Damping

DAMPING CHARACTERISTICS OF REINFORCED
AND PARTIALLY PRESTRESSED CONCRETE BEAMS
A Thesis submitted in fulfilment of the requirements for

the award of the degree of
Doctor of Philosophy
by
SANAUL HUQ CHOWDHURY

B. Eng. (Hons), M. Tech. (Structures), MIEAust, CPEng (Civil)
from
School of Engineering
Faculty of Engineering
GRIFFITH UNIVERSITY
GOLD COAST CAMPUS
April 1999
Dedication
ii
To
My Parents
to whom I owe my very existence on the face of the earth
Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams
Declaration
iii
DECLARATION
This work has not previously been submitted for a degree or diploma in any
university. To the best of my knowledge and belief, the thesis contains no material
previously published or written by another person except where due reference is
made in the thesis itself.
____________________
Sanaul Huq Chowdhury
April 1999
Acknowledgements
iv
ACKNOWLEDGEMENTS
The author expresses his deepest gratitude and acknowledges his heartfelt
indebtedness to his supervisor, Professor Yew-Chaye Loo, Foundation Professor and
Head of the School of Engineering, Griffith University, for having provided this
research opportunity and for his enthusiastic and expert guidance, constant
inspiration and encouragement, and invaluable technical suggestions throughout the
course of this investigation. The author also deeply appreciates the help and support
received from Professor Loo in relation to the enhancement of his research skills and
in making it possible for him to attend various international conferences over the
past four years.
Special thanks are due to Mr. Tingkan Lu, a former research assistant of
Professor Loo, for making available the results of the first-stage beam tests to the
author; to the authors co-supervisor, Dr. Sam Fragomeni, for his general assistance
and, in particular, for going through the first draft of this thesis and making the
necessary corrections; to Miss Angela Salzmann and Messrs Rodney Stewart and
Mathew James, the 1997 Engineering Honours graduates, for their help rendered in
the fabrication and testing of the second-stage beams; to Messrs Gilles Ravenelli,
John Reukers and Charles Allport, the technical staffs of the School of Engineering,
for their invaluable assistance in the laboratory work.
Financial supports provided by Griffith University in form of a Postgraduate
Research Scholarship as well as for the purchase of the TDS 460A Digitizing
Oscilloscope and other related equipment are gratefully acknowledged.
The author is deeply indebted to his wife, Fahima Chowdhury, and daughter,
Nafisa Nawar Chowdhury, for their understanding, constant encouragement and
continuous moral support.
Last but by no means the least, the author is indebted to his parents and other
members of his family living in Bangladesh and Australia, for their understanding
and encouragement throughout the course of this research.
List of Publications v
LIST OF PUBLICATIONS
During the course of research work, the following papers have been published or
submitted for publication.
1. Chowdhury, S.H. and Loo, Y.C., Unified Formula for Prediction of Crack
Widths in Reinforced and Partially Prestressed Concrete Beams, July 1999,
32p. (Submitted for publication in the ACI Structural Journal).
2. Chowdhury, S.H., Loo, Y.C. and Fragomeni, S., Damping Formulae for
Reinforced and Partially Prestressed Concrete Beams, July 1999, 18p.
(Submitted for publication in the ACI Structural Journal).
3. Chowdhury, S.H. and Loo, Y.C., Damping Prediction Formulae for
Reinforced and Partially Prestressed Concrete Beams, Proceedings, Seventh
East Asia-Pacific Conference on Structural Engineering and Construction
(EASEC-7), H. Okamura and H. Shima (eds), Kochi University of Technology,
Kochi, Japan, August 27-29, 1999, Vol. 2, pp. 1063-1068.
4. Chowdhury, S.H. and Loo, Y.C., Formulae for Prediction of Damping in
Reinforced and Partially Prestressed Concrete Beams, Paper No. papj18,
CDROM Proceedings, International Conference on Theoretical, Applied,
Computational and Experimental Mechanics (ICTACEM 98), I.I.T. Kharagpur,
India, December 1-5, 1998, pp. 1-9.
5. Chowdhury, S.H. and Loo, Y.C., Prediction of Damping in Reinforced and
Partially Prestressed Concrete Beams, Proceedings, Sixth East Asia-Pacific
Conference on Structural Engineering and Construction (EASEC-6), Y.-B.
Yang and L.-J. Leu (eds), Taipei, Taiwan, R.O.C., January 14-16, 1998, Vol. 2,
pp. 979-984.
List of Publications vi
6. Waroonkun, T., Loo, Y.C. and Chowdhury, S.H., Precast Reinforced Concrete
Building Construction Procedures for Northern Thailand, Proceedings, Sixth
East Asia-Pacific Conference on Structural Engineering and Construction
(EASEC-6), Y.-B. Yang and L.-J. Leu (eds), Taipei, Taiwan, R.O.C., January
14-16, 1998, Vol. 2, pp. 1023-1028.
7. Chowdhury, S.H. and Loo, Y.C., Crack Width Formula for Reinforced and
Partially Prestressed Concrete Beams, Proceedings, International Conference
on Maintenance and Durability of Concrete Structures, P. Dayaratnam and N.V.
Ramana Rao (eds), JNT University, Hyderabad, India, March 4-6, 1997, pp. 4651.
8. Chowdhury, S.H., Loo, Y.C. and Wu, T.H., Crack Width Predictions of
Reinforced and Partially Prestressed Concrete Beams, Proceedings, Fourteenth
Australasian Conference on the Mechanics of Structures and Materials (14
ACMSM), A. Beasley et. al. (eds), University of Tasmania, Hobart, Tasmania,
Australia, December 11-13, 1995, Vol. 2, pp. 558-563.
9. Wu, T.H., Chowdhury, S.H. and Loo, Y.C., Embedment-Depth Effect
Modelling in Dynamic Analysis of Soil-Structure Interaction, Building for the
21st Century, Proceedings, Fifth East Asia-Pacific Conference on Structural
Engineering and Construction (EASEC-5), Y.C. Loo (ed), Gold Coast,
Queensland, Australia, July 25-27, 1995, Vol. 3, pp. 1865-1872.
10. Loo, Y.C. and Chowdhury, S.H., Inadequacies of Some Existing Code
Provisions for Punching Shear Strength Design and Effective Flange Width
Estimation, Proceedings, Third International Symposium on Concrete Model
Code for Asia, Tokyo Metropolitan University, Tokyo, Japan, March 17, 1995,
11p.
Synopsis
vii
SYNOPSIS
Advances in construction materials and computational methods have made it
possible to design and construct taller masts, buildings with increasingly slender
frames, and bridges (and roof structures) with ever larger spans. In addition, masts,
towers and new forms of construction such as offshore structures are being built in
more hostile environments than previously contemplated. These evolving structures
which keep extending the boundary of normal designs require that the designers
take into account vibration of structures at the design stage to a much greater extent
than they have done in the past.
The slenderness of modern structures and the large magnitude of the loads that
many of them must carry also make it imperative that such structures be designed
for stresses induced by dynamic disturbances. The response of a structure to a
dynamically applied load may be many times greater than its response to the same
load applied statically. The relationship between a structures static and dynamic
responses depends primarily on its damping characteristics and on its natural
periods of vibration. In fact, damping is one of the most significant contributors to
the dynamic response of high-rise buildings, bridges, tall chimneys and other
slender structures considered to be significantly affected by dynamic forces.
Under a severe lateral dynamic loading condition, the structure that is likely to
survive is one whose members are sufficiently ductile to absorb and dissipate energy
by elastic and/or inelastic deformation. This requires the designer to realistically
assess the possible levels of strength in flexural and shear elements. Thus, in
designing such a concrete structure, it is important to understand and determine the
ability of the structure to absorb energy under an external impulsive force. At this
stage, information in this regard is lacking in published literature and the ability of
the constituent elements of the structure to absorb energy is not well understood.
This, for example, is true for reinforced and partially prestressed concrete beam,
especially the cracked ones. In particular, no simple and accurate formulae are
available to evaluate the damping ratios of reinforced and partially prestressed
concrete beams cracked or otherwise, for use in the dynamic design of civil
Synopsis
engineering structures.
viii
It is this area which forms the primary focus of this
research.
In this research, an extensive test programme has been carried out to study the
cracking and damping behaviour of reinforced and partially prestressed concrete
beams.
The tests were carried out in two stages and involved a total of 30
reinforced and partially prestressed beams.
Nine reinforced and 12 partially
prestressed simply supported full-size box beams were tested at the first stage.
Tested at the second stage were 2 simply supported and 3 two-equal span
continuous reinforced full-size box beams and 4 solid rectangular full-size simply
supported reinforced beams.
For all the beams, at each level of loading,
measurements were made of instantaneous and residual crack widths, instantaneous

and residual concrete strains, and mid-span deflections.
Each beam was also
subjected to free vibration tests to measure its logarithmic decrement of damping

corresponding to each load level.
Based on the experimental results, two empirical formulae have been developed
for predicting logarithmic decrement of damping separately in reinforced and
partially prestressed concrete beams. These formulae predict damping from the
residual crack widths of the beams. For these formulae to be of practical use, a
formula relating the residual crack widths of concrete beams to the instantaneous
average crack widths was developed. In addition, a unified formula was derived for
the prediction of the instantaneous average crack widths based on the general beam
parameters. As an alternative, separate formulae are also presented for predicting
residual crack widths using mid-span deflections of reinforced and partially
prestressed beams.
These further enhance the practicability of the proposed
damping formulae.
In an effort to verify the accuracy and reliability of the proposed formulae,
comparative studies are carried out based on the authors own laboratory test
results as well as those available in published literature. In total, 104 full-size
reinforced and prestressed concrete solid and box beams are involved in the
comparative study. In general, good correlations are obtained for instantaneous
and residual average crack widths and for logarithmic decrement of damping
values. These are true for both reinforced and partially prestressed concrete beams.
Table of Contents
ix
TABLE OF CONTENTS
Title Page .. i
Dedication . ii
Declaration iii
Acknowledgements .. iv
List of Publications v
Synopsis vii
Table of Contents .. ix
List of Figures . xiii
List of Tables xx
Notation . xxiv
CHAPTER
1.
Introduction 1-1
1.1
General Remarks .. 1-1
1.2
Damping and Its Significance . 1-3
1.3
Damping as related to Cracking 1-4
1.4
Objectives and Scope 1-5
1.5
Layout of the Thesis . 1-6
2.
Damping - Theory and Concepts .. 2-1

2.1
2.2
Damping Defined .. 2-1
2.3
Nature of Damping 2-2
2.4
Representation of Damping .. 2-3
2.4.1 Viscous damping .. 2-4
2.4.2 Coulomb damping 2-4
2.4.3 Hysteretic damping .. 2-5
2.5
Measurement of Damping 2-5
2.5.1 Decay curve method . 2-6
2.5.2 Bandwidth method 2-7
2.5.3 Energy method .. 2-8
2.5.4 Phase shift method 2-9
2.6
Damping Mechanisms in Reinforced Concrete 2-9
2.7
Overall Damping of a Structure .. 2-11
2.8
Summary . 2-13
3.
Damping in Concrete Elements A Literature Review 3-1

3.1
3.2
Development of Damping Measurement Methods .. 3-2
Table of Contents
3.3
3.4
Review of Research on Damping of Concrete Structural

Elements .. 3-5
3.3.1 Plain concrete elements 3-5
3.3.2 Reinforced concrete elements . 3-14
3.3.3 Prestressed concrete elements . 3-21
Summary . 3-25
4.
Damping Theory of Simply Supported and Continuous Beams .. 4-1

4.1
4.2
Single Degree of Freedom (SDOF) Model .. 4-2
4.3
Undamped Free Vibration of a SDOF System . 4-3
4.4
Damped Free Vibration of a SDOF System . 4-7
4.4.1 Free vibration with viscous damping 4-7
4.4.2 Free vibration with Coulomb damping 4-15
4.4.3 Free vibration with hysteretic damping .. 4-17
4.5
Vibration Quantities and Their Relationships 4-19
4.6
Free Vibration of Beams and Their Damping Determination . 4-21
4.7
Summary . 4-23
5.
Cracking in Reinforced and Partially Prestressed Concrete Beams .. 5-1

5.1
5.2
Cracking in Reinforced Concrete Beams . 5-2
5.2.1 Causes of cracking 5-2
5.2.2 Types of cracking . 5-4
5.2.3 Factors affecting crack widths .. 5-5
5.2.4 Mechanism of flexural cracking ... 5-9
5.2.5 Evaluation of crack spacing and crack width . 5-16
5.3
Cracking in Partially Prestressed Concrete Beams . 5-23
5.3.1 Causes of cracking .. 5-24
5.3.2 Types of cracking 5-26
5.3.3 Factors affecting crack widths 5-26
5.3.4 Evaluation of crack spacing and crack width . 5-28
5.4
Summary . 5-33
6.
The Experimental Programme . 6-1

6.1
6.2
Background Theory .. 6-2
6.3
Reinforced Concrete Specimens ... 6-2
6.3.1 Simply supported beams ... 6-3
6.3.2 Two-equal-span continuous beams ... 6-4
6.4
Partially Prestressed Concrete Specimens .. 6-17
6.5
Materials . 6-22
6.6
Support Systems . 6-23
6.7
Loading System .. 6-26
6.8
Summary . 6-28
7.
Development of Instantaneous Crack Width Formula ... 7-1

7.1
7.2
Test Programme ... 7-2
Table of Contents
7.3
7.4
7.5
7.6
7.7
7.8
7.9
xi
Proposed Equation for Instantaneous Average Crack Width ... 7-4

Data Analysis and Development of the Proposed Formula .. 7-5
Relationship between Average and Maximum Crack Width ... 7-7
Comparison with Test Data .. 7-9
7.6.1 The present test beams .. 7-9
7.6.2 Clarks beams . 7-11
7.6.3 Chi and Kirsteins beams 7-18
7.6.4 Nawys beams . 7-22
Comparison with Code Formulae ... 7-27
Cracking Behaviour of Reinforced and Partially Prestressed
Concrete Beams .. 7-32
Summary . 7-34
8.
Residual Crack Width Measurements and Predictions .. 8-1

8.1
8.2
Test Procedure and Instrumentation . 8-2
8.3
Alternative Procedure of Residual Crack Width Measurement 8-4
8.4
Deflection at Mid-Span of Beam 8-9
8.5
Prediction of Residual Crack Width ... 8-14
8.5.1 Prediction from instantaneous crack width . 8-14
8.5.2 Prediction from mid-span deflection .. 8-16
8.6
Summary . 8-19
9.
Damping in Reinforced Concrete Beams . 9-1

9.1
9.2
Instrumentation and Data Recording 9-2
9.2.1 Loading system . 9-2
9.2.2 Vibration measurement . 9-2
9.2.3 Vibration measurement procedure 9-7
9.3
Presentation of Test Results .. 9-9
9.3.1 Over reinforced simply supported beams ... 9-10
9.3.2 Under reinforced simply supported beams . 9-10
9.3.3 Two-equal-span continuous beams . 9-11
9.4
Damping Formula for Reinforced Concrete Beams ... 9-11
9.4.1 Development of the damping formula 9-11
9.4.2 Comparison with test results ... 9-14
9.4.3 Verification of viscous damping characteristics . 9-25
9.5
Damping Behaviour of Reinforced Concrete Beams . 9-27
9.5.1 Effect of steel ratio .. 9-27
9.5.2 Effect of compressive strength of concrete . 9-28
9.5.3 Effect of span length ... 9-29
9.5.4 Effect of cracking 9-30
9.6
Summary . 9-31
10.
Damping in Partially Prestressed Concrete Beams .. 10-1

10.1 General Remarks 10-1
10.2 Instrumentation and Data Recording .. 10-2
10.3 Presentation of Test Results 10-2
10.3.1 Over reinforced beams 10-3
Table of Contents
10.4
10.5
10.6
11.
xii
10.3.2 Under reinforced beams .. 10-4

Damping Formula for Partially Prestressed Concrete Beams . 10-4
10.4.1 Development of the damping formula 10-4
10.4.2 Comparison with test results .. 10-7
Damping Behaviour of Partially Prestressed Concrete Beams .. 10-13
10.5.1 Effect of steel ratio 10-14
10.5.2 Effect of compressive strength of concrete ... 10-15
10.5.3 Effect of span length . 10-16
10.5.4 Effect of degree of prestressing 10-17
10.5.5 Effect of cracking .. 10-18
Summary ... 10-19
Conclusion 11-1
11.1 General Remarks 11-1
11.2 Prediction of Crack Widths and Damping . 11-1
11.2.1 Crack width formulae . 11-1
11.2.2 Logarithmic decrement of damping 11-3
11.3 Conclusions 11-3
11.3.1 Cracking behaviour . 11-3
11.3.2 Damping behaviour 11-4
11.3.3 Accuracy of the proposed procedures . 11-6
11.4 Recommendations for Further Research . 11-6
References ... R-1

Appendix A Crack Width Results for the Test Beams .. A-1
Appendix B Comparison of Measured and Calculated Crack Widths for the
Test Beams B-1
Appendix C Crack Patterns for the Test Beams C-1
Appendix D Instantaneous Crack Width/Strain Versus Loading Curves for the
Test Beams D-1
Appendix E Response Curves for Reinforced Concrete Test Beams E-1
Appendix F Damping Versus Residual Crack Widths for Reinforced Concrete
Test Beams F-1
Appendix G Response Curves for Partially Prestressed Concrete Test Beams .. G-1
Appendix H Damping Versus Residual Crack Widths for Partially Prestressed
Test Beams H-1
List of Figures
xiii
LIST OF FIGURES
Figure 2.1
Figure 2.2
Figure 2.3
Figure 2.4
Figure 2.5
Figure 2.6
Different types of damping

Free vibration response
Frequency response spectrum
Stress-strain diagram under cyclic loading
Damping ratio of a reinforced concrete element in different
states of cracking
Cracked bending element and the associated damping forces
Page
2-3
2-6
2-7
2-9
2-10
2-11
Figure 3.1
Figure 3.2
The hysteresis loop (after Ashbee et al., 1976)

Approximate logarithmic decrement for creep
(after Ashbee et al., 1976)
Figure 4.1
Figure 4.2
Figure 4.3
Figure 4.4
Figure 4.6
Figure 4.7
Equivalent SDOF mass-spring systems

System of free undamped vibration
SDOF system with viscous damping
Motion of a SDOF system with viscous damping ratio > 1
(overdamped case)
Motion of a SDOF system with viscous damping ratio < 1
(underdamped case)
SDOF model with Coulomb damping
Hysteresis curve
4-17
4-18
Figure 5.1
Figure 5.2
Primary and secondary flexural cracks

Members with cracking
5-5
5-11
Figure 6.1
Figure 6.2
Figure 6.3
Figure 6.4
Figure 6.5
Figure 6.6
Figure 6.7
Figure 6.8
Figure 6.9
Figure 6.10
Figure 6.11
Figure 6.12
Figure 6.13
Figure 6.14
Figure 6.15
Figure 6.16
Figure 6.17
Reinforcement details of beams 1, 2, 3, 4, 8 and 11

Reinforcement details of beam 5
Reinforcement details of beams 6 and 9
Reinforcement details of beams 19, 23 and 27
Reinforcement details of beams 20, 24 and 28
Reinforcement details of beams 21, 25, 26 and 29
Stress-strain curve for reinforcing steel
Stress-strain curve for prestressing steel
6-5
6-6
6-7
6-8
6-9
6-10
6-11
6-12
6-14
6-15
6-16
6-18
6-19
6-20
6-21
6-24
6-24
Figure 4.5
3-4
3-4
4-3
4-4
4-8
4-11
4-13
List of Figures
xiv
Figure 6.18
Figure 6.19
Figure 6.20
Figure 6.21
Beam support system for the first-stage beams

Beam support system for the second-stage beams
The loading system
The loading diagram
6-25
6-25
6-27
6-28
Figure 7.1
Figure 7.2
Crack detection microscope

Measured versus calculated average crack widths for the
test beams
Measured versus calculated average crack widths for Clarks
beams
Measured versus calculated average crack widths for Chi and
Kirsteins beams
Measured versus calculated maximum crack widths for
Nawys beams
Comparison of measured versus calculated average crack
widths for authors reinforced concrete beams
widths for Clarks beams
widths for Chi and Kirsteins beams
7-3
7-10
Figure 7.3
Figure 7.4
Figure 7.5
Figure 7.6
Figure 7.7
Figure 7.8
Figure 8.1
Figure 8.2
Figure 8.3
Figure 8.4
Figure 8.5
Figure 8.6
Figure 8.7
Figure 8.8
Figure 8.9
Figure 8.10
Figure 8.11
Figure 8.12
Figure 8.13
Figure 8.14
Figure 8.15
Figure 8.16
Figure 8.17
Figure 8.18
Figure 8.19
Demec strain gauge

Location of strain measurement
Instantaneous crack width/strain versus loading curves for
beam 5
beam 25
Residual average strain versus removal loading for
beam 2
beam 5
beam 20
beam 26
Load-deflection diagram for beam 5
Relationship between residual and instantaneous average
crack widths
Measured versus calculated average residual crack widths for
the second-stage beams (using Eq. (8.3))
Relationship between average residual crack widths and
mid-span deflections for reinforced concrete
7-17
7-21
7-27
7-28
7-30
7-30
8-3
8-3
8-5
8-6
8-7
8-7
8-8
8-8
8-10
8-10
8-11
8-11
8-12
8-12
8-13
8-13
8-15
8-16
8-17
List of Figures
Figure 8.20
Relationship between average residual crack widths and midspan deflections for partially prestressed concrete beams
Figure 9.1
Different methods of vibration pick-up mountings (after

Major, 1980)
Mounting of accelerometer on concrete beam
2630 Personal Fourier Analyzer
TDS 460A Digitizing Oscilloscope
Hammer excitation positions for the first-stage beams
Hammer excitation positions for the second-stage beams
Stages in a vibration measurement system
Measured versus calculated damping values for the
reinforced concrete beams
Ln (Ao/An) plotted against n for second-stage beams
Effect of steel ratio on the damping characteristics of
Effect of compressive strength of concrete on the damping
characteristics of reinforced concrete beams
Effect of span length on the damping characteristics of
Figure 9.2
Figure 9.3
Figure 9.4
Figure 9.5
Figure 9.6
Figure 9.7
Figure 9.8
Figure 9.9
Figure 9.10
Figure 9.11
Figure 9.12
Figure 10.1
Figure 10.2
Figure 10.3
Figure 10.4
Figure 10.5
Measured versus calculated damping values for the

partially prestressed concrete beams
Effect of steel ratio on the damping characteristics of
characteristics of partially prestressed concrete beams
Effect of span length on the damping characteristics of
Effect of degrees of prestressing on the damping
Figure C.1
Figure C.2
Figure C.3
Figure C.4
Figure C.5
Figure C.6
Figure C.7
Figure C.8
Figure C.9
Figure C.10
Figure C.11
Crack patterns for beam no. 3

Crack patterns for beam no. 7 (middle-third region)
Crack patterns for span-1 of beam no. 16
Crack patterns for negative span of beam no. 16
Crack patterns for negative span of beam no. 18
Figure D.1

beam 7
beam 10
Figure D.2
xv
8-18
9-4
9-4
9-5
9-5
9-6
9-6
9-8
9-15
9-26
9-27
9-28
9-30
10-13
10-14
10-15
10-16
10-18
C-1
C-2
C-3
C-4
C-5
C-6
C-7
C-8
C-9
C-10
C-11
D-1
D-2
List of Figures
Figure D.3
Figure D.4
Figure D.5
Figure D.6
Figure E.1
Figure E.2
Figure E.3
Figure E.4
Figure E.5
Figure E.6
Figure E.7
Figure E.8
Figure E.9
Figure E.10
Figure E.11
Figure E.12
Figure E.13
Figure E.14
Figure E.15
Figure E.16
Figure E.17
Figure E.18
Figure E.19
Figure E.20
Figure E.21
Figure E.22
Figure E.23
Figure E.24
Figure E.25
xvi

beam 11
beam 23
beam 27
beam 29
D-3
Response records for beam no. 1 at zero load (wr = 0)

Response records for beam no. 1 after application of 10 Kn
load (wr = 0.050 mm)
Response records for beam no. 2 after application of 80 kN
Response records for span 1 of beam no. 16 at zero load
(wr = 0)
Response records for span 1 of beam no. 16 after application
of 160 kN load (wr = 0.059 mm)
(wr = 0)
E-1
E-2
D-4
D-5
D-6
E-2
E-3
E-3
E-4
E-4
E-5
E-5
E-6
E-6
E-7
E-7
E-8
E-8
E-9
E-9
E-10
E-10
E-11
E-11
E-12
E-12
E-13
E-13
List of Figures
Figure E.26
Figure E.27
Figure E.28
Figure E.29
Figure E.30
Figure E.31
Figure E.32
Figure E.33
Figure F.1
Figure F.2
Figure F.3
Figure F.4
Figure F.5
Figure F.6
Figure F.7
Figure F.8
Figure F.9
Figure F.10
Figure F.11
Figure F.12
Figure F.13
Figure F.14
Figure F.15
Figure F.16

(wr = 0)
(wr = 0)
(wr = 0)
(wr = 0)
Logarithmic decrements versus residual crack widths for
beam 3
beam 4
beam 6
beam 7
beam 10
beam 11
beam 1
beam 2
beam 5
beam 8
beam 9
span 1 of beam 16
span 2 of beam 16
span 1 of beam 17
span 2 of beam 17
span 1 of beam 18
xvii
E-14
E-14
E-15
E-15
E-16
E-16
E-17
E-17
F-1
F-2
F-2
F-3
F-3
F-4
F-4
F-5
F-5
F-6
F-6
F-7
F-7
F-8
F-8
F-9
List of Figures
xviii
Figure F.17

span 2 of beam 18
Figure G.1
Figure G.2
Response records for beam no. 19 at zero load (wr = 0.050 mm) G-1
G-2
G-2
G-3
G-3
G-4
G-4
G-5
G-5
G-6
G-6
G-7
G-7
G-8
G-8
G-9
G-9
G-10
G-10
G-11
G-11
G-12
G-12
Figure G.3
Figure G.4
Figure G.5
Figure G.6
Figure G.7
Figure G.8
Figure G.9
Figure G.10
Figure G.11
Figure G.12
Figure G.13
Figure G.14
Figure G.15
Figure G.16
Figure G.17
Figure G.18
Figure G.19
Figure G.20
Figure G.21
Figure G.22
Figure G.23
Figure H.1
Figure H.2
Figure H.3
Figure H.4
Figure H.5

beam 20
beam 21
beam 25
beam 28
beam 19
F-9
H-1
H-2
H-2
H-3
H-3
List of Figures
Figure H.6
Figure H.7
Figure H.8
Figure H.9
Figure H.10
Figure H.11
Figure H.12

beam 22
beam 23
beam 24
beam 26
beam 27
beam 29
beam 30
xix
H-4
H-4
H-5
H-5
H-6
H-6
H-7
List of Tables
xx
LIST OF TABLES
Table 2.1
Table 2.2
Material damping of different materials (after Bachman

et al., 1995)
Logarithmic decrements, , for various construction materials
and in structures - system damping contributions
Table 6.1
Table 6.2
Table 6.3
Table 6.4
Details of reinforced concrete simply supported beams

Details of reinforced concrete continuous beams
Details of partially prestressed concrete box beams
Compressive strength of concrete for the test beams
Table 7.1
Parameters used in the development of average crack

spacing formula
Details of Clarks beams
Relation between measured and computed average crack width
for Clarks beams
Details of Chi and Kirsteins beams
for Chi and Kirsteins beams
Details of Nawys pre-tensioned beams
Details of Nawys post-tensioned beams
Measured versus calculated maximum crack width at steel
level of Nawys pre-tensioned beams
Measured versus calculated maximum crack width at steel
level of Nawys post-tensioned beams
Comparison of correlation statistics (Figures 7.6, 7.7 and 7.8)
Table 7.2
Table 7.3
Table 7.4
Table 7.5
Table 7.6
Table 7.7
Table 7.8
Table 7.9
Table 7.10
Table 9.1
Table 9.2
Table 9.3
Table 9.4
Table 9.5
Table 9.6
Table 9.7
Table 9.8
Table 9.9
Table 9.10
Damping data for beam 1

Relation between measured and calculated logarithmic
decrement values for beam 1
Page
2-12
2-13
6-4
6-13
6-17
6-23
7-6
7-12
7-13
7-18
7-19
7-23
7-24
7-24
7-25
7-32
9-12
9-13
9-13
9-14
9-17
9-17
9-18
9-18
9-19
9-19
List of Tables
Table 9.11
Table 9.12
Table 9.13
Table 9.14
Table 9.15
Table 9.16
Table 9.17
Table 9.18
Table 9.19
Table 9.20
Table 9.21
Table 10.1
Table 10.2
Table 10.3
Table 10.4
Table 10.5
Table 10.6
Table 10.7
Table 10.8
Table 10.9
Table 10.10
Table 10.11
Table 10.12
Table 10.13
Table 10.14
Table 10.15
xxi

decrement values for span 1 of beam 16
9-20

10-5
10-5
10-6
10-6
10-8
9-20
9-21
9-21
9-22
9-22
9-23
9-23
9-24
9-24
9-25
10-8
10-9
10-9
10-10
10-10
10-10
10-11
10-11
10-11
10-12
List of Tables
xxii
Table 10.16

10-12
Table A.1
Table A.2
Table A.3
Table A.4
Table A.5
Table A.6
Table A.7
Table A.8
Table A.9
Table A.10
Table A.11
Table A.12
Table A.13
Table A.14
Table A.15
Table A.16
Table A.17
Test results for beam 5

A-1
A-2
A-3
A-3
A-4
A-5
A-5
A-6
A-7
A-8
A-9
A-9
A-10
A-10
A-11
A-11
A-12
Table B.1

for beam 1
for beam 2
for beam 4
for beam 5
for beam 6
for beam 7
for beam 10
for beam 11
for beam 12
for beam 13
for beam 14
for beam 15
for beam 16
for beam 17
Table B.2
Table B.3
Table B.4
Table B.5
Table B.6
Table B.7
Table B.8
Table B.9
Table B.10
Table B.11
Table B.12
Table B.13
Table B.14
B-1
B-2
B-2
B-3
B-3
B-4
B-4
B-5
B-5
B-6
B-6
B-7
B-7
B-8
List of Tables
Table B.15
Table B.16
Table B.17
Table B.18
Table B.19
Table B.20
Table B.21
Table B.22
Table B.23
Table B.24
Table B.25
Table B.26

for beam 18
for beam 20
for beam 21
for beam 22
for beam 23
for beam 24
for beam 25
for beam 26
for beam 27
for beam 28
for beam 29
for beam 30
xxiii
B-8
B-9
B-9
B-10
B-10
B-11
B-11
B-11
B-12
B-12
B-12
B-13
Notation
xxiv
NOTATION
A
A1
Acef
Ae
An+1
As
Ast
B0
=
=
=
=
=
=
=
b
bw
c
=
=
=
c0
cc
ce
DMF
d
dc
E
Ed
Es
FD
f1, f2
=
=
=
=
=
=
=
=
=
=
=
fc
fD, fS
fd
=
=
=
fn
fnt
fr
fs
fsy
ft
ftn
g
h
=
=
=
=
=
=
=
amplitude of motion, area of concrete surrounding

each bar (= Ae/n)
initial amplitude of a free vibration
effective area of concrete in tension (=2.5b(h-d))
effective area of concrete in tension (= 2b(h-d))
amplitude at (n+1) cycle of a free vibration
area of reinforcing steel (in tension)
area of each reinforcing bar
maximum concrete area whose centre of gravity
coincides with that of the principal reinforcement
width of a section
web width
viscous damping constant, concrete clear cover,
distance from the point at which the crack width is to
be determined to the surface of the nearest
reinforcement bar
minimum cover to steel
critical damping constant
equivalent viscous damping constant
dynamic magnification factor
effective depth
concrete cover measured to the bar centre
strain energy at maximum amplitude
dynamic modulus of elasticity of concrete
elastic modulus of elasticity for steel
friction force
frequencies on either side of resonance at which the
amplitude of vibration is 70.7% of the maximum
compressive strength of concrete
damping force (hysteretic)
stress in the prestressing steel corresponding to the
decompression load
natural cyclic frequency of vibration in Hz (cycles per
second)
stress in the prestressing steel at any load level beyond
the first cracking load or decompression load
resonant frequency
stress in reinforcing steel
tensile yield strength of steel
tensile strength of concrete
flexural tensile stress in the concrete
acceleration due to gravity
total depth of a section
Notation
h1
h2
k
kd
L
l
lcr
lmax
lmin
M
m
N
n
n0
Q
S
s
T1
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
tb
Td
Tn
ts
U
u
un
upeak
upeak(0)
upeak(n)
u! n
!!u 0
!!u n
W
wcr
wcr,c
wcr,m
wmax
wmax,c
wmax,m
wr
X
x, u
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
x0 , u0
xmax
xn
=
=
=
xxv
distance from the centroid of tension steel to the

neutral axis
distance from the extreme tension fibre to the neutral
axis
spring constant, stiffness
neutral axis depth
span of the beam
length of the beam, spacing between cracks
average crack spacing
maximum possible crack spacing
minimum crack spacing
age of specimen in months
mass
normal force between sliding surfaces
number of cycles of vibration, number of bars
fundamental resonant frequency of the beam
measurement of damping by bandwidth method
specific damping capacity
average spacing between reinforcing bars
correction factor which was approximately unity for
the standard beams tested
bottom cover measured from the centre of the lowest
bar
period of damped vibration in seconds
natural period of vibration in seconds
side cover measured from the centre of the outer bar
vibrational energy available
average bond stress
displacement of a body (or particle) after n cycles
peak amplitude of the displacement
peak amplitude of the initial displacement
peak amplitude of the displacement after n cycles
velocity of the body (or particle) after n cycles
initial acceleration of the body (or particle)
acceleration of the body (or particle) after n cycles
weight of the mass m
average (instantaneous) crack width
calculated average crack width
measured average crack width
maximum crack width
calculated maximum crack width
measured maximum crack width
average residual crack width
evaporable water content
displacement of a body (or particle) from the reference
point (static equilibrium position)
initial displacement (at t = 0)
amplitude of vibration
amplitude or displacement at the end of nth oscillation
Notation
xxvi
xr
xs
x! , u!
x! 0, u! 0
!!x , !!u
=
=
=
=
=
amplitude or displacement at the end of rth oscillation

amplitude or displacement at the end of sth oscillation
velocity of the body (or particle)
initial velocity (at t = 0)
acceleration of the body (or particle)
c
E
=
=
=
fs
c
dry
=
=
=
=
max, wet
m
U
W
=
=
=
=
=
i
m
=
=
r
s
=
=
=
=
=
=
=
=
ratio of frequency of exciting force to natural

frequency of supporting structure
ratio of mass of rotating part to mass of supporting
structure
static deflection
cracked immediate deflection component
energy which is transformed into heat or internal strain
energy
net stress in the prestressing, or the magnitude of the
tensile stress in the normal nonprestressing steel at any
load level
change in average steel strain from the stage of
decompression
energy dissipated per cycle
logarithmic decrement
calculated logarithmic decrement values
logarithmic decrement of the specimen at dry
condition
logarithmic decrement of the specimen at wet
condition
measured logarithmic decrement values
actual work done, energy loss
load range
deflection range
average tensile strain in concrete at the level of
reinforcement
instantaneous average strain at the surface of the beam
steel strain at a crack less an empirical term due to the
stiffening effect of concrete tension between cracks,
and modified by the strain gradient term to obtain the
average strain at the extreme tension fibre of the
member
residual average strain at the surface of the beam
strain in the steel at a crack, axial tensile strain of bars,
strain of the reinforcement above the decompressed
stage
average bar diameter, diameter of reinforcing bar
phase angle
loss factor (hysteretic damping)
deflection coefficient
coefficients of static and kinetic friction
density of concrete, steel ratio
Notation
e, r
w
0
1 , 2
=
=
=
=
=
=
d
n
, e
=
=
=
=
=
=
xxvii
effective steel ratio (= As/Ae; As/Acef)

steel ratio (= As/bwd)
sum of the perimeters of the bars
sum of bar diameters
frequency of exciting force
frequencies, one either side of the resonant frequency,
where the amplitude is half that at resonance
resonant frequency
angular or circular frequency of vibration
damped angular natural frequency of vibration
natural circular or angular frequency of vibration
damping ratio, equivalent viscous damping ratio
hysteretic damping constant
Chapter 1 : Introduction
1-1
CHAPTER 1
INTRODUCTION
1.1
General Remarks
Structures planned or constructed in modern times are becoming bolder in

conception, more innovative in design, and more slender in dimensions. This is true
irrespective of their field of application, be it buildings, offshore structures, bridges,
aerospace structures, or mechanical structures. This is mainly due to the recent
advances achieved in the understanding of material behaviour and in analytical and
design techniques for structures of complex forms and geometry.
As modern structures tend to be more and more slender and flexible, the
peaks in the load spectrum and transfer function of a structure edge closer together,
and consequently the danger of dynamic problems occurring in a structure becomes
correspondingly greater.
Also, the risks of the load spectrum and the transfer
function of the structure having peaks in the same frequency range are particularly
critical when wind and earthquake loads prevail. Thus, the slenderness of modern
structures and the large magnitude of the loads that many of them must carry make it
imperative that they be designed for the stresses induced by dynamic disturbances.
Furthermore, the demands on a structure for serviceability have been
increased. With the environmental stress level being generally higher, people have
become more sensitive to vibrations. So, in the case of structures which are to be
1-2
used to a greater or lesser extent by people, the human sensitivity to horizontal and
vertical vibrations should be taken into account. The control of this perception of
motion has become a prime design parameter particularly for tall and slender
buildings with high aspect ratios (Grossman, 1990).
The response of a structure to a dynamically applied load may be many times
greater than its response to the same load applied statically.
The relationship
between a structures static and dynamic responses depends primarily on its

damping characteristics and on its natural period of vibration. In fact, the dynamic
response characteristics of a structure depend to a considerable extent on the
damping properties of the structure (James et al., 1964).
Quantification of damping is by far the most vexing problem in structural
engineering. Unlike the inertial and stiffness properties of a structural system,
damping does not refer to a unique physical phenomenon.
The estimates of
damping in a structural system have intrinsic variability as a result of the complexity

of damping mechanisms (Kareem and Sun, 1990). The ability to estimate damping
values accurately, at the design stages, would certainly alleviate a major source of
uncertainty in the design process.
Tests on actual buildings, field test structures and laboratory test structures
have shown that modern reinforced concrete structures with light damage have
modal damping ratios ranging from 2 to 3 percent; with moderate damage, from 3 to
5 percent; and with heavy damage, from 5 to 10 percent (Blume, 1970). Thus it is
evident that the extent of damage or, in turn, cracking in reinforced concrete
structures contribute significantly to structural damping. It should be equally true
for cracked prestressed and partially prestressed concrete structures. Hence, to
provide a fundamental basis for approaching the problem of predicting damping in
structures with any degree of confidence, it is essential to investigate the damping
characteristics of reinforced and partially prestressed concrete beams in terms of
progressive cracking.
1.2
1-3
Damping and Its Significance
Damping is a very complicated and specialized subject in itself but one which needs
a thorough understanding if it is to be incorporated effectively in any analysis.
Relatively few texts are available which deal with the subject in the depth it
deserves, but references can be made of two authoritative books those of
Lazan (1968) and of Nashif et al. (1985) - which are exceptions to this rule and
which deal with the many different facets of structural and machine damping in a
useful and practical manner. Much useful information on damping and its
mechanisms can also be found in the excellent recent text on vibration damping of
structural elements (Sun and Lu, 1995).
Damping is present to some degree in all structural systems, but the nature of
it and its magnitude are not well understood in any detailed sense. As a result of
experimental work and full-scale tests, it is possible to assign damping levels to
structural systems that bear some relation to reality. However, assignment of system
damping still requires educated guesses to be made for the most part (Irvine, 1986).
It is desirable for all structures to possess sufficient damping so that their
response to the expected excitation is acceptable. Increasing the damping in a
structure will reduce its response to a given excitation. Thus if the damping in a
structure is increased there will be a reduction in vibration and noise, and the
dynamic stresses in the structure will be reduced with a resulting benefit to the
fatigue life. Naturally the converse is also true.
However, it should be noted that increasing the damping in a structure is not
always an easy task. It can be expensive and may be wasteful of energy during
normal operating conditions. Before considering the methods for increasing the
damping in a structure, it is therefore necessary to be able to measure or more so to
predict structural damping accurately.
In any structure a number of mechanisms contribute to the total damping.
Different mechanisms may be significant at different stress levels, temperatures or
1-4
frequencies. Thus damping is both frequency and mode dependent, both as to its
mechanisms and its magnitude (Beards, 1983).
Consequently, damping in
engineering structures is not such an easily definable characteristic as the elastic and
inertial forces of a body. The task of finding a suitable method of predicting
damping in structures is a continuous concern in engineering research.
Damping is an important parameter in the analysis of structures under both
dynamic and static loads. The values used for analysis purposes are generally an
educated guess from a suggested range of values.
Therefore, there has been
significant interest in the prediction of damping values and their roles in structural
analysis.
1.3
Damping as related to Cracking
Damping is generally found to be one of the most structure-sensitive properties on

both the micro and macroscopic scale (Lazan, 1968). It has been shown that the
damping characteristics of a reinforced concrete beam may be influenced by many
factors, but depend strongly on the cracking state (Dieterle and Bachmann, 1981).
Material damping in reinforced concrete elements (and similarly in partially
prestressed concrete elements) in the quasi-elastic range (where there is no yielding
of reinforcement) shows some special features mainly due to cracking.
damping depends strongly on the stress intensity.
The
For low stress intensity,
corresponding to the uncracked state, relatively low damping ratio exists (i.e. <
1%). (Note that the damping ratio, , is defined in Section 4.4.1). With formation
of cracks the damping ratio increases. In the final cracked state but still with
relatively low stress intensity the damping ratio is relatively high, perhaps twice or
three times the value of the initial uncracked state (Bachmann et al., 1995).
The influence of cracking on damping is further exemplified by the damping
values given by Bock (1942). He gives values of damping value, presumed to be
damping ratio (), for concrete free from cracks of 0.0032 - 0.0064, and for
concrete with cracks of 0.0127 - 0.0207.
1-5
The mean values for logarithmic
decrement (see Section 2.5.1), , would be of the order of 0.03 (uncracked) and 0.1
(cracked). Cracking has also been identified by Cole (1965), Cole and Spooner
(1965a, 1965b), Jones and Welch (1967), Jordan (1980), and Askegaard and
Langs (1986) as contributing significantly to damping in concrete.
Despite these findings, there has been little or no significant effort given to
investigating the role of cracking in damping let alone establishing the relationships
between damping and cracking. In this research, crack width will be used as a basis
to investigate the effect of cracking on the damping characteristics of reinforced and
partially prestressed concrete beams.
1.4
Objectives and Scope
In addition to other factors, cracking in reinforced and partially prestressed concrete

elements significantly affects the damping characteristics of structures. Codes of
practice from different countries stipulate allowable limiting values on crack widths
in reinforced and partially prestressed concrete flexural members to control cracks or
to improve serviceability. But in practical situations, there will always be some
residual cracks in reinforced and partially prestressed concrete structures. As a
result, the damping characteristics of cracked concrete structures will be different
from those of the uncracked ones. Therefore, it is important to be able to predict the
inherent damping of real-life structures (with ever-present cracks) based on the
residual crack widths.
The main objective of this study was to conduct an experimental
investigation to provide a basis for the development of damping prediction formulae
for reinforced and partially prestressed concrete beams in terms of residual crack
widths. The residual crack widths, in turn, were to be related to instantaneous crack
widths and/or deflections under flexural bending of beams. Both instantaneous
crack width and deflection can be predicted from the general beam parameters.
Thus it would be possible to predict damping in reinforced and partially prestressed
1-6
concrete beams from the beam parameters. The extension of damping formula so
developed for simply supported beams to predict damping of continuous beams was
also envisaged.
These objectives have been achieved in this research through the completion
of the following tasks:
(1)
Development of a new unified formula for predicting instantaneous average

crack widths in reinforced and partially prestressed concrete beams.
relationship between the average and the maximum instantaneous crack widths
is also established.
(2)
Verification of the accuracy and reliability of the proposed instantaneous crack

width prediction formula. This is done by making comparisons with different
published and own laboratory test results, and with the performance of the
Australian, the British, the European and the American code methods.
(3)
Development of equations for predicting the residual crack widths from the
instantaneous crack widths and from mid-span deflections for both reinforced
and partially prestressed concrete beams.
(4)
Development of two damping prediction formulae, one each for reinforced and
partially prestressed concrete beams, and their subsequent verification based
on laboratory test results.
(5)
Establishment of the fact that the instantaneous crack width formula and the
damping prediction formula developed for simply supported beams are equally
applicable to the individual spans of two-equal-span continuous beams.
1.5
Layout of the Thesis
Following the introduction in this chapter, Chapter 2 presents the basic theory and
concepts of damping.
1-7
Apart from the nature and representation of damping
inclusive of the different methods of damping measurement, the discussion also

includes the concepts of damping mechanism in reinforced concrete and damping of
a structure as a whole.
A literature review on damping of concrete structures is given in Chapter 3.
It covers the damping measurement methods developed over the years as well as the
damping studies on different structural elements made up of plain, reinforced and
prestressed concrete.
Chapter 4 deals with damping theories of simply supported and continuous
beams. Special emphasis is given on the free vibration of single degree of freedom
systems with viscous, Coulomb or hysteretic damping.
Also discussed are the
different vibration quantities such as displacement, velocity and acceleration and the
relationships between them.
A theoretical basis for the proposed experimental
investigation is established in this chapter.

Extensive deliberations on the causes and types of cracking and the factors
affecting crack widths are made separately for reinforced and partially prestressed
concrete beams in Chapter 5.
A literature review is also carried out on the
mechanism of flexural cracking in reinforced concrete beams as well as on the

available crack spacing and crack width formulae for both reinforced and partially
prestressed concrete beams.
The test programme carried out to investigate the damping characteristics of
reinforced and partially prestressed concrete beams is described in Chapter 6 along
with the background theory. The description of the beams tested, the materials used,
the support systems, the test set-up and the loading systems are covered in full
detail. The tests were carried out in two stages. While the 21 beams of the first stage
were tested by Lu (1993), beams of the second stage (9 in total, including 3
continuous beams) were designed, fabricated and tested by the author with the
assistance of James (1997), Salzmann (1997) and Stewart (1997). The results from
both stages are included in the present study.
1-8
A new unified formula, to predict instantaneous average crack widths in

reinforced and partially prestressed concrete simply supported beams, is proposed in
Chapter 7. A relationship between the average and the maximum crack widths is
also developed. A comparative study is carried out in this chapter to establish the
accuracy and reliability of the proposed crack width formula.
This involves
different published and own laboratory test results, and the four codes of practice,
namely the Australian Standard (SAA, 1994), the British Standard (BS, 1985), the
European Standard (EC2, 1991) and the American Concrete Institute code (ACI,
1995). The applicability of the proposed formula to two-equal-span continuous
beams is also verified.
Residual crack width is identified as the basis for the damping prediction
formulae. Equations to compute the residual crack widths from the instantaneous
crack widths, and from the mid-span deflections of beams are developed in
Chapter 8. The direct and indirect measurement techniques and the underlying
theories to measure the residual crack widths are also discussed.
Chapter 9 deals with the development of the formula for prediction of
damping in reinforced concrete beams. The instrumentation and the measurement
procedures for recording the vibration data are described. Also presented are the test
results exemplifying the effects of different parameters on the damping values. The
proposed damping formula is verified based on the laboratory test results.
Its
applicability to continuous beams is checked.

