National Academies Press: OpenBook

Proposed AASHTO Seismic Specifications for ABC Column Connections (2020)

Chapter: Chapter 3 - Experimental Programs

« Previous: Chapter 2 - State-of-the-Art Literature Review
Page 101
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 101
Page 102
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 102
Page 103
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 103
Page 104
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 104
Page 105
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 105
Page 106
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 106
Page 107
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 107
Page 108
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 108
Page 109
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 109
Page 110
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 110
Page 111
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 111
Page 112
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 112
Page 113
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 113
Page 114
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 114
Page 115
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 115
Page 116
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 116
Page 117
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 117
Page 118
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 118
Page 119
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 119
Page 120
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 120
Page 121
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 121
Page 122
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 122
Page 123
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 123
Page 124
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 124
Page 125
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 125
Page 126
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 126
Page 127
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 127
Page 128
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 128
Page 129
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 129
Page 130
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 130
Page 131
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 131
Page 132
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 132
Page 133
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 133
Page 134
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 134
Page 135
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 135
Page 136
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 136
Page 137
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 137
Page 138
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 138
Page 139
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 139
Page 140
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 140
Page 141
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 141
Page 142
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 142
Page 143
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 143
Page 144
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 144
Page 145
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 145
Page 146
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 146
Page 147
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 147
Page 148
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 148
Page 149
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 149
Page 150
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 150
Page 151
Suggested Citation:"Chapter 3 - Experimental Programs." National Academies of Sciences, Engineering, and Medicine. 2020. Proposed AASHTO Seismic Specifications for ABC Column Connections. Washington, DC: The National Academies Press. doi: 10.17226/25803.
×
Page 151

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

101 3.1 Introduction Chapter 2 summarized the state-of-the-art literature review of four precast column connections that was conducted for NCHRP Project 12-105 and the knowledge gaps that were identified. After the NCHRP 12-105 project panel reviewed the available test data and connection performance, it selected three types of connection for development of guidelines: mechanical bar coupler connections, grouted duct connections, and pocket/socket connections. Table 3-1 synthesizes all available test data from the literature and presents the experimental program undertaken in the present project to fill the knowledge gaps. The tests performed in the present study included three groups of specimens: 30 mechanical bar splices, a mechanically spliced large-scale column model, and 12 grouted duct specimens. The bulk of the experimental data synthesized from the literature was found sufficient for columns incorporating grouted ducts and pocket/socket connections; thus, no testing was performed on columns that use these connections. This chapter discusses the experimental program, procedures, and key results for specimens tested in the present study. 3.2 Mechanical Bar Splices The objectives of the experimental program for mechanical bar splices (or couplers) were to determine the feasibility of uniform acceptance criteria for seismic couplers and to verify and refine the coupler modeling method. The general experimental parameters and the expected outcome of the coupler testing are presented in Table 3-1. ASTM A706 Grade 60 steel No. 10 reinforcing bars were used in the spliced specimens. The majority of previous studies on couplers were focused on No. 8 or smaller bars. Bridge column longitudinal bars are typi- cally larger, with No. 10 and No. 11 being very common. To generate information for realistic bar sizes, No. 10 bars were utilized in the present study. Table 3-2 presents the test matrix for mechanical bar splices. All the specimens were tested under tensile loads. The testing of the spliced bars consisted of static, half-cyclic, and dynamic loading of 30 spliced specimens to generate information for three types of couplers: grouted sleeves, headed bars, and swaged couplers. 3.2.1 Coupler Rigid Length Ratio A material model was needed for mechanical bar splices to systematically establish their behavior. The coupler material discussed in Section 2.6.2 was adopted in the present study. C H A P T E R 3 Experimental Programs

Type of Connection Test Data to Support Design Equations Objective Experimental Studies (NCHRP 12-105) Experimental Parameters Deliverable Mechanical bar splices No. of available coupler test data: Swaged: 7 Headed bar: 10 Grouted sleeve: 43 No. of new test data from NCHRP 12-105: Swaged: 10 Headed bar: 10 Grouted sleeve: 10 1. Determine the feasibility of the uniform acceptance criteria developed under the present project and make the necessary refinements. 2. Verify and refine the proposed modeling method for couplers. Extensive testing of different couplers according to the proposed acceptance criteria to obtain the mechanical properties of the couplers and refine the acceptance criteria. 10 No. 10 (32 mm) spliced bars per selected coupler type were tested to failure. Of the five types of couplers discussed in Chapter 2, grouted sleeves, headed bars, and swaged couplers were selected. Ten samples were tested for each type: three each under monotonic and half-cyclic loading and four under dynamic loading. 1. Additional data to help finalize the acceptance criteria for couplers for seismic application. 2. Finalized acceptance criteria for seismic couplers. Mechanically spliced column connections No. of available column test data: Swaged: 1 Headed bar: 6 Grouted sleeve: 9 No. of new test data from NCHRP 12-105: Grouted sleeve: 1 1. Verify the design and performance of columns with couplers. 2. Finalize the column testing protocol. Testing of a large-scale mechanically spliced cantilever column with couplers in the plastic hinge following the testing protocol. Grouted sleeves were used to splice the column longitudinal bars. No. 10 (32 mm) reinforcing steel bars were used, and the column was tested under static cyclic loading. 1. Verified coupler design equation(s) and information about coupler performance and performance of mechanically spliced columns. 2. Standard column testing protocol. Grouted duct design equations No. of available grouted duct test data: Pullout test data: 119 No. of new test data from NCHRP 12-105: Grouted ducts: 12 Develop proof-tested, reliable design equations for embedment length for grouted ducts with conventional grout. Pullout testing of No. 10 (32 mm) bars anchored in grouted duct connections. Under monotonic loading, the bond behavior of grouted duct connections was investigated. The test parameters were duct diameter, duct thickness, bar eccentricity, and bundled bars. The test embedment lengths were within a practical range. Verified design equations for embedment length of bars in conventional grouted ducts. Column with grouted duct connection No. of available columns with grouted duct test data: 16 No. of new test data from NCHRP 12-105: None No testing was planned, since sufficient test data were available. Not applicable Not applicable Not applicable Column with pocket connection No. of available columns with pocket connection test data: 37 No. of new test data from NCHRP 12-105: None No testing was planned, since sufficient test data were available. Not applicable Not applicable Not applicable Note: All proposed tests are static unless otherwise specified. Table 3-1. Past and proposed experimental programs for three accelerated bridge construction column connections.

Experimental Programs 103 Details of the model are presented in Tazarv and Saiidi (2016). In this model, a portion of the splice (β.Lsp) is rigid and thus does not contribute to the elongation of the coupler region. Lsp is the coupler length and β is defined as the “coupler rigid length ratio.” 3.2.2 Experimental Studies and Results for Grouted Sleeve Splices Seventeen grouted sleeve couplers to splice No. 10 bars at both ends were ordered. Of these, seven were incorporated in the plastic hinge zone of a large-scale column model described in Section 3.3. The remaining 10 couplers were used to splice No. 10 bars and were instrumented and tested as individual components. Figure 3-1 shows the test setup for a grouted sleeve specimen. The tests on the mechanical bar splices were conducted with new testing methods developed under this project, and then the methods were finalized and written in the format of AASHTO/ASTM (Appendix B of this report). SS Mortar, as specified and supplied by the sleeve manufacturer, was used as grout. Extra bar samples of the same lot were ordered to determine the stress–strain characteristics of the ID Bar Size Coupler Type Description GS GSM1 No. 10 Grouted sleeve coupler Monotonic Loading GSM2 No. 10 Grouted sleeve coupler GSM3 No. 10 Grouted sleeve coupler HBM1 No. 10 Headed bar coupler HBM2 No. 10 Headed bar coupler HBM3 No. 10 Headed bar coupler SWM1 No. 10 Swaged coupler SWM2 No. 10 Swaged coupler SWM3 No. 10 Swaged coupler GSC1 No. 10 Grouted sleeve coupler Cyclic Test HB GSC2 No. 10 Grouted sleeve coupler GSC3 No. 10 Grouted sleeve coupler HBC1 No. 10 Headed bar coupler HBC2 No. 10 Headed bar coupler HBC3 No. 10 Headed bar coupler SWC1 No. 10 Swaged coupler SWC2 No. 10 Swaged coupler SWC3 No. 10 Swaged coupler GSD1 No. 10 Grouted sleeve coupler Dynamic Test SW GSD2 No. 10 Grouted sleeve coupler GSD3 No. 10 Grouted sleeve coupler GSD4 No. 10 Grouted sleeve coupler HBD1 No. 10 Headed bar coupler HBD2 No. 10 Headed bar coupler HBD3 No. 10 Headed bar coupler HBD4 No. 10 Headed bar coupler SWD1 No. 10 Swaged coupler SWD2 No. 10 Swaged coupler SWD3 No. 10 Swaged coupler SWD4 No. 10 Swaged coupler Table 3-2. Test matrix for coupler studies.

