Project ID: 2020-UCD-01 Grant # 69A3551947137

Fiber Reinforced Polymer (FRP) Seismic Retrofit of Reinforced

Concrete Bridge Columns Vulnerable to Long-Duration Subduction
Zone Earthquakes

By

William Nickelson
Ruben Jerves
Christopher Motter, co-PI
Adam Phillips, co-PI

Department of Civil and Environmental Engineering

Washington State University
Pullman, WA 99164

for

National University Transportation Center TriDurLE

Department of Civil & Environmental Engineering
405 Spokane Street PO Box 642910
Washington State University Pullman, WA 99164-2910

May 2022
i
 Acknowledgments

This report is based on a thesis by William Nickelson in partial fulfillment of the requirements

for a M.S. degree from Washington State University. This research was sponsored by the

National Center for Transportation Infrastructure Durability & Life-Extension at Washington

State University and by the U.S. Department of Transportation. This support is gratefully

acknowledged. Simpson Strong Tie (SST) is thanked for providing in-kind donations of fiber

reinforced polymer and resin, in addition to installation of the steel jackets used in the test

columns. Nucor is thanked for in-kind donations of Grade 40 reinforcement used in the test

columns. Scott Lewis, Joshah Jennings, Levi Arnold, and Roshan Ghimire are thanked for

assistance with laboratory testing.

Disclaimer

The contents of this report reflect the views of the authors, who are responsible for the facts and

the accuracy of the information presented. This document is disseminated under the sponsorship

of the Department of Transportation, University Transportation Centers Program, in the interest

of information exchange. The U.S. Government assumes no liability for the contents or use

thereof.

ii
 Table of Contents

Acknowledgments........................................................................................................................... ii

Disclaimer ....................................................................................................................................... ii

Table of Contents ........................................................................................................................... iii

List of Figures ................................................................................................................................. v

List of Tables ................................................................................................................................. xi

Executive Summary ........................................................................................................................ 1

1. Introduction ................................................................................................................................. 3

1.1. Problem Statement ............................................................................................................... 3

1.2. Objectives ............................................................................................................................ 5

1.3. Expected Contributions ........................................................................................................ 6

1.4. Report Overview .................................................................................................................. 7

2. Literature Review........................................................................................................................ 8

3. Column Testing......................................................................................................................... 12

3.1. Methodology ...................................................................................................................... 12

3.1.1. Test Specimen Details................................................................................................. 12

3.1.2. Test Specimen Construction ....................................................................................... 16

3.1.3. Material Properties ...................................................................................................... 20

3.1.4. Test Set-Up ................................................................................................................. 24

3.1.5. Instrumentation ........................................................................................................... 27

iii
 3.1.6. Loading Protocol ......................................................................................................... 30

3.2. Results and Discussion ...................................................................................................... 35

3.2.1. Observed Damage ....................................................................................................... 35

3.2.2. Load-Deformation....................................................................................................... 52

3.2.3. Backbone Modeling .................................................................................................... 57

3.2.4. Effective Secant Stiffness ........................................................................................... 66

3.2.5. Reinforcement Strain .................................................................................................. 67

3.2.6. Column Curvature ....................................................................................................... 72

3.2.7. Shear Sliding and Base Rotation................................................................................. 75

3.2.8. Components of Deformation....................................................................................... 78

4. Column Modeling ..................................................................................................................... 81

4.1. Methodology ...................................................................................................................... 81

4.2. Results and Discussion ...................................................................................................... 88

5. Conclusions ............................................................................................................................... 90

References ..................................................................................................................................... 93

iv
 List of Figures

Figure 3.1: Column Reinforcement Layout .................................................................................. 14

Figure 3.2: General Footing Reinforcement Details (10-#7 Starter Base Depicted) (Dimensions

in Inches)....................................................................................................................................... 15

Figure 3.3: Footing Sleeve Details (10-#7 Starter Bars Depicted) (Dimension in Inches) .......... 16

Figure 3.4: CFRP Jacket Before Testing for C(CFRP)-4.0-#7(1.3)-0.05 ..................................... 19

Figure 3.5: Footing Construction .................................................................................................. 19

Figure 3.6: Construction: a) Bracing for Column Concrete Pour, and b) Bracing for Top Block

Concrete Pour................................................................................................................................ 20

Figure 3.7: Longitudinal Steel Reinforcement Testing: a) #5 Stress-Strain Relationship, b) #7

Stress-Strain Relationship ............................................................................................................. 21

Figure 3.8: Test Set-up, Pd/(Agf’c) = 0.05: a) Schematic, b) Photo.............................................. 26

Figure 3.9: Axial Load Setup, Pd/(Agf’c) = 0.05 .......................................................................... 26

Figure 3.10: Instrumented Starter Bar Layout .............................................................................. 28

Figure 3.11: Strain Gauge Reinforcement Layout (Dimensions in Inches).................................. 29

Figure 3.12: Stationary Reference Measurements ........................................................................ 29

Figure 3.13: LVDT Instrumentation Layout (Dimensions in Inches) .......................................... 30

Figure 3.14: Standard Cyclic Loading Protocol ........................................................................... 31

Figure 3.15: Earthquake Loading Protocol ................................................................................... 32

Figure 3.16. Concrete Damage at Base at . 0 δ/δy . For C(CFRP)-4.0-#7(1.3)-0.05-EQ Protocol,

First Excursion to Exceed Value, Photo Depicts 7.3 δ/δy Excursion:a) C(CFRP)-4.0-#7(1.3)-

v
0.05, b) C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-

EQ, e) C(CFRP)-4.0-#7(1.3)-0.05-2X, f) C(CFRP)-4.0-#7(1.3)-0.15.......................................... 38

Figure 3.17. Concrete Flexural Damage at . 0 δ/δy . For C(CFRP)-4.0-#7(1.3)-0.05-EQ Protocol,

First Excursion to 2.55 δ/δy in Positive Direction is Depicted. For C(CFRP)-4.0-#7(2.7)-0.05

and C(CFRP)-4.0-#7(1.3)-0.05-2X First Excursion to −4.0 δ/δy is Depicted:a) C(CFRP)-4.0-

#7(1.3)-0.05, b) C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-

#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-#7(1.3)-0.05-2X, f) C(CFRP)-4.0-#7(1.3)-0.15 .................... 39

Figure 3.18. Concrete Damage at Base at . 0 δ/δy . For C(CFRP)-4.0-#7(1.3)-0.05-EQ Protocol,

First Excursion to 7.3 δ/δy in the Negative Direction is Depicted:a) C(CFRP)-4.0-#7(1.3)-0.05,

b) C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ,

e) C(CFRP)-4.0-#7(1.3)-0.05-2X, f) C(CFRP)-4.0-#7(1.3)-0.15 ................................................. 40

Figure 3.19. Concrete Flexural Damage at . 0 δ/δy . For C(CFRP)-4.0-#7(1.3)-0.05-EQ Protocol,

First Excursion to 7.3 δ/δy in the Negative Direction is Depicted. For C(CFRP)-4.0-#7(2.7)-0.05

Only a Depiction of the East Face was Available:a) C(CFRP)-4.0-#7(1.3)-0.05, b) C(CFRP)-4.0-

#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-

#7(1.3)-0.05-2X, f) C(CFRP)-4.0-#7(1.3)-0.15 ............................................................................ 41

Figure 3.20. Concrete Damage at Base at 15 δ/δy . For C(CFRP)-4.0-#7(1.3)-0.05-EQ Protocol,

the First Excursion to is 15 δ/δy Depicted:a) C(CFRP)-4.0-#7(1.3)-0.05, b) C(CFRP)-4.0-

#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-

#7(1.3)-0.05-2X, f) C(CFRP)-4.0-#7(1.3)-0.15 ............................................................................ 42

Figure 3.21. Concrete Flexural Damage at 15 δ/δy . For C(CFRP)-4.0-#7(1.3)-0.05-EQ

Protocol, the First Excursion to 15 δ/δy is Depicted. For C(CFRP)-4.0-#7(2.7)-0.05 Only a

Depiction of the East Face was Available:a) C(CFRP)-4.0-#7(1.3)-0.05, b) C(CFRP)-4.0-

vi
#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-

#7(1.3)-0.05-2X, f) C(CFRP)-4.0-#7(1.3)-0.15 ............................................................................ 43

Figure 3.22. Concrete Damage at Base at 20 δ/δy . For C(CFRP)-4.0-#7(1.3)-0.05-EQ Protocol,

the First Excursion to 20 δ/δy is Depicted. C(CFRP)-4.0-#7(1.3)-0.15 is Depicted at Completion

of Testing, Occurring at 15 δ/δy:a) C(CFRP)-4.0-#7(1.3)-0.05, b) C(CFRP)-4.0-#5(1.4)-0.05, c)

C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-#7(1.3)-0.05-2X,

f) C(CFRP)-4.0-#7(1.3)-0.15 ........................................................................................................ 44

Figure 3.23. Concrete Flexural Damage at 20 δ/δy . For C(CFRP)-4.0-#7(1.3)-0.05-EQ

Protocol, the First Excursion to is 20 δ/δy Depicted. For C(CFRP)-4.0-#7(2.7)-0.05 Only a

Depiction of the East Face was Available. For C(CFRP)-4.0-#5(1.4)-0.05 Only a Depiction of

the West Face was Available. For C(CFRP)-4.0-#7(1.3)-0.05-EQ Only a Depiction of the West

Face was Available. C(CFRP)-4.0-#7(1.3)-0.15 is Depicted at Completion of Testing, Occurring

at 15 δ/δy:a) C(CFRP)-4.0-#7(1.3)-0.05, b) C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-

#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-#7(1.3)-0.05-2X, f) C(CFRP)-

4.0-#7(1.3)-0.15 ............................................................................................................................ 45

Figure 3.24. Concrete Damage at Base at Completion of Testing:a) C(CFRP)-4.0-#7(1.3)-0.05,

b) C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ,

e) C(CFRP)-4.0-#7(1.3)-0.05-2X, f) C(CFRP)-4.0-#7(1.3)-0.15 ................................................. 46

Figure 3.25. Concrete Flexural Damage at Completion of Testing. For C(CFRP)-4.0-#7(2.7)-0.05

Only a Depiction of the East Face was Available. For C(CFRP)-4.0-#5(1.4)-0.05 Only a

Depiction of the West Face was Available. For C(CFRP)-4.0-#7(1.3)-0.05-EQ Only a Depiction

of the West Face was Available: a) C(CFRP)-4.0-#7(1.3)-0.05, b) C(CFRP)-4.0-#5(1.4)-0.05, c)

vii
C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-#7(1.3)-0.05-2X,

f) C(CFRP)-4.0-#7(1.3)-0.15 ........................................................................................................ 47

Figure 3.26. Lateral Load Deformation:a) C(CFRP)-4.0-#7(1.3)-0.05, b) C(CFRP)-4.0-#5(1.4)-

0.05, c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-#7(1.3)-

0.05-2X, f) C(CFRP)-4.0-#7(1.3)-0.15 ......................................................................................... 55

Figure 3.27. Backbone Model Fit for C(CFRP)-4.0-#7(1.3)-0.05: a) Base Shear, b) Effective

Base Shear ..................................................................................................................................... 60

Figure 3.28. Backbone Model Fit for C(CFRP)-4.0-#5(1.4)-0.05: a) Base Shear, b) Effective

Base Shear ..................................................................................................................................... 60

Figure 3.29. Backbone Model Fit for C(CFRP)-4.0-#7(2.7)-0.05: a) Base Shear, b) Effective

Base Shear ..................................................................................................................................... 61

Figure 3.30. Backbone Model Fit for C(CFRP)-4.0-#7(1.3)-0.05-EQ: a) Base Shear, b) Effective

Base Shear ..................................................................................................................................... 61

Figure 3.31. Backbone Model Fit for C(CFRP)-4.0-#7(1.3)-0.05-2X: a) Base Shear, b) Effective

Base Shear ..................................................................................................................................... 62

Figure 3.32. Backbone Model Fit for C(CFRP)-4.0-#7(1.3)-0.15: a) Base Shear, b) Effective

Base Shear ..................................................................................................................................... 62

Figure 3.33. Bilinear Model Backbone Slope Parameters ............................................................ 63

Figure 3.34. Normalized Bilinear Backbone Model Plots ............................................................ 65

Figure 3.35. Normalized Test Data Backbone Plots ..................................................................... 65

Figure 3.36. Effective Secant Stiffness Plots ................................................................................ 67

Figure 3.37. Strain Measured in Longitudinal Reinforcement at Cycle Peaks, C(CFRP)-4.0-

#5(1.4)-0.05................................................................................................................................... 69

viii
Figure 3.38. Strain Measured in Longitudinal Reinforcement at Cycle Peaks, C(CFRP)-4.0-

#7(2.7)-0.05................................................................................................................................... 69

Figure 3.39. Strain Measured in Longitudinal Reinforcement at Cycle Peaks, C(CFRP)-4.0-

#7(1.3)-0.05-EQ ............................................................................................................................ 70

Figure 3.40. Strain Measured in Longitudinal Reinforcement at Cycle Peaks, C(CFRP)-4.0-

#7(1.3)-0.05-2X ............................................................................................................................ 70

Figure 3.41. Strain Measured in Longitudinal Reinforcement at Cycle Peaks, C(CFRP)-4.0-

#7(1.3)-0.15................................................................................................................................... 71

Figure 3.42. Measured Curvature:a) C(CFRP)-4.0-#7(1.3)-0.05, b) C(CFRP)-4.0-#5(1.4)-0.05, c)

C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-#7(1.3)-0.05-2X,

f) C(CFRP)-4.0-#7(1.3)-0.15 ........................................................................................................ 73

Figure 3.43. Measured Curvature Excluding Bond Slip:a) C(CFRP)-4.0-#7(1.3)-0.05, b)

C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e)

C(CFRP)-4.0-#7(1.3)-0.05-2X, f) C(CFRP)-4.0-#7(1.3)-0.15 ..................................................... 74

Figure 3.44. Shear Sliding: a) C(CFRP)-4.0-#5(1.4)-0.05, b) C(CFRP)-4.0-#7(2.7)-0.05, c)

C(CFRP)-4.0-#7(1.3)-0.05-EQ, d) C(CFRP)-4.0-#7(1.3)-0.05-2X, e) C(CFRP)-4.0-#7(1.3)-0.15

....................................................................................................................................................... 76

Figure 3.45. Base Rotation: a) C(CFRP)-4.0-#5(1.4)-0.05, b) C(CFRP)-4.0-#7(2.7)-0.05, c)

C(CFRP)-4.0-#7(1.3)-0.05-EQ, d) C(CFRP)-4.0-#7(1.3)-0.05-2X, e) C(CFRP)-4.0-#7(1.3)-0.15

....................................................................................................................................................... 77

Figure 3.46. Components of Deformation: a) C(CFRP)-4.0-#5(1.4)-0.05, b) C(CFRP)-4.0-

#7(2.7)-0.05, c) C(CFRP)-4.0-#7(1.3)-0.05-EQ, d) C(CFRP)-4.0-#7(1.3)-0.05-2X,

e) C(CFRP)-4.0-#7(1.3)-0.15 ........................................................................................................ 80

ix
Figure 4.1. Column Deformation Model ...................................................................................... 83

Figure 4.2. Column Load-Deformation Response for Model and Tests: a) C(CFRP)-4.0-#7(1.3)-

0.05, b) C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-

EQ, e) C(CFRP)-4.0-#7(1.3)-0.05-2X, f) C(CFRP)-4.0-#7(1.3)-0.15.......................................... 89

x
 List of Tables

Table 2.1: Parameters of Past Tests on Reinforced Concrete Circular Bridge Columns Retrofitted

with FRP Jackets ........................................................................................................................... 11

Table 3.1: Test Matrix................................................................................................................... 14

Table 3.2: Measured Properties of Steel Reinforcement Obtained from Tensile Testing ............ 21

Table 3.3: Footing Concrete Compressive Strength from Cylinder Testing ............................... 22

Table 3.4: Column Concrete Compressive Strength from Cylinder Testing ................................ 23

Table 3.5: CFRP Properties in Major Direction ........................................................................... 24

Table 3.6: Main Shock Excursions and Drifts .............................................................................. 32

Table 3.7: After-shock Excursions and Drifts .............................................................................. 34

Table 3.8: Concrete Damage Onset .............................................................................................. 37

Table 3.9: Sequence of Longitudinal Bar Fractures ..................................................................... 49

Table 3.10: Summary of Cycles at which Longitudinal Bars Fractured....................................... 51

Table 3.11: Peak Demands, Displacements, and Drifts ................................................................ 56

Table 3.12: Deformation Capacity at Lateral Failure ................................................................... 56

Table 3.13: Stiffness and Strength of Backbone Models .............................................................. 64

Table 4.1: Effective Stiffness from Model and Tests ................................................................... 88

xi
 Executive Summary

Many bridges in the western United States, including those built for the Interstate Highway

System in the 1950s and 1960s, have seismically vulnerable reinforced concrete (RC) columns.

