Anda di halaman 1dari 283

IIIGH STRENGTH

STEEL

REINFORCEMENT

IN

ORDINARY

REINFORCED

AND FIBRE

REINFORCED

CEMENT

COMPOSITE

LIGHTWEIGHT

CONCRETE

BEAMS

A thesis presented for the degree of Dcctor of Philosophy

by Kosaa Abdul Aziz Al-Sanjary B. Sc.

Department

of Civil Engineering of Salford

The University

May 1975

TO MY FATHER AND MOTHER

ACKNOWLEDGEMENTS

The author is grateful

to Professor

T. Constantine,

B. Sc., Ph. D, C. Eng., and for

F. I. C. E. F. I. M un. E. M. Inst. H. E. Professor of Civil Engineering , , , Chairman of the Civil Engineering Department for providing the facilities the research work. The author wishes to express his gratitude Ph. D, D. Sc., C. Eng., Engineering F. I. C. E., FI Struct. to Professor E., Professor E. R. Bryan, of Structural

M. Sc. (Eng)j

for recommending

and providing

the opportunity

to the author to study of the author

for a higher degree, and also for the encouragement to the progress since his early enrollment at the University.

The author is deeply indebted to his supervisor Ph. D, C. Eng, M. I. C. E. him to the subject , D. C. T.

Dr. N. J. Dave, M. Eng, introduced

A. M. I. E. M. A. S. C. E., who first ,

for his constant encouragement,

guidance and above all

patience throughout the work. The author would like to thank Mr. D. C. O'Leary, for his help during the testing programme. The author is most thankful to the technical the concrete and Fitton structures Messrs. T. Clark, laboratories. staff, for their practical help in M. Sc., C. Eng., M. I. C. E,

In this respect thanks are due to E, K. Naylor, T. Ward,

W. Deakin, C. Eng.,

MI Strut.

R. Smith, H. Whewell and C. Lomax. The author is especially thankful to Mr. N. Beaver, A. R. T. C. S. , for

reading through the thesis and for his most valued remarks. Thanks are also due to the Building Research Establishment rig for the sustained loading tests. Thanks are also due to TAC Construction and supplying the fibre reinforced
The author for the first is grateful

for providing

the

Materials

Limited

for manufacturing

cement channels used in this research.


Council for paying the University fees

to the British

two years

of the enrol: meat.

The author acknowledges with gratitude Iraqi Ministry Finally of Oil and Minerals.

the financial

support provided by the

my special thanks to my wife Neda for typing the draft of the thesis

and for her encouragement and patience throughout the work.

SYNOPSIS When high strength steel is used as reinforcement members, great economies can be achieved. However, in lightweight concrete

because lightweight the working steel states of serviceability

concrete has low tensile strength and modulus of elasticity, stresses hoped for may not be fully utilised (cracking and deflection) and deflection in flexural not being satisfied. concrete members, due to the limit

To control the amount of cracking a new type of construction has been

employed, whereby precast fibre reinforced

cement (f. r. c) units in the form of at the flexural tensile zone of

thin channels are used as a surface reinforcement the concrete members.

The concrete in the tensile zone, confined by the f. r. c to formation and extension of crack; of the member will be

channel, will have a greater resistance consequently the rate of reduction decreased. A total of 27 ordinary lightweight concrete beams, reinforced

in the flexural

rigidity

and fibre reinforced

cement composite

150mm wide, 300mm deep and 5m long were tested,

18 under static load test,

5 under fatigue load test and 4 under sustained load test. to the ordinary beams in every respect, except

The composite beams were similar that f. r. c. channels (150mm width, were incorporated parameters

60mm length of upstands and 6mm thickness) tensile sides. The main

as integral parts on their flexural

employed for both beams were the type and amount of steel provided The various types of reinforcement with the

for the tension reinforcement. corresponding


(275 N/mm2), (590 N/mm2)

nominal yield,
Unisteel and "Kam

or 0.2% proof stress employed were mild steel


Unisteel 550 (550 N/mm2), "Kam 60"

410 (410 N/mm2), 90" (875 N/mm2).

The flexural

behaviour of both types of beams under static,

fatigue and

sustained types of loading has been studied, great emphasis being placed upon the limit states of ultimate strength, cracking and deflection with particular reference to the contribution of the f. r. c. channels in the composite beams. it is concluded that a considerable reduction in the amount

From the results, of deflection

and cracking

can be achieved by using f. r. c. channels at the flexural thus allowing a more efficient use of the high-

tensile zone of concrete members, strength steel.

OUTLINE

OF THESIS

In chapter one the structural and lightweight

and economic aspects of high strength steel are discussed.

concrete when employed in concrete construction aspects and application

Emphasis is placed on the structural

of fibre reinforced The limitations deflection of

concrete and the techniques employed for mixing the fibres. three-dimensional random distribution of fibres on controlling

and

cracking are also presented. In chapter two the work carried lightweight concrete is reviewed, out in the past using high strength steel in The other

and a general conclusion is drawn. development

part of the chapter is concerned with the origin, measures to reduce cracking carried out at the University and deflection of Salford

and use of the proposed Tests

in flexural

concrete members.

and other places are also discussed. used in this

Chapter three covers the design of test beams and materials research. parameters moments. properties Information employed,

given for the test beams includes their dimensions, condition of loading and analysis of the working and ultimate with regard to their

Also included is the design of the f. r. c, units and geometry. The mix proportions

for the concrete and the properties

of the steel used are also given. Chapter four deals with the proposed theoretical research for the stresses in the concrete, relationships analysis developed in this The various

f. r. c. channel and steel.

established together with the idealised curves are also presented. considerations regarding the proposed methods are also discussed.

In chapter five the theoretical for the limit states of ultimate

strength,

cracking and deflection

Chapter six covers the manufacture

and methods of testing employed for the

test beams and the control specimens for the concrete properties. The observations made and the behaviour of the beams tested under static A comparison of the observed values

loading are discussed in chapter seven. with those predicted

in accordance with chapters four and five is also presented. of behaviour between ordinary and

This chapter also includes a direct comparison composite beams.

In chapter eight the behaviour and the observations tested under fatigue and sustained loading are discussed. of behaviour between ordinary Finally,

made for the beams A direct comparison

and composite beams is also presented.

in chapter nine, the conclusions of the research are drawn up, and out in this field.

suggestions are made for future work to be carried

BIITFORCSLT-MIIT HIGH STP. Bi: GT St. 'I, P. IIJ Q'tDIIdA',Y R"IlT? 'MIC';D AD LIGHTMIGHT HIITFO; CM, C3.1-MITT F7BRE P. i'OSITB CO'.. I:IS COi:C^LT:, B:S.

i'

ATA

For units
Vor neutral

of moments substitute
axis depth substitute

"kit . n" for


"X" for "X"

m" throuc
throughout

out .

vA

CONTENTS
PaKe No. ACKNOWLEDGEMENT SYNOPSIS OUTLINE CONTENTS LIST OF SYMBOLS ABBREVIATIONS CODING REFERENCE CHAPTER 1.1 1.2 ONE : FOR TEST INTRODUCTION BEAMS OF THESIS ii iii v xi xv xvi 1 1 Steel 2

Summary High-Strength

1.2.1 1.2.2

General Economic Aspects Lightweight Concrete

2 3 4 4 5 6 8 11

1.3 Structural 1.3.1 1.3.2

General Economic Aspects

1.4 Wealmess of Concrete in Tension 1.5 Fibre-Reinforced Concrete

1.6 The Need for this Investigation CHAPTER TWO : PREVIOUS INVESTIGATIONS 2.1 Use of High-Strength Concrete 2.1.1 2.1.2 2.1.3 General Summary of Past Research Conclusions From Past Research Steel in Structural Lightweight

12 12 12 19 20 20 20 22

2.2 Approaches Employed to Control Cracking and Deflection In Flexural Members Reinforced with High-Strength Steel. 2.2.1 2.2.2 2.2.3 2.2.4 2.2.5 General Closely Spaced Continuous Reinforcement Fibre Reinforced Concrete Using Steel

Composite Concrete Construction Channels Work at Salford

26 27

Page No CHAPTER THREE


3.1 3.2 General Design 3.2.1 3.2.2 3.2.3 Considerations Ordinary Composite Fibre Reinforced Reinforced Lightweight Lightweight Concrete Concrete (f. r. c. Units) Beams Beams

: MATERIALS USED AND DESIGN OF TEST BEAMS


30 30 30 32 33

Reinforced

Cement Units

3.2.3.1 3.2.3.2 3.3 Materials 3.3.1 3.3.2

Properties Geometry

33 34 35

Lightweight

Aggregate Concrete

35 36

Reinforcement : THEORETICAL BASIS OF ANALYSIS FOR STRESSES IN CONCRETE, STEEL AND f. r. c. CHANNELS

CHAPTER FOUR

4.1 4.2 4.3

Introductions Variations Analysis of the Neutral of Compressive Axis Level in Concrete

7 37 41

Stresses

4.3.1 4.3.2 4.3.3

Stress-Strain Relationship in Compression Variations Concrete of the Flexural

of Lightweight Compressive

Concrete Strain in

41 41 42 44 46
46

Variations of the Area and Centroid of the Compressive Stress Distribution in Concrete of Stresses in f. r. c. Channels of Steel Stresses

4.4 Analysis 4.5

Calculation
4.5.1 4.5.2 4.5.3 4.5.4

Assumptions Ordinary Composite Summary Stresses Lightweight Lightweight Concrete Concrete Beams Beams of Steel

47 47 48

for the Theoretical

Calculation

4.6

Suggested Method for Calculation on Experimental Results

of Steel Stresses Based

48

Page No
CHAPTER FIVE CONSIDERATIONS FOR : THEORETICAL LIMIT STATES OF DESIGN FOR ORDINARY AND COMPOSITE BEAMS 50 Strength 51

5.1 5.2

Introduction Limit State of Ultimate

5.2.1 5.2.2. 5.2.3 5.3 Limit 5.3.1 5.3.2. 5.3.3 5.3.4 5.3.5

General Considerations

and Assumptions

51 52 53 54 54 56 58 58, 62

Comments on the Basic Assumptions Composite Beams State of Deflection Introduction Considerations for Ordinary and Composite Lightweight Concrete Beams Behaviour of Ordinary Lightweight Concrete Beams Concrete Beams

Behaviour of Composite Lightweight Methods of Calculation Approach One - Existing Theories

for Ordinary

Beams and

62 63

Approach Two - Empirical Method for Ordinary Composite Beams Approach Three - Proposed Methods (a) Ordinary Beams (b) Composite Beams 5.4 Limit 5.4.1 5.4.2 5.4.3 State of Cracking Introduction Cracking Mechanism, Concrete Beams Methods of Calculation : MANUFACTURE in Composite Lightweight

63 65 67 67 68 71

CHAPTER SIX 6.1 General

AND METHODS OF TESTING 73 73

6.2 Manufacture 6.2.1 6.2.2 6.2.3 6.2.4 Ordinary Reinforced Lightweight Lightweight Concrete Beams Concrete Beams

73 74 74 75

Composite Reinforced

Control Specimens for Concrete Fibre Reinforced Cement Channels (f. r. c. Channels)

Page No. 6.3 Static Loading Tests 6.3.1 6.3.2 6.4 Arrangements and Conditions of Loading 75 75 76 77 and Conditions of Loading 77 78
78 of Loading 78 80 80 81 Tests Tests on Concrete on Steel 81 82 83

Testing Procedure

Fatigue Loading Tests 6.4.1. 6.4.2 Arrangements

Testing Procedure
Loading Tests and Conditions

6.5

Sustained 6.5.1 6.5.2

Arrangements Testing

Procedure

6.6 6.7

Instrumentation Other Tests 6.7.1 6.7.2 6.7.3 Control Tensile Tests

on f. r. c. Units

CHAPTER SEVEN : DISCUSSION OF TEST RESULTS AND COMPARISON OF TEST BEHAVIOUR WITH THEORETICAL PREDICTION FOR ORDINARY AND COMPOSITE BEAMS

7.1 7.2 7.3 7.4 7.5 7.6 7.7

Introduction Variations Flexural Varations Stresses Stresses Limit 7.7.1 7.7.2 of the Neutral Strain Distribution Compressive Strain in Concrete Axis Level

84 84 87 88 91 93 95 Strength 95 97 Beams Beams of 97 101 103

of the Flexural

in the f. r. c. Channels in the Steel Reinforcement

States of Design Limit Limit 7.7.2.1 7.7.2.2. 7.7.2.3 State of Ultimate State of Deflection Behaviour Behaviour of Ordinary of Composite

Comparison Between the Behaviour Ordinary and Composite Beams

7.7.2.4

Comparison with Theory

105

Page No. 7.7.3 Limit 7.7.3.1 7.7.3.2 7.7.3.3 7.7.3.4 CHAPTER EIGHT State of Cracking Behaviour Behaviour of Ordinary Beams 108 108 111 113 114

of Composite Beams

Comparison Between the Behaviour of Ordinary and Composite Beams Comparison with Theory : EFFECTS CP LONG TERM LOADING ON TEST BEHAVIOUR FOR ORDINARY AND COMPOSITE BEAMS

8.1 Introduction 8.2 Fatigue Loading Tests 8.2.1 8.2.2 8.2.3 8.2.4 8.2.5 Variations of the Neutral Axis Depth Compressive

116 116 116 118 120 121 123 Strength 123 124 128 131 Flexural Strain Distribution Strains in the f. r. c. 131 134 135 136 136 139
FOR

Variations of the Maximum Flexural Strain in Concrete Stresses in the f. r. c. Channels Variations Principal 8.2.5.1 8.2.5.2 8.2.5.3 of the Steel Stresses Limit Limit Limit Limit States State of Ultimate State of Deflection State of Cracking

8.3 Sustained Loading Tests 8.3.1 8.3.2 8.3.3 8.3.4 Time-Dependent Variations Channels Variations Limit 8.3.4.1 8.3.4.2
CHAPTER NINE

of the Tensile

of the Steel Stresses

States of Serviceability Limit Limit State of Deflection State of Cracking

: CONCLUSIONS AND "SUGGESTIONS FUTURE WORKS

9.1

Conclusions

142 143 143

9.2 Advantages of Using f. r. c. Channels at the Tensile Sides of Flexural Members 9.3 Suggestions for Future Work

Page No.

REFERENCES TABLES FIGURES PLATES APPENDICIES

145-156

LIST OF SYMBOLS Ach Ae As Asv a a cross section area of f. r. c. channel effective concrete area in tension area of tension reinforcement cross-sectional deflection distance from compression width is being calculated acr distance from the point (crack) considered to the surface of the nearest logitudinal b C Co, Cl, C2 bar face to the point at which the crack area of the two legs of a link

width of section and also width of f. r. c. channel compression ratios force in concrete moment, for the prediction

of applied moment to ultimate

of the neutral axis level Cp ratio of applied moment to ultimate the maximum compressive c Dc d moment for the prediction of

strain in concrete

concrete cover to tension steel density of concrete at time of test total depth of beam height of upstand of f. r. c. channel

da dl d2 E Ea Ec Es ec ei

centroid of stress distribution effective depth of reinforcement effective

from soffit of f. r. c. channel

depth of the tensile force (Ta) in f. r. c, channel

modulus of elasticity initial initial modulus of elasticity modulus of elasticity of f. r. c. units of concrete

modulus of elasticity compressive

of steel

strain in concrete stress of concrete is

strain at which maximum compressive first attained

emax ep

maximum compressive compressive

strain of concrete

strain in concrete at M/Mu = Cp, for the variation strain in concrete with applied moment

of the flexural et fab fat fbs fc

strain in main tension reinforcement tensile stress at soffit of f. r. c. channel tensile stress at top of the upstands of the f. r. c. channel bond stress
compressive compression stress zone in the extreme element of concrete in the

fcu fr frc

characteristic flexural flexural

strength of concrete in compression-cube

tests

tensile stress of plain concrete

(modulus of rupture) for composite

tensile stress in concrete at interface

section fs 41 fs fsave fsmin ft fy stress in main tension reinforcement tensile stress resisted by concrete between cracks average stress in main tension reinforcement minimum stress in main tension reinforcement

average tensile stress in concrete characteristic proof stress) strength of reinforcement. (Nominal yield or 0.2%

fYV I ICC Ico Io Ip

characteristic

strength of link reinforcement

second moment of area-general second moment of area for cracked composite section second moment of area for cracked ordinary second moment of area for untracked second moment of area for partially beams) section

concrete section - general cracked section. (composite

Iuc Iuo
i K, K

for of area uncracked composite section second moment second moment of area for untracked constants for deflection calculations ordinary section

for composite beams for the deflection calculations

Ko. Kl K2, K3

constants used for deriving formula for ordinary beams.

(K3 = Ko Kl K2)

L lc M Ma Mc Mdw mu m ml n
P P1 rb 1

span of beam lever arm of tensile force in concrete applied bending moment cracking moment or f. r. c. channel cracking moment of concrete

design working moment ultimate resistance moment

modular ratio = Es/Ec modular ratio = Ea/Ec ratio of neutral axis depth to effective geometrical geometrical depth = As/bdl

ratio of main tension reinforcement ratio of f. r. c. channel = Ach/bd2

radius of curvature curvature of beam at midspan

rb Sv Tc Ts Ta t u V v VC W w X Xc Xcc Xco XP XU

spacing of links along the member tensile force of concrete in tension tensile force in main reinforcement tensile force in f. r. c. channel thickness of f. r. c. channel perimeter shear force due to ultimate shear stress ultimate shear stress in concrete load

total applied load on beam crack width neutral axis depth-general neutral axis depth of cracked section-general neutral axis depth of cracked composite section neutral axis depth of cracked ordinary neutral axis depth of partially section

cracked composite section

neutral axis depth of uncracked section-general

Xuc Xuo Z" U.

neutral axis depth of untracked neutral axis depth of untracked

composite section ordinary section

moment lever arm for steel reinforcement coefficient for calculating diagram centroid of concrete compressive area of concrete compressive stress

distribution

p Pi

coefficient

for calculating

stress distribution

diagram cracked second moment of area, for

constant depending upon effective and modulus of elasticity

of f. r. c. channel and concrete, of stresses

composite beams (for calculation

in f. r. c. channels)

Y Ym

coefficient partial

=1-

(3

safety factor for strength (Ym = 1.5 for concrete and 1.15

for steel)
aC co

compressive

stress in concrete stress in stress-strain relationship for

maximum compressive concrete = 0.67 fcu

TI

ratio of effective steel = d2/d1

depth of tensile force in f. r. c. channel to that of

TI

ratio of of steel =

length of upstand for f. r. c. channel to effective depth d/dl = d/dl

ratio of total depth to effective depth of steel curvature average curvature

ave

ABREVIATIONS

A. C. I. A. S. C. E. B. R. E. B. S. I.

American American

Concrete Institute Society of Civil Engineers

Building Research Establishment British Standards Institution

C. & C. A.
4'

Cement and Concrete Association


u

C. E. B. C. P. F. I. P.

Comite Europeen du Beton Code of Practice oFederation International

(European Concrete Committee)

de la Precontrainteto (International

Prestressing f. r. c. I. A. B. S. E. I. C. E. I. C. L. C. P. C. A. R. Fibre Reinforced International Institution International Portland

Federation) Cement for Bridge and Structural Engineering

Association

of Civil Engineers Congress on Lighweight Concrete

Cement Association Concrete Review

Reinforced

CODING REFERENCE

FOR TEST

BEAMS

The coding reference below refers concrete beams. manner;

to ordinary

and composite lightweight was given in the following

For each beam the coding reference number and a letter, e. g.

two letters,

ST4-0

The first two letters These are: ST FA SU : :

refer to the type of loading employed.

Beams tested under static loading Beams tested under fatigue loading Beams tested under sustained loading

The number refers the beams, e. g. 1 2 3 The last letter These are: 0 C:

to the amount and type of reinforcements

employed in

2-16mm, 2-16mm, 2-16mm,

mild steel Uni-steel Uni-steel 410 550

refers to the type of the beam tested.

Ordinary

reinforced

lightweight lightweight

concrete beams concrete beams

Composite reinforced

CHAPTER INTRODUCTION

ONE

1.1

Summary

When high strength steel is used as reinforcement great economies can be achieved. There are, however,

in concrete members, certain limitations on

its use; these are imposed mainly to satisfy the limit (cracking and deflection)

states of serviceability When lightweight concrete

at the working load conditions.

is used cracking and deflection becomes even more critical, of elasticity and tensile strength. of the conventional elastic theory, of concrete in tension,

due to its low modulus

Since the introduction made to rectify

attempts have been

the limitations

and also to reduce the amount

load. deflection taking place under and of cracking by: 1) Fully or partially prestressing

This has been partly achieved

the concrete,

basic techniques being of the members. may be

employed to improve the structural 2) Using fibre-reinforced obtained in the flexural modulus fibres. A new concept of fibre-reinforced a means of controlling the University
"

performance

improvements certain whereby concrete

behaviour of the concrete by the addition of high

cement composite concrete construction

as

cracking

and deflection has been evolved at and patented by

of Salford.
investigation was initiated concrete to study the possibility members. of using high of

The present strength

steel in flexural

lightweight

The programme of fibre composite for lightweight cement units

investigation construction concrete employed

then was extended

to employ a new concept cracking investigation

as a means of controlling In the present reinforcement.

and deflection the asbestos

members. as surface

were

In the following paragraphs the structural of lightweight aggregate concrete, high-strength

properties

and economical aspects composite

steel and fibre-reinforced

concrete are discussed.

1.2
1.2.1

High Strength Steel


General

In this investigation

the general term high-strength

steel denotes reinforcing

bars having a yield or 0.2% proof stress of 410 N/mm2 or higher. High-strength steel bars have been produced and used in various parts of the concrete design. strength, These reinforcing

world as a means of economy in reinforced bars differ significantly in their ultimate

extent of surface deformation

and method of manufacture. To control the distribution characteristics occurring. The effect of the manufacturing can be seen in fig. process on the tensile properties of the bars and width of cracks the steel bars should have surface

failure bond be developed high to without any enabling stresses

1, where hot rolled bars have a definite yield point while cold

worked bars yield gradually.


The advantages high-strength claimed by many investigators concrete (1) (2) (3) (4) on the use of in the following

steel in reinforced

may be summarized

1. When the full strength of the reinforcing a significant reduction

bars is considered in the design, The

can be obtained in the amount of steel.

to be equal in with steel would mild steel area compared percentage saving (1 - fy/fy) 100,

where fy = yield stress of mild steel fy = yield or 0.2% proof stress of high-strength 2. Ease and speed of construction due to a reduction steel used

of end hooks and also member.

a possible reduction in the size of the structural


3. Economy. This is discussed in detail in 1.2.2.

There are certain limitations, in reinforced requirements limitations concrete members. (cracking

however,

on the use of high-strength

steel

These are mainly imposed to satisfy the safety at the working loading conditions. These

and deflection)

arise from the fact that high-strength as mild steel,

steel bars have nearly the same steel stresses are

modulus of elasticity employed, greater

so that when high working

strains result which lead to wider cracks and greater deflection.

The Code of Practice

CP110 (5) permits

the use of reinforcing

bars with

0".2% proof or yield stress between (410 - 460 N/mm2), and the nominal size of the reinforcement. permissible conditions. stresses of up to 380 N/mm2 Considering

depending on the type

On the continent (2), however, are being allowed under working load 0.2% proof or yield

a load factor of 1.8, this will permit

stress of 685 N/mm2 to be used.


In the past a considerable and other beams parts of the world, amount of research has been carried of normal out in Britain weight concrete

investigating steel.

the behaviour There

reinforced

with high-strength

has been,

however,

only a

limited

amount of work carried Further

out into the use of high strength steel in lightweight in this respect is, therefore, (cracking deemed at

concrete members.

investigation

to be necessary as the limit

states of serviceability

and deflection) concrete is used.

the working load conditions may be critical This is discussed in 1.3.1. 1.2.2. Economic Aspects Many researchers

when lightweight

have established that by using high-strength

steel in place used

of mild steel, a direct reduction (1) (2) (6). The American

can be made in the amount of reinforcement

Iron and Steel Institute

(7) has presented design analyses based with

on the A. C. I. code recommendations high-strength steel.

for beams and columns reinforced

These analyses showed that when high-strength

steel was

used instead of mild steel a considerable made.

saving in the amount of steel could be'

For beams with balanced section it was suggested that by using Unisteel

550

instead of Unisteel 410 savings of 25% could be obtained in the steel area with an overall saving of 4% in the total cost of the member (4). office building (3) indicated that a saving of steel with a

An analysis of a six-storeyed

28.5% in the amount of steel could be obtained when high-strength

yield stress of 417 N/mm2 replaced mild steel of 255 N/mm2 yield stress.

1.3

Structural

Lightweight

Concrete

1.3.1

General Because of the advantages lightweight concrete possesses, material it has in the last

two decades become an important construction. resistance,

structural

used in various types of greater fire (8) (9) (10). which include

Among these advantages are its lower unit weight, improved thermal concrete, insulation and ease of construction

Lightweight

however,

has certain shortcomings,

a low tensile strength and modulus of elasticity, and creep. These considerations

with high values for shrinkage

together with a lack of available data on its

behaviour have limited When high-strength

its use to a minor role in concrete structures. steel is used as a reinforcement in lightweight concrete

beams and the various limit These beams, however,

be economies can achieved. great satisfied states are amount of cracking and

tend to exhibit a considerable

deflection at working loads. is limited.

It is for this reason that the use of this type of steel

The maximum characteristic accordance with

be in bars for used can which steel strength concrete ranges

CP 110 for both normal weight and lightweight

between 410 - 460 N/mm2. The code also specifies that the design should be based on lower values of limit the to satisfy steel stresses where necessary width. In addition the recommended states of deflection and crack

span-depth ratio for normal weight concrete concrete is used. rigidity It is

should be multiplied

by a factor of 0.85 when lightweight

believed that this takes into account the reduced flexural concrete members compared with normal-weight In the past there has been a considerable structural properties of lightweight however, lightweight

of the lightweight

concrete. amount of research into the A limited

concrete as a material

(11) (12) (13).

amount of research, full scale reinforced of lightweight

has been carried

out into the flexural Considering

behaviour of

concrete members.

the shortcomings and high

concrete (low tensile strength and modulus of elasticity,

values for shrinkage and creep) it was thought that the limit would not be fully satisfied; especially

states of serviceability

when fatigue or sustained types of loading

-4

were considered. flexural

It was for this reason thought necessary to study the lightweight concrete members under static,

behaviour of reinforced

fatigue and sustained types of loading


1.3.2 Economic Aspects

It is rather difficult lightweight

to assess the actual saving in cost brought about by using It may be true that freshly concrete, but

concrete instead of normal weight concrete.

mixed lightweight

concrete is more expensive than normal-weight

this should not form the only basis of comparison.

To make a complete estimation concrete should be

of the savings all the relevant factors for the use of lightweight considered; lighter

these include the lower total dead weight and the consequent use of and lifting equipment.
multi-storey office building Although showed it was was

transportation

An analysis results slightly

of cost made on a typical in favour

of the use of lightweight could be drawn, were

concrete.

concluded competitive

that no set rule

the use of lightweight right. (14)

concrete

when the circumstances

Recently a working party of the Concrete Society was formed to study the economics of lightweight units were considered, in cost to normal-weight e. g., concrete construction. In this study suspended floor concrete was comparable,

lightweight that it concluded was and concrete,

and would even be cheaper if other factors, (15)

possible higher output and suitable plant, were taken into consideration In the C. E. B. recommendations (10) the economies of lightweight

concrete categories: -

in relation

to normal-weight

concrete were divided into the following

(i) Economies due to lower dead weight (ii) Economies due to other properties of lightweight concrete the reduction of

Considering

the lower unit weight of lightweight

concrete,

the total load will depend upon the density of the lightweight of the live load to dead load. The amount of reduction

concrete and the ratio

compared with normal weight

concrete may be calculated as follows: Considering: P= g= live load KN/m2 dead load (normal weight concrete) (KN/m2)

-5

Y2 Y1 a1= o

= density of lightweight

concrete (Kg/m3) (Kg/m3)

= density of normal weight concrete P/g Y2 /Y1

The weight saving in relation

to normal weight concrete being

Y1

Y2

I
I

Y1

100% =(1-

o) 100%
+ a1

The ratio of total load to dead load is p+g=1 g

Then the reduction of the total load, when lightweight


is

concrete is to be used

11+

o
lUU%p

al correlating the reduction in total load and the In this figure, it can be

A graph has been established different values of o and

a1 as shown in fig. 2.

seen that the reduction in total load increases

as the ratios of live load to the to normal weight concrete ( o ) factor in the economic

dead load ( a1), and the density of the lightweight decreases.

This was concluded to be the most important

considerations. Other factors also considered to have a great influence on the economies are the cost reduction of formwork and scaffolding, savings in the cost of foundations,

reduced column sizes, cheaper transport

per unit volume and faster building. resistance to heat, frost and fire,

The other advantages, e. g., additional were considered to cause an indirect


1.4 Weakness of Concrete in Tension

reduction

in building costs.

Concrete widely used as a building

material

has certain shortcomings, When the tensile

among

these are its low tensile strength and limited strength of the concrete is reached, the concrete will crack. deflection of a flexural to corrosion

extensibility.

either due to applied tensile stress or shrinkage,

This will lead to a greater rate of increase in the member with applied load. Wide cracks can also lead strength may

of the steel, consequently the limit

state of ultimate

not be fully satisfied. Much has been done to improve the strength of concrete in compression. Concrete of more than 100 N/mm2 been no significant improvement can now be produced (16). There has, however, which is

in the tensile strength of the concrete, strength.

only about 1/10 to 1/20-the value of its compressive

It is for this reason

that the tensile strength of the concrete is ignored in most design considerations. In 1906 (17) the basic assumptions were formulated of reinforced ignored. different concrete, in which the tensile resistance for the elastic theory of the concrete was to be used by many

This formed the basis of various codes of practices countries. in

In the load factor method introduced theory of design in ultimate CP110, the structure

C P114 (18) and the recent limit

state

being considered is analysed at its

load condition.

At this stage the concrete in tension is ignored in calculating to the is beam, the the carry considered steel and only of

the bending resistance tensile stresses. In

CP110 hypothetical

values are given to include the stiffening the deflection and cracking. stress distribution

effect of the

concrete in tension for calculating the deflection

For calculating for the concrete

at the working load, a triangular

in tension is assumed as 1 N/mm2 at the steel level and zero at the level of the neutral axis. This assumed tensile stress has in fact taken into consideration the

average contribution

the between the to in the tension the cracks zone concrete of For the calculation of the crack width the

stiffness of the concrete member. stiffening

effect of the concrete has also been considered. with high-strength steel high stresses

For concrete members reinforced and strains

are expected at the working load condition which will cause wider Hence there is an essential need to improve the This may be achieved by

cracks and greater deflection. structural improving performance

of the concrete members.

the tensile strength of the concrete. was introduced as a means to improve the structural In prestressed performance

Prestressing of flexural

members at working load conditions.

concrete class 1 in an absence

the concrete is assumed not to carry

any tensile stress; this results

-7-

of cracking.

As there is no loss of rigidity reduced. prestressed

due to cracking the deflection would

be considerably Partially improvement

concrete was then introduced concrete.

as an economical tensile stresses and in

on fully prestressed

This permitted

some cases limited

cracking to occur in the concrete at working load. could well be regarded as a forward of concrete members. step towards

In general prestressing the improvement

of performance

Recently attempts have been made to improve the tensile strength of concrete by the introduction paragraph.
1.5 Fibre-Reinforced Concrete

of fibres

in the concrete mix.

This is discussed in the following

The problem of the low tensile strength of concrete has occupied the attention of many engineers in the past. In 1910 (19) it was suggested that the characteristics

of the concrete could be greatly improved by the addition of short pieces of steel mixed in the concrete matrix. different Recent attempts in this field have incorporated Indeed, fibre-reinforced

types of fibres in a cement or a concrete matrix.

cement and concrete has opened up a whole new field of concrete technology which could result in new concepts of design for concrete structures. , Generally, when fibres are incorporated in a cementatious improvements improvements in the structural properties of the matrix

matrix,

certain These

can be achieved.

however mainly depend upon: of the fibres with regard to their modulus of elasticity, extensibility and bond efficiency. of distribution of the fibres in the

1) The properties " tensile strength,

2) The volume content and uniformity matrix. 3) The degree of effectiveness

of the fibres in the matrix; These are: -

this depends upon

the technique of mixing employed.


a. b. c. Three-dimensional Two-dimensional Unidirectional random random distribution.

distribution. distribution.

In the following

these are discussed at further

length.

