast column initially has a stiffness 16% higher, then 45% lower, next 23% lower
compared to the cast-in-place benchmark column. However, the precast column reached the capacity of
the cast-in-place column at about a 3% drift ratio. The lower stiffness of the precast column is due to the
rocking movement to allow flexibility and prevent concrete cracking at lower drift ratios. The ultimate
displacement ductility for the precast column was nearly 6.5 which was slightly lower than the cast-in-
place (e.g., 7.5) This was mainly due to larger yield drift for the precast column. Table 5 and Table 6 shows
significant values from the cast-in-place testing procedure while Table 9 and Table 10 show significant
values for the precast column testing procedure.
The curvature distribution along the height of the first 54 inches of the cast-in-place and precast columns
are shown in Figure 30 and Figure 45, respectively. Each of the columns is expected to yield and fail within
the plastic hinge of the column. The plastic hinge of the cast-in-place column is 13.7 in. The plastic hinge

for the precast column is 13.07 in. Each of the curvature figures shows that yielding occurred in the bottom
18 inches of the column. From 18 inches to 36 inches, the columns approached the yield point but never
reached it. Because the region from 18 inches to 54 inches never reached the yield point, this region
deformed elastically throughout of the testing procedures. The maximum calculated curvature the cycle
corresponding to the 10th cycle or 7.99% drift for the cast-in-place column is 0.0038 radian compared to
0.0036 radians for the precast column. These values are nearly identical. There were several large cracks
in the plastic hinge length of the cast-in-place column. However, for the precast column, there was
essentially one large crack (e.g., gap opening) in the beginning until the spalling of the cover concrete.
The dissipated energy is obtained by finding the area of each loop in the Force vs. Displacement plot after
the units are converted to newtons and meters. The area of each loop represents the total energy
dissipated in units of joules. The total dissipated energy during the cast-in-place testing procedure is 456
kJ (336,327 ft-lb). The total dissipated energy during the precast testing procedure is 1025 kJ (755,999 ft-
lb) which is more than 2.2 times the cast-in-place column. However, comparing the cumulative dissipated
energy plots at the end of the cycle just before the cast-in-place column’s first rebar break, 8.87% drift,
the cast-in-place column dissipated 293 kJ (216,105 ft-lb) compared to 388 kJ (286,173 ft-lb) for the
precast column. Figure 49 and Figure 50 show the dissipated energy for both the columns. One reason
why the dissipated energy is higher for the precast column is that there is more steel at the column
connection which makes the column more ductile and able to absorb more energy. Also, the precast
column had better confinement (e.g., presence of a steel shell) compared to the cast-in-place benchmark.
Figure 49. Cumulative Dissipated Energy for Both Columns (1 kJ = 737.56 ft-lb)

Figure 50. Dissipated Energy for Each Cycle for Both Columns (1 kJ = 737.56 ft-lb)
Figure 51 shows the corrected area-based hysteretic damping plots for both the cast-in-place and precast
columns. It can be observed that the precast column had higher values of hysteretic damping compared
to the cast-in-place column up until a ductility of nearly 5.5. The initial jump in the hysteretic damping
plot of the precast column is due to contact damping provided by the elastomeric pad. The effects reduce
as the pipe starts yielding with increasing displacement ductility.

Figure 51. Corrected Area-Based Hysteretic Damping for Both Columns
Figure 52 shows the residual drift for the cast-in-place and precast columns. Looking at the image, the two
columns follow a similar trend line throughout their testing procedures. The residual drift of the cast-in-
place column on the last cycle of the testing procedure is 6.64% compared to the 9.27% drift that the
column was pushed. This corresponds to the column maintaining 70% of the displacement applied to the
column when the lateral force was equal to zero. The residual drift of the precast column was 10.91%
compared to the 12.08% drift that the column was pushed or 90% of the displacement of the column was
maintained when the column lateral force was equal to zero. If the cast-in-place column could continue
through more of the test cycles, the residual drift results are expected to continue following the precast
cast column results.

Figure 52. Residual Drift for the Cast-In-Place and Precast Columns
Each of the cantilever columns is designed and constructed having the same overall dimensions and
approximately similar capacities. The cast-in-place and precast columns are tested using identical loading
protocols and displacement progression for each cycle. The columns perform similarly throughout the
testing procedures. The precast column, however, continued through more loading cycles and as a result,
achieved higher deflections and cumulative dissipated energy. Overall testing shows better performance
of the precast column compared to the cast-in-place column. It reached higher displacements with good
strength. The residual deformation for both columns is comparable. Table 1 summarizes some of the key
data from the testing of the two columns.

Bent Pier System: Cast-In-Place
Introduction
This section presents design, construction, and experimental testing of a Cast-In-Place (CIP) bent system
with the intention of establishing a performance level in which to compare the precast bent system using
the proposed pier connection. A review of the construction process is thus presented discussing the
challenges faced during a CIP construction project and the work required. The full testing arrangement
used for the experimental work is presented and discussed. Followed by the experimental testing carried
out on the system and its resulting performance.
Prototype Structure
The first steps in developing the specimen is the determination of the overall size of the specimen. As the
research aims to test the proposed connection at a large scale, the specimen sizing is determined near
the maximum capacity of the testing facility, Idaho State University Structural Laboratory (SLAB). The
overall specimen itself is considered to be sized as a scaled version of a typical mid-to-long span bridge
constructed in south-east Idaho. South-east Idaho is proposed to be the place of construction as it is the
most seismically active area of the state where the proposed connection is to be used. An example of a
typical mid span bridge in south-east Idaho is presented in Figure 53. This particular bridge is constructed
over the Bear River near Preston, Idaho about 70 miles south of Pocatello, Idaho.
The bridge consists of two 137 ft spans. Each span is set between the bridge’s abutments and a center
bent system located approximately in the center of the river. The bent system is comprised of three
octagonal columns measuring 4 ft. in width and having an overall height of 29ft-3in. The column cap
measuring 40 ft. in length, 5 ft. wide, and 4 ft-6 in. deep and the foundation being 40 ft. in length, 22 ft.
wide, and 5 ft-6 in. deep. A detail of the bent is provided in Figure 54.

Figure 53. Elevation and Top View of SH-36 Bridge Over Bear River
Figure 54. Bear River Bent System Elevation View

Through the consideration of the capabilities within the SLAB, maximum dimensions were established for
a height of the specimen and associated setup to not exceed 13 ft. and the specimen’s overall length not
to exceed 15 ft. After factoring in the items required for loading and monitoring the specimen during
testing the specimen height was determined to be 11 ft. in total height and have a pier cap length of 15
ft. For determining the depth of the pier cap and footing segments consideration of the embedment
required by the HSS pipe in the proposed precast connection had to be considered in order to achieve
similar dimensioning between the CIP and precast bent systems. Through this a required depth of 2 ft-6
in. is necessary for both the footing and pier cap. The footing width and length is determined based on
the layout of the SLAB strong floor which has embedded anchors in an 18 in. x 18 in. square pattern
throughout the floor. Considering the centrally located pier in each footing and providing adequate
anchorage the footings are required to be 4 ft. x 4 ft. square.
For sizing of the columns, past experiments performed in the lab were considered to determine the final
width. Previous experiments on single piers acting as cantilevers had been performed with a column width
of 18 in. with steel reinforcing ratios of 2% (r=2%). While also attempting to match the capacity of the
previous experiments a reduced column width of 14 in. was determined suitable as the bent system would
produce higher demands during testing than previous cantilever columns tested. From the column width
and consideration of cap reinforcing and cover concrete a 2 ft. width for the cap is necessary. Through
consideration of the lab limitations, past experiments, typical bent ratios, and requirements for the both
the CIP and precast specimen the resulting specimen depicted in Figure 55 is determined.
The reason behind using much stiff beams and footing is to make sure no non-linear deformation occur
in those elements as cap beam and footing are capacity protected elements. The cap beam and footing
was made rigid.

Figure 55. CIP Bent Specimen
Design of the CIP Specimen
After establishing the overall specimen dimensions to accommodate the limitations and experimental
goals, the design of the system is performed. For properly designing the bent the 8th Edition of AASHTO
LRFD Bridge Design Specification (AASHTO 2017) is considered. Through the design process the reinforcing
steel is determined for accomplishing the targeted strength of system. An analysis of the lab capabilities
and past experiments deemed a safe target strength of 60-70-kip force applied during testing as an
achievable target force applied to the bent system.
Considering each individual pier as an individual system comprised of two connections, footing-to-column
and column-to-cap, each contributing to the overall strength of the bent system, an approach which
considers each connection to constitute 25% of the overall pushover force required for the system was
established. Considering the previously established safe operating force for the lab (60-70-kip) a target
force of 15-kip is considered for the design of each pier connection. This 15-kip force is considered as the
base shear force for each connection. With a reinforcing ratio of 2% the longitudinal reinforcing steel is

determined to be seven #6 Grade 60 rebar. Grade 60 rebar has a yield strength of 60 ksi and modulus of
elasticity of 29,000 ksi. Additionally, in accordance with AASHTO design a continuous reinforcing spiral is
provided throughout the length of the column. The #3 Gr. 60 reinforcing spiral is broken at both the
footing-to-column and column-to-cap interface, with the spiral continuing in both the footing and cap for
the full length of the longitudinal reinforcing. The #3 spiral is terminated at each end by mechanical splice.
The #3 spiral is spaced at a constant pitch of 1.5 in. throughout the full length of the longitudinal
reinforcing. The required cover concrete of 1.5 in. is provided between the reinforcing spiral and column
surface.
The resulting connection capacity is thus calculated considering a normal weight concrete with
compressive strength, f’c, of 4,000 psi. The column design moment capacity of approximately 61.7-kip-ft.
is calculated. Considering the loading height to the actuator center of 83-3/4 in. from the footing surface,
the resulting design base shear is 8.8-kip. The ultimate base shear is 13.6-kip which is 91% of the target
base shear per connection as discussed above. Note threaded terminators are used in the cap to develop
the required embedment strength where reinforcing congestion makes it difficult to utilize hooks or bends
for producing development length. A column cross-section is provided in Figure 56 below, with a column
detail in Figure 57.
Figure 56. CIP Pier Detail

Figure 57. CIP Pier Cross-Section
For the cap design, a computer modeling program, SAP-2000, is used to determine the required moment
and shear demands experienced by the cap. The full bent is developed from the footing surface up. As the
experiment is being carried out as an investigation of the pier connections a conservative approach
toward designing the pier cap is taken in order to ensure the failure is forced to the four pier connections.
The resulting moment and shear demand within the cap are 340-kip-ft. and 111-kip, respectively. As a

conservative approach is taken to ensure failure is forced to the columns the design moment and shear is
taken to be 500-kip-ft. and 150-kip, respectively. The design process using concrete of compressive
strength of 4,000 psi, yields a required 11 #6 Gr. 60 rebar be provided at the top and bottom with #4 Gr.
60 stirrups provided at 4 in. center-to-center spacing. Additionally, as per ITD general practice two
alternating #4 Gr. 60 cross-ties are provided for each stirrup. Figure 58 and Figure 59 provide the final CIP
cap beam drawings. Cap beam and cap beam to column panel zone or connection can be designed using
strut and tie method for shear reinforcing.
Figure 58. CIP Cap Beam
Figure 59. CIP Cap Beam Cross-Section

The footings are similarly designed in a conservative fashion to ensure failure is forced into the columns.
As the footing dimensions are largely controlled by the lab and proposed precast connection
requirements, the reinforcing is determined as 10 #6 Gr. 60 rebar provided at the top and bottom in each
direction. Using Response-2000 the footing is calculated to have an ultimate moment capacity of 532-kip-
ft providing a significant factor of safety over the moment demand created by the pier connection. Figure
60 provides a top view of the footing and rebar layout for the bent system.
Figure 60. CIP Footing
Construction
After determining the final dimensioning and design details, the CIP construction began. Typical CIP
construction is performed completely onsite with multiple in-place concrete pours taking place. For
construction of a bent substructure there are three main pours, footing, columns, and cap. Forming and
rebar fabrication are performed simultaneously throughout the construction. Wood form work is
determined as a suitable material as it can serve for both the bents constructed and is considerably more
cost effective and less labor intensive than producing steel forms for two specimens.

The footing, longitudinal, and spiral reinforcing in the column is prepared as a singular cage for each
individual pier (Figure 61). The column cage is prepared and then lifted into the lower rebar mat of the
footing (Figure 61a, b) with the upper mat finished last (Figure 61c). After the footing and column are
completed, the sleeves are placed for the floor anchors (Figure 61d). Finally, the concrete is placed,
courtesy of Pocatello Ready Mix, finishing the footings up to the footing-to-column interface (Figure 61e)
The interface where the column is to be poured is left rough to assist in bonding. The total concrete order
for pouring of the footings was 3.25 cubic yards.
Figure 61. CIP Footing Construction: a) Column Cage, b) Place Column Cage, c) Tying Top Mat, d)
Anchor Sleeves, e) Footing Pour
After footings have adequately cured the form work is removed and the footings are relocated to the SLAB
where they are anchored in place (Figure 62). The column spiral is checked for proper placement and
secured. For the pouring of the columns in order to follow proper concrete placing procedures the pier
form work is built in two 3 ft. segmental sections, which can be assembled during pouring. This ensures
the concrete is not dropped at too great a distance resulting in segregation and proper vibrating is
accomplished throughout the full column. The full 6 ft. of the two columns are poured so as not to have

a cold-joint present in the length of the column. At 7 days the pier formwork is removed allowing for the
cap pour preparation to begin (Figure 63). The pour for the two columns was a total of 1 cubic yard.
Figure 62. CIP Footing Placement
Figure 63. Completed CIP Pier
As the cap has to be poured in the lab at a height of 8 ft-6 in., false-work for supporting the concrete
during initial curing is necessary. Making use of existing items in the lab proved the most efficient way to
construct false work. Figure 64 provides a view of an assortment of reaction frames and sections serving
to provide the necessary false work for completing the cap pour. The cap reinforcing cage is started on
the ground (Figure 65a) and lifted into place over top the longitudinal column reinforcing (Figure 65b).
Then the final stirrups, cross-ties, form-work, and false-work is placed for pouring (Figure 65c)

Figure 64. Cap False-Work
Figure 65. CIP Cap Construction: a) Cap Cage, b) Cap Cage Placement, c) Cap Complete False-Work,
Cage, and Formwork

The full cap, measured to be a total of 75 cubic feet of concrete and estimated to weigh 11,000 lbs., is
completed in a single pour to eliminate cold joints (Figure 66). The completion of the cap marked the final
pour required for the construction of the CIIP bent system. The total CIP bent is constructed of
approximately seven cubic yards and estimated to weigh in the range of 26,500 lbs. to 28,000 lbs. After
the cap had adequately cured the false-work and formwork is removed, the specimen is painted, and
prepared for instrumentation (Figure 67).
Figure 66. Completed Cap Pour
Figure 67. CIP Bent Prepared for Instrumentation
Testing Arrangement
After the CIP bent construction is completed the testing arrangement is erected. For the purposes of
testing the connection, a uniaxial load is applied in the transverse or perpendicular direction of the bridge
deck. The purpose for loading in this direction is due to a full bridge structure being weak in the transverse
direction. This is based upon the assumption that a full bridge, including the superstructure, has
significantly higher resistance to loading parallel to the superstructure as the bridge integral abutments
provide adequately stable resistance to such loading. An additional vertical load is applied to the system
during the entire testing procedure serving as a gravity simulant.
The lateral load is applied cyclically in accordance with the American Concrete Institute (ACI) (ACI
Committee 374 2013) via a hydraulic servo-valve actuator. The actuator is a 2.5 gallon-per-minute (gpm)
servo-valve actuator with a total achievable stroke of 24 in. A 225-kip tension/compression loadcell is
mounted in-line with the horizontal actuator in order to monitor the actuator force during the
experiment. The actuator is mounted horizontally to the reaction frames. The reaction frames are
comprised of two identical columns fabricated of steel channel, C12x30, laced together with 0.25 in. flat
plate. Additional channel and wide flange angle bracing are provided to support the columns. In total the
reaction frame is anchored to the floor via twenty-four 1 in. diameter high strength threaded rods (Figure
69). The head of the actuator is attached to the cap beam using four 1 in. diameter 120 ksi all thread.
During casting of the cap, embedded anchors were fabricated into the cap end. The embedded anchors
were provided additional anchoring support in the cap via high strength threaded rod connected to an
anchoring plate embedded 12 in. into the cap beam (Figure 68). The specimen itself is secured rigidly to
the SLAB strong floor using nine 1 in. rods. The foundations are assumed to be rigidly fixed disallowing
soil-structure interaction effects during the testing.
Figure 68. Actuator Cap Anchors
The final phase of the testing arrangement is the application of the gravity load applied vertically at the
center of the cap beam. The vertical force of the gravity load corresponds to 4.5% of the axial compressive
capacity of the columns. Typically, the target gravity load is 5% of the axial compressive capacity of the
specimen, but due to equipment limitations and the limited 3000 psi hydraulic pump the highest
achievable axial ratio is 4.5%. For the two 14 in. columns of 4 ksi concrete, the gravity load is determined
as approximately 60-kips. The equation for determining the gravity force is given as:
𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺 𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹𝐹 = 0.045�𝐴𝐴𝑔𝑔 𝑓𝑓𝑐𝑐′ �
Where:
Ag = Gross cross-sectional area of the two columns, in2