The seismic performance of many of these bridges is essential to post-earthquake mobility, as

bridges are relied upon as critical lifelines into urban centers after natural disasters. Some states,

including California and Washington, have introduced retrofit programs to enhance the seismic

ductility of vulnerable columns. The retrofit involves wrapping the column with either a

structural steel or fiber reinforced polymer (FRP) jacket, which enhances the deformation

capacity of the column to improve the seismic performance. Previous research on jacketed

columns has focused on strike-slip earthquakes, rather than long-duration, subduction-type

earthquakes that are characteristic of the Cascadia Subduction Zone.

This research was focused on characterization of the behavior of FRP jacketed bridge columns

under long-duration earthquakes. Six tests were conducted on cantilever bridge columns with

FRP jackets at the base. The FRP jackets were 0.40” and extended over a height of one column

diameter. The FRP had specified strength of 128-ksi and elastic modulus of 14.2ksi in the

circumferential direction. Test variables for the columns included longitudinal reinforcement

ratio, longitudinal bar diameter, axial load ratio, and loading protocol. All tested columns had

lateral deformation capacity of at least 6% drift, with lateral deformation capacity considered to

occur at 20% strength loss. Axial failure was not achieved in any of the test columns, and the

tests were stopped after multiple cycles at 10% drift. Five of the six test columns were nominally

identical to a set of previously tested columns with FRP jackets. The lateral deformation capacity
1
of each column with an FRP jacket met or exceeded that of the equivalent column with a steel

jacket.

A model was formulated to predict the deformation capacity of FRP jacketed columns. The

model included a column model followed by a fatigue model to estimate bar fracture. The

column was modeled in OpenSees using an elastic beam column element over the height of the

column, with two zero-length bond slip elements at the base. The two elements represented bond

slip of longitudinal reinforcement from the footing and from the jacketed column. The OpenSees

model was run to determine the stress-strain history in the outermost longitudinal reinforcement.

The strain history was used in a fatigue model to predict the drift at fracture of the first

longitudinal bar. The modeling approach was validated to test data from the experimental study.

The model was used to estimate the deformation capacity of FRP jacketed bridge columns.

2
 1. Introduction

1.1. Problem Statement

Significant damage to reinforced concrete bridge columns was observed following the 1971 San

Fernando, CA earthquake (Fung et al, 1971). Subsequent research led to an improved

understanding of the design issues, and, in 1983, AASHTO issued new bridge design guidelines

with changes aimed at addressing the issue of nonductile bridge columns in new construction.

The seismic vulnerabilities of pre-1971 bridge columns were again exposed by the 1989 Loma

Prieta, CA (NIST, 1990; EERI, 1990) and 1994 Northridge, CA earthquakes (EERI, 1994;

Buckle, 1994). Based on the damage observed in these earthquakes and research on the topic, it

is clear that there is a need for seismic retrofit of pre-1971 reinforced concrete bridge columns in

areas of high seismicity. Many bridges in the U.S. were constructed prior to 1971, including state

bridges constructed in the 1950s and 1960s as part of the national Interstate Highway System.

These bridges represent important lifelines for cities in the aftermath of a large earthquake.

Deficiencies in pre-1971 reinforced concrete bridge columns include insufficient lap splice

lengths of 20 times the longitudinal reinforcement diameter (i.e., 20db), inadequate shear

strength, and inadequate flexural ductility due to insufficient transverse reinforcement (typically

#4 hoops at 12” hoops) (Chai et al, 1991). Initial research on retrofit of seismically deficient

bridge columns focused on the use of steel jackets (Chai et al, 1991; Priestley et al, 1994a,b), and

this research led to field implementation in Washington and California. For circular columns, the

3
jackets were typically installed by seam welding two half-plates of semi-circular cross-section.

The jackets were oversized, and the gap that was left between the jacket and the column, which

was typically 1-2”, was filled with grout. For rectangular columns, the jackets were elliptical,

and concrete was used to fill the voids between the jacket and the column. If enhancement to

column shear strength was needed to prevent shear failure, the jackets were provided over the

full height of the column. Alternatively, the jackets were provided only in plastic hinge zones

(i.e., at the base of the column for cantilevers, and at the top and the bottom of the column for

fixed-fixed columns) if enhancement to shear strength was not needed. In this case, the jackets

can provide adequate confinement to prevent splice failure of 20db lap splices and to prevent

crushing of concrete. A small gap (~2”) was left between the base of the steel jacket and the

footing in order to avoid contact that would allow load transfer between the footing and the

jacket as the column deforms under earthquake demands.

Recommendations on steel jacket design are provided in the FHWA guidelines (2006) and stem

from research (Chai et al, 1991; Priestley et al, 1994a,b). The recommendations specify the

jacket thickness needed to prevent concrete crushing or splice failure. If these failure modes are

avoided, fatigue failure of longitudinal reinforcement is expected to occur at the gap between the

bottom of the steel jacket and the top of the footing. Chai et al (1991) determined that load

sharing was occurring between the steel jacket and the column, such that the jacket provided an

enhancement to flexural strength. The enhancement to flexural strength limited the spread of

plasticity, and Chai et al (1991) recommended that the zone of plasticity (i.e., plastic hinge

length) could be modeled as being equal to the gap length between the jacket and the footing

plus bond slip of the longitudinal reinforcement into the footing and into the steel jacket.

4
Much of the research on steel jackets was conducted in the 1990s, at a time when steel jackets

were a more cost-efficient option relative to fiber reinforced polymer (FRP). FRP has become

less cost inhibitive with time, and the use of FRP jackets is particularly beneficial in cases where

a steel jacket retrofit may be inadequate. FRP jackets may offer an advantage over steel jackets

by better accommodating the vertical spread of plasticity, which may be achieved by orienting

the FRP fibers in the circumferential direction. The improved spread of plasticity leads to a

larger plastic hinge length. This means that for a given curvature demand at the base of the

column, the FRP jacket accommodates a larger deformation capacity than the steel jacket,

leading to a reduced collapse likelihood in earthquakes. Previous research has not focused on the

behavior of FRP jacketed columns in long duration earthquakes. Existing models to predict the

deformation capacity of FRP jacket retrofitted columns based on fatigue failure of reinforcement

do not exist.

1.2. Objectives

The first objective of the research was to experimentally assess the deformation capacity of FRP

jacketed bridge columns, with properties characteristic of Washington State bridges, subjected to

Cascadia Subduction Zone earthquake demands. The second objective was to formulate a model

to determine the deformation capacity of FRP jacket retrofitted bridge columns. The model was

intended to be suitable for use in nonlinear time history analyses, making it a suitable tool for

analyzing the behavior of FRP jacket retrofitted bridge columns in a given earthquake ground

5
motion. The data needed for model calibration/validation was generated from the experimental

program. Six approximately one-half-scale to two-thirds-scale bridge columns with FRP jacket

retrofits were constructed and tested. The test program was intended to address gaps from

previous studies by including the range of practical parameters that influence column

deformation capacity. A lumped plasticity model for the retrofitted columns was developed and

validated with experimental results. In this model, deformation capacity was based on fatigue

fracture of reinforcement.

1.3. Expected Contributions

Results from testing of FRP jacketed columns were used to provide an assessment of the

deformation capacity of WSDOT bridge columns in Cascadia Subduction Zone earthquakes. The

tests produced a unique dataset for FRP jacketed columns subjected to long duration earthquake

demands that may be used for model calibration/validation. The model formulated in this study

was validated to this data. This model may be used to assess the failure probability of any FRP

jacket retrofitted bridge column for a given earthquake ground motion, making it a useful tool to

aid in the design of FRP retrofits. This study addresses TriDurle research thrust #5: Addressing

natural hazards and extreme disaster events that threaten the durability and service life of

transportation infrastructure.

6
1.4. Report Overview

This report includes five chapters and a list of references. An introduction is provided in Chapter

1, and a literature review is provided in Chapter 2. The experimental study is presented in

Chapter 3, which included two subsections. An overview of the experimental program was

provided in the first subsection. This included methods of construction, testing, and material

properties. Test results were provided and discussed in the second subsection. This included a

summary of damage, load-deformation column response, stiffness, reinforcement strains, column

curvatures, and components of deformation. Information on column modeling was provided in

Chapter 4, which also included two subsections. A description of the column model was

provided in the first subsection, with model validation described in the second subsection. A

summary and list of conclusions are provided in Chapter 5. A list of references follows Chapter

5.

7
 2. Literature Review

A number of previous experimental studies have been conducted on circular reinforced concrete

columns with FRP jackets. Xiao and Ma (1997) tested three columns that had longitudinal

reinforcement at the base of the column with lap splice lengths of 20 times the bar diameter. Two

of the three columns were retrofit with FRP jackets prior to testing. The unretrofitted column

failed due to lap splice failure at low ductility. This column was repaired, retrofit with FRP, and

tested again. For all three retrofitted columns, the level of confinement provided by the jackets

was such that bond failure occurred gradually, allowing for ductility demands in the range of

four to six at 20% loss of lateral load-carrying capacity. Xiao and Ma (1997) used the test results

to develop and validate a deformation capacity model that accounted for the influence of

confinement on bond slip in the splice region.

Haroun and Elsanadedy (2005a,b) conducted tests on FRP jacketed circular columns that also

had lap splice lengths of 20 times the longitudinal bar diameter. The jackets used were such that

the reported failure occurred due to concrete crushing or longitudinal bar buckling rather than

bond. Similar failure modes were reported for the tests of Ghosh and Sheikh (2007).

For FRP jacket retrofit of reinforced concrete bridge columns, failure may occur due to concrete

crushing accompanied by jacket rupture or low-cycle fatigue fracture of longitudinal

reinforcement. The likelihood of jacket rupture is mitigated by providing a jacket of suitable

thickness based on the Mander et al (1988) confined concrete model, which is reflected in the

8
FHWA guidelines (2006). It is shown in Section x on column modeling that for cases in which

reinforcement fatigue fracture governs failure, the parameters that impact column deformation

capacity are the neutral axis depth (influenced by the axial load ratio and longitudinal

reinforcement ratio), plastic hinge length (influenced by the longitudinal bar diameter and

column diameter), and the loading history. Systematic variation of these parameters is not

evident from the summary of previous tests provided in Table 1, and the proposed experimental

program is intended to address this issue.

Previous experimental programs (Xiao and Ma, 1997; Haroun and Elsanadedy, 2005a,b; Ghosh

and Sheikh, 2007; Breña and Schlick, 2007) used loading protocols that were developed using

strike-slip earthquake ground motions (e.g., Krawinkler, 1992). Subduction zone earthquakes

produce longer duration ground motions than strike-slip earthquakes, which results in greater

cumulative plastic deformation in flexure-yielding components and the increased potential for

fatigue failure. The USGS has recently released updated hazard maps that require increased

levels of seismicity for structural design in regions affected by the Cascadia Subduction Zone

(CSZ). These new maps reflect the geologic evidence indicating that the CSZ is capable of

producing M9 megathrust earthquakes at the interface between the Juan de Fuca and North

American plates (Atwater et al. 1995). Such an event has a 10-14% chance of occurring in the

next 50 years (Goldfinger et al. 2012). In an M9 CSZ earthquake, strong ground shaking is

expected in Washington, Oregon, northern California, and Alaska. Due to the lack of an accurate

metric to assess the deformation capacity of FRP jacket retrofitted columns, there is uncertainty

as to the expected performance of retrofitted bridge columns in Cascadia Subduction Zone

earthquakes. The proposed research will result in the development of a new model that may be

9
used to determine if and when a FRP jacketed bridge column fails under a specific seismic

demand, such as a CSZ-type ground motion. The model will be calibrated/validated using test

data generated through an experimental study on FRP jacketed columns. The tests are expected

to generate a unique dataset for model calibration relative to data that is available from previous

tests. Specifically, the columns will be tested under loading protocols reflective of long-duration

earthquakes. Past experimental studies on RC columns with FRP jackets (Xiao and Ma, 1997;

Haroun and Elsanadedy, 2005a,b; Ghosh and Sheikh, 2007; Breña and Schlick, 2007) have not

considered long-duration earthquake demands. In addition to the difference in loading protocol,

the level of FRP confinement used for the test columns in the proposed study will be higher than

that typically used in previous studies. In previous studies, failure typically occurred due to

concrete crushing or longitudinal reinforcement buckling. The combination of increased

confinement and increased cycle content for the columns in the proposed study is expected to

produce fatigue failure of reinforcement. Furthermore, direct comparison between steel jackets

and FRP jackets has not been considered in an experimental study that isolates this test variable.

To address this shortcoming, the six FRP jacketed columns tested as part of the proposed

research were identical to a set of six steel jacketed columns tested by the P.I.’s as part of a study

funded by WSDOT.

10
 Table 2.1: Parameters of Past Tests on Reinforced Concrete Circular Bridge Columns
Retrofitted with FRP Jackets
Total
Longitudinal Shear Lap Splice
Axial Load Diameter Longitudinal FRP FRP FRP Elastic
Specimen Steel Area/ Span Ratio Length /
Experimental Program Ratio, or Bar Diameter Thickness Ultimate Modulus
Name Gross (M /VD or Longitudinal
P/(Agf'c) Length (inches) (inches) Stress (ksi) (ksi)
Concrete M/VL) Bar Diameter
Area
Xiao and Ma (1997) C2-RT4 0.0195 0.050 24 4 0.75 20 0.5 80 7000

Xiao and Ma (1997) C3-RT5 0.0195 0.050 24 4 0.75 20 0.625 80 7000

Haroun and Elsanadedy (2005a) CF-R1 0.0195 0.061 24 6 0.75 20 0.028 604 33568

Haroun and Elsanadedy (2005a) CF-R2 0.0195 0.060 24 6 0.75 20 0.028 642 33365

Haroun and Elsanadedy (2005a) CF-R3 0.0195 0.067 24 6 0.75 20 0.449 108 5293

Haroun and Elsanadedy (2005a) CF-R4 0.0195 0.059 24 6 0.75 20 0.067 635 32770

Haroun and Elsanadedy (2005a) CF-R5 0.0195 0.056 24 6 0.75 20 0.5 93 5278

Haroun and Elsanadedy (2005a) CF-R6 0.0195 0.067 24 6 0.75 20 0.327 136 9135

Haroun and Elsanadedy (2005b) CS-R1 0.0195 0.054 24 2 0.75 20 0.028 604 33568

Haroun and Elsanadedy (2005b) CS-R2 0.0195 0.056 24 2 0.75 20 0.028 642 33365

Haroun and Elsanadedy (2005b) CS-R3 0.0195 0.065 24 2 0.75 20 0.406 61 2683

Haroun and Elsanadedy (2005b) CS-R4 0.0195 0.059 24 2 0.75 20 0.047 181 15051

Ghosh and Sheikh (2007) CAF1-2N 0.0172 0.050 14 5.65 0.75 24.67 0.04 148 11458

Ghosh and Sheikh (2007) CAF1-5N 0.0172 0.270 14 5.65 0.75 24.67 0.04 148 11458

Ghosh and Sheikh (2007) CBF1-6N 0.0172 0.050 14 5.65 0.75 24.67 0.04 148 11458

Brena Schlick (2007) CFRP-05 0.0254 0.050 9.5 4.5 0.5 24 0.0065 550 33000

Brena Schlick (2007) KFRP-05 0.0254 0.050 9.5 4.5 0.5 24 0.011 290 17400

Brena Schlick (2007) CFRP-15 0.0254 0.150 9.5 4.5 0.5 24 0.0065 550 33000

Brena Schlick (2007) KFRP-15 0.0254 0.150 9.5 4.5 0.5 24 0.011 290 17400

11
 3. Column Testing

3.1. Methodology

3.1.1. Test Specimen Details

Each test specimen consisted of a column, a footing, and a loading head. A test matrix for the six

test columns is provided in Table 3.1, with drawings of the columns provided in Figure 3.1. The

test columns were nominally identical to those tested by McGuiness (2021), except that carbon

fiber reinforced polymer (CFRP) jackets were used in place of steel jackets. The reinforced

concrete column parameters matched those tested by McGuiness (2021), except that C(S)-6.0-

#7(1.3)-0.05, the tallest column in the McGuiness (2021) study, was replaced by a column with

details identical to C(S)-4.0-#7(1.3)-0.05 and C(S)-4.0-#7(1.3)-0.05-EQ. Test variables in the

experimental study included longitudinal reinforcement diameter, longitudinal reinforcement

ratio, axial load ratio, and loading protocol. Design parameters among the test columns are

inherent in the naming convention. For example, C(CFRP)-4.0-#7(1.3)-0.05 designates a circular

column with CFRP jacket in “C(CFRP)”, with a “4.0” span to depth ratio (H/D = 4.0), using

“#7” longitudinal reinforcement with 1.3% longitudinal reinforcement ratio (As/Ag = 0.013), and

5% axial load ratio (P/(f’cAg) = 0.05). A fully reversed cyclic loading protocol was used for five

of the six columns, whereas C(CFRP)-4.0-#7(1.3)-0.05-EQ was subjected to a loading history

determined from analysis of a bridge to a specific earthquake. C(CFRP)-4.0-#7(1.3)-0.05-2X

was subjected to twice the number of cycles at each increment as C(CFRP)-4.0-#7(1.3)-0.05.