104 Proposed AASHTO Seismic Specifications for ABC Column Connections bars. The average measured yield and ultimate strength of the unspliced No. 10 bars used in the grouted sleeve specimens were 77.1 ksi and 105.7 ksi, respectively (Table 3-3). On the day of testing (±3 days), the average compressive strength of the SS Mortar was 9.8 ksi (age = 123 days), 13.2 ksi (age = 150 days), and 11.4 ksi (age = 164), in the monotonic static, half-cyclic, and monotonic dynamic tests, respectively. The specified 28-day compressive strength of SS Mortar is 12 to 14 ksi, depending on the temperature. Figure 3-2 shows the mode of failure for the 10 grouted sleeve specimens after the testing. Shown in the figure are also the fracture stresses and strains. Tables 3-3 to and 3-6 list the measured yield stresses as well as the stresses and strains at the ultimate point (the point of maximum stress) for the grouted sleeve (GS) specimens tested under monotonic (GSM), cyclic (GSC), and dynamic (GSD) loading, respectively. The tables also indicate whether the failure was within the gauge length. The averages exclude the data for the specimens that failed within the coupler (GSM1, GSM2, GSC1, and GSC4). The strain rates listed in Table 3-6 are discussed in Section 3.2.2.1, specifically on dynamic loading for all couplers. Figure 3-1. Test setup for a grouted sleeve specimen. Specimen Yield Stress (ksi) Ultimate Stress (ksi) Ultimate Strain (in./in.) Fracture Strain (in./in.) GST1 76.8 105.8 0.119 0.164 GST2 77.4 105.3 0.118 0.166 GST3 77.0 106.0 0.110 0.116 Average 77.1 105.7 0.116 0.165 Mill report 75.7 102.2 — 0.178 Note: GST = control bars used in grouted sleeve coupler tensile tests. Table 3-3. Measured stress–strain properties of control bars used in grouted sleeve splices.

Experimental Programs 105 GSM2 - fu = 99.9 ksi - εu = 23,642 microstrainGSM1 - fu = 97.0 ksi - εu = 31,363 microstrain GSM3 - fu = 104.5 - εu = 73,397 microstrain GSC1 - fu = 100.9 ksi - εu = 24,035 microstrain GSC2 - fu = 96.22 ksi - εu = 51,483 microstrain GSC3 - fu = 96.82 ksi - εu = 45,637 microstrain GSD1 - fu = 107.8 ksi - εu = 64,477 microstrain GSD2 - fu = 106.9 ksi - εu = 51,244 microstrain GSD3 - fu = 107.6 ksi - εu = 68,377 microstrain GSD4 - fu = 100.9 ksi - εu = 23,503 microstrain c) Fracture under dynamic monotonic loading b) Fracture under cyclic loading a) Fracture under static monotonic loading Note: fu = ultimate strength of spliced bar; εu = strain at peak stress. Figure 3-2. Fracture of spliced bars with grouted sleeve couplers.

106 Proposed AASHTO Seismic Specifications for ABC Column Connections Specimen Yield Stress (ksi) Ultimate Stress (ksi) Ultimate Strain (in./in.) Failure Location GSM1 77.0 93.7 0.019 Within gauge length GSM2 76.0 98.7 0.024 Within gauge length GSM3 79.0 104.5 0.051 Outside gauge length Average 79.0 104.5 0.051 — Table 3-4. Measured data for grouted sleeve splices tested under monotonic loads. Specimen Yield Stress (ksi) Ultimate Stress (ksi) Ultimate Strain (in./in.) Failure Location GSC1 75.0 100.9 0.023 Within gauge length GSC2 75.7 102.8 0.049 Outside gauge length GSC3 76.5 104.6 0.041 Outside gauge length Average 76.1 103.7 0.045 — Table 3-5. Measured data for grouted sleeve splices tested under cyclic loads. Specimen Yield Stress (ksi) Ultimate Stress (ksi) Ultimate Strain (in./in.) Achieved Displacement Rate (in./s) Net Length Strain Rate (ld/s) Failure Location GSD1 82.4 107.8 0.045 1.95 72,853 Outside gauge length GSD2 80.9 106.9 0.045 1.93 72,015 Outside gauge length GSD3 80.7 107.5 0.043 1.94 72,433 Within gauge length GSD4 82.2 101.0 0.023 1.94 70,826 Within gauge length Average 81.3 107.4 0.044 1.94 72,434 — Table 3-6. Measured data for grouted sleeve splices tested under dynamic loads. 3.2.2.1 Failure Mode in Grouted Sleeve Couplers All the grouted sleeve spliced specimens failed in a ductile manner, but the ultimate strain (the strain at the peak stress) varied. The failure location varied between specimens, even those with the same loading type. Of the monotonically tested specimens, two experienced fracture in the coupler and one in the bar. The bar stress at fracture was comparable for all three, but the fracture strain over the coupler region in the former two was less than one- half of that in the third specimen. The shape of the stress–strain diagrams is discussed in subsequent sections. Of the three specimens tested under half-cyclic loading, one failed within the coupler and two in the bar. All three specimens failed in a ductile mode. However, similar to what was observed in the monotonically tested specimens, the ultimate strain in the former was approximately one-half that in the other two. The bar stress at failure was comparable for all three. Four grouted sleeve specimens were tested under monotonic dynamic loads. The mode of failure was also ductile in all four. The bar fracture stress was in the same range for all four specimens. The fracture in three of the four specimens was in the bar, but the fourth specimen failed within the coupler. The fracture strain in the fourth specimen was less than one-half of that in the other three.

Experimental Programs 107 Of the 10 grouted sleeve coupler specimens, four failed at the coupler center in between the spliced bar ends. Note that grouted sleeve couplers include a stopper with a central hole that allows for the grout to flow but prevents the two spliced bar ends to bear directly on each other. Coupler fracture occurred regardless of the loading type under forces that were approximately the same for all load types. The splice fracture did not reduce the strength appreciably but reduced the ultimate strain significantly. A close examination of the fractured zone indicated that the material was not homogeneous in the fracture plane (Figure 3-3). The effect of the coupler failure in GSM1 and GSM2 can be seen in the ultimate stresses and strains in Table 3-4 when the data are compared with those for GSM3, in which failure was in the bar. GSM1 and GSM2 were excluded in the average ultimate strain for the spliced bars because of coupler failure in these two specimens. Figure 3-4 shows the stress–strain relationship for the spliced specimens and the average rela- tionship for the control bars. The yield stress was comparable between the two groups, but the post-yield strain in the spliced specimens were significantly lower than those of control bars because of the coupler rigidity. 3.2.2.2 Effect of Load Type on Grouted Sleeve Couplers The acceptance criteria for spliced bars proposed in Appendix B includes monotonic static, half-cyclic, and monotonic dynamic loading, with the latter two intended to evaluate the per- formance of the splices under earthquake-type loading. The fracture of the specimens with grouted sleeve couplers was discussed in the previous section. The stress–strain diagrams for the specimens under different types of loading are shown in Figure 3-5. Several trends are evident. The yield stress under cyclic loading was the same as that under monotonic static loading. However, the yield stress increased slightly under dynamic loading, which is expected even in unspliced bars. Another trend is the lower fracture strain for specimens in which the coupler fractured (GSM1, GSM2, GSC1, and GSD4) as compared with those in which the bars fractured. Figure 3-3. Fracture of grouted sleeve splice. Color Variation Bar End Grout

108 Proposed AASHTO Seismic Specifications for ABC Column Connections The coupler fracture strain was not affected by the loading type. However, when fracture was in the bar (the other six specimens), the cyclic load led to a reduction in the fracture strain as compared with the monotonic static load. Dynamic loading also reduced the fracture strain, but to a lesser extent. To better visualize the effect of cyclic loading on the spliced bar stress–strain relationships, the envelopes of the stress–strain diagrams under cyclic loading were superimposed on the measured diagram for GSM3 (GSM1 and GSM2 were excluded because of fracture within the coupler). The result is shown in Figure 3-6, with fracture points marked by triangles. Recall that GSC1 failed within the coupler. On the basis of the figure and comparison of the data in Tables 3-4 and 3-5, it can be seen that cyclic loading reduced the strength of the coupler by a small amount but reduced the ultimate strain from 0.051 to 0.045, a reduction of approximately 10%. The effect of dynamic loading is captured in Figure 3-7 and by comparison of the data in Tables 3-4 and 3-6. The fracture points are marked by triangles in the figure. The slight increase in the yield and ultimate stresses is evident in the plots and expected under dynamic loading. The yield and ultimate stresses increased by approximately 3%. The ultimate strain was reduced by approximately 10%, similar to the reduction seen under the cyclic loading. Judging on the basis of the ultimate strain being nearly the same in Tables 3-5 and 3-6, it is possible to conclude that cyclic loading sufficiently captured the earthquake-type load effects and that dynamic testing did not produce new information, other than the slight increase in the yield stress and ultimate strengths, which was expected. 3.2.3 Experimental Studies and Results for Headed Bar Splices Ten headed bar couplers were tested as part of the experimental program: three each under static monotonic or cyclic loading and four under dynamic loading. Figure 3-8 shows the test setup for a headed bar specimen before the two displacement transducers were placed on the sides of the coupler. Unspliced bar samples of the same lot were also tested to determine the stress–strain characteristics of the reference bars per ASTM A370. The unspliced bar test data were used to establish the baseline for the bars that were linked Figure 3-4. Stress–strain relationships for grouted sleeve splices and control bars.