The gravity force is applied using a 100-ton 4 in. hollow jack. The jack is place on the bent cap, a 225-kip
tension/compression loadcell is stacked on top of the jack, and a reaction beam is placed across the top
of the loadcell. Two high strength all thread bars are used to bolt the reaction beam to the floor to provide
the resisting downward force to the cap. The full test setup is provided in Figure 69 below.
Figure 69. CIP Bent Testing Arrangement
Instrumentation
Throughout the experiment specific measurements of the system are monitored in order to document
the specimen’s response and characterize its overall performance. Various instruments are used including
loadcells, linear potentiometers, string potentiometers, and strain gages. The instruments are
programmed using a Campbell Scientific data acquisition system. The system is uniquely programmed for
all the instruments and set to take five readings per second throughout the experiment. A total of 95 total
instruments are used during the experiment. Figure 70 to Figure 72 provides a visual of the
instrumentation layout on the CIP specimen. Note all instrumentation specifically associated with either
the north or south pier are indicted as “XX-N” and “XX-S”, respectively.
Figure 70. CIP Instrumentation Layout: Front Elevation View
Figure 71. CIP Instrumentation Layout: Top View
Figure 72. CIP Instrumentation Layout: Rear Elevation View
Specific instruments are mounted as such to measure the global movement of the specimen. These
instruments are mounted independent of the specimen and test arrangement in order to provide a true
displacement of the specimen. The instruments mounted independently include CAP-INPLANE, CAP-OP-
S, CAP-OP-N, R1-N, R2-N, R1-S, R2-S, S1-N, S2-N, S1-S, and S2-S. The instrument labeled CAP-INPLANE is a
string-potentiometer, used to measure the true displacement of the cap. It is mounted directly at the
center of the actuator on the opposite end of the cap beam. As it is assumed the cap does not experience
any noticeable compression during testing, this measurement is used as the true displacement of the
specimen. CAP-OP-S and CAP-OP-N are string potentiometers used to monitor the out-of-plane motion of
the cap. These measurements are predominately monitored to ensure the cap does not move excessively
to one side or the other during testing. Monitoring of this measurement reaffirms the stability of the
specimen throughout the experiment and helps to monitor the risk of out-plane collapse. Linear spring
potentiometers are used for measurements R1-N, R2-N, R1-S, R2-S, S1-N, S2-N, S1-S, and S2-S. The “R”
refers to “rocking” as these instruments are mounted vertically, or parallel to the piers, to monitor any
lifting of the footing edges, both in- and out-of-plane. Similarly, “S” refers to “sliding” as these
potentiometers are used to monitor sliding of the footing in both the direction of loading and the
transverse direction. Additionally, a string potentiometer is attached to the actuator to confirm the true
stroke of the actuator itself. As this is a non-independently mounted instrument it is simply used to
confirm the motion of the actuator head.
Instruments occupying groups “A” through “E” are comprised of both linear spring potentiometers and
linear potentiometers with aluminum extensions. The group “A” potentiometers monitor the plastic hinge
zones located at the base of each column. Group “D” and “E” potentiometers monitor the plastic hinge
zones located at the top of each column. Groups “B” and “C” are used to monitor any curvature
experienced by the column falling outside of the plastic hinge zones.
Additionally, two 225-kip tension/compression loadcells are used to monitor the lateral load and vertical
gravity load induced on the specimen. Finally, elongation of the rebar is monitored using strain gages.
Strain gages are attached to the longitudinal reinforcing bars just above the footing-to-column interface
and just below the cap-to-column interface prior to pouring concrete.
Loading Protocol
As mentioned earlier the loading protocol for the specimen is determined in accordance with the ACI (ACI
Committee 371 2013). The quasi-static cyclic loading protocol is based off the yield displacement of the
bent itself. Figure 73 shows the loading protocol given as a graph of cycle number versus the drift ratio.
Figure 73. ACI Loading Protocol
Determination of the yield displacement is derived using an equation provided by the Priestley et al.
(2007) as shown Figure 74. Given the following two equation yield displacement, ∆, is calculated:
∆𝑦𝑦 = ∆𝑦𝑦1 + ∆𝑦𝑦2
2
�𝐿𝐿1 + 𝐿𝐿𝑠𝑠𝑠𝑠 �
∆𝑦𝑦1 = 𝜑𝜑𝑦𝑦
3
2
�𝐿𝐿2 + 𝐿𝐿𝑠𝑠𝑠𝑠 �
∆𝑦𝑦2 = 𝜑𝜑𝑦𝑦
3
Where:
∆y1, ∆y2 = Yield drift for each short column, in.

φy = Curvature in the column corresponding to the first bar yield point, in-1
L1 = Column height, in.
Lsp = Strain Penetration Length, in.
φy and Lsp were then obtained for a conventional CIP constructed column using equations below:
𝐹𝐹𝑦𝑦𝑦𝑦
𝜑𝜑𝑦𝑦 = 2.25
𝐸𝐸𝐸𝐸
𝐿𝐿𝑠𝑠𝑠𝑠 = 0.15𝐹𝐹𝑦𝑦𝑦𝑦 𝑑𝑑𝑏𝑏
Where:
Fye = 1.1 times the yield strength of steel; Yield strength of steel, 60 ksi
E = Modulus of elasticity of steel, 29,000 ksi
D = Diameter of the column, 14 in.
db = Diameter of reinforcing longitudinal bar, in.
Figure 74. Displacement Capacity of a Pier in a Bent with Fixed-Fixed Supports (Caltrans 2013)
Considering the above equations as applied to the CIP specimen with a column diameter of 14 in. (for
design) and comprised of #6 longitudinal reinforcing, a resulting yield drift total of 0.46 in. is determined.
So, 0.35 in. is used for the programming of the actuator controller to ensure two cycles are performed
prior to reaching yield for instrumentation and test set up tests.
From the determination of the yield displacement the final loading protocol used is determined and
graphed, as shown in Figure 75. During the testing, a loading rate of 1 mm/sec is used. As the yield
displacement is multiplied for each set of additional cycles, time is taken to observe the response of the
specimen during each two-cycle set. The cycles are continually increasing in displacement magnitude until
the bent demonstrates a 20% degradation in strength or is determined to be unsafe for continued loading
due to possible collapse.
Figure 75. CIP Bent Loading Protocol
Material Properties
Following the completion of testing, the concrete samples prepared during each stage of pouring had to
be tested to confirm the concrete properties on test day. Three concrete samples from the footing, pier,
and cap pours were tested to verify the respective compressive strength of the concrete. Also, two split
tension cylinder samples were tested. The 28-day compressive strength of concrete, f’c, is targeted to be
4000 psi. However, the laboratory testing of the concrete did not show the desired compressive strength
of columns and cap which could be because of the problems associated with the way concrete was mixed
in the concrete yard which is common. Table 12 and Table 13 provide a summary of the test day
compressive strength and split tension results, respectively.
Table 12. CIP Bent Test Day Compressive Strength
Sample 1 2 3 Average (psi)

Footing 6974 7186 7121 7094
Columns 2994 3211 3535 3247
Cap 3084 3276 3266 3209
Table 13. CIP Bent Test Day Split Tension
Sample 1 2 Average (psi)

Footing 533 603 568
Columns 352 427 390
Cap 443 374 409
Testing Results
Starting the experiment at the first cycle of ± 0.18 in., cracks formed at the interface of the column-to-
footing and column-to-cap (Figure 76a). Hairline cracking began in only the face of the column away from
the actuator, noted as the “North” pier for instrumentation purposes, during the second cycle. The
interface cracks widened during this cycle as well. During the third cycle hairline cracking developed at all
connections, with the furthest forming up to 19 in. from the interface (Figure 76b). During the fourth cycle
cracks measuring 0.0157 in. (0.4 mm) began to open within 12 in. of the interfaces (Figure 76c).
Additionally, hairline cracking was extended up to 26 in. from the interface. The cracks at the interface
then continued to widen. During the fifth cycle spalling began developing at the column top connections
with cracks developing continuing throughout the column. The cracks in the column face were opening as
wide as 0.0394 in. (1 mm) while in tension. The sixth, seventh, and eighth cycles saw continued crack
development up to 0.0984 in. (2.5 mm) and additional slight spalling at the connection faces (Figure 76d).
The ninth cycle is the first which concrete spalling developed at the base column connections (Figure 76e).
The ninth cycle is also where the largest measured force, approximately 66-kip, takes place. From this
point forward, the specimen strength begins to degrade.
The tenth, eleventh, and twelfth cycles saw cracking continue to develop and spalling develop to exposure
of the spiral reinforcing at the top of the column near the actuator (Figure 76f). It is during the fourteenth
cycle which exposure of the longitudinal reinforcing occurred at the top of the column near the actuator
(Figure 77g). The fifteenth and final cycle results in the fracture of longitudinal reinforcing at the top
connection of the column nearest the actuator resulting in a significant drop in lateral force (Figure 77h).
Having the fifteenth cycle finishing with a max lateral force of 53-kip (80% of greatest lateral force
experienced) the test was completed.
An item of note during the test is the flex experienced by the reaction frame. This constituted to a lower
achieved specimen displacement that targeted at each cycle. This is due to the inability of the reaction
frame to be completely rigid against the lateral force of the actuator. Table 14 provides a recap of the
target values programmed for the actuator and the actual displacements experienced by the specimen,
as measured by the independent string potentiometer labeled CAP-INPLANE. The ultimate drift achieved
by the bent prior to 20% strength degradation was 4.94%. This is comparable to the targeted displacement
programmed for the thirteenth cycle. Also, from observing the actual displacements it can be seen that
the yield drift was achieved on the third cycle rather than the second as targeted by the program. This is
important to note as the specimen experienced an extra cycle at a very low drift of 0.11%.
a) b)
c) d)
e) f)
Figure 76. CIP Bent Testing: a) Cycle 1: Interface Cracking, b) Cycle 3: Hairline Crack Development, c)
Cycle 4: 0.0157 in. (0.4 mm) Crack Development, d) Cycle 8: Spalling and 0.0984 in. (2.5 mm) Cracking,
e) Cycle 9: Base Connection Spalling, f) Cycle 13: Spiral Exposure
Figure 77. CIP Bent Testing: g) Cycle 14: Longitudinal Exposure, h) Cycle 15: Longitudinal Rebar
Fracture
Table 14. CIP Bent Loading Protocol Summary
1 0.18 0.21 0.09 0.11
2 0.35 0.42 0.15 0.18
3 0.7 0.84 0.35 0.42
4 1.06 1.27 0.56 0.67
5 1.41 1.68 0.86 1.03
6 1.76 2.10 1.08 1.29
7 2.11 2.52 1.4 1.67
8 2.46 2.94 1.73 2.07
9 2.81 3.36 2.06 2.46
10 3.17 3.79 2.4 2.87
11 3.52 4.20 2.75 3.28
12 3.87 4.62 3.1 3.70
13 4.22 5.04 3.44 4.11
14 4.57 5.46 3.78 4.51
15 4.92 5.87 4.14 4.94
The maximums achieved during the testing of displacement and lateral load were 4.14 in. and 66-kip,
respectively. The lateral load correlates to total moment capacity of 460-kip-ft. If assumed the four
connections shared the lateral load equally, this equates to base shear at each connection of 16.5-kip.
Figure 78 and Figure 79 provide Force vs. Displacement and Force vs. Drift hysteresis of the full CIP bent
testing. As can be seen the specimen reached its design base shear of 35.2-kip. The hysteresis suggests
the bent yielded at 0.42 in. displacement. Similarly, from the Force-Drift hysteresis it is seen the bent
yielded at a drift ration of 0.5%.
As can be noted from the figures, the positive vertical axis shows the specimen in push. As the test began
by first pulling the specimen, and continued to begin all cycles in pull, it can be observed the bent had
higher strength in pulling than push can be attributed to two factors. The bent underwent softening during
the first pull of the cycle thus exhibiting higher strengths in all cycles. Additionally, the reaction frame
exhibited slightly higher stiffness during the pulling stage as opposed to experiencing higher displacement
during the pushing stage.
Figure 80 provides the resulting Force-Drift backbone curve. The backbone curve is comprised of the peak
loads achieved at each cycle. Observation of the curve provides further evidence of the bent performance
and its yield progression.
Figure 78. CIP Bent Force-Displacement Hysteresis
Figure 79. CIP Bent Force-Drift Hysteresis
Figure 80. CIP Bent Force-Drift Backbone Curve
Further analysis of the potentiometers, specifically located in the plastic hinge regions (groups A, B, C, and
D), provides a close breakdown of each connection’s reaction through moment-curvature. The
progression of the yield that is captured by the instrumentation at each plastic hinge can be observed in
Figure 81 through Figure 84, providing the moment-curvature of each. Observation of the top of the south
column (Figure 81) shows a narrower hysteresis produced as opposed to the other connection. This
correlates to the level of damage and spalling observed at each connection, with this particular connection
being the one to sustain the most and ultimately fail of the longitudinal reinforcing as noted in the
previous section. Further observation of the plots shows a consistent increase in the strength degradation
at each cycle once the bent reached its capacity. Through comparing the two south column plots to the
north plots it can be seen that the south column experienced a higher level of drift correlating to the
increased damage observed on the column during testing. Therefore, the observations made during
testing have a good correlation with the experimental results collected via the instrumentation.
Figure 81. South Column: Top (D1-S – D4-S)
Figure 82. South Column: Bottom (A1-S – A4-S)
Figure 83. North Column: Top (D1-N – D4-N)
Figure 84. North Column: Bottom, (A1-N – A4-N)
The energy dissipated per each cycle drift ratio for the bent is presented in Figure 85. The dissipated
energy was calculated using numerical integration of the hysteresis loop at each cycle considering the
area enclosed within the loop. To accomplish this a MATLAB program is utilized to break the hysteresis
down to individual loops and calculate the enclosed areas, which are then summed together resulting in
a “Cumulative Dissipated Energy”. For the CIP bent specimen the resulting cumulative dissipated energy
is expressed in kilojoules (kJ). The result is 342 kJ.
Figure 85. CIP Bent Dissipated Energy per Cycle and Cumulative
Experimental results are used to determine the experimental yield curvature and yield moment by
Caltrans Idealized Model (Caltrans 2013). A bilinear approximation similar to the example is provided in
Figure 86. The moment capacity can be obtained by balancing the area between the idealized M-ϕ and
actual. The global yield curvature and yield moment is determined to be 0.702 in-1 and 380-kip-ft.
Idealized
Actual
Figure 86. Moment Curvature Curve (Caltrans 2013)
Where:
ϕy = Curvature at the first bar yield point (in-1)