More details on the loading protocols are provided in Section 3.6. The columns were tested as

cantilevers, and the column height, H, was the measured distance from the top of footing to the
12
line of action of the applied lateral load. The height of the circular column section was 9” less

than H to facilitate inclusion of the loading head. Column diameter, D, was 24”. All test columns

had nominally identical 1” clear cover, cb, Grade 40 longitudinal and transverse reinforcement,

and #3 hoops with a lap-splice length, lb, of 16” and center-to-center vertical spacing, s, of 8”

used as transverse reinforcement. No hooks were provided on transverse reinforcement lap

splices.

The footing and loading head were consistent with those used in the McGuiness (2021) tests,

with details provided in Figure 3.2 and Figure 3.3. C(CFRP)-4.0-#7(1.3)-0.05-2X, which did not

have an equivalent test in the McGuiness (2021) study, had the same footing as C(CFRP)-4.0-

#7(1.3)-0.05. Voids were provided in the footing using SCH40 PVC pipe, located as shown in

Figure 3.3, to allow the footing to be post-tensioned to the laboratory strong floor and to

facilitate lifting and moving of the test specimens before and after testing.

The CFRP jackets were designed to provide a level of confinement stiffness that was greater than

that provided by the 3/16” steel jackets used in the columns tested by McGuiness (2021). The

specified CFRP modulus of elasticity was 14,200 ksi in the CFRP direction that was oriented

circumferentially around the columns. Five sheets of 0.08” thick CFPR were used, resulting in a

thickness of 0.40”, which provides circumferential stiffness of 14,200 ksi * 0.40” = 5,680 k/in

relative to 29,000 ksi * 0.1875” = 5,437.5 k/in for the steel jacket. The CFRP jackets were used

over the lower 24” inches of the column, which is the column diameter. The steel jackets used in

the columns tested by McGuiness spanned the full circular portion of the column, with the

exception of a 1” gap at the top and bottom.

13
 Table 3.1: Test Matrix

Long. #
Loading
Column I.D. Bar Long. 𝐴𝑠 /𝐴𝑔 𝑃/(𝐴𝑔 𝑓𝑐′ ) 𝐻/𝐷
Protocol
Size Bars

C(CFRP)-4.0-#7(1.3)-0.05 #7 10 0.0133 0.05 4.0 Cyclic

C(CFRP)-4.0-#5(1.4)-0.05 #5 20 0.0137 0.05 4.0 Cyclic
C(CFRP)-4.0-#7(2.7)-0.05 #7 20 0.0265 0.05 4.0 Cyclic
C(CFRP)-4.0-#7(1.3)-0.05-EQ #7 10 0.0133 0.05 4.0 EQ
C(CFRP)-4.0-#7(1.3)-0.05-2X #7 10 0.0133 0.05 4.0 2x Cyclic
C9CFRP)-4.0-#7(1.3)-0.15 #7 10 0.0133 0.15 4.0 Cyclic

Figure 3.1: Column Reinforcement Layout

14
Figure 3.2: General Footing Reinforcement Details (10-#7 Starter Base Depicted) (Dimensions
in Inches)

15
Figure 3.3: Footing Sleeve Details (10-#7 Starter Bars Depicted) (Dimension in Inches)

3.1.2. Test Specimen Construction

Each column was constructed in two pours, with a cold joint at the base of the column.

Formwork consisted of 3/4 CDX plywood, 2” x 4” framing lumber, and tubular forms, made

from heavy cardboard, for the circular portion of the column. Six sets of forms were constructed,

as depicted in Figure 3.4a, to enable pouring from the same load of concrete. Footing

reinforcement cages were tied externally prior to being placed inside the formwork. Column

starter bars were tied to footing reinforcement while located within the column cross-section

16
using plywood drilled holes for the bars. Strain gauges in the footing concrete were wired away

from the footing-column interface. Further description of the strain gauge layout is provided in

Section 3.5. Vertical PVC sleeves were precisely located in the footing to match the pattern in

the strong floor. Formwork heights allowed direct tailgate delivery of ready-mix concrete for the

footing. The cold joint between the footing and subsequent column pour was intentionally

roughened to magnitudes greater than ¼-in. The remaining exposed footing surfaces were

troweled to a smooth uniform finish. Column reinforcement cages where tied horizontally.

After completion of the footing pours, the column cages were rotated into the vertical position,

and affixed to the footing starter bars with contact lap splices of longitudinal reinforcement to

starter bars. Spacers were installed on reinforcement, and column forms were then installed by

lowering from above. Proper vertical alignment of column forms was provided by timber bracing

between the top of the forms to ground level brace points, as depicted in Figure 3.5a. The wires

of strain gauges cast into the column, similar to the footing, were routed away from the footing-

column interface. Top block formwork was supported by the column formwork and at ground

level. Top block transverse reinforcement was tied to column longitudinal reinforcement, and

PVC sleeves were installed within top block formwork, as depicted in Figure 3.6b. Due to

height, the delivery and placement of concrete to the column formwork required the assistance of

a concrete boom truck pump and 12-ft drop hose. Concrete was pumped incrementally, with

periodic pauses to raise the pump hose and facilitate vibratory consolidation of the concrete. The

exposed top surface was trowel finished. After the appropriate cure time, formwork was

removed, including cutting and removing the single-use circular forms.

17
Several months after column construction, CFRP jackets were installed by a professional team.

Installation began with roughening of the concrete surface, using a grinder, to facilitate bond of

epoxy to concrete. Dust from roughening was removed from the column using acetone. The

CFRP wrap was cut to lengths approximately equal to one circumference of the column plus 12-

inches. The epoxy was mixed in batches due to a 1-hour pot life at 70°F (21°C). After the first

epoxy batch was mixed, an epoxy primer was applied to the columns using a nap roller. CFRP

sheets were hand saturated using plastic trowels. Care was taken to ensure full fiber saturation

without saturation so excessive that the sheet would settle once on the column. After saturating

the fibers, the sheets were wrapped around the column and the plastic hand trowels were used to

remove any entrapped air and excess saturate. The sheet had a lap length of 12-inches, with the

lap intended to provide adequate bond. The seams had additional epoxy applied per manufacturer

specifications. After the first layer, additional layers were applied with the start of the new layer

on the opposite side of the column as the end of the previous layer, such that all seams were

offset. After all 5 layers were applied, an epoxy paste was applied as an outer coating, and the

columns were given at least 72 hours to cure before testing of the first column. A finished jacket

is shown in Figure 3.4. Lacking of bulging in the jacket suggests that epoxy was appropriately

applied, with bonding between layers.

18
Figure 3.4: CFRP Jacket Before Testing for C(CFRP)-4.0-#7(1.3)-0.05

Figure 3.5: Footing Construction

19
 Figure 3.6: Construction: a) Bracing for Column Concrete Pour, and b) Bracing for Top Block
Concrete Pour

3.1.3. Material Properties

3.1.3.1. Steel Reinforcement

A batch of #7, #5, and #3 Grade 40 reinforcement was manufactured specifically for this study,

such that all column reinforcement in a given size was from the same heat. Three samples of #5

and #7 column longitudinal reinforcement were tested, with resulting stress-strain plots provided

in Figure 3.7. Values of the resulting yield strength, 𝑓𝑦 , ultimate strength, 𝑓𝑢 , and percent

elongation, are provided in Table 3.2.

20
 Table 3.2: Measured Properties of Steel Reinforcement Obtained from Tensile Testing

Figure 3.7: Longitudinal Steel Reinforcement Testing: a) #5 Stress-Strain Relationship, b) #7

Stress-Strain Relationship

21
3.1.3.2. Concrete

A single concrete supplier and mix design were used for the project. The mix used a 3/8-in

maximum aggregate to reflect 3/4-in maximum aggregate at full-scale. The footings were poured

separately from the columns and loading heads, as described in Section 3.2. 6” x 12” cylinders

were prepared for the footing and the columns, respectively. Cylinders were stored in close

proximity to the specimens. The footing cylinders were tested at 7-days, with the measured

concrete compressive strength provided in Table 3.3. Four cylinders were tested within 3 days of

each column test, and the results are provided in Table 3.4. A clear trend of strength increase

̅′ )
with time is not evident from the data, and the average concrete compressive strength (𝑓𝑐,𝑡𝑒𝑠𝑡

was 3.949-ksi with a coefficient of variation of 0.059.

Table 3.3: Footing Concrete Compressive Strength from Cylinder Testing

22
 Table 3.4: Column Concrete Compressive Strength from Cylinder Testing

Cylinder Cylinder Cylinder Cylinder

Average
1 2 3 4 SD
Column f’c
f’c f’c f’c f’c [ksi]
[ksi]
[ksi] [ksi] [ksi] [ksi]
C(CFRP)-4.0-#7(1.3)-0.05 3.719 3.552 3.980 3.804 3.764 0.178
C(CFRP)-4.0-#5(1.4)-0.05 4.089 4.119 3.929 - 4.046 0.102
C(CFRP)-4.0-#7(2.7)-0.05 4.053 3.505 3.735 3.678 3.743 0.229
C(CFRP)-4.0-#7(1.3)-0.05-EQ 3.742 3.545 4.232 3.715 3.809 0.295
C(CFRP)-4.0-#7(1.3)-0.05-2X 4.478 4.295 4.038 3.821 4.158 0.288
C(CFRP)-4.0-#7(1.3)-0.15 3.851 3.997 4.447 4.405 4.175 0.296
Cumulative Average 3.949 0.232

3.1.3.3. Carbon Fiber Reinforced Polymer (CFRP)

Each CFRP jacket was comprised of Simpson Strong-Tie CSS-CUCF44, which is a

unidirectional carbon fabric designed to be laminated with CSS-ES and CSS-UES saturant. It is

specified to have 10% of the strength properties in the minor direction than the major direction.

The major direction was oriented around the circumferences of the columns. Table 3.5 provides

properties for the major direction, as specified in the ICC Report ESR-3404 (ICC ES, 2022).

23
 Table 3.5: CFRP Properties in Major Direction

Property Value
Dry Fiber Tensile Strength 670,000 psi
Dry Fiber Tensile Modulus 37,000,000 psi
Dry Fiber Elongation at Break 1.9 %
Dry Fiber Unit Weight 44.0 oz./yd.2
Cured Composite Tensile Strength 128,000 psi
Cured Composite Tensile Modulus 14,200,000 psi
Cured Composite Elongation at Break 0.9 %
Cured Composite Thickness per Layer 0.08 in.

3.1.4. Test Set-Up

Using the set-up shown in Figure 3.8, tests were conducted in the Simpson Strong-Tie

Experimental Testing Laboratory, which is part of the Composite Materials and Engineering

Center (CMEC) at Washington State University. Prior to testing, the footing block of each test

column was post-tensioned to the laboratory strong floor. A pair of steel channels spanning over

each end of the footing blocks were used to engage more floor anchors. During testing, constant

axial load and cyclic lateral displacement were applied to the test column. Lateral load and

displacement was applied using a servo-controlled hydraulic actuator with 40-in stroke and

capacity of 220-k in tension and 328-k in compression. The lateral actuator was post-tensioned to

the top of the test column and was reacted by the laboratory strong wall. Axial load was applied

using 60-ton hydraulic jack(s) manually controlled by a self-contained hydraulic power unit.

Load was controlled through a pressure regulating valve integral with the power unit and

monitored with 100-k low profile load cell(s). As shown in Figure 3.10, a single jack and load

cell were used for five of the six tests, while three jacks and load cells were used for C(CFRP)-

24
4.0-#7(1.3)-0.05 due to the higher axial load. The application of vertically oriented axial load to

simulate P-delta was a unique feature of this program relative to much of the prior research (e.g.,

Haroun and Elsanadedy, 2005), which used tendons anchored to the strong floor to apply axial

load. Utilizing a roller and swiveling knuckle assembly, as depicted in Figure 3.9, the applied

axial load was able to translate with the top of the column to remain vertical. A steel frame,

comprised of four steel framing columns and three beams, was used to react the applied axial

load and to prevent out of plane movement of the test column. The standard axial load

configuration (5 of 6 tests) is depicted in Figure 3.9. The column with large axial load set-up

configuration required an additional axial load roller and larger capacity clevis assembly.

Additionally, steel plates were added between the test column and loading beam to distribute and

collect the greater axial loads.

25
 b)

a)

𝑃𝑑
Figure 3.8: Test Set-up, = 0.05: a) Schematic, b) Photo
𝐴𝑔 𝑓𝑐′

𝑷𝒅
Figure 3.9: Axial Load Setup, = 𝟎. 𝟎𝟓
𝑨𝒈 𝒇′𝒄

26
3.1.5. Instrumentation

Forces, strains, and displacements were recorded during testing. Five of the six tests used a

single 100-k load cell to measure axial load, while the test column with higher axial load used

three 100-k load cells. Displacement measurements of the column and footing relative to a

stationary reference frame were obtained from linear variable differential transducers (LVDTs)

and string potentiometers at the locations shown in Figure 3.10. Rotation of the footing was

determined based on two vertical sensors located at each end of the footing. LVDTs spanning

between the footing and 0.5-in above the footing were used to measure column sliding in the

horizontal direction and bond slip in the vertical direction. Axial-flexural deformations within

the jacketed region were measured using vertical sensors at the locations shown in Figure 3.13.

To attach LVDTs, ¼-in threaded instrumentation rods were installed into the column core

concrete by drilling approximately ½-inch diameter holes approximately 1” deep into the column

and inserting the threaded instrumentation rods with fast drying epoxy. Post-installed ¼-in

wedge anchors were alternatively used at the footing and loading head and at the relative

measurements bridging the gaps between the footing-column and column-loading head.

For each column, 14 strain gauges were installed on each of the two longitudinal starter bars

located closest to the column ends, as shown in Figure 3.10. As shown in Figure 3.11, the gauges

were located in the column and footing and were arranged symmetrically above and below a

location at ½-inch above the column-footing interface. The gauges were spaced at intervals of

every fourth bar deformation. This arrangement of strain gauges in the anticipated plastic hinge

region was intended to enable collection of data that would quantify the extent of strain

27
penetration into the columns and footings. Installation of each strain gauge required removal of

one bar deformation. Reinforcement bar deformation removal was limited to the surface area

required to adhere a gauge to the bar, which was typically one-half of the circumferential

deformation. The exception was full circumferential removal of the bar deformation at a location

of ½–inch above the column-footing interface to facilitate the additional installation of a

redundant gauge on the opposing side. Strain gauge wires were arranged to exit from the top of

the footing.

