25%:
vi». tar. Au.‘ ....

v if: 3., I
i hf? 3
3,.

1.442%; A .
uh: .‘1 131.... .:
,4. .c...

:l . . 4..
. . , . ‘ , . .. .x
, {bu ... ”nu. . ‘ , . w. E
. . . . . . ‘ V3 ‘ sub
‘ ‘ . . uni.
‘ . . 3V
. , . 4..
, ‘ - ‘ . »
kw}:
ll 25»
‘ an}

.

far x
... ::.-..£u.ui s
L... 23?

1.1
J
.v I

:1

in.
u 5. u
u»
n _ 2. 1.. 1.
{1 . 114.";
fifmhyfifi“

 

1.1:... a
A .. )1.

Wk ....:.v . iii! 3. .nth. 14-“..1
ti .. ‘ . A 1.... .1 .57» 23.3.. no

 

u 1.3% {Ff

 

This is to certify that the
thesis entitled

EFFECT OF SELF CONSOLIDATING CONCRETE
MIX PROPORTIONING ON
TRANSFER AND DEVELOPMENT LENGTH

presented by

MAHMOODUL HAQ

has been accepted towards fulfillment
of the requirements for the

MASTER OF

SCIENCE degree in CIVIL ENGINEERING

 

72., 1.3%;

Major Profe or’s SIgnature

<//3/0.S

Date

MSU is an Affinnarive Action/Equal Opportunity Institution

 

LIBRARIES
MICHIGAN STATE UNIVERSITY
EAST LANSING, MICH 48824-1048

-._..-_.—._._-—

PLACE IN RETURN Box to remove this dweckout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

got; 1 2 2009

 

101209

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 c:lCIRC/DateDue.indd-p.15

 

EFFECT OF SELF CONSOLIDATING CONCRETE (SCC)
MIX PROPORTIONING ON TRANSFER AND
DEVELOPMENT LENGTH OF PRESTRESSINGSTRANDS
By

Mahmoodul Haq

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF SCIENCE

Department of Civil and Environmental Engineering

2005

ABSTRACT
EFFECT OF SELF CONSOLIDATING CONCRETE (SCC)
MIX PROPORTIONING ON TRANSFER AND
DEVELOPMENT LENGTH OF PRESTRESSINGSTRANDS
By
Mahmoodul Haq
Self-consolidating concrete (SCC), a special concrete tailored for high
deformability and stability, has recently gained much interest from the precast/prestressed
industry. In spite of many developments on SCC material technology, knowledge on the
performance of elements built using SCC is limited. Of particular relevance in
prestressed concrete is the issue of strand bond. The research objective was thus to study
the effect of SCC mix proportioning on the bond of prestressing strands as it relates to
transfer and development lengths in prescast/prestressed girders. The study focused on
13mm (0.5 in.) diameter strands on laboratory-scale T-beams. Three SCC mix designs
were strategically designed to bound the accepted proportioning methods. Comparison
was made with a conventionally vibrated concrete mix (NCC). Transfer lengths were
determined by strand draw-in and concrete strain measurements and development lengths
were obtained through flexural beam tests. Transfer length results were within the current
ACI-318/AASHTO code recommendations. Experimentally determined development
lengths were longer than the ACI-318/AASHTO values. The source for this discrepancy
solely on the concrete mix is questionable and the quality of the strand is being further
evaluated to fully justify this findings. In spite of these doubts, the effect of SCC mix

proportioning was found to distinctly affect the different mechanisms of strand bond

performance influencing transfer and development length.

Copyright by
Mahmoodul Haq

2005

To my parents
for their continuous prayers, sacnfices and
guidance throughout my life

iv

ACKNOWLEDGEMENTS
Praise be to the LORD, The Most Beneficent, The Most Merciful

It is difficult to overstate my gratitude to my advisor, Dr. Rigoberto Burguefio.
With his enthusiasm, his inspiration, and his great efforts to explain things clearly and
simply, he helped to make research fun for me. Throughout my thesis, he provided
encouragement, sound advice, good teaching, good company, and lots of good ideas. I
would have been lost without him.

I wish to thank the members of my master’s committee, Dr. Ronald Harichandran
and Dr. Parviz Soroushian for their valuable time and constructive comments.

I am grateful to Mr. Siavosh Ravanbakhsh for all the help in the laboratory. His
valuable time, friendship and support helped me through the difficult times in this
research. I also thank James “JC” Brenton for his help in preparing test-setup accessories.

I would like to thank Dana Katherine Nuffer and David A Bendert for their
friendship and help in laborious tasks. I would also like to thank my colleagues who
helped me during the casting of concrete.

Lastly, and most importantly, I wish to thank my parents, Dr. Abdul Haq and
Jowhara; they bore me, raised me, supported me, taught me, and loved me. To them I
dedicate this thesis.

This study was possible thanks to the financial support of the Precast/Prestressed
Concrete Institute (PCI), through a 2003-2004 Daniel P. Jenny Research Fellowship, and
the Department of Civil and Environmental Engineering at MSU. The in-kind financial
and technical support from our industry partners Premarc Corporation (Grand Rapids,

MI) and Degussa Admixtures Inc. (Cleveland, OH) is also thankfully acknowledged.

TABLE OF CONTENTS

 

 

LIST OF TABLES ..... . ..... IX
LIST OF FIGURES--- - -- - ------ - - x1
Chapter 1 INTRODUCTION AND BACKGROUND------ 1
1.1 Background and Problem Definition .................................................................. 1
1.2 Objectives ........................................................................................................... 3
1.3 Self Compacting Concrete (SCC) Vs. Normally Consolidated Concrete (NCC)4
1.4 Thesis Organization ............................................................................................ 5
C hapter 2 LITERATURE REVIEW ...... ..........6
2.1 SCC Material Technology .................................................................................. 6
2.1.1 SCC Mix Behavior and Parameters ............................................................ 7
2.1.2 SCC Fresh Property Evaluation and Quality Control: .............................. 11
2.1.3 SCC Hardened Property and Structural Performance ............................... 16

2.2 Importance of Bond in Prestressed Concrete .................................................... 18
2.2.1 Bond Stresses and Mechanisms ................................................................ 19
2.2.2 Transfer Bond Stresses (at release of prestress) ....................................... 21
2.2.3 Flexural Bond Stresses at Ultimate Strength ............................................ 23

2.3 The Concept of Transfer and Development Length ......................................... 27
2.3.1 Definitions ................................................................................................. 29
2.3.2 ACI-3l8 [3] / AASHTO- LRFD [2] Code Recommendations ................. 30

2.4 Studies on Transfer and Development length of Prestressing strand. .............. 31
2.4.1 Studies on Bond Performance ................................................................... 32
2.4.2 Studies on Transfer and Development Length .......................................... 36
2.4.3 Oustanding issues ...................................................................................... 43

2.5 Concluding Remarks ......................................................................................... 46
Chapter 3 MIX DESIGN DEVELOPMENT AND EVALUATION....47
3.1.1 SCC Mix Design Approaches ................................................................... 47

3.2 Project Mix Design Matrix ............................................................................... 49
3.3 Fresh Property Evaluation: ............................................................................... 52
3.3.1 Slump Spread and Visual Stability Index (VSI) ....................................... 52
3.3.2 J Ring Test ................................................................................................ 53
3.3.3 L — Box Test .............................................................................................. 55

3.4 Challenges in SCC Quality Control and Quality Assurance ............................ 56
3.5 Hardened Concrete Property Evaluation .......................................................... 57
3.5.1 Compressive Strength (1’ c) ........................................................................ 58
3.5.2 Elastic Modulus Test ................................................................................. 60
3.5.3 Split Tensile Test. ..................................................................................... 64
3.5.4 Discussion of Results - Hardened Test properties .................................... 66

vi

 

Chapter 4 TEST PROGRAM — INTRODUCTION ------------------ -69

4.1 Specimen Design & Nomenclature Used. ........................................................ 69
4.1.1 Specimen Design ...................................................................................... 69
4. l .2 Nomenclature ............................................................................................ 72
4.2 Material Pr0perties ............................................................................................ 73
4.2. 1 Concrete .................................................................................................... 73
4.2.2 Prestressing Steel ...................................................................................... 73
4.3 Specimen Fabrication ........................................................................................ 74

Chapter 5 STRAND BOND PERFORMANCE EVALUATION.........78

 

5. 1 Introduction ....................................................................................................... 78
5.2 Background ....................................................................................................... 78
5.3 Specimen Preparation ....................................................................................... 79
5.4 Test Procedure 82
5.5 Results and Discussion ..................................................................................... 84
5.6 Summary and Conclusions ............................................................................... 93
Chapter 6 TRANSFER LENGTH EVALUATION...- ------------ 97
6. 1 Introduction ....................................................................................................... 97
6.2 Test Procedure — Concrete Strains .................................................................... 97
6.2.1 Specimen Preparation ............................................................................... 97
6.2.2 Concrete Surface Strain Measurements .................................................... 99
6.2.3 Construction of Surface Compressive Strain Profile ................................ 99
6.2.4 Determination of Average Maximum Strain (AMS) .............................. 102
6.2.5 Precision of Reported Results ................................................................. 108
6.3 Strand Draw—In Method ................................................................................. 109
6.3.1 Theoretical Background .......................................................................... 109
6.3.2 Test Procedure ........................................................................................ 112
6.3.3 Determination Of Transfer Length - Draw-In Value .............................. 114
6.4 Results and Discussion ................................................................................... 117
6.5 Summary and Conclusions ............................................................................. 120
6.6 Recommendations ........................................................................................... 121

Chapter 7 DEVELOPMENT LENGTH TEST PROGRAM..............122

7.1 Test Approach ................................................................................................. 122
7.2 Test Configuration .......................................................................................... 123
7.3 Analysis of Section at Ultimate - Nominal Capacity ..................................... 127
7.4 Instrumentation ............................................................................................... 130
7.5 Failure Modes and Analysis: .......................................................................... 134
7.6 Presentation and Discussion of Test Results .................................................. 142
7.7 Development Length Test Results as per ACI-318 Method [3] ..................... 144
7.8 Development Length Test Results as per the Strain Compatibility (SC) Method.
145
7.9 Flexural Bond Length: .................................................................................... 149
7.10 Summary and Conclusions ............................................................................. 151

vii

Chapter 8 SUMMARY AND CONCLUSIONS...................................153

 

8.1 Summary ......................................................................................................... 153
8.2 Conclusions ..................................................................................................... 155
APPENDICES ..... ‘ ---:‘------- v 333:: ................................. 2---: .....2. -2 ..... 159
Appendix 1- Stress — Strain Response- Elastic Modulus Tests ............................... 160
Appendix 2— Pull Out Test Response ........................................................................ 169
Appendix 3 - Transfer Length - Concrete Strain Profiles .......................................... 187
Appendix 4 - Development Test Response ................................................................. 193
Appendix 5 — Prestressing Details .............................................................................. 206
REFERENCES .00... ..... O ........ O. ........... O. 0000000000000000000000000 00......OOIOOOOOOOOOOOOOOOOOOO207

viii

LIST OF TABLES

Table 3-1 Mix Design Matrix - Binding of Performance by w/c ratio ............................. 49
Table 3-2 Mix Designs used in the project — Target mixes .............................................. 50
Table 3-3 Changes in Admixtures for Actual Project Mix Designs ................................. 51
Table 34 Actual Project Mix Designs .............................................................................. 51
Table 3-5 Slump Spread and VSI rating of SCC mixes .................................................... 52
Table 3-6 and J — Ring Slump Spread and J-Ring Values ............................................... 55
Table 3-7 L- Box Blocking Ratio ..................................................................................... 56
Table 3-8 Compressive Strength ( f’C ) Test Results — 1 to 28 days ................................. 59
Table 3-9 Compressive Strength at Day of Test ............................................................... 59
Table 3-10 Elastic Modulus Tests at 3days — all mixes .................................................... 61
Table 3-11 Elastic Modulus Tests at 28 days ................................................................... 61
Table 3-12 Split Tensile Strength - 1 to 28 Days ............................................................. 65
Table 3-13 Split Tensile Strength - Day of Test ............................................................... 65
Table 5-1 Maximum (Peak) Pull-out Force - Release (3 days) ........................................ 85
Table 5-2 Maximum (Peak) Pull-out Force - 7 days. ....................................................... 85
Table 5-3 Pull Out Forces at First Slip - Release (3 days) ................................................ 86
Table 54 Pull Out forces at first Slip - 7 days. ................................................................ 86
Table 5-5 Average Maximum Bond Strengths from Peak Pull-out Forces ...................... 91
Table 5-6 Pull Out Tests Results - Performed by Logan .................................................. 94
Table 6-1 Average value of Transfer Length per Mix Type — Concrete Strains ............ 105
Table 6-2 Draw—in and Transfer Length Values at Various Concrete Ages .................. 116
Table 6-3 Transfer Length values from ACI 318 / AASHTO equation ......................... 118

ix

Table 6-4 Comparison of Measured and ACI transfer lengths ....................................... 119

Table 7-1 Test Configurations for Development Length Studies ................................... 126
Table 7-2 Development Length Test Results .................................................................. 143
Table 7-3 Comparision of Development Length Results ............................................... 148
Table 7-4 Flexural Bond lengths ..................................................................................... 150
Table A5 - 1 Prestressing Details ................................................................................... 206

LIST OF FIGURES

Figure 2-1 SCC Parameters and Behavior [20] ................................................................ 10
Figure 2-2 Slump Spread Test and VSI ............................................................................ 13
Figure 2-3 J-Ring Value [30] ............................................................................................ 14
Figure 2-4 Schematic of L- Box Test [30] ........................................................................ 15
Figure 2-5 Hoyer’s Effect [34] ......................................................................................... 20
Figure 2-6. Bond Stress Distribution at Strand End [23] .................................................. 21
Figure 2-7. Section Warping and Forces at End [23] ....................................................... 22
Figure 2-8. Stress Distributions at a Crack Front [23] ...................................................... 24
Figure 2-9. Bond Stresses at Flexural Cracks [23] ........................................................... 26
Figure 2-10 Variation of stresses — ACI-318 equation representation [3][2] ................... 28
Figure 2-11 Effect of Embedded Length in Normal Pull-Out Tests [23] ......................... 32
Figure 2-12. Details of Moustafa Test .............................................................................. 33
Figure 3-1 - Slump Spread test ......................................................................................... 53
Figure 3-2 I — Ring Test ................................................................................................... 54
Figure 3-3 L- Box Test ..................................................................................................... 55
Figure 3-4 Compressive Strength variation with time — All Mixes .................................. 60
Figure 3-5 Typical Stress—Strain Response - NCCB - 28 days ....................................... 62
Figure 3-6 Typical Stress—Strain Response - SCCl — 28 days ........................................ 62
Figure 3-7 Typical Stress—Strain Response — SCC2A — 28 days ..................................... 63
Figure 3-8 Typical Stress-Strain Response — SCC3 — 28 days ........................................ 63
Figure 3-9 Split Tensile Strength variation with Time ..................................................... 66
Figure 3-10 Comparison of Compressive Strength at 28 days ......................................... 67

xi

Figure 3-11 Measured vs. AC1 Predicted Elastic Modulus at 28 days ............................. 68

Figure 4-1 Test Specimen - Cross Section Details .......................................................... 70
Figure 4-2 Reinforcement Details of the test specimens .................................................. 71
Figure 4-3 Nomenclature for Transfer Length .................................................................. 72
Figure 4-4 Strand Condition ............................................................................................. 74
Figure 4-5 Layout of Formwork ....................................................................................... 75
Figure 4-6 Schematic Layout of the casting bed .............................................................. 75
Figure 4-7 Pretensioning of strands .................................................................................. 76
Figure 4-8 Release of Prestress — Both ends of the beam being cut simultaneously ........ 77
Figure 5-1 Pull-out Block Geometry and Reinforcement ................................................. 80
Figure 5-2 Casting of Pull-out Test Block ........................................................................ 81
Figure 5-3 Pull-out Test Setup Overview ......................................................................... 81
Figure 5-4 Schematic of Pull Out Test Setup ................................................................... 82
Figure 5-5 Measurement of Displacements - Pull-out Test ............................................. 83
Figure 5-6 Typical Pull-out Test Response - Release ....................................................... 87
Figure 5-7 Typical Pull-out Test Response at 7 days ....................................................... 87
Figure 5-8 “Close-Up” of First slip occurrence - Release (3 days) ................................. 88
Figure 5-9 “Close-Up” of First slip occurrence 7 days ..................................................... 88
Figure 5-10 Comparison of Peak Pull-out forces ............................................................. 89
Figure 5-11 Comparison of Pull-out forces at first Slip ................................................... 90
Figure 5-12 Comparison of Relative Bond Strengths ....................................................... 92
Figure 5-13 Results from LBPI‘ performed by Logan according to [24] ......................... 95
Figure 6-1 Actual Picture of the DEMEC Instrument ...................................................... 98

xii

Figure 6-2 Schematic Representation of Target point ...................................................... 98

Figure 6-3 Performing measurement with a DEMEC gauge ............................................ 99
Figure 64 Extension Brackets to measure the strains at the ends .................................... 99
Figure 6-5 Smoothening Of Strain Profile ...................................................................... 101
Figure 6-6 Comparison of Smooth and Non- Smooth (RAW) data ............................... 101
Figure 6-7 Location of AMS values ............................................................................... 102

Figure 6-8 Determination of Transfer length from Concrete Strain Profiles — NCCB... 105
Figure 6-9 Determination of Transfer length from Concrete Strain Profiles — SCCl 106
Figure 6-10 Determination of Transfer length from Concrete Strain Profiles — SCC2A
................................................................................................................................... 106
Figure 6-11 Determination of Transfer length from Concrete Strain Profiles — SCCZB 107
Figure 6—12 Determination of Transfer length from Concrete Strain Profiles — SCC3 .. 107
Figure 6-13 Comparison of Transfer Length Values Obtained from Concrete Strains .. 108
Figure 6-14 Strand Draw-in — Instrumentation and Measurement ................................. 113
Figure 6-15 Average Draw—in Values at Transfer (3 days) for all mixes ...................... 114

Figure 6-16 Average Transfer Length Values at transfer (3 days) obtained from Draw-in

................................................................................................................................... 115
Figure 6-17 Variation of Strand Draw-in with time ....................................................... 117
Figure 6-18 Transfer Length — ACI Code — All Mixes [3] ............................................ 118
Figure 6-19 Comparison of Measured Transfer Length with ACI Equation .................. 119
Figure 7-1 Schematic Representation of Flexural Test ................................................... 124
Figure 7-2 Overview of the Flexural Test Setup ............................................................ 125
Figure 7-3 Test setup - View of Spreader Beam ............................................................ 125

xiii

Figure 7-4 ACI-318 code Assumed Stress — Strain Distribution [4] .............................. 128

Figure 7-5 Instrumentation for Overall Unit Deformation ............................................ 132
Figure 7-6 Instrumentation for Support Movement and strand End-Slip ....................... 132
Figure 7-7 Average Strain Measurement at Strand Level .............................................. 133
Figure 7—8 Shear- Slip Failure — Initial stage .................................................................. 136
Figure 7-9 Typical Shear-Slip Failure ............................................................................ 136
Figure 7-10 Test Response for a Typical Shear-Slip Failure .......................................... 137
Figure 7-11 Flexural Failure — Symmetric crack pattern ................................................ 138
Figure 7-12 Flexural Failure — Final Condition (compression) ...................................... 139
Figure 7-13 Flexural Failure -— Final Condition (tension) ............................................... 139
Figure 7-14 Test Response for a Typical Flexural Failure ............................................. 140
Figure 7-15 Test response for atypical Bond-Slip failure .............................................. 141
Figure 7-16 Comparison of Development Length Results ............................................. 149
Appendix 1

Figure Al- 1 Elastic Modulus Test — Transfer — SCC2B — Testl .................................. 161
Figure Al- 2 Elastic Modulus Test — Transfer — SCC2B — Test2 .................................. 161
Figure Al- 3 Elastic Modulus Test - Transfer - SCCZB — Test3 .................................. 162
Figure Al- 4 Elastic Modulus Test — Transfer — SCC3 - Testl ..................................... 162
Figure Al- 5 Elastic Modulus Test - Transfer - SCC3 — Test2 ..................................... 163
Figure Al- 6 Elastic Modulus Test — Transfer — NCCB - Testl ................................... 163
Figure A1- 7 Elastic Modulus Test — Transfer - NCCB - Test2 ................................... 164
Figure Al- 8 Elastic Modulus Test — 28 Days - SCCl — Testl ..................................... 164
Figure Al— 9 Elastic Modulus Test — 28 Days - SCCl — Test2 ..................................... 165

xiv

Figure Al- 10 Elastic Modulus Test — 28 Days — SCC2A — Testl ................................ 165

Figure A1- 11 Elastic Modulus Test — 28 Days — SCC2A — Test2 ................................ 166
Figure Al- 12 Elastic Modulus Test — 28 Days — SCC2A —- Test3 ................................ 166
Figure A1- 13 Elastic Modulus Test — 28 Days — SCC3 — Testl ................................... 167
Figure Al- 14 Elastic Modulus Test — 28 Days — SCC3 — Test2 ................................... 167
Figure Al- 15 Elastic Modulus Test — 28 Days — NCCB — Testl .................................. 168
Figure Al- 16 Elastic Modulus Test — 28 Days — NCCB — Test2 .................................. 168
Appendix 2

Figure A2- 1 Pull Out Test Prestress Release. - SCCl — Strand #1 ............................... 170
Figure A2- 2 Pull Out Test at Prestress Release. — SCCl — Strand #2 ........................... 170
Figure A2- 3 Pull Out Test at Prestress Release. - SCCl — Strand #3 ........................... 171
Figure A2- 4 Pull Out Test at Prestress Release - SCC2A — Strand #1 ......................... 171
Figure A2- 5 Pull Out Test at Prestress Release. - SCC2A — Strand #2 ........................ 172
Figure A2- 6 Pull Out Test at Prestress Release — SCC2A — Strand #3 ......................... 172
Figure A2- 7 Pull Out Test at Prestress Release. — SCC2B — Strand #1 ........................ 173
Figure A2- 8 Pull Out Test at Prestress Release. — SCC2B — Strand #2 ........................ 173
Figure A2- 9 Pull Out Test at Prestress Release. — SCC2B — Strand #3 ........................ 174
Figure A2- 10 Pull Out Test at Prestress Release. — SCC3 — Strand #1 ........................ 174
Figure A2- 11 Pull Out Test at Prestress Release. — SCC3 - Strand #2 ......................... 175
Figure A2- 12 Pull Out Test at Prestress Release. — SCC3 - Strand #3 ......................... 175
Figure A2- 13 Pull Out Test at Prestress Release. — NCCB - Strand #1 ........................ 176
Figure A2- 14 Pull Out Test at Prestress Release. — NCCB — Strand #2 ........................ 176
Figure A2- 15 Pull Out Test at Prestress Release. - NCCB — Strand #3 ........................ 177

XV

Figure A2— 16 Pull Out Test at 7 Days -— SCC 1 — Strand #4 .......................................... 177

Figure A2- 17 Pull Out Test at 7 Days — SCCl — Strand #5 .......................................... 178
Figure A2- 18 Pull Out Test at 7 Days — SCCl — Strand #6 .......................................... 178
Figure A2- 19 Pull Out Test at 7 Days - SCC2A — Strand #4 ........................................ 179
Figure A2- 20 Pull Out Test at 7 Days — SCC2A — Strand #5 ........................................ 179
Figure A2- 21 Pull Out Test at 7 Days — SCC2A — Strand #6 ........................................ 180
Figure A2- 22 Pull Out Test at 7 Days — SCCZB — Strand #4 ........................................ 180
Figure A2- 23 Pull Out Test at 7 Days — SCC2B — Strand #5 ........................................ 181
Figure A2- 24 Pull Out Test at 7 Days - SCC2B — Strand #6 ........................................ 181
Figure A2- 25 Pull Out Test at 7 Days — SCC3 -— Strand #4 .......................................... 182
Figure A2- 26 Pull Out Test at 7 Days — SCC3 - Strand #5 .......................................... 182
Figure A2- 27 Pull Out Test at 7 Days — SCC3 - Strand #6 .......................................... 183
Figure A2- 28 Pull Out Test at 7 Days — NCCB - Strand #4 ......................................... 183
Figure A2- 29 Pull Out Test at 7 Days - NCCB — Strand #5 ......................................... 184
Figure A2- 30 Pull Out Test at 7 Days — NCCB — Strand #6 ......................................... 184
Figure A2- 31 Combined Average Pull-out Test response — At 3 Days ......................... 185
Figure A2- 32 Combined Average Pull-out Test response — At 7 Days ......................... 185

Figure A2- 33 Combined Average First Slip - Pull-out Test Response — At 3 Days ..... 186

Figure A2- 34 Combined Average First Slip - Pull-out Test Response — At 7 Days ..... 186

Appendix 3

Figure A3 - 1 Concrete Strain Profile — SCCl — Beaml ............................................... 188
Figure A3 - 2 Concrete Strain Profile - SCCl — BeamZ ............................................... 188
Figure A3 - 3 Concrete Strain Profile — SCC2A — Beaml ............................................ 189

xvi

Figure A3 - 4 Concrete Strain Profile — SCC2A — Beam2 ........................................... 189

Figure A3 - 5 Concrete Strain Profile — SCC2B - Beaml ............................................ 190
Figure A3 - 6 Concrete Strain Profile — SCC2B - Beam2 ............................................ 190
Figure A3 - 7 Concrete Strain Profile — SCC3 — Beaml ............................................... 191
Figure A3 - 8 Concrete Strain Profile — SCC3 - Beam2 ............................................... 191
Figure A3 - 9 Concrete Strain Profile — NCCB — Beaml .............................................. 192
Figure A3 - 10 Concrete Strain Profile — NCCB — Beam2 ............................................ 192
Appendix 4

Figure A4 - 1 Moment vs. Displacement- SCCl-l-A - L, = 72.38 in ............................ 194
Figure A4 - 2 Moment vs. Displacement- SCCl-l-B - L, = 137.75 in .......................... 194
Figure A4 - 3 Moment vs. Displacement- SCC1-2-A - L, =122.00 in .......................... 195
Figure A4 - 4 Moment vs. Displacement- SCC1-2-B - L, = 118.50 in ......................... 195
Figure A4 - 5 Moment vs. Displacement- SCC2A-1-A - L, = 70.50 in ........................ 196
Figure A4 - 6 Moment vs. Displacement- SCC2A-1-B - L, = 64.50 in ......................... 196
Figure A4 - 7 Moment vs. Displacement- SCC2A-2-A - L, = 80.00 in ........................ 197
Figure A4 — 8 Moment vs. Displacement- SCC2A-2-B - L, = 86.75 in ......................... 197
Figure A4 - 9 Moment vs. Displacement- SCC2B-l-A - L, = 70.50 in ......................... 198
Figure A4 - 10 Moment vs. Displacement- SCC2B-1-B - L, = 102.75 in ..................... 198
Figure A4 - 11 Moment vs. Displacement- SCC2B-2-A - L, = 126.75 in ..................... 199
Figure A4 - 12 Moment vs. Displacement- SCC2B-2-B - L, = 103.80 in .................... 199
Figure A4 - 13 Moment vs. Displacement- SCC3-l-A - L, = 58.50 in ......................... 200
Figure A4 - l4 Moment vs. Displacement- SCC3-l-B - L, = 97.75 in ......................... 200
Figure A4 - 15 Moment vs. Displacement- SCC3-2-A - L, = 106.50 in ...................... 201

xvii

Figure A4 - l6 Moment vs. Displacement— SCC3-2-B - L, = 103.00 in ....................... 201

Figure A4 - 17 Moment vs. Displacement- NCCA-l-A - L, = 76.00 in ........................ 202
Figure A4 - 18 Moment vs. Displacement- NCCA —1-B - L, = 122.75 in ..................... 202
Figure A4 - 19 Moment vs. Displacement- NCCA -2-A - L, = 111.00 in .................... 203
Figure A4 - 20 Moment vs. Displacement- NCCA -2-B - L, = 60.00 in ....................... 203
Figure A4 - 21 Moment vs. Displacement- NCCB -1-A - L, = 63.75 in ........................ 204
Figure A4 - 22 Moment vs. Displacement- NCCB -1-B - L, = 64.00 in ........................ 204
Figure A4 - 23 Moment vs. Displacement- NCCB -2-A - L, = 100.50 in ..................... 205
Figure A4 - 24 Moment vs. Displacement- NCCB -2-B - L, = 93.50 in ........................ 205

xviii

CHAPTER 1 INTRODUCTION AND BACKGROUND

1.1 Background and Problem Definition

First introduced in Japan in the early 19805 (Okamura, 1999), self-consolidating,
or self-compacting concrete (SCC) has been gaining increased attention due its unique
behavior to fill forrnwork and flow around obstructions without blocking or segregating.
This behavior, obtained by tailoring the selection of materials and mix proportioning has
allowed it to be classified as a kind of high-performance concrete, offering the possibility
of designing for both the fresh and hardened properties of concrete to specific project
needs.

Much work has been done since the development of SCC, particularly in the
development of mix designs, characterization of its rheology and mechanical properties,
and evaluation its in-situ properties. The underlying principles governing the
development of SCC-type behavior, i.e., a stable highly flowable concrete mix, are now
generally well accepted and a great variety of SCC mixes have been developed using
different materials and optimized for different purposes. In spite of all the progress made
on the material characterization of SCC, very limited information exists on construction
and design specifications to guide industry, designers, and highway transportation
authorities. If this trend continues, it is likely that even the most extensive and detailed
material characterization efforts on SCC will not address the concerns of the state DOT’s,
whose ultimate interest lies on the structural performance and durability of the built
structure.

The underlying principles governing the development of SCC-type behavior, i.e.,

a stable highly flowable concrete mix, are now generally well accepted and a great

variety of SCC mixes have been developed using different materials and optimized for
different purposes (Khayat, 1999). However, the special mix designs that give SCC its
unique fresh-property advantages significantly deviate from what is currently considered
as ideal and developed through many years of experience and research. This has raised
concerns regarding material and structural performance issues. Among them are material
properties such as creep and shrinkage; structural issues such as prestress losses and
bond; and durability issues such as freeze-thaw behavior. These concerns have limited
the acceptance of SCC in the United States, despite of its increased use in Japan, Canada
and Europe.

Self-consolidating concrete (SCC) has become of high interest to the precast
concrete industry due to the benefits it offers in enhancing construction productivity. In
spite of this interest and rapid developments on SCC technology, its acceptance in the
US. is lagging due to material and structural performance concerns; among these is the
issue of bond. The issue of bond for prestressed concrete members has been debated from
the past six decades. Considerable work has been done regarding the development of a
better understanding of bond and its relationship with transfer and development length for
conventionally consolidated concrete. Consequently, several theories for transfer and
deve10pment length have been formulated. In spite of the multiple efforts, current
provisions by the ACI Code (318-02) [3] and AASHTO [2] are primarily based on the
work of Hanson and Kaar [17]. However, in general, it has been found that these
guidelines underestimate the actual transfer and flexural bond lengths, and their validity
has been questioned for over 25 years [8][10][18][24][31][33][35][41]. Thus, even for

conventional concrete, which is well developed and ideal shows large deviation on results

related to bond issues. In such a scenario, with the absence of specific design codes and

guidelines for SCC, the study of bond in SCC has become inevitable

1.2 Objectives

The design of a concrete mix for SCC-type behavior for specific project needs
grants SCC its quality as a hi gh-perforrnance material. However, this advantage can also
limit the development of all-inclusive guidelines. It is now generally agreed that SCC can
be obtained by two general approaches or the combination thereof: (a) mixtures with high
fines content and high-range-water-reducing admixtures, and (b) mixtures with high-
ran ge-water-reducin g admixtures and viscosity-modifyin g admixtures.

The current study thus focuses on bounding the parameters that allow the
development of SCC-type behavior and on investigating the bond performance of SCC in
precast/pretensioned bridge elements with the objective of providing guidance on the
construction and design of these elements when using self-consolidating concrete.

The specific aims will be:

0 To investigate the material properties of three types of SCC mixes through

small-scale tests.

0 To experimentally investigate the development and transfer lengths for V2”

prestressing strands in SCC through small-scale tests.

Also, the constitutive hardened concrete properties including the compressive
strength, split tensile strength and modulus of elasticity were evaluated at various ages of
concrete. The results from SCC mixes are compared with a normally consolidated

concrete (NCC) mix.

1.3 Self Compacting Concrete (SCC) Vs. Normally Consolidated Concrete (N CC)

The unique self-consolidation behavior of SCC and its tailorable nature to suit

specific project needs has allowed it to be classified as a kind of high-performance

concrete. SCC offers many advantages for the precast/prestressed industry, among them

are:

a)

b)

d)

e)

g)

h)

It can be placed with no mechanical vibration, resulting in savings in placement

COSIS.

Improved and more uniform architectural surface finish with little to no remedial

surface work

Ease of filling restricted sections and hard to reach areas

Opportunities to create structural and architectural shapes and finished not

achievable with conventional concrete.

Improved consolidation around reinforcement and bond with reinforcement.

Improved Pumpability

Improved uniformity of in-place concrete by eliminating variable operator —

related effort of consolidation,

Labor Savings,

Shorter construction periods and resulting cost savings and

j) Quicker concrete turn — around times enabling the producer to service the project

more efficiently.

In order to completely utilize the advantages SCC offers, specifically to the
precast/prestressed industry, bond parameters need to be carefully evaluated. Considering
the extensive mix designs of SCC available, the current work focuses on bounding the

hardened properties by proper mix proportions.

1.4 Thesis Organization

This thesis has been organized into 7 chapters. Chapter 2 includes a brief review of
the existing literature on both SCC and bond issues. Definitions of important parameters
related to this project have also been included. Chapter 3 includes: (a) study of the mix
designs used in the project and their fresh and hardened properties and (b)test Specimen
details and fabrication. Chapter 4 gives details about the pull-out testing program.
Transfer length, measurement, methods and results are discussed in Chapter 5.
Development length testing, measurement and discussion of results are presented in
Chapter6. Chapter 7 gives a brief summary of the entire project and the conclusions

drawn from this research

CHAPTER 2 LITERATURE REVIEW

In order to gain the perspective on the use of Self Consolidating Concrete (SCC)
around the world and to understand the different issues related to bond performance of
prestressing strands, a literature review was carried out. This chapter provides a brief
overview of pertinent work related to this research. This chapter has been subdivided into
five sections: (1) SCC material technology (2) importance of bond in prestressed
concrete, (3) Concept of transfer and development length, (4) studies on transfer and
development length of prestressing strands and (5) concluding remarks. The sections are

explained briefly below:

2.1 SCC Material Technology

Self-consolidating concrete (SCC), or "vibration-free" concrete, was developed in
Japan in the 1980's in response to the gradual reduction in the numbers of skilled workers
required for quality construction work [27]. The unique self-consolidation behavior of
SCC has allowed it to be classified as a kind of high-performance concrete [1][27]. The
design of SCC involves tailoring the selection of materials and mix proportioning to
ensure high deformability and adequate resistance to segregation, to achieve high filling
capability and flow around obstructions without blocking [20]. Such requirements require
different opposing measures, such as reduction in coarse aggregate volume and reduction
of free water content to limit inter-particle friction among coarse aggregate, sand and
cement. Although there is no common definition for SCC, a common understanding is
that a self consolidating concrete is a one that:

a) has the fluidity that allows self—consolidation without external vibration,

b) remains homogeneous during and after placement, and

c) flows easily through reinforcement
Thus, the most important criteria that differentiates SCC from conventional concrete are
those that are related to the properties of fresh concrete. These requirements are [19] :
o Self-placing and self-consolidation
0 Retention of deformability throughout transportation and placement
0 High stability during transportation and placement
0 Resistance to segregation, bleeding, and surface settlement (i.e., stability)
after casting
0 Limited bleeding and settlement
0 Uniform surface quality

0 Homogeneous distribution of in-situ properties.

Achieving all of the above performance requirements is not easy and it requires a
compromise between the many parameters influencing the mixture proportioning [20] . In
order to understand SCC material technology, this study has been subdivided into the
following: ( 1) SCC mix behavior and parameters (2) SCC fresh property evaluation and
quality control (3) Hardened properties, and(4) structural performance Each of these

topics is discussed as follows:

2.1.1 S CC Mix Behavior and Parameters

In rheological terms, SCC is often described as a Bingham fluid (viscoelastic)
where the linear relation between the stress and the shear rate ratio is characterized by
two constants — viscosity and yield stress. The yield stress primarily governs the self-
consolidation behavior, while the viscosity affects the homogeneity (stability) and

flowability [20]. While the viscosity may be modified depending on the application

(flowability requirements), the yield stress must remain much lower than conventional
concrete mixes in order to achieve self-consolidation [38].It is clear that the fresh
properties of SCC are particularly important for its ability to flow under its own weight
without segregation and to fill congested structural sections. Thus it is very important to
ensure the following three main parameters:

0 Fluidity / Deformability
0 Easy Flow / Low Blockage, and
o Homogeneity / Stability.

The above primary parameters need to be properly adjusted to satisfy all the
requirements of SCC. These three primary parameters can be adjusted (as shown in
Figure 2-1 [20]) and achieved by varying the mix proportions of SCC. Thus, the mix

design factors that affect these three basic parameters are as follows:
0 Water to cement ratio (w/c)
0 Amount of high-range-water—reducers (HRWR)
0 Use of viscosity modifying admixtures (VMA)
0 Coarse aggregate content (CAC)
0 Sand to paste content (S/Pt)

o Entrained air (EA).