A formula to predict damping in partially prestressed concrete beams is
presented in Chapter 10. Once again, the instrumentation and the data recording
process are described together with the presentation of the laboratory test results.
The validity of this proposed damping prediction formula is verified using these test
results.
Finally, Chapter 11 summarises the outcomes of this research, draws
associated conclusions and makes recommendations for further studies. Relevant
information and data are given in eight appendices.
Chapter 2 : Damping - Theory and Concepts 2-1
CHAPTER 2
DAMPING - THEORY AND CONCEPTS
2.1
General Remarks
This chapter attempts to define damping and discuss the basic theories and concepts
of damping relevant to the present research. The main theme of this thesis is
concerned with the development of damping prediction formulae for reinforced and
partially prestressed concrete beams. Thus, the damping mechanisms in reinforced
concrete are discussed in detail along with concepts of overall damping in completed
structures. The nature and the different ways of representing damping are also
discussed.
Laboratory measurement of damping constitutes a significant facet of the
experimental programme in this research. Accordingly, the different methods of
damping measurement and the related theories are also reviewed in this chapter.
2.2
Damping Defined
Damping defines the energy dissipation properties of a material or system under

cyclic stress (Lazan, 1968). Damping in a vibrating structure is associated with the
dissipation of mechanical energy, usually by conversion into thermal energy. The
energy dissipation equals the work done by the damping force. Thus, damping is
basically the dissipation of energy which occurs in vibrating systems.
2.3
Nature of Damping
The nature of damping is generally described as one of the following:

(a)
Material damping
(b)
System damping
(c)
Radiation damping.
Material damping is the energy dissipated within the materials of construction
and is due to internal hysteresis in materials arising from nonlinear stress-strain

behaviour, intergranular friction and thermoelasticity (Smith, 1988). It can be small
for most structural steels and some reinforced concrete, although materials like
timber laminates may possess high damping.
System damping is the energy dissipated at structural discontinuities, e.g.
bolted and riveted connections, construction joints in reinforced concrete structures.
This results from friction in sliding or fretting of joints, supports, cladding or various
other parts of the structure during relative motion (Smith, 1988). It is also referred
to as discontinuity damping.
Whereas material damping can be measured and
predicted by testing, it is much more difficult to predict the level of system damping.
Its degree may vary greatly even for the structures which are nominally identical.
Radiation damping is the energy dissipated by the structures environment,
e.g. air and water resistance, and foundations. Again this is difficult to predict.
Dissipation of energy in the foundations occurs even when the foundation material is
linearly elastic because of the propagation of stress waves through the foundation.
More important types of damping are shown in Figure 2.1 after
Bachmann et al. (1995).
Damping
Internal
External
Contact areas
within structures
Material
Hysteresis
(Viscous, friction,
perhaps yielding)
Relative motion
between substructures
(bearings, joints, etc.)
External contact
(non-structural
elements, energy
radiation to the
soil, etc.)
Figure 2.1 Different types of damping
2.4
Representation of Damping
Damping from all the sources, as discussed in Section 2.3, is represented in three
ways (not necessarily specifically related to a particular method of energy
dissipation):
(a)
Viscous damping
(b)
Coulomb damping
(c)
Hysteretic damping.
It must be stressed that these are only ways of representing damping. They do not
imply a mechanism for damping.
2.4.1
Viscous damping
In considering damping forces in the dynamic analysis of structures, it is usually

assumed that these forces are proportional to the magnitude of the velocity, and
opposite to the direction of motion. This type of damping is known as viscous
damping (Paz, 1997). This type of damping is analogous to the damping produced
by the motion of fluids.
As viscous type damping can be expressed in a simple mathematical way,
other more complex types of damping are often expressed as an equivalent viscous
damping in the analysis.
The assumption of viscous damping is often made
regardless of the actual dissipative characteristics of the system. In fact, there is a

widespread belief that if the gross nature of the dissipation is accounted for, the
actual details are irrelevant for engineering calculations. That is, the concept of
equivalent viscous damping holds sway (Irvine, 1986).
2.4.2
Coulomb damping
Coulomb or frictional damping can be regarded as existing when the damping force
is a constant (depending only on the normal reaction) and opposes the motion of the
body.
Damping in real structures is not strictly due to viscosity but is mostly caused
by friction at interfaces such as in bolted connections, in joints of cladding and in the
cracks of reinforced concrete. These frictional forces are independent of amplitude
and frequency, they always oppose the motion, and their magnitude may, to a first
approximation, be considered constant (Beards, 1983).
Experiments show that the vibrations of real structures usually lie between
viscous and frictional responses. However, the viscous assumption is convenient to
use analytically and is sufficiently accurate for most purposes (Smith, 1988).
2.4.3
Hysteretic damping
Experiments on damping that occurs in solid materials and structures which have
been subjected to cyclic stressing have shown the damping force to be independent
of the frequency of the exciting force. In order to model this internal or material
damping, the term hysteretic damping (Bishop and Johnson, 1960) was invented
such that
U = A2
(2.1)
where U is the energy dissipated per cycle, is the hysteretic damping constant,
and A is the amplitude of motion. It should be noted that in case of viscous
damping, U = c A2 where c is the viscous damping constant and is the
frequency of the exciting force.
This definition of hysteretic damping happens to coincide with the definition
of structural damping as stated by Clough and Penzien (1975) for steady state
excitation.
Clough and Penzien (1975) define structural damping for a single degree of
freedom system as being such that the damping force is proportional to displacement
and opposes the motion. Under any conditions other than steady state excitation,
this definition is somewhat dubious (Wilson, 1985). Nevertheless, at steady state, it
gives the same result as hysteretic damping as in Eq. (2.1).
Hysteretic damping is the result of friction between internal planes that slip
and slide during deformation of the body. The energy absorbed in this manner is
dissipated in the form of heat (Fertis, 1973).
2.5
Measurement of Damping
As discussed, there are several ways of defining the damping capacity of materials
and/or structures mathematically; in a given situation the procedure adopted will

depend mainly on the method of measurement employed.
The measurement of damping in concrete has always been difficult because
concrete is a very stiff material with relatively low damping.
Experimental
procedures have generally favoured testing by resonant methods and in particular the
decay of a freely vibrating system. There are a variety of techniques commonly
employed to measure the damping capacity.
2.5.1
Decay curve method
Also known as the free decay method, this is the most direct method. The damping
is obtained by measuring the decrease in amplitude of free vibrations (Figure 2.2)
and then calculating the logarithmic decrement, , as
1
n
A1
ln
An+1
(2.2)
where A1 is the initial amplitude and An+1 is the amplitude after n cycles.
n periods
Amplitude A
A1
An+1
Time t
Figure 2.2 Free vibration response
The structures are excited by an impact hammer for free vibrations and the
response is recorded at one or more points by accelerometers. Such a recorded
response is shown in Figure 2.2 schematically.
2.5.2
Bandwidth method
For an ideal damper and for small damping (i.e. << 1 or < 0.1, respectively), the
damping ratio can be obtained from the bandwidth, (2 - 1), from the resonance
curve due to a harmonic forcing function (see Figure 2.3).
Depending on the frequency response of the specimen around its resonant
frequency (Cole, 1965), the bandwidth is denoted by Q where
1
2 1
=
Q
3 r
(2.3)
in which r is the resonant frequency and 1 and 2 are the frequencies, one on
either side of the resonant frequency, at which level the amplitude is half that at
resonance. The various definitions are illustrated in Figure 2.3.
Amplitude A
Amax
Amax /2
1 r 2
Frequency
Figure 2.3 Frequency response spectrum
Instead of the displacement amplitude curve, the velocity or the acceleration

amplitude curves may also be used but only, of course, in cases where damping is
low. For relatively small values of damping capacity, the equivalent logarithmic
decrement, , can be obtained from the expression:
Q=
(2.4)
In most practical cases damping is evaluated by the decay curve method (as
discussed in Section 2.5.1) rather than by the bandwidth method (Bachmann et al.,
1995).
2.5.3
Energy method
Damping can also be measured by means of the specific damping capacity, S,

defined as the ratio of the strain energy lost per cycle to the total strain energy of the
specimen for that cycle. Or,
S=
E
E
(2.5)
where E is the energy which is transformed into heat or internal strain energy, and
E is the strain energy at maximum amplitude. The value of S can be determined
directly by measuring the area of the hysteresis loop of the stress-strain diagram of
the specimen under cyclic loading, as indicated in Figure 2.4.
Once again, according to Cole and Spooner (1965a), for small values of
damping, the logarithmic decrement, , is obtained as
S
2
(2.6)
Stress
E
E
Strain
Figure 2.4 Stress-strain diagram under cyclic loading
2.5.4
Phase shift method
In the case of oscillatory forcing functions, the stress (or the internal forces) is found
to lead the strain (or the deformation). The phase angle , which represents the lag
of strain behind the applied stress in the forced oscillations of a specimen, can be
used to evaluate the damping capacity.
According to Cole and Spooner (1965a), the logarithmic decrement, , can
be related to this phase angle, , as in the following expression:
= tan
2.6
(2.7)
Damping Mechanisms in Reinforced Concrete
The damping that occurs in reinforced concrete elements before yielding of the
reinforcing steel shows some special features mainly due to cracking. The damping
is also strongly related to the stress intensity. For bending elements or beams mainly
subjected to bending moments, the stress intensity may be characterised by the stress
amplitude in the bending reinforcements or by the displacement amplitude of the

beam, both defined at the point of maximum stress or displacement respectively.
The influence of stress intensity on the equivalent damping ratio, , of a beam
subjected to bending, can be interpreted in terms of cracking as shown in Figure 2.5
(Dieterle and Bachmann, 1981).
Figure 2.5 Damping ratio of a reinforced concrete element in different states of

cracking
As can be seen in Figure 2.5, when the beam is at low stress intensity and has
not yet cracked, a relatively low damping ratio exists. With the formation of cracks
the damping ratio increases. When the beam reaches the fully cracked state but still
with relatively low stress intensity, the damping ratio is at its largest value. This
damping value is probably twice or three times the value at the initial uncracked
state. With further increase in the stress intensity, there is a correspondingly rapid
decrease in the damping ratio values and it may even reach a value smaller than that
in the initial uncracked state.
This damping behaviour of reinforced concrete can be explained as follows:

(a) In the uncracked state, nearly pure viscous damping occurs in the concrete.
(b) In the cracked state, two types of damping occur: viscous damping in the
concrete in the uncracked compression zone; and friction damping due to friction
between the concrete and the reinforcing steel in the cracked tension zone.
Figure 2.6 Cracked bending element and the associated damping forces
The damping forces acting on a cracked bending element (as identified in (b)
above) can be diagrammatically represented as shown in Figure 2.6. It can be
shown that in bending elements, after full cracking has occurred, the viscous
damping component of becomes independent of the stress intensity, while the
friction damping component of decreases hyperbolically with increasing stress
intensity (Bachmann et al., 1995). This is the reason for the rapid decrease in the
values with increasing stress intensity once full cracking has taken place.
2.7
Overall Damping of a Structure
Depending on the location of energy dissipation the total overall damping of a

completed structure is a sum of the following contributions:
(a) damping of the bare structure;

(b) damping by non-structural elements; and
(c) damping by energy radiation to the soil.
While the first contribution always exists, the second and/or the third contribution
may be great or small or absent depending on the type and purpose of the structure.
Energy dissipation in the bare structure consists of material damping and
damping at bearings and joints. Material damping is predominant in most cases.
Table 2.1 shows material damping quantities (in terms of equivalent viscous
damping ratio, ) for different materials of construction.
Table 2.1 Material damping of different materials (after Bachman et al., 1995)
Material
Reinforced concrete
- small stress intensity (uncracked)
0.007 - 0.010
- medium stress intensity (fully cracked)
0.010 - 0.040
- high stress intensity (fully cracked but no yielding
0.005 - 0.008
of reinforcement)
Prestressed concrete (uncracked)
0.004 - 0.007
Partially prestressed concrete (slightly cracked)
0.008 - 0.012
Composite
0.002 - 0.003
Steel
0.001 - 0.002
Contribution from damping by non-structural elements depends on their

number, type and relative dimensions. This contribution may be greater than the
equivalent damping ratio of the bare structure alone. This explains why different
structures or structural types made of the same materials may have very different
values of .
Energy radiation to the soil by travelling waves may also contribute
significantly to the overall equivalent damping ratio. It depends more on the support
and/or foundation conditions than on the structure itself.
It is evident from the above discussions that the overall damping of a

completed structure is very much different form the damping of the material of
construction itself. This is further exemplified by the damping values for the basic
material, for beams with frictionless bearings and for bridges, respectively given in
second, third and fourth columns of Table 2.2 (Tilly et al., 1984). The damping
values are in logarithmic decrements, .
Similar results were also given by
Wheeler (1982).
Table 2.2 Logarithmic decrements, , for various construction materials and in
structures - system damping contributions
Steel
Concrete
Basic material
Beams
Bridges
0.002 - 0.008
0.004 - 0.030
0.02 - 0.06
0.01 - 0.06
0.02 - 0.06
0.02 - 0.20
Despite a considerable scatter, the type of construction produces a systematic

trend from non-prestressed concrete structures having the highest damping to
welded steel structures with the lowest damping (Bachmann and Ammann, 1987).
The determination of the overall damping of completed structures is, therefore, very
important.
2.8
Summary
In this chapter, the theories and concepts of damping relevant to present research are
briefly discussed. These include defining damping, classifying damping as well as
discussing the different damping models used in theoretical analyses.
None of the damping models mentioned so far offers an exact description of
damping in engineering. On many occasions, different models of damping are
required to be applied simultaneously. This complexity induced engineers in the
past to adopt models which provide simple solutions to the differential equations of
motion. Consequently, researchers were led to substitute an equivalent viscous
damping model in place of rather complex damping mechanisms. However, this
model should ensure the same amount of energy dissipation as in the original
damping mechanism (Argyris and Mlejnek, 1991).
Also discussed are the different methods of damping measurement. Inherent
structural damping is very important for the dynamic behaviour. Several damping
mechanisms are identified. Each of these mechanisms contribute to the damping
value (logarithmic decrement, ) measured in a vibration decay test (free decay
method) on structures in situ. Also, this is the most direct method and used in most
practical cases.
In the present research, the free decay method is used in the
experimental work.
Since the investigation of the damping characteristics of reinforced and
partially prestressed concrete beams with progressive cracking constitutes the main
essence of this research, the damping mechanisms in reinforced concrete are
thoroughly explored with special attention to cracking.
Finally, the difference between the overall damping of a completed structure
and the damping of its materials of construction is clarified with a view to justifying
the importance of damping determination of full-size reinforced and partially
Chapter 3 : Damping in Concrete Elements A Literature Review
3-1
CHAPTER 3
DAMPING IN CONCRETE ELEMENTS A
LITERATURE REVIEW
3.1
General Remarks
The damping characteristics of structural systems have been the subject of

investigation for well over a century.
Interest in damping study has increased
significantly in recent years. Engineers are particularly interested in the damping

properties of concrete structures.
This chapter presents an overview of the current levels of research on
damping, including measurement methods and theoretical developments.
This
review provides a brief perspective of damping capacity analysis and measurement

methods used in the studies of damping in general.
Also, the damping studies undertaken so far for different types of plain,
reinforced and prestressed concrete structural elements are discussed in this chapter.
This helps to highlight the dearth of any direct and accurate method available to
predict damping in such structures, especially from the many parameters defining
the component members.
3.2
3-2
Development of Damping Measurement Methods
The dynamic response of a structure is characterised by the modes of its vibration

and the associated natural frequencies and damping.
Damping measurement is
particularly important as the damping of structures cannot be predicted, although

estimates may be made from previously measured data and the calculations must
always be based on experimentally determined values (White, 1980). It was thought
that the determination of properties such as the dynamic modulus of elasticity and
damping might be of particular importance in the study of fatigue, creep, and
durability of concrete (Kesler and Higuchi, 1953).
Because concrete is a very stiff material with relatively low damping
(Ashbee et al., 1976), laboratory derived techniques have not been widely applied to
the damping study of concrete and reinforced concrete structures. Considering the
characteristics of concrete, some measurement and calculation methods have been
developed according to the basic requirements to determine damping capacity of
plain and reinforced concrete structures.
Based on the free decay method, bandwidth method and energy method (see
Section 2.5), some mathematical methods for determination of damping capacity
were developed. They are summarised in the following paragraphs.
Kesler and Higuchi (1953) used the sonic testing of concrete to determine a
relationship between dynamic modulus of elasticity, logarithmic decrement and
compressive strength of concrete. They represented the interrelation of these three
variables by empirically derived curves. The sonic test method has been recognised
for several decades as a useful and powerful non-destructive means for studying the
quality of concrete.
While logarithmic decrement may be obtained in several ways, according to
Kesler and Higuchi (1953) it may be obtained most easily by measuring the
sharpness of the resonance curve. In their experimental method, they computed
logarithmic decrement, , using the following equation given by Thomson (1940):
(f 2 f 1)
fr
3-3
(3.1)
where fr is the resonant frequency and f1and f2 are the frequencies on either side of
resonance at which the amplitude of vibration is 70.7% of the maximum.
The free decay method was chosen by Akashi (1960) to eliminate the
influence of frequency response for measuring the logarithmic decrement.
An
equation similar to Eq. (2.2) has been developed to calculate the logarithmic
decrement.
Akashi (1960) also developed an empirical relationship among the
logarithmic decrement (), the dynamic modulus of elasticity (Ed) and the
compressive strength (fc) of concrete which is independent of mix variation, age and
moisture content:
fc = Ed (1.79 +
0.346
x 10-4
)
(3.2)
This, however, necessitates the measuring of the dynamic modulus of elasticity for
computing the damping values.
Ashbee et al. (1976) described a method of measuring the damping
properties of concrete wherein the stress-strain or load-deflection hysteresis loop is
electronically modified to make it easier to examine the loop or measure its area.
For free vibrations, this is the natural logarithm of the ratio of two successive peaks
of the decay curve. This decay demonstrates that a proportional loss is equal to the
energy which must be supplied to maintain a constant amplitude, or forced vibration.
Thus, for a forced vibration system, the logarithmic decrement, which is shown
diagrammatically in Figure 3.1, is defined as:
Energy supplied (or lost)

Hysteresis loop area
=
Vibrational energy available
Vibrational energy
(3.3)
3-4
Load
U
U
2
hysteresis
loop
equilibrium
load
U
2
actual work done

= energy loss
= U
vibrational energy
available = U
Deflection
Figure 3.1 The hysteresis loop (after Ashbee et al., 1976)
If creep can be positively attributed to the cyclic energy input (see

Figure 3.2), Ashbee et al. (1976) suggested that an approximate contribution to
damping is:
Energy lost
Creep
=
Energy available
Deflection range
(3.4)
deflection range
input
Load range W
Load
recovered
Creep
(< )
Deflection
Figure 3.2 Approximate logarithmic decrement for creep (after Ashbee et al., 1976)
3-5
The uncertainties and the complicacies involved with the determination of creep
render Eq. (3.4) rather impractical for its use.
3.3
Review of Research on Damping of Concrete Structural Elements
There have been substantial investigations into specific damping characteristics of

concrete and/or concrete structural elements in the past.
The investigations,
however, concentrated on relationships between damping characteristics and the

factors that influence them, such as mix composition, age, water content, the
modulus of elasticity and the strength of concrete, to name a few. Strangely, there
has been little or no effort given to investigating the role of cracking in damping
despite its being identified as a significant contributor to damping by many
researchers.
Apart from for full-scale buildings, damping studies have been carried out on
plain concrete, reinforced concrete and prestressed concrete structural elements
using different experimental methods and equipment.
These investigations are
reviewed in the following sub-sections.
3.3.1
Plain concrete elements
It has been suggested that the damping characteristics will provide an additional
criterion for assessing the quality of concrete. This is because the damping of
vibrations in concrete is thought to be associated with the presence of water and air
voids, microcracks and acoustical mismatch at the boundaries of different
components in concrete. Thus the measurement of damping properties, especially at
low stress and low frequency, can aid the understanding of the internal structure of
concrete.
To date, there have been a few investigations on the damping properties of
plain concrete elements with basically two different objectives as follows:
(a)
3-6
to establish relationships between the damping properties, the dynamic

modulus of elasticity and the compressive strength;
(b)
to investigate the effects of various parameters such as composition and

water content of the mix, state of curing and age, frequency of vibration and
longitudinal compressive stress.
Kesler and Higuchi (1954) suggested that the dynamic modulus of elasticity
of concrete decreased by up to 10% for an increase in frequency from 1000 to 4000

Hz. This conclusion was drawn from measurements of the fundamental flexural
resonances of beams made of nominally the same concrete but with different
dimensions.
To study this effect of frequency on the dynamic Youngs modulus and also
on dynamic shear modulus, Jones (1957) carried out experiments to determine the
resonant frequencies of concrete beams in flexural, longitudinal and torsional modes
of vibrations at the fundamental and higher harmonics.
Jones (1957) tested long plain concrete beams having the following nominal
dimensions:
beam A
5 ft (1524 mm) x 6 in. (152 mm) x 3 in. (76 mm)
beam B
7 ft 6 in. (2286 mm) x 6 in. (152 mm) x 3 in. (76 mm)
beam C
6 ft (1829 mm) x 4 in. (102 mm) diameter (circular section)
Flexural vibrations were produced about both the 3 in. (76 mm) and 6 in. (152 mm)
dimensions. Supports were placed at 0.224 times the length of the beam from each
end for the fundamental resonance and at the centre of the beam for even order
harmonics. In the longitudinal and torsional modes of vibration, the beam was
usually supported at its centre, which is the appropriate node for the fundamental
and odd harmonics. The effect of frequency on the dynamic Youngs modulus had
been studied within the frequency range from 70 to 10000 Hz and the dynamic shear
modulus within the frequency range from 700 to 10000 Hz.
3-7
From his experimental results, Jones (1957) observed that for the moist beam
the Q values (see Section 2.5.2) in extension and torsion were approximately equal
and did not change appreciably over the frequency range. The test results implied
that damping in shear and in volume were interdependent and probably arose from
the same mechanism. It seems likely, therefore, that damping of vibrations arises
chiefly at the interfacial boundaries in the concrete.
Earlier Kesler and Higuchi (1953) studied the influence of water-cement
ratio, age and curing on the logarithmic decrement by testing 300 standard 6 in.
(152 mm) by 12 in. (305 mm) plain concrete cylinders. They concluded that if the
curing conditions are the same, the logarithmic decrement decreases with an
increase in age. Also, the logarithmic decrement decreases as the moisture content
of the specimen decreases and the former becomes less dependent on water-cement
ratio as the moisture content decreases.
The effect of age, water content and longitudinal compressive stress on the
damping capacity was studied by Cole (1965). He measured the damping capacity
of small cement paste specimens, small mortar specimens and large concrete
columns at low frequencies. The majority of the measurements have been of the
logarithmic decrement of the decay in the amplitude of the specimens in free
flexural vibration. In the case of the small specimens (45 cm x 2.5 cm x 0.6 cm) a
number of measurements of the damping capacity were obtained from the response
of the specimen to forced oscillations at frequencies near resonance.
The damping capacity of all specimens showed a marked degree of
sensitivity to moisture content. The test results indicated that if a specimen less than
one month old was removed from a water bath or humidity cabinet, the logarithmic
decrement (), when measured within 1 or 2 hours, was approximately = 0.050.
As the specimen dried out, the logarithmic decrement decreased. This fall was quite
rapid until an almost constant value had been reached. Thus the condition of a
specimen within a period of 1 hour of withdrawal from a water bath was defined as
wet and a specimen in which the damping capacity has reached the near constant
value as dry.
3-8
It was found that the logarithmic decrement depended linearly upon the
evaporable water content; that is, the logarithmic decrement, , of a specimen at a
particular water content, X, was
= 0.46 max . X + 0.54 max
(3.5)
where max was the logarithmic decrement of the specimen in the wet condition
and the evaporable water content X was assumed to vary between X = 1 when the
specimen was in the wet condition to X = 0 when the specimen reached the dry
condition. Thus, for a particular specimen the lowest value of the logarithmic
decrement will be in the dry condition and will approximately equal 0.54 max.
A linear regression between the logarithmic decrement values for both dry
and wet specimens resulted in the following relationship between in wet and
dry conditions:
dry = 0.55 wet
(3.6)
This was in good agreement with the figure yielded by Eq. (3.5).
It was also concluded that at amplitudes below 0.2 mm, no amplitude
sensitivity occurred and the logarithmic decrement was only slightly affected by the
longitudinal compressive stress. For a specimen which was cracked, the damping
varied considerably with the degree of longitudinal stress and also showed
considerable amplitude sensitivity at all measured amplitudes.
Cole (1965) also gave an equation to calculate the approximate value for the
logarithmic decrement, , for a specimen of evaporable water content, X, and age,
M, as
= 0.026 + 0.022 X - (0.0039 + 0.0030 X) loge M
(3.7)
where M is the age of the specimen in months.

3-9
One other important finding was that the values of damping capacity
measured for the small specimens agreed well with the values obtained by such
investigators as Jones (1957) and Holand (1962). This happened even though the
specimens differed appreciably in size and composition and the damping capacity
was measured at a different frequency.
Cole and Spooner (1965a) carried out measurements of the logarithmic
decrement of hardened cement paste specimens vibrating in flexure at frequencies
from 4 Hz to low frequency.
The apparatus which was used in the test can give a direct measurement of
the logarithmic decrement. For a given specimen, the frequency of oscillation was
dependent only upon the moment of inertia of heavy beams and an alteration in the
moment of inertia of these beams caused no change in strains occurring in specimen.
It was concluded that the variation of logarithmic decrement with the amplitude was
slight and also the variation of logarithmic decrement with frequency at low
frequency was relatively small. Further, it was generally found that over the range
investigated, the damping increased linearly with the maximum applied strain
amplitude and the increase was more marked at lower frequencies. The logarithmic
decrement remained constant until approximately 2.5 Hz, below which it increased
with decreasing frequency. At 0.18 Hz, the logarithmic decrement was found to be
approximately 45% higher than the value at frequencies above 2.5 Hz.
Cole and Spooner (1965b) also carried out laboratory experiments with the
purpose of obtaining fundamental information about the internal structure of
concrete with a view to obtain the damping coefficient for a particular specimen or
complex structure.
The results of their laboratory measurements on concrete
indicate that the damping of concrete is affected differently by different parameters.

The details are as follows:
(a)
The effect of composition.
It was found that the damping capacity of
concrete varies with the composition of concrete but the variation is

relatively small.
When one considers the very large variations of
3-10
composition that can occur in concrete, the resulting change in damping

capacity appeared very small when compared with those which sometimes
occur for a small impurity in a metal.
(b)
The effect of age and water content.
In general, it appears that the
logarithmic decrement of both wet and dry specimens decreases with age at a
decreasing rate, the logarithmic decrement at thirty months being
approximately half that at one month in both cases. The effect of the water
content is such that the logarithmic decrement of a saturated cement paste
specimen decreases to half its initial value as 15% of its weight is lost by
water evaporation.
(c)
The effect of the mode of vibration. The effect of the mode of vibration on
damping capacity was discussed and it was indicated that the damping
coefficient in moist concrete was approximately the same for all three modes
of vibration (flexural, longitudinal and torsional). However, in some cases
the damping capacity for the torsional mode was, at an early age, lower than
that obtained in longitudinal vibration.
(d)
The effect of static compressive stress. A number of researchers have found

that the value of the damping capacity of a specimen is dependent on the
applied longitudinal compressive stress, but Cole (1965) found little change.
There is agreement, however, on the fact that any effect the compressive
stress may have on the damping capacity is due primarily to prior damage in
the specimen. It would appear that the application of stress tends to restrict
the relative movement of surfaces in cracks.
(e)
The effect of dynamic strain amplitude. The value of the damping capacity is
dependent on the dynamic strain amplitude at which it is measured. It was
found that with small cement paste and mortar specimens vibrating in
flexure, the logarithmic decrement varied linearly with the maximum
dynamic strain amplitude, but the slope appeared to change somewhat
arbitrarily from specimen to specimen, certainly not being simply related to
3-11
age or condition. This linear variation was found for both increasing and
decreasing values of the maximum dynamic strain amplitude.
An extensive experimental investigation on the damping behaviour of
concrete has been carried out by Jones and Welch (1967) at the Road Research
Laboratory (U.K.). They described their experimental work based on measuring the
vibrational damping coefficients using the bandwidth method (see Section 2.5.2).
The object of the investigation was to obtain more information on the effects of
changes of composition of concrete on the damping properties and on the
relationships between damping coefficient, dynamic modulus of elasticity and
strength of concrete.
A total of 195 beams were tested. The specimen sizes adopted for the main
investigations were 6 in. (152 mm) by 6 in. (152 mm) by 28 in. (711 mm) beams for
concretes with 1
1
in. (38 mm) aggregate and 4 in. (102 mm) by 4 in. (102 mm) by
2
20 in. (508 mm) beams for concretes with
3
in. (19 mm) aggregate. All the
4
concrete beams were tested in the longitudinal mode of vibration as they were
supported at the centre, which is a nodal plane for the longitudinal resonances of the
beam at its fundamental and odd harmonics. The fundamental resonant frequency of
the beam (n0) was related to the dynamic modulus of elasticity (Ed) by the relation:
Ed = 4 n02 l2 T1
(3.8)
where l is the length of the beam, is the density of concrete and T1 is a correction
factor which was approximately unity for the standard beams tested.
Measurements were made of the damping in additional mortar and concrete
beams at ages between 1 and 28 days. The mixes tested all had a water-cement ratio
of 0.5. The values of Q (see Eqs. (2.3) and (2.4)) were normalised at 28 days and
the variation of Q/Q28 with age is given by:
Q
1
t
log
=1+
Q28
2
28
3-12
(3.9)
where t is the age of the specimen in days, and Q and Q28 are the Q values measured
at t and 28 days respectively. Changes in the composition of the concrete indicated
that the damping introduced at the interfaces between the coarse aggregate particles
and the sand-cement mortar was less than the damping within the sand-cement
mortar.
Separate relationships were obtained between the Q-value and the dynamic
modulus of elasticity, Ed, for the mortar and for each concrete containing a specific
type of coarse aggregate, irrespective of age and mix proportions.
These
relationships were of the form:

Q Ed2
(3.10)
Thus if all of the variations in the dynamic moduli arose from variations in
composition between nominally identical beams then the expected variations in Q
from this source would be about twice as great.
Finally, no significant difference in damping coefficient could be detected
with changes in frequency of vibration, size of specimen or method of measurement.
Also, there were well-defined relationships between the Q value and the
compressive or flexural strength of the mortar and of concretes containing certain
type of coarse aggregate.
Swamy (1970) studied the damping mechanisms in hardened pastes, mortars
and concrete, and indicated that the magnitude of the damping capacity of paste,
mortar and concrete is known to be related to the percentage of water-filled pores in
the system. Increasing the water content leads to a higher degree of damping for
pastes, mortars and concrete, particularly at early ages. Test results have shown that,
for cementitious materials, the loss of moisture and the presence of microcracks
have opposing influences on damping, and that the presence of microcracks
3-13
increases damping. With drying mortar and concrete specimens the increase in
logarithmic decrement due to microcracking is smaller than the decrease due to loss
of water, so that dry specimens possess smaller damping capacity than wet
specimens.
The results of an investigation to study the effect of stress, frequency, curing,
mix and age upon the damping of concrete were presented by Jordan (1980). The
energy method (see Section 2.5.3) was used to determine the damping capacity.
Forty specimens were tested and all specimens were 450 mm in length and 150 mm
in diameter. One outstanding feature of the results reported by Jordan (1980) was
that they showed the damping of dry concrete as greater than that of wet concrete for
the test conditions selected.
Previous work, however, has shown exactly the
opposite (Cole,1965; Swamy ,1970; Swamy and Rigby, 1971). The principal reason
for this discrepancy appears to be associated with the microcracking and its
contribution to Coulomb damping. The results also indicated that the component of
damping derived from microcracking was by far the most dominant feature.
Because of the large dependence of damping on the maximum applied stress, it is
believed that much of the difference between these results and others can be
attributed to load-induced cracking as well as shrinkage cracking. Thus the figure
for damping adopted for design purposes should reflect the effects of factors that are
associated with microcracking.
In an interesting development, Zech and Setzer (1988) measured the dynamic
elastic modulus and damping coefficient of hardened cement paste by exciting the
natural mode of oscillation of a hardened cement paste beam in a temperature range
between +20 and -160 C. They proposed a new statistical model for combining
the elastic moduli of the constituents (solid matrix, water, ice and air) to calculate
the dynamic elastic modulus of hardened cement paste.
In their experiments, a completely sealed measuring chamber was
constructed which can be cooled to -160 C. The small beam of hardened cement
paste was placed on two knife edged supports at the nodal points. Using a hammer a
natural mode of oscillation was excited. There was almost no damping due to the
3-14
measuring system. The oscillation was measured by a strain gauge glued to the
bottom of the beam. It was recorded via a Wheatstone bridge connected to a
transient recorder and evaluated by a microcomputer.
All specimens tested were
water-saturated.
Earlier, Cole and Spooner (1965b) reported that the damping capacity of
mortar specimens measured at a frequency of 1170 Hz did not change by more than
20% between 0 and 100 C except for a peak which occurred at 77 C. They also
found that at low frequencies, the damping capacity increased with increasing
temperature, the value being doubled as the temperature rose from 20 to 90 C. But
Zech and Setzer (1988) found the damping maximum to be between -60 and
-160 C and after a decrease it started to increase with increasing temperature
beyond -50 C or so. This increase in damping value at lower temperatures was due
to an increased internal friction resulting from the solidification of the last unfrozen
water layer between the ice and the solid matrix. Similarly the dynamic elastic
modulus also increased until -160 C because of the ice formation. The remarkable
increase below -50 C was once again due to the freezing of the last unfrozen thin
water film on the internal surface of the cement paste.
3.3.2
Reinforced concrete elements
To date, a number of investigations had been made into the dynamic behaviour and
response of concrete and, as such, the effects on the damping characteristics of
reinforced concrete beams, slabs and other structural elements. These investigations
are discussed herein to illustrate the significance of and the complexities involved in
determining the damping characteristics of such structural elements.
Penzien and Hansen (1954) carried out laboratory investigations into static
and dynamic elastic behaviour of reinforced concrete beams. The primary objective
of the dynamic phase of their investigation was to study the elastic behaviour of
simple structural elements under the action of impulsive loads.
Information
obtained during a single test were strain versus time records of the concrete and steel
3-15
at various points on the structural element, a deflection-time record, and the applied
load versus time. The dynamic magnification factor (i.e. ratio of maximum effective
strain produced at a point under dynamic loading conditions to effective strain
produced under equivalent static loading conditions) was the principal item studied.
They found that maximum strains produced in a structural element subjected
to an impulsive load might be considerably larger than those produced in the same
element when a static load of equal magnitude is applied. However, the internal
damping existing in reinforced concrete members subjected to dynamic forces
reduces these maximum strains considerably. Therefore, it is necessary to obtain a
measure of damping, which exists in concrete structures, if true maximum strains are
to be calculated for a given dynamic loading condition.
Experimental results showed that the damping, which existed in reinforced
concrete elements, had characteristics similar to those of viscous damping. The
damping coefficient or percent of critical damping, which existed in reinforced
concrete elements of the type tested, varied between 6 and 8 percent. Also, the same
damping coefficient applied reasonably well to all modes of vibration. Furthermore,
a reinforced structure designed to withstand impulsive loads should be underreinforced and both the dynamic magnification factor and increase in yield stress of
reinforcing steel should be considered in the design.
Sixteen concrete beams were subjected to sinusoidal exciting forces of
varying magnitudes by James et al. (1964) for the purpose of evaluating the flexural
rigidity and internal damping properties. Experimental data were obtained for both
reinforced and prestressed concrete beams fabricated with Haydite aggregate and a
siliceous type aggregate. The percent of reinforcement was varied for the reinforced
concrete beams. All of the beams were 4 in. (102 mm) x 8 in. (203 mm) in crosssection and of 10 ft (3048 mm) long spans.
The results obtained from the forced vibrations of prestressed and reinforced
concrete beams indicated that the damping characteristics of prestressed and plain
reinforced beams were in general the same. The damping did not appear to be
3-16
viscous for small amplitudes of vibration but as the amplitudes increased, the
dynamic magnification factors appeared to approach a constant value, which is a
characteristic of viscous damping. Repeated tests on the reinforced concrete beams
showed that the test history had a significant effect on the dynamic response
characteristics.
Jordan (1977) carried out a study to determine tensile stress effects on
damping for reinforced concrete members. The energy method (see Section 2.5.3)
was used to determine the logarithmic decrement. One plain and three reinforced
concrete specimens were tested. All specimens were 600 mm long and 150 mm in
diameter. The results suggested that as the tensile stress increased cracking occurred
which caused failure in plain concrete and led to the transfer of load from the
concrete to the reinforcing rods in the reinforced concrete. In reinforced concrete
specimens this was indicated by a sharp drop in tensile modulus to a level dependent
on the percentage of reinforcement. Jordan (1977) concluded that the material
damping of reinforced concrete increased dramatically as tensile stresses were
increased.
A detailed study of the probabilistic dynamic response of reinforced concrete
beams and frames was presented by Ellyin and Chandrasekhar (1977). They found
that the natural frequencies of beams and frames had generally about 11%
deviation from their corresponding mean values. This result points out that if the
natural frequencies of a structure are calculated by a deterministic method and are at
least 10% removed from the frequency of the exciting force, then the probability
of occurrence of the resonance will be very small. Therefore, the present margin of
20% employed in the design practice is too conservative, and for most cases it is
difficult to achieve. The present margin could thus be safely reduced by 50%.
Another interesting result of this study was the widespread of the distribution
of the response amplitude of beams and frames. Depending upon the value of and
, the extreme deviation of the dynamic response of beams from the mean was
found to vary from approximately -99% to 280%. While was the ratio of the
frequency of the exciting force to the natural frequency of the beam, was the ratio
3-17
of the mass of the rotating parts to that of supporting beam. In the case of frames,
the same variation was found for any < 0.9 and any . But for 0.9 and any ,
the variation was found to be much higher. A value of < 0.9 is suggested for all
practical purpose, both in the case of beams and frames.
The dynamic behaviour of floors was investigated by Rainer and
Pernica (1981) by constructing a 28 ft (8534 mm) x 29 ft (8839 mm) single bay floor
sample. Their research was primarily concerned with an examination of the effect
of four different test methods on the modal damping ratios determined for the
fundamental mode of vibration of the floor sample.
The damping ratios were
evaluated using free decay method (see Section 2.5.1) and half power bandwidth
method (see Section 2.5.2). It was found that the commonly used heel impact test
gave consistently higher modal damping ratios for the floor sample than given by
shaker impact, white noise or steady state shaker test. This was recognised to be the
result of the additional damping contributed to the floor sample by the person
performing the heel impact. Also, increasing the number of persons present on the
floor resulted in a proportional increase in the measured damping ratios and the
values obtained from the heel impact test varied with the location of the impactor.
Their results implied that (a) the damping values will be different using
different excitation methods, and (b) additional loading on the structure will
influence the damping values.
Flesch (1981) studied the behaviour of reinforced concrete cantilever
elements. Material damping as well as hysteretic slip damping between steel and
concrete was taken into account.
The energy dissipated during one cycle of
vibration was calculated and equated to that of an equivalent viscous damping

mechanism in order to evaluate an equivalent damping ratio.
According to Flesch (1981), for uncracked concrete members, damping
increased slightly with increasing cross sectional dimensions of concrete beams,
with increasing steel area and with increasing elastic modulus of concrete. The
3-18
values of damping ratio for the uncracked reinforced concrete cantilever members
were found to be 0.01 0.02.
Unlike for the uncracked members, for cracked concrete members, damping
increased significantly with increasing cross sectional dimensions of concrete beams
and decreased greatly with increasing percentage of reinforcement. The effect of
elastic modulus of concrete on damping values of cracked concrete members,
however, was the same as for the uncracked members.
Dieterle and Bachmann (1981) carried out research to develop damping
models with and without cracking for bending elements. Mathematical models of
reinforced concrete beams and the influence of the crack condition were also
presented.
The two chief causes for the damping of the reinforced concrete
structures as identified in this paper were:

(a)
The material damping of reinforced concrete building components and

reinforced concrete structures is influenced not only by the damping
properties of the material used, but also strongly by the crack condition. For
uncracked structures, especially fully prestressed uncracked structures, the
damping properties in many cases can be described with closely agreeing
values.
(b)
In general, the damping of an entire reinforced concrete structure is

influenced not only by the material damping, but also by the system
damping, that is, the damping properties of the surroundings (building site,
development of static system, etc.).
Dieterle and Bachmann (1981) developed damping models for cracked and
uncracked
bending
elements
as
well
as
their
respective
mathematical
representations. Comparison was made between the theoretical damping model

results and the experimental results, and they were shown to be in good agreement.
Buttmann (1983) studied the damping factors for reinforced concrete walls
3-19
by means of a horizontal excitation. The actual dampings of reinforced concrete

walls were determined on a scale of 1:1.
The difference of frequency was
determined at a point where half the oscillation energy was dissipated and damping
was calculated from this value. According to the measurements, the damping values
of 11% for reinforced concrete walls were determined where amplitudes
corresponded to the loading of a safe shut-down earthquake.
Carydis et al. (1984) studied the free vibrations of a simple reinforced
concrete structure, which is considered as having a single degree of freedom. The
results showed that both the damping and the natural period increase with the
amplitude of the vibration.
Askegaard and Langs (1986) tried to correlate the changes in measured
dynamic parameters and changes in stiffness and allowable working load on the
frost-thaw deteriorated reinforced concrete beams.
They tested two sets of
geometrically identical beams each consisting of 16 beams. One set of beams was
made frost resistant while the other set was made frost-sensitive. Half of a set of
beams were reference beams stored at room temperature, while the other half were
temperature-cycled between -20 and +20 C with a periodicity of about 48 hours.
Half the reference beams and half the temperature-cycle beams were tested to
rupture during the test period, while the other halves underwent dynamic tests at
certain times during the period. For the dynamic tests, the beams were subjected to
impulse load (hammer) and harmonic load (vibrator). The damping was determined
by decay curve method (see Section 2.5.1) and bandwidth method (see
Section 2.5.2).
It was concluded that the dynamic methods give an early indication of
incipient deterioration while it is still difficult to observe changes in carrying
capacity and surface cracking. The formation of cracks added to the changes in
frequencies and damping. Loading tests with the reference beams showed that a
well-developed crack formation changed frequencies by about the same amount as
frost-thaw deterioration and damping by about twice the amount.
Changes in
3-20
transmission time (sound transmission test) were about the same, while the
corresponding damping was only 30-40% of that resulting from frost exposure.
Changes in humidity in the concrete caused considerable changes, especially
in damping.
It was reported that increasing the free water content from
approximately 0 to 4% resulted in an increase of 80% in damping and a reduction of

