Dynamic Behavior of Reinforced Concrete Beams Under Varying Rates of Concentrated and Impact Loadings

This document is downloaded from DR‑NTU (https://dr.ntu.edu.
sg)
Nanyang Technological University, Singapore.
Dynamic behavior of reinforced concrete beams

under varying rates of concentrated and impact
loadings
Satadru Das Adhikary
2014
Satadru Das Adhikary. (2014). Dynamic behavior of reinforced concrete beams under
varying rates of concentrated and impact loadings. Doctoral thesis, Nanyang Technological
University, Singapore.
https://hdl.handle.net/10356/58911
https://doi.org/10.32657/10356/58911
Downloaded on 25 Aug 2024 12:35:26 SGT

DYNAMIC BEHAVIOR OF REINFORCED CONCRETE
BEAMS UNDER VARYING RATES OF
CONCENTRATED AND IMPACT LOADINGS
SATADRU DAS ADHIKARY
SCHOOL OF CIVIL AND ENVIRONMENTAL ENGINEERING
2014
DYNAMIC BEHAVIOR OF REINFORCED CONCRETE
BEAMS UNDER VARYING RATES OF
CONCENTRATED AND IMPACT LOADINGS
Satadru Das Adhikary
School of Civil & Environmental Engineering
A thesis submitted to the Nanyang Technological University in partial

fulfillment of the requirement for the degree of
Doctor of Philosophy
2014
Statement of Originality
STATEMENT OF ORIGINALITY
I, Satadru Das Adhikary, do hereby declare that the research work contained in the
thesis entitled “Dynamic behavior of RC beams under varying rates of concentrated
and Impact loadings” is entirely my original contribution. I also certify here that the
thesis has been solely written by me. This thesis has not been previously submitted
for any other degree in any other university.
All the resources used during the doctoral study have been duly acknowledged. All
testing have been conducted solely by the author with the assistance from Prof.
Kazunori Fujikake at National Defense Academy, Japan.
i
ii
Acknowledgement
ACKNOWLEDGEMENT
The research presented in this thesis on Dynamic behavior of RC beams under

varying rates of concentrated and Impact loadings, was undertaken at School of
Civil and Environmental Engineering in Nanyang Technological University,
Singapore. Herein, I would like to take the opportunity to thank those whose
contributions to this research were indispensable.
First and foremost, I would like to express my most profound gratitude to my

supervisor Prof. Bing Li, for his professional guidance, invaluable advice and
continuous encouragement throughout the doctoral candidature. I owe my earnest
gratitude to Prof. Kazunori Fujikake for his immense assistance and guidance in
various stages of this research.
Special thanks should also confer to the technical staff from Construction
Technology and Protective Engineering Laboratory at NTU. The constructive
suggestions, fruitful technical discussion with fellow research students in NTU had
made this research a most memorable one. This acknowledgement would not be
completed without mentioning the assistance from Mr. Takashi Fukuda in the
course of experimental work.
Last but not least, the author would like to express his deepest gratitude to his wife
Piyali and parents, for their unwavering support, encouragement and understanding.
I would like to dedicate this thesis to my parents.
iii
iv
List of Publications
LIST OF PUBLICATIONS
Adhikary, S.D., Li, B., and Fujikake, K. “Dynamic Behavior of Reinforced

Concrete beams under Varying Rates of Concentrated Loading.” International
Journal of Impact Engineering, 47, 24-38, 2012.
Adhikary, S.D., Li, B., and Fujikake, K. “Strength and Behavior in Shear of
Reinforced Concrete Deep Beams under Dynamic Loading Conditions.” Nuclear
Engineering and Design, 259, 14-28, 2013.
Adhikary, S.D., Li, B., and Fujikake, K. “Effects of High Loading Rate on
Reinforced Concrete Beams.” ACI Structural Journal, Available online.
v
vi
Table of Contents
TABLE OF CONTENTS
STATEMENT OF ORIGINALITY ........................................................................ i

ACKNOWLEDGEMENT ...................................................................................... iii
LIST OF PUBLICATIONS..................................................................................... v
TABLE OF CONTENTS ....................................................................................... vii
ABSTRACT ........................................................................................................... xiii
LIST OF TABLES ................................................................................................. xv
LIST OF FIGURES ............................................................................................. xvii
LIST OF SYMBOLS ........................................................................................... xxv
CHAPTER 1 INTRODUCTION ............................................................................ 1

1.1 Research Background ....................................................................................... 1
1.2 Objective and Scopes ....................................................................................... 6
1.3 Structure of Thesis ............................................................................................ 8
CHAPTER 2 LITERATURE REVIEW .............................................................. 11

2.1 Overview ........................................................................................................ 11
2.2 Strain Rates Effect on Mechanical Properties of Materials ............................ 11
2.2.1 Plain concrete........................................................................................... 11
2.2.2 Steel reinforcement .................................................................................. 20
2.3 Influence of Loading Rates on Bond Properties............................................. 21
2.4 Research on RC Beams under Varying Loading Rates .................................. 22
2.4.1 Experimental investigations..................................................................... 22
2.4.2 Analytical and numerical studies ............................................................. 24
2.5 Study on RC beams under Drop-weight Impact Loading .............................. 25
2.5.1 Laboratory tests ....................................................................................... 26
2.5.2 Numerical and analytical studies ............................................................. 42
2.6 Summary......................................................................................................... 43
CHAPTER 3 EXPERIMENTAL INVESTIGATION OF LOADING RATE

EFFECTS ON RC BEAM ..................................................................................... 45
3.1 Introduction .................................................................................................... 45
vii
Table of Contents
3.2 Analysis of Experimental Database................................................................ 46

3.3 Objective of Test Program.............................................................................. 50
3.4 Experimental Overview .................................................................................. 50
3.5 Experimental Results and Discussions ........................................................... 56
3.5.1 Load-midspan displacement relationships ............................................... 56
3.5.2 Concrete contribution to shear stress ....................................................... 61
3.5.3 Concrete and shear reinforcement contribution to shear resistance ........ 62
3.5.4 Cracking stiffness .................................................................................... 62
3.5.5 Energy absorption .................................................................................... 63
3.5.6 Comparison of time to reach peak load, yielding of tensile and shear
reinforcements & peak strain rate ..................................................................... 64
3.5.7 Deflected shapes ...................................................................................... 66
3.5.8 Crack patterns .......................................................................................... 67
3.6 Summary......................................................................................................... 72
CHAPTER 4 NUMERICAL SIMULATION OF LOADING RATE EFFECTS

ON RC BEAM ........................................................................................................ 75
4.1 Introduction .................................................................................................... 75
4.2 Finite Element Modeling Techniques for Reinforced Concrete..................... 76
4.3 Finite Element Model ..................................................................................... 77
4.3.1 Structural configuration ........................................................................... 77
4.3.2 Boundary conditions, application of load and contact algorithm ............ 80
4.3.3 Constitutive model for concrete............................................................... 80
4.3.4 Constitutive model for reinforcing steel .................................................. 83
4.3.5 Constitutive model for supports and loading plate .................................. 84
4.4 Validation of Finite Element Analysis Results .............................................. 84
4.4.1 Load vs. midspan deflection curves......................................................... 85
4.4.2 Failure pattern .......................................................................................... 91
4.5 Parametric Study ............................................................................................ 93
4.5.1 Numerical simulation matrix ................................................................... 93
4.5.2 Influence of longitudinal reinforcement ratio on DIF of maximum
resistance........................................................................................................... 95
4.5.3 Influence of longitudinal reinforcement ratio on failure mode ............... 97
viii
Table of Contents
4.5.4 Influence of transverse reinforcement ratio on DIF of maximum

resistance........................................................................................................... 99
4.5.5 Influence of transverse reinforcement ratio on failure mode ................. 100
4.5.6 Influence of yield strength of longitudinal reinforcement on failure
mode ............................................................................................................... 102
4.5.7 Influence of grade of concrete on peak load .......................................... 106
4.6 Summary....................................................................................................... 107
CHAPTER 5 EXPERIMENTAL ASSESSMENT OF RC BEAM SUBJECTED

TO DROP-WEIGHT IMPACT LOADING ...................................................... 109
5.1 Introduction .................................................................................................. 109
5.2 Objective of Experimental Program ............................................................. 111
5.3 Outline of Experiment .................................................................................. 112
5.4 Experimental Results and Discussions ......................................................... 119
5.4.1 Static test results .................................................................................... 120
5.4.2 Impact test results .................................................................................. 121
5.4.2.1 Time histories of impact load ........................................................ 122
5.4.2.2 Time histories of midspan deflection ............................................ 124
5.4.2.3 Impact response characteristics ..................................................... 125
5.4.2.4 Crack pattern .................................................................................. 128
5.4.2.5 Strain rate ....................................................................................... 134
5.5 Analysis of Experimental Database.............................................................. 137
5.5.1 Empirical formulas for impact-resistance design .................................. 141
5.6 Summary....................................................................................................... 144
CHAPTER 6 FINITE ELEMENT MODELING OF RC BEAM UNDER

IMPACT LOADING ........................................................................................... 147
6.1 Introduction .................................................................................................. 147
6.2 Finite Element Model ................................................................................... 148
6.2.1 Structural configuration ......................................................................... 148
6.2.2 Boundary conditions, application of load and contact algorithm .......... 151
6.2.3 Constitutive model for concrete............................................................. 151
6.2.4 Constitutive model for reinforcing steel ................................................ 152
ix
Table of Contents
6.2.5 Constitutive model for impactor, support roller and inverted triangular
plate................................................................................................................. 152
6.3 Verification of Finite Element Analysis Results .......................................... 153
6.3.1 Time histories of impact load curves ..................................................... 154
6.3.2 Time histories of midspan deflection curves ......................................... 159
6.3.3 Crack profiles......................................................................................... 164
6.3.4 Impact resistance of test specimens ....................................................... 173
6.4 Parametric Study .......................................................................................... 176
6.4.1 Effect of mass ratio ................................................................................ 176
6.4.2 Effect of longitudinal reinforcement ratio ............................................. 177
6.4.3 Effect of compressive strength of concrete ........................................... 179
6.4.4 Effect of boundary conditions ............................................................... 180
6.5 Summary....................................................................................................... 183
CHAPTER 7 ANALYTICAL MODEL FOR IMPACT RESPONSE

PREDICTION OF RC BEAM ............................................................................ 185
7.1 Introduction .................................................................................................. 185
7.2 Energy-balance Model.................................................................................. 185
7.2.1 Application of energy balance model .................................................... 187
7.3 Single-Degree-of-Freedom (SDOF) model .................................................. 191
7.3.1 Elastic-plastic analysis ........................................................................... 192
7.4 Summary....................................................................................................... 201
CHAPTER 8 PERFORMANCE OF IMPACT-DAMAGED RC BEAM ....... 203

8.1 Introduction .................................................................................................. 203
8.2 Outline of Experiment .................................................................................. 204
8.3 Summary of Experimental Results ............................................................... 205
8.3.1 Load vs. midspan deflection curves....................................................... 205
8.3.2 Residual resistance index (RRI) ............................................................ 209
8.3.3 Residual stiffness index (RSI) ............................................................... 211
8.3.4 Failure pattern ........................................................................................ 212
8.4 Finite Element Framework ........................................................................... 215
8.5 Verification of Finite Element Analysis Results .......................................... 216
x
Table of Contents
8.5.1 Impact loading history and residual resistance vs. midspan deflection . 216
8.5.2 Crack profiles of beam after impact and post impact
quasi-static loading ......................................................................................... 218
8.6 Numerical Simulation Case-studies ............................................................. 220
8.6.1 Effect of mass ratio ................................................................................ 220
8.6.2 Effect of longitudinal reinforcement ratio ............................................. 221
8.6.3 Effect of transverse reinforcement ratio ................................................ 223
8.6.4 Effect of compressive strength of concrete ........................................... 224
8.7 Summary....................................................................................................... 225
CHAPTER 9 CONCLUSIONS AND FUTURE WORK.................................. 227

9.1 Conclusions .................................................................................................. 227
9.2 Future work .................................................................................................. 232
REFERENCES ..................................................................................................... 233

APPENDIX-A ....................................................................................................... 245
APPENDIX-B ....................................................................................................... 249
APPENDIX-C ....................................................................................................... 257
xi
xii
Abstract
ABSTRACT
Reinforced Concrete (RC) structures may experience various types of loading

during their lifespan, which in turn induces a wide spectrum of strain rates. The
materials involved in RC structures are strain-rate sensitive and thus, the behavior
of structural members can be affected by the loading rates but only significantly
when the rate differs by more than one order of magnitude. Majority of past
research on RC beams have focused on assessing their behavior under static and
relatively slow loading rates while limited attention has been paid to the high
loading rates. Therefore, to shed some light in this field, an experimental program
comprising twenty-four RC beams was carried out under four different loading
rates ranging from slow (4  10-4 m/s) to fast (2 m/s) to cover the wide range of
loading scenarios (quasi-static, earthquakes and low velocity impact regime).
Comparative analyses of beams under these varying loading rates highlighted
several important aspects of their dynamic behavior. Furthermore, the influence of
various key parameters on dynamic increase factor (DIF) of maximum resistance
and failure modes of beams under these loading conditions was summarized and
discussed through numerical simulation parametric studies after successful
validation of numerical models in an explicit finite element (FE) program.
Although the loading rate effects covering low velocity impact regime were
considered, however, it would be more practical to consider realistic impacts. Thus,
a drop-weight impact test program was undertaken on thirty RC beams to evaluate
their impact responses. The acquired data was then used in the development and
verification of numerical and analytical methods. Two empirical equations have
been proposed by analyzing a dataset which would aid in determining the required
static bending and shear resistance for input impact energy by specifying the
maximum midspan deflection for each limit state of beam. Moreover, to extent the
knowledge beyond the range of parameters investigated experimentally, FE models
of the beams were also developed. Maximum midspan deflection could be an
important performance index to evaluate the damage levels of beam when subjected
xiii
Abstract
to impact loadings; hence simplified analytical models were employed to predict

the same in less modeling effort and computational time.
Residual performance assessment of damaged structures has been gaining

enormous importance in engineering community to ensure that the damaged
structures will not fail catastrophically and uphold structural integrity. However, to
date no study has been documented in literature on residual performance of impact-
damaged RC beams. Thus, the impact-damaged beams have been retested quasi-
statically to determine their post-impact residual responses. Finally, the FE models
developed earlier for drop-weight impact loading were modified by replacing the
impactor with the loading plate to perform quasi-static simulation by utilizing
resulting deformation and damages of beams from the impact stages. Thereafter,
these models were used further to quantify the effect of various parameters on
residual resistance index (RRI), which could be used effectively to delineate the
extents of damage of beams after impact.
xiv
List of Tables
LIST OF TABLES
Table 3.1: Specimen details, theoretical static capacities, static expected and
observed failure modes ............................................................................................ 53
Table 3.2: Comparison between the time to reach peak load and the yielding of
tensile and shear reinforcements .............................................................................. 65
Table 4.1: Total number of elements and nodal points for FE model ..................... 78
Table 4.2: Comparison of peak resistance of beam between test and FE analysis
results ....................................................................................................................... 90
Table 4.3: Specimen characteristics of simulation matrix ...................................... 94
Table 4.4: Beam characteristics for numerical case-studies ................................. 103
Table 5.1: Test program ........................................................................................ 113
Table 5.2: Details of specimens and static design parameters .............................. 117
Table 5.3: Type of impactor and the interface between the beam and impactor .. 139
Table 6.1: Total number of elements and nodal points for FE model ................... 149
Table 6.2: Comparison of maximum impact load of beam between test and FE
analysis results ....................................................................................................... 157
Table 6.3: Comparison of maximum midspan deflection of beam between test and
FE analysis results .................................................................................................. 162
Table 6.4: Maximum reaction force obtained from numerical simulation ........... 175
Table 6.5: Variations in mass ratios ...................................................................... 176
Table 6.6: Variation in longitudinal reinforcement ratios ..................................... 178
Table 7.1: Actual and effective mass of beam ...................................................... 192
Table 7.2: Modified material properties ................................................................ 196
Table 8.1: Parameter variations in case-studies .................................................... 220
xv
xvi
List of Figures
LIST OF FIGURES
Fig. 1.1: Spectrum of strain rates corresponds to different loading environments

(Bischoff and Perry [B11]) ........................................................................................ 2
Fig. 1.2: Test configurations of RC beams under varying loading rates (Adhikary et
al. 2012 [A3], Kulkarni and Shah [K13]) .................................................................. 3
Fig. 1.3: Impact loading test setups (Kishi et al. [K7], Gopalaratnam et al.
[G1]) ........................................................................................................................... 3
Fig. 1.4: The different tendencies between impact, high rate loading and static
loading (Takeda et al. [T3]) ....................................................................................... 5
Fig. 2.1: Relative enhancement of flexural and shear strengths of RC beams

(Kulkarni and Shah [K13]) ...................................................................................... 24
Fig. 2.2: Missile impact phenomena (Kennedy [K3]) ............................................. 26
Fig. 2.3: Failure mode of the RC beams (Mylrea [M12]) ....................................... 27
Fig. 2.4: Impact test set up and hysteretic loop between reaction force and mid-span
displacement of beam (Kishi et al. [K5]) ................................................................. 28
Fig. 2.5: Impact test setup (Kishi et al. [K7]) .......................................................... 29
Fig. 2.6: Crack patterns for beams (a) A36 and (b) B36 (Kishi et al. [K7]) ........... 30
Fig. 2.7: Simplified loop for reaction force-midspan displacement of RC beams

being collapsed in shear-failure mode (Kishi et al. [K7])........................................ 30
Fig. 2.8: Drop-hammer impact test setup (Fujikake et al. [F4]) .............................. 31
Fig. 2.9: Failure modes: (a) S1616 series; (b) S1322 series; (c) S2222 series
(Fujikake et al. [F4]) ................................................................................................ 32
Fig. 2.10: Drop-weight impact test set up (May et al. [M6]) .................................. 33
Fig. 2.11: Post-test crack pattern of beams (a) [A1]; (b) [A2]; (c) [A3]; (d) [B1]; (e)
[B2]; (f) [B3] and (g) [B4] (Chen and May [C3]).................................................... 34
Fig. 2.12: Impact test set up (Saatci and Vecchio [S1]) .......................................... 35
xvii
List of Figures
Fig. 2.13: Final crack pattern of a-series and b-series beams (Saatci and Vecchio
[S1]) ......................................................................................................................... 36
Fig. 2.14: Impact test set up (Tachibana et al. [T1]) ............................................... 37
Fig. 2.15: Crack pattern of type A2 beam (Tachibana et al. [T1]) .......................... 38
Fig. 2.16: Proposed equation by Tachibana et al. [T1] ........................................... 38
Fig. 2.17: Crack patterns in beams of series G2L, G5 and G10 (Kishi and Mikami
[K10]) ....................................................................................................................... 40
Fig. 2.18: Regression curves: (a) regression curve between coefficient of maximum
deflection  def and static flexural load-carrying capacity Pusc ; (b) regression curve
between coefficient of residual deflection  rs and static flexural load-carrying
capacity Pusc by Kishi and Mikami [K10] ................................................................ 41
Fig. 3.1: Influence of (a) shear span to effective depth ratio ( a d ); (b) longitudinal
reinforcement ratios (  L ); (c) failure modes on DIF of singly reinforced (SR)
beams........................................................................................................................ 47
reinforcement ratios (  L ); (c) shear reinforcement ratios (  T ); (d) failure modes on
DIF of doubly reinforced (DR) beams ..................................................................... 48
Fig. 3.3: Flow chart of test program ........................................................................ 51
Fig. 3.4: Dimensions of RC beams, layout of reinforcements, location of strain

gauges and accelerometers ....................................................................................... 53
Fig. 3.5: (a) Casting of beams; (b) preparation of concrete cylinders ..................... 54
Fig. 3.6: (a) Strain gauging in tensile longitudinal reinforcements; (b) Strain
gauging in transverse reinforcement ........................................................................ 55
Fig. 3.7: Schematic diagram of test set up ............................................................... 56
Fig. 3.8: (a) Typical variation of force history from top and bottom load-cells under
high loading rates ..................................................................................................... 57
xviii
List of Figures
Fig. 3.9: Comparative plot of true resistance history obtained from Eqtn. 3-1 and
approaches proposed by Banthia et al. [B2]............................................................. 58
Fig. 3.10: Load vs. mid-span deflection of (a) a d -3.3; (b) a d -4.4 series
specimens under varying loading rates .................................................................... 59
Fig. 3.11: Dynamic Increase Factor (DIF) of maximum resistance of (a) a d -3.3;
(b) a d -4.4 series specimens under varying loading rates ...................................... 60
Fig. 3.12: Concrete contribution to the shear stress ................................................ 61
Fig. 3.13: Concrete and shear reinforcement contribution to shear resistance ........ 62
Fig. 3.14: Cracking stiffness of (a) a d -3.3; (b) a d -4.4 series beams under
varying loading rates ................................................................................................ 63
Fig. 3.15: Energy absorption of (a) a d -3.3; (b) a d -4.4 series beams ................ 64
Fig. 3.16: Peak strain rates of (a) longitudinal (b) shear reinforcements for all
specimens ................................................................................................................. 65
Fig. 3.17: Deflected shape of (a) a d -3.3; (b) a d -4.4 series specimens under high
loading rates ............................................................................................................. 66
Fig. 3.18: Crack patterns of (a) a d -3.3; (b) a d -4.4 series specimens under
varying loading rates ................................................................................................ 68
Fig. 3.19: Development of cracks in SR3.3_0.84 under high loading rates ............ 69
Fig. 3.20: Development of cracks in DR3.3_0.84_0.11 under high loading rates .. 70
Fig. 3.21: Development of cracks in DR3.3_0.84_0.15 under high loading rates .. 71
Fig. 4.1: Three-dimensional FE model of RC beams (a) a d -3.3;
(b) a d -4.4 series .................................................................................................... 79
Fig. 4.2: General shape of the concrete model yield surface in two dimensions .... 82
Fig. 4.3: Proposed Dynamic Increase Factor (DIF) for reinforcing steel (Malvar
[M2]) ........................................................................................................................ 84
xix
List of Figures
Fig. 4.4: Comparison of load vs. midspan deflection of beams under varying
loading rates: (a) 3.3 series; (b) 4.4 series ................................................................ 85
Fig. 4.5: Comparison of load vs. midspan deflection of RC3_S56 under varying
loading rates (Adhikary et al. [A3]) ......................................................................... 89
Fig. 4.6: Comparison of crack pattern from experiment and fringes of effective
plastic strain from simulation of beams under varying loading rates ...................... 91
Fig. 4.7: General schematic diagram of beam ......................................................... 93
Fig. 4.8: Influence of longitudinal reinforcement ratio on DIF of maximum

resistance of beams (a) a d -2.8; (b) a d -3.3; (c) a d -4.4 .................................... 95
Fig. 4.9: Influence of longitudinal reinforcement ratio on failure mode: (a) a d -

3.3; (b) a d -4.4 ....................................................................................................... 98
Fig. 4.10: Influence of transverse reinforcement ratio on DIF of maximum

resistance of beams (a) a d -2.8; (b) a d -3.3; (c) a d -4.4 .................................... 99
Fig. 4.11: Influence of transverse reinforcement ratio on failure modes of a d -3.3

series beams under medium and high loading rates ............................................... 100

series beams under medium and high loading rates ............................................... 101
Fig. 4.13: Influence of yield strength of longitudinal reinforcement on failure

modes of a d -3.3 series beams under varying loading rates ................................ 104

modes of a d -4.4 series beams under varying loading rates ................................ 105
Fig. 4.15: Influence of grade of concrete on peak load of beams under varying
loading rates: (a) a d -2.8; (b) a d -3.3; (c) a d -4.4 ........................................... 106
Fig. 5.1: Impact loading scenarios on structures ................................................... 110
Fig. 5.2: Impact responses of a RC member (Fujikake et al. [F4]) ....................... 111
Fig. 5.3: Static loading test set up .......................................................................... 113
xx
List of Figures
Fig. 5.4: Drop-weight impact loading test set up .................................................. 114

gauges and accelerometers ..................................................................................... 115
Fig. 5.6: (a) Casting of beams; (b) preparation of concrete cylinders ................... 118
gauging in transverse reinforcements..................................................................... 119
Fig. 5.8: Static load vs. midspan responses of beams (a) 3.8 series;
(b) 5.7 series ........................................................................................................... 120
Fig. 5.9: Failure pattern of beams under static loading (a) 3.8 series;
(b) 5.7 series ........................................................................................................... 121
Fig. 5.10: Time histories of impact load (a) 3.3 series; (b) 3.8 series;
(b) 5.7 series ........................................................................................................... 122
Fig. 5.11: Time histories of midspan deflection (a) 3.3 series; (b) 3.8 series; (b) 5.7
series ....................................................................................................................... 124
Fig. 5.12: Maximum impact load under various drop heights of (a) 3.3 series; (b)
3.8 series; (b) 5.7 series specimens ........................................................................ 126
Fig. 5.13: Maximum midspan deflection under various drop heights of (a) 3.3
series; (b) 3.8 series; (b) 5.7 series specimens ....................................................... 126
Fig. 5.14: Time to reach maximum midspan deflection under various drop heights
of (a) 3.3 series; (b) 3.8 series; (b) 5.7 series specimens ....................................... 127
Fig. 5.15: Crack patters of beams (a) 3.3 series; (b) 3.8 series; (b) 5.7 series ....... 133
Fig. 5.16: Variation of strain rate in longitudinal tensile reinforcements .............. 135
Fig. 5.17: Variation of strain rate in transverse reinforcements ............................ 136
Fig. 5.18: Range of maximum strain rates of longitudinal tensile reinforcements for
beams of various series under various drop heights............................................... 137
Fig. 5.19: Relationship between maximum midspan deflections vs. input impact
energy over static flexural resistance ..................................................................... 140
xxi
List of Figures
energy over static shear resistance ......................................................................... 141
Fig. 5.21: Comparison of current test results with proposed equation for flexure-
critical beams ......................................................................................................... 142
Fig. 5.22: Comparison of current test results with proposed equation for shear-
critical beams ......................................................................................................... 143
Fig. 6.1: Three-dimensional FE model of RC beams (a) a d -3.3; (b) a d -3.8; (c)
a d -5.7 series ........................................................................................................ 150
Fig. 6.2: Variation of various energies in a typical impact simulation .................. 153
Fig. 6.3: Comparison of time histories of impact load of beams for various drop
heights (a) 3.3 series; (b) 3.8 series; (b) 5.7 series ................................................. 154
heights (a) S1616; (b) S1322 ................................................................................. 159
Fig. 6.5: Comparison of time histories of midspan deflection of beams for various
drop heights (a) 3.3 series; (b) 3.8 series; (b) 5.7 series......................................... 160
drop heights (a) S1616; (b) S1322 ......................................................................... 163
Fig. 6.7: Comparison of crack profile from experiment and fringes of effective
plastic strain from simulation of DR3.8_0.8_0.11 at a drop height of 0.6 m ........ 165
Fig. 6.11: Typical variation of force history obtained from top and bottom load-
cells during drop-weight impact loadings .............................................................. 173
xxii
List of Figures
Fig. 6.12: Typical variation of impact and reaction load history (from numerical
simulation) ............................................................................................................. 174
Fig. 6.13: Effect of mass ratios on maximum impact load and maximum midspan
deflection of beams (a) a d -3.3; (b) a d -3.8; (c) a d -5.7 ................................. 177
Fig. 6.14: Effect of longitudinal reinforcement ratios on maximum impact load and
maximum midspan deflection of beams ................................................................ 178
Fig. 6.15: Effect of longitudinal reinforcement ratios on failure pattern of beam

under impact energy of 3.53 kJ .............................................................................. 179
Fig. 6.16: Effect of compressive strength of concrete on maximum impact load and
maximum midspan deflection of beams (a) a d -3.3; (b) a d -5.7 ....................... 180
Fig. 6.17: Effect of boundary conditions on maximum impact load and maximum
midspan deflection of beams (a) a d -3.3; (b) a d -3.8; (c) a d -5.7 ................... 181
Fig. 6.18: Effect of boundary conditions on failure pattern of beams (left: pinned
end; right: fixed-end).............................................................................................. 182
Fig. 7.1: Comparison of maximum midspan deflection of beams for various drop
heights (a) 3.3 series; (b) 3.8 series; (b) 5.7 series ................................................. 188
Fig. 7.2: Comparison of maximum midspan deflection of all tested beams ......... 190
heights (a) S1616; (b) S1322 ................................................................................. 191
Fig. 7.4: Idealized SDOF model ............................................................................ 192
Fig. 7.5: Idealized resistance function of beam with and without strain rate ........ 193
Fig. 7.6: Resistance function of beam for static and various strain rate cases: (a) 3.8
series; (b) 5.7 series ................................................................................................ 197
Fig. 7.7: Comparison of midspan displacement and time histories between

experimental and elastic-plastic SDOF analysis results:
(a) 3.8 series; (b) 5.7 series ................................................................................... 198
Fig. 7.8: Comparison of midspan displacement and time histories between

experimental and elastic-plastic SDOF analysis results:
xxiii
List of Figures
(a) S1616; (b) S1322 .............................................................................................. 199
Fig. 8.1: Post impact residual resistance test set up ............................................... 205
Fig. 8.2: Quasi-static load vs. midspan deflection response of undamaged and
impact-damaged beams of 3.3 series ..................................................................... 206
Fig. 8.5: Residual Resistance Index (RRI) of RC beams (a) 3.3 series; (b) 3.8 series
and (c) 5.7 series under various impact energies ................................................... 210
Fig. 8.6: Residual Stiffness Index (RSI) of RC beams (a) 3.3 series; (b) 3.8 series
and (c) 5.7 series under various impact energies ................................................... 212
Fig. 8.7: Crack pattern of impact-damaged RC beams (a) 3.3 series; (b) 3.8 series
and (c) 5.7 series under various drop-heights ........................................................ 214
Fig. 8.8: FE models of beam (a) impact loading stage; (b) post impact residual
resistance stage ....................................................................................................... 215
Fig. 8.9: Validation of impact load history and post impact residual resistance vs.
midspan deflection ................................................................................................. 217
Fig. 8.10: Comparison of cracking pattern from experiment and fringes of effective
plastic strain from simulation of beams after impact and residual test .................. 219
Fig. 8.11: Effect of mass ratio (  ) on RRI under various impact energies .......... 221
Fig. 8.12: Effect of longitudinal reinforcement ratio (  L ) on RRI under various

impact energies....................................................................................................... 222
Fig. 8.13: Effect of transverse reinforcement ratio (  T ) on RRI under various

impact energies....................................................................................................... 223
Fig. 8.14: Effect of compressive strength of concrete ( f c ) on RRI under various

impact energies....................................................................................................... 225
xxiv
List of Symbols
LIST OF SYMBOLS
 Strain rate
s Static strain rate
DIFc / t Dynamic Increase Factor in compression or tension for concrete
DIFc Dynamic Increase Factor in compression for concrete
DIFt Dynamic Increase Factor in tension for concrete
DIFs Dynamic Increase Factor for reinforcing steel
f cs Static compressive strength of concrete
fy Static yield strength of reinforcement
Pusd , Pu Static bending resistance
Ekd , Ecol , E Input impact energy
 rd Residual displacement
Vusd Static shear resistance
 max , Dmax Maximum displacement
b Balanced reinforcement ratio

a Shear span of beam
d Effective depth of beam section
ad Shear span to effective depth ratio
L Longitudinal reinforcement ratio
T Transverse or shear reinforcement ratio
vc Concrete contribution to shear stress
Vc Concrete contribution to shear resistance
Vs Transverse reinforcement contribution to shear resistance
f c Compressive strength of concrete

h Overall depth of beam section
b Width of beam section
L Total span length of beam
xxv
List of Symbols
lclear Clear span length of beam
Rb t  True resistance of beam without inertia force
PTL t  Load measured by top load-cell

 Density of reinforced concrete
xb t  Acceleration at midspan of beam
 LT Compressive longitudinal reinforcement ratio

 LB Tensile longitudinal reinforcement ratio
Rs Static shear resistance
Rb Static bending resistance
 Loading rate
f Ly Yield strength of longitudinal reinforcement
f Ty Yield strength of transverse reinforcement
J1 First invariant of stress tensor

J 2 Second invariant of deviatoric stress tensor
J 3 Third invariant of deviatoric stress tensor
S ij Deviatoric stress tensor
P Pressure
 Cap hardening parameter
Ff Shear failure surface
Fc Hardening cap
 Rubin three-invariant reduction factor
mb Mass of beam
mi , M Mass of impactor
V Velocity just before impact
K bs Stiffness including bending and shear
Km Membrane stiffness
n Contact stiffness
Ri Radius of impactor
xxvi
List of Symbols
i Poisson‟s ratio of impactor
Ei Young modulus of impactor
Eb Young modulus of beam
Wmax Maximum deflection
me Effective mass of beam

h Drop-height of the impactor
Rm Resistance force at yielding of the beam
ue Elastic limit displacement
xxvii
xxviii
Chapter 1: Introduction
CHAPTER 1
INTRODUCTION
1.1 Research Background
Reinforced Concrete (RC) structures during their service life may be subjected to
various types of dynamic loading in the form of earthquakes, impact or blasts.
Structures in low to moderate seismic or earthquake-prone regions could be
subjected to dynamic loading due to ground oscillations. Furthermore, with the
rapid development of the infrastructural project around the world, there is a chance
that RC structural elements or members may be experienced from various types of
impact loading due to several events or accidents. Typical examples comprise in a
diversified fields ranging from transportation structures (e.g. bridge piers, guard
rails, traffic signal posts and electric poles etc.) subjected to vehicle-crash impact,
falling rocks on rock-sheds in mountainous regions, falling heavy loads on
industrial facilities due to accidents or mishaps resulting from pipe and turbine
breaks, marine and offshore structures exposed to ship and ice impact or subjected
by tornado or tsunami-borne debris impact, columns in multi-story car park or
bridge-pier strike by moving vehicle, protective structures subjected to projectiles
or aircraft impact. Thus, in all these above-mentioned cases, it is of fundamental
importance to understand the effect of loading rates on structures. Bischoff and
Perry [B11] addressed the wide spectrum of strain rates corresponds to different
loading conditions typically encountered in practice, as shown in Fig. 1.1. During
creep deformations, very low strain rates (~10-8 to 10-7 /s) usually generate. For
static loading, strain rates usually ranged from 10-6 to 10-4 /s. Moreover, for the case
of seismic loading, it could be in the domain of 10-3 to 1 /s and for hard impact, it
would be in the range of 1 to 50/s. Strain rates greater than the hard impact region
fall in the blast loading range. In the earthquake range of strain rates, inertial effects
on structural elements are generally considered insignificant. However, for strain
rates higher than the seismic regime, inertial effects dominate the structural
response (Kulkarni and Shah [K13]).
-1-
CREEP STATIC EARTHQUAKE HARD BLAST

IMPACT
-3
10-8 10-7 10-6 10-5 10-4 10 10-2 10-1 1 101 102 103 104
STRAIN RATE (sec-1)
Fig. 1.1: Spectrum of strain rates corresponds to different loading environments

(Bischoff and Perry [B11])
A number of devices or machines are available to perform dynamic tests on

structural elements in laboratory for a particular range of strain rates. Kulkarni
[K12] summarized the details of different machines or impact loading test setups,
suitable for a specific range of strain rates. Hydraulic machines are ideally used to
generate the wide range of loading rates extending from 4 × 10-4 to 2 m/s
Corresponding strain rate induced in RC beams were ranging from 10-4 to 10/s.
However, it could be possible to design hydraulic machines having piston velocities
up to 12.5 m/s which would correspond to strain rate of about 50/s. In such testing,
some steel plates or spreader beams are usually placed in between the specimens
and piston for the case of 3-point or 4-point bending configurations respectively
(Figs. 1.2a and 1.2b). The piston is generally maintained contact with the
specimens via steel plates or spreader beams throughout the test so that sudden
contact or impact could be avoided. For higher piston velocities, pneumatic
machines could be utilized in which compressed air is used to drive the piston.
Attainable piston velocities from this machine might be in the order of 30 m/s
which would result in a strain rate of 120/s. Till now, discussions are made on the
machines which could generate different range of loading rates which
correspondingly produces varying range of strain rates in the structural elements.
However, to precisely simulate a natural impact scenarios, real impact tests (i.e.
striker or impactor strikes the structural member with a certain initial velocity) are
indispensable. These experimental studies attempt to replicate the real impact
loading conditions to be experienced by the actual structures to evaluate their
impact performance. Strain rates develop in structural elements could be in the
range of 1 to 10/s due to the aftermath of the low velocity impact event.
-2-
(a) 3-point (b) 4-point
Fig. 1.2: Test configurations of RC beams under varying loading rates (Adhikary et
al. [A3], Kulkarni and Shah [K13])
Impact tests can be classified mainly into two categories such as drop-weight type
and pendulum type, differing in the way the corresponding hammer strikes the
specimen. In the case of first type, impactor having known weight is dropped on a
specimen from a certain drop-height. In a well-instrumented drop-weight impact
test-setup, both global and local structural responses (i.e. impact, reaction force
history, mid-span deflection and strain history of reinforcements) could be captured
(Fig. 1.3a).
(a) Drop-weight impact setup (b) Charpy test setup
Fig. 1.3: Impact loading test setups (Kishi et al. [K7], Gopalaratnam et al. [G1])
-3-
A pendulum-type hammer is used to deliver impact for the second one by releasing
the swing hammer from a certain drop-height. Conventional Charpy impact test is a
standardized test which determines the amount of energy required for fracturing the
specimen. Gopalaratnam et al. [G1] described the modified instrumented Charpy
test in evaluating the dynamic behavior of cement-based composite (Fig. 1.3b)
specimens. However, pendulum type hammer can also be used for large scale
testing of structural components (Orozco [O1]). Furthermore, it is noted that
impulsive loading on structures can also be generated by gas gun and contact blast
loading.
RC structures are composite constructs, constituting plain concrete and steel

reinforcements. It is well known that constitutive characteristics of plain concrete
and steel reinforcements are strain rate dependent. Resistance, stiffness and ductility
(or brittleness) of structural members can be affected by the loading rates but, in
most cases, the difference only become significant when the rate alters by almost
one order magnitude. Therefore, the response of RC structures to varying rates of
concentrated loading depends on the magnitude of loading rates and the rate
sensitivity of the materials used. Some researchers have pointed out the influence of
loading rate on the modes of failure, cracking patterns and energy absorption
characteristics of RC structural elements. One of the important phenomena
perceived in RC beams under high rates of concentrated loading (by a servo-
hydraulic machine with a maximum piston velocity of about 0.75 m/s) is the
possible change in the failure mode of a beam from a flexure type of failure to a
shear type (Mutsuyoshi and Machida [M11]). A beam designed to fail in flexure in
a static test might fail in shear at higher rates of loading. However, an opposite
phenomenon (i.e., transition in the mode of failure from shear failure at the static
rate to flexure failure at the high loading rate) has been observed by Kulkarni and
Shah [K13]. In this investigation, for high rates loading, piston velocity was
considered to be 0.38 m/s, which can be related to strain rates expected in an
earthquake event.
Drop-weight impact loading cannot be merely adjudged as an extreme case of high

rate loading because of the involvement of severe inertia forces, energy transfer
-4-
mechanism and complex pattern of stress waves. In this case, several key
parameters such as mass, initial velocity of the impactor and the stiffness of the
contact region affect the structural behavior. Moreover, under impact loading, the
structural responses are divided into two components: primary response (local
response) and secondary response (overall response) and these responses are
influenced simultaneously by rate effects and the propagation of stress waves. Fig.
1.4 shows the different tendencies of increase of the stress wave generated and the
rate effects incurred in the structural members for three loading cases: impact, high
rates and static loading (Takeda et al. [T3]).
Impact
Stress wave
increase
Rate effect
increase
Static High rate

loading loading
Fig. 1.4: The different tendencies between impact, high rate loading and static
loading (Takeda et al. [T3])
It is generally admitted that there is an apparent increase in strength when the

concrete-like material is subjected to high strain rate. The dynamic increase factor
(DIFc/t) defined by the ratio of dynamic strength to quasi-static strength has been
widely adopted as an important parameter to measure the rate-dependent material
behavior. There are several hypotheses on physical mechanism that explains the
concrete material response under varying strain rates. One of them states that the
strain rate effects observed is due to the presence of water that may influence the
response of the concrete under high strain rates (Rossi et al. [R5]). It is attributed
that Stefan effect on free water within the concrete may change the cracking
patterns under high loading rates. Under quasi-static loading, cracks occur through
the cement matrix and when it encounters the coarse aggregates, the cracks will
continue to propagate around the boundaries of the aggregate. On the other hand in
dynamic loading, due to Stefan effect cracks would penetrate through the coarse
-5-
aggregates and it may be the reason behind the strength enhancement. Another
angle of view is the lateral inertia force effect which causes an apparent increase in
DIFc/t for concrete. Li and Meng [L2] reported that the increase of dynamic
compressive strength could be only caused by the lateral confinement when the
strain rate is higher than around 102 s-1. They termed this lateral inertia confinement
as pseudo-strain-rate-effect and further opined that the acceptance of this strain rate
effect in design and numerical models may overestimate the dynamic compressive
strength of concrete. Zhou and Hao [Z1] developed homogeneous and mesoscale
model with strain rate sensitive material model to analyze concrete-like material
under high strain-rate compression which corroborate reasonably with test results.
Comparison of DIFc caused by lateral confinement and the DIFc obtained from
dynamic tests shows that the inertial confinement is only one of the two sources
that contribute to the DIFc and this contribution becomes more significant when the
strain rate is higher than 1000 s-1. Material strain rate effect cannot be neglected in
modeling concrete material response to high loading rates, especially when the
strain rate is less than 200 s-1. Most recently Ozbolt et al. [O5] performed some
numerical studies of compact tension specimen (CTS) loaded by varying rates.
After validating the numerical results with experimental results they commented
that for strain rate lower than approximately 50 s-1, the structural response is
controlled by rate-dependent constitutive law. However, for higher strain rate crack
branching and progressive increase in resistance is observed. This is attributed to
the effect of structural inertia and not the rate dependent strength of concrete. From
numerical point of view, assuming micro and mesoscale analysis, the effects of the
rate dependency of the growing micro cracks (e.g. influence of inertia at micro
crack level) and viscous behavior of bulk material between the cracks (e.g. viscosity
due to water content) can be accounted by the constitutive law. Whereas, the
structural inertia effect would be automatically accounted through dynamic analysis
(Ozbolt et al. [O4)]
1.2 Objectives and Scopes
 Experimental outcomes are essential to develop and verify finite element (FE)
and analytical methods and these approaches could be employed further to
-6-
investigate the structural behavior for various case studies. Previous studies
(Mutsuyoshi and Machida [M11] and Kulkarni and Shah [K13]) on high
loading rates of RC beams were confined up to the earthquake loading regime.
Fujikake et al. [F4] considered a much higher piston velocity for high loading
rates which could be linked to the impact loading regime; however consider
only one test parameter (i.e. flexural reinforcement ratio). Thus, there is a lack
of systematic research on RC beams under wide range of varying loading rates
to cover static loading to hard impact regime. To fulfill this, hydraulic machine
was used to generate the wide range of loading rates (4 × 10-4 m/s to 2 m/s) and
the effects of these loading rates on RC beams (in three-point bending
configurations) have been studied. Corresponding strain rate would be in the
range of 10-4/s to 10/s (enclosed by an ellipse in Fig. 1.1). An effort has been
made by collecting data (through comprehensive literature review) on beams
under varying loading rates to observe the overall trend of dynamic increase
factor (DIF) of maximum resistance. Furthermore, to extend the findings from
tests, FE model is developed to study the behavior of beams under varying
loading rates. It is worthwhile to mention that all tests were monotonic in nature
and cyclic deflection reversal case is out of scope of this research. Moreover,
the influence of the characteristics of load itself (the relationship between the
magnitude of load and time) is not taken into account.
 Another objective of this research is to enhance the understanding of RC beams

under drop-weight impact loading which can able to replicate the actual impact
scenarios to be experienced by the structures. Although, numerous researches
(Hughes and Beeby [H3]; Banthia [B1]; Kishi et al. [K5-K9]; Chen and May
[C3]; Fujikake et al. [F4]; Bhatti et al. [B8]; Saatci and Vecchio [S1];
Tachibana et al. [T1]; Kishi and Mikami [K10]) have been conducted on
evaluating the impact response of RC beams; however till now limited efforts
have been sought towards determining the residual performance of beams after
impact damage. So, at first, well-instrumented experimental program was
planned to impart impact loading to RC beams by drop-hammer and their
corresponding responses for different drop-heights have been summarized.
Thereafter, the impact-damaged specimens were shifted to 3-point bending test
-7-
setup to obtain their quasi-static residual performances. Subsequently, some

undamaged specimens were tested under monotonic static loading to compare
their response to the impact-damaged specimens. Moreover, set of data (RC
beam under drop-weight impact) is assembled and analyzed to propose
empirical equations which would aid in determining the required static bending
and shear resistance for input impact energy by specifying the maximum
midspan deflection for each limit state of beam. Maximum midspan deflection
would be an important performance index to evaluate the damage levels (i.e.
from no damage to moderate damage, severe damage and complete collapse) of
beam when subjected to impact loading (Fujikake et al. [F4]). Hence, two
analytical approaches such as energy-balance and single-degree-of-freedom
(SDOF) model have been employed to predict the maximum midspan deflection
of beam. Moreover, a numerical procedure has also been developed in explicit
finite element (FE) program LS-DYNA [L1] to reproduce the two stage
experiments (the impact and 3-point quasi-static loading tests) which could be
valuable in providing some insight about the impact responses as well as the
post impact residual performances of damaged specimens. Finally, it is essential
to enumerate that impulsive loading on structures caused by gas gun and contact
blast loading is not within the scope of the research.
1.3 Structure of Thesis