Proper proportioning of these parameters is essential to achieve SCC and to obtain the

required primary parameters of deformability, passing ability and stability. The effects of

the various mix proportioning on the basic parameters of SCC are explained in detail

below:

The deformability of concrete is closely dependent on the quantity of cement paste,
which can be increased with the use of high-range water reducers (HRWR) that allow
lowering of the yield value with only moderate drops in viscosity. Thus, highly flowable
concrete can be obtained without significant reduction in cohesiveness. Deformability
can also be increased by increasing the water to cement (w/c) ratio, which controls the
deformability of the cement paste. However, high w/c ratios reduce the cohesiveness of
the paste and mortar and lead to fine and coarse aggregate segregation. This means that
care must be taken when lowering w/c ratios to ensure that, it does not lead to large
reductions in cohesiveness. Another factor affecting deformability is the friction
between the various solid particles (sand, gravel, cement, and mineral additives). The
inter-particle friction increases the internal resistance to flow thus reducing the
deformability and rate of flow of the concrete. This effect can be reduced with the use of
HRWR, which, by dispersing cement grains, enable water content reduction with limited
dr0ps in viscosity. Reducing the coarse aggregate and sand volumes and increasing the
paste volume also enhances deformability. In addition, the use of continuously graded

cementitious materials and fillers can also help reduce inter-particle friction.

The second main parameter necessary for SCC properties is that this highly flowable
concrete has proper M11112 during transportation, placement, and after casting. As
shown in Figure 2-1 [20], enhanced stability requires reduction in coarse aggregate
content and reduction of the maximum aggregate size. Cohesion must also be improved

by enhancing bond between the mortar paste and the coarse aggregate. This can be

achieved by reducing the w/c ratio and the use of viscosity modifier admixtures
(V MA). Stability must also be ensured after concrete placement to avoid water migration
(bleeding) that can lead to weak interface between the aggregate and the cement paste,
accumulation of cement paste in the bottom half of horizontal reinforcement, and
segregation of suspended solid particles. Decrease in bleeding can be achieved by
reducing the w/c ratio, addition of HRWR, and the use of VMA and/or increased
volume of cementitious and filler materials (i.e., low w/c ratio) to bind some of the excess

water [20].

    

  

fluidity/Deformability

A. Increase paste deformability
* use of HRWR
* balanced w/c ratio

Easy Flow/Low Blockage

A. Enhance cohesiveness
* low w/c ratio
* use of VMA

    
    
 
 

  

B. Reduce inter-particle friction
* low coarse aggregate volume
* use cont. graded powder

B. Compatible flow space and
aggregate size
* low coarse aggregate volume
* low max. size aggregate .

     
 

 

   

Homogeneity/Stability

A. Reduce solids separation
* limit aggregate content
*reduce max. size aggregate ,
*. increase cohesion & viscosit'

— low w/c ratio '
. - use of VMA

   

fluidity/Stability
Trade-off

Low Viscosity .

Hishmfidi'y .

    

         
     
         
 
  

 
    

High Viscosity
Low Fluidity .
VISCOSITY

 

DEFORMABILITY .

   

    

B. Minimize bleeding
I "' low w/c ratio p > ; .
* use of high- -area powder
* increase VMA

     
 
 

    
    
 

Figure 2-1 SCC Parameters and Behavior [20]

The third essential property for a successful SCC mix is its ability to easily flow

through reinforcement or reduction of the risk of blockage when flowing through

narrow spaces. Such risk can be minimized by providing adequate viscosity, to ensure
good suspension of the solid particles during flow and reducing inter-particle friction.
However, increase in viscosity reduces deformability [20]. Thus, to allow SCC to easily
flow through reinforcement it should have appropriate cohesiveness by reducing the w/c
ratio and/or using VMA. In addition, as the free space between obstacles and
reinforcement reduces, the coarse aggregate volume and the maximum size aggregate

should be reduced.

As discussed earlier, successful SCC mixes can be achieved by varying the
achieving the three parameters discussed above and hence there is no commonly accepted
procedure to proportion SCC mixtures and over the years, several methods have been
developed in research centers around the world [11][12][20][39]. The mix designs
selected for this project and the concept of choosing the particular mix designs is given in

Chapter 3.

2.1.2 SCC Fresh Property Evaluation and Quality Control:

As stated above, various mix design methods for SCC have been proposed.
However, it is clear that the performance criteria to have a highly flowable and stable mix
remains the same. This indicates that quality control and acceptance criteria for SCC
should be based on a performance-based approach. Performance requirements for the
hardened concrete are well established for use of precast/prestressed bridge elements.
However, the performance requirements for the fresh SCC mix can only be evaluated at

the time Of placement.

11

Considerable work has been done around the world in developing and evaluating the self-

compactability of SCC, including deformability, stability, and filling capacity. The most

commonly used tests are [29][30][38]:

Concrete Rheometer. A device applies a range Of shear rates and monitors the force
needed to maintain these shear rates in a plastic material; later converting the force
into stress. However, only a few concrete and mortar rheometers are available since
this type of equipment is very expensive and not easy to use in the job site. Thus,

these devices have been limited for use in large research centers.

Spread Test. The slump spread test also called as Slump flow test is used to asses the
horizontal free flow of SCC in absence of obstructions [30]. This procedure uses the
conventional Abram’s cone (Figure 2-2). The cone is filled in one layer without
rodding and the diameter, instead of the slump, of the concrete sample is measured
after the cone is lifted. The test evaluates self-compactibility as it mainly relates to
yield stress. An evaluation of viscosity can be made by monitoring the time it takes
for the concrete to reach a spread of 50 mm (2 in.). This test was performed in this

study and the results are discussed in Chapter 3.

12

 

Figure 2-2 Slump Spread Test and V81

0 M. This apparatus is used to force the SCC flow through reinforcement in
conjunction with an Abrams cone or the Orimet setup. The size and the spacing
between the bars can be adjusted to simulate any reinforcement configuration. J-Ring
value is obtained from the differences between the spread; with and without the ring
or the height difference between the concrete inside and outside the ring are

measured. The J-Ring value is calculated as follows (Figure 2-3) [30]:

a) Measure the values d1 (Figure 2-3)in the center of the J-Ring and also 4

values of d, and db just inside and outside the ring (measurements in mm)
b) Calculate h1=125-d1 and all h values hm=125-dax (x=1 to 4)
c) Calculate 4 values h l-hax; calculate median value hlm-ham.
(1) Calculate 4 values hu-hbx; calculate median value ham-him.

e) Calculate 2(ham-hbm)-(h1m-h,,,,). This is J—Ring value.

13

 

 

 

 

 

 

 

 

 

 

 

Figure 2-3 J-Ring Value [30]

In general, greater J-ring spread flow results in greater passing ability. Satisfactory
passing ability without blockage is attained when the value 2(ham-hbm)-(h1,,.- an.) is
less than 15mm (0.59 in.). Generally acceptable passing ability is achieved when this
value is around 10 mm (0.39 in.). This test was performed in this study and the results
are discussed in Chapter3.

Visual Stability Index. This method involves the visual evaluation of the SCC patty
resulting from observation of the SCC just prior to placement and after the
performance of the spread (slump flow) test. It is used to evaluate the relative stability

of batches of the same or similar SCC mixes. The test requires the development of

14

considerable judgment and may thus be subjective. This test was performed in this
study and the results are discussed in Chapter3.

L-Box amt U-Box. These tests simulate the casting process by forcing an SCC sample
to flow through obstacles under a static pressure. They provide and indication on the
static and dynamic segregation resistance of SCC as well as its ability to flow through

reinforcement.

1"

23 5/8" 600
"wail

' char 3 - #4 With Gap 0f 1 318" (35) Between

 

      

 

D

. III“ A
: “In!“

 

' I

 

'5 718' (150)

 

 

 

Figure 2-4 Schematic of L- Box Test [30]

L—Box test (Figure 2-4) assesses the flow of SCC and also the extent to which it is
subjected to blocking by reinforcement [30]. The test apparatus comprises of a
rectangular cavity in with reinforcement. The level of reinforcement can be changed
to enforce severe or light restraints depending on the actual reinforcement blocking of

the structure. This opening is controlled by a gate. The SCC is placed in the vertical

15

cavity without any vibration and held there for a minute. The gate is then Opened for
the SCC to flow into the horizontal cavity. After the flow is complete the heights of
SCC in the vertical chamber (H1) and the horizontal chamber (H2) are measured. The
ratio of H2/Hl is termed as the blocking ratio. If the SCC flows as freely as water, at
rest, it will be horizontal, so H2/Hl=l Thus closer the blocking ratio is to unity, the
better is the passing ability. The segregation of aggregates, if any can easily noted in
the vertical chamber. This test was performed in this study and the results are
discussed in Chapter 3.

This test was performed in this study and the results are discussed in Chapter 3.
V—Funnelgid Orimet. These tests measure the time it takes for the SCC concrete to
flow through an orifice under its own weight, these tests give an indication of its

viscosity.

Sieve Stability. This test is used to evaluate the resistance of SCC to static

 

segregation. It consists of pouring a concrete sample over a 5-mm sieve and

measuring the amount of mortar passing through the sieve in a two-minute period.

The above methods are not exclusive and while different research groups around the

world have evaluated these concepts, dimensions and measurement specifications have

not been standardized and/or commonly accepted [30].

2.1.3 SCC Hardened Property and Structural Performance

Since the target engineering properties of hardened SCC should be the same as

those for conventional concrete, the same test and procedures that are used for

conventional concrete are used for SCC. Most of the research work to-date on SCC has

been focused on three fronts: (1) development Of self-compacting mix designs

16

[11][12][26][39], (2) comparison Of mechanical properties of SCC against well
compacted regular concrete [1][28], and (3) evaluation of in- situ properties of SCC in
full-scale structural elements [21][40]. The above-mentioned efforts have shown (a) the
feasibility of mix designs with self-compacting behavior and the development of simple
and optimized combinations and (b) that, overall, the uniformity and value Of the
mechanical properties of SCC do not vary significantly from that Of normal well-
compacted concrete [5].

In spite of the above-mentioned research developments, much less is known on
the performance of structural elements cast using SCC. While similar performance might
be expected based on the similitude of mechanical properties between SCC and normal
concrete, the studies performed thus far have mainly focused on compressive strength,
elastic modulus, and to a lesser extent on creep and shrinkage [28]. While, as mentioned
above, these properties were found to be similar to those corresponding to traditionally
vibrated samples, SCC has been found to exhibit higher early age creep coefficients [28],
and not much has been reported on fracture strength.

Researchers from University of Sherbrooke, Canada [21][22] evaluated the
uniformity of in-situ mechanical properties of wall elements using eight different mixes
of SCC. Cores were obtained at various heights of the wall and normal hardened
properties like compressive strength and elastic modulus were evaluated. The results
from this study show very slight or no variation in the compressive strength and modulus
of elasticity values obtained from the top and bottom portions of the wall.

These researchers [21][22] also compared the mechanical performance of highly

confined columns cast using normally consolidated concrete (NCC) and self compacting

17

concrete (SCC) with stirrup configurations representing different degrees of confinement.
The cores were cut at various heights of the column. Results from these cores were
compared with those of normally cast specimens. The test results showed that the SCC
columns developed similar stiffness but approximately 7% lower load carrying capacity.
Also the SCC showed 10% lower cylinder compressive strength relative to NCC.
Depending on the reinforcement configuration SCC columns exhibited 62% to 23% more
ductility than corresponding NCC columns. The distribution of in-place properties was
found to be more homogeneous in SCC than in NCC.

As previously mentioned, the nature Of SCC proportioning is to deviate from
well-understood mix designs developed over many years of experience. This implies that
compromises must be made with respect to performance. Highly fluid mixes, or hi gh-rate
discharge, can compromise the stability of the concrete mix, which can lead to bleeding
and segregation around reinforcement that can adversely affect the bond characteristics of
conventional and prestressed reinforcement such as transfer and development length.
Another potential adverse effect of SCC on structural performance is that related to the
ability to transfer shear stresses across cracks, or the so-called “aggregate interlock.”
Thus, it is foreseen that bond and aggregate interlock are two types of hardened
behavioral mechanisms that can affect the design and performance of precast/prestressed
bridge elements using SCC. The nature and importance only the only bond type of

behavioral mechanism is discussed next.

2.2 Importance of Bond in Prestressed Concrete

The importance of bond between prestressing steel and concrete has been studied

considerably during the development of prestressed concrete [23]. The concept behind

18

prestressing and reinforced concrete clearly relies on the ability to transfer tensile forces
from the strands into the hardened concrete both during service as well as at ultimate.
This parameter (bond) thus forms part of the design considerations of precast/prestressed
members through semi—empirical formulations of development length to ensure proper
anchorage of the prestressing strand when relying only on the interaction between the
strand and the surrounding concrete. The behavioral mechanism and importance of bond
for both (transfer and ultimate) should be understood if a suitable measure of this

parameter (bond) is to be performed.

2.2.1 Bond Stresses and Mechanisms

The transfer Of stresses from the strand to the concrete by bond can be classified
into three distinct mechanisms: (1) adhesion, (2) Poisson’s (Hoyer’s) effect and
(3) mechanical interlock. Each of these mechanisms is explained briefly explained in the
following:

1. Adhesion: This refers to the chemical bond resisting mechanism by which the
concrete bonds to the strands. Adhesion helps in the bond transfer only when
there is no relative slip Of the strand with concrete. This mechanism is relatively
small and hence often neglected [16].

2. Hoyer’s Effect: When the strand is prestressed there is a lateral contraction in the
strand diameter due to the steel’s Poisson’s ratio. When the strand is released, this
lateral contraction is recovered and the strand swells. This swelling is prevented
by the surrounding hardened concrete preventing the strand to return to its
original diameter. This restraint is in the form of a radial normal force on the

strand inducing frictional force along the axis of the strand (Figure 2-5). At the

19

end of the beam, where the surrounding concrete does not exist, the strand is free
to expand, and hence the strand at the end has a relatively larger diameter than the
portion embedded in concrete. This produces a wedge action. This anchorage
mechanism was first described by Hoyer in 1939 [36] and thus is commonly
referred to as Hoyer’s effect. The concrete resists this wedging effect transferring

part of the stress from the strand to concrete.

f . /
radial pressure or

Figure 2-5 Hoyer’s Effect [34]

      
  
 

02:0

   
 

3. Mechanical Interlock: Seven wire prestressing strands consist of six wires wound
in a helical shape around a single wire. When concrete is cast around the strands
the hardened concrete shape matches that of the stressed strands. Due to the match
casting between the concrete and strand, the concrete resists the unwinding of the
strand providing slip resistance. Mechanical interlock is the main contributor to

bond when the stresses are increased beyond the initial transfer stresses [36].

20

2.2.2 Transfer Bond Stresses (at release of prestress )

The approximate distribution of the bond stress 2' at transfer due to all the
mechanisms cited above is shown in Figure 2-6 [23], where 1 becomes zero, the stress in
the strand becomes equal to the stress due to prestressing (0', = 03,. = constant). The
length associated with this is termed the bond length, u. It will depend on the quality of
the bond and on the transverse pressure determined by the member geometry and
transverse reinforcement. The prestressing force is introduced into the member until the
concrete stresses exhibit a linear distribution over the section. The length needed for

achieving this is referred to as the transfer length, L,.

 

 

 

323...... m... //////////////////////////////% 3:31.

I ransmlsslon length'
l‘——"—_’l

Figure 2-6. Bond Stress Distribution at Strand End [23]

At the face of the concrete unit, the steel and concrete stresses are zero. The shear,
or bond stress between the strand and the concrete increases rapidly until it reaches its

maximum value, beyond which it decreases approximately according to a parabolic

21

curve. Due to the bond effect, compressive stresses radiate from the wire into the
concrete causing warping at the member end (Figure 2-7) [23]. As a consequence of this,
a zone Of compressive stresses develops acting radially towards the strand. This further
enhances the Hoyer effect. However, this deformation also introduces tensile stresses that

require the provision of transverse reinforcement.

ion trajectories

 
 
  
  

 

:Warping
20f cls at
—.end

    

Prestressing
Strand

Transverse
tension

Zone of radial
pressure

 

Transverse
Compression

I

\\\\
\V'
\\§\\

 

S1

 

Figure 2-7. Section Warping and Forces at End [23]

22

Of all of the above-mentioned mechanisms, the Hoyer effect is the greatest
contributing mechanism to “bond” upon prestress release. Mechanical interlock is the
main contributor to “bond” when the stress in the strand is increased above the initial
transfer stresses, i.e., when the concrete cracks and the strand stress levels are increased
over their initial state. The adhesion mechanism is the smallest contributor to developing

bond stresses between strand and the concrete [23][36].

2.2.3 F lexural Bond Stresses at Ultimate Strength

In most cases, full prestressing is provided in the design of prestressed concrete
members, which implies that there will be no cracks under working load. It can be shown
that the demands placed upon bond action before the appearance of cracks is very limited
since the transmission of shear takes place just as in a section made from a homogeneous
material. For this reason, prestressed concrete beams can adequately carry working loads
even without the presence Of bond [23]. If bond does exist, then only a very slight shear
stress occurs between the tendon and the concrete due to the connection between the
small area of steel (times the material modular ratio) and the rest of the total cross-
section. Thus, only when the working load is exceeded and cracks occur in the tensile

zone of the concrete does bond become necessary [23].

The importance Of bond on ultimate strength is better understood by considering
the failure behavior with and without bond. Upon loading and appearance of the first
crack at the location of highest tensile stress, there will be a sudden increase of tensile
stress in the strand due to the abrupt disappearance of the concrete tensile force
contribution. If there is no bond, this increase will extend over the entire length of the

tendon. This will lead to considerable deformations and wide crack spacing. In addition,

23

new flexural cracks will be widely spaced, which will tend to decrease the depth Of the
compression zone and eventually lead to early failure of the compression flange. Thus, in
the absence of bond the ultimate capacity is reduced and the strength of the steel strand

cannot be fully utilized.

Early failure is thus avoided by establishing a shear-resisting mechanism, i.e.,
bond, between the steel and the concrete. The bond stresses between the tendon and the
surrounding concrete, 2'], have the effect of immediately reducing the increased stress that
develops in the steel at the crack location a short distance next to it (Figure 2-8) [23].
Depending on the bond quality, the increased steel stress 0'Z is thus limited to a short
length, which will lead to only slight local elongation of the strand and narrow crack

spacing.

é///////////////////////////////

Wfl/Yl/l/lfl/i’l/fll/l //////////////M’lfl//// [Ml/IllW/f/ ///////I7.’///

    

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F'F‘iifli .T M E tendon
: crack
I, i
«ll bond stress
llulu '
e a .. - :
l
' concrete tensile stress
0' ..- I
2 WT ”W i
u [I steel stress

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 2-8. Stress Distributions at a Crack Front [23]

24

The presence of bond will permit the tensile concrete stresses 0",, to continue to
exist besides the crack location, which can increase further upon further loading. This
will lead to closely spaced cracks and a failure pattern with a large number of cracks that
slowly move upwards. The neutral axis depth is thus slowly reduced, allowing
development Of large steel stresses. Consequently, bond allows for a safe failure mode

and better use of the steel reinforcement [23].

As pointed out above, a series Of thin cracks are expected to be developed in route
to the ultimate flexural capacity Of a prestressed concrete member. Thus, in addition Of
addressing the mechanism of bond at a single crack, the distribution Of bond stresses
along the member length is now of importance. First, it must be understood that bond
stresses in the cracked region cannot be determined by the approach followed in
reinforced concrete. At flexural cracks the bond stress beside it will locally increase up to

the bond strength as shown in Figure 2-8 and

Figure 2-9 [23]. Beyond this peak, the bond stress quickly decreases and in some
cases it even changes into a stress of opposite sign. If there are a number of cracks in
succession, the distribution of bond stresses follows a “saw-tooth” pattern [23]. Although
not described in this manner, this same phenomenon was identified by Hanson and Kaar
[17] (Upon whose work the current AC1 recommendations [3] for strand development
length are based) as a “bond stress wave.” Calculation of these bond stresses, however, is

not possible without an accurate knowledge of the bond strength and bond strains.

25

 

 

 

  

 

F—--—1r—--—-

 
  

 

 

 

 

 

 

 

 

52+—

' State ll

State I

 

 

 

Figure 2-9. Bond Stresses at Flexural Cracks [23]

The development of new flexural cracks towards the end of the member will
continue with increasing load demands due to load redistribution. As seen in

Figure 2-9, the bond stress demands will follow along with localized high steel
stress demands at the crack [23]. Increased tensile stresses in the strand will cause a
reduction of cross sectional area due to Poisson’s effect. Thus, if cracking extends into
the transfer zone region, the reduced cross-section tendon area will compromise the
Hoyer effect, which is the main mechanism for bond in the transfer region. Relative slip
of the strand can then occur leading to a reduction in prestressing force and thus limiting
the attainment Of the section full flexural capacity. In addition, the reduced compression
stress state at the beam end will decrease the section shear capacity. Bond failures and

shear failures at end supports are thus clearly related, an issue that has been identified for

26

some time but yet continues to be topic of debate as to the precedence Of each effect

[811331135]-

2.3 The Concept of Transfer and Development Length

Bond is of importance to the design of prestressed members for both initial, or
service, as well as ultimate, or overload conditions. The bond strength between
prestressing strands and concrete depends on the concrete’s ability to transfer shear forces
along the material interface. The distance over which the effective prestress f,., is
developed in the strand has been traditionally called the transfer length, L, An additional
bond length is required so that the stress fin may be developed in the strand at ultimate
flexural strength of the member. This additional length is termed the flexural bond length,
Lf. The sum of these two lengths is commonly referred to as the development length, L,,
of the strand (Figure 2-10).

Considerable work has been done regarding the development of a better
understanding of bond and its relationship with transfer and development length for
conventionally consolidated concrete. Consequently, several theories for transfer and
development length have been formulated. In spite of the multiple efforts, current
provisions by the ACI—318 Code (318-02) [3] and AASHTO [2] (see Figure 2-10) are
primarily based on the work of Hanson and Kaar [17]. However, in general, it has been
found that these guidelines underestimate the actual transfer and flexural bond lengths,
and their validity has been questioned for over 25 years [8][10][18][24][31][32]
[33][35][41]. In this time several approaches have been developed. In the following, a
brief account of these efforts is given and then the proposed approach is described in

detail.

27

 

 

 

 

 

 

 

 

 

 

 

|
4—Development Length—>
Transfer F lexural Bond
Length Length
T ’7?—
rn fsu 'fse
m
0
33 v
m ——A
E
in" fsu
fl.
7 WV
4—fie:,~[: fsu 'fi: ;
+—l.,:>| e ’2’ >
e "d 4

 

 

Distance from the free end

Figure 2-10 Variation Of stresses — ACI-318 equation representation [3][2]

Although strand development seldom governs the design of prestressed concrete
members, with the exception of cantilevers and short span members, several bond-related
failures have been reported with members using conventionally consolidated concrete
and the current criteria [8]. In addition, the relationship between development length and
shear failures at beam supports has been clearly documented and has become a recent
concern, particularly as it relates to the response of prestressed beams due to earthquake-
induced vertical loading. It is then clear that the issues related to bond between hardened

SCC and prestressing strand will be reflected in design practice through these parameters

28

2.3.1 Definitions

Some of the basic definitions that describe the behavior of bond in the prestressing
strand are described below:
Transfer Length (L, ):

Transfer length is defined as the bonded length Of the strand required to fully
transfer the effective prestress (f,,) from the strand to concrete. In other words, transfer
length is the length from the end Of the beam to the point where the prestressing force is
fully effective [34].

F lexural Bond Length (L,):

Flexural bond length is defined as the distance, in addition to the transfer length
Over which the tendon must be bonded to the concrete to develop the full design strength
of the tendons (fin) to resist flexural stresses at nominal resistance Of the member.
Development Length (L4 ):

Development length is defined as the total length of bond required to develop the
steel stress fi,, at the ultimate strength of the member. Development length is the sum Of
the transfer and flexural bond lengths. In other words, Development length can be
defined as the shortest bonded length of tendon along which the tendon stress can
increase from zero to the stress required for achievement of the full nominal strength at
the section under consideration.

Embedment Length (L,):

Embedment length is defined as the length of bond from the critical section to the
beginning of bond. Critical section can be defined as the section closest to the end of the
member that develops full strength when subjected to external loading. It should be noted

that embedment length is related to development length. If the embedment length is

29

greater than the development length then a general bond failure would occur, else the

strand slip will occur and the nominal stresses will not be developed in the strand.

2.3.2 ACI-318l31/AASHTO- LRFD [2] Code Recommendations

The current provisions by the ACI Code (318-02) [3] and AASHTO Design
Guidelines (Figure 2-10) are primarily based on the work of Hanson and Kaar [17][18].

The current AC1 318 / AASHTO provisions are as follows:

1 2 . .
Ld : @(fps Tgfse)db [fps’fsem MPa; db’Ld m In]
(2-1)
2 . . . .
Ld = (fps —-§fse)db [fps,fsern ksr & db,Ld In rn.]
Equation 2-1 can be re-written as:
Ld =—1- -fS—edb—(f S —fs,)db]. [f s,fsein MPa; db,Ld in m]
6.895 3 P P
(2-
Ld = [%db — (fps — f,,)db], lfps, fsein ksi & db,Ld in in.]
2)

where, db is the diameter of the strand, fl, and fps are the effective and nominal stresses in
the tendon respectively.

The first term Of the Equation 2-2 represents the transfer length and the second
term represents the flexural bond length. In Figure 2-10, the steel stress is shown to vary
linearly along the transfer length. Along the flexural bond length, the slope of this curve
decreases. AASHTO-LRFD specifications state that the flexural bond may be assumed to
vary parabolically [7][3]. The ACI code also states that the transfer length may be taken

as 60 strand diameters.

30

2.4 Studies on Transfer and Development length of Prestressing strand.

The previous discussions have presented the mechanisms and the importance of
bond between prestressing strands and concrete for the design and performance of
prestressed concrete beams. Determining the bond strength, or slip resistance, is
however, not a straightforward task. The reason is that the “bond” phenomena that have
been described for both transfer and ultimate strength rely on different mechanisms to
transfer the shear stresses between concrete and strand. It was explained how the bond
shear stresses follow complex distributions at the member ends and at flexural cracks.
Thus, appropriate determination of bond strength must replicate actual conditions as close
as possible.

It has been found that bond mechanisms are affected by many parameters [41].
Among them are:

0 The type of Steel

0 The diameter of the strands

0 The level of stress in the strand

0 The surface condition of the strand

0 The concrete strength

0 The type of loading (static, repeated, impact)

0 The type Of prestress release (sudden, gradual)

0 Confinement reinforcement

0 Time-dependent effects (losses due to creep, shrinkage etc.,)

0 Concrete cover and spacing of strands

0 Amount of shear reinforcement in the critical zone.

31

2.4.1 Studies on Bond Performance
In order to determine values of the slip resistance of a prestressing strand through
experimental methods, care must be taken to consider the mechanisms affecting bond

response [23]:

l. The bond stress (slip resistance), typically assumed as uniformly distributed, is
higher for shorter embedment lengths in a pull-out test specimen. This is shown in
Figure 2—1 1. This maximum stress value has a decisive part in the shear stress vs.

slip response.

2. The slip resistance increases with the quality, the compaction, and the degree of
hardening of the concrete, thus the relevance of evaluating this mechanical

behavior parameter for SCC.

(a) d d

1 Average.

 

          

 

 

_ W/ ///%////i k MaxT~ 5 Max 't'>
T 1 Average 1- Average 2'

Figure 2-11 Effect of Embedded Length in Normal Pull-Out Tests [23]

3. The slip resistance is significantly dependent on whether a transverse compressive
stress acts on the strand. This effect can be reproduced through normal pull-out

tests (see Figure 2-11a).

32

4. Additional transverse pressure is created by the Hoyer effect.

Because of all of these influencing factors, it is difficult to determine bond lengths
by means Of simple pull-out tests. Tests on strand bond confirmation by means of the
“Moustafa test” (Figure 2-12) have been recommended by the PCI Interim Guidelines for

SCC [30].

CHUCK \4]

 

Hydraulic Ram \

 

 

Steel Chair
\

2” Sleeve
\Eiia ‘ .

 

 

 

A
b
s
a
a
A
a
s
Q
0 4
4
is
ab ‘
a
Q
A
a
a
“b
a
‘ b
G
e

a
be
45
a

bed me

18” "
24"

 

 

a
A
a
e
A
ti»
5
A
s
1;
4
s
4
Its
a
a
n
a
b
e

 

 

 

 

 

 

 

 

 

 

 

 

 

4” Sleeve
36!!

 

 

 

Figure 2-12. Details of Moustafa Test

It should be noted that the bond performance is related to the bond mechanisms
and complex phenomena. The results of these simple pullout tests depend on the bond
resistance created by friction and mechanical interlock. The Hoyer’s mechanism which is

the main contributor for transfer length [7] is not represented in these tests. The

33

correspondence between the results obtained from this test and structural design
parameters such as transfer length and development length have been questioned for
conventional concrete [7][27][33][32] and seem to be of continued debate now for SCC.
In order to qualify the strand for adequate bond performance, the test has to be performed
in a specific manner on a specific mix design of concrete. The following are the

guidelines and test procedure as prescribed by Logan [25]:

1) The hydraulic jack shall be a pull-jack with a center hole assembly at the end
of the ram (similar to those normally used for single-strand stressing). It shall
be tested and calibrated to permit loading upto 50 kips (222 kN), and shall
have a travel of at least 12 in. (305 mm)

2) A specific bridging device should be used [25].

3) On the day of casting the test blocks (with heat curing), the cylinbdrs shall be
tested and the concrete strength recorded. Based on results of past testing, the
concrete strength can range from 24.1 to 40.7 MPa (3500 to 5900 psi ) without
affecting the pull-out strengths.

4) The bridge is slipped over each strand to be tested and placed against the

concrete surface. The strand chucks are slipped over the strand to the top of the

bridge and light pressure is applied to the jack to seat the jaws of the chuck into
the strand.

5) The jacking load shall be applied in a single increasing application of load at

the rate of approximately 20 kips per minute (89kip per minute) until the

maximum load is reached and the load gauge indicator can no longer sustain

maximum load. Do not stop the test at the sign of movement of the strand sample

34

or for any other reason. The strand samples can pullout as much as 203 mm to
254 mm (8 to 10 in.) before maximum load is reached with a poor strand, and
25.5 to 51 mm (1 to 2 in.) with good bonding strand.
6) The pull-out capacity of the strand sample shall be recorded as the maximum
load attained by the strand sample before the load drops off on the gage and
cannot be further increased.
7) The following data shall be recorded for each strand sample:
(a) Maximum capacity
(b) Approximate load at first load movement
(c) Approximate distance the strand pulls out at maximum load
((1) General description of failure
8) Record data and compute average failure load and standard deviation for eah
strand group tested. Compare results with minimum requirements for acceptance
for pretensionin g applications.
As discussed earlier, tests on strand bond confirmation by means of the “Moustafa
test” (Figure 2-12) have been recommended by the PCI Interim Guidelines for SCC [30].
However, the correspondence between the results obtained from this test and structural
design parameters such as transfer length and development length have been questioned
for conventional concrete [7][27][33][32] and seem to be of continued debate now for
SCC. While the response evaluated through the Moustafa test is clearly related to bond
performance, its correlation to the complex phenomena occurring in the transfer zone
region and during development of strand capacity under flexural actions, as previously

discussed, is questionable.

35

The pull-out tests and the Moustafa test are excellent methods to provide a
baseline to qualify the strand bonding characteristics, which may be affected by rust or
manufacturing residues. Thus, bond length evaluation needs to be conducted in a manner

consistent with the stress state in both transfer and flexural regions.

2.4.2 Studies on Transfer and Development Length

Significant research has been done in the past to investigate the bond mechanisms
of prestressing strand in concrete as it relates to transfer and development length. Existing
code provisions for the development length of fully bonded strands are based on the
results of two studies conducted in the Research and Development Laboratories of
Portland Cement Association (PCA) [7]. The results of these studies were reported in
papers by Hanson and Kaar [17] and Kaar, LaFraugh and Mass [18]. An overview of the

research including the work conducted at PCA is described in brief in the following:

a) Paul H. Kaar, Robert W. LaFraugh and Mark A. Mass [18]

This study was presented at the Ninth Annual Convention of the Prestressed
Concrete Institute in 1959. This study investigated the influence of concrete strength on
the stress transfer length of seven—wire strand at the time of prestress transfer. Strands of
6.35mm (MI), 9.53 m (3/8), 12.7 mm (1/2), and 15.24 mm (0.6 in.) diameter strands were
used to prestress rectangular prisms having various concrete strengths. For all strands
except the 15.24 mm (0.6 in.) diameter strands, smooth, unrusted strands were used. The
length of the prisms was 2.44 m (8 ft) for all specimens except 6/10 diameter which had a
length of 3.05 m (10 ft). The cross section of the prisms varied with the size of the strand.

The researchers concluded that the concrete strength has little influence on the

transfer length for strands up to 12.7 mm. ('/2 in.) diameter. The 15.24 mm (0.6 in.)

36

strands had a smaller transfer length for concrete with higher compressive strength and
vice-versa. The influence of strand diameter was also studied in this research. The
researchers found that transfer length varied linearly with respect to strand diameter.
They also found that the average increase in transfer length over a period of one year
following prestress transfer was 6% for all strand sizes and that the increase in transfer
length was independent of the concrete strength at transfer.

The method of measurement of transfer length used in this project was by means
of the DEMEC (DEtachable MEChanical) gauge method which has also been used in

many other research programs to measure transfer length of prestressin g strands.

b) Norman W. Hanson., and Paul H. Kaar. [17]

This study was performed at the Portland Cement Association (PCA) laboratories
in 1959 and is considered to be the backbone for the current approaches to deve10pment
length testing and code provisions. The test program involved 47 beam tests, with
varying diameter of the strands were tested in a series of flexural tests. The main variable
in the study was the strand diameter. Secondary variables included the percentage of steel
reinforcement, the concrete strength, the strand surface condition and the use of
embedded end anchorages on pretensioned strands. The authors propose a hypothetical
shape of the bond wave from the flexural test results before slip.

The researchers found that strand size and embedment length have a considerable
influence on the value of average bond stress at which general bond slip occurs. From the
test results, the researchers determined curves that could be used for design such that
general bond slip could be avoided. They also found out that the increase in percentage of

reinforcement or a reduction in concrete strength reduces the possibility of general bond

37

slip, since the steel stress at flexural failure, and the corresponding bond stresses are
reduced. The results of this research are the basis for the current ACI-318

recommendations (Equation 2-1 and 2-2) for development length of prestressing strands.

c) Paul Zia and Talat Mostafa [41]

This study was performed at the University of North Carolina, Raleigh, United
States. The authors proposed a new equation for transfer length of prestressing strands
that accounts for the effects of strand size, initial prestress and concrete strength at
transfer. This equation is applicable to concrete strengths ranging from 13.8 MPa
(2000 psi) to 55.2 MPa (8000 psi.). The researchers also studied the various parameters
affecting the transfer and development length and reviewed the experimental results of
the previous researchers.

The researchers found that the use of reinforcement to resist the bursting stress
near the end of prestressing steel slightly reduced the transfer length, although the effect
was not significant. Based on the review of the then available research information, the
researchers proposed the following new equations for the transfer ( L, ) (Equation 2-3)
and development length ( L4) (Equation 2-4), which is applicable for concrete strengths

varying from 13.8 MPa (2000 psi) to 55.16 MPa (8000 psi):

1, = 1.5%.1, —4.6, f,,, f,,-(ksi) & db (in.) (23)

Cl

Ld =1-5%db -4.6 +1.25(fps —fse)db, fsi,fci,fps (ksi) & db (in.) (2-4)

Cl
where, f,,- is the initial prestress force, f,,- is the concrete strength at transfer, db is the
nominal diameter of the strand, f,“ is the ultimate stress in the strand and f,, is the

effective stress in the strand after transfer.

38

d) Byung Hwan 0h., and Eui Sung Kim. [9]

This study was performed at the Seoul University, Korea. The main objective of
the research was to study the effects of various important parameters on the transfer
length on pretensioned, prestressed concrete girders. The principal test variables were
strand diameter, concrete strength, concrete cover and strand spacing.

Results of this research showed that the current ACI-318 code equation for
transfer length overestimates the actual measured transfer length. This overestimation is
more significant in high strength concrete with larger concrete cover, which is not
considered in the current equation.

The experimental program included testing of 36 pretensioned, prestressed
concrete beams. The transfer length was measured by the concrete strain profiles
measured using the DEMEC system and end-slip measurements. Results from the
research can be summarized as follows:

0 Strand diameter: The transfer length for 15.2 mm strand was found to be 25%
longer than the 12.7 mm strand. Using the ACI code equation (Equation 2-1), the
15.2 mm strand was found to have a 20% longer transfer length than the 12.7 mm
strand.