4% in the frequency. Under normal working condition, however, the change in total
moisture content in a structure is thought to be small. Temperature also has an
effect on natural frequencies and damping. Corrections for different temperature
conditions should therefore be made before comparing results.
An important finding regarding the human perception of vibration is
attributed to Osborne and Ellis (1990) while studying the vibrations of a long-span
lightweight floor. The floor was analysed using different methods to determine
whether it would have acceptable vibration characteristics in service. The vibration
characteristics were then evaluated experimentally for the floor in its bare and
finished states. The site evaluation included quantitative tests to assess human
reaction to vibrations induced in the floor and qualitative tests to measure the
characteristics of the floor. The theoretical analyses indicated that the floor would
be acceptable in service. Calculated natural frequencies were found to be quite
accurate but the measured damping of the floor was considerably less than the
values generally assumed. The addition of a false floor had little effect on the
measured properties, but produced a considerable difference in human perception to
footfalls.
Heiland et al. (1990) studied the influence of fibre reinforcement on the
dynamic behaviour of reinforced concrete structures by testing beams with fibre
contents ranging from 0 to 1.6% by volume. They adopted the usual method of
snap-back tests and considered the influence of different levels of strain and stress
on the material properties and the effect of the number of load cycles. They also
evaluated the total damping ratios of fibre reinforced concrete beams and directly
compared the same with results known for reinforced concrete beams determined by
Dieterle and Bachmann (1981). It was shown that, damping increases because of the
3-21
inclusion of steel fibres. This was especially true for under-reinforced concrete
members. The reason was that, at the cracked stage, the steel fibres stabilised the
damping behaviour as well as frequency characteristics.
Farrar et al. (1994) reported the results of a series of experimental modal
analyses that were performed to examine the similitude of the dynamic parameters
(resonant frequencies, mode shapes, and modal damping) of reinforced concrete
replica models. Also reported were the similitude requirements for damping forces
that had been developed as part of their study. Although damping is typically
considered a material property, the similitude analysis showed that both material
properties and system geometry must be considered when scaling the damping
forces. Results of the experiments showed that the modal frequencies and the mode
shapes of a prototype structure could be accurately predicted from tests on scalemodel structures. Variations in equivalent viscous damping ratios identified on
models and prototype were greater than variations for the other measured dynamic
parameters.
The distortion in damping forces that were observed between model and
prototype are probably not significant in predicting the elastic dynamic response of
reinforced concrete structures.
The effects of these distortions, however, will
become more pronounced when the structure is cracked and the damping increases.
3.3.3
Prestressed concrete elements
Relatively fewer investigations have been carried out to study the damping
characteristics of prestressed concrete structural elements. These research findings
are summarised in the following paragraphs.
Damping of vibrations in simply supported prestressed beams was studied by
Holand (1962). The investigations were carried out in a laboratory to clarify the
damping mechanisms of the concrete and to find out how damping depends on
various variables, for instance, frequency, stress level and concrete quality. The
3-22
damping values were measured using the free decay method (see Section 2.5.1).
Eight prestressed beams of 4 m long spans and 150 mm x 100 mm in cross section
were tested. The results showed that the damping increased with the tendon stress.
Penzien (1964) carried out an investigation to determine the basic damping
characteristics of prestressed concrete beams under dynamic loading. A total of 20
prestressed concrete beams and 4 standard reinforced concrete beams (no prestress)
having the dimensions 6 in. (152 mm) x 6 in. (152 mm) x 90 in. (2286 mm) were
tested.
The basic structural parameters varied in the experiment were type of
prestress, intensity of prestress and ultimate strength of concrete. The dynamic

conditions imposed on each specimen were steady state forced vibration and free
vibration about the static unloaded equilibrium position. For the steady state forced
vibration tests, the damping ratios, , were based on the usual assumption of viscous
damping and calculated using the equation:
1
2 DMF
(3.11)
where DMF is the dynamic magnification factor defined as the ratio of dynamic
deflection to static deflection and was based on resonance conditions. On the other
hand, for the free vibration tests, the damping ratio, , was obtained from the
logarithmic decrement, , using the approximate relation:
(3.12)
where was measured using the decay curve method (see Section 2.5.1).
Penzien (1964) concluded that the amount of internal damping present in a
prestressed concrete member depends a great deal on loading history and on the
amplitude of displacements produced.
He also concluded that the important
parameter, which appeared to influence damping in prestressed or ordinary

reinforced concrete elements, was the degree to which cracking was permitted to
3-23
develop. Damping values were of the order of 1 percent or less for uncracked
members but increased with the degree of cracking permitted. Therefore, magnitude
and type of prestress had an indirect influence on damping since they control
cracking to a considerable extent. Most of the equivalent viscous damping factors
measured ranged from 0.5% to 7% of critical values depending on the degree of
cracking permitted in each test.
The results obtained from the forced vibrations of prestressed and reinforced
concrete beams were reported by James et al. (1964). A total of 16 beams were
tested and they were subjected to sinusoidal exciting forces of varying magnitudes.
The results indicated that the modulus of elasticity of the prestressed siliceous
aggregate concrete beams was 20 to 30% higher than that of companion reinforced
concrete beams. The damping characteristics of prestressed and plain reinforced
beams were in general the same. The damping did not appear to be viscous for
small amplitudes of vibration, but did seem to approach a viscous state for higher
amplitudes of vibration. Also, prestressed beams fabricated with siliceous aggregate
showed a much greater resistance to cracking than plain reinforced beams. Repeated
tests on the prestressed beams showed that the test history of this type of beam had
little effect on the dynamic response characteristics.
Spencer (1969) studied the damping of prestressed concrete members
(152 mm x 102 mm in cross section and 4013 mm long). The energy method was
used and damping was defined as the energy dissipated by a member during one
cycle of steady-state loading.
It was concluded that the two major sources of damping in prestressed
concrete members might be: (a) the inelastic behaviour of the concrete, especially in
those regions where cracking results in high stress and curvature concentration, and
(b) the slip between steel and concrete associated with tension cracking.
The dynamic response and the fatigue behaviour of rectangular and skew
prestressed concrete waffle slabs were investigated by Grace and Kennedy (1990).
A solution for the dynamic response was derived based on classical orthotropic plate
3-24
theory. The theoretical results were verified and substantiated by experimental

results obtained from tests on two prestressed concrete waffle slab models. The two
models were rectangular and skew in plan and represented 1/8-scale concrete waffle
slab bridge decks. The cross-sectional geometries of the two models were identical
with the transverse ribs being orthogonal to the longitudinal ribs. The rectangular
model was 132 in. (3353 mm) long by 77 in. (1956 mm) wide, while the 45 skew
model was 132 in. (3353 mm) long by 110 in. (2794 mm) wide. The two models
were subjected to dynamic, fatigue and ultimate load tests.
The first three
fundamental modes of vibration were studied, namely the first flexural mode, the
first torsional mode and the second flexural mode. The dynamic tests consisted of
(a) sweep-sine wave test; (b) normal mode test; and (c) logarithmic decrement test.
The logarithmic decrement test was conducted by first exciting each model at
its first natural frequency, and then the hydraulic power was turned off to allow the
model to undergo a decaying free vibration. As a result, the associated geometric
and material damping characteristic of the model was determined using the
logarithmic decrement method (same as decay curve method as described in
Section 2.5.1). The damping ratio for the rectangular model with the transverse
crack (occurred during transportation) at 8 in. (203 mm) from midspan was slightly
above 4%, and for the skew model it was only 2.1%. However, after subjecting the
latter model to fatigue loading, the damping ratio increased to 3.8% because of the
accumulation of extensive cracking. Thus, it was concluded that in prestressed
concrete waffle slabs a damping ratio of less than 3% can be expected if the
prestressing forces are sufficient to prevent tension cracks from developing. Under
fatigue loading a somewhat higher damping ratio should be allowed for in the
design.
It was also shown that prestressing of concrete waffle slabs enhances their
natural frequencies; thus resonance in such structures can be avoided when they are
subjected to a low-frequency excited source of vibrations. Thus, due to increased
stiffness in the use of prestressed waffle slabs versus prestressed solid slabs with the
same platform geometry, considerable enhancement in the natural frequencies can
be realised.
3-25
Recently, Abdalla and Kennedy (1995) presented the dynamic analysis and
design of simply supported and continuous prestressed concrete beams with
rectangular openings. Several design parameters were varied such as opening width
and depth, horizontal and vertical locations of the openings, type of cross section
and presence of more than one opening. The analytical results were compared to
test results in the literature. Good agreement was shown between the theoretical and
the experimental results.
An extensive parametric study was performed on
prestressed beams with openings to determine the natural frequencies and the
associated mode shapes.
It was shown that the fundamental frequency of a simply supported
prestressed beam with a shear opening is less than that for a solid beam. When the
opening is located in the maximum bending field, the fundamental frequency
exceeds that of the solid beam. The width and depth of an opening has a more
significant effect on the fundamental frequency of beams with a shear opening than
those with a bending opening. Also, the horizontal location of an opening has a
significant influence on the natural frequencies of prestressed concrete continuous
beams with openings and especially for beams with unequal spans. Having two
openings in the same span of a two-equal-span continuous beam causes a significant
decrease in the third natural frequency, whereas in the case of one opening in each
span, the first and second natural frequencies are significantly decreased.
3.4
Summary
In this chapter, published research efforts on the damping behaviour of concrete and
reinforced concrete members are briefly reviewed. These include the extensive
investigations into the determination methods of the damping capacity and the
damping characteristics of plain, reinforced and prestressed concrete elements.
From the previous research on damping characteristics of plain and
reinforced concrete members, it is found that the damping capacity of such members
may be influenced by many factors. For plain concrete members, these include
3-26
moisture content, age, longitudinal stress, water-cement ratio, mix composition,

temperature, frequency and mode of vibration, state of curing and such other factors.
For reinforced concrete members, the influencing factors are compressive strength
of concrete, steel ratio, modulus of elasticity, loading history, cross sectional area,
cracking state and similar other factors. But the factor, which affected damping in
both plain and reinforced concrete members, most significantly is acknowledged to
be the cracking condition by most researchers. This is equally true for prestressed
concrete members.
Apart from Dieterle and Bachmann (1981), no other researchers have
attempted to develop any sort of relationships between damping and the cracking
state of concrete structures. Also, the conclusions drawn by other researchers were
mostly qualitative and the majority of the mathematical formulae provided were
developed for single-quantity relationship.
In view of all this, there is a need to develop effective, rigorous and reliable
relationships between the damping capacity and the cracking state of reinforced and
prestressed concrete members. Therefore, the main aim of this research is the
development of damping prediction formulae for reinforced and partially prestressed
concrete beams based on the residual crack widths. The residual crack widths, on
the other hand, will be related to the multi-parameters quantifying the concrete
members. As a result, a relationship is established between the damping capacity
and the parameters describing such members.
This would allow the damping
capacity of a structure to be determined explicitly.
Chapter 4 : Damping Theory of Simply Supported and Continuous Beams
4-1
CHAPTER 4
DAMPING THEORY OF SIMPLY SUPPORTED AND
CONTINUOUS BEAMS
4.1
General Remarks
A structure is said to be undergoing free vibration when it is disturbed from its static
equilibrium position and then allowed to vibrate without any external dynamic
excitation.
In this chapter free vibration leading to the notions of the natural
vibration frequency and damping ratio for a single degree of freedom (SDOF)
system is discussed. It will be shown that the rate at which the motion decays in free
vibration is controlled by the damping ratio. Thus the analytical results describing
free vibration provide a basis for determining the natural frequency and damping
ratio of a structure from experimental data of the type shown in Figure 2.2 (see
Section 2.5.1).
Although damping in actual structures is due to several energy-dissipating
mechanisms acting simultaneously, a mathematically convenient approach is to
idealize them by equivalent viscous damping (Chopra, 1995). Consequently, this
chapter deals primarily with viscously damped system. However, free vibration of
SDOF systems involving Coulomb and hysteretic damping is discussed herein as
well. Also presented is the application of the free vibration theory for a SDOF
system to simply supported and continuous beams.
Further, the relationship
4-2
between different vibration parameters such as displacement, velocity and

acceleration is discussed.
Finally, the theoretical investigation carried out as part of this research
provides a suitable basis to develop the general relationship between damping ratio
(or factor) and other relevant parameters of reinforced and partially prestressed
concrete beams. Accordingly an analytical formula for determining the damping
ratio of concrete beams is presented at the end of the chapter.
4.2
Single Degree of Freedom (SDOF) Model
In general, even the simplest of structures such as simply supported beams and
cantilevers are in reality multiple degree-of-freedom (DOF) systems with infinite
number of DOFs. For practical purposes, however, many simple structures and
structural elements may be considered as SDOF systems, by treating them as simple
mass-spring systems with an equivalent lumped mass and an equivalent elastic
spring. Some examples of this form of simplification are illustrated in Figure 4.1.
In the tests carried out in this research, hammer excitation was used as a
vibrating source to generate the free vibration. With hammer excitation at the top of
the beam, the displacement of the beam occurs only within the vertical plane. This
means that no displacement occurs in the horizontal plane and also no torsion
occurs. As the displacement of the system is specified completely in the same
direction with a single displacement coordinate, the system can be assumed to
exhibit a single DOF.
If a statically loaded elastic system, such as a beam, is disturbed in some
manner from its position of equilibrium, the internal forces and moments in
deformed configuration will no longer be in balance with the external loads. That is,
vibrations will occur.
In general, an elastic system can perform vibrations of
different patterns, or modes (Timoshenko et al., 1974). The simplest vibratory
4-3
Figure 4.1 Equivalent SDOF mass-spring systems

system consists of an elastic spring and a mass element, as represented for the
equivalent SDOF systems in Figure 4.1. Such a system is called a SDOF system
since it can move in but one coordinate. Or, it requires only one coordinate to define
its configuration.
4.3
Undamped Free Vibration of a SDOF System
Let us consider the configuration of the spring-mass system, as shown in Figure 4.2,
representing the undamped free vibration of a SDOF system. The mass hangs at the
lower end of a spring, which in turn is attached to a rigid support at its upper end.
At rest, the mass will hang at a level called the static equilibrium position, in which
the upward spring force exactly balances the downward gravitational force on the
mass. In Figure 4.2, if is the static deflection due to the weight W of the mass m,
then for static equilibrium
W = mg = k
(4.1)
4-4
k
k
Static equilibrium
position
x
W = mg
+x
Figure 4.2 System of free undamped vibration

where g is the acceleration due to gravity and k is the spring stiffness. Let the mass
be deflected a distance + x from its static equilibrium position. Then the spring
force is - k ( + x). The application of Newtons second law of motion to the mass
m gives
m !!x = W - k ( + x)
(4.2)
where !!x = d2x/dt2 is the acceleration of the mass. Since W = k, it follows that
m !!x + kx = 0
(4.3)
This indicates that when a mass moves in a vertical direction, its weight can be
ignored, provided x is measured from its static equilibrium position.
The equation of motion for an undamped oscillator, as given in Eq. (4.3), can
be rewritten as
!!x + n2 x = 0
(4.4)
where n =
4-5
(k m) is the natural circular or angular frequency of vibration in units
of radians per second.

The spring constant k can be expressed in terms of the mass m from
Eq. (4.1) as
k=
W
mg
=
(4.5)
Substitution of Eq. (4.5) into the expression for n given above yields
n =
(4.6)
The time required for the undamped system to complete one cycle of free
vibration is the natural period of vibration of the system, which is denoted by Tn in
units of seconds. It is related to the natural circular frequency, n, as
2
n
(4.7)
A system executes 1/Tn cycles in 1 second.
This natural cyclic frequency of
Tn =
vibration, fn, is denoted by
fn =
1
Tn
(4.8)
The units of fn are Hertz (Hz) or cycles per second (cps). And fn is related to n
through
fn =
n
2
(4.9)
4-6
The term natural frequency of vibration applies to both n and fn.

Following Eq. (4.6), the natural cyclic frequency, fn, and natural period, Tn,
defined in Eqs. (4.7) through (4.9) can be expressed in the alternative form
fn =
1
2
or
Tn = 2
(4.10)
Thus, when the mass vibrates in a vertical direction, the natural frequency and the
period of vibration can be computed by simply measuring the static deflection . It
is not necessary to know the spring stiffness k and the mass m.
The general solution for Eq. (4.4) is
x (t) = A cos (n t) + B sin (n t)
(4.11)
where x (t) is the displacement varying with time, t, and A and B are arbitrary
constants to be determined from the initial conditions. Hence the specific solution
for initial conditions of x0 (displacement at t = 0) and x! 0 (velocity at t = 0) is
x (t) = x0 cos (n t) + ( x! 0 /n) sin (n t)
(4.12)
If a constant is defined in such a way that sin = x0/[ x02 + ( x! 0 /n)2] and cos =
( x! 0 /n) /[ x02 + ( x! 0 /n)2], Eq. (4.12) may be rewritten as
x (t) = [ x02 + ( x! 0 /n)2] {sin cos (n t) + cos sin (n t)}
(4.13)
In view of the addition formula sin ( + ) = sin cos + cos sin , we have
x (t) = xmax sin (n t + )
(4.14)
where xmax = [ x02 + ( x! 0 /n)2] is the amplitude of vibration. This is true since
4-7
the peak values of sin (n t + ) are 1. The constant simply locates the start of
the vibration relative to the origin and is referred to as the phase angle.
4.4
Damped Free Vibration of a SDOF System
All structures dissipate energy when they vibrate. The energy dissipated is often
very small so that an undamped analysis is sometimes realistic. But when the
damping is significant its effect must be included in the analysis. As discussed in
earlier chapters, the damping which occurs in structures is due to frictional effects
such as that occurs at the connection between elements, or internal friction in the
structural members. It is often difficult to model damping exactly because many
mechanisms may be operating in a structure. However, each type of damping can be
analysed, and since in many structures one form of damping predominates, a
reasonably accurate analysis is usually possible. As mentioned in Section 2.4, the
most common types of damping are viscous, Coulomb and hysteretic.
Free
vibration of a SDOF system with viscous, Coulomb and hysteretic damping is,
therefore, discussed in the following sub-sections.
4.4.1
Free vibration with viscous damping
For the free vibration of a SDOF system shown in Figure 4.3, with mass m, spring
constant k, and viscous damping c, the system undergoes a dynamic displacement
x(t) measured from its static equilibrium position of the mass. Applying Newtons
second law, the equation of motion of the system is represented by
m !!x = W - k( + x) - c x!
(4.15)
m !!x + c x! + kx = 0
(4.16)
or
Static equilibrium
position
4-8
.
cx
kx
+x
+x
Figure 4.3 SDOF system with viscous damping

where x! = dx/dt is the velocity of the mass. Eq. (4.16) can be solved by assuming
x = Aest
(4.17)
where A and s are undetermined constants. Substitution of Eq. (4.17) into Eq. (4.16)
yields
ms2 Aest + csAest + kAest = 0
(4.18)
Division of each term in Eq. (4.18) by mAest yields

s2 + (c/m) s + n2 = 0
where n =
(k m) as defined earlier in Section 4.3.
(4.19)
The solutions to Eq. (4.19) are
given by
1 c
s =
2 m
()
c
m
2
4 n
4-9
(4.20)
This solution leads to the definition of critical damping constant. The critical
damping, cc, is defined as the value of the damping constant, c, for which the radical
in Eq. (4.20) becomes zero. That is,
c
m

- 4 n2 = 0
(4.21)
or
cc = 2 m n
(4.22)
For any damped system, the damping ratio (or factor), , is defined as the ratio of the
damping constant, c, to the critical damping constant, cc, i.e.
= c/cc
(4.23)
From Eqs. (4.22) and (4.23), one can write

c = 2 m n
(4.24)
Eq. (4.20) can be stated more concisely by incorporating Eq. (4.24) to give
s = [-
( 1) ]
2
(4.25)
Thus the general solution of Eq. (4.16) can be written as

[
x(t) = A1 e - +
) ] nt + A e[- (
2
) ] nt
(4.26)
4-10
where A1 and A2 are arbitrary constants to be determined from the initial conditions
of the system.
The nature of the roots of s given by Eq. (4.25) and hence the behaviour of
the solution given by Eq. (4.26) depends upon the magnitude of damping. It can be
shown that the case = 0 leads to the undamped vibrations discussed in Section 4.3.
Hence it is assumed that 0 and the following three possible cases are considered.
Case 1. Critically damped system in which = 1 or c = cc or (c/m)2 - 4 n2 = 0. In
this case the two roots of s in Eq. (4.25) are equal. That is
s1 = s2 = -
cc
= - n
2m
(4.27)
where s1 and s2 are the two roots of s. Because of the repeated roots, the solution of
Eq. (4.16) is given after Kreyszig (1993) as
nt
x(t) = (A1 + A2 t) e -
(4.28)
The application of the initial conditions x (t = 0) = x0 and x! (t = 0) = x! 0 for this

case, gives the system response as
nt
x(t) = e -
[( x! 0 + n x0) t + x0]
(4.29)
It can be shown that the motion represented by Eq. (4.29) is non-periodic. Since
nt
e-
0 as t , the motion will eventually diminish to zero and the time
required for the system to stop is
t=
- x0
x! 0 + n x0
(4.30)
4-11
Case 2. Overdamped system in which > 1 or c > cc or (c/m)2 > 4 n2. As
( 1) > 0, Eq. (4.25) shows that the two roots of s are real, distinct and negative.
2
In this case, the solution, Eq. (4.26), can be expressed as

[
x(t) = A1 e - +
) ] nt + A e[- (
2
) ] nt
(4.31)
For the initial conditions x (t = 0) = x0 and x! (t = 0) = x! 0, the constants A1 and A2

can be obtained as:
A1 =
2
x 0 n ( + ( 1)) + x! 0
A2 =
2 n (2 1)
2
- x 0 n ( ( 1)) x! 0
2 n (2 1)
(4.32)
Eq. (4.31) shows that the motion is non-periodic regardless of the initial conditions
imposed on the system.
Since the roots of s are both negative, the motion
diminishes exponentially with time, as shown in Figure 4.4.
A1
A1e(- + ( 1) ) nt
2
x (t)
nt
A2
A2e(-
) nt
Figure 4.4 Motion of a SDOF system with viscous damping ratio > 1
(overdamped case)
4-12
Case 3. Underdamped system in which < 1 or c < cc or (c/m)2 < 4 n2. For this
condition, (2 - 1) is negative and the roots of s can be expressed as
s = [- i
where i =
(1 ) ]
2
(4.33)
1 . At this stage the concept of damped angular natural frequency, d,
is introduced which is defined as
d = n
(1 )
2
(4.34)
Hence, using this definition of damped angular frequency, Eq. (4.33) becomes
s = - n i d
(4.35)
And the solution, Eq. (4.26), can be written in complex form of the sine and cosine
functions, or of amplitude (X) and phase angle (), as follows:
x(t) = e n t [A1 ei d t + A2 ei d t ]
= e n t [ A1 cos (d t) + A2 sin (d t)]
(4.36)
= X e n t [sin (d t + )]
where A1 = (A1 + A2) and A2 = (A1 - A2) which are the constants to be determined
from the initial conditions. In terms of the initial conditions, for example, x(0) = x0
and x! (0) = x! 0, the complete solution for x(t) takes the form
x! 0 + n x0
x(t) = e n t [x0 cos (d t) +
sin (d t)]
d
(4.37)
4-13
The motion described by Eq. (4.37) is a damped harmonic motion of angular

frequency d, but because of the factor e n t , the amplitude decreases
exponentially with time, as shown in Figure 4.5. It can be shown from Eq. (4.34)
that the damped natural frequency d is always less than the undamped natural
frequency n and it decreases with increasing amount of damping.
X(t)
envelope exhibiting exponential decay

X0
Figure 4.5 Motion of a SDOF system with viscous damping ratio < 1
(underdamped case)
From the above discussion, it is found that for Case 1, i.e., if = 1, the
system returns to its equilibrium position without oscillating. For Case 2 ( > 1),
again the system does not oscillate and it returns to its equilibrium position, as in
Case 1, but at a slower rate. For Case 3 ( < 1), the system oscillates about its
equilibrium position with a progressively decreasing amplitude.
Thus, the
underdamped case is very important in the study of mechanical vibrations, as it is

the only case which leads to an oscillatory motion (Caughey and OKelley, 1961).
Also, structures of interest - buildings, bridges, dams, nuclear power plants, offshore
structures, etc. - all fall into this category because typically their damping ratio is
less than 0.10.
This, definitely was the case for the reinforced and partially
prestressed concrete beams studied in this research.

One of the ways of estimating the damping level in a structure is to conduct a
free vibration test and inspect the record or time history of the response recorded, as
4-14
was done in the tests in this research. The theory is given below.
It can be shown from Eq. (4.37) that for every period of damped vibration,
Td, or for every cycle of vibration, the amplitude is diminished by the ratio
(x0 e T ) : (x0)
n d
(4.38)
Thus, if xr = the amplitude at the end of the rth oscillation, and xs = the amplitude at
the end of the sth oscillation, then
xr
= { x0 e n r Td }/{ x0 e nsTd }
xs
(4.39)
or
xr
= e n(s-r)Td
xs
(4.40)
Taking natural logarithms of both sides yields
1
xr
ln
= n Td
s- r
xs
(4.41)
When r = 0 and s = n, this becomes
1
x0
ln
= n Td
n
xn
(4.42)
The left hand side of Eq. (4.42) is nothing but the logarithmic decrement, , as
defined earlier in Section 2.5.1. Thus,
= n Td
(4.43)
4-15
As the period of damped vibration is
Td =
2
2
=
2
d
n 1
(4.44)
it follows that
2
(4.45)
For small damping, which is usually the case with most of the practical structures,
Eq. (4.45) can be approximated as
=2
if << 1
(4.46)
The logarithmic decrement is dimensionless and is actually another form of

the dimensionless damping ratio (Rao, 1990). Once is known, can be found by
solving Eq. (4.45) as
(2 )2 + 2
(4.47)
If Eq. (4.46) is used instead of Eq. (4.45), can be obtained from
4.4.2
(4.48)
Free vibration with Coulomb damping
In many mechanical systems, Coulomb or dry-friction dampers are used because of

their mechanical simplicity and convenience (Sinclair, 1955). Also in vibrating
structures, steady friction forces occur when relative motion takes place between
4-16
adjacent members. Coulomb damping arises when bodies slide on dry surfaces.
Coulombs law of dry friction states that when two bodies are in contact, the force
required to produce sliding is proportional to the normal force acting in the plane of
contact. Thus the friction force Fd = N, where denotes the coefficients of static
and kinetic friction, taken to be equal, and N, the normal force between the sliding
surfaces. The friction force is assumed to be independent of the velocity once the
motion is initiated. The direction of the friction force opposes motion, and the sign
of the friction force will change when the direction of motion changes.
This
necessitates formulation and solution of two differential equations, one valid for
motion in one direction and the other, when the motion is reversed.
Coulomb damping in a SDOF system can be modelled as shown in
Figure 4.6. The friction force Fd always opposes the motion, so that if the body is
displaced a distance x0 to the right and released from rest we have, for motion from
right to left only,
m !!x + kx = Fd
(4.49)
The solution to this equation of motion is

x(t) = A sin nt + B cos nt + (Fd/k)
where n =
(4.50)
k m in radians per second.
The initial conditions were x = x0 at t = 0 and x! = 0 at t = 0. Thus A = 0 and

B = (x0 - Fd/k). Hence,
x(t) = x0 -
Fd
k
) cos t + Fk
(4.51)
At the end of the half cycle right to left, nt = which means cos nt = -1
and x(t = /n) = - x0 +
2Fd
. That is, a reduction in amplitude of 2Fd/k per half cycle
k
4-17
occurs.
+ ve
k
m
Fd
Figure 4.6 SDOF model with Coulomb damping

From symmetry, for motion from left to right when the friction force acts in
the opposite direction to the above, the initial displacement is (x0 - 2Fd/k) and the
final displacement is, therefore, (x0 - 4Fd/k), i.e. the reduction in amplitude is 4Fd/k
per cycle. This oscillation continues until the amplitude of the motion is so small
that the spring force is unable to overcome the friction force Fd. This can happen
whenever the amplitude is (Fd/k). The motion is, therefore, sinusoidal for each
half cycle, with successive half cycles centred on points distant + Fd/k and - Fd/k
from the origin. The oscillation ceases with x Fd/k, and the frequency of the
oscillation is (1/2)
4.4.3
k m Hz.
Free vibration with hysteretic damping
Damping resistance occurring from internal friction is referred to as hysteretic

damping, structural damping, or solid damping. It results from the thermal effect of
repeated elastic strain imposed on the material. Heat flowing across boundaries of
grains causes dissipation of energy and hence damping of the motion
(Humar, 1990). This causes a hysteresis loop to be formed in the stress-strain or
force-displacement curve (see Figure 4.7). The energy loss in one loading and
4-18
unloading cycle is equal to the area enclosed by the hysteresis loop (Humar, 1990;
Pinsker, 1949; Scanlan and Mendelson, 1963; Ungar, 1973). The name hysteretic
damping is thus derived from its relationship to the hysteresis loop.
If hysteretic damping is the only type of damping present in the system, the
equation of motion for free vibration takes the form
m !!x + fS (x) = 0
(4.52)
where fS (x) is obtained from the force-displacement relationship of Figure 4.7. The
force is now a nonlinear function of deformation. In fact, for a given deformation
the force is not unique but depends on the deformation history. In a general case,
the solution of Eq. (4.52) is quite complex, and often the only effective method of
obtaining a solution is to use a numerical technique (Humer, 1990).
Stress (force) fS
Hysteresis
loop
Loading
Unloading
Strain
(displacement) x
Figure 4.7 Hysteresis curve

For harmonic motion, such as, for example, in the case of free vibrations
with small amounts of damping, hysteretic damping can be accounted for by
assuming that in the equation of damped vibration, fS = kx still applies, with k being
an average stiffness, but the damping force is given as
fD =
k
x!
4-19
(4.53)
where is the frequency of vibration and is the loss factor, which is a measure of
the hysteretic damping in a structure. Since for harmonic motion x! is proportional
to the frequency of vibration, the latter cancels out from Eq. (4.53), making the
damping force independent of frequency. For free vibration, n, and if we
define equivalent viscous damping constant ce = k/n, and equivalent viscous
damping ratio e = ce/(2mn) = /2, then because fD = ce x! from Eq. (4.53), the free
vibration equation, Eq. (4.37), still apply, with c replaced by ce and replaced by e.
The free vibration response under hysteretic damping is thus given as
x(t) = ee n t [x0 cos (d t) +
where d = n
x! 0 + e n x0
sin (d t)]
d
1 e . Also, the logarithmic decrement is obtained from
2e =
4.5
(4.54)
(4.55)
Vibration Quantities and Their Relationships
There are three vibration quantities in vibration theory viz., the vibration
displacement, velocity and acceleration.
If vibration has the form of a pure
translation oscillation along one axis only, the instantaneous displacement u of the
particle (or body) from the reference position can be mathematically described by
the equation
u = upeak sin (t)
(4.56)
where (= 2f) is the angular frequency, upeak is the maximum displacement from
the reference position and t is the time.
4-20
As the velocity of a moving particle (or body) is the rate of change of the
displacement with time, the motion can be described in terms of velocity u! as
u! =
du
= upeak cos (t)
dt
(4.57)
Similarly, the acceleration !!u of the motion is the rate of change of the
velocity with time. Or,
!!u =
du!
d2 u
=
= - 2 upeak sin (t)
dt
d t2
(4.58)
In Eqs. (4.56) through (4.58), it can be seen that the form and period of
vibration remain the same whether it is the displacement, the velocity or the
acceleration that is being considered. However, the velocity leads the displacement
by a phase angle of 90 or /2, and the acceleration, in turn, leads the velocity by a
phase angle of 90 or /2 (Broch, 1980).
When the ratios of displacement, velocity and acceleration are taken, the
following relationship can be derived:
u0
un
u! 0
u! n
!!u0
!!u n
u peak ( 0 )
u peak(n)
(4.59)
where u0, u! 0 and !!u 0 are initial values of displacement, velocity and acceleration,
respectively and un, u! n and !!u n are the displacement, velocity and acceleration,
respectively after n cycles. Also, upeak(0) is the initial peak offset from the reference
position while upeak(n) is the peak offset from the reference position after n cycles.
Eq. (4.59) indicates that the same result of logarithmic decrement can be
obtained when the free decay method is used during the free vibration measurement
irrespective of the vibration parameter measured i.e. displacement, velocity or
acceleration.
4.6
4-21
Free Vibration of Beams and Their Damping Determination
The fundamental vibrations of simply supported and continuous beams may be

characterised as those of an equivalent SDOF system. The basis of the analysis of
real continuous systems is one of the following:
(a) a continuum differential equation of motion for the system or
(b) a discrete finite element approximation which can be more or less complex.
In both cases the analysis can be reduced, by suitable coordinate transformations, to
a problem involving a set of simple oscillators each of which describes one of the
characteristic vibrations of the system. This is the basis of the normal mode method
and the details of it can be found in good standard textbooks, such as those by
Clough and Penzien (1975) and Thomson (1988). In certain circumstances only the
fundamental mode of vibration is important and so the continuous system can be
approximated by an equivalent SDOF model.
Because it is not possible to determine analytically the damping ratio, , for
practical structures, this elusive property should be determined experimentally
(Chopra, 1995). Free vibration experiments provide one means of determining the
damping. As such, in this research damping has been determined experimentally for
reinforced and partially prestressed concrete beams from the decay of their free
vibrations. As described in Section 4.2, the hammer excitation at the top of the
beams, during the free vibration tests, results in vibrations in the vertical plane only.
This justifies the assumption of the exhibition of SDOF systems by the vibrating
beams. This is equally true for the simply supported beams and each span of the
two-equal-span continuous beams tested.
It is also assumed that the structural
damping mechanism resulting from the free vibrations of the beams has the
characteristics of viscous damping. This has been verified from experimental results
highlighted later in Section 9.4.3. Accordingly, the free vibration of SDOF system
with viscous damping has been discussed at lengths in Section 4.4.2, especially for
the underdamped case.
The very definition of the logarithmic decrement, , indicates that it can only
4-22
be determined experimentally. Once is known, the damping capacity, , can be

found using Eqs. (4.47) or (4.48) depending on the accuracy required. In this
research, crack width is used as a basis to investigate the effect of cracking on
damping characteristics of reinforced and partially prestressed concrete beams.
Thus, based on experimental work, the logarithmic decrement, , can be defined as
the function of residual crack width, wr, as follows:
= F1 (wr)
(4.60)
There can be two similar but separate relationships for reinforced and partially
Residual crack width (wr) of concrete beams, in turn, may be given as
functions of the instantaneous crack width, wcr, and of mid-span deflection, , as
follows:
wr = F2 (wcr)
(4.61)
wr = F3 ()
(4.62)
and
where the relationships can be determined via experimental means.

Finally, based on the experimental work, the instantaneous crack width (wcr)
of reinforced and partially prestressed concrete beams may be considered to be the
function of six variables as follows:
wcr = F4 (Es, fs, s, c, , )
(4.63)
where Es is the modulus of elasticity for steel, fs is the stress in the reinforcing steel,
s is the average spacing between reinforcing bars, c is the concrete clear cover, is
4-23
the steel ratio, and is the average diameter of the reinforcing bars.
Note that the mid-span deflections of concrete beams can also be obtained
from similar beam variables using any standard method of deflection calculation.
In view of the above proposed relationships, the damping ratios of reinforced
and partially prestressed concrete beams can be evaluated as functions of the general
variables defining the beams, namely Es, fs, s, c, and .
4.7
Summary
This chapter serves as the basis for the development of the underlying theory for the
experimental investigation undertaken.
Having identified the importance and
influence of cracking on the damping characteristics of reinforced and partially

prestressed concrete beams in earlier chapters, damping measured from free
vibration tests of the beams is analysed as being related to their residual crack
widths. Residual crack widths, in turn, are taken to be related to the instantaneous
crack widths and the mid-span deflections.
To ensure the reliability and accuracy of the proposed analysis, it is
necessary to develop a new formula for evaluating the instantaneous crack width and
spacing for the reinforced and partially prestressed concrete beams studied in the
experimental work. This is done in Chapter 7.
Hammer excited free vibration of simply supported reinforced and partially
prestressed concrete beams can be considered as the damped free vibration of a
SDOF system with viscous damping. The same applies to each span of the twoequal-span continuous beams. As such, the underlying theory of free vibration of a
SDOF system is discussed in some detail in this chapter. Also discussed is the
relationship between different vibration quantities.
Chapter 5 : Cracking in Reinforced and Partially Prestressed Concrete Beams
5-1
CHAPTER 5
CRACKING IN REINFORCED AND PARTIALLY
PRESTRESSED CONCRETE BEAMS
5.1
General Remarks
As proposed in Section 4.6, the logarithmic decrement values measured for the
reinforced and the partially prestressed concrete beams are to be related to their
residual crack width values. The latter values, in turn, are to be related to the
instantaneous crack width values. All this necessitates the cracking in reinforced
and partially prestressed concrete beams to be studied in detail.
As such, the causes and types of cracking and the factors affecting crack
widths for both reinforced and partially prestressed concrete beams are discussed in
some detail in this chapter. Since flexural cracks are the only cracks of concern in
this research, a literature review is also carried out on the mechanism of flexural
cracking in reinforced concrete beams.
It is proposed to develop a new formula for evaluating the instantaneous
crack width for reinforced and partially prestressed concrete beams from the
variables defining the beams. Accordingly, this chapter also reviews the available
crack spacing and crack width formulae for both reinforced and partially prestressed
concrete beams inclusive of the different national design code formulae. This will
5-2
facilitate the development of the new crack width formula as well as its comparison
with the others.
5.2
Cracking in Reinforced Concrete Beams
Cracks in reinforced concrete members may be expected under a certain service load
because of the low tensile strength of concrete and the use of high-strength steel.
Also, in normal reinforced concrete structures cracking must occur to allow for the
transfer of tensile stresses from the concrete to the steel (Mier, 1997). On the other
hand, cracking in reinforced concrete structures has a major influence on the
structural performance including tensile and bending stiffnesses, energy absorption
capacity, ductility, and corrosion resistance of reinforcement (Albandar and
Mills, 1974; Base et. al., 1966; Bazant and Oh, 1983a; 1983b; Broms, 1965; Broms
and Lutz, 1965; Clark, 1956; Mathey and Watstein, 1960; Nawy, 1968; Park and
Paulay, 1975). It is, therefore, necessary to control cracking in reinforced concrete
structures.
For efficient control of cracking, the designer must be able to predict the
anticipated crack width under a specific load. To be able to do that or in other words
for the development of the proposed crack width formula, an understanding of the
causes and types of cracking as well as the factors affecting crack widths is essential.
These along with the mechanism of flexural cracking and the different available
crack width formulae are therefore discussed in some detail in the following subsections.
5.2.1
Causes of cracking
The causes of cracking in reinforced concrete members are numerous, but most
cracks occur as a result of one or more of the following actions to which they can be
subject (ACI Committee 224, 1968; Nawy, 1968):
(a)
(b)
5-3
Volumetric change
(i)
Drying shrinkage;
(ii)
Creep under sustained load;
(iii)
Thermal stresses including elevated temperatures;
(iv)
Composition of the concrete;
Internal or external direct stress due to continuity, reversible load, long-term

deflection, or environmental effects including differential movement in
structural systems; and
(c)
Flexural stress due to bending.

While the net result of these three actions is the formation of cracks, the
mechanisms of their development cannot be considered identical.
Volumetric
change generates internal microcracking which ultimately develops into full

cracking, while direct internal or external stress could either generate internal
microcracking such as in the case of fatigue due to reversible load or flexural
microcracking due to the other types of direct stress mentioned. ACI publication
SP-20 (ACI Committee 224, 1968) has provided excellent background materials
relating to the basic mechanism of cracking.
Flexural stress due to bending, on the other hand, initially generates flexural
microcracks. These flexural microcrackings, not visible to naked eye, can start at
very low stress levels. The process of fracture in concrete under uniaxial tensile or
uniaxial compressive stress begins with the spreading of microcracks, most of which
originate in the coarse aggregate-mortar interface.
Under service load condition, the tensile strain in the concrete in the vicinity
of the steel must far exceed the extensibility of the concrete; hence, cracking is
inevitable (Huang, 1975). Beyond the flexural microcracking stage, they suddenly
develop into cracks of measurable widths with only a slight increase in load (ACI
Committee 224, 1972; 1984; Nawy, 1968). These visible initial flexural cracks
5-4
develop when the actual tensile stress due to load exceeds the tensile resistance of
the concrete.
5.2.2
Types of cracking
Tensile stresses induced by loads, moments and shears cause distinctive crack
patterns (MacGregor, 1997). Members loaded in direct tension crack right through
the entire cross section, with a crack spacing ranging from 0.75 to 2 times the
minimum thickness of the member.
Members subjected to bending moments
develop flexural cracks. These are vertical cracks and extend almost to the zerostrain axis (neutral axis) of the member. Cracks due to shear have a characteristic
inclined shape and they extend upward as high as the neutral axis and sometimes
into the compression zone. Torsion cracks are similar. In pure torsion, they spiral
around the beam.
Bond stresses lead to splitting along the reinforcement.
Concentrated loads will sometimes cause splitting cracks or bursting cracks.

When a reinforced concrete member is subjected to tension or bending, two
types of cracks eventually form - primary cracks and secondary cracks. In order to
discuss qualitatively the phenomenon of flexural cracking, let us consider a beam in
simple bending (Figure 5.1). Loading of the beam increases the concrete tension
and when this reaches the tensile strength of the concrete, cracks appear on the beam
surface at discrete intervals. These are called primary cracks and they develop as
bending moment increases. In a general way these cracks are wedge shaped and
have a width varying from a maximum at the bottom of the beam to zero at the
neutral axis (Warner et al., 1989). The crack brings about a drastic drop in the
concrete stress and strain so that some elastic recovery occurs in the tensile concrete,
especially on the face of the beam on either side of the primary crack. However,
close to a reinforcing bar the bond prevents to some extent this elastic recovery.
That is to say in a zone of concrete surrounding the bar the concrete retains much of
its tensile stress (and strain) and in consequence the crack width is less near the bar
(Husain and Ferguson, 1968).
(a)
primary cracks
(b)
secondary cracks
5-5
Figure 5.1 Primary and secondary flexural cracks

The persistence of this concrete tension leads to the formation of secondary
cracks (Figure 5.1b). These form internally near the bar, where the concrete tension
is greatest, and frequently do not extend to the surface of the beam (Broms, 1965;
Gergely and Lutz, 1968). The secondary cracks tend to form vertically and midway
between the primary cracks.
A repetition of the process of secondary crack
formation may lead to the formation of further small cracks between the primary and
secondary cracks.
5.2.3
Factors affecting crack widths
Much work has been carried out by many researchers on the cracking of reinforced
concrete members and factors influencing cracking.
Early investigations on
cracking in reinforced concrete members can be traced back to the 1900s.

Considere (1906) was the first one to establish that the crack spacing in beams
increased with the diameter of the reinforcement and that the crack width decreased
with an increase in roughness of the reinforcement and was also approximately
5-6
proportional to the crack spacing. This finding was confirmed later by Graf (1921)s
tests.
Equations suggesting the relative advantage of the utilisation of small
reinforcing bars were introduced by Westergaard (1933).

Saliger (1936) was the first to derive an analytical solution for calculating the
crack spacing for axially loaded reinforced concrete prisms. The basic form of his
equation,
lmax = K
(5.1)
where lmax is the maximum possible crack spacing, K is a constant, is the diameter
of reinforcing bar and is the steel ratio; postulates that / is a major factor in
determining the crack spacing.
This establishment of the dependence of crack
spacing on the ratio / proved to be of great importance in determining cracking in

reinforced concrete members.
Thomas (1936) in his work, pointed out that the crack widths depended
mainly on bar diameter, limiting bond stress, steel stress, reinforcing percentage and
the concrete strength or extensibility. His analysis also showed that the spacing and
width of cracks are proportional to the / ratio.
From his study of the mechanism of crack formation in flexural members,
Bornemann (1936) concluded that the width of cracks should increase with an
increase in the maximum tensile force in the concrete prior to rupture and decrease
with an increase in the ratio of the steel area to the area of the ruptured concrete.
Colonnetti (1936) made a theoretical analysis of the mechanism of crack
formation in flexural members and concluded that crack width decreased with a
reduction of the diameter of the reinforcing bars.
Investigations concerning cracks in symmetrically reinforced concrete
members were summarised by Watstein and Parsons (1943). They found that for a
5-7
given type of steel at a constant stress level the most prominent factor affecting
crack width and spacing was the ratio /, and that the crack width was almost
completely independent of the concrete strength.
Watstein and Seese (1945) studied the effect of bond efficiency of
different types of reinforcing bars on the width and spacing of cracks in concrete
cylinders subjected to axial tension.
They found that the width of the cracks
decreased with an increase of bond efficiency and that at a given steel stress the
crack width for various bars varied in a nearly linear relationship with the spacing of
the cracks. The latter finding was confirmed by Bjuggren (1948).
Watstein and Parsons (1943) equations for axially loaded reinforced concrete
members were also applied later by Clark (1956) to predict crack widths in flexural
members. The equations were modified incorporating the assumption that crack
width is proportional to (/) (h-d)/d instead of the simple term / where h is the
total depth and d is the effective depth of a reinforced concrete section.
A new concept was introduced in the analysis of the problem of crack
formation in a portion of a beam under pure bending by Chi and Kirstein (1958).
According to them, after some initial cracks had occurred the tensile force in the
concrete was resisted by an effective area of concrete immediately surrounding the
steel, which was less than the total area of the concrete in the tensile zone of the
beam. The average width of the cracks at the steel was given by the product of the
average minimum spacing of the cracks and a function of the computed steel strain.
Their test results also indicated that concrete strength in the range from 2000 to
6000 psi (13.8 to 41.4 MPa) had particularly no effect on the formation of cracks.
Kaar and Mattock (1963) conducted tests on flexural members reinforced
with high strength deformed bars. An analysis of the data resulted in a simple
empirical equation indicating that crack width is essentially proportional to steel
stress and, for a given steel stress, proportional to the area of concrete surrounding
each bar which is similar to the effective area as identified by Chi and
Kirstein (1958).
5-8
According to Gergely (1972), flexural crack width depends on geometrical

factors and on loading. The concrete cover and the spacing of the bars are of
primary importance. Loading affects crack width several ways. The crack width is
proportional to fsn, where fs is the steel stress and n is about 1.4; however, n can be
taken as unity without significant error. The distribution and width of cracks also
depends on the variation of moments along the member. For example, flexural
cracking in negative moment regions is different from that in simply supported
beams. The loading history is also important; repeated or sustained loads increase
the widths of cracks.
The magnitude of crack width depends on many factors, the most important
are the following (Suri and Dilger, 1986):
Type of reinforcement. Each reinforcement differs in bond characteristics and has
different load-slip behaviour with concrete so that the type of reinforcement
employed will have a significant influence on crack width.
Concrete cover. The effective concrete cover is taken as the distance from the
tensile face of the beam to the centroid of the nearest reinforcing bars. Previous
tests indicated a proportional relationship between the cover and the crack width.
Total area of steel. The reinforcement is often well distributed and is placed close to
the tension face. Hence it is most effective in crack control.
Strength of concrete. An increase in the strength of concrete for a beam with the
same reinforcement will increase the cracking moment and the concrete area in
tension.
The problem of crack formation and development is a complex one involving
great many parameters; hence some of the conclusions regarding the significance of
these parameters are questionable. However, general agreement does exist on the
following as was mentioned by Nawy (1968):
5-9
(a)
Crack width is a function of crack spacing up to a certain limit.
(b)
Crack width and spacing follow a normal distribution.
(c)
Crack width is a function of steel strain, hence stress. The relationship,

though nonlinear, sometimes conveniently approximates linear relationship
in case of beams.
(d)
Shrinkage strain and tensile strain in the concrete zone between any two
cracks are very small and can be neglected.
(e)
The magnitude of concrete cover has an important effect on crack width.