ϕY= Curvature at the global yield point (in-1)
ϕu = Ultimate curvature at the failure point (in-1)
My = Moment capacity at the first bar yield point (kip-ft.)
MP = Plastic moment capacity (kip-ft.)
For this experiment the backbone curve is analyzed using displacement in place of curvature providing
the following results. The backbone curve average of the push and pull direction is considered. The global
yield moment capacity produced from the experimental results is used to obtain the base shear yield of
56.2-kip corresponding to a yield displacement of 0.596 in. The ultimate base shear provided from the
backbone curve gives a total base shear of 61.9-kip at a displacement of 2.2 in. The bilinear approximation
is shown in Figure 87, below.
Figure 87. Bilinear Approximation for CIP Bent
Further analysis of the experimental results allows for the determination of the overstrength factor (Ω0),
an important seismic parameter. The overstrength factor is determined as the ultimate base shear
capacity at ultimate (Vultimate) divided by the base shear at initial yield (Vyield). The equation is given below.
The resulting overstrength factor of 1.76 is obtained.
𝑉𝑉𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢𝑢
𝛺𝛺0 =
𝑉𝑉𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦
Where:
Vultimate = Ultimate base shear capacity at ultimate, kips

Vyield = base shear at initial yield, kips
Ω0 = Overstrength factor
Displacement ductility is calculated in a similar fashion as seen in the equation below and provides further
seismic parameters on the performance of the CIP bent.
𝛿𝛿
𝜇𝜇 =
𝛿𝛿𝑦𝑦
Where:
δ = Displacement at the ultimate base shear point on the backbone plot (in.) for the displacement
ductility at the ultimate base shear capacity
δ = Displacement at 0.8Vultimate in the backbone plot (in.) for the ultimate displacement ductility
δy = Deflection at yield (in.)
Resulting in a displacement ductility of 3.69 for ultimate base shear and 7.48 at failure point.
The residual drift of the CIP bent is presented in Figure 88, reflecting the permanent deformation of the
columns after the completion of each cycle. At the point of failure, the CIP bent was maintaining 61.5%
(3.04% drift ratio) of the drift applied, 4.94%.
Figure 88. Residual Drift of CIP Bent
Summary
The CIP bent is fully constructed to simulate the traditional construction and design of a typical mid to
long-span Idaho bridge. For this reason, the design followed closely the requirement in AASHTO LRFD
Bridge Design Specifications (AASHTO 2017). Similarly, the full specimen is poured in a staged fashion that
is seen on a typical CIP construction project using traditional materials and methods. The experimental
loading program determined from ACI “Guide for Testing Reinforced Concrete Structural Elements under
Slowly Applied Simulated Seismic Loads” resulted in a ductile specimen response. The CIP bent achieved
the design base shear of 35.2-kip, correlating with yield drift ratio of 0.5% (0.42 in.) at first yield, with an
approximated global yield point of 0.7% (0.596 in.). The system achieved a maximum base shear and total
displacement of 66-kip and 4.14 in., respectively. The bent responded similarly in all four column
connections, as similar plastic hinges developed at each connection. Ultimately, as the experiment
progressed the column nearest the lateral loading actuator began to experience an accelerated
degradation of cover concrete thus, on further loading, the longitudinal reinforcing ruptured at the top
connection. The failure of the longitudinal reinforcing resulted in a significant strength loss bringing the
strength degradation to more than the targeted 20%. With a shear failure eminent in the top connection,
the experiment was terminated to ensure a proper level of safety was maintained. The Force-
Displacement and Force-Drift plots show the specimen had higher strength in pull as compared to push.
This is because the pull cycle is performed first resulting in a softening effect observable in the push of the
cycles. This difference in strength is also due to the reaction frame being stiffer in pull than push.
Additionally, the total energy dissipated during the experiment by the CIP bent resulted in a total of 342
kJ. The CIP bent resulted in an overstrength factor of 1.76 and displacement ductility values of 3.69 and
7.48 for ultimate base shear and failure point, respectively. Overall, the CIP bent performed as expected.
Bent Pier System: Precast
Introduction
This chapter presents the design, construction, and experimental testing of a precast bent system using a
proposed column connection by ITD. A review of the construction process is presented discussing the
challenges faced during the precast construction and the implementation of the proposed connection and
its accompanying aspects. As the proposed connection is comparable to that of CFSTs; a similar approach
which closely follows the 2019 WSDOT Bridge Design Manual (WSDOT 2019), is used for design. As a
comparison to the CIP bent is to be made aspects such as overall specimen dimensions, testing
arrangement, loading protocol, and instrumentation are repeated as they were carried out for the
benchmark CIP bent.
Overview
The proposed connections are to be tested as a precast ABC technology used in bridge substructure
column connections. The technology makes use of HSS pipe, suggested to be similar pipe as used in
structural piles, with a pipe embedded in both the column and footing/cap. The column pipe is typically
protruding half its length from the pier end (Figure 90). The footing/cap has a fully embedded pipe insert
of larger diameter designed to accommodate the column pipe into the footing/cap. Figure 89 provides a
typical footing detail suited for the proposed connection. The full connection assembly provided in Figure
1 shows a typical footing connection. As opposed to traditional CIP pier it can be seen there is no
longitudinal reinforcing to bridge the interface of the two elements. HSS pipe is the only item passing
through the footing-to-column interface. The HSS provides the flexural capacity, shear capacity, and
confinement for the connection. Figure 2 similarly provides the proposed connection detail for the
column-to-cap connection.
Figure 89. Proposed Connection: Precast Footing Detail
The precast footing was placed directly against the floor. The base connection for this specimen were
assumed to be fixed connection and no soil-structure interaction was considered in this research. Since
the footing is over designed and is bulky, there was no compression strut because the footing was
designed to take that.
Figure 90. Proposed Connection: Precast Column Detail
As this connection provides similar total steel area at the interface of the connection in a more condensed
arrangement than typical CIP connections, consideration for a greater yielding of steel must be accounted
for. The pipe is precast into the center of the column which requires a smaller diameter as opposed to
traditional longitudinal reinforcing. For this an unbonded length of the pipe is provided in order to
establish a greater yield area of the steel to be activated during higher seismic loading. This unbonded
length is identified in Figure 1, Figure 2, and Figure 90 on the column HSS and is positioned just above and
below the interface of the footing and cap, respectively.
Additionally, an elastomeric pad is provided at each interface. The requirement for the elastomeric pad is
due to the grouted state of the precast connection after completion. The connection is finished with a
grout poured between the two pipes filling the gap to the interface at the pier. A non-rigid material is
desired to help eliminate any voids that would be present in a dry concrete-to-concrete interface. The
bearing pad helps to seal this connection for the completion of the grout pour and from natural elements
(water, road salt, etc.) during the life of the connection. The bearing pad also allows for slight non-
destructive movement of the connection during low level loading. Resulting in decreased cracking and
spalling during low cycle seismic activity.
As mentioned above the connection is completed using a grout fed into the footing and pushed up
through the void until sufficient flow out the air outlets is achieved. For the cap connection the grout can
be fed from the top of the cap down into the void until similar sufficient flow is produced out the air valve
in the top of the cap. The grout to be used must be non-shrink with a low metallic content to reduce
interaction with HSS pipe in regard to corrosion. For all the HSS pipe it is suggested that all surfaces are
sand-blasted or similarly prepped prior to concrete or grout application. The gap provided around the
column pipe inserted into the embedded pipe is dependent upon the grout specifications and as to allow
for full flow of the grout throughout the connection.
Regarding the footing and cap reinforcing interrupted by the embedded HSS sections, sufficient
development is to be supplied via rebar bends or terminators to ensure full development of the bar. The
remaining elements of the substructure are designed in accordance with the AASHTO LRFD Bridge Design
Speculations (AASHTO 2017) as similar to the CIP bent benchmark specimen as possible.
Precast Bent System Design
The steps taken to design the proposed connection for the purposes of this experiment are considered to
be similar to CFSTs. The concrete filled HSS pipe is the ductile element at the interface of the connection.
The remainder of the pier is designed to remain elastic throughout loading and is designed as a traditional
pier. A consideration made for the proposed connection is the assumption that the unconfined concrete
cover cast around the HSS pipe does not contribute to the flexural capacity of the connection. The footing
and cap are designed as traditional members with no contribution of strength from the reinforcing
interrupted by the embedded HSS pipe.
As mentioned, following WSDOT (2019) the pier HSS pipe is first sized. As the design relies on a variety of
resistance factors, a factor of one is selected because the bent is to be tested to an extreme limit state.
Sizing of the HSS pipe is begins by ensuring it is not subjected to local buckling prior to developing the pipe
strength. WSDOT (2019) offers the use of the below equation for determination of member sizing as to
assure ductile behavior/buckling criteria:
𝐷𝐷 𝐸𝐸
≤ 0.15
Where:

The selected pipe is determined through a combination of strength capacity, column rebar spacing, and
other requirements. The selected HSS section is determined as an HSS6x0.500, with properties of 42 ksi
yield strength, an ultimate strength (Fu) of 58 ksi and modulus of 29,000 ksi. The actual design pipe
thickness is 0.465 in. For this particular HSS pipe D/t is equivalent to 12.9 and 0.15E/Fy is equivalent to
103.6.
After confirmation that the selected HSS meets the buckling criteria, the moment capacity of the
connection must be determined. The equation provided is used:
𝑀𝑀𝑛𝑛 (𝑦𝑦) = �𝑐𝑐�𝑟𝑟𝑖𝑖2 − 𝑦𝑦 2 � − � ∗ 0.95𝑓𝑓𝑐𝑐′ + 4𝑐𝑐𝑐𝑐 𝐹𝐹𝑦𝑦
3 𝑟𝑟𝑖𝑖
Where:

𝑐𝑐 = 𝑟𝑟𝑖𝑖 cos 𝜃𝜃
Where:

𝑦𝑦
𝜃𝜃 = sin−1 � �
𝑟𝑟𝑚𝑚
Where:
As the neutral axis is expected to be equal at the centroid the variable, y is taken to equal zero. Once y is
determined the variables 𝜃𝜃 and c are calculated as 00 and 2.54 in., respectively. The final resulting moment
capacity of the connection is 56.7-kip-ft. The testing arrangement is identical to the CIP bent which has a
loading height off the top of the footing of 83.75 in. to the center of the actuator, the resulting base shear
is taken as 8.1-kip; with a total design base shear of 32.5-kip when considering all four column connections
present in the precast bent system.
A further design element required for the proposed connection is the embedment length of the HSS pipe
into the footing, cap, and column. For this, two approaches from literature are considered. The two
methodologies are proposed by Edward P. Wasserman (Wasserman and Walker 1996) and WSDOT (2019).
Wasserman and Walker’s approach is based on “Design of Integral Abutments for Jointless Bridges” by
Edward P. Wasserman (1996). The following proposed equation was derived from a method used for
application to develop the plastic moment capacity of piles used in bridge abutments. The original
derivation used 3.78f’c for the concrete bearing capacity, based on research performed by Burdette,
Jones, and Fricke. The derivation used below uses a much more conservative value of 0.7f’c, as allowed
by AASHTO for concrete bearing pressure (C5.5.4.2, Pages. 5-30). The resulting proposed equation is:
𝑀𝑀
𝑙𝑙𝑒𝑒 = 2 � 1�
′
(0.7𝑓𝑓𝑐𝑐 𝑏𝑏)2
𝑀𝑀 = 𝑍𝑍 ∗ 𝐹𝐹𝑦𝑦
𝑑𝑑
𝑏𝑏 = √𝜋𝜋
2
Where:
𝑙𝑙𝑒𝑒 = Required embedment depth, in

M = Plastic pipe moment, lb-in
f’c = Compressive strength of concrete, psi
d = Outside diameter of the pipe, in
Z = Gross plastic section modulus, in.3
Fy = Yield strength of the pipe, ksi
Note that the equations above are empirical and require the parameters to be in the proper units.
The method used by the WSDOT was developed experimentally at the University of Washington by Dawn
E. Lehman and Charles W. Roeder in “Rapid Construction of Bridge Piers with Improved Seismic
Performance”, published January 2012 (Lehman and Roeder, 2012). The method was developed for use
with CFSTs for foundation connections and bridge piers. The connection uses a steel pipe with an annular
ring, as shown in Figure 91, imbedded into a pocket connection either preformed or formed with a
corrugated steel pipe and grouted in place.
Figure 91. WSDOT Pipe Embedment Cross-Section
WSDOT Proposed equation to ensure full plastic behavior of the CFST:

1
𝐷𝐷02 5.27𝐷𝐷𝐷𝐷𝐹𝐹𝑢𝑢 2 𝐷𝐷0
𝑙𝑙𝑒𝑒 ≥ � + � −
4 �𝑓𝑓𝑐𝑐′ 2
Where:

D0 = Outside diameter of the annular ring, in
Again, the equation is empirical. This equation can be reduced for the proposed connection as an
annular ring is not used. The resulting simplified equation is:
1
5.27𝐷𝐷𝐷𝐷𝐹𝐹𝑢𝑢 2
𝑙𝑙𝑒𝑒 ≥ � �
For the purpose of conservative construction for the experimental investigation of the proposed
connection both methodologies are considered with the greater resulting value used for construction.
The controlling resulting embedment length is determined to be 22 in. This results in an HSS section with
an overall length of 49 in.; 22 in. is required in the footing or cap; an additional 2.5 in. is considered for
the elastomeric bearing pad and unbonded length. This leaves 24.5 in. to extend into the column which
ensures effective bonding of the HSS pipe section. For the embedded elements within the cap and footing
the HSS pipe section is required to be a total length of 23 in. to accommodate for the full 22 in. embedment
required and an additional 1 in. for grout flow. Similar to the CIP bent the remainder of the column is
designed as a traditional CIP section. Resulting in 7 #6 Gr. 60 longitudinal reinforcing, and a #3 Gr. 60 spiral
with a pitch of 1.5 in. running the full length of the column. However, the longitudinal reinforcing is
required to terminate within the column as opposed to running continuously, as in the CIP columns. For
this, threaded terminators are used at either end of the longitudinal reinforcing within the column. The
full bent detail is provided in Figure 92.
The elastomeric bearing pad used was 0.5 in. This can be designed in accordance with the AASHTO M251
(Standard Specification for Plain and Laminated Elastomeric Bridge Bearings) for thickness, select material
for durability, and shear performance of it. They can be designed to be replaceable. We are not relying on
the elastomeric bearing pad structurally but does provide damping on smaller earthquake when it starts
rocking and dampens the motion. We are neglecting this damping and is simply referred to be a contact
damping.
Figure 92. Precast Bent Cross-Section
Further examination of Figure 92 provides details to the footing and cap sections. The footing and cap
embedded HSS8.625x0.500 are selected based on the calculated column pipe diameter and the clearance
required for grout flow in the void. This size provides 7/8 in. clearance around the column pipe when
sitting centered in the embedded insert. Beyond the pipe the footing and cap are dimensioned exactly as
the CIP bent specimen was. The footing measures 4 ft. x 4 ft. x 2.5 ft. and the cap measures 15 ft. x 2 ft. x
2.5 ft. (L x W x D). With the presence of the embedded pipe, adjustments to the top reinforcing and bottom
reinforcing of the footing and cap, respectively, had to be made. For the footing this was simply
accomplished through adequate 900 rebar hooks for any of the interrupted top mat reinforcing with the
bars falling outside of the embedded HSS pipe constructed as normal.
The cap design required a unique approach as the experiment aimed to ensure failure occurs in the
column. With the requirement that none of the reinforcing interrupted by the embedded HSS be
considered to constitute to the cap’s moment capacity, the cap is considered to be comprised of two
individual concrete beams on either side of the HSS pipe. This results in design of a beam having a cross-
section of 7.7 in. x 30 in. to conservatively handle one half of the targeted moment capacity of 500-kip-ft.
As can be seen in the cap cross-section in Figure 93, the resulting beam is comprised of three layers of
two #6 Gr. 60 rebar on either side of the HSS pipe with a resulting moment capacity of 608-kip-ft.
Additionally, the top reinforcing is similar to the CIP bent cap with 9 #6 Gr. 60 rebar provided. The cap is
designed with #4 Gr. 60 stirrups provided 4 in. center-to-center spacing. As in the CIP bent specimen and
in compliance with ITD’s general practice, two alternating #4 Gr. 60 crossties are provided for each stirrup.
Also seen in the cap cross section is the grout polyvinyl chloride (PVC) pipe inlet and air vent at the top of
the HSS pipe.
Figure 93. Precast Bent Cap Detail
The final resulting precast bent, having the same overall dimensions as the CIP bent, is shown in Figure
94. The figure shows the full detail of bent, with the implementation of the proposed connection.
Figure 94. Precast Bent Specimen
Construction
The construction of the precast bent is done completely outside the lab with each element, footing,
columns, and cap, poured and moved into the lab for assembly as would be performed on a true bridge
project. The cages for each respective element are constructed and placed in the form work reused from
the CIP bent. The reuse of the form work for the footing is simply done as the footing forms are not
required to be altered or reworked. The HSS pipe is first sand blasted, for improved bond with both the
concrete and grout, then fabricated with suitable rebar risers in order to be secured and sit at the proper
height for pouring. The grout inlet is also secured in place prior to placement. Finally, a wood block is fixed
to the bottom pipe end and sealed with caulk to ensure concrete does not rise into the embedded pipe.
The rebar risers, grout inlet and wood block are all shown in Figure 95 after fabrication and installation.
The welds in Figure 95 are fillet welds which is only intended for the ease of construction or alignment
and does not do anything structurally. Figure 96 shows the completed footing elements after the full
installation of the embedded HSS pipe, grout inlet, rebar cage, and anchoring sleeves.
Figure 95. Embedded HSS Pipe Prepared for Installation
Figure 96. Precast Bent Footing Elements
The pier construction is carried out independently of the footing construction as opposed to the
traditional CIP method used for the benchmark specimen. The rebar cage and HSS pipe sections are
fabricated and placed in the forms for pouring. The columns for the precast bent are poured horizontally
as opposed to vertically for the CIP bent. This greatly reduces the labor involved and increases safety as
all work can be completed on the ground as opposed to lifting concrete to the finished pier height. The
forms from the CIP bent construction are simply reduced to three sides because the piers are poured on
a casting bed outside the lab (Figure 97).
Figure 97. Precast Bent Column Prepared for Concrete
Following the completion of the footing and columns, the cap construction began. Due to available space
on the casting bed the items had to be cast in separate pours. Whereas a true precast operation can likely
handle producing the items at one time or as is necessary for the project. The cap is poured similar to the
CIP bent cap but is constructed and poured on a casting bed making the full process much simpler and
safer. The forms from the CIP bent are again used and are placed after the rebar cage is completed. The
HSS pipe embedment’s are placed and sealed to the precast bed. The rebar cage is then lowered into
place (Figure 98) with the forms and grout ducts placed last. Rebar lifting hooks are also installed to assist
with handling and erecting of the cap. The construction of each element is followed by pouring of the
concrete, form removal, and relocating of the elements.
Figure 98. Cap Rebar Cage Placement
The footings are placed and anchored to the floor followed by the erection of the columns. The columns
are firstly fitted with alignment fins to ensure proper centering of the columns within the HSS pipe
embedded in both the footing and the cap. The alignment fins also serve to ensure a minimum gap is
maintained on all sides of the column pipe to ensure grout application is possible. The columns are then
lowered into place (Figure 99). After placement, the grout air vents located in the base of columns are
ensured to not be blocked by the elastomeric pad and allow air flow for grout application.
Figure 99. Precast Bent Column Erection
The cap is then prepared for placement. Due to limitation in the lab the 10-kip forklift had to be fitted
with an extension frame to place the cap. The extension frame allowed for the cap to reach the necessary
height to pass over the column HSS pipe extending from the column tops and be lowered down (Figure
100). After the placement of the precast cap the connections are grouted using SikaGrout 328. SikaGrout
328 is determined as a suitable material due to its ability to be highly flowable, non-shrink, non-metallic
content, and it has an extended working time. As the grout is applied through a gravity feed method the
extended working time ensured the full void is filled. After sufficient flow is achieved through the grout
vents the specimen is allowed to cure and is prepared for testing.
Practical limits on full size cap beams weight, lifting and shipping can be determined for individual projects
by the precasters. If the weight exceeds what’s allowed on the highways, the cap might have to be made
hollow or partial-solid. If there are limitations in the size, it can be made in sections. The precast pier
system as in Figure 100 is proposed for short to medium span bridges. The size of the cap beam and the
weight of it should not exceed at least for short span bridges. A shallower cap can be used only if further
research is available on the development length of CFST. The joints for the cap beam, at least the shear
reinforcing can be design using strut and tie method.
Figure 100. Precast Bent Cap Erection
Testing Arrangement
The test arrangement, instrumentation, and loading protocol for the precast bent is setup and done the
same as that of the benchmark CIP bent. The one exception in the instrumentation is the strain gages
present on the longitudinal reinforcing in the column of the CIP bent. As the longitudinal reinforcing in
the precast columns does not pass through the interface of the connection, the HSS pipe at each
connection is instrumented with a total of 16 strain gages. The placement of the strain gages can be seen
in Figure 38. Refer to the test arrangement, instrumentation, and loading protocol sections of CIP Bent
for the full details.
Testing Results
The test started with a first cycle targeted displacement of +/- 0.18 in. achieving a maximum of 0.15 in.
No visible cracking developed. The second cycle again developed no visible cracking. A slight gap opening
at the elastomeric pad is observable during the second cycle. On the third cycle hairline cracking appeared
within 18 in. of the footing and cap interfaces on the column face (Figure 101a). The fourth and fifth cycles
resulted in additional hairline cracking expanding up and down the full length of the columns and the
development of vertical hairline cracks near the ends of the columns (Figure 101b). Additionally, slight
spalling developed at the top connections resulting in quarter sized concrete pieces. During the sixth cycle,
more top connection spalling developed, and crack development continued with maximum cracks of
0.0157 in. (0.4 mm) (Figure 102c). The gap opening at the elastomeric bearing on the tension side was
widening to approximately 3/16th in. gap (Figure 102d). The seventh, eight, and ninth cycles continued
crack development with spalling at the base connections resulting during the ninth cycle. Overall crack
development is significantly less than that of the CIP bent. Cracking frequency is reduced with few large
cracks developing after initial hairline crack development resulting in large slab like spalling. It is during
the ninth cycle the maximum lateral force of 71.4-kip is achieved. Figure 102e demonstrates the significant
slab spalling developed during the ninth cycle. The spalling resulted in an approximately 3-kip drop in
force and exposure of spiral reinforcing. The tenth cycle resulted in minimal additional spalling and
continued crack development. The eleventh cycle resulted in continued spalling of slabs, with significant
increase spalling at the column top connections (Figure 102f). At this point the north pier away from the
lateral actuator experienced the majority of spalling. During the twelfth cycle the south pier developed
significant spalling at the bottom connection (Figure 102g). The fourteenth and fifteenth cycles resulted
in continued spalling and crack development with the gap opening at the elastomeric bearing having
increased to approximately ½ in. (Figure 102h). At this point both piers had developed spalling at both,
top and bottom, connections resulting in exposed spiral and longitudinal reinforcing. Additional cracking
had stopped developing with existing cracking continuing to widen. The HSS pipe had yet to be exposed.
Figure 101. Precast Bent Testing: a) Cycle 3: Hairline Cracking, b) Cycle 5: Vertical Hairline Cracking
c) d)
e) f)
g) h)
Figure 102. Precast Bent Testing: c) Cycle 6: Spalling Development, d) Cycle 6: Gap Opening 3/16-inch,
e) Cycle 9: Slab Spalling Bottom, f) Cycle 11: Slab Spalling Top, g) Cycle 12: Slab Spalling North, h) Cycle
14: Gap Opening ½ inch
Spalling continued to develop at all connections from the sixteenth to eighteenth cycles. Exposure of the
pipe resulted on the eighteenth cycle (Figure 103a). During the nineteenth cycle development of a vertical
crack in the column face perpendicular to the loading actuator developed significantly (Figure 103b). The
vertical crack continued to develop over the following cycles indicating the spiral reinforcing losing
confinement allowing the column concrete to begin separating from the HSS pipe (Figure 103c). During
the twenty-fourth cycle the maximum force achieved dropped to 55-kip, achieving the targeted 20%
strength degradation. Figure 103d provides an image of the north column bottom connection at failure.
Figure 103. Precast Bent Failure: a) Cycle 18: HSS Pipe Exposure, b) Cycle 19: Perpendicular Face
Cracking, c) Cycle 24: Confinement Failure, d) Cycle 24: Failure
The HSS pipe had not been exposed enough to evaluate it at the time the test was terminated. Additional
concrete was removed manually to better observe the HSS pipe. Exposed HSS pipe revealed that the top
connections resulted in a higher level of deformation. Figure 104 provides images of the observed
deformation. A significant bulge of the HSS section developed below the cap for both top connections. An
additional observation is the separation of the HSS pipe and grout. The separation is the result of the HSS
pipe fracturing just inside the interface of the cap. The HSS pipe underwent “necking” while in tension
causing it to separate and fracture just inside the interface.
Figure 104. Precast Bent Top Connection Post Test: a) Bulging HSS Pipe, b) HSS and Grout Separation,
c) HSS Bulging, d) HSS Bulging and Separation
The bottom pier connection exhibited far less damage and deformation. Slight bulging was present but
minimal in comparison to the top connections. Figure 105a provides a view of the minimal bulging that
resulted. The bottom connection did not exhibit any signs of the HSS necking. No separation of the grout
and HSS pipe was observed.
Figure 105. Precast Bent Bottom Connection Post Test: a) South Column, b) North Column
During the testing, the bent performed a total of 24 cycles resulting in a maximum actual peak
displacement of 7.6 in. A side view of the bent is provided in Figure 106 showing the bent at the maximum
pushed state during the 24th cycle. The targeted displacement during the 24th cycle is 8.4 in., but similar
to the reaction frame flex during the CIP bent test the frame was not acting completely rigid. Table 15
provides the targeted displacement and drift ratios and the actual values achieved during the testing of
the precast bent.
Figure 106. Cycle 24 Peak State
Table 15. Precast Bent Loading Protocol Summary
1 0.18 0.21 0.15 0.18
2 0.35 0.42 0.27 0.32
3 0.7 0.84 0.53 0.63
4 1.06 1.27 0.78 0.93
5 1.41 1.68 1 1.19
6 1.76 2.10 1.28 1.53
7 2.11 2.52 1.55 1.85
8 2.46 2.94 1.92 2.29
9 2.81 3.36 2.25 2.69
10 3.17 3.79 2.61 3.12
11 3.52 4.20 3.06 3.65
12 3.87 4.62 3.38 4.04
13 4.22 5.04 3.7 4.42
14 4.57 5.46 4.03 4.81
15 4.92 5.87 4.37 5.22
16 5.27 6.29 5.04 6.02
17 5.62 6.71 5.38 6.42
18 5.97 7.13 5.71 6.82
19 6.32 7.55 6.05 7.22
20 6.67 7.96 6.39 7.63
21 7.02 8.38 6.7 8.00
22 7.37 8.8 7.04 8.41
23 7.72 9.22 7.4 8.84
24 8.07 9.64 7.66 9.15
The data captured during the precast bent test is presented in this section in a similar layout as that for
the CIP bent specimen. The maximum values achieved during the testing of displacement and lateral load
were 7.66 in. and 71.4-kip, respectively. The peak lateral load correlates to a total moment capacity of
498-kip-ft. The precast bent demonstrates a stable response through the test progression resulting in
significant energy dissipation. The degradation of strength is fairly stable and consistent through the end
of the cycles. As assumed for the CIP bent, the four connections shared the lateral load equally, this
equates to base shear at each connection of 17.9-kip. Figure 107 and Figure 108 provide the resulting
precast bent Force vs. Displacement and Force vs. Drift hysteresis. The Force-Displacement hysteresis
suggest the precast bent achieved design base shear of 32.5-kip, correlating to a yield displacement of
0.95 in. Similarly, from the Force-Drift hysteresis it is seen the bent yielded at a drift ration of 1.13%. Figure
109 provides the Force-Drift backbone curve developed during testing, highlighting the peak force and
displacements achieved throughout the test. The curve exhibits a stable consistent degradation of the
precast bent strength as the test progressed.
It should be noted for the figures the positive vertical axis shows the specimen in push. The test began by
first pulling the specimen and continued to begin all cycles in pull. It can be observed the bent exhibited
higher strength during pulling as opposed to pushing, which can be attributed to two factors. The bent
underwent softening during the first pull of the cycle thus exhibiting higher strengths in all cycles.
Additionally, the reaction frame exhibited slightly higher stiffness during the pulling stage as opposed to
experiencing higher displacement during the pushing stage.
Figure 107. Precast Bent Force-Displacement Hysteresis
Figure 108. Precast Bent Force-Drift Hysteresis
Base Shear – 32.5-kip
Figure 109. Precast Bent Force-Drift Backbone Curve
Analysis of the potentiometers, specifically located in the plastic hinge regions (groups A, B, C, and D),
again provide a close breakdown of each connection’s reaction through moment-curvature. The
progression of the yield, captured by the instrumentation at each plastic hinge, can be observed in Figure
110, which provides the moment-curvature of each. Observation of four plots shows great stability and
consistency across all four connections. Each connection demonstrated a significant amount of energy
dissipation. All four moment-curvature hysteresis provide easy identification of the 3-kip drop in force
which took place during the ninth cycle. This demonstrates good correlation between the observational
data and instrumental data collected during the test. Further observation of the plots shows a consistency
in the strength degradation once the precast bent reached ultimate lateral capacity.
Similarly, the final figure provided from the instrumental data is the Dissipated Energy (Figure 114). Similar
to the approach for the CIP bent the energy dissipated at each cycle is determined and the cumulative
dissipated energy resulting is 2,125 kJ (1,567,320 ft-lb). This is significantly larger than that of the CIP bent
due to the ability of the precast bent to perform 23 full cycles as opposed to the 16 cycles performed by
the CIP bent.
Figure 110. South Column: Top (D1-S – D4-S)
Figure 111. South Column: Bottom (A1-S - A4-S)
Figure 112. North Column: Top (D1-N – D4-N)
Figure 113. North Column: Bottom (A1-N – A4-N)
Figure 114. Precast Bent Dissipated Energy Per Cycle and Cumulative (1 kJ = 737.56 ft-lb)
Similar to that of the CIP bent the backbone curve is analyzed using displacement in place of curvature for
a bilinear approximation providing the following results. The global yield moment capacity produced from
the experimental results is used to obtain the base shear yield of 62-kip corresponding to a yield
displacement of 1.246 in. The ultimate base shear provided from the backbone curve gives a total base
shear of 70.8-kip at a displacement of 2.53 in. The bilinear approximation is shown in Figure 115. The
overstrength factor is calculated in the same fashion as the CIP bent and results in a factor of 2.18 for the
precast bent. With accompanying displacement ductility’s (μ) of 2.03 at ultimate base shear and 6.02 at
failure point.
Figure 115. Bilinear Approximation for Precast Bent
The residual drift of the precast bent is presented in Figure 116, providing the permanent deformation of
the pier after the completion of each cycle. At the point of failure or test termination, the precast bent
was maintaining 74.8% (6.84% drift ratio) of the drift applied, 9.14%.
Figure 116. Residual Drift of Precast Bent
Summary
The precast bent designed, constructed, and tested with the proposed connection is intended to emulate
a CIP bent under similar conditions. With the proposed connection allowing for the adaptation of ABC in
regions of seismicity. Using previously developed design equations for CFSTs from WSDOT Bridge Design
Manual and the AASHTO LRFD Bridge Design Specifications, the proposed connection using embedded
HSS pipe sections is incorporated into a similarly sized concrete bent as the previously tested benchmark
CIP bent, to be tested in SLAB at ISU. Incorporating specific design details in the proposed connection
assisted in the connections ability to emulate a CIP connection. The use of an elastomeric bearing pad at
the base and top of the piers allows for the piers to have available flexure at small drifts resulting in
significantly reduced cracking and spalling and assisting in the ability to develop similar drift capacities as
a CIP connection. The embedment of the pipe is determined to ensure full plastic capacity is developed
prior to pullout. As the HSS pipe is accountable for providing the full flexural and shear resistance for the
column, an unbonded length is provided to allow yielding of the section to take place over a larger area
of the HSS pipe. The cap and footings of the bent are designed to be capacity protected members. Each
had an embedded HSS pipe section providing a socket for which the column HSS pipe is fitted into using
alignment fins for ensuring proper placement and that the gap is maintained for grouting.
The proposed connection proved to provide a safer pour for all elements of the bent as it could be
performed completely at ground level. The full precast bent was able to be constructed prior to any
grouting with minimal requirement for bracing. The construction did require for additional equipment
capable of handling larger elements during assembly. The grouting process was carried out using a highly
flowable grout mix that did well to fill the void between the two HSS pipe sections.
The precast bent performed well during quasi-static cyclic loading, as it performed a total of 24 complete
testing cycles achieving 20% strength degradation at a drift ratio of 9.15%. The precast bent achieved an
ultimate lateral capacity of 71.4-kip, correlating to a moment capacity of 498-kip-ft. Overall, the precast
bent demonstrated a stable consistent response through the full progression of testing. The precast bent
resulted in less cracking than the CIP bent, but with elevated spalling. The precast bent performance far
exceeded expectations and will compare favorably to the CIP bent.
Bent Pier System: Precast vs. Cast-In-Place