Experimental Programs 109 (a) Static Monotonic Stress–Strain Curves (c) Dynamic Monotonic Stress–Strain Curves (b) Static Cyclic Stress–Strain Curves Figure 3-5. Effect of loading type on stress–strain relationships of grouted sleeve splice specimens.

110 Proposed AASHTO Seismic Specifications for ABC Column Connections Figure 3-6. Effect of cyclic loading on stress–strain relationships of grouted sleeve splice specimens. Figure 3-7. Effect of dynamic loading on stress–strain relationships of grouted sleeve splice specimens. with the couplers and to provide data on the basis of which the rigidity of the coupler region could be quantified by comparing the stress–strain relationships of the spliced and unspliced bars. The average measured yield and ultimate strength of the unspliced No. 10 (32 mm) bars used in the headed bar specimens were 70.8 ksi and 98.0 ksi, respectively (Table 3-7). The tests were conducted with preliminary testing methods that were developed in the present study and then finalized as presented in Appendix B. Figure 3-9 shows the fractured spliced specimens after the tests. Shown in the figure are also the fracture stresses and strains. Note that all the spliced specimens failed away from the couplers, owing to the fracture of the bars. Tables 3-8 to 3-10 list the measured yield stresses as well as the stresses and strains at the ultimate point (the point of maximum stress) for the headed bar (HB) specimens tested under

Experimental Programs 111 monotonic (HBM), cyclic (HBC), and dynamic (HBD) loading, respectively. The tables also indicate whether the failure was within the gauge length. 3.2.3.1 Failure Mode in Headed Bar Couplers All the headed bar coupler specimens failed in a ductile manner outside the splice, but the ultimate stresses and strains varied to different extents. The data were consistent within each loading type. The fracture location in all 10 specimens was in the bar at two or more times the bar diameter away from the coupler ends. There was no coupler fracture. The shape of the stress–strain diagrams and the effect of loading type are discussed in sub- sequent sections. The bar stress at fracture was comparable for all specimens within each loading group. Figure 3-10 shows the stress–strain relationship for the spliced specimens and the average relationship for the control bars. Note that the data for HBM3 could not be obtained because Figure 3-8. Test setup for a headed bar coupler specimen. Specimen Yield Stress (ksi) Ultimate Stress (ksi) Ultimate Strain (in./in.) Fracture Strain (in./in.) HBT1 71.0 98.0 0.103 0.121 HBT2 70.8 97.9 0.105 0.135 HBT3 70.5 98.2 0.111 0.144 Average 70.8 98.0 0.106 0.140 Mill report 74.9 98.1 — 0.130 Note: HBT = control bars used in headed bar coupler tensile tests. Table 3-7. Measured stress–strain properties of control bars used in headed bar splices.

112 Proposed AASHTO Seismic Specifications for ABC Column Connections Figure 3-9. Fracture of spliced bars with headed bar couplers. HBM2 - fu = 97.72 ksi - εu = 73,385 microstrainHBM1 - fu = 97.31 ksi - εu = 112,563 microstrain HBM3 - fu = 98.42 ksi - εu = N/A HBC1 - fu = 96.22 ksi - εu = 69,391 microstrain HBC2 - fu = 96.22 ksi - εu = 69,906 microstrain HBC3 - fu = 96.82 ksi - εu = 65,424 microstrain HBD1 - fu = 101.1 ksi - εu = 120,319 microstrain HBD2 - fu = 100.9 ksi - εu = 97,983 microstrain HBD3 - fu = 101.2 ksi - εu = 82,972 microstrain HBD4 - fu = 101.3 ksi - εu = 80,340 microstrain c) Fracture under dynamic monotonic loading b) Fracture under cyclic loading a) Fracture under static monotonic loading

Experimental Programs 113 Specimen Yield Stress (ksi) Ultimate Stress (ksi) Ultimate Strain (in./in.) Failure Location HBM1 70.3 97.3 0.073 Within gauge length HBM2 70.3 97.7 0.070 Outside gauge length HBM3 NA NA NA Outside gauge length Average 70.3 97.5 0.072 — Note: NA = not available. Table 3-8. Measured data for headed bar splices tested under monotonic loads. Specimen Yield Stress (ksi) Ultimate Stress (ksi) Ultimate Strain (in./in.) Failure Location HBC1 70.1 96.2 0.069 Outside gauge length HBC2 70.5 97.2 0.070 Outside gauge length HBC3 70.0 96.8 0.065 Outside gauge length Average 70.2 96.7 0.068 — Table 3-9. Measured data for headed bar splices tested under cyclic loads. Specimen Yield Stress (ksi) Ultimate Stress (ksi) Ultimate Strain (in./in.) Achieved Displacement Rate (in./s) Net Length Strain Rate (ld/s) Failure Location HBD1 73.0 101.1 0.079 1.93 88,618 Within gauge length HBD2 72.7 100.9 0.098 1.96 89,907 Within gauge length HBD3 73.1 101.2 0.083 1.96 90,094 Outside gauge length HBD4 73.5 101.3 0.080 1.96 90,234 Outside gauge length Average 73.1 101.1 0.085 1.95 89,713 — Table 3-10. Measured data for headed bar splices tested under dynamic loads. Figure 3-10. Stress–strain relationships for headed bar splices and control bars.

114 Proposed AASHTO Seismic Specifications for ABC Column Connections of improper attachment of the transducer. The curves indicate that the stiffness of the spliced bars was lower than the control bar stiffness because of slight settling with the couplers. How- ever, the yield points, the ultimate stresses, and the ultimate strains of the spliced and unspliced specimens were nearly the same, although the fracture strains varied. The data are also listed in Table 3-8. The average ultimate strain in the spliced bars was 0.072 in./in. 3.2.3.2 Effect of Load Type on Headed Bar Couplers The acceptance criteria for spliced bars proposed in Appendix B includes monotonic static, half-cyclic, and monotonic dynamic loading, with the latter two intended to evaluate the perfor- mance of the splices under earthquake-type loading. The fracture of the specimens with headed bar couplers is discussed in the previous section. The stress–strain diagrams for the specimens under different types of loading are shown in Figure 3-11. Several trends are evident. Some of the trends for headed bar couplers are similar to those of the grouted sleeve couplers. The yield stress under cyclic loading was the same as that under monotonic static loading. However, the yield and ultimate strengths increased under dynamic loading, which is expected even in unspliced bars. The ultimate strains under cyclic loading tended to be smaller than those under monotonic static loading, but the difference is in the range of the data scatter. Another trend is the higher failure strain under dynamic loading, but there is significant scatter in the data. The effect of cyclic loading on the spliced bar stress–strain relationships is shown in Figure 3-12 by comparing the envelopes of the stress–strain diagrams under cyclic loading on the measured diagram for HBM1. The fracture points are marked by triangles in the figure. The data for HBM1 and HBM2 are nearly identical (Table 3-8), but HBM1 was selected because fracture in this specimen occurred within the gauge length. On the basis of the figure and comparison of the data in Tables 3-8 and 3-9, it can be seen that the effect of cyclic loading was negligible, reducing the ultimate stress and strain by 5% or less. The effect of dynamic loading is shown in Figure 3-13 and can be quantified by com- paring Tables 3-8 and 3-10. The fracture points are marked by triangles in the figure. The slight increase in the yield and ultimate stresses is evident in the plots and expected under dynamic loading. The yield and ultimate stresses increased by approximately 4%. The ulti- mate strain was also increased by approximately 10%, which may be attributed to scatter in the data. The average ultimate strain under dynamic loading exceeded the strain under cyclic loads (Tables 3-9 and 3-10). This finding suggests that cyclic loading sufficiently captured the earthquake-type load effects and was more critical than dynamic loading for headed bar couplers. The only new information from dynamic testing that could be critical was the increase in the yield stress and ultimate stress, which was relatively small. 3.2.4 Experimental Studies and Results for Swaged Splices Ten swaged couplers splicing No. 10 (32 mm) bars at both ends were used as part of the experimental program, three each under static monotonic or cyclic loading, and four under dynamic loading. Unspliced bar samples of the same lot were also tested to determine the stress–strain characteristics of the reference bars per ASTM A370, and the data were used as a baseline to compare with the data from the spliced bar test. The rigidity of the coupler region was quantified by comparing the stress–strain relationships of the spliced and unspliced bars. The average measured yield and ultimate strength of the unspliced No. 10 bars used in the swaged couplers were 71.4 ksi and 101.8 ksi, respectively (Table 3-11).