Recently, Makhlouf and Malhas (1996) tested 32 reinforced concrete
rectangular beams with thick concrete cover to investigate the effect of the increase
in concrete cover on the maximum crack width. Comparison of their test results
with the computed values for the maximum crack width (according to different code
formulae) showed that the codes overestimate the sensitivity of maximum crack
width to the increase in concrete cover. It has also been shown that the use of a
50-mm concrete cover leads to acceptable levels of crack widths under service loads.
The ability of the codes equations to correlate with the experimental results has
been observed to depend on the reinforcement ratio and the load level.
5.2.4
Mechanism of flexural cracking
The width and spacing of cracks in reinforced concrete members are influenced by
many variables. Because of the complexity of the problem, there are a number of
approximate, semi theoretical and empirical approaches for the determination of the
width of cracks, each approach containing a selection of the variables. Some of the
methods are reviewed below to indicate their background.
(a)
5-10
Classical Theory
In the mechanism of cracking of axially loaded reinforced concrete members

proposed in most of the early studies on cracking, crack control was believed to
depend largely on the quality of bond between the concrete and steel.
In an axially loaded tension member, initial tension cracks form when the
tensile strength of the concrete is exceeded at weak sections that are randomly
distributed. Slip occurs between the concrete and steel at the cracks. At the cracks
the concrete is free from stress, and the reinforcement alone carries the external load.
Flexural tensile stress in the concrete between cracks is present because of bonding
action that takes place between the concrete and the reinforcement. In this area,
stress is transferred by bond from the reinforcement to the concrete. The magnitude
and distribution of bond stress between the cracks determines the distribution of
tensile stress in the concrete and the steel between the cracks.
Additional cracks will form between the initial cracks at higher loads. These
secondary cracks (see Section 5.2.2) will continue to form if enough force is
transferred from the steel to the concrete to develop strains that will exceed the
strain capacity or the tensile strength of the concrete.
The foregoing hypothesis known as the classical theory was formalised by
Watstein and Parsons (1943), and several other theories appeared later. Hognestad
(1962) has described the derivation of the theoretical equations as follows.
For a reinforced concrete axially loaded member (see Figure 5.2a), initial
tension cracks form at irregular intervals along the length of the member when the
tensile strength of the concrete is exceeded at weak sections. Additional cracks form
between the initial cracks at higher loads, but the crack spacing can only be reduced
down to a certain minimum spacing, lmin.
The presence of existing primary cracks affects the formation of secondary
cracks under increasing load. Theoretically, a secondary crack should occur midway
between existing cracks. If cracks form initially at sections A and C (Figure 5.2a),
5-11
lmin
Section
Ae = h b
(a) Member with axial tension
lmin
C
b
l
Section
Ae = 2 (h - d) b
(b) Member with flexure
Figure 5.2 Members with cracking

with a spacing, l, slightly greater than twice the minimum crack spacing, lmin, then a
new crack will form between A and C at section B when there is an increase in the
axial load. The distance between A and B or C and B is at least equal to or greater
than lmin. However, if two initial cracks happen to form at a spacing slightly smaller
than 2 lmin, then a new crack cannot form between these two initial ones. Thus the
crack spacing at ultimate loading can be expected to vary between lmin and 2 lmin
under normal conditions.
The average crack spacing, lcr, under ultimate load,
theoretically, will have the following value:
lcr = 1.5 lmin = 0.75 lmax
5-12
(5.2)
where lmax is the maximum crack spacing. The above reasoning indicates that in
practice there will be a large scatter in crack spacing; crack spacings ranging
between 0.67 and 1.33 of the average spacing are theoretically possible.
The tensile force required to crack the concrete is Aeft, where Ae is the
effective area of concrete in tension and ft is the tensile strength of the concrete.
The tension transferred to the concrete is lmin u 0, where u is the average bond
stress, and 0 is the sum of the perimeters of the bars. Equating these two tension
values gives
lmin =
Ae f t
'
(5.3)
u 0
For bars of the same diameter, 0 = 4 As /, where As is the steel area and
is the bar diameter. Also substituting e = As / Ae into Eq. (5.3) gives
'
lmax = 2 lmin =
ft
2ue
(5.4)
The crack width is given by the elongation of the steel between two cracks minus the
elongation of the concrete. Ignoring the elongation of the concrete as small, the
maximum crack width, wmax, is given by lmax fs / Es, where fs is the steel stress and Es
is the modulus of elasticity for steel. Substituting lmax from Eq. (5.4) into the
expression for maximum crack width gives
wmax =
fS
e K1
(5.5)
where K1 = 2 u Es / ft.
5-13
This basic equation for maximum crack width has been modified by many
researchers. The above derivation assumes that the tensile stress in concrete at
section B of Figure 5.2a is uniform and the effective area of concrete in tension, Ae,
is therefore the whole cross section of the member. This assumption is questionable
because the actual distribution of tensile stress may be highly nonuniform. It is also
assumed that the opening of cracks is due to slip of the concrete relative to the
reinforcement, that the spacing of cracks is determined by the force that can be
transmitted from the steel to the concrete by bond, and that the crack has parallel
sides (i.e. a constant width) through the thickness of the member (Park and
Paulay, 1975).
Further assumptions are required to apply Eq. (5.5) to the bending of a beam
as in Figure 5.2b. The effective area of concrete in tension, Ae, must be suitably
defined. Generally Ae is taken as the area of concrete having the full width of the
beam and having the same centroid as the main reinforcement, as in Figure 5.2b.
Attempts to apply Eq. (5.5) to beams have demonstrated the need to reduce the
effect of and e.
Modified forms of Eq. (5.5) have been suggested based on comparison with
test results.
For example, Kaar and Mattock (1963) of the Portland Cement
Association modified Eq. (5.5) to express the maximum crack width at the level of
reinforcement as
wmax = 0.115
A fs x 10-6 in.
(5.6)
where A = area of concrete surrounding each bar (A = Ae/n, where n is the number
of bars) in square inches, and the steel stress fs is in psi (1 in. = 25.4 mm; 1 psi =
0.006895 N/mm2). The measured maximum crack widths from which Eq. (5.6) was
derived showed a scatter of up to 40% from the equation. To obtain the maximum
crack width at the extreme tension fibre for beams reinforced with deformed bars,
Kaar and Hognestad (1965) modified Eq. (5.6) to
wmax = 0.115
A fs
h2
x 10-6 in.
h1
5-14
(5.7)
where h1 = distance from the centroid of tension steel to the neutral axis and h2 =
distance from the extreme tension fibre to the neutral axis.
(b)
No-Slip Theory
A fundamentally different approach called the no-slip theory was proposed by Base
et al. (1966). They assumed that for the range of crack widths normally permitted in
reinforced concrete, there is no slip of the steel relative to the concrete. The crack is,
therefore, assumed to have zero width at the surface of the reinforcing bar and to
increase in width as the surface of the member is approached. This means that the
crack width is dependent on the deformations of the surrounding concrete. The
theory of elasticity can be used to determine the distribution of stress and strain in
the concrete between the cracks. The stresses so calculated indicate when further
cracking is likely; the strains indicate the deformed shape of the concrete surface,
hence the likely width of the cracks.
Based on results from 107 beam tests at Cement and Concrete Association,
Base et al. (1966) gave the following formula for the prediction of the maximum
crack width at the lower extreme tensile fibres of beams:
wmax = K c
f s h2
Es h1
(5.8)
where c = distance from the point at which the crack width is to be determined to the
surface of the nearest reinforcement bar, and K = 3.3 for deformed bars and 4.0 for
plain bars. Es, fs, h1 and h2 were defined earlier.
(c)
A Statistical Approach
Gergely and Lutz (1968) analysed statistically the flexural crack data taken from six
investigations with the aid of a computer to determine the importance of the
5-15
variables involved. Many combinations of variables were tried, and it was very
difficult to obtain an equation that fitted all sets of data well.
The important
variables were found to be the effective area of concrete in tension, Ae, the number
of bars, the side or bottom cover, the strain gradient from the level of the steel to the
tension face and the steel stress. Of these, the steel stress was the most important.
The following equations were developed for predicting the most probable maximum
crack widths.
For the most probable maximum crack width at the level of the
reinforcement
ts A
(fs - 5) x 10-3
1 + t s / h1
3
wmax = 0.091
(5.9)
where wmax is in in., A is the average effective area of concrete in tension around
each reinforcing bar (= Ae/n, where n is the number of bars) in in.2, ts is the side
cover measured from the centre of the outer bar in in., fs is in ksi and h1 is in in.
(1 in. = 25.4 mm; 1 ksi = 6.895 N/mm2).
For the most probable maximum bottom crack width on the bottom (or
tension) face of the beam
wmax = 0.091
tb A
h2
(fs - 5) x 10-3
h1
(5.10)
where tb is the bottom cover measured from the centre of the lowest bar in in. and h2
is in in. while the other variables are in the same units as in Eq. (5.9).
Nawy (1968) has compared the accuracy of a form of Eq. (5.9) with
experimental maximum crack width data from the tests conducted by himself,
Hognestad (1962), Kaar and Mattock (1963), and Base et al. (1966). The scatter of
data about the predicted maximum crack widths was very considerable.
Nawy (1968) also compared the maximum crack widths predicted by Eqs. (5.6) and
(5.8) with the experimental data and found a wide scatter. From their tests on oneway slabs reinforced by deformed bars, deformed wires, deformed wire fabric and
5-16
smooth wire fabric, Lloyd et al. (1969) concluded that the Gergely and Lutz (1968)
equations (Eqs. (5.9) and (5.10)) satisfactorily predicted the maximum crack width.
(d)
More General Approach
A more general approach has been considered by a few researchers to study the
cracking behaviour of reinforced concrete members.
Bianchini et al. (1968) discussed the cracking phenomenon and three
cracking mechanisms in terms of the simple model of a reinforced concrete member
with the reinforcement loaded in direct tension. One mechanism that has been
proposed or implied in many cracking studies (classical theory as discussed above)
is based on a uniform tensile stress distributed over an effective concrete area. A
second mechanism is based on redistribution of concrete stresses and observed
internal cracking.
A third mechanism based on similarities between internal
cracking of reinforced concrete and fracture toughness is proposed.

A clear understanding of the mechanism of cracking, however, resulted from
the work of Beeby (1970). Beeby (1970) measured crack widths and spacing at
various points across the bottom of one-way reinforced concrete slabs, that is, for
various values of c as defined in Eq. (5.8). It was found that the crack spacing and
width increased with distance from the bar and at some distance from the bar
approached constant values, which were dependent on the crack height rather than
the distance from the bar. Beeby (1970), therefore, concluded that the crack pattern
at any point was the result of interaction between two basic crack patterns:
(1) cracking at a point distant from a reinforcing bar, i.e. the crack pattern controlled
by the crack height; and (2) cracking directly over a reinforcing bar, i.e. the crack
pattern controlled by the concrete cover.
5.2.5
Evaluation of crack spacing and crack width
The use of high strength reinforcement and methods of reducing crack formation
were published by the International Congress for Bridge and Structural Engineering
in 1936 (Bornemann, 1936; Saliger, 1936).
5-17
In the 1930s and 40s extensive
experimental work was done to determine the significance of the various cracking
parameters especially those concerned with the bond characteristics of various types
of bars. Further attempts were made to establish permissible crack widths for
various corrosive conditions. The RILEM (1957) symposium on bond and crack
formation can be considered a significant step towards development of knowledge
on the cracking problem in reinforced concrete structures. The papers indicated that
more studies were necessary to reduce the discrepancy between the cracking
equations and approaches of the various investigators.
It is only recently that several expressions for crack control in beams have
been proposed which seem to offer the least deviation from available test data.
Expressions similar to those presented by Watstein and Parsons (1943) for
axially loaded members were proposed by Clark (1956) to predict crack widths in
flexural members. Clarks equation for computing average crack width, wcr, in
reinforced concrete flexural members, thus, was
h d
wcr = 2.27 x 10-8
[f - 56.6 1 + 8 ]
s
(5.11)
where wcr, h, d and are in in., and fs is in psi (1 in. = 25.4 mm; 1 psi = 0.006895
N/mm2).
Formulae for the estimation of the crack width at the tension face or at the
level of reinforcement of flexural members proposed by Kaar and Mattock (1963),
Base et al. (1966) and Gergely and Lutz (1968) have already been discussed in
Section 5.2.4.
Beeby (1971) simplified his earlier equations to give the maximum crack
width that will be exceeded by approximately 20% of the results as
wmax =
3c m
1 + 2(c c0) / ( h kd)
(5.12)
5-18
where c = distance from point of measurement of crack to surface of the nearest bar,
c0 = minimum cover to steel, h = overall depth of the section, kd = neutral axis depth
and
2.5bh
h kd
m = s
x 10 6
d kd
As
(5.13)
where s = strain in the steel at a crack and As = area of tension steel. m is the steel
strain at a crack less an empirical term due to the stiffening effect of concrete tension
between cracks, and modified by the strain gradient term to obtain the average strain
at the extreme tension fibre of the member.
Earlier Ferry Borges (1966) developed the following equation for the
maximum crack width in beams reinforced by deformed bars:
wmax =
1
2.5 dc + 0.066
Es
w
) (f - 107
) in.
(5.14)
where dc = thickness of concrete cover over bar (in.), Es and fs are in psi, is in
inches, and w = As/bwd; As = steel area (in.2), bw = web width (in.), d = effective
depth (in.) : 1 in. = 25.4 mm, 1 psi = 0.006895 N/mm2.
It has been observed by several researchers that the spacing of cracks
depends largely on the size of reinforcing bars and on the concrete cover. Possibly
the simplest expression for crack spacing and one, which takes account of these
factors, is that proposed by Welch and Janjua (1971):
lcr = (1.5 dc + 3.0 )
(5.15)
where lcr is the average crack spacing, dc is the concrete cover measured to the bar
centre and is the bar diameter. Their final expression for maximum crack width is
fs
( h kd )
wmax = 1.5 (1.5 dc + 3.0 ) 0.0001
Es
(d kd )
5-19
(5.16)
in which all the variables are as defined earlier.

Oh and Kang (1987) proposed accurate prediction formulae for maximum
crack width and average crack spacing in reinforced concrete flexural members.
These formulae were based on the cracking theory developed by Bazant and
Oh (1983c) which, in turn, was based on the energy criterion of fracture mechanics
as well as the strength criterion. Oh and Kang (1987) proposed the maximum crack
width equation that gave the best prediction as
(h kd)
w max
= a0 (s - 0.0002)
(d kd)
(5.17)
where wmax represents the maximum crack width at the extreme tension face, s is
the axial tensile strain of bars, and
( )
4 .5
dc
a0 = 159
h kd
( )
A
+ 2.83
A st
(5.18)
where Ast is the area of each reinforcing bar and other variables are as defined
earlier.
Their equation for the average crack spacing lcr was as follows:
0.236x 10-6
l cr
= c0 +
2
(5.19)
in which
( )
dc
c0 = 25.7
h kd
4 .5
( )
A
+ 1.66
A st
(5.20)
5-20
where all the variables are as defined previously.

On the basis of the test data obtained from a total of 205 reinforced concrete
members, new formulae for predicting the crack spacing and crack width in
reinforced concrete members were developed by Lan and Ding (1992).
Their
equation for predicting average crack spacing in flexural reinforced concrete

members was
lcr = (2.7 c + 0.11 /e)
(5.21)
where e = As/Ae, As, Ae, c and were defined earlier, and = 0.7 for deformed
bars and 1.0 for plain bars. This formula is applicable to compressive members with
eccentric load too.
Their equation for calculating maximum crack width in flexural members
and compression members with an eccentric load was
wmax = 1.41 (fs/Es) (2.7 c + 0.11 /e)
(5.22)
in which
= 1.1 - 0.65 ft / (e fs)
(5.23)
where ft is the tensile strength of concrete. Other variables were defined earlier.
Zhao and Wang (1993) developed a unified formula for evaluating the
maximum crack width of reinforced concrete members subjected to pure tension,
bending, eccentric tension or compression as follows:
0.5
fs
wmax = k1 k2 k3
(1 - f t
Es
e fs
'
) (3 c + 0.12
)
e
(5.24)
5-21
where k1 is the coefficient denoting the load effect and is 1.0 for flexural members,
0.9 for eccentric compression members, 1.1 for eccentric tension members, and 1.2
for axial tension members, k2 is the coefficient denoting the bond effect of the bars
and is 1.0 for deformed bars and 1.3 for plain bars, k3 is the coefficient denoting the
effect of the duration of the loading and is 1.0 for the short-term load and 1.5 for the
long-term load. Other variables were defined earlier.
Recently, Ouyang and Shah (1994) proposed a fracture energy approach
based on nonlinear fracture mechanics to predict cracking of reinforced concrete
members subjected to tension. They derived a closed-form solution to predict the
minimum reinforcement ratio for tensile members, and this minimum reinforcement
ratio was shown to depend on the size of the members.
Finally, the formulae for estimating crack widths in flexural reinforced
concrete members adopted by different national and international codes are
discussed as follows.
(a)
The SAA Approach
The Australian Standard AS 3600 (SAA, 1994) does not recommend any formula
for the calculation of crack widths. Crack control is deemed to be satisfactory in
flexural members (Clause 8.6.1) if the centre-to-centre spacing of bars near the
tension face of the beam does not exceed 200 mm and the distance from the side or
soffit of a beam to the centre of the nearest longitudinal bar is not greater than
100 mm.
In addition, AS 3600 requires that for crack control in the side faces of
beams where the overall depth exceeds 750 mm, longitudinal reinforcement,
consisting of Y12 bars at 200 mm centres, or Y16 bars at 300 mm centres, must be
placed in both side faces of the beam.
(b)
The ACI Approach
Requirements for crack control in beams and thick one-way slabs in the ACI
5-22
Building Code (ACI, 1995) are based on the statistical analysis of maximum crack
width data from a number of sources.
The analysis was due to Gergely and
Lutz (1968). The equation which was considered the most accurate in predicting the
most probable maximum crack widths at the bottom of beam is
wmax = 0.0132 fs
d c A x 10-3
(5.25)
where wmax is the maximum crack width at bottom of beam in mm, dc is in mm, fs is
in N/mm2, and A is the concrete tension area surrounding each reinforcing bar
(= Ae/n, where n is the number of bars) in mm2. This is the crack width formula
adopted in the ACI code.
(c)
The British Standard Approach
The British Standard BS 8110 (BS, 1985) uses the expression developed by
Beeby (1970; 1971) to compute the crack width for a given section with a known
reinforcement. Beeby (1971)s equation, Eq. (5.12) is adopted as the formula to
compute the crack width at any point by the British code. Directly over a bar, the
distance c is equal to the concrete cover c0, and Eq. (5.12) reduces to
wmax = 3 c0 m
(5.26)
where m is obtained using Eq. (5.13).

(d)
The Eurocode Approach
Eurocodes are the European standards for structural design. Like the present UK
codes of practice, Eurocodes will come in a number of parts, covering a range of
applications. These documents are at various stages of development. For instance,
Part 1.1 of Eurocode 2 (EC2, 1991) which is similar in scope to BS 8110 (BS, 1985)
has already been finalised and issued as preliminary standards (ENV) (Prnorme
Europenne) (Arya, 1994). The first draft of Eurocode 2 for concrete structures was
based on the CEB-FIP Model Code (CEB, 1978) of 1978 drawn up by a number of
experts from various European countries.
5-23
The Eurocode 2 (EC, 1991) recommended equation for the calculation of

average crack width in reinforced concrete members is
wcr =

50 + 0.25 k1 k2
r
(5.27)
where is the average bar diameter in mm, k1 is a coefficient that takes account of
the bond properties of the bars and is 0.8 for high bond bars and 1.6 for plain bars, k2
is a coefficient that takes account of the form of the strain distribution and is 0.5 for
bending and 1.0 for pure tension, r is the effective reinforcement ratio, As/Acef,
where As is the area of reinforcement contained within the effective tension area,
Acef, and m is the mean strain in the tension reinforcement allowing for the effects
of tension stiffening, shrinkage, etc. under the relevant load combinations. The
effective tension area Acef is generally taken as Acef = 2.5 b (h-d), where b is the
width, h is the total depth and d is the effective depth of the section, but the height of
the effective area should not be greater than (h kd/3) where kd is the neutral axis
depth.
Thus, the Eurocode 2 (EC2, 1991) equation for evaluating average crack
width in a flexural member reinforced with high bond bars would be
wcr =
5.3
{50 + 0.1 }
(5.28)
Cracking in Partially Prestressed Concrete Beams
In a fully prestressed concrete beam which remains in compression under service

load cracking usually starts only in the overload condition, whereas in a partially
prestressed beam permanent open cracks of limited width may occur at service load
level. Cracks form in these members when the tensile stress exceeds the modulus of
rupture of the concrete. The control of these cracks is necessary mainly for aesthetic
5-24
reasons. The residual crack width, after removal of the major portion of the live
load, is small [about 0.001 to 0.003 in. (0.03 to 0.08 mm)] and therefore, crack
control is usually not necessary if the live load is transitory (ACI Committee 224,
1993). However, due to the fact that steel used for prestressing is more sensitive to
corrosion than ordinary reinforcing steel, smaller crack widths than those in
reinforced concrete members should be the goal of the designer.
As discussed in the previous sections, numerous investigations have been
carried out in the past directed towards developing equations for the spacing and
width of flexural cracks in reinforced concrete members. However, very limited
information exists on crack evaluation in prestressed concrete structures.
The
available experimental data are limited and, at the same time, the number of
variables is greater in prestressed members. It is, therefore, necessary to know about
the causes and types of cracking as well as the factors affecting crack widths in
prestressed concrete members for the purpose of the development of the proposed
crack width formula. These along with the available crack width and crack spacing
formulae are discussed in the following sub-sections.
5.3.1
Causes of cracking
Deformation, cracking and redistribution of moments in reinforced and in

prestressed concrete structures are closely interrelated. The fundamental cracking
mechanism in prestressed concrete elements is in many aspects similar to that of the
reinforced concrete elements. However, the external compressive prestressing force
imposes large lateral compression on the prestressed concrete section. Cracking can
develop only if a decompressive transverse force due to external live load causes the
extreme tensile fibres of the critical section in a beam to reach the rupture strength of
concrete. This concept of decompression allows the engineer to reduce the problem
of cracking in prestressed concrete to the problem of cracking in reinforced concrete
(Huang, 1975).
Cracking in prestressed concrete beams mainly starts either in over-load
5-25
conditions or, in the case of partially prestressed beams, at service loads.
An
analysis of the prestressed beam in the cracked state can be made when the effects of
prestressing are determined, namely, the net stress, strain, eccentricity of prestress,
shrinkage and creep. The equations for the prediction of crack width in nonprestressed beams may also be used, with some modifications, for prestressed
beams. The steel stress should be that caused by difference between the acting
moment and the decompression moment (that results in zero extreme fibre stress).
Somewhat better correlation exists with data if the calculations are based, after
appropriate modifications, on the cracking moment (Gergely, 1972). Repeated load
tests indicate that some residual crack width (about 0.0008 in. or 0.02 mm) is present
in prestressed beams with limited prestress and this value is to be added to
calculated crack widths (Bennett and Chandrasekhar, 1971). The zero-load crack
width is appreciable in prestressed beams with little or no nonprestressed steel
because of the weaker bond properties of most prestressing steels.
Surface conditions of steels seem to affect cracking significantly.
The
calculated values for deformed bars should be increased by a factor of about 1.6 if
strands are used, and by a factor of about 2.5 if smooth prestressing wires are
employed (Bennett and Chandrasekhar, 1971; Holmberg and Lindgren, 1970).
As for reinforced concrete members, the development of cracks in
prestressed members depends mainly on the amount of reinforcement or prestressing
steel and on its distribution. If the amount is small the increase in deformation
which occurs during cracking will be greater than that which occurs with a larger
quantity, and consequently cracks will appear under a smaller stress in the first case
than in the second. Similarly, if a small number of prestressing bars or cables, each
of larger diameter, are used cracking will occur at a smaller load than would be the
case if an equal area of bars or cables, each of smaller diameter, were provided.
Another major influence is the thickness of concrete covering the steel; the greater
the
cover,
the
sooner
the
cracks
will
become
visible
(Abeles
and
Bardhan-Roy, 1981).
5.3.2
5-26
Types of cracking
Types of cracks developed under load in partially prestressed concrete members are
very much similar to those developed in reinforced concrete members. However,
longitudinal cracks frequently occur in the anchorage zones of prestressed concrete
members due to transverse tensile stresses set up by the concentrated forces
(Gergely, 1969; Zielinski and Rowe, 1960). Such cracks may lead to (or in certain
cases are equivalent to) the failure of the member.
Transverse reinforcement
(stirrups) must be designed to restrict these types of cracks.

Two types of cracks may develop: spalling cracks which begin at the end
face (loaded surface) and propagate parallel to the prestressing force, and bursting
cracks which develop along the line of the force or forces, but away from the end
face (ACI Committee 224, 1993).
In this research the cracks of concern are the flexural cracks only. The initial
flexural cracks develop in prestressed concrete members when the actual tensile
stress due to load exceeds both the effective pre-compression and if existent, the
tensile resistance of the concrete. The pre-compression is the normal stress due to
the prestressing force in the concrete cross-section. This normal stress can be
replaced with an axial force and a bending moment. The latter forces are the
decompressive forces which enable a prestressed concrete member to be treated as a
reinforced concrete member. The decompresssed state under which the normal
stresses vanish from the concrete section can be taken as the equivalent of an
unloaded reinforced concrete element and the crack formation development can then
be studied by the theories applying to the reinforced concrete member
(Huang, 1975).
5.3.3
Factors affecting crack widths
Available experimental data for the prediction of crack width in prestressed concrete
5-27
members is limited. Whether prestressing is full or partial there are many common
parameters which influence the cracking problem.
One approach of crack control in prestressed concrete members is to limit the
calculated fictitious stress at the tensile face. If the prestressing steel tendons and
nonprestressed bars are well distributed, the magnitude of the fictitious stresses is a
good indication of the expected maximum crack width (Abeles, 1967).
The concept, originated from Abeles (1967), states that crack width is related
to the flexural tensile stress, ftn, in the concrete. The following are concrete tensile
stresses, ftn, corresponding to crack width limits shown.
(a)
Crack width, w = 0.004 in.

ftn = 800 + 500 ( - 0.3) psi for strands
ftn = 700 + 450 ( - 0.3) psi for round bars and wires
(b)

(c)

where = total steel percentage of the section (1 in. = 25.4 mm; and 1 psi =
0.006895 N/mm2).
The crack width formulae proposed by Ferry Borges (1968), CEB (1970),
Holmberg
(1970),
Holmberg
and
Lindgren
(1970),
and
Nawy
and
Potyondy (1970; 1971) can be considered to have the same concept, namely that the
crack width is related to the stress or strain in the reinforcement.
5-28
The influences of the various parameters affecting cracking in prestressed

concrete members are generally agreed upon by most investigators. The following
in approximate order of significance are the main factors involved in the control of
the crack pattern:
(a)
reinforcement stress,
(b)
the bond characteristics of reinforcement,
(c)
the distribution of reinforcement over the effective concrete area,
(d)
the percentage of reinforcement,
(e)
the concrete cover over the reinforcement and
(f)
the material properties of the concrete.
5.3.4
Evaluation of crack spacing and crack width
The increased use of partial prestressing, allowing limited tensile stresses in the
concrete under service-load and overload conditions while allowing nonprestressed
steel to carry the tensile stresses, is becoming prevalent due to practicality and
economy. Consequently, an evaluation of the flexural crack widths and spacings
and control of their development become essential. Work in this area is relatively
limited because of the various factors affecting crack width development in
prestressed concrete. However, experimental investigations support the hypothesis
that the major controlling parameter is the reinforcement stress change beyond the
decompression stage (Nawy, 1996).
Because of the importance of serviceability behaviour of prestressed concrete
members, several experimental and analytical investigations have been undertaken
and expressions proposed. These are described in the following paragraphs.
lcr = 6.0 + 0.8
B0
5-30
(5.32)
where c is the concrete cover, B0 is the maximum concrete area whose centre of
gravity coincides with that of the principal reinforcement and is the sum of bar
diameters.
The mean crack width, wcr, at the level of the reinforcement should be
wcr = s lcr
(5.33)
where s is the strain of the reinforcement above the decompressed stage.

The maximum crack width was assumed to be 1.7 times the mean crack
width.
The CEB-FIP recommendation (CEB, 1970) states that the maximum crack
width, wmax (in mm), is a function of the net steel stress, fs (in N/cm2), at the level
of steel. The equations are given below.
For non-repetitive loads
wmax = (fs - 4000) x 10-5
(5.34)
and for loads repeated more than 100 times

wmax = fs x 10-5
(5.35)
Huang (1975) proposed the following expression (later on reported by

Nawy, 1996) for calculating maximum crack width, wmax, at the reinforcement level:
( )
wmax = 1.4 x 10-5 lcr f s
1.31
(5.36)
5-31
where wmax is in inches, lcr is in inches and fs is in ksi (1 ksi = 6.895 N/mm2).
Simplification of Eq. (5.36) by linearization and using simplified expression
of crack spacing results in
wmax = 5.85 x 10-5
Ae
0
(fs)
(5.37)
where Ae is the effective concrete area in uniform tension in square inches, and 0 is
the sum of bar perimeters in inches.
After analysing various investigators results (Harajli and Naaman, 1989;
Naaman and Siriaksorn, 1979; Siriaksorn and Naaman, 1979), Naaman produced the
following modified expression for partially prestressed pretensioned members:
Ae
wmax (in.) = 42 + 558

( f s) x 10-5
.
(5.38)
The expression developed by Nawy (1984) for the crack width (at the
reinforcement level closest to the tensile face) in post-tensioned bonded beams
which contain mild steel reinforcement is given as
wmax (in.) = 6.51 x 10-5
Ae
0
(fs)
(5.39)
For non-bonded beams, the factor 6.51 in Eq. (5.39) becomes 6.83.
Suzuki and Ohno (1984) proposed crack width equations for prestressed
concrete beams and slabs introduced by reference to the equation for reinforced
concrete members in the CEB-FIP Code (CEB, 1978). They were based on the
experimental results of about 140 beam specimens and about 50 slab specimens
tested.
Their formula for the evaluation of maximum crack width, wmax, in
reinforced as well as prestressed concrete beams is

1.5 [2(c +
) + 0.1
] s
e
10
5-32
(5.40)
where c is the concrete cover, s is the spacing of reinforcing bars, is the diameter
of reinforcing bar, e (= As/Ae) is the effective steel ratio, and s is the change in
average steel strain from the stage of decompression given by
1
s =
Es
'
f
f s k1 k 2 t
e
(5.41)
where fs is the change in steel stress in a cracked section from the stage of
decompression, k1 and k2 are constants, and other variables are as defined before.
Values of s should be such that s 0.4 fs/Es.
Recently, Scholz (1993) described a new method to compute the crack width
attributable to the loading of prestressed flexural members. He made use of the
similarities that exist between the deflection theory and the crack theory of
prestressed sections and elements. The principal significance of his new method lies
in its compatibility with deflection calculations for the same structural element. The
method lends itself also to the formulation of cracking checks that do not require the
explicit computation of the crack width. Scholz (1993) expressed the average crack
width, wcr, as a function of the cracked immediate deflection component, c, as
follows:
wcr = lcr
(1 kd / d ) c / L
L/d
(5.42)
where lcr is the average crack spacing, kd is the neutral axis depth, d is the effective
depth, L is the span, and is the deflection coefficient. Scholz (1993) suggested to
obtain the average crack spacing, lcr, from the first principle as demonstrated by Park
and Paulay (1975) or Ghali and Favre (1986).
The formulae proposed by the
CEB-FIP Code (CEB, 1978; CEB, 1985) were also recommended.
5.4
5-33
Summary
As demonstrated in this chapter, the calculation of crack width is a complex

problem. There are many factors that cause cracking in concrete. The crack caused
by loading is the main factor considered in design. The other factors are usually
eliminated or reduced by selecting suitable material and improving the quality of
construction. Because of the complexity of the problem, a number of formulae are
currently in use for calculating crack width in reinforced and prestressed concrete
flexural members.
Also, the spacing and width of cracks under immediate and long-term
loading are still not known for all conditions and their predictions are certainly not
agreed upon among the different investigators.
The available expressions to
evaluate the crack widths and those adopted by different codes are approximate and
give a wide scatter of predicted values. Furthermore, there are not many crack width
formulae applicable to both reinforced and prestressed concrete beams.
To ensure the reliability and accuracy of the analysis proposed in this
research, it is necessary to develop a new unified formula for evaluating crack width
and spacing for both the reinforced and the partially prestressed concrete beams
studied in the current experimental work.
The applicability of the formula so
developed, to predict crack widths at the negative and positive regions of the twoequal-span continuous beams tested, will also be verified.
Chapter 6 : The Experimental Programme
6-1
CHAPTER 6
THE EXPERIMENTAL PROGRAMME
6.1
General Remarks
This chapter presents the test programme planned and carried out to investigate the
damping and cracking behaviour of reinforced and partially prestressed concrete
beams. Also presented is the underlying theory as proposed in Chapter 4 which
guided the experimental work.
A total of 26 reinforced and partially prestressed concrete full-size box
beams were tested in the laboratory. The specimens were in three groups: (1)
reinforced concrete simply supported box beams with various levels of steel ratios;
(2) partially prestressed concrete simply supported box beams with different degrees
of prestressing; (3) two-equal-span continuous reinforced concrete box beams with
different steel ratios. In addition, 4 more solid rectangular reinforced concrete
simply supported beams were tested for verification purposes.
The tests were carried out in two stages. The first stage involved the testing
of 9 reinforced and 12 partially prestressed concrete simply supported box beams
and were carried out by Lu (1993), a former research assistant of the principal
supervisor of the present work. The rest of the beams (all nine of which in the
second stage) were designed, fabricated and tested by the author, with assistance
received from James (1997), Salzmann (1997) and Stewart (1997), as part of their
6-2
Honours thesis research. The main reason for using box beams in this research was
that the first-stage beams were originally intended for impact testing. Also, because
of the long spans of the full-size beams, the volume and the weight were of concern
for lifting and handling in the laboratory. Box beams, of course, are widely used in
bridges and other large-span constructions.
In this chapter, details of the specimens are presented including lengths and
cross sections of beams, steel ratios, compressive strengths, degrees of prestressing,
as well as reinforcement details. The support systems used, the test set-ups and the
loading systems are also described.
6.2
Background Theory
As proposed in Chapter 4, the damping values in terms of logarithmic decrement

measured for the reinforced and partially prestressed concrete beams from the free
vibration tests are to be related to the residual crack widths for the beams. The
residual crack widths, in turn, are to be related to the instantaneous crack widths as
well as the mid-span deflections of the beams. The proposed approach is discussed
in detail in Chapter 4, especially Section 4.6.
To produce cracking in the beams, static loading was applied. At each level
of loading, the instantaneous crack widths and the mid-span deflections were
measured. The residual crack widths of the beams were measured at zero loads after
each increasing level of loading had been applied. The beams were subjected to free
vibration tests at the same time, at zero loads, as well. The details of these tests and
measurements are discussed in the relevant chapters and sections.
6.3
Reinforced Concrete Specimens
A total of 18 full-size reinforced concrete beams were constructed and tested. These
comprised of 11 simply supported box beams, 3 two-equal-span continuous box
6-3
beams and 4 simply supported solid rectangular beams. The box beams were of
various lengths and each had an overall cross section of 300 mm x 300 mm. The
void of 180 mm x 180 mm in each beam was created by embedding polystyrene
prisms. The solid rectangular beams were each 2.5 m long and 150 mm x 250 mm
in cross section. The other details of these beams are described in the following subsections.
6.3.1
Simply supported beams
A total of 15 reinforced concrete simply supported beams were tested. The main
design properties of these beams are summarised in Table 6.1. The reinforcing
details and geometry are shown in Figures 6.1 through 6.8.
As can be seen from Table 6.1, the box beams were of three different overall
lengths - 5.5 m, 6.7 m or 8.0 m. They were reinforced with either 3-Y20 bars, or
6-Y20 bars, or 6-Y24 bars, having different steel ratios as shown. Mild steel hot
rolled deformed bars of grade 400Y (with a minimum yield strength, fsy, of
400 MPa) were used as longitudinal reinforcement and 250R plain bars (with
fsy = 250 MPa), for the vertical ties (stirrups). The stirrups used were R6 plain bars
(see Figures 6.1 through 6.5).
As for the four solid rectangular beams, they were all of the same overall
length i.e. 2.5 m, and each was reinforced with 3-Y20 bars. But they had three
different steel ratios because of the different types of stirrups used. As they were
used for the verification of the proposed crack width formula only, they had either
no stirrups or stirrups of different types (see Figures 6.6 through 6.8). As will be
shown later in Section 6.7, each of them was loaded differently as well.
While beams 5, 7 and 12 through 15 were the second-stage beams, the rest
were the first-stage beams.
6-4
Table 6.1 Details of reinforced concrete simply supported beams

Reinforcement details
box beam
box beam
box beam
box beam
box beam
box beam
box beam
box beam
box beam
number
of bars
3
3
3
3
3
6
6
3
6
bar diameter,
(mm)
20
20
20
20
20
20
20
20
20
Steel
area,
As
(mm2)
942
942
942
942
942
1884
1884
942
1884
box beam
box beam
solid beam
solid beam
solid beam
solid beam
6
3
3
3
3
3
24
20
20
20
20
20
2712
942
942
942
942
942
Beam
number
Type of
beam
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
6.3.2
Reinfor
-cement
ratio,
0.01154
0.01154
0.01154
0.01154
0.01163
0.02309
0.02326
0.01154
0.02309
Beam
length
L
(m)
5.5
5.5
5.5
5.5
5.5
5.5
5.5
6.7
6.7
0.03348
0.01154
0.02990
0.02935
0.02855
0.02855
6.7
8.0
2.5
2.5
2.5
2.5
Two-equal-span continuous beams
Three two-equal-span continuous reinforced concrete box beams were constructed

and tested. These were tested to check the applicability of the damping and crack
width prediction formulae developed for simply supported beams to the individual
spans of continuous beams.
Each beam was 300 mm x 300 mm in cross section with a void of 180 mm x
180 mm created by permanently embedding polystyrene prisms. Thus both the web
and the flange thickness were 60 mm each. Each beam was 12 m long with two
equal spans.
2 R6
150 150
6-5
7 R10 @ 50 mm c/c
3 Y20
R6 as
stirrups
150 150
180
60
60
60
Embedded
Polystyrene
as void
12
28
48
L = 8.0 m for beam 11
180
L = 5.5 m for beams 1, 2, 3 and 4

100 25
All dimensions are in mm
30
120
120
30
Section A-A
Figure 6.1 Reinforcement details of beams 1, 2, 3, 4, 8 and 11
150 150
5 R10 @ 50 mm c/c
2 R6
6-6
R6 as
stirrups
3 Y20
150 150
5500
180
60
112
60
21
60
14
30
46
180
Embedded
Polystyrene
as void
30
120
120
30
Section A-A
Figure 6.2 Reinforcement details of beam 5
2 R6
150 150
6-7
7 R10 @ 50 mm c/c
R6 as
stirrups
6 Y20
150 150
60
60
60

180
100 25
12
28
48
180
Embedded
Polystyrene
as void
30 48 48 48 48 48
30
Section A-A
Figure 6.3 Reinforcement details of beams 6 and 9
150 150
5 R10 @ 50 mm c/c
2 R6
6-8
R6 as
stirrups
6 Y20
150 150
5500
180
60
112
60
21
60
14
30
46
180
Embedded
Polystyrene
as void
30
48 48 48 48 48 30
Section A-A
2 R6
150 150
6-9
7 R10 @ 50 mm c/c
R6 as
stirrups
6 Y24
150 150
6700
180
60
60
100 25
60
12
28
48
180
Embedded
Polystyrene
as void
30 48 48 48 48 48
30
Section A-A
2 Y12
62.5
6-10
R10 as
stirrups
3 Y20
62.5
2500
36
78
36
20
36
20
40
210
40
35
35
40
Section A-A
2 Y12
80
6-11
R6 as
stirrups
3 Y20
80
2500
37
76
32
20
37
20
36
210
41
34
34
41
Section A-A
6-12
3 Y20
A
2500
150
230
20
30
No stirrups were used
220
35
40
40
35
Section A-A
6-13
The main design properties of these beams are presented in Table 6.2 and the
reinforcement details and geometry shown in Figures 6.9 through 6.11.
Table 6.2 Details of reinforced concrete continuous beams
Beam
Positive
Negative
reinforcement reinforcement
number num
-ber
of
bars
Positive steel
Negative steel
Total
area
As
mm2
beam
bar
diameter
(mm)
num
-ber
of
bars
bar
diameter
(mm)
area
As
mm2
ratio
ratio
length
(m)
16
20
1
2
2
24
10
20
942
.01163 1238 .01519
12
17
20
12
24
24
20
24
1884 .02361 2160 .02714
18
2
4
5
1808 .02283 2260 .02854
12
As indicated in Table 6.2, the beams were reinforced with different types and
sizes of bars both for the positive and the negative reinforcements having different
steel ratios as shown. Each beam was 12 m long and divided into two equal spans.
Thus each span of each two-equal-span continuous beam was 6 m.
While the main reinforcing bars were either R10 or Y20 or Y24 bars, the
stirrups used were either R6 or R10 bars (see Figures 6.9 through 6.11).
The support systems, the materials used, the loading sequences and other test
details are discussed in the relevant sections later on.
6-14
5 R10 @ 50 mm c/c
2 R10 + 2 Y20 + 1 Y24
2 R10
2 Y20
150 150
3 Y20
R6 as
stirrups
Central
support
150 150
3 Y20
2 R10
12000
30
60
60
31.6
14
Embedded
Polystyrene
as void
180
Embedded
Polystyrene
as void
180
60
180
60
30
240
30
14
30
14
30
46
46
60 60
110 14
60
28.4
30
30
Section A-A
120
120
30
Section B-B
6-15
5 R10 @ 50 mm c/c
4 Y20 + 2 Y24
2 Y20
4 Y20
150 150
6 Y20
R10 as
stirrups
Central
support
150 150
6 Y20
2 Y20
12000
34
60
34
232
34
60
14
34
14
34
46
46
180
Embedded
Polystyrene
as void
180
25.3
14
Embedded
Polystyrene
as void
180
60
125 14
46.4 46.4 46.4 46.4 46.4
34.7
34
34
46.4 46.4 46.4 46.4 46.4
Section A-A
34
Section B-B

6-16
5 R10 @ 50 mm c/c
5 Y24
2 Y24
2 Y24
150 150
4 Y24
R10 as
stirrups
Central
support
150 150
4 Y24
2 Y24
12000
36
60
32
28
60
Embedded
Polystyrene
as void
180
Embedded
Polystyrene
as void
180
60
180
57
36
228
36
14
36
14
36
46
46
57 57
130 10
57
10
36
36
Section A-A
76
76
76
36
Section B-B
6.4
6-17
Partially Prestressed Concrete Specimens
As mentioned earlier, a total of 12 partially prestressed concrete full-size simply

supported box beams were constructed and tested. The beams were of various
lengths and each had an overall cross section of 300 mm x 300 mm. The void of
180 mm x 180 mm in each beam was created by permanently embedding
polystyrene prisms.
The main design properties of these beams are tabulated in Table 6.3. The
reinforcing details and geometry are presented in Figures 6.12 through 6.15.
Table 6.3 Details of partially prestressed concrete box beams
Prestressing
steel
Beam
num- numbber
er of
bars
19
bar
diameter
(mm)
5
20
21
22
23
5
7
10
2
5
5
5
5
24
25
26
27
5
7
7
2
5
5
5
5
28
29
30
5
7
10
5
5
5
Reinforcing
steel
number of
bars
1
2
4
2
1
1
2
4
2
2
1
2
4
2
1
bar
diameter
(mm)
12
16
12
12
12
12
16
12
12
12
12
16
12
12
12
Steel
area,
As
Reinfor
-cement
ratio
Degree
of
prestressing
Beam
length
L
mm2
554
0.00737
0.25
5.5
550
363
309
554
0.00730
0.00511
0.00460
0.00737
0.50
0.75
1.00
0.25
5.5
5.5
5.5
6.8
550
363
363
554
0.00730
0.00511
0.00511
0.00737
0.50
0.75
0.75
0.25
6.8
6.8
6.8
8.0
550
363
309
0.00730
0.00511
0.00460
0.50
0.75
1.00
8.0
8.0
8.0
(m)
2 R6
7 R10 @ 50 mm c/c
150 150
2 Y16 + 1 Y12 +
25
6-18
R6 as
stirrups
150 150
180
60
60
60
Embedded
Polystyrene
as void
38
49.4
Prestressing steel
22
Reinforcing steel
180

100 25
26 61.5 62.5 62.5 61.5 26
Section A-A
Figure 6.12 Reinforcement details of beams 19, 23 and 27
2 R6
150 150
7 R10 @ 50 mm c/c
R6 as
stirrups
25
4 Y12 + 3 5
6-19
150 150
60
60
60
Embedded
Polystyrene
as void
60
33
27
48.75
Reinforcing steel
180
Prestressing steel
180
100 25
26 30 47 47 47 47 30 26
Section A-A
Figure 6.13 Reinforcement details of beams 20, 24 and 28
2 R6
150 150
7 R10 @ 50 mm c/c
R6 as
stirrups
25
2 Y12 + 5 5
6-20
150 150
60
60
Embedded
Polystyrene
as void
60
27
63.25
63
Prestressing steel
60
180

L = 6.8 m for beams 25 and 26
Reinforcing steel
180
100 25
30 40 40 40 40 40 40 30
Section A-A
Figure 6.14 Reinforcement details of beams 21, 25, 26 and 29
2 R6
150 150
25
7 R10 @ 50 mm c/c
25
1 Y12 + 6 5
6-21
R6 as
stirrups
150 150
180
60
60
Prestressing steel
180
33
Embedded
Polystyrene
as void
40
Reinforcing steel
76
30
60
60
100 25
30 40 40 40 40 40 40 30
Section A-A
6-22
Table 6.3 indicates that the beams were of three different overall lengths
5.5 m, 6.8 m or 8.0 m. The main reinforcements comprised of both prestressing
tendons as well as non-prestressed deformed bars. High strength wires of nominal
diameter 5 mm were used as prestressing tendons and four different degrees of
prestressing were used (i.e. 25%, 50%, 75% and 100%).
6.5
Materials
The materials used for the fabrication of all 30 beams tested for the present research
are described in this section.
Concrete: All beams were constructed using concrete ordered from local suppliers.
The maximum size of aggregate used was 10 mm. Ordinary Portland (Type GP)
cement was used for the concrete.
Three standard 100 mm diameter control
cylinders were cast with each specimen, stored in the same manner, and tested at the
same age as the specimen to determine the compressive strength, fc, of the concrete.
Normal curing procedures were followed for both the cylinders and the specimens.
The values of the concrete compressive strength for each specimen are presented in
Table 6.4.
Reinforcement: Mild hot rolled deformed bars of grade 400Y were mostly used as
longitudinal main reinforcement and 250R plain bars, for the vertical ties or stirrups.
The minimum tensile yield strength, fsy, of the main reinforcement was 400 MPa. A
typical stress-strain curve for reinforcing steel is shown in Figure 6.16.
High strength wires of nominal diameter 5 mm were used as prestressing
tendons in the fabrication of the partially prestressed concrete beams.
In all
calculations, yield strength at 0.2% proof stress (fsy = 1550 MPa) was adopted, as the
prestressing wire had no indication of a definite yield point.
The modulus of
elasticity for the steel, Es, was taken as equal to 200000 MPa. A typical stress-strain
diagram for the prestressing wire is shown in Figure 6.17.
6-23
Table 6.4 Compressive strength of concrete for the test beams

Beam number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
6.6
Compressive strength (fc)

(MPa)
25.9
27.1
26.1
30.0
37.7
25.4
32.4
31.0
33.8
28.4
27.6
37.1
36.6
34.4
34.1
34.1
34.2
30.6
25.9
45.8
46.4
31.0
30.7
32.4
33.2
31.3
28.4
39.1
34.4
27.8
Support Systems
To allow the beams to act in a manner in which they were modelled (Single-Degreeof-Freedom), and to allow unrestricted movement of the beams, the specimens were
supported on two concrete abutments with a roller at one end and a knife-edge at the
6-24
800
Stress in MPa
600
400
fsy = 400 MPa
200
0
0.00
.02
.04
.06
.08
.10
Strain in mm/mm
Figure 6.16 Stress-strain curve for reinforcing steel
2000
Stress in MPa
1500
fsy = 1550 MPa

1000
500
0
0.00
.01
.02
.03
Strain in mm/mm
Figure 6.17 Stress-strain curve for prestressing steel
6-25
other as shown in Figures 6.18 and 6.19. While Figure 6.18 represents the support
system used for the first-stage beams, the support system used for the second-stage
beams are represented by Figure 6.19.
Beam
Roller
Beam
Knife-edge
Figure 6.18 Beam support system for the first-stage beams
Figure 6.19 Beam support system for the second-stage beams
6-26
While 8 mm thick steel bearing plates with dimensions of 300 mm x 180 mm

were used at the reaction points for the first-stage beams, the steel plates used for the
second-stage beams were 200 mm x 75 mm x 11 mm. The purpose of providing
these plates was to prevent any localised crushing or splitting of the beam under
high concentrated loads. To increase the stiffness and stability of the support system
for the first-stage beams, the two concrete abutments were bolted to the strong floor
so as to reduce vibration energy loss through the support system when the beams
were under the hammer excitation. For the second-stage beams, each supporting
column was constructed from three Besser blocks put together with the core filled up
with high strength concrete reinforced heavily throughout. The continuous beams
were supported on roller supports at either end and on a knife-edge support at the
centre.
6.7
Loading System
The loading system is illustrated in Figure 6.20. It includes the hydraulic jack, its
support, guide structure, load cell and the loading beam. The loading beams were
steel I-beams of various lengths. These were used to apply loads at two points on
the beam to have a constant moment region to facilitate uniform crack formation.
The loading frame was bolted to the strong floor (see Figure 6.20).
With a
maximum tensile strength of 200 kN per anchorage bolt, each loading frame was
secured by four bolts. The load cell and the hydraulic jack had a maximum loading
capacity of 1000 kN. For the continuous beams, the load cell and hydraulic jacks
were so synchronised that the same load was applied at each span simultaneously.
Except with beams 13 and 15, all other beams were loaded symmetrically at
two points at a distance as detailed in Figure 6.21. This applies to each span of the
continuous beams as well. For beams 13 and 15, the single point load was applied at
the centre of the beam.
6-27
Figure 6.20 The loading system
6-28
Applied load
Loading beam
100 mm
100 mm
Beam length, L
l = 1200 mm for beams 1 through 4, 6, 8 through 11, and 19 through 30

l = 2000 mm for beams 5, 7, and each span of beams 16 through 18
l = 800 mm for beam 12 and 400 mm for beam 14
Figure 6.21 The loading diagram
6.8
Summary
The details of all the 30 beams tested for this research are described in this chapter.
This includes the representation of the lengths and cross sections of beams, steel
ratios, compressive strengths, degrees of prestressing as well as reinforcement
details. Reference has also been made to the background theory as outlined in
Chapter 4. The support systems, the loading systems and the sequence of loading
are also described in some detail. Also discussed are the materials used and their
relevant properties.
Chapter 7 : Development of Instantaneous Crack Width Formula
7-1
CHAPTER 7
DEVELOPMENT OF INSTANTANEOUS CRACK
WIDTH FORMULA
7.1
General Remarks
As discussed in Chapter 5, there are several formulae available for the evaluation of
crack widths in reinforced and/or partially prestressed concrete beams. However,
the values of the predicted crack widths differ, depending on the formulae used.
Such differences arise from the following factors:
(a)
Assumptions concerning the cracking mechanisms of a reinforced or

partially prestressed concrete beam;
(b)
Differing dimensions of the specimens as well as methods of test and

analysis;
(c)
Test errors.
It is, therefore, necessary to develop a new analytical method for determining
the crack widths of the reinforced and the partially prestressed concrete beams
associated with the current work.
7-2
In this work, flexural tensile cracks were studied experimentally in a series of

reinforced and partially prestressed concrete beams (as described in Chapter 6).
This forms part of an attempt to develop a new unified formula for the evaluation of
the crack width of both reinforced and partially prestressed concrete beams. Also, it
is proposed to relate the damping values to the residual crack widths, which in turn,
to relate to the instantaneous crack widths. Hence, the development of the proposed
crack width formula constitutes an essential part of this research.
This chapter describes the development of the proposed average crack width
prediction formula for simply supported reinforced and partially prestressed concrete
beams. Apart from own laboratory test results, other published results are also used
to check the performance of the proposed formula. To underscore the improvement
in accuracy, the proposed formula is also compared with the formulae prescribed by
the major national and international codes namely, the Australian Standard
(SAA, 1994), the British Standard (BS, 1985), the European Standard (EC2, 1991)
and the American Concrete Institute code (ACI, 1995). The applicability of the
proposed formula to the individual span or region of (two-equal-span) continuous
beams is also verified.
A relationship between the average and the maximum crack widths is also
developed. Since different code formulae utilise different crack width values, i.e.
the maximum or the average value, this facilitates the comparison of the different
codes.
Finally, the general cracking behaviour of reinforced and partially
prestressed concrete beams are discussed in the light of the experimental findings.
7.2
Test Programme
All 30 test beams, as described in Chapter 6, were subjected to static loading. The
main objectives were to observe the crack spacings as well as the crack widths in the
7-3
reinforced and the partially prestressed concrete beams and to develop an explicit
formula to determine the instantaneous average crack width.
Details of the specimens, the loading and the support systems, and the
locations for the two-point load applications are presented in detail in Chapter 6.
For all beams, the instantaneous crack widths at each level of loading were
measured using a crack detection microscope. Specifically designed for measuring
crack widths in concrete and manufactured by ELE International, this high definition
microscope (model EL35-2505) operates via an adjustable light source provided by
high-power batteries (see Figure 7.1). The microscope had a magnification factor of
35 times and a measuring range of 4 mm with each division reading 0.02 mm.
Figure 7.1 Crack detection microscope

The crack spacings for each beam were measured at around 60 to 70% of the
ultimate load after they were stabilised.
7.3
7-4
Proposed Equation for Instantaneous Average Crack Width
The opening of a crack is caused by the difference between the elongation of the
reinforcing steel and the elongation of the concrete at the same level. Therefore, the
crack width is given by the elongation of the steel minus the elongation of the
concrete between two adjacent cracks, that is
wcr = s lcr - c lcr
(7.1)
where wcr is the average crack width, lcr is the average crack spacing, s is the
average strain in tensile reinforcement and c is the average tensile strain in concrete
at the same level as the reinforcement.
The elongation of concrete due to the tensile stress and the effects of
shrinkage and creep are very minor and can be neglected (Nawy, 1996). The
average crack width, thus, can be expressed as the elongation of the steel between
two cracks, or
wcr = s lcr
(7.2)
This may be given in terms of the average steel stress, fs as

wcr = (fs/Es) lcr
(7.3)
where Es is the modulus of elasticity for steel.