The two systems performed as expected and designed, with each exhibiting similar capacity having the
same overall dimensions and both being subjected to similar loading. The CIP bent developed significant
cracking and failed due to longitudinal reinforcing fracture after a total of 15 cycles. The precast bent using
the proposed connection exhibited reduced cracking, in comparison, with increased spalling and failed
after 24 total cycles due to the HSS pipe deforming and ultimately tearing. The precast pier demonstrated
a reduced stiffness, 38.7 kip/in., approaching ultimate capacity in comparison to the 56.7 kip/in. exhibited
by the CIP bent. This reduced stiffness is due to the presence of the elastomeric bearing. The precast bent
also achieved a higher ultimate capacity of 71.4-kip than that of the 66-kip achieved by the CIP bent. The
two systems exhibited a great difference in the yield displacements. Resulting in initial yield displacement
drift ratios of 0.5% and 1.13%, for the CIP and precast bent, respectively. However, in comparing global
yield the two systems exhibited differing values. The CIP bent had a global yield of displacement 0.596 in.
and a correlating moment of 392-kip-ft, whereas the precast bent exhibited a global yield displacement
of 1.246 in. and correlating moment of 433-kip-ft. The two systems also exhibited differences in
cumulative energy dissipation. The reduced stiffness of the precast bent caused it to have lower energy
dissipation at lower cycles; it slowly cumulated until the stability and added confinement of the HSS pipe
allowed it to produce a more stable consistent response to the continued cycles. At the point of failure,
the CIP bent demonstrated a cumulative dissipated energy of 342 kJ (252,246 ft-lb), whereas the precast
bent at similar cycles (15 total cycles) demonstrated a cumulative dissipated energy of 466 kJ (343,704 ft-
lb). As the precast bent was capable of withstanding a total of 24 complete cycles the ultimate cumulative
dissipated energy of the systems was 2,125 kJ (1,567,320 ft-lb). Additionally, the precast bent exhibited a
slightly greater overstrength factor, 2.18, verse 1.76 for the CIP bent. However, the precast bent exhibited
reduced displacement ductility than that of the CIP bent. The precast bent having at minimum matched
or exceeded the CIP bent in all categories, except low drift stiffness and displacement ductility due to the
elastomeric bearing presence, the proposed connection used is said to have successfully emulated CIP
construction under seismic loading. Table 2 provides a summarized comparison of the CIP bent and
precast bent.
4. Analytical Investigation
Cantilever Column
This section presents development of appropriate distributed plasticity models for analytical prediction
of the two cantilever specimens. Analytical models for both cast-in-place and precast piers are developed
using the OpenSees software. Results are compared against experimental data. The analytical prediction
in this report is limited in scope and covers only validation of analytical models for the tested specimens.
It should be noted that discussion on performance-based design and displacement based seismic design
are beyond the scope of project. In this project a force-based seismic design philosophy has been primarily
used.
Cast-In-Place Pier
The analytical models were developed using the OpenSees Finite Element software. Half-scale columns
were modeled with nonlinear materials for unconfined concrete, confined concrete, and steel reinforcing
bars. Uniaxial fiber-sections were used in the models. Bond-slip spring elements were used to represent
the bar slip relative to the concrete in the footing. This slip causes rotation at the column footing interface,
resulting in additional displacement (or drift) in the column. The CIP Pier specimen is shown in Figure 117.
Figure 117. Details of CIP Pier Specimen
In order to duplicate the same results as the experimental study, OpenSees’ “Concrete01”, “Concreter04”,
and “Steel02” nonlinear material models were used for unconfined concrete, confined concrete, and
longitudinal steel bars, respectively. The computer model uses distributed plasticity frame elements with
uniaxial fiber sections.
The model shown in Figure 118 has six nodes and five elements. Two elements are nonlinear, and three
elements are linear elastic. The nonlinear element E1 is an OpenSees’ “zeroLength” element that
represents the bond-slip at the base of the column. The other nonlinear element (element E2) is an
OpensSees’ “nonlinearBeamColumn” element that uses fiber sections with the three materials noted
above. E3, E4, and E5 are OpenSees’ “elasticBeamColumn” elements with high stiffness values.
The rotation caused by the bond-slip at the base of the column is modeled using a rotational spring with
a “Hysteretic” material behavior. The rotational spring uses a bi-linear curve describing the moment-
rotation relationship. The idealized (fitted) bi-linear curve is shown in Figure 119. This process employs
the method developed by Wehbe, et al. (1999). The tensile stress and strain in the extreme longitudinal
bar and the location of the neutral axis are determined using a moment-curvature analysis (Haber et al.
2013; Ebrahimpour et al. 2016).
Figure 118. OpenSees Model of the Experimental CIP Column
Figure 119. Schematic of the Fitted Bi-Linear Curve for the Footing Bond-Slip Moment Rotation
One of the primary failure mechanisms of reinforced concrete columns subjected to a strong seismic event
is low-cycle fatigue fracture of longitudinal reinforcing bars. This assumes that adequate confinement
exists to prevent substantial core damage, and longitudinal bar buckling is prevented. Experimental
studies on ASTM A706 and A615 bars have indicated that as little as seven full cycles of 0.06 strain can
result in low-cyclic fatigue fracture of bars (Hawileh et al. 2010). In order to introduce low-cycle fatigue in
OpenSees, the appropriate parameters for the Coffin-Manson curve needed to be introduced.
Brown & Kunnath’s (2000) strain values for the number of half-cycles to failure are found to be
considerably lower than Hawileh’s (2009). Therefore, the number of cycles to failure for ASTM A615 steel
were estimated by combining both data from Hawileh and Brown & Kunnath. The experimental data from
Hawileh and Brown & Kunnath are shown in Figure 120. The mathematical relation for the combined
strain versus number of half-cycles to failure for ASTM A615 plot is shown in Eq. below (Maskey 2017).
−0.522
0.2468�2𝑁𝑁𝑓𝑓 �
The above equation is for strain versus number of half-cycles to failure. However, OpenSees uses efficients
that are for strain versus the number of full cycles to failure. When 2Nf is replaced with Nf (i.e., the format
required by OpenSees), the coefficient of 0.2468 changes to 0.172, but the exponent of -0.522 remains
the same. For the pipe section in the precast model (see next section), the coefficients are 0.198 and -
0.468.
Figure 120. Combined Strain Vs. Number of Half Cycles to Failure Plot (Based on Brown & Kunnath
2000 and Hawileh, et al. 2009)
For the model of the cast-in-place column, an axial compressive force of 50 kips is applied at Node 5 shown
in Figure 118. Node 6 is subjected to a horizontally applied displacement-controlled increasing cyclic load
that was used in the experimental work. The lateral load was applied to the column model in a slow cyclic
fashion. For each drift level, two full push and pull cycles were planned. Figure 121 shows the
experimentally measured and calculated hysteretic force-displacement curves for the cast-in-place
column.
Figure 121. Numerical and Experimental Hysteresis Force-Displacement for CIP Column
The result validates the assumptions made for CIP column because the hysteresis loop for both
experimental and numerical data are on top of each other.
Precast Pier
Detailed sections of the analytical model for the precast column are shown in Figure 122 and Figure 123.
The Zone 1 section of Figure 123 (a) is used for top of the column where there is rebar and concrete. For
the middle section of the column where rebar, concrete as well as steel pipe are present, the section
shown in Figure 123 (b) is used. At the bottom of the column where there is no rebar and only the pipe is
present, the section shown in Figure 123 (c) is used. See Figure 122 for the longitudinal locations of the
Zones 1 to 3 of the precast column.
Figure 122. Details of the Precast Pier Specimen
Figure 123. Detailed Sections of the Model for Precast Column: a) Zone 1, b) Zone 2, and c) Zone 3
Figure 124 shows the comparison of the numerical and experimental hysteretic force-displacement
results for the precast column. As can be seen, the OpenSees model results match the experimental
results until the last cycle in which the pipe buckles right before the complete collapse of the column.
Figure 124. Numerical and Experimental Hysteresis Force-Drift for Precast Column
The result validates the assumptions made for precast column because the hysteresis loop for both
experimental and numerical data are on top of each other.
Bent Pier System
Similar to cantilever pier system, analytical models were developed for the cast-in-place and precast bent
piers using the OpenSees software.
Cast-In-Place Bent
The analytical models were developed using the OpenSees finite element software. Half-scale bents were
modeled with nonlinear materials for unconfined concrete, confined concrete, and steel reinforcing bars.
The dimensions of the bent pier are discussed in the experimental portion of the project in the previous
chapter. All the assumptions and modeling procedure are kept identical to the previously discussed
cantilever pier system. For the cap beam, the assumption made was that it behaved as an elastic member.
Figure 125 shows the experimentally measured and calculated hysteretic force-displacement curves for
the cast-in-place bent pier. As it can be observed, the analytical model is in good agreement with the
experimental data.
80
60
40
20
Force (kip)
-20
-40
-60
-80
-5 -4 -3 -2 -1 0 1 2 3 4 5
Horizontal Displacement (in.)
Figure 125. Numerical and Experimental Hysteresis Force-Displacement for CIP Bent Pier
Precast Bent
Figure 126 shows the experimentally measured and calculated hysteretic force-displacement curves for
the precast bent. The analytical force-displacement for the precast bent is in good agreement with the
experimental data up to a drift ratio of 6%. During cycles of large drift ratio, the analytical model does not
follow the experimental data to a high accuracy. This is thought to be due to excess shear deformation
and Bauschinger effects in the HSS which are not easy to capture analytically.
80
60
40
20
Force (kip)
-20
-40
-60
-80
-10 -8 -6 -4 -2 0 2 4 6 8 10
Horizontal Displacement (in.)
Figure 126. Numerical and Experimental Hysteresis Force-Displacement for Precast Bent Pier
5. Parametric Case Studies
This chapter provides some limited parametric case studies for modeling a typical highway bridge (i.e.,
SH-22 over I-15 at Dubois) in Idaho using OpenSees computer program. Models use both cast-in-place
(CIP) and emulative CIP precast piers. The goal is to provide some guidance for the engineer that the
proposed ITD connection can be modeled and analyzed very similar to CIP. The size of HSS was selected
in order to closely match the capacity of the CIP bridge. For a simplified analysis, the unbonded length of
the HSS was not considered in the analysis. Materials properties for the bridge were mostly kept similar
for a better comparison. Nonlinear static (pushover) and nonlinear dynamic (time-history analysis) are
conducted on the prototype bridge. The global seismic response of the bridge such as formation of plastic
hinges, ductility, strength, base shear demand, force-displacement response has been summarized for
each CIP and precast model and are compared against each other.
Description of the Bridge

The bridge on SH-22 over I-15 at Dubois is a two-span bridge with a four-column bent. The superstructure
is made up of an 8-in. thick concrete deck that rests on eight steel girders. The substructure is composed
of the pier cap, four columns, and footings all cast-in-place (CIP). Figure 127 and Figure 128 show the plan
and elevation views of the bridge, respectively. This bridge also has a non-integral superstructure-pier
connection. The Columns are 3.5 ft in diameter with a height of 14.05 ft. The column section has 13 No.
11 steel reinforcing bars.
Figure 127. Plan View of the SH-22 over I-15 Bridge at Dubois (NTS)
Figure 128. Elevation View of the SH-22 over I-15 Bridge at Dubois (NTS)
Nonlinear Static Analysis

To keep the analysis simple, the single-mode spectral method was used for the bridge analysis. Since the
goal is to compare the behavior of two different types of nonlinear columns (cast-in-place columns versus
precast columns with the proposed pipe connection), a more complex analysis is not conducted.
In accordance with the LRFD Bridge Seismic Reference Manual (2014), the use of the uniform load method
or single-mode method is one of the two options in “nonlinear static analysis.” Obviously, the procedure
in the “nonlinear static analysis” is much less tedious than the “nonlinear dynamic analysis.” Nonlinear
static analysis, also known as pushover analysis, is a static procedure that uses a simplified nonlinear
technique to estimate seismic structural deformations. Structures redesign themselves during
earthquakes. As individual components of a structure yield or fail, the dynamic forces on the building are
shifted to other components. A pushover analysis simulates this phenomenon by applying loads until the
weak link in the structure is found and then revising the model to incorporate the changes in the structure
caused by the weak link.
The nonlinear models of the Dubois bridge with node and element placement are shown in Figure 129
and Figure 130. For the support stiffness in the spring support condition; the assumption made is the
springs at abutments with fixed column bases. The abutment type is the seat type. Similarly, for the
restraint of superstructure, abutments were modeled with linear springs both in longitudinal and
transverse direction along with unrestrained rotation about the z-axis and infinite restraint in all other
DOF’s.
Figure 129. Dubois Bridge Nonlinear Model with Node Numbers
Figure 130. Dubois Bridge Nonlinear Model with Element Numbers
From the analysis, a global force-displacement or drift (backbone curve) is obtained (Figure 131). The
displacement or drift (X axis) is taken at the center of the gravity of the superstructure and is plotted
against the total base shear (Y axis). The curve is extended until hinges form in the piers and collapse,
resulting in at least 20% reduction in the total base shear. The hierarchy of plastic hinge formation is
shown on the backbone curve to highlight progressive collapse. It can be observed that both bridge
models (CIP and Precast) have similar strength. However, the precast bridge achieves higher displacement
capacity compared to CIP.
Figure 131. Base Shear Vs. Displacement for both CIP and Precast Column
Nonlinear Dynamic Analysis

Nonlinear dynamic analysis is carried out to investigate the performance of the bridge with the proposed
pipe connection. The goal is to observe how the prototype bridge would perform compared to CIP when
subjected to ground motions.
Before selecting ground motion, the bridge is assumed to be located in the most seismically active location
in Idaho. The “Step-by-Step Instructions for AASHTO Guide Specifications” and the USGS web site
(https://earthquake.usgs.gov/ws/designmaps/) were used to obtain the parameters needed to construct
the current AASHTO 1000-yr return period response spectrum for the design of a conventional bridge.
For the Upper-Level Ground Motion and 7 percent probability of exceedance in 75 years; the return period
in accordance with AASHTO is about 1,000 years. Location of the bridge is assumed to be in Montpelier in
Idaho (Latitude: 42.32 degrees, Longitude: -111.3 degrees) where soil class is assumed to be D ‘Stiff Soil’.
Figure 132 shows the AASHTO recommended three-point method that was used to obtain the design
spectrum for the analysis. Similarly, Table 16 shows the data point obtained to construct the uniform
hazard spectrum and Figure 133 shows the design spectrum obtained from the data points.
Figure 132. AASHTO Seismic Coefficient Design Spectrum Constructed with the Three-Point Method
Table 16. Uniform Hazard Spectrum Data Points for Montpelier, Idaho
Period, Tm (second) Elastic Seismic Coefficient, Csm