Experimental Programs 115 (a) Static Monotonic Stress–Strain Curves (c) Dynamic Monotonic Stress–Strain Curves (b) Static Cyclic Stress–Strain Curves Figure 3-11. Effect of loading type on stress–strain relationships of headed bar splice specimens.

116 Proposed AASHTO Seismic Specifications for ABC Column Connections Figure 3-12. Effect of cyclic loading on stress–strain relationships of headed bar splice specimens. Figure 3-13. Effect of dynamic loading on stress–strain relationships of headed bar splice specimens. Specimen Yield Stress (ksi) Ultimate Stress (ksi) Ultimate Strain (in./in.) Fracture Strain (in./in.) SWT1 71.5 102.0 0.100 0.120 SWT2 71.8 101.7 0.103 0.110 SWT3 71.0 101.6 0.110 0.126 Average 71.4 101.8 0.104 0.126 Mill report 70.0 100.7 — — Note: SWT = control bars used in swaged coupler tensile tests. Table 3-11. Measured stress–strain properties of control bars used in swaged splices.

Experimental Programs 117 Figure 3-14 shows the test setup for a swaged bar specimen. The tests were conducted according to preliminary testing methods that were developed in the present study and then finalized as presented in Appendix B. Figure 3-15 shows the fractured spliced speci- mens after the tests. Shown in the figure are also the fracture stresses and strains. Note that all the spliced specimens failed away from the couplers as a result of the fracture of the bars and not within the coupler. Tables 3-12 to 3-14 list the measured yield stresses as well as the stresses and strains at the ultimate point (the point of maximum stress) for the swaged (SW) bar specimens tested under monotonic (SWM), cyclic (SWC), and dynamic (SWD) loading, respectively. The tables also indicate whether the failure was within the gauge length. 3.2.4.1 Failure Mode in Swaged Couplers All the swaged spliced specimens failed in a ductile manner outside the coupler with fairly consistent ultimate stresses and strains. This was true regardless of the loading type. The fracture location in all 10 specimens was in the bar away from the coupler ends. There was no coupler fracture. The shape of the stress–strain diagrams and the effect of loading type are discussed in subsequent sections. The bar stress at fracture was comparable for the specimens within each loading group. Figure 3-16 shows the stress–strain relationship for the spliced specimens and the average relationship for the control bars. The curves indicate that the initial stiffness, the yield stress, and the ultimate stress of the spliced and unspliced specimens were nearly the same, but the ultimate strain varied. The data are also listed in Table 3-12. The data for the three spliced specimens were consistent, except for the fracture strain in SWM3, which was higher than that in SWM1 and SWM2. The average ultimate strain in the spliced bars was 0.049, which was approximately one-half of the average ultimate strain in the control bars, owing to the rigidity of the couplers. Figure 3-14. Test setup for a swaged coupler specimen.

118 Proposed AASHTO Seismic Specifications for ABC Column Connections Figure 3-15. Fracture of spliced bars with swaged couplers. SWC2 - fu = 101.2 ksi - εu = 51,262 microstrainSWM1 - fu = 102.0 ksi - εu = 54,396 microstrain SWM3 - fu = 102.1 ksi - εu = 81,754 microstrain SWC1 - fu = 101.2 ksi - εu = 50,386 microstrain SWM2 - fu = 102.3 ksi - εu = 50,837 microstrain SWC3 - fu = 101.4 ksi - εu = 50,933 microstrain SWD1 - fu = 105.0 ksi - εu = 52,311 microstrain SWD2 - fu = 104.9 ksi - εu = 50,972 microstrain SWD3 - fu = 104.8 ksi - εu = 58,208 microstrain SWD4 - fu = 104.9 ksi - εu = 56,357 microstrain c) Fracture under dynamic monotonic loading b) Fracture under cyclic loading a) Fracture under static monotonic loading

Experimental Programs 119 Specimen Yield Stress (ksi) Ultimate Stress (ksi) Ultimate Strain (in./in.) Achieved Displacement Rate (in./s) Net Length Strain Rate (με/s) Failure Location SWD1 74.5 105.1 0.052 1.88 99,339 Within gauge length SWD2 74.3 104.9 0.045 1.88 99,183 Outside gauge length SWD3 74.4 104.8 0.049 1.81 95,650 Outside gauge length SWD4 74.3 104.9 0.049 1.87 99,015 Outside gauge length Average 74.4 104.9 0.049 1.86 98,296 — Table 3-14. Measured data for swaged splices tested under dynamic loads. Table 3-13. Measured data for swaged splices tested under cyclic loads. Specimen Yield Stress (ksi) Ultimate Stress (ksi) Ultimate Strain (in./in.) Failure Location SWC1 73.0 101.2 0.046 Outside gauge length SWC2 69.7 101.2 0.051 Within gauge length SWC3 70.0 101.4 0.048 Outside gauge length Average 70.9 101.3 0.048 — Table 3-12. Measured data for swaged splices tested under monotonic loads. Specimen Yield Stress (ksi) Ultimate Stress (ksi) Ultimate Strain (in./in.) Failure Location SWM1 71.7 102.0 0.048 Outside gauge length SWM2 71.8 102.3 0.045 Outside gauge length SWM3 72.0 101.6 0.053 Within gauge length Average 71.8 102.0 0.049 — Figure 3-16. Stress–strain relationships for swaged splices and control bars.

120 Proposed AASHTO Seismic Specifications for ABC Column Connections 3.2.4.2 Effect of Load Type on Headed Bar Couplers The acceptance criteria for spliced bars proposed in Appendix B include monotonic static, half-cyclic, and monotonic dynamic loading, with the latter two intended to evaluate the performance of the splices under earthquake-type loading. The fracture of the specimens with SW couplers was discussed in the previous section. The stress–strain diagrams for the specimens under different types of loading are shown in Figure 3-17. Several trends are evident. Some of the trends for swaged couplers are similar to those of the other two couplers. The yield stress under cyclic loading was the same as that under monotonic static loading. However, the yield and ultimate strengths increased slightly under dynamic loading. This is expected under dynamic loading of steel bars. The ultimate strains were comparable for all the specimens, regardless of the loading type, but the fracture strain in SW-3 exceeded that of the other two significantly. The effect of cyclic loading on the spliced bar stress–strain relationships is shown in Figure 3-18 by comparing the envelopes of the stress–strain diagrams under cyclic loading on the data for SWM3. The fracture points are marked by triangles in the figure. The data for SWM1, SWM2, and SWM3 were very close (Table 3-12), but SWM3 was selected because fracture in this specimen occurred within the gauge length. On the basis of the figure and com- parison of the data in Tables 3-12 and 3-13, it can be seen that the effect of cyclic loading was negligible, reducing the ultimate stress and strain by 2% or less. The effect of dynamic loading is shown in Figure 3-19 and can be quantified by comparing Tables 3-12 and 3-14. The fracture points are marked by triangles in the figure. The slight increase in the yield and ultimate stresses is evident in the plots and expected under dynamic loading. The yield and ultimate stresses increased by approximately 3%, but the ultimate strain remained the same. The average ultimate strain under dynamic loading was approximately the same as the strain under cyclic loads (Tables 3-13 and 3-14). This suggests that cyclic loading sufficiently captured the earthquake-type load effects. The only new information from dynamic testing that could be critical was the increase in the yield stress and ultimate stress, which was relatively small. 3.2.5 Coupler Rigid Length Ratios The coupler rigid length ratios (β) calculated on the basis of measured bar and coupler strains are listed in Table 3-15. The ratios were determined by using the method described in Appendix B. The ratios for GSM1 and GSM2 were excluded in calculating the average ratio for grouted sleeves because these specimens failed within the coupler. Note that grouted sleeve couplers that were tested did not meet the current acceptance criteria described in Appendix B because the fracture in GSM1 and GSM2 was in the coupler. A ratio of 0.55 means that bar forces were transferred over 45% of the length of the coupler, thereby making the remaining 55% behave essentially as a rigid segment. The values of β for headed bar couplers were negative because of the lower initial stiff- ness that shifted the stress–strain diagrams to the right in the HBM specimens (Figure 3-10). On the basis of the analytical study described in the next chapter of this report, a β of 0.5 is recommended as the minimum coupler rigid length ratio for headed bar couplers and for all other types of couplers. The average β for the swaged couplers was 0.89 with a recommended ratio of 0.9, indicat- ing that the bar forces were transferred to the coupler over 10% of the coupler length, thus

Experimental Programs 121 (a) Static Monotonic Stress–Strain Curves (c) Dynamic Monotonic Stress–Strain Curves (b) Static Cyclic Stress–Strain Curves Figure 3-17. Effect of loading type on stress–strain relationships of swaged splice specimens.