Three-dimensional
for some time

random

distribution

techniques have been employed


in a cement or concrete improvement matrix. can

incorporating

various

types of fibres matrix,

When steel wires be obtained

are used in a concrete ductility

a considerable strength

in the impact,

and flexural

(20) (21) (22) (23).

There are certain points,

however,

which need to be considered when fibres are in plain or reinforced

to be dispersed in three-dimensional concrete members. These are: 1) Difficulty

random distribution

of mixing and handling the fibres.

In some instances

a special

technique may have to be adopted in their use (24) 2) A good workability better for distribution be to ensured a needs This may be achieved by: of fibres

and compaction of the matrix.

(a) Using a low fibre content and aspect ratio which may tend to reduce the strength of the matrix
(b) By controlling in the concrete the size and quantity mix (23)

(21) (23)
of the coarse aggregate used

(c) By the addition of liquid additives and a partial cement by pulverised 3) Flexural fuel ash content (23)

replacement

of

strength and crack control depends on the bond characteristics In the case of steel fibres this may

between the fibres and the concrete. be relatively


4) The possible This

inefficient
corrosion

due to the smooth surface of their finish.


of the steel wires to those wires in the long term bridging crack (22) (25). widths in the

may apply especially

concrete.

5) The fibres will be dispersed in areas where they are not mostly needed, e. g., in the compression zone of the flexural member, i. e., full utilization

of all the fibres would not be achieved. 6) The cost of fibre-reinforced concrete, when incorporating 2% of steel wires.

Is about five times that of plain concrete (22) (25) 7) In three-dimensional which effectively random distribution of fibres in a matrix the fibres

carry the stress in any given direction

are between

0-20% of the total volume included in the mix (25).

In reinforced

concrete,

where only the flexural

strength of the concrete of fibres be not may

needs to be improved, very efficient,

three-dimensional

random distribution

and an economical design may not be achieved. The material

This conclusion

is based on the present state of Inlowledge. stage, and a more efficient understanding

is still in the development

use of the material

will result when there is a better Recently. the Delft about in this and other

of the techniques in dealing with fibres.

Conference (26) (27) has shown how much Engineers were enthusiastic developing and processing field of research, parts of the world.
In two-dimensional of the fibres volume carrying random stress compared distribution of fibres in one plane

this material.

The future seems promising out in the United Kingdom

as work is being carried

the efficiency

in a given direction with 0-

is between

30 - 37% of the total (25). When in a

of the fibres fibres

20% for the three-dimensional random to impact,

high-modulus cement stresses a concrete matrix,

are dispersed

in two-dimensional resistance

distribution fire

the composite

has a better

and tensile in

(28) (29). matrix,

Two-dimensional

distribution

of fibres for

can not be employed to produce high

but has been used successfully

some time

quality fibrous

cement composite units.

Glass or asbestos fibrous cement composite

units are good examples of this. Unidirectional distribution of fibres in a matrix would appear to give the most

effective use of the material,

but it is not possible to employ this method with the It is possible that in the future this type of fibre-reinforced concrete when a

existing techniques of mixing the fibres. of distribution

may dominate the construction

better technique of mixing is lnown. At the University of Salford a new technique in construction employing

precast units of fibre-reinforced

to reinforcements cement . as surface _ concrete members has been evolved by Dr. Dave. This type of fibre-reinforced behaviour forms a major part of the

cement composite concrete and its structural work presented in this thesis.

10 -

1.6

The Need for this Investigation

In carrying

out this investigation

the following

points were considered: amount of research had reinforced considering static,

1) It was realised that in the past been conducted into the flexural lightweight concrete members.

only a limited

behaviour of full-scale A study in this respect,

fatigue and sustained types of loading was deemed to be necessary.


2) The maximum allowed in specified characteristic steel stress for reinforcing bars

CP110 is 425 to 460 N/mm2, used. However, states it is believed

depending

on the size and type can be increased of the

of the bars without structure

that this limit

the limit

of ultimate

strength

and serviceability

being reached. concrete has a low tensile amount of cracking strength and modulus of

3) Since lightweight elasticity, high working controlling

a considerable steel stresses

and deflection is expected.

when employing Means for In

in flexural and cracking

members therefore

this deflection

seems necessary.

this respect

a new form of construction

is to be employed,

in which

precast fibre-reinforced as a surface reinforcement 4) Experimental

cement units in the form of thin channels are used at the tensile zone of the concrete members. values for constants for various that are

data is necessary to ascertain relationships

to be used in proposed empirical behaviour.

aspects of

CHAPTER PREVIOUS INVESTIGATIONS

TWO

2.1 2.1.1

Use of Iiigh-Strength General In the past

Steel in Structural

Lightweight

Concrete

mild steel has been usually employed as reinforcement Very limited tests have been carried

in

lightweight the flexural

concrete members. behaviour

out into

of such members when reinforced

with high strength steel

loading. fatigue types of sustained and especially under In general reinforced most of the investigators have considered the behaviour of for reinforced light-

normal weight concrete as a basis of comparison

weight concrete. In the following investigations paragraph a brief review of the available data from previous however, covers only lightweight aggregate. concrete

is presented.

This,

made with various types of synthetic lightweight


2.1.2 Summary of Past Research

Many researchers and structural

have been involved for some time

investigating

the physical

lightweight types of of various aspects It would be difficult properties

aggregate concrete, which can regarding the

be used in construction. physical and structural


t

to draw conclusions

of lightweight aggregate. However,

concrete due to the great variation This is generally beyond the relevant to the present extensibility,

in the choice of produced lightweight scope of the present investigation. investigation

the properties strength,

are the modulus of elasticity,

flexural

shrinkage and creep. As regards the modulus of elasticity investigators similar of lightweight concrete, many

have compared this value with that of normal weight concrete of The following table is prepared to illustrate this difference.

strength.

Reported by
J. J. Shideler, 1957 (11)

E (lightweight) x 100 E (normal weight)


53 - 82

Types of lightweight aggregate Various types produced in U. S. A.


Ditto

J. A. Hanson, 1958 (30) and 1961 (31) A. Short, 1958 (8) D. C. Teychenne, 1967 (9)

50-67 33.350 - 60 67

Various types produced in U. K. Sintered expanded and sintered pulverised fuel ash.
value

G. B. Welch 1964 (32)

& B. J. F. Patten

No specific given about 60

Sand - expanded shale.


Sintered expanded clay

R. H. Evans & T. R. Hardwick 1960 (33)

One of the very lightweight concrete

early

investigations was carried

on the structural out by Richart of Illinois,

behaviour

of reinforced

members

and Jensen in 1931 (34). for 32 lightweight and

They reported

test results

at the University

normal weight concrete beams reinforced

with mild steel.

These beams

(152.4mm wide, x 304.8mm deep, with 2438.4mm span) were tested under short term static loading. compressive natural sand. The experimental effect on the ultimate under reinforced, resistance results showed that the compressive strength had little The main variables in the investigation were the concrete . strength and the effect of replacing fine lightweight aggregate by

load of the beams.

This was possibly due to the beams being

where the ultimate

strength depended mainly on the moment of

of the steel. the deflecticrisof lightweight concrete members are greater than However, when the fine

In general

those of the corresponding lightweight deflection

normal weight concrete.

aggregate was replaced by natural sand, the difference between the of the lightweight and normal weight concrete bemis was reduced. The

concrete cylinder

compressive

strength of the beams considered ranged between

(12 - 32.6) N/mm2.

It was also concluded that, due to the lower modulus of

elasticity

of lightweight

concrete,

the neutral axis of such beams was lower than

those of normal weight concrete. Hanson carried laboratories out a series of tests at the Portland aspects of lightweight Cement Association In 1958 (30) and lightweight The beams

on the structural

concrete.

1961 (31) he conducted shear investigations

on a total of 57 reinforced static loading tests.

and normal weight concrete beams in short-term

were tested using span lengths of 1981.2mm and 3048mm with a cross section of 152.4mm wide and 304.8mm deep. used was 334 N/mm2 for the lightweight
The strength span. rmin variables which

The yield stress of the steel reinforcement concrete beams.


were the concrete compressive

in the investigation (20.7

ranged between

62 N/mm2), -

the steel area and the shearing

The calculated deflection based on the "Cracked transformed

cross section"

deflection the that the with reasonable actual method could predict method showed accuracy. The deflections of lightweight concrete beams at loads where initial diagonal

than those 35% be found to to 15% of normal weight greater cracldng occurred were concrete beams. The modulus of elasticity of the lightweight concrete, however,

was between 50% to 66% of that for normal weight concrete. Short carried out an extensive research investigation and structural properties at the Building Research aggregate

Establishment concrete.

on the physical.

of lightweight

In 1959 (8). he reported results lightweight

of tests carried

out to study the flexural

behaviour of reinforced

concrete beams.

A wide range of lightweight percentages of

aggregates produced in the U. K. were covered, mild steel employed for the main reinforcement.

with different Similar

beams made of normal

weight concrete were also tested so as to form a basis of comparison. In short-term static tests the deflections of lightweight concrete beams at of normal

the working load were found to be 10%to 50% greater weight concrete beams. The cracking for lightweight

than the deflections

concrete beams was also

found to be more severe than for similar In sustained loading tests and creep, of some lightweight

normal weight concrete beams. due to shrinkage

the total permanent deflections

concrete beams were 15% to 45% greater than those

14 -

of normal weight concrete beams when measured under the same conditions of temperature and humidity.

As regards the use of high strength steel, Short suggested that, this should be limited by the values of deflection and cracking, other wise, the values of

span/depth ratios given in CP114 / 1957 for normal weight concrete should be multiplied by a factor of 0.75. at Leeds University

Evans, also conducted an extensive amount of research on the behaviour of reinforced and prestressed (33) lightweight

concrete. and normal

In 1960 Evans and Hardwick weight concrete beams reinforced

reported tests on 34 lightweight

with mild steel.

High strength steel was also

used for normal weight concrete beams. ultimate strength, deflection

It was mainly intended to study the of concrete beams made

and cracking characteristics

from sintered expanded clay aggregates. In short - term static loading tests lightweight the immediate deflections of reinforced

concrete beams were 10% to 25% greater than those for corresponding It was also observed that the cracks in lightweight

normal weight concrete beams.

concrete beams were about 50% wider than those in normal weight concrete beams, and also that they were spaced about 60% closer. At a working steel stress of concrete beams

138 N/mm2 the average crack width and spacing in lightweight was 0.07mm and 90mm respectively. They also concluded that the ultimate

strength of lightweight

concrete beams

could be obtained with reasonable accuracy by using Whitney's


In 1964 loading beams tests were Evans and Orangun (35). reported the results

theory.
- term beams. static The

of short concrete

on 28 sintered

pulverised

fuel ash "(Lytag)" -

229mm wide and 381mm deep with 2438mm

span.

Two types of steel were usedmild

steel with a yield stress of about 275 N/mm2 various

and high strength square twisted bars of about 410 N/mm2 yield stress, percentages of steel and a wide range of concrete compressive adopted. They concluded that the immediate deflection

strengths were

be load could at working

calculated by the conventional transformed

cracked section method employing a of

modular ratio of 17, a modular ratio of 30 was suggested for the calculations

- 15 -

deflection under sustained load. At a steel stress of about 205 N/mm2 beams reinforced with the largest

the maximum

crack widths in the

size of the square twisted bars (31.75mm) did and cracking, a working steel stress

not exceed 0.254mm.

To control deflection

of 186 N/mm2 was suggested. It was also concluded that the ultimate be satisfactorily loads of "Lytag" theory. concrete beams could

calculated by using Whitney's

The design by either the safety factor. tests carried

load factor method

or the elastic method would ensure the required

In 1967 Evans and Paterson (36), presented a paper describing out to investigate fuel ash "(Lytag)" the long term deformation characteristics

of sintered pulverised and bending.

concrete members in both axial compression

The creep in bending was investigated

in two types of beams, the first being reinforced with different

190.5mm wide and 416mm deep with 2438mm span

dimension had high the sectional of a cross second strength steel, of percentages (101.6mm x 101.6mm) with 660.4mm span reinforced steel. with a similar percentage of Normal

The high strength steel used had a yield stress of about 410 I1/inm2.

basis form to beams tested of comparison. a so as also were weight concrete For the first final to initial type of beams it was observed that after 750 days, the ratio of concrete beams ranged between (2-2.33),

deflection

of lightweight

this however was (2.34 - 3.09) for normal weight concrete.


The maximum to be greater, weight concrete crack width at working load in lightweight concrete was found

but it did not exceed a value of 0.2mm in an exposed condition (37).

recommended

for normal

For the second type of beams was made with normal concrete rapidly after weight

a direct

comparison

for long term

behaviour

concrete.

The initial weight

deformation concrete.

of the lightweight This ratio dropped value beams

was 1.5 to 2.0 times within 100 to 200 days, of 300 days,

that of normal

and then levelled the total weight

off to a reasonable of lightweight

constant concrete

a period

leaving

deflection concrete.

marginally

higher

than that of normal

Generally the improvement

in the long term deformation

of lightweight

concrete

beams was mainly due to the following: -

1) Greater improvement

in the concrete compressive

strength for light:

weight concrete with time. 2) The low value of the modulus of elasticity of lightweight concrete resulting

in a lower neutral axis, consequently a lower average concrete stress for a given applied load.
3) A slow rate tensile Roberts, of shrinkage which will enable the concrete resistant to crack to gain its full formations. College,

strength,

hence it will

be more work

in 1962 (38), presented

carried

out at Northampton

investigating

the bond and crack characteristics

of reinforced

sintered expanded

loading term tests. beams tested static short under clay aggregate concrete The beams were 190.5mm wide and 304.8mm deep; two types of reinforcement N/mm2 275 of stress steel with a mild yield used, were
with 0.2% proof stress of 410 N/mm2.

and high strength steel

It was concluded that steel embedded in lightweight bond strength and therefore bars.

concrete had adequate

there was no need to provide end hooks with deformed

At the level of the working moment spacing of cracks in both lightweight

and

deformed bars 127mm beams were used, when was about normal weight concrete and 165mm when mild steel was used. The average crack width if deformed bars The same crack

206.8 N/mm2. 0.1mm of steel stress a at were used was about

N/mm2 138 when mild steel was used. of stress steel width was observed at a It was concluded, however, that deformed bars with a working concrete beams.
for a series of tests on long - term aggregate

stress of

206.8 N/mm2 could be used in lightweight


In 1968 Kanoh (39) deflection of reinforced presented concrete results beams

employing

artificial

lightweight

made in Japan.

Thirty-one

reinforced

concrete beams were tested, incorporating

various

percentages and types of tension reinforcement. cylinder compressive strength of about 30 N/mm2 It was concluded

The concrete used had an average and a modulus of elasticity of

about 13.8 KN/mm2. deflection of lightweight

that the general behaviour of the long - term to that of normal

aggregate concrete beams was similar However,

weight concrete beams.

it was observed that the ratio of final to initial

deflection weight

for lightweight beams.

concrete This

beams was some what less than that for normal agrees well with that stated previously

concrete

conclusion

by Evans (36).

Swamy and Ibrahim, carried

in 1974 (40), presented results

of an investigation characteristics of U. K. )

out to study the short and long - term deflection and prestressed aggregate. reinforced

reinforced lightweight

concrete beams made from expanded slate ("Solite",

The ordinary

beams were 127mm wide with an effective

depth

equal to 156mm and 2286mm span. Of the ordinary reinforced beams, three were made with all lightweight replacement of lightweight fines with

aggregate and one was made with partial natural sand.

A beam made from normal weight concrete was also tested so as In these beams various percentages of steel the steel bars used had a minimum yield stress of

to form a basis of comparison. reinforcement 410 N/mm2. were employed;

The design concrete cube strength at 28 days was 51 N/mm2.


of 3.16% was employed, with partial concrete the beams made from of lightweight all fines with of 434, concrete

When a steel ratio lightweight natural concrete,

concrete weight

replacement

sand and normal

gave a span to deflection This

ratios

317 and 378 respectively

at the design rigidity

load. compared

showed that lightweight weight

beams had adequate flexural

with normal

concrete.

The maximum crack width at the design load in lightweight was nearly equal to that in the corresponding working steel stress of 207 N/mm2, a value of 0.15mm. In long - term loading, initial

concrete beams At a

normal weight concrete.

the maximum

crack width did not exceed

after a period of 193 days, the ratio of final to concrete, and

deflection was 1.94 for a beam made from all lightweight replacement of lightweight

1.72 for a beam made with partial sand. The investigators of lightweight

fines with natural

have shown that the short and long - term deflection

concrete beams could be calculated by employing CP110 and ACI

318-71 standards.

2.1.3.

Conclusions As stated before,

From

Past Research of the lightweight carried aggregate

due to the wide variation amount of tests concrete the flexural

used,

and also due to the limited

out using high strength it would be difficult for these members. have presented beams. to

steel as reinforcement draw a definite

in lightweight regarding

members, behaviour

conclusion

It has also been noticed their data in a form

that most of the previous with normal

investigators weight concrete

of comparison

However,

the following points on the general behaviour of reinforced

lightweight

concrete members may be drawn: 1) The ultimate load of lightweight concrete members could well be predicted The design by either the elastic safety factor.

by the existing ultimate

load theories.

theory or the load factor method would ensure the required 2) The instantaneous deflection of lightweight

concrete beam is greater than

that of normal weight concrete; this however depends mainly on the modulus of elasticity lightweight transformed for both normal and lightweight concrete. The deflection of

by the be conventional calculated concrete members can cross section method using a suitable modular ratio. concrete beams in general, This, however, are wider

3) The crack widths in lightweight

than those in normal weight concrete. the low tensile

is possibly due to

by tensile the stresses additional caused and strength concrete beams. the total deformation of lightweight concrete

shrinkage in lightweight 4) In sustained load tests "

members may not be significantly

greater than that of normal weight concrete. as

Thi; however, again depends on the shrinkage and creep characteristics well as on the modulus of elasticity concrete under consideration. 5) At this point of time there is not enough data available regarding behaviour of reinforced of loading. The maximum steel strength used in the past was 410 N/mm2. follow that the above conclusions apply only to lightweight This may lightweight the of the particular type of lightweight

concrete members under fatigue type

concrete members

reinforced
2.2.

with steel bars of strength not greater


Employed Reinforced to Control Cracking

than 410 N/mm2.


and Deflection in Flexural

Approaches Members

with High-Strength

Steel

2.2.1

General

The problem of concrete weakness in tension is commonly overcome by the traditional concrete; use of reinforcing the latter bars, or by prestressing the reinforcement in

method is rather costly. the problem of concrete weakness in tension leads to and deflection, especially when high working steel

In flexural a considerable

members

amount of cracking

stresses are employed. Recently various methods to try and overcome this deficiency of the

concrete have been proposed by various investigators.


this problem (41) and fibre employing methods reinforced of employing concrete a closely

In a direct approach to
wire reinforcement method and These

spaced continuous (42) (20).

have been suggested at the soffit

Another fixed (43).

a steel channel placed

of a concrete

member

held in position methods 2.2.2.

by means of shear connectors in the following

has also been adopted

are discussed Closely

paragraphs:

Spaced Continuous

Reinforcement

In conventional reinforced tensile zone at relatively in tension,

concrete members

cracks would form in the the area of concrete

low tensile stresses,

for this reason

which forms more than half of the total area is usually ignored for Past research (44) (45) indicated that good control on the width diameter reinforcing

design purposes.

and spacing of the cracks can be obtained by using smaller bars well distributed over the effective concrete area.
improvement members. in the cracking This is mainly bars.

This concept, however,


mechanism for conventional

has not led to a substantial types of reinforced limitations concrete

due to the traditional

of size and spacing

of reinforcing

An early suggestion for the use of a more fine and"closely spaced reinforcement was first This carried made by Nervi (46). This was reported by Romualdi and Batson (41). of an experimental slab reinforced investigation with a closely

suggestion came

out as a result

out on the behaviour

of a concrete

spaced steel wire the cracks point.

mesh.

The spacing until

of the wires

was about (10.16mm), nearly mechanics,

and

were not observed this research,

the steel was stressed of fracture

to its yield a study was

Following

in the field

made on the effect projected structures. spacing crack

of riveted

stiffeners

placed perpendicular arresters that: "for

to the line of the in stressed certain plate

extension;

these acted as crack (47) indicated

In this

study Romualdi a condition

stiffener is arrested,

and rivet from

sizes,

could be obtained larger

whereby

a crack

or prevented crack

enlarging

at stresses

than those required

to extend the

in the absence of stiffeners".

It was thought, however, concrete construction is similar

that this mechanism could be applied to reinforced in plate structures From this concept spacing less than

where the role of the riveted stiffeners bars in a concrete matrix.

to that of reinforcing

Romualdi and Batson (41) concluded that "at a reinforcement a certain critical groups of bars". Romualdi and Batson behaviour of reinforced (41). presented an experimental value, all potential

between be adjacent contained could cracks

investigation

on the flexural

concrete beams with a closely spaced continuous

reinforcement

A total of 19 small scale beams were tested under short - term static loading tests. The reinforcements employed were steel wires of 1.59 and 0.89mm diameter and 758.6 N/mm2 respectively. The wire The

with a yield stress of 634.5 N/mm2 reinforcement horizontal 4.242mm. diameter

facilitate to into a mesh was woven

placing in the mould.

spacings employed between these wires were 12.7mm, The vertical

8.45mm and

distance between the wires was varied according to the The concrete This was with

of the wires used, for keeping the same percentage of steel. wet, and the only aggregate used was sand.

mix used was relatively mainly to facilitate ordinary

pouring the concrete.

Two of the beams were reinforced

deformed bars of 9.5mm diameter

and 275 N/mm2 yield stress employing

a 1.47% of steel area; this percentage of steel was also employed for most of the wire reinforced beams. the ultimate

It was claimed that for a wire spacing of about 5.08 to 7.62mm, strength of the beams was substantially strength was about 50%. increased;

this increase in ultimate

The beams did not exhibit any noticeable cracks until the very final stages of loading. reinforced The neutral axis was also observed to be lower than that for conventional concrete beams. that for the failure mechanism obtained applications. was that the strength this technique

It was concluded

could be used for a wide variety An important ultimate theories.

of structural

feature claimed for this type of construction

strength was much greater than that predicted by the ultimate

This increase in strength was dependent on the spacing of the reinforcement; load.

the lower the spacing the higher was the ultimate

On this point Broms and Shah (48) pointed out that to maintain the same percentage of reinforcement reinforcement Romualdi and Batson used a small diameter strength values. These values, however, the ultimate were strength

with high ultimate

not used in their calculations.

Broms and Shah recalculated

of the beams based on these higher values, the spacing of the reinforcement

and found that the effect caused by strength was negligible.

wires on the ultimate

discussing the paper by Romualdi and Batson (41) j bond between the the besides the strength, that ultimate spacing effect on suggested Abeles (49) on the same line, the wires and the concrete should also be taken into consideration. efficiency of beams with closely spaced reinforcement The high

may not be obtained, insertion and

because the space between the wires hardly allows satisfactory compaction of the mortar Furthermore difficulties around the wires.

the method has a restricted

usefulness due to practical to offer in economical savings due

in casting and also has very little

to the high cost of material However,

and labour involved (50). widths may

the concept of using this technique for reducing crack

well converge on an idea to employ a new type of construction concrete) which is discussed in the following paragraph.
2.2.3 Fibre Reinforced Concrete

(fibre reinforced

The suggestion to use short pieces of steel to improve the properties concrete was first made in 1910 by Porter

of

(19) as reported by Hannant (22).

This was then followed by Biryukovitch

et al (42) and Romualdi and Mandel (20)

when they initiated

a more practical

investigation

in this field.

In an early study (41) (51) it was suggested that by the use of closely spaced continuous wire reinforcement a substantial increase in the tensile cracking the

strength of concrete could be obtained. growth of cracks which originate theoretical study revealed

This was accomplished by arresting flaws in the concrete. A

from internal

that the tensile strength of concrete was proportional

to the inverse

square root of the wire spacing. technique. Romualdi and Mandel (20) considered

To provide a more practical

that the same concept of crack arrest directly fine of wires short pieces important

be by adding could achieved mechanism It was considered so as

into the concrete mix.

that the percentage of wires added to the mix should be sufficient, In their theoretical

to obtain an adequate average spacing of the wires. they derived an approximate percentage and size of fibres. expression

study.

for the effective wire spacing for a given part of their study included

The experimental

short - term static tests on sand-cement


Most of the beams 905.2mm span length tested in bending

mortar beams.
were 44.45mm wide, 76.2 mm deep with of steel wires. a basis of

reinforced

with various

sizes and percentages

Plain beams without comparison.

any reinforcements

were also tested

so as to form

The ratio for the flexural

tensile strength for beams reinforced

with fibres

to plain beams varied between 1.2 to 2.52 depending on the aspect ratio and percentage of fibres used. A good agreement was obtained between the earlier for the tensile cracking the material applications Untraur stresses and the observed values. theoretical prediction that

It was concluded

had unique features which could be used in a wide variety where fatigue, thermal shock and cracking are important

of structural considerations.

and Works (52),. in their discussion on Romualdi and Mandel's paper] of the claim regarding the effectiveness of the addition of fibre strength as

doubted the validity reinforcement

to the tensile cracking

strength.

The tensile cracking

defined by Romualdi and Mandel is the stress which corresponds deviation from linearity of the load-deformation characteristic

to the first of the material.

If this definition cracking however,

is to be applied, their results

show little

increase in the tensile They did,

strength resulting

from the addition of fibre reinforcement.

confirm Romualdi and Mandel's results that the addition of fibres improved the tensile strength of the concrete. (53), discussing the same paper, placed more emphasis on the stress of fibres within the matrix. He suggested that the low

substantially Abolitz carrying

efficiency

effectiveness

of randomly distributed

wires would not be a favourable factor in concrete construction.

economic considerations

with regards to their use in reinforced

Agbim (54), in discussion of the same paper, drew attention to the difficulty which he experienced in mixing the fibres. He also pointed out that improvements

in the strength over plain concrete were mainly obtained in short - term tests. The time effect on these improvements should also be taken into consideration, obtained may be reduced with as

in some cases it was found that the improvements time.


Due to the conflicting many attempts conclusions drawn

on fibre

reinforced of fibre

cement

and concrete, and

have been made to study the mechanism the various properties of fibre

reinforcement,

also to investigate

reinforced

cement

and concrete.

Shah and Rangamin 1971 (55), reported properties reinforcing of fibre reinforced

an investigation

on the mechanical

concrete and mortar. steel wires.

The intention was to study the

action of randomly distributed various

In this investigation.

small fibrous reinforced flexural

concrete and mortar strengths in

beams were tested to study the tensile, relation to conventional reinforced

and compressive

concrete beams.
of the specimens with a tensile and 25.4mm. tested had a crossstrength of 828 N/mm2.

The steel wires sectional The length dimension

used for the majority of (0.254mm varied

x 0.254mm) 6.4mm

of the wires

between

From the results of wire reinforcement

obtained Shah and Rangan (55)

concluded that the addition effect on

to concrete and mortar beams had a negligible in the matrix.

the load at which cracks initiate

This effect was observed for Considering other investigator's

specimens tested with a steel content up to 1.5%. results, reinforced

Shah and Rangan also suggested that this could apply to specimens with up to 4% of steel wires. They also concluded, that the spacing had

little

effect on the ultimate The conclusions

tensile strength.

suggested by Shah and Rangan differ from that suggested to the effect of addition of fibres

earlier

by Romualdi at al (20) (41) with regard-

and the effect of the spacing of the fibres on the tensile strength of the concrete.
In a comparison dimensional random between distribution fibre reinforced concrete (incorporating reinforced three concrete,

of fibres)

and conventional

Shah and Rangan observed conventional that of fibre reinforcement reinforced

that for a 1% reinforcement gave a maximum flexural

content

beams with a times

load of more than three

concrete.

Swamy and Lankard (56) recently presented a paper describing applications of steel fibre reinforced concrete both in Britain

several practical

and the U. S. A.

Among the practical

, pplications discussed were the use of steel fibrereinforced concrete pipes, pavements, constructions. concrete to resist overlays,

concrete in a slab deck of a car park, marine structures

and mining and tunnelling

It was concluded that the ability cracks propogation and its resistance

of steel fibre reinforced to thermal

shock, fire and dynamic loads material. carried out in the

make fibrous concrete a unique new construction

It is not in the scope of this research to cover investigations field of fibre reinforced concrete; however,

there are very few examples on the members.


on the use of steel wires in

use of fibres in reinforced


In 1972 Hannant reinforced deep with stress lightweight 1.83m (57)

concrete flexural
presented

an investigation A total 2-13mm used were

concrete

beams. with

of four beams diameter bars

(127mm wide and 230mm of 375 N/mm2 yield

span) reinforced

were tested.

The steel wires

of 0.38mm

diameter

x 25ram length,

two percentages

of fibres

being employed;

these were

1.2% and 1.6%a by volume.

Conventionally fibre reinforcement

reinforced

normal

and lightweight

concrete beams with no

were also tested to form a basis of comparison. concrete beams without fibres were between With the addition of to that of normal

The deflection of the lightweight

40% and 50% greater than those for normal weight concrete. fibres to lightweight weight concrete.

concrete beams, the deflection was similar

The cracking tensile stresses for the fibrous lightweight approximately twice that for the plain lightweight

concrete beams were

concrete beams; the crack widths

observed were lower in the fibrous lightweight Hannant

concrete beams. and cracking may well lead to

suggested that the saving in deflection

the use of high strength steel as a reinforcement At the University of London. Samarrai

in concrete beams. (58) have carried out investigation

and Elvery

into the use of steel fibres in reinforced

concrete members.

It was mainly intended concrete

to study the effect of steel fibres on controlling members tested in uniaxial has also been studied. Duoform) tension.

crack widths in reinforced

The possibility

of using high strength steel

Variables

that were also studied included type (plain or

size and quantity of fibre and type of main tension reinforcement. results showed that the addition of steel wires to reinforced

The experimental

concrete had substantially before a particular

increased the stress in the main tension reinforcement The improvements were also found to be

crack width occurs.

greater when deformed high strength steel bars were used for the main tension reinforcement. 2.2.4 Composite Concrete Construction Using Steel Channels emerged from combining the

The idea of employing this method of construction material of the following types of construction

(i) The use of high strength steel in flexural (ii) The composite construction section is incorporated haunches. It was thought (43), for a deep haunch section, instead of a beam section, effective the bursting

concrete members

with deep haunches, where a mild steel beam

by shear connectors to the soffit of the concrete

that by using a channel section

or spalling of the concrete due to the low The

side cover to the shear connectors in the haunch could be prevented.

channel could. also provide an adequate rigidity steel could be used. On the application

to the section so that high strength

of this type of construction

(59)

a total of 9 beams were

tested in short - term static loading tests. incorporating different

These beams were of 4572mm length

dimension and size of the concrete cross section and the steel

channel.

Various amounts of steel were used, the ultimate and 9GGN/mm2.


so that at the working

steel stresses employed

ranging between 283 N/mm2


The beams were stress in the channel

designed did

load condition, stress.

the maximum

not exceed about 90% of its yield

From the results well below the limits

obtained

cracking and deflection However,

at the working load were with mild

specified in CP110.

beams reinforced

steel and the shallow section beams showed a considerable deflection respectively.

amount of cracking and

It was concluded, however, used while satisfying the limit

that steel of 828 N/mm2 ultimate

stress could be

states of serviceability indeeddhas new merits however,

at the working load conditions. of potential importance in the

This type of construction, field of concrete technology; which should be carefully (1) The construction

there are certain shortcomings these are: -

and doubts

considered, has little

resistance

to fatigue;

tests have showed that

after about 500,000 cycles failure

would occur. resistance to fire

(2) The steel channel being exposed would offer little 2.2.5 Work at Salford At the University of Salford

an extensive amount of research has been cement and concrete.

conducted into the structural

aspects of fibre reinforced

In 1971 (60). research was conducted on the use of chopped steel wires in a concrete matrix. fibres The major part of the work was carried out using two sizes of

0.5mm diameter x 38mm length and 0.38mm x 25mm length, the wires were The percentages of steel used varied between 1% and 3% by

plain and Duoform. volume. Properties

investigated

were the cube crushing strength, In carrying

flexural

strength, investigation.

direct and indirect difficulty

tensile strengths.

out the experimental

of mixing and compacting the wires

was experienced.

It was also observed

that the greater the aspect ratio of the wires the greater to knit into balls. The results and the flexural showed that properties certain improvements concrete.

is the tendency for the wires

were obtained in the compressive was

of the fibrous

The peak of the improvement

at a spacing of about 4mm, below which Comparing the performance spacing with that of plain concrete, for the compressive, and 20 respectively. In 1972 (24) of normal a research was carried flexural,

the results

obtained became erratic.