The problem of crack formation and development is a complex one involving
many parameters.
According to Nawy (1968), the average crack spacing in
reinforced concrete members has an obvious relation to the thickness of concrete

cover, the diameter of reinforcing bars, the steel ratio and the concrete strength.
Also many researchers have concluded that the average crack spacing, lcr, increases
approximately linearly with increasing / (see Section 5.2.3). Note that is the
average bar diameter and is the steel ratio defined as the ratio of tensile steel area,
7-5
As, to the effective cross-sectional area (b x d) of the beam. Similar conclusions

were drawn for partially prestressed concrete beams (see Section 5.3.3).
Based on the above findings, four variables were selected to carry out the
statistical analysis employed herein to derive statistically the unified crack spacing
formula applicable to both reinforced and partially prestressed concrete beams.
They are:
(a)
lcr, the average crack spacing;
(a)
the / ratio;
(c)
c, the concrete cover; and
(d)
s, the average spacing between the reinforcing bars.

The regression equation involving these four selected variables takes the
form:
lcr = C1 c + C2 s + C3 (/)
(7.4)
where C1, C2 and C3 are the regression coefficients to be determined from the
statistical analysis.
Once the expression for average crack spacing, lcr, is obtained from the
above-mentioned statistical analysis, its substitution in Eq. (7.3) leads to the
derivation of the proposed average crack width formula.
7.4
Data Analysis and Development of the Proposed Formula
The relevant data from four reinforced and four partially prestressed concrete firststage beams were used in the proposed regression analysis. The relevant data are
presented in Table 7.1. The selected beams were beams 2, 6, 10, 11, 21, 23, 24 and
30. The other details of the beams can be found in Tables 6.1 and 6.3.
7-6
Table 7.1 Parameters used in the development of average crack spacing formula
Beam
number
2
6
10
11
21
23
24
30
Average
bar
diameter,
(mm)
20
20
24
20
6.6
10.8
8.1
5.6
Steel ratio
The
ratio
/
(mm)
0.01154
0.02309
0.03348
0.01154
0.00511
0.00737
0.00730
0.00460
1733
866
717
1733
1292
1465
1110
1217
Average
spacing
between
bars, s
(mm)
120
48
48
120
40
62
38.5
40
Concrete
cover, c
(mm)
Average
crack
spacing, lcr
(mm)
12
12
12
12
27
38
27
40
131.6
223.7
148.7
120.0
162.5
126.9
118.2
142.0
The beams were so selected that they cover all the three beam lengths (5.5 m,
6.7 m and 8.0 m) and the three different steel ratios (0.01154, 0.02309 and 0.03348)
for the reinforced concrete beams. Similarly, the partially prestressed beams cover
the three beam lengths (5.5 m, 6.8 m and 8.0 m) and the four different degrees of
prestressing (0.25, 0.50, 0.75 and 1.00).
The solutions for and rounding offs of the regression coefficients in Eq. (7.4)
using the data from Table 7.1 leads to the following unified equation for calculating
the average crack spacings for reinforced and partially prestressed concrete beams:
lcr = 0.6 (c s) + 0.1 (/)
(7.5)
where lcr, c, s and are all in mm.

Substitution of Eq. (7.5) into Eq. (7.3) yields
wcr = (fs /Es) [0.6 (c s) + 0.1 (/)]
(7.6)
which is the unified formula developed for predicting instantaneous average crack
widths in both reinforced and partially prestressed concrete beams.
7.5
7-7
Relationship between Average and Maximum Crack Width
Of importance in design is the probable width of the largest crack. However,

different researchers have suggested different relationships between the maximum
and the average crack widths.
From his test results, Clark (1956) found the average maximum crack width
for each value of steel stress from 15000 psi (103 MPa) to 40000 psi (275 MPa) to
be about 1.64 times the average width. He also found that, although essentially the
same value of 1.64 was obtained at each value of stress in the steel, it ranged from
1.18 to 2.77 for individual specimens.
Chi and Kirstein (1958) determined the ratios of maximum to average widths
of cracks for 70 beams at each of six stress levels (15000 psi (103 MPa) to 40000 psi
(275 MPa) computed steel stress).
The lengths over which the cracks were
measured for these beams ranged from 36 to 66 in. (914.4 to 1676.4 mm) and the
ratio of maximum to average width of crack ranged from 1.08 to 2.64. The average
value for this ratio was found to be 1.63.
Holmberg (1970), and Holmberg and Lindgren (1970), in the development of
their crack width formula (Eqs. (5.32) and (5.33)), assumed the maximum crack
width to be 1.7 times the average crack width.
Welch and Janjua (1971) have suggested that wmax may be taken as 1.5 wcr
for deformed bars, where wmax and wcr are maximum and average crack width,
respectively.
In the CEB-FIP Model Code (CEB, 1978), the design or characteristic
maximum crack width (95 per cent fractile) is taken to be 1.7 times the average
crack width.
Suzuki and Ohno (1984) considered the value of 1.5 for the ratio wmax/wcr
as reasonable for the crack control of partially prestressed and reinforced concrete
7-8
members.
The Eurocode 2 (EC2, 1991) employs Eq. (5.27) to calculate the average
crack width, wcr; but the code defines the design or characteristic maximum crack
width, wmax, as:
wmax = wcr
(7.7)
For load induced cracking, the value of the coefficient to be used in

Eq. (7.7) is = 1.7 or 1.3, respectively for sections whose minimum dimension
exceeds 800 mm or is smaller than 300 mm.
In this work, for the first-stage beams, the maximum crack width data were
not available for all beams. The maximum and the average crack width data at
different load levels available for the 8 first-stage beams as well as those for all the
second-stage beams are presented in Tables A.1 through A.17 given in Appendix A.
The ratios of maximum to average crack widths (wmax/wcr) calculated at each load
level for these beams are also presented in these tables.
From Tables A.1 through A.7, it can be found that the wmax/wcr ratios ranged
from 1.00 to 2.63 for the simply supported reinforced concrete beams with the
average value for the ratio being 1.33. Similarly from Tables A.8 through A.10,
these ratios ranged from 1.00 to 2.80 for the individual spans of the two-equal-span
continuous reinforced concrete beams, the average value being 1.48. On the other
hand, it can be found from Tables A.11 through A.17 that, for the partially
prestressed concrete beams, the wmax/wcr ratios ranged from 1.00 to 3.05 with the
average value being 1.63. The overall average value for the wmax/wcr ratio from all
the 17 beams and for the different load levels, however, is 1.46.
It is evident from above discussions that the average value of the ratio of
maximum to average crack width varies from 1.3 to 1.7 with different researchers
suggesting different values. The overall average would be 1.5. In the present
experimental work, the average value is found to be 1.46. Hence, for the purpose of
7-9
comparison between different crack width formulae, the ratio of maximum to

average crack width is considered to be equal to 1.5 in this research.
7.6
Comparison with Test Data
The average crack width formula, developed in Section 7.4, is compared with own
laboratory test data as well as various test data available in the literature. These test
data are taken from Clark (1956), Chi and Kirstein (1958), and Nawy (1984). The
comparison of the proposed crack width formula, Eq. (7.6), with the different test
data is summarised under the following sub-sections.
7.6.1
The present test beams
Eq. (7.6) has been applied to different steel stress levels to determine the theoretical
values of the average crack widths for each beam. These theoretical values together
with the measured values corresponding to the various steel stress levels are listed
for each beam in Tables B.1 through B.26 presented in Appendix B. The ratio of the
measured and the theoretical values were computed and are listed in the same tables.
It is seen that the mean values of the ratios are very close to 1.0 and in some cases
exactly equal to 1.0. The standard deviations vary from 4 to 22 percent. The larger
means and standard deviations at the lower stress levels are expected since crack
widths at such levels involve a larger error of observation than at higher stress
levels. This is mainly because cracks at early stages of development are not clearly
visible to the naked eye for identification and measurements.
An inspection of Tables B.1 through B.8 reveals that for the reinforced
concrete simply supported box beams, the means of the ratios of the measured to the
calculated (using the proposed formula) average crack width values were within
+ 30% as they varied from 0.96 to 1.26 and the standard deviations varied from 8 to
19 percent.
As mentioned earlier, 4 solid rectangular beams (beams 12 through 15) were
7-10
specifically designed and tested to verify the applicability of the proposed average
crack width formula developed for box beams to solid beams. From Tables B.9
through B.12, it is evident that the proposed formula is successfully applicable to
solid beams. The means of the ratios of measured to calculated average crack
widths for these 4 beams were within + 30% as they varied from 0.85 to 1.21 with
the standard deviations varying from 9 to 22 percent.
Tables B.13 through B.15 clearly demonstrate the applicability of the
proposed average crack width formula to the individual span/region of continuous
beams. For the 3 two-equal-span continuous beams tested, the means of the ratios of
measured to calculated average crack widths for the positive or the negative spans
varied from 0.99 to 1.23 and the standard deviations varied from 4 to 20 percent.
Finally, for the partially prestressed concrete beams (see Tables B.16 through
B.26), the means of the ratios of measured to calculated average crack widths ranged
from 0.80 to 1.17 and the standard deviations ranged from 4 to 21 percent.
Measured average crack widths in mm
0.4
0.3
- 30% line
+ 30% line
Partially prestressed
concrete beams
0.2
Reinforced concrete
continuous beams
0.1
Reinforced concrete
solid beams
Reinforced concrete
box beams
0.0
0.0
0.1
0.2
0.3
0.4
Calculated average crack widths in mm

Figure 7.2 Measured versus calculated average crack widths for the test beams
7-11
The above findings are further demonstrated by a plot of the measured

average crack widths against the calculated values using the proposed formula
(Eq. (7.6)) for the 26 test beams given in Figure 7.2. Note that the test results for 4
of the first-stage beams were not available to the author. A + 30% band of scatter
envelopes majority of the 258 data points for different steel stress levels for the
beams. Such a wide spread is expected and is considered reasonable for flexural
cracking problems (Huang, 1975; Nawy, 1984).
As all the data points for the continuous beams and most of the data points
for the solid beams lie within + 30% limits, it is proved that the proposed average
crack width formula is applicable to both solid and continuous beams. Also, as all
the data points for the partially prestressed concrete beams lie well within + 30%
limits, the prediction of average crack widths using the proposed formula for both
reinforced and partially prestressed concrete beams are considered accurate.
7.6.2
Clarks beams
To check the validity of the proposed crack width formula, Eq. (7.6) has also been
compared with the test data available from Clark (1956). Clark (1956) tested 28
slabs of different widths and 26 beams having cross sections of 6 in. (152.4 mm) x
15 in. (381.0 mm) and 6 in. (152.4 mm) x 23 in. (584.2 mm). The beams were
tested with spans of 9 ft. (2743.2 mm) and 11 ft. (3352.8 mm), respectively. The
beams were reinforced with deformed bars and tested as simply supported beams
with two point loads applied at quarter points.
Details of the 26 beams tested by Clark (1956) and used in this comparison
are presented in Table 7.2. The theoretical values of the average crack widths are
calculated at the net stress levels of 15000 psi (103 MPa) to 45000 psi (310 MPa) of
steel stress at increments of 5000 psi (34.5 MPa). These theoretical values are
compared with the measured values corresponding to the various stress levels for all
the 26 beams.
The ratios of the measured and the calculated values are also
computed for each beam at each steel stress level as shown in Table 7.3. It is found
7-12
Table 7.2 Details of Clarks beams

Specimen
no.
Effective Reinfordepth, d cement

(mm)
ratio,
Concrete
cover, c
(mm)
Average spacing
between bars, s
(mm)
The ratio
/
(mm)
15-6-8-1
330.20
0.0101
38.10
152.4
2515
15-6-8-2
330.20
0.0101
38.10
152.4
2515
15-6-6-1
339.85
0.0110
31.62
82.6
1730
15-6-6-2
339.85
0.0110
31.62
82.6
1730
15-6-6-3
333.50
0.0112
37.97
82.6
1699
15-6-7-1
357.12
0.0142
12.76
79.4
1565
15-6-7-2
331.72
0.0153
38.16
79.4
1450
15-6-7-3
331.72
0.0153
38.16
79.4
1450
15-6-7-4
331.72
0.0153
38.16
79.4
1450
15-6-7-5
331.72
0.0153
38.16
79.4
1450
15-6-10-1
326.64
0.0165
38.48
152.4
1953
15-6-10-2
326.64
0.0165
38.48
152.4
1953
15-6-8-3
330.20
0.0203
38.10
76.2
1250
15-6-8-4
330.20
0.0203
38.10
76.2
1250
15-6-9-1
344.42
0.0246
22.29
73.0
1163
15-6-9-2
328.68
0.0258
38.04
73.0
1110
15-6-9-3
328.68
0.0258
38.04
73.0
1110
15-6-9-4
328.68
0.0258
38.04
73.0
1110
15-6-9-5
328.68
0.0258
38.04
73.0
1110
23-6-10-1
529.84
0.0101
38.48
152.4
3193
23-6-10-2
529.84
0.0101
38.48
152.4
3193
23-6-11-1
515.62
0.0128
51.12
152.4
2797
23-6-9-1
531.88
0.0159
38.04
73.0
1801
23-6-9-2
531.88
0.0159
38.04
73.0
1801
23-6-11-2
528.32
0.0250
38.42
66.7
1433
23-6-11-3
528.32
0.0250
38.42
66.7
1433
7-13
Table 7.3 Relation between measured and computed average crack width for
Clarks beams
Steel stress, fs = 15000 psi (103 MPa)
Measured
average
crack
width,
wcr,m
(mm)
Calculated
The
average ratio
crack (wcr,m
width, /wcr,c)
wcr,c
(mm)
15-6-8-1
15-6-8-2
15-6-6-1
15-6-6-2
0.0340
0.0472
0.0429
---------
0.0942
0.0942
0.0733
---------
0.36
0.50
0.58
-------
15-6-6-3
15-6-7-1
15-6-7-2
15-6-7-3
15-6-7-4
15-6-7-5
15-6-10-1
15-6-10-2
15-6-8-3
15-6-8-4
15-6-9-1
15-6-9-2
15-6-9-3
15-6-9-4
15-6-9-5
0.0343
0.0328
0.0467
0.0404
0.0488
0.0460
0.0744
0.0859
0.0820
0.0648
0.0399
0.0625
0.0551
0.0480
0.0475
0.0737
0.0600
0.0619
0.0619
0.0619
0.0619
0.0654
0.0654
0.0526
0.0526
0.0442
0.0464
0.0464
0.0464
0.0464
23-6-10-1
23-6-10-2
23-6-11-1
23-6-9-1
23-6-9-2
0.0401
0.0574
0.0785
0.0549
0.0554
23-6-11-2
23-6-11-3
0.0523
0.0495
Specimen
no.
Mean ratio

Measured
average
crack
width,
wcr,m
(mm)
Calculated
The
average ratio
crack (wcr,m/
width,
wcr,c)
wcr,c
(mm)
15-6-8-1
15-6-8-2
15-6-6-1
15-6-6-2
0.0688
0.0838
0.0714
0.0521
0.1262
0.1262
0.0983
0.0983
0.55
0.66
0.73
0.53
0.47
0.55
0.75
0.65
0.79
0.74
1.14
1.31
1.56
1.23
0.90
1.35
1.19
1.04
1.02
15-6-6-3
15-6-7-1
15-6-7-2
15-6-7-3
15-6-7-4
15-6-7-5
15-6-10-1
15-6-10-2
15-6-8-3
15-6-8-4
15-6-9-1
15-6-9-2
15-6-9-3
15-6-9-4
15-6-9-5
0.0800
0.0536
0.0742
0.0663
0.0798
0.0795
0.1102
0.1209
0.1034
0.0841
0.0526
0.0765
0.0605
0.0686
0.0643
0.0988
0.0804
0.0830
0.0830
0.0830
0.0830
0.0876
0.0876
0.0705
0.0705
0.0593
0.0621
0.0621
0.0621
0.0621
0.81
0.67
0.89
0.80
0.96
0.96
1.26
1.38
1.47
1.19
0.89
1.23
0.97
1.10
1.04
0.1292
0.1292
0.1127
0.0819
0.0819
0.31
0.44
0.70
0.67
0.68
23-6-10-1
23-6-10-2
23-6-11-1
23-6-9-1
23-6-9-2
0.0846
0.1039
0.1125
0.0757
0.0663
0.1732
0.1732
0.1511
0.1098
0.1098
0.49
0.60
0.74
0.69
0.60
0.0651
0.0651
0.80
0.76
23-6-11-2
23-6-11-3
0.0772
0.0706
0.0872
0.0872
0.89
0.81
0.82
Specimen
no.
0.88
7-14
Table 7.3 (continued) Relation between measured and computed average crack
width for Clarks beams
Measured
average
crack
width,
wcr,m
(mm)
Calculated
The
average ratio
crack (wcr,m
width, /wcr,c)
wcr,c
(mm)
15-6-8-1
15-6-8-2
15-6-6-1
15-6-6-2
0.1080
0.1186
0.0866
0.0643
0.1573
0.1573
0.1225
0.1225
0.69
0.75
0.71
0.53
15-6-6-3
15-6-7-1
15-6-7-2
15-6-7-3
15-6-7-4
15-6-7-5
15-6-10-1
15-6-10-2
15-6-8-3
15-6-8-4
15-6-9-1
15-6-9-2
15-6-9-3
15-6-9-4
15-6-9-5
0.1034
0.0691
0.0960
0.0884
0.0917
0.0996
0.1448
0.1613
0.1364
0.1044
0.0671
0.1036
0.0782
0.0874
0.0843
0.1231
0.1002
0.1034
0.1034
0.1034
0.1034
0.1092
0.1092
0.0878
0.0878
0.0739
0.0774
0.0774
0.0774
0.0774
23-6-10-1
23-6-10-2
23-6-11-1
23-6-9-1
23-6-9-2
0.1222
0.1560
0.1806
0.0993
0.0897
23-6-11-2
23-6-11-3
0.0859
0.0932
Specimen
no.
Mean ratio

Measured
average
crack
width,
wcr,m
(mm)
Calculated
The
average ratio
crack (wcr,m/
width,
wcr,c)
wcr,c
(mm)
15-6-8-1
15-6-8-2
15-6-6-1
15-6-6-2
0.1267
0.1557
0.1019
0.0810
0.1893
0.1893
0.1474
0.1474
0.67
0.82
0.69
0.55
0.84
0.69
0.93
0.85
0.89
0.96
1.33
1.48
1.55
1.19
0.91
1.34
1.01
1.13
1.09
15-6-6-3
15-6-7-1
15-6-7-2
15-6-7-3
15-6-7-4
15-6-7-5
15-6-10-1
15-6-10-2
15-6-8-3
15-6-8-4
15-6-9-1
15-6-9-2
15-6-9-3
15-6-9-4
15-6-9-5
0.1257
0.0871
0.1133
0.1074
0.1125
0.1250
0.1768
0.1943
0.1669
0.1257
0.0823
0.1217
0.0874
0.1082
0.1062
0.1481
0.1206
0.1245
0.1245
0.1245
0.1245
0.1314
0.1314
0.1057
0.1057
0.0889
0.0932
0.0932
0.0932
0.0932
0.85
0.72
0.91
0.86
0.90
1.00
1.35
1.48
1.58
1.19
0.93
1.31
0.94
1.16
1.14
0.2158
0.2158
0.1883
0.1368
0.1368
0.57
0.72
0.96
0.73
0.66
23-6-10-1
23-6-10-2
23-6-11-1
23-6-9-1
23-6-9-2
0.1529
0.1867
0.1910
0.1237
0.1123
0.2597
0.2597
0.2266
0.1647
0.1647
0.59
0.72
0.84
0.75
0.68
0.1086
0.1086
0.79
0.86
23-6-11-2
23-6-11-3
0.1021
--------
0.1308
--------
0.78
--------
0.93
Specimen
no.
0.94
7-15
Measured
average
crack
width,
wcr,m
(mm)
Calculated
The
average ratio
crack (wcr,m
width, /wcr,c)
wcr,c
(mm)
15-6-8-1
15-6-8-2
15-6-6-1
15-6-6-2
0.1458
0.1920
0.1189
0.0958
0.2204
0.2204
0.1716
0.1716
0.66
0.87
0.69
0.56
15-6-6-3
15-6-7-1
15-6-7-2
15-6-7-3
15-6-7-4
15-6-7-5
15-6-10-1
15-6-10-2
15-6-8-3
15-6-8-4
15-6-9-1
15-6-9-2
15-6-9-3
15-6-9-4
15-6-9-5
0.1537
0.1044
0.1313
0.1255
0.1346
0.1516
0.2070
0.2047
0.1984
0.1471
0.0955
0.1496
--------0.1313
0.1262
0.1725
0.1404
0.1449
0.1449
0.1449
0.1449
0.1530
0.1530
0.1231
0.1231
0.1035
0.1085
--------0.1085
0.1085
23-6-10-1
23-6-10-2
23-6-11-1
23-6-9-1
23-6-9-2
0.1842
0.2159
0.2243
0.1478
0.1331
23-6-11-2
23-6-11-3
-----------------
Specimen
no.
Mean ratio

Measured
average
crack
width,
wcr,m
(mm)
Calculated
The
average ratio
crack (wcr,m/
width,
wcr,c)
wcr,c
(mm)
0.1750
0.2350
0.1397
0.1105
0.2515
0.2515
0.1958
0.1958
0.70
0.93
0.71
0.56
0.89 15-6-6-3
0.74 15-6-7-1
0.91 15-6-7-2
0.87 15-6-7-3
0.93 15-6-7-4
1.05 15-6-7-5
1.35 15-6-10-1
1.34 15-6-10-2
1.61 15-6-8-3
1.20 15-6-8-4
0.92 15-6-9-1
1.38 15-6-9-2
------- 15-6-9-3
1.21 15-6-9-4
1.16 15-6-9-5
0.1770
0.1234
0.1410
0.1466
0.1567
0.1768
0.2385
0.2334
0.2626
0.1684
0.1115
0.1928
--------0.1539
0.1491
0.1968
0.1602
0.1654
0.1654
0.1654
0.1654
0.1746
0.1746
0.1404
0.1404
0.1181
0.1238
--------0.1238
0.1238
0.90
0.77
0.85
0.89
0.95
1.07
1.37
1.34
1.87
1.20
0.94
1.56
-------1.24
1.20
0.3024
0.3024
0.2638
0.1917
0.1917
0.61
0.71
0.85
0.77
0.69
23-6-10-1
23-6-10-2
23-6-11-1
23-6-9-1
23-6-9-2
0.2141
0.2517
0.2576
0.1748
0.1570
0.3451
0.3451
0.3010
0.2188
0.2188
0.62
0.73
0.86
0.80
0.72
-----------------
------- 23-6-11-2
------- 23-6-11-3
-----------------
-----------------
---------------
0.96
Specimen
no.
15-6-8-1
15-6-8-2
15-6-6-1
15-6-6-2
0.99
7-16
Specimen
no.
Calculated average
crack width, wcr,c
(mm)
0.2835
0.2835
0.2207
0.2207
0.2218
0.1806
0.1864
0.1864
The ratio
(wcr,m/wcr,c)
15-6-8-1
15-6-8-2
15-6-6-1
15-6-6-2
15-6-6-3
15-6-7-1
15-6-7-2
15-6-7-3
Measured average
crack width, wcr,m
(mm)
0.2385
0.3081
0.1593
0.1278
0.2035
0.1410
0.1554
0.1748
15-6-7-4
15-6-7-5
15-6-10-1
15-6-10-2
15-6-8-3
15-6-8-4
15-6-9-1
15-6-9-2
15-6-9-3
15-6-9-4
15-6-9-5
23-6-10-1
23-6-10-2
23-6-11-1
------------------0.2819
0.2687
------------------0.1280
------------------------------------0.2446
0.2885
----------
------------------0.1968
0.1968
------------------0.1331
------------------------------------0.3890
0.3890
----------
------------------1.43
1.37
------------------0.96
------------------------------------0.63
0.74
----------
23-6-9-1
23-6-9-2
23-6-11-2
23-6-11-3
-------------------------------------
-------------------------------------
-------------------------------------
Mean ratio
0.84
1.09
0.72
0.58
0.92
0.78
0.83
0.94
0.91
that the means of the ratios are very close to unity. In fact they varied from 0.82 at a
steel stress of 15000 psi (103 MPa) to 0.99 at a steel stress of 40000 psi (275 MPa),
the overall mean being 0.92. The larger variations at the lower stress levels (i.e. at
15000 psi (103 MPa)) are as expected since the crack width measured at such a level
7-17
involved a larger error of observation than those taken at higher stress levels.
Note that the beams for which the ratios of measured to calculated average
crack widths were the most offset from unity were those reinforced with single bars.
In such cases the average spacing between bars was taken to be equal to the total
width of the beam for computing the average crack widths using the proposed
formula.
A plot of the measured average crack widths against the calculated values for
these 26 beams is given in Figure 7.3. The data points representing different steel
stress levels are shown separately in the diagram.
0.4
0.3
- 30% line
+ 30% line
fs= 45000 psi (310 MPa)
0.2
fs = 40000 psi (275 MPa)

fs = 35000 psi (241 MPa)
fs = 30000 psi (207 MPa)
0.1
fs = 25000 psi (172 MPa)
fs = 20000 psi (138 MPa)
fs = 15000 psi (103 MPa)
0.0
0.0
0.1
0.2
0.3
0.4

Figure 7.3 Measured versus calculated average crack widths for Clarks beams
It may be seen from Figure 7.3 that good correlations exist between the
calculated and the measured values for Clarks beams as majority of the 161 data
points are within + 30% limits and the scattering is acceptably small. Once again, as
expected the most data points falling outside + 30% limits are those for lower stress
levels.
7.6.3
7-18
Chi and Kirsteins beams
The proposed average crack width formula was further validated by applying
Eq. (7.6) to the test data of Chi and Kirstein (1958). Chi and Kirstein (1958) tested
16 reinforced concrete beams having cross sections of 7.5 in. (190.5 mm) x 6 in.
(152.4 mm), 11 in. (279.4 mm) x 6 in. (152.4 mm), 6 in. (152.4 mm) x 15 in. (381.0
mm), and 6 in. (152.4 mm) x 23 in. (584.2 mm). The beams having depths of 6 in.
(152.4 mm), 15 in. (381.0 mm) and 23 in. (584.2 mm) had span lengths of 72 in.
(1828.8 mm), 108 in. (2743.2 mm) and 132 in. (3352.8 mm), respectively.
Details of these 16 beams tested by Chi and Kirstein (1958) are presented in
Table 7.4. Theoretical values of the average crack widths are calculated at the net
stress levels of 15000 psi (103 MPa) to 40000 psi (275 MPa) of steel stress at
increments of 5000 psi (34.5 MPa). These theoretical values are compared with the
measured values corresponding to the various stress levels for 14 of the beams. Two
of the beams, specimen nos. 3 and 4 were actually slabs and hence, were excluded
Table 7.4 Details of Chi and Kirsteins beams
Specimen
no.
Effective
depth, d
(mm)
Reinforcement
ratio,
Concrete
cover, c
(mm)
Average spacing
between bars, s
(mm)
The ratio
/
(mm)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
133.35
133.35
130.05
130.05
331.72
331.72
331.72
331.72
330.20
330.20
330.20
330.20
530.10
530.10
531.88
531.88
0.0101
0.0101
0.0156
0.0156
0.0153
0.0153
0.0153
0.0153
0.0101
0.0101
0.0101
0.0101
0.0097
0.0097
0.0159
0.0159
12.70
12.70
12.83
12.83
38.16
38.16
38.16
38.16
38.10
38.10
38.10
38.10
38.23
38.23
38.04
38.04
127.0
127.0
209.6
209.6
79.4
79.4
79.4
79.4
152.4
152.4
152.4
152.4
152.4
152.4
73.0
73.0
1257
1257
1221
1221
1450
1450
1450
1450
2515
2515
2515
2515
3273
3273
1801
1801
7-19
Table 7.5 Relation between measured and computed average crack width for Chi
and Kirsteins beams
Specimen
no.
Measured
average
crack
width,
wcr,m
(mm)
Calculated
The
average ratio
crack (wcr,m
width, /wcr,c)
wcr,c
(mm)
1
0.0211 0.0257
0.82
2
------- --------- ------5
0.0432 0.0539
0.80
6
0.0432 0.0539
0.80
7
0.0460 0.0539
0.85
8
0.0467 0.0539
0.87
9
0.0358 0.0816
0.44
10
0.0333 0.0816
0.41
11
-------------- ------12
0.0455 0.0816
0.56
13
-------------- ------14
0.0521 0.1209
0.43
15
0.0485 0.0746
0.65
16
0.0429 0.0746
0.57
Mean ratio
0.65
1
0.0500 0.0430
1.16
2
0.0521 0.0430
1.21
5
0.0810 0.0900
0.90
6
0.0886 0.0900
0.98
7
0.0932 0.0900
1.04
8
0.0960 0.0900
1.07
9
0.0833 0.1363
0.61
10
0.0869 0.1363
0.64
11
0.0762 0.1363
0.56
12
0.1016 0.1363
0.75
13
0.1229 0.2020
0.61
14
0.1189 0.2020
0.59
15
0.1110 0.1246
0.89
16
0.0726 0.1246
0.58
Mean ratio
0.83
Specimen
no.
Measured
average
crack
width,
wcr,m
(mm)
Calculated
The
average ratio
crack (wcr,m/
width,
wcr,c)
wcr,c
(mm)
1
2
5
6
7
8
9
10
11
12
13
14
15
16
0.0376 0.0345
1.09
0.0338 0.0345
0.98
0.0719 0.0722
1.00
0.0635 0.0722
0.88
0.0683 0.0722
0.95
0.0749 0.0722
1.04
0.0533 0.1094
0.49
0.0605 0.1094
0.55
------------------0.0721 0.1094
0.66
0.0757 0.1620
0.47
0.0925 0.1620
0.57
0.0838 0.1000
0.84
0.0564 0.1000
0.56
0.77
1
0.0719 0.0517
1.39
2
0.0643 0.0517
1.24
5
0.1016 0.1084
0.94
6
0.0950 0.1084
0.88
7
0.1067 0.1084
0.98
8
0.1214 0.1084
1.12
9
0.1118 0.1641
0.68
10
0.1153 0.1641
0.70
11
0.0853 0.1641
0.52
12
0.1295 0.1641
0.79
13
0.1669 0.2430
0.69
14
0.1440 0.2430
0.59
15
0.1313 0.1499
0.88
16
0.0914 0.1499
0.61
0.86
7-20
width for Chi and Kirsteins beams
Specimen
no.
1
2
5
6
7
8
9
10
11
12
13
14
15
16
Mean ratio
Measured
average
crack
width,
wcr,m
(mm)
Calculated
The
average ratio
crack (wcr,m
width, /wcr,c)
wcr,c
(mm)
0.0930
0.0805
0.1245
0.1118
-------0.1486
0.1481
0.1417
0.1158
0.1580
0.2032
0.1651
0.1580
0.1115
0.0602
0.0602
0.1262
0.1262
-------0.1262
0.1910
0.1910
0.1910
0.1910
0.2830
0.2830
0.1746
0.1746
0.90
1.54
1.34
0.99
0.89
------1.18
0.78
0.74
0.61
0.83
0.72
0.58
0.91
0.64
Specimen
no.
1
2
5
6
7
8
9
10
11
12
13
14
15
16
Measured
average
crack
width,
wcr,m
(mm)
Calculated
The
average ratio
crack (wcr,m/
width,
wcr,c)
wcr,c
(mm)
0.1133
0.0998
0.1379
0.1295
-------0.1712
0.1732
-------0.1389
-------0.2428
0.1961
0.1872
0.1285
0.96
0.0687
0.0687
0.1440
0.1440
--------0.1440
0.2180
-------0.2180
-------0.3229
0.3229
0.1992
0.1992
1.65
1.45
0.96
0.90
------1.19
0.79
------0.64
-----0.75
0.61
0.94
0.65
from the comparison as the proposed formula is not applicable to slabs. The ratios
of measured to calculated values are also computed for each beam at each steel
stress level as shown in Table 7.5.
Note that the measured crack width values for Chi and Kirsteins beams were
at the steel levels. In order to obtain crack width values at steel levels theoretically
using the proposed formula, which gives values at the tension face of the beams, the
values were multiplied by the factor (d/h) where d and h are the effective depth and
the total depth, respectively.
It is found from Table 7.5 that the means of the ratios are very close to unity.
They varied from 0.65 at a steel stress of 15000 psi (103 MPa) to 0.96 at a steel
stress of 40000 psi (275 MPa), the overall mean being 0.83. The larger variations at
7-21
the lower stress levels (i.e. at 15000 psi (103 MPa) and 20000 psi (138 MPa)) are as
expected for reasons explained before.
Note that six of the beams (specimen nos. 9 through 14) were reinforced with
single bars for which the ratios of measured to calculated average crack widths were
the most offset from unity. The reason for this was that in these cases the average
spacings between bars were considered to be equal to the total widths of the beams
for computing the average crack widths using the proposed formula.
The
assumption of such a value for the spacing between bars is somewhat ambiguous
and in reality there is hardly any beam reinforced with a single longitudinal bar.
A plot of the measured average crack widths against the calculated values for
these 14 beams is given in Figure 7.4. The data points representing different steel
stress levels are shown separately in this figure.
Measured average crack widths at steel level in mm
0.4
0.3
Legend
- 30% line
+ 30% line
0.2
fs = 40000 psi (275 MPa)

fs = 35000 psi (241 MPa)
0.1
fs = 30000 psi (207 MPa)

fs = 25000 psi (172 MPa)
fs = 20000 psi (138 MPa)
0.0
fs = 15000 psi (103 MPa)
0.0
0.1
0.2
0.3
0.4
Calculated average crack widths at steel level in mm

Figure 7.4 Measured versus calculated average crack widths for Chi and Kirsteins
beams
7-22
As explained, because of large number of beams reinforced with single bars

(6 out of 14), the scatter is relatively large in Figure 7.4. Yet most of the 76 data
points lie within + 30% limits. The rest of the points mostly lie underneath the
+ 30% line underscoring the fact that in this particular case the proposed formula
over estimates the average crack width and hence is on the safe side.
7.6.4
Nawys beams
Finally, the applicability of the proposed crack width formula to prestressed concrete
beams has been further verified by comparing Eq. (7.6) with the test data of
Nawy (1984). Nawys tests were conducted on 20 simply supported pre-tensioned
9-ft. (2743.2 mm) span beams and 4 two-span continuous beams of effective 9-ft.
(2743.2 mm) span. He also tested 22 simply supported post-tensioned beams of 7-ft.
6-in. (2286 mm) span.
Nawy (1984), however, provided crack width data for 20 pre-tensioned and
14 post-tensioned beams. The details of these beams are given in Tables 7.6 and
7.7, respectively.
Pre-tensioned beams were 10 in. (254 mm) deep and post-
tensioned beams were 11 in. (279.4 mm) deep.
The 20 pre-tensioned beams
consisted of the following: (a) B-1 to B-6 were T-sections with pre-tensioned
prestressed tendons only; (b) B-7 to B-18 were also T-sections reinforced with both
pre-tensioned prestressed tendons and non-prestressed mild steel; and (c) B-19 to
B-20 were I-sections with both prestressed and non-prestressed reinforcement. All
post-tensioned beams were T-section beams.
The crack width data were maximum crack widths at steel level at three
different net steel stress levels for the 34 beams. Note that the proposed crack width
formula calculates average crack widths at the tension face of the beam. Hence for
the purpose of comparison, the computed average crack width values were
multiplied by a factor of 1.5 to get the maximum crack widths (see Section 7.5).
The values thus obtained were also multiplied by the ratio (d/h) to get the value at
steel level where d is the effective depth and h is the total depth of the beam.
7-23
Table 7.6 Details of Nawys pre-tensioned beams
B-1
B-2
B-3
B-4
B-5
B-6
B-7
*Effective
depth,
d
(mm)
203.20
196.85
185.40
177.80
166.88
158.75
223.52
B-8
223.52
B-9
219.71
B-10
219.71
B-11
212.60
B-12
212.60
B-13
214.88
B-14
214.88
B-15
208.79
B-16
208.79
B-17
203.20
B-18
203.20
B-19
228.60
B-20
228.60
Beam
Reinforcement
details
no.
bar
of
diameter
bars
(mm)
3
6.35
4
6.35
5
6.35
6
6.35
7
6.35
8
6.35
3
6.35
2
9.53
3
6.35
2
9.53
4
6.35
2
9.53
4
6.35
2
9.53
5
6.35
2
9.53
5
6.35
2
9.53
6
6.35
2
12.7
6
6.35
2
12.7
7
6.35
2
12.7
7
6.35
2
12.7
8
6.35
2
12.7
8
6.35
2
12.7
3
6.35
2
9.53
4
6.35
2
12.7
Steel
area, ReinfoAs
rcement
(mm2)
ratio,
69.7
0.0017
92.9
0.0023
116.1 0.0031
139.4 0.0039
162.6 0.0048
185.8 0.0058
211.6 0.0047
Concrete
cover, c
(mm)
Spacing
betweebars, s
(mm)
The
ratio
/
31.75
34.93
49.53
57.15
64.9
69.85
8.25
76.20
76.20
76.20
76.20
76.20
76.20
76.20
3735
2761
2048
1628
1323
1095
1621
211.6
0.0047
8.25
76.20
1621
234.8
0.0053
8.75
76.20
1398
234.8
0.0053
8.75
76.20
1398
258.1
0.0060
12.83
76.20
1210
258.1
0.0060
12.83
76.20
1210
397.4
0.0091
16.89
76.20
873
397.4
0.0091
16.89
76.20
873
420.6
0.0099
22.99
76.20
784
420.6
0.0099
22.99
76.20
784
443.9
0.0108
28.58
76.20
706
443.9
0.0108
28.58
76.20
706
211.6
0.0061
12.07
42.86
1249
351.0
0.0101
11.64
41.28
839
Total depth = 10 in. (254 mm); Web thickness = 3 in. (76.2 mm); Flange
width = 8 in. (203.2 mm) for T-sections; 6 in. (152.4 mm) for I-sections.
7-24
Table 7.7 Details of Nawys post-tensioned beams
Beam
Effective
depth,d
(mm)
B3, B4
218.95
B5, B6
226.31
O-II-1
O-II-2
O-III-1
O-III-2
O-IV-1
O-IV-2
U-II-1
U-II-2
U-IV-1
U-IV-2
215.65
224.54
237.24
208.03
239.01
Reinforcement
details
no.
bar
of
diameter
bars
(mm)
4
6.35
2
9.53
4
6.35
2
12.70
2
6.40
2
12.70
2
9.53
2
12.70
2
12.70
2
15.88
2
6.40
1
12.70
1
12.70
2
15.88
Steel
area,
As
(mm2)
Reinforcement
ratio,
Concrete
cover, c
(mm)
Spacing
between
bars, s
(mm)
The
ratio
/
233.8
0.0053
38.43
76.2
1398
349.9
0.0076
33.99
76.2
1115
261.9
0.0060
38.25
76.2
1592
339.4
0.0074
26.24
76.2
1503
597.4
0.0124
13.54
76.2
1152
163.2
0.0037
49.07
76.2
2297
498.7
0.0103
11.76
76.2
1439
Table 7.8 Measured versus calculated maximum crack width at steel level of
Nawys pre-tensioned beams
Beam
B-1
B-2
B-3
B-4
B-5
B-6
B-7
B-8
B-9
B-10
B-11
B-12
B-13
B-14
B-15
B-16
B-17
B-18
B-19
B-20
Mean
Measured
maximum
crack width,
wmax,m
(mm)
0.2261
0.1803
0.1575
0.1346
0.1194
0.1067
0.1016
0.0991
0.0940
0.1041
0.0914
0.0889
0.0762
0.0813
0.0635
0.0660
0.0686
0.0533
0.1041
0.0660
Calculated
maximum
crack
width,
wmax,c
(mm)
0.4308
0.3024
0.2140
0.1645
0.1280
0.1026
0.1487
0.1487
0.1213
0.1213
0.0996
0.0996
0.0632
0.0632
0.0557
0.0557
0.0492
0.0492
0.1334
0.0840
0.92
The ratio
(wmax,m/
wmax,c)
Beam
Measured
maximum
crack width,
wmax,m
(mm)
0.52
0.60
0.74
0.82
0.93
1.04
0.68
0.67
0.77
0.86
0.92
0.89
1.21
1.29
1.14
1.19
1.39
1.08
0.78
0.79
B-1
B-2
B-3
B-4
B-5
B-6
B-7
B-8
B-9
B-10
B-11
B-12
B-13
B-14
B-15
B-16
B-17
B-18
B-19
B-20
0.3099
0.2642
0.2286
0.1905
0.1702
0.1473
0.1397
0.1372
0.1270
0.1422
0.1270
0.1321
0.1143
0.1194
0.1067
0.0991
0.1067
0.0787
0.1448
0.0991
Calculated
maximum
crack
width,
wmax,c
(mm)
0.5723
0.4017
0.2843
0.2185
0.1701
0.1362
0.1975
0.1975
0.1611
0.1611
0.1323
0.1323
0.0839
0.0839
0.0740
0.0740
0.0654
0.0654
0.1772
0.1116
The ratio
(wmax,m/
wmax,c)
0.54
0.66
0.80
0.87
1.00
1.08
0.71
0.69
0.79
0.88
0.96
1.00
1.36
1.42
1.44
1.34
1.63
1.20
0.82
0.89
1.00
7-25
Table 7.8 (continued) Measured versus calculated maximum crack width at steel
level of Nawys pre-tensioned beams
Beam
B-1
B-2
B-3
B-4
B-5
B-6
B-7
B-8
B-9
B-10
B-11
B-12
B-13
B-14
B-15
B-16
B-17
B-18
B-19
B-20
Mean ratio

Calculated
Measured
maximum crack
maximum crack
width, wmax,c (mm)
width, wmax,m (mm)
0.5029
0.8615
0.4369
0.6048
0.3785
0.4279
0.3226
0.3290
0.2743
0.2561
0.2413
0.2051
0.2413
0.2973
0.2261
0.2973
0.2057
0.2425
0.2286
0.2425
0.2032
0.1992
0.2184
0.1992
0.1956
0.1264
0.2032
0.1264
0.1956
0.1114
0.1702
0.1114
0.1880
0.0985
0.1600
0.0985
0.2286
0.2667
0.1778
0.1680
1.14
The ratio
(wmax,m/wmax,c)
0.58
0.72
0.88
0.98
1.07
1.18
0.81
0.76
0.85
0.94
1.02
1.10
1.55
1.61
1.76
1.53
1.91
1.62
0.86
1.06
Table 7.9 Measured versus calculated maximum crack width at steel level of
Nawys post-tensioned beams
Calculated
Measured
maximum The ratio
maximum
crack
(wmax,m/
Beam
crack
width,
width,
wmax,c)
wmax,c
wmax,m
(mm)
(mm)
B3
0.1397
0.1294
1.08
B4
0.1473
0.1294
1.14
B5
0.1372
0.0985
1.39
B6
0.1346
0.0985
1.37
O-II-1
0.1372
0.1635
0.84
O-II-2
0.1245
0.1635
0.76
O-III-1
0.1118
0.1311
0.85
O-III-2
0.1219
0.1311
0.93
O-IV-1
0.0889
0.0986
0.90
O-IV-2
0.094
0.0986
0.95
U-II-1
0.1829
0.2467
0.74
U-II-2
0.1651
0.2467
0.67
U-IV-1
0.0965
0.1331
0.72
U-IV-2
0.1092
0.1331
0.82
Mean
0.94

Calculated
Measured
maximum
The ratio
maximum
crack
(wmax,m/
Beam
crack
width,
wmax,c)
width,
wmax,c
wmax,m
(mm)
(mm)
B3
0.2184
0.1719
1.27
B4
0.2388
0.1719
1.39
B5
0.2159
0.1309
1.65
B6
0.2032
0.1309
1.55
O-II-1
0.2032
0.2172
0.94
O-II-2
0.1880
0.2172
0.87
O-III-1
0.1778
0.1742
1.02
O-III-2
0.1854
0.1742
1.06
O-IV-1
0.1422
0.1310
1.09
O-IV-2
0.1499
0.1310
1.14
U-II-1
0.2819
0.3277
0.86
U-II-2
0.2718
0.3277
0.83
U-IV-1
0.1549
0.1768
0.88
U-IV-2
0.1702
0.1768
0.96
1.11
7-26
Table 7.9 (continued) Measured versus calculated maximum crack width at steel
level of Nawys post-tensioned beams
Beam
B3
B4
B5
B6
O-II-1
O-II-2
O-III-1
O-III-2
O-IV-1
O-IV-2
U-II-1
U-II-2
U-IV-1
U-IV-2
Mean ratio
Measured
maximum crack
width, wmax,m
(mm)
Calculated
maximum crack
width, wmax,c
(mm)
0.3454
0.3658
0.3505
0.3251
0.3404
0.3150
0.2794
0.2997
0.2235
0.2438
0.4521
0.4267
0.2489
0.2642
0.2588
0.2588
0.1970
0.1970
0.3269
0.3269
0.2622
0.2622
0.1972
0.1972
0.4934
0.4934
0.2662
0.2662
1.18
The ratio
(wmax,m/wmax,c)
1.33
1.41
1.78
1.65
1.04
0.96
1.07
1.14
1.13
1.24
0.92
0.86
0.93
0.99
The theoretical values thus computed are compared with the measured values
corresponding to the various stress levels for all 34 beams. The ratios of measured
to calculated maximum crack widths are also computed for each beam at each steel
stress level. The comparison results for the pre-tensioned and the post-tensioned
beams are summarised separately, respectively in Tables 7.8 and 7.9.
It can be seen from Table 7.8 that for the pre-tensioned beams the ratios of
measured to calculated maximum crack widths varied from 0.92 to 1.14 for the three
different steel stress levels. The overall mean was 1.02.
Similarly from Table 7.9, for the post-tensioned beams the ratios varied from
0.94 to 1.18 and the overall mean was 1.07.
Once again, measured maximum crack widths at steel level for all 34 Nawys
7-27
beams were plotted against calculated maximum crack widths as shown in

Figure 7.5. The data points representing the pre-tensioned and the post-tensioned
Measured maximum crack width at steel level in mm
beams are shown separately in this figure.