The thesis comprises of nine chapters and three appendices. The contents of the
chapters are briefly illustrated as follows:
Chapter 2 presents a comprehensive survey of available literature in the field of RC

beams subjected to varying rates of concentrated and drop-weight impact loading.
To understand the structural response under these above-mentioned loading
conditions, knowledge on mechanical properties of steel and plain concrete under
various ranges of strain rates is essential. Therefore, a brief review of strain rate
effects on constitutive characteristics of steel and plain concrete was accomplished
in parallel.
Chapter 3 focuses on experimental investigation of loading rate effects on RC
-8-
beams which consists of the detailing of the specimens, experimental program,

material properties, instrumentations and test results and discussions. Comparative
analyses of beams under these varying loading rates highlighted several key aspects
of their dynamic behavior.
Three dimensional non-linear FE simulations of beams under varying loading rates

are presented in Chapter 4. This FE model is then used to conduct some case
studies to quantify the effects of various parameters on DIF of maximum resistance
and failure modes.
Chapter 5 describes the experimental assessment of RC beam under drop-weight

impact loading. This chapter elaborates the beam details, test program,
instrumentation; material properties and results. In addition, empirical equations
have been formulated by analyzing a pool of data which would aid in determining
the required static bending and shear resistance for input impact energy by
specifying the maximum midspan deflection for each limit state of beam.
Chapter 6 discusses the development of the numerical model of beams subjected to

drop-weight impact loadings and parametric studies are carried out to determine the
effects of several parameters on impact responses.
Chapter 7 presents some analytical models aimed at attaining the key impact
responses of the beams in less modeling efforts and computational time.
Chapter 8 demonstrates both experimental and numerical evaluation of residual

performance of impact-damaged beams. A numerical procedure comprising two
steps is developed in explicit FE program to capture the residual resistance of
damaged specimens. These models are used further to quantify the effect of various
parameters on residual resistance index (RRI), which could be used effectively to
delineate the extents of damage of beams after impact loading.
Chapter 9 summarizes the conclusions drawn from the research work and provides
the recommendations for future study. Appendix A tabulates the database of RC
beams under various loading rates, accumulated from the literature. Similarly,
databank corresponds to RC beams subjected to drop-weight impact loading is
-9-
summarized in Appendix B and C.
- 10 -
Chapter 2: Literature Review
CHAPTER 2
LITERATURE REVIEW
2.1 Overview
To understand the behavior of RC beams under varying rates of concentrated and

drop-weight impact loading, efforts are devoted for both experimental and
numerical studies. This chapter presents a comprehensive survey of the strain rates
effect on the material behavior (i.e. plain concrete and steel reinforcement) and the
response of structural components (i.e. RC beam only). At first, a brief review of
available literature on strain-rate dependent mechanical properties of plain concrete
and reinforcing steel is performed as the behavior of these materials under dynamic
conditions differ significantly from their behavior under static conditions. Influence
of strain rates on bond properties are reviewed successively. This is followed by the
summary of the behavior of RC beams under varying rates of concentrated and
drop-weight impact loading.
2.2 Strain Rate Effect on Mechanical Properties of Materials
Since, the mechanical properties of plain concrete and steel are strain-rate
dependent, the behavior of structural members under various dynamic loading
conditions can only be accurately predicted by considering the rate dependent
properties of materials. Therefore, to evaluate the structural performances in terms
of resistance and behavior, the constitute properties of concrete and steel over a
wide range of strain rates are required.
2.2.1 Plain concrete
Three factors mainly influence the behavior of structures under varying loading
rates. These are (a) the rate dependency of growing micro crack; (b) viscous
behavior of bulk material between cracks; (c) structural inertia forces which can
significantly change the state of stresses and strains of the material [O3]. There are
various methodologies for accounting rate-dependent material behavior. Mihashi
- 11 -
and Wittmann [M8] employed the rate theory to explain the first two above-
mentioned reasons to explain the rate-sensitivity. This theory asserts that crack
initiation on an atomic scale is controlled by the activation energy. More bond
breaking occurs than bond healing when an external force acts on the material.
Long duration loading causes more bond-rupture than short duration loading as the
number of bond-breaking steps is assumed to be constant over time. This means
that sustained loading causes a reduction in strength whereas very short duration
loading causes an increase in strength. This is the logical explanation behind the
increase of strength of the material under high rates. The following expression has
been given to express increased strength in compression, tension and flexure:

f d f s    0 

(2-1)
where, f d is dynamic strength; f s is static strength under monotonic loading;  is
stress rate under dynamic load;  0 is stress rate under static load;  is the
parameter depends on load, types of material and way of loading. Reinhardt and
Weerheijm [R1] took into account the inertia effect at material level. The authors
look at a cluster of penny shaped cracks in a plane which is perpendicular to the
tensile loading direction. Cracks face move when the cracks propagate into the
material with a certain velocity and for this movement energy balanced is
computed. The calculation showed that the rate of energy supplied becomes too
high to be absorbed in the fracture process and thus major part of energy supplied is
stored at the crack tip. As a result, stress distribution around the crack tip changes
and the stress intensity factor decreases with increasing loading rates. This resulted
in the strength increases of material with increasing loading rates. In the model of
Bazant et al. [B5, B6], the influence of loading rates is considered in two parts: the
effect of viscosity is accounted by simple visco-elastic model and the influence of
strain-dependent growth of micro-cracks is based on the theory of activation
energy. Moreover, this model did not consider the influence of inertia forces. The
expression proposed is as follows:
- 12 -
  2 
 ( )   0 ( )1  C1 ln   (2-2)
  C2 
where,  is stress at dynamic loading;  0 stress at static loading;  is strain rate ;

C1 and C 2 are constant determine from experiments. Ozbolt et al. [O3] compared
the theoretical models with experimental results and from their plot, it can
concluded that Reinhardt and Weerheijm [R1] model is better supported by test data
than others. This has been recently demonstrated that for low and medium strain
rate up to 10 s-1, the total resistance is accounted by viscosity and strain dependent
crack growth. However, for high strain rates above 10 s-1, the inertia forces
dominate and cause progressive increase of structural resistance. Moreover,
dominant influence on failure mode and crack propagation is observed under high
rates [O4, O5].
Over the last several decades, numerous studies have been carried out on the
compressive behavior of concrete over a wide range of strain rates. Fu et al. [F2]
and Bischoff and Perry [B11] provides excellent synopsis on compressive behavior
of plain concrete subjected to varying strain rates by reviewing the existing
literature on it. Some key findings are summarized as follows: both compressive
strength and stiffness increase with increasing strain rates; however disharmony
prevails among researchers regarding the effect of strain rates on ultimate strain and
strain at peak stress; higher strain rates appear to have a more profound effect on
normal strength concrete than on high-strength concrete; dry concrete is less rate
sensitive than wet concrete; slope of the descending branch in the stress-strain
diagram increases with increasing rate of straining.
As compared to the compressive behavior of concrete over a wide range of strain

rates, test data of the dynamic behavior of concrete in tension is rather limited.
Several methods (split-cylinder tests, direct uniaxial tension tests, Split-Hopkinson
Pressure Bar [SHPB] tests etc.) usually are employed in laboratory to determine the
rate sensitivity of concrete in tension. Sauris and Shah [S3] investigated the rate
sensitivity of concrete in tension, compression and flexure using drop-weight
impact machine. It was concluded that the tensile strength of concrete is more
- 13 -
strain-rate sensitive than its compressive strength. Zielinski and Reinhardt [Z2]
used the Split-Hopkinson-Bar technique to examine the tensile behavior of concrete
and mortar at high stress rates (5000-30000 MPa/sec). It was concluded that the
significant increase in the tensile strength in mortar and concrete was due to the
excessive micro cracking in the whole volume of stressed specimens. Larger impact
strain was observed than static ones which confirmed the above-mentioned
hypothesis. Moreover, in impact loading, the specimens had fractures into three
pieces. The higher impact tensile strength of concrete than mortar was explained by
direct crack arresting action of the tougher aggregates which increased the amount
of energy absorbed in the impact tensile fracture process. Furthermore,
comprehensive review of strain rate effects for concrete in tension has been
performed by Malvar and Ross [M5].
The most comprehensive model for strain rate enhancement of concrete both in
tension and compression is presented by the CEB model code [C5]. Strain rate
effect on compression and tension is typically reported as dynamic increase factor
(DIFc/t) - i.e. ratio of dynamic to static strength. In compression CEB model
equations come out to be properly fit with the available data. The DIFc for
compressive strength is given by:
 ( s )1.026 s   30s 1

DIFc   (2-3)
 s ( s ) s  30s 1
1/ 3
where  is the strain rate in the range of 30  10-6 to 300 s-1; s  30  10-6 s-1
(static strain rate); log  s  6.156 s  2 ;  s  1 ( 5  9 f cs f co ) ; f co  10 MPa;
f cs is the static compressive strength of concrete.
Soroushian et al. [S8] proposed an equation for DIFc of concrete in compression by

fitting the test results reported by different researchers. The DIFc is depicted as
follows:
DIFc  1.48  0.160 log10   0.0127 log10  
2
(2-4)
where  is the strain rate (s-1) greater than 10-5. However, after observing the wide
spectrum of test data, the authors tried to figure out the main source of scatter.
- 14 -
Finally, they pointed out that the moisture content of concrete is the reason for the
variability of results. It was concluded that the strain rate effect on increasing the
compressive strength of concrete becomes more significant as the concrete moisture
content increases. The available test data did not show any considerable influence
of strain rates on static compressive strength of concrete. Furthermore, the effect of
strain rate on concrete compressive strength was found to be independent of the age
of the specimens if their moisture contents are identical. DIFc for both dry and wet
concrete were suggested as follows:
For dry concrete,

DIFc  1.48  0.206 log10   0.0221log10  
2
(2-5)
For wet concrete,

DIFc  2.54  0.580 log10   0.0543log10  
2
(2-6)
Ross et al. [R2, R3] and Tedesco and Ross [T5] conducted a series of SHPB test to
investigate the effect of strain rates and moisture content on concrete strength. The
DIFc equations for compression suggested by Tedesco and Ross [T5] are as follows:
DIFc  0.00965 log10   1.058  1.0 for   63.1 s-1 (2-7)

and
DIFc  0.758 log10   0.289  2.5 for   63.1 s-1 (2-8)
An experimental examination of the dynamic behavior of concrete and mortar at

very high strain rate and under high hydrostatic pressure was reported by Grote et al.
[G2]. Quasi-static compression, SHPB and plate impact experiments were used
involving strain rate from 10-3 to 104 s-1 and confining pressure from 0 to 1.5 GPa.
Following formulae is suggested to measure the strain-rate dependent DIFc.
DIFc  0.0235 log10   1.07 for   266 s-1 (2-9)

and
DIFc  0.882log10    4.4log10    7.22log10    2.64 for   266 s-1 (2-10)

3 2
- 15 -
Li and Meng [L2] examined the application of SHPB to determine the dynamic
strength of concrete-like materials whose compressive strength is hydrostatic-stress-
dependent. It was shown that the apparent dynamic strength enhancement beyond
the strain-rate of 102 s-1 is strongly influenced by the hydrostatic stress effect due to
lateral inertia confinement. Following equations were derived to calculate DIFc in
compression.
DIFc  1  log10   3 0.03438 for   100 s-1 (2-11)

and
DIFc  8.5303  7.1372 log10   1.729log10     100 s-1

2
for (2-12)
Katayama et al. [K1] demonstrated a numerical simulation methodology to analyze

the local damage and dynamic response of structures composed of RC and /or other
geological material subjected to high explosive loads. Moreover, they simulated the
dynamic behavior of the RC structures subjected to impulsive loading (i.e.
velocity> 100 m/s) and numerical analysis results were compared to those reference
experiments. To incorporate strain rate effect into Drucker-Prager‟s equation for
compressive strength in concrete, they used the DIFc formula proposed by
Yamaguchi et al. [Y1].The corresponding equation is as follows:
DIFc  1.021  0.05076 log10   0.2583log10  

2
(2-13)
Shkolnik [S6] demonstrated the influence of high strain rates on the stress-strain
relationships, strength and modulus of elasticity using thermo fluctuation theory,
principle of accumulation and development of damages as well as nonlinear
behavior of concrete. The obtained equations describe and unify the influence of the
high strain rates on constitutive properties of concrete under various uniaxial
stresses. Comparative study has been performed with existing experimental results
and the general equation given by CEB model code [C5]. It was concluded that
substantial agreement found among them. The proposed DIFc formula in
compression is as follows:
DIFc  1  2kT Wa  2  c ln d s E0  E0 2 c  2  (2-14)
- 16 -
where, k is Boltzmann constant; kT  404 10 23 J at the usual temperature of test
( T  293 K);  c is the strain of breaking the interatomic bonds; E0 is tangent
modulus of elasticity; Wa is volume of atom; d , s are strain rates at dynamic and
quasi-static loading.
Zhou and Hao [Z1] developed homogenous and mesoscale model to analyze the
behavior of concrete-like material under high strain-rate compression. Both strain
rate insensitive and strain rate sensitive materials were considered in the numerical
model to quantify the relative contribution of inertia effect and strain rate effect on
the compressive strength DIFc. The proposed compressive DIFc is as follows:
DIFc  0.0225 log10   1.12 for   10 s-1 (2-15)

and
DIFc  0.2713log10    0.3563log10    1.2275 for 10    2000 s-1

2
(2-16)
Fujikake et al. [F5] carried out tri-axial rapid compressive loading tests on concrete
specimens to formulate the dynamic constitutive model for concrete. DIF c in
compression was proposed as follows:
0.006log sc 1.05

  
DIFc    (2-17)
 sc 
where, sc  12 10 5 s-1 and   10 s-1.
Uniaxial compressive test on concrete cubes with grade C30 and C50 was
performed by Li and Li [L3] using electro-hydraulic servo-controlled testing
machine (displacement-controlled load). Four strain rates (e.g., 10-5, 10-4, 10-3 and
10-2) were considered in this test program. Based on results, the DIFc of
compressive strength of grade C30 is reported as follows:
  
DIFc  1.0  0.0648 log c  (2-18)
 so 
Similarly, the DIFc of compressive strength of C50 was obtained as follows:
- 17 -
  
DIFc  1.0  0.0314 log c  (2-19)
 so 
where, c implied the strain rate of concrete and co denotes the quasi-static strain
rate of concrete.
Moreover, the DIFt for tension as per CEB model code [C5] is furnished as:
( s )1.016 s   30s 1

DIFt   (2-20)
  s ( s ) s  30s 1
1/ 3
where  is the strain rate in the range of 3x10-6 s-1 to 300 s-1; s  3  10-6 s-1 (static
strain rate); log  s  7.11  2.33 ;  s  1 ( 10  6 f cs f co ) ; f co  10 MPa; f cs is the

static compressive strength of concrete.
However, it was found by Malver and Ross [M5] that the available data in literature
and their additional new data on concrete in tension differed somewhat from the
CEB [C5] recommendations, mostly for strain rates beyond 1 s-1. Thus, they
modified the DIFt equations and the change in slope occurs to strain rate of 1 s-1
instead of 30 s-1. The proposed formulations then becomes
 ( s )   1s 1
DIFt   (2-21)
 ( s ) s  1s 1
1/ 3
where  is the strain rate in the range of 10-6 to 160 s-1; s  10-6 s-1 (static strain
rate); log   6  2 ;   1 ( 1  8 f cs f co ) ; f co  10 MPa; f cs is the static

compressive strength of concrete.
A series of dynamic tensile tests were conducted by Tedesco et al. [T4] and based
on the results they proposed following equations:
DIFt  0.1425 log10   1.833  1.0 for   2.32 s-1 (2-22)

and
DIFt  2.929 log10   0.814  6 for   2.32 s-1 (2-23)
- 18 -
Direct tension, splitting tensile and direct compression test of mortar and concrete
were performed by Ross et al. [R4] at strain rate 10 and 102 s-1 on a SHPB. Quasi-
static tests on the same kind of specimens were conducted using a standard material
testing machine so that DIFt values can be computed. The suggested DIFt in tension
is as follows:
   
3.373

DIFt  exp 0.00126 log10  d   (2-24)
  s  
where, s  10 7 s-1 and d ,s = subscript for dynamic and static loading.
An experimental investigation on the strain rate dependent behavior of concrete in

tension was carried out by Yan and Lin [Y3] in a servo-hydraulic testing machine.
The strain rates ranged from 10-5 to 10-0.3 s-1. Specimens were grouped according to
their water contents and temperature environment. By least square curve fitting of
the experimental data, the DIFt has been expressed as a linear function of strain rate
on the semi-log scale. Formulas for various groups are listed below:
For Group A:
DIFt  1.0  0.134 logt ts  (2-25)
For Group B:
DIFt  1.0  0.135 logt ts  (2-26)
For Group C:
DIF t  1.0  0.265 logt ts  (2-27)
For Group D:
DIFt  1.0  0.115 logt ts  (2-28)
where, t is strain rate in the range from 10-5 to 10-0.3 s-1; ts is the quasi-static
strain rate, 10-5 s-1. Comparison between Eq. (2-21) and Eq. (2-23) reveals that the
strength increases in fully saturated concrete are much greater than those
experienced by concrete with normal moisture content.
- 19 -
The proposed tensile DIFt as per Zhou and Hao [Z1] is as follows:
DIFt  0.0225 log10   1.12 for   0.1 s-1 (2-29)

and
DIFt  0.7325log10    1.235log10    1.6 for 0.1    50 s-1

2
(2-30)
Xiao et al. [X1] performed some dynamic tensile testing of plain concrete
specimens with axial strain rate ranging from 10-5 to 10-1 s-1. As compared to the
quasi-static strain rate of 10-5 s-1, the dynamic tensile strength of concrete at strain
rates of 10-4, 10-3, 10-2 and 10-1 s-1 increase 6.37%, 13.08%, 20.48% and 25.47%
respectively. They concluded that this result is almost similar to the several
researchers. From the test results, they suggested following equation:
DIFt  1.0  0.0653 logt ts  (2-31)
where, t is dynamic strain rate; ts is quasi-static strain rate (10-5 s-1)
Therefore, it is seen that there are plentiful of equations to calculate the DIF c/t of
concrete in compression and tension. Numerous factors may affect the constitutive
behavior and DIFc/t of concrete under varying strain rates such as mix proportion,
cement content, aggregate shape and size, water-cement ratios, age and curing
conditions etc. Thus, more test data are indeed needed in future to estimate the
DIFc/t more precisely.
2.2.2 Steel reinforcement
Several studies have been documented on the effect of strain rate on reinforcing
bars, structural steel and steel wires (Keenan and Feldman [K2]; Wakabayashi et al.
[W1]; Soroushian and Choi [S7]. A detailed review of the available work has been
conducted by Fu et al. [F3] and Malvar [M2]. According to Wakabayashi et al.
[W1], yield stress of steel bar increases with increasing strain rate but the behavior
in the strain hardening region is not affected largely by a strain rate. Soroushian and
Choi [S7] concluded that the yield strength of steel is more strain-rate sensitive than
the ultimate strength. The modulus of elasticity is independent of rate of straining.
- 20 -
According to them, the most important factor influencing the strain rate effects is
the static yield strength. The mechanical properties of steel with lower yield
strength are more strain rate-sensitive than higher one. Malvar [M2] narrated that
the DIFs of yield and ultimate stress is inversely related to the yield stress itself. A
formulation was proposed to find out the DIFs as a function of strain rate and yield
stress by fitting the available data in literature. This formulation is valid for yield
stresses in between 290 and 710 MPa and for strain rates in between 10-4 s-1 and 10
s-1. The formulation which gives the DIFs for both yield and ultimate stress is as
follows:
DIFs  ( / 10 4 ) (2-32)
where for yield stress,    fy ;  fy  0.074  0.04 f y 414 ; for ultimate stress,
   fu ;  fu  0.019  0.009 f y 414 ;  is the strain rate in s-1 and f y is static

yield strength of reinforcement in MPa.
2.3 Influence of Loading Rates on Bond Properties

Bond behavior in between reinforcement and concrete plays a vital role in
evaluating the response of RC structures. Hitherto, plentiful research has been
conducted to understand the bond phenomenon under static loading conditions.
However, for the case of bond behavior under varying strain rates, very less
information exists. Von and Reinhardt [V1], Takeda [T2], Yan [Y2] and
Weathersby [W2] made important contributions in this regard. Neither the bond
resistance nor the shape of the bond stress vs. displacement of plain reinforcing bars
is significantly affected by the loading rates. However, there is an immense
influence of the loading rate on the bond resistance and the bond stress vs.
displacement behavior of deformed bars. Bond stiffness also increases with
increasing strain rates (Von and Reinhardt [V1]). The following expression has
been given by Von and Reinhardt [V1] by analyzing the test data.

   
  (2-33)
 0 0 
- 21 -
where,  and  0 are the average bond stress belonging to certain displacement; 
and 0 are the respective loading rates;  0 and 0 can be conceived as static
reference value;  is a parameter which depends upon material and environmental

conditions and in this case  is a function of concrete quality and the reinforcement
0.7(1  2.5 )
displacements;   with  is the rebar displacement in mm and f c is
f c0.8
cube compressive strength in MPa and the relation is valid within the limits
0    0.2 mm.
Yan [Y2] designed a testing program aimed at attaining the bond behavior between
reinforcement and concrete under impact loading. It was concluded that for smooth
bars, the bond resistance is due to the chemical adhesion and the frictional force at
the interface and there exists a linear bond stress-slip relationship under both static
and high rate loading. Whereas, for deformed bars, bond stress-slip relationship
under high rates varies with time and is different at various points along the
reinforcements. High loading rates significantly increases the bond resistance
capacity. Not only high loading rates but also higher compressive strength of
concrete significantly increases the bond resistance capacity and fracture energy at
bond failure. At the same time, these factors greatly influence the stress distribution
in concrete, the slips at the interface between reinforcements and concrete, and the
crack development. Moreover, under high rates, the stress distribution along the
reinforcements is not uniform, and not even linear; there is more stress
concentration along the reinforcements than under static loading.
2.4 Research on RC Beams under Varying Loading Rates

This section will provide an overview on limited experimental and analytical
investigation of on loading rate effects on RC beams.
2.4.1 Experimental investigations
Mutsuyoshi and Machida [M11] tested twenty seven singly and doubly RC beams
under different loading rates, noting that the final mode of failure shifted from
- 22 -
flexure at a low rate to shear at high loading rate for two pairs of specimens having
a d - 4 and 5.7, where a - shear span and d - effective depth of the beams. Seven
pairs of singly reinforced beams were tested by Kulkarni and Shah [K13]. For each
pair, one beam was tested at a „static‟ rate (7.1 × 10-6 m/s) while the other at a „high‟
rate (3.8  10-1 m/s). In this study, the strain rates generated in the reinforcements of
beams by high loading rates can be correlated to the earthquake-induced strain
rates. At the high rate, there was no recognizable „yield point‟ and „yield plateau‟ in
the load-midspan deflection curves of a beam failing in flexure. Furthermore, for
the three pairs of beams ( a d - 4, 4.5 & 5) tested by them, an opposite phenomenon
(i.e. transition in the mode of failure from shear failure at the static rate to flexure
failure at the high loading rate) has been observed. It was concluded that the
difference in the observed failure tendencies in these studies might be attributed to
the strain rate effect on the yield strength of steel used in the two studies
(Mutsuyoshi and Machida [M11] and Kulkarni and Shah [K13]). From Fig. 2.1, it
is apparent that the flexural strength of RC beams used by Mutsuyoshi and Machida
[M11] increased at a much higher rate than the shear strength, indicating an
increased tendency towards shear failure at high rates. However, the converse is
true for RC beams used by Kulkarni and Shah [K13]. Two different grades of steel
used in these studies affects the relative capacity (e.g. flexure and shear strength) of
beams under varying strain rates which would be the reason behind the opposite
phenomena observed in changes of failure mode from static to high rates.
Furthermore, Fujikake et al. [F4] performed the testing of three pairs of doubly RC
beams with varying amount of longitudinal reinforcements, but the shear
reinforcement ratios were kept identical for all specimens. All beams were designed
as under-reinforced and with a shear to bending resistance ratio greater than one.
For each pair, one beam was tested at a “static” rate (5 × 10-4 m/s) while the other at
a “high” rate (2 × 100 m/s). As compared to Mutsuyoshi and Machida [M11] and
Kulkarni and Shah [K12], Fujikake et al. [F4] considered a much higher piston
velocity for high loading rates which could be linked to the impact loading regime.
The main intention of this test program was to validate loading rate dependent
analytical model which would aid in to acquire the spring characteristics (i.e. load-
midspan deflection relationship of RC beam with loading rate effects) for two-
- 23 -
degree-of- freedom model. All specimens failed in flexure under both static and
high loading rates. The ultimate resistance of the beams was enhanced by the
increment of longitudinal tensile reinforcements for both static and high loading
rates. A database of RC beams under different loading rates has been assembled
from the literature (Mutsuyoshi and Machida [M11]; Kulkarni and Shah [K13];
Fujikake et al. [F4]; Adhikary et al. [A3 and A4]). The database containing the
particulars and response of the specimens (i.e. geometric characteristics of beam,
reinforcement ratios, material properties, peak resistance and failure modes etc.) is
succinctly tabulated in the Appendix A.
1.5
Relative increase in capacity
flexural (present study)

shear (both studies)
1.4
flexural (Mutsuyoshi and Machida)
1.3
1.2
1.1
1
1 100 10000 1000000
Relative midspan velocity
Fig. 2.1: Relative enhancement of flexural and shear strengths of RC beams

(Kulkarni and Shah [K13])
2.4.2 Analytical and numerical studies
A simple sectional analysis was first attempted by Kulkarni and Shah [K13] to
predict the load-midspan deflection of RC beams under high rates. However,
standard sectional analysis with rate-dependent material properties did not capture
the shape of the curves adequately. Extreme localized yielding of reinforcing bars
due to enhanced bond properties at high rates could be the reason behind that. Thus,
the sectional analysis was augmented by incorporating the shape of the average
stress-strain curve of reinforcing bars consistent with the characteristics of localized
- 24 -
yielding which leads to improved prediction of load-midspan deflection curves.
Fujikake et al. [F4] proposed a section based nonlinear analytical model to capture
the load-midspan deflection relationship of RC beams under rapid flexural loading.
Load-midspan deflection relationship was obtained through moment-curvature
relationship by sectional analysis technique, in which the strain rate effects of
concrete and steel reinforcement were duly considered.
2.5 Study on RC Beams under Drop-weight Impact Loading
Impacts are generally classified as hard and soft impact. For the case of soft impact,
the missile itself also deforms significantly whereas in hard impact, the missile
undergoes almost no deformation compared to impacted structures. Depending on
the nature of impact, the impacted structure may respond in several ways: (a) it may
suffer local damage only, dissipating the majority of impact energy at or around the
impact zone; (b) it may respond to the impact loading globally through the
deformation of the entire member; (c) it may respond in such a way that it suffers a
combination of both local and global damage. Local damage is usually categorized
as follows: (a) penetration of the missile in the front face and scabbing of small
pieces of concrete at the back face; (b) significant scabbing of the concrete at both
faces; (c) perforation of the element, with the missile existing from the back face
with a residual velocity. Local and overall impact phenomena for hard missile
impact are schematically presented in Fig. 2.2.
- 25 -
Fig. 2.2: Missile impact phenomena (Kennedy [K3])
This section gives a brief summary of research on the global response of RC beam
under drop-weight impact loads. It is worthwhile to mention that research on the
use of special materials, such as high strength concrete or fibre reinforced concrete
for increasing the impact resistance and the local impact response of beam are
omitted in this review for brevity.
2.5.1 Laboratory tests
One of the earliest impact test program was carried out by Mylrea [M12], in which
254 x 406.4 mm (width by depth) RC beams with 2.44 m spans having varying
amounts of reinforcing steels of various grades subjected to various falling weights
(254 kg and 925 kg drop weight hammer were used). The beams contained different
amounts and grades of longitudinal reinforcement and no shear reinforcement.
Although the beams were severely damaged with diagonal shear cracks, it was
concluded that the impact resistance of beams was quite significant in the presence
of even small amounts of reinforcement, since it was impossible to rupture any of
the longitudinal reinforcement. The failure mode of beams is shown in Fig. 2.3.
- 26 -
Fig. 2.3: Failure mode of the RC beams (Mylrea [M12])
The factor of safety against rupture of the reinforcement was too high. Moreover,
there was enormous reserve strength against collapse after the yielding of the
reinforcing bars. Similar to the impact performance of beam containing the
structural steel reinforcement, the impact behavior of beam containing the rail-steel
and other brittle steel reinforcement was quite satisfactory.
Kishi et al. [K5] conducted drop-weight impact tests on eight RC beams having
clear span length of 2 m to establish a rational impact-resistant design procedure for
flexural-failure-types specimens. Impact load was imparted onto the midspan of
specimen by free-falling 200 kg steel-impactor, as exhibited in Fig. 2.4(a). Cross-
sectional dimensions, reinforcement ratios and impact velocity were considered as
experimental variables. Ultimate resistance of flexural-failure type beams subjected
to drop-weight impact loading was estimated by using the maximum reaction force
at failure. Configuration of the hysteretic loop between reaction force and the mid-
span displacement at failure was approximated by a parallelogram, as shown in Fig.
2.4(b).
- 27 -
(a) Impact test set up (b) Hysteretic loop
Fig. 2.4: Impact test set up and hysteretic loop between reaction force and mid-span
displacement of beam (Kishi et al. [K5])
This is suggested that flexure-failure type beams under impact load may be
designed with a margin of safety by assuming dynamic response ratio as 2.0 and
ratio of absorbed energy to input kinetic energy as 0.7. A simple equation was
proposed to calculate the required static bending resistance of beams against impact
loading. The formulation to estimate the required static bending resistance is as
follows:
Ekd
Pusd  0.35 (2-34)
 rd
where Pusd is static bending resistance, Ekd is input kinetic energy and  rd is
residual displacement.
Kishi et al. [K7] carried out falling-weight impact tests on shear-failure type RC
beams to establish a rational impact-resistant design procedure. In total, twenty-
seven simply supported rectangular beams (150 mm width and 250 mm depth)
without shear reinforcement were constructed to perform that test program.
Longitudinal reinforcements, shear span to effective depth ratios, static shear to
bending resistance ratio and clear span length were taken as test variables. An
impact load was applied at the mid span of the beam by dropping a free-falling 300
kg steel-weight, as depicted in Fig. 2.5. The striking face of the steel-weight was
spherical with a radius of curvature of 1407 mm. This has been confirmed by the
- 28 -
same author that the effects of the striking face of the impactor on the dynamic
response and failure mode of the beam are very small under similar impact velocity.
Fig. 2.5: Impact test setup (Kishi et al. [K7])
Fig. 2.6 presents the crack patterns of two series specimens under various impact
velocities. It was addressed that if the static shear to bending resistance ratio is less
than unity (e.g. 0.64 for beam A36), the beams clearly collapse in a shear-failure
mode under impact loading. On the contrary, if the static shear to bending
resistance ratio is greater than one (e.g. 1.03 for beam B36); the RC beams collapse
in a bending-failure mode under low-velocity impact loading but may collapse in a
shear-failure mode under high-velocity impact loading. The shape of the hysteretic
loop between the reaction force and the mid span displacement was assumed to be
triangular as shown in Fig. 2.7 when the beams just collapsed in shear-failure
mode. Kishi et al. [K7] recommended that shear-failure type beams without shear
reinforcement under impact loading may be designed with a certain safety margin
by assuming a dynamic response ratio of 1.5 and absorbed input energy ratio of 0.6.
Required static shear resistance for beams against impact loading could be
evaluated by a simple equation as follows:
- 29 -
Ekd
Vusd  0.8 (2-35)
 rd
where Vusd is static shear resistance, Ekd is input kinetic energy and  rd is residual
displacement.
Fig. 2.6: Crack patterns for beams (a) A36 and (b) B36 (Kishi et al. [K7])
Fig. 2.7: Simplified loop for reaction force-midspan displacement of RC beams

being collapsed in shear-failure mode (Kishi et al. [K7])
Bhatti et al. [B8] reported a test program of shear-failure type beams, aimed at
developing a simple elasto-plastic impact response analysis by finite element (FE)
method. The test program consisted of twelve simply supported beams, having 200
mm wide, 400 mm deep and 2400 mm long. Impact load was applied at mid span of
RC beam by dropping a 400 kg steel-weight from predefined height. Following
experimental results were compared with numerical analysis results to check the
- 30 -
reliability of the FE method: 1) time histories of impact force, reaction force and
mid-span displacement; 2) hysteretic loops of impact force vs. mid-span
displacements and reaction force vs. mid-span displacements; 3) crack profiles on
side surface of beam.
Fujikake et al. [F4] examined the impact responses of twelve RC beams through an
experimental study which involves a drop-hammer impact test. The beams were of
250 mm in depth, 150 mm in width and 1700 mm in length. The specimens used in
this research consisted of under-reinforced sections with sufficient amount of
transverse reinforcements to allow for an overall flexural failure. Investigation was
carried out to evaluate the influence of drop-height and the amount of longitudinal
reinforcement ratio to the structural response. The static shear to bending resistance
ratios of beam were varied from 1.5 to 2.6. Fig. 2.8 displays the drop-hammer
impact test setup where a hammer with a mass of 400 kg was dropped freely onto
the top surface of the beam at midspan from four different heights. The striking
head of the hammer had a hemispherical tip with a radius of 90 mm.
Fig. 2.8: Drop-hammer impact test setup (Fujikake et al. [F4])
Typical failure modes obtained from this test program are shown in Fig. 2.9. Beams
of series S1616 exhibited an overall flexural failure at all the drop-heights
considered for this test program. However, for specimens of series S1322 and
S2222, the overall flexural failure was observed up to a drop-height of 0.6 m. Local
- 31 -
failure with extensive concrete crushing near the impact region was monitored at a
drop height of 1.2 m and 2.4 m.
Fig. 2.9: Failure modes: (a) S1616 series; (b) S1322 series; (c) S2222 series
(Fujikake et al. [F4])
It was concluded that the amount of longitudinal reinforcement ratio significantly

affected the failure modes of beams. Beams with comparatively lower amounts of
longitudinal steel reinforcement exhibited only overall flexural failure, while beam
having comparatively higher amount of longitudinal reinforcement not only
exhibited the overall flexural failure but also local failure located near impact
loading point. Local failure was substantially reduced when the beam contained
heavy longitudinal compression reinforcement. Impact response characteristics such
as: the maximum impact load, the impulse, the duration of impact load, the
maximum midspan deflection and the time taken for the maximum midspan
deflection increased as the drop-height was increased. However, the duration of
impact load, the maximum midspan deflection and the time taken for the maximum
midspan deflection were affected by the flexural rigidity of beams.
Chen and May [C3] accomplished a test program to investigate the high mass and
low velocity impact behavior of RC beams. Fourteen 2.7 m and four 1.5 m clear-
span length specimens were tested under impact loads using drop-weight facility.
All the tests were conducted under a drop-weight of 98.7 kg with an impact velocity
of 7.3 m/s, as shown in Fig. 2.10. Support conditions (e.g. pin-ended and simply
supported), type of impactors (e.g. hemispherical and flat) and impact interface (e.g.
plywood placed in between beam and impactor and direct impact) were taken as
- 32 -
test variables in this investigation. Measurements were taken during testing which
includes transient impact force-history, acceleration-history, strain-history of
longitudinal reinforcements and video footage of crack profiles generated from the
images captured by high-speed camera.
Fig. 2.10: Drop-weight impact test set up (May et al. [M6])
Three types of failure modes were observed by the authors:
Failure mode – a: Flexural failure with some crushing beneath the impactor and
some shear cracking in the impact zone were observed. Vertical cracks starting
from the top of a beam were found along the beam section away from the impact
zone as presented in Figs. 2.11(a) and (b) for beams A1 and A2 respectively.
Further shear cracking and short vertical cracking occurred in beam A3, as shown
in Fig. 2.11(c). There was less damage in this type of failure than in modes „b‟ and
„c‟, described below, owing to some impact energy being absorbed in deforming the
plywood.
Failure mode – b: Local failure at the impact zone with excessive concrete crushing
below the impactor and yielding of the tensile reinforcements were noticed as
shown in Figs. 2.11(d) and (e) for beams B1 and B2 respectively. Away from the
- 33 -
impact zone the cracking was similar to that of mode „a‟ failure. This failure mode
was observed in beams impacted directly with the hemispherical impactor.
Failure mode - c: This is similar to mode „a‟ failure but was accompanied by loss of
the concrete cover at the bottom of the beam owing to scabbing as shown in Figs.
2.11(f) and (g) for beams B3 and B4 respectively. It was occurred on beams
impacted directly with the flat impactor.
Figure 2.11: Post-test crack pattern of beams (a) [A1]; (b) [A2]; (c) [A3]; (d) [B1];
(e) [B2]; (f) [B3] and (g) [B4] (Chen and May [C3])
The beam test results revealed that the supports conditions had less influence on the
impact force than the span length. Moreover, this has been noticed that the plywood
interface distributed the impact force in a similar manner like a flat impactor.
Saatci and Vecchio [S1] reported a well instrumented experimental program to

contribute to the understanding of the effects of shear mechanisms on the behavior
of RC beams under impacts loading. Eight specimens were tested under free-falling
drop-weights impact loading at their midspan. Two different drop-weights were
used for the impact testing: a lighter weight of 211 kg and a heavier weight of 600
kg. The contact surface of the drop-weights were flats and a 50 mm thick, 305mm
square steel plate was placed on the top of the beam at impact zone to obtain a well-
distributed impact force. The weights were dropped from a clear height of 3.26 m
above the specimen, resulting in an 8 m/s impact velocity. For the a-series
specimens, the beams were first tested once with the smaller drop-weight, followed
by two tests with the larger one. However, for b-series specimens, the order was
- 34 -
reversed; they were tested twice with the larger drop-weight and the one last time
with the smaller drop-weight. All the specimens had identical longitudinal
reinforcement but varying amount of shear reinforcement ratios, intended to
investigate the effects of shear resistance on the impact behavior. Fig. 2.12 displays
the test setup which was specially devised to prevent the uplift of the specimen
without creating restrain moment at the supports.
Fig. 2.12: Impact test set up (Saatci and Vecchio [S1])
Crack patterns of a-series and b-series specimens are shown in Fig. 2.13. Static test
results divulged that SS3 and SS2 exhibited a ductile flexure-critical behavior,
whereas, SS1 and SS0 were shear-critical. During impact testing, regardless of their
projected static behavior, all specimens developed severe diagonal cracks,
originating at the impact point and propagating downward with an angle
approximately 45 degrees, forming shear plugs. Additionally, numerous diagonal
cracks parallel to shear-plug cracks also formed, along with some vertical flexural
cracks at the midspan and the supports. In the flexure-critical beams, shear-plugs
developed faster than the formation of support cracks. On the other side, failure of
shear-critical specimens occurred due to the development of shear-plug as well as
the formation of diagonal cracks from loading point to support point. The shear
resistance of flexural-critical specimens was sufficient to carry the shear forces
beyond the shear plug to the supports. Subsequent impact loading pressed the shear-
plug down further without causing significant damage in the other region. However,
shear-critical members did not have enough strength to transmit the shear forces to
- 35 -
the supports. As a result, the shear forces beyond the shear-plug region caused
massive damage between the shear-plug and the supports, finally failing the beams
in that region.
Fig. 2.13: Final crack pattern of a-series and b-series beams (Saatci and Vecchio
[S1])
Shear characteristics of the specimens played a pivotal role in their overall impact
response. All specimens, regardless of their shear capacity, developed severe
diagonal cracks and shear-plug cracks under the impact point. Specimens with
higher shear capacity were able to sustain more impacts and absorb more energy,
whereas the ones with lower shear capacity suffered extensive damage under the
same or smaller impact loads. Shear mechanism must be considered during the
method development in predicting the impact responses. Impact forces at the initial
stages of response were mainly resisted by the inertia of the specimens before the
forces reached the supports. Therefore, the mass and geometric properties of a
structure, such as the span length of a beam, are important factors in resisting the
impact forces. After gaining some experience from their test program, Saatci and
Vecchio provided some recommendations for future experimental studies on
impact. To cover wider range of parameters (i.e. impact parameters, such as mass
and velocity of impactor; specimen parameters, such as geometry, reinforcement
ratios and material properties etc.) that would affect the overall behavior, higher
number of specimens should be considered. Measurement of impact responses
should be recorded with a higher sampling rate. To measure the displacement
during impact test, optical technique should be utilized. Moreover, to better
understand the crack formation and propagation mechanisms, high-speed camera
- 36 -
should be employed during testing.
Tachibana et al. [T1] documented an investigation on a series of low-speed impact

experiments of RC beams with varying span length, cross-section and longitudinal
reinforcements. Beams were impacted by a steel-weight which was dropped from a
specific height, as shown in Fig. 2.14. The steel-weight used in this experiment had
a curved contact surface with a length and radius of 565 and 75 mm respectively.
Steel-weights having various masses such as 150, 300 and 450 kg were employed
in this test program. Static shear to bending resistance ratios for all beams were
greater than one; that means flexure-failure will precede shear-failure under static
loading conditions.
Fig. 2.14: Impact test set up (Tachibana et al. [T1])
Fig. 2.15 shows the crack pattern of type A2 beam after a 300 kg impactor struck
the beam with a velocity of 5 m/s. This is a typical crack pattern for flexure-failure
with little concrete fragmentation at the impact region.
- 37 -
Fig. 2.15: Crack pattern of type A2 beam (Tachibana et al. [T1])
The impulse resulting from impact loading is proportional to the momentum of the
impacting mass. Similarly, the duration of impact force is proportional to the ratio
of the momentum of impacting mass and static ultimate flexure resistance of beam.
Moreover, an equation was proposed (Tachibana et al. [T1]) to calculate the
maximum mid-span deflection of beam based on impact energy and static ultimate
flexure resistance. The proposed formulation is as follows:
Ecol
 max  0.522 (2-36)
Pu
where  max is maximum displacement (mm), Ecol is kinetic energy (J) and Pu is
ultimate flexure resistance (kN). The validity of the proposed equation was
corroborated by comparing with other experimental results and numerical results
from FE simulations, as shown in Fig. 2.16.
Fig. 2.16: Proposed equation by Tachibana et al. [T1]
The static bending resistance of considered beams ranged from 16.7 to 66.7 kN and
the impact energies varied from 150 to 5400 J based on the variation of mass and
impact velocity. This has been concluded that the proposed equation to estimate the
maximum displacement of beam under impact loading, allows for a performance-
- 38 -
based design of protective structures.
Kishi and Mikami [K10] developed an impact-resistance design methodology for

RC beams following performance-based design based on falling-weight impact test
results. In total, 36 beams were tested and cross-sectional dimensions, clear span
length, amount of longitudinal and shear reinforcement ratios, compressive strength
of concrete, yield strength of longitudinal reinforcements, mass and velocity of the
drop-hammer were the test variables. All beams had rectangular cross-section, 150
to 250 mm wide, 200 to 400 mm deep and 2 to 3 m clear span length. The amount
of longitudinal reinforcements was varied from 0.8 to 3.17% and the shear
reinforcements were placed with varying spacing of 100, 125 and 150 mm. For all
beams, the static shear to bending resistance ratio was greater than 1, signifying that
they were statically flexure-critical. In this study, three different mass of the drop-
hammer which were 300, 400 and 500 kg used and impact velocities were ranged
from 3.1 to 7.7 m/s. All drop-weights had identical shape of spherical bottom with
1407 mm radius and 2 mm taper. Crack pattern of beams corresponding to series
G2L, G5 and G10 is shown in Fig. 2.17. Flexural cracks were observed not only in
the bottom fibre but also in the upper fibre throughout the clear span. Moreover,
diagonal shear cracks formed around the mid-span area, mainly below the impact
point and these cracks became prominent with increasing impact velocity. The
maximum and residual deflections are proportional to the input impact energies and
the gradients are empirically formulated using the inverse of static flexural load-
carrying capacity of the beam.
- 39 -
Fig. 2.17: Crack patterns in beams of series G2L, G5 and G10 (Kishi and Mikami
[K10])
The maximum and residual deflections per unit input impact energy E (coefficient
of maximum and residual deflection and static flexural load-carrying capacity Pusc
are  def and  rs ) plotted in Fig. 2.18. On the basis of these relationships, empirical
design formulae were proposed which are as follows:
0.63
 def  (2-37)
Pusc
0.42
 rs  (2-38)
Pusc
Rearranging the above two equations, the followings can be obtained
E
Pusc  0.63 (2-39)
Dmax
E
Pusc  0.42 (2-40)
 rs
- 40 -
where Pusc is static flexural load-carrying capacity (kN), E is input impact energy
(J), Dmax is maximum displacement and  rs is the residual displacement (mm).
Fig. 2.18: Regression curves: (a) regression curve between coefficient of maximum
deflection  def and static flexural load-carrying capacity Pusc ; (b) regression curve
between coefficient of residual deflection  rs and static flexural load-carrying
capacity Pusc by Kishi and Mikami [K10]
These above-mentioned two equations allow for a significant simplification in

design approach for structures subjected to impact loading. If the input impact
energy and the static flexural load-carrying capacity in both equations are the same,
the ratio of the maximum deflection to residual deflection would be 1.5, which
could be adjudged as an important design parameter. However, there are some
limitations in the application which are: 1) input impact energy E < 15 kJ; 2) Eqs.
(2-8) and (2-9) are applicable for RC beams having static flexural load-carrying
capacity Pusc < 240 kN and static shear-flexural capacity ratio  > 1.5.
A database of RC beams under drop-weight impact loading has been accumulated

from the literature (Kishi [K7], Fujikake et al. [F4], Bhatti et al. [B8], Chen and
May [C3], Saatci and Vecchio [S1], Tachibana et al. [T1], Kishi and Mikami
[K10]). The database containing the particulars and the impact responses of the
- 41 -
specimens (i.e. geometric characteristics of beam, reinforcement ratios, material

properties, mass & velocity of the impactor, maximum impact load, maximum mid-
span displacements and failure modes etc.) is concisely tabulated in the Appendix B
and C.
2.5.2 Numerical and analytical studies
Bhatti et al. [B8] demonstrated a simple elasto-plastic finite element (FE) analysis
method for simulating the experimental impact behavior. Concrete material model
was assumed by using a bilinear model in compression side and a cut-off model in
tension side. The steel reinforcing bars were modeled as elasto-plastic with
isotropic hardening. Moreover, strain rate effect on concrete and reinforcements
was not considered in this study. Finally the numerical results were compared to the
experimental outcomes such as impact, reaction force history, mid-span
displacement history, impact force vs. mid-span displacement loops and reaction
force vs. mid-span displacement loops. Moreover, energy absorption and crack
patterns were predicted by this method within a reasonable accuracy.
Saatci and Vecchio [S1] proposed a nonlinear finite element analysis (NLFEA)
procedure to numerically simulate the impact responses of RC beam. To express the
effectiveness of the procedure, the computed responses (i.e., time histories of mid-
span displacements, crack profiles and longitudinal reinforcement strains) were
compared to the measured responses from an experimental program. The
methodology based on distributed stress field model (DSFM) was successful in
predicting the shear-dominant behavior of the specimens under impact loads.
However, accurate prediction of test specimens was obtained in this research by
considering strain rate dependent material properties; thus, this issue needs further
investigation.
Fujikake et al. [F4] modeled the RC beam under drop-hammer impact loading as
two-degree-of-freedom mass-spring-damper system. This model could represent not
only global response of the beam but also the local response at the contact point
between the drop-hammer and beam. Here, the impact scenario has been presumed
as a perfectly plastic collision. The spring property corresponding to the overall
- 42 -
response of the beam was determined from the loading-rate dependent load vs. mid-
span deflection relationships whereas the contact spring stiffness value was
assumed from Hertz‟s contact theory. The global damping coefficient was
considered to be zero and contact damping coefficient was assumed to be one-half
of the critical damping coefficients. The analytical results in terms of impact force
history and mid-span deflection history were in good agreement with the
experimental results when the beams predominantly failed in flexure. However,
large discrepancies were observed in mid-span deflection responses when both local
failure at impact point and flexure failure occurred.
2.6 Summary
Based on literature review, some conclusions can be summarized as follows:
(1) Review of literature related on the performance of RC beams under varying

loading rates reveals a paucity of works in this field. A limited number of tests on
beams show that the increased flexural strength due to the increase in yield strength
of steel reinforcement is a primary reason for change in failure mode of beams
subjected to loading rate expected in an earthquake event. This change in failure
mode phenomenon is more pronounced in beams of intermediate slenderness.
Moreover, investigation on effects of loading rates corresponds to low velocity
impact on RC beam is found to be inadequate in literature. Thus, a systematic test
program on intermediate-slender beams subjected to wide spectrum of concentrated
loading rates at their mid-span is worthy of investigation.
(2) There is a lack of numerical research on identifying the effects of various

parameters on the behavior of beams under a wide range of varying loading rates.
Therefore, numerical model (three-dimensional) is essential to validate the
experimental results first and further utilized to expand the knowledge beyond the
range of parameters investigated experimentally.
(3) Assessment of the behavior of RC beams under drop-weight impact loading has
been well documented in the literature; however, evaluation of the residual
performance of impact-damaged beams is a lacking area. To contribute towards a
better understanding in this area, an organized experimental program (i.e. impact
- 43 -
tests followed by quasi-static tests) is essential.
(4) Due to the scarcity of test results, it is obvious that there will no numerical
research on residual performance of impact-damaged beams. Therefore,
development of three-dimensional FE model to numerically simulate these two-
stage experiments is valuable to examine.
- 44 -
Chapter 3: Experimental Investigation of Loading Rate Effects on RC beams
CHAPTER 3
EXPERIMENTAL INVESTIGATION OF LOADING
RATE EFFECTS ON RC BEAMS
3.1 Introduction
Over the past few decades, there has been a growing interest among the researchers
and designers in analyzing and designing RC structures against extreme loads, such
as earthquakes, blasts and impact. These loading scenarios induce varying strain
rates in the structural elements. In this study, the strain-rate effect on RC beams
falls in the wider range of static to hard impact (10-4 to 10/s) regime. A variety of
devices such as hydraulic machines, pneumatic machines and impact systems (mass
swings or falls from specific heights to strike a specimen) have been constructed to
perform high loading rate tests on structural elements in laboratory. Previous
studies (Mutsuyoshi and Machida [M11] and Kulkarni and Shah [K13]) on high
loading rates of RC beams were confined up to the earthquake loading regime.
Fujikake et al. [F4] considered a much higher piston velocity for high loading rates
which could be linked to the impact loading regime; however consider only one test
parameter (i.e. flexural reinforcements ratio). There is a lack of systematic research
on RC beams under wide range of varying loading rates to cover static loading to
hard impact regime. Therefore, a well-instrumented experimental program was
initiated to investigate the effects of varying loading rates on the structural behavior
of intermediate slender RC beams with limited amount of flexural reinforcements
(approximately 0.3 b where b refers to balanced reinforcement ratio in
percentage). In this study, hydraulic machine was used to generate the wide range
of loading rates extending from 4 × 10-4 m/s to 2 × 100 m/s. Corresponding strain rate
induced in the reinforcements of RC beams were ranging from 10-4/s to 10/s.
- 45 -
3.2 Analysis of Experimental Database
A database of RC beams under different loading rates has been assembled from the
literature to conduct the parametric studies with an objective of gaining some
insight on the influence of various parameters such as shear span to effective depth
ratio ( a d ), longitudinal reinforcement ratios (  L ), shear reinforcement ratios (  T )
and failure modes on the dynamic increase factor (DIF) of maximum resistance.
Due to the scarcity of data, it was very difficult to compare the data which could be
grouped in such a way that they would present respective variable as the only
variable, i.e. beam with identical properties except varying a d . The database
containing the particulars of the specimens is tabulated in the Appendix-A.
Most of the RC beams shown in the Appendix-A, can be classified into singly RC
beams without shear reinforcement (SR) and doubly RC beams with shear
reinforcement (DR). Thus, parametric study was carried out for SR and DR beams
separately to determine the effect of several parameters on their corresponding DIF
of maximum resistance. Fig. 3.1 (a) shows the influence of shear span to effective
depth ratio on the trend of variation of DIF of SR beams. Effects of longitudinal
reinforcement ratios and failure mode on DIF under different loading rates are
shown in Figs. 3.1 (b) and (c). Due to scattering of data points, it is very tough to
ascertain the effects of single parameters on DIF of SR beams. Failure modes (shear
or flexure) and the time of yielding of tensile reinforcements (before or after the
peak resistance of shear critical beams) have some significant effect on DIF. In Fig.
3.1 (c) for 0.1 to 1 m/s, few DIF are higher than the clustering data points. In these
cases, yielding of tensile reinforcements occurred after their peak resistance.
Similarly, influence of shear span to effective depth ratio ( a d ), longitudinal
reinforcement ratios (  L ) and shear reinforcement ratios (  T ) and failure mode on
DIF of DR beams are shown in Figs. 3.2 (a), (b), (c) and (d) respectively. From the
trend of DIF, it can be concluded that beams with lower a d ratios (1.9) have
higher DIF under varying loading rates as compared to slender beams ( a d -3.3 and
4.4). Beams of low shear reinforcement had a higher DIF value as compared to
other specimens.
- 46 -
(a) 2.2
a/d :- 5.5-6
2
a/d :- 4.5-5
a/d :-4-4.5
1.8 a/d :- 3-3.5
DIF
1.6
1.4
1.2
1
0.01 0.1 1 10
Loading rate (m/s)
(b) 2
L :- 0.8-1%
1.8
L :- 1-1.4%
L :- 1.7-2%
1.6
DIF
1.4
1.2
1
0.01 0.1 1 10
Loading rates (m/s)
(c) 1.6
1.5
1.4
DIF
1.3
1.2
1.1 Shear
Flexure
1
0.01 0.1 1 10
Loading rates (m/s)
reinforcement ratios (  L ); (c) failure modes on DIF of singly reinforced (SR)

beams
- 47 -
When the specimen contains less shear reinforcements, it may fail in shear where
the tensile strength of concrete (high rate sensitive than compressive strength) plays
a significant role in increasing the ultimate resistance under high loading rates and
correspondingly increases the DIF. Finally based on the comparative plot of DIF vs.
failure modes, it can be concluded that shear critical beams produced higher DIF as
compared to flexure critical for all loading rates.
(a) 1.7 (b) 1.8