4mm fibres fibrous at steel concrete containing plain of the improvements expressed as a percentage

indirect

and direct tensile strength were 11,83,57

out investigating

the structural

properties

and lightweight

concrete and cement mortar

reinforced

with three dimensional

random distribution

of asbestos fibres.

Various types and percentages of asbestos fibres by weight of cement were incorporated The workability . as the amount of fibres increased. in the matrix Certain improvements cement mortar of the mixes was observed to decrease

in the compressive,

tensile and impact strengths of the These improvements were fairly

and concrete were also obtained.

small and largely

influenced by the volume content of fibres,

the type of aggregate

and the water cement ratio of the mix. Other reasons why the improvements , 'ere low are as follows: (i) The low directional of fibres. (ii) The short length of the asbestos fibres used (less than 6mm) allowed little stress transfer from the matrix to the fibre. that ways of preferentially aligning the efficiency of three dimensional random distribution in the properties of the fibrous matric as

It was suggested in this investigation fibres should be found.


Investigation partially has also been carried composite tensile T-beams zone (61).

out into the flexural with three This

behaviour

of full

scale

prestressed in their

dimensional

random

distribution

of steel fibres reinforcement prestressed

investigation in cracking

showed that steel fibre and deflection of partially

can result composite

in some improvement

T-beams.

Conclusions drawn from previous investigations were obtained in the properties However,

show that substantial

improvements

of the concrete when high modulus fibres were used. could be utilized in

it is still doubtful to what extent these improvements in reinforced

reducing the cracking and deflection

concrete members.

The material,

is still in the development stage, and there are not enough the effect of the fibres on the control of deflection and

data available regarding

long term loading conditions. especially cracking, under . The initial work at Salford and other places has revealed that the addition of fibre reinforcement in three dimensional in reinforced random distribution concrete construction. is not a practical or

an economical proposition

A new concept has

been developed at Salford which involves the use of precast units of fibre reinforced cement cracking as surface and deflection reinforcement (62) (63). cement composite concrete construction in which for concrete members with a view to controlling

This type of fibre reinforced

precast asbestos cement units are used as surface reinforcement work presented in this thesis.

forms the basis of

CHAPTER MATERIALS

THREE BEAMS

USED AND DESIGN OF TEST

3.1

General The main aim of this investigation was to study the structural behaviour of with high - strength

ordinary steel.

and composite lightweight

concrete members reinforced

It was deemed necessary to provide experimental

evidence regarding

the cracldng

and deflection behaviour of these members for a comparison to be made with the predicted values. Furthermore the experimental relationships. consisted of testing full-scale reinforced results would also help to ascertain

values of constants in empirical The programme lightweight

of investigation

concrete beams, containing different

types and percentages of steel bars

for the main tension reinforcement. The size of beams was chosen to represent construction and to enable the performance members commonly used in building examined.

be to critically of such members

The beams adopted were 150mm wide, 300mm deep and 5m long. simply supported over a span of 4.5m. shown in Fig. 3. to the ordinary as integral The loading arrangement

These were

of the beams is

The composite beams were similar

beams in every respect, parts at their tensile zones. the and test

except that the f. r. c. units were incorporated Control specimens, compressive procedures

e. g. cubes and prismswere

also cast to ascertain The design, manufacture

and tensile strengths of the concrete.

employed in this research were in accordance with the specifications Standards. used and the design considerations

given in CP110 and the various British In the following

paragraphs the materials

for the f. r. c. units and the test beams are discussed.


3.2 Design Considerations

3.2.1

Ordinary Reinforced As mentioned in

Lightweight

Concrete Beams

3.1, the beams adopted were 150mm wide, 300mm deep with beams. The nominal yield or 0.2%

4.5m span and were designed as under reinforced

proof stresses of the reinforcing 550 , l/mm2,590

bars used were 275 N/mm2,410

N/mm2,

N/mm2 and 875 N/mm2;

the steel ratios employed were 0.582%, are given in

1.044%, 1.483% and 1.643%. tables 1,2

Details of the beams reinforcements

and 3; these are also shown in Figs. 4 and 5.


characteristic cover cube strength reinforcements members of concrete was 50 N/mm2. this being the conditions of exposure The

The designed nominal minimum concrete allowed

to the main

was 35mm, in severe

for lightweight

concrete

as per CP110.

For calculating of ultimate design. strength,

the design working moment of the various beams, the limit as per CP110, was considered to be the main criterion a rectangular - parabolic stress distribution for

state

In this calculation

incorporating The is

the partial

safety factor for the concrete at the ultimate relationship for the lightweight

condition was used.

adopted stress-strain shown in Fig. 6. Fig.

concrete in compression

7, shows the stress-strain

curves as per CP110. resistance moments would

The beams being under reinforced,

their ultimate

be those based on the yielding of the steel. The values of the ultimate MU = Ym ym where Z= =partial dl -P safety factor for steel (1.15) fy As (3.2) As Z follows: be as calculated moments can (3.1)

fcu ayb _m The values of and a can be derived by considering stress distribution for the concrete in compression; a rectangular - parabolic

this is shown in appendix B.

The working moment was obtained by subtracting load of the beam from the ultimate values 1.4 and 1.6 being the partial resistance

1.4 times the moment due to dead

moment, then dividing by (1.6), the

safety factors for dead and live load respectively. for the lightweight concrete was 1800 kg/m3. and

The air dry density used in the calculation The details of the reinforcement

and the calculated values of the ultimate

working moments for the various beams tested under static, of loading are given in tables 1,2 and 3 respectively. is given in appendix C.

fatigue and sustained types beam a of analysis

A typical

For the shear reinforcements

rectangular

mild steel stirrups

of 6mm diameter

were provided for all the beams at a uniform

spacing of 100mm.

None, however,

was provided in the constant moment zone, mainly to avoid any effect which the stirrups might have on the initiation and distribution of the cracks. This practice

has also been adopted by previous investigators stirrups was carried

(35) (64) (65).

The design of the

out in accordance with the ultimate

shear stress requirements

for lightweight

concrete as given in CP110.

It was necessary to ensure that the beams was to be

would not fail in shear, as the flexural studied.

behaviour of the beams up to failure

The local and anchorage bond stresses for the various beams were calculated in accordance with CP11O. When mild steel was used for the main tension reinforcement, U-type hooks were provided; however, no hooks or any type of anchorage devices wer e It was also intended to flexural resistance moment '

provided for the deformed bars as it was not necessary. prevent bond failure at the ultimate

load, since the ultimate

based on yielding of the steel was to be observed.


3.2.2 Composite Reinforced Lightweight Concrete Beams

The parameters the ordinary behaviour beams.

employed for the composite beams were the same as those for To allow a direct comparison to be made between the flexural to the ordinary as integral

of both types of beams, the composite beams were similar except that the f. r. c. units were incorporated

beams in every respect,

parts at their tensile sides.


For the various composite beams, the nominal yield or 0.2% proof N/mm2 stresses and 590 N/mm2. The details of

used for the reinforcements The steel ratios employed

were were

275 N/mmn2,410 0.582%, 0.874%,

N/mm2,550

1.044% and 1.643%.

the reinforcements

are given in tables 1,2 and 3 and also are shown in Figs 4 and 5. the limit state of ultimate strength was

In deriving the working moment, considered to be the main criterion contribution made by the f. r. c. units.

for design, no account being taken of the The contribution made by the f. r. c. units,

if any, was considered to be an additional The calculated values of the ultimate beams tested under static,

factor of safety. and working moments for the various

fatigue and sustained types of loading are given in

tables 1,2

and 3 respectively. and design considerations for the f. r. c units are discussed

The properties in detail in 3.2.3


3.2.3.1

3.2.3. Cement Units (f. r. c. Units)

Fibre Reinforced
Properties

It was most important tensile strength and greater also considered desirable insulation, fire resistance,

that the f. r. c. units to be used should have much higher extensibility compared with those of concrete. thermal from Properties

for these units were the impact strength, durability and resistance

to the attack by alkalanity

the atmosphere. The choice of fibres was limited and which could be easily incorporated sheet or channel forms. The types of fibres which in general fitted most of these requirements glass and asbestos. Steel wires also have a high modulus, but they cannot be easily incorporated into a cement matrix and be fabricated fibres, into sheets or channel forms. durability in were to those having a high modulus of elasticity into a cement matrix and be fabricated into

As regard the E-glass

there is a problem of long-term E-glass fibre when incorporated

the presence of an alkali environment. portland portland resistant

in ordinary of ordinary

cement would loose most of its strength due to the alkalanity cement.

This problem can be overcome to some extent by the use of alkali in a cement matrix by a spray and to be fabricated. The

glass fibre which can be incorporated

suction technique, durability

allowing f. r. c, sheets or channel forms

of these f. r. c, units in long term application,

however,

is not known. and having

Asbestos fibres being the cheapest form of high modulus a greater durability

fibres,

under moist conditions were considered to be the most favourable The asbestos cement sheets are in

type that could be used in a cement matrix. manufactured predominantly in layers.

This process results in the ad estos fibres being distributed random fashion.

two dimensional

For normal asbestos cement products the length of fibre used is 6mm and less, the fibre content normally being 10-12% by weight of cement, giving a product density

of 1700 - 1900 Kg/m3 (GG) (67).


To satisfy and economy asbestos the requirements of this investigation the channels were for ease of construction as normal

in cost per unit, products.

manufactured

cement

The mechanical properties

of the fibre reinforced shown in Fig.

cement units used are 8. were carried out,

given in table 4 with the tensile properties

For the various composite beams, initial

calculations

tensile the bending to theory stresses which of the amount obtain simple using beams. the the f. soffits of be developed in the at placed r. c. units when could f. the tensile the then stress r. c. units. of These stresses were checked against The important consideration in the use of the f. r. c. units was that they would

load the conditions. at working crack not 3.2.3.2 Geometry fibre type reinforced of one cement unit with

Throughout the investigation similar properties

and geometry was used.

A channel section was chosen to

provide a surface reinforcement flexural

to maximum the subjected was area which on The length and width of a channel depended

tensile stresses and strains.

on the dimensions

of the concrete beam. points

following the the height the thickness upstands, In choosing of and were considered: -

1) The thickness and height of the upstand should be such, that an adequate tensile strength is provided so as to avoid surface cracking load conditions.
2) Consideration load, was also given to eliminate of the beams as well cracking under design working

at the working

at the soffit

as at the level The concrete

of the reinforcement cover to the main

to prevent

the corrosion

of the steel.

reinforcement

was 35mm and the maximum dimension of upstand of 55mm.

size of bar used was 20mm; For the purpose of this research

this gave an overall the average height

chosen was 60mm.

Considering

these points,

the cross section chosen for the f. r. c. channels the length of the

was 150mm wide, 6mm thickness with 60mm height of upstands, channel being 4900mm. The cross section of the f. r. c. channels is shown in Fig. 5.

3.3. 3.3.1

Materials Lightweight Aggregate Concrete

The lightweight commercially

aggregate chosen was the sintered pulverised

fuel ash

known as "Lytag".

The choice of this type of aggregate was made showed, that it had good prospects in

mainly because previous investigations structural applications (35).


of the lightweight cement, natural aggregate.

The constituents were "lytag" rapid-hardening

aggregate

concrete

used in the investigation and medium

sand zone 3 as a fine aggregate

13 - 10mm as coarse

The use of natural the structural pieperties

sand as a fine aggregate of the concrete (12) (3).

was mainly

intended

to improve

The intended 28 days cube strength was 50 N/mm2 with a concrete air dry density of 1700 - 1800 Kg/m3. the aggregate, research When choosing the size and mix proportions of the manufacturers (68) and a previous adopted were of

the recommendations

(69) were considered.

The mix proportions

finally

1: 0.83: 1.64 by weight for rapid-hardening "Lytag" respectively.

cement, natural sand and medium

The grading of the aggregate used was checked according to the B. S. 3797: 1964 (70) for the lightweight for the natural were carried Trial aggregate; aggregate and B. S. 882 and 1201: 1965 (71) The sieve analysis tests

this is shown in Fig. 9.

out according to B. S. 812 (72). out to fix the water cement The water cement ratio was for the mix. The air-dry

mixes for the concrete were carried workability and strength.

ratio for the required

found to be 0.59, which gave a medium workability

density for the concrete varied between 1675 Kg/m3 and 1860 Kg/m3. To improve the workability ratio and to maintain the required grade was added to the mix. of the mix without increasing cube strength the admixture the water cement "Febflow" standard

The amount of "Febflow"

added was 140cm3 per recommendations.

50 Kg cement; this was according to the manufacturers'

3.3.2

Reinforcement The sizes and types of the various were as follows: bars used in this research as main

tension

reinforcements

1) mild steel; plain round bars with a nominal yield stress of 275 N/mm2. The sizes used were 12mm, 16mm and 20mm.
2) Unisteel 410; hot rolled deformed bars with a nominal yield stress of

410 N/mm2.

The sizes used were

12mm,

16mm and 20mm.

3) Unisteel

550; cold worked deformed bars,

produced by cold stretching The sizes

Unisteel 410, with a nominal 0.2% proof stress of 550 N/mm2. used were 16mm and 19.05mm. 4) Kam 60; natural hard ribbed steel, manufactured 60 refers to the yield stress in Kg/mm2. in this research was 590 N/mm2.

in Sweden. The figure

The nominal yield stress adopted

The sizes used were 12mm and 16mm. Kam 60 steel by The The

5) Kam 90; cold worked steel, produced by cold stretching 5%. The number 90 refers

to the 0.2% proof stress in Kg/mm2.

nominal 0.2% proof stress adopted in the research was 875 N/mm2. size of the bars used was 16mm. 6) Lancasteel 60; hot rolled heavily deformed steel bars,

the number 60 refers

to the yield stress in Ksi. was 410 N/mm2. Typical in Fig. 1 stress-strain

The nominal yield stress adopted in this research

The size of the bars used was 16mm. curves for the various types of reinforcement are given in table 5. are shown

their tensile properties

CHAPTER

FOUR
STEEL

THEORETICAL

BASIS OF ANALYSIS FOR STRESSES IN CONCRETE,

AND f. r. c. CHANNELS
4.1 Introduction

In reinforced

concrete flexural

members the steel stress is an important of the beam.

factor when determining

the width of the cracks and the curvature

f. the in beams the in r. c. stresses When determining the steel stresses composite channels are also an important consideration. by beams are calculated and composite of moments. In using this equation shape, centroid

The steel stresses for ordinary

employing the general equation of compatibility it is important

that the level of the neutral axis and the geometric distribution stress

and area of the compressive

in the concrete are determined.

the distribution in the determine to beams is stress it For composite necessary tensile the of the to magnitude f. r. c. channels in order position and calculate force. Relationships based on theoretical considerations and analysis of test results flexural

have been established with the applied moment for the neutral axis level, compressive distribution strain in concrete, area and centroid of the compressive

stress

in concrete and stress in the f. r. c. channels.

These relationships for both

have been established to enable the stress in the tension reinforcement ordinary 4.2

level be beams to any of applied moment. at calculated and composite of the Neutral Axis Level axis is an important of reinforced factor when determining members. the In

Variations

The level of the neutral steel stress and the curvature general,

concrete flexural

the position of the neutral axis depth at any level of applied moment

depends upon the geometry of the section (this includes the percentage of steel), the properties of equilibrium of the materials and whether the member is cracked or not. Conditions

for the member should also be satisfied.

The depth of the neutral axis for a fully cracked or an untracked section can be determined by employing the elastic theory approach. cracked concrete section is that when most of flexural and that the contribution

concrete A fully

cracks have been developed

of the concrete in the tensile zone has become negligible. and fully cracked conditions the level of the based on experimental results.

At the stage between the untracked neutral

axis can be calculated by relationships

The variation

of the neutral axis level in relation

to the applied moment at this cracks towards the

stage is primarily compression

flexural by the of caused propagation

face of the member.

For the stage between the uncracked and fully cracked conditions of a concrete member a simplified straight line relationship results has been suggested (73). However,

analysis of the experimental curvilinear


In this

of previous investigations (4) (74) (75).


the variation

suggests that a

relationship
research

could be used.

a study is made regarding moment

of the neutral

axis

depth in relation is suggested ultimate

to the applied the level

up to failure.

An idealised

relationship to

between

of the neutral

axis and the ratio

of the applied

moment

as shown in Fig. 10 .

The relationship

following be divided the into can

stages: -

(The values of

Co, C1 and C2 are discussed in 7.2) Untracked stage From 112/Alu =0 to M/Mu = Co and the neutral section

At this stage it is assumed that the concrete is untracked axis level can be determined by considering employing the elastic theory approach. For ordinary
n=

an untracked

transformed

The equations are:

beams:
Xuo dl =2+2 2 (m - 1)p (m - 1) ]p (4.1)

[A+

For

composite

beams: Xuc

0.5

X+ X+

n=d=

(m - 1)p + (ml - 1)pl (m - 1)p + (m1 - 1)pi q

2 T,

(4.2)

1 d/d1,9

Where

A-

= d2/d1

The derivation

of equation 4.2 is shown in appendix D1, equation 4.1


concrete

a similar

approach

can be employed for deriving


For is very This neutral ordinary lightweight because

beams the range of the untracked formation of cracks

behaviour

limited

mainly

of the early

in these members.

suggests

that the value of Co in Fig. 10 would rise above the level

can be taken as zero and that the of the calculated value for the untracked evidence, however,

axis level

transformed in this respect

section

once loading

has commenced.

Experimental

is necessary.

Transition

Stage

From

M/Mu

= Co to M/1M

= Cl is When

At this stage the movement of the neutral axis as mentioned earlier greatly cracking influenced by the formation occurs, and extension of the flexural cracks.

the steel will be strained considerably,

mainly because of the This immediate axis towards

sudden transfer increase

of the tensile stresses from the concrete to the steel. can further raise the levelof the neutral

in the steel strains

the compression Considering neutral

face of the member. a gradual propagation of flexural cracks, the variation relationship. of the This

axis with the applied moment can follow a curvilinear as mentioned before can also be seen in results suggested that a parabolic

behaviour

of previous investigations. at

It is therefore

curve may be used for the relationship

this stage between the neutral axis depth and the ratio of applied to ultimate The assumptions used and the derivation The equations are: For ordinary X= beams: Xuc C1 Xcc (R2

moment.

of the equations can be seen in appendix D3.

-2

C1) + Xuc

(4.3)

For

composite

beams:

X=

Xuc

Xcc 0C1 R= + C12 M/Mu

( R2 -2

C1 R+

C12 )+

Xcc

(4.4)

Co2 -2C For both equations

Fully

Cracked Stage

From

M/Mu

= C1

to 117/Mu = C2 and the neutral section axis at this

At this stage the concrete is assumed to be fully cracked, axis depth can be determined by considering employing the elastic theory approach.

a cracked transformed of the neutral

The variation

stage is very small and can almost be neglected. The equations are:
For ordinary beams:

n=

Xco dl

= mp

(+2mp

1)

(4.5)

For composite beams:


Xcc dl = (mp + mlPi q 1+ (mp+mlPl7 The derivation of equation (4.6) is given in appendix equation (4.5) (D2), a similar approach 2 (mp+m1P1 2 -1 (4.6)

can also be employed

for deriving

Non-elastic

stage

From

M/Mu

= C2

to

M/Mu

=1

At this stage it is assumed that steel and/or concrete are in their non-elastic range of behaviour, is considerably steel, and that. therefore the rate of strain due to the applied moment in the

increased.

This increase in the rate of strain, especially effect on raising the neutral

can have a considerable

axis level towards

the compression

face of the member.

Another factor which also may be considered

is that the rate of increase in the steel stress at this stage is much lower than The level of the neutral axis will therefore behaviour. that at the elastic range of rise to provide a greater conditions of equilibrium. the neutral axis depth calculated by the lever arm to the forces in the steel, thus satisfying the

An attempt is made to correlate strain compatibility

method at the ultimate

moment with the value of AT/Mu = 1. applicable to under reinforced beams.

The behaviour described above is generally

4.3 4.3.1

Analysis

of Compressive Relationship

Stresses

in Concrete Concrete in Compression

Stress-Strain

of lightweight

The rectangular compression

- parabolic

design stress-strain

curve for concrete in incorporates the properties

given in CP110, and as shown in Fig. 7 This is shown by the initial

of normal weight concrete.

tangent modulus drawn to for normal

the curve at zero strain given in terms of the modulus of elasticity weight concrete, and also by the value of strain at the junction (e j). concrete having a similar

For lightweight

strength to that of normal weight Assuming a similar stress-strain

concrete the modulus of elasticity relationship in compression,

would be lower.

the initial

tangent modulus at zero strain for lightweight which

concrete will be lower.

This will affect the value of strain at the junction (ej), Considering

for be than that normal weight concrete. greater will

both concretes

have the same maximum value of strain (0.0035), the area under the curve for lightweight concrete, be level the smaller stressv. would of at same than that of normal

weight concrete. In this research the stress-strain derived, employing the initial curve for the lightweight for lightweight concrete was concrete as

modulus of elasticity

recommended by CP110. relationship in Fig. 6a. by the partial


4.3.2

This is shown in appendix (A). the partial

The stress-strain is shown

without incorporating

safety factor for the material

The design relationship safety factor (ym=1.5).


of the Flexural

can be obtained by dividing the values of Fig. 6a

Variations

Compressive

Strain

in Concrete

For determining

the shape, centroid and area of the stress distribution zone it is important is also important that the strain profile in the determination

across

the depth of the compression The compressive curvature

is lmown. of the

strain profile member.

of a flexural

In this research strain in the concrete simplified

it Is intended in relation bilinear

to study the variation moment

of the flexural up to failure. between

compressive As shown in of the in the

to the applied relationship

Fig. 11 aa applied concrete.

is suggested

the ratio strain

moment This

to ultimate

moment well

and the maximum with experimental

compressive results

concept agrees

of a previous

investigation

flexural weight concrete on normal

members (4).

The variation

of the compressive relationship

line have is to in the assumed straight concrete a strain = Cp (The value of

with the applied moment up to a value of M/Mu

Cp js discussed in 7.4).

The values of the concrete strain at this stage can be of moments together with the For the composite beams the moment of the f. r. c, channels.

predicted by employing the equation of compatibility assumption of a straight line strain distribution.

moment of the tension side also includesthe resisting However,

beams in f. the the the M/Mu Cp, channels c. composite r. at value = If this was the case, the analysis of the composite beams the effect of the f. r. c. channels.
= Cp to M/Mu non-elastic = 1) it is assumed that the

might have been cracked. should be carried

out without considering


phase (From

In the non-elastic concrete therefore, moment. rise zone.

M/Mu

and/or

steel are in their in the concrete the values

range of behaviour increased strain

and that,

the rate of strain

is considerably of the concrete reduction

due to applied

What also may affect

is the considerable

of the neutral In order

axis and the consequent the conditions which will

in the area of the compression higher stress values will be

to satisfy

of equilibrium, lead to a greater

developed

in the concrete,

rate of strain

up to failure.

At the moment of failure (emax) will be reached.

(M/Mu

the in 1) the strain concrete maximum =

The factors which may affect the maximum compressive As per CP110 the value of e, max , at the non-elastic phase as shown in Fig. 11a ec=ep Cp at = and M/Mu =1

discussed in 5.2.1 in the are concrete strain is equal to 0.0035. is represented The relationship

by a straight line between M/Mu

at e.max = 0.0035.
4.3.3. Variationsof the Area and Centroid of the Compressive Stress Distribution

in Concrete The area and centroid are important members. assumed lightweight considerations of the compressive when determining stress distribution stresses in the concrete in flexural in compression relationship is for

the steel

In CP110 the stress-strain to have a parabolic-rectangular concrete is shown in Fig.

relationship distribution. 6a

for concrete

The derived

Values of coefficients

for the centroid and area of the stress distribution strain in the concrete. compressive zone. strain in deemed

diagram depend upon the magnitude of the compressive In flexural flexural the depend this upon would members

the concrete at the extreme element of the compression necessary the centroid to derive relationships

It is therefore

between the concrete compressive stress distribution

strain and

and area of the compressive

diagram in the concrete.

From thcseit would then be possible to predict the values of for the centroid of the stress distribution diagram and a for the area of the stress distribution diagram

flexural level the once maximum of moment at any moment is known. relation Fig 12 b shows the variation

strain in the concrete at that and a in

of the values of

to the compressive relationship

strain in the concrete.

The geometrical

shape of the

stress-strain the derivation

in Fig. in 1 12 the was considered concrete as shown of This is shown in appendix E. )

of the equations.

The equations are:

(Reference can be made to Fig 12

For (1) when the compressive ei (0< ec <_ ej) 4 ej - cc 12 ej -4ec

less is ec in or equal than concrete strain

(4.7)

3 ej a=

cc
3ec

(4.8)

ei

For

(ii.)

(e;

ec c ems) 4 ee e, + e32
_4 e ee ej

6 ee2

p
-3 a

(4.9)

12 eel

cc

(4.10)

3 ee see appendix (A) Fig. 6 strain in concrete at failure)

ej a 0.00282 emax = 0.0035

(assumed compressive

Values of ec which represent

the maximum

concrete compressive

strain

at the extreme element of the compression determined calculated 4.4

zone of flexural

members can be data or

at any level of applied moment either from experimental as shown in 4.3.2. of Stresses in f. r. c. Channels distribution

Analysis

Magnitude, are important

and centroid of tensile stresses in the f. r. c channels the steel stresses for composite beams. The

in determining

assumed distribution is shown in Fig.


Considering stresses

of tensile stress in the f. r. c, channels for the composite beams

13.
the basic assumptions as stated in 4.5.1 remain to calculate the steel and perfect

in composite

beams

(plane sections and concrete),

plane after bending section

bond between by the simple the relationship moment,

the f. r. c, channel bending theory.

the composite

can be analysed is used for

As shown in Fig. 11b, a bilinear stress

concept

between

the maximum

in the f. r. c, channels section

and the applied or not.

taking

into consideration

whether

the concrete

is cracked

For the first concrete section.

stage the composite section is analysed assuming an untracked Once the concrete has cracked (at Me) the composite section the

should then be treated as a cracked concrete section taking into consideration initial untracked

behaviour and an effective values for the second moment of area for the f. r. c, channels and concrete. of the maximum stresses in

of a cracked section and modulus of elasticity The mathematical expressions

for the calculation

the f. r. c. channel at the two stages are given in the following:


(i) For (M < Me)

fab =M

(d - Xuc) Iuc

ml

(4.11)

Where Mc : is the cracking Mc = frc d' Iuc

moment for the concrete at the interface. (4.12) Xuc

(ii)

For

(Mc

<M

<_ Ma) than the cracking at the soffit moment of the concrete Me)

At an applied

moment
Ii

Al (greater

as shown in Fig.

11b the stress

(fab) is equal to:

fab=fab+fab Where fab and fab are the stresses corresponding and \1 - Mc) respectively. Therefore: fab = Mc (d
Iuc

to the bending moment (Me)

Xuc)

m1 + (M - MZc) (d - Xcc) m1 P, Ice

(4.13)

Where

, is a factor depending on an effective

second moment of area for

a cracked section (Icc) and the modular ratio (mi). The assumed cracking beams the is that in the composite concrete of moment This is mainly because cracks which effect on the flexural

when cracks are observed at the interface.

have form in the no significant will confined concrete may behaviour of the member due to the restraint

f. the r. c units. action of concrete are between

Average values of the modulus of rupture for lightweight 2 and 2.5 N/mm2. This, however,

for the N/mm2 3 to be increased composite can f. the r. c. channels. of action for f. r. c. channel as shown

section at the interface Considering in Fig. 13 as follows: fat =d-d-x i

dae to the restraint

the straight

line stress distribution

the tensile stress at the top element of the upstand (fat) can be obtained

fab

(4.14)

'

d-x The tensile force in the f. r. c channel (Ta) can be determined Ta = (fat + fab) (d - t)t To determine
the stress

from the following: (4.15)

+ fab.

b. t

the effective

depth d2 for the tensile force Ta, the centroid of


The distance (da) for the centroid the following

distribution distribution

should be determined. from the soffit

of the stress expression:

can be obtained

from

da = fat

(d2

fat) t2) (fab + (fat + fab)

(d2

td -

2t2) /3 + fab b. t/2 .

(4.1G)

( 'd - t) + fab .b in the f. r. c.

This was obtained by taldng moments of area of the stress distribution channel shown in Fig. 13, about the soffit of the channel.

Therefore
4.5 4.5.1

d2

=d-

da

Calculation

of Steel Stresses

Assumptions

The assumptions

made for the calculation

of the steel stresses for ordinary

and composite beams are as follows: (1) Average strain distributions for concrete in compression, reinforcement

and the f. r. c. channels are derived from the assumption that plane sections bending, irrespective after remain plane is cracked or not. (2) The stress-strain relationship for concrete in the compression distribution zone of a of whether the concrete section

beam follows a parabolic-rectangular (3) The contribution (4) The variation to ultimate

as shown in Fig. Ga

of the concrete in tension is ignored. axis depth in relation to the ratio of applied as

of the neutral

moment follows a parabolic

straight line relationship

shown in Fig. 10. (5) The relationship between the flexural compressive strain and the applied

moment is based on a bilinear Additional assumptions

concept as shown in Fig. 11

follows: beams for the are as composite made

(1) Stresses in the f. r. c. channels can be obtained either from the bilinear relationship between the stress and the applied moment suggested earlier stress-strain curve of the f. r. c. units as shown

or from the experimental in Fig. 8

(2) The stress gradient for the f. r. c. channels is a straight through the neutral axis (3) The contribution Fig. 13.

line which passes

of the f. r. c. channels on the tensile

side is ignored once

cracking occurs in them. (4) There is a perfect bond between the f. r. c. channels and the concrete enabling the units to be fully co-operative The mathematical for the ordinary procedure in the composite action. of the steel stress

adopted for the calculation

and composite beams is discussed in the following paragraphs.

4.5.2

Ordinary

Lightweight

Concrete

Beams

The steel stress at any level of moment is calculated by employing the general equation of equilibrium of resistance M= based on steel. As fs (d1 - for the applied moment and the internal The equation is: x) (4.18) for the centroid at any level of moment

The values of the neutral axis depth (X) and the coefficient of the compressive stress distribution () can be determined

moment as shown in the previous paragraphs. For For X paragraph paragraph (4.2) (4.3.3) for the

The steel stress may also be checked by the equation of equilibrium tensile and compressive
As fs =a

forces
fcb X an accurate prediction of the value (4.19) (fc) at any (ec)

This, however, requires calculated or experimental

value used for the strain

in the concrete

The stress-strain zone of flexural relationship

relationship

the in for the compression concrete assumed It should be noted that this

6a. Fig. in is shown members

is based on direct compressive relationship

tests on concrete cylinders. 5.2.2, discussed in as the results - may be

The stress-strain different

in flexure, Therefore,

from that assumed above. stress-strain

obtained by using

the idealised
4.5.3.

relationship,
Concrete

Fig. 6a,
Beams beams

may not be very accurate.

Composite The steel

Lightweight

stress

in the composite

is calculated

by the following

stages: (1) For applied (M S Ma). tension moment At this up to the cracking stage the contribution in the calculation. moment of the f. r. c. channels in is:

of the f. r. c. channels The equation employed

is considered

M=

fs. As

(dl

X) + Ta

(d2 -

X)

(4.20) manner as that for

The values of X and can be determined the ordinary beams (explained in 4.2,

in a similar

). 4.3.3. The values of the tensile force and

in the f. r. c. units (Ta) is dependent upon the state of the composite section discussed in 4.4. is 'Iltis is the cracked or not. concrete whether depth d2 can also be determined
(2) For an applied channel moment

The effective

as shown in the same paragraph.


greater than the cracking moment f. the r. c. of

(M >_ Ma).