0.9
0.8
0.7
0.6
0.5
- 30% line
0.4
+ 30% line
0.3
Post-tensioned beams
0.2
Pre-tensioned beams
0.1
0.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
Calculated maximum crack width at steel level in mm

Figure 7.5 Measured versus calculated maximum crack widths for Nawys beams
It can be seen from Figure 7.5 that the proposed crack width formula
performs relatively better for post-tensioned beams than for pre-tensioned beams as
the number of data points falling outside + 30% limits are less for post-tensioned
beams. Overall the applicability of the proposed formula to prestressed concrete
beams has been well validated as the majority of the 102 data points lie well within
+ 30% limits. Also, the successful predictions of maximum crack widths using the
proposed average crack width formula with a multiplying factor of 1.5 validates the
relationship between maximum and average crack widths developed in Section 7.5.
7.7
Comparison with Code Formulae
To show the improvements in accuracy of the proposed average crack width

7-28
formula, the predicted values are plotted along with those predicted by the formulae
adopted by different codes of practice in the same correlation plots. The formulae
considered are those adopted by the ACI Building Code (ACI, 1995), the British
Standard (BS, 1985) and the Eurocode (EC2, 1991). It should be noted here that the
Australian Standard
(SAA, 1994) does not recommend any formula for the
calculation of crack widths.

Formulae adopted by the above mentioned codes have been discussed in
detail in Chapter 5 (see Section 5.2.5). Note that while the ACI code and the British
standard formulae (Eqs.(5.25) and (5.26), respectively) give the maximum crack
widths, the Eurocode formula (Eq. (5.28)) gives the average crack widths. As
mentioned before, the values obtained using the ACI code and the British standard
formulae were divided by the factor 1.5 (= wmax/wcr) to compute average crack
widths for the beams (see Section 7.5).
0.4
0.3
0.2
Legend
- 30% line
0.1
+ 30% line
ACI formula
Proposed formula
0.0
0.0
0.1
0.2
0.3
0.4
(a) Comparison between the proposed and the ACI formulae

Figure 7.6 Comparison of measured versus calculated average crack widths for
authors reinforced concrete beams
7-29
0.4
0.3
0.2
Legend
- 30% line
+ 30% line
0.1
British standard
formula
0.0
Proposed formula
0.0
0.1
0.2
0.3
0.4
(b) Comparison between the proposed and the British standard formula
Figure 7.6 (continued) Comparison of measured versus calculated average crack
0.4
0.3
0.2
Legend
- 30% line
0.1
+ 30% line
Eurocode formula
0.0
Proposed formula
0.0
0.1
0.2
0.3
0.4
(c) Comparison between the proposed and the Eurocode formula

Figure 7.6 (continued) Comparison of measured versus calculated average crack
7-30
0.4
Legend
0.3
- 30% line
0.2
+30% line
Eurocode formula
British standard
0.1
formula
ACI formula
Proposed formula
0.0
0.0
0.1
0.2
0.3
0.4
Clarks beams
0.4
Legend
0.3
- 30% line
0.2
+ 30% line
Eurocode formula
British standard
0.1
formula
ACI formula
Proposed formula
0.0
0.0
0.1
0.2
0.3
0.4

Chi and Kirsteins beams
7-31
In order to compare the proposed formula with the three code formulae,
measured average crack widths are plotted against the calculated values predicted by
all four formulae (the proposed formula and the three code formulae) in the same
correlation plots for each type of beams. Three such plots resulted for the three
types of beams namely, the authors reinforced concrete beams, Clark (1956)s
beams and Chi and Kirstein (1958)s beams. These are presented in Figures 7.6, 7.7
and 7.8, respectively. However, due to large number of data points for the authors
beams, to get a clearer picture Figure 7.6 is divided into three separate plots as
shown.
Since none of the code formulae are recommended to be applied to
prestressed or partially prestressed concrete beams, neither the authors partially
prestressed beams nor Nawy (1984)s beams have been used in this comparison.
Note that the same multiplying factor of (d/h) has been applied to the calculated
values for Chi and Kirstein (1958)s beams while using the code formulae, as has
been applied for the proposed formula to obtain crack width values at the level of
steel (see Section 7.6.3).
An inspection of Figures 7.6 through 7.8 reveals that while the ACI formula
tends to overestimate average crack widths for the authors reinforced concrete
beams, it correlates well with Clarks and Chi and Kirsteins beams. The proposed
formula, on the other hand, correlates well with all three types of beams except with
a few data points for Clarks beams. The Eurocode formula seems to correlate
reasonably well with most of the data points but tends to underestimate average
crack widths for few of the beams from all three types.
The British standard
formula, however, does not correlate well at all with any of the test results. In fact,
it grossly underestimates average crack width values for all three types of beams.
To have a clear picture of the relative performance of the four crack width
formulae, statistical analyses (Nie et al., 1975) have been carried out. Table 7.10
gives the Pearson correlation coefficients, the intercepts, and inclinations of the
regression lines for the twelve sets of correlation points (four sets each for each of
Figures 7.6 through 7.8).
7-32
Table 7.10 Comparison of correlation statistics (Figures 7.6, 7.7 and 7.8)
Types of
beams
Formula
Authors
reinforced
concrete
beams
Clark
(1956)s
beams
Proposed formula
ACI formula
British standard formula
Eurocode formula
Proposed formula
ACI formula
Eurocode formula
Proposed formula
ACI formula
Eurocode formula
Chi and
Kirstein
(1958)s
beams
Pearson
correlation
coefficient
0.946
0.856
0.683
0.896
0.987
0.946
0.869
0.922
0.855
0.908
0.853
0.846
Intercept
(mm)
Slope
0.0014
0.0106
0.0328
0.0036
0.0280
-0.0231
-0.0165
-0.0146
0.0243
-0.0085
0.0021
-0.0107
1.075
0.813
1.865
0.955
0.876
1.074
1.944
1.264
0.881
0.856
1.695
1.133
It is evident from Table 7.10 that the proposed formula is statistically equally
reliable as the ACI formula and the Eurocode formula.
The British standard
formula, on the other hand, is statistically very much unreliable. However, as

compared to the other methods, the proposed formula is more versatile in that it can
predict average crack widths for both reinforced and prestressed concrete beams
more accurately (as shown in Section 7.6).
7.8
Cracking Behaviour of Reinforced and Partially Prestressed Concrete

Beams
Under static loading, crack widths of reinforced and partially prestressed concrete
beams accumulated with increasing levels of loading.
The average spacings
between cracks decreased with the loading at initial stage, and reached a constant
value eventually once stabilised crack patterns had been established. This is called
stabilised average crack spacing and occurred when the 60-70% of ultimate loading
for each beam was reached.
During the tests, crack patterns on the front, back and bottom of the beams
were marked on the beam at each stage of loading for each reinforced and partially
7-33
prestressed concrete beam tested. The typical crack patterns for some of the beams
are shown in Appendix C (see Figures C.1 through C.11). In some of the figures in
Appendix C, the load level at which the particular cracks were formed are indicated
by representing the load in kN within circles adjacent to the cracks.
Some of the conclusions regarding the cracking behaviour of reinforced and
partially prestressed concrete beams based on the experimental findings are as
follows:
(a) Based on the two-point loading condition, the cracks were mainly distributed
within the middle two-third regions of the beams.
(b) Cracking patterns show that the cracks are parallel to and equidistant from each
other, and perpendicular to the reinforcing bars. This indicates that all cracking
were produced by tensile loading and the cracks were flexural cracks.
(c) For under reinforced beams, the lengths of cracks propagated gradually with
increasing levels of loading (see Figures C.2, C.10 and C.11). Tensile failure of
the beams took place when the ultimate load was reached.
(d) For over reinforced beams, the crack propagation was not significant when
applied load varied from cracking moment to ultimate moment level (see Figures
C.1 and C.3). Beams failed due to crushing of concrete in the compression zone
when the ultimate load was reached.
(e) For continuous beams, similar crack patterns were observed for each of the
positive spans as well as the negative regions over the central support (see
Figures C.4 through C.9). In the negative zone, the cracks started from the top
of the beam and extended towards the bottom while in the positive zones the
reverse were the cases.
7.9
7-34
Summary
It has been mentioned that crack width is an important factor influencing the
damping characteristics of reinforced and partially prestressed concrete beams. The
crack width is a key factor in this investigation. The objective of the work as
presented in this chapter is to develop an accurate formula for evaluating the average
crack width of a reinforced or partially prestressed concrete beam based on the
current experimental work.
The proposed formula (Eq. (7.6)) will be used in
determining the residual crack widths of concrete beams.

The instantaneous average crack width of reinforced and partially prestressed
concrete beams at the tension face of the beam can be predicted accurately using
Eq. (7.6). The validity of the proposed crack width formula has been verified by
applying it to the test results of 26 present test beams, 26 Clark (1956)s beams, 14
Chi and Kirstein (1958)s beams and 34 Nawy (1984)s beams totalling 597 data
points.
The proposed formula has also been compared with the formulae
recommended by different national and international codes, and found to perform

equally or better than those formulae.
A relationship between maximum and average crack width was also
developed. The relationship was verified from the test results as well.
Finally certain conclusions were drawn regarding the cracking behaviour of
reinforced and partially prestressed concrete beams based on the experimental
results.
Chapter 8 : Residual Crack Width Measurements and Predictions
8-1
CHAPTER 8
RESIDUAL CRACK WIDTH MEASUREMENTS AND
PREDICTIONS
8.1
General Remarks
The main factor influencing the damping characteristics of a reinforced or partially

prestressed concrete beam, is the residual crack width. It was therefore necessary to
determine the residual crack width of the beams at zero loads (after removal of
loads), after various levels of loading had been applied, and prior to conducting the
vibration tests. During the vibration tests, the response signals produced by the
vibration in the beam were measured at various levels of residual crack widths under
condition of free vibration
The residual crack width is much smaller than the instantaneous crack width,
and is too small to measure accurately, especially in the initial stages of loading. As
such, an alternative procedure for determining residual crack widths at early stages
of loading is described in this chapter. The test procedure and the instrumentation
involved are described as well.
A formula for predicting instantaneous average crack width from the beam
parameters has been presented in Chapter 7. A relationship between residual and
8-2
instantaneous crack widths is developed in this chapter which facilitates the

prediction of residual crack widths from the beam parameters.
As an alternative approach, residual crack widths are also related to the midspan deflections of the beams. There are many well-established standard formulae
available for calculating deflections in beams.
Thus, by making use of such
formulae, residual crack widths can readily be predicted.
8.2
Test Procedure and Instrumentation
As mentioned in the preceding section, residual crack widths at early stages of

loading were too small to measure using the crack detection microscope in certain
cases. As a solution to this problem, instantaneous and residual strains at the
concrete surface of the beams were measured and a relationship between them
developed. Residual crack widths, which could not be measured directly, were
obtained indirectly from this relationship and the measured instantaneous crack
widths.
Static loads of varying magnitudes were applied to each of the beams during
the tests. At each level of loading, the instantaneous crack widths, the strains at the
concrete surface and the deflections at mid-span were recorded. In addition, the
relevant residual strains at the concrete surface of the beams were measured during
the unloading operations.
In the tests, a mechanical Demec strain gauge was used to measure the strain
at the surface of the beam.
This mechanical device manufactured by ELE
International Ltd. (model EL35-2838) enabled strain measurements to be made at

different parts of a structure using a single instrument. As shown in Figure 8.1, the
instrument consists of a main beam manufactured from a high quality steel alloy of
negligible thermal expansion, two conical locating points, one in a fixed position
and the other pivoting on a special knife edge. The strain gauge used was of 200mm gauge length. To obtain the average surface strain, measurements were made at
8-3
three points on the front surface of the beam. The locations of the measurement
points on the beam are shown in Figure 8.2. Drilled stainless steel discs were glued
to the beam surface, using a setting-out bar to locate these gauge points.
Figure 8.1 Demec strain gauge
Reinforcement
L/4
L/4
L/4
L/4
10
20 200
50
20
(a)
200
(b)
Figure 8.2 Location of strain measurement
8-4
To measure the deflection of the beam at mid-span, dial gauges were used.
In addition, for the second-stage beams, Digimatic Scales connected with Mitutoyo
Series 572 Display Units to provide a remote display of the values shown on the
scales, were used to measure the deflections. The Digimatic Scales (model 572-233)
manufactured by M.T.I. Qualos Pty. Ltd. gave direct readings with a resolution of
0.01 mm in the range of 300 mm on liquid crystal displays.
8.3
Alternative Procedure of Residual Crack Width Measurement
Both instantaneous and residual crack widths were measured using the crack
detection microscope. But in certain cases, residual crack widths were too small to
measure directly, especially at the early stages of loading. To address this problem,
the relationship between instantaneous and residual concrete surface strains of the
beams was used to define the relationship between instantaneous and residual crack
widths quantitatively.
The experimental results indicated that both the mean instantaneous beam
surface strain and crack width increased with increasing loads. To see how these
two quantities vary with load, they are plotted against the applied load for some of
the beams. To facilitate a comparison, two such plots for the same beam are given
in the same diagram. Typical plots are presented in Figures 8.3 and 8.4. Other
similar plots are presented in Appendix D. It can be seen that both the crack width
and the strain vary the same way with the applied load.
Similarly, within the elastic range, a similar pattern in the variation of the
mean residual strains at the concrete surface as the loading was removed was
observed. This is illustrated by the plots of the residual strains at the beam surface
versus the loads at which the loading was removed for some of the beams as shown
in Figures 8.5 through 8.8. Note that the residual strains were measured at each
loading stage after the removal of the load.
From the above observations, it can be concluded that at any given level of
8-5
Instantaneous average crack widths in mm
.20
.15
.10
.05
0.00
0
20
40
60
80
100
120
Load in kN
(a) Instantaneous average crack width versus loading

Instantaneous average strain in (x10-6) mm/mm
1
0
0
20
40
60
80
100
120
Load in kN
(b) Instantaneous average concrete surface strain versus loading

Figure 8.3 Instantaneous crack width/strain versus loading curves for beam 5
8-6
.4
.3
.2
.1
0.0
0
10
20
30
40
Load in kN
(a) Instantaneous average crack width versus loading
-5
Instantaneous average strain in (x 10 ) mm/mm
2.0
1.5
1.0
.5
0.0
0
10
20
30
40
Load in kN
(b) Instantaneous average concrete surface strain versus loading

Figure 8.4 Instantaneous crack width/strain versus loading curves for beam 25
8-7
Residual surface strain in (x 10 -6) mm/mm
2.0
1.5
1.0
.5
0.0
0
10
20
30
40
50
60
70
'Removal load' in kN
Figure 8.5 Residual average strain versus removal loading for beam 2
2.0
-6
Residual surface strain in (x 10 ) mm/mm
2.5
1.5
1.0
.5
0.0
0
20
40
60
80
100
120
8-8
12
Residual surface strain in (x 10
-6
) mm/mm
10
0
0
10
20
30
40
50
60
.9
-6
Residual surface strain in (x 10 ) mm/mm
1.2
.6
.3
0.0
0
10
20
30
40
8-9
loading, the proportional variation between instantaneous and residual strains ( i

and r, respectively) at the concrete surface should be approximately the same as the
proportional variation between instantaneous and residual crack widths (wcr and wr ,
respectively).
In other words, mathematically speaking the following relationship can be
set up:
i = w cr = c
wr
r
(8.1)
where c is a constant for any given loading.

For each specimen and, at any given loading condition, the mean
instantaneous strain and crack width can be obtained for the loading. Relevant
residual strain can also be obtained when the loading is removed. For the cases
where residual crack widths cannot be measured directly, can thus be determined
using Eq. (8.1). That is to say that for those cases, residual crack widths can be
obtained using the following:
wr = wcr r
i
8.4
(8.2)
Deflection at Mid-Span of Beam
Along with the measurements of crack widths and beam surface strains at each level
of loading, the deflection at mid-span of the beam was recorded by a dial gauge
fitted to the bottom of the beam.
To check the accuracy of dial gauge
measurements, a Digimatic Scale was used parallel with the dial gauge for the
second-stage beams.
Based on the experimental results the load-deflection diagrams for some of
the beams are drawn and presented in Figures 8.9 through 8.16. These figures
8-10
further confirm the accuracy of deflection measurements. Since direct relationship

between deflections and residual crack widths is developed in this research and is
recommended for use in predicting residual crack widths from deflections, the
accuracy in deflection measurement was important.
120
100
Load in kN
80
60
40
20
0
0
10
20
30
40
50
Mid-span deflection in mm
Figure 8.9 Load-deflection diagram for beam 5

300
250
Load in kN
200
150
100
50
0
0
10
20
30
40
50

8-11
30
25
Load in kN
20
15
10
0
0
50
Load in kN
40
30
20
10
0
0
8-12
40
Lpoad in kN
30
20
10
0
0
10
40
Load in kN
30
20
10
0
0
8-13
25
Load in kN
20
15
10
0
0

30
25
Load in kN
20
15
10
0
0
12
16
20
8.5
8-14
Prediction of Residual Crack Width
As outlined in Chapter 4 (see Section 4.6), relationships between residual and

instantaneous crack widths as well as between residual crack widths and mid-span
deflections are developed in this section. These relationships result in formulae for
predicting residual crack widths, respectively from instantaneous crack widths and
mid-span deflections.
The development of these formulae is detailed in the
following sub-sections.
8.5.1
Prediction from instantaneous crack width
The test results indicate that residual crack widths are only a fraction of
instantaneous crack widths and there exists a direct relationship between these two
parameters. In order to obtain this relationship, residual crack widths for the test
beams are plotted against their corresponding instantaneous crack widths.
The
equation of the best-fit line passing through the origin for all the data points in such
a plot gives the formula.
While the first-stage beam results are used for the development of the
residual crack width formula, the results for the second-stage beams are used for the
verification of the formula. The plot of residual crack widths against instantaneous
crack widths for the first-stage beams, for which the results were available, is
presented in Figure 8.17. The beams whose results were used in this plot are beams
1, 4, 10, 11 and 20 through 30.
The equation of the best-fit straight line as shown in Figure 8.17, or in other
words the formula for the prediction of residual crack widths is,
wr = 0.312 wcr
(8.3)
where wr and wcr are residual and instantaneous average crack widths, respectively.
8-15
Residual average crack widths in mm
.20
.15
.10
.05
0.00
0.0
.1
.2
.3
.4
Figure 8.17 Relationship between residual and instantaneous average crack widths
The
coefficient
of
determination
(R2)
and
the
beta
coefficient
(Hair et. al., 1992) for this relationship were 0.822 and 0.907, respectively. Use of
the expression for wcr from Eq. (7.6) into Eq. (8.3) yields,
wr = 0.312 (fs /Es) [0.6 (c s) + 0.1 (/)]
(8.4)
where fs, Es, c, s, and are as defined before.

Thus, Eq. (8.4) gives the formula for predicting residual crack widths from
the general variables defining the beams.
In order to verify the accuracy of the relationship developed above, average
residual crack widths are calculated using Eq. (8.3) for the second-stage beams and
compared with the corresponding measured values. The measured average residual
crack widths are plotted against the calculated values as shown in Figure 8.18. The
beams used in this comparison are beams 5, 7, 16, 17 and 18. Note that, residual
crack widths were not measured for beams 12 through 15, which were used for the
8-16
verification of the proposed instantaneous average crack width formula only. Also
note that, for beams 16 through 18, the residual crack widths are for each of the
positive spans and the negative span of each beam.
Measured average residual crack widths in mm
.12
.09
Legend
.06
- 30% line
+ 30% line
.03
Continuous beams
Simply supported
0.00
beams
0.00
.03
.06
.09
.12
Calculated average residual crack widths in mm
Figure 8.18
Measured versus calculated average residual crack widths for the

second-stage beams (using Eq. (8.3))
Data points for simply supported and two-equal-span continuous beams are
shown separately in Figure 8.18. As seen from Figure 8.18, most of the 120 data
points lie within + 30% limits. This, along with high values for the coefficient of
determination (0.822) and the beta coefficient (0.907), confirms the validity of the
relationship developed in Eq. (8.3).
8.5.2
Prediction from mid-span deflection
Using the test data for residual crack widths and deflections at mid-span, a
relationship between these two parameters may be determined.
Two such
8-17
relationships are developed one for the reinforced concrete beams and the other for
the partially prestressed concrete beams.
Residual crack widths for the reinforced concrete beams are plotted against
their corresponding mid-span deflections and presented in Figure 8.19. The data
used in this plot are for beams 5, 7, 11, 16, 17 and 18. The deflection data for most
of the first-stage reinforced concrete beams were not available to the author. Like
residual crack widths, deflections also were not measured for beams 12 through 15,
which were used for the verification of the proposed instantaneous average crack
width formula only.
Average res idual crack widths in mm
.08
.06
.04
.02
0.00
0
10
20
30
40
50
Mid-span deflections in mm
Figure 8.19
Relationship between average residual crack widths and mid-span

deflections for reinforced concrete beams
The line of best-fit is shown in Figure 8.19 and the equation for this line or
the relationship between average residual crack widths (wr) and mid-span
deflections () for reinforced concrete beams is:
wr = 0.0014
(8.5)
8-18
Eq. (8.5) gives the residual crack widths in terms of mid-span deflections for
reinforced concrete beams. The coefficient of determination (R2) and the beta
coefficient for this relationship were 0.863 and 0.929, respectively.
Such high
values for these coefficients are indicative of the correctness of the relationship.
Similarly, for the partially prestressed concrete beams, average residual
crack widths are plotted against their corresponding mid-span deflections and
presented in Figure 8.20. The data used in this plot are for beams 20, 24, 25, 28,
29 and 30. Once again, deflection data for other first-stage partially prestressed
concrete beams were not available to the author.
Average res idual crack widths in mm
.14
.12
.10
.08
.06
.04
.02
0.00
0
10
12
Mid-span deflections in mm
Figure 8.20
Relationship between average residual crack widths and mid-span

deflections for partially prestressed concrete beams
The equation for the line of best-fit, as shown in Figure 8.20, or in other
words, the relationship between average residual crack widths (wr) and mid-span
deflections () for partially prestressed concrete beams is obtained as:
wr = 0.0084
(8.6)
8-19
Thus, Eq. (8.6) is the formula for predicting residual crack widths in terms of
mid-span deflections for partially prestressed concrete beams. The coefficient of
determination (R2) and the beta coefficient for this relationship were 0.883 and
0.940, respectively. The values for these coefficients are even higher to justify the
accuracy of the relationship developed in Eq. (8.6).
There are many formulae available for accurately predicting mid-span
deflections in simply supported beams or individual spans of continuous beams.
Such formulae are available for both reinforced and partially prestressed concrete
beams. Probably the most extensively documented studies regarding deflection
calculations can be found in Branson (1977). Various other methods have been
proposed for the calculation of deflections (ACI Committee 435, 1966; 1974; 1985;
Gilbert, 1983). There are deflection calculation formulae recommended by different
codes of practice as well.
Making use of any of the formulae given by the researchers to obtain
deflection, once the deflection is known, residual crack widths can then be
determined using Eq. (8.5) or (8.6), respectively for reinforced or partially
8.6
Summary
In this chapter, the main work involved the determination of the residual crack
widths of concrete beams, especially at the early stages of loading. To do this,
measurements of instantaneous and residual surface strains of the concrete beams
were carried out. The residual crack widths, then can be determined using Eq. (8.2)
for any given level of loading within the elastic range. The relationship between
residual crack widths and the general beam parameters has been developed as shown
in Eq. (8.4). Alternatively, relationships between residual crack widths and midspan deflections have also been developed, separately for reinforced and partially
prestressed concrete beams (Eqs. (8.5) and (8.6)). Any of these proposed formulae
can be used for the prediction of residual crack widths.
8-20
As vibration measurements of the concrete beams were performed after the

removal of the applied loading, the residual crack width is the only parameter related
to the measured vibration parameter logarithmic decrement. Accordingly, residual
crack widths are used as the basis to propose a computational procedure for the
determination of the damping capacity of concrete beams in the chapters to follow.
Chapter 9 : Damping in Reinforced Concrete Beams
9-1
CHAPTER 9
DAMPING IN REINFORCED CONCRETE BEAMS
9.1
General Remarks
The rapid developments in instrumentation, measurement and computer systems

have made it possible to carry out detailed investigation of the dynamic behaviour of
prototype civil engineering structures readily and accurately (Severn et al., 1988).
Such investigations have an important role to play in understanding the dynamic
behaviour of real structures and providing the necessary information to determine
the accuracy of the assumptions used to formulate a mathematical model. From the
complexity of several factors influencing the dynamic responses of elements or
buildings, it is obvious that prototype studies are necessary to yield the information
of the dynamic characteristics of the structure (Paskalov et al., 1980;
Williams, 1988).
The most severe earthquake likely to happen can only be survived by a
building if the members of the building are sufficiently ductile to absorb and
dissipate seismic energy by inelastic deformation.
This requires a designer to
realistically assess the possible levels of strength in flexural and shear elements
(Rosenblueth, 1980). So, in designing earthquake resistant concrete structures, it is
important to understand and determine the ability of the structure to absorb energy
under an external impulsive force. At this stage, information in this regard is
lacking in published literature and the ability of building elements to absorb energy
9-2
is not well understood. This, for example, is true of reinforced concrete beams,
particularly cracked reinforced concrete beams.
In particular, no simple and
accurate formulae are available to evaluate the damping ratios of reinforced concrete
beams cracked or uncracked, for use in the dynamic design of concrete structures.
In this study, experiments were carried out on full-scale reinforced concrete
beams with simple supports to investigate the effect of cracking on damping
behaviour. The instrumentation and the measurement procedures for the recording
of the vibration data are described in detail in this chapter. A simple and accurate
formula for predicting damping ratios in terms of logarithmic decrements for
reinforced concrete simply supported beams is developed in this chapter.
The
proposed damping formula is verified based on the laboratory test results.
Its
applicability to continuous beams is also verified. Finally, the effects of different

parameters on the damping values are also discussed aided by the experimental
results.
9.2
Instrumentation and Data Recording
All the equipment and instruments used in carrying out the experimental work for
the current study are briefly described in the following sub-sections.
9.2.1
Loading system
Static loading was used in this study to produce cracks in concrete beams. The
loading system used is described in Chapter 6 (see Section 6.7). Widths of cracks in
the concrete beams were measured using the crack detection microscope (see
Section 7.2).
9.2.2
Vibration measurement
As described in Chapter 2, there are several methods for determining the damping
9-3
capacity, but the most direct method is the free decay method. This involves the
measurement of the decrease in amplitude of free vibrations and calculating the
logarithmic decrement, , using Eq. (2.2).
The free decay method with hammer excitation was used to measure the
response signals of the concrete beams. The hammer excitation method is popular
for vibration measurements. It is a relatively simple means of exciting the structure
into vibration.
An ICP (Integrated Circuit Piezoelectric) impulse-force hammer (Model
086C04) was used. The hammer consists of an integral, ICP, quartz force sensor
mounted on the striking end of the hammer head. The sensing element functions to
transfer impact force into an electrical signal for display and analysis.
It is
structured with rigid quartz crystals and a built-in, microelectronic unity gain,
amplifier. The cable is connected to the end of the handle for convenience and to
avoid connector damage should a miss hit occur. The hammer is 20.3 cm long, has a
head dimeter of 1.5 cm, tip diameter of 0.63 cm, a mass of 0.14 kg and a sensitivity
of 1.2 mV/N in a range of 4400 N (5V output).
To record the signal produced by beam vibration, an accelerometer was set
up in the middle of the beam for each test. A vibration pick-up can be mounted by
using different methods as shown in Figure 9.1 (Major, 1980). Mounting with steel
studs is the best stiff solution (see Figure 9.1 (a)). When electrical isolation is
necessary, an isolated stud with mica washers should be adopted (see
Figure 9.1 (b)). Sometimes, a thin layer of wax is applied to stick the accelerometer
to the vibrating surface as shown in Figure 9.1 (c). As shown in Figure 9.1 (d), a
permanent magnet also gives electrical isolation.
There is a permanent mounting magnet on the bottom of the accelerometer to
attach it to the steel, if this is being used. In the current tests, a thin steel washer of
the same diameter as the accelerometer was fixed to the top/bottom of the concrete
beam using cyanoacrylate glue (for the first-stage beams) or Petro wax (for the
9-4
second-stage beams). Details of the accelerometer and its mountings on the beam
are shown in Figure 9.2.
The accelerometers used were quartz shear mode accelerometers and offer
high performance for precision accelerometer measurements. The use of quartz
sensing crystals operating in the shear mode provides very low sensitivity to
environmental conditions such as transient thermal effects, transverse motion and
base strain. The built-in ICP conditioning electronics allow ease of use and test setup while the naturally polarized quartz sensing element offers long term stability and
repeatability. The sensitivity of the accelerometers varied from 10 mV/g [or 1.02
mV/(m/s2)] to 100 mV/g [or 10.19 mV/(m/s2)].
Figure 9.1 Different methods of vibration pick-up mountings (after Major, 1980)
Signal cable
Concrete beam
Accelerometer
Steel washer
Figure 9.2 Mounting of accelerometer on concrete beam
9-5
Figure 9.3 2630 Personal Fourier Analyzer
Figure 9.4 TDS 460A Digitizing Oscilloscope
9-6
Top of beam
Location of accelerometer
250
mm
l = 1000 mm for beams 1 through 4, 6, and 19 through 22

l = 1100 mm for beams 8,9,10, and 23 through 26
l = 1250 mm for beams 11, and 27 through 30
Points A, B and C are hammer excitation points

Figure 9.5 Hammer excitation positions for the first-stage beams
Figure 9.6 Hammer excitation positions for the second-stage beams
9-7
To measure and record the vibration signal, a 2630 Personal Fourier

Analyzer (see Figure 9.3) was used for the first-stage beams.
The analyzer
provides a visual display for making accurate displacement measurements. It has

four channels, optional zoom, optional source, complete alias protection and
selectable resolution from 25 to 1600 spectral lines, and it can be configured to
meet the structural measurement requirements.
For the second-stage beams, a TDS 460A Digitizing Oscilloscope (see
Figure 9.4) was used for the recording and measurement of vibration signals. The
digitizing oscilloscope is a tool for acquiring, displaying and measuring
waveforms. Its performance addresses the needs of both laboratory and portable
applications.
It has four channels and four eight-bit digitizers, 400 MHz
maximum analog bandwidth with 100 Megasamples/second maximum digitizing

rate, up to 30000-point record length per channel, and complete measurement and
documentation ability. Apart from simple mathematical functions, it can integrate
or differentiate a single waveform or perform an FFT (Fast Fourier Transform) on
a waveform to display its magnitude or phase versus its frequency.
The
oscilloscope supports five different acquisition modes, has zoom option, full
software programmability and on-board printer capability.
During the tests, three excitation positions for each damping measurement
were set up to check the reliability and accuracy of the vibration recordings.
Theoretically, the damping value of the beam should be the same irrespective of the
hammer excitation positions. The locations of the excitation points on the top of
the beam for the first-stage beams are shown in Figure 9.5. The hammer excitation
positions for the second-stage beams are shown in Figure 9.6. To determine the
value for the logarithmic decrement at each test, the average of the values for the
three excitation positions was taken.
9.2.3
Vibration measurement procedure
Current rapid development in instrumentation, measurement and computer systems
9-8
have made it possible to carry out detailed dynamic testings at a reasonable cost.
Such testings have an important role to play in the development of our
understanding of the dynamic behaviour of real-life structures and ultimately in the
development of more reliable design and analytical procedures. Various techniques
for dynamic testing and monitoring of prototype civil engineering structures have
become well established research tools (Severn et al., 1988).
Figure 9.7 Stages in a vibration measurement system

Figure 9.7 shows the major stages of a vibration measurement system. The
system consists of a motion detector-transducer, an intermediate signal modification
system (e.g. amplifier) and a display system (e.g. oscilloscope). The type of motion
to be measured may be displacement, velocity or acceleration. The purpose of
detector-transducer is to detect the desired mode of motion and, in most cases, to
produce an output that is proportional to the input motion but of different form. The
most widely used motion transducer is accelerometer, a device that senses
9-9
acceleration and produces an output voltage proportional to the input acceleration

(Craig, 1981).
Vibration signals were picked up by an accelerometer attached to the
top/bottom of the beam (stage 1). After passing through a signal amplifier (stage 2),
they were converted into displacements to be displayed on the oscilloscope/signal
analyzer (stage 3).
9.3 Presentation of Test Results

In this investigation, the damping characteristics of 14 full-size reinforced concrete
beams with various residual crack widths present were studied in the laboratory.
The major objectives were (a) to examine the effect of cracking on damping
behaviour, and (b) to develop a simple and accurate formula for determining the
damping ratios of reinforced concrete beams with or without cracking.
Static loading was applied to reinforced concrete beams at two points (see
Figure 6.21) to produce cracking. At various levels of loading, instantaneous crack
widths and concrete surface strains were measured. At each level of loading, after
the removal of the load, measurements were taken to determine average residual
crack widths and/or concrete surface strains, and the corresponding free decay
vibration was also measured. The quantitative relationship between average residual
crack widths in the beam and the logarithmic decrement values can then be
determined.
The damping data may be presented under three categories to show the effect
of cracking on the damping values.
These are for (a) over reinforced simply
supported beams, (b) balanced and under reinforced simply supported beams, and
(c) two-equal-span continuous beams. Typical vibration response curves for the
reinforced concrete beams at various levels of cracking are presented in Appendix E.
9.3.1
9-10
Over reinforced simply supported beams
It is well known that there are three types of failure for reinforced concrete beams:
balanced failure, compression failure and tension failure.
For over reinforced simply supported beams, compression failure would
occur when ultimate loading is reached. For the whole range of load application, the
tension steel remained in the elastic range, and resulted in a very small residual
crack width when the load was removed. The variation in the logarithmic decrement
values was very small and usually less than 15%.
Logarithmic decrement values have been plotted against the corresponding
residual crack widths for the over reinforced simply supported beams and presented
in Figures F.1 through F.6 given in Appendix F to illustrate this point.
9.3.2
Under reinforced simply supported beams
For this type of reinforced concrete beams, tension failure would occur when the
ultimate loading is reached. The tensile reinforcing steel yields when the bending
moment reached the yielding moment. As the reinforcing steel did not return to its
original shape, it resulted in relatively large residual crack widths when the loading
was removed. Thus, the logarithmic decrement values increased significantly with
increasing residual crack widths for this type of beams. Similar behaviour was
exhibited by the beams which had a balanced failure.
The above findings are demonstrated by plotting logarithmic decrement
values against the corresponding residual crack widths for the balanced and under
reinforced simply supported beams as presented in Figures F.7 through F.11 in
Appendix F.
9.3.3
9-11
Two-equal-span continuous beams
For the two-equal-span continuous reinforced concrete beams tested, the damping
behaviour of each individual span as related to cracking was very much similar to
those of simply supported concrete beams. To demonstrate this particular feature of
the damping behaviour of continuous beams, logarithmic decrement values for each
individual span of the beams are plotted against their corresponding average residual
crack widths. These are presented in Figures F.12 through F.17 given in Appendix F.
An inspection of Figures F.12 through F.17 reveals that the individual spans
of the continuous beams behaved more like over reinforced simply supported beams.
This may be attributed to the constraints provided by the continuity of the beam over
the central support and/or the redistribution of moments taking place during the final
stages of loading.
9.4 Damping Formula for Reinforced Concrete Beams

One of the major objectives of the research was to develop a computational
procedure for evaluating the damping capacity of reinforced concrete beams. As
such, a formula for computing damping in reinforced concrete beams is developed
and verified from the laboratory test results in this section.
9.4.1
Development of the damping formula
In any structure a number of mechanisms contribute to the total damping. Damping

occurs within the structure due to viscosity, hysteresis, yielding and friction, and
also externally because of actions at supports, radiation of energy into the ground
and fluid damping (normally by air resistance).
Thus damping in reinforced
concrete beams may be influenced by many factors.
Observation of the
experimental results indicates the main factor influencing the damping values
(measured as logarithmic decrements) to be the residual crack width (as shown in
9-12
Section 9.3). The effects of other factors such as, steel ratios, span lengths and
compressive strengths of concrete, are found to be negligible and inconclusive as
will be shown in Section 9.5.
Accordingly, the logarithmic decrement values and the average residual
crack widths of selected reinforced concrete beams are used to carry out a regression
analysis (Nie et al., 1975), resulting in a damping prediction formula for simply
supported reinforced concrete beams. The relevant data for beams 1, 6, 8 and 11 are
used in the regression analysis. The reasons behind selecting these beams are that
they represent the three different span lengths (5.5 m for beams 1 and 6; 6.7 m for
beam 8; and 8.0 m for beam 11) and the two major types of reinforcements used
(3-Y20 bars for beams 1, 8 and 11; and 6-Y20 bars for beam 6). The damping data
for these four beams used in the regression analysis, are tabulated in Tables 9.1
through 9.4.
Table 9.1 Damping data for beam 1
Removal
Load
(kN)
0
10
15
20
25
30
Average
residual
crack width
(mm)
0
0.050
0.080
0.132
0.176
0.198
35
40
45
50
55
60
65
70
80
0.220
0.220
0.264
0.308
0.374
0.396
0.418
0.500
0.525
Logarithmic decrement,
for hammer for hammer for hammer
excitation at excitation at excitation at
point A
point B
point C
0.074
0.077
0.075
0.083
0.082
0.080
0.079
0.082
0.081
0.080
0.083
0.083
0.082
0.083
0.083
0.085
0.086
0.084
0.084
0.085
0.079
0.085
0.088
0.089
0.100
0.104
0.108
0.084
0.084
0.086
0.089
0.091
0.098
0.103
0.106
0.107
0.085
0.086
0.088
0.088
0.090
0.094
0.098
0.107
0.108
Average
logarithmic
decrement,
0.075
0.082
0.081
0.082
0.083
0.085
0.084
0.085
0.085
0.087
0.090
0.094
0.101
0.106
0.108
9-13

Removal
Load
(kN)
5.8
9.8
19.6
30
40
60
80
97
Average
residual
crack width
(mm)
0
0.035
0.050
0.064
0.067
0.078
0.088
0.094
point A
point B
point C
0.092
0.088
0.087
0.089
0.088
0.089
0.091
0.089
0.089
0.096
0.094
0.089
0.098
0.094
0.097
0.098
0.097
0.097
0.099
0.097
0.098
0.099
0.098
0.098
Average
logarithmic
decrement,
0.089
0.089
0.090
0.093
0.096
0.097
0.098
0.098

Removal
Load
(kN)
0
5
10
15
20
25
30
Average
residual
crack width
(mm)
0
0.004
0.008
0.132
0.176
0.220
0.264
35
40
45
50
55
60
65
70
75
80
85
0.308
0.352
0.440
0.352
0.440
0.660
0.792
0.880
1.012
1.100
1.144
point A
point B
point C
0.063
0.064
0.066
0.070
0.071
0.070
0.071
0.073
0.070
0.068
0.074
0.075
0.070
0.072
0.074
0.073
0.075
0.077
0.085
0.082
0.080
0.090
0.091
0.094
0.097
0.098
0.100
0.104
0.109
0.110
0.118
0.121
0.088
0.090
0.097
0.098
0.100
0.103
0.105
0.110
0.114
0.117
0.123
0.086
0.090
0.094
0.097
0.099
0.100
0.104
0.108
0.111
0.117
0.121
Average
logarithmic
decrement,
0.064
0.070
0.071
0.072
0.072
0.075
0.082
0.088
0.090
0.095
0.097
0.099
0.101
0.104
0.109
0.112
0.117
0.122
9-14