0 0.38
0.1072 0.908
0.2 0.908
0.535 0.908
0.6 0.81
0.7 0.694285714
0.8 0.6075
0.9 0.54
1 0.486
1.1 0.441818182
1.2 0.405
1.3 0.373846154
1.4 0.347142857
1.5 0.324
1.6 0.30375
1.7 0.285882353
1.8 0.27
1.9 0.255789474
2 0.243
2.1 0.231428571
2.2 0.220909091
2.3 0.211304348
2.4 0.2025
2.5 0.1944
1.8
Arithmetic Mean Landers - USA
1.6
Duzce - Turkey Darfield - New Zealand
1.4
El Mayor - Mexico AASHTO Design Spectrum
Spectral Acceleration (g)
1.2
0.8
0.6
0.4
0.2
0
0 0.5 1 1.5 2 2.5
Period (Sec)
Figure 133. Design Spectrum
The Pacific Earthquake Engineering Research Center (PEER) website (https://ngawest2.berkeley.edu/)

was used to obtain earthquake records and scale the ground motion to the design spectrum. The search
criteria for the earthquake records were set as follows:
• Using NGA-West2 database (Western United States)
• Identifying up to 10 ground motion records for strike-slip/normal faults; magnitude 7-7.5; rupture
distance (R_rup) 0-20 km; and any average seismic shear-wave velocity (Vs30)
• Assuming a natural period of 1.5 sec for the bridge (transverse direction); the period of interest
for scaling can be taken as 0.2Tn – 1.5 Tn or 0.3 – 2.25 sec.
• Spectral ordinate of H1 (not-rotated); 5% damping ratio; arithmetic suite average; and minimize
MSE scaling method with weight factor of 1.
• Neglecting vertical excitation and rotational components of ground motions
Four earthquake records from different parts of the world were selected. These records and their scaling
factors are summarized in Table 17. The bridge model is subjected to ground displacement from each
record for both components of the ground motion (e.g., first run one component in the transverse
direction, obtain the results and then run the second component in the transverse direction). Similar
procedure for longitudinal direction is carried out.
Table 17. Earthquake Records form Different Regions
Scale Earthquake Rjb Vs30

Year Magnitude Mechanism Rrup (km)
Factor Name/Location/Station (km) (m/sec)
1.239 Landers – USA - Joshua Tree 1992 7.28 Strike-slip 11.03 11.03 379.32
0.6596 Duzce – Turkey - Bolu 1999 7.14 Strike-slip 12.02 12.04 293.57
1.4092 Darfield - New Zealand - DFHS 2010 7 Strike-slip 11.86 11.86 344.02
El Mayor-Cucapah – Mexico -
1.0718 2010 7.2 strike-slip 9.98 11.26 196.88
El Centro Array #12
Cast-In-Place Bridge Response
A typical dynamic input and response of the model is shown in Figure 134 and Figure 135. In these figures,
Direction 1 is a longitudinal direction and Direction 2 is transverse direction. Table 18 provides a summary
of the nonlinear dynamic analysis for the CIP bridge. The last column in the table shows the yield stress
which is obtained from the static pushover analysis of the bridge explained previously. AASHTO
requirement for finding maximum base shear and displacement was used to obtain the load combination
for base shear demand as follows:
Where:
A = Larger displacement or base shear between two directions (longitudinal and transverse)
B = Smaller displacement or base shear between two directions (longitudinal and transverse)
In this research, matching of horizontal ground motion components to an orientation-independent

spectrum (RotDnn) was not considered. However, this could be a good future research to investigate the
nonlinear dynamic performance of bridges incorporating such connections.
Figure 134. Displacement vs. Time for Landers Earthquake
1500
1000
Base Shear (kips)
500
Dynamic Response
0
Pushover
-500
-1000
-1500
-0.6 -0.4 -0.2 0 0.2 0.4 0.6
Relative Displacement of the Deck w.r.t. Bottom of Columns (ft)
Figure 135. Pushover vs. Dynamic Response for Landers Earthquake for the Bridge with CIP Columns in
Direction 2
Table 18. Maximum Base Shear and Displacement for CIP Bridge (Absolute Values Shown)
Max. Long. Max. Trans. Max. Base Shear Max. Max. Max. Displ. Pushover
Direction of
Base Shear Base Shear Demand, Combo Long. Trans. Demand Yield Strength
EQ
(kips) (kip) (kips) Displ. (ft) Displ. (ft) Combo (ft) (kips)
Landers
Dir1-Long, 296 780 869 0.02 0.05 0.056 940
Dir2-Trans
Landers
Dir1-Trans, 413 749 873 0.025 0.047 0.055 940
Dir2-Long
Duzce
Dir1-Long, 530 657 816 0.03 0.04 0.05 940
Dir2-Trans
Duzce
Dir1-Trans, 315 974 1067 0.021 0.098 0.1 940
Dir2-Long
Darfield
Dir1-Long, 972 966 1262 0.094 0.082 0.12 940
Dir2-Trans
Darfield
Dir1-Trans, 714 978 1192 0.051 0.11 0.05 940
Dir2-Long
El Mayor
Dir1-Long, 622 770 957 0.043 0.049 0.062 940
Dir2-Trans
El Mayor
Dir1-Trans, 648 931 1125 0.044 0.067 0.08 940
Dir2-Long
From Table 18, it can be observed that excluding the Landers record, all other earthquake records pushed
the CIP bridge beyond the yield capacity. The displacement demand from all four records are smaller. This
is due to the Idaho’s seismicity where earthquakes are not expected to cause large displacements as
compared to other locations such as California which has a higher seismic demand.
Precast Bridge Response
Similar to the CIP model, the bridge with precast pier was subjected to the four ground motion records.
Results are summarized in Table 19. As it can be observed, all four earthquake records pushed the bridge
beyond the yield capacity.
Table 19. Maximum Base Shear and Displacement for Precast Bridge (Absolute Values Shown)
Max. Long. Max. Trans. Max. Base Shear Max. Max. Max. Displ. Pushover
Direction of
Base Shear Base Shear Demand, Combo Long. Trans. Demand Yield Strength
EQ
(kips) (kip) (kips) Displ. (ft) Displ. (ft) Combo (ft) (kips)
Landers
Dir1-Long, 315 872 967 0.02 0.047 0.053 960
Dir2-Trans
Landers
Dir1-Trans, 421 761 887 0.024 0.04 0.047 960
Dir2-Long
Duzce
Dir1-Long, 544 696 860 0.029 0.037 0.05 960
Dir2-Trans
Duzce
Dir1-Trans, 340 975 1077 0.02 0.083 0.09 960
Dir2-Long
Darfield
Dir1-Long, 977 973 1269 0.088 0.08 0.11 960
Dir2-Trans
Darfield
Dir1-Trans, 788 984 1220 0.05 0.1 0.12 960
Dir2-Long
El Mayor
Dir1-Long, 661 965 1163 0.043 0.061 0.07 960
Dir2-Trans
El Mayor
Dir1-Trans, 755 952 1179 0.047 0.054 0.07 960
Dir2-Long
Cast-In-Place Vs. Precast Bridge

Results from nonlinear static and dynamic analyses showed that the bridge with the precast pier closely
emulates the cast-in-place bridge behavior. Both models had similar strength, ductility, and stiffness. Since
the location of the Dubois bridge was assumed to be in Montpelier, nonlinear dynamic analysis showed
that both cast-in-place and precast bridge models would reach their yield capacity under the four
earthquake records. This is to be expected since the design seismic accelerations for Montpelier are larger
than those of Dubois. In summary, existing nonlinear analysis techniques for cast-in-place bridges can be
used to perform a system level analysis of a bridge with the proposed ITD connection.
6. Design and Detailing Considerations
Introduction
This chapter discusses flexural analysis, interaction diagrams, design/detailing considerations,
construction/assembly technology, limitations, inspection, and potential post-earthquake repair methods
for the precast pier.
Flexural Analysis
Existing resources that can be used to design the precast pier include the AASHTO LRFD Bridge Design
Specifications (AASHTO 2017), WSDOT Bridge Design Manual (WSDOT 2019), and Integral Abutments for
Steel Bridges (Wasserman and Walker 1996). The AASHTO LRFD Bridge Design Specifications are used as
a basis for all footing and column reinforcement. The WSDOT Bridge Design Manual is used to determine
the capacity and size the HSS member used to connect the column to the footing. The WSDOT Bridge
Design Manual and Integral Abutments for Steel Bridges are used to determine the embedment length of
the HSS member.
The first step is to design the column as cast-in-place and identify the required nominal capacity. In the
next step, the size of the HSS for the precast solution is identified. The WSDOT Bridge Design Manual
provides several equations to select an appropriate HSS member. Appendix D of this report provides a
variety of different pipe sizes as a means to quickly determine if a pipe will work and to estimate the
moment capacity of an HSS pipe having a yield strength of 42 ksi, ultimate strength of 58 ksi, and
compression strength of concrete of either 4 or 8 ksi.
WSDOT (2019) provides equations to check the diameter to pipe thickness ratio and to determine the
moment capacity of a CFST. The following equation is used to ensure the pipe is not subject to local
buckling before developing the strength of the pipe:
𝐷𝐷 𝐸𝐸
≤ 0.15
Where:

According to WSDOT (2019), the pipe thickness should be reduced as a result of corrosion over a 75-year
minimum design life. The corrosion rates vary according to the environment as described in Table 20 and
assumes that the soil is not highly corrosive. The minimum corrosion reduction shall be taken as 1/16 in.
The interaction diagrams and tables in this report do not incorporate the corrosion reduction factor.
Table 20. Common Corrosion Rates from WSDOT Bridge Design Manual
Location Corrosion Rate

Soil embedded zone (undisturbed soil) 0.001 in./year
Soil embedded zone (fill or disturbed soils) 0.003 in./year
Immersed zone (freshwater) 0.002 in./year
Immersed and tidal zone (saltwater) 0.004 in./year
Splash zone (saltwater) 0.006 in./year
Atmospheric zone 0.004 in./year
In the absence of research data on displacement-based design of the bridges with such connections, the
current equation from WSDOT can be used. If future research demonstrates that an expected nominal
moment as per AASHTO Seismic Guides can be used for this connection; it would likely influence the
presented design guidelines. The equation WSDOT provides to determine the nominal moment capacity
is shown below:
𝑀𝑀𝑛𝑛 (𝑦𝑦) = �𝑐𝑐�𝑟𝑟𝑖𝑖2 − 𝑦𝑦 2 � − ′
� ∗ 0.95𝑓𝑓𝑐𝑐 + 4𝑐𝑐𝑐𝑐 𝐹𝐹𝑦𝑦
3 𝑟𝑟𝑖𝑖
Where:

𝑐𝑐 = 𝑟𝑟𝑖𝑖 cos 𝜃𝜃
Where:

𝑦𝑦
𝜃𝜃 = sin−1 � �
𝑟𝑟𝑚𝑚
Where:
Because the neutral axis is expected to be approximately equal to the centroid when the structure is
assembled, the variable y is equal to zero. If the column is not plumb when the column is assembled to
the footing, the variable y is used to reduce the column capacity.
To select the column receiving pipe, use a size that will allow a tolerance gap. The gap needs to be large
enough to allow the grout to easily flow between the two HSS members. This tolerance gap is larger
compared to grouted ducts or similar connections. It is expected that the gap can be up to 2 in. on each
side of the pipe without compromising the structural integrity of the connection. For gaps larger than 2
in. experimental and analytical testing should be performed to demonstrate the integrity of the
connection beyond the yield point. The gap used in this research was equal to 0.5 in. on each side of the
pipe.
Interaction Diagrams
The equations used to develop the interaction diagram are the same equations that are discussed in the
previous section for flexural analysis for CFSTs from the WSDOT Bridge Design Manual (WSDOT 2019). To
create the interaction diagram, a variety of different pipe thicknesses, pipe diameters, and two different
compression strengths of concrete. The pipe thicknesses used are values found in the American Institute
of Steel Construction Manual (AISC Committee 2010) and vary from 0.174 in. to 0.581 in. The Pipe
diameters range from 10 in. to 60 in. The compression strength of the concrete used is 4 and 8 ksi and
does not account for an axial load. The interaction diagram is shown in Figure 136 and displays the
variables used to develop it. Looking at Figure 136, the moment capacity increases as the pipe diameter,
pipe thickness, and compression strength of concrete increase. It should be noted that the interaction
diagrams in this report do not account for an axial load.
To determine the moment capacity of a specific HSS member, the user only needs to have predetermined
the pipe diameter, pipe wall thickness, and the compression strength of concrete. For example, if a 36 in.
diameter pipe having a thickness of 0.465 in. were to be selected to be embedded in concrete having a
compression strength of 8 ksi, the resulting moment capacity can be determined from Figure 136 to be
roughly 4500 kip-ft.
20000
t = 0.581 in.
18000 f'c = 8 ksi
16000
14000
Moment Capacity (kip-ft)
f'c = 4 ksi
12000 t = 0.465 in.
10000 Interaction Diagram

fy = 42 ksi f'c = 8 ksi
8000 Es = 29000 ksi
f'c = 8 ksi
6000 t = 0.349 in. f'c = 4 ksi
f'c = 8 ksi
4000 t = 0.291 in.
t = 0.233 in. f'c = 4 ksi
2000
t = 0.174 in.
f'c = 4 ksi
f'c = 4 ksi
0
10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64
Pipe Diameter (in.)
Figure 136. HSS Member Moment Capacity Interaction Diagram
Embedment Length
In the absence of extensive experimental data on embedment length, two methodologies are proposed
to determine the embedment of the pipe into the column and footing. The methodologies used are
proposed by WSDOT Bridge Design Manual and (Wasserman and Walker 1996). The equation used in this
research is shown below.
5.27𝐷𝐷𝐷𝐷𝐹𝐹𝑢𝑢
𝑙𝑙𝑒𝑒 ≥ �
Where:

Where, le is the calculated embedment length, D is the diameter of the embedded pipe, t is the wall
thickness of the embedded pipe, Fu is the ultimate strength of the embedded pipe, and f’c is the
compression strength of the concrete with all variables in terms of kip and inches.
For the embedment length proposed by Wasserman, the following equations are used:
2𝐹𝐹𝑦𝑦 𝑍𝑍
𝑙𝑙𝑒𝑒 =
�700𝑓𝑓𝑐𝑐′ 𝑏𝑏
𝑑𝑑√𝜋𝜋
𝑏𝑏 =
2
Where:
Fy = Yield strength of the embedded pipe, ksi

Z = Plastic section modulus of the embedded pipe, in3
To select the appropriate embedment length, simply follow the flow chart shown in Figure 137. A
summary of embedment lengths for commercially available HSS and pipes are shown in Figure 138 and
Figure 139, respectively with f’c is 4 ksi, Fy is 46 ksi, and Fu is 62 ksi for both graphs.
Figure 137. Embedment Length Determination Flow Chart
35
30
25
20
15
10
Wasserman (1996) WSDOT (2018)
Figure 138. Embedment Length Comparison for HSS Round (in.)
40
35
30
25
20
15
10
Series1 Series2
Figure 139. Embedment Length Comparison for Pipe (in.)
Unbonded Length
The unbonded length in the plastic hinge is a way to force the HSS to yield at a designated location (above
the footing) that can be inspected following an earthquake. The unbonded length of the HSS member is
determined using the methodology presented in the PRESSS Design Handbook (Pampanin et al. 2010).
The unbonded length can be selected such that the strain in the HSS member would be approximately
equal to a cast-in-place column or say 6% at the design level. To begin, the strain penetration needs to be
calculated using the following equation:
Where:
Fye = 1.1 times the yield strength, ksi

dbl = Pipe thickness, in
Using a value of 60 ksi (413.7 MPa) for Fy and 0.75 in. (19.05 mm) for dbl, the strain penetration is
calculated to be 7.43 in. (188.7 mm). Strain in the steel is then calculated using the following equation:
∆
𝜀𝜀𝑠𝑠 =
𝑙𝑙𝑢𝑢𝑢𝑢 + 2𝑙𝑙𝑠𝑠𝑠𝑠
Where:
Δ = Elongation of the rebars, in

lub = Unbonded length, in
It is important to estimate the elongation of the HSS member to determine an appropriate unbonded
length. In this report, data from the cast-in-place column was used to predict the elongation of the HSS
member.
Experimental results from testing of the precast piers showed that an unbonded length equal to 30-35%
diameter of the HSS/pipe improves the ultimate displacement capacity of the pier. Further experimental
and analytical investigations are needed to quantify the unbonded length for various commercially
available HSS and pipe sections.
Construction Technology and Assembly