At this stage the contribution

is ignored in f. tension and the channels r. c. of beams. This is mainly because the

the section is analysed as that for ordinary maximum

steel stress will occur in the vicinity

of cracks where the restraint

is f. the not effective. channel r. c. of action as that for the ordinary

The equation employed is the same

beams (e. g. equation 4.18 in 4.5.2) manner as explained

The values of 3 and X can be determined in similar previously. 4.5.4 Summary for the Theoretical (1) Calculate the ultimate method. Calculation

of Steel Stresses

by the a suitable plastic member moment of

depth in as shown the level axis For neutral (2) of moment calculate any 4.2 and Fig. 10. (3) For any level of moment calculate the maximum flexural 11a. Fig. 4.3.2 in and strain in the concrete as shown
(4) Calculate the values stress of and a for the centroid in the concrete,

compressive

and the area of the as shown in 4.3.3.

compressive and Fig. (5) 12.

distribution

For ordinary 4.5.2.

beams the stresses

are calculated

by equation

(4.18)

in'

(6) For composite beams the magnitude and the centroid of the tensile force in the f. r. c. channels are calculated in accordance with 4.4. (7) Steel stresses for the composite beams are calculated in accordance with 4.5.3. 4.6. Suggested Method for Calculation of Steel Stresses Based on Experimental

Results In order to demonstrate the validity of the theoretical predictions for the

steel stresses, compressive

calculations

using the experimental

results

(Neutral axis depth,

strain in concrete and tensile strain in the f. r. c. channels) out. In these calculations the general equations for equilibrium can be used.

should be carried (4.18) and (4.20)

for the ordinary

and composite beams respectively

CHAPTER THEORETICAL FOR ORDINARY 5.1 Introduction CONSIDERATIONS AND COMPOSITE

FIVE STATES OF DESIGN

FOR LIMIT BEAMS

Until recently the design of an individual

reinforced

concrete member was

based on either the elastic theory or the load factor method as recommended by CP114. The limit various limit safety factors state theory recently states in design. adopted in CP110 has the concept for considering states partial

In the assessment of these limit

are employed for the strength of the materials safety factors

and magnitude of loads. data; they

The values of these partial take into consideration materials life.


The concept investigators in more detail. of limit

are predicted from statistical

the probable variation

in the strength of the constructional throughout its specified design

and magnitude of the loads on the structure

state design,

however,

has been well

explained

by many this

(64) (7G) (77) (78); hence it is considered

nctnecessary

to discuss

The principal relevant

limit

states considered in design as per CP110, which are are the following: Strength.

to this research

(1) Limit (2) Limit

State of Ultimate

states of serviceability

(i) Deflection. (ii) Cracking. state of ultimate strength an adequate safety factor is provided safety factors for the materials

For the limit

for the working load condition by employing partial and loads. limit

This will also ensure that the stress in the steel does not exceed its

of proportionality. Apart from limiting the span - depth ratios and the spacings of the reinforcements, and

the code does not recommend any suitable method for the control of cracking deflection.

In general,

methods of calculating

the various

limit

states proposed in the of fibre

past should be reviewed, reinforced

and if possible modified to fit the requirements

cement composite concrete construction. the limit state of deflection proposed

In appendix (F) methods for calculating in the past are reviewed.

In the following paragraphs proposed methods, behaviour with the prediction of ultimate strength, cracking

and points associated and

and deflection for ordinary

composite lightweight 5.2


5.2.1

concrete beams are discussed. Strength


and Assumptions

Limit

State of Ultimate
Considerations

General

The flexural

failure

of reinforced

concrete members occurs strain capacity.

when either

the concrete or steel reaches its maximum

Since steel usually failure often occurs

has a higher strain capacity compared to that of concrete, by the concrete being crushed in the compression In case of under reinforced beams failure zone.

the by steel which of yielding starts The maximum compressive strain

will be followed by the crushing of the concrete.

in concrete may vary between 0.19 to 0.52% depending on the shape of the compression zone for the beam, the position of the neutral loading (37). of rate a flexural axis, the quality of concrete and the stress distribution diagram for

The shape of the compressive

the the depends quality of upon concrete and concrete member mainly

also on the rate of loading (37). For under reinforced almost entirely beams, where the load-carrying capacity is determined the shape

by the tensile force that the steel is capable of resisting, stress distribution would not significantly

of the compressive resistance internal

affect the ultimate

of the beams; at the most, it may slightly

alter the lever arm of the

forces.
assumptions for calculating the ultimate strength as per CP110 are

The basic as follows:

(1) The strain distribution reinforcement

for concrete in compression

and the strains in the

are derived from the assumption that plane sections remain

plane after bending.

(2)

For the concrete distribution using curve

in compression

a rectangular-parabolic

stress version design (Ym=1.5)

is assumed.

- An alternative block.

to this is a simplified

a rectangular incorporating

stress

The rectangular-parabolic safety factor for the concrete maximum

the partial

as per CP110 is shown in Fig. strain in the concrete being

7a, the assumed

compressive

0.0035.

(3) The concrete does not resist

any tension. relationship for the steel is given This is shown in

(4) A specified design stress-strain incorporating Fig. 7b. a partial

safety factor

Ym=1.15.

From the above considerations the neutral axis is adjusted by a trial

at the ultimate and error

load condition the depth of so that the compression

procedure

forces balance the tensile forces across the section. equilibrium the ultimate

With the attainment of

design moment can be determined by taking moments for stress block. The working

the tensile forces about the centroid of the compressive

moment can then be obtained by dividing the design ultimate partial 5.2.2. safety factors for the live and dead loads. Comments on the Basic Assumptions Considering of flexural

moment by the appropriate

the previous basic assumptions for predicting

the ultimate

strength accurate

members,

the method as used in CP110 would give only a fairly points are considered: -

value for flexural

members when the following

(1) The assumption of a straight line strain distribution the ultimate

may not be valid at

condition according to the theory proposed by Baker (79),

where he assumes that there is a slip between the steel and the concrete.
(2) Partial safety factors for the materials strain and loads should not be used. should be carefully discussed in

(3) The maximum assessed, 5.2.1.

compressive

in the concrete the factors

taking

into consideration

previously

(4) The stress-strain

relationship

for the reinforcement

should be that obtained

from a direct tensile test.

It should also be emphasised here that the

design stress-strain Fig.

curve for the reinforcement

suggested by CP11O which some

7b, does not allow for any strain hardening properties are capable of exhibiting.

of the steel reinforcements

It can also be seen and cold-

that there is no proper distinction worked steels. The differences

made between hot-rolled in their stress-strain

characteristics

are that the cold-worked

steel barq in general,

have no yield plateau,

and are capable of exhibiting compared with hot-rolled

high stresses with rather low strains when The code, however, range of diameters. does allow a 12.2% when cold-worked

bars.

increase in stress for a limited steel bars are used.


(5)

The distribution can be different

of the concrete stress in the compression from that obtained from compressive

zone in flexure

tests on cylinders. levels

This is mainly due to the fact that in a flexural of the compression proportional
be rather zone.

member the different rates, which are

zone undergo strain at different

to their distance from the neutral axis.


to assess the true stress distribution theories

It would therefore
for the compression

difficult

In this respect data.

most of the previous However,

depended on distribution of an underdiagram

experimental diagram reinforced suggested

since the shape of the stress affect the ultimate stress strength

would not significantly beam,

the rectangular-parabolic

distribution

by the code may be used.

Additionally,

CP110 does not give

in particular

the compressive

stress-strain

relationship

for lightweight however, can

design. be in used concrete which can be derived from the assumptions recommendations

This relationship,

stated by CP110 and the CEB (10) tangent

for the shape of curve and the value of initial This is discussed in 4.3.1.

modulus of elasticity.

the derivation and ,

is given in appendix (A).


5.2.3 Composite Beams

An assessment of the ultimate by following a similar

strength for composite beams can be obtained beams. By

approach to that employed for the ordinary

assuming a full interaction

between the concrete and the f. r. c. channels,

assumption

(1) in 5.2.1.

is still valid.

This follows that the strain in the line strain

f. r. c. channels can be obtained from the assumption of a straight distribution. However, at the ultimate

load condition the tensile strain induced

at the soffit of the composite beams will be greater than the tensile strain capacity of the f. r. c. units. Therefore, at this stage, and when only the ultimate the contribution strength

of the beam is to be calculated, The contribution factor of safety.


5.3 5.3.1. Limit State of Deflection

of the f. r. c. units can be ignored.

of the f. r. c. channels, if any, can be considered as an additional

Introduction

The limitations

on the amount of deflection by various codes of practice deformation which neither impaires

are

mainly based on the smallest of the structure carried

the appearance In research of

nor causes any damage to finishes or partitions.

for deflection the majority found the that in Germany it (80). out was

structures

which had given rise to complaint had a deflection more than span/250. that this limit might vary according to the capacity of the or resist strain, since the damage to these had (81). state theory, while employing

It was suggested partitions

or finishes to absorb

often been the criterion

for limiting

the deflection

The design of concrete members using the limit high strength materials

has made it possible for the size of such members to be is taken into consideration. Since

reduced when only the strength of the material the increase in strength of the material increase in the elastic modulus, reduced,

is not associated with a corresponding rigidity of such members is deflection for the members

the flexural

considerably

and this will result in a greater In addition,

under the same applied load. higher strains width of cracks. out deflection

by employing high working steel stresses, curvature and possibly a greater

are expected leading to a greater Therefore,

it would seem necessary in certain cases to carry amount of deflection

calculations

in order to assess the correct

that would result under load.

Various

methods have been proposed for the calculation

of deflection;

these can predict the deflection for laboratory 20% (81) (82). In general, distribution the straight

specimens within an accuracy of

These methods are reviewed in appendix (F). the deflection of a beam can be calculated if the magnitude and are known along the span of the beam. of a homogeneous flexural By employing

of the curvature line theory

the curvature

be can member

obtained from the following


1= rb

expression: (5.1)

ee + et dl

Where ec and et are the maximum compressive and steel respectively. The curvature following = of an untracked

and tensile strains in the concrete

concrete section can also be expressed in the

form: NI Ec Iu section. becomes rather


(5.2)

Iu : second moment of area for an untracked For a cracked section difficult. distribution relationship

the assessment of the curvature

Once the concrete is cracked, of the curvature with the moment.

be created where the will a condition

does have linear beam the the not a of span along This is mainly due to the fact that the values of affect the curvature vary between a maximum value

the steel strains which directly in the vicinity cracks.

of the cracks and a minimum

value at a point between two adjacent and

This variation

for the steel strain is mainly caused by the restrained

non-restrained

action of concrete at and between the cracks.

Since the deflection

depends more upon the curvature

along the span of a beam than at a particular should be based on the average strain action of the concrete in the tension zone I

point, the assessment of the curvature values for the steel where the restrained is considered. The restraint in the calculation

action of the concrete in the tensile of the curvature by various

zone has been considered in different ways;

existing theories

these are discussed in detail in appendix (F).

In the following methods of calculation

paragraphs the considerations, for the deflection of ordinary

behaviour

and proposed

and composite lightweight

concrete beams are discussed.


5.3.2. Considerations for Ordinary and Composite Lightweight Concrete Beams

Lightweight of elasticity

concrete,

in general,

is characterised

by having a lower modulus

compared with normal weight concrete.

This will to a certain extent However, the deflection

affect the deflection behaviour of the structural depends upon the flexural rigidity,

member.

of the member,

i. e, the modulus of elasticity

(Ec) as well as the second moment of area (1). For the same geometrical concrete section analysed by the elastic theory, will lead to a greater neutral i>, results. rigidity of the member due to the axis depth;

concrete of lower modulus of elasticity consequently a greater

moment of inertia

This indicates that the reduction in the flexural lower modulus of elasticity of inertia. To illustrate is partially

compensated by an increase in the moment 14,

this point,

a graph is established as shown in Fig.

for both ordinary

and composite beams relating

the ratio of the neutral axis depth (m = Es/Ec)

to the effective depth (n =. I) dl

to the modular ratio

The equations used are based on the elastic theory approach for a cracked transformed section incorporating the properties of the test beams. The

equations are discussed in 4.2 For the ordinary


n= Xco d1

beams
_ mp (I I+2mp 1) (4.5)

For the composite beams


n= Xcc if-, _ (nip + mlplr) )(1+2 (mp + mLE1 i12)2 -1)5 (mp + m1 pl q) in these equations are: (4.6)

The values p:

of the parameters

incorporated

As shown in the graph = 0.7, p1 = Ach = 0.03628, Ti = d2 = 1.1

m1 : Ea

Ec

bd2

d1

From the graph it can be seen that the greater ratio neutral (i. e. lower concrete modulus of elasticity) axis depth. 14 a direct comparison

the value of the modular

the lower is the level of the

In Fig. neutral

can also be made between the level of the For a concrete section with

axis depth of ordinary

and composite beams.

a modular

ratio of 12 and a steel percentage area of 0.582, the neutral axis depth beam. The difference

of a composite beam is 16%o greater than that of an ordinary between the level of the neutral axis depth of ordinary

and composite beams decreases

when higher percentages of steel are employed; percentage of 1.643 the difference is 6%.

for a modular ratio of 12 and a steel

It can also be seen from the same graph that when lower values of the modular ratio. (m) are used (Le.,, high modulus of elasticity of concrete), the difference

between the neutral axis level of the composite and ordinary greater;

beams becomes

this, however,, may be considered as an advantage for normal weight concrete

composite construction.
In general, in the following a lower position of the neutral axis in concrete beams may result

advantages: -

(1) Greater flexural (2) A more efficient

rigidity

for a cracked section. that is by subjecting a larger area

use of the material,

of concrete in compression. (3) A smaller Fig. concrete area in the tensile zone and reduction in cracking. the effect of the ratio of the modulus of elasticity of the

15 illustrates

f. r. c. units to that of concrete on the level of the neutral axis depth. graph, equation 4.6 discussed earlier
the graph with respect

For this

was used.
of change in the neutral = Ea/Ec) is very small. axis It

From (n = Xcc) -Fl-r-

it can be seen that the rate to the modular ratio (m1

can also be seen that the rate of change becomes smaller of steel are employed. It must be emphasised here

when high percentages

that the above considerations

are limited

to the

type and shape of the f. r. c. channels employed in this research.

5.3.3.

Behaviour

of Ordinary

Lightweight

Concrete

Beams

The load-deflection is similar

behaviour of the ordinary

lightweight However,

concrete beams with the existence range

to that of normal weight concrete beams. in the lightweight concrete,

of shrinkage cracks

the load-deflection

within which the beam behave Furthermore, the contribution

as an untracked

member may be very limited.

of the concrete in the tensile zone to the stiffness

of the member may differ from that of normal weight concrete. To justify 5.3.4. these arguments experimental of Composite Lightweight results are necessary.

Behaviour

Concrete Beams to that of an

The load deflection ordinary

curve of a composite beam in relation represented

beam can be schematically

as shown in Fig 16a. The load

deflection behaviour these are: (1) Uncracked


(ii) Partially

of the composite beam can be divided into three main stages,

) stage (0 to 1.
cracked stage (1 to 3) but not the channel)

(Concrete (iii) Fully cracked

cracked

stage (3 to 4)

(both concrete and channel are cracked) The corresponding cross sections for each of these stages are shown in Fig. 17.

These stages are discussed in the following: (i) Uncracked The behaviour untracked relationship, stage, (section No. 1 in Fig. 17)

at this stage is based on the assumption that the section is are in their elastic range of behaviour. The

and that the materials therefore,

is assumed to follow a straight

line for the loading and

unloading process. When a comparison is to be made between the flexural properties rigidities for ordinary

and composite beams of similar

and cross sections it must be realised

that part of the concrete in the tensile zone of a composite beam is replaced by a material (f. r. c. units) with a relatively lower modulus of elasticity. rigidity This will composite

result- in an overall section.

reduction in the flexural

(EI) for an untracked

This may give the impression

that the composite beams give a greater

deflection

at the untracked

stage than the ordinary

beams.

However,

this is not so

because of the following: (1) The second moment of area based on an untracked transformed section for the composite beams (Iuc) is only 1.67% and 2.1% less than that for the ordinary respectively. beams (Iuo) for steel percentages of 1.643 and 0.852 Therefore, the overall difference in the flexural rigidity

(EI) between ordinary little

beams is minimal composite and

and will have

influence on the deflection.

For the above values the area and modulus

of elasticity

for into taken f. the account calculating were of r. c. units The neutral axis

the second moment of area for the composite beams.

depths were calculated by employing equations 4.1 and 4.2 for the ordinary and composite beams respectively. type of construction it is quite possible that cracks may Furthermore,

(2) In the ordinary initiate

in flexural

loading commences. once members

the tensile strength of the concrete in the ordinary fully utilized

beams may not be

due to the existence of micro cracks in the concrete.

tensile the in the zone large beams the concrete In area of a composite formation delay the This of cracks f. is confined by the will r. c. channels. behave beam the loading as an the which range within and consequently untracked member is increased. The cracking mechanism of composite

beams is discussed in more detail in 5.4.2. (3) The formation can significantly however, flexural in of shrinkage cracks lightweight concrete members

behaviour. deflection the affect ignored in ordinary

The effect of these cracks, beams. In composite

is normally

reinforced

beams the formation reduced.

be delayed will cracks and their width of shrinkage

This is probably due to a slow rate of shrinkage in the composite strength and be more

beams, thus allowing the concrete to have a greater resistant to the induced tensile stresses. the above points,

This is discussed in 5.4.2. reasonable to assume that in composite

Considering the properties

it is therefore

of concrete in tension will be fully utilized

beams and modified values for the tensile properties for ordinary beams can be considered in design.

over those employed

The above points may

also suggest that for ordinary

types of construction especially

a fully untracked lightweight

behaviour may not exist in reality concrete construction. (ii) Partially

for ordinary

cracked stage (sections Nos. 2 and 3 in Fig.

17) 16a,

After the cracking load for the concrete is reached at point (1) in Fig. the slope of the curve will gradually depend entirely on the rate of change.

The rate of this change will and propagation of the cracks therefore, will depend on the action that It appears of the

formation

in the tensile zone. properties

The saving in deflection,

of the f. r. c. channels and on the amount of restraint

these channels will exert on the width and extension of the cracks. therefore that the contribution of the f. r. c. channels to the stiffness

member is more pronounced at this stage. With the composite beams having a large area of concrete confined by the f. r. c. channels, the growth and propagation gradual and better controlled the transition of the cracks will be more Therefore,

than with the ordinary beams.

behaviour fully the between the of cracked and uncracked zone longer than that for the ordinary zone for

composite beams will be considerably reinforced the ordinary beams. Comparison

between the transition be made can

beams (a - ) and that for the composite beams (1 - 3), where the behaviour of the composite beams at this stage can be characterised by a curvilinear Considering " relationship. the curvilinear load-deflection relationship for composite beams,

this stage can be divided for analytical (a) Partially cracked concrete

purposes into two parts:

(1 to 2)

Where only part of the concrete in the tensile zone is assumed to be cracked. The corresponding cross section for this stage is No. 2 in Fig. 17. The

uncracked concrete area in the tensile zone is equivalent to the confined concrete by the presence of the f. r. c. channel. The logic behind this the width and height of cracks towards

assumption is that the f. r. c. channel is controlling travel of the cracks. Therefore,

the extension of the flexural

the neutral axis level is limited

and will mainly be affected by the height of

the upstands of the f. r. c. channel.

(b) Fully cracked concrete

(2 to 3). cracked

Where concrete in the tensile zone is assumed to be completely and only the contribution of the flexural No. 3 in Fig. rigidity,

of the f. r. c. channel is considered in the calculation the corresponding cross section for this stage is along the span of the

17. The distribution

of the curvature

be lower beam and more uniform can composite beams. (iii) Fully cracked stage (section No. 4 in Fig. at this stage is represented

when compared with ordinary

17)

The behaviour

from points 3 to 4 on the curve

the tensile that At 3 is it strength Fig. maximum in 16a. assumed point shown form in the f. is channels. the will cracks reached and of r. c. units At this stage the tensile stresses which are carried by the f. r. c. channels to the steel at cracked concrete sections. This increase

will be transferred

in the steel strain could affect the average curvature deflection However, is expected. the deflection behaviour of the following

of the beam and a greater

by be this each, or can affected stage at

a combination

factors: flexural to the f. the r. c. channels of rigidity

(1) The degree of contribution of the member.

(2) The magnitude of the cracking load of the f. r. c. units in relation ultimate strength of the beam. f. the of r. c. units, mechanism

to the

(3) The effect of the fracture of flexibility

and the amount

that these channels exhibit before cracking. between the f. r. c. units and the concrete. and composite beams

(4) The bond efficiency At point 4 in Fig.

1Ga the deflection for both ordinary This is especially

is expected to be equal. severely cracked, flexural rigidity

so when the f. r. c. channels are to the

and they are not making any more contribution

of the member.

5.3.5.

Methods

of Calculation

Approaches employed for predicting and composite lightweight


Approach one - Existing

the short-term

deflection for ordinary

concrete beams are discussed in the following: Theories for Ordinary Beams

To illustrate for lightweight

the extent of the existing theories in predicting

the deflections

concrete beams, a comparison

is to be made between the predicted The methods used

values of deflections

using these methods and those observed. These are: -

are discussed in appendix (F).


Based on a cracked (37). (82). section.

CEB formula Branson

method

Yu and Winter method (83). Beeby and the Unified draft code method (81) (84). CP110 method (5). The calculated values obtained by employing the above methods are discussed in 7.7.2.4. Approach two - Empirical Method for Ordinary and Composite Beams of the beam is to be used the

In this approach the flexural in the calculation maximum

strain distribution

of the average curvature.

From the strain distribution

compressive

strain in the concrete at the top element of the compression axis depth can

zone, the strain at the level of the steel and the level of the neutral be determined. From these values the experimental expression: curvature

(4 exp) can be

obtained from the following


' exp = ecexp

(5.4)

Xexp or
exp = ec exp + etexp dl . (5.5)

Where ecexp : experimental compressive strain in concrete at the extreme zone.

element of the compression

etexp : experimental
Xexp : average

steel strain
neutral axis depth

experimental

The accuracy of this method in predicting on the following points: -

the amount of deflection

depends

(1) The measured strain distribution the average values.

across the depth of the beam represents

(2) The general accuracy of the readings taken deflection and the determination

regarding

strain measurements,

of the position of the neutral axis level.

This method also proves that a logical approach can be made to calculate the average curvature flexural the beam from the the of values observed of strains

and the neutral axis depth.


Approach three - Proposed Method

(a) - Ordinary

Beams of flexural members is calcualted This average steel stress,

In this approach the average curvature by considering

an average value for the steel stress. the restraint

is calculated by considering in the tensile zone. the vicinity

between the cracks the concrete action of

As stated earlier

of a crack and minimum

in 5.3.1. , the steel stress is maximum in at a point mid-way between two adjacent cracks. 18. the average contribution 18.

This is represented To simplify

in a diagram shown in Fig.

the derivation

of the average steel stress,

of the concrete in the tensile zone is represented

in section 2.. 2 of Fig.

It is assumed that the average steel stress can be obtained by deducting the average stresses i. e. from the between the the cracks maximum concrete resistedby steel stress

(save = fs - fs Hence the average curvature ave=fs


.1

can be calculated from the following

expression: (5.6)

-fs
Es (dl - X)

it can also be expressed by


e

ave = (fs /Es fs) -

11

+ fc/Ec

(5.7)

dl
f

-63-

Where fs : maximum steel stress


fs X: fc : average neutral tensile stress resisted by concrete section at top element the contribution of the compression of concrete in tension between cracks

axis depth for a cracked stress in concrete

: compressive zone calculated

by considering of concrete

Ec

: modulus

of elasticity

To obtain the value of fs, the following

procedure is adopted: considering

Fig.

18;

the tensile force Tc for the concrete in tension is equal to To = koAeft Where ko : constant depends on the distribution of the steel Ae : effective concrete area in tension ft : average tensile strength of concrete of the stress and bond efficiency

By expressing Ae = K1 bd1 ft k2 fr (fr : modulus of rupture) in the concrete becomes (5.8)

The equation of the tensile force Tc = KO Kl K2 bdi fr

= K3 bdl fr

The moment of resistance of the force Tc.

due to the force T. is Tc lc, le being the lever arm

The stress due to the contribution fs = Te 1e As(dl X) -

of concrete in tension (fs) is equal to: (5.9)

By assuming 1c = dl -Px
fs = K3 bdl a K3 fr/P Where p= As/bdl fr/As

(5.10)

The average curvature

as given in equation (5.6) is

4ave

= (fs - fs) Es (dl - x)

By substituting

the value of fs in equation (5.6) fs - 1c3fr/p Es (dl -x) (5.11)

ave =

ave Es (d1 - x) = fs - K3 fr/P K3 = p/fr { fs - $' ave E. (d1 - x) }

(5.12) (5.13) ave, X, fr and fs the value

From this equation with the experimental of K3 can be determined.

values of

It must be emphasised here that the assumption given in the CEB recommendations with regard to the calculation in this method. calculated for the average steel stresses is similarly (as the CEB recommended adopted

The average curvature

(79), can be

from the following


_

expression: (5.14)

ec + et dl

et = et -

et

(5.15)

Where cc : compressive

strain in concrete

et : strain in steel at a cracked section


et`' : reduction of that strain including the contribution of the tension zone

of the concrete.

The sequence employed in the derivation minor assumptions)

of equation (5.11) (i. e.., some of the of

is also influenced by a method suggested in the journal the average curvature of flexural members.

the A. C. I. (85) for calculating


(b) Composite Beams

The load deflection behaviour of the composite beams has been fully discussed 5.3.4' . According to that, an idealised trilinear moment deflection relationship up to the level of the cracking moment of the f. r. c. channels (Ma) is assumed. in
This is shown in Fig. 17. For 16b. The corresponding cross sections for analysis are

shown in Fig. calculated

a given moment For

(M) the curvature reference see

of a beam can be

as in the following:

5.3.4.

(I) Uncracked For M

stage <_ Me ; from 0 to 1 in Fig. 16b, and cross section No. 1 in Fig. (5.16) 17.

4=M Ec 10

Where

Mc : cracking

moment of concrete which can be calculated as

shown in 4.4. Io based area on an untracked of second moment : section Ec : initial (ii) Partially modulus of elasticity of concrete
/

concrete

Cracked Stage cracked concrete M! 5 Mp, from 1 to 2 in Fig. 16b and cross section No. 2 in Fig. 17. Mc Ec io +K MMc (5.17)

(a) Partially For Me >

Ec IP
(5.18)

MP

Ma

+ Mc

2 Ip : is the second moment of area based on a partially


cross section No. 2 in Fig. 17.

cracked section,

The neutral axis depth (dip) and the second moment of area (Ip) can be calculated by considering an untracked area of concrete equal to that confined by

the presence of the f. r. c. channel in the tensile zone as shown in section No. 2 Fig. 17. This is discussed in 5.3.4 for the partially cracked stage of the loadof the equations are given

deflection behaviour. in appendix H.


(b) Fully For Fig. 17. Mc Ec Io Cracked MpC M'

The assumptions and derivations

Concrete Ma: from 2 to 3 in Fig. 16b, and cross section No. 3 in

Mp
K Ec Ip

Mc
K Ec Icc

(5.19)

Where
(5.20)

Ma : cracldng moment of the f. r. c. channel K: flexural depending K, effective on an constants at the stage in consideration
Values experimental calculate for constants results. K, K can be determined the foregoing

rigidity

for the section

by applying it will

the equation be possible

to to

Using

relationship,

the curvature

at any section

along the member,

consequently

the deflection

integration found by be normal can


5.4 Limit State of Cracling

procedure.

5.4.1

Introduction The limitations on the width of cracks in reinforced of the reinforcement, concrete members are and also to preserve

mainly intended to prevent corrosion

the aesthetic appearance of the structure. travel

A reduction in the width and length of rigidity of the member.


should not exceed

of the cracks will also improve the flexural


CP110 recommends that the surface crack

width in general

0.3mm.

For members exposed to certain categories

of environments,

the allowable

the as given times cover 0.004 concrete nominal exceed should not widths crack in the code for that particular In general, members category. cracks in ordinary reinforced concrete

the width of flexural

depend upon the magnitude of the tensile strain in the steel and concrete. while not associated with a would result in greater The limited steel

Thus the use of high working steel stresses, corresponding strains

increase in the modulus of elasticity

and consequently greater width of cracks. difficulties regarding

tensile strain of

the concrete presented further The limit state of cracking,

the control of crack widths. especially for

therefore,

became more critical In flexural tensile

members reinforced

with high strength steel.

concrete members cracks stresses.

flexural by formed be applied shrinkage and/or can


Si) Shrinkage cracks

These cracks form due to the stresses that are set up by the varying rates of contraction of the cement matrix during the hydration and curing period. This

is mainly caused by the drying shrinkage that takes place in the concrete matrix. In order to control the drying shrinkage the most effective way found was
to reduce difficulties could the water content of the concrete. This, however, may present methods certain that

in the mixing

and compaction

of the concrete.

Other

also be utilized surface

are the use of a larger coating to the concrete

size of aggregate (86).

or the application

of a special

(ii)

Flexural Flexural

cracks cracks form initially at the soffit of the member once the tensile When the applied stresses are increased, zone of the

strength of the concrete is reached.

these cracks become wider and propagate towards the compression flexural member.

The incompatibility

of tensile strain between the steel and concrete is a major and width of these cracks. It has been suggested

factor which affects the formation that the state of incompatibility 62 N/mm2,

can be reached at a steel stress lower than can develop (86) (87). cracking in ordinary reinforced

at which microcracking

The analysis and mechanism of flexural beams has been well studied in the past. necessary to discuss the cracking reference following 5.4.2
(a)

In this research it is thought only

beams the in composite with special mechanism This is discussed in the

to the contribution paragraph.

of the f. r. c. channels.

Cracking Mechanism in Composite Lightweight


Cracks

Concrete Beams

Shrinkage

The formation

of shrinkage cracks in flexural

members depends mainly on

the amount and rate of drying shrinkage that takes place in the concrete.
In composite confined beams a large volume This of the concrete a reduction in the tensile zone is

by the f. r. c. channels. that takes place,

permits

in the rate and amount to reach its full tensile

of shrinkage properties

thus allowing to cracks

the concrete formation.

and be more resistant

Furthermore,

the surface shrinkage strains

maybe distributed

through

bond stresses along the length and area of the f. r. c. channels which enable more fine cracks to occur instead of a few wide cracks.

(b) Flexural

Cracks tensile strength of concrete, and extending. To illustrate flexural cracks should improvements

To improve the flexural be prevented from forming in the tensile resistance

the theoretical

of the concrete due to the restraint

action of the f. r. c. is confined by 19. The example

units a simple example is considered where a concrete prism f. r. c. units and subjects to a direct tension, is mainly intended to simulate the restraint flexural composite beams. as shown in Fig.

action of the f. r. c. channels in the

The direct tensile stresses acting on the concrete will be transferred means of bond to the f. r. c. units. By assuming a perfect bond relationship

by between

the concrete and the f. r. c. units the strain in the concrete and the units are expected to be equal at any level of applied stress. in the'f. r. c. units, therefore, The corresponding stress

depends upon the modular ration.

m1 = Ea/Ec

(the ratio of the modulus of elasticity The formation

for the f. r. c. units to that of the concrete).

of cracks due to applied tensile stresses and the corresponding

action of the f. r. c. units is discussed in the following: (1) At rather low applied stresses concrete), microcracks (lower than the tensile strength of the

can form at the weak sections in the concrete,

an increase in width and extension of these cracks may lead to a complete concrete failure. Nicrocracks which originate either from the outer to a great

surface or the inside mass of the concrete can be controlled extent by the restraint action of the f. r. c. units.

For cracks originating 19), the f. r. c.

from the outer surface of the concrete (cracks (a) in Fig. units will arrest will be controlled. them from further width increase,

hence their extension

For the cracks that originate

from the inside concrete

mass (cracks (b) in Fig.

19), their extension towards the outer surface mainly because the elements near the due to the restraint action of the f. r. c. The restraint

of the concrete will be limited, outer surface of the concrete, units, will have a greater action of the f. r, c. units, concrete,

resistance therefore,

to crack propagation. allows greater

tensile stresses in

thus allowing values nearer the full tensile strength to be achieved.

(2) When the theoretical

tensile strength of the concrete is reached tensile

local bond failure between the form in the concrete and concrete cracks This, these in the f. of cracks. the vicinity r. c. units may occur and however, failure lead to a complete will not of the concrete prism since the

the restraint

f. help by in arresting the c. r. units action exerted

these the cracks. of width and growth The restraint depends this f. the on stage mainly at r. c. units action of the the f. the of absorbing capacity and r. c. units of of the cracks.

the bond efficiency

strain produced by the formation For effective crack control. should have: (a) A higher (b) A greater tensile strength.

the f. r. c. units compared with the concrete

extensibility.

An increase in the modulus of elasticity help in arresting

further f. the units will r. c. of When the concrete cracks, to the f. r. c. units. will have This may of f. r. c.

the phenomenon of cracking.

the tensile stresses in concrete will be transferred Fibre reinforced

low modulus of elasticity of a cement units

high those local modulus. a of strain compared with a greater result in a reduction units in controlling of the restraint the cracks. action and the effectiveness

(3) At the point where the concrete is heavily cracked, Fig.

as shown in (4) of resisted by the strength or of a crack

19, the applied tensile stresses will be completely

f. r. c. units, extensibility

When the units reach their maximum tensile cracks will occur. This can be at the vicinity

that has already occurred in the concrete. failure of the prism would be expected.