Removal
Load
(kN)
0
5
10
15
20
25
Average
residual
crack width
(mm)
0
0.005
0.009
0.010
0.022
0.075
point A
point B
point C
0.057
0.055
0.057
0.058
0.057
0.058
0.059
0.057
0.058
0.061
0.060
0.059
0.063
0.061
0.062
0.064
0.063
0.065
Average
logarithmic
decrement,
0.056
0.058
0.058
0.060
0.062
0.064
The regression resulted in the following damping prediction formula for

simply supported reinforced concrete beams:
= 0.075 x 10 0.205 w r
(9.1)
where is the damping in terms of logarithmic decrements and wr is the average

residual crack widths in mm.
In Eq. (9.1), the average residual crack widths (wr) are obtained using
Eq. (8.4) or (8.5). Thus obtained logarithmic decrement () values can then be used
to compute the damping ratios, , using Eq. (4.47) or (4.48), depending on the
accuracy required. Consequently, the damping capacity/ratio of reinforced concrete
simply supported beams can be obtained from the general beam parameters.
9.4.2
Comparison with test results
In order to check the validity of the regression analysis, that is, to verify the
accuracy of the proposed damping prediction formula, predicted logarithmic
decrement values are compared with the measured values for all 14 reinforced
concrete beams. This includes the data from the three two-equal-span continuous
beams as well, to extend the applicability of the proposed formula to continuous
beams.
9-15
A plot of measured logarithmic decrement values against the calculated

values using Eq. (9.1) for the 14 reinforced concrete beams is given in Figure 9.8.
The first-stage and the second-stage beams as well as the simply supported beams
and the continuous beams are shown separately in this figure.
Measured logarithmic decrements, m
.16
.12
Legend
- 30% line
.08
+ 30% line
Second-stage continuous beams
.04
Second-stage simply
supported beams
First-stage beams
0.00
0.00
.04
.08
.12
.16
Calculated logarithmic decrements, c
Figure 9.8 Measured versus calculated damping values for the reinforced concrete
beams
As can be seen from Figure 9.8, an excellent correlation exists between the
calculated and the measured damping values as all but 8 of the 191 correlation
points lie well within + 30% limits. Considering the fact that the first- and the
second-stage beams were tested at different test environments such as different
support types were used, different spacings for the two-point loadings were
employed, different vibration recording devices were used, etc., the performance of
the proposed damping formula was even more satisfactory. Note that the proposed
formula is developed from the data of only 4 of the first-stage beams.
9-16
Eq. (9.1) is applied to different levels of residual crack widths corresponding

to different removal loads to determine the theoretical logarithmic decrement
values for each beam.
Note that both residual crack widths and logarithmic
decrement values were measured at each level of loading after the removal of the
load. These theoretical values together with the measured values corresponding to
the various removal loads are listed for each beam in Tables 9.5 through 9.21
(these were plotted in Figure 9.25 as well). The ratios of measured to calculated
logarithmic decrement values were also computed and are listed in the same tables.
It can be seen that the value of this ratio is very close to 1 and in certain cases
exactly equal to 1. The standard deviations were well within 10 percent for all
beams except with beam 9 and span 1 of beam 18.
For the first-stage beams, which were all simply supported beams, the means
of the ratios of measured to calculated logarithmic decrement values varied from
0.79 to 1.22 with standard deviations varying from 3 to 7 percent (see Tables 9.5
through 9.8, 9.10, 9.12, 9.14 and 9.15). The only exception was beam 9 for which,
the mean ratio and the standard deviation were 1.31 and 21%, respectively (see
Table 9.13). In fact, all the 8 correlation points falling outside + 30% limits in
Figure 9.8 belonged to beam 9.
For the second-stage simply supported beams, the mean ratios varied from
0.98 to 0.99 and the standard deviations varied from 3 to 5 percent (see Tables 9.9
and 9.11). Thus the proposed formula performed extremely well for the secondstage simply supported beams.
Finally, for the individual spans of the second-stage continuous beams, the
mean ratios varied from 1.05 to 1.14 and the standard deviations varied from 5 to 13
percent only. Such low values for the standard deviations and the values of the
mean ratios so close to unity (within 15 percent) confirms the validity of the
applicability of the proposed damping formula developed for simply supported
beams to continuous beams.
9-17
Table 9.5 Relation between measured and calculated logarithmic decrement values
for beam 1
Removal
load
(kN)
0
10
15
20
25
30
35
40
45
50
55
60
65
70
80
Average
residual
crack
width
(mm)
Measured logarithmic decrement

for
for
for
hammer
hammer
hammer
excitation excitation excitation
at point A at point B at point C
0
0.074
0.050
0.083
0.080
0.079
0.132
0.080
0.176
0.082
0.198
0.085
0.220
0.084
0.220
0.085
0.264
0.079
0.308
0.085
0.374
0.088
0.396
0.089
0.418
0.100
0.500
0.104
0.525
0.108
Mean ratio
Standard deviation
0.077
0.082
0.082
0.083
0.083
0.086
0.084
0.084
0.086
0.089
0.091
0.098
0.103
0.106
0.107
0.075
0.080
0.081
0.083
0.083
0.084
0.085
0.086
0.088
0.088
0.090
0.094
0.098
0.107
0.108
Measured
average
logarithmic
decrement
m
0.075
0.082
0.081
0.082
0.083
0.085
0.084
0.085
0.085
0.087
0.090
0.094
0.101
0.106
0.108
1.04
0.04
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.075
0.077
0.078
0.080
0.081
0.082
0.083
0.083
0.085
0.087
0.089
0.090
0.091
0.095
0.096
1.00
1.07
1.04
1.03
1.02
1.03
1.01
1.02
1.00
1.00
1.01
1.04
1.11
1.12
1.12
for beam 2
Removal
load
(kN)
0
10
20
30
40
50
60
70
80
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
0
0.064
0.176
0.070
0.198
0.080
0.616
0.089
0.660
0.087
0.880
0.102
1.100
0.125
1.320
0.130
1.540
0.138
Mean ratio
Standard deviation
0.065
0.074
0.080
0.091
0.092
0.100
0.127
0.131
0.135
0.061
0.069
0.081
0.090
0.093
0.100
0.126
0.133
0.136
Measured
average
logarithmic
decrement
m
0.063
0.071
0.080
0.090
0.091
0.101
0.126
0.131
0.136
0.91
0.05
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.075
0.081
0.082
0.100
0.102
0.114
0.126
0.140
0.155
0.84
0.87
0.97
0.90
0.89
0.89
1.00
0.94
0.88
9-18
for beam 3
Removal
load
(kN)
0
10
20
30
40
50
55
60
65
70
75
80
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
0
0.077
0
0.079
0
0.081
0
0.080
0
0.084
0
0.087
0
0.085
0.002
0.087
0.003
0.087
0.008
0.086
0.015
0.087
0.018
0.088
Mean ratio
Standard deviation
0.076
0.078
0.079
0.080
0.082
0.084
0.085
0.087
0.088
0.088
0.089
0.089
0.078
0.079
0.079
0.081
0.082
0.082
0.084
0.085
0.086
0.087
0.088
0.089
Measured
average
logarithmic
decrement
m
0.077
0.079
0.080
0.080
0.083
0.084
0.085
0.086
0.087
0.087
0.088
0.089
1.12
0.05
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.075
0.075
0.075
0.075
0.075
0.075
0.075
0.075
0.075
0.075
0.076
0.076
1.03
1.05
1.07
1.07
1.11
1.12
1.13
1.15
1.16
1.16
1.17
1.18
for beam 4
Removal
load
(kN)
Average
residual
crack
width
(mm)
0
10
20
0
0.018
0.020
0.083
0.092
0.089
0.080
0.088
0.089
30
40
50
60
70
80
0.025
0.028
0.034
0.078
0.088
0.096
0.091
0.096
0.098
0.098
0.099
0.099
0.089
0.094
0.094
0.097
0.097
0.098
Measured
average
logarithmic
decrement
m
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.081
0.087
0.088
0.081
0.089
0.089
0.075
0.076
0.076
1.08
1.18
1.18
0.089
0.089
0.097
0.097
0.098
0.099
0.090
0.093
0.096
0.097
0.098
0.099
0.076
0.076
0.076
0.078
0.078
0.078
1.19
1.22
1.26
1.25
1.25
1.26

for
for
for
hammer
hammer
hammer
Mean ratio
Standard deviation
1.21
0.06
9-19
for beam 5
Removal
load
(kN)
Average
residual
crack
width
(mm)
0
25
35
45
55
65
75
85
95
105
0
0.027
0.025
0.032
0.025
0.028
0.030
0.027
0.036
0.034

for
for
for
hammer
hammer
hammer
0.064
0.070
0.070
0.071
0.063
0.066
0.079
0.071
0.086
0.088
0.098
0.078
0.073
0.062
0.077
0.082
0.072
0.076
0.091
0.076
0.058
0.073
0.081
0.079
0.067
0.073
0.072
0.074
0.072
0.073
Mean ratio
Standard deviation
Measured
average
logarithmic
decrement
m
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.073
0.074
0.075
0.071
0.069
0.074
0.074
0.074
0.083
0.079
0.075
0.076
0.076
0.076
0.076
0.076
0.076
0.076
0.076
0.076
0.97
0.97
0.98
0.93
0.91
0.97
0.98
0.97
1.09
1.04
0.98
0.05
for beam 6
Removal
load
(kN)
Average
residual
crack
width
(mm)
5.8
9.8
19.6
30
40
60
80
97
0
0.035
0.05
0.064
0.067
0.078
0.088
0.094

for
for
for
hammer
hammer
hammer
0.092
0.089
0.091
0.096
0.098
0.098
0.099
0.099
Mean ratio
Standard deviation
0.088
0.088
0.089
0.094
0.094
0.097
0.097
0.098
0.087
0.089
0.089
0.089
0.097
0.097
0.098
0.098
Measured
average
logarithmic
decrement
m
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.089
0.089
0.09
0.093
0.096
0.097
0.098
0.098
0.075
0.076
0.077
0.077
0.077
0.078
0.078
0.078
1.19
1.17
1.17
1.20
1.24
1.25
1.25
1.25
1.22
0.04
9-20
for beam 7
Removal
load
(kN)
0
25
40
55
70
85
100
120
140
160
180
200
230
Average
residual
crack
width
(mm)
0
0
0.012
0.015
0.011
0.015
0.019
0.016
0.019
0.023
0.028
0.043
0.09

for
for
for
hammer
hammer
hammer
0.077
0.070
0.071
0.071
0.073
0.067
0.075
0.074
0.077
0.086
0.077
0.074
0.070
0.076
0.074
0.074
0.076
0.073
0.078
0.070
0.078
0.073
0.078
0.077
0.071
0.071
0.069
0.076
0.074
0.076
0.074
0.078
0.081
0.078
0.077
0.076
0.084
0.072
0.078
Mean ratio
Standard deviation
Measured
average
logarithmic
decrement
m
0.073
0.070
0.075
0.079
0.073
0.074
0.075
0.076
0.070
0.075
0.078
0.077
0.078
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.075
0.075
0.075
0.076
0.075
0.076
0.076
0.076
0.076
0.076
0.076
0.077
0.078
0.97
0.94
1.00
1.04
0.97
0.98
1.00
1.00
0.93
0.99
1.02
1.00
0.99
0.99
0.03
for beam 8
Removal
load
(kN)
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
Average
residual
crack
width
(mm)
0
0.004
0.008
0.132
0.176
0.220
0.264
0.308
0.352
0.440
0.352
0.440
0.660
0.792
0.880
1.012
1.100
1.144

for
for
for
hammer
hammer
hammer
0.063
0.064
0.066
0.070
0.071
0.070
0.071
0.073
0.070
0.068
0.074
0.075
0.070
0.072
0.074
0.073
0.075
0.077
0.085
0.082
0.080
0.090
0.088
0.086
0.091
0.090
0.090
0.094
0.097
0.094
0.097
0.098
0.097
0.098
0.100
0.099
0.100
0.103
0.100
0.104
0.105
0.104
0.109
0.110
0.108
0.110
0.114
0.111
0.118
0.117
0.117
0.121
0.123
0.121
Mean ratio
Standard deviation
Measured
average
logarithmic
decrement
m
0.064
0.070
0.071
0.072
0.072
0.075
0.082
0.088
0.090
0.095
0.097
0.099
0.101
0.104
0.109
0.112
0.117
0.122
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.075
0.075
0.075
0.080
0.081
0.083
0.085
0.087
0.089
0.092
0.089
0.092
0.102
0.109
0.114
0.121
0.126
0.129
0.85
0.93
0.94
0.90
0.88
0.90
0.97
1.01
1.02
1.03
1.10
1.07
0.99
0.95
0.96
0.93
0.93
0.95
0.96
0.07
9-21
for beam 9
Removal
load
(kN)
0
10
20
30
35
40
45
50
55
60
65
70
75
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
0
0.074
0.010
0.080
0.132
0.083
0.132
0.085
0.176
0.097
0.220
0.120
0.264
0.123
0.264
0.123
0.308
0.126
0.308
0.129
0.352
0.128
0.396
0.135
0.440
0.138
Mean ratio
Standard deviation
0.073
0.081
0.082
0.084
0.098
0.121
0.123
0.123
0.127
0.128
0.128
0.136
0.137
0.075
0.080
0.084
0.084
0.100
0.120
0.124
0.125
0.126
0.127
0.129
0.133
0.138
Measured
average
logarithmic
decrement
m
0.074
0.080
0.083
0.084
0.098
0.120
0.123
0.124
0.126
0.128
0.128
0.135
0.138
1.31
0.21
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.075
0.075
0.080
0.080
0.081
0.083
0.085
0.085
0.087
0.087
0.089
0.090
0.092
0.99
1.06
1.04
1.05
1.20
1.44
1.45
1.46
1.45
1.48
1.45
1.49
1.49
for beam 10
Removal
load
(kN)
0
10
20
30
40
50
60
70
80
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
0
0.075
0
0.075
0
0.076
0
0.077
0
0.077
0.001
0.079
0.003
0.081
0.003
0.085
0.006
0.086
Mean ratio
Standard deviation
0.076
0.076
0.077
0.077
0.079
0.081
0.082
0.086
0.086
0.075
0.076
0.076
0.077
0.078
0.078
0.083
0.087
0.089
Measured
average
logarithmic
decrement
m
0.075
0.076
0.076
0.077
0.078
0.079
0.082
0.086
0.087
1.06
0.06
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.075
0.075
0.075
0.075
0.075
0.075
0.075
0.075
0.075
1.00
1.01
1.01
1.03
1.04
1.05
1.09
1.15
1.16
9-22
for beam 11
Removal
load
(kN)
Average
residual
crack
width
(mm)
0
5
10
15
20
25
0
0.005
0.009
0.010
0.022
0.075

for
for
for
hammer
hammer
hammer
0.057
0.058
0.059
0.061
0.063
0.064
0.055
0.057
0.057
0.060
0.061
0.063
0.057
0.058
0.058
0.059
0.062
0.065
Mean ratio
Standard deviation
Measured
average
logarithmic
decrement
m
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.056
0.058
0.058
0.060
0.062
0.064
0.075
0.075
0.075
0.075
0.076
0.078
0.75
0.77
0.77
0.80
0.82
0.82
0.79
0.03
for span 1 of beam 16
Measured
average
logarithmic
decrement
m
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.073
0.072
0.079
0.075
0.082
0.081
0.078
0.078
0.075
0.075
0.076
0.076
1.09
1.07
1.03
1.03
0.077
0.076
0.074
0.080
0.075
0.083
0.081
0.094
0.076
0.075
0.072
0.087
0.085
0.076
0.077
0.089
0.077
0.075
0.074
0.083
0.077
0.079
0.079
0.094
0.076
0.076
0.076
0.076
0.076
0.076
0.076
0.076
1.01
0.99
0.97
1.09
1.01
1.03
1.04
1.23
0.086
0.076
0.085
0.077
1.10
Removal
load
(kN)
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
0
30
40
50
0
0.010
0.019
0.022
0.084
0.085
0.083
0.083
0.087
0.086
0.073
0.077
60
70
80
90
100
110
120
140
0.023
0.026
0.024
0.029
0.030
0.035
0.034
0.037
0.078
0.075
0.076
0.082
0.072
0.077
0.080
0.098
160
0.059
0.093
Mean ratio
Standard deviation
1.05
0.07
9-23
Removal
load
(kN)
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
0
30
40
50
60
70
80
90
100
110
120
140
160
0
0.101
0.018
0.092
0.018
0.089
0.023
0.074
0.023
0.065
0.021
0.072
0.023
0.086
0.026
0.090
0.03
0.097
0.032
0.090
0.040
0.075
0.047
0.082
0.053
0.095
Mean ratio
Standard deviation
0.075
0.070
0.078
0.077
0.075
0.084
0.079
0.092
0.078
0.096
0.073
0.101
0.084
0.075
0.081
0.063
0.082
0.080
0.069
0.078
0.075
0.077
0.078
0.077
0.108
0.104
Measured
average
logarithmic
decrement
m
0.084
0.081
0.077
0.078
0.073
0.075
0.081
0.086
0.084
0.088
0.075
0.097
0.094
1.09
0.09
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.075
0.076
0.076
0.076
0.076
0.076
0.076
0.076
0.076
0.076
0.076
0.077
0.077
1.12
1.07
1.01
1.02
0.97
0.99
1.07
1.13
1.11
1.15
0.98
1.26
1.23
Removal
load
(kN)
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
0
20
45
70
95
120
145
170
195
220
245
270
295
0
0.081
0.020
0.080
0.028
0.076
0.031
0.076
0.031
0.098
0.028
0.075
0.030
0.085
0.028
0.076
0.039
0.105
0.036
0.097
0.036
0.080
0.044
0.087
0.037
0.084
Mean ratio
Standard deviation
0.081
0.096
0.083
0.080
0.081
0.092
0.093
0.082
0.101
0.083
0.078
0.081
0.094
0.067
0.077
0.070
0.109
0.063
0.081
0.070
0.080
0.087
0.077
0.105
0.083
0.089
Measured
average
logarithmic
decrement
m
0.076
0.084
0.076
0.088
0.081
0.083
0.083
0.079
0.098
0.086
0.088
0.084
0.089
1.11
0.07
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.075
0.076
0.076
0.076
0.076
0.076
0.076
0.076
0.076
0.076
0.076
0.077
0.076
1.02
1.10
1.00
1.16
1.06
1.09
1.09
1.04
1.28
1.12
1.15
1.09
1.17
9-24
Removal
load
(kN)
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
0
20
45
70
95
120
145
170
195
220
245
270
295
0
0.071
0.019
0.076
0.031
0.081
0.032
0.078
0.038
0.081
0.032
0.095
0.038
0.075
0.038
0.077
0.044
0.075
0.050
0.079
0.034
0.079
0.057
0.085
0.056
0.085
Mean ratio
Standard deviation
0.079
0.096
0.083
0.084
0.081
0.081
0.079
0.071
0.080
0.079
0.075
0.080
0.086
0.086
0.089
0.077
0.081
0.083
0.080
0.080
0.077
0.078
0.087
0.079
0.083
0.084
Measured
average
logarithmic
decrement
m
0.079
0.087
0.080
0.081
0.082
0.085
0.078
0.075
0.078
0.082
0.078
0.083
0.085
1.06
0.05
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.075
0.076
0.076
0.076
0.076
0.076
0.076
0.076
0.077
0.077
0.076
0.077
0.077
1.05
1.15
1.06
1.06
1.07
1.12
1.02
0.98
1.02
1.07
1.02
1.07
1.10
Removal
load
(kN)
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
0
5
15
20
50
80
110
140
170
200
230
260
0
0.066
0
0.073
0
0.073
0.018
0.084
0.028
0.096
0.034
0.105
0.03
0.085
0.026
0.084
0.044
0.090
0.043
0.077
0.038
0.098
0.05
0.074
Mean ratio
Standard deviation
0.072
0.078
0.076
0.084
0.076
0.108
0.073
0.103
0.107
0.116
0.105
0.089
0.075
0.071
0.074
0.089
0.077
0.080
0.107
0.069
0.105
0.104
0.089
0.084
Measured
average
logarithmic
decrement
m
0.071
0.074
0.074
0.086
0.083
0.098
0.088
0.085
0.101
0.099
0.097
0.082
1.14
0.13
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.075
0.075
0.075
0.076
0.076
0.076
0.076
0.076
0.077
0.077
0.076
0.077
0.95
0.99
0.99
1.13
1.09
1.28
1.16
1.12
1.31
1.29
1.27
1.07
9-25
Removal
load
(kN)
Average
residual
crack
width
(mm)
0
5
15
20
50
80
110
140
170
200
230
260
0
0.102
0
0.079
0
0.076
0.047
0.075
0.031
0.081
0.028
0.075
0.030
0.079
0.036
0.073
0.040
0.073
0.049
0.092
0.050
0.083
0.054
0.092
Mean ratio
Standard deviation
9.4.3

for
for
for
hammer
hammer
hammer
0.078
0.073
0.077
0.073
0.083
0.075
0.072
0.083
0.102
0.091
0.083
0.088
0.074
0.077
0.074
0.078
0.078
0.073
0.090
0.075
0.075
0.083
0.092
0.081
Measured
average
logarithmic
decrement
m
0.085
0.076
0.076
0.075
0.081
0.074
0.080
0.077
0.083
0.089
0.086
0.087
1.06
0.06
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.075
0.075
0.075
0.077
0.076
0.076
0.076
0.076
0.076
0.077
0.077
0.077
1.14
1.02
1.01
0.98
1.06
0.98
1.06
1.01
1.09
1.16
1.12
1.13
Verification of viscous damping characteristics
It has been mentioned that the damping capacity (damping ratio, ) of reinforced
concrete beams can be obtained from logarithmic decrement of damping using either
Eq (4.47) or (4.48). But the expression for the logarithmic decrement of damping
given by Eq. (4.48) assumes that the resulting structural damping mechanism has the
characteristics of viscous damping, i.e. that the damping force resisting the motion at
any time is proportional to the velocity of vibration.
This assumption can be checked by plotting the values of ln (Ao/An) against
n, the number of oscillations, where Ao and An are amplitudes of vibration at the
start and at n-cycles after the start, respectively. The plotted values will lie along a
straight line if the damping is proportional to the velocity, and along a curve, if it is
not. When the former is the case the damping is independent of, and if not it is
dependent on, the amplitude of response (Buchholdt, 1997).
9-26
Not all the details for the first-stage beams were available to the author.
However, ln (Ao/An) values are plotted against the number of oscillations, n, for the
second-stage beams as shown in Figure 9.9. These include values for the three
hammer excitation positions for each of the beams totalling 297 data points. The
lines of best fit and the corresponding values for the coefficients of determination
(R2 values) for each beam are shown separately in this figure. The data points
belonging to different beams are shown separately as well.
Beam 18
R2 = 0.9877
Beam 17
R2 = 0.9915
ln (Ao/An)
Beam 16
R2 = 0.9888
Beam 7
R2 = 0.9973
1
Beam 5
R2 = 0.9849
0
0
10
20
30
40
50
60
Number of oscillations, n
Figure 9.9 Ln (Ao/An) plotted against n for second-stage beams
It can be seen from Figure 9.9 that, for each beam, the plotted points mostly
lie along a straight line (best-fit line). Values of the coefficients of determination
(R2 values) very close to unity for the best-fit line for each beam (varying from
0.9849 to 0.9973) further confirms the close proximity of the points to the best-fit
lines. This verifies the fact that the damping characteristics exhibited by reinforced
concrete beams are predominantly viscous as often assumed.
9-27
9.5 Damping Behaviour of Reinforced Concrete Beams

As discussed in Chapter 3, many factors may influence the damping characteristics
of reinforced concrete beams. In the present work, a total of 14 reinforced concrete
beams with various parameters of various levels were tested. The effects of the
various main variables on the damping characteristics are presented in the following
sub-sections.
9.5.1
Effect of steel ratio
In order to see the effect of steel ratio on the damping characteristics of the
reinforced concrete beams, the measured logarithmic decrement values at zero load
(no cracks) are plotted against the steel ratios for all 14 reinforced concrete beams
tested. This is presented in Figure 9.10. Beams of same span lengths and their
respective lines of best fit are shown separately in this figure. Note that there was
only 1 beam with span length of 8.0 m for which no best-fit line was drawn.
.15
Logarithmic decrements,
.12
.09
Continuous beams
(Beams 16, 17 & 18)
.06
Beam of 8.0 m span

(Beam 11)
Beams of 6.7 m span
.03
(Beams 8,9 & 10)

Beams of 5.5 m span
0.00
0.00
(Beams 1 through 7)
.01
.02
.03
.04
.05
Steel ratios,
Figure 9.10 Effect of steel ratio on the damping characteristics of reinforced

concrete beams
9-28
It is found that while the logarithmic decrement values increased slightly

with increase in steel ratios for the simply supported beams, they decreased with
increasing steel ratios for the individual spans of continuous beams. Thus, the
variation of damping values with steel ratios for reinforced concrete beams are
inconclusive. Hence, there is no conclusive effect of steel ratio on the damping
characteristics of reinforced concrete beams.
9.5.2
Effect of compressive strength of concrete
To see the effect of compressive strengths of concrete on the damping characteristics

of reinforced concrete beams, the measured logarithmic decrement values for the
uncracked beams (at zero load) are plotted against the compressive strengths of
concrete for all 14 beams as shown in Figure 9.11. Once again, beams of same span
lengths are grouped together and shown separately in this figure. The lines of best
fit for beams of each span length variety are shown as well. As there was only one
beam of span 8.0 m (one data point), no best-fit line was drawn for that beam.
.15
.12
.09
Continuous beams
(Beams 16, 17 & 18)
.06
Beam of 8.0 m span

(Beam 11)
Beams of 6.7 m span
.03
(Beams 8,9 & 10)

Beams of 5.5 m span
(Beams 1 through 7)
0.00
0
10
20
30
40
50
Compressive strength of concrete, f'c in MPa
Figure 9.11 Effect of compressive strength of concrete on the damping

characteristics of reinforced concrete beams
9-29
It is found from Figure 9.11 that the variation of logarithmic decrement

values with changes in the compressive strengths of concrete is rather mixed and
varied. While the damping values decrease with increases in compressive strength
of concrete values for the simply supported beams, the reverse is the case for the
individual spans of continuous beam.
9.5.3
Effect of span length
Three different lengths have been used for the reinforced concrete test beams in this
study: 5.5 m, 6.7 m and 8.0 m (see Table 6.1). In addition, each individual span of
each two-equal-span continuous beam was 6.0 m long (total beam length 12.0 m, see
Table 6.2).
Logarithmic decrement values of the reinforced concrete beams are plotted

against their respective span lengths as shown in Figure 9.12 to see the latters effect
on damping values.
It has been shown that the effects of steel ratios and
compressive strengths of concrete on the damping characteristics of reinforced

concrete beams are inconclusive. Also, as can be seen from Figures 9.10 and 9.11,
the effects of steel ratios and compressive strengths of concrete are completely
opposing in nature and thereby will cancel each others effect. Hence, beams having
similar steel ratios and concrete strengths are grouped together and shown separately
in Figure 9.12. The lines of best fit for such similar beams are also shown in this
figure.
As can be seen from Figure 9.12, while for most of the beams the
logarithmic decrement values decreased with increasing span lengths, they increased
with increasing span lengths for certain beams. Also, as most of the beams tested
were of 5.5 m or 6.0 m span lengths, there were not enough data to make any
conclusive statement about the effect of span lengths on the damping characteristics
of reinforced concrete beams. Note that only one beam was of 8.0 m span and three
beams of 6.7 m span out of a total of 14 beams.
9-30
.12
.10
.08
Beams with higher value
.06
Beams with = 0.01163
.04
.02
0.00
0
10
12
Span lengths in m
Figure 9.12 Effect of span length on the damping characteristics of reinforced

concrete beams
9.5.4
Effect of cracking
One of the main concerns of this study was to investigate the effect of cracking on
the damping characteristics of reinforced concrete beams. When cracking occurred
in the beams, the damping capacities increased with increased crack widths. This
has been discussed in detail in Section 9.3.
The damping prediction formula
developed and verified in Section 9.4 had cracking as its very basis of development.
Thus, the effect of cracking had already been well explained and discussed in
detail in Sections 9.3 and 9.4.
9.6
9-31
Summary
A total of 11 reinforced concrete simply supported and 3 continuous beams were

tested to investigate the effect of cracking on the damping characteristics of
reinforced concrete beams. Test data from 4 of these simply supported beams were
used to develop a simple yet accurate formula (Eq. (9.1)) for predicting damping in
terms of logarithmic decrements in simply supported reinforced concrete beams.
The proposed damping formula has been verified using the test results from all 14
beams including the continuous beams, thereby proving the applicability of the
formula to continuous beams.
It has been shown that, using the proposed damping formula and other
formulae developed in earlier chapters, it is possible to predict damping in
reinforced concrete beams from the general beam parameters.
Often assumed
viscous damping characteristics prevailing in reinforced concrete beams has been

verified from the test results as well.
Since, neither any damping prediction formula nor any crack related
damping data are available for reinforced concrete beams in the published literature,
the proposed damping formula can serve as a new breakthrough in this field.
Finally, the effects of different parameters on the damping characteristics of
reinforced concrete beams have been discussed in some detail. As it turned out, no
conclusive statement about the effects of steel ratios, compressive strengths of
concrete, or the span lengths on the damping characteristics of reinforced concrete
beams could be made.
Overall, their effects were varied and ignored in the
formulation. In the end, it was cracking which had definitive effects on the damping
of reinforced concrete beams. Hence, the proposed damping formula has been
developed on that basis.
Chapter 10 : Damping in Partially Prestressed Concrete Beams
10-1
CHAPTER 10
DAMPING IN PARTIALLY PRESTRESSED
CONCRETE BEAMS
10.1
General Remarks
Partially prestressed concrete beams are more susceptible to cracking than

reinforced concrete beams.
As such, the effect of cracking on their damping
characteristics would be more pronounced. Use of partially prestressed beams is on

the increase and no simple and accurate formula to evaluate damping in such
structures is available in the literature.
A total of twelve partially prestressed concrete beams were tested in this
study to investigate the effect of cracking on the damping characteristics of partially
prestressed concrete beams. The approach and measurement methods used for
testing the reinforced concrete beams were also employed for the partially
A simple yet accurate formula for the prediction of damping in terms of
logarithmic decrements in partially prestressed concrete beams, based on the test
results of a number of selected beams, is developed in this chapter. The effects of
different parameters on the damping characteristics of partially prestressed concrete
beams are also discussed.
10.2
10-2
Instrumentation and Data Recording
All the equipment and instruments used in carrying out the experimental work for
the partially prestressed concrete beams were the same as were used for the
reinforced concrete beams (see Section 9.2).
As for the reinforced concrete beams, the free decay method with hammer
excitation was used to measure the vibration response signals of the partially
The details of the impact hammer and the
accelerometers used, as well as the mounting of the accelerometer on the beams, are
described in Section 9.2.2.
As all the partially prestressed concrete beams tested were first-stage beams,
a 2630 Personal Fourier Analyzer (see Figure 9.3) was used to measure and record
the vibration signals.
The technical specifications of the analyzer have been
described in Section 9.2.2. The locations of the hammer excitation points on the top
of the beam for the partially prestressed concrete beams have also been shown in
Figure 9.5 (see Section 9.2.2).
10.3
Presentation of Test Results
The damping characteristics of 12 full-size partially prestressed concrete simply

supported beams with various residual crack widths present were investigated in this
study. The major objectives were (a) to investigate the effect of cracking on the
damping behaviour, and (b) to develop a simple yet accurate formula for the
determination of damping ratios for partially prestressed concrete beams (with or
without cracking).
As with the reinforced concrete beams, static loads were applied at two
points as shown in Figure 6.21 to produce cracking in the partially prestressed
concrete beams. While instantaneous crack widths, concrete surface strains and
deflections were measured under the load at each level of loading, the residual crack
10-3
widths and/or concrete surface strains and the corresponding free decay vibrations
were measured after the removal of the load. Thus, average residual crack width
was the only parameter that can be directly related to the logarithmic decrement of
damping values.
The effect of cracking on the damping characteristics of partially prestressed
concrete beams can be shown by presenting the test results under two different
categories as shown in the following sub-sections. These are for (a) over reinforced
beams, and (b) under reinforced beams. Typical vibration response curves for the
partially prestressed concrete beams at various levels of cracking (as obtained from
the analyzer) are presented in Appendix G.
10.3.1 Over reinforced beams

As explained in Section 9.3.1, for over reinforced partially prestressed concrete
beams also, compression failure takes place when ultimate load is reached. The
tensile steel remained in the elastic range for the whole range of load application and
the resulting residual crack widths were small upon removal of the load. The
variation in logarithmic decrement values with changes in residual crack widths was
also small. The variation in logarithmic decrement values were mostly less than
15% for individual beams.
To show the effect of cracking on the damping characteristics of over
reinforced partially prestressed concrete beams, logarithmic decrement values for
such beams are plotted against their corresponding average residual crack widths.
These are presented in Figures H.1 through H.4 given in Appendix H. The near
horizontal features of, or the small inclinations for the lines of best fits for the data
points in Figures H.1 through H.4 prove the point that the variations in logarithmic
decrement values with cracking were rather small.
10-4
10.3.2 Under reinforced beams

For under reinforced partially prestressed concrete beams, tensile failure of the
beams took place due to the yielding of the tensile reinforcements.
As the
reinforcing steel did not return to its original form or shape, it resulted in relatively
larger residual crack widths after the removal of the loads.
The logarithmic
decrement values increased significantly with increase in residual crack widths for
such beams. Similar behaviour was observed for the beams, which had a balanced
failure.
Once again, the logarithmic decrement values are plotted against their
corresponding residual crack widths for the balanced and under reinforced partially
prestressed concrete beams to demonstrate the above damping behaviour. These are
presented in Figures H.5 through H.12 given in Appendix H.
10.4
Damping Formula for Partially Prestressed Concrete Beams
As for the reinforced concrete beams, the development of a computational procedure

for evaluating damping in partially prestressed concrete beams was also a major
objective of this research. Accordingly, a simple yet accurate formula for predicting
damping in partially prestressed concrete beams is developed and verified from the
laboratory test results in this section.
10.4.1 Development of the damping formula

Damping in partially prestressed concrete simply supported beams may be
influenced by many factors. In the present work, a total of 12 partially prestressed
concrete beams with varying steel ratios, span lengths, compressive strengths of
concrete and degrees of prestressing were tested. As will be shown in Section 10.5,
the experimental results show that the effects of steel ratios, span lengths,
compressive strengths of concrete and degrees of prestressing on the damping values
10-5
measured as logarithmic decrements are either negligible or inconclusive. The main

factor which influences the damping values is found to be the residual crack width
as shown in Section 10.3.
As such, the logarithmic decrement values and the average residual crack
widths of selected partially prestressed concrete beams are considered to carry out a
regression analysis (Nie et al., 1975), which resulted in the damping prediction
formula for partially prestressed concrete simply supported beams. The relevant
data for beams 19, 22, 26 and 28 are used in this regression analysis. The reasons
for the selection of these particular beams are that they incorporate the four degrees
of prestressing (0.25 for beam 19, 0.50 for beam 28, 0.75 for beam 26 and 1.00 for
beam 22) and the three different span lengths (5.5 m for beams 19 and 22, 6.8 m for
Removal
Load
(kN)
0
5
10
15
20
25
30
35
40
Average
residual
crack width
(mm)
0.050
0.075
0.092
0.125
0.217
0.233
0.283
0.300
0.433
point A
point B
point C
0.077
0.071
0.075
0.079
0.076
0.077
0.083
0.084
0.079
0.085
0.085
0.081
0.098
0.101
0.097
0.096
0.105
0.100
0.101
0.110
0.107
0.122
0.124
0.118
0.152
0.148
0.149
Average
logarithmic
decrement,
0.074
0.077
0.082
0.084
0.099
0.100
0.106
0.121
0.150

Removal
Load
(kN)
0
9
12
15
18
21
24
Average
residual
crack width
(mm)
0
0.010
0.015
0.015
0.017
0.020
0.025
point A
point B
point C
0.072
0.073
0.073
0.074
0.073
0.074
0.073
0.074
0.074
0.075
0.076
0.075
0.079
0.081
0.077
0.077
0.082
0.080
0.082
0.083
0.083
Average
logarithmic
decrement,
0.073
0.074
0.074
0.075
0.079
0.080
0.083
10-6

Removal
Load
(kN)
0
10
20
30
40
45
Average
residual
crack width
(mm)
0
0.121
0.268
0.894
0.964
1.237
point A
point B
point C
0.082
0.084
0.085
0.085
0.085
0.085
0.085
0.087
0.088
0.087
0.087
0.089
0.091
0.094
0.092
0.141
0.138
0.144
Average
logarithmic
decrement,
0.084
0.085
0.087
0.088
0.092
0.141

Removal
Load
(kN)
0
3
6
10
13
14
17
Average
residual
crack width
(mm)
0
0.049
0.056
0.057
0.064
0.133
0.249
point A
point B
point C
0.047
0.049
0.050
0.047
0.049
0.051
0.048
0.051
0.051
0.049
0.052
0.053
0.051
0.053
0.055
0.052
0.053
0.055
0.055
0.056
0.058
Average
logarithmic
decrement,
0.049
0.049
0.050
0.051
0.053
0.053
0.056
beam 26 and 8.0 m for beam 28) used. The damping data for these four beams,
which were used in the regression analysis, are presented in Tables 10.1 through
10.4.
The damping prediction formula for simply supported partially prestressed
concrete beams resulting from the above-mentioned regression analysis is:
= 0.070 x 10 0.220 w r
(10.1)
where is the damping in terms of logarithmic decrements and wr is the average

residual crack widths in mm.
The average residual crack widths (wr) in Eq. (10.1) are obtained using either
Eq. (8.4) or Eq. (8.6). On the other hand, using the logarithmic decrement () values
10-7
thus obtained, the damping ratios () can be computed from Eq. (4.47) or (4.48),
depending on the accuracy required. Thus, the damping capacity (damping ratio) of
partially prestressed concrete simply supported beams can be obtained from the
general beam parameters making use of the above equations.
10.4.2 Comparison with test results

No damping data for partially prestressed concrete beams were available in the
published literature.
Hence, to verify the accuracy of the proposed damping
formula, the logarithmic decrement values predicted by Eq. (10.1) are compared
with the experimental values for all 12 partially prestressed concrete beams tested in
this study.
Eq. (10.1) is applied to different levels of residual crack widths
corresponding to different removal loads to determine the theoretical logarithmic
decrement values for each beam.
Note that both residual crack widths and
logarithmic decrement values were measured at each level of loading after the
removal of the load. These theoretical values together with the measured values
corresponding to the various removal loads are listed for each beam in Tables 10.5
through 10.16. The ratios of measured to calculated logarithmic decrement values
were also computed and are listed in the same tables. It can be seen that the mean
values of this ratios are very close to unity for most of the beams. The standard
deviations were within 20 percent for all the beams except with only one beam
(beam 27).
For the partially prestressed concrete simply supported beams, the variations
of the means of the ratios of measured to calculated logarithmic decrement values
were within + 30% as they varied from 0.70 to 1.28 with standard deviations varying
from 2 to 20 percent (see Tables 10.5 through 10.12 and 10.14 through 10.16). The
only exception was beam 27 for which, the mean ratio and the standard deviation
were 1.58 and 26 percent, respectively (see Table 10.13).
10-8
for beam 19
Removal
load
(kN)
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
Measured
average
logarithmic
decrement
m
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.050
0.077
0.071
0.075
0.074
0.072
1.03
0.075
0.079
0.076
0.077
0.077
0.073
1.06
10
0.092
0.083
0.084
0.079
0.082
0.073
1.12
15
0.125
0.085
0.085
0.081
0.084
0.075
1.13
20
0.217
0.098
0.101
0.097
0.099
0.078
1.27
25
0.233
0.096
0.105
0.100
0.100
0.079
1.27
30
0.283
0.101
0.110
0.107
0.106
0.081
1.31
35
0.300
0.122
0.124
0.118
0.121
0.081
1.49
40
0.433
0.152
0.148
0.149
0.150
0.087
1.72
Mean ratio
1.27
Standard deviation
0.22
for beam 20
Measured
average
logarithmic
decrement
m
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.077
0.076
0.070
1.09
0.076
0.077
0.076
0.070
1.09
0.077
0.077
0.078
0.077
0.071
1.09
0.042
0.078
0.078
0.078
0.078
0.072
1.09
30
0.047
0.079
0.078
0.079
0.079
0.072
1.10
39
0.048
0.082
0.080
0.081
0.081
0.072
1.13
40
0.051
0.085
0.086
0.088
0.086
0.072
1.20
Removal
load
(kN)
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
0.074
0.076
10
0.074
20
0.018
25
Mean ratio
1.11
Standard deviation
0.04
10-9
for beam 21
Removal
load
(kN)
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
Measured
average
logarithmic
decrement
m
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.080
0.080
0.078
0.079
0.070
1.13
0.080
0.081
0.081
0.081
0.070
1.16
10
0.082
0.082
0.082
0.082
0.070
1.17
20
0.001
0.082
0.084
0.085
0.084
0.070
1.20
25
0.010
0.085
0.086
0.086
0.086
0.070
1.22
30
0.014
0.088
0.087
0.090
0.088
0.070
1.25
40
0.020
0.096
0.089
0.095
0.093
0.071
1.32
50
0.021
0.096
0.095
0.097
0.096
0.071
1.36
Mean ratio
1.23
Standard deviation
0.08
for beam 22
Measured
average
logarithmic
decrement
m
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.073
0.073
0.070
1.04
0.073
0.074
0.074
0.070
1.05
0.073
0.074
0.074
0.074
0.071
1.05
0.015
0.075
0.076
0.075
0.075
0.071
1.06
18
0.017
0.079
0.081
0.077
0.079
0.071
1.12
21
0.020
0.077
0.082
0.080
0.080
0.071
1.13
24
0.025
0.082
0.083
0.083
0.083
0.071
1.17
Removal
load
(kN)
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
0.072
0.073
0.010
0.074
12
0.015
15
Mean ratio
1.09
Standard deviation
0.05
10-10
for beam 23
Removal
load
(kN)
0
10
15
19
Average
residual
crack
width
(mm)
0
0.072
0.012
0.089
0.055
0.100
0.069
0.108
Mean ratio
Standard deviation
Table 10.10
Removal
load
(kN)
0
10
20
25
30
35
40
0
10
20
30
35
40
42
0.070
0.087
0.096
0.106
0.070
0.088
0.100
0.108
Measured
average
logarithmic
decrement
m
0.071
0.088
0.099
0.107
1.28
0.20
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.070
0.070
0.072
0.072
1.01
1.25
1.38
1.48
Relation between measured and calculated logarithmic decrement

values for beam 24
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
0
0.054
0
0.055
0.007
0.058
0.011
0.061
0.021
0.067
0.028
0.073
0.033
0.084
Mean ratio
Standard deviation
Table 10.11
Removal
load
(kN)

for
for
for
hammer
hammer
hammer
0.055
0.055
0.057
0.062
0.065
0.070
0.085
0.053
0.056
0.056
0.063
0.067
0.071
0.088
Measured
average
logarithmic
decrement
m
0.054
0.055
0.057
0.062
0.066
0.071
0.086
0.91
0.15
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.070
0.070
0.070
0.070
0.071
0.071
0.071
0.77
0.79
0.81
0.88
0.93
1.00
1.21

values for beam 25
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
0
0.048
0
0.049
0.010
0.050
0.021
0.051
0.037
0.052
0.039
0.053
0.266
0.054
Mean ratio
Standard deviation
0.050
0.051
0.052
0.052
0.053
0.054
0.055
0.050
0.050
0.052
0.053
0.053
0.055
0.055
Measured
average
logarithmic
decrement
m
0.049
0.050
0.051
0.052
0.053
0.054
0.055
0.72
0.02
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.070
0.070
0.070
0.071
0.071
0.071
0.080
0.70
0.71
0.72
0.74
0.74
0.76
0.69
Table 10.12
Removal
load
(kN)
0
10
20
30
40
45
Removal
load
(kN)
0
3
6
9
12
0
3
6
10
13
14
17

for
for
for
hammer
hammer
hammer
0
0.082
0.121
0.085
0.268
0.085
0.894
0.087
0.964
0.091
1.237
0.141
Mean ratio
Standard deviation
0.084
0.085
0.087
0.087
0.094
0.138
0.085
0.085
0.088
0.089
0.092
0.144
Measured
average
logarithmic
decrement
m
0.084
0.085
0.087
0.088
0.092
0.141
1.02
0.17
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.070
0.074
0.080
0.110
0.114
0.131
1.20
1.14
1.09
0.80
0.81
1.08

values for beam 27
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
0
0.086
0.037
0.101
0.039
0.115
0.049
0.121
0.102
0.140
Mean ratio
Standard deviation
Table 10.14
Removal
load
(kN)

values for beam 26
Average
residual
crack
width
(mm)
Table 10.13
10-11
0.086
0.097
0.120
0.122
0.137
0.087
0.100
0.121
0.122
0.141
Measured
average
logarithmic
decrement
m
0.086
0.099
0.119
0.122
0.139
1.58
0.26
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.070
0.071
0.071
0.072
0.074
1.23
1.39
1.67
1.70
1.89

values for beam 28
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
0
0.047
0.049
0.047
0.056
0.048
0.057
0.049
0.064
0.051
0.133
0.052
0.249
0.055
Mean ratio
Standard deviation
0.049
0.049
0.051
0.052
0.053
0.053
0.056
0.050
0.051
0.051
0.053
0.055
0.055
0.058
Measured
average
logarithmic
decrement
m
0.049
0.049
0.050
0.051
0.053
0.053
0.056
0.70
0.02
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.070
0.072
0.072
0.072
0.072
0.075
0.079
0.70
0.68
0.69
0.71
0.73
0.71
0.71
Table 10.15
Removal
load
(kN)
0
5
10
15
20
25
30
32
Removal
load
(kN)
0
10
20
25
30
35

values for beam 29
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
0
0.053
0
0.057
0
0.057
0.030
0.058
0.036
0.065
0.047
0.069
0.107
0.083
0.173
0.092
Mean ratio
Standard deviation
Table 10.16
10-12
0.056
0.057
0.060
0.061
0.067
0.071
0.082
0.094
0.054
0.055
0.058
0.060
0.067
0.070
0.084
0.090
Measured
average
logarithmic
decrement
m
0.054
0.056
0.058
0.060
0.066
0.070
0.083
0.092
0.93
0.16
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.070
0.070
0.070
0.071
0.071
0.072
0.074
0.076
0.77
0.80
0.83
0.84
0.93
0.98
1.12
1.20

values for beam 30
Average
residual
crack
width
(mm)

for
for
for
hammer
hammer
hammer
0
0.049
0
0.049
0.004
0.051
0.011
0.052
0.018
0.056
0.052
0.058
Mean ratio
Standard deviation
0.050
0.050
0.051
0.053
0.055
0.057
0.050
0.051
0.051
0.054
0.056
0.058
Measured
average
logarithmic
decrement
m
0.050
0.050
0.051
0.053
0.056
0.058
0.75
0.04
Calculated
logarithmic
decrement
c
The
ratio
(m/
c)
0.070
0.070
0.070
0.070
0.071
0.072
0.71
0.71
0.73
0.75
0.79
0.81
To further highlight the above findings, a plot of the measured logarithmic

decrement values against the calculated values using Eq. (10.1) for the 12 partially
prestressed concrete beams is presented in Figure 10.1.
It can be seen from
Figure 10.1, that an excellent correlation between measured and calculated values
exists as all but 9 of the 81 correlation points lie well within + 30% limits. Out of
these 9 points, 4 points belonged to beam 27.
10-13
Measured logarithmic decrements, m
.16
- 30% line
.12
.08
+ 30% line
.04
0.00
0.00
.04
.08
.12
.16
Calculated logarithmic decrements, c
Figure 10.1 Measured versus calculated damping values for the partially
prestressed concrete beams
Another important finding was that, at no or low values of residual crack

widths, the measured damping values for partially prestressed concrete beams varied
widely for the different beams of similar geometrical properties. This may be
because of the variations in internal cracking due to the different levels of
prestressing in the beams.
10.5 Damping Behaviour of Partially Prestressed Concrete Beams

The twelve partially prestressed concrete beams tested in this study were of varying
steel ratios, span lengths, compressive strengths of concrete and degrees of
prestressing. They underwent various levels of cracking under varying loads as
well. The effects of these main variables on the damping characteristics of partially
prestressed concrete beams are discussed in the following sub-sections.
10-14
10.5.1 Effect of steel ratio

In order to see the effect of steel ratio on the damping characteristics of partially
prestressed concrete beams, the measured logarithmic decrement values at zero
loads or zero cracking (for uncracked beams) are plotted against the steel ratios for
all 12 beams tested. This is presented in Figure 10.2. Beams of same span lengths
are grouped together and shown separately in the figure. The line of best fit for each
span length group is also shown separately.
.15
.12
.09
.06
Beams of 8.0 m span