Figure 3.10: Instrumented Starter Bar Layout

28
Figure 3.11: Strain Gauge Reinforcement Layout (Dimensions in Inches)

Figure 3.12: Stationary Reference Measurements

29
 Figure 3.13: LVDT Instrumentation Layout (Dimensions in Inches)

3.1.6. Loading Protocol

A fully-reversed cyclic loading protocol was used for four of the six test columns, and one test

was conducted using an earthquake loading protocol. These protocols were identical to those

used by McGuiness et al (2021). The reversed-cyclic protocol began with force-controlled

cycles, with three cycles each at 5-k, 10-k, and additional intervals of 10-k prior to the yield drift.

Displacement control was employed thereafter, with six full cycles each at 1.0, 1.25, 1.5, 1.75,

2.0, and 2.5 times the yield drift, followed by two cycles each at 3.0, 3.5, 4.0, 5.0, 6.0, 8,0, 10.0,

12.5, 15.0, 20.0, and 25.0 times the yield drift, or until the test was completed. For cases in

which testing was continued, additional cycles were applied at 25.0 times yield drift, 𝛿𝑦 /𝐻.

30
C(CFRP)-4.0-#7(1.3)-0.05 used a modification of the reversed cyclic protocol that had twice the

number of cycles at each increment. Consistent with the steel jacketed columns tested by

McGuiness (2021), the yield drift was taken as 0.4% for C(S)-4.0-#7(1.3)-0.05, C(S)-4.0-

#5(1.4)-0.05, and C(S)-4.0-#7(1.3)-0.05-2X and 0.5% for C(S)-4.0-#7(2.7)-0.05 and C(S)-4.0-

#7(1.3)-0.15. C(S)-4.0-#7(1.3)-0.05-EQ was tested using an earthquake response history protocol

that consisted of a main-shock and aftershock, as shown in Figure 3.36 with values provided in

Table 3.6 and 3.7, respectively. For displacement-controlled cycles, the drift used to control the

test was corrected to account for footing sliding and rotation.

Figure 3.14: Standard Cyclic Loading Protocol

31
 Figure 3.15: Earthquake Loading Protocol

Table 3.6: Main Shock Excursions and Drifts

# Drift [%] # Drift [%] # Drift [%] # Drift [%] # Drift [%]
1 0.000 71 -2.396 141 -0.445 211 -0.167 281 -0.418
2 -0.011 72 -1.836 142 0.545 212 0.212 282 0.395
3 -0.005 73 -3.165 143 -0.550 213 -0.242 283 -0.333
4 -0.006 74 1.565 144 0.502 214 0.208 284 0.181
5 0.016 75 -0.240 145 -0.542 215 -0.165 285 -0.148
6 -0.003 76 2.746 146 0.468 216 0.099 286 0.249
7 -0.001 77 0.611 147 -0.571 217 -0.035 287 -0.212
8 -0.004 78 2.301 148 0.355 218 0.059 288 0.168
9 -0.002 79 -3.229 149 -0.185 219 -0.141 289 -0.146
10 -0.008 80 -0.268 150 0.252 220 0.140 290 0.179
11 0.000 81 -1.303 151 -0.290 221 -0.066 291 -0.247
12 -0.004 82 0.567 152 0.257 222 0.079 292 0.162
13 0.005 83 -0.468 153 -0.325 223 -0.077 293 -0.187
14 -0.002 84 0.569 154 0.350 224 0.035 294 0.165
15 -0.005 85 -2.447 155 -0.257 225 -0.082 295 -0.190
16 0.000 86 4.087 156 0.145 226 0.135 296 0.146
17 0.000 87 -1.566 157 -0.225 227 -0.232 297 -0.068
18 0.004 88 0.957 158 0.353 228 0.259 298 0.042
19 -0.011 89 -1.340 159 -0.452 229 -0.217 299 -0.118
32
20 -0.005 90 1.496 160 0.455 230 0.141 300 0.080
21 -0.006 91 -1.944 161 -0.442 231 -0.103 301 -0.038
22 0.004 92 2.784 162 0.491 232 0.249 302 0.040
23 0.003 93 -1.460 163 -0.337 233 -0.421 303 -0.060
24 0.004 94 1.911 164 0.276 234 0.376 304 0.044
25 0.004 95 -2.256 165 -0.296 235 -0.304 305 -0.018
26 0.023 96 1.260 166 0.190 236 0.219 306 0.021
27 -0.026 97 -2.155 167 -0.249 237 -0.185 307 -0.077
28 0.044 98 2.183 168 0.139 238 0.174 308 0.140
29 -0.055 99 -2.025 169 -0.033 239 -0.230 309 -0.132
30 0.025 100 1.794 170 0.018 240 0.168 310 0.102
31 -0.021 101 -1.856 171 -0.184 241 -0.082 311 -0.130
32 -0.021 102 1.752 172 0.217 242 0.060 312 0.058
33 -0.039 103 -1.727 173 -0.204 243 -0.127 313 0.008
34 0.119 104 1.510 174 0.187 244 0.136 314 0.011
35 -0.130 105 -0.758 175 -0.256 245 -0.145 315 -0.063
36 0.057 106 0.473 176 0.300 246 0.119 316 0.034
37 -0.004 107 -0.771 177 -0.210 247 -0.127 317 -0.023
38 0.251 108 0.869 178 0.184 248 0.063 318 0.016
39 -0.323 109 -0.852 179 -0.226 249 -0.081 319 -0.016
40 0.073 110 0.672 180 0.213 250 0.071 320 0.072
41 -0.195 111 -0.495 181 -0.221 251 0.000 321 -0.047
42 0.447 112 0.345 182 0.232 252 0.001 322 0.029
43 -0.460 113 -0.433 183 -0.253 253 0.000 323 -0.033
44 0.493 114 0.424 184 0.221 254 0.004 324 0.019
45 -0.593 115 -0.674 185 -0.211 255 -0.023 325 -0.024
46 0.625 116 0.785 186 0.157 256 0.070 326 0.013
47 -0.746 117 -0.541 187 -0.232 257 -0.183 327 -0.018
48 0.980 118 0.384 188 0.216 258 0.170 328 0.008
49 -0.855 119 -0.465 189 -0.132 259 -0.191 329 -0.014
50 1.019 120 0.482 190 0.086 260 0.261 330 0.004
51 -0.659 121 -0.566 191 -0.097 261 -0.340 331 -0.011
52 0.405 122 0.478 192 0.025 262 0.330 332 0.002
53 -0.590 123 -0.340 193 -0.070 263 -0.347 333 -0.009
54 0.930 124 0.021 194 0.087 264 0.373 334 0.000
55 0.924 125 0.011 195 -0.203 265 -0.319
56 0.987 126 0.395 196 0.254 266 0.206
57 -1.095 127 -0.708 197 -0.224 267 -0.223
58 0.841 128 0.714 198 0.192 268 0.310
59 -2.925 129 -0.590 199 -0.117 269 -0.342
60 4.524 130 0.245 200 0.236 270 0.175
33
61 -6.148 131 -0.117 201 -0.284 271 -0.048
62 3.341 132 0.066 202 0.159 272 0.076
63 -2.340 133 -0.045 203 -0.179 273 -0.205
64 1.358 134 -0.036 204 0.183 274 0.251
65 -2.432 135 -0.055 205 -0.310 275 -0.080
66 3.553 136 0.161 206 0.202 276 0.021
67 -1.108 137 -0.453 207 -0.277 277 -0.031
68 -0.795 138 0.468 208 0.310 278 0.058
69 -0.800 139 -0.402 209 -0.181 279 -0.196
70 2.616 140 0.306 210 0.142 280 0.423

Table 3.7: After-shock Excursions and Drifts

# Drift [%] # Drift [%] # Drift [%] # Drift [%] # Drift [%]
1 -0.017 46 -0.01 91 -0.3 136 0.029 181 0.062
2 -0.013 47 -0.648 92 0.256 137 -0.092 182 -0.062
3 -0.024 48 0.503 93 -0.207 138 -0.098 183 0.057
4 -0.003 49 -0.161 94 0.129 139 0.114 184 -0.035
5 -0.024 50 -0.077 95 -0.209 140 -0.105 185 0.054
6 0.043 51 -0.079 96 0.364 141 0.111 186 -0.212
7 -0.034 52 0.263 97 -0.272 142 -0.16 187 0.155
8 -0.032 53 -0.351 98 0.231 143 0.149 188 -0.118
9 -0.035 54 0.666 99 -0.144 144 -0.139 189 0.06
10 -0.028 55 -0.77 100 0.17 145 0.063 190 -0.049
11 -0.028 56 0.752 101 -0.311 146 0.05 191 0.06
12 0.007 57 -0.215 102 0.13 147 0.086 192 -0.106
13 0.003 58 -0.147 103 -0.202 148 -0.102 193 0.075
14 0.043 59 -0.271 104 0.271 149 -0.009 194 -0.078
15 -0.031 60 0.773 105 -0.197 150 -0.026 195 0.035
16 -0.028 61 -0.32 106 0.196 151 -0.018 196 -0.078
17 -0.075 62 0.593 107 -0.309 152 -0.075 197 0.004
18 -0.016 63 -0.527 108 0.39 153 0.083 198 -0.006
19 -0.023 64 0.438 109 -0.362 154 -0.093 199 0.03
20 0.002 65 -0.492 110 0.312 155 0.035 200 -0.036
21 -0.021 66 0.34 111 -0.233 156 -0.045 201 0.079
22 -0.02 67 -0.579 112 0.181 157 0.018 202 -0.09
23 -0.021 68 0.982 113 -0.359 158 -0.088 203 0.025
24 -0.02 69 -1.005 114 0.277 159 0.105 204 -0.131
25 -0.092 70 0.855 115 -0.213 160 -0.08 205 0.108
26 0.098 71 -0.26 116 0.18 161 0.11 206 -0.87
27 -0.218 72 0.132 117 -0.154 162 -0.192 207 0.062

34
 28 0.333 73 -0.225 118 0.04 163 0.157 208 0
29 -0.309 74 0.235 119 -0.125 164 -0.196
30 0.095 75 -0.169 120 0.145 165 0.143
31 -0.096 76 0.005 121 -0.232 166 -0.146
32 0.317 77 -0.098 122 0.219 167 0.184
33 -0.053 78 0.477 123 -0.117 168 -0.159
34 -0.052 79 -0.169 124 0.142 169 0.11
35 -0.058 80 0.212 125 -0.155 170 -0.139
36 -0.057 81 -0.159 126 0.174 171 0.033
37 -0.157 82 -0.011 127 -0.277 172 -0.078
38 0.369 83 -0.117 128 0.161 173 0.061
39 -0.225 84 0.001 129 -0.175 174 -0.013
40 0.566 85 -0.1 130 0.194 175 0.02
41 -1.321 86 0.105 131 -0.223 176 0.001
42 1.37 87 -0.069 132 0.182 177 0.028
43 -2.02 88 0.198 133 -0.191 178 -0.074
44 1.103 89 -0.328 134 0.151 179 0.071
45 -0.017 90 0.326 135 -0.107 180 -0.085

3.2. Results and Discussion

3.2.1. Observed Damage

3.2.1.1 Concrete Damage

Photos that show damage at 4 𝛿/𝛿𝑦 , 8 𝛿/𝛿𝑦 , 15 𝛿/𝛿𝑦 , 20 𝛿/𝛿𝑦 , and the completion of testing are

provided in Figure 3.16 and Figure 3.17, Figure 3.18 and Figure 3.19, Figure 3.20 and Figure

3.21, Figure 3.22 and Figure 3.23, and Figure 3.24 and Figure 3.25, respectively. Concrete

crushing was not observed within the jacketed region, indicating that the jackets adequately

confined the concrete. Damage generally concentrated at the top of the footing in the vicinity of

the column, with spalling observed in the footing. Cracking was observed above the jacket.

Strength degradation was associated with fracture of longitudinal reinforcement, which is

described in more detail in Section 3.2.1.2, as spalling of concrete and longitudinal

35
reinforcement buckling were not observed above the jacket. The exception was C(CFRP)-4.0-

#7(1.3)-0.15, in which significant concrete damage and longitudinal reinforcement buckling,

characteristic of a plastic hinge, occurred above the jacket. Spalling of concrete was first

observed at the first cycle to 12.5(𝛿/𝛿𝑦 ) in the negative direction, with longitudinal bars visible

at 15𝛿/𝛿𝑦 (Figure 3.21.f). Spalling of concrete and buckling of longitudinal reinforcement

occurred, as shown in Figure 3.24.f and Figure 3.25.f, until a state of axial failure was reached

during the final cycle to 15(𝛿/𝛿𝑦 ). Extending the jacket further up the height of the column

would likely have prevented this damage.

The cycles at which footing cracks, horizontal flexural cracks, vertical shear cracks, and spalling

of footing concrete were first observed are provided in Table 3.8. Damage in all tests included

splitting cracks on the top surface of the footing that would propagate down the long sides and

horizontal cracks on the tension face of the column just above the jacket. The horizontal cracks

closed when the load was reversed and would propagate as the test progressed. In cycles after

8𝛿/𝛿𝑦 , horizontal cracks extended up to 64-inches from the column base. The extent of flexural

cracks varied significantly between tests in both severity and how far they propagated up the

column. Flexural cracks began to develop between 30-kip and 3.5 𝛿/𝛿𝑦 for all columns.

C(CFRP)-4.0-#7(1.3)-0.15 is the only column in which flexural cracks developed prior to footing

cracks. The flexural cracks progressed faster in terms of quantity, length, and severity for

C(CFRP)-4.0-#7(1.3)-0.15 with a measured 0.068 inch wide crack at the first cycle to 8𝛿/𝛿𝑦 in

the negative direction (Figure 3.19.f). The flexural cracks continued to grow as the test

progressed, with vertical cracks connecting the horizontal flexural cracks being noticed at the

first cycle to 12.5 𝛿/𝛿𝑦 .

36
Diagonal cracks near mid-depth of the column, which were characterized as shear cracks, were

observed for C(CFRP)-4.0-#5(1.4)-0.05, C(CFRP)-4.0-#7(2.7)-0.05, C(CFRP)-4.0-#7(1.3)-0.05-

2X, and C(CFRP)-4.0-#7(1.3)-0.15. For C(CFRP)-4.0-#7(1.3)-0.05-2X, a shear crack was first

observed at 12.5 𝛿/𝛿𝑦 , while shear cracks were not observed for C(CFRP)-4.0-#7(1.3)-0.05 and

C(CFRP)-4.0-#7(1.3)-0.05-EQ, which differed from C(CFRP)-4.0-#7(1.3)-0.05-2X only in

loading protocol.

Significant torsion was observed for C(CFRP)-4.0-#7(1.3)-0.05-EQ, and test was stopped before

reaching the third cycle at 10% drift, as the base of the column had ratcheted roughly 1.5 inches

out of plane. The Final photos of the column shown in figure 3.9.d, show that the column

scraped off the cover of the footing all the way down to the upper reinforcement as it moved out

of plane.