At this stage a complete

It can be seen from this example, be obtained from incorporating (a) The full utilisation (b) Additional In flexural which originate members,

that the possible advantages that can

f. r. c. units are: of the tensile strength of the concrete.

tensile strength. the flexural rigidity will be reduced mainly by cracks The rate of reduction in

from the outer tension face of the beam.

the flexural flexural

rigidity

depends mainly on the rate of formation the flexural

and propagation of the

cracks.

In general,

cracks have their maximum width at members) It

the outer tension surface of the concrete (soffit of the beam in flexural and gradually diminish

in size as the crack extends throughout its length.

would then follow member.

that the f. r. c. units should be placed at the tensile zone of the action exerted by the f. r. c. units in arresting at this position. of a member can also be affected by the cracks which These cracks can be controlled in similar manner the cracks

The restraint utilised rigidity

will be effectively The flexural originate

from internal explained.

flaws.

as previously

The restraint

action of the f. r. c. units,

therefore,

ban increase the loading member. This range

range within which the beam would behave as an untracked

could be adjusted by the design of the f. r. c. units so that no surface cracking would occur up to the design working load. Although the modulus of elasticity of the concrete, affected. the stiffness of these units is slightly lower than that

of the member at the untracked

stage is not severely

This is discussed in 5.3.4. The

When the concrete reaches its full tensile strength cracks will occur. action of the f. r. c. units as explained earlier opening excessively; in a gradual manner. diminishing

from these be to prevent cracks will zone will. be for

hence their extension towards the compression

The role of the f. r. c. units at this stage is important rigidity of the member.

the rate of reduction in the flexural

Furthermore,

the reduction in crack widths will also diminish and preserve

the risk of corrosion

to the steel

the aesthetic appearance of the concrete member.

The final stage is when the f. r. c. units reach their maximum tensile strength or extensibility be completely and cracking occurs.. However, the composite action may not Provided that there is a

lost at the formation

of the first

crack.

good bond between the f. r. c. units and the concrete the f. r. c. units will still be effective for the rest of the beam in arresting 5.4.3 Methods of Calculation Crack widths for ordinary (Ordinary lightweight flexural Beams) concrete beams can be calculated cracks.

employing the methods suggested in C and CA (88) and CP110


The CP110 formula is: w air em w=3 1+2 (acr - c)

(5.21)

d-x Where em = e1-1.2bd(a-x) As (d-x) fy el x10 3 (5.22)

: is the strain at the level considered,

calculated ignoring

the stiffening

effect of the concrete in the tension zone. The C and CA formula


w= 3.3.; fs Es The term d-x dl -x

for the maximum crack width is: (5.23)

(d - x) (dl - x) the 1.0 when crack width at the steel level

reduces

is sought

The calculated values obtained by employing the above methods are discussed in 7.73.3

CIIAPrER MANUFACTURE G. 1 General

SIX OF TESTING

AND METHODS

A total of 27 ordinary configuration conditions

beams The loading tested. were and composite

to tests simulate types adopted mainly were employed of and

met in practice. and the loading was applied the span length of the beam.

All the beams tested were simply supported, through two symmetrical points acting at one-third

This provided a constant moment zone for the middle third region of the beam. With this system of loading that produced by a uniformly Static, the central deflection is 2.04% greater than load for the same applied moment. out under the same the manufacture of beams

distributed

fatigue and sustained loading tests were carried of loading. In the following paragraphs

configuration and specimens


6.2 6.2.1

and the tests employed are discussed.

Manufacture Ordinary Reinforced Lightweight Concrete Beams

The beams were cast in the concrete laboratory type mixer with a capacity of 0.25 m3 was used. aggregate were first mixed dry for two minutes,

where a horizontal

pan

The cement and air dried then the water was added and

the total time for mixing was about five minutes. The concrete mix was carried batch the workability slump test. out for each beam in two batches. For each

factor by the compacting apparatus and the was measured out in accordance with B. S. 1881

These tests were carried

Values obtained for the compacting factor and slump tests for all the beams 40mm between 20 between 0.95, 0.8 respectively. and varied The mould used for the test beams consisted of two steel channels, which formed the sides of the beam. steel table. Cover to the main reinforcement at the soffit of the beam was obtained by These channels were 150mm apart fixed to a

temporarily

lifting

the reinforcement

cage to the desired height by wires fixed 35mm thickness were then placed to fix the appropriate side

to the top of the mould.

Pieces of timber

between the steel channels and the reinforcement cover. casting. The wires and timbers

were then withdrawn

during the process of spacers, of the

This method was adopted to avoid the use of reinforcement

as it was thought that their use might affect the initiation cracks.

and distribution

Compaction of the concrete was achieved by the use of three vibrators diameter by 25mm the bed to the the a of use of mould and also clamped vibrator. poker

When the concrete had been fully compacted the top face of the beam levelled off. The beams were then left to cure after casting,

was carefully

covered with wet hessian and polythene sheeting. After 24 hours. the sides of the mould were stripped and the beam was three to four days, after which it was then

kept on the steel table for a further marked for positions


discs These 200mm. were placed Demec

of support and load.

At the middle of the span "Demec"


This is shown in Fig. of prism 20.

in position

over the depth of the beam. by an'Araldite" also fixed

discs were locating

affixed

glue over a gauge length on the sides of one small

Demec

discs were

and three cylinders

so that shrinkage measurements

and the modulus of elasticity

for the concrete could be obtained. 6.2.2 Composite Reinforced The manufacturing as that for the ordinary Lightweight Concrete Beams

process of the composite beams was nearly the same beams, the difference being that fibre reinforced channels

were placed at the soffit of the mould before the reinforcement

cage was fixed.

Before casting of the beams commenced the channels were wetted with plenty of water. concrete.
6.2.3 Control Specimens for Concrete

This prevented water being absorbed from the freshly

mixed

From the first the following control

batch of concrete which formed the lower half of the beam specimens were cast: -

(i) Six beams of 100 mx tests


(ii) Three

100mm x 500mm for flexural

and direct tensile

100mm cubes for the concrete

compressive

tests.

These were in the upper half.

only for a comparison From the following

with the strength

of the concrete the upper

the second batch of concrete control specimens were

which formed

half of the beam

cast: -

(i) Six 100mm cubes for the concrete compressive (ii) Three cylinders of elasticity 6.2.4 of 150mm diameter

strength tests

and 300mm long for the static modulus strength tests.

and the concrete compressive

Fibre Reinforced
The process

Cement Channels (f. r. c. channels)


for manufacturing asbestos the f. r. c. channels was basically the

employed

same as that applied

for normal

cement products.

The f. r. c channels were first required

manufactured

in the form of thin sheets to the

length, total width and thickness.

While still wet, the sheets were placed In the bending process employed to form wcurred at the corners.

on a wooden mould and formed into shape. the upstands of the channel 6.3
6.3.1

longitudinal

cracks occasionally

Static Loading Tests


Arrangements and Conditions of Loading

The static tests for most of the beams were carried laboratory with the loading arrangement

out in the concrete and plate (1).

as shown in Fig. 3,

The rig used for testing the beams ccausisted of a steel portal frame whose columns were made of steel channel section, bolted to the sides and floor of a 1.27 m width trench. The end supports were Universal steel section beams (165mm wide x 310mm deep x2m long). These were placed, fixed and supported across the trench, their centres being symmetrically frame. At each support the beam was rested on a 25mm diameter steel roller

placed 4.5 m apart from the centre line of the portal

sandwiched by two steel plates of 25mm thickness. glued with "Evostick" the soffit of the beam. to the support,

The lower steel plate was to

while the upper one was fixed likewise

At one of the supports the roller other support the roller

was welded to the lower plate, and at the to accommodate the longitudinal

was allowed free rotation

load. beam deflected the under movement when Load was applied by a hydraulic portal frame. The hydraulic jack which reacted against the beam of the model T GOJ

jack was operated by a "Denison"

Console machine.

The jack load was transferred

to the beam by means of a steel

load the by beam, at plates welded point steel stiffened which was spreader positions. The spreader beam was placed on two 75mm diameter on steel plates. freely. One of the rollers steel rollers resting

the fixed the to plate while other rested was which allowed a fair alignment. Laboratory had a similar

The plates were bedded on to the beam with plaster,

level face to be achieved so that the plates were in correct The static tests carried arrangement out in the Fitton Structures described,

to those previously steel beams.

except that stools were used as supports

instead of Universal
6.3.2 Testing

Procedure

A total of 18 beams were tested under a static type of loading.

Before testing

the beams in the the the constant moment zone were of region sides of commenced facilitate to 100mm in squares a grid pattern of marked and when they occurred.. tracing of the cracks as

The beams were also examined for shrinkage cracks if

any, and their width was measured. All static tests were carried (1) Up to the working load. loading three in of as follows: cycles out As defined in 3.2 and then to zero.

(2) Up to 1.5 times the working load and then to zero. (3) Up to the calculated ultimate In the first loads of the beams and then on to failure. of 2KN for all the beams, were 4KN. At every

cycle the load was applied by increment

except for beams ST 5-0 and ST 9-0, for which the increments increment deflection

readings were taken, while strains and cracks were generally

measured at every alternate increment. The deflection readings were taken at the centre line of the beam and at the Flexural strains were measured over the depth and extension of cracks on

third point along the span length.

of the beam on both sides at mid span.

The formation

both sides of the beam in the constant moment zone were carefully The width of cracks for the ordinary

observed.

beams were measured at the level of the For composite beams the cracks

steel and at the bottom edge of the beams.

in the concrete were measured at the interface

(top of the upstands of the channels) Cracks that formed in

formed. they the the and when soffit of channel as and at the shearing zone were also observed,

and their widths checked to see whether

they exceeded crack widths that had formed in the constant moment zone.
In the unloading with readings process the decrements and strains were twice the order of the increments, as

for deflections

being taken in a similar

manner

employed in the loading process, only taken at zero load. In the second cycle a similar

measurements

for width of cracks however, were

approach to that employed in the first

cycle

for loading and unloading the beam was followed. In the third cycle the beams were loaded by increments of 4KN up to 1.5 times

the working load, after which the increments failure load.

to to 2KN then up reduced were increments were 8KN and crack widths and

For beams ST 5-0 and ST 9-0 the first All the measurements

thereafter deflections

reduced to 4KN.

for strains,

were taken in a similar

two. in one cycles and employed as manner failure approaching of when warning and crack widths, was

At the very final stages of loading, clearly

imminent by the increased rate of deflection

the dial

gauges were removed and replaced by a vertical

scale rule fixed and placed at the to be observed. strains

mid span of the beam. so as to enable the final amount of deflection

For some beams an attempt was made to measure the compressive in the concrete as near as possible to failure. 0.97 of the failure
6.4 6.4.1 Fatigue

Final readings were taken at 0.95 and

load for beams ST 10-0 and ST 2-C respectively.


Tests and Conditions of Loading

Loading

Arrangements

The arrangements adopted for fatigue tests. hydraulic jack,

and conditions of loading used for static tests were similarly The difference was that the load was applied through a

connected to either an S. B. E. 120, or an S. B. E. 80 Losenhausen

fatigue tests machine.

The range of loading cycles consisted of an upper limit working load and a lower limit

equal to the applied The rate

equal to half the applied working load. deflection

of cycling employed was dependent on the maximum the upper and lower limits.

of the beams at

For the beams tested the rate of load employed varied

between 60 and 100 cycles per minute.


6.4.2 Testing Procedure

A total of five beams were tested under fatigue loading conditions. beams initially were subjected to a static test similar to the first

These

cycle as previously

explained in 6.3.2. After the completion cyclic loading. of the first static test the beams were subjected to a repetitions was applied to the beams tested of about

A total of three million

under fatigue loading. every half million, relevant

The sequence of cycling was stopped at intervals

be loading test that carried could so a static

out with all the

data being obtained as was the procedure previously

mentioned.

On completion

of the total number of cycles the beams were subjected to a The test was carried out in three cycles similar

static loading test up to failure. to that explained in 6.3.2. Occasional

stoppages due to mechanical

failure

of the machine occurred

during the test, but apart from the time being lengthened for the test the testing procedure was satisfactory. Table (6) shows the number of cycles and age of the

various beams tested. 6.5 6.5.1 Sustained Loading Tests Arrangements and Conditions of Loading out by setting two beams (ordinary

The sustained loading tests were carried and composite)

back to back; the load was applied by means of tension springs The two beams

acting at the end supports of the beams, as shown in Plate (5). (ordinary and composite) had the same working moment,

and were separated

by spacing units which were positioned

1.5 m apart. the distance between

The two beams were supported on two steel trestles; each support and the mid span point was calculated

so as to produce a similar

moment at the mid point of the beams.

In calculating

this distance the dead

weight of the beams and the weights of the spring assembly acting at the end supports of the top beam was taken into consideration.
Since the deflection the spacing total units, of the beams

See appendix

Q.
relative to

at the mid point was measured configuration

it can be shown that for the loading

adopted the

deflection

of the beam is equal to 7.7 times units. See appendix (G).

the deflection

at the mid point

relative

to the spacing

After the supporting trestles

had been positioned and levelled

the lower

beam was placed on these supports. roller

At each support the beam was rested on a For one of these rollers at the other movement

which was sandwiched between two steel plates.

movement was restricted support the roller of the beam.

by V shaped grooves machined in the plates,

was free to rotate,

thus compensating for anylengitudinal

The spacing units separating the two concrete beams consisted of a square steel bar sandwiched between two pieces of square steel block with aV shaped incision. These blocks allowed free rotation mix. and were fixed to the concrete beam

with a plaster

The upper beam was placed in position with its compressive

face resting on the spacing units. At each end of the beams two tension springs each of 35XN capacity were connected to steel plates by means of riveted shackles. transmitted The applied load was

to the beams through a steel ball and plate assembly.

When load was applied to the beam, demountable loading plates as shown in plate (5) were attached by steel rods to the springs.
rM

The load was then applied

by the use of a calibrated

EPCO hydraulic

jack which was placed between the When the required

plate at the spring connection and the demountable plate. load was reached by the jack

it was held in position by means of screws that

were part of the spring assembly. Each time measurements initial were taken the load was brought up to its as the load could be

value (applied working load).

This was necessary

reduced due to creep

despite the compensating

effect of the springs.

6.5.2

Testing

Procedure

For each pair of beams was applied by increments deflection

a static test was carried

out, in which the load At each increment at the level of the

of 2KN up to their working load.

readings were taken by means of dial gauges positioned deflection measurements

spacing units,

were also taken at the supports and one-third were also

the span length of the lower beam.. Strains and cracks measurements taken.

The applied working load was then sustained on the beams, with further readings for strains, deflections and crack width being taken at periodic intervals.

These were more frequent in the early stages of the sustained loading period.
6.6 Instrumentation Measurements of the test beams were measured by setting span loading three points. dial gauges, The dial

(a) Deflection

The deflections

one at the mid span and one at each of the one-third gauges of 50mm travel a magnetic with base. 0.01mm scale divisions

were fixed by means of a with system

stand with thin

The spindles

of the gauges came into contact For the sustained 0.01mm loading

steel plates

glued to the soffit

of the beams. with

the dial gauges used were

of 25mm travel to light

scale divisions,

the gauges units.

at the mid span being fastened

steel channels

fixed to the spacing

(b) Strain Measurements The flexural strains across the depth of the beam were measured by means "Demec" strain gauge of 200mm length with scale gauge was used

of demountable mechanical divisions indicating

a strain of (0.82 x 10-5).

The same "Demec"

to measure strains due to shrinkage, of elasticity.


(c) Crack Measurements

creep and also for tests to find the modulus

For measuring the width of the cracks with scale divisions of 0.1mm was used.

a small illuminated

hand microscope

6.7

Other Tests

6.7.1
All

Control Tests on Concrete


the control specimens for the concrete control properties carried were tested according the of are

to the B. S 1881: 1970 (89). These structural the concrete presented properties

tests were used,

out to investigate the uniformity

of the concrete tested.

and also to ensure values

for all the beams (6).

The average are: -

of these properties

in table

The tests

employed

(a) Cube Compressive

Strength

From the second batch of concrete which formed the upper half of the beam six 100mm cubes were cast as control an age of 7 days specimens. Of these, three were tested at The cubes

and three were tested at the time of testing the beams.

were tested in an Avery hydraulic 1800121. (b) Modulus of Elasticity Three cylinders

testing machine with a maximum loading range of

and Cylinder

Compressive

Strength

of concrete size 150mm diameter x 300mm length were

tested from the batch of concrete placed in the upper half of the beam.
The cylinders measurements alumina cement were marked out at 900 intervals, testing, tested at which positions were strain

were taken. mortar

Before

the cylinders in a Denison strain

capped with high testing were machine taken at had been

and then were

Universal

of 3000KN capacity. the four positions taken the cylinders the total testing

At each 5KN increment, on the cylinder, were then tested after

measurements

an adequate number and their

of readings

to failure

maximum

load recorded,

time was about (15 minutes).

A typical experimental in Fig. 6b.

stress-strain

curve of the lightweight modulus of elasticity

concrete is shown from the experiment, -

Values obtained for the initial

curves for the various beams are given in table (6).


(c) Direct Tensile Strength

To obtain values for the tensile strength of the concrete three prisms 100mm x 100mm x 500mm were tested in a "Denison T. 42. B4 machine" of 500KN

capacity.

To avoid eccentricity

of loading during the testing of the prisms

special

jaws were used.


(d) Modulus

The values obtained for the various beams are given in table (G)

of Rupture values for the modulus of rupture were at the time of testing the beams

To obtain three

100mm x 100mm x 500mm prisms testing

tested

in flexure

in a "Denison at

Universal one-third

machines" of 500KN capacity

by which the load was applied

the span length.

(e) Shrinkage Values for shrinkage were obtained by means of a "Demec" 200mm length with readings being taken on four sides of a prism 500mm). conditions shrinkage These prisms laboratory kept in the concrete were strain gauge (100mm x 100mm x

under similar

of exposure as the full scale beams. strain measured on these prisms

The maximum values of the free

were 0.0005 and 0.00066 after a period

of about 30 days and 60 days respectively. For beams tested under sustained loading. the shrinkage strains were also measured on the sides of the beams near their end supports. a beam was 5m and the span length adopted was 4.5mm. unloaded at each end, at which the measurements values obtained are discussed in
6.7.2 Tensile To obtain Tests on Steel properties of the various steel reinforcement in direct of 200KN. tension used using

The total length of

This left a 250mm length The

of shrinkage were made.

8.3.1.

the tensile

in this Investigation, a "Denison T42.

samples

500mm long were tested with a capacity a "Lindley" with

C2 testing were

machine"

The strain gauge a strain

measurements length

made by using extension

extcn meter

of 50.8mm indicating

and maximum

of 2.54mm,

scale divisions

of2.5x10-5.

The mechanical properties Typical stress-strain

of the steel bars used are given in table (5) 1

curves for the various types of bars used are shown in Fig. in general, differed

The yield or 0.2% proof stress of the various bars tested,

Uy + 5%.

6.7.3

Tests

on f. r. c. Units

In order to investigate used in this investigation,


and bending. used or from properties deformation These sheets

the tensile and bending properties various


were

units of the f. r. c-. '

samples were tested in. direct tension


obtained either by cutting from

samples aipplied

the channels

especially tested

for this. purpose. are given in table and bending

The mechanical (4); the load 8

of the various characteristics out are: Tensile

samples

in direct

tension

are shown in Fig.

The testscarried (i) Direct

Test

Samples of 50mm width,

5mm thickness and 250mm long were tested in a The samples were fixed by an epoxy during the process

"Denison T. 42 machine" of (65KN) capacity. resin mortar of loading. Strains in the direction to special clamps,

so as to avoid any eccentricity

of the applied stress were measured at increments gauge of 50.8mm length with scale division of

of 0.5KN by means of a "Demec" 2.48 . x 10-5. The tensile properties


(ii) Bending Test

of the various

8a in Fig. tested are shown samples

Samples of 30mm width, bending.

6mm thickness

and 240mm span were tested in where the load was applied the span

The samples were supported on steel trestles The load was transferred

manually in kgs. length.

to two points at one-third

At each increment,

which was 1 Kg, the central deflection was measured and a scale division load were recorded. of 0.01mm. The

with a dial gauge of 50.8mm travel deflection deflection near failure and the failure

A typical load

curve is shown in Fig. 8b. modulus of elasticity obtained from these tests are

Values of the initial given in table (4). ,

CHAPTER

SEVEN

DISCUSSION THEORETICAL 7.1

OF TEST RESULTS PREDICTION

AND COMPARISON

OF TEST

BEHAVIOUR BEAMS

W IThI

FOR ORDINARY

AND COMPOSITE

Introduction

beams, test the behaviour the discusses This chapter of actual various aspects of behaviour with the theoretical predictions

the compares and

in accordance with

chapters four and five. neutral

This includes the variation compressive

with the applied moment for the stresses in the

axis depth, flexural

strains in the concrete,

f. r. c. channels and the stresses in the steel. states of ultimate strength, cracking

Emphasis is placed upon the limit

and deflection. behaviour of ordinary

A direct comparison beams. composite and 7.2 Variations

is also made between the flexural

of the Neutral Axis Level axis depth for ordinary and composite beams at any

The level of the neutral

level of applied moment was obtained from average values of strains measured on both sides of a beam. for ordinary The "Demec" the for values of strains measuring points used

20. Fig. in beams shown are and composite


prediction for the level of the neutral axis depth in relation relationships for to

The theoretical the applied ordinary of values percentage These moment

has been explained beams

in 4.2 and the idealised 10.

and composite

are shown in Fig.

To facilitate

the determination

Co, C1 and C2 shown in Fig. of steel are plotted

10, the results irrespective

of beams with a similar of the type of steel used.

on one graph,

are shown in Figs

21 to 25.

From an examination values of 0.1,0.6 however,


(a)

of the experimental

results

it was found suitable to assume For ordinary beams,

C2 for C1 Co, 0.8 respectively. and and

the value of Co was assumed to be zero.


Beams

Ordinary

In Figs.

21 and 22 are plotted the theoretical

and experimental

values of the moment.

neutral axis depth versus the ratio of applied moment to the observed ultimate

The values of the observed ultimate col. 9 of table (7). The theoretical

in the in given used graphs are moments values of the neutral axis depths were obtained and 4.5. These equations

in accordance with 4.2 are:

by employing equations 4.1,4.3

For the uncracked stage


n= Xuo dl = A2 2X+ +2m1) p (m - 1) p]

(4.1)

For the transition


S-YCXCC

stage (R2 -2 where R= Cl R) + Xuc M/Mu (4.3)

C12

For the fully cracked stage


J1 n= Xco = mp ( +2 mn mp (4.5)

-ri r,

The values of the parameters p and d1 are as given in table (1), From Figs.

incorporated A= d/dl

in these equations were as follows: = 300/d1 and m= 12.

21 and 22 it can be seen that the level of the neutral axis rises for that higher than an untracked level loading calculated to a of at an early stage have does depth stable position a the not that This axis neutral shows clearly section. due in 4.2 is behaviour This possibly loading. suggested initial as the stage of at to an early formation of cracks in the ordinary reinforced lightweight concrete beams. The theoretical 0.8) " graphs at the transition and cracked stages (i. e. up to M/Mu

agree well with the experimental

values. = 1, some scatter in the results involved in the determination (at M/Mu = 1), where the

For values between M/Mu = 0.8 was observed. of the neutral neutral

and M/Mu

This could be mainly due to an error axis level at the ultimate load conditions

by the strain compatibility depth was calculated axis However,

method using the a definite conclusion

actual values of strength for concrete and steel. on the behaviour

of the neutral axis level at values of M/Mu greater than 0.8 cannot number of observations made at this stage and also by accompanied

be drawn because of the limited because of the variation

of the non-elastic

strain of the materials

creep under high stresses.

(b)

Composite

Beams

In Figs. of the neutral ultimate

23,24

and 25 are plotted the theoretical

and experimental

values

axis depth versus the ratio of applied moment to the observed The values of the observed ultimate The theoretical moment used in these graphs axis depth

moment.

are given in col, 9 of table (7).

values of the neutral

were obtained in accordance with 4.2 by employing equations (4.2) (4.4) and (4.6). These equations are: For the uncracked stage n= Xcu
ul
-3-

= 0.5 A+

A2 + (m - 1) p+
(m - 1)p + (ml stage -

(m1 - 1)pl 1)pl 71

92

(4.2)

For the transition X= Xuc Co -Xcc 2C

C+C 011

(R2_2clR+c12)xC
(4.4)
1+ 2(mp + m1pi T12) (mp + mip1T) )2 1 4. G

For the fully cracked stage


n= XCC -d1 = (mp

The values of parameters as follows:

incorporated

in these equations for the beams were

X= d/d1 = 300/dl 12, dl in table (1), p and m= as given where Ach = 1548mm2, ml = Ea/Ec = 0.7 and

P1 = Ach/bdl T1=d2/d1 = 284/d1.

It can be seen that the level of the neutral axis depth does not rise higher than that calculated for an untracked section at the early stage of loading. composite beam.
is obtained transition between the stages

This clearly

shows the distinct behaviour for an untracked


From theoretical the graphs

it can be seen that good agreement values for the uncracked,

and experimental = 0.8).

and cracked > 0.8)

(i. e. up to M/Mu in the results The arguments

In the non-elastic the behaviour

stage (when M/Mu being similar beams

some scatter beams.

was observed, suggested

to that for ordinary can be similarly

earlier

for the ordinary

applied

to the composite beams.

(c)

Comparison

Between

Ordinary

and Composite

Beams

For both ordinary between the theoretical

beams composite and and experimental

a good correlation

was obtained In

values of the neutral

axis depth.

most cases the experimental the theoretical In general, ones.

values of the neutral axis depth were greater than

the neutral axis depths for the composite beams were greater ordinary beams as long as the f. r. c. channels did neutral axis depths for composite beams

than those of the corresponding not crack.

The ratios of the experimental ordinary

to those of corresponding

beams, at applied working load, varied between The exception to slightly

1.09 and 1.213 depending on the percentage of steel employed.

this was beam ST 4-C which had a neutral axis depth. at the working moment less than that of the corresponding ordinary beam ST 4-0.

This was mainly due to

the f. r. c. channel of beam ST 4 -C cracking at a level lower than the working moment. in the theoretical suggested as predictions behaviour therefore The agrees well with 5.3.2. 7.3 Flexural Strain Distribution the depth the members were obtained of strains across 20.

The values of the flexural

from the "Demec" readings at the mid span of the member as shown in Fig.
(a) Ordinary A typical Fig. 26. Beams flexural strain distribution for an ordinary beam is shown in

For this it can be seen that the distribution of a straight in the flexural line distribution. tensile Erratic

of strains results, started

agrees with the were observed

assumption for strains

however, to form.

zone when cracks

The distribution within and

of these outside concrete

strains

then depended on the disposition gauge length". The linearity load.

and propagation

of cracks strains

the "Demec

of the compressive

in the

was maintained

up to the failure

Values for strain at the level of the working affected by the second and third cycles of loading.

moment were not significantly Similarly the values of strain

at 1.5 times the working moment (the highest level at the second cycle) were nearly the same in the second and third cycles of loading.

(b)

Composite In Fig.

Beams 27 a typical strain distribution strains for a composite in the concrete beam is shown. strains

The linearity

of the compressive

and the tensile

in the f. r. c. channels was well maintained up to the cracking channel, irrespective in justifying

moment of the f. r. c. This is important

of whether the concrete was cracked or not.

the assumption of a straight line strain distribution

used in the calculation

in 4.4. f. discussed for the channels r. c. as of stresses The disposition of cracks in the flexural tensile zone of the concrete in the

f. the between the top r. c. channels and the neutral axis of of upstands portion level caused some erratic cracking erratic results for strains measured in this region. When the

moment of the f. r. c. channel was reached, results

as can be seen in Fig. 27,

for the tensile strains in the f. r. c. channel were observed Provided that the f. r. c. channel did not show any signs of cracking, effect on the strain values.
Beams

(e. g. at 113KN).

the second cycle of loading had a negligible


(c) Comparison Between Ordinary

and Composite

The strain distribution for ordinary beams.

for composite beams in general was more uniform

than

The composite beams also showed lower values of strain and moment of the f. r. c.

greater values for the neutral axis depth up to the cracking channels. For the same cycle of loading the remaining

strains, in the composite beams ordinary beams. The

were lower compared with those of the corresponding remaining strains after the first

beams loading ST 3-0 and ST 3-C in of cycle


At the outer element of the compression with 0.0133%

are shown in Figs zone beam ST 3-C for beam ST 3-0.

26 and 27 respectively. had a remaining strain

equal to 0.0083%

compared

This shows that the remaining

strains in the ordinary The cracking

beams are usually much moment of the f. r. c.

greater than those in the composite beams. channels in some beams had a considerable strain in the concrete. 7.4 Variations

effect on the values of the compressive

This is discussed in 7.4 Compressive Strain in Concrete strain in the concrete with applied

of the Flexural

The variation

of the maximum compressive

moment is explained in 4.3.2;

the idealised relations lla.

ips of the ordinary

and

composite beams are shown in Fig. results

From an examination

of the experimental lla).

it was found suitable to assume a value of 0.8 for Cp (shown in Fig. results

This agreed well with experimental carried

obtained in a previous investigation

out on normal weight concrete beams (4). Beams and experimental values of the maximum flexural compressive

(a) Ordinary

The calculated

the to the observed ultimate in moment of applied ratio concrete are plotted against strain moment as shown in rigs. The theoretical 28 and 29. by employing

4.3.2. in with accordance values were obtained

the equation of compatibility strain distribution. The equation employed is: M= fs As (d1 - x)

line the of straight a assumption of moments and

(4.18)

The values of ec (compressive ec=fs Es (dl X - x)

from: be the in obtained can concrete) strain

Values of the parameters M/11 = 0.8) were as follows

level to the (up the in of equation employed :x= the calculated value of neutral 21 and 22, axis depth

for cracked transformed P=0.345. maximum

section given in Figs.

Es = 200 N/mm2,

The value of used was an average value where the measured compressive strains in the concrete up to M/Mu = 0.8 were considered. strain in the concrete

The relation between the value of and the compressive is explained in 4.3.3.

From the graphs it can be seen that a good agreement is obtained between the predicted and the experimental values. However, above. the value of M/Mu = 0.8

some scatter in the results was observed.

This could be due to the difference strain in the strain for

between the assumed and actual values for the maximum compressive concrete at failure.

The values of the measured maximum compressive

the test beams are discussed in section C below.

(b) Composite Beams In Figs. 30 and 31 are shown the experimental and the theoretical values

for the maximum compressive moment to the observed ultimate The theoretical

strain in the concrete versus the ratio of applied moment. the equations

values were calculated in accordance with 4.3.2, beams.

employed being the same as for the ordinary formation

This was mainly due to the = 0.8, and beams.

lower than M/Mu f. in the values of cracks r. c. channels at of the members was regarded as similar 30 and 31 that the theoretical ones obtained at the initial

the behaviour

to that of the ordinary

It can be seen in Figs. than the experimental because the contribution Thereafter values. Similar

values are slightly

greater

stages of loading.

This was mainly in the calculation.

of the f. r. c. channels : was not included

a good agreement is obtained between the theoretical

and the experimental

to the behaviour of ordinary

beams,

some scatter in the results was = 1. The compressive affected by the

observed at the stage between M/MU strains

M/Mu 0.8 and =

in the concrete for the composite beams can be greatly of cracks in the f. r. c. channels. and ST 12 -C Typical

formation

examples are beams ST 7-C, 31, where they show a = 0.8 and M/Mu = 1.

ST 10 - C, ST 11 -C considerable
(c)

as shown in Fig.

amount of strain at the stage between M/Mu


Between Ordinary and Composite Beams

Comparison

As mentioned earlier phase of behaviour

in section (a), the scatter of the results in the non-elastic = 0.8 and M/Mu = 1) could be

(i. e. at values between M/Mu

caused by the difference between the assumed and the actual values of the maximum compressive strain in the concrete at failure. For both types of beams strain

values greater than 0.0035 (assumed in CP110) were obtained.