122 Proposed AASHTO Seismic Specifications for ABC Column Connections Figure 3-18. Effect of cyclic loading on stress–strain relationships of swaged splice specimens. Figure 3-19. Effect of dynamic loading on stress–strain relationships of swaged splice specimens. making the remaining 90% of the coupler length rigid. These ratios should be interpreted carefully. The higher β in swaged couplers should be viewed in light of the fact that swaged couplers are significantly shorter than grouted sleeve couplers. 3.2.6 Strain Rate for Dynamic Loading of Spliced Bars Dynamic testing of spliced bars was required as part of the preliminary coupler testing methods proposed in this project, as follows: “Load the spliced specimen with a strain rate of 75,000 micro-strain/sec (7.5%/sec) to fracture. A deviation of up to 33% between the target and the achieved strain or displacement rates was allowed.” The recommended strain rate was based on engineering judgement and the rate used in previous studies. The measured strain rates for different types of splices are listed in Table 3-16. It can be seen that the actual rates varied slightly between the couplers within each type but more significantly

Experimental Programs 123 β Sample Failure Type Measured Recommended GSM1 In coupler 0.64 0.55 GSM2 In coupler 0.59 GSM3 Outside coupler 0.51 Average — 0.51 HBM1 Outside coupler –0.42 0.5HBM2 Outside coupler –0.89 Average — –0.65 SWM1 Outside coupler 0.85 0.9 SWM2 Outside coupler 0.97 SWM3 Outside coupler 0.84 Average — 0.89 Table 3-15. Measured and recommended coupler rigid length ratios. Specimen Strain Rate (με/s) Specimen Strain Rate (με/s) Specimen Strain Rate (με/s) GSD1 72,853 HBD1 88,618 SWD1 99,339 GSD2 72,015 HBD2 89,907 SWD2 99,183 GSD3 72,433 HBD3 90,094 SWD3 95,650 GSD4 70,826 HBD4 90,234 SWD4 99,015 Average GS 72,032 Average HB 89,713 Average SW 98,296 Table 3-16. Measured dynamic strain rates for spliced bars. between different types. Nonetheless, they were all within the 33% deviation that was recom- mended. The testing machine at the University of Nevada, Reno, is one of the most complete testing systems, with a state-of-the-art control system, and is operated by very knowledgeable technicians. The fact that the target and achieved strain rates were different and varied between different types of couplers for No. 10 (32 mm) bars is an indicator of the limits of precise con- trol of loading in dynamic testing, an issue that the research team was aware of in developing the preliminary acceptance criteria for couplers. Further investigation of realistic strain rates in earthquake testing conducted as part of the current study led to the revision of the target strain rates in dynamic loading, as presented in Appendix B. A discussion of this issue is also presented in the next chapter. 3.2.7 Summary of Mechanical Bar Splice Testing Three types of mechanical bar splices were tested in the present study to fill the knowledge gaps detailed in Table 3-1 and to identify the necessary refinement of the proposed testing method. By using the preliminary testing methods proposed in this study for couplers, 10 specimens were tested per coupler type: three under monotonic loading, three under cyclic loading, and four under dynamic loading. After all the test data were collected, the proposed testing method was further refined and finalized in the ASTM/AASHTO format (Appendix B). The appendix

124 Proposed AASHTO Seismic Specifications for ABC Column Connections also includes acceptance criteria and a material model for couplers. Overall, it was found that consistent and reliable measurement could be obtained with the proposed testing method. Furthermore, the proposed criteria were sufficient to comment on the suitability of a bar coupler type for use in the plastic hinge region of bridge columns. 3.3 Mechanically Spliced Bridge Columns 3.3.1 Introduction A half-scale bridge column model incorporating No. 10 (32 mm) grouted sleeve couplers in the plastic hinge region was tested to generate information on the cyclic response of columns with longitudinal bar sizes that represent field application. The testing also helped determine the need for any refinement of the column testing protocol for qualification of different types of couplers that might emerge. This section describes the column model, instrumentation, test procedure, and key test results. 3.3.2 Column Model The test column was labeled GC10, which stands for grouted sleeve with No. 10 (32 mm) bars. Figure 3-20 shows the details of the column. Figures 3-21 to 3-27 show different Figure 3-20. Details of GC10 column model with grouted sleeves.

Experimental Programs 125 Figure 3-22. Column cage after installation of grouted sleeves. Figure 3-23. Transportation of precast column. Figure 3-24. Lowering of column onto footing dowels. Figure 3-21. Column reinforcement cage.

126 Proposed AASHTO Seismic Specifications for ABC Column Connections Figure 3-25. Close-up of lowering column onto footing dowels. Figure 3-26. Completed grouting of sleeves. construction stages, including the completed column. The specified concrete compressive strength was 4.5 ksi. The reinforcement bars were all of ASTM 706 Grade 60 steel. Standard 6- by 12-in. cylinders were tested to determine the concrete compressive strength. The com- pressive strength of the SS Mortar used to fill the sleeve was determined by using standard 2-in. cube samples. The compressive strength of the concrete and SS Mortar on the 7th day, 28th day, and test day are listed in Table 3-17. The reinforcement steel properties are listed in Table 3-18. The material properties in GC10 met or exceeded the specified strengths.

Experimental Programs 127 Figure 3-27. Completed GC10 column. Material and Element Specified Strength (ksi) Measured Strength (ksi) 7 days 28 days Test Day Conventional concrete Footing 4.5 3.8 5.1 6.5 Column and head 4.5 4.0 5.4 6.1 Grout SS Mortar 13.9–18.2a 11.4 12.4 17.3 aDepending on cure temperature. Table 3-17. Specified and measured material properties for concrete and SS Mortar used in GC10. Bar Size Yield Stress (ksi) Ultimate Stress (ksi) Ultimate Strain (in./in.) Fracture Strain (in./in.) No. 10 76.46 105.69 0.107 0.149 No. 3 74.50 103.0 0.106 0.194 Table 3-18. Measured material properties for reinforcing steel bars used in GC10.

128 Proposed AASHTO Seismic Specifications for ABC Column Connections Figure 3-28. Test setup for GC10 column model. The test setup is shown in Figure 3-28. The geometry, dimensions, test setup, and loading history for this column were the same as those in cyclic testing of eight other column models tested previously at the University of Nevada, Reno (Haber et al. 2013; Tazarv and Saiidi 2014). Included in those tests were a cast-in-place (CIP) column built by using conventional construc- tion methods and an accelerated bridge construction (ABC) column with grouted sleeve splices in the plastic hinge with No. 8 (25-mm) longitudinal bars. No. 10 (32 mm) bars were used for the column longitudinal bars in the current project. However, the longitudinal reinforcement ratio of all the columns was nominally the same, at approximately 2%. The column model was secured to the structural floor and tested as a cantilever subjected to cyclic horizontal forces by using a hydraulic actuator. A reaction wall was constructed by using a post-tensioned concrete block system. A nominally constant axial load of 200 kips, correspond- ing to an axial load index (the ratio of the axial load and the product of the gross cross-sectional area and the specified compressive strength of concrete) of 0.10 was applied to the column with two vertical jacks through a transfer steel beam. The hydraulic system for these jacks was con- nected to an accumulator to control and minimize fluctuation of the axial load as the column underwent horizontal displacement. The column model was instrumented with an array of 44 strain gauges and various displacement transducers to measure lateral displacement and local deformation of the plastic hinge (Figures 3-29 and 3-30). The gauges were installed on critical locations of the column longitudinal and transverse reinforcement, slightly below the top of the footing and in the plastic hinge zone. The target loading history is shown in Figure 3-31. The high-amplitude load cycles shown in the figure were not applied because the column failed at a lower drift level, as explained in subsequent sections. The column was loaded under drift control, with increasing drift levels. For columns in which the plastic hinge detail varies from that of conventional CIP columns, the drift ratio rather than the displacement ductility ratio is a better indicator of the behavior, because the drift ratio is independent of the yield displacement, which could be affected by the plastic hinge detail. The loading was continued until the column failed. 3.3.3 Test Results for GC10 Column Model Key tests results on the column apparent damage, force-displacement relationships, and strains are presented in this section.