Table 3.8: Concrete Damage Onset

Footing
Column ID Footing Cracks Flexure Cracks Shear Cracks
Spalling
C(CFRP)-4.0-
5-k (2nd) 2(δ/δy ) (1st) N/A 15(δ/δy ) (1st)
#7(1.3)-0.05
C(CFRP)-4.0-
1(δ/δy ) (1st) 1.25(δ/δy ) (1st) 4(δ/δy ) (1st) 15(δ/δy ) (1st)
#5(1.4)-0.05
C(CFRP)-4.0- 1.75(δ/δy )
3.5(δ/δy ) (2nd) 5(δ/δy ) (1st) 10(δ/δy ) (1st)
#7(2.7)-0.05 (2nd)
C(CFRP)-4.0-
Start of test 1.12(δ/δy ) (1st) N/A 12.5(δ/δy ) (1st)
#7(1.3)-0.05-EQ
C(CFRP)-4.0-
Start of test 30-k (1st) 12.5(δ/δy ) (1st) 10(δ/δy ) (1st)
#7(1.3)-0.05-2X
C(CFRP)-4.0-
1.25(δ/δy ) 40-k (1st) 8(δ/δy ) (2nd) 10(δ/δy ) (1st)
#7(1.3)-0.15

37
a) d)

b) e)

c) f)

Figure 3.16. Concrete Damage at Base at 𝟒. 𝟎𝜹/𝜹𝒚 (2.55𝜹/𝜹𝒚 for C(CFRP)-4.0-#7(1.3)-0.05-

EQ) for: a) C(CFRP)-4.0-#7(1.3)-0.05, b) C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-
0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-#7(1.3)-0.05-2X, and f) C(CFRP)-4.0-
#7(1.3)-0.15

38
a) d)

b) e)

c) f)

Figure 3.17. Concrete Damage Above Jacket at 𝟒. 𝟎𝜹/𝜹𝒚 (2.55𝜹/𝜹𝒚 for C(CFRP)-4.0-#7(1.3)-
0.05-EQ) for: a) C(CFRP)-4.0-#7(1.3)-0.05, b) C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-
#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-#7(1.3)-0.05-2X, and f)
C(CFRP)-4.0-#7(1.3)-0.15

39
 a) d)

b) e)

c) f)

Figure 3.18. Concrete Damage at Base at 𝟖. 𝟎𝜹/𝜹𝒚 (7.3𝜹/𝜹𝒚 for C(CFRP)-4.0-#7(1.3)-0.05-EQ)

for: a) C(CFRP)-4.0-#7(1.3)-0.05, b) C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-0.05,
d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-#7(1.3)-0.05-2X, and f) C(CFRP)-4.0-
#7(1.3)-0.15

40
a) d)

b) e)

c) f)

Figure 3.19. Concrete Damage Above Jacket at 𝟖. 𝟎𝜹/𝜹𝒚 (7.3𝜹/𝜹𝒚 for C(CFRP)-4.0-#7(1.3)-
0.05-EQ) for: a) C(CFRP)-4.0-#7(1.3)-0.05, b) C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-
#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-#7(1.3)-0.05-2X, and f)
C(CFRP)-4.0-#7(1.3)-0.15

41
a) d)

b) e)

c) f)

Figure 3.20. Concrete Damage at Base at 𝟏𝟓𝜹/𝜹𝒚 for: a) C(CFRP)-4.0-#7(1.3)-0.05, b)

C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e)
C(CFRP)-4.0-#7(1.3)-0.05-2X, and f) C(CFRP)-4.0-#7(1.3)-0.15

42
a) d)

b) e)

c) f)

Figure 3.21. Concrete Damage Above Jacket at 15𝜹/𝜹𝒚 for: a) C(CFRP)-4.0-#7(1.3)-0.05, b)

C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e)
C(CFRP)-4.0-#7(1.3)-0.05-2X, and f) C(CFRP)-4.0-#7(1.3)-0.15

43
a) d)

c) e)

d) f)

Figure 3.22. Concrete Damage at Base at 20𝜹/𝜹𝒚 (at 𝟏𝟓𝜹/𝜹𝒚 for C(CFRP)-4.0-#7(1.3)-0.15)
for: a) C(CFRP)-4.0-#7(1.3)-0.05, b) C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-0.05,
d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-#7(1.3)-0.05-2X, and f) C(CFRP)-4.0-
#7(1.3)-0.15

44
a) d)

b) e)

c) f)

Figure 3.23. Concrete Damage Above Jacket at 20𝜹/𝜹𝒚 (at 𝟏𝟓𝜹/𝜹𝒚 for C(CFRP)-4.0-#7(1.3)-
0.15) for: a) C(CFRP)-4.0-#7(1.3)-0.05, b) C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-
0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-#7(1.3)-0.05-2X, and f) C(CFRP)-4.0-
#7(1.3)-0.15

45
 a) d)

b) e)

c) f)

Figure 3.24. Concrete Damage at Base at Completion of Testing for: a) C(CFRP)-4.0-#7(1.3)-

0.05, b) C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-
EQ, e) C(CFRP)-4.0-#7(1.3)-0.05-2X, and f) C(CFRP)-4.0-#7(1.3)-0.15

46
a) d)

b) e)

c) f)

Figure 3.25. Concrete Damage Above Jacket at Completion of Testing for: a) C(CFRP)-4.0-
#7(1.3)-0.05, b) C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-
#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-#7(1.3)-0.05-2X, and f) C(CFRP)-4.0-#7(1.3)-0.15

47
3.2.1.2 Fatigue Fracture of Longitudinal Reinforcement

In all cases, low-cycle fatigue fracture of longitudinal reinforcement occurred near the footing-

column interface. Concrete cracks that formed at the footing-column interface, as shown in

Figure 3.16, were attributed to longitudinal reinforcement bond slip. The sequence of

reinforcement fractures in each loading direction is provided in Table 3.9 and Table 3.10. For

each of the five tests with fully reversed-cyclic loading, a level of variability is evident in Table

3.9. For C(CFRP)-#7(1.4)-0.05-EQ, the first positive and negative fractures occurred at the same

cycle, despite higher demands in the negative direction during the earthquake and aftershock

protocols. In the columns containing more than one bar fracture, the second fracture occurred in

the same or immediately succeeding excursion. Due to torsional ratcheting at the base, shown in

Figure 3.24.d, testing of C(CFRP)-4.0-#7(1.3)-0.05-EQ was stopped before more than six bars

had fractured. Only one bar fractured for C(CFRP)-4.0-#7(1.3)-0.15 due to plastic hinging above

the CFRP jacket, as shown in Figure 3.25.f.

48
 Table 3.9: Sequence of Longitudinal Bar Fractures

Column Name
Bar C(CFRP)- C(CFRP)- C(CFRP)- C(CFRP)- C(CFRP)- C(CFRP)-
Fracture 4.0- #7(1.3)- 4.0-#5(1.4)- 4.0-#7(2.7)- 4.0-#7(1.3)- 4.0-#7(1.3)- 4.0-#7(1.3)-
0.05 0.05 0.05 0.05-EQ 0.05-2X 0.15
@ 4.26% 1st @ 5.22% 1st @ 6.39% @ 4.75% @ -1.62%
1st
to to 2nd to 1st to 2nd to ̶
Pos. (+)
+8.0% +8.0% +7.5% +10.0% +6.0%
@ -5.28% @ -5.02% @ -4.20% @ -4.74% @ -0.22% @ -0.26%
1st
2nd to 2nd to 2nd to 1st to 1st to 1st to
Neg. (–)
-6.0% -6.0% -7.5% -10.0% -6.0% -7.5%
@ 2.10% @ 5.92 % @ 0.83% @ 7.06% @ 5.80%
2nd
2nd to 1st to 1st to 1st to 1st to ̶
Pos. (+)
+8.0% +8.0% +10.0% +10.0% +8.0%
@ -2.18% @ -5.77% @ -6.08% @ -6.82% @ 0.24%
2nd
2nd to 2nd to 2nd to 1st to 3rd to ̶
Neg. (–)
-8.0% -6.0% -7.5% -10.0% -6.0%
@ 7.39% @ 3.00% @ 6.42% @ 1.10 @ 4.61%
3rd
1st to 2nd to 1st to 2nd to 2nd to ̶
Pos. (+)
+10.0% +8.0% +10.0% +10.0% +8.0%

@ -4.38% @ -3.63% @ 0.80% @ -9.09% @ -3.83%

3rd
2nd to 1st to 1st to 1st to 3rd to ̶
Neg. (–)
-8.0% -8.0% -10.0% -10.0% -6.0%

@ 2.61% @ 3.73% @ 7.97%

4th
3rd to 2nd to 1st to ̶ ̶ ̶
Pos. (+)
+10.0% +8.0% +10.0%
@ -9.14 @ -4.30% @ -6.62% @ -7.08%
4th
4th to 1st to 1st to ̶ 4th to ̶
Neg. (–)
-10.0% -8.0% -10.0% -8.0%
@ 8.88% @ 6.70 % @ 8.40%
5th
4th to 2nd to 1st to ̶ ̶ ̶
Pos. (+)
+10.0% +8.0% +10.0%
@ -4.49% @ -5.51% @ -9.72%
5th
̶ 2nd to 2nd to ̶ 2nd to ̶
Neg. (–)
-8.0% -10.0% -10.0%

49
 @ 7.38% @ 6.46%
6th
N.A. 2nd to 2nd to N.A. N.A. N.A.
Pos. (+)
+8.0% +10.0%
@ -7.25% @ -7.83
6th
N.A. 2nd to 2nd to N.A. N.A. N.A.
Neg. (–)
-8.0% -10.0%
7th @ 7.11% 1st @ 9.89%
Pos. (+) N.A. to 2nd to N.A. N.A. N.A.
+10.0% +10.0%
@ -7.68% @ -7.85%
7th
N.A. 2nd to 6th to N.A. N.A. N.A.
Neg. (–)
-8.0% -10.0%
@ 9.34% @ 0.15%
8th
N.A. 3rd to 4th to N.A. N.A. N.A.
Pos. (+)
+10.0% +10.0%
@ -9.59% @ -9.25%
8th
N.A. 2nd to 7th to N.A. N.A. N.A.
Neg. (–)
-10.0% -10.0%
@ 6.46% @ 8.23%
9th
N.A. 4th to 4th to N.A. N.A. N.A.
Pos. (+)
+10.0% +10.0%
@ -8.03%
9th
N.A. 3rd to ̶ N.A. N.A. N.A.
Neg. (–)
-10.0%
@ 9.01%
10th @ -5.89%
N.A. 4th to N.A. N.A. N.A.
Pos. (+) 5th to +10.0%
+10.0%

10th
N.A. ̶ ̶ N.A. N.A. N.A.
Neg. (–)

50
 Table 3.10: Summary of Cycles at which Longitudinal Bars Fractured

Cycle at Bar Fracture Column Name

𝜹
C(CFRP)- C(CFRP)-
𝜹𝒚 C(CFRP)-4.0- C(CFRP)-4.0- C(CFRP)-4.0-
Drift #7(1.3)- 4.0-#7(1.3)-
(Cycle #5(1.4)-0.05 #7(2.7)-0.05 #7(1.3)-0.15
0.05 0.05-2X
Number)
6.0%*,
15 (1)- 1 1
7.5%†
15 (2)+ 7.5%† 1 2
6.0%*,
15 (2)- 1 1, 2 2, 3
7.5%†
15 (3)- 6.0%‡ 3, 4
8.0%*,
20 (1)+ 2 3, 4 4, 5, 6, 7 5
10.0%†
8.0%*,
20 (1)- 5, 6 8, 9
10.0%†
8.0%*,
20 (2)+ 3 7, 8, 9, 10 10, 11 6
10.0%†
8.0%*,
20 (2)- 4, 5 11, 12, 13 12, 13
10.0%†
20 (4)+ 10.0%† 14, 15
20 (4)- 8.0%* 7
20 (5)+ 10.0%† 16
20 (6)- 10.0%† 17
20 (7)- 10.0%† 18
25 (1)+ 10.0%* 6 14
25 (2)- 10.0%* 15 8
25 (3)+ 10.0%* 7 16
25 (3)- 10.0%* 17, 18
25 (4)+ 10.0%* 8 19
25 (4)- 10.0%* 9
* for C(CFRP)-4.0- #7(1.3)-0.05, C(CFRP)-4.0-#5(1.4)-0.05, and C(CFRP)-4.0-#7(1.3)-0.05-2X
† for C(CFRP)-4.0-#7(2.7)-0.05 and C(CFRP)-4.0-#7(1.3)-0.15
‡ 3rd and 4th cycles of 15 𝛿/𝜹𝒚 only completed in the extended loading protocol of C(CFRP)-4.0-#7(1.3)-0.05-2X

51
3.2.2. Load-Deformation

The lateral load-deformation response is provided in Figure 3.26 for all tests. All columns were

flexure-yielding. In general, minimal pinching is evident in the hysteretic responses, indicative of

favorable energy dissipation. The level of pinching for C(CFRP)-4.0-#7(1.3)-0.15, which

incurred damage above the jacketed region, was greater than the other tests. The sequence of

longitudinal bar fractures is indicated on the plots, and the degradation in lateral load resistance

occurred primarily due to bar fractures. P-delta demands also reduced lateral load resistance,

with quantification of P-delta effects provided in Section 3.3. The values for peak shear demand,

𝑉𝑚𝑎𝑥 , normalized peak base moment, 𝑀𝑚𝑎𝑥 /𝑀𝑛 , displacement at peak demand, 𝛿𝑚𝑎𝑥 , and drift

𝛿
at peak demand, ( ) , are provided in Table 3.11 for each test column in both the positive and
𝐻 𝑚𝑎𝑥

negative loading direction. 𝑀𝑚𝑎𝑥 is defined as:

𝑀𝑚𝑎𝑥 = 𝑉𝑚𝑎𝑥 𝐻 + 𝑃Δ (3-1)

where 𝐻 is the height of the column, 𝑃 is the axial load, Δ is the column lateral displacement

measured at the point of lateral load application, and 𝑉𝑚𝑎𝑥 is the maximum shear. As shown in

Table 3.11, 𝑀𝑚𝑎𝑥 /𝑀𝑛 ranged from 1.244 to 1.435 for the six columns and ranged from 1.279 to

1.345 for the four tests with identical loading protocol. 𝑀𝑚𝑎𝑥 /𝑀𝑛 was smallest for C(CFRP)-

4.0-#7(1.3)-0.05-2X, suggesting that the increased number of cycles led to a reduction in

strength. 𝑀𝑚𝑎𝑥 /𝑀𝑛 was highest for C(CFRP)-4.0-#7(1.3)-0.05-EQ, likely due to a large cycle

earlier in the cycle sequence than that of the fully reversed cyclic tests. For all tests 𝑀𝑚𝑎𝑥 /𝑀𝑛

was within 4% for both the positive and negative loading directions.
52
Lateral failure was defined to have occurred when the lateral load at a maximum cycle peak first

dropped below 20% of the peak lateral load and did not return to this level during subsequent

cycles. A maximum cycle peak is defined as a cycle peak that had lateral displacement greater

than or equal to any previous peak in that loading direction. The maximum cycle peak prior to

the cycle at which lateral failure occurred in the positive and negative loading direction is

provided in Table 3.12 for each column. All columns reached lateral failure between 12.5 𝛿/𝛿𝑦

and 20.0𝛿/𝛿𝑦 . C(CFRP)-4.0-#7(1.3)-0.05-EQ had the greatest displacement before lateral failure

occurred in either loading direction. C(CFRP)-4.0-#7(1.3)-0.05 reached the same level of

displacement before lateral failure as C(CFRP)-4.0-#7(1.3)-0.05-EQ in the positive direction but

not the negative direction. C(CFRP)-4.0-#7(1.3)-0.05-EQ and C(CFRP)-4.0-#7(1.3)-0.05-2X had

the greatest difference between the positive and negative loading directions. Both columns had

increased strength in the negative loading direction with a 3.7% of 𝑀𝑛 increase for C(CFRP)-

4.0-#7(1.3)-0.05-EQ and a 3.4% of 𝑀𝑛 increase for C(CFRP)-4.0-#7(1.3)-0.05-2X. The

discrepancies between the positive and negative cycles of C(CFRP)-4.0-#7(1.3)-0.05-EQ may be

attributed to asymmetry in the loading cycles of the earthquake protocol. Of the other four tests,

none had greater than a 1% difference in peak moment between the corresponding positive and

negative loading directions.

The first bar fracture occurred during the excursion at which lateral failure occurred, with the

exception of C(CFRP)-4.0-#7(1.3)-0.15, in which the first bar fractured after lateral failure

occurred and the loss in strength was attributed to the concrete damage described in Section

3.2.1.1. The majority of longitudinal reinforcement fractures occurred during the portion of the

cyclic excursion beyond zero lateral displacement. While some fractures were reported during

53
unloading, they were in the minority, occurring twice for C(CFRP)-4.0-#7(1.3)-0.05-2X and

once for C(CFRP)-4.0-#7(2.7)-0.05. Fractures occurring on the loading excursions typically were

associated with an immediate loss of strength. Fractures occurring on the unloading excursions

led to a reduction in the rate of loading increase.