The last readings with the corresponding At levels between taken for the maximum levels of moments compressive strain in the concrete (7).

are given in cols. ultimate

11 and 12 of table the values

0.9 and 0.97 of the observed and 0.00395.

moment

of strains

ranged between

0.00217

A previous investigation strains

(90), reported values for the maximum compressive

ranging between 0.003 and 0.0055 at 90% to 100% of the observed ultimate

moment. failure

The values of the maximum compressive

strain in the concrete at such as the shape of the rate of loading

as suggested in 5.2.1 depends upon various factors zone, the type and strength of the concrete, Therefore,

the compressive

and the percentage of steel employed. are only applicable to beams of similar A definite conclusion therefore results

the values obtained in this work

conditions to those tested in this investigation.

on this point can not be made. of the present investigation showed that the values stages

The experimental of compressive

strain in the concrete for the composite beams at the initial beams. However,

of loading were lower than those for the ordinary

when the cracking

moment of the f. r. c. channel was reached the strain values of the composite beams became greater than those for the corresponding mainly due to the following reasons: ordinary beams. This could be

(a) When cracks form in the f. r. c. channel the tensile stresses resisted by the channel will be transferred the tensile strain of the steel. line strain distribution, lead to a corresponding strain in the concrete. (b) When cracks form in the f. r. c. channel more cracks will propagate in the concrete. Consequently the neutral axis level will rise and this will confine zone to a smaller greater area. Due to a higher concentration strains can be expected. of to the steel; this will cause an increase in Considering the assumption of a straight

the increase in the tensile strain in the steel can increase in the maximum flexural compressive

the compressive compressive

stresses,

compressive

7.5 Stresses in the f. r. c. Channels


In Figs. tensile values stresses were 32 and 33 are plotted the theoretical against and experimental moment. values of the

in the f. r. c. channels by correlating

the applied

The experimental

obtained

the measuredstrains

in the f. r. c. channels

with the experimental The theoretical

stress strain relationship

of the f. r. c. units shown in Fig.

8.

values of the stresses in the f. r. c. channels were obtained in The equations employed are:

accordance with 4.4. For M< Me fab =M

(d - Xuc) Iuc

m1

Mc

frc

Iuc

d dXIc For M.: 29 M<


fab = Mc

(4.12)

Ma
(d - Xvc) + Q1T Mc) (d - Xcc) ri1 (4.13)

Iuc

ml

1 icc 1

The parameters m= 12, ml

employed in these equations are as follows: = 0.7. The values of Xcc are given in Figs.

frc

=3

N/mm2 1=0.9 and 25.

23,24

In Figs. the theoretical channel (Ma).

32 and 33 it can be seen that a good agreement is obtained between and experimental values up to the cracking moment of the f. r. c.

The stresses due to the self weight of the beams are shown in the

same figures and also given in col. 6 of table (8).


The observed corresponding given in cols. values of strains at the soffits at levels just of the f. r. c. channels before the channels varied from with the are

values

of stresses, (8).

cracked,

4 and 5 of table tensile obtained

The values varied

of strains

800 to 1700 x 10 The maximum varied between

the corresponding tensile strains

stresses from testing

between

8 and 14.5 N/mm2. in direct varied tension

the f. r. c. units tensile

1250 and 1800 x 10-6; the corresponding 18.4 N/mm2. values These are already

stresses (4).

between

14.5 and these

given in table

A comparison

between

and those measured properties

at the soffits

of the f. r. c. channels had not been fully reasons:

would utilized.

seem to indicate This,

that the tensile however,

of the f. r. c. units

could not be the case for the following

(1) The strain measurements

for the f. r. c. channels were obtained over a 20, the values, therefore, represent

"Demec" gauge length of 200mm Fig.

average surface strains over this region. the f. r. c. channel just before cracking initial length. What also made it difficult

The maximum local strain in

could only be measured if the gauge

crack in the f. r. c. channel had formed within the "Demec"

to measure the actual and local strain in the strain might occur on the inside to the outer surface might not have

f. r. c. channel was that the maximum face of the channel and transference

occurred due to a possible slip or shear within the layers of the f. r. c. channels.

(2) The existence of some weak sections along the length of the f. r. c. channels. This could be due to the presence of longitudinal cracks along

the corner of the f. r. c. channels which might have formed at the time of manufacture during the process of forming the upstands. of

Another factor which may be considered is the probable variation

strength through the length of the f. r. c. channel due to a non-uniform distribution of fibres.

The f. r. c. channels for all the beams cracked in the second cycle of loading at values greater than the working moment, the exception reinforced with Kam 60 steel where t1i being beam ST 4-C cycle of moments of

charnel cracked during the first The ratios of the cracking

loading (below the working moment level), the f. r. c. channels to the corresponding given in col. 8 of table (8).

working moments for the test beams are

The ratios for the beams excluding beam ST 4-C

ranged between 1.14 and 1.81. It may be pointed out here that the f. r. c. channels for the beams tested under fatigue and sustained loading. 7.6 also did not crack during the first cycle of loading.

Stresses in the Steel Reinforcement Beams 34 and 35 the theoretical graphs of the steel stresses calculated on the experimental values obtained in

(a) Ordinary In Figs.

in accordance with 4.5 are superimposed accordance with 4.6. M=Asfs

The equation employed is:

(d1 - x) 9 (4.18) It can be seen that full agreement Is obtained between the theoretical and experimental values. The variation of the steel stress in relation by a straight line. results to the applied

moment can almost be represented steel stresses based on experimental

The values of the working

are given in col. 3 of table (7).

These values are compared with the stresses in the composite beams in section c. (b) Composite Beams In Figs. 36 and 37 the theoretical graphs for the stresses calculated in accordance values obtained in accordance with 4.6.

with 4.5 are superimposed

on the experimental

The equations employed in calculating M_ M= AS fs (dl As fs (dl - -

the steel stresses are: x)


3 (4.20)

x) + Ta (d2 - X)

36 and 37.):

(4.18)

The various as follows

stages at which the stresses were calculated in each beam are (Reference can be made to Figs.

(1) At the cracking

moment of the concrete where the contribution (equation 4.20)

of the

f. r. c. channel is considered

(2) At a level of moment just before the cracking channel, where the contribution (equation 4.20).

moment of the f. r. c.

of the f. r. c. channel is considered

(3) At a level of moment just after the cracking moment of the f. r. c. channel, where the contribution of the f. r. c. channel is not considered (equation 4.18)

(4) At a level greater than the cracking


0.8 times channel the ultimate moment),

moment of the f. r. c. channel (at about


the contribution of the f. r. c.

where

is not considered

(equation

4.18).

From Figs. the theoretical (c) Comparison

36 and 37 it can be seen that good agreement is obtained between values of the steel stresses. and Composite Beams moments for both Col. 4 of table (7)

and experimental

Between Ordinary

The values of the steel stresses at the level of the working ordinary table 3 (7). in beams of col. given and composite are

shows the reduction in the steel stress due to the contribution

of the f. r. c. channels

in the composite beams ranging between 9% and 16.7% of the stresses in the corresponding ordinary beams. The amount of reduction Beam ST 10 -C depends mainly on the type with 0.582% of No reduction

and percentage of the steel employed.

reinforced

Kam 60 steel showed the highest level of reduction in the steel stress. beam ST 4 -C. in the stress of obtained working steel was
f. r. c. channel being The moments stress moment cracked at which at a stage earlier the composite

This was due to the


of the working moment. steel

than the level attained

beams

the same working the working

for the ordinary of the beams.

beams varied

between

1.12 and 1.2 times

The reduction at levels greater

in the steel stress in the composite beams became even more At the levels just before the cracking

than the working moment.

beams in the the f. were the stresses composite steel r. c. channels moments of between 10.7% and 18.4% lower than those in the corresponding ordinary beams. The values of the steel stresses and 6 of table (7). and the amounts of reduction are given in cols. 5

This reduction in the steel stress was a major factor in reducing of the composite beams.

the width of the cracks and deflection


7.7 7.7.1 Limit Limit States of Desi 'n Strength

State of Ultimate

The mechanism of failure

for the test beams was initially

started by yielding

in the the by followed the compression concrete the of crushing steel, which was of to Beams being beams due the near to This under-reinforced. was mainly zone. the failurestage tested typical 3a 2. In be in plate plate seen can failure for the beams

is shown. by beams greater a for observed the the was various steel of

The yielding

rate of increase in crack widths and deflections. At loads near to the failure 100mm (giving span to deflection of impending failure In the following deflections 2mm with stage crack widths of ratio of 45) were observed, of

thus a good warning

2). (sec plate obtained was sections the observed and calculated values of the ultimate

moment for the test beams are discussed.


(a) Ordinary Beams

The ultimate approaches,

strength of the members was calculated by employing two one was in accordance with the recommendations of CP110

the first

employing partial

safety factors for materials

and using nominal design strengths Values of the ultimate strengths

for concrete and steel.

This is discussed is 3.2.1.

for the test beams employing this approach are given in col. 7 of table (7). The second approach was based on the strain compatibility actual values of strength for the materials described in 5.2.1. without the partial method, using the safety factors as

The stress strain relationship

for the reinforcement

shown in

Fig.

1 and for the concrete shown in Fig.

Ga were used.

The concrete cube strengths

used were those obtained at the time of testing the beams as given in table (6). The values of the ultimate strength of the various beams employing this approach

are given in col. 8 of table (7). A comparison between the observed and calculated values of the ultimate moments cols. 7,8 and 9 in table (7), shows that the approach based on the actual gives a better agreement than that based on

values of strengths described in 5.2.1. CP110 recommendations. strengths

It can also be seen that the approach based on the actual underestimates the observed values of ultimate

described in 5.2.1.

strength for all the beams except ST 3-0,, where the values are nearly equal.
The ratios the actual between observed values of the observed of strength values of ultimate moments to those calculated 10, table using

for the ordinary ranging moments

beams

(given in col.

(7)) varied of the It has

0.98 and 1.39. to calculated

Values ultimate

between

1.05 and 1.25 for the ratio earlier (90).

have been reported strength

been suggested difference

(90) that this variation the values

in the ultimate steel

could be due to the and those used

between

of the actual

stresses

at failure

in the calculation. It is of interest exhibit an ultimate conventional to mention here that steel bars embedded in concrete can

strength or yield stress greater than that obtained from a

tensile test (91). the increase in the ultimate strength is difficult to predict and safety

However, therefore factor.


(b) Composite

can not be used in design but should be considered as an additional

Beams

The analysis employed for the composite beams was similar ordinary

to that for the of

beams explained in section (a), that is, by neglecting the contribution load conditions.

the f. r. c. channels at the ultimate Calculated values of ultimate values of strength with the partial

strength employing the CP110 method (nominal safety factors) are given in col. 7 of table (7). underestimates the observed values

This shows that the CP110 method considerably of the ultimate strength.

The strain compatibility values of strengths the observed values. 9 of table (7).

method described in 5.2.1. safety factors)

(based on the actual

without partial

gave a better agreement with 8 and

The observed and calculated values are given in cols.

It can be seen that the observed values are greater than the calculated ones for six out of the eight beams tested. of the ultimate The ratios of the observed to calculated values

moments for the composite beams ranged between 0.95 and 1.38. in section (a) with regard to the increase in beams are also applicable to the composite

The arguments discussed earlier the ultimate beams. Considering strengths of the ordinary

the observed values of ultimate

moments,

a factor of safety of

more than 2 was ensured for the worldng moment against collapse. This may suggest that an increase in the working moment may be utilised provided that the limit states of cracking
(c) Comparison

and deflection
Between Ordinary

are satisfied.
and Composite Beams

The observed values of the ultimate differ significantly

beams did the not composite strengths of ordinary beams. This can be

from those of the corresponding

seen in cols. However,

9 of table (7). an increase in the ultimate strength was obtained for beams reinforced with 0.582% of

with a low percentage of steel, Kam 60 had an ultimate

e. g. beam ST 10 -C reinforced

strength equal to 17% greater than that of the corresponding

beam ST 10 - 0. This may suggest that the f. r. c. channels makes a greater contribution conclusion, to beams reinforced however, with a low percentage of steel. A definite amount

in this respect can not be drawn

due to the limited

of test results.
7.7.2 7.7.2.1. (a) Limit State of Deflection of Ordinary Beams

Behaviour

General

In theory the load deflection has two initial

curve of an ordinary These represent

beam, when first an untracked

loaded,

phases of behaviour.

and a cracked

section.

The idealised relationship 16.

for ordinary

beams is represented

by the dotted line shown in Fig.

The observed load deflection in Figs. 38,39 and 40.

curves for the test beams are shown rigidity of an untracked

In these the flexural

tangent by the is indicated (Eclo) section figures.

01 shown on the same

When the tensile strenth of the concrete is reached and the first crack appears, the load deflection With increasing is achieved. gradient load, further linearity. from deviate curve will

cracks form until a stable crack pattern change occurs in the

When this point is reached very little

of the load deflection

curve until the yield point of the steel rigidity flexural of a cracked section rigidity) is indicated

is reached. (kEcIc;

At this stage the flexural

k is a constant for an effective

by the tangent

0 2. To be consistent these tangents are drawn at to the working moments as shown in the figures. however, that all the beams cracked long before the

levels corresponding It may be noted,

level of the working moment. Examination of Figs. rigidity 38,39 difference the that 40 shows and section ( 01) and that

between the flexural

of an untracked

beams for is reinforced (0 pronounced more of a cracked section 2) ST 7-0 for Beams low g, percentage of steel; e. with a reinforced and ST 8-0 reinforced

beam ST 9-0 1.643%o and of ratio with a steel the ratios O1/ 02

with a steel ratio of 1.483% and 1.08 respectively,

were 1.17,1.25

for beams containing 1.044% of steel whereas

the ratio varied between 1.48 and 2.02. In general, the flexural this behaviour is mainly due to the difference of cracked (kEclc) and untracked between

rigidities

(EcIu) sections of steel.

being lower for members reinforced To illustrate this point the relative

with high percentages ratios EcIu kEclc

for beams ST 10 -0

(0.584%),

ST 1-0

(1.044%),

ST 9-0

(1.483%)

and ST 7-0 ratio

(1.643%) of 12

are 3.28,2.27,1.9 was employed

and 1.8 respectively.

A modular

in these calculations.

Other factors which might have also affected this behaviour beams. in the took that the place amount of shrinkage was flexural the to tend form due to reduce shrinkage which of concrete members. It can also be seen that the range of an uncracked section in the load deflection curves is small. This indicated an early formation shrinkage and the lower of cracks Cracks

rigidities

which could be due to the effect of greater tensile strength of lightweight


As mentioned of deflection approaching in 7.7.1

concrete.
the beams showed a significant giving amount of

at the final failure.

stages of loading, 2a an ordinary

a good warning

In plate

beam near the failure

stage is shown.

(b) Working Moment Condition The observed deflections for the test the moments worldng at

beams (given in col. 2 of table (9)) varied between 7.6mm and 27.2mm9 which gave a corresponding span to deflection ratios of 592 and 165 respectively.

The beams which gave a span to deflection ratio lower than 250 were
ST 5-0, table highest (9). ST9 -0 and ST 3-0. The values are given in col. 7 of the gave a Beams

Beam ST 5-0

was reinforced

with Kam 90 steel, investigation, value.

grade of steel employed ratio

in the present

span to deflection ST 9-0

of 165, which was the lowest reinforced with

and ST 5-0

1.483% and 1.044% of Unisteel It 550

550 gave a span to deflection can therefore be assumed

ratios

of 212 and 245 respectively. percentage of Unisteel

that if a lower

is employed,

a span to deflection

ratio greater

than 250 can be expected.

For the sake of comparison, ST 10 -0 reinforced with ratio

a reference

is made to beam which gave a Similarly

0.582% of Kam 60 steel, load.

span to deflection beam ST 4-0 ratio greater

of 317 under working Kam 60 steel

containing

gave a span to deflection reinforced ratio with a similar for beams

than that of beam ST 3-0

amount

of Unisteel

550 . The span to deflection

ST 4-0 flexural

and ST 3-0 rigidity

were 256 and 245 respectively.

The improved

for beam ST 4-0

could mainly be due to the surface

deformation

of Kam 60 steel having a better control on the crack widths.


investigation with (35) beams with a span to depth ratio nominal yield strength and

In a previous of 7.6. reinforced percentages ratio ranging

steel of 410 N/mm2 between

of steel varying between

1 and 4.2 gave a span to deflection Another investigation (4g employing

1010 and 461.

a greater between between varied

span to depth ratio

of 14.6 with percentages

of steel varying ratio

1.14 and 3.16 of Unisteel 291 and 434. between The concrete

410 gave a span to deflection cube strength

for both investigations

44 N/mm2

and 59.6 N/mm2.

A direct comparison between the values obtained investigations and those obtained in this research

in previous

can not be made

mainly due to the difference Other factors,

in the span to depth ratio employed. concrete was ratios.

e. g.., the type and amount of steel employed,

deflection the level the the stress at which steel of strength and deflection the to the span of values could also measured affect For each beam tested in the present investigation at the working first the the at same nearly moment was Similarly the deflection

the deflection and second

cycles of loading.

at 1.5 times the working

moment was nearly equal at the second and third cycles of loading.

(c) Relationship

With Steel Stress between the experimental as shown in Fig. 41

For various percentages of steel a relationship steel stresses and the observed deflection

can be established

in which three percentages of steel were considered.

From this figure it can be

seen that the deflection of a span/250 was reached at steel stresses of 400 N/mm2, 312 N/mm2 and 280 N/mm2 corresponding to steel percentages at 0.582 (Kam 60),

1.044 (Unisteel

550), and 1.643 (Unisteel 410) respectively.

(d) Remaining Deflection The remaining col. 3 of table (9). deflections for the test beams after the first in the amount of deflection at the working moment.
that recoveries of about 80% for lightweight concrete (92) of the

cycle are given in

The recovery

ranged between 74%

and 87% of the respective


Previous concrete respective investigations

deflection
indicated

(40) and between working

62% and 83% for normal at the first

weight

deflection

cycle were obtained.

7.7.2.2. S)

Behaviour of Composite Beams

General The load deflection behaviour for the composite beams based on theoretical

considerations established,

is discussed in 5.3.4., as shown in Fig. 1Gb.

consequently an idealised

relationship

is

The observed load deflection and 43. behaviour

for beams test are shown in Figs. curves

42

In these figures the beams show a steeper gradient for the load-deflection at the initial stages of loading. When the concrete cracks the gradient a

of the load deflection behaviour reduction in the flexural rigidity

changes in a gradual manner; of the members.

this indicates

The curvilinear

behaviour between the levels of the cracking

moments of the

concrete (Mc) and the cracking

moments of the f. r. c. channels (Ma) is mainly due This

to the control of the f. r. c. channels on the width and extension of the cracks. is discussed in 5.3.4. When the f. r. c. channels cracked some scatter in the load-deflection

behaviour

was observed, a noticeable

e. g., beams ST 1-C,

ST 7-C, rigidity.

ST 10 -C

and ST 12 -C

showed

reduction in their flexural

The factors which might have moment of

affected the load deflection behaviour

at levels greater than the cracking

the f. r. c. channels are discussed in 5.3.4. For beams with a similar percentage of steel the load deflection behaviour was moment of the f. r. c. channels. differed depending upon the effect

level the to the of the cracking same up nearly Thereafter the behaviour of the beams slightly

that the formation individual beams.

behaviour had the the f. the in of on r. c. channels of cracks

An example deflectionthose

of this is that due to the cracking containing mild

of the f. r. c. channel, at a greater with Unisteel

the rate than 410 and

of beam ST 1-C ST 2-C

steel increased

of beams

and ST 3-C This

which were reinforced

Unisteel deflection

550 respectively. curves

behaviour at levels

can be seen by comparing than the cracking the deflection

the load of the

of the beams

greater

moments

f. r. c. channel increased

(Ma) as shown in Fig.

42, where

of beam ST 1-C This could had

at a greater

beam ST 2-C those than of rate that the formation of cracks

and ST 3-C.

be due to the greater

effect

in the f. r. c. channels

1-C. ST beam in the the of concrete cracks on

Reference can be made to Figs. was due to

51

beam ST 1-C the in of 52, the concrete width where maximum crack and increased considerably 2-C beams ST in those than greater and ST 3-C

the formation Similar significant

of a crack in the f. r. c. channel. to the behaviour of ordinary beams, the composite beams showed a which gave a good stage can be seen

amount of deflection

at the final stages of loading,

warning of impending failure. in plate 2b.


(b) Working Moment Condition

A composite beam near the failure

The observed deflections between 4.2mm and 18.4mm,

at the working moments for the test beams varied which gave a corresponding span to deflection ratios

ranging between 1071 and 245. The values for the various beams are given in cols. 2 and 7 of table (9). for beam ST 4-C The lowest value for the span to : deflection reinforced ratio obtained was

with Kam 60 steel, where the f. r. c. channel cracked

at a stage before the working moment of the beam. Considering the recommended span to deflection ratio of 250 the composite at the level

beams except beam ST 4-C of the working moment.

showed an adequate degree of stiffness

For each beam tested the deflection

at working moment in the second loading loading cycle. Similarly the

first the in that the obtained same as cycle was nearly deflections same.

the third the in the 1.5 times second and cycle were moment working at moment of the f. r. c. channel was

For both cases, when the cracking

reached some scatter in the results was observed. (c) Remaining Deflection The remaining col. 3 of table (9). deflections for test beams after the first recovery first the after cycle are given in

The deflection

cycle ranged between

77.4% and 86% of the respective 7.7.2.3

deflection

at the working moment. of Ordinary and Composite Beams to that of an for both

Comparison Between the Behaviour

The load deflection behaviour of a composite beam in relation ordinary beam is explained in 5.3.4. Typical

and idealised relationships

types of beam are shoven in Fig. 16. In Figs. ordinary made. ordinary 44,45 deflections the of 46, values observed plotted arc and for the

direct beams, that comparison so a and composite In these the composite beams show a greater

of behaviour can be

degree of stiffness than the

beams up to and well above the working moment. the level deflection in at obtained of the working was saving This could be due to the ordinary beams being completely controlled by

The greatest

moment of the beams. cracked,

beams the in the are effectively composite cracks whereas action of the f. r. c. channels.

the restraint

The ratios of the deflections

for the composite beams to those of the ordinary The

beams at the level of the working moment, varied between 0.96 and 1.54. lowest value was obtained for beam ST 4-C

containing Kam 60 steel, where the f. r. c. moment of the beam. It can

channel cracked at a value lower than the working

also be seen that when a higher percentage of steel was employed the saving in

the amount of deflection was reduced, e. g., beam ST 7-C 1.643% gave the least amount of saving.
The moments deflection at which the composite ordinary beams attained

reinforced

with

the same working

for the corresponding the working

beams were

at 1.2,1.17,1.15,1.22 ST 2-C, ST 3-C, when an

and 1.03 times ST 10 -C increase

moment

for beams ST 1-C, These values

and ST 7-C in the working

respectively. moment

are important

of a composite

beam is to be considered.

The saving in deflections level of the cracking rigidity

for the composite beams was maintained up to the Thereafter the flexural

moment of the f. r. c. channels. decreased.

of the composite beams gradually

When more cracks were formed rigidity increased and

in the f. r. c. channel finally flexural

the rate of reduction in the flexural

a stage was reached where the ordinary rigidity.

and composite beam had the same

This stage was reached at steel stresses of 230 N/mm2 and

220 N/mm2 for beams containing mild steel with steel ratios of 1.044% and 1.643%. beam bars deformed (except beams that For the were reinforced with respectively. C) this stage was reached at steel stresses with the corresponding ratio and 550) Uni N/mm2 (1.044% 460 410), steel Uni 405 N/mm2 (1.044% type of steel of steel ST 4fnm2 560 N, (0.582% Kam 60). and At the levels well above the cracking the composite beams showed a slightly ordinary in Figs. beams, e. g., beams ST 1-C, 44,45 and 46. This was mainly f. the r. c. channels of moments rigidity some of

lower flexural ST 4-C,

than the corresponding and ST 10 -C shown

ST 7-C

due to. the formation

of cracks in the

f. r. c. channels and the sudden transfer

from the f. r. c. channels tensile stresses of of the members.

to the steel, which consequently increased the curvature In general,

the saving in deflection when employing the same type of f. r. c. factors: member; whether the concrete of the member is

channel depends upon the following (1) condition of the flexural cracked or not.

(2) the percentage of steel employed. (3) the level of the cracking (4) the level of steel stress. moment of the f. r. c. channels,

7.7.2.4.

Comparison With Theory

The central deflection for the test beams was calculated by employing the approaches parameters suggested in 5.3.5. In these calculations fr =2 the values of the various Ec = 17 kN/mm2, frc =3 N/mm2 and

employed were as follows; Ea = 12 kN/mm2,

N/mm2,

Es = 200 kN/mm2,

m2 = 0.7,

Ma = as given in col. 3 of table (8). The approaches employed are as follows: Approach One : Existing Theories
The calculated obtained by employing against theories deflections

(Ordinary

Beams)
moments of the ordinary in appendix beams (F)

at the working existing values

the various

theories

reviewed 47.

are plotted the existing accuracy

the observed can predict

as shown in Fig.

It can be seen that to within an

the actual

deflection of amount

of + 20%.

Approach Two : Empirical

Method (Ordinary

and Composite Beams)

In this approach the deflection was calculated by employing the empirical method discussed in 5.3.5. the experimental exp = In this calculation strains the curvature was obtained from

values of flexural ecexp XeXP

following: in the the axis as neutral and (5.4)

Where ecexp : experimental Xexp : experimental

maximum flexural neutral axis depth

compressive

strain in concrete

The values for 0cexp and Xexp used were those obtained from the experimental results. The calculated are shown in Figs. and observed load deflection 38,39 and 40. curves for the ordinary beams

Good agreement is obtained between the calculated and the observed values. Similar calculations were carried out for the composite beams. and a good In col. 4 of table (9) The ratios

agreement has also been obtained with the observed values. the predicted

values at the level of the working moment are shown.

deflections to the of the values observed calculated of

for the test beams varied The good agreement as discussed in 5.3.5

between 0.96 and 1.15 with an average value equal to 1.06. deflections the the between observed calculated and obtained indicates the following points:

(1) The measured strain distribution the average values.

across the depth of the beam represents

(2) The readings taken with regard to strain measurements, the determination were sufficiently

deflection

and

beams the the depth the of central point at neutral axis of accurate. of a flexural from be the calculated can member across the depth of the beam and could be due

(3) The average curvature

observed values of the flexural the neutral axis depth. to the formation
Approach Three : Proposed

strains

The occasional scatter in the results

length. "Demec" the gauge of cracks within


Methods

Ordinary

Beams

in beams the accordance calculated was The deflection at the worldng mcments of following: from the with 5.3.5, where the average curvature was obtained fs - k3 fr/P Es(d1 -x) 0 (5.11)

ave

From the experimental k3 was 0.3.

for found that it suitable a value was values obtained, (f5) employed for the beams were calculated in 34 and 35. The value of the neutral section by employing

The steel stresses

accordance with 4.5.2

Figs. in shown are and

transformed for depth that a cracked calculated used was axis equation 4.5. The theoretical deflections for the and observed values

at the level of working

(9). table 5 in 2 of and cols. moments are given the observed values of deflections

The ratios of the calculated to These varied indicates the

are given in col. 6 of table (9). This clearly

between 0.94 and 1.06 with an average value of 1.01. reliability in the predicting method proposed of concrete beams.

the actual amount of deflection

for lightweight

Composite _(b)

Beams

The central deflections

for the composite beams were calculated throughout moment of the f. r. c. channels The idealised relationship

the loading stages and up to the level of the cracking by employing the proposed method suggested in 5.3.5. is shown in Fig. 1Gb. In this method the curvature three stages as in the following: (1) For
_

of a beam can be calculated by

mC
Al EcIo

Me,

cross section No. 1 in Fig.

17. 17.
(5.17)

(5.16)

(2) For
_

Mc-Z-- R'I7: 5: 1LIp, cross section No. 2 in Fig. M+ EcIo bi k LcIn Ma + Mc Mc

where

mp P2

(3) For Alp :cM

Ma,

Fig. 3 in No. section cross Al -lip k Ec Tcc

17.

mc
ECIo

+MP

MC -

(5.19)

k EeIp

Ip

H be in can shown appendix calculated as : From an examination of the experimental results obtained it was found

suitable to use values of 0.75 and 0.7 for k and k respectively. In Figs. 42 and 43 are plotted the theoretical values for the deflections obtained

by employing this method together with the observed values. been obtained between the experimental seen that the experimental represented by the trilinear and the theoretical

A good agreement has values. It can also be

load deflection behaviour

of the composite beams is well

concept suggested for this method. at the level of the working moments are given in

The calculated deflections

col. 5 of table (9). The ratios of the calculated to observed values of the deflections, as given in col. 6 of table (9), varied between 0.98 and 1.05 with an average value of 1.02. It can be observed that the proposed method is quite reliable actual amount of deflection for composite beams. in predicting the

7.7.3 7.7.3.1

Limit

State of Cracking of Ordinary Beams

Behaviour

(a) General

The mechanism of cracks formation limitations put on theiAvidth

in reinforced

concrete members and the In general the surface crack

are discussed in 5.4.1.

width under normal conditions of exposure should not exceed 0.3mm. In this research the width of the cracks at the level of the steel reinforcement as well as at the bottom edge of the beams was measured. Some of the beams showed surface cracking after a period of three to four weeks after casting. occurred This could mainly be due to the shrinkage strain which The maximum values of shrinkage strain

in the concrete members.

obtained after four weeks from casting were 0.0005 in the small concrete prisms (100mm x 100mm x 500mm). The maximum crack width observed was 0.08mm In a previous investigation a maximum crack concrete members (90). to

with an average value of 0.04mm.

for lightweight due 0.05mm to shrinkage of observed was width The early formation

difficult beams it the test in made of some of cracks cracking occurs.

observe the load at which flexural

The values of the direct tensile strength given in table (6) ranged between 0.94 N/mm2 and 3.5 N/mm2 with an average value of 2.02 N/mm2. The values

of the modulus of rupture varied from 1.6 N/mm2 to 4.2 N/mm2 with an average value of 2.8 N/mm2. The average ratio of the direct tensile strength to the

modulus of rupture obtained was equal to 0.7. The low values obtained for the tensile strength and the modulus of rupture for the concrete could mainly be due to the formation concrete prisms. In the calculation of the cracking moment of the concrete the average value of the This gave a good agreement between
moments. These are given in

of shrinkage cracks in the

direct tensile strength (2 N/mm2) was used.


the calculated cols. and the observed (10). values

of cracking

2 and 3 of table

(b)

Worldng

Moment

Condition

The observed crack widths at the steel level as well as at the bottom edge of the beams at the working moment level are given in table (10). The maximum

crack widths varied between 0.13mm and 0.2mm at the level of the steel, and between 0.15mm and 0.3mm at the bottom edge of the beams. The average crack

widths ranged between 0.05mm and 0.14mm at the steel level and between 0.09mm and 0.22mm at the bottom edge of the beams. This indicated that the crack widths It can therefore

at the level of the working moment did not exceed a value of 0.3mm. be assumed that the limit the beams tested.

state of cracking under static loading was satisfied for

The ratios of maximum to average crack widths at the level of the steel for the test beams varied between 1.1 and 2.4 with an average value pf 1.68; the ratios for the crack widths at the bottom edge of the beams varied between 1.22 and 1.67 with an average value of 1.52. The cracks observed in the region of the constant moment zone gave average spacings of 129mm and 145mm at' he bottom edge of the beams and at the level of the steel respectively. With regard to the effect of surface deformation reinforced ST 3-0 with Kam 60 steel showed smaller reinforced with Unisteel of the bars, beam ST 4-0

crack widths compared with beam

550. This could be due to the surface deformation The values of the

of Kam 60 steel having a better control on the crack widths. crack widths are given in table (10).
The crack not differ from widths at the level of the working moment

at the first Similarly

cycle the

did

that at the second cycle by more thin 0.02mm.

crack widths at 1.5 times the working moment were nearly equal at the second and third cycles of loading.

(c) Steel Stress The variation of the maximum crack width at the steel level in relation 48,49, and 50. These figures clearly to the

steel stress can be seen in Figs.

show

the effect that the steel stress has on the width of the cracks. steel stresses were calculated in accordance with 4.6,

The values of the 34 and 35.

and shown in Figs.

The observed maximum crack widths at the bottom edge of the beams with the corresponding steel stresses in the test beams were as follows:

Beam mark

p%o

Type of steel

Steel stress (N/mm2) at a crack width of


0.2mm 0.3mm

ST 1-0
ST2-0, ST3-0,

1.044 ST5-0 1.044 1.643 1.643


1.583

M. S.

235

305

ST4-0, ST6-0 ST 7-0 ST8-0


ST9-0

Deformed bars M. S. Deformed bars


Deformed bars

240 -

325

360 - 440 296 384

160 230
220

ST10-0

0.582

Deformed bars

240

472

From the results

it can be seen that the deformed bars showed better control

on the width of the cracks than did the mild steel bars. In a previous investigation (35) an average maximum crack widths of 0.2mm

bars twisted 0.18mm were observed at a steel mild steel with and with square stress of 207 N/mm2. 0.99% and 4.62%. research The percentages of steel employed ranged between however, with the values obtained in this in type and percentage of steel

A direct comparison,

to the variation can not be made mainly oavmg

and the type of concrete used. reinforcement


(c) Remaining

The amount of concrete cover to the main tension

is archer factor which should be considered.