(Beams 27 through 30)
Beams of 6.8 m span
.03

Beams of 5.5 m span
0.00
0.000
.002
.004
.006
.008
.010
Steel ratios,
Figure 10.2
Effect of steel ratio on the damping characteristics of partially

It can be seen from Figure 10.2 that while the logarithmic decrement values
increased with increasing steel ratios for beams having 8.0 m long spans, they
decreased with increasing steel ratios for other beams.
Also, the variations in
logarithmic decrement values with changes in steel ratios were different for different
beams. Overall, there may be a general decrease in damping values with increasing
steel ratios but the effect would be less significant.
10-15
10.5.2 Effect of compressive strength of concrete

To see the effect of compressive strengths of concrete on the damping characteristics
of partially prestressed concrete beams, the measured logarithmic decrement values
for the uncracked beams (at zero load) are plotted against the compressive strengths
of concrete for all 12 beams as shown in Figure 10.3. Once again, beams of same
span lengths are grouped together and shown separately in this figure. The lines of
best fit for beams of each span length variety are also shown separately.
.20
.16
.12
.08
Beams of 8.0 m span

Beams of 6.8 m span
.04

Beams of 5.5 m span
0.00

0
10
20
30
40
50
60
Compressive strength of concrete, fc in MPa
Figure 10.3

It is found from Figure 10.3 that the variations of logarithmic decrement

values with changes in the compressive strengths of concrete are rather mixed and
varied. While the damping values increased slightly with increases in compressive
strength of concrete values for the beams having spans of 5.5 m, the reverse was the
case for the other beams. There was a sharp decrease in damping values with the
slightest increase in compressive strengths of concrete for beams having 6.8 m long
10-16
spans. Also the variation in damping values is strikingly different for different
beams.
10.5.3 Effect of span length

The partially prestressed concrete beams tested were of three different span lengths.
These were 5.0 m, 6.8 m and 8.0 m (see Table 6.3). Logarithmic decrement values
of the partially prestressed concrete beams are plotted against their respective span
lengths as shown in Figure 10.4 to see the latters effect on damping values.
.20
.16
.12
.08
.04

0.00

0.0
2.0
4.0
6.0
8.0
10.0
Span lengths in m
Figure 10.4
Effect of span length on the damping characteristics of partially

It has already been shown that the effects of steel ratios and compressive
strengths of concrete on the damping characteristics of partially prestressed concrete
beams are inconclusive. Hence, beams having similar steel ratios and concrete
10-17
strengths are grouped together and shown separately in Figure 10.4. The lines of
best fit for such similar beams are also shown in this figure.
As can be seen from Figure 10.4, while for the most of the beams the
logarithmic decrement values decreased with increasing span lengths, they increased
with increasing span lengths for beams with higher steel ratios ( = 0.00737). Also,
as the variations in damping values were so dissimilar for different beams, it seems
there were inadequate data to make any conclusive statement about the effect of
span lengths on the damping characteristics of partially prestressed concrete beams.
10.5.4 Effect of degree of prestressing
The partially prestressed concrete beams tested had four different degrees of
prestressing, namely 0.25, 0.50, 0.75 and 1.00 (see Table 6.3). To see the effect of
degrees of prestressing on the damping characteristics of partially prestressed
concrete beams, the measured logarithmic decrement values at zero load or zero
cracking i.e. for uncracked beams are plotted against the degrees of prestressing for
all 12 beams tested. This is presented in Figure 10.5.
Once again, beams of same span lengths are grouped together and shown
separately in this figure. The lines of best fit for beams belonging to each span
length category are drawn separately and shown in the figure as well.
As can be seen from Figure 10.5, there were very slight changes in damping
values with degrees of prestressing for most of the beams as exhibited by the best-fit
lines for beams of 5.5 m and 6.8 m spans which are nearly horizontal. On the other
hand, there was a decrease in damping values with increasing degrees of prestressing
for beams with 8.0 m spans. Overall there is no significant effect of degrees of
prestressing on the damping characteristics of partially prestressed concrete beams.
10-18
.14
.12
.10
.08
.06
Beams of 8.0 m span
.04
Beams of 6.8 m span

.02
Beams of 5.5 m span

0.00
0.00

.25
.50
.75
1.00
1.25
Degrees of prestressing
Figure 10.5
Effect of degrees of prestressing on the damping characteristics of

10.5.5 Effect of cracking
The parameter, which had the most definitive and conclusive effect on the damping
characteristics of partially prestressed concrete beams, was cracking.
This has
already been discussed in detail in Section 10.3. It has also been shown that when
cracking occurred in beams, the damping capacity increased with increased crack
widths. Thus, cracking has quite justifiably been used as the basis of the damping
formula developed and verified in Section 10.4.
The effect of cracking on the damping characteristics of partially prestressed

concrete beams has, thus already been well explained and discussed in detail in
Sections 10.3 and 10.4.
10-19
10.6 Summary
Twelve partially prestressed concrete simply supported beams were tested to
investigate the effect of cracking on the damping characteristics of partially
prestressed concrete beams. Test data from four of these beams were used to
develop a simple yet accurate formula (Eq. (10.1)) for the prediction of damping in
terms of logarithmic decrements in simply supported partially prestressed concrete
beams. Due to the non-availability of test data in the published literature, the
proposed damping formula has been verified from the test results of all 12 beams
tested in this study.
It has been shown that, using the proposed damping formula and other
formulae developed in earlier chapters, it is possible to predict damping in partially
prestressed concrete beams from the general beam parameters. Since, neither any
damping prediction formula nor any crack related damping data are available for
partially prestressed concrete beams in the published literature, the proposed
damping formula can serve as a new breakthrough in this field.
Finally, the effects of different parameters on the damping characteristics of
partially prestressed concrete beams have been discussed in some detail. Based on
the experimental results, no conclusive statement about the effects of steel ratios,
compressive strengths of concrete, the span lengths or the degrees of prestressing on
the damping characteristics of partially prestressed concrete beams could be made.
Overall, their effects were too varied to be considered in the formulation. It was
cracking which was found to have definitive effects on the damping of partially
prestressed concrete beams. Hence, the proposed damping formula has rightly been
developed on that basis.
Chapter 11 : Conclusion
11-1
CHAPTER 11
CONCLUSION
11.1
General Remarks
In this concluding chapter, after giving a summary of the major findings of this
thesis, the relevant conclusions are drawn. These are followed by recommendations
for further research.
11.2
Prediction of Crack Widths and Damping
11.2.1 Crack width formulae

A best-fit equation is proposed for the stabilised average crack spacing in reinforced
and partially prestressed concrete beams. It takes the form,
lcr = 0.6 (c s) + 0.1 (/)
(7.5)
This led to the development of the unified formula for predicting average
instantaneous crack widths at the tension face of both reinforced and partially
prestressed concrete beams. Or,
wcr = (fs /Es) [0.6 (c s) + 0.1 (/)]
(7.6)
11-2
A relationship between the maximum and average crack width has also been
derived with the aid of the experimental results. The average ratio of maximum to
average crack width has been found to be 1.33 for reinforced concrete simply
supported beams, 1.48 for reinforced concrete two-equal-span continuous beams,
and 1.63 for partially prestressed concrete simply supported beams. The overall
average value for this ratio, as obtained from the test results of 17 beams, was 1.46.
A value of 1.5 is recommended for general use. Otherwise the individual values
mentioned above can be used for specific purposes.
The residual crack width is much smaller than instantaneous crack width,
and is too small to measure accurately, especially in the initial stages of loading.
Based on the experimental results, the following formula is proposed for computing
residual crack widths (wr) from measured instantaneous crack width (wcr) and
concrete surface strains (r and i):
wr = wcr r
i
(8.2)
The formula developed for predicting average residual crack widths (wr) for
both reinforced and partially prestressed concrete beams from the general beam
parameters is, therefore:
wr = 0.312 (fs /Es) [0.6 (c s) + 0.1 (/)]
(8.4)
A formula has also been derived to predict average residual crack widths (wr)
from mid-span deflections () of reinforced concrete beams as:
wr = 0.0014
(8.5)
A similar formula developed for partially prestressed concrete beams is:

wr = 0.0084
(8.6)
11-3
11.2.2 Logarithmic decrement of damping

The formula developed for predicting damping in terms of logarithmic decrements
in simply supported reinforced concrete beams is:
= 0.075 x 10 0.205 w r
(9.1)
This formula is applicable to two-equal-span continuous beams as well.

The formula derived for the prediction of damping in terms of logarithmic
decrements in simply supported partially prestressed concrete beams is:
= 0.070 x 10 0.220 w r
11.3
(10.1)
Conclusions
Conclusions from this research can be drawn for the following categories:
(a)
Cracking behaviour of reinforced and partially prestressed concrete beams,
(b)
Damping behaviour of reinforced and partially prestressed concrete beams,
(c)
Accuracy of the proposed computational procedures.
11.3.1 Cracking behaviour

It has been observed from the test results that the initial flexural cracks randomly
form at irregular spacings at low steel stress levels. Flexural cracking usually
initiates at the bottom fibre and propagates in a vertical direction. Also, crack
widths of the reinforced and partially prestressed concrete beams accumulate with
increasing levels of loading.
11-4
Visible crack spacings stabilise at approximately 60-70% of the ultimate

load. No new cracks form after this stabilized condition is reached and crack
spacings remain unchanged. For partially prestressed concrete beams, the cracks are
more uniformly distributed, stabilized crack spacings are smaller and crack widths
are finer. Also, for loads above the cracking load, deflections are smaller for
partially prestressed concrete beams.
Cracks form perpendicular to the direction of the principal tensile stress for
two-equal-span continuous beams.
Initial cracks usually develop first near the
centre of both spans and above the middle support. Cracks are generally wider
above the middle support.
Scattering is the nature of cracking. Variations in crack spacing and crack
width range between + 30%. The same degree of scattering in cracking as shown in
simply supported beams applies for two-equal-span continuous beams. Variations
in cracking and deflection between different spans of a two-equal-span continuous
beam are relatively small.
11.3.2 Damping behaviour

Current published methods for evaluating the damping capacity of reinforced
concrete beams consist of single parameter formulae, which are not suitable for the
design of structures subjected to dynamic or impact loadings. The research work
described herein is to provide appropriate approaches for determining and dealing
with the damping characteristics of reinforced and partially prestressed concrete
beams.
In this investigation, crack width is identified as a basis for the evaluation of
damping capacity of reinforced and partially prestressed concrete beams. According
to this consideration, the relationship between damping capacity and the variables
defining reinforced and partially prestressed concrete beams is developed
empirically.
11-5
The experimental work included measurement of instantaneous and residual

crack widths, instantaneous and residual strains, mid-span deflections, as well as the
vibration response signals of the beams produced by a hammer excitation at free
vibration.
Based on the test results and observations, the following conclusions are
drawn:
(a)
Compressive strength of concrete, steel ratio, span length of beam, and

degree of prestressing may influence the damping characteristics of
reinforced and partially prestressed concrete beams, but they are
inconclusive and less significant than the effect of cracking.
(b)
Initial cracking of the concrete beams may not influence the damping
characteristics, but the effect becomes more pronounced as the crack
widths increase significantly.
(c)
A comparison of the damping capacities of the reinforced and the

partially prestressed concrete beams indicates that there is no significant
difference in the damping characteristics of the two types of beams.
(d)
Damping characteristics of the individual span of two-equal-span

continuous beams is similar to that of simply supported beams. Like
cracking and deflection, variations in damping between different spans of
a two-equal-span continuous beam are also small.
(e)
The damping characteristics of reinforced and partially prestressed

concrete beams are predominantly viscous.
11-6
11.3.3 Accuracy of the proposed procedures

Based on the original test data, formulae for prediction of instantaneous and residual
crack widths as well as damping in reinforced and partially prestressed concrete
beams are developed.
These are summarised in Section 11.2.
To check the
accuracy of each of the formulae developed, comparisons between measured and

computed results are carried out. Wherever applicable, the proposed formulae are
verified from the test results available in the published literature as well.
Incorporating the formulae developed, computational procedures are
proposed for determining the damping capacity (damping ratio) of reinforced and
partially prestressed concrete beams with or without cracking. Despite scattering is
the nature of both cracking and damping, the scattering of cracking and damping
ranged between + 30% using the formulae developed in this study. These indicate
that the proposed procedures are accurate and suitable for design purposes.
11.4
Recommendations for Further Research
The average crack width formula developed for simply supported reinforced
concrete beams has been found to apply equally satisfactorily to the two-equal-span
continuous beams. For the partially prestressed concrete beams, the same cannot be
said as there was no available data to check whether the formula developed for
simply supported beams is applicable to continuous beams or not. Hence, further
research is required to verify the applicability of the crack width formula developed
herein to the individual spans of multi-span continuous beams for both types of
beams.
Similarly, further research is also needed to verify the applicability of the
damping prediction formulae developed for simply supported beams to the
individual spans of multi-span reinforced and partially prestressed concrete
continuous beams.
11-7
The damping predicted by the proposed formulae is predominantly material

damping.
It is found that the maximum damping ratio () exhibited by the
reinforced or partially prestressed concrete beams was less than 3%. Thus, the
material damping in concrete structures is relatively small. Therefore, if it is desired
to increase the damping capacity of concrete structures, it has to be the system
damping resulting from different types of joints and connections. Hence, damping
studies for different types of joints and connections including beam-column, beampanel, and panel-panel, may be considered for further study.
References
R-1
REFERENCES
Abdalla, H. and Kennedy, J.B. (1995), Dynamic Analysis of Prestressed Concrete
Beams with Openings, Journal of Structural Engineering, ASCE, Vol. 121,
No. 7, July, pp. 1058-1068.
Abeles, P.W. (1967), Design of Partially Prestressed Concrete Beams, Journal of
the ACI, Proceedings, Vol. 64, No. 10, October, pp. 669-677.
Abeles, P.W. and Bardhan-Roy, B.K. (1981), Prestressed Concrete Designers
Handbook, Third edition, Cement and Concrete Association, Wexham
Springs, 556p.
ACI Committee 224 (1968), Causes, Mechanism, and Control of Cracking in
Concrete, SP-20, American Concrete Institute, Detroit, 244p.
ACI Committee 224 (1972), Control of Cracking in Concrete Structures, ACI
Journal, Proceedings, Vol. 69, No. 12, December, pp. 717-753.
ACI Committee 224 (1984), Causes, Evaluation and Repair of Cracks in Concrete
Structures, ACI Journal, Proceedings, Vol. 81, No. 3, May-June.
ACI Committee 224 (1993), Control of Cracking in Concrete Structures, ACI
Manual of Concrete Practice, Part 3 : Use of Concrete in Buildings Design, Specifications, and Related Topics, American Concrete Institute,
Detroit, pp. 224R-1 - 224R-43.
ACI Committee 435 (1966), Deflections of Reinforced Concrete Flexural
Members, ACI Journal, Proceedings, Vol. 63, No. 6, June, pp. 637-674.
ACI Committee 435 (1974), Deflections of Concrete Structures, ACI Special
Publication SP-43, American Concrete Institute, Detroit.
ACI Committee 435 (1985), Deflections of Concrete Structures, ACI Special
Publication SP-86, American Concrete Institute, Detroit.
Akashi, T. (1960), On the Measurement of Logarithmic Decrement of Concrete,
General Meeting Reviews, Cement Association of Japan, May, pp. 103-104.
References
R-2
Albandar, F.A-A., and Mills, G.M. (1974), The Prediction of Crack Widths in
Reinforced Concrete Beams, Magazine of Concrete Research (London),
Vol. 26, No. 88, September, pp. 153-160.
American Concrete Institute (ACI) (1995), Building Code Requirements for
Structural Concrete (ACI 318-95) and Commentary (ACI 318R-95),
Farmington Hills, Michigan, 369p.
Argyris, J.H. and Mlejnek, H.-P. (1991), Dynamics of Structures, Elsevier Science
Publishers B.V., Amsterdam, 606p.
Arya, C. (1994), Design of Structural Elements : Concrete, Steelwork, Masonry and
Timber Design to British Standards and Eurocodes, 1st edition, E & FN
Spon, London.
Ashbee, R.A., Heritage, C.A.R. and Jordan, R.W. (1976), The Expanded Hysteresis
Loop Method for Measuring the Damping Properties of Concrete, Magazine
of Concrete Research, Vol. 28, No. 96, September, pp. 148-156.
Askegaard, V. and Langs, H.E. (1986), Correlation between Changes in Dynamic
Properties and Remaining Carrying Capacity: Laboratory Tests with
Reinforced Concrete Beams, Materials and Structures, RILEM, Vol. 19,
No. 109, January-February, pp. 11-19.
Bachmann, H. et al. (1995), Vibration Problems in Structures: Practtical
Guidelines, Birkhaser Verlag Basel, 234p.
Bachmann, H. and Ammann, W. (1987), Vibrations in Structures Induced by Man
and Machines, International Association for Bridge and Structural
Engineering, Zurich.
Base, G.D., Read, J.B., Beeby, A.W. and Taylor, H.P.J. (1966), An Investigation of
the Crack Control Characteristics of Various Types of Bar in Reinforced
Concrete Beams, Research Report, No. 18, Cement and Concrete Association, London, December, Part 1, 44p., and Part 2, 31p.
Bazant, Z.P. and Oh, B.H. (1983a), Deformation of Cracked Net-Reinforced
Concrete Walls, Journal of Structural Engineering, ASCE, Vol. 109, No. 1,
January, pp. 93-108.
Bazant, Z.P. and Oh, B.H. (1983b), Microplane Model for Cracking Analysis of
Concrete Structures, Symposium on the Interaction of Non-Nuclear
Munition with Structures, U.S Air Force Academy, Colorado Springs, May
1983, pp. 49-55.
Bazant, Z.P. and Oh, B.H. (1983c), Spacing of Cracks in Reinforced Concrete,
Journal of Structural Engineering, ASCE, Vol. 109, No. 9, September, pp.
2066-2085.
References
R-3
Beards, C.F. (1983), Structural Vibration Analysis: Modelling, Analysis and

Damping of Vibrating Structures, Ellis Horwood Limited, Chichester, 153p.
Beeby, A.W. (1970), An Investigation of Cracking in Slabs Spanning One Way,
Technical Report TRA 433, Cement and Concrete Association, London,
April, 33p.
Beeby, A.W. (1971), Prediction and Control of Flexural Cracking in Reinforced
Concrete Members, Cracking, Deflection and Ultimate Load of Concrete
Slab Systems, SP-20, American Concrete Institute, Detroit, pp. 55-75.
Bennett, E.W. and Chandrasekhar, C.S. (1971), Calculation of the Width of Cracks
in Class 3 Prestressed Beams, Proceedings, Institution of Civil Engineers,
Vol. 49, July.
Bianchini, A.C., Kesler, C.E. and Lott, J.L. (1968), Cracking of Reinforced
Concrete Under Flexural Load, Causes, Mechanism, and Control of
Cracking in Concrete, SP-20, American Concrete Institute, Detroit, pp. 7385.
Bishop, R.E.D. and Johnson, D.C. (1960), The Mechanics of Vibrations, Cambridge
University Press, London.
Bjuggren, Ulf. (1948), discussion of Bond and Anchorage by Mylrea, T.D., ACI
Journal, Proceedings, Vol. 44, Part 2, December, pp. 552-1 - 552-4.
Blume, J.A. (1970), The Motion and Damping of Buildings Relative to Seismic
Response Spectra, Bulletin, Seismological Society of America, Vol. 60, No.
1, February.
Bock, E. (1942), Behaviour of Concrete and Reinforced Concrete Subjected to
Vibrations Causing Bending, Zeitschrift des Vereins Deutscher Ingenieure,
Vol. 86, No. 9/10, pp. 145-147.
Bornemann, E. (1936), Increasing the Tensile Strength and Avoidance of
Formation of Cracks in Concrete, Proceedings, Second Congress,
International Association for Bridge and Structural Engineering, BerlinMunich, pp. 169-190.
Branson, D.E. (1977), Deformation of Concrete Structures, McGraw Hill, New
York.
British Standards Institution (BS) (1985), BS 8110: 1985 Structural Use of Concrete
Part 2. Code of Practice for Special Circumstances, London.
Broch, J.T. (1980), Mechanical Vibrations and Shock Measurements, Bruel & Kjaer.
Broms, B.B. (1965), Crack Width and Crack Spacing in Reinforced Concrete
Members, ACI Journal, Proceedings, Vol. 62, No. 10, October, pp. 1237-
References
R-4
1256.
Broms, B.B., and Lutz, L.A. (1965), Effects of Arrangement of Reinforcement on
Crack Width and Spacing of Reinforced Concrete Members, ACI Journal,
Proceedings, Vol. 62, No. 11, November, pp. 1395-1410.
Buchholdt, H.A. (1997), Structural Dynamics for Engineers, Thomas Telford
Publications, London, 305 p.
Buttmann, P. (1983), Experimental Determination of Damping Factor for Walls of
Masonry and Reinforced Concrete, Proceedings, 7th International
Conference on Structural Mechanics in Reactor Technology, pp. 507-511.
Carydis, P.G., Psycharis, I.N. and Kuribayashi, E. (1984), Experimental
Determination of Damping in Reinforced Concrete Frame Structures,
Proceedings, Eighth World Conference on Earthquake Engineering, San
Francisco, 21-28 July, 1984, Prentice-Hall Inc., Vol. VI, pp. 379-386.
Caughey, T.K. and OKelly, M.E.J. (1961), Effect of Damping on the Natural
Frequencies of Linear Dynamic Systems, Journal of the Acoustical Society
of America, Vol. 33, pp. 1458-1461.
Chi, M. and Kirstein, A.F. (1958), Flexural Cracks in Reinforced Concrete Beams,
Journal of the ACI, Proceedings, Vol. 54, Title No. 54-48, April, pp. 865878.
Chopra, A.K. (1995), Dynamics of Structures : Theory and Applications to
Earthquake Engineering, Prentice Hall, Englewood Cliffs.
Clark, A.P. (1956), Cracking in Reinforced Concrete Flexural Members, ACI
Journal, Proceedings, Vol. 52, No. 8, April, pp. 851-862.
Clough, R.W. and Penzien, J. (1975), Dynamics of Structures, McGraw Hill, New
York.
Cole, D.G. (1965), The Damping Capacity of Hardened Cement Paste, Mortar and
Concrete Specimens, Vibration in Civil Engineering; Proceedings of a
Symposium, London, April, 1965, Skipp, B.O. ed., Butterworths, London,
1966, pp. 235-247.
Cole, D.G. and Spooner, D.C. (1965a), The Damping Capacity of Hardened
Cement Paste and Mortar in Specimens Vibrating at Very Low Frequencies,
Proceedings, ASTM, Vol. 65, pp, 661-667.
Cole, D.G. and Spooner, D.C. (1965b), The Damping Capacity of Concrete, The
Structure of Concrete and Its Behaviour Under Load : Proceedings of an
International Conference, London, September, 1965, Brooks, A.E. and
Newman, K. eds., Cement and Concrete Association, London, 1968, pp. 217225.
References
R-5
Colonnetti, G. (1936), The Tensile Strength of Stressed Parts in Reinforced

Concrete, Proceedings, Second Congress, International Association for
Bridge and Structural Engineering, Berlin-Munich, pp. 191-196.
Comit Euro-International du Bton (CEB) (1978), CEB-FIP Model Code for
Concrete Structures, English Edition, Cement and Concrete Association,
London.
Comit Euro-International du Bton (CEB) (1985), CEB Manual : Cracking and
Deformations, CEB, Lausanne.
Comit Europen du Bton (CEB) (1970), International Recommendations for the
Design and Construction of Concrete Structures, Comit Europen du
Bton /Fdration International de la Prcontrainte, Paris, France, (English
edition published by the Cement and Concrete Association, London,
England), pp. 1-80, Appendix Vol., pp. 1-47.
Considere, A. (1906), Reinforced Concrete, 2nd edition, pp. 110-118.
Craig, R.R. (1981), Structural Dynamics An Introduction to Computer Methods,
John Wiley & Sons, New York.
Dieterle, R. and Bachmann, H. (1981), Experiments and Models for the Damping
Behaviour of Vibrating Reinforced Concrete Beams in the Uncracked and
Cracked Condition, International Association for Bridge and Structural
Engineering Report of the Working Commissions, Vol. 34, pp. 69-82.
Ellyin, F. and Chandrasekhar, P. (1977), Probabilistic Dynamic Response of Beams
and Frames, Journal of the Engineering Mechanics Division, ASCE, Vol.
103, No. EM3, June, pp. 411-421.
ENV 1992-1-1: Eurocode 2 (EC2) (1991), Design of Concrete Structures, Part 1:
General Rules and Rules for Buildings, European Prestandard, December,
1991.
Farrar, C.R., Baker, W.E. and Dove, R.C. (1994), Dynamic Parameter Similitude
for Concrete Models, ACI Structural Journal, Vol. 91, No. 1, JanuaryFebruary, pp. 90-99.
Ferry Borges, J. (1966), Cracking and Deformability of Reinforced Concrete
Beams, Publications, International Association for Bridge and Structural
Engineering, Zurich, Vol. 26, pp. 75-95.
Ferry Borges, J. (1968), Preliminary Report, CEB, Commission IV, Laboratorio
Nacional de Engenharia Civil, Lisbon, Portugal, March, 40p.
Fertis, D.G. (1973), Dynamics and Vibration of Structures, John Wiley & Sons,
New York.
References
R-6
Flesch, R. (1981), The Damping Behaviour of Reinforced Concrete Cantilever

Elements, International Association for Bridge and Structural Engineering
Report of the Working Commissions, Vol. 34, pp. 83-98.
Gergely, P. (1969), Anchorage Systems in Prestressed Concrete Pressure Vessels :
Anchorage Zone Problems, ORNL-TM-2378, Oak Ridge National
Laboratory, U.S. Atomic Energy Commission, Oak Ridge, pp. 1-49.
Gergely, P. (1972), Cracking and Crack Control, State of Art Report No. 1,
Proceedings, International Conference on Planning and Design of Tall
Buildings, Lehigh University, Bethlehem, August 21-26, 1972, ASCEIABSE, Vol. III-24, pp. 645-654.
Gergely, P. and Lutz, L.A. (1968), Maximum Crack Width in Reinforced Concrete
Flexural Members, Causes, Mechanism, and Control of Cracking in
Concrete, SP-20, American Concrete Institute, Detroit, pp. 87-117.
Ghali, A. and Favre, R. (1986), Concrete Structures : Stresses and Deformations,
Chapman and Hall, London.
Gilbert, R.I. (1983), Deflection Calculations for Reinforced Concrete Beams, Civil
Engineering Transactions, Institution of Engineers Australia, Vol. CE25,
No. 2, May, pp. 128-134.
Grace, N.F. and Kennedy, J.B. (1990), Dynamic Characteristics of Prestressed
Waffle Slabs, Journal of Structural Engineering, ASCE, Vol. 116, No. 6,
June, pp. 1547-1564.
Graf, O. (1921), Crack Development, (Translation), Handbuch fur Eisenbetonbau,
3rd edition, Vol. 1, pp. 104-115.
Grossman, J.S. (1990), Slender Concrete Structures - The New Edge, ACI
Structural Journal, Vol. 87, No. 1, January-February, pp. 39-52.
Hair, J.F., Anderson, R.E., Tatham, R.L. and Black, W.C. (1992), Multivariate Data
Analysis with Readings, Third edition, Macmillan Publishing Company, New
York, 544 p.
Harajli, M.H. and Naaman, A.E. (1989), Cracking in Partially Prestressed Beams
under Static and Fatigue Loading, ACI SP-113, American Concrete
Institute, Farmington Hills, Michigan, pp. 29-56.
Heiland, D., Bausch, S. and Stangenberg, F. (1990), Experimental Studies on the
Dynamic Behaviour of Reinforced Concrete Beams with Additional Steel
Fibers, Structural Dynamics, Proceedings of the European Conference on
Structural Dynamics EURODYN 90, Bochum, 5-7 June, 1990, Krtzig,
W.B. et al. Eds., A.A. Balkema, Rotterdam, 1991, Vol. 2, pp. 1055-1061.
Hognestad, E. (1962), High Strength Bars as Concrete Reinforcement, Part 2.
References
R-7
Control of Flexural Cracking, Journal of the Portland Cement Association

Research and Development Laboratories, Vol. 4, No. 1, January, pp. 46-63.
Holand, I. (1962), Damping of Vibrations in Simply Supported Prestressed Beams,
Institute of Structural Mechanics, The Technical University of Norway,
Trondheim, Experimental Investigation Report, April.
Holmberg, A. (1970), Flexural Crack Width, Nordish Beton.
Holmberg, A. and Lindgren, S. (1970), Crack Spacing and Crack Width due to
Normal Force of Bending Moment, Document D2 : 1970, National Swedish
Building Research.
Huang, P-T. (1975), Serviceability Behavior and Crack Control in Pretensioned
Prestressed Concrete I and T Beams, Ph D Thesis, Rutgers University, New
Jersey, 207p.
Humar, J.L. (1990), Dynamics of Structures, Prentice Hall, Englewood Cliffs.
Husain, S.I. and Ferguson, P.M. (1968), Flexural Crack Width at the Bars in
Reinforced Concrete Beams, Research Report No. 102-1F, Centre for
Highway Research, University of Texas, Austin.
Irvine, H.M. (1986), Structural Dynamics for the Practising Engineers, 1st edition,
reprinted 1993, E & FN Spon, London, 208p.
James, M. (1997), Prediction of Deflection in Simply Supported and Continuous
Reinforced Concrete Box Beams, B. Eng. Thesis, Griffith University,
Queensland, Australia, 228p.
James, M.L., Lutes, L.D. and Smith, G.M. (1964), Dynamic Properties of
Reinforced and Prestressed Concrete Structural Components, ACI Journal,
Proceedings, Vol. 61, No. 11, November, pp. 1359-1381.
Jones, R. (1957), The Effect of Frequency on the Dynamic Modulus and Damping
Coefficient of Concrete, Magazine of Concrete Research, Vol. 9, No. 26,
August, pp. 69-72.
Jones, R. and Welch, G.B. (1967), The Damping Properties of Plain Concrete:
Effect of Composition and Relations with Elasticity and Strength, Road
Research Laboratory Report, No. LR 111, 16p.
Jordan, R.W. (1977), Tensile Stress Effects on Damping, Concrete, March,
pp. 31-34.
Jordan, R.W. (1980), The Effect of Stress, Frequency, Curing, Mix and Age upon
the Damping of Concrete, Magazine of Concrete Research, Vol. 32, No.
113, December, pp. 195-205.
References
R-8
Kaar, P.H. and Hognestad, E. (1965), High Strength Bars as Concrete

Reinforcement, Part 7. Control of Cracking in T-Beam Flanges, Journal of
the Portland Cement Association Research and Development Laboratories,
Vol. 7, No. 1, January, pp. 42-53.
Kaar, P.H. and Mattock, A.H. (1963), High Strength Bars as Concrete
Reinforcement, Part 4. Control of Cracking, Journal of the Portland
Cement Association Research and Development Laboratories, Vol. 5, No. 1,
January, pp. 15-38.
Kareem, A. and Sun, W.-J. (1990), Dynamic Response of Structures with Uncertain
Damping, Engineering Structures, Vol. 12, January, pp. 2-8.
Kesler, C.E. and Higuchi, Y. (1953), Determination of Compressive Strength of
Concrete by Using its Sonic Properties, Proceedings, American Society for
Testing and Materials, Vol. 53, pp. 1044-1051.
Kesler, C.E. and Higuchi, Y. (1954), Problems in the Sonic Testing of Plain
Concrete, Proceedings of the International Symposium on the NonDestructive Testing of Materials and Structures, January, Paris, RILEM,
Vol. 1, Paper A-4, pp. 45-53.
Kreyszig, E. (1993), Advanced Engineering Mathematics, 7th edition, John Wiley &
Sons, New York.
Lan, Z. and Ding, D. (1992), Crack Width in Reinforced Concrete Members,
International Journal of Structures, Nem Chand & Bros., Roorkee (India),
Vol. 12, No. 2, July-December, pp. 137-163.
Lazan, B.J. (1968), Damping of Materials and Members in Structural Mechanics,
Pergamon, Oxford.
Lloyd, J.P., Hassan, M.R. and Kesler, C.E. (1969), Crack Control in One-Way
Slabs Reinforced with Deformed Welded Wire Fabric, ACI Journal, Vol.
66, No. 5, May, pp. 366-376.
Lu, T. (1993), Damping Characteristics of Reinforced Concrete Beams, An
Unpublished Report Presented to Professor Y.C. Loo, March, 164p.
MacGregor, J.G. (1997), Reinforced Concrete : Mechanics and Design, Third
edition, Prentice-Hall, New Jersey.
Major, A. (1980), Dynamics in Civil Engineering Analysis and Design, Vol. I,
Akadmiai Kiad, Budapest.
Makhlouf, H.M. and Malhas, F.A. (1996), The Effect of Thick Concrete Cover on
the Maximum Flexural Crack Width under Service Load, ACI Structural
Journal, Vol. 93, No. 3, May-June, pp. 257-265.
References
R-9
Mathey, R.G. and Watstein, D. (1960), Effect of Tensile Properties of

Reinforcement on the Flexural Characteristics of Beams, ACI Journal,
Proceedings, Vol. 56, No. 12, June, pp. 1253-1273.
Mier, J.G.M. van. (1997), Fracture Process of Concrete : Assessment of Material
Parameters for Fracture Models, CRC Press, Boca Raton.
Naaman, A.E. and Siriaksorn, A. (1979), Serviceability Based Design of Partially
Prestressed Beams. Part I - Analysis, PCI Journal, Vol. 24, No. 2, April,
pp. 64-89.
Nashif, A.D., Jones, D.I.G. and Henderson, J.P. (1985), Vibration Damping, Wiley,
New York.
Nawy, E.G. (1968), Crack Control in Reinforced Concrete Structures, ACI
Journal, Proceedings, Vol. 65, No. 10, October, pp. 825-836.
Nawy, E.G. (1984), Flexural Cracking of Pre- and Post-Tensioned Flanged
Beams, Proceedings of the NATO Advanced Research Workshop on
Partial Prestressing, From Theory to Practice, Paris, France, June 18-22,
1984, Cohn, M.Z. ed., Vol. 2, Martinus Nijhoff Publishers, Dordrecht, 1986,
pp. 137-156.
Nawy, E.G. (1996), Prestressed Concrete : A Fundamental Approach, Second
edition, Prentice Hall, New Jersey.
Nawy, E.G. and Potyondy, J.G. (1970), Moment Rotation, Cracking and Deflection
of Spirally Bound, Pretensioned Prestressed Beams, Engineering Research
Bulletin, No. 51, Rutgers University, November, 96p.
Nawy, E.G. and Potyondy, J.G. (1971), Flexural Cracking Behavior of Pretensioned Prestressed Concrete I- and T-Beams, ACI Journal, Proceedings, Vol.
68, No. 5, May, pp. 355-360.
Nie, N.H., Hull, C.H., Jenkins, J.G., Steinbrenner, K. and Bent, D.H. (1975), SPSS:
Statistical Package for the Social Sciences, Second edition, McGraw-Hill
Book Company, New York, 675p.
Oh, B.H. and Kang, Y.-J. (1987), New Formulas for Maximum Crack Width and
Crack Spacing in Reinforced Concrete Flexural Members, ACI Structural
Journal, Vol. 84, No. 2, March-April, pp. 103-112.
Osborne, K.P. and Ellis, B.R. (1990), Vibration Design and Testing of a Long-Span
Lightweight Floor, The Structural Engineer, The Institution of Structural
Engineers, London, Vol. 68, No. 10, 15 May, pp. 181-186.
Ouyang, C. and Shah, S.P. (1994), Fracture Energy Approach for Predicting
Cracking of Reinforced Concrete Tensile Members, ACI Structural Journal,
Vol. 91, No. 1, January-February, pp. 69-78.
References
R-10
Park, R. and Paulay, T. (1975), Reinforced Concrete Structures, John Wiley & Sons,
New York.
Paskalov, T., Taskov, L. and Taskovaki, B. (1980), Damping Determination from
Full-Scale Experiments, Proceedings, Seventh World Conference on
Earthquake Engineering, September 8-13, 1980, pp. 551-558.
Paz, M. (1997), Structural Dynamics: Theory and Computation, 4th edition,
Chapman and Hall, New York, 825p.
Penzien, J. (1964), Damping Characteristics of Prestressed Concrete, ACI Journal,
Proceedings, Title No. 61-61, Vol. 61, No. 9, September, pp. 1125-1148.
Penzien, J. and Hansen, R.J. (1954), Static and Dynamic Elastic Behaviour of
Reinforced Concrete Beams, Journal of the ACI, Proceedings, Title No. 5032, Vol. 50, No. 7, March, pp. 545-567.
Pinsker, W. (1949), Structural Damping, Journal of the Aeronautical Sciences,
Vol. 16, pp. 699.
Rainer, J.H. and Pernica, G. (1981), Damping of a Floor Sample, Proceedings of
the Second Specialty Conference on Dynamic Response of Structures:
Experimentation, Observation, Prediction and Control, Atlanta, January 1516, 1981, Hart, G. ed., ASCE, pp. 859-873.
Rao, S.S. (1990), Mechanical Vibrations, Second edition, Addison-Wesley
Publishing Company, New York.
RILEM (1957), International Symposium on Bond and Crack Formation in
Reinforced Concrete, Stockholm, 1957, Vol. 1-4, RILEM, Paris (Published
by Tekniska Hogskolans Rotaprinttryckeri, Stockholm, 1958), 880 p.
Rosenblueth, E. (1980), Design of Earthquake Resistant Structures, Pentech Press,
London.
Saliger, R. (1936), High Grade Steel in Reinforced Concrete, Proceedings, Second
Congress, International Association for Bridge and Structural Engineering,
Berlin-Munich, pp. 293-315.
Salzmann, A. (1997), Experimental Investigation of Damping Characteristics of
Simple and Continuous Reinforced Concrete Box Beams, B. Eng. Thesis,
Griffith University, Queensland, Australia, 70p.
Scanlan, R.H. and Mendelson, A. (1963), Structural Damping, AIAA Journal, Vol.
1, pp. 938-939.
Scholz. H. (1993), Alternative Crack Width Computation for Prestressed
Members, Proceedings of the Institution of Civil Engineers Structures and
Buildings, Vol. 99, Issue 2, May, pp. 135-140.
References
R-11
Severn, R.T., Brownjohn, J.M.W., Dumanoglu, A.A. and Taylor, C.A. (1988), A
Review of Dynamic Testing Methods for Civil Engineering Structures,
Civil Engineering Dynamics, Proceedings of the Conference organised by
the University of Bristol, in association with the Society for Earthquake and
Civil Engineering Dynamics, 24-25 March 1988, University of Bristol, 1989,
pp. 1-24.
Sinclair, D. (1955), Frictional Vibrations, Journal of Applied Mechanics, Vol. 22,
pp. 207-214.
Siriaksorn, A. and Naaman, A.E. (1979), Serviceability Based Design of Partially
Prestressed Beams. Part II - Computerized Design and Parametric
Evaluation, PCI Journal, Vol. 24, No. 3, May, pp. 40-60.
Smith, J.W. (1988), Vibration of Structures: Applications in Civil Engineering
Design, Chapman and Hall, London, 338p.
Spencer, R.A. (1969), Stiffness and Damping of Nine Cyclically Loaded
Prestressed Concrete Members, PCI Journal, Vol. 14, No. 3, May-June,
pp. 39-52.
Standards Association of Australia (SAA) (1994), AS 3600-1994: Concrete
Structures, Standards Australia, Homebush, NSW, Australia.
Stewart, R.A. (1997), Crack Widths and Cracking Characteristics of Simply
Supported and Continuous Reinforced Concrete Box Beams, B. Eng. Thesis,
Griffith University, Queensland, Australia, 203p.
Sun, C.T. and Lu, Y.P. (1995), Vibration Damping of Structural Elements, Prentice
Hall, Englewood Cliffs, 372p.
Suri, K.M. and Dilger, W.H. (1986), Crack Width of Partially Prestressed Concrete
Members, ACI Journal, Proceedings, Vol. 83, No. 5, September-October,
pp. 784-797.
Suzuki, K. and Ohno, Y. (1984), Crack Width and Crack Control of PPC
Members, Proceedings of the NATO Advanced Research Workshop on
Partial Prestressing, From Theory to Practice, Paris, France, June 18-22,
1984, Cohn, M.Z. ed., Vol. 2, Martinus Nijhoff Publishers, Dordrecht, 1986,
pp. 157-173.
Swamy, R.N. (1970), Damping Mechanisms in Cementitious Systems,
Proceedings of a Conference on Dynamic Waves in Civil Engineering,
University College, Swansea, July 1970, Howells, D.A. ed., WileyInterscience, London, 1971, pp. 521-542.
Swamy, R.N. and Rigby, G. (1971), Dynamic Properties of Hardened Paste, Mortar
and Concrete, Materials and Structures : Research and Testing, Vol. 4, No.
19, January-February, pp. 13-40.
References
R-12
Thomas, F.G. (1936), Cracking in Reinforced Concrete, The Structural Engineer,

Vol. 14, No. 7, July, pp. 298-320.
Thomson, W.T. (1940), Measuring Changes in Physical Properties of Concrete by
the Dynamic Method, Proceedings, American Society for Testing and
Materials, Vol. 40, p. 1113.
Thomson, W.T. (1988), Theory of Vibration, Third edition, Prentice-Hall
International Inc., Englewood Cliffs.
Tilly, G.P., Cullington, D.W. and Eyre, R. (1984), Dynamic Behaviour of
Footbridges, IABSE Surveys S-26/84, May.
Timoshenko, S., Young, D.H. and Weaver, W. (1974), Vibration Problems in
Engineering, Fourth edition, John Wiley & Sons, New York.
Ungar, E.E. (1973), The Status of Engineering Knowledge Concerning the
Damping of Built-Up Structures, Journal of Sound and Vibration, Vol. 26,
pp. 141-154.
Warner, R.F., Rangan, B.V. and Hall, A.S. (1989), Reinforced Concrete, 3rd edition,
Longman Cheshire, Melbourne.
Watstein, D. and Parsons, D.E. (1943), Width and Spacing of Tensile Cracks in
Axially Reinforced Concrete Cylinders, Journal of Research, National
Bureau of Standards, Vol. 31, No. RP 1545, July, pp. 1-24.
Watstein, D. and Seese, N.A., Jr. (1945), Effect of Type of Bar on Width of Cracks
in Reinforced Concrete Subjected to Tension, ACI Journal, Proceedings,
Vol. 41, February, pp. 293-304.
Welch, G.B. and Janjua, M.A. (1971), Width of Spacing of Tensile Cracks in
Reinforced Concrete, UNICIV Report No. R76, University of New South
Wales, Kensington.
Westergaard, H.M. (1933), Stresses at a Crack, Size of the Crack, and the Bending
of Reinforced Concrete, Journal of the ACI, Proceedings, Vol. 30,
November, pp. 93-102.
Wheeler, J.E. (1982), Prediction and Control of Pedestrian Induced Vibration in
Footbridges, Journal of Structural Division, ASCE, Vol. 108, No. ST 9,
September, pp. 2045-2065.
White, R.G. (1980), Vibration Testing, An Advanced Course in Noise and
Vibration, Institute of Sound and Vibration Research, University of
Southampton, pp. 31.1-31.15
Williams, C. (1988), Determination of Structural Parameters from Full-Scale
Vibration Tests, Civil Engineering Dynamics, Proceedings of the
References
R-13
Conference organised by the University of Bristol, in association with the

Society for Earthquake and Civil Engineering Dynamics, 24-25 March 1988,
University of Bristol, 1989, pp. 43-60.
Wilson, J. (1985), Transient Response of Structures, Chapter 9: Vibrations of
Engineering Structures, Brebbia, C.A. and Orszag, S.A. ed., Springer-Verlag
Berlin, Heidelberg, pp. 151-175.
Zech, B. and Setzer, M.J. (1988), The Dynamic Elastic Modulus of Hardened
Cement Paste. Part I: A new Statistical Model - Water and Ice Filled Pores,
Materials and Structures, RILEM, Vol. 21, No. 125, September, pp. 323328.
Zhao, G.-F. and Wang, Q.-X. (1993), Evaluation of Crack Width of Concrete
Structural Members, Proceedings, Fourth East Asia-Pacific Conference on
Structural Engineering & Construction, Seoul, Korea, September 20-22,
1993, Vol. 2, pp. 1243-1248.
Zielinski, J.L. and Rowe, R.E. (1960), An Investigation of the Stress Distribution in
the Anchorage Zones of Post-Tensioned Concrete Members, Technical
Report No. 9, Cement and Concrete Association, London, September, 32p.

Damping

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

Damping

Uploaded by

Copyright:

Available Formats

DAMPING CHARACTERISTICS OF REINFORCED

AND PARTIALLY PRESTRESSED CONCRETE BEAMS

A Thesis submitted in fulfilment of the requirements for

SANAUL HUQ CHOWDHURY

Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams

Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams

Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams

Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams

Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams

It is this area which forms the primary focus of this

reinforced and partially prestressed beams.

Nine reinforced and 12 partially

For all the beams, at each level of loading,

measurements were made of instantaneous and residual crack widths, instantaneous

Each beam was also

subjected to free vibration tests to measure its logarithmic decrement of damping

These further enhance the practicability of the proposed

Damping - Theory and Concepts .. 2-1

Damping in Concrete Elements A Literature Review 3-1

Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams

Review of Research on Damping of Concrete Structural

Damping Theory of Simply Supported and Continuous Beams .. 4-1

Cracking in Reinforced and Partially Prestressed Concrete Beams .. 5-1

The Experimental Programme . 6-1

Development of Instantaneous Crack Width Formula ... 7-1

Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams

Proposed Equation for Instantaneous Average Crack Width ... 7-4

Residual Crack Width Measurements and Predictions .. 8-1

Damping in Reinforced Concrete Beams . 9-1

Damping in Partially Prestressed Concrete Beams .. 10-1

Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams

10.3.2 Under reinforced beams .. 10-4

References ... R-1

Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams

Different types of damping

The hysteresis loop (after Ashbee et al., 1976)

Equivalent SDOF mass-spring systems

Primary and secondary flexural cracks

Reinforcement details of beams 1, 2, 3, 4, 8 and 11

Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams

Beam support system for the first-stage beams

Crack detection microscope

Demec strain gauge

Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams

Different methods of vibration pick-up mountings (after

Measured versus calculated damping values for the

Crack patterns for beam no. 3

Instantaneous crack width/strain versus loading curves for

Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams

Instantaneous crack width/strain versus loading curves for

Response records for beam no. 1 at zero load (wr = 0)

Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams

Response records for span 1 of beam no. 17 at zero load

Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams

Logarithmic decrements versus residual crack widths for

Logarithmic decrements versus residual crack widths for

Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams

Logarithmic decrements versus residual crack widths for

Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams

Material damping of different materials (after Bachman

Details of reinforced concrete simply supported beams

Parameters used in the development of average crack

Damping data for beam 1

Damping Characteristics of Reinforced and Partially Prestressed Concrete Beams