0 Concrete Strength: The measured transfer lengths for high strength concrete were
shorter than the lower strength concrete. Transfer length of prestressed members
with a compressive strength of 45 MP3 (6500 psi) was found to be approximately
12% shorter relative to similar members with a compressive strength of 35 MPa
(5075psi) for 12.7 mm (0.5 in.) diameter strands. Similarly, for 15.2 mm (0.6 in.)

diameter strands, this decrease was found to be approximately 15%.

39

Cut End Vs. Dead End Efi‘ect: The cut end refers to the end of the specimen
where the ends are cut and the dead end refers to the undisturbed end where the
stresses are relieved by the cutting of the strands in the other end. The transfer
lengths at the cut end were found to be consistently longer than the dead end due
to the sudden release of the prestress. On average, the researchers found 16% and
13% increase in transfer lengths at out ends for 12.7 mm (0.5 in.) and 15.2 mm
(0.6 in.) strands respectively

Strand spacing: The strand spacing studied was 2d,, 3d,, and 4d,, where db is the
nominal diameter of the prestressing strands. The spacing was varied and the
transfer lengths were measured for both the 12.7 mm (0.5 in.) and 15.2 mm
(0.6 in.) strands. Reduction of clear strand spacing from 3d,, to 4d,, resulted in a
remarkable increase in transfer length. The increase in strand spacing from 3d,, to
4d,, , showed a very little reduction in transfer length.

Concrete Cover: The side cover was kept constant at 5cm and the clear cover was
varied. The beams contained only one strand and the clear bottom cover values
tested were: 3,4 and 5 cm. The researchers found that the transfer length increased
with reduction in clear bottom cover.

Time dependent efi’ect: The transfer length was measured at release, 7 days, and
periodically up to 90 days. The authors found a shift in the transfer length profiles
but not significant increase in transfer length values, this was due to the increase
in axial concrete strains due to creep and shrinkage. The increase in transfer

length at 90 days was found to be 5%.

4O

0 Prediction from End Slip: The researchers predicted the transfer length from end-
slip measurements at the cut and dead ends. They found that theoretically
calculated transfer lengths from end-slip measurements correlated well with the

test data.

e) Mohsen Sahawy [35]

In this work the author performed a critical evaluation of the existing proposals of
calculating the development length of prestressing strands. He also discusses on the
extensive test programs on variety of prestressed concrete members and the modifications
to the existing equation made by various researchers over the past 10 years (since 2001)
are discussed. The objectives of this research were to compare and contrast the
development length equations given by AASHTO specifications (Equations 2-1 and 2-2)
and the equations proposed by other researchers, compare the federal highway agency
(FHWA) results with findings of florida department of transportation structures research
center (FSRC) and to present a rational method for calculating development length of
prestressing strands. Based on the study, the author proposes two equations for
development length, depending on the depth of the girder:

(a) For members with depth equal to or less than 610 mm (24 in.)

 

Ld = (£351)D+ (fsu l—fe)D , (fsi’fsu ,fse) (ksi) & db (in.) (2-5)

(b) For members with depth greater than 610 mm (24 in.)

 

L, =(I;%)D+ Us" '1' :56”) +1.47h, f,,-,f,,,,f,, (ksi) and D,h (in.) (2-6)

41

where, D is the nominal diameter of the strand, f,“ is the stress in the strand (in ksi) at
nominal strength of the critical section, f,,- is the stress in the strand at time of initial
prestress (ksi) and h is the overall depth ( in.) of the member.
The brief features and results from this investigation are as follows:
0 Shear — flexural interaction affects the development length of prestressing
strands and that it should be included in the design recommendations.
0 The effects of shear are more pronounced in deeper members and the
author proposes his new equation taking this effect into consideration
0 For prestressed concrete members with depth equal or greater than 610
mm (24 in.), the existing AASHTO [2] equation and the proposed
equation (Equation 2-3) yields the closest prediction, while FHWA [35]
and Buckner predictions [8] (increase in development length by a factor of
1.6) are extremely conservative.
0 For prestressed concrete members with a depth greater than 24 in., the
AASHTO equation [2] yields unacceptable low values for most cases and

the new proposed equation (Equation 2-6) gives conservative values.

I) Deatherage, J.H., Burdette, E.G., and Chew, C.K., [14]

This study was performed at the University of Tennessee, Knoxville. Twenty full
scale AASHTO Type-I girders with large strand diameters: 12.7 m (1/2 in.), 14.3 mm
(9/16 in.) and 15.2 mm (0.6 in.) were statically tested to failure. Transfer and
development lengths were measured and factors affecting both transfer and development
length are discussed and evaluated. Based on the test data, new equations are proposed.

From the existing equation, the authors proposed replacing the initial stress instead of the

42

effective stress in the transfer length term and introduce a conservative factor of 1.5 in the

flexural bond length term. The proposed equation is given as follows:
Ld = (%)D+l.5(fps —fse)D (2-5)

Based on the study, the authors concluded that the 15.2 mm (0.6 in.) strand should be
accepted as a common practice. The authors found that 15.2 mm (0.6 in.) strands have
much shorter transfer lengths relative to any other strand diameter used in their research.
Also, the measured development lengths from 15.2 mm (0.6 in.) strand were comparable
with those of other diameter strands used. The ultimate capacity of members with 15.2

mm (0.6 in.) were substantially higher relative to all other strands used in the study

2.4.3 Oustanding issues

With the exception of cantilevers and short span members, strand development
seldom governs the design of pretensioned concrete members. Nevertheless, several
bond-related failures of pretensioned members have been reported since the adoption of
the current criteria.

Diameter of the strand

In current practice, Grade 1860 MPa (270 ksi) strand with a higher tensile
strength, 1.860 MPa (270 ksi) and larger cross sectional are is used. Low-relaxation
strand with higher yield stress has replaced stress relieved strand. The current ACI-318 /
AASHTO LRFD expression for transfer length was derived using a bond stress of 2.76
MPa (400 psi), which represents the average values form the tests conducted by Hanson
and Kaar [17]. This stress applies to the actual perimeter of the seven wire perimeter of

the strand. For Grade 2705trand, this constant is about 6% larger. Despite wide variations

43

in measured values, several researchers have concluded that the transfer length increases
directly with strand diameters ranging up to 15.7 mm (0.6 in.).
Shear - Bond interaction

The issue of web-shear cracking and bond of prestressing strands has been
discussed and debated for a long time. The interaction between shear and bond has been
considered to be cause for slip failures[17][7][8] It has been found that the initial slip
occurred coincident with the web shear cracking[7][8]. However, there is a doubt on
whether web-shear cracking initiates strand slip or vice versa. Researchers from
University of Texas at Austin [7][8] found that the results indicate a direct interaction
between shear and bond with the initial slip occurring immediately or shortly after the
appearance of first shear crack. The best documented evidence found to explain
interaction between shear and bond gives a strong indication that general bond slip
occurred prior to the sudden shear failure [7][8].

Shahawy [35] proposed two different equations (Equations 2-5 and 2-6) for
development length depending on the depth of the member. He found that the effects of
shear on development length cannot be neglected on members greater than 610 mm (24
in.) as there exists interaction between shear and flexure and the slippage of the strand is
most likely to occur before the flexural capacity of the member is achieved.

The wide variations and the confusion in the validation of the current code
provisions may also be due to the inconsistency in the amount of shear reinforcement
used in the various research studies performed. Researchers have used different amounts
of shear reinforcements in their studies on transfer and development lengths. Taking into

account the shear-bond interaction, researchers who have had relatively higher shear

reinforcement are more likely to have lower transfer and development lengths and vice
versa. This may also be a reason for the wide variation in the results in past studies.
Shear-bond interaction must be taken into consideration to properly correlate the
variations in transferand development length results available.
Failure strains in test specimens

The wide variation in the results of development lengths have also been attributed
to the level of strand strains at section failure. Apparently to simplify testing, most
development specimens have been proportioned to fail at relatively low strains. The
exceptions have been girders with cast—in place composite slabs [8]. Experimental results
from most test programs suggest that average bond strength is lower in specimens with
large strand strains at failure (eg. strains near the guaranteed ultimate minimum
elongation of 0.0350) as compared to specimens that failed with strains near the yield
strain. Studies indicate that the studies on non-composite sections with failure strains
much lower than the guaranteed ultimate minimum elongation showed that the ACI-
318/AASHTO LRFD recommendations were conservative. At the same time, composite
sections with failure strains near the guaranteed minimum elongation showed 1.7 times
longer development length than the ACI—318/AASHTO LRFD recommendations [8][35].
A possible reason for this discrepancy has been attributed to the relative difference in
strain levels at failure. Members with composite sections represent the actual bridge
applications and hence it has been recommended that development length studies should
be made with cross sections that would develop strand strains near the minimum

guaranteed ultimate elongation.

45

2.5 Concluding Remarks

From the above-mentioned studies, it is clear that SCC is of high interest due to the
many advantages it provides to the precast industry for improved construction efficiency
and quality. However, it becomes obvious that most efforts to date have focused on the
material aspect of SCC and only limited efforts have validated its structural performance.
For precast/prestressed concrete construction, the bond between concrete and prestressing
strands is of primary importance. Given the complex nature of bond stresses and the
mechanisms controlling them; along with the varied results and continuous debate over
the existing code recommendations for conventional concrete, it becomes essential to
evaluate the bond performance of prestressing strand in members built using SCC. The
evaluation of bond performance on SCC is not an easy task taking into account that there
is no commonly accepted procedure to proportion SCC mixes. Proper evaluation of the
bond performance and evaluation of the respective bond parameters (transfer and
development lengths) is essential to take the advantages SCC offers into safe

implementation in precast/prestressed concrete construction.

46

CHAPTER 3 MIX DESIGN DEVELOPMENT AND
EVALUATION

Due to the many options available for obtaining SCC, the goal of this research was to
bound the effects of SCC mix proportioning by considering extreme conditions for its
proportioning. This chapter describes the philosophy behind the selected mix designs for
this project and provides the specific proportioning used. Also included herein is the
evaluation of the fresh and hardened properties of the experimental mix design matrix.
SCC mix development and test results are compared with a reference normally

consolidated concrete (NCC) mix.

3.1.1 SCC Mix Design Approaches
As previously discussed in Chapter2, SCC achieves its fresh property advantages

through specially proportioned mix designs that significantly deviate from what is
considered ideal and developed through many years of research and development. The
tailorable design of SCC mixes for fresh and hardened properties gives infinite
possibilities to obtain SCC. While there is no commonly accepted procedure to
proportion SCC mixes, over the years several methods have been developed in research
centers around the world [20][11][12][27]. In spite of the different methods of achieving
SCC, it is commonly agreed that all methods are bounded by two main approaches [20] :
0 Approach 1: Proportioning concrete with moderate w/c ratios (e.g., 0.45), and use
of HRWR and VMA to provide fluidity and increase stability, respectively. The
VMA increases both the yield value and viscosity, while the HRWR reduces the
yield value. The resulting combination provides a mix with relatively low yield and

moderate viscosity.

47

0 Approach 2: Mixes without any viscosity-enhancing admixture, but with lower w/c
ratios (e.g., 0.33) to reduce free water content and provide stability and use of a

relatively high content of HRWR to provide hi gh-fluidity.

Due to the wide variety of mix designs that have been proposed, and that can be
developed to create SCC, this research focused on bounding the proportioning techniques
for SCC. Hence an approach was adopted in which the mix design characteristics of SCC
and their effect on material and structural properties were bounded, to provide designers
with knowledge on the compromises made through optimization of an SCC mix for
fresh-concrete behavior. This will allow design freedom to tailor the proportioning of
SCC to match fresh performance objectives while giving guidance to design engineers on
the compromises that the mix design will have on short term and long term hardened
properties as well as the structural response of structural elements.

In this research, an SCC mix was selected from each of the above mentioned
approaches. The first SCC mix (SCCl) with low w/c ratios (0.35) was designed after
Approach 1, the second SCC mix (SCC3) with high We ratios (0.45) followed Approach
2 and the third SCC mix (SCC2) with moderate w/c ratio (0.40) was obtained from the
combination of these two approaches. Also a normally consolidated concrete (NCC) mix
(w/c ratio = 0.40) was used as a reference. These mix designs and the controlling
parameters are described in the following section. The SCC2 and NCC mixes had to be
repeated due to poor performance of the mix and testing equipment. The first and second

attempts are designated by the letters “A” and “B,” respectively.

48

3.2 Project Mix Design Matrix

Based on the parameters governing the proportioning of a SCC mix and
implementing the idea of bounding the performance of all SCC mixes, the mix design

matrix for the test program is shown conceptually in Table 3-1.

Table 3-1 Mix Design Matrix - Binding of Performance by w/c ratio

 

 

 

 

 

Mix w/c HRWR VMA CAC S/Pt EA
Design
SCCl 0.35 + -— Less more +
SCC2 0.40 + — +
SCC3 0.45 + + More less +
NCC 0.40 — — 0.50 0.50 +

 

 

 

 

 

 

 

 

 

All mixes used Type III cement with a design compressive strength of 48.3 MPa
(7,000 psi). Local natural aggregates in agreement with the Michigan Department of
Transportation specifications for use in bridge elements were used, namely 6AA coarse
aggregates and 2NS sand. The level of entrained air for all mixes was 6%. The high range
water reducer, viscosity modifying admixture, air entraining admixture and set-retarding
admixture were provided by Degussa Admixtures Inc. The set retardants were used to
delay the initial setting time of concrete since it had to be delivered to the laboratory via a
ready mix delivery truck. In addition the casting process was long as it included multiple
test units and material testing samples. Thus it should be noted that since these
requirements of ready mix concrete delivery, these mix designs may vary slightly from
those used in situ in the precast/prestressed plants.

The mix designs with respect to the approaches, were obtained from consultation

with Degussa Admixtures Inc. The mix designs in Table 3-2 were used as target to be

49

achieved to cast the test specimens at MSU’S Civil Infrastructure Laboratory. However,
during the actual casting of the specimens, some of the mix designs had to be modified
on site due to variations in delivery time, moisture and temperature.The mix design
proportions used in this research for all the mixes are given in. Table 3-2 The in-situ
admixtures and w/c ratio which slightly varied from the target mix design (Table 3-2) for
the various mix designs used in this research are expressed in terms of percentage change
from the original (Table 3-3) for all the mixes. The negative values indicate that less
quantity of particular admixture has been added in the mix relative to the target mix.
There was no change in any other mix proportion component of the mix design. Table
34 gives the final mix designs used in the test beams.

Table 3-2 Mix Designs used in the project — Target mixes

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MIX TYPE
SSD weights (lbs / cyd)
& SCCl & SCC3
NCCB SCC2B
Cement - Type III 700 700 700 700
Fine Aggregates. 1216 1519 1426 1275
Coarse Aggregates. 1580 1380 1380 1435
Water 280 245 280 3 l 5
Air 6% 6% 6% 6%
w l c ratio - Target 0.40 0.35 0.40 0.45
ADMIXTURES fl.oz./cwt
Air Entraining Admixture l 0.5 0.5 0. 5
High Range Water Reducer 6 6 7 8
Viscosity Modifying Admixture 0 0 1 2
Set Retardant 6 6 6 6
llb/yd3=0.593 kg/m3
1 fl oz./cwt = 65 mUlOO kg

 

 

50

Table 3-3 Changes in Admixtures for Actual Project Mix Designs

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MIX TYPE
Chan es in Admixtures relative to target mix (%)
ADMIXTURES
NCCA NCCB SCC1 SCC2A scczg SCC3
Air Entraining 0.00 0.00 0.00 -5714 0.00 81.63
Admixture
High Range Water -3333 -66.67 33.65 21.28 0.00 0.00
Reducer
Viscosity Modifying 0.00 0.00 0.00 293.88 0.00 350.00
Admixture
Set Retardant -100.00 0.00 16.67 -100.00 11.56 0.00
wlcRatio 0.00 0.00 0.00 5.00 0.00 0.00
Table 3-4 Actual Project Mix Designs
NCCA | NCCB [ sccr [SCC2A l scczgj SCC3
COMPONENT SSD WTS (lstcyd)
cement ' Type I" 700 700 750 700 700 700
Fm" Aggregates‘ 1216 1216 1627.5 1426 1426 1275
coarse Aggregates‘ 1580 1580 1478.57 1380 1380 1435
water 280 280 262.5 280 280 315
Air 6% 6% 6% 6% 6% 6%
w I c ratio - Target 0.4 0.4 0.35 0.42 0.4 0.45
ADMIXTURES oncwt
Air Entraining
Admixture 3.50 3.50 1.75 0.75 1.75 3.18
High Range Water
Reducer 8.06 4.03 12.93 14.59 12.03 15.37
Viscosity Modifying
Admixture 0.00 0.00 0.00 6.99 1.78 15.37
se‘Re‘a'dam 0.00 52.50 70.00 0.00 58.57 46.67
1 lb/yd3=0.593 kg/mj
lfloz./cwt=65 mIJlOO kg
lin. =25.4mm

 

 

51

 

3.3 Fresh Property Evaluation:

As previously discussed in Chapter 2, The acceptability of the SCC mix resides
on the ability for the mix design to satisfy performance criteria that defines a concrete as
self-consolidating, namely to have a highly flowable and stable mix. A brief discussion

and results of the tests performed in this research is given as follows:

3.3.1 Slump Spread and Visual Stability Index (V5!)

The Slump spread was the first test performed. If the concrete performance was
satisfactory then the next tests were performed. A flat base plate was kept on a level
ground and was slightly moist with water. All tests were consistently performed with an
inverted cone, by the same person. The concrete was filled in the cone with a scoop; no
tamping was done. Any excess concrete was removed and the cone was lifted vertically
allowing the SCC to flow freely. The diameter of the spread was measured in
perpendicular directions. The average of the two measured diameters was calculated. The
stability of the mixture was rated by visual examination of the spread concrete. The
results of the slump spread values before cast for all the SCC mixes and their VSI ratings
are shown in Table 3-5.

Table 3-5 Slump Spread and VSI rating of SCC mixes.

 

 

 

 

 

 

 

 

MIX Agfiiaeiisiii'ip xiii;
SCC1 27.0 0.0
SCC2A 25.0 0.5
scczg 24.5 0.0
SCC3 27.0 1.0
l in. = 25.4 mm

 

 

 

52

 

c) Step 3 - Cone lifted vertically (I) Step 4 - Measurement of spread

Figure 3-1 — Slump Spread test

3.3.2 J Ring Test

The J Ring test is used to determine the passing ability of SCC. For this test, J
Ring is used in conjunction with the slump flow test procedure. The combination of these
two tests enables to study the flowing ability and the passing ability of SCC. The
procedure is essentially the same as for the slump spread test with the exception that a J
ring is placed concentrically around the slump cone. (Figure 3-2). The cone is lifted and
the SCC is allowed to pass through the J-Ring. The spread in perpendicular directions

was measured as before. Visual examination of any segregation or bleeding was also

53

carried out. The values of concrete spread with J — Ring were compared with the values
of slump spread without the J-Ring. The smaller these differences showed the better
passing ability of concrete. Table 3-6 shows the results obtained from the slump spread in
conjunction with the J -Ring. The J-Ring measurements were not taken for SCCZB mix as
the mix was stiffening rapidly. Table 3-6 also shows the J-Ring value calculated as

explained in Chapter2.

 

     

b) J — Ring test in Progress
Figure 3-2 J — Ring Test

54

Table 3-6 and J — Ring Slump Spread and J-Ring Values

 

 

 

 

 

 

 

 

Average Slump J- Ring Value
MIX Spread (in.) (mm)
SCC1 25.0 6.35
SCC2A 20.5 9.53
SCC2B - -
SCC3 23.5 0
l in. = 25.4 mm

 

 

 

3.3.3 L — Box Test

L-Box test assesses the flow of SCC and also the extent to which it is subjected to
blocking by reinforcement [30]. The test apparatus comprises of a rectangular cavity in
with reinforcement. The level of reinforcement can be changed to enforce severe or light
restraints depending on the actual reinforcement blocking of the structure. The L—Box
used in this test had the same level of reinforcement as given in PC] guidelines [30]

(Figure 2-4). This opening is controlled by a gate.

 

Figure 3-3 L— Box Test

55

The L-Box results obtained for the various SCC mixes are reported in Table 3-7

Table 3-7 L— Box Blocking Ratio

 

 

 

 

 

Blocking

Ratio
MIX (HZ/H1)
SCC1 0.80
SCC2A 0.86
SCC2B 0.77
SCC3 0.69

 

 

 

 

3.4 Challenges in SCC Quality Control and Quality Assurance

Ready Mix Concrete was used for the SCC in this project. The travel time for the
concrete mixer to reach to MSU’s Civil Infrastructure Laboratory was approximately 30
to 45 minutes. In the first SCC mix (SCC2A), all the admixtures were added in the plant
and by the time SCC reached the lab, its behavior was not of typical of SCC (low
fluidity). This mix did not have any set-retarding agents (stabilizers). Substantial addition
of HRWR’S was needed to achieve SCC behavior. It was thus determined that it was
difficult to achieve proper SCC behavior with long delivery times and use of Type III
cement. In order to delay the initial setting time, set retarding agents (stabilizers) were
used in all the future mixes. Also, it was found that the order of admixtures addition plays
a vital role in the setting time and behavior of the SCC mix. In order to avoid the issues
of rapid setting and to achieve the target SCC mixes, all of the admixtures were not added
at the Ready Mix Plant. Aggregates, cement, water, air entraining agent and set retarding
agent were added in the plant. The high range water reducers (HRWR) and viscosity
modifying admixtures (VMA) were added at the laboratory, and in that order. These
admixtures were added individually and mixed in the truck drum for approximately 5-10

minutes before the next admixture was added. The fresh behavior of SCC was studied

56

after all the admixtures were added. In some cases it was found that proper fluidity of the
SCC mix was not achieved or not satisfactory. Hence additional quantities of admixtures
were added to achieve the desired SCC behavior. As a result, the actual SCC mixes
deviated from the target mix design (see Table 3-3). This deviation was mainly in the
content of admixtures.

The desired target properties for the NCC mix were achieved without any problem.
This reflects the significant experience with the conventional concrete (NCC) mix
designs. Conversely, even though SCC was first introduced in 1980, it is still relatively
new to the industry. In spite of much research work done in fresh concrete properties and
mix designs of SCC, achieving the same target mix is not easy in the field, as climate
conditions and aggregate properties may vary significantly. More research on various
practical difficulties needs to be performed in collaboration with industry and admixtures

experts to make SCC achievable, controllable and available to industry as easily as NCC.

3.5 Hardened Concrete Property Evaluation

Since the target engineering properties of hardened SCC should be the same as
those for conventional concrete, the same test and procedures that are used for
conventional concrete were used for SCC. Tests were performed on concrete cylinders of
following dimensions: 101.6 mm x 203.2 mm (4”x8”). The compressive strength tests,
modulus of elasticity tests and split tensile strengths were performed in accordance with
ASTM C39, ASTM C469 and ASTM C496 standards, respectively. Compressive and
Split tensile strength tests were performed at 1, 3 (day of release of prestress), 7, 14, and

28 days. Tests were also performed at the day of flexural testing, where in the average

57

age of concrete was approximately 110-130 days Elastic modulus tests were performed

only at the day of release (3 days) and at 28 days of age.

3.5.1 Compressive Strength (f’,)

The target compressive strength for all mix designs used in this project was
48.26 MPa (7000 psi) at 28 days. The compressive strength tests were performed in
accordance with the ASTM C39 standards. Three concrete cylinders were tested for each
test.

Table 3-8 shows the average compressive strength values and their standard
deviation for concrete ages varying from 1 to 28 days for all the mixes. Table 3-9 shows
the compressive strength and the standard deviation for all mixes the day of flexural
testing (~1 10-130 days). Figure 3-4 the variation of compressive strength with the age of
concrete. Due to improper functioning of the equipment, 14 day test results are not

available for SCC1.

58

Table 3-8 Compressive Strength ( f’c ) Test Results — 1 to 28 days

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MIX NCCB SCC1 SCC2A
Age Of Average Standard Average Standard Average Standard
Concrete f’c Deviation f’c Deviation f’c Deviation
(Days) (psi) (psi) (Psi) (psi) (psi) (psi)
1 4899.2 191.1 7003.6 194.37 N/A N/A
3 5545.1 93.41 7685.0 129.42 7693.25 607.29
7 6478.2 146.9 8500.7 350.76 7778.93 1307.9
14 6535.8 232.9 - - 8161.18 323.00
28 6864.6 1131.6 8686.9 374.77 8646.88 392.77
MIX SCC2B SCC3
Age Of Average Standard Average Standard
Concrete f’c Deviation f’c Deviation
(Days) (psi) (psi) (psi) (psi)
1 6132.4 124.9 6132.4 124.9
3 6703.8 344.2 6703.8 344.2 1 MPa = 145.04 psi
7 7065.1 603.2 7665.4 157.3
14 7671.9 29.8 7671.9 29.8
28 8038.4 1388.8 7810.0 924.0

 

 

 

 

 

 

 

Table 3-9 Compressive Strength at Day of Test

 

 

 

 

 

 

 

Age of Concrete Compressive Standard
at Day of Test Strength Deviation
MIX . .
(daLS) (pSI) (1)81)

NCCB 133 7018.28 302.69
SCC1 124 8973.29 322.27
SCC2A 118 9208.98 1121.69
SCCZB 109 9378.28 1131.81
SCC3 120 7971.25 199.62

 

 

 

 

 

59

Age Of Concrete (days)
0 20 4o 60 80 100 120 14%

r T Viv V

 

V T Y ' V V I U I V ' ' V V V V U I

  
   

d
d
d
- 9500
1
‘
1
1
a - 9000
d
1

4. «28000
@7500
a 17000
gesoo

81833291888

 

b
M
'n

—o— NCCB

Compressive Strength (MPa)
Compressive Strength (psi)

 

 

 

 

 

39 + SCC1 E 5000
36 + SCC2A _ 550°
33 + 80028 3
30 - —-I— sccs 5000
I . . r . . I . I I I I x I . . . I r I I x I l I I I . I . . L. ‘ 45w
0 20 40 60 80 100 120 140

Age Of Concrete (days)

Figure 3-4 Compressive Strength variation with time - All Mixes

3.5.2 Elastic Modulus Test
The average Elastic modulus (E,) was calculated as per the ASTM C469

standards and was compared with the equation recommended by the ACI code

E mode = 57000412. (psi). Table 3-10 and Table 3-11 Show the values of the modulus of

C

elasticity obtained both experimentally and from the ACI code expression and their
relative comparison for 3 days and 28 days respectively. Due to improper functioning of
testing equipment, reliable values of E, could not be obtained for SCC1 and SCC2A
mixes at 3 days and for SCC2B mix at 28 days. A sample plot of the compressive stress—
strain response (28 days) for each of the mix designs showing the calculation of the
modulus of elasticity are given in Figure 3-5 to Figure 3-8. Detailed individual plots are

shown in Appendix 1.

60

Table 3-10 Elastic Modulus Tests at 3days — all mixes.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MIX Ec,meas Encode Eameas / Encode
(ksi) (ksi)
Avera e Standard Avera e Standard
g Deviation g Deviation
NCCB 4,053.20 24.81 4,244.46 35.75 0'95
SCC1 - - 4,996.86 648.46 ‘
SCC2A - - 4,999.54 1,404.67 ‘
SCC2“ 4,333.26 313.59 4,912.91 341.70 0'88
SCC3 4,090.24 494.76 4,611.07 104.08 0‘89
Table 3-11 Elastic Modulus Tests at 28 days
MIX Ec,meas Encode Earneas / Encode
(ksi) “(50
Average Standard Average Standard
Deviation Deviation
NCCB 4,383.35 0.00 4,693.45 391.68 0'93
SCC1 4,624.80 530.86 5,311.99 114.61 0'87
SCC2A 5,312.35 221.35 5,299.69 102.23 1'00
SCC2” - - 4,992.59 311.28 '
SCC3 4,463.93 130.92 5,032.90 298.24 0'89

 

 

 

 

 

 

61

 

 

Compressive Strength ( MPa )

Compressive Strength ( MPa )

54

42

36

30

24

18

12

Compressive Strain ( micro strains )

250 500 750 1000 1250

O

1500

1750 2000 2250 2500

8000

 

 

0 250 500 750 1000 1250

7 V V V V V Y Y Y Y 77 1’ Y Y Y 7 V V V V V
' r I ' r ' i l

1500

V T

   

T 'VYlTrV'V YT

Ec = 4,383,350 psi

1 l

s...§¥%€5¥§149F1.

: 7000

: 5000
1 4000
: 3000

:2000

 

0

1750 2000 2250 2500

Compressive Strain ( micro strains )

Figure 3-5 Typical Stress—Strain Response — NCCB - 28 days

66

54

42
36
30
24
18
12

6

0

Compressive Strain ( micro strains )

0

 

Y—Y—Y—YIY'TYYI V7 VY'IVTVYYYYYYIYYT'TYVYYT YT

' Y

Ec=4,,249249psi
=: 129”?0L’I‘FMI.

A AAAAAA llAALlJllAllAlAll‘LLJLl‘lelALILLLLIAALA

AILJLJIAAA

1
4

 

250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000 32500000

YYYIYVTTTYYY YTV—TWYIV

9000

-j 2000
Ec = Elastic Modulus:
‘ 1000

0

0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000 3250

Compressive Strain ( micro strains )

Figure 3-6 Typical Stress—Strain Response - SCC1 — 28 days

62

Compressive Strength ( psi )

Compressive Strength ( psi )

Compressive Strength ( MPa )

Compressive Strength ( MPa)

Compressive Strain ( micro strains )

 

 

 

 

 

0 250 500 750 1000125015001750 2000 2250 2500 2750 3000 3250 3501130000
66} 1

E O 49000
60: 1

; 1
54E :8000
48:— €7000
42g £6000
36: 4:5000
30E 3

: 14000
24:— g

E 13000
18: I
12: 7 £2000

_ Ec = Elastic Modulus?

6: Ec=5,468,869 psi 31°00
0’- Ir..IIiIIIILIIIIiIIIIiIII.1.IIIiIIII11I=IIi.I3I7I!7IOI7I|.v|Ea,.,1o

  

 

Compressive Strain ( micro strains )

250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000

Figure 3-7 Typical Stress—Strain Response - SCC2A — 28 days

54

48

42

36

30

24

18

12

Compressive Strain ( micro strains)

0 250 500 750 1000 1250 1500 1750 2000

2250 2500

 

YfYIXYTY YT VIWfiV 77" T

'77 Y

or

.
E0 = Elastic Modulus:
E0 = 4,371,360 psi 3
= 30,140 Mpa

ALLIAAALILLLLILIAAIA‘lA‘

  
 

 

4
L

 

I l L

 

   

LLLJILIA

  

Compressive Strain ( micro strains)

0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750

8000

7000

6000

5000

4000

3000

‘ 2000

1000

0

Figure 3-8 Typical Stress—Strain Response — SCC3 — 28 days

63

Compressive Strength ( psi )

Compressive Strength ( psi )

3.5.3 Split Tensile Test.

As previously mentioned, split tensile strength (fl) tests were performed in
accordance with the ASTM C469 standards for all mixes. Table 3-12 shows the average
split tensile strength values and their standard deviation for concrete ages varying from 1
to 28 days. Table 3-13 shows the split tensile strength and the standard deviation for all
mixes at their respective days of flexural test. Figure 3-4 shows the variation of split
tensile strength with the age of concrete. Due to large variation and improper functioning
of testing equipment, some data was considered unreliable and not included in the results

given here. Thus, Table 3-12 and Table 3—13 below have some blank spaces.

Table 3-12 Split Tensile Strength — l to 28 Days

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MIX NCCB SCC1 SCC2A
Average Average Average
Age Of Split Standard Split Standard Split Standard
Concrete Tensile Deviation Tensile Deviation Tensile Deviation
(Days) Strength (psi) Strength (psi) Strength (psi)
(psi) (psi) (psi)
1 434.43 65 514.97 73 - -
3 — - 519.94 0 574.95 68
7 424.28 120 571.96 58 574.95 68
14 540.63 33 540.63 33 583.57 8
28 - - 581.11 33 596.33 88
MIX SCCZB SCC3
Average Average
Age Of Split Standard Split Standard
Concrete Tensile Deviation Tensile Deviation
(Days) Strength (psi) Strength (psi)
(psi) (psi)
528.46 64 509.01 73 1 MPa __. 145.043psi
- - 532.20 121
534.26 11 569.09 82
14 572.89 111 569.33 36
28 589.47 40 - -

 

 

 

 

 

 

Table 3-13 Split Tensile Strength — Day of Test

 

 

 

 

 

 

 

 

 

Age at Average
Day of Split Tensile Standard
MIX (01:3; 813181511. D6623“
NCCB 133 544.54 113
SCC1 124 615.73 117
SCC2A 118 623.76 218
SCCZB 109 629.47 219
SCC3 120 580.33 92

 

 

 

65

 

Age Of Concrete (days)

 

 

 

 

 

 

 

 

 

 

0 20 40 60 80 100 120 140
4.75~ 2
fi 1- ‘ c
“.450 4550“
E 1 9
£4.25 . .c
‘5, : 450061
. ‘ C
54.05 0
h
b : a;
$3.75: 15502
E350 ‘ E
. L 0
'1: ‘500,_
53.25 +50“ 3 E
‘g- +scczA;4sog-
-+-SCCS
1.1111111IAAJAIAAAAIAJAlljnlnlnLAI 4m
0 20 40 60 80 100 120 140

Age Of Concrete (days)
Figure 3-9 Split Tensile Strength variation with Time

3.5.4 Discussion of Results - Hardened Test properties
The target compressive strength at 28 days for all mixes was 48.3 MPa (7000 psi).

It was found that all the mixes had compressive strengths much beyond this target value
at 28 days. Figure 3-10 shows the variation of the compressive strengths at 28 days. The
results show that the compressive strength at 28 days was approximately 2% lower than
the designed target for NCCB and was greater than the design target by 11% — 25% for
the SCC mixes. Considerable research in all aspects and extensive use of NCC has made
it possible to control the hardened properties of NCC with more confidence. The same
amount of control and confidence has not been achieved with SCC. From the elastic
modulus (Table 3-11) at 28 days of concrete age it was observed that the measured value

of NCCB was 7% lower than that predicted by the ACI code, whereas SCC1 and SCC3

66

had lower measured modulus by 13% and 11%, respectively, relative to the code
predicted values. Figure 3-11 shows the relative comparison of the elastic modulus for
the various mixes Due to technical problems with the testing equipment, SCC2B modulus
tests at 28 days could not be performed. SCC2A showed no variation with the code
predicted values. It should be noted that SCC2A was a very stiff mix and the performance

of the mix was poor and hence the data reliability is questionable.

 

    

10000

65'MixDesign

60- Target
4355- «3000 a
IL a
:50- ___ 9;
V
545- i,
a ........... -m C
:40. 2
2
17535 35
330- 3
a «4000 3
§25- g
n o.
E2°' E
o
015' -2000 o

10-

5

 

 

NCCB SCC1 SCC2A SCC2B SCC3
MIX TYPE

Figure 3—10 Comparison of Compressive Strength at 28 days

Overall, the SCC mixes showed a larger variation in elastic modulus compared to
the code predicted values than the NCC mix. The results reiterate the fact that much
control in hardened properties has not yet been gained in SCC mix proportioning relative
to NCC mix design. It should also be noted that the variations in SCC hardened

properties may also be due to the various mix proportions used. Infact, SCC1 and SCC3

67

504-5

:2: 89.260 .o 2:23: ozosm

MMMMMO

 

 

  

_ q . _ .

     
    

   

     

    

                         
        

 

0.43.... .. 459,232.84? ...Z._..../....56
yfflgéxofifififle

are... 4.449.553.“ (rum/.5. 10... has}: 1.4/fin. . 4.4.30... 2.3%... in?
"swaggwo figsmfigawgufiafifi

    

      

SCCZA SCC23 SCC3

MIX TYPE

      

                

SCC1

gggafl/xm/mfllfifi/M

 

 
 

d

m 8

mm m
a a 2%2/52848 m
n .. r _.

_ _ ‘

 

are mixes obtained from two different approaches and the variation is expected. SCC2B

is the only mix relatively similar to NCCB mix.

83......

€55 22260 .8 3:505. oznsm

40000

Figure 3-11 Measured vs. ACI Predicted Elastic Modulus at 28 days
68

CHAPTER 4 TEST PROGRAM - INTRODUCTION

The effect of bond on transfer and development length of precast/prestressed
girders built using SCC was evaluated through a structural testing program described
herein. The experimental study is based on evaluating transfer and development lengths
for 13 mm (0.5 in.) diameter strand in laboratory scale precast/prestressed beams. Two
beams for each of the program mix designs (NCCA, NCCB, SCC1, SCC2A, SCC2B and
SCC3 Table 3-4) were constructed and used in the study. This section discusses the test

unit (beams) design, their naming convention and their fabrication.
4.1 Specimen Design & Nomenclature Used.