1.6 a/d - 1.9 1.7 L - 0.82%
a/d - 3.3 1.6 L - 0.84%
1.5 a/d - 4.4
1.5
L - 2.4%
1.4
DIF
DIF 1.4
1.3
1.3
1.2 1.2
1.1 1.1
1 1
0.01 0.1 1 10 0.01 0.1 1 10
Loading rate (m/s) Loading rate (m/s)
(c) 2 (d) 1.6

T :- 0.11-0.15% Shear
1.5
1.8 T :- 0.19% Flexure
T :- 0.42% 1.4
1.6 T :- 0.56%
DIF
DIF
T :- 0.84% 1.3
1.4
1.2
1.2
1.1
1 1
0.01 0.1 1 10 0.01 0.1 1 10
Loading rate (m/s) Loading rates (m/s)
reinforcement ratios (  L ); (c) shear reinforcement ratios (  T ); (d) failure modes on
DIF of doubly reinforced (DR) beams
The review of existing literature revealed variety of high loading rate tests and it is
very difficult to make effective comparisons due to the involvement of numerous
parameters at a time. As such, it will be beneficial if the researchers can agree on a
preferred way to perform high loading rate testing of RC beams. There are some
- 48 -
major differences in the testing methods worth notable. At first, the test setup may
be designed for 3- or 4-point bending depending on the objective of the test
program. For high loading rate testing using hydraulic machines, 3-point
configurations were recommended by Kulkarni and Shah [K13] from the point of
view of reducing inertial effects. Furthermore, attempt should be made to test
identical specimens (i.e., similarities not only in structural level but also in material
level, such as aggregate size, compressive strength of concrete and yield strength of
reinforcements) using different test machines. Time to reach peak displacement in
loading history (for a particular loading rate in displacement control) during
experiment should be kept nearly unvarying. The data acquisition system
(particularly the frequency response characteristics) should be nearly identical to
facilitate better analysis and comparisons. In addition, the type of anchorage (i.e.,
by nuts, welded with plates or development length by hooking) at the end of
flexural reinforcements has a profound effect on the shear failure mechanism of
beams (which will be governed by arch or beam action) under varying loading rates
and should as such be kept identical across the tests.
- 49 -
3.3 Objective of Test Program
Review of literature in the field of RC beam under varying loading rates reveals that
there is a need to assess the behavior of beam under wide range of varying loading
rates. The primary objective of this research is to study the loading rate effect on the
load vs. mid-span displacement relationships, concrete contribution to the shear
stress, and peak strain rate at longitudinal tensile and shear reinforcements.
Moreover, Dynamic increase factors (DIF) of maximum resistance, cracking
stiffness, energy-absorption and failure mechanism of beams are also discussed by
comparative analysis. In addition, this test program will supplement the literature
with detailed test data that can assist further in the development and verification of
numerical models.
3.4 Experimental Overview

Till date, limited research has been attributed to explore the behavior of RC beams
under wide range of loading rates. Meticulous inspection of the database provided
in the appendix reveals that the behavior of RC beams with intermediate
slenderness under varying loading rates is still deficient. With an intention of
contributing to that scarce research area, an experimental program was undertaken
for RC beams with limited amount of flexural reinforcements ( 0.31b and 0.3 b
for 3.3 and 4.4 series respectively; where  b refers to balanced reinforcement ratio
in percentage) subjected to a wide range of loading rates (Static: 4 × 10-4, Low:
4 × 10-2, Medium: 4 × 10-1 and High: 2 × 100 m/s). Comparative analyses of
specimens under these varying loading rates highlighted several key aspects of their
dynamic behavior.
The experimental program consisted of 24 RC beams which were divided into two
groups in terms of their shear span-to-effective depth ratios ( a d -3.3 and 4.4).
Each group had three types of specimens distinguished by their shear reinforcement
ratio. Each type had four identical specimens subjected to four types of varying
rates of concentrated loading at their mid-span. The entire test program was
summarized in the form of a flow chart as shown in Fig. 3.3. It is noted that all test
- 50 -
specimens have been designed and cast at NTU, Singapore however; the testing
was conducted at NDA, Japan. The specimen denoted by SR3.3_0.84 corresponds
to „singly-reinforced‟ beam (SR) followed by the shear span-to effective depth ratio
and percentage of longitudinal tensile reinforcement ratios. Again,
DR3.3_0.84_0.11 refers to „doubly-reinforced‟ beam (DR) followed by the shear
span-to effective depth ratio, percentage of longitudinal tensile and compressive
reinforcement ratio and percentage of shear reinforcement ratio.
Test Program
a d  3.3 a d  4.4
a/d- 3.3
SR3.3_0.84 SR4.4_0.82
DR3.3_0.84_0.11 DR4.4_0.82_0.13
DR3.3_0.84_0.15 DR4.4_0.82_0.19
Loading rate
S-Static (4  10-4 m/s)

L-Low (4  10-2 m/s)
M-Medium (4  10-1 m/s)
H-High (2  100 m/s)
Fig. 3.3: Flow chart of test program
The layout of the longitudinal reinforcements, spacing of shear reinforcements and

the measuring points (acceleration and strain for steel reinforcing bars) are shown in
Fig. 3.4. Here, „SA‟ refers to the strain gauges attached at the midspan of
longitudinal tensile reinforcements (e.g. two strain gauges for two bottom tensile
reinforcements) where „SS‟ refers to the strain gauges attached to transverse
reinforcements (e.g. one strain gauge at one leg of stirrups). Longitudinal
reinforcements consisted of deformed steel bars of T13 (diameter-13mm) and T10
(diameter-10 mm) for a d -3.3 and 4.4 series specimen respectively while R6
(diameter- 6 mm) plain round bars were used as shear reinforcements for all
- 51 -
specimens. Target concrete compressive strength at 28 days was 40 MPa with

maximum aggregate size of 10 mm for all specimens. The yield strengths of T13,
T10 and R6 were 520, 510 and 310 MPa respectively. All specimens were cast at
the same time using ready mix concrete of same batch; hence, all specimens had an
identical concrete strength. The compressive strength at 28 days obtained from
three concrete cylinders was 39.0, 39.6 and 42.2 MPa respectively. The specimen
details, theoretical static capacities, static expected and observed failure modes are
listed in Table 3.1, where the static flexural and shear resistances are calculated
according to ACI 318-08 [A2]. According to this table, except SR3.3_0.84, all
specimens are theoretically flexure-dominant (shear to bending resistance ratio is
greater than one) under monotonically increasing static loads. However, for DR3.3
series shear failure mode (e.g. Shear*1 and Shear*2) was observed under static
loading. Shear cracks were developed just after the yielding of tensile
reinforcements and after the crushing of compression concrete for the case of
shear*1 and shear*2 respectively. Thus, this type of failure could be visualized as
mixed type failure mode. The reason of inconsistency between the expected and the
observed failure modes in DR3.3 specimens is the wider interval of the stirrups.
The maximum spacing of vertical stirrups as the smaller of d 2 or 610 mm is set by
ACI 318-08 [A2]. Following ACI Code, the vertical stirrups surely intersect at a
diagonal shear crack formed in the RC beam. Thus, stirrups were not effective at all
in preventing shear failure.
- 52 -
Strain gauge for rebars Unit: mm

Accelerometer
SA-1,2 210 250 SA-1,2 160 190

60
70 250 200 200 250 2T10
250 200 200 250 2T13
150 1400 120
1400
1700
1700
SR4.4_0.82
SR3.3_0.84
R6 2T13 R6 2T10
SS-2 SA-1,2 SS-1 170 210 250 SS-2 SA-1,2 SS-1 130 160 190
1400 70 2T13 1400 60 2T10

4@350=1400 150 4@350=1400
120
1700 1700
DR3.3_0.84_0.11 DR4.4_0.82_0.13
R6 2T13 R6 2T10
SS-2 SA-1,2 SS-1 170 210 250 SS-2 SA-1,2 SS-1 130 160 190
70 1400 60 2T10
1400 2T13
6@250=1500
6@250=1500 150 120
1700
1700
DR3.3_0.84_0.15 DR4.4_0.82_0.19

gauges and accelerometers
Table 3.1: Specimen details, theoretical static capacities, static expected and
observed failure modes
Longitu- Shear Shear to
Flexural Shear Expected Observed
-dinal rebar bending
Specimens a/d resistance resistance failure failure
rebar ratio ratio resistance
(kN) (kN) modes modes
(%) (%) ratio
SR3.3_0.84 0.84T 0.00 77.1 66.2 0.86 Shear Shear*1
DR3.3_0.84_0.11 3.3 0.84T&C 0.11 77.6 87.1 1.12 Flexure Shear*2
DR3.3_0.84_0.15 0.84T&C 0.15 77.6 95.5 1.23 Flexure Shear*1
SR4.4_0.82 0.82T 0.00 33.7 40.2 1.19 Flexure Flexure
DR4.4_0.82_0.13 4.4 0.82T&C 0.13 34.0 56.1 1.65 Flexure Flexure
DR4.4_0.82_0.19 0.82T&C 0.19 34.0 62.5 1.83 Flexure Flexure

Shear*1=Shear failure following yielding tensile reinforcements
Shear*2=Shear failure following flexure compression failure
Note:- T-only tensile reinforcements; T&C- Tensile and compressive reinforcements
To construct the specimens, steel formwork was used. Before the placement of
reinforcing cages, oiling was carried out on the inside surfaces of formwork so that
specimens can be removed easily and the surface of the specimens would not
- 53 -
damage. Before casting, concrete spacer blocks were placed in the underside as well
as the side faces of transverse reinforcements to achieve clear concrete cover.
Lifting hooks were mounted on both sides of the specimens to facilitate lifting
process during testing. Fig. 3.5 shows the photos of the casting of beams and the
preparation of concrete cylinders (150mm  300 mm). After casting, damp hessian
fabrics were employed to cure the specimens for two weeks.
(a) (b)
Fig. 3.5: (a) Casting of beams; (b) preparation of concrete cylinders
All specimens were well instrumented to capture the loads, displacements,

accelerations and the strains of steel reinforcement. The instrumentation for this test
program included a load cell, strain gauges and accelerometers. Strain gauges with
3 mm gauge length were installed in the mid-span of the longitudinal tensile
reinforcements and in the mid-point of the two legs of the shear reinforcements to
measure the local strain. These are general purpose foil strain gauges (KFG types)
with resistance of 120  and linear expansion coefficient of 11  10-6/ 0 C . Soldering
to the terminal of the gauges was not required due to the pre-attachment of three
parallel vinyl-coated lead wires with 10m long. Before the installation of strain
gauges at the specific locations, the reinforcement surfaces were grinded lightly and
cleaned with alkaline and acidic chemicals. Fig. 3.6 shows the photos the glued
strain gauges in longitudinal tensile and transverse reinforcements. Although in Fig.
3.6, the strain gauge is shown in top face of the tensile reinforcements however, the
reinforcement cages were inverted during the placement into the moulds; then the
gauges would be in the bottom face of the tensile reinforcements. A load cell of
- 54 -
capacity of 980 kN and measuring frequency of 5 kHz was attached to the actuator
to measure the load. Five accelerometers (capacity of 1000 times gravity and
resonance frequency greater than 70 kHz) were mounted on the specimens for each
test (except for static loading) to measure the accelerations for low, medium and
high rates of concentrated loading. The midspan deflection of RC beams was
measured by laser-type variable displacement transducers (LVDTs) which have a
measuring range of 80 mm and sampling rate of 50 kHz. Data from the sensors
were collected by a digital data acquisition system which has a sampling rate of 100
Hz, 10 kHz, 100 kHz, and 200 kHz for static, low, medium and high loading rates
respectively. A steel plate having 40 mm thickness was placed on the top of the
beam in between loading bar and specimen to transfer well-distributed force to the
specimen. The use of digital photography and high-speed video recording proved to
be valuable in providing insights into the cracking patterns and failure modes. A
schematic diagram of three point bending test setup is shown in Fig. 3.7. All
experiments were conducted in displacement control as this is known to be
indispensible for obtaining the entire post-peak response of structural elements.
Moreover, in varying loading rates testing, displacement control is even effective to
reduce the inertia effects during the pre-peak stage.
Strain gauges
Strain gauge
(a) (b)
gauging in transverse reinforcement
- 55 -
Load cell
Loading plate
Specimen
Load cell
Rigid support
Laser type LVDT
Fig. 3.7: Schematic diagram of test set up
3.5 Experimental Results and Discussions
Following the experimental overview sections which consist of specimen design,

material properties, test-setup and instrumentation; experimental results and major
observations are comprehensively presented and discussed in this section.
3.5.1 Load-midspan displacement relationships
The input load by the actuator was measured by the load cell attached to it and the
reaction forces were measured by two load cells placed at the two pin supports of
the beams. For static, low and medium loading rates, almost identical responses (i.e.
load-midspan deflection) were obtained from the top and two bottom load cells.
However, for high loading rates, higher response was obtained from two bottom
load cells as compared to the top load cell. The 100 kg mass of each supporting
apparatus attached to each support load cell might induce additional responses.
Besides, the wider range of frequency of the load acting on the lower load cells as
compared to its natural frequency could be the another reason. So, the reaction
force (summation of two bottom load cells) obtained from high loading rates might
- 56 -
overestimate the true resistance of the beam as shown in Fig. 3.8. This figure
displays the typical variation of force history obtained from top and bottom load-
cells under high loading rates. Moreover, there is a time lag (2.98 ms) in between
the two responses which is clearly visible in the figure. This is due to the
propagation time required of stress wave to travel from loading point to support
region.
250
2.98 ms
200 SR3.3_0.84_High
Load (kN)
150
Top load cell
100
Two bottom load cells
50
-50
0 0.005 0.01 0.015 0.02 0.025 0.03
Time (s)
Fig. 3.8: Typical variation of force history from top and bottom load-cells under
high loading rates
Thus, inertia force should be deducted from the top load cell data (i.e., using
following equation) by considering the beam as a SDOF system to evaluate its true
resistance.
Rb t   PTL t   M b  xb t  (3-1)
where, Rb t  : true resistance of RC beam without inertia force; PTL t  : load
measured by top load-cell; M b : equivalent mass of RC beam (i.e. 17 Al clear / 35 ;
 =density of RC beam; A =cross-sectional area of beam; lclear = span length in

between supports); xb t  : acceleration at midspan obtained from accelerometers;
moreover, another 25 kg extra mass (considering the mass of the loading bar) has
been added to the equivalent mass of the beam to calculate the true resistance.
Banthia et al. [B2] proposed two approaches (i.e. linear for plain and sinusoidal for
conventionally reinforced beams) to calculate the inertia force based on the
- 57 -
distribution of accelerations along the beam. Efforts have been made to compare the
current methodology and the approaches provided in literature. Fig. 3.9 shows the
comparative plot of true resistance history obtained from Eqtn. 3-1 and approaches
recommended by Banthia et al. [B2]. From this figure, it can be concluded that true
resistance history obtained from current approach is almost similar to the sinusoidal
case coined by Banthia et al. [B2]
(a) 250 (b) 250
200 Eqtn 3-1 200 Eqtn 3-1

Linear case [B2] Linear case [B2]
Sinusoidal case [B2] Sinusoidal case [B2]
150 150
Load (kN)
Load (kN)
100 100
50 50
0 DR3.3_0.84_0.11_High 0 DR4.4_0.82_0.19_High
-50 -50
-0.005 0 0.005 0.01 0.015 0.02 0.025 -0.005 0 0.005 0.01 0.015 0.02 0.025
Time (s) Time (s)
Fig. 3.9: Comparative plot of true resistance history obtained from Eqtn. 3-1 and
approaches proposed by Banthia et al. [B2]
The comparisons of the load-midspan displacement relationship of each specimen

to the corresponding loading rates are illustrated in Fig. 3.10. Due to some
unforeseen problem (i.e., fault in hydraulic machine), the response of the beam
naming DR4.4_0.82_0.19 under low loading rates could not be captured and hence,
the corresponding beam response has been omitted. Load-carrying capacity of all
specimens increases with incremental loading rates. Dynamic increase factors (DIF)
of a d -3.3 and 4.4 series specimens are shown in Figs. 3.11 (a) and (b)
respectively. DIF is calculated as a ratio of maximum resistance of beam at any
loading rates (i.e. low, medium and high) to the corresponding static maximum
resistance at a loading rate of 4 × 10-4 m/s. From the comparative analysis of the
above-mentioned two figures, DIF for 3.3 series seems to be higher than 4.4 series
due to their contrast nature of failure mode. It is noteworthy that DR beams had less
DIF as compared to SR; however for DR3.3_0.84_0.15, resistance for low loading
rate has been used to calculate DIF. As there is no sign of ductile behavior with
strain hardening (i.e. sudden drop in load-deformation curves after peak load
- 58 -
although the longitudinal tensile reinforcements yielded before the beam attained
peak resistance) in the load vs. mid-span deflection curves of DR3.3_0.84_0.15,
thus the peak resistance obtained from testing under static loading would
overestimate the DIF.
(a)
140 160
High High DR3.3_0.84_0.11
120 SR3.3_0.84 140
100 120 Medium
Medium
Load (kN)
Load (kN)
80 100 Low
60 Low 80
40 60
20 Static 40 Static
0 20
-20 0
0 5 10 15 20 25 30 0 10 20 30 40 50
Mid-span deflection (mm) Mid-span deflection (mm)
160
140 High Medium
120
100
Load (kN)
80
60
Low
40
20 Static DR3.3_0.84_0.15
0
0 5 10 15 20 25 30 35
Mid-span deflection (mm)
specimens under varying loading rates (continued)
- 59 -
(b)
70 80
High SR4.4_0.82 High DR4.4_0.82_0.13
60 70
50
60
Load (kN)
Load (kN)
50
40
40
30
Medium 30
20 Low 20 Medium
10 Static 10 Static Low
0 0
0 10 20 30 40 50 0 10 20 30 40 50
Mid-span deflection (mm) Mid-span deflection (mm)
80
70 High DR4.4_0.82_0.19
60
Load (kN)
50
40
30
Medium
20
10 Static
0
0 5 10 15 20 25 30 35 40
Mid-span deflection (mm)
specimens under varying loading rates
(a) (b)
1.7 1.7
Dynamic Increase Factor (DIF)
SR3.3_0.84 SR4.4_0.82
1.6 1.6
DR3.3_0.84_0.11 DR4.4_0.82_0.13
1.5 DR3.3_0.84_0.15 1.5 DR4.4_0.82_0.19
1.4 1.4
1.3 1.3
1.2 1.2
1.1 1.1
1 1
0.01 0.1 1 10 0.01 0.1 1 10
Fig. 3.11: Dynamic Increase Factor (DIF) of maximum resistance of (a) a d -3.3;
(b) a d -4.4 series specimens under varying loading rates
- 60 -
3.5.2 Concrete contribution ( vc ) to the shear stress
For evaluating the concrete contribution to the shear resistance, only SR beams
were considered and collected from the published database and the current testing
data. ACI318-08 [A2] only provides the coefficients of concrete contribution to
shear stress normalized by the square root of concrete compressive strength (i.e.
0.17 for SI unit system) under static loading only. ACI methodology of concrete
contribution to shear stress for static loading has been extending further to
incorporate the loading rate effects on the same. The testing shear resistance values
obtained from varying loading rates are normalized by the square root of concrete
compressive strength, the width and the effective depth of beams. Fig. 3.12 displays
the variation (e.g. increasing trend under high rates) of concrete contribution to
shear stress normalized by the square root of concrete compressive strength
v c /  
f c under different loading rate  . Few erratic data was observed in the
data provided by Mutsuyoshi and Machida [M10]. Yielding of tensile

reinforcement after the shear failure of the beam could be the reason behind this
inconsistent behavior.
0.5
Current test
Kulkarni & Shah
Mutsuyoshi & Machida
0.4
/f ' ^ 0.5 (SI unit)
0.3
c
-1
*b *d
0.2
-1
test
V
0.1
0
-6 -5 -4 -3 -2 -1 0 1
10 10 10 10 10 10 10 10
Loading rate (m/s)
Fig. 3.12: Concrete contribution to the shear stress
- 61 -
3.5.3 Concrete ( Vc ) and shear reinforcement ( Vs ) contribution to shear
resistance
Fig. 3.13 shows the concrete and shear reinforcement contribution to shear strength
of DR beams of a d -3.3 series, as the specimens of this series failed in shear
irrespective of loading rates. Some crude estimation has been done to calculate the
corresponding contribution to shear resistance. Concrete contribution has been
taken from SR beams of the same series as shear force was solely resisted by
concrete. To account for the stirrup contribution, shear resistance of the SR beams
has been deducted from the shear resistance of DR beams of same series by
neglecting the contribution of compression reinforcement to shear strength. From
Fig. 3.13, it is evident that the shear reinforcement contribution as compared to
concrete contribution is quite small under these four varying loading rates.
70 70
60 Vc DR3.3_0.84_0.11 60 Vc DR3.3_0.84_0.15
Vs Vs
50 50
Vc & Vs (kN)
Vc & Vs (kN)
40 40
30 30
20 20
10 10
0 0
4E-4 4E-2 4E-1 2E+0 4E-4 4E-2 4E-1 2E+0
Fig. 3.13: Concrete and shear reinforcement contribution to shear resistance
3.5.4 Cracking stiffness
Cracking stiffness of the beams was calculated by using the secant of the load vs.
midspan deflection curve passing through the point from which the curve changes
its slope. Then, it was simply computed by taking the quotient of the resistance and
the deflection of that corresponding point. Fig. 3.14 displays the variation of
cracking stiffness of the two series (i.e. a d -3.3 and 4.4) of specimens.
Enhancement in cracking stiffness of a d -3.3 series specimens was observed when
- 62 -
the loading rate increased from static to low and medium rates. Similarly, for a d -
4.4 series specimens, stiffness increases as the loading rate progresses from static to
medium. However, the above-mentioned figure lacks the stiffness under high rates
as the load vs. midspan deformation curves was obtained in this case by deducting
the inertia force from the top load cell (described in section 3.5.1). Due to the time-
lag between the top and bottom load-cell signals, the load vs. midspan deformation
response obtained from the bottom load cells under high rates may not be similar to
the same from the top load cell. As a result, the computed cracking stiffness for
high rates is not compared with other loading rates. The reason behind this time-lag
could be due to the inertia forces, as the initial force applied is solely resisted by the
inertia of the beam before the bottom load-cell attained any load.
(a) (b)
100 40
Cracking stiffness (kN/mm)
Cracking stiffness (kN/mm)
SR3.3_0.84 35 SR4.4_0.82
80 DR3.3_0.84_0.11 DR4.4_0.82_0.13
DR3.3_0.84_0.15 30 DR4.4_0.82_0.19
60 25
20
40 15
10
20
5
0 0
-4 -3 -2 -1 0 -4 -3 -2 -1 0
10 10 10 10 10 10 10 10 10 10
Fig. 3.14: Cracking stiffness of (a) a d -3.3; (b) a d -4.4 series beams under
varying loading rates
3.5.5 Energy absorption
To calculate the amount of energy absorption of the specimens under these varying
loading rates, the area under the load vs. midspan deflection curves needs to be
calculated. Before that, the generalized mid-span displacement range should be
defined to impose the same conditions for all the specimens. The deformation range
was defined up to peak load displacement (for shear-failure type beams) or three
times of the yield deformation (for flexure-failure type beams), solely depending on
- 63 -
the characterization of their load vs. midspan deflection curves. However, for
medium and high loading rate, the load-midspan deformation curves did not
displayed the „yield plateau‟. For these cases, peak load displacement has been
taken as benchmark point. Fig. 3.15 clearly illustrates the variation of the energy
absorption capacity of RC beams for static, low, medium and high loading rates.
The energy absorption capacity of the RC beams increased with the enhancement of
loading rates from static to high.
(a) (b)
3000 800
Energy absorption (kN-mm)
Energy absorption (kN-mm)

SR3.3_0.84 750 SR4.4_0.82
2500 DR3.3_0.84_0.11 DR4.4_0.82_0.13
700 DR4.4_0.82_0.19
DR3.3_0.84_0.15
2000 650
1500 600
550
1000
500
500
450
0 400
-4 -3 -2 -1 0 1 -4 -3 -2 -1 0 1
10 10 10 10 10 10 10 10 10 10 10 10
Fig. 3.15: Energy absorption of (a) a d -3.3; (b) a d -4.4 series beams
3.5.6 Comparison of time to reach peak load, yielding of tensile and shear
reinforcements & peak strain rate
Table 3.2 represents the comparison between the time to reach peak load and
yielding of tensile and shear reinforcements for 3.3 series specimens as these series
specimens failed in shear at their ultimate resistance. The term „Average‟ in Table
3.2 refers to the mean value of two strain gauge (e.g., SA1,2 and SS1,2) readings. It
is worthwhile to know that whether the longitudinal and shear reinforcements
yielded before or after attainment of the peak resistance. From this table, it is clear
that except static loading, the tensile reinforcements yielded at the very early stages
for all loading rates. When the tensile reinforcements yielded before the shear
failure, the DIF would be less than the case of yielding of tensile reinforcements
after shear failure. However, only few cases were seen where shear reinforcement
yielded before shear failure or not yielded at all. Due to the larger spacing of shear
- 64 -
reinforcements, diagonal shear cracks did not intersect (i.e. propagated in between
stirrups) them. Additionally, strain rate was calculated by differentiating the strain
history data of longitudinal and shear reinforcements. To define the materials (e.g.
concrete and steel) behavior, most of the formulations were developed for certain
strain rate intervals; therefore, it is essential to determine the range of strain rates.
Table 3.2: Comparison between the time to reach peak load and the yielding of
tensile and shear reinforcements
Static
Low Medium High
Specimens Time
(s)
(s) (ms) (ms)
SR3.3_0.84 362.5
0.22 20.0 3.30
DR3.3_0.84_0.11 Tpl 181.4
0.50 34.0 5.70
DR3.3_0.84_0.15 188.4
0.57 36.0 5.40
SA1,2
SA1,2 SA1,2 SA1,2
Average
SR3.3_0.84 359.3 0.125 11.4 1.84
DR3.3_0.84_0.11 Tylt 174.9 0.084 12.2 1.49
DR3.3_0.84_0.15 183.6 0.095 11.6 1.68
SS1,2 SS1,2 SS1,2 SS1,2
Average
DR3.3_0.84_0.11 152.3 0.32 NY 4.20
Tys
DR3.3_0.84_0.15 NY NY NY 1.21
Note:-Tpl-Time to reach peak load; Tylt-Time to yield longitudinal tensile
reinforcement; Tys-Time to yield shear reinforcement; NY-Not Yielded
(a) (b)
1 1
10 10
Maximum strain rate (s )
-1
-1
0
0 10
10
-1
10
-1
10
-2
10
-2
10 -3
10
-3
10 Tensile longitudinal -4
reinforcements 10 Shear reinforcements
-4 -5
10 10
-4 -3 -2 -1 0 1 -4 -3 -2 -1 0 1
10 10 10 10 10 10 10 10 10 10 10 10
Fig. 3.16: Peak strain rates of (a) longitudinal (b) shear reinforcements for all
specimens
- 65 -
Thus, the variations of peak strain rates (i.e. choosing the maximum value from the
differentiation of strain history data) over these loading rates for longitudinal and
shear reinforcements are indicated in Figs. 3.16 (a) and (b) respectively.
3.5.7 Deflected shapes
As mentioned previously, accelerometers were placed to the test specimens along

the entire length on their top face in between the support points. Assuming zero
displacement at support, the displacement time history of each point could be
determined by double integrating the acceleration data.
2.5 ms 10ms 17.5 ms 2.5 ms 10 ms 17.5 ms

5 ms 12.5 ms 20 ms 5 ms 12.5 ms 20 ms
(a) 7.5 ms 15 ms 22.5 ms 7.5 ms 15 ms 22.5 ms
5 5
DR3.3_0.84_0.11_H DR3.3_0.84_0.15_H
0
Displacement (mm)
0
Displacement (mm)
-5
-5
-10
-15 -10
-20
-15
-25
-20
-30
-35 -25
0 200 400 600 800 1000 1200 1400 0 200 400 600 800 1000 1200 1400
Specimen coordinate (mm) Specimen coordinate (mm)
2.5 ms 10 ms 17.5 ms 2.5 ms 10 ms 17.5 ms

5 ms 12.5 ms 20 ms 5 ms 12.5 ms 20 ms
(b) 7.5 ms 15 ms 22.5 ms 7.5 ms 15 ms 22.5 ms
5 5
SR4.4_0.82_H DR4.4_0.82_0.13_H
0 0
Displacement (mm)
Displacement (mm)
-5 -5
-10 -10
-15 -15
-20 -20
-25 -25
-30 -30
-35 -35
0 200 400 600 800 1000 1200 1400 0 200 400 600 800 1000 1200 1400
Specimen coordinate (mm) Specimen coordinate (mm)
Fig. 3.17: Deflected shape of (a) a d -3.3; (b) a d -4.4 series specimens under high
loading rates
- 66 -
For a particular time, these data points are plotted and joined together by
smoothened lines to get the deflected shape of the beams. Fig. 3.17 displays the
deformed shapes of the beams in 2.5 ms interval for high loading rates. Due to the
damage of some data (i.e., data from accelerometers), two types of specimens from
each series are presented here. These figures distinctly show the one-sided shear
failure of a d -3.3 series specimens and flexure-type failure of a d -4.4 series
specimens. So, the failure pattern can be observed in the deflected shapes of the
beams in any time of the loading history. However, to get more precise deflected
shape of beams, closely-placed accelerometers need to be installed.
3.5.8 Crack patterns
Fig. 3.18 shows the crack patterns of all specimens except DR4.4_0.82_0.19_L
under these loading rates. SR beams of 3.3 series exhibited brittle shear-critical
behavior irrespective of loading rates. Diagonal shear cracks originated from the
mid-height of beam and propagated towards loading and support points for all
loading rates. However, for DR specimens, diagonal shear cracks under medium
and high loading rates, propagated with much steeper angles towards the bottom
surface of the beam as compared to static and low loading rates. In these cases,
diagonal cracks under medium and high loading rates tried to punch the mid-part of
the specimens towards bottom surface of the beam at steep angles as the shear
reinforcements gave resistance to transmit these cracks towards supports. In
general, for 3.3 series of specimens, diagonal tension and/or shear tension type
failure was observed in SR beams whereas diagonal tension and/or shear
compression were the dominant failure mode for DR specimens. It was observed
that all 3.3 series specimens failed in shear at their peak resistance, however
measurement of steel strain data exhibited the yielding of tensile reinforcements
before peak load or shear failure occurred for low, medium and high loading rates.
Development and propagation of cracks around different points (yielding of tensile
reinforcement, first hairline shear crack, peak load etc.) in load vs. time curves were
portrayed in Fig. 3.19 for SR3.3_0.84 under high loading rates. Similarly,
development and propagation of cracks around different points in the load vs. time
curves were presented in Fig. 3.20 and Fig 3.21 for DR3.3_0.84_0.11 and
- 67 -
DR3.3_0.84_0.15 respectively, under high loading rates. Two explanations could

be used to justify the shear failure of DR3.3 series specimens under high loading
rate. One is the wider spacing of web reinforcements and the other one is the
increment of flexural resistance of beams under low, medium and high loading rates
due to strain rate sensitivity of yield stress of tensile reinforcements.
(a)
SR3.3_0.84 DR3.3_0.84_0.11 DR3.3_0.84_0.15
Static
Low
Medium
High
(b)
SR4.4_0.82 DR4.4_0.82_0.13 DR4.4_0.82_0.19
Static
Low
Medium
High
Fig. 3.18: Crack patterns of (a) a d -3.3; (b) a d -4.4 series specimens under
varying loading rates
- 68 -
Considering the DIF of yield stress of tensile reinforcements, it was calculated that
flexural resistance boosted around 18.5, 21.3 and 24.8% for the above-mentioned
loading rates as compare to assuming static yield stress for all rates. However,
strain rate sensitivity of yield stress of stirrups did not contribute much (in majority
of cases stirrups did not yield) in their shear resistance. As a result, the shear to
bending resistance ratios become in the range of 0.9-1 which was 1.1-1.2 at the
initial design stage (considering static yield stress).
120
C A-Yielding of tensile rebars
100 B-First hairline shear crack
B
C-Peak load
80 D
Load (kN)
D-80% of Peak load

E-50% of Peak load
60 A
E in descending branch
40
SR3.3_0.84_H
20
0
0 0.005 0.01 0.015 0.02 0.025 0.03
Time (s)
A B
C D
Fig. 3.19: Development of cracks in SR3.3_0.84 under high loading rates
For 4.4 series specimens, ductile flexure type failure with some spalling and
crushing of concrete below the loading plate was observed for all loading rates. In
- 69 -
these cases, flexural cracks usually started from the bottom surface of the beam and
propagated vertically upwards and successively crushed the compression concrete.
140
E A-Yielding of tensile rebars
120
B-Hairline bending crack
C
100 C-Shear crack
Load (kN)
D D-Yielding of shear rebar
80 E-Peak load
B F
60 F-50% of Peak load
in descending branch
A
40
20 DR3.3_0.84_0.11_H
0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07
Time (s)
A B
C D
E F
Fig. 3.20: Development of cracks in DR3.3_0.84_0.11 under high loading rates
- 70 -
120
D DR3.3_0.84_0.15_H
100
C
B E
80
Load (kN)
A A-Yielding of tensile rebars
60 B- Shear crack
C-Bending crack
40 D-Peak load
E-Crushing of compression
20 concrete
0
0 0.01 0.02 0.03 0.04
Time (s)
A B
C D
Fig. 3.21: Development of cracks in DR3.3_0.84_0.15 under high loading rates
- 71 -
3.6 Summary
This chapter presents the experimental investigation on RC beams under a wide

spectrum of loading rates. Based on the observation and analyses of experimental
results, the following conclusions can be drawn:
(1) With increasing loading rates, the ultimate load resistance, cracking stiffness
and energy absorption of RC beams were found to increase. When the loading rate
progressed from low to high, an increasing trend in DIF was observed. In addition,
specimens having smaller a d ratio produced higher DIF as compared to the
specimens having higher a d ratio for low, medium and high loading rates. For
instance, 3.3 series specimens had higher DIF as compared to 4.4 series. Moreover,
for low, medium and high loading rates if the tensile reinforcements yielded before
the shear failure, the DIF would be lesser than the case where yielding of tensile
reinforcements was followed by the shear failure. For the first case, strain rate
sensitivity of yield stress of reinforcements controls the magnitude of DIF whereas
strain rate sensitivity of the tensile strength of concrete controls the magnitude of
DIF for the second case. As, the tensile strength of concrete is more rate sensitive
than yield stress of reinforcements, thus the DIF is higher for second case.
(2) High loading rates seem to have beneficial effect on shear resistance as concrete
contribution increases with increasing loading rates. Analysis response revealed that
compared to shear reinforcement contribution, concrete contribution governs in
shear resistance contributions under these loading environments. Moreover, peak
strain rate was amplified in one order of magnitude (approximately, 10 times) as the
loading rates progressed from low to high.
(3) For 3.3 series specimens, diagonal tension and/or shear tension type failure was
observed in SR beams whereas diagonal tension and/or shear compression were the
dominant failure mode for DR specimens. In all loading rates of the above-
mentioned series, diagonal shear cracks originated from the mid-height of beam and
propagated towards loading and support points. However, for DR specimens,
diagonal shear cracks under medium and high loading rates propagated with much
steeper angles towards the bottom surface of the beam as compared to static and
- 72 -
low loading rates. On the other hand, for 4.4 series specimens, ductile flexure type
failure was observed for all loading rates.
(4) After observing the wide scatter of results, it is being realized that there is a
great need of consensus among researchers in high-rate testing of RC beams. Test
set up, frequency response of measurement system, beam geometry, reinforcing
detailing and material properties should be kept nearly identical to facilitate better
analysis and effective comparisons.
- 73 -
- 74 -
Chapter 4: Numerical Simulation of Loading Rate Effects on RC beams
CHAPTER 4
NUMERICAL SIMULATION OF LOADING RATE
EFFECTS ON RC BEAMS
4.1 Introduction
From the literature review and experimental investigation chapter on loading rate
effects of RC beams, it is quite evident that research in this field in terms of
experimental and numerical means is quite limited. Although the experimental
investigation presented in the previous chapter can provide significant information
about the tested specimens; however it could be a time-consuming and costly
option. Therefore, it is worthwhile to develop reliable numerical model that can
serve as an alternative to experimental studies. The objective of this chapter is to
numerically predict the experimental behavior (e.g. load-mid-span deflection
characteristics and failure modes etc.) of beams under varying loading rates to
investigate the potentiality of explicit finite element (FE) program. Adhikary et al.
[A3] developed three-dimensional non-linear FE model to analyze the structural
response of beams under these loading conditions. Although this FE model can
capture the structural response up to medium loading rate within reasonable
accuracy; however for the case of high loading rates, the prediction deviates from
the test results. Moreover, they recommended for a numerical parametric study to
investigate the influence of various grade of steel reinforcement and concrete. Some
researchers suggested for the investigation of the effects of longitudinal
reinforcement ratio. It could be of some interest to observe the behavior mainly the
failure mode when the longitudinal reinforcement is increased in an under-
reinforced beam with intermediate-slenderness. Thus, to expand the knowledge
beyond the range of parameters investigated experimentally and their effects on the
structural response, three-dimensional nonlinear FE model is developed. LS-DYNA
[H1] was employed in this study because of its proven effectiveness in geometric
modeling and analysis capability for high rates of concentrated loading. However,
prior to trusting the results of any numerical model, it is of paramount interest to
validate the numerical results with the test results. Therefore, the proposed FE
- 75 -
model is validated by the test results and then employed further to conduct
parametric studies.
4.2 Finite Element Modeling Techniques of Reinforced Concrete
Finite element models of RC structures have generally been based on replacing the
composite continuum by an assembly of elements representing the concrete and the
steel reinforcement. Three alternative approaches are mainly used for modeling
reinforcement in a three-dimensional finite element model of a concrete structure:
the discrete model, the embedded model, and the smeared model (El-Mezaini and
Citipitiouglu [E1]). In discrete model, reinforcement is modeled using beam
element connected to the nodes of concrete mesh. In this model, perfect bond is
assumed in between the concrete and reinforcement. Hence, concrete and
reinforcement mesh share the common nodes. This model is simple in concept and
it has advantage of representing different material properties more precisely. The
only shortcoming of using the discrete model is that concrete mesh patterns are
restricted by the location of reinforcement. To overcome the problem of mesh
dependency in discrete model, the embedded model could be used where the
concrete mesh could be chosen independently. The embedded formulation is based
on the idea of assessing the stiffness of reinforcing elements individually from the
concrete elements. The strain compatibility is managed by an assumed bond-slip
relationship. To model the bond-slip effects, reinforcing bars are modeled using
special elements connected to concrete through fictitious springs. However, the
additional nodes requirement for the reinforcement increases the total number of
degrees of freedom (DOF) and the computational efforts. In the smeared
formulation, the reinforcing bars are assumed to be uniformly distributed over the
concrete elements. Assuming perfect bond in between concrete and reinforcement,
the constitutive relations of the elements are derived from the composite theory. In
large structural model where the reinforcement details are not so critical on overall
response, this model could be employed.
- 76 -
4.3 Finite Element Model
The nonlinear explicit FE program LS-DYNA [L1] was employed in the numerical
simulation process. The advantage of using explicit method is that there is no need
to calculate stiffness and mass matrices for the entire system. The solution can be
carried out at the element level and thus, relatively less storage is required as
compared to implicit method. However, explicit methods are conditionally stable
and thus small time step would be needed. To make the computation stable, the
selection of time step should be such that it is less than the time required for a stress
wave to travel through the shortest element of a FE model. Hence, excessive
computational time will be needed when the level of discretization increases. The
description of modeling includes structural geometry of beam, loading plate,
support roller, boundary conditions, application of load, and relevant material
properties.
4.3.1 Structural configuration
To simulate the same experimental conditions, it is essential to develop a FE model

as close to the physical system as possible. Therefore, a three-dimensional model is
the obvious choice. The three-dimensional FE mesh was developed which consist
of solid elements to represent concrete, support rollers and loading plate and beam
elements to characterize longitudinal and transverse reinforcements. Different mesh
sizes have been considered to study mesh sensitivity during the initial stage of
numerical study. Here, a mesh size of 15 mm was used to create the solid elements
which seemed to be ideal as further decrease in mesh size almost generated the
same results however augmented the computational time. Table 4.1 presents total
number of elements and nodes for each beam model. Fig. 4.1 shows the three-
dimensional FE model of beams (i.e. one type from each series), support rollers and
loading plate. Moreover, Fig. 4.1 also displays the schematic of the reinforcing
cages of the beam. Eight node solid elements with a single integration point were
used to represent concrete in the models. One of the disadvantages of single-point
integration is the appearance of zero energy modes which are also called hourglass
modes. LS-DYNA [L1] has several hourglass control types to avert the occurrence
- 77 -
of these modes. The Flanagan-Belytschko integration form was used in this study.
The 2-node Hughes-Liu beam element formulation with 2  2 Gauss quadrature
integration was used to model all reinforcing bars (i.e. longitudinal and transverse
reinforcements). This element formulation is simple and allows for computational
efficiency and robustness. It is also compatible with the brick elements, because the
element is based on a degenerated brick element formulation and moreover, it
includes finite transverse shear strains. This element contains six translational and
six rotational degrees of freedom. Furthermore, a reference node is required for
each element to determine the initial orientation of the cross-section. The steel
reinforcing bars were modeled explicitly using beam elements connected to the
concrete mesh nodes. The nodes that linked the concrete and reinforcement mesh
were shared and therefore unable to slip. Due to this assumption of complete
compatibility of strains between the concrete and steel nodes, they formed a perfect
bond. Ultimate dynamic bond at failure is 70-100% higher than that of quasi-static
loading conditions for the case of deformed reinforcing bars (Weathersby [W2]).
Moreover, there is not enough time to develop extensive bond slip along the length
of the reinforcements during high loading rates. Thus the assumption of perfect
bond in between the reinforcement and concrete is quite reasonable. Moreover,
loading plate and support rollers were also modeled by solid elements.
Table 4.1: Total number of elements and nodal points for FE model
Solid elements
Beam designation Beam elements Nodal points
for concrete
SR3.3_0.84 21317 228 35060