54
a) d)

b) e)

c) f)

Figure 3.26. Lateral Load Deformation: a) C(CFRP)-4.0-#7(1.3)-0.05, b) C(CFRP)-4.0-#5(1.4)-

0.05, c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-#7(1.3)-
0.05-2X, f) C(CFRP)-4.0-#7(1.3)-0.15

55
 Table 3.11: Peak Demands, Displacements, and Drifts

C(CFRP) C(CFRP) C(CFRP) C(CFRP) C(CFRP) C(CFRP)

-4.0- -4.0- -4.0- -4.0- -4.0- -4.0-
Load Parameter
#7(1.3)- #5(1.4)- #7(2.7)- #7(1.3)- #7(1.3)- #7(1.3)-
0.05 0.05 0.05 0.05-EQ 0.05-2X 0.15
+
𝑉𝑚𝑎𝑥 [K] 45.46 48.7 82.0 50.8 45.7 60.6
−
𝑉𝑚𝑎𝑥 [K] 48.0 49.3 81.3 50.6 46.9 60.5
+
𝛿𝑚𝑎𝑥 [IN] 3.91 2.30 4.74 3.57 3.09 2.41
Peak −
𝛿𝑚𝑎𝑥 [IN] 4.10 3.13 4.90 5.90 3.04 2.33
Shear 𝛿 +
( ) [%] 3.92 2.36 4.85 3.64 3.18 2.51
𝐻 𝑚𝑎𝑥
𝛿 −
( ) [%] 4.41 3.30 5.10 6.24 3.21 2.43
𝐻 𝑚𝑎𝑥
+
𝑀𝑚𝑎𝑥
1.331 1.317 1.345 1.398 1.244 1.327
𝑀𝑛
−
𝑀𝑚𝑎𝑥
1.279 1.315 1.336 1.435 1.278 1.327
Peak 𝑀𝑛
+
Base 𝛿𝑚𝑎𝑥 [IN] 7.88 4.73 4.74 4.24 3.09 3.77
−
Moment 𝛿𝑚𝑎𝑥 [IN] 9.85 4.64 4.90 5.90 4.72 5.64
𝛿 +
( ) [%] 8.05 4.89 4.85 4.33 3.18 3.92
𝐻 𝑚𝑎𝑥
𝛿 −
( ) [%] 10.4 4.87 5.10 6.24 4.96 5.88
𝐻 𝑚𝑎𝑥

Table 3.12: Deformation Capacity at Lateral Failure

Column I.D. Lateral Failure

𝜹 Drift [%]
(Cycle Number)
𝜹𝒚
(+) (̶) (+) (̶)
C(CFRP)- #7(1.3)-0.05 20.0 (1) + 15.0 (1) ̶ 8.0 6.0
C(CFRP)-4.0-#5(1.4)-0.05 15.0 (2) + 15.0 (1) ̶ 6.0 6.0
C(CFRP)-4.0-#7(2.7)-0.05 15.0 (1) + 15.0 (1) ̶ 7.5 7.5
C(CFRP)-4.0-#7(1.3)-0.05-EQ 20.0 (1) + 20.0 (1) ̶ 8.0 8.0
C(CFRP)-4.0-#7(1.3)-0.05-2X 15.0 (1) + 12.5 (4) ̶ 6.0 5.0
C(CFRP)-4.0-#7(1.3)-0.15 12.5 (2) + 12.5 (2) ̶ 6.25 6.25

56
3.2.3 Backbone Modeling

Bilinear backbone models, suitable for implementation into computer software used for

conducting nonlinear time history analyses, were fit to the load-displacement response of each

tested column in both the positive and negative loading direction, as shown in Figure 3.27.a

through Figure 3.32.a for the applied shear demand and Figure 3.27.b through Figure 3.32.b for

an effective shear demand that accounts for P-delta. Effective shear was computed as the base

moment, including P-delta, divided by the column height of 96 inches. The procedure in

ASCE/SEI-41 Section 7.4.3.2.4 (ASCE, 2017) was used for bilinear modeling. Although this

procedure is prescribed for fitting a backbone model to results of a pushover analysis of a

building, it was used for component modeling in this study. The first step in the procedure was

the formulation of a test data backbone, which consisted of a piecewise linear fit to peaks of

initial cycles, defined as any cycle that has a larger displacement that any previous cycle. The

bilinear model was formulated by placing the terminal points at the origin and maximum load.

The first line in the bilinear model intersected the test data backbone at 60% of the yield force.

The yield point, located at the intersection of the lines in the bilinear model, was determined such

that the area under the bilinear model was equal to the area under the test data curve between the

origin and maximum load. Base moment, including P-delta, was used to determine the excursion

with maximum loading.

From the backbone models, the effective stiffness and post-yield stiffness, Ke and Kp,

respectively, were taken as the slope of the first and second lines of the bilinear model,

respectively, as shown in Figure 3.33, and computed as:

57
 𝑉𝑦
𝐾𝑒 = (3-2)
𝛿𝑦

𝑉𝑚𝑎𝑥 − 𝑉𝑦
𝐾𝑝 = (3-3)
𝛿𝑚𝑎𝑥 − 𝛿𝑦

where 𝑉𝑦 and 𝛿𝑦 are the base shear and displacement at yielding, respectively, and 𝑉𝑚𝑎𝑥 and

𝛿𝑚𝑎𝑥 are the maximum shear and corresponding displacement, respectively, all determined by

the bilinear backbone modeling procedure. Values of 𝐾𝑒 and 𝐾𝑝 for both the applied shear and

effective shear backbones are provided in Table 3.13 for each of the six test columns in each of

the two loading directions. Assuming all deformation in the cantilever column was due to

bending, 𝐾𝑒 was converted to an effective flexural rigidity, 𝐼𝑒𝑓𝑓 , as:

𝑉𝑦 𝐻 3 𝐻 3
𝐸𝐼𝑒𝑓𝑓 = = 𝐾 (3-4)
3𝛿𝑦 3 𝑒

where H is the height of the column. Resulting values for (𝐸𝐼)𝑒𝑓𝑓 and (𝐸𝐼)𝑒𝑓𝑓 /(𝐸𝑐 𝐼𝑔 ) are

additionally reported in Table 3.13 where 𝐸𝑐 is the modulus of elasticity of concrete, and 𝐼𝑔 is the

moment of inertia of the gross concrete column section without the CFRP jacket included. The

average tested concrete strength from all column cylinder tests, shown in Table 3.4, was 3.95-

ksi, was used to compute 𝐸𝑐 = 3580-ksi using:

𝐸𝑐 [𝑘𝑠𝑖] = 57√𝑓𝐶′ [𝑝𝑠𝑖] (3-5)

58
ACI 318-19 Section 19.2.2.1. C(CFRP)-4.0-#7(1.3)-0.15 had the largest (𝐸𝐼)𝑒𝑓𝑓 in both

directions, likely attributed to increased concrete strength from the increased axial loads.

C(CFRP)-4.0-#7(1.3)-0.05 had the lowest (𝐸𝐼)𝑒𝑓𝑓 in the positive direction and C(CFRP)-4.0-

#7(1.3)-0.05-EQ had the lowest (𝐸𝐼)𝑒𝑓𝑓 in the negative direction. Since these columns only

differ in loading protocol it is not surprising that their (𝐸𝐼)𝑒𝑓𝑓 values are similar. The three

columns with the same layout, C(CFRP)-4.0-#7(1.3)-0.05, C(CFRP)-4.0-#7(1.3)-0.05-EQ, and

C(CFRP)-4.0-#7(1.3)-0.05-2X, had an average (𝐸𝐼)𝑒𝑓𝑓 /(𝐸𝑐 𝐼𝑔 ) of 0.433 with a COV of 10.5% in

the positive direction and 0.458 with a COV of 19.7% in the negative direction.

Bilinear models and test data backbones, normalized by lateral load at yielding determined from

the bilinear model fit, are provided in Figure 3.34 and Figure 3.35, respectively. Directional

variability was evident from the asymmetry of (𝐸𝐼)𝑒𝑓𝑓 values for C(CFRP)-4.0-#7(1.3)-0.05,

C(CFRP)-4.0-#5(1.4)-0.05, C(CFRP)-4.0-#7(2.7)-0.05, and C(CFRP)-4.0-#7(1.3)-0.05-EQ.

C(CFRP)-4.0-#7(1.3)-0.15 had the lowest directional variability of the 6 columns with the

negative direction having only 0.03% increase in (𝐸𝐼)𝑒𝑓𝑓 compared to the positive direction.

C(CFRP)-4.0-#5(1.4)-0.05 and C(CFRP)-4.0-#7(1.3)-0.05-EQ are the only columns to have a

lower (𝐸𝐼)𝑒𝑓𝑓 in the negative direction than the positive direction.

59
 a) b)

Figure 3.27. Backbone Model Fit for C(CFRP)-4.0-#7(1.3)-0.05: a) Base Shear, b) Effective
Base Shear

a) b)

Figure 3.28. Backbone Model Fit for C(CFRP)-4.0-#5(1.4)-0.05: a) Base Shear, b) Effective
Base Shear

60
 a) b)

Figure 3.29. Backbone Model Fit for C(CFRP)-4.0-#7(2.7)-0.05: a) Base Shear, b) Effective
Base Shear

a) b)

Figure 3.30. Backbone Model Fit for C(CFRP)-4.0-#7(1.3)-0.05-EQ: a) Base Shear, b) Effective
Base Shear

61
 a) b)

Figure 3.31. Backbone Model Fit for C(CFRP)-4.0-#7(1.3)-0.05-2X: a) Base Shear, b) Effective
Base Shear

a) b)

Figure 3.32. Backbone Model Fit for C(CFRP)-4.0-#7(1.3)-0.15: a) Base Shear, b) Effective
Base Shear

62
Figure 3.33. Bilinear Model Backbone Slope Parameters

63
 Table 3.13: Stiffness and Strength of Backbone Models

C(CFRP) C(CFRP) C(CFRP) C(CFRP) C(CFRP) C(CFRP)

-4.0- -4.0- -4.0- -4.0- -4.0- -4.0-
Load Parameter
#7(1.3) #5(1.4) #7(2.7) #7(1.3) #7(1.3) #7(1.3)
-0.05 -0.05 -0.05 -0.05-EQ -0.05-2X -0.15
𝐾𝑒 (+)† 75.23 104.32 99.78 90.74 90.98 156.18
𝐾𝑝 (+)† 1.217 4.587 1.564 2.915 3.511 5.573
𝐾𝑒 (−)† 107.68 93.79 137.27 72.03 91.89 168.39
𝐾𝑝 (−)† 4.393 3.422 4.792 1.134 3.786 5.918
Base (𝐸𝐼)𝑒𝑓𝑓 (+)‡ 2.219 3.077 2.943 2.676 2.683 4.606
Shear (𝐸𝐼)𝑒𝑓𝑓
(+)§ 0.3803 0.5274 0.5045 0.4587 0.4600 0.7896
𝐸𝐶 𝐼𝑔

(𝐸𝐼)𝑒𝑓𝑓 (−)‡ 3.176 2.766 4.048 2.124 2.710 4.966

(𝐸𝐼)𝑒𝑓𝑓
(−)§ 0.5443 0.4741 0.6940 0.3641 0.4646 0.8513
𝐸𝐶 𝐼𝑔

𝐾𝑒 (+)† 68.36 96.19 96.76 91.67 92.05 131.13

𝐾𝑝 (+)† 86.08 212.73 169.49 389.23 453.15 295.79
𝐾𝑒 (−)† 86.64 94.87 110.67 73.01 92.88 131.16
𝐾𝑝 (−)† 281.22 446.02 156.50 210.75 481.44 246.10
Effective
Base (𝐸𝐼)𝑒𝑓𝑓 (+)‡ 2.106 2.837 2.854 2.703 2.715 3.867
Shear (𝐸𝐼)𝑒𝑓𝑓 (−)‡ 2.555 2.798 3.264 2.153 2.739 3.868
(𝐸𝐼)𝑒𝑓𝑓
(+)§ 0.3456 0.4863 0.4892 0.4634 0.4654 0.6629
𝐸𝐶 𝐼𝑔
(𝐸𝐼)𝑒𝑓𝑓
(−)§ 0.4378 0.4796 0.5595 0.3691 0.4696 0.6631
𝐸𝐶 𝐼𝑔
†
[kip/in]
‡
[kip-in2 × 107 ]
§
[UL]

64
Figure 3.34. Normalized Bilinear Backbone Model Plots

Figure 3.35. Normalized Test Data Backbone Plots

65
3.2.4. Effective Secant Stiffness

Effective secant stiffness, 𝐾𝑠𝑒𝑐 , was determined for each initial cycle (i.e., each point on the test

data backbone) as the slope of the line from the origin to that point. Assuming all deformation in

the cantilever column was due to bending, 𝐾𝑠𝑒𝑐 was converted to an effective flexural rigidity,

(EI)sec, as:

𝐾𝑠𝑒𝑐 𝐻 3 𝑉𝐻 3
(𝐸𝐼)𝑠𝑒𝑐 = = (3-6)
3 3𝛿

Plots of 𝐸𝐼𝑠𝑒𝑐 /(𝐸𝐶 𝐼𝑔 ) versus drift with data points connected by lines are provided in Figure

3.36. As expected, C(CFRP)-4.0-#7(2.7)-0.05 and C(CFRP)-4.0-#7(1.3)-0.15, which had greater

longitudinal reinforcement ratio and higher axial load, respectively, than the other columns,

exhibited greater stiffness at given drift levels. Minor variation was observed between C(CFRP)-

#7(1.3)-0.05, C(CFRP)- #7(1.3)-0.05-EQ, and C(CFRP)-4.0-#7(1.3)-0.05-2X, which had the

same longitudinal reinforcement layout and axial load.

66
 Figure 3.36. Effective Secant Stiffness Plots

3.2.5. Reinforcement Strain

Measurements obtained from the strain gauges shown in Figure 3.10 and Figure 3.11 were

plotted in Figures 4.22 through Figures 4.26 for initial cycle peaks at various drift levels.

Negative values on the y-axes indicate strain gauges in the footing. Yield strain, 𝜖𝑦 , is shown on

the plots and was determined from the reinforcement stress-strain data provided in Figure 3.7.

During positive loading excursions, the South reinforcement was in tension and the North
67
reinforcement was in compression. Gauges tended to become damaged as testing progressed,

leading to sparser data at increased drift levels. Some gauges malfunctioned before testing

commenced, and these data were omitted. Gauge results for C(CFRP)-4.0-#7(1.3)-0.05 were not

available due issues with gauges. Two gauges were used at 0.05-in above the column-footing

interface, and, when both gauges were functioning, average values were reported.

Generally, strains increased with greater magnitude excursions. Most columns had larger strain

in the tension than compression direction, as expected, and this trend was most prominent for

C(CFRP)-4.0-#5(1.4)-0.05 and C(CFRP)-4.0-#7(1.3)-0.05-EQ. C(CFRP)-4.0-#7(1.3)-0.05-EQ,

C(CFRP)-4.0-#7(1.3)-0.05-2X, and C(CFRP)-4.0-#7(1.3)-0.15 all exhibited higher strains further

from the column-footing interface, whereas C(CFRP)-4.0-#5(1.4)-0.05 exhibited higher strains

closer to the column-footing interface. Reinforcement yielding was reached in all columns.

C(CFRP)-4.0-#7(1.3)-0.05-EQ and C(CFRP)-4.0-#7(1.3)-0.15 both exhibited reinforcement

yielding at the sensor farthest below the column-footing interface. C(CFRP)-4.0-#7(1.3)-0.05-

EQ also exhibited reinforcement yielding at the sensor farthest above the column footing

interface. All other columns exhibited at least 80% of the yield strain in one of the sensors at

least 8-in away from the column-footing interface.