4.1.1 Specimen Design

The test units consisted of precast/prestressed T-beams with two 13mm (0.5 in.)
diameter prestressing stands and nominal compression and shear reinforcement (Figure
4-1). The cross-section and reinforcement amounts were chosen so that the strand strain
at the section nominal capacity was close to the guaranteed ultimate strain of the strand,
taken to be 0.0035. The strands with a nominal diameter of 13mm (0.5 in.), Seven wire
low relaxation 1860 MP8 (270 ksi) Grade strands with a nominal diameter of 13 mm (0.5
in.) were used. The beam length was 11.58 m (38 ft) with the goal of being able to
perform two flexural tests per beam (one for each end). The prestressing strands were
completely bonded. Nominal shear reinforcement was provided with 6mm (0.25 in.)
diameter smooth stirrups. The U-shaped stirrups were placed at 305 mm (12 in.) spacing

through out the length of the beam except at the ends where they were spaced at 152 mm

69

(6 in.) for a length of 610 mm (24in.) An elevation drawing is showing the reinforcement

details is given in Figure 4-2.

 

 

'4 15" J
l l

A J) _1_.J.5_"_]l ______ :::3‘ a2 - #4 bars

I #2 stirrups
/ @ 12" clc

 

 

 

/

15"

 

12"

2 - 0.5"
diameter

1 / strands

 

 

 

 

 

 

 

 

Figure 4-1 Test Specimen - Cross Section Details
1 in. = 25.4 mm

The strain level in the prestressing strand has been identified to be an important
parameter in the discrepancy of experimentally obtained values for development length
[8]. Experimental values from most test programs suggest that the average bond strength
is lower in test units with large strand strains at failure (e.g., near the guaranteed ultimate
elongation) as compared to specimens that failed with strains near the yield strain (0.010)
[8].Therefore, the beams for this study were designed such that the strain demand on the

strands was closer to the guaranteed ultimate strength.

70

.82 m 1 mean?

so ..m m_ g 6..ch
93> ce>om 380 Ohm mcfimgmoi 56861.6 as.

3.895 E 5098 £95. Emma 05 So

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5:25 so .9 6 325m «5 - m 69:. < 09»...
- _ / 268 2 u N \ ._ \ 1 .
swam :2 850 / \m F
mficobfloEcSm I 1 N.. 1+1 N_. 11.10 mlrlmllo -
/ \ h
/ \ N
4 It / 0 ii JHPI
a. 4 I 0
IV IV
AV 0 l
8.44 _ 11.1.1-1, 1.1111... .. 1...... 1.... “mun...
_ 2L1

 

 

 

 

 

 

 

 

 

 

9033 65 E0: EN .2

#:0802256. 02860 >_co ob am © 825m we .. m $9»...

1.818.. 0 69?

Figure 4-2 Reinforcement Details of the test specimens

71

4.1.2 Nomenclature

Two test specimens were cast at a time for each mix. For transfer length
measurements, strains on both sides of the beam web were measured. The specimen
identification nomenclature used through out this project for transfer length studies is

summarized in Figure 4-3.

 

Type of Concrete: End at which measurement
SCC — Self Consolidating Concrete taken:
NCC - Normally Consolidated Concrete A: West end

B: East end

Side of the web where transfer length is measured with respect
to the cut end. 1: North side, 2: South side

 

 

 

1,
:SCC1X- B #- 8 DE — END

Mix Number:

SCC has three mixes Beam Number:

viz. 1, 2 81 3 1 or 2. Each mix has two beams
Repeat Number “A or B”

NCC 81 8002 have two mixes: NCCA,
NCCB, SCCZA, SCC2B

Example: SCC28 - 1 - A : which means, transfer length reported is measured at the
east end on side 1 (facing north) from the 2"d beam of the second SCC mix.

 

 

 

Figure 4-3 Nomenclature for Transfer Length

The nomenclature used for the development length tests was similar to the
transfer length nomenclature. The flexural tests to determine the development length
were performed on both the ends of the beams. The only difference in the nomenclature

for development length is the removal of the “side” term from the nomenclature for

72

transfer length. Thus, “SCCZB—l—A” refers to the development length test of the first

beam of SCCZB mix and the test being performed at beam end A..
4.2 Material Properties

4.2.1 Concrete

Fresh property tests on concrete were performed for every mix before acceptance
for use in the beam units. Results on the fresh concrete properties of the SCC mixes are
described in Section 3.3. The hardened concrete material properties were determined at
various ages of concrete. The target f’c at 28 days for all mix designs was 48.3 MPa (7000
psi) and was achieved or exceeded by all the mixes. Results of the hardened concrete

properties are given in Section 3.5

4.2.2 Prestressing Steel

The pretensioning reinforcement used in the test specimens was 13 mm (0.5 in.)
diameter Grade 1860 MP3 (270 ksi) low—relaxation seven wire strand. The nominal cross
sectional area was 97.870 mm2 (0.152 inz). The modulus of elasticity and guaranteed
minimum elongation provided by the manufacturer was 196 GPa (28400 ksi) and 0.035
in./in. (3.5%) respectively.

Same strands were to be used for all test beam specimens for all the mix designs.
But, due to the bad performance of the first SCC2 mix (SCC2A), the mix design was
repeated. Hence, a new set of strands from the same pool, but two months later were
obtained. This new set of strand pieces had slight rust on its surface. The rust condition
was minor and could be removed if wiped off with a cloth. Nonetheless, in order to avoid

disturbing the relative performance of SCC mixes, clean non- rusted strands (Figure 4-4a)

73

were used for all SCC mix designs and the slightly pitted strand (Figure 4-4b) was used

for NCCB beam specimens.

   

(a) Clean Strand (b) Slightly pitted Strand

Figure 4-4 Strand Condition

4.3 Specimen Fabrication

The beam units were fabricated at Michigan State University’s Civil Infrastructure
Laboratory. The fabrication process can be grouped into four steps: a) Assembly of
formwork, b) prestressing operation, c) placement and curing of concrete, and (1) release
of prestress. A brief description of the fabrication process is explained as follows:

The formwork assembly with the reinforcement and prestressin g tendons placed is
shown in Figure 4-5. The strands were then pretensioned by anchoring them to reaction
concrete blocks post tensioned to the laboratory strong floor. The strands were
pretensioned individually using a hydraulic jack to a level of approximately 75% of the
ultimate after anchor-set losses. Electrical resistance strain gages were attached to the

strands to monitor the prestress operation and to measure the forces before and after

74

release of the strands. Figure 4-6 shows the schematic layout of the casting bed and

Figure 4-7 shows the prestressing operation carried out for one of the strands.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TOP N
— Side# 2 '—
Beam # 2
Side# 2 Side#1
Beam # 1
Side#1
AnIhor H—_L = 11,513 m (38 ftp—>1 An hor
Bl ck Bl k
ELEVATION

 

 

I iikfiix‘l‘t‘u‘liflél‘t‘fim-

suroaszsszozozozz '2:o:ore:o:oms:s:o:ozozozozozozszoszozozozo333353335.as:

 

Figure 4-6 Schematic Layout of the casting bed
The concrete was mixed at a ready mix plant and brought to MSU’s Civil

Infrastructure Laboratory. As discussed in Section 3.4 , the admixtures were added on site

75

in order to achieve proper SCC behavior. After appropriate fresh property tests and
approval from the research team, the concrete was poured into the test specimens. Pull
out blocks cast and material testing cylinders were cast in the same operation. After
casting the beams were left to cure at room temperature and humidity conditions. The
formwork was removed the next day to place instrumentation for transfer length

measurements.

 

Figure 4-7 Pretensioning of strands

After instrumentation of the test specimens had been completed for the transfer
length measurements and the initial measurements taken, the release of prestress was
carried out. Prestress release was done by flame cutting on both the ends simultaneously.
However, the strands were first gradually heated with a broad flame, until most of the
prestressing force was completely transferred by thermal elongation of the strands. The

strands were heated over a distance of 305 mm (12 in.) by slowly moving the flame

76

above and below the strand in gradual strokes. This process was done simultaneously on
both ends and was coordinated by a team member. The annealing process was performed
for approximately 5 minutes to release as much of the prestress as possible. The release
of prestress was also monitored by resistance strain gages attached to one of the wires of
the strands. These strain gages were installed during the loading of tendons to measure
the amount of stress in the strands. During the heating process, the strain in the strands
was verified to drop and reach near zero values. The strands were then cut simultaneously
on both the ends (Figure 4-8). In some cases, one of the wires of the strands would
fracture during the heating process. In such cases the heating process was terminated and
the strand at both beam ends was cut simultaneously immediately after such event. The

process just described was repeated for each strand.

 

Figure 4-8 Release of Prestress — Both ends of the beam being cut simultaneously

77

CHAPTER 5 STRAND BOND PERFORMANCE EVALUATION

5.1 Introduction

This Chapter deals with the evaluation of strand bond performance with the
different concrete mixes in the test program. This evaluation was done by means of
simple pull-out tests on unstressed strands. The test description, procedure and results are
presented in detail. These tests were performed in series with the transfer length study.
Pull out tests were performed at the day of transfer (mostly 3 days) and at 7 days. Results

for the SCC mixes are compared to those obtained for the NCC mix.

5.2 Background

The need to have a standardized test to measure the bond performance of
prestressing strands lead to the development of various tests such as simple Pull-out tests,
and tensioned Pull-out tests. Tests on strand bond confirmation by means of the
“Moustafa test” have been recommended by the PCI Interim Guidelines for SCC [30].
Previous research has shown that the bond quality of strands from different
manufacturers varied significantly. Hence, a modified version of the Moustafa test, the
large block pull out test (LBPI‘) has been recommended by Logan [25] to qualify strand
for prestressing use.

As discussed in Section 2.4.1, the test recommended by the PCI Interim Guidelines
for SCC [30], has to be performed in a specific manner to qualify the strand for use in
pretensioning purposes. In this research, the pullout test is being performed to study the

relative bond performances on different SCC mixes.

78

5.3 Specimen Preparation

The test procedure used for the pull-out tests was similar to that proposed by
Moustafa [41] and Logan [24], 1974. The pull-out block details are shown in Figure 5-1.
Each pull-out block contained six non-tensioned prestressing strands, except for the
NCCA mix, for which the pull-out block was made with 12 strands. A block was cast for
each concrete mix. The prestressing strands used were of the same type as that used in the
test beam specimens. The total depth of the block was 610 mm (24 in.). Plastic sleeves of
50 mm (2in.) and 101 mm (4 in.) were provided in the top and bottom of the strands
respectively. The prestressing strands had an embedment length of 457mm (18 in.) The
strands had a side cover of 115 mm (4.5 in.) and a center to center spacing of 229 mm (9
in.). The total strand length was 1.83 m (6 ft), with approximately 300mm (1 ft)
extending below the block and 915 mm (3 ft) extending above the block. The longer end
was used as the jacking end to attach the pull-out equipment and instrumentation was
done on both the ends of the strands while performing the pull-out test. Nominal
reinforcement was provided to the block to prevent any temperature or shrinkage effects
on the block. The casting procedure and the concrete used were the same as that used in
the test beam specimens. Forrnwork for the block was removed at the same time as the
formwork for the test beams. Figure 5-2 shows the casting of the Pull-out block and
Figure 5-3 shows the finished pull-out block with a test in progress. No vibration was

used for the SCC mixes, while the NCC mix was conventionally vibrated.

79

 

 

 

 

 

 

 

 

 

. . . ‘\ TOP
°2 ‘ #5 Ties
O O K
II \
4 5 9" 9 4 5 0.5" Strand

 

 

 

 

 

 

 

 

 

 

 

 

 

 

|| [11 II T F (FF—4H
5:513

4" Plastic

Sleeve /.
11 11 11 V 311 11.

H I II I 11 1

FRONT ‘ .f‘ .. N J
ELEVATION 4'5 8,905 4'5
ELEVATION

 

1" = 25.4 mm

2" Plastic
Sleeve

 

 

 

 

 

 

 

 

Figure 5-1 Pull-out Block Geometry and Reinforcement

80

 

 

 

out Test Block

Figure 5-2 Casting of Pull

.v.

.4}...

GEL":

if.”

 

Figure 5-3 Pull-out Test Setup Overview

81

5.4 Test Procedure

The Pull-out tests were performed at the Civil Infrastructure Laboratory in
Michigan State University. The pull-out test were performed after the release of prestress
in the test beam specimens. This was done so that the bond strength at the time of transfer
in the actual beam specimens could be correlated with the pull-out test data obtained. The
test procedure slightly differed from that proposed by Logan [25] (See Section 2.4.1.)
The schematic pull-out test setup is shown in Figure 5—4. The actual Pull-out test setup is

also shown in Figure 5—5

Hollow Core Cylinder i = 8%
Piston (Stroke 6 in.) p = 5

  
    
  

Steel Plate at = 4
Load Cell 0 = 2%

 

Prestressing Chuck ¢ = 15/8

 

Strand ,0 = V2

 

 

13_5 0.5—>1 1+1— ! k—H

 

ALL DIMENSIONS IN INCHES

Figure 5-4 Schematic of Pull Out Test Setup

After removal of forms the pull-out block was turned on its side so that both the
free and jacking (end from which the strand is pulled) ends of the strand could. be easily

accessed and instrumented. A hollow core hydraulic cylinder with a capacity of 100 ton

82

(220 kips) and a ram stroke of 150 mm (6 in.) was used to pull-out the strands. The pull—
out test setup was assembled as described next: A hollow core cylinder’s piston was
brought to zero position (completely intruded), the strand was inserted and the cylinder
was attached to the face of the block. A load cell assembly consisting of a center-hole
load cell [capacity = 334 kN (75 kips.)] sandwiched between two center-hole steel plates
was then placed next to the cylinder ram. A prestressing chuck was then placed over the
load cell assembly to anchor the strand against the cylinder piston. During the loading
process the piston of the cylinder would extrude thereby pushing the load cell assembly
against the chuck. The chuck, in turn would pull the strand out and the corresponding
load was measured by the load cell. Two linear potentiometers placed in line with the

stand were used to measure the strand movements at both the free and jacking ends

(Figure 5-5).

   

Front Displacement Back Displacement

Figure 5-5 Measurement of Displacements — Pull-out Test

83

The Pull-out rate was calculated from the peak load and the total time taken to
complete the test. The loading rate of 90 kN (20 kip) per minute as proposed by
Logan [25] could not be achieved with the hydraulic jack used. An average pull-out rate
varying from 13 kN (3 kip) to 27 kN (6 kip) per minute was obtained with this jack. As
expected, past researchers have noted that slower jacking rates will result in loer pull-out
force measurements [31]The time taken to complete the test varied from 3 to 6 minutes.
The test was stopped after the peak load was recorded and when there was no significant
increase in load corresponding to the increase in the displacements, in other words when
excessive slip was observed.

A total of 12 trial tests were performed on strands embedded in the NCCA mix at
different concrete ages. As expected, there was an increase in the measured peak pull-out
force measured with the increase in age of concrete. Also, there was very little variation
observed in the peak pull—out forces of multiple strands tested at the same age of
concrete. From the insights gained from these trial tests, the following pull-out-tests were
performed on only three strands at concrete ages corresponding to the release of prestress
in the beam test specimens. The remaining three strands were tested at 7 days. Also, the
results obtained from the individual set of three tests showed very little variation. A detail

discussion about the results is presented in the next section.

5.5 Results and Discussion

Peak pull-out forces were recorded for all mix designs from the tests performed at 3
and 7 days. Table 5-1 shows the peak pull-out forces and the standard deviation for all
mixes at 3 days. Table 5-2 shows the peak pull-out forces and its standard deviation for

all the mixes at 7 days. In the pullout tests performed by Rose and Russell [13] the onset

84

of general bond slip is defined as the load at a free end slip of 0.005 in. In this project the
loads at first slip were determined by examining the measured test force—displacement
response data. It was observed that the first slip was noticeable by a pronounced drop in
load and increase rate of deformation at both the free and jacking ends. Table 5-3 shows
the pull-out forces and the coefficients of variation for all the mixes corresponding to the
first slip at 3 days. Table 54 shows the pull-out forces and the coefficients of variation
for all the mixes corresponding to the first slip at 7 days. The values of the pull—out forces
that deviated significantly from the average were neglected and such values are reported

as “n/a” in the following tables.

Table 5-1 Maximum (Peak) Pull-out Force — Release (3 days)

 

 

 

 

 

 

 

 

Maximum Pull-out Force (kips) 1 kip = 4.448
Mix kN

Strand # 1 2 3 Average Deviation

NCCB 28.66 32.81 29.73 30.40 2.15

SCC1 17.61 14.91 17.30 16.61 1.48

SCC2A 27.06 n/a 24.95 26.01 1.49

SCC2B 18.28 20.92 18.98 19.39 1.37

SCC3 16.30 20.25 23.80 20.12 3.75

 

 

 

 

 

 

Table 5-2 Maximum (Peak) Pull-out Force - 7 days.

 

 

 

 

 

 

 

 

Maximum Pull-out Force (kips.) 1 kip = 4.448
Mix kN

Strand # 4 5 6 Average SD
NCCB n/a 32.23 33.05 32.64 0.58
SCC1 18.03 16.75 n/a 17.39 0.91
SCC2A 37.69 29.45 26.94 31.36 5.62
SCC2B 27.28 25.03 21.60 24.64 2.86
SCC3 28.74 31.85 28.78 29.79 1.7 8

 

 

 

 

 

 

 

 

Table 5-3 Pull Out Forces at First Slip - Release (3 days)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Pull-out Force (kips.) @ First Slip 1 kip = 4.448
Mix kN
Strand # 1 2 3 Average SD
NCCB 18.83 20.68 19.49 19.67 0.94
SCC1 10.30 7.62 8.74 8.89 1.35
SCC2A 6.46 n/a 6.96 6.71 0.35
SCC2B 7.50 6.94 6.94 7.13 0.32
SCC3 5.35 6.87 8.01 6.74 1.34
Table 54 Pull Out forces at first Slip - 7 days.
Pull-out Force (kips.) @ First Slip 1 kip = 4.448
Mix kN
Strand # 1 2 3 Average SD
NCCB 14.11 15.68 15.01 14.93 0.47
SCC1 9.86 8.85 n/a 9.36 0.71
SCC2A 8.52 7.42 6.04 7.33 1.24
SCC2B 9.05 6.19 6.13 7.12 1.67
SCC3 6.72 6.55 6.57 6.61 0.09

 

 

 

 

 

 

 

 

All the pull-out tests displayed a gradual load-slip behavior and no fracture of
strand was achieved. It took approximately 4 to7 minutes to complete each test. A typical
load-slip response for all the mixes at release and 7 days is shown in Figure 5-6 and
Figure 5-7, respectively. A “close-up” of the response at first movement for all the mixes
at release and 7 days are shown in Figure 5-8 and Figure 5-9 respectively. The Individual
Pull-out Plots for all the strands are given in Appendix 2. The average peak pull-out
forces obtained at 7 days were compared with those obtained at release (3 days), and as
expected, it was found that the 7 day pull-out forces were slightly higher The smallest
increase of approximately around 6% was found for the SCC1 and NCCB mixes, a
moderate increase of around 24% was found for the SCC2 mixes and a large increase of
around 48% was observed for SCC3 mixes. Figure 5-10 shows the comparison of the

peak pull-out forces at release (3days) and 7 days for all mixes.

86

Strand Slip {front} (in.)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.035

WWW—“WWW

4

150g
140:—
130:—

 
 
 
 
   

   

 

 

A 120 :—
g 110;
§ 100;-
a 90%
.. 80 E
3
(_3 70 :—
'5
o. 60
x 50
(B O NCCB
5 40 I scc1
30 A SCC2A
20 ,_ e sccze
10; + SCC3
of:
o 10 20 30 4o 50 60 7o 80 90 100110120 130140150160170
Strand Slip {front} (mm)
Figure 5-6 Typical Pull-out Test Response - Release
Strand Slip {front} (in.)
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.
1601‘. ........ ,....,...rrv...,.........,....1...1,.1TT,.......n:
150 E
140—
130;
21205 P“ .;
51105-
I$100 E . j
5 80% l/ 5
g 70:— / 1‘
:1 1: -f
X 50 E
g l: /, o NCCB .;
4°. ' - scc1
30; , A 5cm 5
20 I e scczs
10 + SCC3
0

 

 

0 10 20 30 40 50 60 70 80 90100110120130140150160170

Strand Slip {front} (mm)

Figure 5-7 Typical Pull-out Test Response at 7 days

87

6:
Max PullOut Force (kip)

'

N
O

lllllll
ONDQQ-fi-fi-fi-fid

 

D
88388

N
A
Max PullOut Force (kip)

ION
ON

ONbOQ

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

\1
O

50

Max PullOut Force (kN)

‘o" 8 (EN)

Max PullOut Force
0

 
     
   

30,

Strand Slip {front} (in.)

 

l‘r‘r v v I

._

 

Strand Slip {front} (mm)
Figure 5-8 “Close-Up” of First slip occurrence — Release (3 days)

Strand Slip {front} (in.)
.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

 

 

 

L ,.. ...........,...-:22
i 20
I: _18
E :16
:_ 314
:12
.10
: _is
o NCCB ‘6
. I 3cm 3
f A scc2A {4
9 sccza j
+ secs 32
I l [L l l I I | l I All I 0

Strand Slip {front} (mm)

Figure 5-9 “Close-Up” of First slip occurrence 7 days

88

Max PullOut Force (kip)

Max PullOut Force (kip)

 

_A A N N
O 01 0 0|
Max. Pullout Force (Kip.)

O1

 

 

 

NCCB SCC1 SCC2A SCCZB SCC3
MIX TYPE

Figure 5—10 Comparison of Peak Pull-out forces

The results of the peak pull-out forces both at release and at 7 days follow the
same trend (Figure 5—10). The NCCB mix had the highest average peak pull-out force
while SCC1 (high-paste SCC mix), had the lowest. Of the SCC mixes, the highest pull-
out force was for SCC3 (high aggregate content mix). SCC2A was a stiff concrete that
had to be repeated and its performance was closer to the NCCB mix. Thus, the results
support the concept of bounding the response of SCC behavior with the selected mix

designs.

As expected, an increase in peak pull-out forces was observed for all mixes for 7
days relative to those at transfer. The same trend was not observed for the first peak loads
(Figure 5-11). The first peak loads at 7 days were smaller than the first peak loads at 3

days for the NCCB and SCC3 mixes. This could be due to a variety of reasons as the

89

bond phenomena is quite complex. The surface conditions of the concrete at 3 days and 7
days could vary and may affect the bond properties of the strand. These surface
conditions depend on various factors including the age of concrete, the mix design,
admixtures added etc. However, in general, it was found that the first peak load for all

SCC mixes was relatively the same.

 

     

22
90 — At Transfer
———————— "—— _‘Ft‘l Gay‘sr‘ 2°
80 H» ~ ~ 18
£70 ? - ———————————————— 16g
0 I-
— ‘ ************* 12 o
250 ’ a:
2 _7 - 10 3
'540 g
a. — — — a 3
§30 : x'
E _, * ~ 6 «so
20 :__ _‘ 4
10 ; 7 _ 2
1’ o

 

 

 

NCCB SCC1 SCCZA SCC2B SCC3
MIX TYPE

Figure 5-11 Comparison of Pull-out forces at first Slip
The average pull-out forces corresponding to the first slip for all the SCC mixes
were found to be relatively the same at both release and at 7 days. Of all the SCC mixes,
SCC1 had the highest average first slip pull-out force, approximately 40 RN (8.9 kips)
while all others had an average first slip pull-out force of around 30 kN (6.75 kips). The
NCCB mix block yielded the highest average first slip loads for both release and 7 days

of 67 kN (14.93 kips.) and 88 kN (19.67 kips) respectively.

90

The average maximum bond strength (U) for all the mixes was obtained from the
average peak pull-out forces, given the embedment length, and the surface area of the
strand. The following Equation 5-1 shows the calculation of this average maximum bond

strength

0 = :mzzb (5-1)
,.

 

. . . 4 . . .
where, Dn IS the nonunal Circumference =§*7r* db , db is the diameter of the prestressmg

strands [17], and Lb is the embedment length, distance of the strand contributing to
bond = 457 mm (18 in.).

The average maximum bond strength was calculated only from the peak pull-out
force and not the first slip force. The bond strength calculated from the peak pull—out
forces at release for each mix type is reported in Table 5—5. In order to effectively
correlate the bond strength for all mix designs, the bond strength value was normalized
with respect to concrete compressive strength at the time of test. The variation of bond
strength of SCC mixes with respect to the NCC mix was evaluated by observing the
relative ratios of the normalized bond strength of the SCC mixes against the NCC mix.

Table 5-5 Average Maximum Bond Strengths from Peak Pull-out Forces

 

 

 

 

 

 

 

Bond Compressive ' U U .
Peak strength strength U = _S_CC_
Load (U) (Fe) f '. U 'Ncc
(kip) (1751') (psi) J0?»
NCCB 30.40 806.39 5545.12 10.83 1.00
SCC1 16.61 440.51 7685.02 5.02 0.46
SCC2A 26.01 689.80 7693.25 7.86 0.73
SCCZB 19.39 514.42 6703.80 6.28 0.58
SCC3 20.12 533.61 6703.80 6.52 0.60

 

 

 

 

 

 

 

 

91

 

o _s _L
on O M
i
l
I
l
i
l
I
i
i
l
l
1
ui d
O N

Relative Bond Strength
9 c
-b O)

P
M
Relative Bond Strength

 

 

 

NCCB SCC1 SCC2A SCCZB SCC3
MIX TYPE

Figure 5-12 Comparison of Relative Bond Strengths

Comparison of the normalized maximum bond strength (Figure 5-12) shows that all
SCC mixes had less bond resistance than the NCC mix. The SCC1 mix (high fines)
showed the least average bond strength, with approximately 54% less bond resistance
than the NCC mix. The SCC3 (high coarse aggregates) mix showed the best performance
of all the SCC mixes with a 40% reduction in bond strength relative to NCC. The SCC2A
mix being a stiff mix, showed only 27% reduction in bond relative to NCC. However,
this test is not being considered as the reliability of the SCC2A mix is in question. A the
same time the SCCZB showed a reduction of 42% in bond strength relative to NCC,
Hence, the results support the concept of bounnding the response of the SCC mixes with

selective mix designs.

92

5.6 Summary and Conclusions

As discussed earlier bond phenomena depends on different mechanism to transfer
the shear stresses between concrete and strand. Bond shear stresses follow complex
distributions at member ends and at flexural cracks. Because of all these factors
influencing the slip resistance of a prestressing strand in concrete, it is difficult to
determine bond lengths by means of simple pull-out tests [23]. Thus, the correspondence
between the results obtained from this test and structural design parameters such as
transfer length and development length have been questioned for conventional concrete
[8][24][31][32] and seem to be of continued debate now for SCC. While the response
evaluated through simple pull—out tests is clearly related to bond performance, its
correlation to the complex phenomena occurring in the transfer zone region and during
development of strand capacity under flexural actions, as previously discussed, is
questionable. Nonetheless, pull-out tests are good methods to provide a baseline to
qualify the strand bonding characteristics and can serve as a relative performance
measure between normally consolidated concrete and the different SCC mixes under
evaluation (Figure 5—12).

As presented in Section 2.4.1, simple pullout tests can also serve as a good index
tests to qualify strand bond performance. According to Logan’s [24] LBPT guidelines,
the peak pull-out strength and the first slip load for 13mm (0.5 in.) daimeter strands must
be over 160 kN (36 kip) and 71 kN (16 kip), respectively for the strand to be qualified for
the desired bond performance. It should be emphasized, however, that these values have
been proposed based on precise recommendations for concrete age, mix, and test
procedure[24] . As noted earlier with the pull-out tests performed in this research (Figure

5-10 and Figure 5-11), the peak and first slip pull-outs were considerably lower than

93

these values for both the smooth and slightly rusted strand for all mixes. The lower pull-
out forces are attributed to the different concrete mixes (both NCC and SCC) and
modifications in the test setup and test procedure.

In order to check the qualification of the strand based on the pull-out tests, the
strand samples used in this research were independently tested by Logan [24]. Table 5-6
and shows the results of the pull-out tests on strand specimens used in this research as
performed by Logan [24] in full accordance to his LBPT protocol. The “B-control” strand
is the benchmark strand used by Logan to compare to other strands. The slightly rusted
strand (used in NCCB mix) met the peak pull-out force requirement of 160 kN (36 kip),
but did not reach the first slip requirement. The new (i.e clean and shiny) strand (used in

all SCC mixes) did not meet the pull-out force requirements prescribed by Logan [24].

Table 5-6 Pull Out Tests Results - Performed by Logan

 

 

 

 

 

 

 

 

 

 

2']? Standard Peak PuII- Standard
Strand Type Load Deviation out Force Deviation
(kips) (kips) (kips) (kips)
IB-Control 22. 8 3 .37 40.5 2.31
New/Clean 7.9 0.83 31.3 2.91
SR - Slightly rusted 12.9 1.23 37.7 1.43
1 kip: 4.448 kN

 

94

 

 

Maximum Pullout Force (kN)
Maximum Pullout Force (kip)

 

 

B-Control New/Clean SR - Slightly rusted
MIX TYPE

Figure 5-13 Results from LBPT performed by Logan according to [24]

As seen in the results of figure 5—13, the strand used in this research does not seem
to qualify the bond quality requirements with respect to the criteria prescribed by Logan
[24]. Unfortunately, the LBPT evaluation by Mr.Logan came as an afterthought to the
research team upon noticing the low pull-out values and longer development lengths
(Chapter 7) observed in this research. The research team did not pursue this qualification
tests earlier as they ad no reason to doubt the quality of the strand being used. This new
information has obviously raised concerns regarding the validity of the results from this
research program as discussed in this chapter and results presented in Chapters 6 and 7 on
transfer and development length studies, respectively.

Another factor influencing interpretation of the presented results is the effect of rust

in the strand used for the NCCB mix. The surface condition has been recognized as an

95

important parameter to bond performance. Pull-out tests performed on clean and
weathered strands have shown higher first movement and peak load values for weathered
strand by 100% and 24% respectively. The strand used in NCCB was only slightly rusted,
whereby most of the rust was superficial and not complete. The influence of rust in the
testes strand in this program is thus expected to be lower. Under such conditions, the
large difference between the NCC and SCC bond performance can be attributed to the
concrete mix and not the strand. Further evaluations of the test data and consideration of
further testing using a pre—qualified strand are being discussed while completing this
work. Nonetheless, the research team believes that even in the case that the presented
research has been affected by the strand quality the results are still applicable for
assessment of the relative effect of SCC mix proportioning on bond behavior and its
relationship to transfer and development length parameters.

Thus, from the results of the pull-out tests it was found that all SCC mixes have
less bond strengths relative to NCC. Among the SCC mixes, SCC1 had the least bond
strength followed by SCC2 and SCC3 mixes, thereby the behavior of SCC mixes was

bound by the concept of selective mix design.

96

CHAPTER 6 TRANSFER LENGTH EVALUATION

6.1 Introduction

This chapter deals with the evaluation of bond performance of strands in terms of
transfer length in precast/prestressed beams. Two different techniques were used to
experimentally determine transfer length: (i) measurement of concrete strains along the
length of the beam, and (ii) measurement of strand draw-in. Results, observations and

discussion are also included in this chapter.

6.2 Test Procedure - Concrete Strains

As previously defined in Section 2.3.1, transfer length is the distance from the end
of the beam to the point in the concrete member where the entire stress from the strands
is transferred to the concrete member. As discussed before, steel stresses along the beam
length increase rapidly from the beam end until becoming constant once equilibrium
between concrete and steel stresses is achieved. The strain in the concrete can be thus
measured as a means to locate where the strain becomes constant, and hence can be used
to measure the transfer length. This section describes the procedure of instrumentation

and measurement of transfer length using concrete strains.

6.2.1 Specimen Preparation

The DEMEC (DEtachable MEChanical) strain measurement system (Figure 6-1)

was used to measure the strains on the surface of concrete. The DEMEC system consists
of a mechanical gauge used in conjunction with small stainless discs ((1) = 6.3 mm), each
with a small hole (¢ = 1.0 mm) in the center designated to fit the mechanical gage (Figure

6-2.). The stainless discs are glued to the surface of interest at a given spacing over which

97

the strain needs to be measured. These discs thus define strain measuring points or target
points obtained from the change in length between target points (gage length) measured

by the mechanical gage.

lmm
(0T 04 In.

Figure 6- 1 Actual Picture of the DEMEC Figure 6— 2 Schematic Representation of
Instrument Target point

[~—6.3 mm (0.25 in.)—el

For placement of the DEMEC target points the forms were removed after a day
(18-24 hours) and the specimen was allowed to dry at ambient conditions to obtain
surface dry condition. The strain profile strand centerline was marked. The strand
centerline is to be measured along the strand centroid and thus this defines the placements
of the target points. The strand centerline was 51mm (2 in.) from the base of the beam
(Figure 4-1). The concrete surface was grinded and then cleaned along the prestressing
centerline to prepare the surface for bonding of the target points. The target points were
attached using a rapid setting adhesive on both sides of the beam. Since the variation of
stresses is more pronounced in the transfer zone (end of the beam), the spacing of the
target points was 51 mm (2 in) along the expected transfer zone of 1.52 m (60 in.). For
the rest of the beam excluding the transfer zones the spacing of the target points was

increased to 203 mm (8 in.). In order to measure the strains in the target points close to

98

the face of the beam, extension brackets were attached as shown in Figure 6-4. The target
points were attached to both sides of the beam in order to capture any unbalanced effects

from the pair of prestressing tendons. Initial strain readings were taken (Figure 6-3) prior

to the release of the prestressing strands.

   

Figure 6—3 Performing measurement with a Figure 6-4 Extension Brackets to
DEMEC gauge measure the strains at the ends

6.2.2 Concrete Surface Strain Measurements

Concrete surface strain measurements were taken approximately 4 hours after the
release of prestress. Two sets of readings were recorded for each side to increase
confidence in the readings. Readings were taken by the same person and care was taken

to maintain the same amount of pressure and posture while taking the measurements.

6.2.3 Construction of Surface Compressive Strain Profile

The first step to determine transfer length from concrete strains involves the
construction of a concrete surface profile for each end for each side of the beam. The
compressive strain for each measured gauge length of 203 mm (8in.) was obtained by

dividing the gauge length with the difference in the recorded values of the measurements

99

taken prior and after the release of prestress. The value obtained from the measurement of
two DEMEC points is assigned to the middle of these two points. In the transfer zone,
where the spacing of the target points is reduced, these middle values overlap. Hence an
average is taken of three consecutive readings and this value is applied to the middle of
these three points. This procedure has been termed “smoothing the data” [34]. A general
equation for the strain data smoothing procedure is represented as follows:

8- _fi4+%+fin
r,smooth " 3

 

(6-1)

Once the strain values are assigned to each target disc location, the concrete strain
profile is plotted against the distance of the particular target point from the end of the
beam. The data obtained from the DEMEC points tends to have considerable scatter.
Smoothing techniques (Figure 6-5) has been shown to lessen the scatter and reduce the
effect of data points that have values higher or lower than the average. By smoothing the
data it is easier to define the plateau at which the constant strain in the beam is
established [36]. A plot comparison of the smoothed and non- smoothed (raw data) is

shown in Figure 6-6.

100

811-824-193 _

 

 

1”?

..... @MHMGDC‘D

 

 

3 _ €2.5m00th

1

@ ...... ® ......

...... @...,.....@
C3) 2 1

\1\///M/J’/i/”

Figure 6-5 Smoothening Of Strain Profile

Distance from end of the beam (in.)

   
     

   

0 10 20 30 40 50 60 70 80 90 100 1101212,0

 

 

 

 

 

 

 

 

0.00065 0065
0.00060 0.00060
3 0.00055 0.00055
5 0.00050 0.00050
E 0.00045 0.00045
‘5’ 0.00040 0.00040
,5 0.00035 0.00035
5:; 0.00030 0.00030
., 0.00025 0.00025
g 0.00020 0.00020
5 0.00015 0.00015
0 0.00010 ‘ Smem 0.00010
0.00005 — Unsmoothcd 0m 1 0.00005
0.00000 .1121..I.11.211.11.12a4.1...11. .11....1..Lmrm41.4o.ooooo
0 300 600 900 1200 1500 1300 2100 2400 2700 3000

Distance from end of the beam (mm)

Figure 6-6 Comparison of Smooth and Non- Smooth (RAW) data

101

’?
C

InJi

Concrete Strains (

6.2.4 Determination of Average Maximum Strain (AMS)

Due to the scatter in the concrete strain profile, a uniform constant strain value in
the concrete is difficult to define. Thus a representative value, given by 95% of the
average maximum strain (AMS), is commonly used as the effective transfer prestress
level. To determine the AMS, strain values in the likely plateau region were visually
inspected and then the arithmetic mean of these values was calculated. This method is
subjective as it requires visual definition of the plateau region. Care was taken to be

consistent in this approach for all the transfer zones.