It is recommended that the materials used in this report be a minimum for practical applications. When
constructing the footing, care should be taken to ensure the inside of the receiving HSS member remains
clear of concrete. If concrete propagates into the receiving HSS member, the undesired concrete will need
to be removed with a grinder before the assembly of the precast column. To keep concrete out of the
footing HSS member, a ¾ in. piece of plywood is sized and secured to the inside bottom edge of the HSS
member and gorilla tape is placed at the top of the HSS member (Figure 140 a-c). While the footing is
curing, water should not be allowed to pool at the bottom of the receiving HSS member. Any rust that
accumulates on the inside of the receiving HSS member should be removed before assembling the precast
column and the grouting process.
The precast column is poured horizontally for ease of construction, the restrictions over dropping height
for concrete, and to make sure it is not knocked over accidentally while the concrete is being poured or
while the concrete is curing. The outside of the HSS pipe is sandblasted before pouring concrete to achieve
a better bond between the HSS member to the concrete. The unbonded length of the pipe is accomplished
by wrapping the desired unbonded pipe length with gorilla tape. The grout vents used are PVC pipes. The
PVC pipes are sealed using gorilla tape while the concrete is being poured to ensure concrete does not
enter and clog them. A level should be used to ensure the HSS member is oriented correctly in the
formwork (Figure 140 d). If the HSS member is not level before the concrete is being poured, then the
column will not be plumb when the precast column is assembled to the cast-in-place footing. Once the
concrete is poured, the column should be allowed to cure sufficiently so premature cracking does not
occur. The centering fins should be welded onto the protruding HSS member after it is removed from the
formwork (Figure 140 e). Refer to Chapter 4 of this report for more construction images.
a) b) c)
d)
e)
Figure 140. a) Receiving HSS Member, b) Footing Reinforcement, c) Finished Footing, d) Column
Reinforcement, and e) Welded Centering Fins
For assembly, the precast column needs to be rotated from the horizontal orientation to a vertical
orientation. To rotate the column to a vertical orientation, care should be taken to ensure the HSS
member that is protruding from the precast column is not damaged. A crane should be used at each end
of the column, without coming in contact with the HSS member, until the column is vertical. The HSS
member should not come in contact with the ground at any time to prevent any dents or scratches that
could compromise the structural performance and long-term durability. Once the column is in a vertical
position the precast column’s protruding HSS member should be promptly inserted into the cast-in-place
footing receiving HSS pipe and grouted into place.
Non-shrinkage grouting should be used to fill the void between the two HSS members that has a higher
compressive strength compared to the concrete used in the column and footing. The grout can be either
pumped or gravity feed into the footing grout inlet pipe. Once all the voids between the two HSS members
are filled, grout would flow out of the grout vents located on each side of the column. Allow the grout to
flow unobscured out of each grout vent for approximately 30 seconds to ensure all air pockets are
removed that may have been trapped. Once the grout has been sufficiently cured, the connection is
complete.
Limitations
The limitations of the precast column connection are nearly the same as a traditional cast-in-place column.
One of the limitations to the precast column that varies from a cast-in-place column is the column
capacity. The capacity of the precast column is controlled by the compressive strength of concrete used,
the size of the selected HSS member, and the material properties of the HSS member. Vertical pour of the
precast column may not be possible due to concrete drop height restrictions as well as bracings for the
formwork. Another limitation is the ability to lift and rotate the column to a vertical position. Equipment
capable of lifting the column is required to transport the column to its designated location. As the column
height and diameter increase, the total weight of the column also increases, and as a result, larger
equipment may be required. For precasting, it is recommended to stay with the available formwork sizes
or standardizing the sections so multiple precast columns can be produced using the same formwork. This
will help in reducing the cost of the precast columns.
Transportation limitations are likely to govern by the size and height of the precast cap beam. Options for
the precast cap range from hollow to partially solid or fully solid. Precasting of the piers would likely
require horizontal pour in the factory. Assembly of the pier on-site and any concrete pour (e.g., filling the
cap) would be similar to existing precast and cast-in-place practices. Footings will be cast-in-place, welding
of rebars to the receiving pipe can be avoided by providing 90-degree bends at the end of the rebars near
the pipe (footing or cap socket).
In areas susceptible to chlorides and moisture, further research investigation is required to determine the
suitability of the system including its components such as elastomeric joint compounds for wider
application.
Inspection
Inspection during construction and regular service would be similar to a cast-in-place pier with special
attention to column-to-cap and column-to-footing interfaces (e.g., damage to elastomeric pad, moisture
ingress, cracking etc.) For better durability, leaving unbonded length in the pipe can be eliminated.
The plastic hinges are the hot spots in the proposed precast pier during an earthquake. In post-earthquake
inspection, smaller cracks without spalling of cover concrete could indicate the status of the pipe inside
the pier. No spalling of the cover concrete can signal that there is little to no yielding of the pipe. However,
loss of cover concrete in potential plastic hinge zone may require further intrusive inspection of the pipe
by removing the elastomeric pad and cleaning out the debris to inspect the pipe. Non-destructive
hardness test of the steel pipe can be carried out to indicate the amount of inelastic deformation during
the earthquake and the residual capacity and fatigue life. The “elephant leg” buckling of the pipe would
be an indicator of substantial inelastic deformation of the HSS/pipe. A rupture/crack to the pipe means
the need of extensive repair or possible replacement of the pier.
Proposed Repair Methodologies

If a seismic event were to occur that damaged the precast column, the precast column may be able to be
repaired. (Bumstead et al. 2019) explored a variety of repair options for Concrete Filled Steel Tubes (CFST)
after a major seismic event. Each of the repair methods used a concrete pedestal,1.5 times the diameter
of the steel tube, and a height equal to the steel tube, around the damaged section of the column. The
concrete pedestal is encased in a steel tube to confine the concrete. The three methods are a welded
embedded ring, the use of shear studs, and weld beads on the outside of the damaged pipe as shown in
Figure 141. Using a finite element model, via ABAQUS software, the three repair methods are analyzed
with a damaged concrete steel pipe as shown in Figure 142.
Figure 141. Repair Methods, a) Embedded Ring Repair, b) Shear Stud Repair, and c) Weld Bead Repair
Figure 142. Finite Element Model for Repair Methods
The results from the analysis indicated that all of the repair methods are effective in restoring the column's
original stiffness and strength. The welded bead repair is the most effective. The top four analysis results
were: two embedded rings having a width of x4 the pipe thickness (ER-2x4t), four shear studs equally
spaced (SS-x4), three welded beads equally spaced (WB-x3), and three welded beads with a thicker
pedestal confinement tube (WB-x3-1). The graphical results of the described methods are shown in Figure
143 and Figure 144 while Figure 145 depicts the various repair methods that exhibited the highest results.
Figure 143. Moment vs. Drift for Repair Methods
Figure 144. Pedestal Rotation for Repair Methods
a) b) c)
Figure 145. Repair Methods with the Highest Results, a) ER-2x4t, b) SS-x4, c) wb-x3 and wb-x3-1
Some of the repair methods proposed by Bumstead (2019) could be applied to the precast column
connection investigated in this report with some adjustments. Figure 146 presents a proposed repair
methodology for the pipe connection based on existing literature. A concrete pedestal is cast around the
base of the precast column. This repair methodology aims to push the plastic hinge further up the column
and essentially transform the column behavior to a cast-in-place column. The plastic hinge is expected to
form on top of the pedestal during a future earthquake. The diameter of the pedestal should be 1.5 times
the diameter of the precast column. The height of the pedestal should be extended up until the
termination point of the HSS member inside the column as shown in Figure 146. The longitudinal rebars
of the pedestal are epoxied or grouted to the drilled holes in the foundation. A combination of an
embedded annular ring at the base and epoxied studs around the circumference of the column are used
to connect the precast column to the pedestal.
To secure the threaded bars to the column and the pedestal rebars to the footing, the column and footing
should be scanned to locate the existing rebar, so the drill does not damage any rebar. The embedded
ring should be welded to the exposed headed rebars instead of the HSS member because of the difficulty
of welding in a confined area. An alternative to welding the embedded ring to the column after a seismic
event is to weld the embedded ring before the precast column is poured. The embedded ring will not
provide any support to the column until after the column has been damaged and a pedestal is poured.
Testing should be carried out to validate the proposed repair methodologies in Figure 146.
Figure 146. Proposed Repair Method for the ITD Precast Pier: a) Embedded Ring, b) Shear Stud, c)
Weld Bead
The use of a concrete jacket in order to push the plastic hinge zone at the top of the jacket could be one
of the most practical approaches to repair the piers after an earthquake. Recently Mission Bridge in
Canada was seismically retrofitted using this methodology (Figure 147). The jacket can be made of normal
concrete, Ultra-High-performance concrete (UHPC) or structural steel. Similar to concrete jacketing for
the cast-in-place piers, Figure 148 shows a proposed repair method that uses concrete jacketing for the
precast pier. This would turn the precast pier to an equivalent cast-in-place column, but with shorter
height. The proposed repair using concrete jacketing has to be experimentally tested to investigate its
adequacy in restoring stiffness and strength to the precast pier.
Figure 147. Mission Bridge, BC, Canada
Figure 148. Proposed Concrete Jacketing Repair Method for Precast Column Connection
Summary
The flexural analysis of the pier can be conducted similar to CFSTs. Some design considerations and
recommendations for construction/assembly are proposed. A series of interaction diagrams were created
using the WSDOT Bridge Design Manual. The interaction diagrams show the moment capacity increasing
as the HSS member diameter and wall thickness increase. The interaction diagrams do not account for an
axial load but can be used as an effective tool in sizing the HSS/pipe to match a certain cast-in-place
capacity. Some limitations of the precast column were presented with respect to the capacity of the
connection, precasting technology, and the weight of the precast element. It should not be used in areas
with highly corrosive environments until further data/information is available on the performance of
unbonded length and elastomeric bearings. Other areas it should not be used include places where the
plastic hinging goes under the water, where plastic hinging cannot be inspected easily, and the places with
liquifiable soil until further research is available in soil-structure interaction of the system. It is also
recommended to avoid in places where Quality Assurance/Quality Control (QA/QC) cannot be maintained.
Several potential repair methods using an embedded ring, welded beads, and concrete jacketing were
discussed. The proposed repair methods should be experimentally tested to verify their effectiveness in
restoring stiffness and strength to the precast pier. Some post-earthquake inspection and indication of
damage to the HSS/pipe was also discussed.
7. Conclusions
Both the cast-in-place and precast columns were designed with nearly the same materials properties,
flexural capacity, overall shape, and dimensions. The only material that varied was the HSS member. The
cast-in-place cantilever column was determined to have a moment capacity of 165 kip-ft compared to the
precast column connection which had a moment capacity of 143 kip-ft. Although the cast-in-place column
was designed to have a larger moment capacity, its overall seismic performance was lower compared to
the precast column.
The precast cantilever column endured more drift cycles (16 cycles) compared to the 13 cycles that the
cast-in-place column endured. The maximum displacement of the cast-in-place column during the testing
procedure was 7.7 in. which corresponded to a 9.9% drift. The maximum load applied to the cast-in-place
column during the testing procedure was 37.8 kip which corresponded to a 245.7 kip-ft moment capacity.
The maximum displacement of the precast column during the testing procedure was 9.6 in. which
corresponded to a 12.3% drift. This is 20% higher compared to the cast-in-place benchmark. The maximum
load applied to the precast column during the testing procedure was 41.2 kip which corresponded to a
267.8 kip-ft moment capacity. This was 9% higher compared to the cast-in-place benchmark. The precast
column also dissipated more energy 1025 kJ (756,001 ft-lb) compared to 456 kJ (336,328 ft-lb) for the
cast-in-place column which was just below half of what the precast column dissipated. The residual drift
for both columns were similar.
Based on the results above, several conclusions can be made when comparing a precast cantilever column
and cast-in-place column having identical overall dimensions and moment capacities.
• The precast column pipe connection offers ABC advantages as well as better tolerances and
performance compared to other connections.
• During a seismic event, the precast column absorbs more of the seismic energy. The precast
column can absorb more energy because there is more steel located at the column to footing
interface as a result of the HSS pipe which makes the connection more energy dissipative and
ductile.
• The precast column has better confinement and shear resistance.
• The precast column connection will maintain its structural integrity during a major seismic event.
This is evident because the precast column was able to endure more of the cyclic loads during the
testing procedure.
• Based on the observations during each of the testing procedures, the precast column does not
get damaged during smaller seismic events due to lower initial stiffness and flexibility to
accommodate small displacements at the top. This provides better performance for the
serviceability of the bridge. Testing showed that cracks appeared on the cast-in-place column
after the first cycle (0.16 in. of displacement or 0.2% drift) compared to the third cycle (0.47 in. of
displacement or 0.6% drift) for the precast column.
• Overall, the precast column performs better during a seismic event compared to the cast-in-place
column.
• It is possible to repair the precast column using traditional methodologies such as concrete
jacketing after an earthquake
With the CIP and precast bent piers having been designed identical, in both dimensional and performance
capacities, the two are easily compared. The two systems performed similar in nature, with the precast
bent pier showing less overall damage than that of the CIP bent pier. The precast bent pier also achieved
a higher moment capacity of 498 kip-ft., than that of the CIP bent pier, having achieved 468 kip-ft. The
precast bent pier withstood a total of 24 complete loading cycles, culminating in a final ultimate drift ratio
of 9.15%. The CIP bent pier failed far short of the precast achievement, having withstood a total of 15
loading cycles, resulting in an ultimate drift ratio of 4.94%. The two specimens exhibited similar ultimate
capacities at similar drift ratios during the experiment. The precast bent pier achieved a maximum force
of 71.4 kip at a drift ratio of 2.69%. While the CIP bent pier achieved a maximum force of 66 kip at a drift
ratio of 2.46%. Similarly, the precast bent pier dissipated a much higher amount of energy than that of
the CIP bent pier, as it endured more loading cycles. The total dissipated energy of the precast and CIP
bent pier was 2,125 kJ (1,567,320 ft-lb) and 342 kJ (252,246 ft-lb). The precast bent pier had however
dissipated more energy at the fifteenth cycle than that of the CIP bent pier, having dissipated a total of
466 kJ (343,704 ft-lb) to that point.
From the experimental investigation and comparison of the bent specimens made in the research
presented, here several conclusions can be made.
• Compared to an equivalent cast-in-place pier, the precast pier with the proposed pipe connection
achieved higher strength and ductility.
• The precast bent withstood small displacement without suffering hairline cracking due to the
presence of the elastomeric bearing.
• The precast bent displayed better confinement through reduced cracking observed throughout
the experimental program.
• The precast bent displayed far more resilience during the loading as it withstood an additional 9
loading cycles than that of the CIP bent.
• The resilience of the precast bent is demonstrated by the cumulative energy dissipation levels it
was capable of achieving during testing.
• Buckling and tearing of the HSS pipe is observed during large drift ratios to be the failure
mechanism of the proposed precast connection.
• The precast pier demonstrates a reduced stiffness as it approaches ultimate capacity than that of
the CIP bent.
• The precast bent also exhibited a higher yield displacement of 1.13% drift ratio compared to the
0.5% drift exhibited by the CIP bent.
• The precast bent displayed more energy dissipation at the point of failure for the CIP bent as once
the precast bent had achieved ultimate capacity it displayed far less degradation than that of the
CIP bent pier, constituting its ability to withstand additional cycles.
• The precast bent with the proposed precast connection proved capable of successfully emulating
the CIP bent under similar quasi-static loading.
The analytical modeling in this report is aimed to provide a practical tool for bridge engineers when
considering new connection details. Analytical models were created for the CIP column and precast
column with proposed pipe connection using the Open System for Earthquake Engineering Simulation
(OpenSees) software. To predict the experimental results, low-cycle fatigue data were included in the
OpenSees model. Results show good agreement between the analytical and experimental data. Similarly,
analytical models were created for the CIP bent pier and precast bent pier with proposed pipe connection
using OpenSees software. The results showed good agreement between the analytical and experimental
data. It was difficult to capture shear deformation and Bauschinger effects for the precast bent pier after
6% drift ratio.
As far as parametric studies are concerned, some limited nonlinear static (pushover) and nonlinear
dynamic (time-history analysis) were run on an actual bridge model to compare the global seismic
response (e.g., formation of plastic hinges, ductility, strength, force-displacement response etc.) for both
cast-in-place and precast models. Results showed that the precast pier emulates the cast-in-place
behavior closely under nonlinear static and dynamic loads.
Based on the experimental program, flexural analysis, interaction diagrams, design and detailing
considerations, construction/assembly technology, limitations, inspection, potential post-earthquake
repairs, were presented. Most of these guidelines are following existing cast-in-place and precast
technologies for a wider adoption of the proposed ITD connection.
Future Research Work