Experimental Programs 129 Figure 3-29. Strain gauge layout in GC10 column model. 3.3.3.1 Observed Damage and Strains in GC10 Column Model A summary of the observed damage in and near the plastic hinge zone under different drift levels is presented in Tables 3-19 and 3-20. Figures 3-32 to 3-39 show the damage under dif- ferent drift ratios. Figure 3-40 shows the overview of the column under maximum drift. The dark line on the column surface above the grout vents indicates the location of the top of the grouted couplers. It can be seen in Table 3-19 and Figures 3-32 and 3-33 that apparent damage under small drift ratios was mostly flexural cracking spread over and beyond the coupler region in the column. This is the type of damage normally seen in cyclic testing of conventional reinforced concrete columns that meet current seismic design codes. Spalling of concrete under 3% drift ratio is also typical of damage in well-designed CIP columns. The minor vertical cracks that were formed under 4% drift ratio were over the grouted couplers because the concrete cover in those areas was lower than elsewhere. These cracks are insignificant and might be due to the fact that the column model was of half-scale, with reduced concrete cover as compared with the prototype column cover. Under the second cycle loading toward +5% drift ratio, one of the couplers fractured, as indicated by a significant drop in the lateral load, which is discussed

130 Proposed AASHTO Seismic Specifications for ABC Column Connections Figure 3-30. Displacement transducer layout in GC10 column model. in the next section. The strain data for the bar with the fractured coupler also dropped and remained the same, even as the drift ratio was increased. This can be seen in the part of the strain plot marked by “A” in Figure 3-41. The plot shows that the residual strain that had developed in the bar as a result of yielding prior to the coupler fracture remained unchanged. The fractured coupler could not be seen at this stage because the concrete cover blocked the view. Also, under 5% drift ratio, substantial spalling of concrete that exposed the transverse reinforcement was seen. Cyclic loading was continued despite the coupler fracture. No other couplers fractured in subsequent cycles. A portion of the couplers was visible under 6% drift ratio, but not the grouted coupler fracture location. The flexural concrete crack above the coupler became substantially wider under 6% drift. The location of the fractured coupler became visible under 8% drift ratio (Figure 3-39). The fracture occurred at midheight of the coupler, which was consistent with the fracture location in the four grouted couplers that had failed in the spliced bar tests reported earlier.

Experimental Programs 131 D ri ft (% ) D is pl ac em en t ( in ) Cycle Number Figure 3-31. Target loading history for GC10 column model. Drift Ratio (%) Observed Damage +0.25 Minor vertical and flexural cracks –0.25 Minor flexural cracks +0.50 Vertical, flexural and inclined cracks Cracks in column –0.50 Flexural cracks along top of coupler +0.75 Inclined cracks –0.75 Flexural cracks +1.00 Flexural and inclined cracks –1.00 Flexural and inclined cracks +2.00 Inclined cracks in coupler region Widening of cracks along top of coupler Widening of cracks at column base Minor cracks in top surface of footing –2.00 Cracks at column base Vertical and flexural cracks New inclined cracks Note: Positive drifts were based on displacements away from the reaction wall (east to west). Table 3-19. Summary of damage in GC10 at up to 2% drift ratio.

132 Proposed AASHTO Seismic Specifications for ABC Column Connections Drift Ratio (%) Observed Damage +3.00 Widening of cracks above coupler region Initiation of spalling of concrete in coupler region on west face of column –3.00 Initiation of spalling of concrete in coupler region on east face of column +4.00 Vertical cracks in coupler region Extensive spalling in coupler region on east and west faces of column Spalling of concrete on top of coupler region Multiple transverse bars visible on west face of column Widening of flexural cracks along coupler region –4.00 Widening of flexural cracks +5.00 Fracture of one coupler Multiple transverse bars visible on east face of column More minor cracks in footing Wide gap between column and footing –5.00 Extensive spalling of concrete on east face of column Wide gap between column and footing +6.00 Coupler visible at lower part of column (east face) More transverse bars visible on east face of column –6.00 No further damage +8.00 Location of coupler failure visible –8.00 No further damage Note: Positive drifts were based on displacements away from the reaction wall east to west). Table 3-20. Summary of damage in GC10 at 3% drift ratio and higher. (a) Southeast face (b) Northwest face Figure 3-32. Damage observed at second cycle of 1.0% drift.

Experimental Programs 133 (a) Southeast face (b) Northeast face Figure 3-33. Damage observed at second cycle of 2.0% drift. a) Southeast face (b) Northwest face Figure 3-34. Damage observed at second cycle of 3.0% drift.

134 Proposed AASHTO Seismic Specifications for ABC Column Connections (a) Southeast face (b) Northwest face Figure 3-35. Damage observed at second cycle of 4.0% drift. (a) Southeast face (b) Northwest face Figure 3-36. Damage observed at second cycle of 5.0% drift.

Experimental Programs 135 (a) Southeast face (b) Northwest face Figure 3-37. Damage observed at second cycle of 6.0% drift. (a) East face (b) West face Figure 3-38. Damage observed at second cycle of 8.0% drift.

136 Proposed AASHTO Seismic Specifications for ABC Column Connections Figure 3-39. Fractured coupler as seen under 8% drift ratio (fracture occurred under 5% drift). Figure 3-40. Overview of column model under maximum drift. Figure 3-41. Strain data for the bar with fractured coupler.

Experimental Programs 137 The maximum longitudinal bar and coupler strains under low drift ratios of up to 2% and high drift ratios of 3% or more are plotted in Figures 3-42 and 3-43. Under low drift ratios, the strain increased gradually with an expected trend. The strain data at approximately 10 in. above the footing were obtained from gauges installed on the surface of the grouted sleeve couplers at midheight. It can be noted that this strain was always lower than the bar strain within the footing and above the coupler. The difference was more pronounced under 2% drift ratio. The smaller strain in the coupler was due to the larger cross-sectional area of the couplers. The figure also marks the measured yield strain of the bars. None of the bars yielded under low drift Figure 3-42. Longitudinal bar and coupler strain profile for GC10 column model under drifts of 2% or less. Figure 3-43. Longitudinal bar and coupler strain profile for GC10 column model under drifts of 3% or more.

138 Proposed AASHTO Seismic Specifications for ABC Column Connections levels. In the reference CIP column, the longitudinal bars yielded under 1% drift ratio. The data for GC10 indicate that rigidity of the coupler inhibited yielding in the plastic hinge zone. The strain in all the bars outside the coupler yielded at 3% and the strain continued to increase more visibility in the footing (Figure 3-43). The strain in the coupler remained unchanged as the drift ratio increased because the column had yielded and lateral force did not increase significantly with an increase in the drift ratio. Again, the effect of the coupler rigidity on the plastic hinge strain is evident. The maximum transverse bar tensile strain in the column plastic hinge region under dif- ferent drift ratios is shown in Figure 3-44. This strain is influenced by column shear forces or the core confinement mechanism. The overall trend is that the strain increases with drift. However, a reduction in strain was observed as the drift ratio increased from 5% to 6%. Spiral strains are sometime influenced by new cracks that form near but away from the gauge loca- tion, thus reducing the strain in that location. The measured yield strain for No. 3 (10-mm) bars that were used in the spirals was 2,570 µe. It can be seen that the spirals did not yield until the drift ratio reached 7%. This means that the spirals continued to confine the core, even under high drifts. This trend in spiral strains is typical of what is seen in CIP column responses. 3.3.3.2 Lateral Force-Displacement Relationship for GC10 Column Model The measured lateral force-drift relationship for the column model is shown in Figure 3-45. It is clear that up to the first cycle of 5% drift ratio, the hysteretic loops were stable and wide, which indicates good energy dissipation. As the model was pushed to the second cycle of +5% drift ratio, one of the couplers on the tension side of the column fractured, as indicated by the sharp drop marked by “A” in the figure. This led to a significant reduction of the lateral load in the positive displacement zone. There was no fracture in any other coupler. The lateral load for the negative drift side remained stable. The asymmetric behavior is due to the fact that the pattern of the longitudinal reinforcement was not symmetric (Figure 3-20). The drift capacity of a similar column with No. 8 (25 mm) longitudinal bars and grouted sleeve couplers (GCNP) was 6%, but fracture in that test model occurred in the bars below the coupler in the footing as a result of stress concentration. The envelopes of the hysteresis curves in the positive and negative displacement regions and the average of the two are shown in Figure 3-46. Both the positive and negative displacement Figure 3-44. Maximum transverse bar strains for GC10 column model under different drift ratios.