Distance from end of the beam (in.)
0 10 20 30 40 50 60 70 80 90 100110120)

 

 

 

0.00065
0.00060 Region over which AMS value obtained 0.00060
A 0.00055 0.00055
E 0.00050 0.00050 2
3 0.00045 0.00045 i
E, 0.00040 0.00040 7:”
3 0.00035 _ 0.00035 -§
§ 0.00030 0.00030 5,
‘3 0.00025 0.00025 9
§ 0.00020 0.00020 §
2 0.00015 0.00015 8
8 0.00010 0.00010 0
0.00005 . 0.00005
0.00000 -444 0.00000

0 300 600 900 1200 1500 1800 2100 2400 2700 3000

Distance from end of the beam (mm)

Figure 6-7 Location of AMS values

In this project, the 95% AMS method was used to obtain the transfer length from

the concrete strain profiles. This method defines the transfer length as the distance at

102

which the measured strain profile crosses a horizontal line representing 95% of the AMS
value. For this, the smoothed concrete strain profiles were plotted along the length of the
beam and a horizontal line representing the 95% AMS value for that particular transfer
zone was also plotted. The transfer length was obtained as the value of the distance along
the length of the beam where the 95% AMS line intersected the concrete strain profile.
With two beams for each mix and concrete strains measurements taken on both
sides of the beam, four transfer zones were evaluated for each beam. Moreover, two trials
of readings were taken for each side of the beam. Hence, a total of 8 transfer length
values were obtained for each beam. A total of 16 values of transfer lengths were
obtained for each mix design. A single plot of concrete strains was obtained for each
beam by averaging 8 sets of concrete strains. Also, a single plot of concrete strains was
obtained by averaging all the 16 sets of concrete strain measurements, thereby obtaining a
single transfer length value for each concrete mix. In this section, plots for each mix
(average of 16 values) are shown as follows: (NCCB (Figure 6-8), SCC1 (Figure 6-9),
SCC2A (Figure 6-10), SCCZB (Figure 6-11), SCC3 (Figure 6-12)). The plots for each
beam (averages of 8 plots) are given in Appendix 3. In order to obtain a single value of
transfer length for each mix, selective numerical average of the transfer length values
obtained from individual concrete strain profiles of the 16 sets of plots was used. The
single concrete strain profile for each mix obtained from the averages of the concrete
strains does not represent the actual transfer length for that particular mix, since it may

include bad data points which may skew the overall plot. As a result, individual transfer

103

length values were obtained from the 16 concrete strain plots and the outlierl values were
not considered in the determination of a single value of the transfer length for a particular
mix. Figure 6-8 compares the transfer length values for each mix obtained from the
numerical averages of all the 16 set of values and the selective numerical average.
Removal of outlier values in the selective averages reduced the standard deviation. It
should be noted that for all the mixes, out of the 16 values only 2 to 3 values were
outlying the average transfer length values, thereby increasing confidence in the data.
Figure 6-13 summarizes the transfer length value for each mix by the method of concrete
strain measurements. It can be noted that the obtained transfer length values show a trend,
where in the transfer length values of the SCC2 mix seems to be bounded by the results

of SCC1 and SCC3 mixes.

 

‘ An outlier is a data point that is located far from the rest of the data. Given a mean and standard deviation,
a statistical distribution expects data points to fall within a specific range. Those values that do not fall in
the specific range are called outliers and should be investigated

104

Table 6-1 Average value of Transfer Length per Mix Type - Concrete Strains

 

 

 

 

 

 

 

 

Transfer Length (inches) 1 in. = 25 .4 mm
Non - Selective Selective Numerical
Numerical Averages Averages

Standard Standard

Mix Type Average Deviation Average Deviation
NCCB 20.53 4.25 19.65 2.56
SCC1 28.82 3.64 29.81 3.25
SCC2A 30.88 4.07 27.00 4.07
SCC2B 31.56 5.15 29.81 3.85
SCC3 30.02 5.54 30.13 3.78

 

 

 

 

 

 

 

Distance from end of the beam (in. )
0 51015202530354045505560

 

 

   
 

 

 

 

 

 

 

 

000050 : ........................................................ ‘ 000050
E 000045 : L 4973 mm i 0.00045
80.00040; €0.000405'
E 0.00035 000035.. =
E 0.00030 0.00030 ‘5
§ 0.00025 g 95%AM3 0.00025 g
<75 0.00020 0.000203
‘3 0.00015 0.00015 §
‘5; 0.00010 0.000105;
0 0.00005 ; NCCB - ALL 0.00005

0.00000 - .................................................. 4 0.00000

0 150 300 450 600 750 900 1050 1200 1350 1500
Distance from end of the beam (mm)

Figure 6-8 Determination of Transfer length from Concrete Strain Profiles - N CCB
(Average of all 16 transfer zones)

105

Distance from end of the beam (in. )

 

 

 

 

 

 

 

 

 

0 5 10 15 20 25 30 35 404550 5560
0.00050 _. ................................................... ”000050
A 0.00045 3 “”7366” J «i 0.00045
; (29.0ln.) ‘ ’T
E 0.00040 _—_ ........................................ 4 0.00040 5
E 0.00035 '\ 000035 E,
2,“ 0.00030 3 95% ms 0.00030 3
0.00025 :- 5 0.00025 g
E E (D
0: 0.00020 : -, 0.00020 0
: : *-
§ 0.00015 ;» g 0.00015 g
‘5’ 000010 0.00010‘5J
0 0.00005 . SCC1 .ALL 0.00005
0.00000 : 1111111111111111111111111111111111111111111111111 4 0.00m0
0 150 300 450 600 750 900 10501200 13501500

Distance from end of the beam (mm)

Figure 6-9 Determination of Transfer length from Concrete Strain Profiles — SCC1
(Average of all 16 transfer zones)

Distance from end of the beam (in. )

0 5 1015 20 25 303540 45 50 55 60
0.00050 . 0. 00050
7:" 0.00045 0.00045
E 0.00040 3 0.00040 5:
\ I W
E 0.00035 ; 0.00035 3
73’ 0.00030 . 0.00030 3
a 0. 00025 3 0.00025 g
3 3 (n
g 0. 00020 I 0.00020 2
‘9' 0.00015 0.00015 g
g 0.00010 030010053
0 0. 00005 x 30ch .ALL 0.00005
o.mooo_ 111111111111111111111111111111111111111111111111 : 0.0m
0 150 300 450 600 750 900 1050120013501500

 

 

 

 

 

 

 

Distance from end of the beam (mm)

(Average of all 16 transfer zones)

106

Figure 6-10 Determination of Transfer length from Concrete Strain Profiles — SCC2A

0 5 10 15 20 25 ”30 35 40 45 50 55 60

000050 p finwfiwfinn”.fln,” ....................... ‘ 000050

E L,= 812.8 mm 3

A 0.00045 , > 1 0.00045
E : (32.0 in.) ; ,7
E 0.00040 3 0.00040 ,5
\ ,_ ..................... x
2 0.00035 5 ’\ -; 0.00035 5
V I 2 m
2 0.00030 C 950/0 AMS ': 0.00030 .5
'5 0. 00025 : -: 0.00025 g
a g 3 w
to 0.00020 .~ -. 0.000020
0) Z I ""
‘0' 0.00015 3 «1 0.00015 1’
'6 ; = 2
5 0.00010 g 0.00010 8
0 0.00005 . 30023 - ALL 0.00005

0.1.00000_.in_1_ai_141...i... .. 11.44441411 41144111111 000000

Distance from end of the beam (in. )

 

 

 

    
  

 

 

 

 

 

 

 

0

0.00050
E 0.00045
E 0.00040
\

E
E 0.00035

7;; 0.00030 g
E 0.00025
('5 0.00020
0.00015
0.00010 ~.
0.00005 g
0.00000 '

Concrete

0

150 300 450 600 750 900 1050 1200 1350 1500

Distance from end of the beam (mm)

(Average of all 16 transfer zones)

Distance from end of the beam (in. )
1015 20 25 30 35 40 45 50 55

 

 

vrfi‘vjvvvvfrvivlvvvv'uvvv vaW vvvvvvvvvvvvvvvvvvvvvvvv

L,= 787. 4 mm
(31.0 in.)

 

A

SCC3 - ALL

AALLLLLAiAAAL AAAAAAAAAA LLL xxxxxxxxxxxx l 2222222222

 

 

 

 

 

0

150 300 450 600 750 900 1050 1200 1350 1500
Distance from end of the beam (mm)

(Average of all 16 transfer zones)

107

Figure 6-11 Determination of Transfer length from Concrete Strain Profiles — SCCZB

0.00050
0.00045A
i 0. 000405
0. 00035.
0. 00030
0.00025
0.00020
0.0001 5
0.00010

Concrete Strains (inJin.

§

0.00000

Figure 6-12 Determination of Transfer length from Concrete Strain Profiles - SCC3

 

Transfer Length (mm)
Transfer Length (in)

 

 

 

\

NCCB SCC1 SCC2A SCCZB SCC3

MIX TYPE

Figure 6-13 Comparison of Transfer Length Values Obtained from Concrete Strains

6.2.5 Precision of Reported Results

The degree of accuracy of the transfer length values obtained by the concrete strain

measurement method depends on various factors as follows:

Spacing of DEMEC points: The minimum spacing of the target points was
51mm (2in.), thus assigning a precision less than that value would
completely rely on the smoothening process and interpolation.

Strain measurements: The DEMEC readings needed to be taken in hard-
to—reach places, which compromised the quality and consistency of the
readings.

Temperature: Almost all the DEMEC measurements were made

consistently at approximately same ambient temperature, since the casting

108

of the test specimen and measurements were all done indoors. But it
should be noted that temperature fluctuations of the concrete surface
between readings will introduce strains that will be incorporated in the
measurements. This can be especially a problem when instrumenting high
strength concrete specimens at early age because of the very high

hydration temperatures [16].

The accuracy of the DEMEC measurements may have been decreased by any of
these factors. The overall accuracy of the DEMEC system is reported to be approximately

16 micro strains. [16].

6.3 Strand Draw-In Method

The second method used to determine transfer length of the 13 mm (0.5 in.)
prestressing strands was the draw-in method. This method follows from the premise that
when prestress is released, the strand at the face of the member is pulled into the concrete
member. Draw-in is the measurement of this pulling-in phenomena, hence it is also
referred to as “suck-in” or ‘free end slip” [7]. In this thesis the term “draw-in” is used to
refer to this phenomenon and the term “end slip” is used to refer the strand slip due to
extemal loads. The draw in measurement can be correlated with transfer length as they

both involve the bonding of the strand with concrete [36] .

6. 3.1 Theoretical Background

The relationship of strand draw-in with the transfer length (L,) was proposed by

Guyon (1953) A simple account of its derivation is provided next. Since there is no

109

displacement between the steel and the concrete at the end of the transfer length, then the

strand draw-in may be calculated as:

A d = as — 6,. (6-2),
where:
A; = Strand Draw in measured at the face of concrete
6; = Contraction of steel tendon in the transfer length zone
(5} = Contraction of concrete in the transfer length zone.

The contractions of the tendon and the concrete resulting from the release of prestress can

be obtained by integrating the strains over the transfer length.

55 = I Aesdx
L! (6-3)
dc = I Agcdx
1,
where:
A8, = Change in steel strains resulting from prestress release
A6} = Change in concrete strains resulting from prestress release.
Equation (6-3) can be re-written in the following form:
A, = j (Ass — Age) dx (6-4)

L,
At the beginning of the transfer zone (in this case, at the face of the beam where bonding
of the tendon with the concrete begins), the change in concrete strains is zero. Hence, the

change in steel strain may be calculated as:

110

= —4 (6-5)

 

where:
6f, = Change in steel stress resulting from prestress release
f,.,- = Intial Stress in the strand — J acking strain minus relaxation of Steel
E, = Modulus of Elasticity of Steel.

At the end of the transfer zone the steel and concrete strains are compatible to each other.
The strain remaining in the section after the transfer zone is the effective stress (fie),

which is related to the concrete and steel by:

(6'6) 2

where:

f“. = effective stress immediately after transfer

If the variation of both concrete and steel strains is assumed to be linear along the
length of the transfer zone, then equation varies from 12,- at the initiation of the transfer
zone to zero at the termination of the transfer zone. Then, a relationship between the
jacking stress, strand draw—in and transfer length can thus be obtained from Equation (6-

4) as:

f.
A =—“— 6-7.
d a.EpL' ( )

The above equation can be rearranged to obtain transfer lengths from measured strand

draw—in as:

111

 

L, = ”3 Ad (6-8)

where:
a: constant that depends on the load stress distribution (on = 2 for constant load
stress distribution and on = 3 for linear load stress distribution [6] ) and
EPS = Modulus of Elasticity of the strand.
In this report, the variation of strains from the beam end is assumed to be lkinear.
With this assumption, the value of on is taken as 2 for the determination of transfer length.

Equation 5-2 can be written in terms of stress as:

= 2.Eps.Ad

(6-9)
fsi

6.3.2 Test Procedure

Draw-in measurement involves determining the relative movement of the strand
into the concrete member at the end of the beam. In order to avoid irregularities in the
concrete surface, a glass target plate was attached with rapid setting glue on the place
where the draw-in measurements were to be taken. The draw-in measurements were
made using a digital vemier with a precision of 0.00254 mm (0.0001 in.). The draw-in
measurements were made possible by mounting reference brackets on to the strands
approximately 50 mm (2in.) from the face of the beam. These brackets (U-channels) were
attached to the strands with metal clamps. The metal clamps were tightened such that

there was no relative motion of the clamp, U channel and the strand. The channel webs

112

had a cavity in them through which the movable end of the vemier was inserted such that
the vemier came in contact with the target glass plate. Since the vemier had to pass
through the two cavities in the channel webs, it made it possible for the readings to be

taken from approximately the same point of consideration, thus making the

measurements more consistent and accurate.

 

Figure 6-14 Strand Draw-in - Instrumentation and Measurement

It should be noted the strands were flame cut after annealing. However, in some
cases, the strands unwound or splayed violently thereby affecting the U—channel
mounting. This problem was overcome by tightly attaching three or more metal clamps
before the measurement device. In such cases, these metal clamps prevented the effect of
the strand unwinding from reaching the measurement device. For most of the test
specimens, prestress release was gradual, and without any disturbance to the reference U
channels.

The draw—in measurements were taken prior and after release (~4 hours later), and

then at 7, 14, 28 days and finally at the day of flexural test (Chapter7). The beams had to

113

be moved from the casting bed to the storage yard and brought back from the yard to the
test setup for flexural test (Chapter7). In some cases, during this process, the glass target
plates for draw-in measurements were damaged. In such cases, new glass target plates
were attached at the same location in the best possible way and draw-in measurements
were carried out. Effect on accuracy from these instances is difficult to estimate but care

was taken in the data analysis to consider the possibility of error sources such as this one.

6.3.3 Determination 0f Transfer Length - Draw-In Value

The draw-in value (A, in Equation 6-7) was obtained as the difference of the
readings taken prior to release with respect to all future readings. This value was used in
Equation 5-3 to obtain the transfer length at various ages of concrete. Figure 6-15 shows

the average draw-in values at transfer for each of the mix types.

 

  

 

 

 

4.0 . 0.16

3.5
A ‘ 0.12 A
E 3.0 E:
E 2 5 ....... ._ Q10 .5
E 2.0 .. ' 0-08 E
5 n
g 1.5 .7 ' 0.06 .2
5 2
iii 1.0 - - 0.04 ‘7’

0.5 ~ 0.02

0.0 553* __ ‘ -- 0.00

NCCB scc1 SCC2A sccza scca
MIX TYPE

Figure 6-15 Average Draw—in Values at Transfer (3 days) for all mixes

114

Although the stress in the strands was almost the same for all strands and for all of
test mix designs, the stress values for each of the test specimens was used to obtain the
transfer length for each mix. Figure 6-16 shows the average transfer length values

obtained at transfer for each of the mixes from the draw—in values.

 

1100-
1000 5‘ ———————————————————————————— 40
900 -
800 »
700 -
500 -
500 -
400 -
300 »
200 ~
100 7

Transfer length (mm)
Transfer length (in.)

 

 

 

NCCB SCC1 SCC2A SCCZB SCC3
MIX TYPE

Figure 6-16 Average Transfer Length Values at transfer (3 days) obtained from Draw-in

Draw-in measurements taken at several times showed that the draw-in values
increased with time. This variation was different for every mix. The least variation was
for the NCC mix and the maximum variation was for the SCC1 (high fines) mix.

Table 6-2 shows the variation of draw—in and the corresponding transfer length

values for various mixes at different ages of concrete. The day of test (DOT) reading for

115

the SCC2A mix was not taken due to some miscommunication between the team

members and hence the corresponding values in Table 5-2 is shown as “n/a.”

Table 6-2 Draw—in and Transfer Length Values at Various Concrete Ages

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Age of Strand Draw - in (inches) Transfer Lenggr (inches)
Concrete Standard Standard
(days) Average Deviation Average Deviation
NCCB
At transfer * 0.0791 0.0185 22.22 5.20
7 0.0934 0.0244 26.22 6.85
14 0.1063 0.0187 29.85 5.24
28 0.1100 0.0186 30.90 5.23
134.5 0.1236 0.0169 34.72 4.74
SCC1
At transfer 0.1086 0.0370 30.79 10.48
7 0.1100 0.0342 31.18 9.70
14 0.1155 0.0477 32.76 13.53
28 0.1210 0.0441 34.32 12.51
126 0.1992 0.0572 56.49 16.22
SCC2-A
At transfer 0.0754 0.0324 19.97 8.58
7 0.1022 0.0198 27.06 5.24
14 0.1105 0.0226 29.26 5.97
28 0.1239 0.0296 32.80 7.85
DOT n/a n/a n/a n/a
SCC2-B
At transfer 0.1012 0.0330 27.21 8.88
7 0.1177 0.0212 31.66 5.69
14 0.1228 0.0068 33.02 1.83
28 0.1301 0.0034 35.01 0.90
126 0.1962 0.0166 52.78 4.47
SCC3
At transfer 0.1314 0.0202 31.19 4.80
7 0.1364 0.0171 32.37 4.05
14 0.1430 0.0102 33.93 2.43
28 0.1551 0.0156 36.79 3.70
129.5 0.2109 0.0479 50.04 1 1.36
* For all beams, release was 3 days after cast 1 in. = 25.4 mm

 

116

 

The variation of draw-in measurements with time is shown in Figure 6-17. The
NCCB beams showed a very gradual and small variation with time, while all the SCC
mixes showed significant increase. The NCCB beam showed an increase of 56% more
transfer length relative to the value at release where as SCC1, SCC2B and SCC3 showed
an increase of 83%, 94% and 60% respectively. Draw-in measurements for SCC2A were

not taken at the day of test.

Age Of Concrete (days)
0 20 40 60 80 100 120 140

 

  
 

 

 

 

 

 

 

5.5 fit I v , 1 , v v , v r v—T , v ‘
E 1 0.20
E 4.0% : 0-18 .s'
g E E 0.16 g
3 ; 3 B
E : 1 0.14 E
D . E : D
'2 2.0 ' i 0-12 '2
“i E _.. NCCB 3 at
'5', 1'55 + SCC1 101023
1.0 E _.t— SCC2A}
0 I + SCCZB: 0.08
-5 _ —+— 5003 3
0.0 ’ ......... r . 1 1 L 1 . I I 1 r . 1 1 L 1 1 . 0.06
0 20 40 60 80 100 120 140

Age Of Concrete (days)

Figure 6-17 Variation of Strand Draw-in with time

6.4 Results and Discussion

Transfer length values obtained from the concrete strains and draw-in
measurements are discussed in this section. Transfer length was also calculated from the
ACI 318 / AASHTO recommendations {3] (Equation 2—6) and the results are compared in

Table 6-3 .

117

Table 6—3 Transfer Length values from AC] 318 / AASHTO equation

 

 

 

 

 

 

 

 

 

 

 

 

Transfer length (fu / 3 )db
Mix (inches)
NCCB 28.74
SCC1 28.60
SCC2A 30.67
SCC2B 30.13
SCC3 34.09
1 in. = 25.4 mm
1100 - 45
1000 7"'—‘—_'"-‘*" ______________ “40
A 900 - G
E 800 » g
g 700 - g,
g 600 - 2
g 500 - " E,
«g, 400 - g
g 300 - I:
" 200 »
100 - 7

 

 

 

NCCB SCCl SCC2A SCCZB SCC3
MIX TYPE

Figure 6-18 Transfer Length — ACI Code — All Mixes [3]

Figure 6-19 shows the overall comparison of the measured average transfer length
values with the values obtained from the ACI code [3]. Both of the measurement methods
(concrete strains and draw-in) show consistent trend in the obtained transfer length
values. The transfer length value of the SCC2 mix is bounded by the SCC1 and SCC3

mixes. When compared with the predicted values from the ACI equation (Table 6-4), it

118

can be seen that all of the measured values are less than code estimate except for the
SCC1 (high fines) mix. Thus the ACI equation appears to be applicable and conservative
with respect to all transfer length for all mixes except SCC1 mix.

Table 6-4 Comparison of Measured and ACI
transfer

    

/

The transfer length values obtained from draw-in and concrete strain
measurements had some variation. This variation was not consistent, transfer length
values from draw-in measurements were generally lower than the values obtained from

concrete strain measurements, except for the SCC1 and NCCB mixes.

45

 

1100 - — Concrete Strains
— Strand Draw-in _ 40
— ACI /AASHTO

1000» "" '

 

 

Transfer length (mm)
Transfer length (in.)

 

 

 

NCCB SCC1 SCC2A SCCZB SCC3

MIX TYPE
Figure 6-19 Comparison of Measured Transfer Length with ACI Equation

119

The average of transfer length values obtained form both the experimentally methods
were compared with the ACI equation for all the mixes. A ratio of measured transfer
length (L, mm) with transfer length predicted by ACI equation (LMCI) was found. Table
6-4 shows this ratio (L, "was / mer) for all the mixes. The table results show that the ACI
code recommendation is conservative in determining the transfer lengths except for

SCC1, which is under predicted by 2%.

6.5 Summary and Conclusions

Transfer lengths were experimentally found for a total of 12 beams (24 ends) of
different concrete mixes. Initial transfer length was determined by concrete strain profiles
and draw-in measurements. Long term transfer lengths were determined only with the
draw- in values.

The overall average transfer length values at release were found to be less than
those predicted by the ACI code provisions, with an exception for the SCC1 mix which
was under predicted by 2%. Transfer length values seemed to increase with time based on
draw-in measurements. Draw-in values were measured for a period of approximately
120-130 days from the date of release. It was observed that the draw-in values increased
from 50% - 100% for NCCB to SCC1 mixes, respectively. It should be noted, however,
that in many cases, the last reading was affected by the condition of the beam end. This
decreases the confidences on the readings at the day of test (~l30 days). In spite of the
insight gained from this part of the project, the ACI code recommendations cannot be

completely validated for SCC because of the limited scope of this project.

120

6.6 Recommendations

The determination of transfer length by draw-in measurement has been supported as
well as questioned in previous research. Draw-in measurements give only the transfer
length value directly and the stress variation along the length of the member cannot be
obtained as in the case of concrete strain measurements. However, draw-in measurements
are much simpler and less time consuming. In addition, the instrumentation is also quite
simple thus cost efficient. This technique is also more practical for the prestress concrete
industry as, it can be easily performed and tracked in the field. This is particularly
advantageous when long term transfer length need to be determined since concrete strain
profiles may not be practically feasible.

However, in order to make draw-in measurements acceptable to the prestress
industry, a detailed statistical study should be performed considering as many factors

like: type of release, strand condition, concrete type and concrete time dependent effects.

121

CHAPTER 7 DEVELOPMENT LENGTH TEST PROGRAM

Evaluation of strand bond on SCC precast/prestressed girders with respect to
development length was made through flexural load tests on laboratory scale beams. The
beam units are those described in Section 4.1 and shown diagramatically in Figure 4-1. As
previously mentioned, two beams for each of the project concrete mix designs (see Table
3-2) were built. The beams were 11.58 m (38 ft) in length such that two tests (one per
beam end) were conducted per beam unit. Thus, a total of 4 flexural tests for each
concrete mix were performed. This chapter provides details on testing configuration,

procedure, observations and results.

7.1 Test Approach

As discussed in section 2.4.2, development length is defined as the total length of
bond required to develop the steel stress fps at the ultimate strength of the member.
Development length consists of two components: transfer length and flexural bond
length. Determination of transfer length is described in Chapter 6. It is not possible to
evaluate flexural bond length separately and hence the total development length was
determined from flexural tests. From these tests and the known transfer length values, the

flexural bond length can hence be calculated.

By definition, the development length of a prestressing strand will depend on
achievement of its design stress level at the section flexural capacity. Section capacity
depends on several factors, which makes development length measurements, or
estimates, difficult to determine in a single test. Thus, a trial and error or bounding

approach has been typically used [14] and was hence also used in this project. The

122

distance from the end of the member to the critical section is defined as the embedment
length (L,). In this project, since the strands are completely bonded throughout the length
of the beam, development length is the minimum embedment length required to develop
nominal stresses at the critical section. The critical section can be defined as the section
closest to the end of the member that develops full strength when subjected to external
loading.

The testing approach thus consists in determining the minimum distance from the
beam end that will allow attainment of the strand design stress level at the section
nominal flexural capacity. The process is thus iterative, where the resulting failure mode
(flexure, flexure—slip, or shear) defines whether the evaluated embedment length was
sufficient. If the test response reveals that the ultimate moment at the critical section was
equal or greater than the nominal capacity, then the next evaluated embedment length
was reduced. Conversely, if the moment nominal capacity was not reached, then the
embedment or development length to the critical section was increased. With beams long
enough for two tests at each end and to beams per concrete mix, the current project could
afford four trials. In this way, the range within which the actual development length may

lie for the particular mix can be obtained.

7 .2 Test Configuration

The flexural test setups consisted of a simple span beam loaded under a pair of
concentrated loads at the critical section and a cantilever overhang that was unaffected by
the test. A schematic of the test setup is shown in Figure 7-1. For most of the tests, the
beam was supported over a span of 7.32 m (24 ft) leaving a cantilevered length of 4.27 m

(14 ft). In the first few tests (NCCA and SCC2A), the span length was kept as a variable

123

and the support blocks were not moved. In all other tests (SCC1, SCC2B, SCC3 and
NCCB), the span kept made constant at 24 feet. The beam was supported on two
neoprene pads of dimensions 150 x 305 x 19 mm (6.125” x 12” x LVi”) on each support.
Loading was applied by means of a servo controlled hydraulic actuator mounted on a
reaction frame. The actuator load was transferred to the girder through a loading beam
with two contact points in order to create a constant moment demand region. Since the
development length testing requires that the embedment length be varied from test to test,
the support blocks were moved to get the required embedment length. Figure 7-2 shows
the overall test setup at the MSU Civil Infrastructure laboratory with some of the

components of the test setup labeled.

Embedment length or Test Development

 
  
 

 

 

 

 

 

 

Length
P .
C . . l l Cantilevered end unaflected by
rztlca -
a lzed loads
Section l—® pp
A-A
l
32 N
Maximum Moment
Region Lab Floor
4_ 6n
+—-——24 ' 0"
Center to Center of Bearing pads
4 38' 0” h

 

 

 

Figure 7-1 Schematic Representation of Flexural Test

124

. .
. .5...r..:gz:.: ..=

. §=§§fia§§§

 

Figure 7-2 Overview of the Flexural Test Setup

 

3 Test setup — View of Spreader Beam

Figure 7—

125

Figure 7-3 shows a close-up view of the spreader beam attached to the actuator to
create a constant moment region. Two aluminum plates of 305 x 76 x 6.4 mm (12” x 3”x
Mt”) were used to transfer the loads from the spreading beam loading points onto the
beam flange. Two tilt prevention blocks 457 x 610 x 1372 mm (18” x 24” x 54”) were
tied to the strong floor and used on either side of the test specimen to prevent for the
event that the beam may become unstable. The different test configurations: effective
span, embedment length (LdW) , the shear span (considering the center of the support),
the test date for all the mixes are given in Table 7-1 for the 24 tests performed.

Table 7-1 Test Configurations for Development Length Studies

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Mix Unit End Test Date Emmi" 5"“ La“, Shea”
Span, a
l in. = 25.4 mm (ft) (in) (ft)
NCCa 1 A 6-Oct-04 23.83 76.00 6.08
1 B ll-OCt-04 27.67 122.70 9.98
2 A l4-OCt-04 26.67 1 1 1.00 9.00
2 B 20-OCt-04 24.25 60.00 4.75
SCC2A l A 30-OCt-O4 23.50 70.50 5.63
l B 3-NOV—04 23.50 64.50 5.13
2 A 9-NOV-04 23.50 80.00 6.42
2 B l6-NOV—04 23.50 86.75 6.98
SCC2B l A l7-Nov-04 24.00 70.50 5.63
l B l9-NOV-04 24.00 102.75 8.31
2 A 23-NOV-04 24.00 126.75 10.31
2 B 30-NOV-04 24.00 124.50 10.13
SCC3 1 A 2-DeC-04 24.00 58.00 4.58
1 B 7-DeC-04 24.00 97.75 7.90
2 A lO-DeC-04 24.00 106.50 8.63
2 B 21-DeC-04 24.00 103.00 8.33
NCCB 1 A 22-Dec-04 24.00 63.75 5.06
l B 25-DeC-04 24.00 64.00 5.08
2 A 29-DeC-04 24.00 103.50 8.38
2 B 30-DeC-04 24.00 93.50 7.54
SCC1 l A 4-Jan-05 24.00 72.38 5.78
l B 5-Jan—05 24.00 133.75 10.90
2 A 4-1 an-OS 24.00 122.00 9.92
2 B 4-Jan—05 24.00 1 18.50 9.63

 

126

 

7 .3 Analysis of Section at Ultimate - Nominal Capacity

The analysis of the section for its ultimate strength is necessary to determine the

nominal moment resisting capacity of the section. The cross-section dimensions, material

properties, amount of reinforcement and the amount of prestress force in the strands must

be known to calculate the moment at ultimate. The ultimate strength of the section is

achieved outside the linear range of the behavior (load—deflection) of the prestressed

section. Actual analysis of this response is quite cumbersome and may not be feasible in

daily design practice. Simplifications have been made by most code provisions which

allow a fast but sufficiently accurate evaluation of the nominal capacity of the section [4].

The following assumptions are made by the ACI code recommendations (Figure 7—4) [4]:

1.

2.

Plane sections remain plane before and after loading; strain distribution is linear
Perfect bond exists between steel and concrete

The limiting compressive strain of concrete is 0.003 for all cross-sections, types
of concrete and amount of reinforcement.

Tensile strength of concrete is neglected.

The total force in the concrete compressive zone can be approximated by
considering a uniform stress of magnitude 0.85f’c over a rectangular block
(Whitney’s block) of width b and depth a=fl1c, where c represents the depth of
the neutral axis and ,6; (Equation 7-1 ) depends on the compressive strength of

concrete as:

,6] = 0.85 for f 'c S 4000psi
,6] = 0.65 for f'c 2 8000psi (7-1)
[3, = 0.85 —5x10’5(f'c —4000) for f'c 5 40005 8000psi

127

 

 

 

 

 

 

 

 

 

ASSUMEI: STRAIN ACTUAL STRESS ACI ASSUMEE
DIAGRAM DIAGRAM STRESS DIAGRAM

Figure 7-4 ACI-318 code Assumed Stress — Strain Distribution [4]

Calculation of the nominal capacity of a prestressed section according to the ACI
method involves the assumption of the Whitney’s rectangular stress block. In order to
satisfy the horizontal equilibrium of the section, the resultant tensile (T) and compressive
forces must balance each other. The nominal moment capacity (Mn) is obtained from the

couple created by the resultant tensile (T) and compressive (C) forces, then thus giving:
, a
M, = 0.85 f c ba(d — '2'} (7-2)

where, d is the distance from the extreme compression fiber to the centroid of the tensile
force. If the section includes any passive compressive reinforcement, the contribution of
such reinforcement must be included in the total compressive force. The centroid (d) of
the tensile reinforcement can be estimated accurately from the locations, area and yield

strengths of the prestressing tendons (dp, Aps) and passive reinforcement ((1,, A,):

d = Apsfpsdp + Asfyds

(7-3)
Apsfps + Asfy

 

128

The ACI-318 code provides a simplified approach to determine the value of stress in the
strand at the nominal capacity of the member (fps) if the effective stress (fse) is not less

than 0.5 times the ultimate strength of the tendon (5..) to be given by:

7,, fpu d .
fpszfpu 1"‘31' pp—'—+‘J-(w_w) , (7-4)
c p
where:

)1, = factor for type of prestressing steel

2 0.55 for jig/f” not less than 0.80

= 0.40 for jig/f” not less than 0.85

= 0.28 for fp/fpu not less than 0.90
j}, = specified yield strength of prestressing steel (psi).
pp = ratio of prestressed reinforcement = Ap/bdp
dp = distance from extreme compression fiber to centroid of prestressed

reinforcement

a),a)’ = reinforcement indices, 0): pfy/fl' & (0' = ,O'fy /fl'
p’ = ratio of compression reinforcement = A/bd

Clearly, the nominal capacity can also be calculated from the resultant tensile forces:
a
Mn = (Apsfps + Asfy )(d _5)‘ (7—5)

Calculation of the nominal capacity by the ACI-318 code as discussed above is a
simplified method. A more accurate analysis involves the use of the actual stress
distributions in the section by considering strain compatibility and realistic material
constitutive models. In this approach, the strain distribution in the section is still assumed

to be linear. (plane sections remain plane). However, no assumptions for fin are made.

129

Rather, at every point of loading equilibrium between resultant compressive and tensile
forces is checked and fps is calculated accordingly.

In this project, the nominal capacity was calculated by three methods: a) ACI-318
code equations, b) a custom program using a strain compatibility approach using
Rambcrg-Osgood constitutive model and c) a research software for analysis of concrete
sections (RESPONSE 2000) [15], which also uses strain compatibility method and
refined stress-strain models for both concrete and steel.

As discussed earlier, the ACI equations represent a simplified method of
calculating the nominal section moment capacities. The custom program code used to
find M, by strain compatibility was validated by comparing its results to those from
Response 2000 [15]. However, Response 2000[15] is a more powerful program that takes
into account the detailed material constitutive models and can predict strain demands at
various stages of loading. Taking these advantages to the best use, the moments
developed during the test program (MW) were compared with the nominal moment

capacities obtained from both the ACI equations (MMCI) and those of Response 2000

(MRes)-

7 .4 Instrumentation

Instrumentation for the development length tests can be broadly classified into two
types: (1) primary instrumentation — used to study the response of parameters essential to
the development length study, and (2) secondary instrumentation - used for monitoring
the overall test response and safety of the specimen and the crew during testing. All

instrumentation readings were automatically recorded via a data acquisition system

130

The servo—controlled hydraulic actuator has in-built transducers that measure the
load in the actuator and its displacement. The actuator has a capacity of 1450 kN (328
kips) with a stroke of 1016 mm (40 in.). The actuator load and displacement signals were
recorded both at the controller computer and the data acquisition system.

Potentiometers were used to measure all the displacement responses of the test
specimen. Two types of displacement transducers were used: (1) devices with a stroke of
305 mm (12 in.) were used to measure deformation under the points of application of the
loads, and (2) devices with a stroke of 38 mm (1.5 in.) were used to measure support
movements and strand end—slip. Figure 7-5 shows the three 305 mm (12 in.) devices that
were used to measure the test unit displacements; two were used at the two points of
application of the load and the third one at the center. A total of six 38 mm (1.5 in.)
displacement transducers (38mm) were used: Two to measure the vertical deformation at
the supports, two to measure the horizontal motion of the supports, and the last two were
used to measure end-slip of the strands. The transducers used to measure strand end slip
were attached to the strands with mounting of brackets and clamps. Figure 7-6 shows one
of the supports with the instrumentation to measure these displacements.

Compressive strains at the top surface of the flange were monitored throughout
the test with a 60 mm (2.36 in.) foil type strain gage placed at the top compression fiber

of the section in the middle of the constant moment region.

131

fit!"

llm‘t'tllilnitik

L

 

Figure 7-6 Instrumentation for Support Movement and strand End—Slip

132

In order to monitor the strains developed in the strand, average strains were
measured on the concrete surface at the strand level on the constant moment region by
means of a mechanical gage. DEMEC target points attached at strand level were spaced
at 152.4 mm (6 in.) for a distance of 762 mm (30 in.) centered in the constant moment
region. Initial readings were taken before the start of the experiment. Loading was
applied in displacement-control at a rate of 2.5 mm (0.1 in.) per minute. A set of readings
were taken after each displacement loading increment. Figure 7-7 shows the strain

measurements being taken at the end of a displacement loading cycle.

 

Figure 7-7 Average Strain Measurement at Strand Level

133

7.5 Failure Modes and Analysis:

The failure mode and definition of the nominal moment capacity at the critical
section performed play a vital role in determining the development length of the strands
in test unit. The ultimate moment achieved at the critical section for a given test (MW)
was that obtained from the test response and the recorded data. Since all the beams tested
in this project were of the same cross section, the nominal moment capacity (M,.) of the
section for each test specimen depends on the amount of effective prestress in the strands
and the concrete strength (Table 3-9) at the day of test. As discussed earlier in section
7.3, the nominal moment capacities were calculated by two methods: (1) ACI 318
method (Equation 7-2) [3], and (2) refined strain compatibility analysis (Response 2000)
[15].