The following topics are recommended for future research:
• Refining the embedment length of the HSS using experimental pull-out tests and analytical Finite
Element modeling. In this report, two embedment lengths were calculated to determine an
appropriate length. The two calculations were from the WSDOT LRFD Bridge Manual (WSDOT
2019) and Integral Abutment for Steel Bridges (Wasserman and Walker 1996). Additional research
should be conducted to not only verify the methods, but also to close the gap between the two
existing methods.
• Quantifying the appropriate gap between the column and footing pipes that is optimized for
construction tolerance and structural performance.
• Experimental and analytical work to quantify the accurate unbonded length of the pipe. More
information is needed on this topic to understand the effects of the unbonded length of a pipe
embedded in concrete. Leaving an unbonded length is not a common practice and needs to be
understood to clarify its use in CFST.
• The durability aspects of the pier with an unbonded length in the plastic hinge zone should be
investigated.
• The proposed repair methods should be validated experimentally. The literature regarding CFST
is a good resource, but they are based on analytical work only.
• Bi-directional quasi-static cyclic, shake table testing, and consideration of soil-structure

interaction for the proposed pipe connection will provide valuable data about its performance for
a wider application and various seismic hazard/ground motion type.
Cited Works
AASHTO. (2020). AASHTO LRFD Bridge Design Specifications. American Association of State Highway and
Transportation Officials, 9th Edition.
AASHTO. (2011). AASHTO LRFD Seismic Bridge Design. American Association of State Highway and
Transportation Officials, 2nd Edition.
AASHTO M 251. (2015). AASHTO Standard Specification for Plain and Laminated Elastomeric Bridge
Bearings. American Association of State Highway and Transportation Officials, 2015 Edition.
ACI Committee 374. (2013). Guide for testing reinforced concrete structural elements under slowly applied
simulated seismic loads (ACI 374.2R-13). American Concrete Institute, Farmington Hills, MI.
AISC Committee. (2010). Steel Construction Manual American Institute of Steel Construction. Chicago, Ill
Al-Kaseasbeh, Q., and Mamaghani, I. H. P. (2019). “Performance of Thin-Walled Steel Tubular Circular
Columns with Graded Thickness under Bidirectional Cyclic Loading.” ASCE Structures Congress 2019.
Ameli, M. J., Brown, D. N., Parks, J. E., and Pantelides, C. P. (2016). “Seismic column-to-footing connections
using grouted splice sleeves.” ACI Structural Journal.
ASCE. (2017). "Infrastructure Report Card." ASCE's Infrastructure Report Card,
<https://www.infrastructurereportcard.org/cat-item/bridges/> (Jan. 8, 2020)
ASCE. (2018). Report Card for Idaho’s Infrastructure." ASCE's Infrasture Report Card,
<https://www.infrastructurereportcard.org/cat-item/bridges/> (Jan. 8, 2020)
Brown, J. and S. K. Kunnath (2000). Low-cycle Fatigue Behavior of Longitudinal Reinforcement of
Reinforced Concrete Bridge Columns. Report No. MCEER 00-0007. University at Buffalo, The State
University of New York.
Bumstead, J., Korat, J., and Stephens, M. T. (2019). “Repair Strategies for Earthquake-Damaged CFST
Bridge Columns.” ASCE Structures Congress 2019, 154–164.
Chapel Steel. (2018). Structural, Carbon & HSLA Steel Plate. <https://www.chapelsteel.com/a709-
grade50-grade-345.html> (Feb. 22, 2020).
Caltrans. (2013). Seismic Design Criteria-Version 1.7. Caltrans. California Department of Transportation,
Sacramento, CA
Chopra, Anil K. (2017). Dynamics of Structures: Theory and Applications to Earthquake Engineering.
Pearson Education Limited.
Collins, M. P., and Mitchell, D. (1991). Prestressed Concrete Structures. Prentice-Hall. Englewood Cliffs, NJ.
Ebrahimpour, A., B. E. Earles, S. Maskey, M. Tangarife, and A. D. Sorensen (2016). Seismic performance of
columns with grouted couplers in Idaho accelerated bridge construction applications. Report No.
FHWA-ID-16-246. Idaho Transp. Department, Boise, ID.
Galvis, F., and Correal, J. F. (2017). “Characterization of the seismic behavior of a column-foundation
connection for accelerated bridge construction.”
Haber, Z. B., M. S. Saiidi, and D. H. Sanders (2013). Precast Column-Footing Connections for Accelerated
Bridge Construction in Seismic Zones. Report No. CCEER 13-08. Department of Civil Engineering,
University of Nevada, Reno, NV.
Hawileh, R., A. Rahmen, and H. Tabatabai (2010). “Evaluation of the Low-cycle Fatigue Life in ASTM A706
and A615 grade 60 Steel Reinforcing Bars,” Journal of Materials in Civil Engineering, V. 22(1), pp. 65-
76.
Kim, C. S., Lim, W. Y., Park, H. G., and Oh, J. K. (2016). “Cyclic loading test for cast-in-place concrete-filled
hollow precast concrete columns.” ACI Structural Journal.
Marsh, L. P. (2018). “Seismic Design and Accelerated Bridge Construction (ABC).”
<http://onlinepubs.trb.org/onlinepubs/webinars/180711.pdf>.(Jan. 10, 2020)
Marsh, M. L. et al. (2011). Application of Accelerated Bridge Construction Connections in Moderate-to-
High Seismic Regions.
Mashal, M., and Palermo, A. (2019a). “Low-damage seismic design for accelerated bridge construction.”
Journal of Bridge Engineering.
Mashal, M., and Palermo, A. (2019b). “Emulative seismic resistant technology for Accelerated Bridge
Construction.” Soil Dynamics and Earthquake Engineering, Elsevier Ltd, 124(December 2018), 197–
211.
Mashal, M., White, S., and Palermo, A. (2016). “Quasi-static Cyclic Testing of Emulative Cast-in-place
Connections for Accelerated Bridge Construction in Seismic Regions.” Bulletin of the New Zealand
Society for Earthquake Engineering, 49(3).
Maskey, S. (2017). “Low-cycle Fatigue Behavior of Reinforcing Steel in Idaho Bridge Columns During
Earthquakes.” M.S. thesis, Dept. of Civil and Environmental Engineering, Idaho State Univ.
McCormac, J.C., and Brown, R.H. (2014) Design of Reinforced Concrete. Wiley, Hoboken, NJ
Mehraein, M., and Saiidi, M. S. (2016). Seismic Performance of Bridge Column-Pile-Shaft Pin Connections
for Application in Accelerated Bridge Construction. Reno, NV.
Open System for Earthquake Engineering Simulation (OpenSees) [Computer software]. Univ. of California,
Berkeley, Berkeley, CA. <http://opensees.berkeley.edu/OpenSees/user/index.php> (October 21,
2019).
Ou, Y. C., Alrasyid, H., Haber, Z. B., and Lee, H. J. (2015). “Cyclic behavior of precast high-strength
reinforced concrete columns.” ACI Structural Journal.
Pampanin, S. et al. (2010). PRESSS Technology Design Handbook. NZCS, Auckland N.Z.
PEER. (2020). "Ground Motion Database." PEER Ground Motion Database, University of California,
Berkeley <http://ngawest2.berkeley.edu>
Priestley, M. J. N., Calvi, G. M., and Kowalsky, M. J. (2007). Displacement-Based Seismic Design of
Structures. Earthquake Spectra, IUSS Press.
Tazarv, M., and Saiidi, M. S. (2015a). Design and Construction of Precast Bent Caps with Pocket
Connections for High Seismic Regions. Reno, NV.
Tazarv, M., and Saiidi, M. S. (2015b). Design and Construction of Bridge Columns Incorporating Mechanical
Bar Splices in Plastic Hinge Zones. Reno, NV.
U.S. Department of Transportation/Federal Highway Administration. (2019). “Accelerated Bridge
Construction.” <https://www.fhwa.dot.gov/bridge/abc/> (Jan. 8, 2020).
Vielma, J. C., and Mulder, M. M. (2017). “Procedure for Assessing the Displacement Ductility Based on
Seismic Collapse Threshold and Dissipated Energy Balance.” 16th World Conference on Earthquake
Engineering, (January).
Wasserman, E. P., and Walker, J. H. (1996). Integral Abutments for Steel Bridges. American Iron and Steel
Institute, Chicago, Ill
Wehbe, N. I., M. S. Saiidi, and D. H. Sanders (1999). “Seismic Performance of Rectangular Bridge Columns
with Moderate Confinement,” ACI Structural Journal. 96(2), pp. 248-258.
WSDOT. (2019). "Bridge Design Manual (LRFD)." Engineering and Regional Operations Bridge and
Structures Office
Zaghi, A. E., and Saiidi, M. S. (2010). Seismic Design of Pipe-Pin Connections in Concrete Bridges. Reno, NV.
Zollman, C. C., Depman, F., Nagle, J., and Hollander, E. F. (1992). “Building and Rebuilding of Philadelphia’s
Walnut Lane Memorial Bridge.” PCI Journal, 37(4), 66–82.
Appendices
Appendix A. [Design Calculations of the Column Specimen]
A.1 Cast-In-Place Column Calculations
A.2 Precast Column Calculations
Appendix B. [Design Calculations of the Bent Specimen]
B.1 Cast-In-Place Bent Calculations

Circular Column Dimensions Compression strain = -0.003
Diameter = 14 in Tensile strain = 0.002
Cover = 1.5 in fy = 60 ksi
Stirrup = 3 f'c = 4 ksi
Long. Bar #= 6 Es = 29000 ksi
Total bars = 7 B1 = 0.85 (see pg 68)
Ds = 9.5 in
For equivalent rectangular Column (Figure 10.9)
B = Ag/(.8D) = 13.74
H = .8*D = 11.20
Hs = Ds*2/3 = 6.33
A's = 1.55
Cover = 2.43
Determine values of y and of the steel stains
y = 0.003 / ( 0.003 + 0.002 ) * 11.2= 6.72 in.

e's = (( 6.72 - 2.43 ) / 6.72 ) * 0.003 = 0.00191 <
0.00207 Compression Steel Does Not Yield
es = (( 11.2 - 6.72 - 2.43 ) / ( 11.2 - 6.72 ))*0.002 = 0.00091 <
0.00207 Tension Steel Does Not Yield
a = B1*c = 0.85 * 6.72 = 5.71 in.

Cc = B1*a*B*f'c = -266.93 k Fe = 0.75
C's = fy*A's-B1*A's*f'c = -87.52 k
Ts = es*E*A's = 40.97 k
Column height = 6.48 ft
B.2 Precast Bent Calculations
Pipe Connection fy = 42 ksi AISC Table 2-4 pg 2-48
Pipe Diameter = 6 in. fu = 58 ksi AISC Table 2-4 pg 2-48
Pipe Thickness= 0.465 in. f'c = 8 ksi
Concrete Diameter = 5.07 in. Es = 29000 ksi
D/t = 12.90322581 < Good 103.57
As = 8.09 in2
Ac = 20.19 in2
Determine y value by changing the following values:
Applied axial load (P) = 50.0 kips Ts = 355.02 kips P + Ts =

405.0225634 kips
Ast = 8.453 in2 Cc = 143.38 kips Cc + Cs = 405.0225634 kips
Acc = 18.866 in2 Cs = 261.64 kips 0 kips ** Set this

value equal to zero using solver.
Asc = 6.230 in2
Determine y value by using the solver

Summing the moments about the c.g.c of the column:
y= 0.0 in. M= 32.81250014 kip-in.

** Set this value equal to zero using solver.
Appendix C. [Grout Product Data Sheet]
Appendix D. [Interaction Diagrams and Tables for the Precast Column]
Screenshots of table calculations for various pipe connections and thicknesses follows. To view
accessible tables displaying these calculations, please contact the research group in the Disaster
Response Complex at Idaho State University (208-282-2902), or the Idaho Transportation Department
Research Program (research@itd.idaho.gov).
D.1 HSS Pipe Connection f’c = 4 ksi
D.1.1 HSS thickness = 0.174 in.
D.2 HSS Pipe Connection f’c = 8 ksi
D.3 Cast-in-place Moment Capacities with f’c = 4 ksi

Dot 63530 DS1-1

Uploaded by

Copyright:

Available Formats

Dot 63530 DS1-1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dot 63530 DS1-1

Uploaded by

Copyright:

Available Formats

IDAHO TRANSPORTATION DEPARTMENT

A Precast Pier System for Accelerated

Mustafa Mashal, Arya Ebrahimpour, Mahesh Acharya, Jared Cantrell, Corey

Idaho State University

Idaho Transportation Department

ITD Research Program, Contracting Services

Highways Construction and Operations

This report does not constitute a standard, specification, or regulation.

A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 2

6. Performing Organization Code

7. Author(s) 8. Performing Organization Report No.

A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 3

• Matt Farrar, P.E.

• Dan Gorley, P.E.

• Leonard Ruminski, P.E.

• Michael Johnson, P.E.

• Craig Wielenga, P.E.

A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 4

A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 5

A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 6

A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 7

A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 8

A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 9

A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 10

A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 11

A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 12

A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 13

A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 14

ABC Accelerated Bridge Construction

ACI American Concrete Institute

ASTM American Society for Testing and Materials

CFST Concrete Filled Steel Tubes

HSS Hollow Structural Section

ITD Idaho Transportation Department

PBES Prefabricated Bridge Elements and Systems

PVC Polyvinyl Chloride

SLAB Structural LAB

SIBC Slide-in Bridge Construction

UHPC Ultra-High Performance Concrete

WSDOT Washington State Department of Transportation

A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 15

Task 1. Literature review

a) Literature review on ABC in moderate-to-high seismicity

b) Available technologies for seismic resistant connections between precast elements

c) Comparison of past studied connections from the perspectives of transportation, seismic

d) Experimental testing techniques for ABC bridge substructure systems

A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 16

Task 2. Experimental Investigation

c) Construction of the specimens

f) Instrumentation and data acquisition system for experimental testing

g) Experimental testing of the specimens up to failure point

h) Disposal/demolition of the tested specimens

i) Processing of testing results (e.g., force-displacement hysteresis, moment-rotation etc.)

A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 17

A Precast Pier System for Accelerated Bridge Construction (ABC) in Idaho 18

Performance Factors Cast-in-place column Precast column

Ultimate Base Shear 37.8 kip 41.2 kip

Total Energy dissipated 456 kJ 1025 kJ

Table 2. Summarized Comparison of CIP and Precast Bent