68
Figure 3.37. Strain Measured in Longitudinal Reinforcement at Cycle Peaks, C(CFRP)-4.0-
#5(1.4)-0.05

Figure 3.38. Strain Measured in Longitudinal Reinforcement at Cycle Peaks, C(CFRP)-4.0-

#7(2.7)-0.05
69
Figure 3.39. Strain Measured in Longitudinal Reinforcement at Cycle Peaks, C(CFRP)-4.0-
#7(1.3)-0.05-EQ

Figure 3.40. Strain Measured in Longitudinal Reinforcement at Cycle Peaks, C(CFRP)-4.0-

#7(1.3)-0.05-2X
70
Figure 3.41. Strain Measured in Longitudinal Reinforcement at Cycle Peaks, C(CFRP)-4.0-
#7(1.3)-0.15

71
3.2.6. Column Curvature

Curvature was determined using the vertically oriented LVDTs in the column, shown in Figure

3.13. At each height increment, curvature was determined using the two North-South Sensors,

with plots of the values provided in Figure 3.42. Results were more limited at advanced

deformation levels due to damage interfering with instrumentation. It is evident that curvature

demand concentrated in the lower 6 inches of the columns. However, at a height of 0.5-in, the

deformation is primarily due to bond slip and extension, such that the strain determined from the

LVDTs over the lower 0.5-in is not reflective of true curvature. Column curvatures excluding

data for the lower 0.5-in are provided in Figure 3.43. The calculated yield curvatures shown in

Figure 3.42 and Figure 3.43 were computed using moment-curvature analysis with the steel

reinforcement material properties provided in Table 3.2 and the concrete material properties

provided in Table 3.3. Measured curvature did not exceed calculated yield curvature at locations

more than 4-in above the column-footing interface. The exception was C(CFRP)-4.0-#7(1.3)-

0.15, in which curvature exceeded the calculated yield curvature at 12-in above the column-

footing interface at 2% and greater drift in the positive direction and 4% and greater drift in the

negative direction. It is noted that data was excluded from Figure 3.42 and Figure 3.43 due to

issues with sensors.

72
Figure 3.42. Measured Curvature: a) C(CFRP)-4.0-#7(1.3)-0.05, b) C(CFRP)-4.0-#5(1.4)-0.05,
c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e) C(CFRP)-4.0-#7(1.3)-0.05-
2X, f) C(CFRP)-4.0-#7(1.3)-0.15

73
 Figure 3.43. Measured Curvature Excluding Bond Slip: a) C(CFRP)-4.0-#7(1.3)-0.05, b)
C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-EQ, e)
C(CFRP)-4.0-#7(1.3)-0.05-2X, f) C(CFRP)-4.0-#7(1.3)-0.15

74
3.2.7. Shear Sliding and Base Rotation

Shear sliding and base twisting were determined using the horizontally oriented sensors spanning

from the footing to the base of the column, with sensor locations shown in Figure 3.12. Plots of

shear sliding and base twisting versus base shear are provided in Figure 3.43 and Figure 3.44,

respectively, with clockwise and counter-clockwise indicating column twist direction relative to

the footing. Data for C(CFRP)-4.0-#7(1.3)-0.05 were omitted due to sensor error. Included data

for all columns were more limited at advanced deformation levels due to damage interfering with

instrumentation. C(CFRP)-4.0-#7(2.7)-0.05 had the most shear sliding of any test, reaching

almost 0.7-in of sliding in both directions. C(CFRP)-4.0-#7(1.3)-0.15 had the least shear sliding

and base rotation, with less than 0.15-in of shear sliding in both directions and a maximum base

twisting rotation of 0.0042 radians in the clockwise direction and 0.0000 radians in the counter

clockwise direction. C(CFRP)-4.0-#5(1.4)-0.05, C(CFRP)-4.0-#7(2.7)-0.05, C(CFRP)-4.0-

#7(1.3)-0.05-EQ, and C(CFRP)-4.0-#7(1.3)-0.05-2X had more shear sliding in the negative

loading direction, while C(CFRP)-4.0-#7(1.3)-0.15 had more shear sliding in the positive loading

direction. C(CFRP)-4.0-#7(1.3)-0.05-EQ had the most discrepancy in shear sliding in the

positive and negative directions, with 1.92 times as much shear sliding in the negative direction

than in the positive direction. This column had the most base rotation of any of the columns, with

a maximum base twisting rotation of 0.094 radians in the clockwise direction and 0.033 radians

in the counter clockwise direction. The twisting caused out-of-plane translation due to ratcheting,

as described in Section 3.2.1.1.

75
 Figure 3.44. Shear Sliding: a) C(CFRP)-4.0-#5(1.4)-0.05, b) C(CFRP)-4.0-#7(2.7)-0.05, c)
C(CFRP)-4.0-#7(1.3)-0.05-EQ, d) C(CFRP)-4.0-#7(1.3)-0.05-2X, e) C(CFRP)-4.0-#7(1.3)-0.15

76
 Figure 3.45. Base Rotation: a) C(CFRP)-4.0-#5(1.4)-0.05, b) C(CFRP)-4.0-#7(2.7)-0.05, c)
C(CFRP)-4.0-#7(1.3)-0.05-EQ, d) C(CFRP)-4.0-#7(1.3)-0.05-2X, e) C(CFRP)-4.0-#7(1.3)-0.15

77
3.2.8. Components of Deformation

The components of deformation for the six tests at peaks of initial cycles are provided in Figure

3.45. The components are provided as a percentage of the overall column displacement and drift.

“Flexure in the Jacket” was determined from the curvature over the height of the jacket, using

curvature values shown in Figure 3.42 with constant curvature assumed over the length of the

sensor (i.e., center of rotation at mid-height of the sensor length). This did not include the

curvature in the bottom 0.5-in of the column, which contained bar slip and elongation. The lower

0.5-in of the column, which was included separately in the figure as “Bond Slip/Elongation”,

was determined from the rotation measured over the lower 0.5-in of the column. The shear

sliding described in Section 3.2.7 is included in Figure 3.45 as “shear sliding”. Estimated shear

deformation, presented in Figure 3.45 as “Estimated Shear”, was computed as:

𝑉ℎ
Δ𝑠ℎ𝑒𝑎𝑟 = (3-8)
(𝐴𝑉 𝐺)𝑒𝑓𝑓

where V is the shear force at the base of the column, h is the height of the column shear span, and

(AvG)eff, is the effective shear rigidity, which was computed using the recommended value in

Appendix A of ACI 318-19 as:

(𝐴𝑉 𝐺)𝑒𝑓𝑓 = 0.4𝐸𝑐 𝐴𝑔 (3-7)

78
where 𝐸𝑐 is the concrete modulus of elasticity of concrete and 𝐴𝑔 is the gross cross-sectional

area. The “Other” category is the displacement or drift not accounted for by the previously

mentioned components. This includes flexural deformation above the jacket, as well as any

sources of error. For C(CFRP)-4.0-#5(1.4)-0.05 and C(CFRP)-4.0-#7(2.7)-0.05 the “Other”

category constitutes everything but shear sliding and estimated shear displacement, due to sensor

error described in Section 3.2.6. Similarly, C(CFRP)-4.0-#7(1.3)-0.05 was omitted due to the

sensor malfunctions described in Sections 4.2.6 and 4.2.7.

As evident from Figures 4.31.c. through 4.31.e., the majority of the deformation occurred at the

bottom 0.5-in of the column, with shear sliding contributing minimally to the overall

deformation. Shear sliding between -5.6% and 12.7%for all columns. Flexure in the jacket also

contributed minimally to the overall deformation with a minimum contribution of -10.8% and a

maximum contribution of 7.0% for C(CFRP)-4.0-#7(1.3)-0.05-EQ, C(CFRP)-4.0-#7(1.3)-0.05-

2X, and C(CFRP)-4.0-#7(1.3)-0.15. The estimated shear deformation had a maximum

contribution of 11.8% across all columns. The majority of the deformation was due to bond

slip/elongation and other sources. The contribution from bond slip and elongation generally

increased as drift increased. An increase in “other” with increase in damage above the jacket was

evident in the negative direction for C(CFRP)-4.0-#7(1.3)-0.15, which failed above the jacket,

as described in Section 3.2.1.1.

79
Figure 3.46. Components of Deformation: a) C(CFRP)-4.0-#5(1.4)-0.05, b) C(CFRP)-4.0-
#7(2.7)-0.05, c) C(CFRP)-4.0-#7(1.3)-0.05-EQ, d) C(CFRP)-4.0-#7(1.3)-0.05-2X,
e) C(CFRP)-4.0-#7(1.3)-0.15

80
 4. Column Modeling

4.1. Methodology

Failure of the FRP jacket retrofitted bridge columns was due to low-cycle fatigue fracture of

longitudinal reinforcement. Predicting the drift and cycle at failure for a given column is

contingent upon predicting low-cycle fatigue failure of longitudinal reinforcement. Many

existing models for low-cycle fatigue fracture of reinforcement are based on plastic strain history

(Uriz and Mahin, 2008; Huang and Mahin, 2008; Kanvinde, 2004; Padilla-Llano et al., 2018). A

column hinge rotation model was used to formulate a relationship between the strain in the

reinforcement and the column drift.

The column model, shown in Figure 4.1, consisted of a linear elastic line element, a fiber section

element over the 1” clear cover to the footing top reinforcement, and a bond slip element at the

footing column interface and at 1” into the footing. The spread of plasticity into the column was

not modeled, as it was evident from test data in Section 3.2.6 that the spread of plasticity in the

jacketed region was minimal. The use of a fiber section in the concrete cover was intended to

account for the concrete spalling that occurred, as described in Section 3.2.1.1. The bond slip

elements were intended to model strain penetration into the footing and into the jacket. In this

model, the plastic hinge length, Lp, was 1”, which was concrete cover dimension, and plastic

deformation was modeled to occur in the fiber section element and in the two zero-length bond

slip elements. Rotation in the plastic hinge, θp, is related to drift ratio as:

81
 (𝛥−𝛥𝑦 ) 𝛥𝑝
𝜃𝑝 = 𝐿𝑝 = 𝐿𝑝 (4-1)
(𝐻+ 2 ) (𝐻+ 2 )

where Δ is the total lateral displacement at the top of the column, Δy is the lateral displacement at

yield, Δp is the plastic lateral displacement, H is the column clear height, and Lp/2 is the height of

the plastic hinge center of rotation below the base of the column. Previous research (Chai et al,

1991) has shown that plastic curvature in steel jacket retrofitted columns concentrates at the base

at the gap between the steel jacket and the foundation. Because Lp is small relative to the height

of the column, plastic hinge rotation may be approximated as plastic drift ratio:

𝛥𝑝 𝛥𝑝 𝛥𝑝
𝜃𝑝 = = 𝐿𝑔𝑎𝑝 ≈ (4-2)
(𝐻−𝐻ℎ𝑖𝑛𝑔𝑒 ) (𝐻− 2 ) 𝐻

Resulting in a linear relationship between plastic rotation and drift ratio.

Assuming plane section behavior and uniform strain over the height of the fiber section element,

plastic tensile strain in the outermost longitudinal reinforcement, εp,t, is:

𝜃𝑝 𝛥𝑝 /𝐻 𝛥𝑝 /𝐻
𝜀𝑝,𝑡 = 𝜀𝑠 − 𝜀𝑦 = 𝜙𝑝 (𝑑 − 𝑐) = (𝑑 − 𝑐) ≈ (𝑑 − 𝑐) = (𝑑 − 𝑐) (4-3)
𝐿𝑝 𝐿𝑝 𝐿𝑔𝑎𝑝 +𝐶𝑑𝑏

where εs is the tensile strain in the outermost tensile longitudinal reinforcement, εs is the yield

strain of the outermost tensile longitudinal reinforcement, ϕp is the plastic curvature, d is the

depth to the outermost tensile longitudinal reinforcement, and c is the neutral axis depth.

Similarly, the plastic compression strain in the outermost longitudinal reinforcement, εp,c, is:

82
 𝜃𝑝 𝛥𝑝 /𝐻 𝛥𝑝 /𝐻
𝜀𝑝,𝑐 = 𝜀𝑠 − 𝜀𝑦 = 𝜙𝑝 (𝑐 − 𝑑′) = (𝑐 − 𝑑′) ≈ (𝑐 − 𝑑′) = (𝑐 − 𝑑′) (4-4)
𝐿𝑝 𝐿𝑝 𝐿𝑔𝑎𝑝 +𝐶𝑑𝑏

where d’ is the compression strain in the outermost compressive longitudinal reinforcement and

Lp is from Equation (4-2) based on the recommendation by Chai et al (1994).

Figure 4.1. Column Deformation Model

83
The column model was formulated in OpenSees (McKenna, year) as a linear elastic beam

element with a displacement based fiber element of length Lp = 1-in. The model is intended to be

efficient for use in nonlinear time history analyses and enabled the recording of stress and strain

in longitudinal reinforcement. In this manner the strain history in the longitudinal reinforcement

could be directly related to the drift history of the column, without needing to explicitly

implement Equations (4) and (5) in the code.

The stiffness of the unjacketed column was determined using the method of Elwood and

Eberhard (2009), which accounts for the contribution from flexure, shear, and bond slip as:

𝛥 = 𝛥𝑓𝑙𝑒𝑥𝑢𝑟𝑒 + 𝛥𝑠𝑙𝑖𝑝 + 𝛥𝑠ℎ𝑒𝑎𝑟 (4-5)

𝐻 2 𝜙𝑦
𝛥𝑓𝑙𝑒𝑥𝑢𝑟𝑒 = (4-6)
3

𝐻𝑑𝑏 𝑓𝑦 𝜙𝑦 𝐻𝑑𝑏 𝑓𝑦 𝜙𝑦
𝛥𝑠𝑙𝑖𝑝 = = (4-7)
8𝑢
8(9.6√𝑓𝑐′ )

𝑀𝑦 𝑀𝑦
𝛥𝑠ℎ𝑒𝑎𝑟 = = (0.85𝐷)(0.2𝐸 (4-8)
𝐴𝑣 𝐺𝑒𝑓𝑓 𝑐)

where H is the height of the column, is the yield curvature, db is the diamater of longitudinal

reinforcement, fy is the yield strength of longitudinal reinforcement, u is the bond stress, My is

the yield moment, Av is the area of the cross-section resisting shear, D is the diamater of the

column, Geff if the effective shear modulus, and Ec is the modulus of elasticity of concrete. My

was determined from moment-curvature analysis at first yield of reinforcement. For the bond slip

contribution, Zhao and Sritharan (2007) was used in place of Eq. (5-8):

84
 1
𝑑𝑏 𝑓𝑦 𝛼
Δ𝑠𝑙𝑖𝑝 = 0.1 [ (2𝛼 + 1)] + 0.013 (4-9)
4√𝑓 ′ 𝑐

where 𝛼 is the parameter used in local bond-slip relation and was taken as 0.4 as done in Zhao

and Sritharan (2007). The increase in stiffness from the FRP jacket was determined using the

approach recommended by Chai et al (1994) for steel jackets, which accounts for the bond

transfer length needed to develop full composite action of the jacket and column. As this was an

FRP jacket, the gap length, 𝐿𝑔𝑎𝑝 , and grout thickness, 𝑡𝑔𝑟𝑜𝑢𝑡 , were both zero. The CFRP jacket

cured composite tensile modulus and tensile strength shown in Table 3.5 were used in place of

the steel jacket elastic modulus and yield strength. Once the stiffness of the column, k, was

determined, it was implemented into the model as the elastic flexural rigidity, (EI)elastic, of the

elastic element, computed as:

𝑀𝑦 𝐿𝑔𝑎𝑝 𝐿𝑝 3 𝑀𝑦 𝐿𝑔𝑎𝑝 𝐿𝑝 3
− 2 − 2) −
𝐻 (𝐻 𝐻 (𝐻 2 − 2)
(𝐸𝐼)𝑒𝑙𝑎𝑠𝑡𝑖𝑐 = = (4-10)
3∆𝑒𝑙𝑎𝑠𝑡𝑖𝑐 𝑀𝑦 𝐿𝑔𝑎𝑝
3( − 𝜙𝑦 𝐿𝑝 (𝐻 −
𝐻𝑘 2 ))

where My and ϕy was determined from moment-curvature analysis using the Mander et al (1988)

confined concrete model. This computation of (EI)elastic accounts for the flexibility in the fiber

element based on the displacement at yield due to curvature in the fiber, such that the elastic

stiffness in the model is expected to match the computed value for k at yield.