Crack Width

The maximum remaining

crack widths in the test beams after the first

cycle of loading varied between 0.05mm and 0.09mm at the level of the steel, and between 0.06mm and 0.1mm at the bottom edge of the beams.
Valued were reported ranging between 0.02mm and 0.05mm (35). for the remaining crack widths

in a previous

investigation

7.7.3.2

Behaviour

of Composite

Beams

(a) General The mechanism of cracks formation of the f. r. c. channels in arresting in the composite beams and the role

these cracks are discussed in 5.4.2. observed at the interface

The cracks in the composite beams were initially (top of the upstands of the f. r. c. channels).

Cracks at the soffits of the beams

were only observed after the f. r. c. channels cracked. As mentioned in 7.7.3.1., the average values of the tensile strength and the For the

modulus of rupture were 2.02 N/mm2 and 2.84 N/mm2 respectively. calculation of the cracking

moment of the concrete by employing equation 4.2, a

value of 3 N/mm2 for the tensile strength of the concrete at the level of the interface was used. This gave a good agreement between the calculated and the observed values moment of the concrete as can be seen in cols. 2 and 3 of table (10).

of the cracking

With increasing

load (greater than the cracking load of the concrete) further length in increase and of the width and an

cracks were formed in the concrete, cracks were also observed. cracked.

A stage, then, was reached where the f. r. c. channel formation by the occurred of

The cracking of the f. r. c. channel normally

one crack at a position where there was a crack already formed in the concrete. This could be due to the tensile stress in the f. r. c. channel being maximum in the vicinity minimum of an existing crack in the concrete (at a cracked concrete section) and at a point between two adjacent crack (at an uncracked concrete section).

Furthermore

the cracks which might have formed in the confined concrete

by the presence of the f. r. c. channels could have induced additional local tensile strains in the f. r. c. channels. This might have contributed to the development of

cracks in the f. r. c. channels at positions of existing cracks in the confined concrete. Another factor is that the neutral axis depth at a cracked concrete section is less than that at an untracked distribution, a greater concrete section. Considering a straight line strain

tensile strain in the f. r. c. channels would be expected at

the cracked section. The maximum first surface crack width measured on the f. r. c. channels when cracked ranged between 0.1mm and 0.4mm depending on the level of the moment of the f. r. c. channels. The corresponding cracks in the confined

cracking

concrete,

however,

might be smaller.

This is discussed in section (d).

At higher levels of loads more cracks were formed in the f. r. c. channel. This indicated of the first carrying was still that the composite action was not completely lost by the formation

f. that the f. the in r. c. channel was still also r. c. channel and crack stresses at the untracked sections. It also indicated that there

tensile

the f. between bond the and concrete which enabled channels r. c. a good

the f. r. c. channels to carry the tensile stresses. The cracking behaviour of the beams can be seen in Figs 51,52,53 interface the the at widths crack observed maximum where plotted against the applied moment. The figures and 54,

(level of the steel) are

show that the width of the cracks

the the level beyond the are well within to working moments of and well up f. the the limits. However, of r. c. channels moments cracking when recommended increased. the the suddenly cracks of width were reached transfer be due the to sudden crack widths could mainly This increase in the of the tensile stresses from

the in the tensile steel, increased strain f. to the the which steel, c. channels r. resulting in a greater crack width.

(b) Working Moment Conditions for the test the level the moment The observed crack widths at working of beams are given in table (10). soffits, The beams did not show any visible cracks at their

the exception being beam ST 4-C.

This was mainly because the f. r. c. values of working where the

their than levels did to respective greater not crack up channels moments, the exception being beam ST 4-C

containing Kam 60 steel,

f. r. c. channel cracked at a level lower than the working moment. The maximum crack width at the level of the steel (measured at the interface) ranged from 0 to 0.1mm, 0.08mm. between 0 and the widths crack varied average whilst

The average ratio of the maximum to the average crack widths for the

various beams was 1.5. It may be concluded, therefore, reduction that the composite beams skived a considerable

in the amount of cracking in concrete at the levels of the working moments.

(c) Remaining Crack Width The remaining maximum crack width at the steel level for the test beams

after the first

cycle of loading are given in col. 10 of table (10). Only beam ST 4-C

These values

varied between 0 and 0.04mm.

had a maximum residual

crack width of 0.1mm at the bottom edge of the beam.


(d) Cracks in the Confined Concrete

To study the formation

of the cracks in the confined concrete

parts of the and

f. r. c. channel of a composite beam which had been subjected to the first second cycles of loading

were cut and removed while the beam was unloaded.

The removed parts were 50mm to 100mm on either side of existing cracks in the concrete (observed at the interface) and also in the f. r. c. channel.

One of the interesting

points observed at zero load was that at positions the crack widths in the confined concrete After reloading A

where cracks had formed in the channel were smaller the beam

than those measured on the surface of the channel.

similar

behaviour was observed at the level of the working moment.

f. had the in the r. c. channel crack confined concrete observed at a position where been cracked working level the the be in, 4b. This of at observed was can seen plate

moment of the beam. there in the that were no cracks areas where point was This was

Another interesting

in the f. r. c. channel the confined concrete did not show vis: ble cracks. observed at zero load as well as at the level of the working moment. at the level of the working moment can be seen in plate 4a.
7.7.3.3 Comparison Between Ordinary and Composite Beams

This behaviour

For a comparison between the cracking behaviour beams, the following points were observed.

of the ordinary

and composite

(1) Shrinkage cracks did not appear on the composite beams, whereas a maximum crack width of 0.08mm was observed due to shrinkage in the ordinary beams. (2) An increase in the apparent cracking were observed at the interface moment of concrete at which cracks The ratios of the cracking

was obtained.

moments of concrete for the composite beams to those for corresponding ordinary beams varied between 2.2 and 4. This clearly indicated the contribution of the f. r. c. channel in controlling flexural cracks in the

composite beams.

(3) At the working moment conditions the widths of the cracks in the composite beams were much smaller ordinary beams Figs. 51,52,53 than those in the corresponding

and 54. The ratios of the maximum beams to those in

crack widths at the level of the steel in the ordinary the composite beams, except beam ST 4-C,

varied between 2 and 6.

At the soffits of the beams no cracks were observed in the composite beams, whereas the maximum crack widths in the corresponding ordinary

beams ranged between 0.15mm and 0.26mm. (4) When the f. r. c. channel cracked, slightly ordinary greater some of the composite beams showed than those in the corresponding and ST 7-C (Figs. 51,

crack widths at the interface

beams, e. g. beams ST 1-C, The maximum

ST "4 -C

52 and 53).

surface crack widths measured on the surface

of the f. r. c. channels at the soffits of the composite beams were greater than those in the corresponding (5) The number of visible in the ordinary beams. ordinary beams.

less that beams than the in was composite cracks This can be seen in col. 6 of table (10), where beams in the composite at cracks
1.

in some beams the number of visible

the level of the working moment was nearly half of that in the ordinary beams. The average spacings of cracks were 250mm and 145mm for the composite and the ordinary beams respectively. due to the between the

(6) The length of the cracks in the composite beams was limited restraint f. the r. c. channels. of action A direct comparison

length of the cracks in ordinary Fig.


7.7.3.4 Ordinary

and composite beams can be seen in

55.
With Theory

Comparison Beams

As mentioned in 5.4.3.,

the crack widths for the ordinary

beams were

calculated by employing the proposed method in CP110 and also by employing the C. and C. A. formula. The C P110 formula is: 3 acr em
w=

1+2

(acr - C) d-x 1.2 bd ( - x) x 1U-3 AS (d - x) fy

(5.21)

Wherem e

= el

(5.22)

In this formula the experimental

the neutral axis depth and the steel stresses obtained from

results were used. the maximum crack widths at the steel level and The calculated

The method underestimated

at the bottom edge of the beams for seven out of the ten beams tested. values of the crack widths are given in cols. given in cols. 4 and 7 of table (10). Calculations were also carried out using the C and C. A. formula

12 and 13 and the observed values are

(88).

The

equation for the maximum crack width at the bottom edge of the beam is
W=3.3c fs Es (d - x) dl -x (5.23)

c: cover the main reinforcement The term d- x/dl The calculated -x reduces to 1.0 when crack width at the steel is sought.

values obtained by employing this method for the maximum

crack width at the steel level together with the observed values for the test beams are plotted against the applied moment as shown in Figs. figures values. 48,49 and 50. These

show that a good agreement is obtained between the calculated and the observed The calculated values of the crack widths at the steel level and at the 14

bottom edge of the beams at the level of the working moment are given in cols. and 15 of table (10). The ratios of the calculated to observed values of the maximum

crack widths

at the steel level ranged between 0.7 and 1.1 with an average value of 0.98, the ratios at the bottom edge of the beams ranged between 0.9 and 1.8 with an average

value of 1.3.

CHAPTER EFFECTS AND OF LONG TERM BEAMS LOADING

EIGHT ON TEST BEHAVIOUR FOR ORDINARY

COMPOSITE

8.1

Introduction

This chapter discusses the effect that long-term of the ordinary and the composite beams.

loading had on the behaviour

A total of nine beams were tested, five beams under fatigue and four beams under sustained loading tests. The properties of the various beams tested are

given in tables (2) and (3). The methods of testing employed are explained in Chapter Six. loading test. limit Beams tested under fatigue loading were initially Thereafter tested under static

they were subjected to a cyclic loading with an upper equal to half the

equal to the working moment of the beam and a lower limit moment. The number of repetitions cycles. loading initially were -

working

applied to each of the test beams

was about three million

Beams tested under sustained loading up to the working moment.

tested under static

The beams then were kept under sustained The discussion of the test behaviour compressive More

loading for a minimum period of 500 days.

includes the effect of long term loading on the neutral axis depth, flexural strain in the concrete and the stresses in the steel and the f. r. c. channels. emphasis is placed on the limit states of ultimate strength, cracking

and deflection. and the

A direct comparison between the flexural composite beams is also presented.


8.2 8.2.1 Fatigue Loading Tests Axis Depth

behaviour of the ordinary

Variations

of the Neutral

(a) Ordinary

Beams

There are two major factors which can affect the position of the neutral axis depth of flexural members subjected to fatigue loading. These are: (1) An increase in width and height of travel of the cracks due to repeated loading can result in raising face. the neutral axis level towards the compression

(2) The increase in the flexural

compressive

strain in concrete due to axis depth.

repeated loading can lower the position of the neutral However, appreciable by referring to Fig.

56, it can be seen that cyclic loading had no

effect on the level of the neutral axis depth for the beams tested in At the levels of the working moments the ratios of the neutral cycles to those obtained at the first (1.044% mild steel),

this investigation.

axis depths measured after three million cycle were FA 2-0 0.98,1.05

and 1.17 for beams FA 1-0

(1.044 Unisteel 410) and FA 4-0

(1.044% Kam 60) respectively.

It can be seen that the higher the level of the working moment the greater is the ratio of the final to initial that the initial moments. increase neutral axis depth. This was mainly due to the fact at low levels of working

values of the neutral axis depth were greater

Additionally,

with higher levels of repeated working load a greater compressive strain in the concrete. This

was observed in the flexural

consequently increased the neutral axis depth. the flexural compressive

The effect of repeated loading on

strain in the concrete is explained in 8.2.2.

(b) Composite Beams In Fig. 56 are plotted the experimental values of the neutral axis depth at

the level of the working moment against the number of load repetitions. In the early stages of the repeated loading process, beam FA 1-C reinforced Thereafter

with mild steel showed a noticeable reduction in the neutral axis depth. no practical however, change in the neutral axis depth was observed.

Beam FA 2-C,

showed a slight increase in the neutral axis depth. axis depths cycle were 0.84

At the levels of the working moments the ratios of the neutral measured after three million and 1.1 for beams FA 1-C cycles to those obtained at the first and FA 2-C respectively.

The factors which might have affected the position of the neutral axis in the ordinary beams, as discussed in section (a), also have a similar effect on the composite beams. Additionally, the formation of cracks in the f. r. c. channels axis depth. This is due to repeated loading could result in reducing the neutral mainly due to the reduction in the tension stiffening and also due to the propagation cracking

effect of the f. r. c. channels face once

of the cracks towards the compression

occurs in these channels.

For beam FA 1-C,

the cracking of the f. r. c. channel had considerably This can be seen in Fig. 56, where most of

depth. the axis neutral reduced the reduction loading, resulted

in the neutral axis depth occurred in the early stages of the repeated This

the f. r. c. channel being cracked after 450,000 cycles of loading. in a lower ratio for the final to initial axis depth of beam FA 2-C,

neutral axis depth for beam FA 1-0. however, was not greatly affected

The neutral by the cracking The relatively of increase neutral to initial


(c)

of the f. r. c. channels which occurred higher level of moment of beam FA 2-C compression

after 300,000 repetitions. had led to a greater rate

in the flexural

strain in the concrete,

consequently the ratio for the final

axis depth was increased. neutral

This resulted in a greater

axis depth for beam FA 2 -C.


Between Ordinary and Composite Beams

Comparison

beams the test showed a greater neutral axis composite static depth than the corresponding ordinary beams. After three million cycles the At the first ratios of the neutral axis depths of the ordinary beams to those of the corresponding and FA 1-C and 0.83 for beams

1-0 for beams FA beams 1.04 were composite FA 2-0 and FA 2-C.

This indicated that only in one case did a composite beam ordinary beam.

show a greater neutral axis depth than that of a corresponding 8.2.2 Variations of the Maximum Flexural Beams Compressive

Strain in Concrete

(a) Ordinary In Fig.

57 are plotted the observed values of the maximum flexural

compressive

load the level the the number of versus repetitions. moment strain at of working
In general, in the concrete; cycles for beams repeated loading caused an increase however, in the compressive during cycles the first strain 600,000

most of the increase, FA 1-0 and FA 2-0

occurred

and 1.5 million

for beam FA 4-0.

Thereafter

no practical

change was observed.

The ratios of the maximum flexural the working first

compressive

strain in the concrete at

moment measured after three million

cycles to that measured at the

cycle were 1.18 for beam FA 1-0

(1.044% mild steel), 125 for beam FA 2-0 (1.044% Kam 60). The initial

(1.044% Unisteel 410) and 1.4 for beam FA 4-0

values of stresses in the concrete at the upper limit

of the cyclic loading were

23% (beam FA 1- 0), 30% (beam FA 2- 0) and 34% (beam FA 4respective cylinder compressive strength. The stresses

0) of the

limit lower the at

limit. half the those to upper at nearly were equal These stresses were obtained by correlating compressive strain the measured maximum experimental stress-

strains in concrete with the corresponding of concrete in compression.

relationship

This indicated that the greater the greater is the Similar

the level of stress at the upper limit rate of increase behaviour concrete in the flexural

of the loading cycles,

compressive

strains in the concrete. investigation

was also observed in a previous, members (93).

on normal weight

fib) Composite Beams The experimental flexural the maximum values of compressive strain in

the concrete at the level of the working moment are plotted against number of load repetitions in Fig. 57. strain in change

This shows that repeated design load increased the compressive the concrete. was observed. The ratios of the maximum flexural the working first compressive However, after about 1.5 million repetitions

no practical

strain in the concrete at

moment measured after three million

cycles to that obtained at the and FA 2-C respectively.

1-C FA beams for 1.78 1.8 and cycle were

for the corresponding those than obtained These ratios are considerably greater because f. the is This (a). beams discussed mainly r. c. in section ordinary channel did not crack at the first for the flexural compressive lower in initial resulted which cycle In addition, values

strain in the concrete.

cracks were This

formed in the f. r. c. channels during the process of repeated loading. considerably concrete.
(c)

increased the values of the flexural This behaviour is explained in 7.4 (c).
Between Ordinary and Composite

compressive

strains in the

Comparison

Beams

The composite beams, at the first flexural ordinary compressive beams.

static test,

showed lower values for the

strains in the concrete than those in the corresponding of the f. r. c. channels cycles the

This was mainly due to the contribution cycle of loading.

which did not crack at the first

After three million

ratios of the maximum flexural compressive strains in concrete in the ordinary beams to those in the corresponding composite beams were 0.7 and for beams FA 1-0 and FA 1-C (1.044%mild steel) and 0.87 for buns FA 2-0 and FA 2-C

(1.044% Unisteel 410). flexural beams. compressive

This indicated that the composite beams had greater strains in the concrete than the corresponding ordinary

This was mainly due to the formation

of cracks in the f. r. c. channel

during the process of repeated loading which consequently led to an increase in the flexural 8.2.3. compressive strains.

Stresses in the f. r. c. Channels The stresses in the f. r. c. channels were obtained by correlating the values

of strains relationship

measured on these channels with their experimental shown in Fig. 8.

stress-strain

At the first FA 1-C

for f. beams in the loading the c. channels r. stresses cycle of loading fatigue to were nearly the subjected which were beams ST 1-C and ST 2-C. At the level stresses

and FA 2-C

same as those for the corresponding of the working moments in the f. r. c. 8.93 N/mm2 (upper limit

of the loading cycles) the maximum

channels for beams FA 1-C respectively. and ST 2-C maximum

and FA 2- C were 5.83 N/mm2 and

The stresses in the f. r. c. channels of the corresponding were 5.33 N/mm2 and 9.13 N/mm2 respectively.

beams ST 1-C The initial

stresses in the f. r. c. channels in the composite beams for beam FA 1 -C

subjected to fatigue loading ranged between (2.3 - 5.83 N/mm2) and between (3.2 - 8.93 N/mm2) for beam FA 2-C. percentages (see Fig.

These values expressed as

of a nominal tensile strength equal to 15 N/mm2 for the f. r. c. units and between 21.3 - 59.5%

8) are between 15.3 - 38.9% for beam FA 1-C

for beam FA 2-C. The behaviour of the f. r. c. channels in the composite beams tested under

fatigue loading could be mainly affected by the level of stresses at the lower and upper limits of the loading cycles.

It is worth pointing out that the increase in the strain of the f. r. c. channels under the stresses at the lower limit was mainly due to creep, as the channels for the period of repeated loading. at the lower limit (2.3 N/mm2

were subjected to these stresses continuously Referring to Fig.

8, it can be seen that the stresses

and 3.2 N/mm2) of the behaviour behaviour

are low and may be considered to be within the elastic range of the f. r. c. units. Thus the effect of these stresses on the

of the f. r. c. channels may not be considerable. therefore, that the increase in strain and the consequent

It is most likely, formation loading. The formation

of cracks in the f. r. c. channels was mainly due, to the effect of repeated

of cracks in the f. r. c. channels was observed after 450,000 and FA 2-C respectively. This

cycles and 300,000 cycles for beams FA 1-C indicated

that the range of stresses in the f. r. c. channels between the lower and of the loading cycles which were 2.3 - 5.83 N/mm2 for beam directly FA 1-C

upper limit

and 3.2 - 8.93 N/mm2 for beam FA 2-C

affected the number of cycles

at which cracks were formed (i. e., the higher the range of applied stresses the lower is the number of repetitions One of the important at which the channel crack).

features in the behaviour of the f. r. c. channels under

the effects of repeated loading was the good bond maintained between these channels and the concrete throughout the process of the fatigue test. No separation occurred

between the concrete and the f. r. c. channel even when the beams were finally tested to destruction 8.2.4. Variations after the application of the Steel Stresses of three million cycles.

(a) Ordinary

Beams

The main factor which might have affected the values of the steel stresses " in beams subjected to fatigue loading was the variation which alter the lever arm of the steel. The working at the first steel stresses calculated in accordance with 4.6 for the beams values) and after three million cycles (final values) are of the neutral axis depth,

cycle (initial

given in cols.

3 and 4 of table (11).

equation (4.18) was used, where the only variable was the neutral axis depth. The value of corresponded to the stress distribution of the concrete in compression at the first cycle of loading. The ratios of the final to initial col. 5 of table (11), were 0.99,1.02 values of the steel stresses, and 1.03 for beams FA 1-0, as given in FA 2-0 and

In this calculation

FA 4-0

respectively.

This,

however,

indicated that repeated loading did not

affect significantly

the values of the steel stresses.

(b) Composite Beams The initial and final values of the steel stresses 3 and 4 of table (11). of the f. r. c. units, values, calculated in accordance equation 4.20

with 4.6 are given in cols. was used, the contribution the calculation of the initial

In this calculation

however,

was only considered in

since the channels for both beams cracked The ratios of final to initial values of

during the process of repeated loading. the steel stresses, FA 1-C

as given in col. 5 of table (11), were 1.17 and 1.22 for beams respectively. ordinary It can be seen that these ratios are greater

and FA 2-C

than those for corresponding

beams as discussed in section (a).

This is mainly due to the following: i) The initial beams the in the composite were steel stresses values of ordinary beams. This was due

lower than those in the corresponding to the contribution cycle of loading. ii) The formation

first did the f. the at not crack channels which r. c. of

during the f. in the process of c. channels r. of cracks When the channels cracked the tensile stresses which to the tension

repeated loading. were carried reinforcement

by the f. r. c. channels were transferred of the member at cracked sections.

Another factor which might have affected the values of the steel stresses was the variation of the neutral axis depth due to repeated loading and the in the member.

consequent change in the lever arm of the tension reinforcement (c) Comparison Between Ordinary and Composite Beams cycle of loading,

Col. 3 of table (11) shows that, at the first

the steel

stresses in the composite beams are lower than those in the corresponding ordinary beams. This, as explained in section (a) earlier, was due to the cycle. After beams

contribution three million

of the f. r. c. channel which did not crack at the first cycles

the values of the ratio of steel stress in the ordinary

to that in the corresponding

composite beams, as given in col. 4 of table (11), and FA 1- C) and 0.96 for beams (FA 2-0 and

were 1.022 for beams (FA 1-0 FA 2- C).

This indicated that the steel stresses

in the composite beams did not

differ

significantly

from those in. the corresponding

ordinary

beams even after

three million
8.2.5.

cycles and :hi spite of the formation


Limit States

of cracks in the f. r. c. channels.

Principal

8.2.5.1.

Limit

State of Ultimate

Strength

(a) Ordinary

Beams

At the end of the fatigue loading test all the beams were tested up to failure under static loading, Beam FA 1-0 as none of the beams tested failed under fatigue loading. moment equal to 36.125 kN. m.; the ultimate was 39 M. nl. Beams FA 2-0 and

failed at an ultimate

moment of the corresponding FA 4-0

beam ST 1-0

exhibited nearly the same ultimate and ST 4-0 respectively.

strength as for the corresponding Previous investigations on prestressed

beams ST 2-0 and reinforced

design that beams repeated worldng showed normal weight concrete effect on the ultimate strength of the members

load did not have an appreciable (64), (94), (95). working

In some cases, however,

it was found that repeated design strength of the members (73), (96). mechanism was similar

load caused an increase in the ultimate

The behaviour of the beams with regard to the failure

to that observed for the beams which were tested under static loading as explained in 7.6.1. Col. 6 of table (11) gives the calculated ultimate using the strain compatibility moments of the beams In this method the

method explained in 5.2.1.

actual strength values of concrete and steel, without the partial of the material,
The ratios given in col.

safety factors

were used.
of observed to calculated 1.13 values 1.15 of the ultimate and 1.12 moment, as

8 of table

(11), were respectively. greater

FA 2-0 ultimate

and FA 4-0 'moments were

, This shows that the observed ones.

for beams

FA 1-0, of the might

values

than the calculated in 7.7.1.

The factors

which

have affected

this behaviour

are discussed

(b) Composite Beams The composite beams tested did not fail under fatigue loading. An excellent

bond between the f. r. c. channels and the concrete was maintained throughout the

process

of repeated loading. tested to destruction, beam FA 1-C beam ST 1-C failed failed

When the beams were finally at an ultimate at 40.5 kN. m. compared concluded,

the M. 38.375 corresponding m; of moment Beam FA 2-C had an ultimate

strength equal to 50.5 kN. m beam ST 2-C. It may be effect

for the corresponding M. 46.5 m with therefore,

that fatigue loading did not have an appreciable

on the ultimate The failure static

beams. the of composite strength mechanism was similar to that for the beams tested under

loading only as explained in 7.6.1. The calculated ultimate the beams, the strain compatibility using moments of

method and incorporating without the partial

the actual values of strengths for concrete and steel table 6 (11). in of col. given are 8 table (11) in of as given col. moments The factors which

safety factors of materials

The ratios

of observed to calculated ultimate

were 1.18 for beams FA 1-C

FA 2-C. beam for 1.28 and

the ultimate have influenced the of values observed could 7.6.1. in beams for those the explained as as ordinary
Comna. risonBetween Ordina

moments are the same

Beams Composite and for the ordinary moments and composite

The observed values of the ultimate beams are given in col. 7 of table (11).

This shows that the observed ultimate greater than those of the corresponding to the

moments of the compositebeanis were slightly ordinary ultimate beams.

This may indicate that the f. r. c. channels contributed However,

beams. the strength of composite

it must be emphasised that

at the end of fatigue loading the channels were cracked at many sections and that the contribution of the f. r. c. channels if any would be negligible.
8.2.5.2. (a) Limit State of Deflection

Ordinary

Beams

The remaining and FA 4-0

and working load deflections

for beams FA 1-0,

FA 2-0

at the first

cycle (given in cols. 2 and 3 of table (12)) agree well with beams ST 1-0, However, ST 2-0 and ST 3-0 was that

those observed for the corresponding (given in cols. 2 and 3 of table (9)). beam FA 2-0

the maximum difference

had a working deflection

10% greater

than that observed for the

corresponding tests carried in Fig. 58.

beam ST 2-0.

Typical

load deflection

curves for static loading are shown

out at frequent intervals

during the repeated loading test

This shows that the difference between the values of deflection and unloading Typical stages diminishes load-deflection

at the loading

with an increase in number of repetitions.

behaviour for the loading and unloading stages when is shown in Fig. 59. The behaviour in

the beams were finally general was similar only.

tested to failure

to that observed for the beams tested under static loading

For each beam the deflection the same at the first

at the level of the working moment was nearly Similarly the deflection at

loading. of cycles and second

1.5 times the working moment was nearly equal at the second and third cycles of loading. The variations in the remaining deflection and the deflection FA 2-0 at the working are

moment with number of repetitions shown in Fig.

for beams FA 1-0,

and FA 4-0

60. It can be seen that most of the increase in the remaining first during the six the deflection and the deflection at working moment occurred hundred thousand repetitions. It is important to emphasise here that the first static test carried of loading. occurred increase 500,000 loading fatigue about cycles after during was the out process of Hence the greater rate of increase in deflection might well have few thousand repetitions. was observed. cycles Thereafter no practical

during the first

in the amount of deflections

The ratios of the deflection


to that at the first FA 1-0 FA 4-0 were (mild cycleeas

three the after moment million working at


G of table (12), were 1.55 for beam

given in col.

steel),

1.31 for beam FA 2-0 This indicates

(Unisteel

410) and 1.22 for beam steel stresses in the because of steel. flexural of travel

(Kam 60 steel).

that when high working of steel a lower rate

employed

with the same percentage with number steel stresses

of increase

amount

of diflection

of repetitions were employed cycle

resulted.

This was mainly

when high working a greater members deflection

with the same percentage Additionally, and height

resulted

at the first steel stresses

of loading.

with high working

had greater

width

of cracks at the first concrete members.

cycle of loading.

In a previous investigation

on normal weight

an increase between 20 and 25% in the amount of deflection was Col. 4 of table (12) shows that

observed in the early stage of repeated loading (97).

the deflection

at the working moment after three million limit

cycles did not exceed with either mild satisfied

the recommended

of span/250 (i. e. 18mm) in beams reinforced This indicates that beams FA 1-0 even alter three million and FA 2-0

steel or uni steel 410. the limit

state of deflection Beams

cycles of loading.

(b) Composite

The working FA 1-C observed

and remaining

deflection at the first

static test for beams

and FA 2-C

(given in cols. 2 and 3 of table (12)) agree well with those beams ST 1-C and ST 2-C (given in cols. 2

for the corresponding However,

and 3 of table (9)). had a working ST2 - C. Fig. carried

the maximum difference

was that beam FA 2-C beam

deflection

8% lower than that observed for the corresponding

58 shows a typical load deflection

from static tests obtained curve The behaviour

out at frequent intervals to that of the ordinary

during the repeated loading test.

was similar of deflection

beams, where the difference between the values with an increase in

at the loading and unloading ranges diminishes A typical load-deflection is shown in Fig.

number of repetitions. finally tested to failure

behaviour for the beams when

59. For each beam tested the deflection and second moment was

first in the the level the the same at of working moment was nearly cycles of loading. Similarly the deflection at 1.5 times the working

the same in the second and the third cycle of loading. The variations in the remaining deflections and deflections at the working are was

moments with the number of repetitions shown in Figs.

for beams FA 1-C

and FA 2-C

60. It can be seen that the rate of increase in both deflections 1.5 million cycles. Thereafter no practical

more pronounced during the first

change in the deflection was observed. A major factor which could have influenced the rate of increase in the deflection was the formation However, of cracks in the f. r. c.. channels during the process of examination of Figs. 60 shows that the formation in the deflection

repeated loading.

of cracks in the f. r. c. channels did not cauae a. sudden increase of the members.

This could be due to the fact that the channels cracked at the

level of the working moment where the stresses in the f. r. c. channel were relatively low. It may follow from this that the transference of the stresses from

the f. r. c. channels to the steel did not affect significantly member. The values of the remaining moment at the first deflections

the curvature

of the

and deflections

at the working

and after three million

cycles are given in table (12). deflection at the working moment

The values of the ratios of final to initial

were 1.97 for beam FA 1- C and 1.87 for beam FA 2-C. greater than those observed for the ordinary beams.

These ratios were

This was mainly due to the

f. r. c. channels did not crack at the first values for deflection. process In addition,

cycle which resulted in lower initial of the f. r. c. channels during the

the cracking

of repeated loading caused in

increase - in the amount of deflection. span/250 (i. e. 18mm) was not cycles. Therefore, the with

For both beams the recommended limit exceeded even after the application composite beams FA 1-C regard to the limit
(c) Comparison

of three million

and FA 2-C

were considered to be serviceable

state of deflection under fatigue loading.


Ordinary and Composite Beams

between

The deflection behaviour in relation ordinary

to the number of repetitions G0.

for the

and composite beams is shown in Fig.

In the first in the deflections

static tests the composite beams showed considerable at the working moment and also in the remaining for the ordinary the moment at working

saving

deflections. beams to and

The ratios of the deflection that of the corresponding FA 1-C (mild steel),

composite beams were 1.3 for beams FA 1-0 and FA 2-C

and 1.53 for beams FA 2-0

(Unisteel 410).

The ratios of the remaining corresponding

deflection for the ordinary

beams to that of the and FA 1-C, and

composite beams were 1.29 for beam FA 1-0

1.53 for beams FA2-OandFA2-C. From the examination working of Fig. 60 it can be seen that at the levels of the stiffness than the

moment the composite beams showed a greater ordinary

corresponding Thereafter

beams within a range of 1.5 million

repetitions. as for beam FA 1 -0, up

beam FA 1-C

exhibited nearly the same stiffness maintained a greater stiffness

whereas beam FA 2-C to three million cycles.

than beam FA 2-0

The values of deflection

observed after three million

cycles for ordinary

and composite beams are given in col. 4 of table (12).

The ratios of the deflection

at the working

moment of the ordinary beams to that of the composite beams were and FA 1-C, and 1.07 for beams FA 2-0 and FA 2-C. deflections ordinary of

1.02 for beams FA 1-0

From the same figures it can also be seen that the remaining the composite beams were lower than those of the corresponding up to 1.4 million cycles. Thereafter the remaining deflections

beams

of the composite ordinary beams.

beams did not differ After three million

significantly

from those of the corresponding deflection

cycles the ratios of the remaining

of the ordinary and

beams to that of the corresponding FA 1-C,


8.2.5.3. (a)

composite beams were 1.15 for beams FA 1-0

and 0.93 for beams FA2-OandFA2-C.


Limit State of Cracking

Ordinary

Beams

The maximum and average crack widths at the working moment for beams FA 1-0, FA 2-0 and FA 4-0 at the first cycle are given in table (13). These

values fairly ST 2-0

agree with those observed for the corresponding

beams ST 1 -0,

and ST 4-0,

the values of which are given in table (10).