Experimental Programs 139 Figure 3-46. Force-drift envelope and idealized curve for GC10 column model. Figure 3-45. Measured force-drift relationship for GC10 column model. zones were considered, because failure occurred during loading in one direction. Also shown is the bilinear idealization of the average envelope calculated according to AASHTO (2014). The elastic part was determined by passing a line from the origin to the point on the envelope associated with the first longitudinal bar yielding. The plastic segment was obtained by pres- ervation of energy (maintaining the same areas under the envelope and the idealized curve after the yield point). The average displacement ductility capacity for the two directions of GC10 was 5. The displacement ductility capacity of GCNP with No. 8 (25 mm) longitudinal bars was 4.5. To demonstrate the effects of using No. 10 (32 mm) longitudinal bars versus No. 8 (25 mm) bars and the effect of using grouted couplers, the research team superimposed the average envelope of the measured force–displacement relationship for GC10 on the average enve- lopes for CIP and GCNP. The concrete and steel properties measured for the three specimens were different. Therefore, to allow for comparison of the results, the pushover curves for the three test models were normalized relative to the peak lateral force in each. Figure 3-47 shows the superimposed envelopes. Unconfined concrete spalled in CIP at approximately 2.5% drift, leading to a slight drop in the lateral force. The same did not occur in GCNP and GC10 because

140 Proposed AASHTO Seismic Specifications for ABC Column Connections Figure 3-47. Normalized pushover response for cast-in-place, GCNP, and GC10 column models. the large cross-sectional area of grouted couplers reduced the compressive forces on concrete in the plastic hinge. The response of GC10 and GCNP was nearly the same up to 5% drift ratio. The fracture of one of the couplers in GC10 led to a drop in the lateral load capacity afterward. Failure in GCNP did not occur in the coupler. Rather, the spliced bars fractured inside the footing under 6% drift ratio. The CIP drift ratio capacity was 10%. 3.3.4 Summary of Mechanically Spliced Column Testing A half-scale bridge column utilizing grouted couplers in the plastic hinge zone was tested to failure to fill the knowledge gaps detailed in Table 3-1. The data for a conventional CIP column were used to evaluate the effect of the couplers on the local and global response of the bridge column. Furthermore, the measured data were used to verify the proposed design methods for mechanically spliced columns (see Chapter 4). The proposed design methods for spliced columns in the AASHTO format is presented in Appendix C. It was found that long and rigid couplers might reduce the column displacement capacity and displacement ductility capacity by 20% and 30%, respectively. Further, the current testing methods for reinforced concrete columns [e.g., ACI 374.2R-13 (ACI Committee 374 2013)] were found sufficient to experimentally establish the seismic performance of mechanically spliced bridge columns. 3.4 Grouted Duct Connections 3.4.1 Introduction Past research on grouted duct connections has generated a great deal of data, mostly for No. 8 (25 mm) bars or smaller, as reviewed in Chapter 2. On the basis of the data collected from the literature, the present study developed a preliminary design equation for anchorage of steel bars in grouted ducts. Additional testing on large bars was deemed necessary to verify and, if necessary, revise the proposed equation (Table 3-1). No. 10 (32 mm) bars were used in the present experimental study on grouted duct connections, since No. 10 and No. 11 bars are commonly uses as longitudinal bars in bridge columns. The proposed preliminary design equation was further refined after testing of the grouted duct specimens and finalized as detailed in Appendix C.

Experimental Programs 141 Group No. and Sp. ID Bar Size Embedment Length, in. (ratio) Bundled Duct Inner Diameter (in.) Steel Gauge Test Variable Short Anchorage G1 1 No. 10 15.5 (0.5lag) No 4 26 Reference case 2 No. 10 15.5 (0.5lag) No 5.26 26 Duct diameter 3 No. 10 15.5 (0.5lag) No 4 24 Duct thickness 4 No. 10 15.5 (0.5lag) No 4 26 Bar eccentricity G2 5 No. 10 24 (0.5lag) Yes 5.26 26 Two bundled bars 6 No. 10 24 (0.5lag) Yes 5.26 26 Two bundled bars, bar eccentricity Full Anchorage G3 7 No. 10 31 (1.0lag) No 4 26 Reference case 8 No. 10 31 (1.0lag) No 5.26 26 Duct diameter 9 No. 10 31 (1.0lag) No 4 24 Duct thickness 10 No. 10 31 (1.0lag) No 4 26 Bar eccentricity G4 11 No. 10 48 (1.0lag) Yes 5.26 26 Two bundled bars 12 No. 10 48 (1.0lag) Yes 5.26 26 Two bundled bars, bar eccentricity Note: Sp. = specimen. Table 3-21. Test Matrix for pullout tests of grouted duct connections. 3.4.2 Grouted Duct Test Models Twelve specimens were tested to generate data for No. 10 (32 mm) bars that were anchored in grouted ducts of different diameters and thicknesses, some with eccentricities. The test param- eters are listed in Table 3-1, and the grouted duct test matrix is presented in Table 3-21. Note that the ratios shown for bundled bars are based on the equivalent diameter of two bundled bars. All steel bars anchored in the grouted duct specimens were ASTM A706 Grade 60. The test models listed in Table 3-21 were placed in four groups, G1 to G4, depending on the number of bars and the anchorage length in the ducts. G1 and G3 included single bars with short and full anchorage, respectively (Figure 3-48). G2 and G4 each had two bundled bars, which were bundled with short and full anchorage, respectively (Figure 3-49). The anchor- age length in G3 and G4 was according to preliminary equations developed in this project by using data from the literature, whereas the anchorage length in G1 and G2 was one-half of the required length. The potential failure mode in G1 and G2 was bond failure. Figure 3-50 shows the formwork, steel cage, and corrugated ducts placed in the forms prior to casting of the con- crete. The completed specimens are shown in Figure 3-51. The models were tested under tensile load to failure. The pull force and slippage at the surface of the models were continuously measured until failure. The data from these groups were used to assess the effect of different parameters on the bond strength. The expected failure mode in G3 and G4 was bar fracture outside the connection. The test data were used to determine if and how the proposed anchorage length equations should be modified. The final version of the proposed equation is presented in Appendix C.

142 Proposed AASHTO Seismic Specifications for ABC Column Connections Figure 3-48. Pullout grouted duct test specimen detail for single bars (Groups G1 and G3). 3.4.3 Material Properties, Instrumentation, and Testing Procedure for Grouted Duct Connections Normal weight concrete with a specified 28-day compressive strength of 4.5 ksi was used in the specimens outside the grouted ducts. Three standard cylindrical samples a diameter of 6 in. and height of 12 in. were tested to determine the actual concrete compressive strength. The measured 28-day concrete compressive strength was 5.11 ksi. The testing took place at approximately 5 months after casting. The measured compressive strength of concrete on test days was 6.56 ksi. A nonshrinkage, mineral-based, high-strength grout was used in the corrugated ducts. The specified 28-day compressive strength of this grout is 9.0 ksi. Three standard cube samples with

Experimental Programs 143 Figure 3-49. Pullout grouted duct test specimen detailing for bundled bars (Groups G2 and G4). a side dimension of 2 in. were tested to determine the actual compressive strength of the grout. The measured 28-day compressive strength for the grout was 8.98 ksi. The measured compres- sive strength of the grout on the test dates was 9.05 ksi and 12.0 ksi for Specimens 1 and 2–12, respectively. ASTM A706 Grade 60 No. 10 (32 mm) bars were anchored in the grouted ducts. The measured yield stress and ultimate strength were 76.5 ksi and 105.7 ksi, respectively. The measured ultimate strain was 0.107 in./in. and the fracture strain was 0.149 in./in. The corrugated galvanized steel ducts conformed to ASTM A653. The wall thickness of the galvanized ducts was either 24-gauge (0.025 in.) or 26-gauge (0.018 in.).