DR3.3_0.84_0.11 22967 634 37327
DR3.3_0.84_0.15 22967 702 37395
SR4.4_0.82 11856 228 14718
DR4.4_0.82_0.13 12063 594 25353
DR4.4_0.82_0.19 12063 646 25405
- 78 -
Fig. 4.1: Three-dimensional FE model of RC beams (a) a d -3.3; (b) a d -4.4

series
- 79 -
4.3.2 Boundary conditions, application of load and contact algorithm
To simulate the actual experimental conditions, the beams were supported on two
rigid cylinders made of solid elements. Constraints were defined to the support
cylinder, so that it could rotate about its own longitudinal axis but would not be
able to translate. The load was prescribed (by
BOUNDARY_PRESCRIBED_MOTION_RIGID keyword) to the beam through a
rigid plates located at the top of the beam at its mid-span by specifying the same
vertical displacement-history used during testing. The rigid loading plate was
allowed to move only in the vertical direction. The prescribed displacement-history
was similar to the displacement-history applied during experiment which
successively generated the desired rate of loading. The corresponding load due to
the prescribed displacement was then determined by monitoring the vertical
reaction forces at the concrete nodes in contact with the support solid cylinders.
The algorithm „CONTACT AUTOMATIC SURFACE TO SURFACE‟ in LS-
DYNA [L1] was used to model the contact between the support cylinders, loading
plate and beam. This algorithm automatically generates slave and master surfaces
and uses a penalty method where nominal interface springs were used for
interpenetration between element and surfaces.
4.3.3 Constitutive model for concrete
One of the main challenges in modeling the behavior of RC structures is the

difficulty of incorporating realistic material model that can precisely represent the
behavior of the physical system. Concrete behaves quite differently from the other
homogeneous material such as metals, due to its heterogeneous composition and
inelastic behavior. Clear distinction exists in behavior between metals and concrete
due to the random distribution of fine and coarse aggregate particles throughout the
hardened cement matrix in concrete and the nonlinear behavior of it during the
process of loading. To understand with the complex behavior of concrete, extensive
research over the past two decades has resulted in a variety of constitutive models
that are capable of representing the various aspects of concrete behavior (Bazant et
al. [B4]; Kim and Lee [K4]; Chan et al. [C1]; Foster et al. [F1]; Kwak and Kim
- 80 -
[K14]). Despite the extensive research efforts, especially in the field of concrete
materials, still there is a demand for a reliable material model capable of
representing the complex behavior of concrete, and therefore it is an active research
area. Concrete material formulation can be categorized into elasticity-based models,
plasticity-based models, plastic-fracturing models, elasto-plastic damage models
and endochronic models. This is beyond the scope of this research to discuss all
material formulations in details and only the material model employed in the
current simulation will be demonstrated.
A number of material models from LSDYNA [L1] material library can be used to
represent concrete, such as Material type 005 (MAT SOIL AND FOAM); Material
type 014 (MAT SOIL AND FOAM FAILURE); Material type 016 (MAT
PSEUDO TENSOR); Material type 025 (MAT GEOLOGIC CAP MODEL);
Material type 072 (MAT CONCRETE DAMAGE); Material type 084 (MAT
WINFRITH CONCRETE); Material type 096 (MAT BRITTLE DAMAGE);
Material type 159 (MAT CSCM CONCRETE); Material type 072 R3 (MAT
CONCRETE DAMAGE REL3). In this study, MAT CSCM CONCRETE was
employed to model concrete. This model was developed mainly for roadside safety
application, such as concrete bridge rail and portable barrier (made of normal
strength concrete) impacted by vehicles, but it could also be applicable to other
dynamic applications. This is a cap model with a smooth intersection between the
shear yield surface and hardening cap, as shown in Fig. 4.2. The initial damage
surface coincides with the yield surface and the rate effects are modeled with
viscoplasticity. The yield surface is formulated in terms of three stress invariants:
J 1 , the first invariant of stress tensor; J 2 , the second invariant of the deviatoric
stress tensor; and J 3 , the third invariant of the deviatoric stress tensor. The
invariants are defined in terms of the deviatoric stress tensor, S ij and pressure, P as
follows:
J 1  3P (4-1)
1
J 2  S ij S ij (4-2)
2
- 81 -
1
J 3  S ij S jk S ki (4-3)
3
Fig. 4.2: General shape of the concrete model yield surface in two dimensions
(Murray [M10])
The three-invariant yield function is delineated in terms of these three invariants

and the cap hardening parameter,  , as shown in following equation:
f J 1 , J 2 , J 3 ,   J 2   2 F f2 Fc (4-4)
Here F f is the shear failure surface; Fc is the hardening cap; and  is the Rubin
three-invariant reduction factor. The cap hardening parameter  is the value of the
pressure invariant at the intersection of cap and shear surfaces. For the shear failure
surface, the strength of concrete is modeled by the shear surface in the tensile and
low confining pressure regimes. The shear failure surface F f is defined as follows:
F f J 1      exp  J1  J 1 (4-5)
Here the values of  ,  ,  and  are selected by fitting the model surface to the
strength measurements from tri-axial compression tests conducted on plain concrete
cylinders. The strength of the concrete is modeled by the combination of cap and
shear surfaces in the low to high confining pressure regimes. The cap is used to
model plastic volume change related to pore collapse (although the pores are not
explicitly modeled). The initial location of the cap determines the onset of plasticity
in isotropic compression and uniaxial strain. The elliptical shape of the cap allows
- 82 -
the onset for isotropic compression to be greater than the onset for uniaxial strain,
in agreement with shear enhanced compaction data. Without ellipticity, a “flat” cap
would produce identical onsets. The motion of the cap determines the shape
(hardening) of the pressure-volumetric strain curves via fits with data. Without cap
motion, the pressure-volumetric strain curves would be perfectly plastic. Concrete
usually exhibits softening (strength reduction) in the tensile and low to moderate
compressive regimes and softening behavior is modeled through a damage
formulation. Without the damage formulation, the cap model predicts perfectly
plastic behavior for laboratory tests such as direct pull, unconfined compression,
triaxial compression and triaxial extension. This behavior is not realistic although
perfectly plastic response is typical of concrete at high confining pressures; it is not
representative of concrete at lower confinement and in tension. The damage
formulation models both strain softening and modulus reduction. Damage initiates
and accumulates when strain-based energy terms exceed the damage threshold.
Moreover, the strain rate effects on concrete strength are duly considered in this
model. Default parameter generation option (by providing some key specifications)
has been used herein instead of the traditional method where all material parameters
were needed. Default parameters are provided for the concrete model based on
three input specifications: the unconfined compressive strength of concrete,
aggregate size and units. This option is valid for unconfined compressive strength
from 20 to 58 MPa and aggregate sizes from 8 to 32 millimeters. Further details of
this material model can be found somewhere (Murray [M10]).
4.3.4 Constitutive model for reinforcing steel
The steel reinforcement bars (longitudinal and transverse reinforcements) within

the beam were modeled as a strain sensitive uniaxial elastic-plastic material to
account for its strain rate sensitivity as well as stress-strain history dependence.
Material model MAT PIECEWISE LINEAR PLASTICITY (MAT 024) was used
in this study to model reinforcing bars where the strain rate effects was duly
considered. The expressions proposed by Malvar [M2] on strain rate effects on
yield and ultimate stress of steel reinforcements were utilized in this study. Fig. 4.3
shows this proposed DIFs for both yield and ultimate stress. The DIFs for steel can
- 83 -
be expressed as follows:
DIFs  ( / 10 4 ) (4-6)
   fu ;  fu  0.019  0.009 f u 414 ;  is the strain rate in s-1 and f y is the

reinforcement static yield strength in MPa. Eq. (4-6) is valid with yield stresses
between 290 and 710 MPa, and for strain rates between 10-4 and 10 s-1.
1.7
Grade 40 yield
1.6 Grade 60 yield
Dynamic Increase Factor
Grade 75 yield
1.5
Grade 40 ultimate
1.4 Grade 60 ultimate
Grade 75 ultimate
1.3
1.2
1.1
1.0
0.0001 0.001 0.01 0.1 1 10
Strain Rate (1/s)
Fig. 4.3: Proposed Dynamic Increase Factor (DIFs) for reinforcing steel (Malvar
[M2])
4.3.5 Constitutive model for supports and loading plate
MAT RIGID (MAT 020) was used from LS-DYNA [L1] material library to model
the supports and loading plate. Realistic values of Young‟s modulus and Poisson‟s
ratio of the rigid material should be defined since unrealistic values may contribute
to numerical problems in contact. Young‟s modulus and Poisson‟s ratio of steel
material were used for the rigid material in the numerical simulation.
4.4 Validation of Finite Element Analysis Results
Before proceed further to carry out any case studies with the developed FE model,
- 84 -
it is essential to validate or verify the FE results with the experimental results to

establish the reliability of the model. Therefore, this section provides the
comparison of FE results with the test results presented in the previous chapter.
Load vs. midspan deflection characteristics and crack pattern of the beams for
various loading rates (e.g. static, low, medium and high) was used here for
verification purpose.
4.4.1 Load vs. midspan deflection curves
The comparison of load vs. midspan deflection curves between test results and FE
predictions is illustrated in Fig. 4.4. From this figure, it can be concluded that the
FE models can predict the load vs. midspan deflection response of beam under
varying loading rates within a reasonable accuracy.
140 140
120 Static DR3.3_0.84_0.11 120 Low DR3.3_0.84_0.11
100 100
Load (kN)
Load (kN)
80 80
60 60
40 40
Experiment Experiment
20 Simulation 20 Simulation
0 0
0 5 10 15 20 0 5 10 15 20 25 30
Midspan deflection (mm) Midspan deflection (mm)
140 160
120 Medium DR3.3_0.84_0.11 140 High DR3.3_0.84_0.11

120
100
Load (kN)
Load (kN)
100
80
80
60
60
40
Experiment 40 Experiment
0 0
0 5 10 15 20 0 5 10 15 20 25
loading rates: (a) 3.3 series; (b) 4.4 series (continued)
- 85 -
120 140
Static DR3.3_0.84_0.15 120 Low DR3.3_0.84_0.15
100
100
80
Load (kN)
Load (kN)
80
60
60
40
40
0 0
0 5 10 15 0 5 10 15 20
140 160
120 Medium DR3.3_0.84_0.15 140 High DR3.3_0.84_0.15

120
100
Load (kN)
Load (kN) 100

80
80
60
60
40
0 0
0 5 10 15 20 0 5 10 15 20 25
(a)
60 60
50
Static DR4.4_0.82_0.13
50 Low DR4.4_0.82_0.13
40 40
Load (kN)
Load (kN)
30 30
20 20
0 0
0 5 10 15 20 25 0 5 10 15 20 25
loading rates: (a) 3.3 series; (b) 4.4 series (continued)
- 86 -
60 70
Medium DR4.4_0.82_0.13 60 High DR4.4_0.82_0.13
50
50
40
Load (kN)
Load (kN)
40
30
30
20
20
0 0
0 5 10 15 20 25 0 5 10 15 20 25 30 35 40
60 60
Static DR4.4_0.82_0.19 Medium DR4.4_0.82_0.19
50 50
40 40
Load (kN)
Load (kN)
30 30
20 20
0 0
0 5 10 15 20 0 5 10 15 20
70
60 High DR4.4_0.82_0.19
50
Load (kN)
40
30
20
Experiment
10 Simulation
0
0 5 10 15 20 25 30 35 40
Midspan deflection (mm)
(b)
loading rates: (a) 3.3 series; (b) 4.4 series
Although, the FE models predicts the peak resistance for all loading rates quite
accurately; however, some deviation in stiffness is observed mainly for some cases
in high loading rates (e.g. DR3.3_0.84_0.15, DR4.4_0.84_0.13 and
DR4.4_0.84_0.19). This may be due to the simplification made in the previous
- 87 -
chapter that for high loading rates inertia force was deducted from the top load cell
data by considering the beam as a SDOF system to evaluate its true resistance.
Accurate calculation of the inertia force is a challenging issue as it is the
distribution of forces along the length of the specimens. Moreover, for
DR3.3_0.84_0.15 under static loading, the load vs. midspan deflection
characteristic predicted by FE analysis did not project the experimental behavior.
Although the longitudinal tensile reinforcements yielded before the beam attained
peak resistance, still sudden drop in experimental load-midspan deformation curves
after peak load was observed. After yielding, the tensile reinforcements did not
exhibit strain-hardening characteristics and in the meantime due to the wider
spacing (i.e. greater than the limit specified by ACI 318-08 [A2]) of transverse
reinforcements, sudden drop in load-midspan deformation after peak resistance was
perceived. Comparison of peak load in between test and numerical analysis results
was summarized in Table 4.2 for all beams expect DR4.4_0.82_0.19 for low rates
due to the lack of test result. It is evident that the FE models could capture the peak
load under varying loading rates within an acceptable accuracy. The average value
of the ratio of peak load in between test and simulation is 0.98 and the coefficient
of variation is less than 10%. It was mentioned in the introduction section of this
chapter that the FE model developed by Adhikary et al. [A3] did not capture the
load vs. midspan deflection response of the beam under high loading rates quite
satisfactorily. Thus, comparison of load vs. midspan deflection of RC3_S56 under
varying loading rates was also made to check the accuracy of the current FE model,
as shown in Fig. 4.5.
- 88 -
250 300
Static 250 Low
200
200
Load (kN)
Load (kN)
150
150
100
Exp._RC3_S56_S1 100 Exp._RC3_S56_L1
Exp._RC3_S56_S2 Exp._RC3_S56_L2
0 0
0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35
300 350
250
Medium 300 High
250
200
Load (kN)
Load (kN)
200
150
150
100 Exp._RC3_S56_M1 Exp._RC3_S56_H1
100
Exp._RC3_S56_M2 Exp._RC3_S56_H2
0 0
0 5 10 15 20 25 30 35 40 0 10 20 30 40 50
Fig. 4.5: Comparison of load vs. midspan deflection of RC3_S56 under varying
loading rates (Adhikary et al. [A3])
- 89 -
Table 4.2: Comparison of peak resistance of beam between test and FE analysis
results
Loading rate Peak Load (kN)

Specimens Test/Simulation
(m/s) Test Simulation
4  10-4 76.7 81.0 0.95
4  10-2 82.9 90.2 0.92
SR3.3_0.84
4  10-1 93.4 99.0 0.94
2  100 103.0 108.3 0.95
4  10-4 90.5 95.8 0.94
4  10-2
96.7 105.1 0.92
DR3.3_0.84_0.11
4  10-1
104.6 106.0 0.99
2  10 0
115.1 119.5 0.96
4  10-4 78.1 92.0 0.85
4  10-2 100.8 103.8 0.97
DR3.3_0.84_0.15
4  10-1 106.6 107.8 0.99
2  100 113.3 115.0 0.99
4  10-4
40.0 40.2 1.00
4  10-2
40.8 43.8 0.93
SR4.4_0.82
4  10-1
46.8 45.9 1.02
2  100 51.0 52.3 0.98
4  10-4 45.1 41.5 1.09
4  10-2 45.1 44.0 1.03
DR4.4_0.82_0.13
4  10-1 47.0 45.8 1.03
2  100 53.1 54.3 0.98
4  10-4
46.7 41.5 1.13
4  10-2
No data - -
DR4.4_0.82_0.19
4  10-1
47.5 46.1 1.03
2  100 54.0 55.7 0.97
Mean 0.98
COV 0.06
- 90 -
4.4.2 Failure patterns
After comparing the load vs. midspan deflection responses, this section is presented
the comparison of crack pattern of beams under varying loading rates. The damage
of the beams obtained from numerical simulation is shown by plotting the fringes
of effective plastic strain in the right hand side of Fig. 4.6. For DR3.3_0.84_0.11,
cracking patterns are almost similar to the experimental one for all loading rates.
However, for static loading the diagonal crack developed during experiment from
the loading point to the support point cannot be predicted by numerical simulation.
DR3.3_0.84_0.11
Static
Low
Medium
High
DR3.3_0.84_0.15
Static
Low
Medium
High
plastic strain from simulation of beams under varying loading rates (continued)
- 91 -
DR4.4_0.82_0.13
Static
Low
Medium
High
DR4.4_0.82_0.19
Static
Medium
High
plastic strain from simulation of beams under varying loading rates
To avoid excessive run time for the case of static loading in explicit method, the
event simulation time (i.e. in seconds) is much lesser than the experiment (i.e. in
few minutes). That means there was no enough time for the diagonal crack to
- 92 -
extend from loading to support points for the numerical simulation case. Moreover,
it is well recognized that for quasi-static problem, implicit methods are usually
more effective than explicit methods. To keep the consistency, here explicit
methods are employed for all loading cases. Similarly, for DR3.3_0.8_0.15, crack
patterns were well captured for all loading rates except static one. For 4.4 series
specimens, ductile flexure type failures were arrested effectively by the FE model
for all loading rates.
4.5 Parametric Study
4.5.1 Numerical simulation matrix
After verifying the numerical simulation results against the experimental results,
this section presents a parametric investigation to elucidate more information about
the behavior of beams under varying loading rates. The effects of some key
parameters such as longitudinal reinforcement ratio, transverse reinforcement ratio,
yield strength of longitudinal reinforcement and concrete grades on the response of
beam were studied. The influence of these parameters on DIF of maximum
resistance and failure modes was summarized and discussed. Fig. 4.7 illustrates the
general schematic diagram of the beams and Table 4.3 summarizes the specimen
characteristics of the simulation matrix.
d h
2a b
L
Fig. 4.7: General schematic diagram of beam
- 93 -
Table 4.3: Specimen characteristics of simulation matrix
 LT &  LB T
f c h b d a L
Beam mark ad Ratio f Ly Ratio f Ty Rs Rb
(MPa) (mm) (mm) (mm) (mm) (mm)
(%) (MPa) (%) (MPa)
DR2.8_0.89_0.13 0.13 0.93
DR2.8_0.89_0.26 0.89 0.26 1.19
DR2.8_0.89_0.42 0.42 1.52
290 180 250 2.8
DR2.8_1.69_0.13 0.13 0.51
DR2.8_1.69_0.26 1.69 0.26 0.65
DR2.8_1.69_0.42 0.42 0.84
DR3.3_0.84_0.19 0.19 1.32
DR3.3_0.84_0.38 0.84 0.38 1.80
DR3.3_0.84_0.56 0.56 2.20
40 250 150 210 700 1700 3.3 520 310
DR3.3_1.62_0.19 0.19 0.73
DR3.3_1.62_0.38 1.62 0.38 0.99
DR3.3_1.62_0.56 0.56 1.21
DR4.4_0.82_0.19 0.19 1.84
DR4.4_0.82_0.47 0.82 0.47 2.82
DR4.4_0.82_0.94 0.94 4.46
190 120 160 4.4
DR4.4_1.6_0.19 0.19 0.98
DR4.4_1.6_0.47 1.6 0.47 1.51
DR4.4_1.6_0.94 0.94 2.38
Note: All side cover 40 mm for 2.8 and 3.3 series and 30 mm for 4.4 series (from the centre of tensile and compression reinforcements)
- 94 -
4.5.2 Influence of longitudinal reinforcement ratio on DIF of maximum

resistance
Fig. 4.8 illustrates the effects of longitudinal reinforcement ratio on the DIF of
maximum resistance of beam under varying loading rates (low: 4 × 10-2, medium:
4 × 10-1 and high: 2 × 100 m/s). Three types of beams were considered here in terms
of their slenderness ratios ( a d -2.8, 3.3 and 4.4).
(a)
1.3 1.3

1.25 a/d-2.8 1.25 a/d-2.8
T-0.13% T-0.26%
1.2 1.2
1.15 1.15
1.1 1.1
L- 0.89% L- 0.89%
1.05 L- 1.69% 1.05
L- 1.69%
1 1
0.01 0.1 1 10 0.01 0.1 1 10
1.25
a/d-2.8
1.2
T-0.42%
1.15
1.1
L- 0.89%
1.05
L- 1.69%
1
0.01 0.1 1 10
Loading rate (m/s)
(b) 1.35 1.35

1.3 a/d-3.3 1.3 a/d-3.3

1.25  -0.19%
T
1.25 T-0.38%
1.2 1.2
1.15 1.15
1.1 1.1
1.05 L- 0.84% 1.05
L- 0.84%
1 L- 1.62% 1 L- 1.62%
0.95 0.95
0.01 0.1 1 10 0.01 0.1 1 10

resistance of beams (a) a d -2.8; (b) a d -3.3; (c) a d -4.4 (continued)
- 95 -
1.3

1.25 a/d-3.3
1.2 T-0.56%
1.15
1.1
1.05
1 L- 
0.95 L- 1.62%
0.9
0.01 0.1 1 10
Loading rate (m/s)
(c) 1.35 1.3

1.3 a/d-4.4 1.25 a/d-4.4

1.25
 -0.19% 1.2
T-0.47%
T
1.2 1.15
1.15 1.1
1.1 1.05
L- 0.82% L- 0.82%
1.05 L- 1.6% 1 L- 1.6%
1 0.95
0.01 0.1 1 10 0.01 0.1 1 10
1.4
1.3
a/d-4.4
T-0.94%
1.2
1.1
1 L- 0.82%
L- 1.6%
0.9
0.01 0.1 1 10
Loading rate (m/s)

resistance of beams (a) a d -2.8; (b) a d -3.3; (c) a d -4.4
From each type three cases were taken into account by varying transverse
reinforcement ratios. Then, finally for each distinct case, two amounts of
longitudinal reinforcements (approx. 0.3 and 0.6 times of balanced reinforcement
ratio) were considered to quantify the effect on the DIF of maximum resistance.
The amount of balanced reinforcement ratio is estimated to be 2.7%. Moreover, this
- 96 -
is worthwhile to mention that for all cases f c are taken as 40 MPa and f Ly , f Ty are
considered to be 520 and 310 MPa, respectively. It was observed that DIF are on
the higher side for beam with low reinforcement ratio. When the beam contains low
amount of longitudinal reinforcements, the response is mainly influenced by the
strain rate effect of longitudinal reinforcing steel which in turn causing an
enhancement of the DIF.
4.5.3 Influence of longitudinal reinforcement ratio on failure mode
To describe the effects of longitudinal reinforcement ratio on failure mode of beam

under varying loading rates, specimens from a d -3.3 and 4.4 series were chosen.
Left hand side of Fig. 4.9(a) displays the failure pattern of 3.3 series beam having
longitudinal reinforcements of 0.3 times of balanced reinforcement ratio whereas
right hand side shows longitudinal reinforcements of 0.6 times of balanced
reinforcement ratio and for all cases transverse reinforcement ratios were kept
identical (i.e. 0.56%). Similarly, Left hand side of Fig. 4.9(b) presents the failure
pattern of 3.3 series beam having longitudinal reinforcements of 0.3 times of
balanced reinforcement ratio whereas right hand side shows longitudinal
reinforcements of 0.6 times of balanced reinforcement ratio and for all cases
transverse reinforcement ratios are kept constant (i.e. 0.47%). This was observed
that when the beam contains relatively small amount of longitudinal reinforcements
(i.e. 0.3 times of balanced reinforcement ratio), only the formation of flexural
cracks were observed for all loading rates; whereas for the case of beam having
relatively higher amount of longitudinal reinforcements (i.e. 0.6 times of balanced
reinforcement ratio), both flexural-shear and some diagonal cracks were noticed.
Specifically, for high loading rates, extent of damage was more severe when the
beam contained relatively higher amount of longitudinal reinforcements. Therefore,
it can be concluded that change in failure mode may be observed (i.e. mainly in
high rates considered here) when the amount of longitudinal reinforcement ratio is
increased (i.e. approaching towards balanced-reinforcement ratio) in an under-
reinforced beam.
- 97 -
(a) DR3.3_0.84_0.56 DR3.3_1.6_0.56
Static
Low
Medium
High
(b) DR4.4_0.82_0.47 DR4.4_1.6_0.47
Static
Low
Medium
High
Fig. 4.9: Influence of longitudinal reinforcement ratio on failure mode: (a) a d -
3.3; (b) a d -4.4
- 98 -
4.5.4 Influence of transverse reinforcement ratio on DIF of maximum

resistance
The effects of transverse reinforcement ratio on the DIF of maximum resistance of

beam under varying loading rates (low: 4 × 10-2, medium: 4 × 10-1 and high: 2 × 100
m/s) are shown in Fig. 4.10.
(a) 1.3 1.3

1.25 a/d-2.8 1.25 a/d-2.8

L-0.89% 1.2 L-1.69%
1.2
1.15
1.15
1.1
1.1 T- 0.13% T- 0.13%
1.05
T- 0.26% T- 0.26%
1.05
T- 0.42% 1 T- 0.42%
1 0.95
0.01 0.1 1 10 0.01 0.1 1 10
(b) 1.4 1.3

1.25
1.3 a/d-3.3 a/d-3.3
 -0.84%
L
1.2 L-1.62%
1.2
1.15
1.1 1.1
1.05
1 T- 0.19% T- 0.19%
T- 0.38% 1
T- 0.38%
0.9 T- 0.56% 0.95 T- 0.56%
0.8 0.9
0.01 0.1 1 10 0.01 0.1 1 10
(c) 1.4 1.4

1.3 a/d-4.4 a/d-4.4

L-0.82% 1.3
L-1.6%
1.2
1.2
1.1
1.1
1 T- 
T- 0.47% T- 0.19%
0.9 1 T- 0.47%
T- 0.94%
T- 0.94%
0.8 0.9
0.01 0.1 1 10 0.01 0.1 1 10
Fig. 4.10: Influence of transverse reinforcement ratio on DIF of maximum

resistance of beams (a) a d -2.8; (b) a d -3.3; (c) a d -4.4
- 99 -
Here, for all cases f c were taken as 40 MPa and f Ly , f Ty were considered to be 520
and 310 MPa, respectively. It is evident that DIF is on the lower side when the
beam contains high amount of transverse reinforcements.
4.5.5 Influence of transverse reinforcement ratio on failure mode
The effects of transverse reinforcement ratio on failure mode of beam for medium
and high loading rates are shown in Fig. 4.11 and Fig. 4.12.
Medium
DR3.3_0.84_0.19
DR3.3_0.84_0.38
DR3.3_0.84_0.56
High
DR3.3_0.84_0.19
DR3.3_0.84_0.38
DR3.3_0.84_0.56

series beams under medium and high loading rates
- 100 -
Medium
DR4.4_0.82_0.19
DR4.4_0.82_0.47
DR4.4_0.82_0.94
High
DR4.4_0.82_0.19
DR4.4_0.82_0.47
DR4.4_0.82_0.94

series beams under medium and high loading rates
Here only medium and high loading rates were selected due to that the effects of
transverse reinforcement on cracking patterns are minimal for the considered beams
under static and low loading rates. Fig. 4.11 shows the failure patterns of 3.3 series
beam having three different amounts of transverse reinforcements (i.e. 0.19, 0.38
- 101 -
and 0.56%). From this figure, it was clear that when the beam contained lower
amount of transverse reinforcements (i.e. 0.19%), diagonal cracks were formed
with flexure cracks; whereas flexure type of failure modes were observed in beam
having high amount of transverse reinforcements (i.e. 0.56%). Similar type of
phenomena was noticed in the failure patterns of 4.4 series beam, as clearly
presented in Fig. 4.12.
4.5.6 Influence of yield strength of longitudinal reinforcement on failure

mode
Literature review on RC beams under varying loading rates revealed that rate
sensitivity of yield strength of longitudinal reinforcements plays a pivotal role in
determining the mode of failure. Kulkarni and Shah [K13] reported that the final
failure mode shifted from shear failure at static loading to flexure failure at the high
rate for three pairs of singly reinforced beams and may be due to relatively small
rate sensitivity of longitudinal reinforcement used. The above-mentioned
phenomenon runs counter to the brittle transition in the mode of failure of doubly
reinforced beam without shear reinforcements under high loading rates documented
by Mutsuyoshi and Machida [M11] and may be due to high rate sensitivity of
longitudinal reinforcement used by them. Therefore, this section provides some
case-studies to highlight the effect of yield strength of longitudinal reinforcement
on failure modes. Four types of beam were taken from Table 4.3 and some case-
studies were carried out by changing the yield strength of longitudinal
reinforcements, keeping all others parameters identical. The beam designations,
considered yield strengths of longitudinal reinforcements and corresponding static
shear to bending resistance ratios are shown in Table 4.4. It is noted that all
considered beams contain longitudinal reinforcement ratio of 0.6 times of balanced
reinforcement ratio.
- 102 -
Table 4.4: Beam characteristics for numerical case-studies
Type Beam mark f Ly (MPa) Rs Rb

DR3.3_1.62_0.19 420 0.89
(A)
DR3.3_1.62_0.19 520 0.73
DR3.3_1.62_0.38 420 1.21
(B)
DR3.3_1.62_0.38 520 0.99
DR4.4_1.6_0.19 420 1.17
(C)
DR4.4_1.6_0.19 520 0.98
DR4.4_1.6_0.47 420 1.79
(D)
DR4.4_1.6_0.47 520 1.51
Fig. 4.13 shows the influence of yield strength of longitudinal reinforcement on

failure modes of a d -3.3 series beams under varying loading rates. From Table
4.4, it is clear that type-A beams are statically shear-critical. There was no evidence
of flexure type failure (Fig. 4.13a) when the loading rates progressed from static to
high; however the severity of shear failure increased for high rates. First beam from
type-B (i.e. having yield strength of 420 MPa of longitudinal reinforcements) is
statically flexure critical as static shear to bending resistance ratio (i.e. 1.21) is
greater than one and from Fig. 4.13b, ductile type failure was observed for static
and low rates whereas punching shear cracks at midspan was noticed for medium
loading rate. With the increment of the loading rate to high, severe failure in terms
of both punching and diagonal shear cracks was the ultimate consequences. Due to
high rate sensitivity of low yield strength steel, the increment of yield strength in
medium and high rates was such a way that the bending resistance surpassed the
shear resistance and eventually failure mode changed. The static shear to bending
resistance ratio of second beam of type-B (i.e. having yield strength of 520 MPa of
longitudinal reinforcements) is almost equal to one, thus any type of failure (i.e.
ductile or brittle) could be possible for static loading. Here, ductile type failure was
perceived for static and low loading rates whereas for medium and high rates, both
punching and diagonal shear cracks were observed (Fig. 4.13b). Similarly, the
influence of yield strength of longitudinal reinforcement on failure modes of a d -
4.4 series beams under varying loading rates are presented in Fig. 4.14. Change in
failure mode from ductile at static loading to punching and diagonal shear failure at
high rates was seen for the first beam of type-C (static shear to bending resistance
- 103 -
ratio- 1.17). For both beams of type-D (the static shear to bending resistance ratio is
greater than 1.5), ductile type of failure is observed for all loading rates. Although
the severity in flexure failure increases in large extent, however, there was no
change in failure mode.
(a) DR3.3_1.62_0.19 ( f Ly - 420 MPa) DR3.3_1.62_0.19 ( f Ly - 520 MPa)
Static
Low
Medium
High
(b) DR3.3_1.62_0.38 ( f Ly - 420 MPa) DR3.3_1.62_0.38 ( f Ly - 520 MPa)
Static
Low
Medium
High

modes of a d -3.3 series beams under varying loading rates
- 104 -
(a) DR4.4_1.6_0.19 ( f Ly - 420 MPa) DR4.4_1.6_0.19 ( f Ly - 520 MPa)
Static
Low
Medium
High
(b) DR4.4_1.6_0.47 ( f Ly - 420MPa) DR4.4_1.6_0.47 ( f Ly - 520MPa)
Static
Low
Medium
High

modes of a d -4.4 series beams under varying loading rates
- 105 -
4.5.7 Influence of grade of concrete on peak load
Due to the limitation of constitutive model of concrete applied in this FE model, the
unconfined compressive strength of concrete used in this parametric study is within
58 MPa. Therefore, to delineate the effect of grade of concrete on peak load under
varying loading rates, the unconfined compressive strength of concrete was varied
from 30 to 50 MPa keeping all other parameters constant. For all cases, f Ly and
f Ty are considered to be 520 and 310 MPa, respectively. Influence of various grade
of concrete (e.g. 30, 40 and 50 MPa) on peak load of beams under varying loading
rates is shown in Fig 4.15. Increasing trend in peak load was observed when the
loading rate progresses from static to high. Moreover, beam having high strength
concrete (i.e. 50 MPa) yielded higher peak load for all loading rates as compared to
relatively low strength concrete (i.e. 30, 40 MPa) beams.
(a) 220
210 a/d-2.8
200 L-0.89%
T-0.42%
Peak load (kN)
190
180
170
160 30 MPa
40 MPa
150 50 MPa
140
0.0001 0.001 0.01 0.1 1 10
Loading rate (m/s)
(b) 140
130 a/d-3.3
L-0.84%
120 T-0.56%
Peak load (kN)
110
100
90 30 MPa
40 MPa
80 50 MPa
70
0.0001 0.001 0.01 0.1 1 10
Loading rate (m/s)
loading rates: (a) a d -2.8; (b) a d -3.3; (c) a d -4.4 (continued)
- 106 -
(c) 60
55 a/d-4.4
L-0.82%
Peak load (kN)

50 T-0.94%
45
40
30 MPa
40 MPa
35
50 MPa
30
0.0001 0.001 0.01 0.1 1 10
Loading rate (m/s)
loading rates: (a) a d -2.8; (b) a d -3.3; (c) a d -4.4
4.6 Summary
FE model was developed in this chapter to evaluate behavior of beams under
varying rates of concentrated loading at midspan. The comparisons of experimental
and analytical results indicated that this model is able to predict both the load vs.
midspan deflection characteristics and cracking profile with reasonable accuracy.
After verifying the numerical simulation results with the experimental outcomes,
parametric investigation was carried out to elucidate the influence of some key
parameters on DIF of maximum resistance and failure modes.
This was observed that DIF were on the higher side for beam with low
reinforcement ratio. When the beam contains low amount of longitudinal
reinforcements, the response was mainly influenced by the strain rate effect of
longitudinal reinforcing steel which in turn causing an enhancement of the DIF.
This was noticed that when the beam contains relatively small amount of
longitudinal reinforcements (i.e. 0.3 times of balanced reinforcement ratio), only the
formation of flexural cracks were observed for all loading rates; whereas for the
case of beam having relatively higher amount of longitudinal reinforcements (i.e.
0.6 times of balanced reinforcement ratio), both flexural-shear and some diagonal
cracks were noticed. Specifically, for high loading rates, extent of damage was
more severe for the case of beam containing relatively higher amount of
- 107 -
longitudinal reinforcements. Therefore, it can be concluded that change in failure

mode may be observed (i.e. mainly in high rates considered here) when the amount
of longitudinal reinforcement ratio was increased (i.e. approaching towards
balanced-reinforcement ratio) in an under-reinforced beam.
Beam containing higher amount of transverse reinforcements, DIF would be on the

lower side. Furthermore, for medium and high loading rates, the beam containing
lower amount of transverse reinforcements, diagonal cracks were formed with
flexure cracks; whereas flexure type of failure was observed in beam having high
amount of transverse reinforcements. Hence, the amount of transverse
reinforcements had significant influence in preventing the formation of diagonal
cracks under medium and high loading rates. For beams having various concrete
grades (30, 40 and 50 MPa), increasing trend in peak load was observed when the
loading rate progressed from static to high. Moreover, beam having high strength
The yield strength of longitudinal reinforcements of beams is the key parameter,

affecting the change in failure mode from flexure at static loading to shear at high
rates. Punching and diagonal shear cracks were observed in medium and high rates
whereas only flexure cracks were perceived in static and low rates in beams
containing longitudinal reinforcements of yield strength 420 MPa and the static
shear to bending resistance ratio less than 1.5. Due to high rate sensitivity of low
yield strength steel (i.e. 420 MPa yield strength steel was more sensitive than 520
MPa yield strength steel), the increment of yield strength of longitudinal
reinforcements in medium and high rates was such a way that the bending
resistance surpassed the shear resistance and eventually failure mode changed.
- 108 -
Chapter 5: Experimental Assessment of RC Beams Subjected to Drop-weight Impact Loading
CHAPTER 5
EXPERIMENTAL ASSESSMENT OF RC BEAMS
SUBJECTED TO DROP-WEIGHT IMPACT LOADING
5.1 Introduction
With the rapid development of the infrastructural project around the world, there is
a chance that RC structures may be experienced from various types of impact
loading during their service life due to several events. Moreover, concrete structures
have been widely used over the years by engineers in the design and construction of
protective structures to resist impact loads. Typical examples comprise in a
diversified fields ranging from transportation structures (i.e., bridge piers, guard
rails, traffic signal posts and electric poles etc.) subjected to vehicle-crash impact,
falling rocks on rock-sheds in mountainous regions, falling heavy loads on
industrial facilities due to accidents, marine and offshore structures exposed to ship
and ice impact or subjected by tornado or tsunami-borne debris impact, columns in
multi-story car park or bridge-pier strike by moving vehicle, protective structures
subjected to projectiles or aircraft impact and structures sustaining shock and
impact loads during explosions or earthquakes. Fig. 5.1 depicts some of the impact
loading scenarios on structures. Impact loading is an extremely severe loading
conditions characterized by its application of a force of great intensity within a
short duration. Impact loading can be categorized into two basic types: single point
impact loading and distributed impact loading. When a structures strike by an
impactor or striker at a particular point that is called single point impact; whereas
explosions or blasts would bring about distributed impact loading. Banthia [B1]
summarized three distinct situations in a typical impact event by considering the
size and mass of the impacting body: a very large object struck by a small
impacting mass, an impact involving comparable masses, a small object struck by a
large impactor. Although, the third case is comparatively rare, the first and second
cases are often encountered. Furthermore, in terms of mass and velocity of the
impactor, impact can be classified as low and high velocity impact. Low velocity
impact can be referred to the impact load caused by a larger impactor with low
- 109 -
velocity whereas low-mass impactor at high velocity can cause a high velocity
impact. Gas gun can be used to generate high velocity impact which will cause
local failure (i.e. perforation and penetration etc.) of the structures. To simulate the
low velocity impact scenario, drop-weight impact test set up would be ideal, which
will cause the overall failure (i.e. shear and flexure failure with some local failure at
or nearby impacting point) of the structures.
(a) (b)
(c) (d)
Fig. 5.1: Impact loading scenarios on structures: (a) Yoshida et al. [Y4]; (b) Yuan
and Harik [Y5]; (c) Dahlberg [D1]; (d) Sugano et al. [S9]
The behavior of RC beam under low velocity impact may consist of two response
phases as shown in Fig. 5.2: the local response due to the stress wave that occurs at
the impacting point during a very short period after impact; and the overall response
include the free vibration of the whole structural member (Fujikake et al. [F4]).
Load rate effects and dynamic behavior of the structural components primarily
- 110 -
control the overall response. It is recognized that the structural members behave
differently under impact loading as compared static loading. Furthermore, the effect
of impact loading can result in catastrophic and sudden structural failure. So the
understanding of the performance and vulnerability of RC structures under drop-
weight impact loading has become an emerging topic in recent years. Although,
numerous researches (Hughes and Beeby [H3]; Banthia [B1]; Kishi et al. [K5-K8];
Chen and May [C3]; Fujikake et al. [F4]; Bhatti et al. [B8]; Saatci and Vecchio
[S1]; Tachibana et al. [T1]; Kishi and Mikami [K10]) have been conducted on
evaluating the impact response of RC beams; however till now little effort has been
made towards determining the impact responses of beams by considering various
parameters at a time. Therefore, this chapter is devoted on the assessment of the
structural response of beam induced by low velocity impact (<10 m/s) at its
midspan. The impact loading is produced by dropping a steel impactor on the
beams from specific heights.
Fig. 5.2: Impact responses of a RC member (Fujikake et al. [F4])
5.2 Objective of Experimental Program
Due to the wide prevalence of unintentional, man-made accidents and natural
- 111 -
calamities, demand for impact-resistant structures has increased tremendously.

However, there is a lack of systematic research on structural behavior of beams
under drop-weight impact loading. Considering the effect of various parameters at a
time, drop-weight impact test was performed for various drop heights on beams by
varying span length, cross-sections, shear span to effective depth ratios,
longitudinal and transverse reinforcement ratios. The objective is to evaluate the
impact responses (e.g. maximum impact load and maximum midspan deflections)
and the failure modes of beams. Moreover, this experimental program will provide
experimental results that can contribute further in the development and validation
of analytical and numerical models. In addition, these impact-damaged beams will
be used further to evaluate their post impact residual performances (e.g. residual
resistance and stiffness) which are summarized in Chapter 8.
5.3 Outline of Experiment
The experimental program carried out in two phases: drop-weight impact tests and
static tests. Impact program consisted of 30 RC beams which were divided into two
groups in terms of their shear span-to-effective depth ratios ( a d -3.3, 3.8 and 5.7).
Each group had three types of specimens distinguished by their shear reinforcement
ratios. Each type of 3.3 series had four identical specimens whereas 3.8 and 5.7
series had three identical specimens. Six additional specimens identical to 3.8 and
5.7 series were cast to determine their experimental static behavior. However, the
static behavior of 3.3 series specimens was determined from FE analysis due to the
unavailability of experimental data. Table 5.1 presents the overview of the entire
test program including the considered drop-heights. Mass ratios (   mb mi ) i.e.
ratios of the mass of the beams ( mb ) to the impactor-mass ( mi ) for a d -3.3, 3.8
and 5.7 specimens were 0.42, 0.49 and 0.26 respectively. The specimen denoted by
SR5.7_1.6 corresponds to „singly-reinforced‟ beam (SR) followed by the shear
span-to effective depth ratio and percentage of longitudinal tensile reinforcement
ratios. Again, DR5.7_1.6_0.20 refers to „doubly-reinforced‟ beam (DR) followed
by the shear span-to effective depth ratio, percentage of longitudinal tensile and
compressive reinforcement ratio and percentage of shear reinforcement ratio. It is
- 112 -
noted that all test specimens have been designed and cast at NTU, Singapore
however; the drop-weight impact testing was conducted at NDA, Japan. Before
commencing the impact loading tests, quasi-static tests were performed in a 3-point
bending test set up by hydraulic machine, as shown in Fig. 5.3. For impact loading,
a drop-weight impact loading machine was used, as shown in Fig. 5.4.
Table 5.1: Test program
Beam Static Impact velocity

Drop height (m) Impact energy (kJ)
designation test (m/s)
DR3.3_2.4 0.15, 0.30, 0.60, 1.72, 2.43, 3.43, 0.44, 0.88, 1.77,
1.20 4.85 3.53
DR3.3_2.4_0.12 * 0.30, 0.60, 0.90, 2.43, 3.43, 4.20, 0.88, 1.77, 2.65,
1.20 4.85 3.53
DR3.3_2.4_0.56 0.60, 0.90, 1.2, 3.43, 4.20, 4.85, 1.77, 2.65, 3.53,
1.60 5.60 4.71
SR3.8_0.8 0.30, 0.60, 0.90 2.43, 3.43, 4.20 0.88, 1.77, 2.65
DR3.8_0.8_0.11 √ 0.60, 0.90, 1.20 3.43, 4.20, 4.85 1.77, 2.65, 3.53
DR3.8_0.8_0.15 0.60, 0.90, 1.20 3.43, 4.20, 4.85 1.77, 2.65, 3.53
SR5.7_1.6 0.30, 0.45, 0.60 2.43, 2.97, 3.43 0.88, 1.32, 1.77
DR5.7_1.6_0.15 √ 0.30, 0.45, 0.60 2.43, 2.97, 3.43 0.88, 1.32, 1.77
DR5.7_1.6_0.20 0.30, 0.45, 0.60 2.43, 2.97, 3.43 0.88, 1.32, 1.77
Note: √= Tested; *= FEM determined
Fig. 5.3: Static loading test set up
A drop-weight with a mass of 300 kg was dropped freely onto the top surface of the
- 113 -
beam at midspan from different heights. For 3.3 series four and for 3.8 and 5.7
series specimens three different drop heights were considered. The striking head of
impactor had a hemispherical tip with a radius of 90 mm. Specially designed
support devices were mounted over the clear span of the beam, allowing it to rotate
freely while preventing the uplift of beam during impact loading.
Fig. 5.4: Drop-weight impact loading test set up
All beams had rectangular cross section 120 to 160 mm in width, 170 to 250 mm in
depth. Beams of 3.3 series were supported over the clear span of 1400 mm whereas
specimens from 3.8 and 5.7 series were supported over the clear span of 1600 mm.
The longitudinal tensile reinforcement ratios are in the range of 0.8 to 2.4%. For
DR beam section equal amount of reinforcements were provided in compression
side. When the longitudinal tensile reinforcement ratios were calculated in terms of
balanced reinforcement ratios, these values would be around 0.3 pb , 0.6 pb and
0.9 pb for 3.3, 3.8 and 5.7 series respectively; where pb refers to balanced
reinforcement ratio in percentage. The layout of the longitudinal reinforcements,
spacing of shear reinforcements and the measuring points (accelerometers for
measuring accelerations and strain gauges for measuring strain of steel reinforcing
bars) are shown in Fig. 5.5. Here, „SA‟ refers to the strain gauges attached at the
midspan of longitudinal tensile reinforcements whereas „SS‟ refers to the strain
gauges attached to transverse reinforcements and A1, A2, A5 and A6 refers to the
- 114 -
accelerometers.
Unit: mm Strain gauge for rebars

Accelerometer
2@40=80
A1 A2 A5 A6 2T22
SA-1,2 170 210 250
250 200 200 250 70 2T22

150
1400
1700
DR3.3_2.4
150 R6 2T22
SS-3,4 SA-1,2 SS-1,2 170 210 250
1400 70 2T22
4@350=1400 150
1700
DR3.3_2.4_0.12
80 R6 2T22
170 210 250

SS-3,4 SA-1,2 SS-1,2 70 2T22
1400 150
22@70=1540
1700
DR3.3_2.4_0.56
2@50=100
A1 A2 A5 A6
SA-1,2 210 240
250 250 250 250 100

2T13
1600 160
2000
SR3.8_0.8

gauges and accelerometers (continued)
- 115 -
40 R6 2T13
SS-3,4 SA-1,2 SS-1,2 180 210 240
1600 100 2T13

6@320=1920
160
2000
DR3.8_0.8_0.11
80
R6 2T13
SS-4 SS-3 SA-1,2 SS-1 SS-2 180 210 240
1600 100 2T13

8@230=1840
160
2000
DR3.8_0.8_0.15
2@50=100
A1 A2 A5 A6
SA-1,2 140 170

250 250 250 250 60
2T13
1600 120
2000
SR5.7_1.6
40 R6 2T13
SS-3,4 SS-1,2
SA-1,2 110 140 170
1600 60 2T13
6@320=1920
120
2000
DR5.7_1.6_0.15
80 R6 2T13
SS-4 SS-3 SS-1 SS-2
SA-1,2 110 140 170
1600 60 2T13
8@230=1840
120
2000
DR5.7_1.6_0.20
gauges and accelerometers
- 116 -
Longitudinal reinforcements consisted of deformed steel bars of T22 (diameter-

13mm), T13 (diameter-13mm) and T13 (diameter-13mm) for a d -3.3, 3.8 and 5.7
series specimen respectively while R6 (diameter-6mm) plain round bars were used
as transverse reinforcements for all specimens. Target concrete compressive
strength at 28 days was 40 MPa with maximum aggregate size of 10 mm for all
specimens. The yield strengths of T22, T13 and R6 were 520, 520 and 310 MPa
respectively. All specimens were cast at the same time using ready mix concrete of
same batch; thence, all specimens had an identical concrete strength. The
compressive strength at 28 days obtained from three concrete cylinders was 39.0,
37.6 and 38.8 MPa respectively. The specimen details, theoretical static capacities,
static expected and observed failure modes are listed in Table 5.2, where the static
flexural and shear resistances are calculated according to ACI 318-08 [A2].
Table 5.2: Details of specimens and static design parameters
Shear span Shear to

Longitudinal Transverse Bending Shear
to bending
Designation reinforcement reinforcement resistance resistance
effective resistance
ratio (%) ratio (%) (kN) (kN)
depth ratio ratio
DR3.3_2.4 0.00 66.20 0.33
DR3.3_2.4_0.12 3.3 2.4 0.12 203.70 87.10 0.43
DR3.3_2.4_0.56 0.56 170.40 0.84
SR3.8_0.8 0.00 70.45 67.83 1.04
DR3.8_0.8_0.11 3.8 0.8 0.11 93.32 1.38

67.83
DR3.8_0.8_0.15 0.15 102.30 1.51
SR5.7_1.6 0.00 35.32 0.83
DR5.7_1.6_0.15 5.7 1.6 0.15 42.40 50.52 1.19
DR5.7_1.6_0.20 0.20 56.40 1.33
To construct the specimens, steel formwork was used. Before the placement of
reinforcing cages, oiling was carried out on the inside surfaces of formwork so that
specimens can be removed easily and the surface of the specimens would not
damage. Before the casting, concrete spacer blocks were placed in the underside as
well as the side faces of transverse reinforcements to achieve clear concrete cover.
Lifting hooks were mounted on both sides of the specimens to facilitate lifting
process during testing. Fig. 5.6 shows the photos of the casting of beams and the
- 117 -
preparation of concrete cylinders (150mm x 300 mm). After casting, damp hessian
fabrics were employed to cure the specimens for two weeks.
(a) (b)
Fig. 5.6: (a) Casting of beams; (b) preparation of concrete cylinders
All specimens were well instrumented to capture the loads, displacements,

accelerations and the strains of steel reinforcements. The instrumentation for this
test program included a load cell, strain gauges and accelerometers. Strain gauges
with 3 mm gauge length were installed in the mid-span of the longitudinal tensile
reinforcements and in the mid-point of the two legs of the shear reinforcements to
measure the local strain. These are general purpose foil strain gauges (KFG types)
with resistance of 120  and linear expansion coefficient of 11x10-6/ 0 C . Soldering
to the terminal of the gauges was not required due to the pre-attachment of three
parallel vinyl-coated lead wires having length of 10 m. Before the installation of
strain gauges at the specific locations, the reinforcement surfaces were grinded
lightly and cleaned with alkaline and acidic chemicals. Fig. 5.7(a) and (b) shows
the photos of the glued strain gauges in longitudinal tensile and transverse
reinforcements. Although in Fig. 5.7, the strain gauge is shown in top face of the
tensile reinforcements however, the reinforcement cages were inverted during the
placement into the moulds; then the gauges would be in the bottom face of the
tensile reinforcements. The impact load developed between the impactor and beams
was measured by a dynamic load cell (frequency response 5 kHz), which was
rigidly attached to the drop-weight. Four accelerometers (capacity of 1000 times
gravity and resonance frequency greater than 70 kHz) were mounted on the
- 118 -
specimens for each test (except for static loading) to measure the accelerations. The
midspan deflection of RC beams was measured by laser-type variable displacement
transducers (LVDTs) which have a measuring range of 80 mm and sampling rate of
50 kHz. Data from the sensors were collected by a computer-based data acquisition
system which has a sampling rate of 100 kHz. For getting insights about the
cracking patterns and failure modes of beams, digital photography and high-speed
video recording systems were employed.
Strain gauges
(a)
Strain gauges
(b)
gauging in transverse reinforcements
5.4 Experimental Results and Discussions
Following the outline of experimental sections which consist of specimen design,
- 119 -
material properties, test-setup and instrumentation; experimental results and major

observations are comprehensively presented and discussed in this section.
5.4.1 Static test results
Load vs. midspan deflection responses obtained from static tests are presented in
Fig. 5.8 and simultaneously failure patterns are depicted in Fig 5.9. SR beam of 3.8
series failed in shear-critical mode with the formation of diagonal shear crack.
Sharp fall of the post-peak branch in the load vs. midspan deflection curves of
SR3.8_0.8 is clearly seen whereas, DR beams from same series exhibited ductile
failure response. For specimens of 5.7 series, ductile failures were observed. From
the picture of the crack pattern of SR5.7_1.6 diagonal crack was perceived,
however this crack was formed at the ending time of the testing process or at a
large deflection.
(a)
100
80
Load (kN)
60
40
SR3.8_0.8
20 DR3.8_0.8_0.11
DR3.8_0.8_0.15
0
0 5 10 15 20 25 30 35 40
(b)
60
50
40
Load (kN)
30
20
SR5.7_1.6
DR5.7_1.6_0.15
10
DR5.7_1.6_0.20
0
0 10 20 30 40 50 60
Fig. 5.8: Static load vs. midspan responses of beams (a) 3.8 series; (b) 5.7 series
- 120 -
SR3.8_0.8
DR3.8_0.8_0.11
DR3.8_0.8_0.15
(a)
SR5.7_1.6
DR5.7_1.6_0.15
DR5.7_1.6_0.20
(b)
Fig. 5.9: Failure pattern of beams under static loading (a) 3.8 series; (b) 5.7 series
5.4.2 Impact test results
This section summarizes the test results obtained from the drop-weight impact tests.
Time histories of impact load, midspan deflection, impact response characteristics
(i.e. maximum impact load, maximum midspan deflection and time to reach
maximum midspan deflection etc.) and crack patterns for each specimen at each
drop height are summarized.
- 121 -
5.4.2.1 Time histories of impact load
Measuring the impact load in a drop-weight impact experiment poses some

challenges. Here, the impact load history was measured using a load-cell attached
to the impactor. Fig. 5.10 shows the time histories of impact load with time
duration of 10 ms from the beginning of impact. The impact load history comprises
two initial successive peak parts within short time interval followed by steady part.
The first peak was due to the initial shock of impact which was generated by the
inertia of beam. Few cases the second peak surpassed the first one; this one could
be due to the presence and the arrangement of load-cell which may itself affect the
contact and the measurements (Bischoff [B8]). Furthermore, due to the direct
contact in between the impactor and beam, local failure (i.e. cracking and spalling
of concrete at or around the impact zone) played a significant role in controlling the
characteristics of impact load history. Here, maximum value corresponding to first
peak has taken as peak impact load.
(a) 600 600