For a given embedment length, if the moment achieved in the critical section was
greater than the calculated nominal capacity of the section, then the embedment length
was considered equal to or greater than the actual development length and the
embedment length for the next test was reduced. Conversely, if the moment achieved in
the critical section was less than the calculated nominal capacity of the section, then the
embedment length was considered to be insufficient relative to the actual development
length and the embedment length for next test was increased. As a result, a range within
which the actual development may occur for a particular mix was obtained. Three distinct
failure modes were observed in the development length tests. 1) shear—slip failure,
2) flexural failure (no slip), and 3) flexure-slip failure. These failure modes are discussed

as follows:

134

Shear-Slip failure:

This type of failure was initiated by large slip (or draw-in) at the free end of the
beam. The nominal capacity of the section was not achieved under this type of failure. In
this type of failure, the test behavior is as follows: Initially, as the loading was increased,
flexural cracks symmetrical to the points of application of load were observed. The crack
propagation ceased to remain symmetric after the first slip occurred. The Crack closest to
the end (support from which embedment length was measured), grew relatively much
faster than the other cracks. Initiation of the strand end-slip is characterized by the
formation of a horizontal crack at the strand level(Figure 7-8). This horizontal crack
usually occurs at the crack closest to the support. As the load was increased, the strand
end-slip increased rapidly until there was a compressive failure at the top flange. The
strains in the strands were much lower (~ 0.220 strains for SCC3-l-A) than the ultimate
strain capacity of the strand (0.035 strains) and the moment capacity of the section was
not achieved. Figure 7-8 shows the initial stages of one of the shear-slip failure tests.
Figure 7-9 shows the top flange compressive failure with the expansion of the crack close

to the support. Figure 7-10 shows the test response corresponding to this type of failure.

135

 

Figure 7—8 Shear- Slip Failure — Initial stage

 

Figure 7-9 Typical Shear-Slip Failure

136

Displacement at the section (in.)
3

 

 

 

 

 

 

o 1 2 5 6 7 8 10
120 -_‘_‘__'___‘_‘.‘___'.T_:_LEL‘LLL‘;L‘.‘.‘_:‘_'___'_'_‘_'-, 9°
3 Mn = 117 kN-m (86.3 k-ft) _
110 ‘. _ 80
100 ‘ Slip Onset
- 70
E 90 — g
2' so - A ’ 6° 3
5 70 P12 P12 g
560— v-._, -— / {1&1 40$
5 I‘ «Ax-5 ————— — =- a
5 bl—da—fi . — 30
30 ~‘ ' ) 1r
5 l/ : 20
20 —‘
: A - 10
10 1
o . , . Y . . I 1 1 y 1 v v I v I r v V ' ' 1 r I I 0
0 30 so 90 120 150 180 210 240 270 300
Displacement at the section (mm)
Figure 7-10 Test Response for a Typical Shear—Slip Failure

This type of failure is basically a pure flexural failure wherein no end slip is

F lexural Failure:

137

observed. The nominal capacity of the section was mostly achieved in this type of failure.
In this type of failure, the test behavior is as follows: Initially, as the loading was
increased, flexural cracks symmetrical to the points of application of loads were observed
(Figure 7-11). The cracks propagate symmetrically throughout the test and large crack
openings and deformations were observed. As the load was increased, the crack pattern
remained symmetric and the crack grth was proportional to the time of occurrence of
the crack during the test. The failure, in most cases was reached first in the top
compression flange. The compression flange failed within the maximum moment region

(Figure 7-12) and not at the point of application of load as observed in test units

displaying shear—slip failures. In most cases, for this type of failure, the measured strain
in the strand was beyond the guaranteed ultimate strain of 0.035 strains. Out of all the 24
development tests performed, the strands fractured in only two units. Figure 7-13 shows
the flexural failure for one of the units failing with strand fracture. Figure 7-14 shows the
moment—displacement response corresponding to this type of failure. In most cases, this
type of failure occurred at long embedment lengths. Even though no strand slip is
associated to this type of failure, this type of response does not guarantee that the nominal

capacity of the section is achieved.

 

Figure 7-11 Flexural Failure — Symmetric crack pattern

138

.l
U

‘3
ll

' "‘11‘1fi

:.

r ‘12-. y

is.“
. 2.2..

2__=.__.._.:.._,
1. a;

. 3...:

:.=

:_,.

=3.

1
~

u: :1...

a

2...:

 

Figure 7-12 Flexural Failure — Final Con 'tion (compression)

 

)

ion

Final Condition (tens

Figure 7-13 Flexural Failure

139

Displacement at the section (in.)
0123456789101112

 

 

    

   

 

 

 

120 _____________________________ {90
=1 - . - I

110 Mn 17.27kNm(865kfl) :80
NoSllpOccured :

A100 Mnachleved :70;

59° 1 E g

¥80 3600

v t :1

E P12 P12 i '0

g 60 "~.-. 1 -.-—°" . E 40;

50 L 5'5" E V
30 1 V i I
. l f
20 ’ .
A L1

10 0
0,,fifis o

 

o 30 60 90 120 150 180 210 240 270 300
Displacement at the section (mm)

Figure 7-14 Test Response for a Typical Flexural Failure

F lexural-Slip Failure:

In this type of failure a combined effect of flexure and strand slip effects were
observed. The nominal capacity was achieved at the critical section only in some cases,
depending on the dominance of either the flexure or the strand slip contributions. If the
test embedment length was slightly greater than the actual development length, then the
bond component was dominant and the moment capacity at the critical section would be
achieved with some amount of slip. Similarly if the test embedment length was slightly
less than the actual development length then the strand-slip component would dominate
and the moment capacity in the section would not be achieved. Only the cases in which
the nominal capacity of the section was achieved are considered in this type of failure. If
the nominal capacity was not achieved, such failure types were classified as shear-slip

failures. In this type of failure, the test behavior was a combination of the first two cases.

140

Initially, as the loading was increased, flexural cracks symmetrical to the points of
application of loads were observed (Figure 7-11). As the loading was increased,
depending upon the dominance of the flexural bond or strand-slip contributions, the crack
propagation varied slightly. The variation was mainly in the shear crack causing the
strand slip. If the strand-slip component was larger than the flexural bond component,
then this crack would grow relatively faster than the flexural cracks. If the flexural bond
component dominated, then the flexural cracks would grow faster than the shear crack. It
should be noted that strand-slip was recorded in all cases, irrespective of the dominance
of the components.

Displacement at the section (in.)
o 1 2 3 4 5 6 7 8 9 1o 11 12

 

 

 

 

r
120 M9 _=_1 1.7411011; (scum _________________ 5 9°
110 :80
100 .

A ~70A

E 90 : i

:2 3° 3603

v SllpOnset : .-

‘E 70 :50 g

0 E E

goo . o

E 50 :40:
40 24””
20 , :

"A :10
10 .
o ,....,....,....,..-rr....,....,....,....,....,....,....,....,’o

 

0 25 50 75 100125 150175 200 225 250 275 300
Displacement at the section (mm)

Figure 7-15 Test response for a typical Bond-Slip failure
In most cases, this type of failure was accompanied with the top flange failing in
compression (Figure 7-12). The measured steel strains in this type of failure were very

close to the strand guaranteed ultimate strain of 0.035 strains.

141

Figure 7-15 shows the test response for the case wherein the flexural response
was dominant and the moment capacity was achieved. The test response for the case
wherein the strand-slip component was dominant and moment capacity was not achieved

is similar to atypical shear-slip failure as shown in Figure 7-10.

7 .6 Presentation and Discussion of Test Results

A total of 24 development length tests were performed at MSU’s Civil
Infrastructure Laboratory. Four tests per concrete mix were performed. Excluding the
NCCB test specimens, all other test specimens had unrusted clean strands The NCCB test
specimens used a slightly rusted strand, since it was acquired a couple of months later
due to the need to repeat the NCC mix. A clean cloth was used to wipe off the strands
with one stroke before the cast of concrete. The N CC and SCC2 mixes were repeated due
to the poor performance of the mix and the equipment, hence the first trial is represented
by suffix “A” and the second trial is represented by the suffix “B.” A brief description of
the test results for individual concrete mixes are given in this section. The test results are
tabulated in Table 7-2. The values of LA“, , MMC, were obtained from the ACI-318
equations using a value for fin (nominal stress in the strand) as per the ACI 318 provisions
(Equation 7-4) [3]. The value of Muse was obtained from the strain compatibility. The
nominal strain at the nominal strength of the section (cps) was obtained from strain
compatibility analysis and the corresponding fin was evaluated from a constitutive model
for prestressing steel (Ramberg Osgood model). Table 7-2 also shows the failure type of
the test unit, where in “F” represents flexural failure, “S” represents the shear-slip failure

and “FS” represents flexural slip failure. Results of the flexural tests for each mix design

142

are explained in detail in two cases: the ACI 318 method and the strain—compatibility
method in the following:

Table 7-2 Development Length Test Results

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Mnfps Strain
Mmfi” ACI-318 Compatibilty (SC)
Ldtest L .
Test ID L. M L—m. M F :3?
(m. ) —Ld_AC, M "_ACI d—ACHSC Mn-SC
SCC1-1-A 72.38 1.07 1.057 1.01 0.981 FS
SCC1-LB 133.75 2.03 1.137 1.92 1.056 F
SCC1-2-A 122.00 1.79 1.121 1.70 1.039 F
SCC1-2-B 118.50 1.74 1.217 1.65 1.128 F
SCC2A-l-A 70.50 1.09 1.105 1.03 1.024 F S
SCC2A-1-B 64.50 1.00 0.966 0.94 0.896 S
SCC2A-2-A 80.00 1.27 1.127 1.19 1.045 F S
SCC2A-2-B 86.75 1.37 1.137 1.29 1.054 F S
SCCZB-l-A 70.50 1.21 1.007 1.13 0.934 FS
SCC2B-l-B 102.75 1.76 1.140 1.65 1.058 FS
SCCZB-Z-A 126.75 1.78 1.101 1.69 1.021 F
SCC2B-2-B 124.50 1.75 1.187 1.66 l. 100 F
SCC3-LA 58.00 1.06 0.953 0.98 0.884 S
SCC3-LB 97.75 1.79 1.090 1.66 1.010 FS
SCC3-2-A 106.50 1.80 1.100 1.67 0.994 FS
SCC3-2-B 103.00 1.74 1.132 1.61 1.050 F S
NCCA-l-A 76.00 1.06 0.952 1.03 0.910 F S
NCCA-l-B 122.70 1.71 1.158 1.67 1.107 FS
NCCA-Z-A 111.00 1.55 1.201 1.51 1.149 FS
NCCA-2-B 60.00 0.84 0.907 0.82 0.867 S
NCCB-l-A 63.75 1.06 1.036 1.00 0.972 F
NCCB-l-B 64.00 1.07 1.049 1.00 0.983 F
NCCB-Z-A 103.50 1.31 1.145 1.25 1.076 F
NCCB-Z-B 93.50 1.22 1.137 1.16 1.068 F
TYPE OF FAILURE:
S - Shear Slip Failure . _
F — Flexural Failure 1 1n. _ 25'4 mm
FS - Flexural Slip failure

 

 

143

7.7 Development Length Test Results as per ACI-318 Method [3]

Results of the development length tests for each mix design with the nominal
moment capacity (Mn) and the nominal stress in the strand at the nominal capacity of the
section (fin) calculated as per the ACI-318 recommendations are explained next. Specific
test details and failure type are given only for the test with the minimum embedment
length that satisfied the moment requirements:

§C_Cl; All the four test beams achieved the nominal moment capacity. The least

test embedment length that achieved the moment capacity had an Lam/Lama ratio of

1.07, corresponding to L4,“, of 1.83 m (72.38 in.). The maximum average value of

strain in the steel at strand level measured on the concrete surface was 0.039. The

corresponding Mun/Mm“, ratio was evaluated to be 1.057. This test had a flexural-

slip failure

513% Three out of the four test beams achieved the nominal moment capacity.
The least test embedment length that achieved the moment capacity had an
Ldm/LdAC, ratio of 1.09, corresponding to Ldm, of 1.79 m (70.50 in.). The
corresponding Mus/Mm“, ratio was evaluated to be 1.105. This test had a flexure-

slip failure

639211; All the four test beams achieved the nominal moment capacity. The least
test embedment length that achieved the moment capacity had an Lam/Lac: ratio of
1.21, corresponding to a de of 1.79 m (70.50 in.). The average value of strain in
the strand as measured on concrete surface was 0.0183 strains. The corresponding

M,,._,,/M,.-Ac, ratio was evaluated to be 1.007. This test had a flexure-slip failure

144

7.8

m Three of the four test beams achieved the nominal moment capacity. The
least test embedment length that achieved the moment capacity had an Mus/Lac,
ratio of 1.79, corresponding to Em, of 2.48 m (97.75 in.). The maximum average
strain value in the steel at strand level was 0.0371 strains. The corresponding

M,,.,,/M,,-AC, ratio was evaluated to be 1.090. This test had a flexure-slip failure

NCCA: Two of four test beams achieved the nominal moment capacity. The least
test embedment length that achieved the moment capacity had an Lanes/Lac, ratio of
1.55, corresponding to L4,“, of 2.82 m (111 in.). The corresponding M,,,_,/M,,-AC,

ratio was evaluated to be 1.201. This test had a flexure-slip failure

m All the four test beams achieved the nominal moment capacity. The least
test embedment length that achieved the moment capacity had an Ldm/LdACI ratio of
1.07, corresponding to a Ldm, of 1.63 m (64.00 in.). The maximum average value of
strain in the strand was 0.0264. The corresponding Mus/Mm“, ratio was evaluated

to be 1.049. This test had a flexural failure

Development Length Test Results as per the Strain Compatibility (SC)
Method.

Results of the development length tests for each mix design with the nominal

moment capacity (Mn) and the nominal stress in the strand at the nominal capacity of the

section (flu) calculated as per the strain compatibility (SC) method are explained as

follows:

145

& Three of the four test beams achieved the nominal moment capacity. The
least test embedment length that achieved the moment capacity had an
Ldm, 1410,50 ratio of 1.65, corresponding to a Lam, of 3.01 m (118.50 in.). The
maximum average value of strain in the strand was 0.039. The corresponding

Mus/Music; ratio was evaluated to be 1.128. This test had a flexural failure

_SC_C22_1; Three out of the four test beams achieved the nominal moment capacity.
The least test embedment length that achieved the moment capacity had an
LdtesfladACNSC) ratio of 1.03, corresponding to a L4,“, of 1.79 m (70.50 in.). The
corresponding M,,s/M,,-AC,(5C) ratio was evaluated to be 1.024. This test had a

flexural-slip failure

w Three of the four test beams achieved the nominal moment capacity. The
least test embedment length that achieved the moment capacity had an
Ldm/LdAcuscjratio of 1.65, corresponding to a Ldm, of 2.61 m (102.75 in.). The
maximum average value of strain in the strand was 0.0295 . The corresponding

M,,s,/M,,-Ac1(5c) ratio was evaluated to be 1.058. This test had a flexure-slip failure

SCC3: Two of the four test beams achieved the nominal moment capacity. The
smallest test embedment length that achieved the moment capacity had an
Ld,e,./LdAc,(sc) ratio of 1.66, corresponding to L4,“, of 2.48 m (97.75 in.). The

maximum average value of strain in the strand was 0.0371 strains. The

146

corresponding M,,,/M,,-AC,(5C) ratio was evaluated to be 1.010. This test had a

flexure-slip failure

NCCA: Two of four test beams achieved the nominal moment capacity. The least
test embedment length that achieved the moment capacity had an Ld,,s/LdACI(SC)
ratio of 1.51, corresponding to a Ldm, of 2.82 m (111 in.). The corresponding

M,e,,/M,,-AC,(5C) ratio was evaluated to be 1.149. This test had a flexure-slip failure

m Two of the four test beams achieved the nominal moment capacity. The
least test embedment length that achieved the moment capacity had an
Ldm/LdACHSC) ratio of 1.16, corresponding to Ldm, of 2.29 m (93.50 in.). The
maximum average in strand was 0.0379. The corresponding Mm, /M,,-AC,(5C) ratio

was evaluated to be 1.068. This test had a flexural failure

The refined method (strain compatibility method) of determining the nominal
moment capacity gives higher moment capacities and longer development lengths
relative to those based on the values taken from ACI-318 recommended equations. As a
result, it was observed that in some cases that the nominal capacity as per the ACI-318
code was achieved in the development length tests but the same tests did not meet the
nominal capacity as determined by strain compatibility approach. Out of the twenty four
development length tests, twenty tests achieved the moment capacity requirements of the
ACI-318 recommendations, but only fifteen of them met the refined analysis
requirements. Although, the refined analysis methods may not be available or practically

feasible for everyday design, these methods are more conservative.

147

The development length as discussed was found iteratively by bounding the
failure modes. In order to obtain the relative performances of each mix type, the test
development lengths need to be normalized. The test data (ratios of L4 and M") were fit to
a third order polynomial and the value of L, ratio corresponding to a value of unit ratio of
nominal capacities (Mm, / Mmdwwy = I) was found. Table 7-3 shows the values of
normalized Ld ratios calculated from the ACI-318 and strain compatibility method.

Table 7-3 Comparision of Development Length Results

 

 

 

 

 

 

 

 

ACI - 318 Strain Compatibility (SC)

Must Ld-test Ld—“Pi- M test Lil-test Ld-expr.

M n-ACI Lei-ACI Ld—ACI M n-SC Ld—ACI(SC) Lat—ACI
SCC1 1.06 1.07 1.03 1.128 1.650 1.2958
SCC2a 1.1 1 1.09 1.04 1.024 1.030 1.0300
SCC2b 1.01 1.21 1.17 1.058 1.650 1.6600
SCC3 1.09 1.79 1.42 1.010 1.660 1.6600
NCCb 1.04 1.06 0.97 1.076 1.160 1.0500

 

 

 

 

 

 

 

 

 

Results from the ACI-318 method show that the SCC beams required longer
development lengths relative to the NCCB beams. The SCC1, SCC2B and SCC3 beams
had 3%, 17% and 42% longer development lengths while NCCB beam had 3% shorter
length than recommended by the ACI-318 code. Conversely, the results based on strain
compatibility analyses showed that ACI-318 may be under-predicting the values of
development length for all mixes. The SCC1, SCC2B, SCC3 and NCCB beams required
30%, 66%, 66% and 5% longer development lengths than that predicted by the ACI-318
recommendatons. Figure 7-16 shows the comparison of the experimental and ACI-318

predicted development lengths for all mix designs.

148

 

— Ld-expt/Ld-ACI
’—7 ' ‘ LdiA’éfl Lit-261 ’ 7 7
. _ Ld-expt/Ld-Ae1-(SC)—-

 

 

l'd-expi / Ld - theory

 

 

 

NCCB SCC1 SCC2A SCCZB SCC3
MIX TYPE

Figure 7-16 Comparison of Development Length Results

7.9 Flexural Bond Length:

As discussed earlier, a direct experimental method of determining only the flexural
bond length is not possible. Hence, the total development length is obtained from flexural
tests as as presented in this chapter and the flexural bond length is obtained as the
difference of the development and length and the measured transfer lengths (Chapter 6).
The development length was studied in two ways: (1) based on the ACI-318 equations
and (2) based on strain compatibility analyses. The theoretical transfer length was
obtained only by the ACI-318 recommendations. The experimental flexural bond length
for each mix was obtained from the difference of the test development length (LAW) and
the average experimental transfer length. The ratio of measured flexural bond lengths

with that of ACI-318 predicted values for the test units corresponding to the least

149

deve10pment length for each of the mix design are given in Table 7-4. Similar to the
development length, in order to compare the flexural bond lengths for different mixes, the
test flexural bond lengths were normalized for unit nominal moment ratios.

Table 7-4 Flexural Bond lengths

 

 

 

 

 

 

[frail/1441C! Lj—em/Lf/lCI
SCC1 1.10 1.21
SCC2A 1.32 0.99
SCC2B 1.60 1.53
SCC3 3.53 2.29
NCCB 1.58 1.00

 

 

 

 

 

The experimental flexural bond lengths for NCCB and SCC2A (stiff mix) test
units measured experimentally agree with the ACI-318 predictions. However, for all
other SCC mixes the experimental flexural bond lengths are longer than the predictions
of the ACI-318 code. It is observed that the SCC1, SCC2B and SCC3 test beam units had
21%, 53% and 229% longer flexural bond lengths respectively. It should also be noted
that the values of flexural bond length discussed here were obtained from the transfer
lengths measured at the release of prestress. The development length tests were
performed approximately 120 days from the day of release. The transfer lengths in
prestressed members increase slightly with time and creep [9]. The increase in transfer
length with time was found experimentally with draw-in measurements. The draw-in
measurements were taken at time of release, 7, 14, 28 days and the day of flexural testing.
Measurements were not taken between 28 days and the day of flexural testing. As
discussed earlier, the reliability of the draw-in measurements at the day of test are in
question due to the damage of the glass target plate during the movement of the beam in

the yard, hence were not used for the prediction of the transfer length.

150

7.10 Summary and Conclusions

Determining the development length of prestressing strand involves the study of
complex bond phenomena between concrete and steel tendons. As discussed earlier, bond
phenomena depend on various factors like material properties, industry practices, quality
control etc. Hence a definite or accurate relationship of all the parameters cannot be
obtained, but a relative study can be made.

The current ACI-318 equation (Equation7-2) [3] was studied, and in the previous
chapter on transfer length it was found that the ACI equation applied conservatively for
transfer length for all beams and the mixes in the study. The development length studies
showed that the ACI equation was un-conservative for SCC mixes and thus the flexural
bond length component of the ACI equation is underpredicted. The test flexural bond
lengths (Limo were obtained for all test units by calculating the difference of the test
embedment length and the average measured transfer length values. As shown in Table
7-4, the experimental values of flexural bond lengths are longer than that predicted by the
ACI-318 code.

The criteria for defining whether the development length for a particular beam
depends on the determination of the ultimate moment capacity of the critical section. As
discussed earlier, refined models tend to give larger moment capacities and the
development length as per ACI-318 code recommendations may not be met in some
cases. For NCCB, the ACI nominal capacity was reached with the recommended ACI
development length. However, the SCC beams required slightly longer lengths: 3, 17, and
42% longer for the SCC1, SCC2B, and SCC3 test beams, respectively. The development
lengths required to achieve the member capacity according to a strain compatibility

analysis were longer: 30% longer for SCC1 beams and 66% longer for SCC2B and SCC3

151

beams. The NCCB beam achieved the higher nominal capacity with only a 5% increase
in L4.

An interesting observation was that mix designs that had relatively smaller transfer
lengths had larger flexural bond lengths and vice versa. For example SCC1 test units, had
a ans / me’ ratio of 1.02 (Table 6-4), where as it had the least Lfm, /L,,,C1 ratio (Table
7-4) for the SCC mix beams. The different bond transfer mechanisms contributing to the
transfer and flexural bond components respectively seem to nullify each other producing
approximately the same effect on development length for all SCC mixes. This indicates
that the SCC mix proportioning has a distinct effect on the different bond mechanisms

contributing to the development length of prestressing strands.

152

CHAPTER 8 SUMMARY AND CONCLUSIONS

8.1 Summary

Self compacting concrete has become of high interest specially to the precast!
prestressed industry because of the many advantages it offers. At the same time, not
much work has been performed on the structural performance of members built using
SCC. This study examined the bond of 13 mm (0.5 inch.) diameter strand in self
consolidating concrete (SCC) in terms of transfer and development length in
precast/prestressed beams. Three SCC mix designs were strategically selected to bound
the different proportioning methods: (1) a design based on controlled coarse-to-fine
aggregate ratios (2) a design based on the use of chemical admixtures and (3) an
intermediate design. A baseline comparison is made with a conventionally vibrated
concrete mix (NCC). Laboratory scale test beams with a T-cross section were designed in
such a way that the prestressing strand experienced strains close to its guaranteed
ultimate capacity during testing. The cross-section had a top flange width of 381 mm (15
in.) and the flange had a thickness of 76 mm (3 in.). The total depth of the section was
381 mm (15 in.) and the web was 152.4 mm (6 in.) thick. The section had two seven-wire
low-relaxation 1860 MPa (270) Grade steel prestressing strands of 13mm (0.5 in.)
nominal diameter. The strands were placed at a depth of 51 mm (2 in.) from the bottom
of the beam. The effective stress for design was taken as 1100 Mpa (l60ksi). For actual
tests, the effective stress was calculated from the initial stressing loads (as determined
from the strain gage readings) and estimated losses. Also a passive compressive
reinforcement consisting of two 13 mm (#4) bars was provided at the top flange to
control cracking at transfer. Nominal shear reinforcement, U stirrups of 6 mm diameter

(#2 bars) at 305 mm (12 in.) center to center spacing, was provided. Concrete strengths at

153

release were greater than 28 MPa (4000 psi) for all test units. The strand condition was
smooth and non-rusty for all mixes except for NCCB mix which used a strand that may
be classified as slightly rusted.

Transfer length of the prestressing strand was assessed using two methods: (a)
concrete strain profile and (b) draw-in measurements. The concrete strain profile
measurements were taken using the DEMEC strain gage system. Small steel discs were
affixed to the concrete surface on both sides of the beam test unit and measurements were
taken using the mechanical gage before and after the release of prestress. The average
strain was obtained from the changes in the locations of these discs. The transfer length
was obtained using the 95% average maximum strain (AMS) method. The draw-in
method involved the measurement of the strand movement at the end of the beam. A
small U—channel was tightly clamped to the strand at the face of the beam and a glass
target plate was fixed to the concrete face. The relative movement of the strand with
respect to the face of the concrete was measured and the transfer length was calculated.
The draw-in measurements were found to be less time consuming and required less-skill
for the relatively same accuracy of the results. The strand profile method had been widely
used and accepted as it gives the entire strand profile throughout the length of the beam.
The results from both the methods compare favorably.

Strand bond was evaluated by means of simple pull-out tests on the used strands for
the all concrete mixes. Six strands were pulled-out for each mix design, three at the time
of release and the remaining were tested on 7 days. The pull-out tests were designed and
conducted in a similar way to the test developed by Moustafa in 1974 and modified by

Logan in 1997. The bonded length was 457 mm ( 18 in.). A hollow-core hydraulic jack

154

was used to pull the strands out of the block. Instrumentation to measure the movement
of the strand was done at both the ends of the strands. The pull-out load and the strand
movement at both the ends (free and jacking) were recorded.

Development length of the prestressing strand was determined through an iterative
process of flexural tests. The beam units were loaded at various embedment lengths and
the failure mode was recorded and observed. Loading was performed by a servo-
controlled hydraulic actuator. The applied load, deflections at points of loading and at the
resultant location of loading, strand-end slips, top fiber concrete compressive strains were
measured continuously throughout the test.

Fabrication, instrumentation and testing for this test program was carried out at the

Civil Infrastructure laboratory at Michigan State University, East Lansing, USA.

8.2 Conclusions

The following conclusions were drawn from this study:

0 The designed concrete compressive strength at 28 days was 48.3 MPa (7000 psi).
The compressive strength at 28 days achieved by the conventional concrete was
exactly 48.3 MPa (7000) psi. At the same time, the compressive strengths of SCC
mixes were much higher than design strength. The high strength of SCC may be
advantageous, but at the same time, the same amount of confidence and control in
designing the mixes may still not have achieved by SCC mix proportioning.

o The ACI 318/AASHTO code equations for transfer length were found to be
conservative only for transfer length measurements for all the mix designs and

only marginally (2%) under estimated the results for the SCC1 mix.

155

The SCC1 mix had the largest transfer length with a (Lynn, / LMCI) ratio of the
measured transfer length (L,,,,,.,,,,) to the value predicted by the ACI code (LMCI) of
1.02.

The (Luna, / L,-AC,) ratio for NCCB, SCC2A, SCC2B, SCC3 was found to be 0.86,
0.82, 0.90 and 0.91, respectively. The NCCB and SCC2A mixes were stiff and
hence a significant reduction in transfer length was found.

It was found that SCC mixes, had larger transfer length values relative to the NCC
mixes.

Results of transfer length studies supported the concept of bounding the hardened
properties by proper mix selection. The transfer length values for the SCC2 mix
(moderate w/c) was bounded by the SCC1 (high fines) and SCC3 (high
aggregates) mixes.

The high values of transfer length for SCC1 could be attributed to the presence of
high paste content and large amount of admixtures.

Draw-in values were found to increase significantly with time. Draw-in
measurements at approximately 110 — 130 days after release showed an increase
of approximately 50% for NCC mixes and about 80% - 100% for SCC mixes.
Correspondence of the increase in draw—in measurements to increase transfer
lengths with time can be inferred. However, further evaluation is needed in order
to assess other factors influencing the increase in draw-in values such as concrete
shrinkage and creep

The peak pull-out test results at release show that SCC mixes had low pull-out

strengths relative to the NCC mixes.

156

The peak pull-out strengths at release are discussed as follows: The SCC1 (high
fines) mix was found to have the least peak pull-out strength of 74. kN (16.61
kip.). The SCC3 (high aggregates) had the highest peak pull-out strength of all
SCC mixes with a value of 89.53kN (20.12 kip.). The peakpull~out strength of the
SCC2 mix was bounded between SCC1 and SCC3 and found to be 86.29 kN
(19.39 kip)

The development length values of SCC mixes were found to be significantly
greater than the values predicted by the ACI-318/AASHTO recommendations.
Development length test results were studied in two cases: based on nominal
capacities according to the ACI-318 equations and based on capacities obtained
from a refined strain-compatibility analysis.

Conclusions regarding the development lengths required to achieve the member
capacity according to the ACI-318 equations are: for NCCB, the ACI nominal
capacity was reached with the recommended ACI development length. However,
the SCC beams required slightly longer lengths: 3, 17, and 42% longer for SCC1,
SCC2B, and SCC3, respectively.

The development lengths required to achieve the member capacity according to a
strain compatibility analysis were longer: 30% longer for SCC1 and 66% longer
for SCC2B and SCC3. The NCCB beam achieved the higher nominal capacity
with only 5% increase in Ld.

All SCC mixes had longer development length than that predicted by the ACI-318
code. Thus the code was found to be unconservative with respect to the

development length for members built using SCC.

157

The strands used in this research did not meet the pull-out (bond qualification)
requirements as proposed by Logan [25]. Hence, the under-performance of
strands in SCC elements based on predictions of the ACI-318 code cannot be
completely confirmed.

Since the same strand and relatively the same parameters being maintained for all
the SCC mixes, the relative performances of SCC mixes should remain the same.
Thus, inspite of the inconclusive outcome set forth by the possibility of low
quality strand, the relative bond performance of strand in different SCC mix
designs is still valid.

The concept of bounding the hardened bond behavioral properties (in this study,
transfer and development length) by proper mix pr0portions was supported by the
obtained experimental transfer lengths. In the transfer length study, the
experimental values of transfer length for SCC2 mixes were bounded by the
values of the SCC1 and SCC3 mixes.

The end result for development length seems to be fairly uniform for all SCC
mixes; however, the transfer and flexural bond length components were not. This
indicates that SCC mix proportioning has clear and different effects on the bond
mechanisms (hoyer’s effect and mechanical interlock) contributing to the
development length of prestressing strands. This supports the concept that bond
performance of prestressing strand is influenced by the hardened behavioral

properties defined by different mix proportioning.

158

APPENDICES

159

Appendix 1 - Stress — Strain Response — Elastic Modulus Tests

The elastic modulus of the different concrete mixes used in this project was
evaluated in accordance with the ASTM C469 test method. The stress—strain responses
measured for the different mixes are provided in this section

The elastic modulus tests were performed at the release of prestress
(approximately 3 days of age) and at 28 days. Elastic modulus tests at release were not
performed for all mixes due to technical and equipment problems. In this section the
elastic modulus plots at release (approximately 3 days) are followed by the elastic
modulus plots at 28 days. The plots for the various mixes are given in the following
order: SCC1, SCC2A, SCC2B, SCC3 and NCCB for both transfer and 28 days of

CODCI’CtC age.

160

Compressive Strength ( MPa )

Compressive Strain ( micro strains )

 

  
    

Compressive Strength ( psi )

 

 

0 250 500 750 1000 1250 1500 1750 2000 2250 2500

48 ....,....,....,....,....,..s.,-e-.,....,....,.e-.‘7ooo
{6000

l
« 5000

i
:4000
{3000

1
{2000
Ec=Elastic Modulus: 1000

EC: 4,318,802 psi ‘

= 29,7771Mpa Z

AlLLL llAll‘lllleALllJAA ALLA

 

A l L l 1 o
0 250 500 750 1000 1250 1500 1750 2000 2250 2500

Compressive Strain ( micro strains)

Figure Al- 1 Elastic Modulus Test — Transfer - SCC2B — Testl

Compressive Strain ( micro strains)

0 250 500 750 1000 1250 1 500

YTrfYIV" ['Tfil

 

fWTIVVY YTTTfi'

1750 2000 2250 2500

 
  

   

60

   

54 . 8000

42 6000

36 5000

LAAlALA‘lAAJA

{4000
24 ‘

18 .
ézooo
Ec = Elastic Modulus}
Ec=4,027,140 psi 1 1000
= 27,766Mpa 3
1111111111111111111111111111111111111LL1_1111.111 O

0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000

Compressive Strength ( MPa )
8

12

 

Compressive Strain ( micro strains)

Figure Al— 2 Elastic Modulus Test — Transfer — SCCZB — Test2

161

Compressive Strength ( psi )

Compressive Strength ( MPa )

Compressive Strength ( MPa )

Compressive Strain ( micro strains)

 

 

  
   

 

 

0 250 500 750 1000 1250 1500 1750 2000 2250 2500
_ ...,....,+T..,....,r.....fi.,.fi.,.fi.,.efl,.... 7000
48 _ .
42 ~ 6000
t 1
35 L l
i f 5000
t .
30 E I
. - 4000
24 f
i 1 3000
18 f 2
. l 2000
12 _. ‘
6 I Be == Elastic Modulus: 1000
; Ec = 4,653,828 psi
- -.- 32,087 Mpa 1
0 1 1 1 l 1 L 1 1 1 1 1 1 1 1 1 1 4 1 1 1 1 1 1 1 1 1 1 1 l 1 1 1 1 l 1 1 1 1 l 1 1 1 1 o
0 250 500 750 1000 1250 1500 1750 2000 2250

CompressiveStrain ( micro strains )

Figure Al- 3 Elastic Modulus Test — Transfer - SCC2B — Test3

Compressive Strain ( micro strains)

0 250 500 750 1000 1250

1500 1750 2000 2250

 

Y I Y Y Y

YTVTYV‘FYIrY—Ffirf T

42}
36 i
30 f
zti
18 f

12}

 

lllALILLLAAEAALAl

0 250 500 750 1000 1250

L A A A

'rfllfYYlYTIVVV

   
 

Be 3 4,440,085 psi

Be 3 Elastic Modulus:

d
l
-
— 30,613Mpa1
lllAllAAAAEALLA‘

d

AlLuAlAAAA AA‘

 

2500
v. 6500

6000

1500 1750 2000 2250

Compressive Strain ( micro strains)

Figure Al- 4 Elastic Modulus Test — Transfer — SCC3 - Testl

162

Compressive Strength ( psi )

Compressive Strength ( psi )

Compressive Strength ( MPa )

Compressive Strength ( MPa )

Compressive Strain ( micro strains )

 

 

  

 

 

0 250 500 750 1000 1250 1500 1750 2000 2250 2500
43 yvrrrrfvr,,Y,,,Ymvrrr,.Tvfiv,,YHYYTY.T.W,T.,YV‘ 7000
1 l
42 f - 6000
: 1 5000
30 3 i
I 1 4000
24 :- J
C 1 3000
18 _— 1
: i 2000
12: ‘
6 E0 = Elastic Modulus? 1000
Be = 3,470,393 psi
= 25,790 Mpa
o 1 1 l 1 1 1 1 l 1 l 11 LL 1 1 1 1 l 1 111 L 1 1 1 1 l 1 1 1 1 1 1 1 1 1 l 1 1 1 1 o

 

0 250 500 750 1000 1250 1500

Compressive Strain ( micro strains)

1750 2000 2250 2500

Figure Al- 5 Elastic Modulus Test — Transfer — SCC3 — Test2

Compressive Strain ( micro strains)

 

 

 
        

 

o 250 500 750 1000 1250 1500 1750 2000 2250 2500
42:. 44,, ,Y ,..,,,,.,..,.fi,€6000
>- 2
: -Z 5500
36 1 3
: 1 5000
so; 45°°
: i 4000
24L 3500
. 3000
18 5 5 2500
I «f 2000
12 ~ 3
Z 1 1500
5 ' Ec = Elastic Modulus“; 1°00
’ 5° Ec = 4,035,552 psi .3 500
= 27,825 Mpa ;
o 1 1 LA 1 1 l 1 1 1 1 l 1 1 1 1l1 1 1 1 l 1 1 1 1 l 1 1 1 L1 1 1 1 1 l 1 1 1 1 o

   

 

 

  

750

0 250

500 1000 1250 1500

Compressive Strain ( micro strains)

1750 2000 2250 2500

Figure A l - 6 Elastic Modulus Test - Transfer — NCCB — Testl

163

Compressive Strength ( psi )

Compressive Strength ( psi )

Compressive Strength ( MPa )

Compressive Strength ( MPa )

Compressive Strain ( micro strains )

 

  
  

 

 

0 250 500 750 1000 1250 1500 1750 2000 2250‘ 2500
48hrtv.,,...4,rrrw,yr+4,,v4vq,ififir,,rr,,r.,,, r,,, . 7000
42 E j 6000

L 1
36 ; - 5000
30 E j

: 1 4000
24 3

I 1 3000
18 f j

: i 2000
12 : 1

6 : Ec = Elastic Modulusi 1000

: Ec = 4,070,740 psi

1lfiF11 flgflxr§7raqut

 
 

1L+111114LLL11214_ALAA|II

 

LJngLAA

50

 

  