144 Proposed AASHTO Seismic Specifications for ABC Column Connections Figure 3-50. Formwork prior to casting concrete for grouted duct specimens. (a) G1 (b) G2 (c) G3 (d) G4 The ratios of the required anchorage lengths (0.5 or 1.0) listed in Table 3-21 were based on the specified material properties. When the actual material properties are used, the ratios change to 0.55 and 1.1 for single bars, respectively, and 0.57 and 1.14 for bundled bars, respectively. In addition to the measurement of the pullout force, strain data on the bars and the ducts were recorded with strain gauges. Figures 3-52 and 3-53 show the location of the strain gauges on different specimens. Two gauges were attached on opposite faces of the single bars. In addition, two gauges were installed on opposite faces of the ducts. Figures 3-54 and 3-55 show a sketch and a photo of the test setup, respectively. A Franken ram with a capacity of 300 kips was used in the tests. The bars were anchored above the ram by shear screw couplers. The need for threading the bar ends to provide anchorage and the risk of bar fracture at the threads were eliminated by using these couplers. The couplers proved to be effective, in that there was no failure at the coupler location. The loads were applied at a rate of 0.05 in./s until one of the three expected modes of failure occurred: (1) bar fracture, (2) failure

Experimental Programs 145 (a) G1 (b) G2 (c) G3 (d) G4 Figure 3-51. Completed blocks for grouted duct specimens. Figure 3-52. Instrumentation details for grouted duct specimens. (a) Specimens 1, 2, 3, 7, 8, and 9 (b) Specimens 4 and 10

146 Proposed AASHTO Seismic Specifications for ABC Column Connections Figure 3-53. Instrumentation details for grouted duct specimens. (a) Specimens 5 and 11 (b) Specimens 6 and 12 Figure 3-54. Front view of grouted duct test setup.

Experimental Programs 147 Figure 3-55. Photograph of grouted duct test setup. of the bar–grout bond, or (3) failure of the duct–concrete bond. The test method was consistent with ASTM E488. 3.4.4 Results of Grouted Duct Connection Tests A summary of the failure modes for different specimens is listed in Table 3-22. Figures 3-56 to 3-59 show the specimens after failure. Note that the anchorage length in G1 and G2 was one-half of the length required by the preliminary design equations, whereas the full anchorage length was provided in G3 and G4. The anchorage lengths had been based on the specified compressive strength of 4.5 ksi for concrete and 9 ksi for the grout. The actual concrete strengths were 6.56 ksi and 12.0 ksi (except for SP1, in which the compressive strength was 9.05 ksi) for the concrete and grout, respectively. It can be seen in Table 3-22 that none of the specimens exhibited bond failure in the interface of bars and grout, even SP1, in which the grout strength was lower than that of the others (9.05 ksi versus 12.0 ksi). Recall that single bars were anchored in G1 but two bundled bars were anchored in G2. Failure of the bond between the duct and the surrounding concrete was observed in SP4 to SP6. The common feature of these specimens was that the bars were eccentric relative to the ducts. The bond failure occurred regardless of the number of bars. The duct pullout can be seen in Figures 3-56d and 3-57. Table 3-22 indicates that the bars were fully developed in G3 and G4, regardless of eccen- tricity, duct diameter, duct thickness, and the number of bars in the duct. Figures 3-58 and 3-59 show that surface cracks were observed in the grout in G3 and G4, but there was no bar or duct pullout. The measured maximum forces and displacements for different specimens are listed in Table 3-23. Recall that no bar pullout from the grout in the ducts was observed, but SP4, SP5,

148 Proposed AASHTO Seismic Specifications for ABC Column Connections Table 3-22. Mode of failure in grouted duct specimens. Sp. ID Mode of Failure G1 1 Bar rupture 2 Bar rupture 3 Bar rupture 4 Duct pullout G2 5 Duct pullout 6 Duct pullout G3 7 Bar rupture 8 Bar rupture 9 Bar rupture 10 Bar rupture G4 11 Bar rupture 12 Bar rupture and SP6 failed as a result of duct pullout (Table 3-22). The peak force was nearly the same in all the specimens in which the bar(s) fractured outside the grouted duct connections. This was observed within three groups: G1 (SP1 and SP3), G3 (SP7–SP10), and G4 (SP11 and SP12). The failure displacements listed in the table are displacements of the bar at the loading plate (Figure 3-54) when either the bar fractured or the duct pulled out. The distance from the top surface of the concrete to the plate was 55 in. Most of the deformation in the specimens in which the bar fractured was in the unanchored part of the bar (over the 55-in. length). The fracture strain of the bars was 0.149. Therefore, the estimated displacement due to the bar strain alone was approximately 8.2 in. The measured maximum displacements listed in the table were smaller in SP4, SP5, and SP6 because the bars did not fracture, even though they yielded (the yield force was 97.2 kips in one bar and 194.3 kips in two bars). The measured maximum duct strain was relatively small, ranging from 200 to 1,900 µe, with the average being 1,000 µe. 3.4.5 Summary of Grouted Duct Connection Testing Twelve grouted duct specimens filled with conventional grout were monotonically tested to failure to fill the knowledge gaps detailed in Table 3-1. The test results confirmed that the pre- liminary proposed design equation for grouted ducts was reliable, and all the specimens with full anchorage failed due to the bar fracture. The additional data collected in this study were used to further refine the proposed design equation. The next chapter discusses the methodology used to develop the grouted duct design equation and verification. The proposed design method for grouted duct connections in the AASHTO format is presented in Appendix C.

Experimental Programs 149 Figure 3-56. Damage in Group G1 of grouted duct specimens. (a) G1-1 (b) G1-2 (c) G1-3 (d) G1-4 (a) G2-5 (b) G2-6 Figure 3-57. Damage in Group G2 of grouted duct specimens.

150 Proposed AASHTO Seismic Specifications for ABC Column Connections (a) G3-7 (b) G3-8 (c) G3-9 (d) G3-10 Figure 3-58. Damage in Group G3 of grouted duct specimens. (a) G4-11 (b) G4-12 Figure 3-59. Damage in Group G4 of grouted duct specimens.

Experimental Programs 151 Table 3-23. Measured peak loads and displacements at failure of grouted duct specimens. Sp. ID Number of Bars Load, F (kips) Displacement (in.) G1 1 1 134.9 8.59 2 1 134.5* 7.85 3 1 134 7.78 4 1 124 5.81 G2 5 2 198.6 2.87 6 2 242.5 3.8 G3 7 1 133.5 7.52 8 1 136.2 7.83 9 1 133.6 7.87 10 1 133.5 8.09 G4 11 2 268 8.58 12 2 269.8 8.35 *Estimated on basis of single bar fracture in other specimens. The force could not be measured due to improper setting of a valve. 3.5 Testing of Other Types of Connections Considering the bulk of available experimental data on the seismic response of columns with grouted duct connections, pocket/socket connections, and the analytical studies presented in Chapter 4, no new testing was deemed necessary for precast columns incorporating grouted duct or pocket/socket connections. 3.6 References AASHTO. (2014). AASHTO Guide Specifications for LRFD Seismic Bridge Design. American Association of State Highway and Transportation Officials, Washington, D.C. 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, Mich. Haber, Z. B., Saiidi, M. S., and Sanders, D. H. (2013). Precast Column-Footing Connections for Accelerated Bridge Construction in Seismic Zones. CCEER-13-08. Center for Civil Engineering Earthquake Research, Department of Civil and Environmental Engineering, University of Nevada, Reno. Tazarv, M., and Saiidi, M. S. (2014). Next Generation of Bridge Columns for Accelerated Bridge Construction in High Seismic Zones. CCEER-14-06. Center for Civil Engineering Earthquake Research, Department of Civil and Environmental Engineering, University of Nevada, Reno. Tazarv, M., and Saiidi, M. S. (2016). Seismic Design of Bridge Columns Incorporating Mechanical Bar Splices in Plastic Hinge Regions. Engineering Structures, Vol. 124, pp. 507–520.

Next: Chapter 4 - Analytical Programs »
Proposed AASHTO Seismic Specifications for ABC Column Connections Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Accelerated bridge construction (ABC) utilizes rigorous planning, new technologies, and improved methods to expedite construction. Prefabricated columns and their connections to adjoining bridge members (cap beams, footings, pile caps, and pile shafts) are the most critical components of ABC in moderate- and high-seismic regions.

The TRB National Cooperative Highway Research Program's NCHRP Research Report 935: Proposed AASHTO Seismic Specifications for ABC Column Connections develops AASHTO specifications for three types of precast column connections to facilitate ABC implementation in moderate- and high-seismic regions.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!