500 DR3.3_2.4 0.15 m 500 DR3.3_2.4_0.12 0.3 m
0.3 m 0.6 m
400 400
Impact load (kN)
Impact load (kN)
0.6 m 0.9 m
300 1.2 m 300 1.2 m
200 200
100 100
0 0
-100 -100
-200 -200
0 0.002 0.004 0.006 0.008 0.01 0 0.002 0.004 0.006 0.008 0.01
Time (s) Time (s)
800
DR3.3_2.4_0.56 0.6 m
600 0.9 m
Impact load (kN)
1.2 m
400
200
-200
0 0.002 0.004 0.006 0.008 0.01
Time (s)
Fig. 5.10: Time histories of impact load (a) 3.3 series; (b) 3.8 series; (c) 5.7 series
(continued)
- 122 -
(b) 500 600
400 SR3.8_0.8 0.3m DR3.8_0.8_0.11 0.6m

0.6m 400 0.9m
0.9m
Impact load (kN)

1.2m
Impact load (kN)
300
200 200
100 0
0
-200
-100
-200 -400
0 0.002 0.004 0.006 0.008 0.01 0 0.002 0.004 0.006 0.008 0.01
Time (s) Time (s)
600
DR3.8_0.8_0.15 0.6m
400 0.9m
Impact load (kN)
1.2m
200
-200
-400
0 0.002 0.004 0.006 0.008 0.01
Time (s)
(c) 300 250

250 SR5.7_1.6 0.3m DR5.7_1.6_0.15 0.3m
0.45m 200
0.45m
200 0.6m
Impact load (kN)
Impact load (kN)
150 0.6m
150
100 100
50
50
0
0
-50
-100 -50
0 0.002 0.004 0.006 0.008 0.01 0 0.002 0.004 0.006 0.008 0.01
Time (s) Time (s)
300
250 DR5.7_1.6_0.20 0.3m
0.45m
200
Impact load (kN)
0.6m
150
100
50
0
-50
-100
0 0.002 0.004 0.006 0.008 0.01
Time (s)
Fig. 5.10: Time histories of impact load (a) 3.3 series; (b) 3.8 series; (c) 5.7 series
- 123 -
5.4.2.2 Time histories of midspan deflection
Fig. 5.11 shows the time histories of midspan deflection of beam. Almost similar
type of shapes can be seen from these figures. The initial response exhibited a half-
sine wave followed by steady response. Maximum and residual midspan deflection
of the beam in the same series increased with the increment of the drop heights.
(a) 60 40
50 DR3.3_2.4 0.15 m 0.3 m
DR3.3_2.4_0.12

0.3 m 30 0.6 m
40
0.6 m 0.9 m
30 1.2 m 1.2 m
20
20
10 10
0
0
-10
-20 -10
0 0.05 0.1 0.15 0 0.05 0.1 0.15
Time (s) Time (s)
40
DR3.3_2.4_0.56
0.6 m
30 0.9 m
1.2 m
20 1.6 m
10
-10
0 0.05 0.1 0.15
Time (s)
(b) 50 70
SR3.8_0.8 0.3 m
DR3.8_0.8_0.11
40 60 0.6 m
0.6 m 0.9 m
0.9 m 50 1.2 m
30
40
20
30
10
20
0 10
-10 0
0 0.05 0.1 0.15 0 0.05 0.1 0.15
Time (s) Time (s)
Fig. 5.11: Time histories of midspan deflection (a) 3.3 series; (b) 3.8 series; (c) 5.7
series (continued)
- 124 -
60
DR3.8_0.8_0.15 0.6 m

50
0.9 m
1.2 m
40
30
20
10
0
0 0.05 0.1 0.15
Time (s)
(c) 60 60
SR5.7_1.6

50 0.3 m 50 DR5.7_1.6_0.15 0.3 m

0.45 m 0.45 m
40 0.6 m 0.6 m
40
30
30
20
20
10
0 10
-10 0
0 0.05 0.1 0.15 0 0.05 0.1 0.15
Time (s) Time (s)
60
DR5.7_1.6_0.20 0.3 m
50
0.45 m
0.6 m
40
30
20
10
0
0 0.05 0.1 0.15
Time (s)
Fig. 5.11: Time histories of midspan deflection (a) 3.3 series; (b) 3.8 series; (c) 5.7
series
5.4.2.3 Impact response characteristics
Fig. 5.12 shows variation of maximum impact load under various drop heights of
three series specimens. Similarly, the variation of maximum midspan deflection
and time to reach maximum deflection at each drop heights are presented in Fig.
5.13 and 5.14 respectively. With the enhancement of the drop heights, incremental
- 125 -
response was observed for each beam.
(a) 500
(b) 700
Maximum impact load (kN)

450 DR3.3_2.4 SR3.8_0.8
600
DR3.3_2.4_0.12 DR3.8_0.8_0.11
400
DR3.3_2.4_0.56 DR3.8_0.8_0.15
500
350
300 400
250
300
200
200
150
100 100
0 0.5 1 1.5 2 0.2 0.4 0.6 0.8 1 1.2 1.4
Drop height (m) Drop height (m)
(c) 220
200 SR5.7_1.6
DR5.7_1.6_0.15
180 DR5.7_1.6_0.20
160
140
120
100
0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65
Drop height (m)
Fig. 5.12: Maximum impact load under various drop heights of (a) 3.3 series; (b)
3.8 series; (c) 5.7 series specimens
(a) 50
(b) 70
Max. midspan deflection (mm)
DR3.3_2.4 60 SR3.8_0.8
40 DR3.3_2.4_0.12 DR3.8_0.8_0.11
DR3.3_2.4_0.56 50 DR3.8_0.8_0.15
30 40
20 30
20
10
10
0 0
0 0.5 1 1.5 2 0.2 0.4 0.6 0.8 1 1.2 1.4
series; (b) 3.8 series; (c) 5.7 series specimens (continued)
- 126 -
(c) 60

SR5.7_1.6
50 DR5.7_1.6_0.15
DR5.7_1.6_0.20
40
30
20
10
0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65
Drop height (m)
series; (b) 3.8 series; (c) 5.7 series specimens
(a) 0.03 (b) 0.03

max. midspan deflection (s)
0.025 DR3.3_2.4 SR3.8_0.8

DR3.3_2.4_0.12 DR3.8_0.8_0.11
0.025
DR3.3_2.4_0.56
Time to reach
DR3.8_0.8_0.15
Time to reach
0.02
0.015 0.02
0.01
0.015
0.005
0 0.01
0 0.5 1 1.5 2 0.2 0.4 0.6 0.8 1 1.2 1.4
(c) 0.03
0.028 SR5.7_1.6
DR5.7_1.6_0.15
0.026 DR5.7_1.6_0.20
Time to reach
0.024
0.022
0.02
0.018
0.016
0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65
Drop height (m)
Fig. 5.14: Time to reach maximum midspan deflection under various drop heights
of (a) 3.3 series; (b) 3.8 series; (c) 5.7 series specimens
- 127 -
5.4.2.4 Crack patterns
Besides, data collected by the sensors, the cracks developed after each test was
marked and the crack widths were measured. The widths of the diagonal crack were
measured at mid-height whereas for the case of flexural cracks, measurement was
taken at the bottom surface of beam at midspan. Fig. 5.15 shows the sketched
cracks profiles of beams at each drop height. This was observed that the crack
patterns and the failure modes varied between the specimens depending on their
static bending and shears resistance. Detailed description of cracks profiles of beam
at each drop height is summarized below:
(1-a) DR3.3_2.4 (Drop height - 0.15 m)
No damage was observed in this beam at a drop height of 0.15 m.
DR3.3_2.4 (Drop height - 0.30 m)
At 0.30 m drop height, a 0.75 mm diagonal crack was observed in the right hand
side whereas another hairline diagonal crack appeared in other side. Local failure
did not happen at or around the impact region.
With the increment of the drop height to 0.60 m, two diagonal cracks formed in the
right hand side having width of 0.65 and 1.2 mm, respectively. In the other side,
hairline diagonal crack was perceived. Massive crushing and spalling of concrete
occurred at the impact region.
At 1.2 m drop height, the specimen suffered extensive damage with significant
spalling of concrete at top, bottom and side surface of the beam in one side,
exposing the longitudinal tensile and compressive bars. In the other side, spalling of
concrete occurred at top and side surface of the specimen. The damage was so
catastrophe that the beam was disintegrated into two components.
- 128 -
(1-b) DR3.3_2.4_0.12 (Drop height - 0.30 m)
Two hairline diagonal cracks were observed in this beam at a drop height of 0.30
m.
DR3.3_2.4_0.12 (Drop height - 0.60 m)
At 0.60 m drop height, the diagonal cracks in one side widened, reaching up to 1
mm. Moreover, diagonal cracks forming a shear-plug (crack width 1.3 mm in one
side) under the impact point were observed.
DR3.3_2.4_0.12 (Drop height - 0.90 m)
With the further increment of the drop height to 0.90 m, the diagonal cracks in one
side widened significantly, reaching up to 6.5 mm and in the other side, diagonal
crack formed with 1.2 mm width. Like previous case, shear-plug developed at the
impact point. Furthermore, local failure in terms of crushing and spalling of
concrete at the impact region occurred.
DR3.3_2.4_0.12 (Drop height - 1.2 m)
At 1.2 m drop height, diagonal cracks in both sides broadened extensively, attaining
up to 12.5 and 4.5 mm, respectively. The local damage was quite extensive with
crushing and spalling of concrete. Moreover, few horizontal cracks propagated in
the longitudinal direction of beam from the impact point.
(1-c) DR3.3_2.4_0.56 (Drop height - 0.60 m)
Hairline flexure-shear cracks were observed in this beam at a drop height of 0.60 m
without any local damage.
DR3.3_2.4_0.56 (Drop height - 0.90 m)
At 0.90 m drop height, the number of flexure-shear cracks increased and flexural
cracks at midspan developed. Still, at this drop height, no damage in the impact
region occurred.
DR3.3_2.4_0.56 (Drop height - 1.2 m)
With the further increment of the drop height to 1.2 m, the flexure-shear cracks
- 129 -
became much flatter as compared to previous two cases. Moreover, some local
damage has been occurred in the impact region. Some splitting cracks were
identified in the side cover of the beam.
DR3.3_2.4_0.56 (Drop height - 1.6 m)
At 1.6 m drop height, the severity of damage increased significantly mainly at or

around impact region, with crushing and spalling of concrete and number of
flexure-shear cracks also increased.
(2-a) SR3.8_0.8 (Drop height - 0.30 m)
Typical flexure failure was observed in this beam for a drop height of 0.30 m.
Flexural cracks developed from the bottom surface and propagated towards top
surface of the beam. Flexural cracks (crack width 0.5 mm) underneath the impact
point formed at the bottom surface and propagated towards top surface and finally
crushed the compression concrete.
SR3.8_0.8 (Drop height - 0.60 m)
Like previous case, almost similar type of crack pattern has been perceived at a
drop height of 0.60 m. However, flexural cracks underneath the impact point
widened further, reaching up to 1.4 mm.
With the further increment of the drop height to 0.90 m, no too much difference in
the crack pattern has been observed. However, flexural crack underneath the impact
point became wider, reaching up to 1.7 mm. Moreover, local failure occurred at the
impact region which combined with the flexural crack underneath the impact point
disintegrated beam into two parts.
(2-b) DR3.8_0.8_0.11 (Drop height - 0.60 m)
Flexural cracks (width 5 mm) underneath the impact point developed from the
bottom surface and propagated towards top surface. Moreover, some hairline
flexure crack was perceived near surrounded region of midspan.
- 130 -
DR3.8_0.8_0.11 (Drop height - 0.90 m)
At 0.9 m drop height, flexural cracks underneath the impact point became widened,
attaining up to 9 mm width and successively some crushing at the impact region
was noticed.
DR3.8_0.8_0.11 (Drop height – 1.2 m)
With the further increment of the drop height to 1.2 m, both flexure crack (width
around 15 mm) and shear plug (crack width around 11 mm at bottom side of beam)
were observed. Moreover, massive local damage occurred at or around the impact
region.
(2-c) DR3.8_0.8_0.15 (Drop height - 0.60 m)
Flexural cracks (width 1.1 mm) underneath the impact point developed from the
bottom surface and propagated towards top surface. Moreover, few hairline cracks
were perceived in the surrounding region.
DR3.8_0.8_0.15 (Drop height - 0.90 m)
Flexural cracks underneath the impact point became widened at 0.9 m drop height.
The crack almost attained 9 mm in width at bottom surface of beam. At the same
time some cracking and crushing of concrete at impact region were noticed.
DR3.8_0.8_0.15 (Drop height - 1.2 m)
With the further increment of the drop height to 1.2 m, flexural cracks underneath
the impact point became widened significantly, reaching up to 30 mm in width.
Moreover, there was significant damage occurred at the impact region.
(3-a) SR5.7_1.6 (Drop height - 0.30 m)
Flexural cracks (crack width 1.4 mm) underneath the impact point formed at the
bottom surface and propagated towards top surface and connected to the crushed
and spalled concrete at the impact point.
- 131 -
Like previous case, almost similar type of crack pattern has been perceived at a
drop height of 0.45 m. However, flexural crack underneath the impact point
widened more; reaching up to 6 mm. Crushing and spalling of concrete at the
impact region were also noticed. These crushed and spalled concrete combined
with the flexure crack crumbled the beam into two parts.
With the further increment of the drop height to 0.60 m, more flexural cracks
developed and the flexure cracks beneath the impact point became more broaden,
reaching up to 7 mm. This crack propagated vertically upwards and disintegrated
the beam into two parts.
(3-b) DR5.7_1.6_0.15 (Drop height - 0.30 m)
bottom surface and propagated towards top surface.
DR5.7_1.6_0.15 (Drop height - 0.45 m)
Flexure cracks (width 5 mm) at the midspan and crushing of concrete at the impact
region were perceived at drop height of 0.45 m.
DR5.7_1.6_0.15 (Drop height - 0.60 m)
With the increment of the drop height to 0.60 m, both flexure crack (width 4 mm)
and shear-plug were observed. Significant crushing of concrete at the impact region
has also been occurred.
(3-c) DR5.7_1.6_0.20 (Drop height - 0.30 m)
bottom surface and propagated towards top surface.
- 132 -
0.15m 0.30m 0.60m
0.90m
0.30m 0.60m
0.90m 1.2m
0.60m
1.6m
1.2m 1.2m
DR3.3_2.4 DR3.3_2.4_0.12 DR3.3_2.4_0.56
(a)
0.60m 0.60m
0.30m
0.90m
0.60m 0.90m
0.90m 1.2m 1.2m
SR3.8_0.8 DR3.8_0.8_0.11 DR3.8_0.8_0.15
(b)
0.30m 0.30m 0.30m
0.45m 0.45m 0.45m
0.60m 0.60m 0.60m
SR5.7_1.6 DR5.7_1.6_0.15 DR5.7_1.6_0.2
(c)
Fig. 5.15: Crack pattern of beams (a) 3.3 series; (b) 3.8 series; (c) 5.7 series
- 133 -
DR5.7_1.6_0.20 (Drop height - 0.45 m)
Flexural cracks underneath the impact point widened more as compared to previous
case, reaching up to 6 mm and some hairline flexure cracks were observed in the
surrounded region. Moreover, some crushing and spalling of concrete at impact
region were noticed.
DR5.7_1.6_0.20 (Drop height - 0.60 m)
Extensive local damage (chunk of concrete spalled off from the both side of the
impact point) has been occurred at a drop height of 0.60 m. Two major flexure
cracks underneath the impact point were seen with width of 4.5 and 0.85 mm,
respectively.
5.4.2.5 Strain rates
Effect of strain rate on material behavior (concrete and reinforcing steel) has been
well documented in the literature (CEB model code [C5], Malvar and Ross [M5],
Malvar [M2]). Constitutive material properties for structural materials are
dependent of loading rates. Impact loads introduce considerably higher strain rate
as compared to static loading, resulting in different behavior in concrete and steel.
Hence, it is worthwhile to determine the maximum strain rate and their variation in
the beam for various drop heights considered in this test program. This is
mentioned earlier that strain gauges were mounted in all longitudinal tensile
reinforcements and some transverse reinforcements; however, no initiative was
taken to measure the strain in concrete. Strain rates are calculated by differentiating
the strain-history data provided by the strain gauges. One specimen for a particular
drop height was taken from each series (i.e. 3.3, 3.8 and 5.7) to present their strain
rate-history data for longitudinal tensile reinforcements and transverse
reinforcements. Here, „SA‟ refers to the strain rates in longitudinal tensile
reinforcements whereas „SS‟ refers to the strain rates in transverse reinforcements.
Maximum strain rates in longitudinal tensile reinforcements were found to be
around 2.5, 6 and 5.9 s-1 for DR3.3_2.4_0.56, DR3.8_0.8_0.15 and
DR5.7_1.6_0.15, respectively for the corresponding drop heights mentioned in Fig.
5.16. Similarly, maximum strain rates in transverse reinforcements were found to
- 134 -
be around 3.4, 3 and 1.75 s-1 for DR3.3_2.4_0.56, DR3.8_0.8_0.15 and

DR5.7_1.6_0.15, respectively for the corresponding drop heights mentioned in Fig.
5.17.
(a) 4
3.5 DR3.3_2.4_0.56
SA-1
3 SA-2 Drop height - 1.2 m
Strain rate (s )
-1
2.5
2
1.5
1
0.5
0
0 5E-4 1E-3 1.5E-3 2E-3 2.5E-3 3E-3 3.5E-3 4E-3
Time (s)
(b) 10
SA-1 DR3.8_0.8_0.15
8 SA-2 Drop height - 0.6 m
Strain rate (s )
-1
0
0 5E-4 1E-3 1.5E-3 2E-3
Time (s)
(c) 10
DR5.7_1.6_0.15
8
Drop height - 0.6 m
Strain rate (s )
-1
6 SA-1
SA-2
0
0 5E-4 1E-3 1.5E-3 2E-3 2.5E-3
Time (s)
Fig. 5.16: Variation of strain rates in longitudinal tensile reinforcements
- 135 -
(a) 5
DR3.3_2.4_0.56
4
Drop height - 1.2 m
Strain rate (s )
-1
3
SS-1
SS-3
2 SS-4
0
0 5E-3 1E-2 1.5E-2
Time (s)
(b) 6
5 SS-1 DR3.8_0.8_0.15
SS-2 Drop height - 0.6 m
Strain rate (s )
-1
0
0 5E-3 1E-2 1.5E-2 2E-2 2.5E-2 3E-2
Time (s)
(c) 3.5
3 SS-1 DR5.7_1.6_0.15
SS-2 Drop height - 0.6 m
2.5
Strain rate (s )
-1
1.5
0.5
0
0 5E-3 1E-2 1.5E-2 2E-2 2.5E-2 3E-2
Time (s)
Fig. 5.17: Variation of strain rates in transverse reinforcements
- 136 -
Furthermore, Fig. 5.18 presents the range of maximum strain rates of longitudinal
tensile reinforcements for beams under various drop-heights. This will help the
researchers to gain some idea regarding the development of maximum strain rates
and correspondingly strain-rate dependent material properties can be applied in the
analytical and numerical model. In general, maximum strain rates of longitudinal
tensile reinforcements varied in between 1 to 7 s-1 for the corresponding considered
drop heights of 0.15 to 1.2 m (i.e. impact velocity varies from 1.72 to 4.85 m/s).
9
8 3.3 series
-1
3.8 series
7 5.7 series
6
5
4
3
2
1
0 0.2 0.4 0.6 0.8 1 1.2 1.4
Drop height (m)
Fig. 5.18: Range of maximum strain rates of longitudinal tensile reinforcements for
beams of various series under various drop heights
5.5 Analysis of Experimental Database

This is mentioned in the literature review chapter that an equation was proposed by
Tachibana et al. [T1] to estimate the maximum midspan deflection of beam based
on impact energy and static flexural resistance. Similarly, empirical design
formulae following the performance based design concept were suggested by Kishi
and Mikami [K10], which were the equations involving the static flexural
resistance, maximum and /or residual deflection and input impact energy. These
equations can be used as design formulae following the performance-based design
concept of RC beams. As limited database was employed by the previous
researchers, thus, efforts have been devoted to find the relationship among static
resistance (both flexure and shear), maximum midspan deflection and input impact
- 137 -
energy by using a databank documented by several researchers from the literature.

A database of RC beams (including the data from the current test program too)
subjected to drop weight impact loading was assembled from the literature. In total,
174 beams under drop weight impact loading at their mid span have been collected.
The beam details, material properties and static shear to bending resistance ratio
were tabulated in Appendix-B. Out of 174 specimens, 53 of them are statically
shear failure type and remaining 121 are statically flexural failure type beams. All
impact responses were sorted out based on their first impact. Furthermore, mass of
the impactor, impact velocity and impact responses are succinctly presented in
Appendix-C. Geometry of the test specimens, amount of their longitudinal and
shear reinforcement, compressive strength of concrete and yield strength of
longitudinal reinforcement spans a wide range. All beams have rectangular cross
section 100 to 300 mm wide, 150 to 560 mm height and 1 to 4 m clear span length.
Longitudinal reinforcement ratio varies from 0.84 to 2.75%, whereas the shear
reinforcement ratio ranges from 0.11% to 0.75%. Concrete strength of the
specimens mostly falls within 24 to 50 MPa. Diameter of longitudinal
reinforcement among the test specimens in the database varies from 13 to 35 mm
and diameter of transverse reinforcements for 90% of beam is 6 mm. The yield
strength of longitudinal reinforcement varies from 345 MPa to 520 MPa. Moreover,
various types, shape of the impactors and the interface between the beam and
impactors are used by various researchers, as shown in Table 5.3. Fig. 5.19 shows
the relationship between maximum midspan deflections  max versus input impact
energy over static flexural resistance E Rb . About 90 out of 121 specimens of

experimental data from statically flexural failure type beam are plotted in Fig. 5.19
as there is insufficient experimental data given by previous researchers for the
remaining 31 specimens. The static bending resistance of the considered beams
varies from 11.4 to 237.5 kN and the input impact energy varies from 0.1 to 19.2 kJ
due to the variation of mass of impactor (98.7 to 600 kg) and impact velocity (1 to
8 m/s). Although variations exist in beam geometry, types of impactor and the
interface, it was observed that values of maximum midspan deflection fits well with
the linear line. However, this was noticed that the proposed equation can be applied
up to the value of 150 of E Rb . More data is required in order to extend the
- 138 -
applicability of the equation beyond the value of 150 of E Rb .
Table 5.3: Type of impactor and the interface between the beam and impactor
Investigators Type of Impactor Impact Interface

Kishi et al. [K5] N/P N/P
Spherical with a radius of curvature of 1407
Kishi et al.[K7] Direct Contact
mm
Spherical steel weight with a radius of
Bhatti et al. [B8] Direct Contact
curvature of 1407 mm
Fujikake et al. Drop hammer with a hemispherical tip with
Direct Contact
[F4] radius of 90 mm
Stainless steel impactor with a 90 mm
diameter and hemispherical profile of 125
Chen and May 12 mm plywood and
mm radius and second one made of Mild
[C3] direct Contact
steel with a 100 mm diameter and a flat
contact surface
50 mm thick 305
Saatci and 305 mm square hollow structural steel
mm square steel
Vecchio [S1] weight
plate
Tachibana et Steel weight with a curve contact surface
Direct Contact
al.[T1] with a length of 565 mm, a radius of 75 mm
Kishi and Mikami Steel weight with a spherical bottom of
Direct Contact
[K10] radius in 1407 mm and a 2 mm taper
Note: N/P-Not Provided
- 139 -
120
(mm)
E
100 max = 0.574
R
max
b
Maximum midspan deflection,  80
60
40
20
0
0 50 100 150 200 250
Input impact energy/Static bending resistance, E/R (mm)
b
energy over static flexural resistance
Similarly, for statically shear critical beams, the relationship between maximum
midspan deflections  max versus input impact energy over static shear
resistance E Rs is shown in Fig. 5.20. As mentioned earlier that there is insufficient

impact response data for some shear failure type beam from the database, thus the
data points shown in Fig. 5.20 consists of 42 out of 53 specimens. The input impact
energy varies from 0.15 to 19.2 kJ due to the variation of mass of impactor (98.7 to
600 kg) and impact velocity (1 to 9.3 m/s). It was observed that the maximum
displacement can be accurately predicted by using the proposed equation. Finally
this can be concluded that more data would be essential in future in order to extend
the applicability of the equation beyond the value of 50 of E Rs .
- 140 -
60
(mm)
E
50 max = 0.614
max
R
Maximum midspan deflection, 

s
40
30
20
10
0
0 10 20 30 40 50 60 70 80
Input impact energy/Static shear resistance , E/R (mm)
s
energy over static shear resistance
5.5.1 Empirical formulas for impact-resistant design
Two empirical equations can be proposed from the aforementioned analysis of

experimental database.
For statically flexure failure type beam,
E
 max  0.574 (5-1)
Rb
Rearranging the above equation, it would be
E
Rb  0.574 (5-2)
 max
Similarly, for statically shear failure type beam,
E
 max  0.614 (5-3)
Rs
Rearranging the above equation, it would be
E
Rs  0.614 (5-4)
 max
- 141 -
where, Rb and Rs are in kN; E and  max are in J and mm, respectively.
Therefore, by specifying the maximum midspan deflection  max for each limit state
of the beam, the required static bending ( Rb ) or shear resistance ( Rs ) for impact
resistant design can be determined by applying Eq. (5-2) and (5-4). To demonstrate
the applicability of the proposed equations, comparison has been made with the
current test results. Figs. 5.21 and 5.22 (for flexure-critical and shear-critical
beams) show the comparative plot of current experimental results of maximum
deflection with the proposed equations. Thus, these two proposed equations could
be used in design with the before-mentioned limitations.
45
(mm)
E
40 max = 0.574
R
max
b
35
30
25
20
15
10
10 20 30 40 50 60
Input impact energy/Static bending resistance, E/R (mm)
b
Fig. 5.21: Comparison of current test results with proposed equation for flexure-
critical beams
- 142 -
40
(mm)
35 E
max = 0.614
max
R

30 s
25
20
15
10
0
0 10 20 30 40 50 60
Input impact energy/Static shear resistance , E/R (mm)
s
Fig. 5.22: Comparison of current test results with proposed equation for shear-
critical beams
The equations proposed for statically flexure failure type beam by Tachibana et al.
[T1] and Kishi and Mikami [K10] are almost similar to the proposed Eq. (5-2)
except for the constant values which were specified as 0.522 and 0.63, respectively.
The constant value proposed in Eq. (5-2) is 0.574, which is almost average of the
two values mentioned above; therefore Eq. (5-2) gives a less conservative but more
economical calculation of required static bending resistance as compared to Kishi
and Mikami [K10] and vice versa for Tachibana et al. [T1]. Although there are
some variations in beam dimensions, test setup, type of impactor and impactor
interface among researchers, it can be concluded that the required static bending
resistance for impact-resistant design is highly dependent on input impact energy
and the limit state of maximum midspan deflection.
The required static shear resistance for statically shear failure type beam against
impact loading can be evaluated by a simple equation proposed by Kishi et al. [K7]
as follows:
Ekd
Vusd  0.8 (5-5)
 rd
- 143 -
where Vusd is static shear resistance, Ekd is input kinetic energy and  rd is residual
displacement. The constant value proposed by them was 0.8 whereas Eq. (5-4)
provides a constant value of 0.614. The discrepancy is due to the reason that Kishi
et al. [K7] considered limit state of residual deflection,  rd as an input parameter
while in Eq. (5-4) limit state of maximum midspan deflection,  max is considered.
Therefore, it is difficult to directly compare these two equations.
5.6 Summary
Experimental assessment of RC beam under drop-weight impact loading and the

analysis of experimental dataset obtained from the literature are presented in this
chapter. Following conclusions are obtained:
(1) Maximum impact load, maximum midspan deflection and time to reach
maximum deflection were found to increase with the increment of drop-heights or
impact velocities. Here, maximum value of impact load corresponding to first peak
had taken as peak impact load. The first peak was due to the initial shock of impact
which was generated by the inertia of beam. However, for few cases the second
peak surpassed the first one; this one could be due to the presence and the
arrangement of load-cell which may itself affect the contact and the measurements.
Furthermore, due to the direct contact in between the impactor and beam, local
failure (i.e. massive cracking and spalling of concrete at or around the impact zone
at higher drop heights) played a significant role in affecting the response of impact
load history.
(2) The crack pattern and failure modes varied among the specimens depending on
their static resistance. Specimens of 3.3 series are statically shear-critical, therefore
in DR3.3_2.4 (without transverse reinforcements) hairline diagonal cracks were
observed at 0.3 m drop height. With the enhancement of the drop-heights, width of
the diagonal cracks and severity of local failure increases. At drop height of 1.2 m,
the damage was so severe that the beam was disintegrated into two components.
Similarly, for DR3.3_2.4_0.12, hairline diagonal cracks formed at low drop height
(i.e. 0.3 m) and with the increment of the drop heights, both widened diagonal
- 144 -
cracks and shear plug were noticed. At 1.2 m drop height, diagonal cracks in both
sides broadened extensively and extensive crushing and spalling of concrete
occurred at the impact region. Although, DR3.3_2.4_0.56 is statically shear-critical,
however, only few flexure-shear cracks were perceived at a drop height of 0.6 m.
With the further increment of drop heights, number of flexure-shear cracks
increases without any formation of diagonal cracks. This could be due to the
narrower spacing (i.e. 70 mm) of the transverse reinforcements in this beam. As
much as at 1.6 m drop heights, severity of damage increased significantly mainly at
or around impact region, with crushing and spalling of concrete. Specimens of 3.8
series are statically flexure-critical, therefore in SR3.8_0.8 (without compression
and transverse reinforcements) flexural cracks beneath the impact point were
observed at 0.3 m drop height. With further increment of drop heights, flexural
cracks became much wider and propagated vertically and crushed the compression
concrete. At a drop height of 0.9 m, local failure occurred at the impact region
which combined with the flexural crack underneath the impact point disintegrated
beam into two parts. Therefore, it can be concluded that compression
reinforcements plays an important role in preventing the disintegration of beam into
two parts. Another two types of beams of the same series (DR3.8_0.8_0.11 and
DR3.8_0.8_0.15) exhibited typical flexure failure at a drop height of 0.6 m. With
the further increment of the drop height to 1.2 m, flexural cracks underneath the
impact point became widened significantly. Moreover, there was massive local
damage occurred at or near the impact region. Almost similar type of failure was
observed for SR and DR beams of 5.7 series.
(3) Strain rates were computed from the strain history obtained from strain gauge.
The range of maximum strain rates of longitudinal tensile reinforcements for beams
of various series under various drop-heights is provided. This will help researchers
to gain some idea regarding the development of maximum strain rates and
correspondingly strain-rate dependent material properties can be applied in the
analytical and numerical model. In general, maximum strain rates of longitudinal
tensile reinforcements vary in between 1 to 7 s-1 for the corresponding considered
drop heights of 0.15 to 1.2 m (i.e. impact velocity varies from 1.72 to 4.85 m/s).
- 145 -
(4) Two empirical equations are proposed by analyzing a dataset on RC beams

under drop-weight impact loading, documented by several researchers from the
literature. By specifying the maximum midspan deflection for each limit state of
beam, the required static bending and shear resistance can be determined for
designing the beam subjected to input impact energy.
- 146 -
Chapter 6: Finite Element Modeling of RC Beams under Impact Loading
CHAPTER 6
FINITE ELEMENT MODELING OF RC BEAMS
UNDER IMPACT LOADING
6.1 Introduction
From the literature review on FE analysis of RC beams under drop-weight impact

loading, it was seen that most of the studies (Kishi et al. [K6], Bhatti et al. [B8],
Saatci and Vecchio [S2], May and Sangi [M7]) concentrated their efforts on
validating the numerical results and little efforts was expended towards
understanding the effect of various parameters on impact responses. Moreover,
impact testing of structural members to quantify the effect of several parameters is
cumbersome, expensive and requires special laboratory facilities. Thus, to expand
the knowledge beyond the range of parameters investigated experimentally and
their effects on the structural responses, three-dimensional nonlinear FE model was
developed in this chapter. Explicit finite element (FE) program LS-DYNA [H1]
was employed in this study because of its proven effectiveness in geometric
modeling and analysis capability for impact loading. However, prior to trusting the
results of any numerical model, it is of vital interest to validate the numerical
results with the test results. Therefore, the proposed FE model is first validated by
the test results and then used to conduct parametric studies. Therefore, the first
objective of this chapter is to numerically predict the experimental behavior (e.g.
time histories of impact and midspan deflections and failure modes etc.) of beams
under impact loading for various drop-heights. The second objective is that the
model will be employed further in Chapter 8 to numerically predict the residual
resistance of beam after impact damage. Moreover, to carry out the residual
resistance simulation in two steps, the resulting deformation and damages of beams
from the impact stages will be utilized.
Finite element models of RC structures have generally been based on replacing the
composite continuum by an assembly of elements representing the concrete and the
steel reinforcement. Three alternative approaches are mainly used for modeling
- 147 -
reinforcement in a three-dimensional finite element model of a concrete structure:

the discrete model, the embedded model, and the smeared model (El-Mezaini and
Citipitiouglu [E1]). Existing literature suggests the improvement of bond between
concrete and steel under impact loading. Thus, in this model, perfect bond is
assumed in between the concrete and reinforcement. Hence, concrete and
reinforcement mesh share the common nodes. However, to model the bond-slip
cases, reinforcing bars are modeled using special elements connected to concrete
through fictitious springs representing bond. This model is simple in concept and it
has advantage of representing different material properties more precisely. The only
shortcoming of using the discrete model is that concrete mesh patterns are restricted
by the location of reinforcement.
6.2 Finite Element Model
The nonlinear explicit FE program LS-DYNA [L1] was employed in this numerical
simulation study.
6.2.1 Structural configuration
To replicate the same experimental conditions, efforts are devoted here to develop a
FE model as close to the physical system as possible. Therefore, a three-
dimensional model is the obvious option to adopt. Fig. 6.1 shows the three-
dimensional FE model of beams (i.e. one type from each series). Eight node solid
hexahedron elements with a single integration point and viscous hourglass control
were used to represent beams (concrete), support rollers, inverted triangular plates
and impactor while beam elements (2-node Hughes-Liu beam element formulation
with 2  2 Gauss quadrature integration) were used to model steel reinforcing bars.
The FE model of the impactor divided into three sections to reproduce the same
used in the experiments. First section consists of hemispherical striking head
having 90 mm radius whereas solid cylinder and rectangular block were used to
model the remaining part. Different mesh sizes have been considered to study mesh
sensitivity during the initial stage of numerical study. Here, a mesh size of 15 mm
was used to create the solid elements which seemed to be ideal as further decrease
- 148 -
in mesh size almost generated the same results however augmented the
computational time. Table 6.1 presents total number of elements and nodes for
each beam model. The beams were supported on two cylinders made of solid
elements and to prevent the uplifting of the beam after impact loading two inverted
triangular plates were modeled at the top of the beam at support positions. A
numerical convergence study has shown that further decrease in mesh size had only
a very little effect on the accuracy of the results, while considerably increasing the
computational time. The mesh discretization was established in such a way that the
reinforcement nodes coincided with the concrete nodes. The steel reinforcing bars
were modeled explicitly using beam elements connected to the concrete mesh
nodes. Due to this assumption of complete compatibility of strains between the
concrete and steel nodes, they formed a perfect bond. Ultimate dynamic bond at
failure is 70-100% higher than that of quasi-static loading conditions for the case of
deformed reinforcing bars (Weathersby [W2]). Steel deformation under impact load
is limited to a region beneath the point of impact. Moreover, there is not enough
time to develop extensive bond slip along the length of the bar during an impact as
the peak impact load is being reached within few milliseconds. Thus, the
assumption of perfect bond in between the reinforcement and concrete was quite
reasonable.
Table 6.1: Total number of elements and nodal points for FE model
Solid elements
Beam designation Beam elements Nodal points
for concrete
DR3.3_2.4 22572 456 41266

DR3.3_2.4_0.12 22968 634 41900
DR3.3_2.4_0.56 24156 1258 46790
SR3.8_0.8 23760 270 42212
DR3.8_0.8_0.11 24288 818 40922
DR3.8_0.8_0.15 24640 902 52657
SR5.7_1.6 12960 270 27631
DR5.7_1.6_0.15 13248 720 28432
DR5.7_1.6_0.20 13440 776 28722
- 149 -
Fig. 6.1: Three-dimensional FE model of RC beams (a) a d -3.3; (b) a d -3.8; (c)
a d -5.7 series
- 150 -
6.2.2 Boundary conditions, application of impact velocity and contact

algorithm
Constraints were defined to the support cylinder and triangular plates, so that it
could rotate about its own longitudinal axis but would not be able to translate.
Triangular plates are placed 1 mm above the beam surface instead of giving the
direct contact to the beam to avoid any development of restraining moment at the
supports. After placing the impactor 1 mm apart from the top surface of beam at
mid-span, initial velocity was defined to it to save computational time which was
calculated from the free-falling formula. Constraints were defined to the impactor
in such a way that it could only move in vertical downward direction. Moreover,
the gravitational acceleration of impactor and self weight of beam was taken into
account in this numerical simulation. Global damping was not considered in the
impact simulation here as the impact responses lasted only 30 ms. The algorithm
CONTACT AUTOMATIC SURFACE TO SURFACE was used to model the
contact between the support cylinder, impactor and beam. The contact non-linearity
was stabilized by assigning a value of 30 for the viscous damping coefficient. This
algorithm automatically generates slave and master surfaces and uses a penalty
method where nominal interface springs were used to interpenetrate between
element and surfaces. The corresponding impact force due to the prescribed initial
velocity to impactor was then determined by monitoring the contact forces at the
concrete nodes in contact with impactor.
6.2.3 Constitutive model for concrete
MAT CSCM CONCRETE (MAT 159) is commonly known as Continuous Surface

Cap Model, was used to model concrete. This model was primarily developed to
simulate the deformation and damage of roadside safety concrete structures (i.e.
bridge rails, safety barriers and guard rails) impacted by moving vehicles. The
salient features of the models are: isotropic constitutive equations, three stress
invariant yield surface with translation for pre-peak hardening, a hardening cap that
expands and contracts, damage-based softening with erosion and modulus
reduction, rate effects for increasing strength in high-strain rate application. Further
- 151 -
details of the concrete model theory could be found elsewhere (Murray [M10]).
Moreover, the strain rate effects on concrete strength are duly considered in this
model. Default parameter generation option (by providing some key specifications)
was used herein instead of the traditional method where all material parameters
were needed. Default parameters were provided for the concrete model based on
three input specifications: the unconfined compressive strength of concrete,
aggregate size and units. This option is valid for unconfined compressive strength
from 20 to 58 MPa and aggregate sizes from 8 to 32 millimeters. The unconfined
compressive strength affects all aspects of the fit, including stiffness, three-
dimensional yield strength, hardening and damage-based softening whereas the
aggregate size affects only the softening behavior of damage formulation.
6.2.4 Constitutive model for reinforcing steel
The steel reinforcement bars (longitudinal and transverse reinforcements) within

the beam were modeled as a strain sensitive uniaxial elastic-plastic material to
account for its strain rate sensitivity as well as stress-strain history dependence.
Material model MAT PIECEWISE LINEAR PLASTICITY (MAT 024) was used
in this study to model reinforcing bars where the strain rate effects are duly
considered. The expressions proposed by Malvar [M2] about strain rate effects on
yield and ultimate stresses of steel reinforcements were employed in this study
(already discussed in Chapter 4).
6.2.5 Constitutive model for impactor, support rollers and inverted

triangular plate
MAT RIGID (MAT 020) was used from LS-DYNA [L1] material library to model
the impactor, support rollers and inverted triangular plates. Realistic values of
Young‟s modulus and Poisson‟s ratio of the rigid material should be defined since
unrealistic values may contribute to numerical problems in contact. Young‟s
modulus and Poisson‟s ratio of steel material were used for the rigid material in the
numerical simulation.
- 152 -
6.3 Verification of Finite Element Analysis Results
To establish the reliability and accuracy of the FE model, it is essential to validate

or verify the FE results with the experimental results. Therefore, this section
provides the comparison of FE results with the test results presented in the previous
chapter. Time histories of impact load and midspan deflection and crack pattern
(i.e. at various stages of impact loading duration) of the beams for various drop
heights would be used here for verification purpose. But, before that one
conservation of energy should be checked in simulation results. As mentioned
earlier, eight node solid hexahedron elements with a single integration point were
used to model concrete. Although these elements have the advantage of requiring
less computational time, however, hourglass modes (i.e. nonphysical, zero-energy
modes of deformation that produce zero strain and no stress) occur in these
elements. LS-DYNA [L1] has various algorithms for inhibiting hourglass modes. A
probable way to reduce hourglassing is to refine the mesh. However, to entirely
eliminate hourglass concern, the efficient way is to switch the element formulation
with fully-integrated or selectively reduced integration. To satisfy the conservation
of energy during an impact event, the kinetic energy of impactor before impact
should be equal to the sum of the current kinetic energy, the internal energy of
beam and the energy loss due to friction and hourglass modes. Energy loss due to
friction was very small and can be neglected. The variation of various energies in a
typical impact simulation is shown in Fig. 6.2.
1.5
Energy (kJ)
1 Total
Kinetic
Internal
0.5 Hourglass
0
0 0.005 0.01 0.015 0.02 0.025 0.03
Time (s)
Fig. 6.2: Variation of various energies in a typical impact simulation
- 153 -
Here, hourglass energy was around 3.5% of total energy at peak impact load and
around 8.5% of total energy at maximum midspan deflection. Therefore, the low
amount of hourglass energy in impact simulation corroborates the acceptance of
analysis results.
6.3.1 Time histories of impact load curves
The comparison of time histories of impact load curves between test results and FE
predictions is illustrated in Fig. 6.3. Moreover, comparison of first peak load in
between test and numerical analysis results is summarized in Table 6.2 for all
tested beams. From Fig. 6.3 and Table 6.2, it can be concluded that the FE models
can predict the first peak load in impact load history under various drop heights
within a reasonable accuracy.
500 600
400 DR3.3_2.4_0.56 DR3.3_2.4_0.56
400
300 Drop height - 0.60 m Drop height - 0.90 m
Impact load (kN)
Impact load (kN)
200 200
100
0 0
-100 Experiment Experiment

-200
-200 FEM FEM
-300 -400
0 0.005 0.01 0.015 0.02 0 0.005 0.01 0.015 0.02
Time (s) Time (s)
600
DR3.3_2.4_0.56
400
Drop height - 1.2 m
Impact load (kN)
200
Experiment
-200
FEM
-400
0 0.005 0.01 0.015 0.02
Time (s)
(a)
heights (a) 3.3 series; (b) 3.8 series; (c) 5.7 series (continued)
- 154 -
400 600
DR3.8_0.8_0.11 DR3.8_0.8_0.11
300
Drop height - 0.6 m 400
Drop height - 0.9 m
Impact load (kN)

Impact load (kN)
200
200
100
0
0
-100 -200
FEM FEM
-200 -400
0 0.005 0.01 0.015 0.02 0 0.005 0.01 0.015 0.02
Time (s) Time (s)
600
DR3.8_0.8_0.11
400 Drop height - 1.2 m
Impact load (kN)
200
-200 Experiment
FEM
-400
0 0.005 0.01 0.015 0.02
Time (s)
(b)
200 200
DR5.7_1.6_0.20 DR5.7_1.6_0.20
150 150
Drop height - 0.3 m Drop height - 0.45 m
Impact load (kN)
Impact load (kN)
100 100
50 50
0 0
Experiment
-50 -50 Experiment
FEM FEM
-100 -100
0 0.005 0.01 0.015 0.02 0 0.005 0.01 0.015 0.02
Time (s) Time (s)
- 155 -
300
DR5.7_1.6_0.20
Impact load (kN)

100
0
Experiment
-100
FEM
-200
0 0.005 0.01 0.015 0.02
Time (s)
(c)
heights (a) 3.3 series; (b) 3.8 series; (c) 5.7 series
However, the discrepancy in first post peak response characteristics is due to the
presence and the arrangement of load-cell which may itself affect the contact and
the measurements (Bischoff [B8]). Furthermore, due to the direct contact in
between the impactor and beam, local failure (i.e. cracking and spalling of concrete
at or around the impact zone) played a significant role in controlling the
characteristics of impact load history. Moreover, the local oscillation and stress
wave propagation through the impactor could affect the measurements.
- 156 -
Table 6.2: Comparison of maximum impact load of beam between test and FE
analysis results
Drop Impact Maximum impact load (kN)

Specimens heights velocity Test/FEM
Test FEM
(m) (m/s)
0.15 1.72 128.4 147.0 0.87
0.30 2.43 195.3 209.2 0.93
DR3.3_2.4
0.60 3.43 245.0 284.0 0.86
1.20 4.85 257.0 345.0 0.74
0.30 2.43 195.0 208.0 0.94
0.60 3.43 236.0 284.0 0.83
DR3.3_2.4_0.12
0.90 4.20 315.0 315.3 1.00
1.20 4.85 - - -
0.60 3.43 289.0 289.5 1.00
0.90 4.20 336.0 305.0 1.11
DR3.3_2.4_0.56
1.20 4.85 - - -
1.60 5.60 - - -
0.30 2.43 201.0 170.0 1.18
SR3.8_0.8 0.60 3.43 247.9 269.0 0.92
0.90 4.20 291.0 304.0 0.96
0.60 3.43 245.9 258.0 0.95
DR3.8_0.8_0.11 0.90 4.20 347.0 303.0 1.15
1.20 4.85 374.0 347.0 1.08
0.60 3.43 264.5 271.0 0.98
DR3.8_0.8_0.15 0.90 4.20 297.0 343.0 0.87
1.20 4.85 315.0 360.0 0.88
0.30 2.43 126.0 117.0 1.08
SR5.7_1.6 0.45 2.97 146.6 143.0 1.03
0.60 3.43 170.6 170.0 1.00
0.30 2.43 127.2 132.0 0.96
DR5.7_1.6_0.15 0.45 2.97 145.9 153.0 0.95
0.60 3.43 160.5 162.0 0.99
0.30 2.43 117.0 116.0 1.01
DR5.7_1.6_0.20 0.45 2.97 137.2 155.0 0.89
0.60 3.43 168.9 170.0 0.99
Mean 0.97
COV 0.10
- 157 -
Moreover, two types of RC beam specimens (i.e. S1616 and S1322) were taken
from Fujikake et al. [F4] to examine the accuracy of numerical model. The
comparison of time histories of impact load curves between test results and FE
predictions is presented in Fig. 6.4. From this figures, it can be opined that the
numerical model can reasonably predict the impact load history of beams.
250 300
S1616 250 S1616
200
Drop height - 0.15 m 200 Drop height - 0.30 m
Impact load (kN)
Impact load (kN)