0 250 500 750 1000 1250 1500 1750 2000 2250 2500

Compressive Strain ( micro strains)

Figure Al- 7 Elastic Modulus Test — Transfer — NCCB -— Test2

Compressive Strain ( micro strains)

0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000 3250

 

 

firwrfipnnm ,..... ............ ,..,., ,Hfifif... ‘ 0000
66 :
‘9000
60 ’ :
48 £ 7000
42 {6000
36 {5000
30 a
q 4000
24 g
1 3000
18 :
12 {2000
Ec = Elastic Modulus:
5 A 5° Ec= 4,249,249 psi :1000
= 29,300Mpa ,
o111111111111111111111l1111l111111111111111111 11111111 o

 

0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000 3250

Compressive Strain ( micro strains )

Figure Al- 8 Elastic Modulus Test - 28 Days — SCC1 — Testl

164

Compressive Strength ( psi )

Compressive Strength ( psi )

Compressive Strength ( MPa )

Compressive Strength ( MPa)

Compressive Strain ( micro strains )

 

0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000

mepmwqu, ,rfirqufmwrwfifl,”qfiflwn”Hrs. 10000
66: 1

E -‘ 9000
60 :- I
543 48000
48:— {7000
42;— £8000
36:5 15000
30:» 3

: 1 4000
24:— 3

5 1 3000
18 :- 1
12:- . €2000

: EC Ec = Elastic Modulus:
3 ;. Ec = 5,000,173 psi 1 1000

’ = 34,475 Mpa 3
c 11111111111111111111111111111111111111111l11111141111111

 

 

 

Compressive Strain ( micro strains )

1 1 1 o
0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000

Figure Al- 9 Elastic Modulus Test - 28 Days — SCC1 — Test2

Compressive Strain ( micro strains )

750

 

YT' Y

0 250 500

33§88£§£§88

 

YTjY

750

1000 1250 1500 1750 2000 2250 2500

TTTY Y Y Y

I V Y T TT r Y I' V l V Y Y

 

Ec -.- Elastic Modulus}

Ec = 7,875,771 psi
“=1 1 1591310?.MP“

A1L4111111111

Compressive Strain ( micro strains)

-

4
J
u
q
4
q

.1
.1
1
1
4

 

7000

5000

3000

1000 1 250 1 500 1750 2000 2250

Figure Al- 10 Elastic Modulus Test — 28 Days — SCC2A - Testl

165

Compressive Strength ( psi )

Compressive Strength ( psi )

Compressive Strain ( micro strains)

0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000 3250 350(1)000

 

 

 

 

66 :

60 O €9000
’3 38000
v 48 {7000
5 I
g 42 {6000
e 2
6'5 36 {5000
0 30 I
> 1
a 24 ”000
9 33000
a 18 1
g :

o 12 ‘ ?2000
Ec = Elastic Modulus:

s Ec= 5,,468869psi 31°00
111111111111111L1111J. .1.1111111111111?11113171!71071M.Ea 110

o 1
0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750 3000

Compressive Strain ( micro strains )

Figure Al- 11 Elastic Modulus Test — 28 Days — SCC2A — Test2

Compressive Strain ( micro strains )

0 250 500 750 1000 1250 1500 1750 2000 2250 2500

 

  

 

,nfirfln ”2,. eqfl. ,4 ,2,“. 0000
66 :
49000
60 3
A ;
g 54 38000
V 48 £7000
.5 :
g’ 42 {6000
2 36 :
a; {5000
0 3o 1
> 4
g 24 14000
g 3000
E 18 ;
8 12 {2000
E0 = Elastic Modulus:
6 Ec=5,,155837psi {1000
= 35,548Mpa *

 

111111l1111l1111l114_LJ114_111J_11-LJ_1L110

0 250 500 750 1000 1250 1500 1750 2000 2250 2500

CompressiveStrain ( mlcro strains )

Figure Al- 12 Elastic Modulus Test — 28 Days — SCC2A — Test3

166

Compressive Strength ( psi )

Compressive Strength ( psl )

Compressive Strength ( MPa )

Compressive Strength ( MPa )

Compressive Strain ( micro strains )

2000

l

 

1000 1250 1500 1750

i

2250 2500
1 . , 10000

firYII [Yfiv

§ 9000
i 8000
«37000
€ 6000
§ 5000
4000

«j 3000

  
   
   

~j 2000

Ec = Elastic Modulus:

Ec = 4,556,505 psi 1 1000
= 131,4181Mpa 0

0 250 500 750 1000 1250 1500 1750 2000 2250

 

 

Compressive Strain ( micro strains)

Figure Al- 13 Elastic Modulus Test - 28 Days - SCC3 — Testl

Compressive Strain ( micro strains)

0 250 500 750 1000 1250 1500 1750 2000 2250 2500

 

 

 

54 L ‘ 3000
48 f { 7000
42 E J 6000
36 3 f 5000
30 E 1

C j 4000
24 E :

E . 3000
18 f ,

E i 2000
12 f I

~ E0 = Elastic Modulus:
s : Ec = 4,371,360 psi 3 100°

. = 30,140 Mpa ,

1111L111111111111111111111111111114111111111 o

   

 

0 250 500 750 1000 1250 1500 1750 2000 2250 2500 2750

Compressive Strain ( micro strains )

Figure Al- 14 Elastic Modulus Test — 28 Days — SCC3 —- Test2

1(17

Compressive Strength ( psi )

Compressive Strength ( psi )

Compressive Strength ( MPa )

Compressive Strength ( MPa )

48_

42

36

24

18

12

Compressive Strain ( micro strains )

 

 

  
  

 

 

 

750 1000 1250 1500

Compressive Strain ( micro strains )

0 250 500 750 1000 1250 1500 1750 2000 2250 2509000

3 6000

._ i

i '1 5000

E 3 4000

:‘_ i

; 1

E « 3000

: i

; 7 2000
Ec Ec = Elastic Modulus; 1000

I Be = 3,545,480 psi ‘

: = 24,445 Mpa 1

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 o
0

1750 2000 2250 2500

Figure Al- 15 Elastic Modulus Test — 28 Days — NCCB — Testl

54

48

42

36

30

24

18

12

Compressive Strain ( micro strains )

 

 

 

‘7000

6000

5000

.1000

0 250 500 750 1000 1250 1500 1750 2000 2250 2500
H”,HH,.,..,W,,,HH.. ,,,rfi,...,,..,.,.n,.3000
,
E 0 E6 = Elastic Modulus}
Ec = 4,383,350 psi
1 1141111114_11L11111111111111111111=11113PL122121LMP1811:o
0 250 500 750 1000 1250 1500 1750 2000 2250 2500

Compressive Strain ( micro strains)

Figure Al- [6 Elastic Modulus Test -— 28 Days — NCCB — Test2

168

Compressive Strength ( psi )

Compressive Strength ( psi )

Appendix 2 - Pull Out Test Response

The Pull out test was performed on the same strands used in the beam specimens.
The pull out block and test procedure was similar to that proposed by Moustafa (1974)
(see Chapter 6). A total of six strands were pulled out for each type of mix. Three strands
were pulled out at the time of release of prestress (approximately 3 days of age) and the
remaining three strands were pulled out at 7 days for all mixes.

This appendix contains the pull-out force-slip response of each of the strands of
all the mix types. Firstly, the response of the pull-out test at release for all mixes is given
followed by the response at 7 days. The plots are provided in the following order: SCC1,
SCC2A, SCC2B, SCC3 and NCCB for both ages of concrete (release and 7 days). The
combined average plots for each mix at release (3 days) and at 7 days are given at the
end. Also, a combined plot of the average responses of each mix showing the peak loads

at first strand slip is also given.

169

Max. PullOut Force (kN)

Max. PullOut Force (kN)

Strand Slip (in.)
83.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00 3.258

 

 

 

 

 

.-"f ~‘1'r 1' Y1K
7o ' T 16
60 a 14 a
123
so 3
h
- 103
40 . .5
€ 8 Q
30 I 3
i -‘ 6 a:
I I X
20 -' , _‘ 4 g
Peak Pull Out Force =78.33 kN (17.61 kip)
10 Front Slip vs PullOut Force‘: 2
— Back Slip vs Pullout Force *
o 1 1 1 1 l 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0
0 10 20 30 4O 50 60 70 80

Strand Slip (mm)

Figure A2- 1 Pull Out Test Prestress Release. — SCC1 — Strand #1

 

 

 

 

 

 

Strand Slip (in.)
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
. 16
70 .
I , ‘ I. I
. ‘ - 14

60 ' , ‘ A I

’ f . .‘ ' ’ ' v 4
50 i l . 3
i 1 V
/ - 1o 8
" h
40 ’ If
.1 8 g
30 _ j =
. : 5 5’.

20 L 4
t 4 g

; Peak Pull Out Force =66.32 kN (14.91 kip;

10 - -

: Front Slip vs PullOut Force 1 2
? — Back Slip vs Pullout Force 3
o _1 1 1 1 1 1 1 1 1 1 1_1_ 1 1 1 1 1 4 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 o
0 10 20 30 40 50 60 70 80

Strand Slip (mm)

Figure A2- 2 Pull Out Test at Prestress Release. - SCC1 — Strand #2

170

Strand Slip (in.)

00.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.018

. 1 .. I E
")1 1‘1,- ’ 1" ‘1‘ . ,*. ,' .315
f 514

Max. PullOut Force (kip)

 

 

 

 

 

f 60
£5
.. 12
o
‘5’ 5° *
u, - 10
‘5 40 V .
9 ’ « a
E 30 . ,
- - 6
1’15 A .
Peak Pull Out Force =76.95 kN (17.30 kip):
1° ' — Front Slip vs PullOut Force“: 2
? — Back Slip vs Pullout Force 3
o 111111111111111114111114111111111141111111111111111111111111111111111111111111111111111
0 10 20 30 40 50 60 70 80 90 100110120130140150160170
Strand Slip (mm)

Figure A2- 3 Pull Out Test at Prestress Release. — SCC1 — Strand #3

 

 

Max. PullOut Force (kN)

 

 

 

Strand Slip (in.)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.028

120:....,..... ...,.....fi1....fi,. .......,..,.,sfi.f
E 1 25
110 E .: 24
100 : 22
90 E 20
so 3 18
70 f 16
60 : " 14
50 e: 12
E -: 1o

40 r 5
E 1 8
3°: 6

20 * Peak Pull Out Force =120.37 kN (27.06 kip)?
-: 4

10 Note: Front Displacement LVDT was Front Slip V3 PullOut Force 3
— Back Sli vs Pullout Force 1 2
0 1 unable to record data, hence missing ‘ . p l . . : 0

 

 

0102030405060708090100110120

Strand Slip (mm)
Figure A2- 4 Pull Out Test at Prestress Release — SCC2A — Strand #1

171

Max. PullOut Force (kip)

Max. PullOut Force (kN)
8

Strand Slip (in.)

 

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.518

'1

V V Y V V V

(1)
O

N
O

 

10
unabieto
o 1111111111111111111

'UYY'VY
p

T

 

rec

YIY l I I ‘ I I V. I I'YrTI'YYV
4

 

 

 

j 16
5 14A
.9.-
125
1 o
: o
‘ 6
1 1011
‘ ue-o
; 3
1 8 Q
: 2
16 .
: x
. a
S 4 2
Peak Pull om Force =73.35 kN (16.49 kip):
Note: Front Displacement LVDT was Front 3'19 V3 PullOut Force 1 2
ord data, hence missing — Back Slip vs Pullout Force .
1111111111111111111111111111 111111111111111111111111111111111111111111 o

1111

 

 

0 10 20 30 40 50 60 70 80 90100110120130140150160170180190

Strand Slip (mm)

Figure A2- 5 Pull Out Test at Prestress Release. — SCC2A — Strand #2

Max. PullOut Force (kN.)

 

 

 

Strand Slip (in.)
1200.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
-I 26
100 3 22
90 203‘
5 183
80 E 8
7o 16 5
: 11
50 2: 12g
5 10
4o 3 8 2'6
30 _is g
20 ' Peak Pull om Force =110.98 1111 (24.95 kip); 4
10 Front Slip vs PullOut Force: 2
— Back Slip vs Pullout Force 3
o 11111111111111111111111111111111111111111111111 1111111111111111111111111111111111111 o

 

 

0 10 2O 30 40 50 60 70 80 90 100110120130140150160170

Strand Slip (mm)

Figure A2- 6 Pull Out Test at Prestress Release — SCC2A — Strand #3

172

Strand Slip (in)
00 05 1b 15 2.0 2.5 3.0 35 4.0 4.5 5.0 5.5 6.0 6.5 7'%o

 

 

 

 

 

TTrrrrrT' 1'! "r’TVrT r'71 '71 'vrrrv 1' 'VTTTTT'Ir 'rr' "1
so I - 1a
A I - 16
E 70: a
8 so ’ - 145
h - 0
O : '1 12 2
.“3 50. .2
3 .
3 40 _ I g
(L , 1 3 3
x' 30 E “Z
to ~ 5 x
20 3
Peak Pull Out Force =81.31 kN (18.28 kip): 4
10 — Front Slip vs PullOut Force-3 2
-— Back Slip vs Pullout Force“ ‘

 

 

o 10 20 30 4o 50 60 7o 80 90 100110120130140150160170
Strand Slip (mm)

Figure A2- 7 Pull Out Test at Prestress Release. -— SCCZB - Strand #1

 

 

Strand Slip (in.)

0.0 0.5 1 .0 1 .5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.022

90 ' .‘ ' - "\. ': 20

A 80 \ j 13
2 4 A
5 7o 1 15.9
8 : 5
1 01
3 60 : 14 e
u' I 12 °
5 3 ""
O 50 . 15
= l ‘ 1°C
3 40 =
Q 3
x' ‘. 8 a.
u 30 ’ 2 X.
5 1] £6 a
. 1 2

2° Peak Pull om Force=93.06 kN (20.92 klpH 4

10 Front Slip vs PullOut Force} 2

— Back Slip vs PullOut Force:
0 11111111111111111111111111111111111111111111 11411111141111”! 111111111111111111 o

 

 

 

0 1o 20 30 40 so 60 7o 80 90100110120130140150160170
Strand Slip (mm)

Figure A2- 8 Pull Out Test at Prestress Release. — SCC2B — Strand #2

173

Max. PullOut Force (kN)

Strand Slip (in.)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0

 

90:

Y 1“

N
0

88888

 

 

TT'IIT'VUIVIrIII'VI'VU"rt‘IUr'V'V'VTVV'VI"rTWTTITY—YTTTV"l'Iij—rTVT'.

Peak Pull Out Force = 84.42 kN (18.98 kip) .

I
d

 

1o — Front Slip vs PullOut Force
Back Slip vs PullOut Force 1

 

 

0 111111111111111111111111111111111111111111 1* ‘‘‘‘‘‘‘‘‘‘‘‘

4

 

 

o 10 20 30 4o 50 60 7o 80 90100110120130140150160170
Strand Slip (mm)

20

Max. PullOut Force (kip)

Figure A2- 9 Pull Out Test at Prestress Release. -— SCC2B - Strand #3

Max. PullOut Force (kN)

Strand Slip (in. )

800.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.018

 

TVV'ITIVVT‘I’V'TT VVVVVVV II'1"[ IIIIIIIIIIII IWfITIVIUTI'V'V'UI‘II'Yr"
p

q
C
I 1 I I ‘

 
 

YV'II [III '1‘

II 111'

 

N
O

Peak Pull Out Force = 7250 m (16.30 kip)?

d

111111

09
Max. PullOut Force (kip)

111111

 

.1
O

— Front Slip vs PullOut Force

 

 

111111111111111111111111111111 1 ._1_._111_._1_1

--- Back Slip vs PullOut Force 3

111

 

 

0
0 10 20 30 40 50 60 70 80 90100110120130140150160170

Strand Slip (mm)

b

'0

Figure A2- 10 Pull Out Test at Prestress Release. — SCC3 — Strand #1

174

Max. PullOut Force (kN)

Max. PullOut Force (kN)

Strand Slip (in. )
0.0 0.51015 2.0 2.5 ‘30. 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0

 

Y—Y—YrrTTIY'YT—rrrYTrY ITYTTII III 'I IIIIj'II II II III II1'III'I

  

 

 

 

 

90? {20
so} - 18
70 Peak Pull Out Foroe=90.07kN(20.25klp)‘; 16g
I x
j 14"
6° E §
1120
so > . u.
E 1103
40_ : =
.1 3
30 :8 a
~26 ii
.1 2. 2
10 — Front Slip vs PullOut Force: 2
Back Slip vs PullOut Force 1
4111111111111.111111111111111111-111111111111111 1l1111111L1111111111111111111111111 1111

 

 

o 10 20 30 40 so 60 7o 80 90100110120130140150160170
Strand Slip (mm)

Figure A2- 11 Pull Out Test at Prestress Release. — SCC3 — Strand #2

 

 

 

 

 

Strand Slip (in.)
0.0 0.5 1 .0 1 .5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
110 :
1 24
100 A/ ' V‘ E 22
90 I“ 20
5 Ti
70 1 16 8
= a
60 E 1“In.
-: 125
so = 9
-‘ 10's
40 3 a.
3. 3°:
1 a. 2
20 i Peak Pull Out Force: 105.87 N (23.8 klp)? 4
10 : Front Slip vs PullOut Force: : 2
— Back Slip vs mPullOut Force 3
o 1.111....11...1....1..111....1....1.1111..I.. 1o
0 10 20 30 40 50 60 70 80 90 100110120130140150160170

Strand Slip (mm)

Figure A2- 12 Pull Out Test at Prestress Release. — SCC3 - Strand #3

175

Max. PullOut Force (kN)

Max. PullOut Force (kN)

888%83

Strand Slip (in.)
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0

 

 

 

 

 

 

 

 

 

130 I III 'T—I—T‘r—I—Ij I I I I II I I I I TI I I I I l I I I I I I I I I I I IfiI] I I I frII I I I I ITI: 30
’ - 23
120 ' ‘ 26
110 . j 24
100 : '2 22
90 i 20
so _ 5 18
70 f - 16
50 : -‘ 12
40 h 3 1°
a
30 L- _2 6
20 5 Peak Pull Out Force: 127.49 kN (28.66 kip);
E i 4
10 :_ Front Slip vs PullOut Force:
4 Back Slip vs Pullout Force 3 2
O#4111111111111.111L1111.1111111111111....1 ........................ 1 o
0

1o 20 30 4o 50 60 7o 80 90 100110120 130140150
Strand Slip (mm)

Figure A2- 13 Pull Out Test at Prestress Release. — NCCB — Strand #1

Strand Slip (in)

1600-01” 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5

150
140
130
120
110
100

 

80

 

 

 

— Front Slip vs PullOut Force}
10 — Back Slip vs Pullout Force 3

0 llllllILIIllLAlLLLLlLLLLiJLllllLJlllllllllll MJWJAIAALLWIAAIA"AAAA‘ALAJLA:

o 10 2o 30 4o 50 60 7o 80 90 100110120130140150160
Strand Slip (mm)

 

. 9

Peak Pull Out Force = 145.95 W (32.31 kIpfi s
3

o

 

 

Figure A2- 14 Pull Out Test at Prestress Release. — NCCB -— Strand #2

176

Max. PullOut Force (kip)

Max. PullOut Force (kip)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.033
140
130 .. “:30
120 “5‘.“ 527
g 110 4‘ €24
I 100 E
2 so, 321
o i
ll- 80 j 18
5 70 i
9 “ «:15
E, 60 31
. 50 'j 2
3:5 ‘9
2 4°' .
3° 5 Peak Pull om Force -.- 132.25 kN (29.73 kIpyj 5
20 .
Front Slip vs PullOut Force .‘ 3
1° — Back Slip vs Pullout Force 4
o 1111111111111l1111111111111111111l1111l1111L11 1111111111111111111111;;1111111111111111 o
0 1o 20 30 40 so so 70 80 90100110120130140150160170 .

Strand Slip (in.)

 

 

 

 

 

 

Strand Slip (mm)

Max. PullOut Force (kip)

Figure A2- 15 Pull Out Test at Prestress Release. — NCCB - Strand #3

Strand Slip (in.)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5

80

N
O

O
C

Max. PullOut Force (kN)
00
O

.s
O

0

8

l’_-

Peak Pull Out Force =80.20 RM (18.03 kip

 

7.0

1111

d
)1
1

 

Note: Front Displacement LVDT was

 

L1I11411111 1111L111Ll11LL1111

— Front Slip vs PullOut Force 5
unable to record data, hence missing — 330" Slip V8 PUlIOUt Force 1

‘
11111111111ll11111111l11111111111111114

 

 

010 20 3O 40 50 60 70 80 90100110120130140150160170
Strand Slip (mm)

Figure A2- 16 Pull Out Test at 7 Days — SCC1 — Strand #4

177

18

Max. PullOut Force (kip)

0.

8

8

8

Max. PullOut Force (kN)
8

8

10

Strand Slip (in.)

0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5

IIII'IIIIIIIII'IIIIITIIIIIIIIIITII'IIIITIjII'ITrI[IIII‘IIIIrIIII
d

1 Peak PuII om Force =74.so kN (16.75 kIp):

 

 

$1 unable to record data, hence missing — 330k 8le V8 PUHOIR Force 1
0 L1 1 1

— Front Slip vs PullOut Force :

vv'v

Note: Front Displacement LVDT was

 

 

11l11 111111 I 111111111111111111111111 llLLLll‘lAlllAAlALAAl 11111111 l1

 

0110 20 30 40 50 60 70 80 90100110120130140150160

Strand Slip (mm)

Figure A2- 17 Pull Out Test at 7 Days - SCC1 — Strand #5

Strand Slip (in. )

 

 

 

 

 

.a . . .m. . . .
Max. PullOut Force (kip)

h

18

16

on 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 10
sci
2" 5°.
5 .
0 I
5 40.
u.
w p
:3 .
Q 30
'5
a.
1’: 2° .
2
10 . Peak Pull Out Force =56.85kN(12.78 kip)‘
' —— Front Slip vs PullOut Force:
+ — Back Slip vs Pullout Force .
o 11111111111L111111111111111111141LL111111111l11111111.1111111111111111.1111111111111111

 

 

o 10 20 30 40 50 60 7o 80 so 100110120130140150160170
Strand Slip (mm)

Figure A2- 18 Pull Out Test at 7 Days — SCC1 — Strand #6

178

Max. PullOut Force (kip)

Strand Slip (in.)
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 45 4.5 5.0 5.5 6.0 6.5 7.0

 

 

 

 

 

170 , ’ 1-. - - 3‘ 39
160 .v An!" €36
150 333
1140 3
g 130 330g
3 110 ~124§
LL 100 j o
8" 9° =2‘.‘t
= 80 E 183
3 1 _
a, 70 ‘ 153
I,"- 60 d 120:
40 .- 5 9 E
30 1 Peak Pull Out Force: 167.65kN(37.69 kip! 6
2° ; Front Slip vs PullOut Force} 3
10 — Back Slip vs Pullout Force 3
“1.1...111...;.1.11.11LL111111111L1111111111.1..I1...1.11.11.1111...11.111111111111111: o
O 10 20 30 40 50 60 70 80 90 100110120130140150160170
Strand Slip (mm)

Figure A2- 19 Pull Out Test at 7 Days — SCC2A — Strand #4

Strand Slip (in.)
0.0 0.5 1.0 1.5 2.0 2.5 35 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7530

130
120
110
100

 

~j1o

Max. PullOut Force (kN)
*4
O

 

8

Peak Pull om Force =131.oo kN (29.45 up)“; 6
1 4

— Front Slip vs PullOut Force 5 2

— Back Slip vs Pullout Force 5

d
o 11.111111111111111111111111111111111111J11111111 I11111111111111111111111.111111111111111 o

o 10 20 30 4o 50 60 70 so so 100110120130140150160170
Strand Slip (mm)

 

 

 

Figure A2- 20 Pull Out Test at 7 Days — SCC2A — Strand #5

179

Max. PullOut Force (kN)

Max. PullOut Force (kN)

Strand Slip (in)
05 05 15 1.5 2.0 2.5 3.0 3.5 4.0 45 5.0 5.5 6.0 6.5 7.%8

 

 

 

 

 

 

120 vwwrivrva' "I rrvl' I—fTIV WIY—t—rvyr ITVI 'Wl rvvlvrv IV"[ E
*5 26
110 g 24
100 3 22
90 “i 20.3
80 1 18 In
5 2
70 _2 168
60 14g
40 1 100.
58 fi
30 ‘1 E 5 E
20 Peak Pull Out Force: 119.84 kN(26.94 kip);
1 4
10 ' Front Slip vs PullOut Force 5 2
Back Slip vs Pullout Force '
O

 

0 lllllLJAllLLLLll|ALlAlILllLLJLLALJLLLLJILIIALLAI 11111111111111111141111111111111111111.

 

0 10 20 30 40 50 60 70 80 90100110120130140150160170

Strand Slip (mm)

Figure A2- 21 Pull Out Test at 7 Days - SCC2A — Strand #6

Strand Slip (in. )
0.00 0.25 0.50 0.75 150 1.25 150 1.75 2.00 2.25 2.50 275 ”35%

 

 

 

 

 

 

 

130 4. ,.... H...4.....r.,-rfir.....,....,....,.”were,. :
128
120 3
126
110 g 24
100 €22?-
90 120£
so 516§
4 O
70 1 16;
50 1123
. £10".
40 :8 n
30 3 E
20 Peak Pull Out Force=121.35 kN (27.28 kip)? 6
1 4
10 Front Slip vs PullOut Force ‘
-— Back Slip vs PullOut Force: 2
0

o WLJALALLIll11AAllilllll‘llllLlllLLlAll 1111L1L1u11111111111 111111
0 5 1o 15 20 25 30 35 4o 45 50 55 60 65 70 75
Strand Slip (mm)

Figure A2- 22 Pull Out Test at 7 Days — SCCZB — Strand #4

180

Max. PullOut Force (kN)

Max. PullOut Force (kN)

Strand Slip (in.)
3.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00

 

 

 

 

 

 

 

 

2 .4..,.........,.......-.,....,...-,....,....,..-......,.-.-‘
“:26
110 E24
100 522
9° 120
80 118
70 116
50 :12
{10
4o 1
1:8
so. _. 6
20 Peak Pull Out Force=111.34kN(25.03 klp)i
I4
10 Front Slip vs PullOut Forcej 2
Back Slip vs PullOut Force;
0 11111111111111111111111111111111111111l1 11111111111111111111111111111111l o

o 51015202530354045505560657075
Strand Slip (mm)

Figure A2- 23 Pull Out Test at 7 Days — SCC2B — Strand #5

 

 

 

 

 

Strand Slip (in.)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 45 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8524
10° 1 22

.0 N " :
-j 20
so I '5 5 5 18
70 E 16
60 1 14
1 12

50 I
5 1o

40 3
': 8

3° . .

I _..
20- Peak Pull Out Force=96.08kN(21.60 kIp)_£ 4
10 Front Slip vs PullOut Force 2

— Back Slip vs PullOut Force 3
o 111111111111114114441111L1111111111111111111]11414111 .1u11111111..111....1..1.1.u111...111“I: o
o 10 20 30 4o 50 60 7o 80 90100110120130140150160170180190200

Strand Slip (mm)

Figure A2- 24 Pull Out Test at 7 Days — SCCZB — Strand #6

181

Max. PullOut Force (kN)

Max. PullOut Force (kN)

Strand Slip (in.)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5

130
120
110
100

388

88888

0

Peak Pull Out Force = 127.84 kN (28.74 kip

N
O

)‘S

 

— Back Slip vs PullOut Force”?

1111111111111Ll411LL41MLJL.11111111111111 111411 11111111H1111111111111H4111 LJJLAA l1

 

— Front Slip vs PullOut Force: 3

 

 

0 10 20 30 4O 50 60 70 80 90100110120130140150160170

Strand Slip (mm)
Figure A2- 25 Pull Out Test at 7 Days — SCC3 - Strand #4

Strand Slip (in.)

8138838

3: .5
Max. PullOut Force (kip)

.1
GO

_..-3
ON

ounuama

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0

140
130
120

.L

ss's‘s

8888

~-..| \

10

t, 1111
0 10 20 30 40 50 60 70 80 90100110120130140150160170180190200

C‘-

‘

Peak Pull Out Force = 141.68 kN (31.85 up):

1
d
d
d
-
-

-l

d
d
d

I
q
q

 

— Back Slip vs PullOut Force]?

111111111111111l111111111111L111111JJ111111111111111111111111111111 H111111111 J1Ul

 

— Front Slip vs PullOut Force: -2

 

 

Strand Slip (mm)

Figure A2- 26 Pull Out Test at 7 Days — SCC3 - Strand #5

182

8182863853

Max. PullOut Force (kip)

dab—Id
ONbOG

ONhOG

Max. PullOut Force (kN)

Max. PullOut Force (kN)

Strand Slip (in.)

 

130
120
110

§

Peak Pull Out Force = 128.02 RM (28-78 ”PT:

1 l 1 1 1 1 1 1,4 1,1 l 1 1 1 1 1 1 1 1 1 1 1 1

 

 

 

 

1 1 1 1 1 1 1

Front Slip vs PullOut Force -
—- Back Slip vs PullOut Force ‘5

141‘

 

 

0.0 0.51.01.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5

120

110:
100;
.;
003
70;
60:
so;
40:

30
20
10

0

0 25 50 75 100 125

150 175 200 225 250
Strand Slip (mm)

Figure A2- 27 Pull Out Test at 7 Days - SCC3 — Strand #6

Strand Slip (in)

 

 

Peak Pull om Force = 106.00 m (23.33 kip)§
— Front Slip vs PullOut Force: '

1111L11111111L1111111111111111111111111111111111

 

 

 

— Back Slip vs Pullout Force

111111111111111111111111111111111_1_11l

11:

1

 

 

010 20 30 4o 50 60 70 so 90 100110120130140150160170
Strand Slip (mm)

Figure A2- 28 Pull Out Test at 7 Days — NCCB - Strand #4

183

ONbOO

3363833888

_1-1
NO

3
Max. PullOut Force (kip)

Max. PullOut Force (kip)

Max. PullOut Force (kN)

Max. PullOut Force (kN)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 70

Strand Slip (in)

 

 

 

 

 

 

 

 

 

 

140 f 32
130 E 3°
120 E 28
I 26
110 : : 24
100 E -; 22
90 ’ -: 20
so 18
7o :' 1s
60 14
so ' 3 12
4o - 1°

_ f 3

30 Peak Pull Out Force: 143.37 kN (32.23 kip); 5
20 . E 4

Front Slip vs PullOut Force

1° — Back Slip vs Pullout Force ‘3 2

0 1111111111..1.1111.1111111111111.1111111..1.11.11111.11...11111.1..1.11.1.l...Jl...1l...1 o
o 10 20 3o 40 50 60 7o 80 90 100110 120130 140 150160 170
Strand Slip (mm)
Figure A2- 29 Pull Out Test at 7 Days — NCCB — Strand #5
Strand Slip (in.)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
160 . 36
150 I

. 1 33
13° \ 3°
120 *s \ -I 27
110 / _3 24
100 3

90 I j 21
so _. ' E 18
7o ' 5 15
so .3 .
50 - 12
40 5 9
30 Peak Pull Out Force: 143. 37 RM (32. 23 rap): 6
20 Front Slip vs PullOut Force_‘ 3
1° — Back Slip vs Pullout Force 3
07....1....1....1..-L1...444. .1....1....1....1.. ...1-. ..1 ... LL. ..1.. .1... ...1-.-1- ..1. 'l 0

 

 

 

0 10 20 30 40 50 60 70 80 90100110120130140150160170

Strand Slip (mm)

Figure A2- 30 Pull Out Test at 7 Days - NCCB — Strand #6

184

Max. PullOut Force (kip)

Max. PullOut Force (kip)

Strand Slip [front] (mm)
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.032

 

 

 

150% 30

1409 _: 28

130E 26
-120 ...
z 3 24 a.
£110 22 :2
o 100 20 o
0 90 5 O
u. -' 16 “-
= . - 14 =
o 70 o
= . 12 =
:l 3
0. 60 1: 10 O.
x 50' E x
g 0 o NCCB 3 8 a:

4 I 5001 a 5

30 ‘ SCC2A .: 4

20 o sccza 2

10 + soc: 0

o

0 10 20 30 40 50 60 70 80 90 100110120130140150160170
Strand Slip [front] (mm)

Figure A2- 3] Combined Average Pull-out Test response — At 3 Days

Strand Slip [front] (in.)
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.036

160

150 34

140 j 32

.. _..— €22
'3‘ €33 ' > {26 E
. -: 24 v
§ 100 €221 §
.2 so ,1 21° .2
*5 so I f 18 3
g 70 / l‘ -2 16 Q
a so / -. 14 E
x 50 / 12 x
g 40 / o NCCB .210 g

g I SCC1 .: 8

30; A scczA ' s

20 I o sccza 4

10 + see: ‘ 2

o 5 o

 

0 10 20 30 40 50 60 70 80 90 100110120130140150160170
Strand Slip [front] (in.)

Figure A2- 32 Combined Average Pull-out Test response — At 7 Days

185

Strand Slip [front] (in.)

 

 
   

 

 

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
E1 1 r 1 1 1 1 1 1 l 1 1 1 I 1 1 W 1 1 1 1 1 1 1 f1 1 1 1 fT'r T 1 1 1 1 1 I 1 1 1 1 1 1 1 1 1 ‘ 20
so 3 i 18
A70 1- j 16
z : : a
'5 - 2 143
- ‘ 0
2 2 * 0
° 50 ’ 1 12 3
u. : :
+- ~ ‘ LL
3 : j 10 *5
(=3 40 : : o
3 ’ j =
a. ’ :8 a
x 30 I x
g o NCCB 1 6 a
l 1 E
20 l ' SCC1 1
g A seen 3 4
0 some :
10 - 2
+ SCC3 ;
o ‘ o

 

0 2 4 6 81012141618202224
Strand Slip [front] (in.)

Figure A2- 33 Combined Average First Slip - Pull-out Test Response — At 3 Days

Strand Slip [front] (in.)
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

VVYI'VYYY'VVVTIY YYIfjtroYlIIVITI‘I'VT'VIVI'IUI‘Jzz

 

90:

8

so} 516

r
.5
$

70?

...s
b

so}
so}

40'

to
§
1

 

Max PullOut Force (RN)

8 8
+ O b I o\
2
§ 8
-1 u:
.. $5“; ; 3
Max PullOut Force (kip)

 

 

m
D O
- O
1- 00
2441 LLL
'0

Strand Slip [front] (in.)
Figure A2- 34 Combined Average First Slip - Pull-out Test Response - At 7 Days

186

Appendix 3 - Transfer Length - Concrete Strain Profiles

The transfer length measurement was done using two methods: a) concrete strain
profiles (95% average maximum strain method) and b) draw-in measurements. This
section includes the plots for concrete strain profiles for each test specimen. The concrete
strain profiles were measured on both the sides of the test specimen. Two trials of
readings were taken on each side. A total of 8 plots were obtained for each test specimen.
Only the average of the eight plots (each test specimen) are provided here. Two
specimens per mix type were cast. The average of the concrete strain profiles for both of
the beams cast per mix (average of 16 profiles, each mix type) were given in Chapter 6.
The concrete strain profiles for each beam (average of 8) for various mix types are given

in this section in the following order: SCC1, SCC2A, SCC2B, SCC3 and NCCB.

187

Concrete Strains (mm/mm)

Concrete Strains (mm/mm)

Distance from end of the beam (in.)

 

 

 

      

 

  

  

  

 
 

 

 

 

 

  
 

 

     

 
   

 

 

0 5 1o 15 20 25 30 35 40 45 50 55 60
000050 .. ..... ,.fi.,..-.,....,..... -wfi- 1-552 ....‘000050
0.00045 '- L1 = 762'" mm ‘ 0.00045

. (30.0 in_.) 1
o.mo40 >P- _1- r - 1— v - r - — -- _ - - - - _ _ _ - -_: 0.00040
0.00035 3 95% AMS +2 0.00035
0.00030 E 5 0.00030
0.00025 3 € 0.00025
0.00020 I 5 0.00020
0.00015 I -j 0.00015
0.00010 I -j 0.00010
0.00005 -5 0.00005
0.wooo_....1..-.1. .1....1. .1...;1....1..441....1....1‘0.W

0 150 300 450 600 750 900 1050 1200 1350 1500

Distance from end of the beam (mm)
Figure A3 — 1 Concrete Strain Profile — SCC1 — Beaml
Distance from end of the beam (in.)

0 5 10 15 20 25 30 35 40 45 50 55 60
0.00050 (WTTVTfirwnfin..."...m..." 22222 .. 0.00050
0.00045 1 L1 = 527" mm 2' -‘ 0.00045

_ (24.7 in.) €
0.00040 ; . 0.00040

Lb 1 - -1-1-1-1- ‘

0.00035 3 j 0.00035

E 95% AMS 3
0.00030 _ 1 0.00030
0.00025 : 5 0.00025

i 1
0.00020 I -j 0.00020
0.00015 I —j 0.00015
0.00010 . -j 0.00010
0.00005 -j 0.00005
0.00000-‘ .1.L.41.4..1....1....1.. .1....1....1....1....1‘0-ooooo

o 150 300 450 600 750 900 1050 1200 1350 1500

Distance from end of the beam (mm)

Figure A3 - 2 Concrete Strain Profile - SCC1 - Beam2

188

Concrete Strains (in.lin.)