85
Concrete was modeled using Concrete02 in OpenSees. The Mander et al (1988) model for

confined concrete was used to determine the confined concrete compressive strength, f’cc, and the

strain at which f’cc is reached, εcc. The ultimate concrete stress and strain were modeled as 0.2f’cc

and 5εcc, respectively. λ in the Concrete02 model, which is the ratio between unloading slope and

initial slope, was taken as 0.1.. Reinforcement was modeled using ReinforcingSteel, with the

tangent at initial strain hardening taken as 0.01𝐸𝑠 . Strain hardening was initiated at 𝜀𝑠ℎ and strain

at peak stress was reached at 𝜀𝑢𝑙𝑡 . Reinforcement in the bond-slip elements was modeled with

Bond SP01. Yield slip, 𝑠𝑦 , was computed using Eq. (5-8) from Elwood and Eberhard (2008)

utilizing the yield curvature determined from a moment-curvature analysis of the column not

accounting for confinement. Slip at ultimate strength, 𝑠𝑢 , was taken as 30 times the yield slip,

consistent with Zhao and Sritharan (2007). The values for initial hardening slope in the

monotonic slip versus bar stress response, b, and the pinching factor for the cyclic slip versus bar

response, R, were taken as 0.3 and 0.7, respectively.

The stress-strain response of the outermost longitudinal reinforcement at both ends of the column

was recorded during the OpenSees analysis. The strain history of the reinforcement was then

implemented in an existing low-cycle fatigue model that was used to estimate the point of

fracture of the outermost longitudinal bar based on the accumulated plastic strain. The low-cycle

fatigue model was implemented in Matlab, such that failure was determined through post-

processing of the data collected from the OpenSees analysis.

Modeling of low-cycle fatigue was based on the Coffin (1954 and 1971) and Manson (1965)

formulation:

86
 −𝑚
𝜀𝑝 = 𝜀0 (2𝑁𝑓 ) (4-11)

where εp is the plastic strain amplitude of each constant amplitude half cycle, ε0 is a material

constant that approximately indicates the plastic strain amplitude at which one half cycle will

cause failure, 2Nf is the number of constant amplitude half cycles to failure, and m is a material

constant that indicates the sensitivity between εi and Nf. Equation (4-11) may be re-arranged to

determine the number of constant amplitude half cycles at εp needed to reach failure:

𝜀𝑝
(−𝑚)−1 log( )
2𝑁𝑓 = 10 𝜀0 (4-12)

The accumulation of damage was based on Minor’s Rule:

2𝑛𝑖
𝐷𝐼 = ∑ (4-13)
2𝑁𝑓

where 2ni is the number of half cycles at a specific value of εp, and 2𝑁𝑓 is determined for that

same value of εp using Equation (4-12). For an individual half cycle, 2ni = 1, such that the

damage of each half cycle is (2Nf)-1. When the accumulation of half-cycles causes the damage

index, DI, to exceed 1.0, fatigue failure occurs. Although rainflow counting is often used to

define full cycles, the use of half-cycles enables the analysis to progress sequentially without a

need for rainflow counting. In this case, a half-cycle is defined to be bounded by two load

reversals, such that the amplitude of a half cycle is one-half of the strain bounded by two load

reversals.

87
4.2. Results and Discussion

The method to compute the stiffness of each column, described in the previous section, was

validated to data from the column tests described in Chapter 4. The stiffness values from the

model and test data are provided in Table 4.1 for each test. The stiffness from test data was

consistent with the values in Table 3.13, which were determined from fitting a backbone model

to test data, as desribed in Section 3.2.3. It is evident from Table 4.1 that the model predicted the

stiffness with a percent error that was less than 25% for all tests. The average percent error was

2.04% in the positive direction (i.e., overprediction of stiffness). Given the level of variability in

stiffness observed in the tests, the use of this method to determine stiffness was deemed

appropriate.

Table 4.1: Effective Stiffness from Model and Tests

Effective Stiffness
Column Name
Test [kip/in] Model [kip/in] % Error
C-#7(1.3)-0.05 77.5 89.63 15.65
C-#5(1.4)-0.05 95.53 98.19 2.78
C-#7(2.7)-0.05 103.715 115.65 11.51
C-#7(1.3)-0.05-EQ 82.34 89.63 8.85
C-#7(1.3)-0.05-X 92.465 89.63 -3.07
C-#7(1.3)-0.15 131.145 100.34 -23.49

The model was validated to the six columns tested in this study, which were described in Chapter

4. The fit between model and tests is provided in Figure 4.2. The model provided a reasonable fit

to the test data, with the exception of an underestimate of the strength for x and an overestimate

of the level of pinching in the load-deformation hysteresis of x. The model did not capture the

strength degradation observed in the test of C(CFRP)-#7(1.3)-0.15, as damage occurred above

88
the jacketed region. The model prediction of fracture of the first longitudinal bar is indicated in

the figures. The model under predicted fracture for each of the six tests.

a) d)

b) e)

c) f)

Figure 4.2. Column Load-Deformation Response for Model and Tests: a) C(CFRP)-4.0-#7(1.3)-
0.05, b) C(CFRP)-4.0-#5(1.4)-0.05, c) C(CFRP)-4.0-#7(2.7)-0.05, d) C(CFRP)-4.0-#7(1.3)-0.05-
EQ, e) C(CFRP)-4.0-#7(1.3)-0.05-2X, f) C(CFRP)-4.0-#7(1.3)-0.15

89
 5. Conclusions

Retrofit of reinforced concrete bridge columns with jackets is a commonly implemented strategy

to increase column ductility in earthquakes. Six FRP jacket retrofitted reinforced concrete bridge

columns were designed, constructed and tested. Five of the six columns were nominally identical

to a set of recently tested steel jacket retrofitted columns. Test variables included bar size of

longitudinal reinforcement, longitudinal reinforcement ratio, axial load ratio, and loading

protocol, with inclusion of a variable amplitude earthquake time history for one of the columns.

These test variables were selected due to the influence on the strain history in the longitudinal

reinforcement, as strength degradation in the test columns was expected to be due to fatigue

fracture of longitudinal reinforcement. The range of values for the test variables was intended to

reflect the range of variation of these parameters in the Washington State DoT inventory.

Using test results from the columns and reinforcement tests, a model was developed to estimate

the load-deformation response and fatigue fracture of longitudinal reinforcement in steel jacket

retrofitted reinforced concrete columns. The model consisted of a linear elastic element with a

plastic hinge at the base. The plastic hinge length included the gap between the bottom of the

steel jacket and the footing as well as additional length to account for bond slip of reinforcement

due to strain penetration into the footing and into the steel jacketed region. Strain history

determined from the model was used in an existing fatigue model to estimate the drift at fatigue

fracture of longitudinal reinforcement. The model was validated with existing test data.

90
The following conclusions on FRP jacket retrofitted reinforced concrete bridge columns were

reached:

 Concrete damage is limited to the cover concrete at the top of the footing. Cracking

concentrated at the column-footing interface, and the crack at this location became wide

at increased deformation demand levels. This crack is indicative of bond slip of

longitudinal reinforcement. The extent of spalling of cover concrete at the top of the

footing varied for the test columns, with the extent of damage most limited for the

column with the largest axial load.

 Fracture of the first longitudinal bar in the test columns occurred during cycles at 12.5 or

15.0 times the yield displacement. Axial failure was reached for one of the test columns

upon fracture of the final longitudinal bar. This occurred at 20.0 times the yield

displacement, which was a drift of 10% for these two columns.

 Minimal pinching, indicative of favorable hysteretic energy dissipation, was observed in

the load-deformation responses of the test columns. Strength degradation was primarily

due to low-cycle fatigue fracture of longitudinal reinforcement.

 Peak base moment on the column was larger by as much as 40% for the test column with

variable amplitude earthquake protocol. The largest excursions in the earthquake protocol

were associated with a significant reduction in strength upon the next occurrence of

reaching this drift level, and larger drift in the excursion was associated with a larger

reduction in strength. The level of strength drop observed for this test prior to bar fracture

was not observed in tests with fully reserved cyclic protocols prior to bar fracture, and

this is attributed to the difference in cycle content.

91
 90% of column deformation was attributable to bond slip of longitudinal reinforcement.

The next greatest contribution was column sliding shear at the footing--column interface.

Flexure and shear deformation in the retrofit columns, beyond the plastic hinge region,

were minimal, becoming increasingly insignificant at larger drifts. Measured curvature

over a distance of 0.5” to 8.0” above the column base was significantly larger than

locations further up in the column. This was indicative of yielding, lack of composite

action, or both, at this location.

 Lateral failure was defined to have occurred when the lateral load at a maximum cycle

peak first dropped below 20% of the peak lateral load and did not return to this level

during subsequent cycles. A maximum cycle peak is defined as a cycle peak that was

greater than or equal to any previous peak in that loading direction. Lateral failure

corresponded to fracture of the first longitudinal bar. All test columns reached 12.5 to

15.0 times the yield displacement prior to lateral failure.

 The predicted stiffness used in the model matched the measured stiffness from the test

columns with a percent error that was less than 25% for all tests and an average percent

error of 2.04% overprediction. Overall, the retrofitted column model provided a

reasonable fit to test data with an underprediction of first bar fracture.

92
 References

ACI 374.2R-13 (2013). “Guide for Testing Reinforced Concrete Structural Element under

Slowly Applied Simulated Seismic Loads”. American Concrete Institute, Farmington Hills,

Michigan.

ACI 440.2R-17 (2017). “Guide for the Design and Construction of Externally Bonded FRP

Systems for Strengthening Concrete Structures”. American Concrete Institute, Farmington Hills,

Michigan.

Atwater, B. F., Nelson, A. R., Clague, J. J., Carver, G. A., Yamaguchi, D. K., Bobrowsky, P. T.,

Bourgeois, J., Darienzo, M. E., Grant, W. C., Hemphill‐Haley, E., Kelsey, H. M., Jacoby, G. C.,

Nishenko, S. P., Palmer, S. P., Peterson, C. D., and Reinhart, M. A. 1995. “Summary of Coastal

Geologic Evidence for Past Great Earthquakes at the Cascadia Subduction Zone.” Earthquake

Spectra, 11(1), 1–18.

Breña, S. F., and Schlick, B. M. (2007). "Hysteretic Behavior of Bridge Columns with FRP

Jacketed Lap Splices Designed for Moderate Ductility Enhancement." Journal of Composites for

Construction, 11(6): 565-574.

Buckle, I.G. (1994). The Northridge, California Earthquake of January 17, 1994: Performance of

Highway Bridges. Technical Report NCEER-94-0008, National Center for Earthquake

Engineering Research, State University of New York at Buffalo, Buffalo, NY, 136 pp.

93
Chai, Y.H., Priestley, M.J.N., and Seible, F. (1991). Seismic Retrofit of Circular Bridge Columns

for Enhanced Flexural Performance. ACI Structural Journal, 88(5): 572-584.

Chai, Y.H., Priestley, M.J.N., and Seible, F. (1994). Analytical Model for Steel-Jacketed RC

Circular Bridge Columns. Journal of Structural Engineering, 120(8): 2358-2376.

EERI (1990). Loma Prieta Earthquake Reconnaissance Report-(Supplement to Volume 6).

Earthquake Spectra, Earthquake Engineering Research Institute, 448 pp.

EERI. (1994). Northridge Earthquake January 17, 1994, Preliminary Reconnaissance Report,”

No. 94-01, J. F. Hall, ed., Earthquake Engineering Research Institute, Oakland, Calif., 1994, p.

104.

Federal Highway Administration. (2006). Seismic Retrofitting Manual for Highway Structures:

Part 1 – Bridges. U.S. Department of Transportation, Federal Highway Administration,

Research, Development, and Technology, McLean, VA.

Fung, G. G., Lebeau, R. J., Klein, E. D., Belvedere, J., and Goldschmidt, A. F. (1971). Field

Investigation of Bridge Damage in the San Fernando Earthquake. Technical Report, Bridge

Department, Division of Highways, California Department of Transportation, Sacramento, CA,

209 pp.

94
Ghosh, K. K., and Sheikh, S. A. (2007). “Seismic upgrade with carbon fiber-reinforced polymer

of columns containing lap-spliced reinforcing bars.” ACI Structural Journal, 104(2): 227-236.

Goldfinger, C., Nelson, C. H., Morey, A. E., Johnson, J. E., Patton, J. R., Karabanov, E.,

Gutiérrez-Pastor, J., Eriksson, A. T., Gràcia, E., Dunhill, G., Enkin, R. J., Dallimore, A., and

Vallier, T. (2012). Turbidite Event History—Methods and Implications for Holocene

Paleoseismicity of the Cascadia Subduction Zone, U.S. Geological Survey Professional Paper

1661–F, 170 p.

Haroun, M. A., and Elsanadedy, H. M. (2005a). "Behavior of Cyclically Loaded Squat

Reinforced Concrete Bridge Columns Upgraded with Advanced Composite-Material Jackets".

Journal of Bridge Engineering, 10(6): 741-748.

Haroun, M. A., and Elsanadedy, H. M. (2005b). "Fiber-Reinforced Plastic Jackets for Ductility

Enhancement of Reinforced Concrete Bridge Columns with Poor Lap-Splice Detailing". Journal

of Bridge Engineering, 10(6): 749-757.

Huang, Y. and Mahin, S. A. (2010). “Simulating the inelastic seismic behavior of steel braced

frames including the effects of low-cycle fatigue.” PEER Report 2010/104, University of

California, Berkeley, CA.

Kanvinde, A. (2004). “Micromechanical simulation of earthquake-induced fracture in steel

structures.” PhD Dissertation, Stanford University, Stanford, CA.

95
Krawinkler H. (1992). Guidelines for Cyclic Seismic Testing of Components of Steel Structure,

ATC‐24. Redwood City, California: Applied Technology Council.

Mander, J.B., Priestley, M.J.N., and Park, R. (1988). Observed Stress-Strain Behavior of

Confined Concrete. Journal of Structural Engineering, 114(8): 1827-1849.

National Institute of Standards and Technology. (1990). Performance of Structures during the

Loma Prieta Earthquake of October 17, 1989. NIST Publication 778, Jan. 1990.

Padilla-Llano, D. A., Hajjar, J.F., Eatherton, M.R., Easterling, S.W., and Schafer, B.W. (2018).

“Cyclic fracture simulation framework for stability and collapse simulation in steel structures,”

Proc. of the Structural Stability Research Council, Baltimore, MD.

Priestley, M. J. N., and Seible, F. (1991). "Seismic assessment and retrofit of bridges." Struct.

Sys. Res. Pro)., Rep. No. SSRP·91/103, University of California at San Diego, Calif.

Priestley, M.J.N., Seible, F., Xiao, Y., and Verma, R. (1994a). Steel Jacket Retrofitting of

Reinforced Concrete Bridge Columns for Enhanced Shear Strength—Part 1: Theoretical

Considerations and Test Design. ACI Structural Journal, 91(4): 394-405.

96
Priestley, M.J.N., Seible, F., Xiao, Y., and Verma, R. (1994b). Steel Jacket Retrofitting of

Reinforced Concrete Bridge Columns for Enhanced Shear Strength—Part 2: Test Results and

Comparison with Theory. ACI Structural Journal, 91(5): 537-551.

Uriz, P., and Mahin, S.A. (2008). Toward Earthquake-Resistant Design of Concentrically Braced

Steel-Frame Structures. PEER Report 2008/08, Pacific Earthquake Engineering Research Center,

College of Engineering, University of California, Berkeley, CA, 401 pp.

Xiao, Y., and Ma, R. (1997). Seismic Retrofit of RC Circular Columns Using Prefabricated

Composite Jacketing. Journal of Structural Engineering, 123(10): 1357-1364.

97