150
150
100 100
50
50
0
0 Experiment Experiment
FEM -50 FEM
-50 -100
0 0.005 0.01 0.015 0 0.005 0.01 0.015
Time (s) Time (s)
400
S1616
Impact load (kN)
200
100
0
Experiment
FEM
-100
0 0.005 0.01 0.015
Time (s)
(a)
300 300
S1322 S1322
250 250
Impact load (kN)
Impact load (kN)
200 200
150 150
100 100
50 Experiment 50 Experiment
FEM FEM
0 0
0 0.005 0.01 0.015 0 0.005 0.01 0.015
Time (s) Time (s)
- 158 -
400
350 S1322
Impact load (kN)

250
200
150
100
Experiment
50 FEM
0
0 0.005 0.01 0.015
Time (s)
(b)
heights (a) S1616; (b) S1322
6.3.2 Time histories of midspan deflection curves
The comparison of time histories of midspan deflection curves between test results
and FE predictions is shown in Fig. 6.5. Furthermore, Table 6.3 summarized the
comparison of maximum midspan deflection of test and numerical results for all
tested beams.
- 159 -
20 25
DR3.3_2.4_0.56

15 DR3.3_2.4_0.56 20
15
10
10
5
5
0
Experiment 0
Experiment
-5 FEM -5 FEM
-10 -10
0 0.005 0.01 0.015 0.02 0 0.005 0.01 0.015 0.02 0.025
Time (s) Time (s)
30
25 DR3.3_2.4_0.56

15
10
5
0 Experiment
-5 FEM
-10
0 0.005 0.01 0.015 0.02 0.025
Time (s)
(a)
30 50
DR3.8_0.8_0.11 DR3.8_0.8_0.11
25 40
20
30
15
20
10
10
5 FEM
FEM
0 0
0 0.005 0.01 0.015 0.02 0.025 0.03 0 0.005 0.01 0.015 0.02 0.025 0.03
Time (s) Time (s)
drop heights (a) 3.3 series; (b) 3.8 series; (c) 5.7 series (continued)
- 160 -
60

50 DR3.8_0.8_0.11
Drop height - 1.2 m
40
30
20
Experiment
10 FEM
0
0 0.005 0.01 0.015 0.02 0.025 0.03
Time (s)
(b)
30
40

DR5.7_1.6_0.20 DR5.7_1.6_0.20
25
20 30
15
20
10
10
5 FEM FEM
0 0
0 0.005 0.01 0.015 0.02 0.025 0.03 0 0.005 0.01 0.015 0.02 0.025 0.03
Time (s) Time (s)
50
DR5.7_1.6_0.20
30
20
Experiment
10 FEM
0
0 0.005 0.01 0.015 0.02 0.025 0.03
Time (s)
(c)
drop heights (a) 3.3 series; (b) 3.8 series; (c) 5.7 series
- 161 -
Table 6.3: Comparison of maximum midspan deflection of beam between test and
FE analysis results
Impact Maximum midspan

Drop
Specimens velocity deflection (mm) Test/FEM
heights (m)
(m/s) Test FEM
0.15 1.72 3.8 3.2 1.19
0.30 2.43 6.9 5.2 1.33
DR3.3_2.4
0.60 3.43 12.7 8.5 1.49
1.20 4.85 - - -
0.30 2.43 6.4 5.2 1.23
0.60 3.43 11.5 8.4 1.37
DR3.3_2.4_0.12
0.90 4.20 - - -
1.20 4.85 - - -
0.60 3.43 9.0 7.4 1.22
0.90 4.20 13.4 10.0 1.34
DR3.3_2.4_0.56
1.20 4.85 15.7 12.0 1.31
1.60 5.60 19.2 15.2 1.26
0.30 2.43 10.8 13.0 0.83
SR3.8_0.8 0.60 3.43 20.2 25.0 0.81
0.90 4.20 29.5 38.0 0.78
0.60 3.43 18.6 16.9 1.10
DR3.8_0.8_0.11 0.90 4.20 34.5 25.5 1.35
1.20 4.85 41.0 33.0 1.24
0.60 3.43 19.6 17.4 1.13
DR3.8_0.8_0.15 0.90 4.20 28.8 25.0 1.15
1.20 4.85 39.2 33.6 1.17
0.30 2.43 20.0 19.9 1.01
SR5.7_1.6 0.45 2.97 28.8 32.7 0.88
0.60 3.43 37.1 42.7 0.87
0.30 2.43 20.0 18.3 1.09
DR5.7_1.6_0.15 0.45 2.97 30.0 25.8 1.16
0.60 3.43 39.1 33.5 1.17
0.30 2.43 19.1 18.4 1.04
DR5.7_1.6_0.20 0.45 2.97 28.8 26.0 1.11
0.60 3.43 37.9 33.2 1.14
Mean 1.14
COV 0.16
- 162 -
Similarly, the comparison of time histories of mid-span deflection curves between

test results and FE predictions is shown in Fig. 6.6. RC beam specimens (i.e. S1616
and S1322) were adopted from the experimental program conducted by Fujikake et
al. [F4].
20 30
S1616 S1616

25
20
10 FEM 15 FEM
10
5
5
0 0
0 0.005 0.01 0.015 0.02 0 0.005 0.01 0.015 0.02
Time (s) Time (s)
60
S1616
50
Drop height - 0.60 m
40
Experiment
30 FEM
20
10
0
0 0.005 0.01 0.015 0.02 0.025
Time (s)
(a)
20 25
S1322 S1322

15
15 Experiment
Experiment FEM
10 FEM
10
5
5
0 0
0 0.005 0.01 0.015 0.02 0 0.005 0.01 0.015 0.02
Time (s) Time (s)
Fig. 6.6: Comparison of time histories of mid-span deflection of beams for various
drop heights (a) S1616; (b) S1322 (continued)
- 163 -
40
35 S1322

25
20
15
10 Experiment
5 FEM
0
0 0.005 0.01 0.015 0.02
Time (s)
(b)
Fig. 6.6: Comparison of time histories of mid-span deflection of beams for various
drop heights (a) S1616; (b) S1322
6.3.3 Crack profiles
This section presents the comparative plot of crack profiles (e.g. side surface) of
beams at various stages (at 2.5, 5, 10, 15, 20 and 25 ms) of impact loading. The
damage of the beams obtained from numerical simulation is shown by plotting the
fringes of effective plastic strain in Figs. 6.7 to 6.10. Fig. 6.7 shows the comparison
of crack profiles of DR3.8_0.8_0.11 at a drop height of 0.6 m. In this case, the
beam failed in flexure with minor local damage. At a drop height of 1.2 m, massive
local damage occurred at or around the impact region with the widening of flexure
cracks which is shown through the comparative plot in Fig. 6.8. Similar two
comparative cases for DR5.7_1.6_0.20 under drop heights of 0.3 and 0.6 m are
presented in Figs. 6.9 and 6.10, respectively where in the first case flexure cracks
(with mild local damage) were observed and for the second extensive local
damages occurred. This is evident from these figures that the crack patters were
captured quite effectively at various stages of impact loading.
- 164 -
at 2.5 ms
at 5 ms
at 10 ms
plastic strain from simulation of DR3.8_0.8_0.11 at a drop height of 0.6 m
(continued)
- 165 -
at 15 ms
at 20 ms
at 25 ms
- 166 -
at 2.5 ms
at 5 ms
at 10 ms
(continued)
- 167 -
at 15 ms
at 20 ms
at 25 ms
- 168 -
at 2.5 ms
at 5 ms
at 10 ms
(continued)
- 169 -
at 15 ms
at 20 ms
at 25 ms
- 170 -
at 2.5 ms
at 5 ms
at 10 ms
(continued)
- 171 -
at 15 ms
at 20 ms
at 25 ms
(continued)
- 172 -
6.3.4 Impact resistance of test specimens
Impact resistance (i.e. reaction force) of specimens can be obtained from the
support load cells. However, in this experimental investigation, due to the
attachment of 100 kg mass to the support load cell, the test reaction force would not
be the actual response. Hence, the test reaction force was not used for validation of
simulation results. Fig. 6.11 displays the typical variation of force history obtained
from top and bottom load-cells during drop-weight impact testing.
(a) 400 (b) 250

1.99 ms 2.2 ms
300 Impact 200 Impact
Reaction Reaction
200 150
Load (kN)
Load (kN)
100 100
0 50
-100 DR3.8_0.8_0.11_0.60 m 0 DR5.7_1.6_0.2_0.30 m

-200 -50
0 0.005 0.01 0.015 0.02 0 0.005 0.01 0.015 0.02
Time (s) Time (s)
Fig. 6.11: Typical variation of force history obtained from top and bottom load-
cells during drop-weight impact testing
- 173 -
Moreover, there is a time lag (2.98 ms) in between the two responses which is
clearly visible in this figure. This is due to the propagation time required of stress
wave to travel from loading point to support region.
300
2.45 ms
250 Impact
Reaction
200
Load (kN)
150
100
50
0
0 0.005 0.01 0.015 0.02
Time (s)
Fig. 6.12: Typical variation of impact and reaction load history (from numerical
simulation)
Still, efforts are made (after gaining confidence from the verification of others
impact responses) to present the variation of maximum reaction force (obtained
from numerical simulation) and the ratio of maximum reaction force to calculated
static resistance. Fig. 6.12 displays the typical variation of impact and reaction
force history obtained from numerical simulation and the time lag (2.45 ms) in
between the two responses is clearly visible.
- 174 -
Table 6.4: Maximum reaction force obtained from numerical simulation
Maximum
Maximum Calculated
reaction force
Drop heights reaction force static
Specimens after impact/
(m) after impact resistance
Calculated static
(kN) (kN)
resistance
0.15 176.4 2.66
0.30 235.1 3.55
DR3.3_2.4 66.2
0.60 314.9 4.76
1.20 377.8 5.70
0.30 242.2 2.78
0.60 315.9 3.63
DR3.3_2.4_0.12 87.1
0.90 327.2 3.76
1.20 358.1 4.11
0.60 317.0 1.86
0.90 331.2 1.94
DR3.3_2.4_0.56 170.4
1.20 367.5 2.16
1.60 420.0 2.46
0.30 115.4 1.70
SR3.8_0.8 0.60 150.8 67.8 2.22
0.90 202.7 2.99
0.60 165.3 2.44
DR3.8_0.8_0.11 0.90 209.6 67.8 3.10
1.20 214.3 3.16
0.60 166.7 2.46
DR3.8_0.8_0.15 0.90 224.9 67.8 3.32
1.20 262.8 3.88
0.30 133.1 3.77
SR5.7_1.6 0.45 168.6 35.3 4.77
0.60 193.7 5.48
0.30 132.1 3.12
DR5.7_1.6_0.15 0.45 160.4 42.4 3.78
0.60 182.2 4.29
0.30 126.7 2.99
DR5.7_1.6_0.20 0.45 168.5 42.4 3.97
0.60 197.0 4.65
Table 6.4 presents the maximum impact resistance of beam for each drop-heights
and the corresponding DIF. Here, DIF is defined here as the ratio of maximum
reaction force to calculated static resistance of beam. This table may provide some
insight about the variation of DIF under various drop-heights. However, more
efforts should be spent in retrieving the reaction force-history accurately from
experiment so that actual impact resistance of beam could be assessed.
- 175 -
6.4 Parametric Study
Parametric investigation is presented in this section to discuss more information

about the behavior of beams under varying drop heights by varying some key
parameters such as mass ratio (  ) i.e. ratio of the beam-mass ( mb ) to impactor-
mass ( mi ), longitudinal reinforcement ratios (  L ), compressive strength of
concrete ( f c ) and boundary conditions. The influence of these parameters on

impact responses is summarized and discussed. The specimens considered here
having clear span length and cross sections are identical to the DR beams ( a d -3.3,
3.8 and 5.7) used in the experiment.
6.4.1 Effect of mass ratio (  )
Table 6.5 shows the variation in mass-ratios (i.e. 0.4 to 1.0) considered for each
series specimens. Here the mass of the beams was kept constant; however, different
mass ratios were obtained by varying the impactor-mass. To compare, the input
kinetic energy is maintained constant whereas the impactor mass and drop-height is
allowed to vary (e.g. impact energy: 2.65 kJ; impactor mass: 300, 170 and 125 kg;
drop-height: 0.9, 1.59 and 2.16 m). Fig. 6.13 illustrates the effect of the mass ratios
on maximum impact load and maximum midspan deflection of beams under
varying impact energies.
Table 6.5: Variations in mass-ratios
f c f Ly f Ty mb mi
ad
(kg) (kg)   mb mi
(MPa) (MPa) (MPa)
3.3 126.0 300, 170, 125 0.42, 0.74, 1.00
3.8 40 520 310 147.5 300, 200, 150 0.49, 0.74, 1.00
5.7 78.3 150, 105, 78 0.52, 0.75, 1.00
It was observed that under certain impact energy, maximum impact load increases
and maximum midspan deflection decreases with the increment of the mass-ratios.
This physically implied that under constant impact energy, impacts caused by high
mass with low velocity resulted in smaller maximum impact load but higher
- 176 -
maximum midspan deflection of beam and vice versa.
(a) 700
Maximum midspan deflection (mm)

18
- 0.42 - 0.42
a/d-3.3 16 a/d-3.3
600 L-2.4% - 0.74 L-2.4% - 0.74
T-0.56% - 1.0 14 T-0.56% - 1.0
500 12
400 10
8
300
6
200 4
1.5 2 2.5 3 3.5 4 1.5 2 2.5 3 3.5 4
Impact energy (kJ) Impact energy (kJ)
(b) 550
50
- 0.49 - 0.49
a/d-3.8 45 a/d-3.8
500
L-0.8% - 0.74 L-0.8% - 0.74
40
450 T-0.15% - 1.0 T-0.15% - 1.0
35
400 30
25
350
20
300
15
250 10
1.5 2 2.5 3 3.5 4 1.5 2 2.5 3 3.5 4
(c)
350 50
- 0.52 - 0.52
a/d-5.7 45 a/d-5.7
300
L-1.6% - 0.75 40
L-1.6% - 0.75
T-0.20% - 1.0 T-0.20% - 1.0
35
250 30
25
200 20
15
150 10
0.8 1 1.2 1.4 1.6 1.8 0.8 1 1.2 1.4 1.6 1.8
Fig. 6.13: Effect of mass ratios on maximum impact load and maximum midspan
deflection of beams (a) a d -3.3; (b) a d -3.8; (c) a d -5.7
6.4.2 Effect of longitudinal reinforcement ratio (  L )
The influence of longitudinal reinforcement ratio on maximum impact responses

under various impact energies is presented in Fig. 6.14. The percentage of
- 177 -
longitudinal reinforcement ratios (i.e. 0.8, 1.5 and 2.0%) and these ratios in terms of
balanced reinforcement ratios (0.3, 0.56 and 0.75) are depicted in Table 6.6. Here,
the mass of the impactor was considered to be 300 kg. Maximum impact load
increases and maximum midspan deflection of beam decreases with the
enhancement of the longitudinal reinforcement ratios. Fig. 6.15 exhibits the effect
of longitudinal reinforcement ratio on failure pattern of beam under impact energy
of 3.53 kJ. Severity in failure pattern (including diagonal and punching shear
cracks) is observed with increasing amount of longitudinal reinforcements. This
could be the reason behind the decrement in maximum midspan deflection when the
beam contains higher amount of longitudinal reinforcements.
Table 6.6: Variations in longitudinal reinforcement ratios
f c f Ly f Ty b L
ad (%) (%)  L b
(MPa) (MPa) (MPa)
0.8 0.30
3.8 40 520 310 2.7 1.5 0.56
2.0 0.75
500
60
L-0.8 L-0.8
a/d-3.8 50 a/d-3.8
450 L-1.5 L-1.5
 -0.24%  -0.24%
T L-2.0 T L-2.0
40
400
30
350
20
300
10
250 0
1.5 2 2.5 3 3.5 4 1.5 2 2.5 3 3.5 4
Fig. 6.14: Effect of longitudinal reinforcement ratios on maximum impact load and
maximum midspan deflection of beams
- 178 -
DR3.8_0.8_0.24 (25 ms) DR3.8_1.5_0.24 (25 ms)
DR3.8_2.0_0.24 (25 ms)
Fig. 6.15: Effect of longitudinal reinforcement ratios on failure pattern of beam

under impact energy of 3.53 kJ
6.4.3 Effect of compressive strength of concrete ( f c )
Beam having concrete compressive strength of 30, 40, 50 MPa was used to
determine the effect of f c on maximum impact load and maximum midspan
deflection under varying impact energies (considered mass of the impactor is 300
kg). Two case studies are depicted in Fig. 6.16. With the enhancement of concrete
compressive strength from 30 to 50 MPa, an increasing trend in maximum impact
load is observed whereas the opposite phenomenon was noticed for the case of
maximum midspan deflections.
- 179 -
(a) 500

Maximum impact load (kN) 18
a/d-3.3 30 MPa 16 a/d-3.3 30 MPa
450 L-2.4% 40 MPa L-2.4% 40 MPa
T-0.56% 50 MPa
14 T-0.56% 50 MPa
400
12
350
10
300 8
250 6
1.5 2 2.5 3 3.5 4 1.5 2 2.5 3 3.5 4
(b) 300

70
280 a/d-5.7 30 MPa

30 MPa a/d-5.7
260
L-1.6% 40 MPa
60
L-1.6% 40 MPa
T-0.67% 50 MPa T-0.67% 50 MPa
240 50
220
200 40
180
30
160
140 20
1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8
Fig. 6.16: Effect of compressive strength of concrete on maximum impact load and
maximum midspan deflection of beams (a) a d -3.3; (b) a d -5.7
6.4.4 Effect of boundary conditions
To investigate the effect of boundary conditions on impact response of beam,

pinned-end and fixed-end conditions were considered. Fig. 6.17 clearly presents
that maximum impact load is in the higher side however maximum midspan
deflection is in lower side for fixed-end beam as compared to pinned-end beam.
Development and propagation of cracks in beams (pinned-end and fixed-end) at
various stage of impact loading are shown in Fig. 6.18.
- 180 -
(a) 650

18
Maximum impact load (kN) 600 a/d-3.3 Pinned-pinned

16 a/d-3.3 Pinned-pinned
550 L-2.4% Fixed-fixed 14 L-2.4% Fixed-fixed
T-0.56% T-0.56%
500 12
450 10
400 8
350 6
300 4
250 2
1.5 2 2.5 3 3.5 4 1.5 2 2.5 3 3.5 4
(b) 500

50
a/d-3.8 Pinned-pinned a/d-3.8 Pinned-pinned

450 L-0.8% Fixed-fixed 40 L-0.8% Fixed-fixed
T-0.11% T-0.11%
400 30
350 20
300 10
250 0
1.5 2 2.5 3 3.5 4 1.5 2 2.5 3 3.5 4
(c)
240 50
220 a/d-5.7 Pinned-pinned a/d-5.7 Pinned-pinned

L-1.6% Fixed-fixed
40 L-1.6% Fixed-fixed
200 T-0.20% T-0.20%
180 30
160 20
140
10
120
100 0
0.8 1 1.2 1.4 1.6 1.8 0.8 1 1.2 1.4 1.6 1.8
Fig. 6.17: Effect of boundary conditions on maximum impact load and maximum
midspan deflection of beams (a) a d -3.3; (b) a d -3.8; (c) a d -5.7
- 181 -
at 2.5 ms
at 5 ms
at 10 ms
at 15 ms
Fig. 6.18: Effect of boundary conditions on failure pattern of beams (left: pinned-
end; right: fixed-end) (continued)
- 182 -
at 20 ms
at 25 ms
DR3.8_0.8_0.11 (Drop height-0.9 m/Impact energy-2.65 kJ)
Fig. 6.18: Effect of boundary conditions on failure pattern of beams (left: pinned-
end; right: fixed-end)
6.5 Summary
FE model was developed in this chapter to evaluate behavior of beams under drop-
weight impact loading. After verifying the numerical simulation results against the
experimental results, parametric investigation was carried out to elucidate the
influence of some key parameters on impact responses.
- 183 -
Under constant impact energy, impacts caused by high mass with low velocity
result in smaller maximum impact load but higher maximum midspan deflection of
beam and vice versa. Maximum impact load increases and maximum midspan
deflection of beam decreases with the enhancement of the longitudinal
reinforcement ratios. With increasing amount of longitudinal reinforcements
severity in failure pattern (including diagonal and punching shear cracks) is
observed. Moreover, with the enhancement of concrete compressive strength from
30 to 50 MPa, an increasing trend in maximum impact load was observed whereas
the opposite phenomenon was noticed for the case of maximum midspan
deflections. Finally, to assess the effect of boundary condition on impact responses,
fixed-end beam has been compared with pinned-end beam. Maximum impact load
is in the higher side however maximum midspan deflection is in lower side for
fixed-end beam as compared to pinned-end beam under considered impact energies.
- 184 -
Chapter 7: Analytical Model for Impact Response Prediction of RC Beams
CHAPTER 7
ANALYTICAL MODEL FOR IMPACT RESPONSE
PREDICTION OF RC BEAM
7.1 Introduction
Maximum midspan deflection could be an important performance index to evaluate

the damage levels (i.e. from no damage to moderate damage, severe damage and
complete collapse) of beam when subjected to impact loadings (Fujikake et al.
[F4]). Moreover, based on the relationships among maximum mid-span deflection
of, static flexural resistance of beam and input impact energy empirical design
formula following the performance-based design concept has been proposed by
Tachibana et al. [T1], Kishi and Mikami [K10]. Therefore, prediction of maximum
midspan deflection of beam subjected to impact loading could be an important
damage index and design parameter. Numerical modeling and prediction of impact
responses have been demonstrated in previous chapter. However, to reduce the
modeling efforts and computational time, simplified analytical models are
presented in this chapter.
7.2 Energy-balance Model
The balance of energy in the system is considered for analyzing the impact
response. The initial kinetic energy of the impactor is used to deform the structure
during impact. When the structure reaches its maximum deflection, the velocity of
the impactor becomes zero and all the initial kinetic energy has been used to deform
the structure, assuming that the structure behaves quasi-statically. The overall
deformation of the structures usually involves bending, shear deformation, and for
large deflections, membrane stiffening effects. Local deformations in the contact
zone also are to be considered. For impacts that induce only small amount of
damage, the energy needed to create damage can be neglected. Therefore, the
energy-balanced equation can be written as
- 185 -
1
MV 2  Eb  E s  E m  Ec (7-1)
2
where M = mass of the impactor, V = velocity just before impact and the subscripts
b, s and m refer to the bending, shear and membrane components of the overall
structural deformation, and Ec is the energy stored in the contact region during
indentation. Moreover, this is possible to express the load-deflection relation in the
form presented below
P  K bsW  K mW 3 (7-2)
where P is the impact force, K bs = stiffness including bending and transverse shear
deformation effects, K m = membrane stiffness and W is the deflection at the impact

point.
1 1
Eb  E s  E m  2
K bsWmax  K mWmax
4
(7-3)
2 4
Contact deformation and contact force can be described by the well-known

Hertzian Contact Law
3
P  n 2 (7-4)
where n represents the contact stiffness related to the material properties of

1
4 12 
contacting bodies and is given by n  Ri 1  i / Ei 
3 
2 1  
Eb  (Sun and Yang
[S8]) Ri , i and Ei are radius, Poisson‟s ratio and young modulus of impactor and
Eb is the young modulus of beam;  represents the relative motion or indentation

of the structure by the impactor
5
 max 2
Ec   Pd  n max
2
(7-5)
0 5
Using equations (7-2), (7-4) and (7-5), the maximum indentation can be expressed
in terms of the maximum displacement of the structure at the impact location:
2
 P 3 2 2
 max     n  3 K bsWmax  K mWmax
3 3 (7-6)
 
n
- 186 -
After substituting into equation (7-5), the contact energy becomes
2  23
 
5
Ec  n K bsWmax  K mWmax
3 3 (7-7)
5
Using equations (7-3) and (7-7), the energy balance equation becomes
2 2
 
5
1 1 1
MV 2  K bsWmax
2
 K mWmax
4
 n  3 K bsWmax  K mWmax
3 3 (7-8)
2 2 4 5
After rearranging equation (7-8),
 2 5

  3 
2 
Wmax  
1 1 1 K 4 K W K 3
MV  K bsWmax 1 
2 m
W 
 2 K bs max 5 
2 bs max
  1  m 2
(7-9)
2 2 n   K  

bs

Neglecting the membrane effects ( K m  0 ), above equation simplifies to
 2

1 1 2  4  K W 
 
3
MV  K bsWmax 1  
2 bs max
(7-10)
2 2  5 n  
 
This equation can be solved numerically for Wmax , maximum deflection of structures
at impact point and maximum impact load can be calculated from the following
equation
5
2 3
1 1P 4P
MV 2  
max max
2
(7-11)
2 2 K bs 5 3
n
This should be noted that the energy method is not dependent on the mass of the
structures. It can only provide the maximum impact responses such as maximum
midspan deflection and maximum impact load; however, cannot capture the effect
caused by various levels impactor mass and velocity when impact energy remains
constant. Furthermore, time-history or time-to-peak response value cannot be
evaluated by using this method.
7.2.1 Application of energy-balance model
The applicability of the energy-balance model is judged by comparing the predicted
- 187 -
response with the experimental results. Fig. 7.1 shows the comparison of maximum
midspan deflection of beams for various drop heights. For each beam in various
series (3.3, 3.8 and 5.7), three drop heights are considered and this was observed
that for the case of lower drop height, this model can capture the results quite
accurately; however for the cases of intermediate to higher drop heights, the degree
of accuracy decreases. Moreover, comparison of all tested data is presented in Fig.
7.2 and discrepancies are mainly observed for midspan deflection more than 20 mm
and obviously these are cases resulted from higher drop heights. This is due to the
reason that in the theory it is assumed that the strain energy stored in the impacted
structure equals the kinetic energy of the impactor at the instance of time when the
maximum impact load occurs.
30 40
(a) DR3.3_2.4 DR3.3_2.4_0.12
Experiment 30
20 Experiment
Energy method Energy method
20
10
10
0 0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
30
DR3.3_2.4_0.56
Experiment
20 Energy method
10
0
0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3
Drop height (m)
- 188 -

60 80
(b) SR3.8_0.8 DR3.8_0.8_0.11
40 Energy method Energy method
40
20
20
0 0
0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3
80
DR3.8_0.8_0.15
60 Experiment
Energy method
40
20
0
0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3
Drop height (m)
80 80
(c) SR5.7_1.6 DR5.7_1.6_0.15
60 Experiment 60 Experiment
Energy method Energy method
40 40
20 20
0 0
0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65
- 189 -

80
DR5.7_1.6_0.20
60 Experiment
Energy method
40
20
0
0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65
Drop height (m)
heights (a) 3.3 series; (b) 3.8 series; (c) 5.7 series
This assumption is solely valid for quasi-static impact (i.e. low velocity or low drop
heights) events as the amount of energy lost due to large plastic deformations and
generation of heat and noise in high-speed impacts (i.e. higher drop heights) would
significantly alter the energy balance.

50
40
Experimental results
30
20
10
0
0 10 20 30 40 50
Results predicted by energy method
Fig. 7.2: Comparison of maximum midspan deflection of all tested beams
Furthermore to examine the effectiveness of energy method, the analytical results

are compared with other researcher‟s test results. Two types of beams were taken
from Fujikake et al. [F4] for validation purpose. From the comparative plots shown
in Fig. 7.3 it can be stated that for the case of lower drop height, this model can
capture the results reasonably well.
- 190 -

20 25
(a) S1616 (b) S1322
Experiment 20
15 Experiment
Energy method
Energy method
15
10
10
5 5
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.2 0.4 0.6 0.8 1 1.2 1.4
heights (a) S1616; (b) S1322
7.3 Single-Degree-of-Freedom (SDOF) Model

Single-degree-of-freedom system is defined as one in which only one type of
motion is possible, or in other words the position of the system at any instant can be
defined in terms of a single coordinate. Here, the mass of the beam assumed to be
lumped at the midspan. In this study, the impactor mass (300 kg) is much greater
than the effective mass of the beam. Therefore, it is assumed that the beam and the
impactor will move together after impact. Table 7.1 shows the actual and effective
mass of beams from various series considered in drop-weight impact test program.
The natural circular frequencies are computed to be 968.4 and 686.3 rad/s for
beams of 3.8 and 5.7 series, respectively. The idealized SDOF system of beam
subjected to drop-weight impact loading is shown in Fig. 7.4. It is noted that
damping is not accounted in this SDOF system. The initial velocity of the SDOF
system is
V0  miV me  mi  (7-12)
where, V  2 gh ; h is the drop height of the impactor, g is the gravitational
acceleration, mi is mass of the impactor, me is effective mass of beam.
- 191 -
Table 7.1: Actual and effective mass of beams
Effective mass, me (kg)

Actual mass, mb
Beam series
(kg) Elastic stage, Plastic stage,
0.5mb 0.33mb
3.8 series 147.5 73.8 48.7
5.7 series 78.4 38.2 25.9
me+mi
mi mi
V0
h me
k k
Fig. 7.4: Idealized SDOF model
7.3.1 Elastic-plastic analysis
It is well known that elastic analysis of SDOF system (i.e. the resistance function
has been a straight line with slope k and without any upper limit) cannot predict the
midspan deflection history of beam under impact loading. Therefore, elastic-plastic
SDOF system is adopted to reasonably predict the impact response. The assumed
bilinear resistance functions of beam with and without strain rate effect are
illustrated in Fig.7.5.
- 192 -
Static
With strain rate effect
R
ue_sr u
m_sr
R
m_sr
Rm_st u
m_st
u
e_st
k
1
Fig. 7.5: Idealized resistance function of beam with and without strain rate
The equations of motion are as follows:
The elastic response equation can be written as follows:
( me  mi )u  ku  0 ; 0  u  ue (7-13)
The plastic response equation can be written as follows
( me  mi )u  Rm  0 ; ue  u  u m (7-14)
( me  mi )u  Rm  k( um  u )  0 ; ( um  2ue )  u  um (7-15)
where, Rm is the resistance force at yielding of the beam, ue is the elastic limit
displacement and um is the maximum deflection. In Fig. 7.5, the subscripts „st‟ and
„sr‟ of R and u are related to static and strain rate effect cases, respectively.
Using the assumed shape functions for the elastic ( i  sin(ix / L) and the plastic
stage ( i  (2x / L) , the equivalent beam mass ( me ) is as follows

L
Elastic stage, me  2 2 ( mb / L )i2 dx
0
L
1
= 2 ( mb / L )(sin( ix / L ))2 dx 
2 mb (7-16)
0 2
L
Plastic stage, me  2 ( mb / L )i2 dx
2
0
- 193 -
L
1
 2 2 ( mb / L )( 2 x / L )2 dx  mb (7-17)
0 3
To obtain the strain rate dependent resistance function (as shown in Fig. 7.5), strain
rate dependent constitutive properties of concrete and reinforcing steel need to be
incorporated in the sectional analysis. This is generally known that strain rates are
in the rage of 1 to 10 for low velocity impact. The dynamic increase factor (DIFc/t
and DIFs) and strain rate relationships of concrete and reinforcing steel are well
documented in the literature.
The most comprehensive model for strain rate enhancement of concrete both in
tension and compression is presented by the CEB model code [C5]. Strain rate
effect on compression and tension is typically reported as dynamic increase factor
(DIFc/t) - i.e. ratio of dynamic to static strength. In compression CEB model
equations come out to be properly fit with the available data. The DIF c for
compressive strength is given by:
 ( s )1.026 s   30s 1

DIFc   (7-18)
 s ( s ) s  30s 1
1/ 3
where  is the strain rate in the range of 30  10-6 to 300 s-1; s  3  10-6 s-1 (static
strain rate); log  s  6.156 s  2 ;  s  1 ( 5  9 f cs f co ) ; f co  10 MPa; f cs is the

static compressive strength of concrete.
Moreover, the DIFc for ultimate strain in compression is as follows:
DIFc   s 

0.02
(7-19)
where  is the strain rate and s  3  10-6 s-1.

Finally, the DIFt for tension is furnished as:
( s )1.016 s   30s 1

DIFt   (7-20)
  s ( s ) s  30s 1
1/ 3
- 194 -
where  is the strain rate in the range of 3x10-6 to 300 s-1; s  3x10-6 s-1 (static
strain rate); log  s  7.11  2.33 ;  s  1 ( 10  6 f cs f co ) ; f co  10MPa ; f cs is

the static compressive strength of concrete.
However, it was found by Malver and Ross [M5] that the available data in literature
and their additional new data on concrete in tension differed somewhat from the
CEB recommendations, mostly for strain rates beyond 1 s-1. Thus, they modified
the DIFt equations and the change in slope occurs at strain rate of 1 s-1 instead of 30
s-1. The proposed formulations then becomes
 ( s )   1s 1
DIFt   (7-21)
 ( s ) s  1s 1
1/ 3
where  is the strain rate in the range of 10-6 to 160 s-1; s  10-6 s-1 (static strain
rate); log   6  2 ;   1 ( 1  8 f cs f co ) ; f co  10MPa ; f cs is the static

compressive strength of concrete.
Several studies have been documented on the effect of strain rate on reinforcing
bars, structural steel and steel wires (Keenan and Feldman [K2]; Wakabayashi et al
[W1]; Soroushian and Choi [S8]). A detailed review of the available work has been
conducted by Fu et al. [F3] and Malvar [M2]. According to Wakabayashi et al.
[W1], yield stress of steel bar increases with increasing strain rate but the behavior
in the strain hardening region is not affected largely by a strain rate. Soroushian and
Choi [S8] concluded that the yield strength of steel is more strain-rate sensitive than
the ultimate strength. The modulus of elasticity is independent of rate of straining.
According to them, the most important factor influencing the strain rate effects is
the static yield strength. The mechanical properties of steel with lower yield
strength are more strain rate-sensitive than high strength steel. Malvar [M2]
narrated that the DIFs of yield and ultimate stress is inversely related to the yield
stress itself. A formulation was proposed to find out the DIFs as a function of strain
rate and yield stress by fitting the available data in literature. This formulation is
valid for yield stresses in between 290 and 710 MPa and for strain rates in between
10-4 s-1 and 10 s-1. The formulation which gives the DIFs for both yield and ultimate
- 195 -
stress is as follows:
DIFs  ( / 10 4 ) (7-22)
   fu ;  fu  0.019  0.009 f y 414 ;  is the strain rate in s-1 and f y is static

yield strength of reinforcement in MPa.
Strain-rate dependent material properties are succinctly presented above. Here, in

tri-linear moment-curvature analysis, modified material properties are employed to
obtain strain-rate dependent resistance function. The formulas proposed by CEB
model code [C5] are used for compressive strength and ultimate compressive strain
of concrete. For the DIFt of tensile strength of concrete, the formula suggested by
Malver and Ross [M5] is employed here. Furthermore, for yield strength of
reinforcing bars, the equation narrated by Malvar [M2] has been used. The
modified material properties are shown in Table 7.2 for three considered strain
rates (e.g., 1, 5 and 10 s-1). The resistance functions for static and three strain rates
(e.g., 1, 5 and 10 s-1) are shown in Fig. 7.6 for 3.8 and 5.7 series, respectively.
Here, in the elasto-plastic SDOF analysis, resistance function for strain rate 1 s-1
will be used to evaluate the midspan deflection history as almost similar responses
were obtained for these three strain-rate based resistance functions. Successively,
comparison of the midspan deflection history will be made with the results by using
static resistance function to assess the effect of strain rates on impact response.
Table 7.2: Modified material properties
Ultimate Tensile Yield strength

Strain rate Compressive
compressive strength of of
strength of
(s-1) strain of concrete longitudinal
concrete (MPa)
concrete (MPa) steel (MPa)
Static 40.0 0.00350 3.16 520.0

1 54.8 0.00430 4.80 644.8
5 57.2 0.00445 8.18 670.8
10 58.0 0.00452 10.30 681.2
- 196 -
(a) 120
100 3.8 series
80
Load (kN)
60
40 Static
Strain rate-1
Strain rate- 5
20 Strain rate- 10
0
0 5 10 15 20 25 30 35
(b) 80
70 5.7 series
60
Load (kN)
50
40
30 Static
20 Strain rate-1
Strain rate-5
10 Strain rate-10
0
0 5 10 15 20 25 30 35
Fig. 7.6: Resistance function of beam for static and various strain rate cases: (a) 3.8
series; (b) 5.7 series
Hence, to solve the dynamic responses, central difference method (i.e. explicit
method) is adopted, which is suitable for impact problems. Fig. 7.7 shows the
comparison of the midspan displacement vs. time histories between experimental
and elastic-plastic SDOF analysis results with and without strain rates. This is
observed that the SDOF model can capture up to maximum midspan displacement
reasonably well which occurs almost within 30ms. It is observed that the assumed
strain rate (i.e. 1 s-1) dependent SDOF model underestimate the maximum midspan
displacement for some cases which could be unsafe in design. On the other hand,
prediction by SDOF without strain rate always in the safe side. Therefore, strain
rate dependent SDOF model must be used with cautious in design.
- 197 -
(a) Elastic-Plastic SDOF w/o strain rate Elastic-Plastic SDOF w/o strain rate
Elastic-Plastic SDOF w/ strain rate Elastic-Plastic SDOF w/ strain rate
40 60
Midspan displacement (mm)

35 DR3.8_0.8_0.15 DR3.8_0.8_0.15
50
40
25
20 30
15
20
10
10
5
0 0
0 0.005 0.01 0.015 0.02 0.025 0 0.005 0.01 0.015 0.02 0.025
Time (s) Time (s)
Experiment
Elastic-Plastic SDOF w/o strain rate
Elastic-Plastic SDOF w/ strain rate
70
60 DR3.8_0.8_0.15
Drop height - 1.2 m
50
40
30
20
10
0
0 0.005 0.01 0.015 0.02 0.025
Time (s)
(b) Elastic-Plastic SDOF w/o strain rate Elastic-Plastic SDOF w/o strain rate
Elastic-Plastic SDOF w/ strain rate Elastic-Plastic SDOF w/ strain rate
40 50
35 DR5.7_1.6_0.20 DR5.7_1.6_0.20
30
25 30
20
15 20
10
10
5
0 0
0 0.005 0.01 0.015 0.02 0.025 0 0.005 0.01 0.015 0.02 0.025
Time (s) Time (s)
Fig. 7.7: Comparison of midspan displacement vs. time histories between

experimental and elastic-plastic SDOF analysis results: (a) 3.8 series; (b) 5.7 series
(continued)
- 198 -
Experiment
60

50 DR5.7_1.6_0.20
Drop height - 0.6 m
40
30
20
10
0
0 0.005 0.01 0.015 0.02 0.025
Time (s)

experimental and elastic-plastic SDOF analysis results: (a) 3.8 series; (b) 5.7 series
Moreover, to examine the usefulness of SDOF model, the analytical results are
compared with other researcher‟s test results. Two types of beams were taken from
Fujikake et al. [F4] for validation purpose. From the comparative plots shown in
Fig. 7.8 it can be demonstrated that the model can capture the test displacement-
history quite satisfactorily.
(a) Elastic-Plastic SDOF w/o strain rate
20 20
S1616 S1616
15 Drop height - 0.15 m 15 Drop height - 0.30 m
10 10
5 5
0 0
0 0.005 0.01 0.015 0.02 0 0.005 0.01 0.015 0.02
Time (s) Time (s)

experimental and elastic-plastic SDOF analysis results: (a) S1616; (b) S1322
(continued)
- 199 -
Experiment
40

S1616
20
10
0
0 0.005 0.01 0.015 0.02
Time (s)
(b) Elastic-Plastic SDOF w/o strain rate
20 30
S1322 S1322
15 Drop height - 0.30 m 22.5 Drop height - 0.60 m
10 15
5 7.5
0 0
0 0.005 0.01 0.015 0 0.005 0.01 0.015
Time (s) Time (s)
Experiment
40
S1322
20
10
0
0 0.005 0.01 0.015 0.02
Time (s)

experimental and elastic-plastic SDOF analysis results: (a) S1616; (b) S1322
- 200 -
7.4 Summary
Simplified analytical models such as energy-balance and SDOF are employed in

this chapter to capture the maximum impact responses of beam. For the case of
lower drop heights, energy-balance model can capture the results quite accurately;
however for the cases of intermediate to higher drop heights, the degree of accuracy
decreases. The reason behind that in the theory it is assumed that the strain energy
stored in the impacted structure equals the kinetic energy of the impactor at the
instance of time when the maximum impact load occurs. This assumption is solely
valid for quasi-static impact (i.e. low velocity or low drop heights) events as the
amount of energy lost due to large plastic deformations and generation of heat and
noise in high-speed impacts (i.e. higher drop heights) would significantly alter the
energy balance.
Furthermore, elastic-plastic analysis of SDOF system of the beam is proposed to

predict the midspan deflection-time history. Bilinear resistance functions of beam
with and without strain rate effect are derived from tri-linear moment-curvature
analysis. For a given drop-height and beam configuration, this is observed that the
SDOF model can capture up to maximum midspan displacement with reasonable
accuracy, which occurs almost within 30 ms. This is noticed that the assumed strain
rate (i.e. 1 s-1) dependent SDOF model underestimate the maximum midspan
displacement for some cases which could be unsafe in design. On the other hand,
prediction by SDOF without strain rate is always in the safe side. Therefore, strain
rate dependent SDOF model must be used with caution in design.
- 201 -
- 202 -
Chapter 8: Performance of Impact-damaged RC Beams
CHAPTER 8
PERFORMANCE OF IMPACT-DAMAGED RC BEAMS
8.1 Introduction
RC structures have been the main structural type of choice in new infrastructural
projects across the world. Due to the wide distribution and prevalence of such
structures, it is almost inevitable that RC structures may be exposed to various
types of impact loading during their service life. Progressive collapse could be
triggered in these systems due to some impact loading thrusts in the structural
elements. Therefore, it is of utmost importance to evaluate the residual resistance of
structural components of an impact-damaged structure. A detailed understanding of
the residual properties of the structural components would be helpful in
determining the integrity and stability of impact-damaged structures during search
and rescue operation. Moreover, to ensure that damaged structures will not fail
catastrophically during their service life and maintain structural efficiency, it is
essential to study the residual properties of the structures after impact. This can be
further extended to evaluating proper strengthening or rehabilitation scheme.
Assessment of the behavior of RC beam under drop-weight impact (Hughes and
Beeby [H3]; Banthia [B1]; Kishi et al. [K5-K8]; Chen and May [C3]; Fujikake et
al. [F4]; Bhatti et al. [B8]; Saatci and Vecchio [S1]; Tachibana et al. [T1]; Kishi
and Mikami [K10]) loading has been well documented in literature and research on
residual strength of blast-damaged RC columns (Bao and Li [B3], Wu et al. [W3])
has been prolific. However, evaluation of the residual response of impact-damaged
beams is a lacking area. Thus, quasi-static response of the undamaged and impact-
damaged specimens has been investigated. First, the beams were subjected to free-
falling impact loadings from various drop-heights. Subsequently the pre-damaged
specimens were tested quasi-statically to determine the load-midspan deflection
responses. Finally, one beam from each group without any prior damage was tested
under monotonic static loading to evaluate their original load-midspan deflection
behavior. Comparative load vs. mid-span deflection profiles of undamaged and
impact-damaged beams were plotted and residual resistance index (RRI) and
- 203 -
residual stiffness index (RSI) was computed. RRI is defined as the ratio of the
ultimate resistance of undamaged and impact-damaged specimens; similarly RSI is
the normalization of the secant stiffness of undamaged and impact-damaged beams.
Here, the secant stiffness was computed by joining the origin to the peak load in the
load vs. mid-span deformation curves brittle failure-type beams; whereas the yield
resistance was taken as reference point to calculate the secant stiffness for
specimens having ductility.
To expand the knowledge beyond the range of parameters investigated

experimentally, a methodology to numerically simulate the drop-weight impact
tests and post impact quasi-static residual resistance tests of damaged beams was
developed. To simulate the impact response of beams, explicit finite element (FE)
code LS-DYNA [L1] was used herein by considering the same boundary conditions
and the exact shape of the impactor which were used during experiment. Afterward,
restart analysis was performed by replacing the impactor with the plate to perform
quasi-static simulation. The resulting deformation and damages of beam from the
impact stages have been utilized to carry out residual resistance simulation. Results
of the explicit FE simulations were compared against impact and post impact quasi-
static residual resistance tests results. Simulation results seem to match the
experimental results with reasonable accuracy. Hence, this numerical procedure
could be valuable in providing some insight about the behavior of impact-damaged
specimens before performing the actual tests in laboratory. Moreover, newly
designed structural elements could be analyzed prior to their actual construction in
site and existing impact-damaged structural elements could be retrofitted
effectively by calculating their residual resistance following the above-mentioned
numerical scheme.
8.2 Outline of Experiment
As mentioned earlier in Chapter 5, drop weight impact test program consisted of

30 RC beams which were divided into two groups in terms of their shear span-to-
effective depth ratios ( a d -3.3, 3.8 and 5.7). Each group had three types of
specimens distinguished by their shear reinforcement ratios. After finishing the
- 204 -
impact test of each beam, the damaged specimens were placed in the 3-point
bending test set up at NDA, Japan to give quasi-static loading for capturing their
residual resistance vs. midspan deflections curves. Details of the post impact
residual resistance test set-up are shown in Fig. 8.1.
Fig. 8.1: Post impact residual resistance test set up
8.3 Summary of Experimental Results