Concrete Strains (in.lin.)

Concrete Strains (mm/mm)

Distance from end of the beam (in.)
0510152025303540455055

 

      
  

 

  

 

  
  

0.00055 'rr'I‘rTTYVTVVU'r'rr*Tr"T" TTTTY‘T'*"ITTV‘T'V'YI""P‘"'. 0.00055
0.00050 L‘ =.711'2 mm
2 . in.
0.00045 ( 8 O )
0.00040 2
o 2
0.00035 95 /° AMS 1 0.00035
1
0.00030 é 0.00030
0.00025 E 0.00025
1
0.00020 1 0.00020
0.00015 - 0.00015
0.00010 5 0.00010
0.00005 ~j 0.00005
o.m ‘1 1 A A l 1 .I L l A .L n J 1 L A A l 1 L 1 L l 1 A A 1 1 A 4 1. L l 1 A 14 I 1 1 1 1 l A A 1 l l o.m

 

 

0 150 300 450 600 750 900 1050 1200 1350 1500

Distance from end of the beam (mm)

Figure A3 - 3 Concrete Strain Profile - SCC2A — Beaml

Distance from end of the beam (in.)
051015202530354045505500

 

000050 ......... ,.. .. ..-...-.,....r..-.........-.. ...,....»0,00050
: L,=299.7mm (11.8 in.)
0.00045- 0.00045
0.000403— 0.00040
0.000353~ 0.00035
...- _ _. ._. ._,_._. _. - ._._._, ,_, . _ ._
0.000303» 0.00030
0.00025 : 95% AMS 0.00025
0.000203» 0.00020
0.00015; 0.00015
0.00010? 0.00010
0.00005é 0.00005
0.000001“ --1.-- 0.00000

 

 

0 150 300 450 600 750 900 1050 1200 1350 1500

Distance from end of the beam (mm)

Figure A3 - 4 Concrete Strain Profile — SCC2A — Beam2

189

Concrete Strains (in.lin.)

Concrete Strains (in.lin.)

Concrete Strains (mm/mm)

Concrete Strains (mm/mm)

0.00050 _

0.00045

0.00040

0.00035

0.00030 ,

0.00025

0.00020

0.00015

0.00010

0.00005

0.00000—‘ ‘ ‘

0.00050

0.00045 j

0.00040

0.00035

0.00030

0.00025

0.00020

0.0001 5

0.00010 ’

0.00005

0.00000—* ‘

Distance from end of the beam (in.)
10 15 20 25 30 35 40 45

l

50 60
0.00050

3 L1 = 810.3 mm 0.00045

(31.9 in.)

 

0.00040

0.00035

3 95% AMS 0.00030
3 0.00025
’— 0.00020

.. 0.00015

Concrete Strains (in.lin.)

 

0.00010

0.00005

lLllLl‘

1350

ALILLLL‘AA‘

1050 1200

11+4111111115

750 900

LIALLAIJLAIILLAA

1 50 300 450 600

0.00000

0 15m

Distance from end of the beam (mm)

Figure A3 - 5 Concrete Strain Profile —- SCCZB — Beaml

Distance from end of the beam (in.)
15202530354045

'YY'rfi'rmv‘U ""

60
0.00050

50

V'YVY

10

Y'YY'Y‘YVI

55

7" TV

 

YVI'YV

L, = 863.6 mm

, (34.0in.)
,_ _ _.- _ _._._ _._._ _,

P . ‘ i . ‘ 0.00040
L 95%» AMS

L
7

0.00045

0.00035

E

Concrete Strains (in.lin.)

» 0.00025
3 0.00020
3 0.00015

0.00010

 

0.00005

LJLJJJI4LJLJLLLLJALAA1111

750 900 1050 1200 1350

Alljlllllll

450 600

4114111111

150 300

‘4 0.00000

0 1500

Distance from end of the beam (mm)

Figure A3 - 6 Concrete Strain Profile — SCCZB — Beam2

190

Distance from end of the beam (in.)

 

 

 

 

 

   

 

 

 

 

 
 
   

 

 

0 5 10 15 20 25 30 35 40 45 50 55 60
000050.--. , H,.........,.........,.”...........,....,-..-,-.fi»0.00050
.
0.00045 3 0.00045
E 0.00040 3 L1= 759-6 mm \ 0.00040
E 0.00035 (30.3 '"') 0.00035
g r I - 1 - 1 — — 1- - v - -— 2 — _ - - - - a - o — - - v -
a, 0.00030 7 0.00030
g g 95% AMS
3 0.00025 ; 0.00025
01 ;
.3 0.00020 3 0.00020
1. I
g 0.00015 3 0.00015
0 E
0.00010 _- 0.00010
0.00005 ' 0.00005
o.W-A A A A l A A A A j A A A A l A A A A 1 A A A j l A A A A l A A A A l A A A A l A A A A l A A A A l o.m
o 150 300 450 600 750 900 1050 1200 1350 1500
Distance from end of the beam (mm)
Figure A3 - 7 Concrete Strain Profile — SCC3 — Beaml
Distance from end of the beam (in.)
0 5 10 15 20 25 30 35 40 45 50 55 60
0.00050 V...,....15...,....,....T 1.......,.........,............52 000050
E L. = 692.2 mm 3
0.00045 , > 0.00045
A (27.3 in.) :
50.00040 9— —=—.—.—.———.— 20.00040
E : :
E 0.00035 : 95% AMS 3 0.00035
‘5 0.00030 ' «j 0.00030
C 1 .
g 0.00025 I € 0.00025
(0 : i
.3 0.00020 ’ 3 0.00020
0. I 4
g 0.00015 1 0.00015
0 2
0.00010 1 0.00010
0.00005 «i 0.00005
olooooo_....1....1....1.A..1....1....1....1....1....1...;1‘0.00000
0 150 300 450 600 750 900 1050 1200 1350 1500

Distance from end of the beam (mm)

Figure A3 - 8 Concrete Strain Profile — SCC3 — Beam2

191

Concrete Strains (in.lin.)

A
0

Concrete Strains (infin

Concrete Strains (mm/mm)

Concrete Strains (mm/mm)

Distance from end of the beam (in.)

0 5 10 15 20 25 30 35 40 45 50 55 60
0.00050 555-.--..,....,....,....,...-.....,....,....,....,....,....‘0,00050

L. = 500.4 mm _] 5 0.00045

1 0.00040

 

0.00045

 

0.00040 E

1 0.00035

  

{—
0.00035 .1
r 95% AMS

  

0.00030 ' 5 0.00030

 

0.00025 0.00025
0.00020 0.00020
0.00015 0.00015
0.00010 0.00010
0.00005 0.00005

 

 
  

 

AAAAIAAAAAAA_AA1A44AJLAAAIAAA1lALAAAAAAALAAAAJAAAA1:0.W
0 150 300 450 600 750 900 105012001350 1500

Distance from end of the beam (mm)

Figure A3 - 9 Concrete Strain Profile — NCCB — Beaml

Distance from end of the beam (in.)
0 51015202530354045505560
;

 

0.00050

"""""""""7'

0.00045 1,, = 508.0 mm _|

 

     

 

 

 

E (20.0 in.)
0.00040: 0.00040
L -r— _.—,—._. - .
0.00035: .0.00035
0.00030 I 95% AMS 2 0.00030
0.00025, {0.00025
0.00020: {0.00020
0.00015: 90.00015
0.00010: {0.00010
I 1
0.00005 : 30.00005
: 1
o.mw' .1... 1....1....1....1....1....1..A.L....1....A‘0.om

 

0 150 300 450 600 750 900 1050 1200 1350 1500

Distance from end of the beam (mm)

Figure A3 - 10 Concrete Strain Profile — NCCB — Beam2

192

Concrete Strains (inJin.)

Concrete Strains (inJin.)

Appendix 4 - Development Test Response

Development length for the prestressing strands was measured by performing
flexural tests on the beam units. For each test specimen two development length tests,
(one per beam end) were obtained. Thus, a total of four development length tests were
performed for each concrete mix type. This section provides the moment-displacement
response at the critical section obtained from these development length tests. The
nominal moment (Mn) that can be developed in the cross section was calculated using the
computer program Response 2000 [15], which is a research-oriented program for the
analysis of concrete sections with refined concrete and steel constitutive models. In the
following plots, this nominal moment is shown as a horizontal line. If the test response
intersects with this line, then the moment capacity for that particular embedment length
was achieved and the embedment length was considered to be a sufficient development
length for that beam unit.

The development length test responses are provided in the following order of mix
types: SCC1, SCC2A, SCC2B, SCC3, NCCA, and NCCB. For each mix type, the tests
responses are provided in the following order: Unitl-EndA, Unit-l-EndB, UnitZ-EndA

and Unit2-End2.

193

 

Displacement at the section (in.)

 

 

 

0 1 2 3 4 5 6 7 8 9 10
120 :;L;;;‘_;'_'.‘_;Z.‘_'_.‘_'_‘_'_‘_‘.L;_‘_';‘_'.'_"_'_‘::T_‘_’;‘:,; 9°
Mn: 117 kN-m (86.3 k-ft) (
110 r80
100 SlipOnset ;
:70
€90 .
z' 80 5°
i
E70 _50
26° .
o _40
25° :
40 ’30
30 20
20 D
10
10 I
L
0 ,.-fi,...-,....,....,.........,....,.........,-....‘0

 

 

 

0 30 60 90120 150 180 210 240 270300
Displacement at the section (mm)

Figure A4 - 1 Moment vs. Displacement- SCC1-l-A - I...: = 72.38 in

Displacement at the section (in.)
0 1 2 3 4 5 6 7 8 9 10 11 12

100

Moment (KN-m)
3' 8 8

88888

 

 

TVIVVVV‘ 7"

 

.0
O
T I
..A
O

 

 

 

'7" I

 

 

O

oI'I‘lrYTlT'VYTII‘fo’TTY']'TV I'YTT‘YVVUI'UTIU‘YVU'I

0 30 60 90 120 150 180 210 240 270
Displacement at the section (mm)

§

Figure A4 - 2 Moment vs. Displacement— SCC1-l-B - Le = 137.75 in

194

Moment (kip-ft)

Moment (kip-ft)

Displacement at the section (in.)

 

     
 
   

 

 

 

 

9 1, “:1 1 1 € :1 z 1
120 023127.1101042655-0). _________________ "-2? 9°
110 E80
100 E

Ego 17°

32‘ NoSlipOccured E50

: Mnnotachieved;

570 350

5111 E...
so E
40 {3°
30 E20
20 I
10 E10

0 fi-fifi...,.........,..............j....,...-...... 0

 

 

0 30 60 90120 150 180 210 240 270300
Displacement at the section (mm)

Figure A4 - 3 Moment vs. Displacement- SCC1-2-A - L, =122.00 in

Displacement at the section (in.)
0 1 2 3 4 5 6 7 8 9 10 11 12

 

b

120

Mn = 1 17.27 kN-m (86.5 1..-11)

No Slip Occured
Mn achieved

5
ss'

 

.5
8 8 8
.....1rnv 1
N
O

Ir 'V‘T‘I'

Moment (KN-m)
a s 5' s s

 

'VIYYVVr'j‘I

 

 

l
.L
O

 

 

 

O

o [YrfiVIIY'V' ' 'T' I'Ifivt' "Vl'v " 'TYfiY—rr‘ Ti'TI‘I'T‘IYf'T

0 30 60 90 120 150 180 210 240 270 300
Displacement at the section (mm)

Figure A4 - 4 Moment vs. Displacement- SCC1-2-B - L. = 118.50 in

195

Moment (kip-ft)

Moment (kip-ft)

120
110

Moment (KN-m) ..
B ‘6' 8 8 8 3 8 8 8

.6
O

O

Displacement at the section (in.)

 

 

 

 

 

 

0 1 2 3 4 5 6 7
0 _=. 111-Z7_KN:UL(8_5-.5_L(-BL _______________________ a:
.

SlipOnset

5" ~
”Lda” P.
A :)M 4:
I
M E
11 1111111111111111'1r11111r1 1111111fi111 11 111T111111111IP
0 15 30 45 60 75 90 105 120 135 150 165 180

Displacement at the section (mm)

338

0

Figure A4 - 5 Moment vs. Displacement- SCC2A-l-A - L. = 70.50 in

1 20

Moment (KN-m) .. ..
8 3 8 8 ‘5‘ 8 8 8 3

Displacement at the section (in.)

 

 

 

 

 

0 1 2 3 4 5 6 7
I A A J J l A A A A L L A L A l A k; A l A A A A l A A A A l A A A A

5 flZ-ZLKNMQGJ .3213). _______________________ 5:.

A E

E

1:12 912 i

. l-

Slip Onset _ """""" l: E

‘-~-- -.- _,_ ............ 7 ll 1

1 5\5 E

r—Ldaq )M AR—E

.11 ’-

0 25 50 75 1 00 1 25 1 50 1 75 200 225

Displacement at the section (mm)

88882188

8

Figure A4 - 6 Moment vs. Displacement- SCC2A-l-B - L, = 64.50 in

196

Moment (kip-ft)

Moment (kip-ft)

Displacement at the section (in.)

0.0 0.8 1 .6 2.4 3.2 4.0 4.8 5.6 6.4 7.2 8.0

IAJAAIAAAAIAJAAIAAAAIAAA A A AAlAAAAlAAAA

  
 
 

120 Mg 5 111.;7_kflinJi6é3£L ________________ _ - f 9°
11o so
.100 A sup 9 Mn . 8.64 mm (0.34 in): 70
5 so E
12" so Slip Onset I: E 6°
E 70 E
E so :

8

 

 

 

V'YTYWTIIIVIIITTTT'IVIIIrUrIT‘VVITrIYITTITT

Displacement at the section (mm)

Figure A4 - 7 Moment vs. Displacement- SCC2A-2-A - Le = 80.00 in

Dlsplacement at the section (in.)

 

0 20 40 60 80 100 120 140 1 60 180 200

 

 

 

 

 

 

o 1 2 3 4 s s 7 s 9 10
120 MEEL‘Z-EUH'IMQQE'EL ________________ -5 9°
110 £80
,1” A swam-5.33 mm (0.21 in.) i 70
£80 Slip Onset |,_ L 60

‘E E
“’70 :50

$60 2
5 ~40

so E
40 E30
30 320

20 ;
~10

10 E
o ,....,....,nrr,”.....errqm.......,........‘o

 

 

o 25 50 75 100 125 150 175 200 225
Displacement at the section (mm)

Figure A4 - 8 Moment vs. Displacement- SCC2A-2-B - Le = 86.75 in

197

Moment (kip-ft)

Moment (kip-ft)

12° €M25111.14.kL4-_ _(_6._1(-_ L _______________________ i 90
110% :_ so
100 é ;

A E r 70

E 90‘: .

z' 3

as 8° 3

‘5 7°“:

0 I

(E) 60 1: Slip Onset

: so—g

Displacement at the section (in.)
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5

 

1

 
  
 
 
   
  

 

 

*A i 310

 

 

 

or‘Y'VTYYrTj'V'Y'TTTVIVUIVTU'V'TT 'l'VY'IIUTVT'VV—rerfiYV o

o 15 30 45 so 75 90 105 120 135 150 165
Displacement at the section (mm)

Figure A4 - 9 Moment vs. Displacement- SCC2B-1-A - L. = 70.50 in

Displacement at the section (in.)
o 1 2 3 4 5 6 7 8 9 1o 11 12

 

 

 

 

120 Mn _---.1 1741mm. 1.854.115; _________________ E 9°
110 E30
100 A SlinMn=10.67 mm (0.42 in.) g
A 370
E so A :
' L
§ so :60
: SlipOnset t
70 .
C .-
0 _50
560 .
~40
2 so :
4o 930
so :
:20
20 :
»-1o
10 i
o ,.........,....,.........,....,....,....,fi.r,....,....,....,'o

 

 

o 25 50 75 100 125 150 175 200 225 250 275 300
Displacement at the section (mm)

Figure A4 - 10 Moment vs. Displacement- SCCZB-l-B - L8 = 102.75 in

198

Moment (kip-ft)

Moment (kip-ft)

Displacement at the section (in.)

 

 

 

 

 

o 1 2 3 4 5 s 7 a 9
120 n511Z-Z7_kN:EL(§§25_k-fli_ _______________________ 9°
110 so
100

70 .2

E 60 e

x

1‘. 80 :7

‘5 70 so 5

0 E

E 50 o

o 40

s so 5

so
3° 20
20
1o 10
o..fifi.,.r-.,........+re...........-.,....,...., o
o 25 so 75 100 125 150 17s zoo 225

Displacement at the section (mm)
Figure A4 - 11 Moment vs. Displacement- SCCZB-Z-A - Le = 126.75 in

Displacement at the section (in.)
o 1 2 3 4 5 6 7 8 9 1o

AJAIIAII‘IILA lIllllJLAAlAlllLLLLLIIIIAllLAAIIII

 

120 éfln = 117.27 kN-m (86.5 k-ft)

110
100

'7"

D
D
'80

AAAAIAAAAJAA

L7o

: A No Sllp Occured :
80 € Mn Achieved :- 60

 

Moment (KN-m)
N
O

AAIAAAA AAAA AAAA AAA

40
so
20
1o

 

I‘LL

 

 

 

 

o IIIVIIIUYIjIl1[ITIVTII'IIUVVI[VI'V'T'Y’TIIVIU'VITII'o

0 25 50 75 100 125 150 175 200 225 250
Displacement at the section (mm)

Figure A4 - 12 Moment vs. Displacement— SCC2B-2-B - Le = 103.80 in

199

Moment (kip-ft)

Displacement at the section (in.)

 

 

 

 

0.0 0.5 1 .0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5
120 M3 3.1 1.631 museum _______________________ 9°
110 80
A100 Slip Onset 70 ...
E t
2' 9° / .2
x 80 60 5
... 1..!
C
5 70 50 g
g“ .0 s
E 50
40 30
30 20
20
10 10
o, ......4.,....,. .,....,......fi.....r,.o
0 20 40 60 80 100 120 140 160

Displacement at the section (mm)

Moment (KN-m)

Figure A4 - 13 Moment vs. Displacement- SCC3-1-A - Le = 58.50 in

Displacement at the ssection7(in.)

 

 

 

 

 

o 1 2 3 4 5 10
120 M2311§-§7_k.81nJ§6-_1’5£L ________________________ 3 9°
110 I_ so

100 E
_' 70

9° :
so 1,3 3 so
70 Slip Onset : so

so :
40

so

40 3o
30 I 20

20 E

A .
10 E 10

0 l ' 'fi—r l T T V 7 I T T T V I I I v v 1 v t v v I r . 1 v [ v u v r I v r v 1 T—rT—v—r I v u v v 1 E o

 

o 25 50 75 100 125 150 175 200 225 250
Displacement at the section (mm)

Figure A4 - 14 Moment vs. Displacement- SCC3-1-B - Le = 97.75 in

200

Moment (kip-ft)

Moment (KN-m)

Displacement at the section (in.)

 

 

 

 

 

 

0 1 2 3 4 5 6 7 8 9 10
120 M2 : 11§-§7_k§-1n_(§.6._22 k_-f_o _______________________ 9°
110 80
100
A 70
590
0
E30 60
E70 50
EGO
0 40
250
40 30
30 20
20
10 10
o ..-fir.r.1...........r.............,.......f.,.....o

o 25 so 75 100 125 150 175 zoo 225 250
Displacement at the section (mm)

Figure A4 - 15 Moment vs. Displacement- SCC3-2-A - Le = 106.50 in

Displacement at th: section 01;.)
7

 

     
 

 

 

 

 

 

0 1 2 3 4 5 10 11
A A A 1444 L1 LA A 11 A A A A l A A A A 142 A A A I A A A A l A A A A l A A A A l A I A A A A
E 90
4 I-
120 1M2 :11§-§7_kB-2n_l8_6-_22|2f.9 ________________ _ _ _.
.
- r
.
1
110 1 _. so
100 J ’
< r-
: : 70
.
90 1 *
: Z
80 1 f 60
d D
70 I 1
3 : 50
so 5 i
1 :40
50 1 »
< v-
1 r
40 1 1' 3°
30 E
- P
2 r 20
20 1 E
.
p-
10
10
o ' r f I I ' I l I U I I Y W T I ' V V Y r Y I V—V—f V I V I l' V I Tfi' 7 U I V I r' I r '1 U I I I I 'T o

 

0 25 50 75 100 125 150 175 200 225 250 275
Displacement at the section (mm)

Figure A4 - l6 Moment vs. Displacement- SCC3-2-B - L. = 103.00 in

201

Moment (kip-ft)

Moment (kip-ft)

Displacement at the section (in.)

 

 

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
120 21111111 12_.4_k§1n_(§2_.9_k-_ft_)_ _______________________ 9°
1 so
A100 : Slip Onset
70
E 1
z . A
5 30 e 60
E .
as: ‘ 50
:5 f 40
4o 3 3 30
‘ 20
20 ~
2 1o
4
o ..4.,....,....,....,H... o

 

 

 

 

0 10 20 30 40 50 60 70 80 90 100110 120

Displacement at the section (mm)

Figure A4 - 17 Moment vs. Displacement- NCCA-l-A - Le = 76.00 in

Displacement at the section (in.)

 

 

 

o 1 2 3 4 s s 7 s 9 1o 11 12
120% {90
EMEJflJhM-mJQZJJS-fi). _________ _ _______
11o<1 Lso
1001 E
2. 1 -70
39071 t
a: so? E60
2' 1 C
J L
E70} :50
g ”‘3

2 F
5°: E
405 330
so? 320
20g E
105 31°

ASIinMn=2.54 mm (0.1 in.) ‘
o ..........r...,.fi.,....,....,....,..T-,....,....,..,.,.fi..’o

 

 

 

0 25 50 75 100 125 150 175 200 225 250 275 300

Displacement at the section (mm)

Figure A4 - 18 Moment vs. Displacement- NCCA -1-B - L8 = 122.75 in

202

Moment (kip-ft)

Moment (kip-ft)

Moment (KN-m)

Displacement at the section (in.)

0 1 2 3 4 5 6 7 8 9 10 11 12

120

A Slip Q Mn 8 5.58 mm (0.22 in.)

'7:
3o

Slip Onset

388

 

 

 

 

  
     

'V'VVTWr'fi—VYr‘rT

 

90

70

Moment (kip-ft)

 

 

oIf,'V'I'IY'VYUII'V‘TI'I'VIY '7 1"VVITVIUIrfiTjYVIYITT'V"VI’I’I

0 25 50 75 100 125 150 175 200 225 250 275 300

Displacement at the section (mm)

Figure A4 - 19 Moment vs. Displacement- NCCA -2-A - L. = 111.00 in

Moment (KN - m)

Displacement at the section (in. )

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 5.5 7.0

 

     

 

 

120 «j _- so
thn = 112.4 kN-m (82.9 k-ft) :
1101 3 so
100 1 i
‘ - - 70 A
90 1 Onset ofSlnp t a;
1 r .9
80 1 f 60 5
i ; E
70 1 1' 50 g
nos
50 i ;
40 a I- so
so ‘3 E 20
20 € :
1o 3 r 10
o 1 T I I I l V Y rfi' 1 I I I I I I I I I I I I I I 1 r r I I l V V I I I I T T 1 r T 1’ V Y P o

 

 

 

o 20 40 so so 100 120 140 160
Displacement at the section (mm)

Figure A4 - 20 Moment vs. Displacement- NCCA -2-B - L. = 60.00 in

203

Moment (KN-m)

Moment (KN-m)

Displacement at the section (in.)

0.0 0.4 0.8 1 .2 1 .6 2.0 2.4 2.8 3.2 3.6 4.0

llll‘llllAlAJlLlJLl ALLIALAA ‘IAILAAIJLJA

‘20 M_n_=_115_.6§ mm 192-2|sz _______________________
1 1 0

85‘

No Slip Occured
mg [3.2 Mn not Achieved

38

 

 

38888

 

 

o 111T1't'7 '1' frtTTl " l I I I I l I I V '1 'I U V l I 'Y—Ujjfirj ' [Tfir]

0 10 20 30 40 50 60 70 00 90 100

Displacement at the section (mm)

3‘88

888

Figure A4 - 21 Moment vs. Displacement- NCCB -l-A - L. = 63.75 in

Displacement at the section (in.)

 

 

 

 

 

 

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
12° "35.115225: mm £823.st _______________________ .:
110 L
100 E
90 E
80 P NoSllpOccured

m m MnnotAchleved ;

7o ._

60 E

r

so .

t

40 r

1

3° L

20 E

10 E
o fireflmfiWH.....m....,T............erfinwnsfifier-

o 102030405060708090100110
Displacement at the section (mm)

38388388

..A
O

D

Figure A4 - 22 Moment vs. Displacement— NCCB -1-B - Lc = 64.00 in

204

Moment (kip-it)

Moment (kip-ft)

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5
lLALlAAAJlAJIAlLALLJJAA‘l AAl I AlAAA-l 1|. All 1.1 A III I 1

Displacement at the section (in.)

 

 

 

 

 

 

 

120 3 ’ 9°
-,_MD.=J 1.538.191; _(§5_.1_k'_fl.1_ ...... _ ._ ........
E - so
100 1 .
A . ' 70
g 1
§ 80 _1 A No Slip _Occured '- 60
v . Mn Achieved *
o:- - 1:12 p12
0 1 ' 50
+ ’Vd
.... 1
2 1 £'\1 4°
“1 "Li“ M ,\1
1 \/ an .20
20 1
I ..A ' 10
0 . xvvvvyrvvvyv YIIVVIIITWYI| 1 1 1 1 l" 'I ‘V 0
0 15 30 45 50 75 90 105 1 20 135 1 50 1 65 1 80

Displacement at the section (mm)

Moment (kip-ft)

Figure A4 - 23 Moment vs. Displacement— NCCB -2-A - L, = 100.50 in

Displacement at the section (in.)
3 4 5 6 7 s

 

 

 

 

 

 

 

 

o 9 10
120 i _ _ - -... 9°
. Mg;_11_sssk_N_n1(25_.1m ..... ,, - . - _
11o -; r *- so
100 -f
A 3 f 70
g so-g :
g so-f A No Slip Occured Eso
:7 1 m m Mn Achieved :
70 1 '-
5 s - 5°
g 60 1 ~ __ “““ 1111
E 50.: ‘ ~ ...... 1V8 ............. El. 111340
E ~—L A, :
4o 1 d8 M : 30
2 E/_/ E 111 — 20
20 g 3
10 'A - 10
o 4 . r I Vfi rfi v v - 1 ' l ' i 1 ' 1 ' ' r l ' l E 0
o 25 so 75 100 125 150 175 200 225 250

Displacement at the section (mm)

Figure A4 - 24 Moment vs. Displacement- NCCB -2-B — L, = 93.50 in

205

Moment (kip-ft)

Appendix 5 - Prestressing Details

 

_8 £8.52" 82.

.38 DE hat of 9 a: 388— st»: weE 03205 8.3 :32: 82? 05 .323 gauche 2: .2 8x *

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3-83 3.83 3-w=<-Nm 3. N : 338 333 33 NN N

3-83 3.8.3.3 3-3-«-3 Q. N : 338 383 £2 < N

3-83 3-83 3.324 N N3. N t 8.38 33 £3 a N

3-83 3.33 3-3-2-: 93: $38 33 £3 < N Gem
3-898 3-3-«-2 382-2 N2: Nam: 3e 3% m N
3.89% 3.32.2 3-3-«-2 N2: Sm: 3e 3% < N
3-89-2 3.3-«3N 3.32:: 32% NN . N MN 5% Q?- m N
3-8e-NN 3-32-2 3-3-«-2 5.5 NNNNN 3% 9% < N muuz
3-89-3 3-3-«3 3.3: 382 8.38 :3 3% m N
3-89-2 3-3-N3 3-323 322 8.32 N5 3% < N

3803 3-3-«3 3-323 NSNN 2:3 :2 3% m N

3-85-N 3-3-N3 3-323 3N NN 3.2-N :3 3% < N 80m
3-32-8 3-3N3N 3378 No. EN 3.3: 383 3% m N
3-82-8 3-3N3N 3-378 3. NB 3.3: 38 35 < N
3.22.2 3378 3378 8.32 eNQN 38 3% m N
3.22.2 3-373 3.378 3.82 838 33 3% < N mNuom
3.82.: 3.3.2 3-33 332 SEN N33 33 m N

3-823 3-3N-NN 3-33 3.8: SEN N3; 32 < N

3-823 3.3.2 333 3. :2 3.NNN N33 38 NN N

3-80-2 3.3.2 3-32 33: woNNN N3; 83 < N <88

as: as: as
$588 3.3—om mesa-H . wage-_- :25
3&5qu mwabmoham many—«moun— _a._=xo= $82.— 3:53: ham—3.3 % mun—m #anm— NE
3 3.5 3233 u.
.3 3.5 .No 3.5 me than i. ..e .95 Na K.
an 8.x... 3 9%,

 

 

 

 

 

 

 

 

 

 

 

$825 $2323on N - m< «Esp.

206

[1]

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

REFERENCES

Abrishami, H.H., and Mitchell, D.(1993, June), “Bond Characteristics of
Pretensioned Strand,” ACI Materials Journal, 90(3), 228—235.

American Association of State Highway and Transportation Officials (AASHTO),
AASHTO LRFD Bridge Design Specifications, Second Edition, 1998.

American Concrete Institute (ACI), Building Code Requirements for Structural
Concrete and Commentary, ACI 318-02, Farmington Hills, Michigan, 2002.

Antoine E Namaan. (1982) “Prestressed Concrete Analysis and Design
Fundamentals.”

Attiogbe, E., See, H., and Daczko 1., “Engineering Properties of Self-Consolidating
Concrete,” Proceedings of the First North American Conference on the Design and
Use of Self- Consolidating Concrete, Evanston, IL, November 2002.

Balazs,G.L.(1993). “Transfer Lengths of prestressing strands as a function of draw-
in and initial prestress.” PCI Journal 38(2) 86—93.

Barnes,R.W., Bums,N.H., & Kreger, ME. (1999, December). “Development
Length of 0.6 inch prestressing strand in standard I shaped pretensioned concrete
beams.” Research Report [388-1. Center of Transportation Research, Bureau of
Engineering Research, University of Texas at Austin,

Buckner, C.Dale., (1995, April) “A Review of Strand Development Length for
Pretensioned Concrete Members,” PCI Journal, 40(2), 84-105.

Byung Hwan Oh., and Eui Sung Kim.,(2000, December). “Realistic Evaluation of
Transfer Lengths in Pretensioned. Prestressed Concrete Members.” ACI Structural
Journal. 97(6). 821-830.

Cousins, T.E., Badeaux, M.H., and Mostafa, S..(1992, February), “Proposed Test
for Determining Bond Characteristics of Prestressing Strand,” PCI Journal, 6673

Daczko, J., and Constantiner, D. (2002, November), “Not All Applications are
Created Equal; Designing and Selecting the Appropriate SCC Mixture,”
Proceedings of the First North American Conference on the Design and Use of Self-
Consolidating Concrete, Evanston, IL.

Daczko, J., and Kurtz, M.A. (2001, October), “Development of High-Volume
Coarse Aggregate Self-Compacting Concrete.” Proceedings of the Second
lntemational Symposium on Self-Compacting Concrete, Tokyo, Japan.

207

[13]

[14]

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

Dallas R. Rose, Bmce W. Russell. (1997, August). “Investigation of standardized
tests to measure the bond performance of prestressing strand.” PCI Journal, 42 (4),
56—80.

Deatherage, J.H., Burdette, E.G., and Chew, CK. (1994, February), “Development
Length and Lateral Spacing Requirements of Prestressing Strand for Prestressed
Concrete Bridge Girders,” PCI Journal, 39(1), 70-83.

Evan C. Bentz., Michael P.Collins, (2000). “Response—2000” Version 1.0.5 —
Reinforced Concrete sectional analysis using the modified compression field theory.

Gross, P. Shawn and Ned H.Bums. (1995, June). “Transfer and Development
Length of 15.2mm (0.6 in.) diameter prestressing strand in high performance
concrete: Results of Hoblitzell—Buckner Beam tests.” Research report 580 -2.
Center of Transportation Research, Bureau of Engineering Research, University of
Texas at Austin.

Hanson, NW. and Kaar, PH. (1959) “Flexural Bond Tests of Pretensioned
Prestressed Beams, ” ACI Journal, 55(7), 783-803.

Kaar, P.H., LaFraugh, R.W., and Mass, M.A. (1963, October) “Influence of
Concrete Strength on Strand Transfer Length,” PCI Journal, 47-67

Khayat, K.H, and Daczko, J .(2002, November) “The Holistic Approach to Self-
Consolidating Concrete,” Proceedings of the First North American Conference on
the Design and Use of Self-Consolidating Concrete, Evanston, IL,

Khayat, K.H. (1999, June) "Workability, Testing, and Performance of Self-
Consolidating Concrete," ACI Materials Journal, 96(3), 346-353

Khayat, K.H., K. Manai and A.Trudel.(1996, December) “In Situ Mechanical
Properties of Wall Elements cast using Self Consolidating Concrete”. ACI
Materials Journal. 94(6). 491-500

Khayat, K.H., Patrick Paultre., and Stephan Tremblay. (2001, October), “Structural
Performance and In-Place properties of Self-Consolidating Concrete used for
casting highly reinforced columns”. ACI Materials Journal. 98(5), 371 — 378

Leonhardt, F. (1964) “Prestressed Concrete — Design and Construction”, Wilhem
Ernst & Sohn, Berlin.

 

Logan, DR, (1996, April) Discussion of “A Review of Strand Development
Length for Pretensioned Concrete Members.” PCI Journal, 41(2), 112-116.

Logan, DR. (1997, April). “Acceptance criteria for bond quality of strand for
pretensioned prestressed concrete applications.” PCI Journal, 42(2), 52-90

208

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[341

[35]

[36]

Nan Su, Kung-Chung Hsu, His-Wen Chai.(2001) “A simple mix design method for
Self Compacting Concrete”. Cement and Concrete Research, V.3l, 1799-1807

Okamura, H. (1999, June), "Self-Compacting High-Performance Concrete," ACI
Materials Journal, 96(3), 346-353.

Persson, B., "A Comparison between Mechanical Properties of Self-Compacting
Concrete and the Corresponding Properties of Normal Concrete," Cement &
Concrete Research, 31 (2001) 193-198

Petersson, O., Gibbs, J ., and Bartos, P.(2002, November) “Testing-SCC,”
Proceedings of the First North American Conference on the Design and Use of Self-
Consolidating Concrete, Evanston, IL.

Precast/Prestressed Concrete Institute (PCI), “Interim Guidelines for the use of Self;
Consolidating Concrete in Precast/Prestressed Concrete Institute Member Plants, ”

TR-6-03, Chicago, Illinois, April 2003.

Rose, DR, and Russel, B.W. (1997, August) “Investigation of Standardized Tests
to Measure the Bond Performance of Prestressing Strand,” PCI Journal, 42 (4), 56-
80.

Russell, B.W., and Burns, NH. (1997, May) “Measurement of Transfer Lengths on
Pretensioned Concrete Elements,” Journal of Structural Engineering, 123(5), 541-
549.

Russell, B.W., and Burns, NH, (1996, April) Discussion of “A Review of Strand
Development Length for Pretensioned Concrete Members,” PCI Journal, 41(2),
1 16-1 19.

Russell, B.W., and Ned H.Bums. (1993, January). “Design Guidelines for Transfer,
Development and Debonding of large Diameter Seven wire strands in pretensioned
concrete girders.” Research Report 1210-5F. Center pf Transportation Research,
Bureau of Engineering Researgh, University of Texas at Austin.

Shahawy, M. (2001, August), “A Critical Evaluation of the AASHTO Provisions
for Strand Development Length of Prestressed Concrete Members,” PCI Journal,
46(4), 94-117.

Thatcher, D.B., Heffington, J.A, Kolozs, R.T., Sylva HI, G.S., Breen, J.E., and
Bums,N.H. (2002, January). “Structural Lightweight concrete prestressed girgers
and panels.” Research report 1852-] Center for transportation research. Bureau of
Engineering Research, University of Texas at Austin.

209

[37]

[38]

[39]

[40]

[41]

Transportation Research Board, “Research Problem Statements,” A2C03
Committee on Concrete Bridges, Group 2 — Design and Construction of
Transportation Facilities, Technical Activities Div.-

Vachon, M. (2002, July) “ASTM Puts Self-Consolidating Concrete to the Test,”
ASTM Standardization News, July 2002, url: http://astmbrg/ggi-
bin/SoftCart.exe/SNEWS/JULY 2002/vachon iu102.htm

Xie, Y., Lin, B., Yin, J ., and Zhou, 8., "Optimum Mix Parameters of High-Strength
Self-Compacting Concrete with Ultrapulverized Fly Ash," Cement & Concrete
Research, 32 (2002) 477-480.

Zhu, W., Gibbs, J .C., and Bartos, P.J.M., "Uniformity of In Situ Properties of Self-
Compacting Concrete in Full-Scale Structural Elements," Cement & Concrete
Composites, 23 (2001) 57-64.

Zia, P., and Mostafa, T. (1977, October) “Development Length of Prestressing
Strands,” PCI Journal, 55—65.

210

   

ulllliliinlliljl‘lmill