In this section, load vs. midspan deflection curves of undamaged and impact-
damaged beams are presented first, followed by the variation of residual resistance
index (RRI) and residual stiffness index (RSI) in terms of various impact energies.
8.3.1 Load vs. midspan deflection curves
Quasi-static load vs. mid-span characteristics of undamaged and impact damaged

specimens (damaged by impactor from different drop-heights) are shown in Fig.
8.2, 8.3 and 8.4. Fig. 8.2 shows the curves of RC beams having shear span to
effective depth ratio 3.3. It was observed that residual resistance of DR3.3_2.4
(without transverse reinforcements) almost reached the same resistance as
undamaged specimens with same stiffness when the impactor struck the specimen
from a drop-height of 0.15 m. However, for 0.3 and 0.6 m drop-height, both
resistance and stiffness degradation was observed notably. Similarly, for
DR3.3_2.4_0.12, drop-heights ranging from 0.6 to 1.2 m caused significant
strength and stiffness reduction. For the same series of specimen with higher
- 205 -
amount of transverse reinforcements (i.e. DR3.3_2.4_0.56), resistance and stiffness

degradation was less than that of above-mentioned two beams. It is noted that the
static response of undamaged beams of this series was not evaluated experimentally
rather the static response was determined by FE simulation. For 3.8 and 5.7 series
beams, specimens with greater transverse reinforcements experienced less strength
degradation compared to the other specimens.
160
140 Undamaged
0.15 m
120 0.3 m (Impact-
0.6 m -damaged)
Load (kN)
100
80
60
40
20 DR3.3_2.4
0
0 5 10 15 20 25 30 35 40
240
Undamaged
200 0.3 m
0.6 m (Impact-
160 0.9 m -damaged)
Load (kN)
1.2 m
120
80
40
DR3.3_2.4_0.12
0
0 10 20 30 40 50
impact-damaged beams of 3.3 series (continued)
- 206 -
350
300 Undamaged
0.6 m
250 0.9 m (Impact-
1.2 m -damaged)
Load (kN)
200 1.6 m
150
100
50 DR3.3_2.4_0.56
0
0 20 40 60 80 100 120
impact-damaged beams of 3.3 series
160
140 Undamaged
0.3 m
120 0.6 m (Impact-
0.9 m -damaged)
Load (kN)
100
80
60
40
20 SR3.8_0.8
0
0 10 20 30 40 50
160
140 Undamaged
0.6 m
120 0.9 m (Impact-
1.2 m -damaged)
Load (kN)
100
80
60
40
20 DR3.8_0.8_0.11
0
0 5 10 15 20 25 30 35 40
- 207 -
160
140 Undamaged
0.6 m
120 0.9 m (Impact-
1.2 m -damaged)
Load (kN)
100
80
60
40
20 DR3.8_0.8_0.15
0
0 10 20 30 40 50
100
Undamaged
80 0.3 m
0.45 m (Impact-
0.6 m -damaged)
Load (kN)
60
40
20
SR5.7_1.6
0
0 10 20 30 40 50 60
100
Undamaged
80 0.3 m
0.45 m (Impact-
0.6 m -damaged)
Load (kN)
60
40
20
DR5.7_1.6_0.15
0
0 5 10 15 20 25 30 35 40
- 208 -
100
Undamaged
80 0.3 m
0.45 m (Impact-
0.6 m -damaged)
Load (kN)
60
40
20
DR5.7_1.6_0.20
0
0 10 20 30 40 50
8.3.2 Residual resistance index (RRI)
RRI is defined as the ratio of the peak resistance of impact-damaged to undamaged

specimens ( RRI  Rdamaged Rundamaged ). Peak resistance can be defined as the
maximum load in load vs. midspan deflection curves of undamaged and impact-
damaged beams. Fig. 8.5 shows the variation of RRI under different impact
energies. For DR3.3_2.4 (without transverse reinforcements), when the input
impact energies was increased from 0.44 to 1.77 kJ, the reduction in ultimate
residual resistance (URR) was 4 and 46% respectively. URR is the maximum
resistance of impact-damaged beams. When transverse reinforcement was around
0.12%, it can absorb more impact energy and the reduction in URR was lesser as
compared to above-mentioned specimens. With a further increment of transverse
reinforcement to 0.56%, URR was reduced 5 to 24% even when the input impact
energy increased from 1.77 to 4.71 kJ. Moreover, the URR of singly reinforced
beam for a d -3.8 series specimens did not only reduce at all but also greater than
the ultimate resistance of undamaged specimens. This is due to the change in the
nature of failure mode where the undamaged specimen failed in shear whereas
damaged one failed in flexure (i.e. strain hardening phenomenon was significant in
impact-damaged beams) under the impact energies (i.e. 0.88 to 2.65 kJ) considered
during testing. For doubly reinforced beams of the same specimen series, under low
- 209 -
impact energies (i.e. 1.77 and 2.65 kJ) their RRIs were almost similar; however for
higher input energy (i.e. 3.53 kJ), specimens having higher amount of transverse
reinforcements performed better. For a d -5.7 series specimens, a systematic trend
of degradation in RRI was observed; and with increasing impact energies, better
performance was observed in specimens with more transverse reinforcements. In
general, with the increasing amount of impact energy, decreasing trend in RRI is
being observed. This trend is more pronounced in specimens with less transverse
reinforcement. Few cases were observed where the URR of damaged specimens
was more than the undamaged one; this may be due to the strain hardening effect of
the steel reinforcing bars.
1.4 1.6
Residual Resistance Index (RRI)
(a) DR3.3_2.4 (b) SR3.8_0.8

DR3.3_2.4_0.12 1.4
1.2 DR3.8_0.8_0.11
DR3.3_2.4_0.56 DR3.8_0.8_0.15
1.2
1
1
0.8
0.8
0.6 0.6
0.4 0.4
0 1 2 3 4 5 6 0.5 1 1.5 2 2.5 3 3.5 4 4.5
1.4
(c) SR5.7_1.6
1.2 DR5.7_1.6_0.15
DR5.7_1.6_0.20
1
0.8
0.6
0.4
0.2
0.5 1 1.5 2
Impact energy (kJ)
Fig. 8.5: Residual Resistance Index (RRI) of RC beams (a) 3.3 series; (b) 3.8 series
and (c) 5.7 series under various impact energies
RRI could be used effectively to define the extents of damage of RC beam after
impact loading. The damage index D is defined as:
- 210 -
D  1  RRI (9-1)
The degrees of damage are defined as follows:
D  0  0.2 Low damage (LD); D  0.2  0.5 Medium damage (MD);

D  0.5  0.8 High damage (HD); D  0.8  1 Severe damage (SD)
Shi et al. (2008) proposed almost similar type of damage index for RC columns
damaged to blast loads and also mentioned that this definition is solely subjective.
Thus, here the damage term for D  0.8  1 has been changed from „collapse‟ to
„severe damage‟ as some researchers believe the shear failure mode as collapse
stage. Hence, the degree of damage of the beam for particular impact energy can be
calculated from Eq. (9-1). When RRI >1, it could be concluded that the beam did
not acquire any substantial damage or the developed hairline cracks might be closed
as soon as the impact load died off.
8.3.3 Residual stiffness index (RSI)
RSI is defined as the ratio of the secant stiffness of impact-damaged to undamaged

specimens ( RSI  K damaged K undamaged ). Here, the secant stiffness was computed
by joining the origin to the peak load in the load vs. mid-span deformation curves
brittle failure-type beams (i.e. sharp fall in load vs. deformation curves after peak
load); whereas the yield resistance was taken as reference point to calculate the
same for specimens having ductility. The variation of residual stiffness index (RSI)
of beams under different impact energies is shown in Fig. 8.6. For DR3.3_2.4
(without transverse reinforcements), when the input impact energies increased from
0.44 to 1.77 kJ, the reduction in stiffness was 0 and 81% respectively. However,
around 70-80% reduction in stiffness was observed for higher impact energies
around 2.65 and 3.53 kJ respectively for specimens having 0.12% transverse
reinforcements. The rate of reduction in stiffness for DR3.3_2.4_0.56 was much
less drastic with the increasing impact energies (i.e. 1.77 to 4.71 kJ) as compared to
the specimens of the same series. Thus, it is evident that transverse reinforcements
inhibited significant degradation in stiffness. Around 20% stiffness reduction was
perceived of SR3.8_0.8 for all input energies (0.88 to 2.65kJ) considered during
- 211 -
experiment. However, for the case of impact energy of 2.65 kJ, the stiffness
reduction was insignificant for doubly-reinforced sections of the same series.
However, for a d -5.7 series specimens, consistent stiffness-degradation was seen
for increasing drop-heights. For the same impact energy, stiffness-degradation was
less in DR beams as compared to SR of 5.7 series beams.
1.6 1.8
Residual Stiffness Index (RSI)

DR3.3_2.4 SR3.8_0.8
1.4 (a) DR3.3_2.4_0.12 1.6
(b)
DR3.8_0.8_0.11
1.2 DR3.3_2.4_0.56 DR3.8_0.8_0.15
1.4
1
0.8 1.2
0.6
1
0.4
0.8
0.2
0 0.6
0 1 2 3 4 5 6 0.5 1 1.5 2 2.5 3 3.5 4 4.5
1.4
1.2 (c) SR5.7_1.6

DR5.7_1.6_0.15
1 DR5.7_1.6_0.20
0.8
0.6
0.4
0.2
0
0.5 1 1.5 2 2.5
Impact energy (kJ)
Fig. 8.6: Residual Stiffness Index (RSI) of RC beams (a) 3.3 series; (b) 3.8 series
and (c) 5.7 series under various impact energies
8.3.4 Failure pattern
Assessment of crack pattern or failure mode is of paramount importance in both

impact and post impact quasi-static loading stages, as it would assist in determining
the integrity and stability of impact-damaged structures. Fig. 8.7 shows the crack
profile of beams after post impact residual resistance test. For DR3.3_2.4, impact
energies were varied from 0.59 to 4.7 kJ (i.e. drop-height varies from 0.15 to 1.2
m). No visible cracks were observed at impact energy of 0.59 kJ (i.e. 0.15 m);
- 212 -
however, diagonal shear cracks were perceived in post impact quasi-static test.
With further increment of impact energies, diagonal cracks were observed and
successively in residual test these cracks widened with much damaged in the
impact regions. At 1.2 m (i.e. impact energy of 4.7 kJ) drop height, the beam was
so severely damaged that post impact residual test could not be performed. For
DR3.3_2.4_0.12, hair-line diagonal cracks were seen under low velocity impact
whereas much wider cracks were discerned for high impact velocity. In residual
tests, further development of diagonal cracks and widening of the existing diagonal
cracks with concrete spalling were noticed. Impact energies were varied from 2.35
kJ to 6.28 kJ (i.e. 0.6 to 1.6 m) for drop-weight impact test of DR3.3_2.4_0.56.
This beam showcases the significance of higher amount of transverse
reinforcements by not only absorbing the higher amount of impact energies (i.e.
approx. 34% higher than specimen having 0.12% shear reinforcements) but also
performing better (i.e. by resisting catastrophic failure mode) in residual resistance
test. However, some spalling in side cover concrete nearby impact region was
observed under impact energy of 6.28 kJ (i.e. 1.6 m). Flexure failure was observed
in all beams of a d -3.8 series for all impact energies (i.e. 1.18-4.7 kJ) considered
during impact tests. Ductile nature (i.e. no change in failure mode) persisted during
residual test of the damaged beams however some additional flexural cracks
developed or the existing flexural cracks became wider. Thus, beam having a d of
3.8 and lower amount of longitudinal reinforcements (although less amount of
shear reinforcements) could withstand the impact and post impact quasi-static load
quite satisfactorily. Similarly, all beams of a d -5.7 series failed in flexure under
impact (i.e. considered impact energies of 1.18-2.35 kJ) and residual tests.
- 213 -
0.15m 0.30m 0.60m
0.30m 0.60m 0.90m
0.60m 0.90m 1.2m
1.2m 1.6m
DR3.3_2.4 DR3.3_2.4_0.12 DR3.3_2.4_0.56
(a)
0.30m 0.60m 0.60m
0.60m 0.90m 0.90m
0.90m 1.20m
1.20m
SR3.8_0.8 DR3.8_0.8_0.11 DR3.8_0.8_0.15
(b)
0.30m 0.30m 0.30m
0.45m 0.45m 0.45m
0.60m 0.60m
0.60m
SR5.7_1.6 DR5.7_1.6_0.15 DR5.7_1.6_0.20
(c)
Fig. 8.7: Crack pattern of impact-damaged RC beams (a) 3.3 series; (b) 3.8 series
and (c) 5.7 series under various drop-heights
- 214 -
8.4 Finite Element Framework
Detailed description of the simulation of beam subjected to drop-weight impact

loading can be found in Chapter 6. After finishing the impact loading simulation,
the impactor and two triangular plates were removed from the model. Then, rigid
loading plate was placed at the mid-span of the beam to initiate residual resistance
simulation through restart analysis. The rigid loading plate was allowed to move
only in the vertical direction and displacement-controlled loading was given to it.
Fig. 8.8 shows the FE model in impact loading and post-impact residual resistance
stage. Here, in the restart analysis environments, resulting deformation and
damages of beam from the impact simulation have been utilized to carry out
residual resistance simulation.
(a)
(b)
Fig. 8.8: FE models of beam (a) impact loading stage; (b) post impact residual
resistance stage
- 215 -
Similar to the determination of impact force history, residual resistance due to the
prescribed displacement was then achieved by monitoring the contact forces at the
concrete nodes in contact with the support solid cylinders. MAT CSCM
CONCRETE (MAT 159) is usually recognized as Continuous Surface Cap Model,
was used to model concrete and material model MAT PIECEWISE LINEAR
PLASTICITY (MAT 024) is used to model reinforcing bars. Furthermore, to model
the impactor, support rollers and triangular plates MAT_RIGID (MAT 020) was
used from LS-DYNA [L1] material library.
8.5 Verification of Finite Element Analysis Results
The numerical simulation results of RC beams subjected to drop-weight impact

loading are validated with the experimental outcomes. Successively, the numerical
results of impact-damaged beam are validated to the test results. Verification of
impact loading history under impact and residual resistance vs. mid-span deflection
profile of impact damaged specimens and cracks patterns is presented in the
following two sections.
8.5.1 Impact loading history and residual resistance vs. midspan deflection
Two types of tested beams are considered for validation of numerical results.
Comparison of impact load history and post impact residual resistance vs. midspan
deflection of DR3.8_0.8_0.11 under 0.6 m drop height is presented in Figs. 8.9 (a)
and (b) respectively. Similarly, above-mentioned two responses are validated for
DR5.7_1.6_0.20 under 0.45 m drop height through Figs. 8.9 (c) and (d). It is
observed that the results from the three-dimensional nonlinear FE model match the
test results within an acceptable accuracy.
- 216 -
400
300
(a) DR3.8_0.8_0.11
Drop height - 0.6 m
Impact load (kN)

200
100
-100 Experiment
FEM
-200
0 0.005 0.01 0.015 0.02 0.025
Time (s)
100
(b)
Residual resistance (kN)
80
60
40
20
Experiment
FEM
0
0 5 10 15 20
250
200 (c) DR5.7_1.6_0.20

Drop height - 0.45 m
Impact load (kN)
150
100
50
0
Experiment
-50
FEM
-100
0 0.005 0.01 0.015 0.02
Time (s)
midspan deflection (continued)
- 217 -
50
(d)
Residual resistance (kN)

40
30
20
10 Experiment
FEM
0
0 5 10 15 20 25
midspan deflection
8.5.2 Crack profiles of beams after impact and post impact quasi-static
loading
Assessment of crack patterns or failure mode of beam under drop-weight impact

and post impact quasi-static loading bears utmost importance. The damage of the
beams obtained from numerical simulation is shown by plotting the fringes of
effective plastic strain. These effective plastic strain contours reveal the strain
localization where failure propagates.
Impact test: drop-height-0.60m
plastic strain from simulation of beams after impact and residual test (continued)
- 218 -
Residual test
(a) DR3.8_0.80_0.11
Impact test: drop-height-0.45m
Residual test
(b) DR5.7_1.6_0.20
plastic strain from simulation of beams after impact and residual test
- 219 -
Fig. 8.10 shows the comparison of damage plot in between experimental and
numerical simulation results. Comparative damage plots of two types of beams
mentioned above are shown. From these comparisons, it can be enunciated that the
damage plot of numerical simulation results can capture the experimental crack
profiles (after impact and residual tests) quite satisfactorily.
8.6 Numerical Simulation Case-studies
After verification of the FE model against the experimental results, this section
presents a parametric investigation to elucidate about the residual resistance of
beams after impact damage. Some key parameters such as mass ratio (  ) i.e. ratio
of the mass of the beams to the impactor-mass (  : 0.26, 0.42 and 0.49),
longitudinal reinforcement ratios (  L : 0.8-2.4%), transverse reinforcement ratios
(  T : 0.12-0.67%), compressive strength of concrete ( f c : 30, 40 and 50 MPa) are
taken into account to study their influence on residual performance under various
impact energies ( E : 0.88-4.70 kJ). The parameter variation in various case studies
is shown in Table 8.1. The beams used for numerical case studies have the same
geometric properties (i.e. span length and cross-sections etc.) as experimental
specimens.
Table 8.1: Parameter variations in case-studies
Impact
ad   L (%)  T (%) f c (MPa)
Energy, E (kJ)
3.3 0.42 1.62, 2.40 0.12, 0.19, 0.38, 0.56 2.35, 3.53, 4.70
3.8 0.49 0.80, 1.50 0.11, 0.15, 0.24, 0.35, 0.50 2.35, 3.53, 4.70 30, 40 and 50
5.7 0.26 0.90, 1.60 0.15, 0.20, 0.31, 0.47, 0.67 1.17, 1.77, 2.35
8.6.1 Effect of mass ratio (  )
Fig. 8.11 illustrates the effect of the mass ratios (  : 0.26, 0.42 and 0.49) on the
RRI of beams under varying impact energies. It is noted that here the mass of the
impactor has been kept constant (300 kg); however, different mass ratios were
obtained due to the variation in span length and cross-sections of beams. Here, the
- 220 -
mass of beam of 3.3, 3.8 and 5.7 series are 126, 147.5 and 78.3 kg, respectively.
Considering same amount of longitudinal reinforcement ratio and two amounts of
transverse reinforcement ratio, two cases are presented in Figs. 8.11 (a) and (b). It
is evident that residual resistance will be in the higher side for the case of higher
mass ratio under same impact energy.
(a) 1.2
L-1.6%
1.1 - 0.26
T-0.20% - 0.42
1 - 0.49
0.9
RRI
0.8
0.7
0.6
0.5
1 1.5 2 2.5 3 3.5 4 4.5 5
Impact energy (kJ)
(b) 1.3
1.2 L-1.6% - 0.26
T-0.50% - 0.42
1.1
- 0.49
1
RRI
0.9
0.8
0.7
0.6
0.5
1 1.5 2 2.5 3 3.5 4 4.5 5
Impact energy (kJ)
Fig. 8.11: Effect of mass ratio (  ) on RRI under various impact energies
8.6.2 Effect of longitudinal reinforcement ratio (  L )
Fig. 8.12 shows the variation of RRI under various impact energies. To understand
the influence of longitudinal reinforcement ratio on RRI, two cases are considered
here. Under low impact energies, the effect of  L seems to be insignificant.
However, for high impact energies, specimen having low amount of  L performs
- 221 -
better in terms of achieving residual resistance. Beams of comparatively lower

amounts of longitudinal reinforcements exhibited only overall flexure failure,
whereas flexure as well as local damage at impacting point occurred in beams with
relatively higher amount of longitudinal reinforcements. This local damage initiates
the further development of diagonal cracks during residual tests which successively
reduce the residual resistance; while in the case of overall flexure-failure type
beam, much ductile response was observed due to the strain hardening of the
longitudinal tensile reinforcements.
(a) 1.4
0.8%( )_0.24%( )
L T
1.5%( )_0.24%( )
1.2 L T
0.8%( )_0.50%( )
L T
1.5%( )_0.50%( )
L T
1
RRI
0.8
a/d-3.8
0.6
2 2.5 3 3.5 4 4.5 5

Impact energy (kJ)
(b) 1.6
0.9%( )_0.20%( )
1.4 L T
1.6%(L)_0.20%(T)
1.2 0.9%( )_0.47%( )
L T
1.6%( )_0.47%( )
L T
RRI
0.8
0.6 a/d-5.7
0.4
1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6
Impact energy (kJ)
Fig. 8.12: Effect of longitudinal reinforcement ratio (  L ) on RRI under various

impact energies
- 222 -
8.6.3 Effect of transverse reinforcement ratio (  T )
Effect of transverse reinforcement on RRI under varying amount of impact energies

is presented in Fig. 8.13. It is evident that specimens having relatively higher
amount of transverse reinforcements possess higher residual resistance. Under
increasing height of the drop-weights, the difference in RRI between beam having
low and high amount of shear reinforcements would be significant. Higher amount
of transverse reinforcement contained beam resist the excessive impact damage and
can produce much higher residual resistance by confinement to the core concrete
and supplementing lateral restraint capacity against the buckling of the longitudinal
reinforcements.
(a) 1.2
a/d-3.3 0.12%
1.1
 -1.62% 0.19% T
1 L 0.38%
0.56%
0.9
RRI
0.8
0.7
0.6
0.5
2 2.5 3 3.5 4 4.5 5
Impact energy (kJ)
(b) 1.8
1.6 0.11%
0.15%
a/d-3.8
1.4 0.24% T L-1.5%
0.35%
1.2 0.50%
RRI
1
0.8
0.6
0.4
0.2
2 2.5 3 3.5 4 4.5 5
Impact energy (kJ)
Fig. 8.13: Effect of transverse reinforcement ratio (  T ) on RRI under various

impact energies
- 223 -
8.6.4 Effect of compressive strength of concrete ( f c )
Beam of concrete compressive strength of 30, 40, 50 MPa was used to determine
the effect of f c on RRI under varying impact energies. Three case studies are
depicted in Fig. 8.14. Under low impact energy, the effect of f c on RRI quite
minimal; however for increasing the impact energies, specimens having high
concrete compressive strength perform better (i.e. yielding higher residual
resistance) as compared to beam with low concrete compressive strength. It is noted
that beam having high strength concrete is out of scope of this research.
(a) 1.2
30 MPa a/d-3.3
1.1
40 MPa L-1.62%
1 50 MPa T-0.56%
0.9
RRI
0.8
0.7
0.6
0.5
2 2.5 3 3.5 4 4.5 5
Impact energy (kJ)
(b) 1.1
1.05 30 MPa a/d-3.8
40 MPa L-1.5%
1 50 MPa T-0.5%
0.95
RRI
0.9
0.85
0.8
0.75
2 2.5 3 3.5 4 4.5 5
Impact energy (kJ)

impact energies (continued)
- 224 -
(c) 1.2
1.1 a/d-5.7
L-1.6%
1
T-0.67%
0.9
RRI
0.8
0.7
0.6 30 MPa
40 MPa
0.5
50 MPa
0.4
1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6
Impact energy (kJ)

impact energies
8.7 Summary
In this chapter, experimental and numerical studies of impact-damaged of RC

beams are demonstrated comprehensively and following conclusions can be drawn:
(1) Two indices, residual resistance index (RRI) and residual stiffness index (RSI)
were defined and discussed thoroughly to show their variation under various impact
energies by analyzing the experimental data. RRI and RSI could be used effectively
to delineate the extents of damage (i.e. resistance and stiffness degradation) of the
beams after impact loading.
(2) A methodology to numerically simulate the drop-weight impact tests and post
impact quasi-static residual resistance tests of damaged beams was developed. The
FE model has been successfully validated with the test results.
(3) Upon successful verification of the FE model against the experimental results, a
parametric study has been carried out to quantify the effect of various parameters
on RRI. RRI will be on the higher side in the case of higher mass ratio under same
impact energy. Under high impact energies, specimen having low amount of
longitudinal reinforcement performs better in terms of achieving residual
resistance. Specimens having relatively higher amount of transverse reinforcements
- 225 -
possess higher residual resistance after impact damage. Moreover, under increasing
amount of impact energies, specimens having high concrete compressive strength
perform better (i.e. yielding higher residual resistance) as compared to beam with
low concrete compressive strength.
- 226 -
Chapter 9: Conclusions and Future Work
CHAPTER 9
CONCLUSIONS AND FUTURE WORK
9.1 Conclusions
The dynamic behavior of RC beams under varying rates of concentrated and drop-
weight impact loading was investigated experimentally and numerically.
Conclusions drawn from this research program are presented in the following
section.
RC beam under varying rates of concentrated loading
Experimental investigation:
Conclusions drawn from the experimental investigation on RC beam under varying

rates of concentrated loading are as follows:
(1) With the enhancement of loading rates, the ultimate load resistance, cracking
stiffness and energy absorption of RC beams were found to increase
correspondingly. When the loading rate progressed from low to high, an increasing
trend in DIF was observed. In addition, specimens having smaller a d ratio
produced higher DIF as compared to the specimens having higher a d ratio for
low, medium and high loading rates. Peak strain rate was amplified by one order of
magnitude (approximately, 10 times) as the loading rates progressed from low to
high.
(2) For 3.3 series specimens, diagonal tension and/or shear tension type failure was
observed in SR beams whereas diagonal tension and/or shear compression were the
dominant failure mode for DR specimens. In all loading rates of the above-
mentioned series, diagonal shear cracks originated from the mid-height of beam and
propagated towards loading and support points. However, for DR specimens,
diagonal shear cracks under medium and high loading rates propagated with much
steeper angles towards the bottom surface of the beam as compared to static and
- 227 -
low loading rates. On the other hand, for 4.4 series specimens, ductile flexure type
failure was observed for all loading rates.
(3) After observing the wide scatter of results, it can be concluded that there is a
need of consensus among researchers in high-rate testing of RC beams. Test set up,
frequency response of measurement system, beam geometry, reinforcing detailing
and material properties should be kept similar to facilitate better analysis and
effective comparisons.
Numerical investigation:
FE model was developed to evaluate the behavior of beams under varying rates of
concentrated loading at midspan. Numerical parametric study yielded following
conclusions:
(1) It was observed that DIF were on the higher side for beam with low
reinforcement ratio. When the beam contained low amount of longitudinal
reinforcements, the response was mainly influenced by the strain rate effect of
longitudinal reinforcing steel which in turn causing an enhancement of the DIF.
Specifically, for high loading rates, extent of damage was more severe for beams
containing relatively higher amount of longitudinal reinforcements. Therefore, it
can be concluded that change in failure mode may be observed (i.e. mainly in high
rates considered here) when the amount of longitudinal reinforcement ratio is
increased (i.e. approaching towards balanced-reinforcement ratio) in an under-
reinforced beam.
(2) For beam containing more transverse reinforcements, DIF would be on the
lower side. Furthermore, for medium and high loading rates, the beam containing
lower amount of transverse reinforcements, diagonal cracks were formed with
flexure cracks; whereas flexure type of failure was observed in beam having high
amount of transverse reinforcements. Hence, the amount of transverse
reinforcements had a significant effect in preventing the formation of diagonal
cracks under medium and high loading rates. For beams having various concrete
grades (30, 40 and 50 MPa), increasing trend in peak load was observed when the
- 228 -
loading rate progressed from static to high. Moreover, beam of high strength
(3) The yield strength of longitudinal reinforcements of beams is the key parameter,
affecting the change in failure mode from flexure at static loading to shear at high
rates. Due to the high rate sensitivity of low yield strength steel, the increment of
yield strength of longitudinal reinforcements in medium and high rates was such a
way that the bending resistance surpassed the shear resistance and eventually
failure mode changed.
RC beam under drop-weight impact loading
Experimental assessment:
From the experimental assessment of RC beam under drop-weight impact loading,

following conclusions can be made:
(1) Maximum impact load, maximum midspan deflection and time to reach
maximum deflection were found to increase with the increment of drop-heights or
impact velocities. Maximum strain rates of longitudinal tensile reinforcements vary
between 1 and 7 s-1 for the corresponding considered drop heights of 0.15 to 1.2 m
(i.e. impact velocity varies from 1.72 to 4.85 m/s).
(2) The crack pattern and failure modes varied among the specimens depending on
their static resistance. For statically shear-critical beams, (without transverse
reinforcements) hairline diagonal cracks were observed at lowest drop height
considered. With the enhancement of the drop-heights, width of the diagonal cracks
and severity of local failure increases. At a maximum drop height considered, the
damage was so catastrophic that the beam disintegrated into two components with
extensive crushing and spalling of concrete at the impact region. For statically
flexure-critical beams, flexural cracks beneath the impact point were observed at
lowest drop height considered. With further increase in drop heights, flexural
cracks became much wider and propagated vertically and crushed the compression
concrete. At a maximum drop height considered, massive local failure occurred at
- 229 -
the impact region.
(3) Two empirical equations are proposed by analyzing a pool of data on impact
loading documented by several researchers from the literature for both statically
flexure and shear failure type beams. By specifying the maximum midspan
deflection for each limit state of beam, the required static bending and shear
resistance can be determined for designing the beam subjected to input impact
energy.
Numerical investigation:
After verifying the numerical simulation results against the experimental results,
parametric investigation was carried out to elucidate the influence of some key
parameters on impact responses. Under constant impact energy, impacts caused by
high mass with low velocity result in smaller maximum impact load but higher
maximum midspan deflection of beam and vice versa. Maximum impact load
increases and maximum midspan deflection of beam decreases with the
enhancement of the longitudinal reinforcement ratios. With increasing amount of
longitudinal reinforcements severity in failure pattern (including diagonal and
punching shear cracks) is observed. Moreover, with the enhancement of concrete
compressive strength from 30 to 50 MPa, an increasing trend in maximum impact
load was observed whereas the opposite phenomenon was noticed for the case of
maximum midspan deflections. Maximum impact load is in the higher side and
maximum midspan deflection is in lower side for fixed-end beam as compared to
pinned-end beam.
Analytical investigation:
Simplified analytical models such as energy-balance and SDOF are employed to

capture the maximum impact responses of beam. For the case of lower drop
heights, energy-balance model can capture the results quite accurately; however for
the cases of intermediate to higher drop heights, the degree of accuracy decreases.
Furthermore, elastic-plastic analysis of SDOF model is proposed to predict the
midspan deflection-time history of beam under impact loading.
- 230 -
Residual performance of impact-damaged RC beam
(1) Two indices, residual resistance index (RRI) and residual stiffness index (RSI)
were defined and discussed thoroughly to show their variation under various impact
energies by analyzing the experimental data. RRI and RSI could be used effectively
to delineate the extents of damage (i.e. resistance and stiffness degradation) of the
beams after impact loading.
(2) A methodology to numerically simulate the drop-weight impact tests and post
impact quasi-static residual resistance tests of damaged beams was developed.
Upon successful verification of the FE model against the experimental results, a
parametric study has been carried out to quantify the effect of various parameters
on RRI. RRI will be on the higher side for the case of higher mass ratio under same
impact energy. Under high impact energies, specimen having low amount of
longitudinal reinforcement performs better in terms of achieving residual
resistance. Specimens having relatively higher amount of transverse reinforcements
possess higher residual resistance after impact damage. Moreover, under increasing
amount of impact energies, specimens having high concrete compressive strength
perform better (i.e. yielding higher residual resistance) as compared to beam with
low concrete compressive strength.
In summary, this study has explored various engineering methodology to assess the
performance of RC beams under varying rates of concentrated and impact loadings.
Useful information has been provided in these two loading environments through
experimental tests, FE analyses and analytical approaches. Substantial amount of
test data are provided in this thesis that can be used further for the development of
new analytical methods. Some points are drawn regarding the testing of beams
under varying loading rates which would be useful during the planning of future
test program. Furthermore, this study could be employed for the development of
future model code and guidelines regarding low velocity impact effect on the
performance assessment of RC structural components.
- 231 -
9.2 Future work
In the light of experience gained from this research program, the following
recommendations can be made for future study:
(1) An investigation should be carried out on the structural behavior of RC beam

under varying rates of concentrated loading, in which the following variables need
to be considered:
 Concrete compressive strength and yield strength of reinforcement bars;

 Size and geometry of beam;
 Boundary conditions;
 Asymmetric loading;
(2) Influence of loading history (mainly the loading rates corresponds to impact
regime) on structural behavior should be explored.
(3) Efforts can be driven towards determining the impact resistance of RC beam
through a drop-weight test program, covering a wider range of parameters.
(4) The effects of various impact parameters, such as mass and velocity of
impactor, impactor shape and impact interface should be investigated. Moreover,
the influence of structural mass and geometry should be evaluated by testing RC
beams with varying span lengths.
(5) Behavior of other structural components such as RC columns, slabs under drop-
weight impact loading should be examined in details. Moreover, impact response of
prestressed structural components can be investigated.
- 232 -
References
REFERENCES
[A1] Abbas, A.A., Pullen A.D., and Cotsovos, D. M. “Structural response of RC

wide beams under low-rate and impact loading.” Magazine of Concrete Research,
62(10), 723-740, 2010.
[A2] ACI Committee 318 “Building Code Requirements for Structural Concrete
(ACI 318-08) and Commentary.” American Concrete Institute, Farmington Hills,
MI, pp. 465, 2008.
[A3] Adhikary, S.D., Li, B., and Fujikake, K. “Dynamic Behavior of Reinforced
Concrete beams under Varying Rates of Concentrated Loading.” International
Journal of Impact Engineering, 47, 24-38, 2012.
[A4] Adhikary, S.D., Li, B., and Fujikake, K. “Strength and Behavior in Shear of
Reinforced Concrete Deep Beams under Dynamic Loading Conditions.” Nuclear
Engineering and Design, 259, 14-28, 2013.
[A5] Adhikary, S.D., Li, B., and Fujikake, K. “Effects of high loading rate on
Reinforced Concrete Beams.” ACI Structural journal, Available online.
[B1] Banthia, N. “Impact resistance of concrete.” PhD thesis, Department of Civil

Engineering, University of British Columbia, Vancouver, B.C., Canada, 1987.
[B2] Banthia, N., Mindess, S., Bentur, A., and Pigeon, M. “Impact Testing of
Concrete Using a Drop-weight Impact Machine.” Experimental Mechanics, 29(1),
63–69, 1989.
[B3] Bao, X., and Li, B. “Residual strength of blast damaged reinforced concrete
columns.” International Journal of Impact Engineering, 37, 295-308, 2010.
[B4] Bazant, Z.P., Pan, J., and Pijaudier, G. “Softening in Reinforced Concrete
Beams and Frames.” Journal of Structural Engineering, 113(12), 2333-2347, 1987.
[B5] Bazant, Z.P., Adley, M.D., Carol, I., Jirasek, M., Akers, S.A., Rohani, B.,
Cargile J.D. and Caner, F.C. “Large-strain generalization of micro plane model for
- 233 -
References
concrete and application.” Journal of Engineering Mechanics, 126(9), 971-980,

2000a.
[B6] Bazant, Z.P., Caner, F.C., Adley, M.D., and Akers, S.A., “Fracturing rate
effects and creep in micro plane model for dynamics.” Journal of Engineering
Mechanics, 126(9), 962-970, 2000b.
[B7] Bentur, A., Mindess, S., and Banthia, N. “The behavior of concrete under
impact loading: Experimental procedures and method of analysis.” Material and
Structures, 19(5), 371-378, 1986.
[B8] Bhatti, A. Q., Kishi, N., Mikani, H., and Ando, T. “Elasto-plastic impact
analysis of shear-failure-type RC beams with shear reinforcements.” Materials and
Design, 30, 502-510, 2009.
[B9] Biggs, J.M. “Introduction to Structural Dynamics.” McGraw-Hill, New York,

1964.
[B10] Bischoff, P.H. “Role of physical testing in impact analysis of concrete

structures.” ACI Symposium, SP 175-13, 1998.
[B11] Bischoff, P.H., and Perry, S.H. “Compressive Behavior of Concrete at High
Strain Rates.” Materials and Structures, 24, 425-450, 1991.
[C1] Chan, H.C., Cheung, Y.K., and Huang, Y.P. “Nonlinear Modeling of
Reinforced Concrete Structures.” Computers and Structures, 53(5), 1099-1107,
1994.
[C2] Chen, W.F. “Plasticity in Reinforced Concrete.” McGraw Hill, Inc., New
York, 1982.
[C3] Chen, Y., and May, I.M. “Reinforced Concrete Members under drop-weight
impacts.” Proceedings of the ICE-Structures and Buildings, 162(1), 45-56, 2009.
[C4] Chopra, A.K. “Dynamics of Structures, Prentice Hall.” Inc., New Jersey,
2001.
- 234 -
References
[C5] Comite Euro-International du Beton, CEB-FIP Model Code 1990, Redwood

Books, Trowbridge, Wiltshire, UK, 1993.
[C6] Cotsovos, D.M., and Pavlovic, M.N. “Numerical investigation of concrete

subjected to compressive impact loading. Part 1: A fundamental explanation for the
apparent strength gain at high loading rates.” Computers & Structures, 86, 145-163,
2008.
[C7] Cotsovos, D.M., and Pavlovic, M.N. “Numerical investigation of concrete

subjected to compressive impact loading. Part 2: Parametric investigation of factors
affecting behavior at high loading rates.” Computers & Structures, 86, 164-180,
2008.
[C8] Cotsovos, D.M., Stathopoulos, N.D., and Zeris, C.A. “Behavior of RC Beams
Subjected to High Rates of Concentrated Loading.” Journal of Structural
Engineering, 134(12), 1839-1851, 2008.
[D1] Dahlberg, D. “Piers: Bridge Design Manual and Policy Engineer.” MnDOT
Bridge Office LFRD Workshop, 2012.
[E1] El-Mezaini, N. and Citipitiouglu, E. “Finite Element Analysis of Prestressed

and Reinforced Concrete Structures.” Journal of Structural Engineering, 97(2), 252-
258, 1991.
[E2] El-Tawil, S., Severino, E., and Fonseca, P. “Vehicle Collision with Bridges
piers.” Journal of Structural Engineering, 10(3), 345-353, 2005.
[F1] Foster, S.J., Budiono, B., and Gilber, R.I. “Rotating Crack Finite Element
Model for Reinforced Concrete Structures.” Computers and Structures, 58(1), 43-
50, 1996.
[F2] Fu, H.C., Erki, M.A., and Seckin, M. “Review of Effects of Loading rate on
concrete in Compression.” Journal of Structural Engineering, 117(12), 3645-3659,
1991.
- 235 -
References
[F3] Fu, H.C., Erki, M.A., and Seckin, M. “Review of Effects of Loading rate on
Reinforced Concrete.” Journal of Structural Engineering, 117(12), 3660-3679,
1991.
[F4] Fujikake, K., Li, B., and Soeun, S. “Impact Response of Reinforced Concrete
Beam and Its Analytical Evaluation.” Journal of Structural Engineering, 135(8),
938-950, 2009.
[F5] Fujikake, K., Mori, K., Uebayashi, K., Ohno, T., and Mizuno, J. “Constitutive
model for concrete materials with high rates of loading under tri-axial compressive
stress states.” Proc. 3rd Int. Conf. on concrete under severe conditions, Vol. 1,
JSCE, Tokyo, 636-643, 2001.
[G1] Gopalaratnam, V.S., Shah, S.P., and John, R. “A modified instrumented

Charpy test for cement-based composites.” Experimental Mechanics, 24(2), 102-
111, 1984.
[G2] Grote, D.L., Park, S.W., and Zhou, M. “Dynamic behavior of concrete at high
strain rates and pressures: I Experimental Characterization.” International Journal of
Impact Engineering, 25, 869-886, 2001.
[H1] Hallquist, J.O. “LS-DYNA 3D Theoretical manual.” Livermore, Livermore

Software Technological Corporation, 2007.
[H2] Ho, D. “Impact Response of Reinforced Concrete: An Experimental and

Numerical Investigation.” MASc thesis, Department of Civil Engineering,
University of Toronto, Toronto, ON, Canada, 2004.
[H3] Hughes, G., and Beeby, A. W. “Investigation of the effect of impact loading
on concrete beams.” The Structural Engineer, 60B (3), 45–52, 1982.
[K1] Katayama, M., and Itoh, M., Tamura, S., Beppu. M., and Ohno. T. “Numerical
analysis method for the RC and geological structures subjected to extreme loading
by energetic material.” International Journal of Impact Engineering, 34 (9), 1546-
1561, 2007.
- 236 -
References
[K2] Keenan, W. A., and Feldman, A., “The Yield Strength of Intermediate Grade
Reinforcing Bars under Rapid Loading.” Behavior and Design of Deep Structural
Members, Part 6, Technical Report ASFWC-TR-59-72, Research Directorate, Air
Force Special Weapons Centre, Kirtland Air Force Base, New Mexico, March
(also: University of Illinois, Structural Research Series 197), 1960.
[K3] Kennedy, R.P. “A review of procedures for the analysis and design of concrete
structures to resist missile impact events.” Nuclear Engineering and Design, 37,
183-203, 1976.
[K4] Kim, J.K., and Lee, T.G. “Nonlinear Analysis of Reinforced Concrete Beams
with Softening.” Computers and Structures, 44(3), 567-593, 1992.
[K5] Kishi, N., Nakano, O., Matsouka, K.G., and Ando, T. “Experimental Study on
Ultimate Strength of Flexural-Failure-Type RC Beams under Impact Loading.”
Transactions, SMiRT 16, Washington DC, August, 2001.
[K6] Kishi, N., Mikami, H., and Ando, T. “An applicability of the FE impact
analysis on shear-failure-type RC beams with shear reinforcement.” Proceedings,
4th Asia-Pacific Conference on Shock and Impact Loads on Structures, 309-315,
Singapore, 2001.
[K7] Kishi, N., Mikami, H., Matsuoka, K. G., and Ando, T. “Impact behavior of
shear-failure type RC beams without shear reinforcement.” International Journal of
Impact Engineering, 27, 955-968, 2002.
[K8] Kishi, N., and Bhatti, A.Q. “An equivalent fracture energy concept for
nonlinear dynamic response analysis of prototype RC girders subjected to falling-
weight impact loading.” International Journal of Impact Engineering, 37, 295-308,
2010.
[K9] Kishi, N., and Khasraghy, S.G. and Kon-No, H. “Numerical simulation of
Reinforced Concrete beams under Consecutive Impact Loading.” ACI Structural
Journal, 108(4), 444-452, 2011.
- 237 -
References
[K10] Kishi, N., and Mikami, H. “Empirical Formulas for Designing Reinforced
Concrete Beams under Impact Loading.” ACI Structural Journal, 109(4), 509-519,
2012.
[K11] Krauthammer, T., Shahriar, S., and Shanaa, H.M. “Analysis of reinforced
concrete beams subjected to severe concentrated loads.” ACI Structural Journal,
84(4), 473-480, 1987.
[K12] Kulkarni, S.M. “Effect of High Strain Rate on Reinforced Concrete Beams.”
PhD thesis, Field of Civil Engineering, Northwestern University, Evanston, Illinois,
1997.
[K13] Kulkarni, S.M., and Shah, S.P. “Response of Reinforced Concrete Beams at
High Strain Rates.” ACI Structural Journal, 95(6), 705-715, 1998.
[K14] Kwak, H.G., and Kim, D.Y. “Nonlinear Analysis of RC Shear Walls
Considering Tension Stiffening Effect.” Computers and Structures, 79, 499-517,
2001.
[L1] Livermore Software Technology Corporation (LSTC) “LS-DYNA Keyword

User‟s Manual.” version 971, Livermore, CA, 2007.
[L2] Li, Q.M., and Meng, H. “About the dynamic strength enhancement of
concrete-like material in a Split Hopkinson Pressure Bar test.” International Journal
of Solid and Structures, 40, 343-360, 2003.
[L3] Li, M., and Li, H. “Effects of loading rate on Reinforced Concrete Beams.”
15th World Conference on Earthquake Engineering, 2012.
[M1] Macgregor, J.G., and Wight, J.K. “Reinforced Concrete: Mechanics and
Design” 4th Edition, Pearson Prentice Hall, New Jersey, 2005.
[M2] Malvar, L. J. “Review of static and dynamic properties of steel reinforcing

bars.” ACI Material Journal, 95(5), 609-616, 1998.
- 238 -
References
[M3] Malvar, L.J. and Crawford J.E. “Dynamic Increase Factor for Concrete.”
Twenty-Eighth DDESB Seminar, Orlando, FL, 1998.
[M4] Malvar, L. J., Crawford, J. E., Wesevich, J. W., and Simons, D. “A Plasticity
Concrete Material Model for DYNA3D.” International Journal of Impact
Engineering, 19, 847-873, 1997.
[M5] Malvar, L. J., and Ross, C. A. “Review of Strain Rate Effects for Concrete in
Tension.” ACI Materials Journal, 95(6), 735-739, 1998.
[M6] May, I.M., Chen, Y., Owen, D.R.J., Feng, Y.T., Bere, A.T. “Experimental
testing and finite element simulation of the behavior of Reinforced Concrete beams
under impact loading.” Proceedings of VIII International Conference on
Computational Plasticity, COMPLAS VIII, Barcelona, Spain, 2005.
[M7] May, I.M., and Sangi, A.J. “Numerical simulation of Response of Reinforced
Concrete Members under Falling weight Impacts.” The Third International
Conference on Design and Analysis of Protective Structures,” 10-12 May, 2010,
Singapore, 2010.
[M8] Mihashi, H., and Wittmann, F. H. “Stochastic approach to study the influence
of the rate loading on strength of concrete.” HERON, 25(3), 1980.
[M9] Mindess, S., Bentur, A., Yan, C., and Vondran, G. “Impact Resistance of
Concrete containing both conventional steel reinforcements and Fibrillated
Polypropylene Fibres.” ACI Material Journal, 86(6), 545-549, 1989.
[M10] Murray, Y.D. “Users manual for LS-DYNA concrete material model 159.”
Report No. FHWA-HRT-05-062, Federal Highway Administration, Washington,
DC, 2007.
[M11] Mutsuyoshi, H., and Machida, A. “Properties and Failure of Reinforced

Concrete Members Subjected to Dynamic Loading.” Transactions of the Japan
Concrete Institute, 6, 521-528, 1984.
- 239 -
References
[M12] Mylrea, T. D. “Effect of impact on reinforced concrete beams.” Journal of

American Concrete Institute, 11(6), 1940.
[O1] Orozco, L.A. “Development of an Experimental Setup for Testing Scaled

Versions of AASHTO Type IV Girders under Dynamic Impact Loadings.” MS
thesis, The University of Texas at Austin, 2006.
[O2] Ozbolt, J., and Reinhardt, H.W. “Rate dependent fracture of notched plain
concrete beams” Proceedings of the Seventh International Conference,
CONCREEP-7, 57-62, 2005.
[O3] Ozbolt, J., Rah, K.K. and Mestrovic, D. “Influence of loading rate on concrete
cone failure” International Journal of Fracture, 139, 239-252, 2006.
[O4] Ozbolt, J., Sharma, A., and Reinhardt, H.W. “Dynamic fracture of concrete
compact tension specimen” International Journal of Solids and Structures, 48,
1534-1543, 2011.
[O5] Ozbolt, J., Bosnjak, J., and Sola, E. “Dynamic fracture of concrete compact
tension specimen: Experimental and Numerical study.” International Journal of
Solids and Structures, 50, 4270-4278, 2013.
[P1] Park, R., and Paulay, T. “Reinforced Concrete Structures.” John Willey &
Sons, N.Y., 744 pp., 1975.
[R1] Reinhardt, H.W., Weerheijm, J. “Tensile fracture of concrete at high loading

rates taking account of inertia and crack velocity effects.” International Journal of
Fracture, 51, 31-42, 1991.
[R2] Ross, A., Tedesco, J.W. and Kuennen, S.T. “Effect of strain rate on concrete
strength.” ACI Material Journal, 92(1), 37-47, 1995.
[R3] Ross, A., Jerome, D.M., Tedesco, J.W. and Hughes, M.L. “Moisture and strain
rate effects on concrete strength.” ACI Material Journal, 93(3), 293-300, 1996.
- 240 -
References
[R4] Ross, C.A., Thompson, P.Y., and Tedesco, J.W. “Split-Hopkinson pressure
bar tests on concrete and mortar in tension and compression.” ACI Material Journal,
86(5), 475-481, 1989.
[R5] Rossi, P., Van Mier, J.G.M., Toutlemonde, F., Le Maou, F., and Boulay, C.
“Effect of loading rate on strength of concrete subjected to uniaxial tension.”
Material and Structures, 27, 260-264, 1994.
[S1] Saatci, S., and Vecchio, F. “Effects of Shear Mechanisms on Impact Behavior
of Reinforced Concrete Beams.” ACI Structural Journal, 106(1), 78-86, 2009.
[S2] Saatci, S., and Vecchio, F. “Nonlinear Finite Element Modeling of Reinforced
Concrete Structures under Impact Loads.” ACI Structural Journal, 106(5), 717-725,
2009.
[S3] Sauris, W., and Shah, S.P. “Properties of concrete subjected to impact.”
Journal of Structural Engineering, 109(7), 1727-1741, 1983.
[S4] Shi Y., Hao H., and Li Z-X. “Numerical derivation of pressure-impulse
diagrams for prediction of RC column damage to blast loads.” International Journal
of Impact Engineering, 35, 1213-1227, 2008.
[S5] Shirai, K., Ito, C., and Onuma, H. “Numerical studies of impact on reinforced
concrete beam of hard impact.” Nuclear Engineering and Design, 150, 483-489,
1994.
[S6] Shkolnik, I.E. “Influence of high strain rates on stress-strain relationship,

strength and elastic modulus of concrete.” Cement and Concrete composites, 30,
1000-1012, 2008.
[S7] Soroushian, P., and Choi, K. “Steel mechanical properties at different strain
rates.” Journal of Structural Engineering, 113(4), 663-672, 1987.
[S8] Soroushian, P., and Choi, K-B., Alhamad, A. “Dynamic constitutive behavior
of concrete.” ACI Journal, 251-259, March-April 1986.
- 241 -
References
[S9] Sugano, T., Tsubota, H., Kasai, Y., Koshima, N., Orui, S., von Riesemann,
W.A., Bickel, D.C., Parks, M.B. “Full-scale aircraft impact test for evaluation of
impact force.” Nuclear Engineering and Design, 140, 373-385, 1993.
[S10] Sun, C.T., and Yang, S.H. “Contact law and impact response of laminated
composites.” Report No.-NASA CR-159884, 1980.
[T1] Tachibana, S., Masuya, H. and Nakamura, S. “Performance based design of

reinforced concrete beams under impact.” Natural Hazards and Earth System
Sciences, 10, 1069-1078, 2010.
[T2] Takeda, J.I. “Dynamic fracture of concrete structures due to severe

earthquakes and some consideration on countermeasures.” Proceedings of Eighth
World Conference on Earthquake Engineering, San Francisco, CA, 1984.
[T3] Takeda, J.I., Tachikawa, H. and Fujimoto, K. “Basic concept of the responses
of structural members and structures under impact or impulsive loadings.” Inter-
Association Symposium on Concrete Structures, Berlin, BAM, 1982.
[T4] Tedesco, J.W., Powell, J.C., Ross, C.A. and Hughes, M.L. “A strain-rate
dependent concrete material model for ADINA.” Computer and Structures, 64(5/6),
1053-1067, 1997.
[T5] Tedesco, J.W., Ross, C.A. “Strain-rate-dependent constitutive equations for

concrete.” ASME Journal of Pressure Vessel Technology, 120, 398-405, 1998.
[T6] Thabet, A., and Haldane, D. “Three-dimensional simulation of nonlinear

response of reinforced concrete members subjected to impact loading.” ACI
Structural Journal, 97(5), 689-702, 2000.
[V1] Vos, E., and Reinhardt, H.W. “Influence of loading rate on bond behavior of
reinforcing steel and prestressing strands.” Material and Structures, 15(85), 3-10,
1982.
[W1] Wakabayashi, M., Nakamura, T., Yoshida, N., Iwai, S., and Watanabe, Y.
“Dynamic loading effects on the structural performance of concrete and steel
- 242 -
References
materials and beams.” Proc. Seventh World Conference on Earthquake

Engineering, Istanbul, Turkey, 271-278, 1980.
[W2] Weathersby, J.H. “Investigation of bond slip between concrete and steel
reinforcement under dynamic loading conditions.” Louisiana State University and
Agricultural and Mechanical College, 2003.
[W3] Wu, Ke-Chiang., Li, B., and Tsai Keh-Chyuan. “Residual axial compression
capacity of localized blast-damaged RC columns.” International Journal of Impact
Engineering, 38, 29-40, 2011.
[X1] Xiao, S., Li, H. and Monteiro, P.J.M. “Influence of strain rate and loading
histories on the tensile damage behavior of concrete.” Magazine of Concrete
Research, 62(12), 887-894, 2010.
[Y1] Yamaguchi, H., Fujimoto, K., and Nomura. S. “Stress-strain relationship for
concrete under high tri-axial compression part 2 rapid loading.” Transaction
Architectural Institute of Japan, 396:50-9, 1989.
[Y2] Yan, C. “Bond between reinforcing bars and concrete under impact loading.”
PhD thesis, Department of Civil Engineering, University of British Columbia,
1992.
[Y3] Yan, D., and Lin, G. “Dynamic properties of concrete in direct tension.”
Cement and Concrete Research, 36, 1371-1378, 2006.
[Y4] Yoshida, H., Nomura, T., Wyllie, D.C., and Morris, A.J. “Rock-fall shed-
Application of Japanese Design in North America.” 1st North American Landslide
Conference, 2007.
[Y5] Yuan, P., and Harik, I.E. “Equivalent Barge and Flotilla Impact Forces on
Bridge Piers.” Journal of Bridge Engineering, 15(5), 523-532, 2010.
[Z1] Zhou, X.Q., and Hao, H. “Modeling of compressive behavior of concrete-like

materials at high strain rate.” International Journal of Solid and Structures, 45,
4648-4661, 2008.
- 243 -
References
[Z2] Zielinski, A.J., and Reinhardt, H.W. “Stress strain behavior of concrete and
mortar at high rates of tensile loading.” Cement and Concrete Research, 12(3), 309-
319, 1982.
- 244 -
Appendix-A
APPENDIX-A: Beam Details and Response under Varying Rates of Concentrated Loading
Shear Longitudinal
span reinforcement
Shear
to
Beam dimensions reinforcement Maximum
f c eff. Loading Failure
Reference Beam mark depth Tensile (T) & resistance
(MPa) rate (m/s) modes
ratio Compressive (C) (kN)
h b d a Ratio f Ly Ratio f Ty
ad (%)
(mm) (mm) (mm) (mm) (%) (MPa) (MPa)
A 1-S - - 1  10-4 23.6
29.0 0.89T 382
A 1-D - - 2.63  10-1 30.8
A 2-S - - 5  10-5 45.0 Flexure
32.7 880 5.5 1.70T 377
A 2-D - - 2.2  10-1 55.0
A 3-S - - 5  10-5 57.9
31.9 2.39T 361
A 3-D - - 2.16  10-1 70.8
A 4-S - - 3  10-5 51.7
34 200 150 160 448 2.8 0.89T 432
A 4-D - - 2.33  10-1 72.4 Shear
A 5-S T - - 3  10-5 50.3
35.1 496 3.1 1.05 381
A 5-D - - 1.15  10-1 71.5
A 6-S - - 2  10-5 45.4
Mutsuyoshi 34 752 4.7 360
A 6-D - - 2.18  10-1 59.8 Flexure
& Machida 1.70T
A 7-S - - 5  10-5 49.0
[M10] 31.6 784 4.9 393
A 7-D - - 2.3  10-1 61.6 Shear
B 1-S T - - 2  10-5 23.6 Flexure
26.8 684 5.7 2.11 456
B 1-D - - 7.63  10-1 36.5 Shear
B 2-S - - 3  10-5 30.4
28.6 564 4.7 1.78T 398 Flexure
B 2-D - - 1.67  10-1 38.2
B 3-S - - 2  10-5 25.5 Flexure
23.2 150 100 120 480 4.0 1.19T 398
B 3-D - - 2.67  10-1 31.3 Shear
B 4-S - - 3  10-5 27.9
28.6 588 4.9 1.78T 393 Flexure
B 4-D - - 5.67  10-1 33.7
B 5-S - - 1.43  10-1 49.2
28.0 456 3.8 2.11T 506 Shear
B 5-D - - 2.77  10-1 47
- 245 -
Appendix-A
Shear Longitudinal
span reinforcement
Shear
to
ad (%)
C 1-S - - 2  10-5 33.4
4.78T
C 2-D 23.7 150 100 120 684 5.7 320 - - 1.89  10-1 50.3 Shear
C 3-D - - 4.08  10-1 52.6
B4JL25-S 41.5 836 5.5 7.1  10-6 41.7
B4JL25-H 41.5 836 5.5 3.8  10-1 46.4
Flexure
B3OC25-S 46.2 760 5 7.1  10-6 40.5
B3OC25-H 46.2 760 5 3.8  10-1 46.7
B4JL20-S 41.9 760 5 7.1  10-6 39.1 Shear
Kulkarni B4JL20-H 41.9 760 5 3.8  10-1 44.9 Flexure
and Shah B3SE03-S 45 684 4.5 1.37T 7.1  10-6 45.9 Shear
178 102 152 518 - -
[K12] B3SE03-H 45 684 4.5 3.8  10-1 51.6
B3DE03-S 43 684 4.5 7.1  10-6 45.3 Flexure
B3DE03-H 43 684 4.5 3.8  10-1 50.9
B3NO15-S 43 608 4 7.1  10-6 43.9 Shear
B3NO15-H 43 608 4 3.8  10-1 56.8 Flexure
B3NO30-S 45 532 3.5 7.1  10-6 49.9 Shear
B3NO30-H 45 532 3.5 3.8  10-1 55.8 Shear
S1616 1.27T&C 426 5  10-4 115
S1616 1.27T&C 426 2  100 135
Fujikake S2222 2.41T&C 418 5  10-4 200
42 250 150 210 700 3.33 0.5 295 Flexure
et al. [F4] S2222 2.41T&C 418 2  100 243
S1322 2.4&0.84 397 5  10-4 180
S1322 2.4&0.84 397 2  100 246
- 246 -
Appendix-A
Shear Longitudinal
span reinforcement
Shear
to
ad (%)
RC3_S0_S1 4  10-4 93.2
RC3_S0_S2 4  10-4 81.3
RC3_S0_L1 4  10-2 105.8
RC3_S0_L2 4  10-2 102.5
0
RC3_S0_M1 4  10-1 138
RC3_S0_M2 4  10-1 126.1
RC3_S0_H1 2  100 140.6
RC3_S0_H2 2  100 147.1
Shear
RC3_S12_S1 4  10-4 120
RC3_S12_S2 4  10-4 120.5
RC3_S12_L1 4  10-2 148.8
Adhikary et RC3_S12_L2 4  10-2 139.5
0.12
al. [A3] RC3_S12_M1 40 250 150 210 700 3.33 2.41T&C 371 342 4  10-1 156.2
RC3_S12_M2 4  10-1 164.7
RC3_S12_H1 2  100 168.7
RC3_S12_H2 2  100 180.9
RC3_S56_S1 4  10-4 200.4
RC3_S56_S2 4  10-4 202
RC3_S56_L1 4  10-2 211.7
RC3_S56_L2 4  10-2 217.9
RC3_S56_M1 0.56 4  10-1 222.6 Flexure
RC3_S56_M2 4  10-1 223.1
RC3_S56_H1 2  100 235.6
RC3_S56_H2 226.5
2  100
- 247 -
Appendix-A
Shear Longitudinal
span reinforcement
Shear
to
ad (%)
RC1.9_S0_S1 4  10-4 138.4
RC1.9_S0_S2 4  10-4 170.1
RC1.9_S0_L1 4  10-2 215.6
RC1.9_S0_L2 4  10-2 246.5
0
RC1.9_S0_M1 4  10-1 290.5
RC1.9_S0_M2 4  10-1 -
RC1.9_S0_H1 2  100 353.6
RC1.9_S0_H2 2  100 -
RC1.9_S42_S1 4  10-4 271.9
RC1.9_S42_S2 4  10-4 -
RC1.9_S42_L1 4  10-2 -
Adhikary et RC1.9_S42_L2 4  10-2 300.6
al. [A4] 40 250 150 210 700 1.9 2.41 T&C 371 0.42 342 Shear
RC1.9_S42_M1 4  10-1 361.3
RC1.9_S42_M2 4  10-1 -
RC1.9_S42_H1 2  100 417
RC1.9_S42_H2 2  100 -
RC1.9_S84_S1 4  10-4 329.8
RC1.9_S84_S2 4  10-4 333.8
RC1.9_S84_L1 4  10-2 -
RC1.9_S84_L2 4  10-2 378.2
0.84
RC1.9_S84_M1 4  10-1 419.4
RC1.9_S84_M2 4  10-1 422.1
RC1.9_S84_H1 2  100 447.9
RC1.9_S84_H2 2  100 445.5
- 248 -
Appendix-B
APPENDIX-B: Beam Geometry and Reinforcement Details for Drop-weight Impact Loading
Beam characteristics Longitudinal reinforcement Shear reinforcement
Beam ad Top Bottom

Mark f c h b d a L Lh Dia. Ratio f Ty Rs Rb
(mm) (mm) (mm) (mm) (mm) Dia. Ratio f Ly Dia. Ratio f Ly (mm) (%)
(MPa) (mm) (%) (mm) (%) (MPa)
(MPa) (MPa)
Kishi et al. [K4]
A19 26.5 240 160 190 1000 2500 5.26 10.42 19 1.88 373 19 1.88 373 6 0.396 373 3.1
A22 26.5 240 160 190 1000 2500 5.26 10.42 22 2.55 373 22 2.55 373 6 0.396 373 1.9
B10 26.5 220 200 170 1000 2500 5.88 11.36 10 0.42 373 10 0.42 373 6 0.317 373 5.85
B13 26.5 220 200 170 1000 2500 5.88 11.36 13 0.75 373 13 0.75 373 6 0.317 373 3.96
B19 26.5 220 200 170 1000 2500 5.88 11.36 19 1.69 373 19 1.69 373 6 0.317 373 2.13
C10 26.5 160 160 120 1000 2500 8.33 15.63 10 0.74 373 10 0.74 373 6 0.396 373 6.04
C13 26.5 160 160 120 1000 2500 8.33 15.63 13 1.32 373 13 1.32 373 6 0.396 373 4.36
C19 26.5 160 160 120 1000 2500 8.33 15.63 19 2.98 373 19 2.98 373 6 0.396 373 2.3
Kishi et al. [K6]
33 250 150 210 500 1400 2.4 5.6 N/A N/A N/A 19 1.8 393 N/A N/A N/A 0.43
33 250 150 210 500 1400 2.4 5.6 N/A N/A N/A 19 1.8 393 N/A N/A N/A 0.43
A24 33 250 150 210 500 1400 2.4 5.6 N/A N/A N/A 19 1.8 393 N/A N/A N/A 0.43
33 250 150 210 500 1400 2.4 5.6 N/A N/A N/A 19 1.8 393 N/A N/A N/A 0.43
33 250 150 210 500 1400 2.4 5.6 N/A N/A N/A 19 1.8 393 N/A N/A N/A 0.43
33 250 150 210 750 1900 3.6 7.6 N/A N/A N/A 19 1.8 393 N/A N/A N/A 0.64
33 250 150 210 750 1900 3.6 7.6 N/A N/A N/A 19 1.8 393 N/A N/A N/A 0.64
A36
33 250 150 210 750 1900 3.6 7.6 N/A N/A N/A 19 1.8 393 N/A N/A N/A 0.64
33 250 150 210 750 1900 3.6 7.6 N/A N/A N/A 19 1.8 393 N/A N/A N/A 0.64
33 250 150 210 1000 2400 4.8 9.6 N/A N/A N/A 19 1.8 393 N/A N/A N/A 0.86
A48 33 250 150 210 1000 2400 4.8 9.6 N/A N/A N/A 19 1.8 393 N/A N/A N/A 0.86
33 250 150 210 1000 2400 4.8 9.6 N/A N/A N/A 19 1.8 393 N/A N/A N/A 0.86
- 249 -
Appendix-B
Beam f c h b d a L ad Top Bottom Dia. Ratio f Ty

Mark
Lh Rs Rb
(MPa) (mm) (mm) (mm) (mm) (mm) Dia. Ratio f Ly Dia. Ratio f Ly (mm) (%) (MPa)
(mm) (%) (MPa) (mm) (%) (MPa)
Kishi et al. [K6]
A48 33 250 150 210 1000 2400 4.8 9.6 N/A N/A N/A 19 1.8 393 N/A N/A N/A 0.86
33 250 150 210 500 1400 2.4 5.6 N/A N/A N/A 13 0.84 393 N/A N/A N/A 0.68
33 250 150 210 500 1400 2.4 5.6 N/A N/A N/A 13 0.84 393 N/A N/A N/A 0.68
B24 33 250 150 210 500 1400 2.4 5.6 N/A N/A N/A 13 0.84 393 N/A N/A N/A 0.68
33 250 150 210 500 1400 2.4 5.6 N/A N/A N/A 13 0.84 393 N/A N/A N/A 0.68
33 250 150 210 500 1400 2.4 5.6 N/A N/A N/A 13 0.84 393 N/A N/A N/A 0.68
33 250 150 210 750 1900 3.6 7.6 N/A N/A N/A 13 0.84 393 N/A N/A N/A 1.03
33 250 150 210 750 1900 3.6 7.6 N/A N/A N/A 13 0.84 393 N/A N/A N/A 1.03
B36
33 250 150 210 750 1900 3.6 7.6 N/A N/A N/A 13 0.84 393 N/A N/A N/A 1.03
33 250 150 210 750 1900 3.6 7.6 N/A N/A N/A 13 0.84 393 N/A N/A N/A 1.03
33 250 150 210 1000 2400 4.8 9.6 N/A N/A N/A 13 0.84 393 N/A N/A N/A 1.37
33 250 150 210 1000 2400 4.8 9.6 N/A N/A N/A 13 0.84 393 N/A N/A N/A 1.37
B48 33 250 150 210 1000 2400 4.8 9.6 N/A N/A N/A 13 0.84 393 N/A N/A N/A 1.37
33 250 150 210 1000 2400 4.8 9.6 N/A N/A N/A 13 0.84 393 N/A N/A N/A 1.37
33 250 150 210 1000 2400 4.8 9.6 N/A N/A N/A 13 0.84 393 N/A N/A N/A 1.37
Bhatti and Kishi [B6]
A37 41.2 400 200 350 1000 2400 2.86 6 35 2.75 373 35 2.75 373 6 0.21 373 0.57
A46 41.2 400 200 350 1000 2400 2.86 6 35 2.75 373 35 2.75 373 6 0.21 373 0.57
A56 41.2 400 200 350 1000 2400 2.86 6 35 2.75 373 35 2.75 373 6 0.21 373 0.57
A65 41.2 400 200 350 1000 2400 2.86 6 35 2.75 373 35 2.75 373 6 0.21 373 0.57
A74 41.2 400 200 350 1000 2400 2.86 6 35 2.75 373 35 2.75 373 6 0.21 373 0.57
- 250 -
Appendix-B
Beam f c h b d a L ad Top Bottom Dia. Ratio f Ty

Mark
Lh Rs Rb
(mm) (%) (MPa) (mm) (%) (MPa)
A84 41.2 400 200 350 1000 2400 2.86 6 35 2.75 373 35 2.75 373 6 0.21 373 0.57
B37 41.2 400 200 350 1000 2400 2.86 6 35 2.75 373 35 2.75 373 6 0.42 373 0.76
B46 41.2 400 200 350 1000 2400 2.86 6 35 2.75 373 35 2.75 373 6 0.42 373 0.76
B65 41.2 400 200 350 1000 2400 2.86 6 35 2.75 373 35 2.75 373 6 0.42 373 0.76
B74 41.2 400 200 350 1000 2400 2.86 6 35 2.75 373 35 2.75 373 6 0.42 373 0.76
B84 41.2 400 200 350 1000 2400 2.86 6 35 2.75 373 35 2.75 373 6 0.42 373 0.76
B93 41.2 400 200 350 1000 2400 2.86 6 35 2.75 373 35 2.75 373 6 0.42 373 0.76
Fujikake et al. [F4]
42 250 150 210 700 1700 3.3 6.8 16 1.28 426 16 1.28 426 10 1.4 295 2.55
42 250 150 210 700 1700 3.3 6.8 16 1.28 426 16 1.28 426 10 1.4 295 2.55
S1616
42 250 150 210 700 1700 3.3 6.8 16 1.28 426 16 1.28 426 10 1.4 295 2.55
42 250 150 210 700 1700 3.3 6.8 16 1.28 426 16 1.28 426 10 1.4 295 2.55
42 250 150 210 700 1700 3.3 6.8 13 0.84 397 22 2.41 418 10 1.4 295 1.51
42 250 150 210 700 1700 3.3 6.8 13 0.84 397 22 2.41 418 10 1.4 295 1.51
S1322
42 250 150 210 700 1700 3.3 6.8 13 0.84 397 22 2.41 418 10 1.4 295 1.51
42 250 150 210 700 1700 3.3 6.8 13 0.84 397 22 2.41 418 10 1.4 295 1.51
42 250 150 210 700 1700 3.3 6.8 22 2.41 418 22 2.41 418 10 1.4 295 1.51
42 250 150 210 700 1700 3.3 6.8 22 2.41 418 22 2.41 418 10 1.4 295 1.51
S2222
42 250 150 210 700 1700 3.3 6.8 22 2.41 418 22 2.41 418 10 1.4 295 1.51
42 250 150 210 700 1700 3.3 6.8 22 2.41 418 22 2.41 418 10 1.4 295 1.51
- 251 -
Appendix-B
Beam f c h b d a L Top Bottom Dia. Ratio f Ty

mark
ad Lh Rs Rb
(mm) (%) (MPa) (mm) (%) (MPa)
Chen and May [C3]
49.2 200 100 175 1350 3000 7.7 15 6 0.32 510 12 1.29 510 6 0.28 250 1.48
49.2 200 100 175 1350 3000 7.7 15 6 0.32 510 12 1.29 510 6 0.28 250 1.48
49.2 200 100 175 1350 3000 7.7 15 6 0.32 510 12 1.29 510 6 0.28 250 1.48
A1
49.2 200 100 175 1350 3000 7.7 15 6 0.32 510 12 1.29 510 6 0.28 250 1.48
45.8 200 100 175 1350 3000 7.7 15 6 0.32 510 12 1.29 510 6 0.28 250 1.44
45.8 200 100 175 1350 3000 7.7 15 6 0.32 510 12 1.29 510 6 0.28 250 1.44
A1* 33.6 200 100 175 1350 3000 7.7 15 6 0.32 510 12 1.29 510 6 0.28 250 1.26
45.8 200 100 175 1350 3000 7.7 15 6 0.32 510 12 1.29 510 6 0.28 250 1.44
A2
42.8 200 100 175 1350 3000 7.7 15 6 0.32 510 12 1.29 510 6 0.28 250 1.38
A3 35.6 200 100 175 750 1800 4.3 9 6 0.32 510 12 1.29 510 6 0.28 250 0.72
33.6 200 100 175 1350 3000 7.7 15 6 0.32 510 12 1.29 510 6 0.28 250 1.26
B1
33.6 200 100 175 1350 3000 7.7 15 6 0.32 510 12 1.29 510 6 0.28 250 1.26
B1* 45.8 200 100 175 1350 3000 7.7 15 6 0.32 510 12 1.29 510 6 0.28 250 1.44
35.6 200 100 175 750 1800 4.3 9 6 0.32 510 12 1.29 510 6 0.28 250 0.72
B2
35.6 200 100 175 750 1800 4.3 9 6 0.32 510 12 1.29 510 6 0.28 250 0.72
B3 42.8 200 100 175 1350 3000 7.7 15 6 0.32 510 12 1.29 510 6 0.28 250 1.38
B3* 33.6 200 100 175 1350 3000 7.7 15 6 0.32 510 12 1.29 510 6 0.28 250 1.26
B4 35.6 200 100 175 750 1800 4.3 9 6 0.32 510 12 1.29 510 6 0.28 250 0.72
- 252 -
Appendix-B

mark
ad Lh Rs Rb
(mm) (%) (MPa) (mm) (%) (MPa)
Saatci and Vecchio [S1]
SS0a 50.1 410 250 357 1500 4880 4.2 11.9 30 1.58 464 30 1.58 464 6 N/A N/A 0.67
SS0b 50.1 410 250 357 1500 4880 4.2 11.9 30 1.58 464 30 1.58 464 6 N/A N/A 0.67
SS1a 44.7 410 250 357 1500 4880 4.2 11.9 30 1.58 464 30 1.58 464 6 0.1 605 1.08
SS1b 44.7 410 250 357 1500 4880 4.2 11.9 30 1.58 464 30 1.58 464 6 0.1 605 1.08
SS2a 47 410 250 357 1500 4880 4.2 11.9 30 1.58 464 30 1.58 464 6 0.2 605 1.28
SS2b 47 410 250 357 1500 4880 4.2 11.9 30 1.58 464 30 1.58 464 6 0.2 605 1.28
SS3a 46.7 410 250 357 1500 4880 4.2 11.9 30 1.58 464 30 1.58 464 6 0.3 605 1.89
SS3b 46.7 410 250 357 1500 4880 4.2 11.9 30 1.58 464 30 1.58 464 6 0.3 605 1.89
Tachibana et al. [T1]
A1 24 250 150 210 500 1400 2.38 5.6 13 0.843 345 13 0.843 345 6 0.75 295 1.4
A2-1 24 250 150 210 1000 2400 4.76 9.6 13 0.843 345 13 0.843 345 6 0.75 295 2.7
A2-2 24 250 150 210 1000 2400 4.76 9.6 13 0.843 345 13 0.843 345 6 0.75 295 2.7
A2-3 24 250 150 210 1000 2400 4.76 9.6 13 0.843 345 13 0.843 345 6 0.75 295 2.7
A2-4 24 250 150 210 1000 2400 4.76 9.6 13 0.843 345 13 0.843 345 6 0.75 295 2.7
A2-5 24 250 150 210 1000 2400 4.76 9.6 13 0.843 345 13 0.843 345 6 0.75 295 2.7
A2-6 24 250 150 210 1000 2400 4.76 9.6 13 0.843 345 13 0.843 345 6 0.75 295 2.7
A2-7 24 250 150 210 1000 2400 4.76 9.6 13 0.843 345 13 0.843 345 6 0.75 295 2.7
A2-8 24 250 150 210 1000 2400 4.76 9.6 13 0.843 345 13 0.843 345 6 0.75 295 2.7
A2-9 24 250 150 210 1000 2400 4.76 9.6 13 0.843 345 13 0.843 345 6 0.75 295 2.7
A2-10 24 250 150 210 1000 2400 4.76 9.6 13 0.843 345 13 0.843 345 6 0.75 295 2.7
A2-11 24 250 150 210 1000 2400 4.76 9.6 13 0.843 345 13 0.843 345 6 0.75 295 2.7
- 253 -
Appendix-B

mark
ad Lh Rs Rb
(mm) (%) (MPa) (mm) (%) (MPa)
A2-12 24 250 150 210 1000 2400 4.76 9.6 13 0.843 345 13 0.843 345 6 0.75 295 2.7
A2-13 24 250 150 210 1000 2400 4.76 9.6 13 0.843 345 13 0.843 345 6 0.75 295 2.7
A2-14 24 250 150 210 1000 2400 4.76 9.6 13 0.843 345 13 0.843 345 6 0.75 295 2.7
A4 24 250 150 210 2000 4400 3.52 17.6 13 0.843 345 13 0.843 345 6 0.75 295 5.5
B 24 150 300 110 1000 2400 9.09 16 13 0.804 345 13 0.804 345 6 0.37 295 2.1
C 24 250 150 210 1000 2400 4.76 9.6 16 1.277 345 16 1.277 345 6 0.75 295 1.9
D 24 250 150 210 1000 2400 4.76 9.6 10 0.499 345 10 0.499 345 6 0.75 295 4.3
E 24 400 150 360 1000 2400 2.78 6 13 0.492 345 13 0.492 345 6 0.75 295 2.4
F 24 400 150 360 1000 2400 2.78 6 10 0.291 345 10 0.291 345 6 0.75 295 4
Kishi and Mikami [K9]
G1-1 33.7 300 200 260 1500 3400 5.77 11.33 19 1.1 379 19 1.1 379 6 0.07 295 2.81
G1-1S 33.7 300 200 260 1500 3400 5.77 11.33 19 1.1 379 19 1.1 379 6 0.07 295 2.81
G2-1 32.2 250 150 210 1000 2400 4.76 9.6 13 0.8 373 13 0.8 373 6 0.13 295 3.67
G2-2 32.2 250 150 210 1000 2400 4.76 9.6 13 0.8 373 13 0.8 373 6 0.13 295 3.67
G2-3 32.2 250 150 210 1000 2400 4.76 9.6 13 0.8 373 13 0.8 373 6 0.13 295 3.67
G2L-1 32.2 250 150 210 1000 2400 4.76 9.6 13 0.8 373 13 0.8 373 6 0.13 295 3.67
G2L-2 32.2 250 150 210 1000 2400 4.76 9.6 13 0.8 373 13 0.8 373 6 0.13 295 3.67
G2L-3 32.2 250 150 210 1000 2400 4.76 9.6 13 0.8 373 13 0.8 373 6 0.13 295 3.67
G3-1 34.6 250 150 210 1000 2400 4.76 9.6 13 0.8 393 13 0.8 393 6 0.13 295 3.51
G3-2 34.6 250 150 210 1000 2400 4.76 9.6 13 0.8 393 13 0.8 393 6 0.13 295 3.51
G3-3 34.6 250 150 210 1000 2400 4.76 9.6 13 0.8 393 13 0.8 393 6 0.13 295 3.51
- 254 -
Appendix-B

f c h b d a L Top Bottom Dia. Ratio f Ty
Beam mark ad Lh Rs Rb
(mm) (%) (MPa) (mm) (%) (MPa)
G4-1 32.3 250 150 210 1000 2400 4.76 9.6 13 0.8 373 13 0.8 373 6 0.13 295 3.52
G4-2 32.3 250 150 210 1000 2400 4.76 9.6 13 0.8 373 13 0.8 373 6 0.13 295 3.52
G5-1 39.2 300 200 260 1500 3400 5.77 11.33 19 1.1 379 19 1.1 379 6 0.07 295 2.85
G5-2 39.2 300 200 260 1500 3400 5.77 11.33 19 1.1 379 19 1.1 379 6 0.07 295 2.85
G6-1 34.7 250 250 210 1000 2400 4.76 9.6 19 1.09 392 19 1.09 392 6 0.13 295 2.19
G7-1 34.7 250 250 210 1500 3400 7.14 13.6 19 1.09 392 19 1.09 392 6 0.09 295 2.78
G7-2 34.7 250 250 210 1500 3400 7.14 13.6 19 1.09 392 19 1.09 392 6 0.09 295 2.78
G8-1 34.7 200 200 160 1000 2400 6.25 12 25 3.17 383 25 3.17 383 6 0.16 295 1.55
G9-1 34.7 200 200 160 1500 3400 9.38 17 25 3.17 383 25 3.17 383 6 0.12 295 2
G9-2 34.7 200 200 160 1500 3400 9.38 17 25 3.17 383 25 3.17 383 6 0.12 295 2
G10-1 23.5 250 200 210 1500 3400 7.14 13.6 19 1.36 404 19 1.36 404 6 0.09 295 5.11
G10-2 23.5 250 200 210 1500 3400 7.14 13.6 19 1.36 404 19 1.36 404 6 0.09 295 5.11
G10-3 23.5 250 200 210 1500 3400 7.14 13.6 19 1.36 404 19 1.36 404 6 0.09 295 5.11
G10-4 23.5 250 200 210 1500 3400 7.14 13.6 19 1.36 404 19 1.36 404 6 0.09 295 5.11
G11-1 23.6 300 200 250 1350 3100 5.4 10.33 22 1.55 401 22 1.55 401 6 0.08 295 1.66
G11-2 23.6 300 200 250 1350 3100 5.4 10.33 22 1.55 401 22 1.55 401 6 0.08 295 1.66
G11-3 23.6 300 200 250 1350 3100 5.4 10.33 22 1.55 401 22 1.55 401 6 0.08 295 1.66
G11-4 23.6 300 200 250 1350 3100 5.4 10.33 22 1.55 401 22 1.55 401 6 0.08 295 1.66
G11-5 23.6 300 200 250 1350 3100 5.4 10.33 22 1.55 401 22 1.55 401 6 0.08 295 1.66
G11-6 23.6 300 200 250 1350 3100 5.4 10.33 22 1.55 401 22 1.55 401 6 0.08 295 1.66
- 255 -
Appendix-B

f c h b d a L Top Bottom Dia. Ratio fTy
Beam mark ad Lh Rs Rb
(mm) (%) (MPa) (mm) (%) (MPa)
G12-1 23.6 300 200 250 1350 3100 5.4 10.33 25 3.04 407 25 3.04 407 6 0.08 295 1.52
G13-1 23.6 400 200 350 1350 3100 3.86 7.75 25 1.45 406 25 1.45 406 6 0.06 295 2.11
G14-1 23.6 350 200 300 1350 3100 4.5 8.86 25 1.69 406 25 1.69 406 6 0.07 295 1.96
G15-1 23.6 400 200 350 1350 3100 3.86 7.75 25 1.11 406 25 1.11 406 6 0.06 295 3.58
G16-1 23.6 370 200 320 1350 3100 4.22 8.38 25 1.58 406 25 1.58 406 6 0.06 295 2.16
Note: N/A- Not Applicable
- 256 -
Appendix-C
APPENDIX-C: Impact Response Characteristics
Mass of Height of Impact Maximum Maximum Maximum

Input kinetic
Beam mark Impactor Impactor Velocity impact force reaction force displacement Failure modes
energy (kJ)
(kg) (m) (m/s) (kN) (kN) (mm)
Kishi et al. [K4]

A19 200 N/P 1.00 0.1 N/P 41.9 N/P N/P
A22 200 N/P 1.00 0.1 N/P N/P N/P N/P
B10 200 N/P 1.00 0.1 21.9 29.2 0.45 N/P
B13 200 N/P 1.00 0.1 N/P N/P N/P N/P
B19 200 N/P 1.00 0.1 N/P 46.9 N/P N/P
C10 200 N/P 1.00 0.1 N/P N/P N/P N/P
C13 200 N/P 1.00 0.1 N/P 21.6 N/P N/P
C19 200 N/P 1.00 0.1 N/P N/P N/P N/P
Kishi et al. [K6]
300 0.05 1.00 0.15 90.9 93.5 1.1 N/P
300 0.46 3.00 1.35 144.4 113.4 10.6 N/P
A24 300 0.82 4.00 2.4 183.5 162.9 15.1 N/P
300 1.27 5.00 3.75 194.3 161.2 29.6 N/P
300 1.84 6.00 5.4 197.3 161.4 53.6 N/P
300 0.05 1.00 0.15 62.2 60.3 1.6 Flexure
300 0.46 3.00 1.35 124.2 121.5 11.3 Diagonal tension (One side)
A36
300 0.82 4.00 2.4 117.9 125.6 28.8 Diagonal tension (Both side)
300 1.27 5.00 3.75 142.2 105.1 70 Disintegration
300 0.05 1.00 0.15 40.3 40.3 4 N/P
300 0.46 3.00 1.35 90.1 90.1 13.7 N/P
A48
300 0.82 4.00 2.4 125 107.9 18.4 N/P
300 1.27 5.00 3.75 141.6 104.6 145.8 N/P
- 257 -
Appendix-C
Mass of Height of Impact Input kinetic Maximum Maximum Maximum

Beam mark Impactor Impactor Velocity energy impact force reaction force displacement Failure modes
(kg) (m) (m/s) (kJ) (kN) (kN) (mm)
Kishi et al. [K6]

300 0.05 1.00 0.15 69.3 70.1 2.3 N/P
300 0.2 2.00 0.6 101.3 95 6.6 N/P
B24 300 0.46 3.00 1.35 124.3 98.8 15.1 N/P
300 0.82 4.00 2.4 125.7 94.5 17 N/P
300 1.27 5.00 3.75 136.4 110.3 53.3 N/P
300 0.05 1.00 0.15 44.9 43.1 2.9 Flexure
300 0.46 3.00 1.35 87 71.5 17.1 Flexure
B36
300 0.82 4.00 2.4 104.7 72.2 27.1 Flexure-Compression
300 1.27 5.00 3.75 128.8 80.1 109.3 Diagonal tension (One side)
300 0.05 1.00 0.15 39.4 31.6 4.5 N/P
300 0.46 3.00 1.35 94.5 56.7 21.8 N/P
B48 300 0.82 4.00 2.4 111.7 86 36.5 N/P
300 1.27 5.00 3.75 133 110.3 52.8 N/P
300 1.84 6.00 5.4 148.4 114.5 135.1 N/P
Diagonal cracks and shear plug from
A37 384 0.7 3.67 2.7 765 657 8.13 impactor point to bottom portion of
beam
More diagonal cracks, formation of
A46 384 1.08 4.58 4.2 900 732 10.13
shear plug and flexural cracks
Diagonal cracks and more pronounced
A56 384 1.6 5.61 6.3 1007 938 13.13
shear plug
A65 384 2.15 6.52 8.5 1125 929 16 Shear plug and diagonal tension failure
A74 384 2.79 7.42 11 1250 1125 22 More Diagonal cracks and shear plug
- 258 -
Appendix-C

(kg) (m) (m/s) (kJ) (kN) (kN) (mm)

Diagonal tension failure and spalling of
A84 384 3.6 8.40 14.1 1350 1088 24.38
concrete on top of the beam
Diagonal cracks and flexural cracks
B37 384 0.7 3.67 2.7 782 635 7.5
appear
More diagonal and flexural cracks
B46 384 1.08 4.58 4.2 900 816 11.25
appear
Diagonal cracks and formation of shear
B65 384 2.15 6.52 8.5 1219 954 16
plug
Width of diagonal crack in one side
B74 384 2.79 7.42 11 1350 1160 19.8 increases and spalling of concrete in
impact zone starts
One sided diagonal cracks and number
B84 384 3.6 8.40 14.1 1382 1214 24
of cracks increases in impact zones
More wide diagonal cracks on both
sides and more number of cracks and
B93 384 4.41 9.30 17.3 1519 1160 28
more spalling of concrete at the top
surface appears
400 0.15 1.73 0.6 119 N/P 5.9 Flexure
400 0.3 2.45 1.2 164 N/P 11.01 Flexure
S1616
400 0.6 3.46 2.4 245 N/P 20.11 Flexure
400 1.2 4.85 4.7 314 N/P 36.17 Flexure
400 0.3 2.45 1.2 177 N/P 7.79 Flexure
S1322
400 0.6 3.46 2.4 262.6 N/P 11.44 Flexure
- 259 -
Appendix-C

(kg) (m) (m/s) (kJ) (kN) (kN) (mm)

Local failure with heavily crushed
400 1.2 4.85 4.7 303.3 N/P 23.45
concrete at impact point
S1322
400 2.4 6.86 9.4 314.8 N/P 27.03
400 0.3 2.45 1.2 204.2 N/P 6.99 Flexure
400 0.6 3.46 2.4 262.7 N/P 11.04 Flexure
S2222 Local failure with heavily crushed
400 1.2 4.85 4.7 309.9 N/P 21.42
400 2.4 6.86 9.4 353.2 N/P 31.82
Chen and May [C3]
Flexure failure and no spalling of
98.7 2.72 7.30 2.63 N/P N/P N/P
impact zone
98.7 2.72 7.30 2.63 223 N/P N/P Same as above
A1 98.7 2.72 7.30 2.63 234 N/P N/P Same as above
98.7 2.72 7.30 2.63 N/P N/P N/P Same as above
A1* 98.7 2.72 7.30 2.63 128 N/P N/P Same as above
A2
Flexure failure and no spalling of
A3 98.7 2.72 7.30 2.63 194 N/P N/P
impact zone
- 260 -
Appendix-C

(kg) (m) (m/s) (kJ) (kN) (kN) (mm)
Chen and May [C3]
High yielding or rupture of tension
98.7 2.72 7.30 2.63 N/P N/P N/P
B1 steel , spalling of impact zone
B1* 98.7 2.72 7.30 2.63 183 N/P N/P Same as above
High yielding or rupture of tension
B2 steel, spalling of impact zone spalling
98.7 2.72 7.30 2.63 171 N/P N/P
of impact zone also scabbing and
spalling
B3 98.7 2.72 7.30 2.63 654 N/P N/P Scabbing and spalling
B3* 98.7 2.72 7.30 2.63 215 N/P N/P Scabbing and spalling
B4 98.7 2.72 7.30 2.63 241 N/P N/P Scabbing and spalling
Several diagonal cracks on both sides
SS0a 211 3.26 8.00 6.75 N/P 305 9.3 but one side slightly wider 0.5 mm in
width
Extensive damage: massive concrete
spalling at top and bottom, longitudinal
SS0b 600 3.26 8.00 19.18 N/P N/P N/P
reinforcements were exposed and bent,
bond failure at support
Diagonal cracks on both sides and
SS1a 211 3.26 8.00 6.75 N/P 356 12.1 shear plug with 0.25 mm wide under
impact point
Diagonal cracks up to 5 mm wide and
SS1b 600 3.26 8.00 19.18 N/P 625 39.5 0.9 mm wide vertical cracks at
midspan
Several minor cracks and 0.2 mm wide
SS2a 211 3.26 8.00 6.75 N/P 327 10
diagonal crack
- 261 -
Appendix-C

(kg) (m) (m/s) (kJ) (kN) (kN) (mm)

Several diagonal cracks, one 1.8 mm
SS2b 600 3.26 8.00 19.18 N/P 592 37.9 wide and vertical cracks (1.3 mm) in
midspan
One sided 0.15 mm wide diagonal
SS3a 211 3.26 8.00 6.75 N/P 431 10.7 crack with approx. 45 degree
inclination
Several diagonal cracks (0.4 mm) and
SS3b 600 3.26 8.00 19.18 N/P 682 35.3
vertical cracks (1 mm) in midspan
A1 300 N/P 5.00 3.75 434 N/P 24.1 N/P
A2-1 150 N/P 3.46 0.9 320.5 N/P 13.6 N/P
A2-2 300 N/P 2.45 0.9 293.4 N/P 25.4 N/P
A2-3 450 N/P 2.00 0.9 245.3 N/P 37 N/P
A2-4 150 N/P 4.90 1.8 453.4 N/P 16.3 N/P
A2-8 300 N/P 4.24 2.7 513.3 N/P 33.3 N/P
A2-9 450 N/P 3.46 2.7 444.6 N/P 48.4 N/P
A2-10 300 N/P 1.00 0.15 65.4 N/P 4.5 N/P
A2-11 300 N/P 2.00 0.6 253.2 N/P 12.6 N/P
A2-12 300 N/P 3.00 1.35 426.2 N/P 26.9 N/P
A2-13 300 N/P 4.00 2.4 489.3 N/P 41.4 N/P
A2-14 300 N/P 5.00 3.75 466.2 N/P 58.3 N/P
A4 300 N/P 5.00 3.75 451.5 N/P 114.9 N/P
B 300 N/P 5.00 3.75 667.1 N/P 77 N/P
C 300 N/P 5.00 3.75 650.3 N/P 42.4 N/P
D 300 N/P 5.00 3.75 638.7 N/P 94 N/P
- 262 -
Appendix-C

(kg) (m) (m/s) (kJ) (kN) (kN) (mm)
E 300 N/P 5.00 3.75 742.2 N/P 29.1 N/P
F 300 N/P 5.00 3.75 663.5 N/P 43.9 N/P
G1-1 300 2.5 7.00 7.35 1495.6 278.5 64.3 N/P
G1-1S 300 2.5 7.00 7.35 1600.6 287.6 58 N/P
G2-1 300 0.82 4.01 2.41 510.2 95.3 28.3 N/P
G2-2 300 1.28 5.01 3.77 779.3 92.9 44 N/P
G2-3 300 1.84 6.01 5.41 853.5 149.6 57 N/P
Flexural Cracks from lower and upper
G2L-1 400 0.82 4.01 3.22 446.6 99 44.2 fibre, diagonal shear cracks around
midspan area
G3-1 300 0.82 4.01 2.41 1208.5 80.6 36.7 N/P
G3-2 300 1.28 5.01 3.77 1469.5 126.4 52 N/P
G3-3 300 1.84 6.01 5.41 1038.6 169 70.6 N/P
G4-1 300 0.82 4.01 2.41 800.3 122.5 39.7 N/P
G4-2 300 1.28 5.01 3.77 985.8 153.6 56.1 N/P
G5-1 400 1.84 6.01 7.22 1313.2 317.2 63.5 fibre, diagonal shear cracks around
midspan area
midspan area
G6-1 300 1.28 5.01 3.77 1335.9 268.4 26.4 N/P
G7-1 300 1.28 5.01 3.77 1242.8 181.1 45.8 N/P
G7-2 300 1.84 6.01 5.41 1588.3 192.8 60.9 N/P
G8-1 300 1.84 6.01 5.41 1192.1 277.5 36.5 N/P
- 263 -
Appendix-C
Mass of Height of Impact Input Maximum Maximum Maximum

Beam mark Impactor Impactor Velocity kinetic impact force reaction force displacement Failure modes
(kg) (m) (m/s) energy (kJ) (kN) (kN) (mm)
G9-1 300 1.28 5.01 3.77 931.2 208.5 43.2 N/P
G9-2 300 1.84 6.01 5.41 1102.7 248.2 57.9 N/P
midspan area
midspan area
midspan area
midspan area
G11-1 500 0.5 3.13 2.45 702.8 230.8 20.5 N/P
G11-2 500 0.9 4.20 4.41 971 380 33.2 N/P
G11-3 500 1.3 5.05 6.37 1461.5 372.4 43.1 N/P
G11-4 500 1.7 5.78 8.34 1877.8 379 55.5 N/P
G11-5 500 2.1 6.42 10.3 1764.8 379.3 67.2 N/P
G11-6 500 2.5 7.00 12.26 1906.6 408.1 83.4 N/P
G12-1 500 3 7.67 14.71 1675 397.5 85.4 N/P
G13-1 500 3 7.67 14.71 2150 598.7 60.6 N/P
G14-1 500 3 7.67 14.71 2258.1 539.6 63.7 N/P
G15-1 500 3 7.67 14.71 2062.9 630.4 40.5 N/P
G16-1 500 3 7.67 14.71 2022.6 567.7 52.9 N/P
Note: N/P- Not Provided
- 264 -

Dynamic Behavior of Reinforced Concrete Beams Under Varying Rates of Concentrated and Impact Loadings

Uploaded by

Copyright:

Available Formats

Dynamic Behavior of Reinforced Concrete Beams Under Varying Rates of Concentrated and Impact Loadings

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dynamic Behavior of Reinforced Concrete Beams Under Varying Rates of Concentrated and Impact Loadings

Uploaded by

Copyright:

Available Formats

This document is downloaded from DR‑NTU (https://dr.ntu.edu.

Dynamic behavior of reinforced concrete beams

Satadru Das Adhikary

Downloaded on 25 Aug 2024 12:35:26 SGT

SATADRU DAS ADHIKARY

SCHOOL OF CIVIL AND ENVIRONMENTAL ENGINEERING

Satadru Das Adhikary

School of Civil & Environmental Engineering

A thesis submitted to the Nanyang Technological University in partial

The research presented in this thesis on Dynamic behavior of RC beams under

First and foremost, I would like to express my most profound gratitude to my

Adhikary, S.D., Li, B., and Fujikake, K. “Dynamic Behavior of Reinforced

STATEMENT OF ORIGINALITY ........................................................................ i

CHAPTER 1 INTRODUCTION ............................................................................ 1

CHAPTER 2 LITERATURE REVIEW .............................................................. 11

CHAPTER 3 EXPERIMENTAL INVESTIGATION OF LOADING RATE

3.2 Analysis of Experimental Database................................................................ 46

CHAPTER 4 NUMERICAL SIMULATION OF LOADING RATE EFFECTS

4.5.4 Influence of transverse reinforcement ratio on DIF of maximum

CHAPTER 5 EXPERIMENTAL ASSESSMENT OF RC BEAM SUBJECTED

CHAPTER 6 FINITE ELEMENT MODELING OF RC BEAM UNDER

CHAPTER 7 ANALYTICAL MODEL FOR IMPACT RESPONSE

CHAPTER 8 PERFORMANCE OF IMPACT-DAMAGED RC BEAM ....... 203

CHAPTER 9 CONCLUSIONS AND FUTURE WORK.................................. 227

REFERENCES ..................................................................................................... 233

Reinforced Concrete (RC) structures may experience various types of loading

to impact loadings; hence simplified analytical models were employed to predict

Residual performance assessment of damaged structures has been gaining

Table 4.3: Specimen characteristics of simulation matrix ...................................... 94

Table 4.4: Beam characteristics for numerical case-studies ................................. 103

Table 5.1: Test program ........................................................................................ 113

Table 6.5: Variations in mass ratios ...................................................................... 176

Table 6.6: Variation in longitudinal reinforcement ratios ..................................... 178

Table 7.1: Actual and effective mass of beam ...................................................... 192

Table 7.2: Modified material properties ................................................................ 196

Table 8.1: Parameter variations in case-studies .................................................... 220

Fig. 1.1: Spectrum of strain rates corresponds to different loading environments

Fig. 2.1: Relative enhancement of flexural and shear strengths of RC beams

Fig. 2.2: Missile impact phenomena (Kennedy [K3]) ............................................. 26

Fig. 2.3: Failure mode of the RC beams (Mylrea [M12]) ....................................... 27

Fig. 2.5: Impact test setup (Kishi et al. [K7]) .......................................................... 29

Fig. 2.7: Simplified loop for reaction force-midspan displacement of RC beams

Fig. 2.14: Impact test set up (Tachibana et al. [T1]) ............................................... 37

Fig. 2.16: Proposed equation by Tachibana et al. [T1] ........................................... 38

between coefficient of residual deflection  rs and static flexural load-carrying

capacity Pusc by Kishi and Mikami [K10] ................................................................ 41

Fig. 3.3: Flow chart of test program ........................................................................ 51

Fig. 3.4: Dimensions of RC beams, layout of reinforcements, location of strain

Fig. 3.7: Schematic diagram of test set up ............................................................... 56

Fig. 3.12: Concrete contribution to the shear stress ................................................ 61

Fig. 3.20: Development of cracks in DR3.3_0.84_0.11 under high loading rates .. 70

Fig. 3.21: Development of cracks in DR3.3_0.84_0.15 under high loading rates .. 71

Fig. 4.1: Three-dimensional FE model of RC beams (a) a d -3.3;

(b) a d -4.4 series .................................................................................................... 79

Fig. 4.7: General schematic diagram of beam ......................................................... 93

Fig. 4.8: Influence of longitudinal reinforcement ratio on DIF of maximum

Fig. 4.9: Influence of longitudinal reinforcement ratio on failure mode: (a) a d -

Fig. 4.10: Influence of transverse reinforcement ratio on DIF of maximum

Fig. 4.11: Influence of transverse reinforcement ratio on failure modes of a d -3.3

Fig. 4.12: Influence of transverse reinforcement ratio on failure modes of a d -4.4