The effect of repeated loading on the maximum and average crack widths at the steel level can be seen in Figs. and FA 4-0. These figures 61 and 62 for beams FA 1-0, FA 2-0

show that the crack widths increased with the number

of load repetitions. However, a stabilised million condition was reached within one million cycles for beam FA 2-0 cycles for

beam FA 1-0,1.7

and six hundred thousand

cycles for beam FA 4-0. An examination of Figs. 61 and 62 shows that, in some cases, a reduction This could be due to the formation of new cracks

in the crack width was observed.

which tended to reduce the width of adjacent cracks. Col. 6 of table (13) shows that even after three million the maximum maximum repetitions of load the

crack width at the steel level did not exceed 0.2mm.

However,

crack widths at the bottom

edge of the beams, as given in col. 8 of in, b nFA 2-0 and 0.24mm in

table (13), were 0.2mm in beam FA 1-0,0.28niin beam FA 4-0. The average crack width, not exceed a value of 0.2mm.

as can be seen in cols 7 and 9 of table (13), did Cols. 10 and 11 of table (13) show that the ratios

-128-

of the maximum to that at the first at the steellevel beams.

crack width at the working moment after three million

repetitions

cycle of loading ranged between 1.25 and 1.43 for the cracks for the cracks at the bottom edge of the between 1.43 1.2 and and increases in crack width ranged between 20

In a previous investigation

and 25% in normal weight concrete members tested under fatigue loading were observed (97).

At the level of the working moment, the maximum crack width in the ordinary beams after three million indicate million cycles did not exceed a value of 0.3mm. state of cracking This may

that the beams had satisfied the limit cycles of loading.


Beams

even after three

V2) Composite

When the beams were first were observed at their soffits. did not crack in the first beam FA 1-C

loaded up to their working moment no cracks This was mainly because the f. r. c. channels At the interface (level of the steel) had a maximum

cycle of loading.

did not show any crack whereas beam FA 2-C

crack width of 0.04mm under the working moment. Figs. Gl and 62 show the variation in the maximum and average crack for beams FA 1-C and

widths in relation FA 2-C. within

to the number of load repetitions

For both beams repetitions.

the in increase the crack widths occurred most of It is worth pointing out here that most of the also occurred within the same and

1.5 million

increase

in the deflection as mentioned in 8.2.5.2.

number of repetitions. cracking behaviour

This indicated a direct relation between the deflection

in composite beams subjected to fatigue loading.

One of the major factors which could have affected the crack widths in the concrete was the formation of repeated loading. of cracks in the f. r. c. channels during the process was not significant compared with 61

This effect, however,

that observed in beams tested under static test only. and 62 that when the channels cracked increase excessively.

It can be seen in Figs.

the crack width in concrete did not due to the fact

This could be as mentioned in 8.2.5.2.

that the f. r. c. channels cracked at the level of the working stresses in the f. r. c. channels were relatively low.

moment where the crack width

The maximum

observed at the time when cracks formed did not exceed a value of 0.2mm.

Examination

of Figs.

61 and 62 shows that at some stages a reduction This as explained in case of the ordinary beams

in the crack width was observed. could also be due to the formation of adjacent cracks.

of new cracks which tended to reduce the width

Col. 6 of table (13) shows that the values of the maximum steel level after three million in beam FA 2-C.

crack widths at the and 0.18

cycles were 0.15mm in beam FA 1-C

The values for the average crack widths given in col. 7 of table and 0.1 in beam FA 2-C. This clearly indicates of

(13) were 0.06 in beam FA 1-C

that the widths of the cracks measured at the steel level even after three millions cycles were not excessive but well within the recommended limits. Cracks at the bottom after three million 0.38mm 0.26mm FA 1-C edge of the beams (on the f. r. c. channels) observed

cycles' had a maximum value of 0.32mm in beam FA 1-C the average value were 0.27mm in beam FA 1-C The remaining

and and

in beam FA 2-C; in beam FA 2-C. and FA 2-C

crack widths at the bottom edge of beams

as given in col. 12 of table (13) were 0.1mm and 0.14mm

respectively. To study the crack formation that carried in the confined concrete a similar exercise to

out for the composite beams tested only under static loading as was carried out for the composite beam FA 2-C. had been subjected to three million This was cycles. (d)

explained in 7.7.3.2. carried

out after the beam FA 2-C and results

The observations (c) Comparison At the first

obtained were similar

to those explained in 7.7.3.2

Between Ordinary cycle of loading

Beams Composite and the composite beams FA 1-C while the corresponding at the working and FA 2-C ordinary beams

did not show any cracks at their soffits, FA 1-0 and FA 2-0

had maximum crack widths, Similarly

momentequal

to 0.14mm and 0.22mm respectively. were much greater beams. Figs. in the ordinary

cracks at the level of the steel

beams compared with those in the composite

61 and 62 show the cracking behaviour

of the ordinary Examination

and compcsite

beams which were tested under fatigue loading. shows that the composite beams had smaller within a range of 1.5 million cycles.

of these figures

crack widths at the level of the steel beam FA 1-C showed a slightly

Thereafter

greater

maximum

crack width than that in the corresponding showed a lower maximum up to three million

beam FA 1-0,

whereas beam FA 2-C corresponding

crack width than the cycles. The average crack ordinary

beam FA 2-0

widths in the composite beams were lower than those in the corresponding beam up to three million cycles.

The crack widths observed at the bottom edge of the composite beams, after the f. r. c. channel had cracked, ordinary beams. (see cols. were greater than those in the corresponding However, the actual crack widths

8 and 9 of table (13)).

in the confined concrete could be lower than those measured on the f. r. c. channel as explained in 7.7.3.2(d) The remaining crack widths after three million cycles as given in col. 12 from those observed

of table (13) in the composite beams did not differ in the corresponding 8.3 ordinary beams.

significantly

Sustained Loading Tests The sustained load for the test beams was maintained for a minimum period

of 500 days. table (6). 8.3.1.

The periods of sustained loading for the various beams are given in

Time-dependent Beams

Flexural

Strain Distribution

(a) Ordinary

The flexural

compressive

beams in the in tested under concrete strains strain in the

sustained loading increased steel. of time. the tensile

at a much higher rate than the tensile

This resulted in an increase in the neutral

axis depth with the passage in

At a later stage in the sustained loading period a slight reduction strains at the level of the steel was observed. of the neutral

Another explanation for the variation due to plastic flow of concrete, This necessitates a greater that the neutral

axis depth given was that

the compressive

stresses in the concrete reduced. in order to provide

axis depth should increase (98).

area of concrete in compression strain distribution 63. A similar reinforced

A typical flexural loading is shown in Fig. distribution

for a beam tested under sustained pattern for the time-dependent and prestressed strain

for conventionally

concrete beams has also

been reported by many investigators The initial (at the first

(64) (73) (96) (99).

cycle) and final (at the end of sustained loading) compressive strain in concrete at the working The ratios of the final to initial and 2.34 for

values of the maximum flexural moment are given in cols.

3 and 4 of table (14).

for beam SU 1-0 3.5 in 5 table (14)) (as were values given col. of beam SU 2-0.

The ratio for beam SU 1- 0 was greater than that of beam SU 2-0; value of beam SU 1-0 and also to the

this was mainly due to the lower initial fact that the increase in the compressive significantly compressive from that of beam SU 2-0.

strain in concrete did not differ The increases in the maximum flexural

strain in the concrete are given in col. 6 of table (14).

The shrinkage strains measured on the sides of the beams (as explained in 6.7.1 6 (e)) after 500 days of loading were 410 x 10- for beam SU 1-0 and 350 x 10-6 for beam SU 2-0. of the respective SU 1-0 This indicated that the shrinkage strains formed 45% and 37% increase in the flexural respectively. compressive concrete strains in beams 6,7 and 8 of

and SU 2-0

The values are given in cols.

table (14). Fig. After 63 shows the increase in the neutral axis depth with the passage of time.

500 days of loading the neutral axis depths were 62% and 58% greater than values in beams SU 1-0 and SU 2-0 respectively. 63. The values of the

the initial neutral (b)

axis depth for beam SU 2-0

can be seen in Fig.

Composite Beams The behaviour of the composite beams under sustained loading was basically beams. flexural strain distribution for a composite beam strains in the The

similar

to that of the ordinary time-dependent

A typical

is shown in Fig.

64, where it can be seen that the compressive

concrete and the neutral axis depth were increased with the passage of time. variation

of the tensile strain in the f. r. c. channels with the passage of time is

discussed in 8.3.2. The values of the maximum flexural are given in cols. 3 and 4 of table (14). compressive strains in the concrete values

The ratios of the final to initial and SU 2-C respectively.

were 3.55 and 2.82 for beams SU 1-C beam SU 1-C was greater

The ratio for

than that of beam SU 2-C;

this was mainly due to the

relatively

lower initial

value of beam SU 1-0.

The shrinkage strain in concrete measured on the sides of the beams (as explained in 6.7.1(e) after 500 days are given in col. 7 of table (14). indicates This increase

that the shrinkage strains formed 41% and 42% of the respective compressive strains in beams SU 1-C 6,7 and SU 2-C

in the flexural

respectively.

The values can be seen in cols.

and 8 of table (14). the final values were 28% and 48% greater and SU 2-C respectively. initial The lower value of the

As regards the neutral axis depth than the initial percentage neutral values for beams SU 1-C

for beam SU 1-C

was mainly due to its greater

axis depth. strain distribution When the in

A major factor which might have affected the flexural beam SU 2-C channel cracked were affected. was the formation

of a crack in the f. r. c. channel.

the neutral axis depth decreased,

consequently the strain values where the increase was greater than

This can be seen in the case of beam SU 2-C, compressive strain in beam SU 2-C

in the maximum flexural that of beam SU 1-C. (c) Comparison

The values are given in col. 6 of table (14). and Composite Beams

Between Ordinary

When the beams were first flexural compressive

loaded, the composite beams showed lower

strains in the concrete and greater values for the neutral ordinary beams. The values

axis depths than those observed in the corresponding of the initial After maximum flexural compressive strains

are given in col. 3 of table (14). maintained a lower value

500 days of sustained loading, beam SU 1-C compressive

for the maximum ordinary

strain in the concrete than the corresponding had a slightly greater value than

beam SU 1-0,

whereas beam SU 2-C beam SU 2-0,

the corresponding table (14). was greater formation contributed

ordinary

the values can be seen in col. 4 of

This could be mainly due to the shrinkage strain of beam SU 2-C than that of beam SU 2-0 (col. 7 of table (14)). Additionally the

of a crack in the f. r. c. channel of beam SU 2-C to the increase in the flexural compressive

might also have

strain. and composite beams

The values of the neutral

axis depth for the ordinary

after 500 days were found to be nearly the same.

The final values of the neutral and SU 2-0 respectively;

axis depths were 178mm and 182mm for beams SU 1-0

for beams SU 2-0


64. These

and SU 2-C

the final values are given in Figs. 63 and


and SU 2-C respectively.

beams for SU 2-0 178mm 174mm and are Strains

8.3.2

Vari %tions of Tensile

in the f. r. c. Channels maximum tensile stresses of the beam for beams in the

When the beams were first induced in the f. r. c. channels of the working

loaded the total including

the stress

due to selfweight

at the level SU 1-C

moment

were 4.3 N/mm2 The corresponding

and 10.2 N/mm2 values

and SU 2-C

respectively.

of strains

f. r. c. channels those obtained beams FA 1-C

were 311 x 10-G and 940 x 10-6. at the first static

These values

agree well with

test in the f. r. c. channels

for the corresponding loading as discussed

and FA 2-C

fatigue tested under which were

in 8.2.3. The initial percentages maximum tensile tensile stresses strength in the f. r. c. channels equal to 15 N/mm2 expressed as

of a nominal

were 29% and 68'oin

beams SU 1-C

and SU 2-C

respectively. tensile strain in the f. r. c. channels with Fig. in shown are 65.

The variations

in the maximum

the passage of time for beams SU I-C

and SU 2-C

It can be seen that the tensile strains in the f. r. c. channels increased at a diminishing rate as time elapsed. Most of the increase in the tensile strains occurred during the first 100 days of loading, after which the change was not tensile in the however, strains A observed appreciable. was slight reduction, after a period of one year. This could be due to the increase in the neutral axis depth caused by the creep and shrinkage ad discussed earlier.
After observed strain eight months of sustained loading a fine crack of 0.16mm width was tensile however, in the f. r. c. channel of beam SU 2-C. The maximum

measured This,

in the f. r. c. channel at the time of cracking than the maximum by direct tensile strain

was 1300 x 10-6.

is lower

(1800 x 10-6) of the f. r. c. channel 8). 1he reasons which might have its maximum tensile strain are

as measured

tensile

test (see Fig. reaching

caused the channel to crack before discussed a period in 7.5. No more cracks

appeared in the f. r. c. channel even after This behaviour strain may be explained by

of 500 days of sustained

loading.

the fact that most of the increase 100 days as mentioned earlier.

in the tensile Beam SU 1-C

occurred

during the first on the

did not show any crack

f. r. c. channel for the period of the sustained loading test. The formation of a crack in the f. r. c. channel of beam SU 2-C mainly due to the high initial strength of the channel.

was

tensile stress which was equal to 68% of the tensile

In addition the shrinkage strain which occurred in the

concrete might have also contributed to the development of cracking in the f. r. c. channel. 8.3.3
(a)

Variations

of the Steel Stresses

Ordinary

Beams

The steel stresses in beams tested under sustained loading could mainly be affected by the movement of the neutral axis depth. Due to creep of the concrete the neutral axis depth dropped, and consequently the lever arm of the steel was reduced. This resulted in an increase in the steel stresses for the same level of moment. Fig. GGshows the increase in the steel stress as time elapsed.
using the equation of compatibility results obtained

The values

in this graph were calculated (equation The value initial

for moments in 4.5.2.

4.18) based on the experimental for centroid distribution between

as explained

of the compression

stress

distribution loaded to their strain

was based on the working moment.

stress

for the beams when first P and the maximum

The relationship is explained depth. From

compressive

in the concrete axis

in 4.3.3. the figure

The only variable

the in the was neutral equation used increased at a

it can be seen that the steel stresses

diminishing

rate with the passage of time. and final values for the steel stresses This shows that the increases are given in cols. after 9 and 10

The initial of table (14). sustained SU 2-0

in the steel stress

500 days of and

loading

are 11.8% and 11.4% of the initial

values for beams SU 1-0

respectively. Beams loading on the steel stresses in the composite earlier. beams

Sb) Composite

The effect of sustained is basically the formation steel stresses stresses similar

to that in the ordinary

beams discussed

Furthermore

of cracks

in the f. r. c. channel could also increase of these cracks. In Fig.

the local tensile

in the vicinity

66 It can be seen that the In calculating the steel

in the steel increased

with the passage of time.

stresses

equation 4.20 for the compatibility the experimental results

of moment was used taking into The values

consideration

obtained as explained in 4.6. stress distribution

used for the centroid stresses

of the compressive

and the tensile

in the f. r. c. channel were those obtained when the beams were first

loaded to their working moments. In cols. are given. 9 and 10 of table (14) the initial This shows that the increases and final values of the steel stresses after 500 days of and SU 2-C

in the steel stresses

loading are 8% and 12.4% of the initial respectively. It is important

values for beams SU 1-C

to point out that the contribution

of the f. r. c. channel As mentioned

was included in the calculation earlier beam SU 2-C

of the steel stresses at all the stages.

had the f. r. c. channel cracked after eight months of the contribution of the f. r. c. channel

sustained loading. the increase initial values.

Without considering

in the steel stress of beam SU 2-C

would be equal to 33% of the

(c) Comparison From Fig.

Between Ordinary

and Composite Beams

66 it can be seen that during the period of sustained loading

the steel stresses in the composite beams are lower than those in the corresponding ordinary beams. After 500 days of loading, beam SU 1- C had a steel stress beam SU 1-0; beam SU 2-C beam SU 2-0. had a If the e. g. for and

8.8% lower than that in the corresponding

steel stress 15.7% lower than that in the corresponding contribution beam SU 2-C

of the f. r. c. channel is not considered in the calculation, where the channel cracked, the stresses in the ordinary

composite beams will be equal.


8.3.4. 8.3.4.1. (a) Limit Limit States of Serviceability State of Deflection

Ordinary

Beams

Deflection

under sustained loading is mainly affected by the variation compressive

taking

place with the passage of time in the flexural axis depth.

strain and the neutral

The increase in deflection under sustained loading for beams SU 1-0 SU 2-0 is shown in Fig.

and

67. For both beams it can be seen that with the passage

of time the deflection increases values of deflection remembered

at a diminishing

rate.

The initial

and final

are given in cols. 12 and 13 of table (15).

It should be

that the total deflection for a beam is equal to 7.7 times the value This is explained in 6.5.1, also see appendix (G). and SU 2-0 The greater as ratio

given in table (15).

The ratios of final to initial

deflection for beams SU 1-0

given in col. 15 of table (15) were 2.4 and 1.42 respectively. of beam SU 1-0 compressive was mainly due to its greater

rate of increase in the flexural rate of deflection.

strain in concrete,

which consequently led to a greater

In a previous investigation working

(36) it was found that after 25 months of sustained deflection ranged between (2 - 2.33) for

load the ratios of final to initial

lightweight beams. periods

concrete beams and between (2.34 - 3.09) for normal weight concrete ratios for final to initial deflection for various as follows: -

In other Investigations

of sustained loading for normal weight concrete were reported

3 for a period of two years (99), 3.14 to 3.94 for a period of about five years (100), 2.15 for a period of five months (101). Factors which could have affected these ratios were the level of steel stress, of steel and the surrounding temperature and humidity for the beams

percentage tested. After immediate

500 days of sustained loading the beams were unloaded and the remaining deflections at zero load were measured. These values are recoveries were 18% and 50% of the and

given in col. 14 of table (15). corresponding SU 2-0 final deflection

The immediate

at the working moment for beams SU 1-0

respectively. of beam SU 1-0 did not reach SU 2-0

Col. 13 of table (15) shows that the deflection the limit L/ 250 which for the test beam equal to exceeded this value.

18 2.34mmBeam a -. 7.7

however marginally

This indicates that the beams had satisfied the limit after 500 days of sustained loading. (b) Composite Beams The time-dependent Similar to the behaviour deflections

state of deflection

even

for the composite beams are shown in rig. beams the deflection

67.

of the ordinary

of the composite beams

increased factors

at a decreasing rate with the passage of time.

One of the major of beam SU 2-C

which could have affected the increase in deflection

was the formation loading. However,

of a crack in the f. r. c. channel after eight months of sustained it can be seen from Fig. 67 that the formation of a crack

in the f. r. c. channel did not have an immediate This was probably

effect on the deflection behaviour.

due to the fact that the channel cracked at the level of the working low and that the crack

moment where the stresses in the channel were relatively width in the f. r. c. channel was very small The values of deflection first (15). (0.16mm).

at the level of the working

moment obtained at the

test and after 500 days of sustained loading are given in col. 12 and 13 of table The ratios of final to initial deflection for beams SU 1-C and SU 2-C

were 3.5 and 2.0 respectively. to its greater

The greater ratio of beam SU 1. -C could be due compressive strain in the concrete

rate of increase in the flexural

which led to a consequent increase in deflection

(see col. 5 of table (14)).

When the beams were unloaded after the sustained loading period the remaining deflections indicated recoveries and SU 2-C of 25% and 50% of the respective respectively. final deflection

for beams SU 1-C

The final deflectionsof the composite

beams (as given in col. 13 of table (15)) when compared with the recommended limit L/250 (i. e. 2.34mm) indicated that the composite beams had satisfied the limit state of deflection (c) Comparison loading. days 500 sustained of even after Between Ordinary and Composite Beams

The deflections

beams the level the the when moment of working were first at The ratios of the deflection for the

loaded are given in col. 12 of table (15). ordinary

beams to that of the corresponding

composite beams were 1.78 for beams

SU1-0andSU1-C,

SU2-OandSU2-C. beams for 1.46 and can be seen in Fig. 67 between the time-dependent This figure shows that the composite This up to the

A direct comparison deflection for ordinary

beams. and composite

beams had a greater

stiffness in the early stages of sustained loading.

behaviour was maintained. by beam SU 1-C end of the sustained loading period.

compared with beam SU 1-0 the greater

However,

stiffness for beam SU 2-

in relation

to beam SU 2-0

gradually

diminished

with the passage of time.

The

main factors

which could have contributed

to this behaviour were the cracldng compared with

of the f. r. c. channel and the greater beam SU 2-0.

shrinkage of beam SU 2-C

The shrinkage strain for beam SU 2-C

as given in col. 7 of

table (14) was 1.4 times that for beam SU 2-0. Another factor which contributed rigidity to the diminishing of the greater flexural

of the composite beams in relation to the ordinary

beams was the greater

rate of reduction in the tension stiffening composite beams. travel

effect of the concrete with time in the loaded the height of

When the composite beams are first

of the cracks is limited,

hence the area of concrete between the neutral

axis level and the level of the tips of the cracks forms a major part in the tension stiffening drops, effect of the concrete. With the passage of time- the neutral axis

resulting

in a decrease in the untracked

concrete area in the tensile zone. effect of the concrete.

This will lead to a reduction in the tension stiffening The tension stiffening effect resulting

from the concrete between flexural and composite beams. beams to and

cracks may be of the same magnitude in the ordinary After

500 days of loading the ratios of deflection of the ordinary

that of the corresponding SU 1-C,


8.3.4.2.

SU for beams 1-0 1.22 beams were composite and SU 2-C.

and 1.03 for beams SU 2-0


Limit State of Cracl ing

(a) Ordinary

Beams reported that crack widths in prestressed and conventionally initially

Many investigators reinforced with

concrete beams tested under sustained loading were increased thereafter a stabilised

the passage of time, A similar

condition could be reached (64) of the crack widths with the This is shown in

(73) (102).

behaviour for the variation

passage of time is also observed in the present investigation. Figs.

68 and 69, where the maximum and average crack widths at the steel level are plotted against time. The initial and final values of the crack widths are given

in table (15). From the figures it can be seen that most of the increase in the crack widths occurred during a period of 200 days, thereafter no appreciable change was

observed.

At some stages a slight reduction in the crack widths was observed. of new cracks which tended to reduce the

This could be due to the formation width of adjacent cracks. However, during the first

it was observed that 90% of the total number of cracks was formed month of loading. maximum crack widthat respectively. the steel level were 2 and

The ratios of final to initial 2.2 for beams SU 1-0 maximum SU 1-0

and SU 2-0

The ratios of final to initial

crack width at the bottom edge of the beams were 1.83 and 1.4 for beams and SU 2-0 respectively. crack width for both beams after 500 days of loading, as given

The maximum in col.

8 of table (15), did not exceed 0.22mm.

This value is lower than a This indicated that the

recommended

maximum crack width of 0.3mm by CP110. the limit

beams had satisfied

state of cracking under sustained loading.

(b) Composite Beams The composite beams when first signs of cracking at their soffits. (at the steel level) did not crack in loaded to their working moment showed no

The concrete at the level of the interface beam SU 1-C SU2-C. Similar to the behaviour of the ordinary

whereas a maximum crack width of 0.08 was observed in beam

beams the width of cracks increased

with the passage of time for a period of about 240 days, after which no practical change was observed; this is shown in Figs.
After interface loaded. 500 days of sustained was 0.1mm loading

68 and 69.
crack width observed at the

the maximum which originally crack

for beam SU 1-C however, after

had no cracks width of 0.08mm,

when first which

Beam SU 2-C, to 0.14mm

had an initial

then increased

500 days of sustained

loading.

The maximum and average crack widths are given in table (15). mentioned in 8.3.2, beam SU 2-C

As previously

after eight months of sustained loading the f. r. c. channel of The width of the crack in the f. r. c. channel when first after 500 days of sustained loading the crack width reached

cracked.

observed was 0.16mm; 0.2mm.

This increase could be due to an increase in the crack width of the of the member.

confined concrete and also due to the increase in the curvature

Again the important the cracking concrete

consideration

as previously

mentioned in 8.3.2 is that affect the cracks in the

of the f. r. c. channel did not significantly

as it did in beams tested under static load only.

The composite beams after a period of 500 days of loading did not become unserviceable with regard to the limit state of cracking. This was due to the fact

that no cracks formed at the soffit of beam SU 1-C width observed in the f. r. c. channel of beam SU 2-C Similar to the behaviour of ordinary

whereas the maximum crack was 0.2. mm.

beams, 90% of the final number of the

cracks were formed within the first (c) Comparison Between Ordinary

month of loading. and Composite Beams 68 and 69 in the ordinary

The maximum and average crack widths as can be seen in Figs. composite beams are considerably beams. period. After less than those in the corresponding

This behaviour was maintained during the 500 days of the sustained loading

500 days of loading the ratios of the maximum crack width at steel beams to that in the corresponding and SU 1-C, composite beams were 2 and SU 2-C.

level in the ordinary for beams SU 1-0

SU for beams 2-0 1.6 and

The ratios of the average crack width in the ordinary corresponding

beams to that in the and SU 1-C, and

composite beams were 1.7 for beams SU 1-0 and SU 2-C.

1.9 for beams SU 2-0

When the beams were unloaded after 500 maximum crack widths in the composite ordinary beams.

days of sustained loading the remaining beams were much smaller

than those in the corresponding 10 and 11 of table (15).


pattern for ordinary This

The values are given in cols.


In Fig. subjected 70 a typical

cracks

and composite

beams of travel beams. the

to sustained

loading

is shown.

shows that the height

of the cracks This behaviour

in the composite

beam was lower

than that in the ordinary value

for the composite

beams is of great zone.

since it reduces

cracked

concrete

area in the tensile

CHAPTER

NINE

CONCLUSIONS AND SUGGESTIONS FOR FUTURE WORK


9.1 Conclusions

The Conclusions drawn from the present investigations (1) The prediction ultimate sufficient

are as follows:

of the stresses in the steel and the f. r. c.. channels, deflections and width of cracks can be made to a

strengths,

degree of accuracy by the methods suggested in this thesis.

(2) The bond between the f. r. c. channels and the concrete was very good (3) The f. r. c. channels did not crack up to and well above the level of the working moments, steel. (4) Adequate warning of impending failure tested. (5) There was a considerable reduction in width of cracks and deflections fatigue and sustained loading was obtained for all the beams the exception being one beam containing "Kam 60"

of composite beams tested under short term, as long as the f. r. c. channels did not crack.

(6) Deformed steel bars had a better control on flexural mild steel bars. (7) There was no significant

cracking

than did

change in the steel stresses for ordinary whereas a maximum increase

beams

subjected to fatigue loading,

of 22% was

obtained in the composite beams after the application (8) There were no significant

of 3x 106 cycles.

increases in the steel stresses for the beams period of 500 days. strength of the beams

tested under sustained loading fora

(9) Fatigue loading did not greatly affect the ultimate tested.

(10) Fatigue loading rather than sustained loading was a critical the cracking of the f. r. c. channels.

condition for

(11) Sustained loading rather than fatigue loading was a critical for the limit (12) Deflection state of deflection,.

condition

under short-term,

fatigue and sustained loading was found

to be a critical 9.2

factor in the design of the beams.

Advantages of Using f. r. c. Channels at the Tensile Sides of Flexural Members

Concrete

The use of f. r. c. channels in the composite beams resulted in the following advantages: (1) Full utilization of the tensile properties of concrete.

(2) Better control on shrinkage cracks. (3) Lower flexural axis depth. (4) A reduction in the steel stresses for the same level of load. (5) A considerable a more efficient 9.3 reduction'in deflection and crack width, steel.
4b.r

compressive

strains in concrete and greater neutral

thus allowing

use of the high-strength

Suggestions for Future Work


(1) The present investigation

used a span to depth ratio of 17 (maximum concrete members

permitted reinforced

in CP110 for simply supported lightweight

with 1% of a steel of 410 N/mm2 nominal yield stress and no steel. the

compressive

It was found that for some beams containing high strength steel limit state of deflection L/250 was exceeded. It is therefore

suggested that beams with a span to depth ratio lower than

This would result in a lower amount of deflection under load, thus allowing better use of the high strength steel. (2) The present investigation It is therefore structural was carried out using rectangular shaped beams. shape on the

17 should be investigated.

suggested that the effect of the geometrical should be investigated.

behaviour

In this respect,

a composite lightweight

concrete T beams with normal

weight concrete flange can be used.

The use of normal weight concrete

flange will help to a great extent in reducing the amount of deflection under load. At this point it would be interesting lightweight concrete web. to use f. r. c. channels on the

The good bond achieved between the f. r. c. concrete will ensure the restraint action

channel and the lightweight

of the f. r. c. channels, thus enabling high-strength (3) The present investigation stiffness cracking

steel to be used.

has shown that composite beams exhibited a beams for levels lower than the

greater than that of ordinary

moments of the f. r. c. channels. to the cracking of the

One of the factors which might have contributed f. r. c. channel was the formation These cracks make an important flexural rigidity

of cracks in the confined concrete. contribution to the reduction in the

of the composite beams. suggested regarding the formation of the cracks in

A study is therefore the confined concrete.

(4) In the fatigue loading tests of the present investigation, load hundred few thousand repetitions. cracked after a required

the f. r. c. channels A study is therefore

into the effect of fatigue loading on f. r. c. units with various A relationship should be established stresses at which

ranges of applied tensile stresses. between the number of repetitions cracking Similarly occurs.

and applied tensile

the tensile creep and the consequent formation

of cracks in the

f. r. c. units, Following this,

employing various levels of 'stressshould also be investigated. fatigue and sustained loading tests should be carried out on

composite beams. be examined.

The effect of long-term

weathering

and shrinkage should also

LIST OF REFERENCES

1.

Hognestad,

Development of High-Strength Steel Journal of Metals, June 1966.

2.

Abeles, P. W. and Gill, V. L. High-Strength Strand Reinforcement Concrete, April 1969.

for Concrete

3.

Ramaswamy, G. S., Chandre, It. and Naraynan, R. Savings in Reinforced Concrete Structures by Using Ultimate Design Procedure and High Strength Deformed Bars.
Indian Concrete Journal. October 1967.

Strength

4.

Clark, L. A. The Flexural Strength of Concrete Beams Reinforced Strength Steel. PhD. Thesis. University of Sheffield, 1968.

with Very High

5.

British Standards Institution. Code of Practice The Structural Use of Concrete. Part 1. November 1972.

CP110 -

6.

Grauholm, H. Kam 40, Kam 60, Och Kam 90. Trans of Chalmers, University of Technology, Gothenberg, Sweden, 1959.

No. 213.

7.

American Iron and Steel Institute Committee of Concrete Reinforcing Bar Producers. High Strength Reinforcing Bars, Vol. 1,1964, revised

edition.

8.

Short, A. The Use of Lightweight Concrete for Reinforced R. C. R., September 1959.

Concrete Construction.

9.

Teychenne, D. C. Structural Concrete Made with Lightweight Concrete, April 1967.

Aggregates.

10.

Manual Lightweight Concrete C. E. B., Bulletin d Information No. 85. Second draft Cand C. A. - May 1972.

11.

Shideler, J. J. Concrete for Structural Lightweight-Aggregate A. C. I. Journal, October 1957.

Use.

12.

Hanson, J. A. Replacement of Lightweight Aggregate Fines with Natural Structural Concrete. A. C. I. Journal, July 1964.

Sand in

13.

Baron-Hay,

J. K.

A Study of Physical Properties of Structural Quality of Lightweight Aggregate (Aglite and Terlite). PhD. Thesis, University of London, July 1960

14.

Zunz,

G. J.

Some Examples of When and How Lightweight I. C. L. C., May 1968.

Concrete be Used.

15.

Skoyles,

E. R.
Structural Concrete

Aggregate The Economics of Lightweight B. R. E. CP7/73, March 1973. ,

16.

Parrott, L. J. Properties for Workable Concrete. Selection of Constituent London, C and CA., Technical Report 417, May 1969.

17.

Abeles, P. W. Introduction to Prestressed Concrete Vol. 1, Concrete Publication Ltd., London,

1964.

18.

British Standards Institution, Code of Practice CP114 The Structural Use of Reinforced Concrete in Building 1965. reset and reprinted,

19.

Porter, II. F. Preperation of Concrete From Selection of Materials to Final Disposition. Proceedings of the National Association of Cement Users. A. C. I., Vol. 6,1910, p 296.

20. "

Romual. di, J. P. and Mandel, J. A. Tensile Strength of Concrete Affected by Uniformly Distributed Closely Spaced Short Lengths of Wire Reinforcement. A. C. I. Journal, June 1964.

and

21.

Hughes, B. P. Advances in Concrete


proceedings of a symposium September 1971. held at the University of Birmingham,

22.

Hannant, D. J. . Steel Fibre Reinforced Concrete Prospects for fibre reinforced construction materials B. R. E. Department of the Environment, 1972 , Swamy, R. N.
of Steel Fibre Reinforced Concrete for Practical

23.

The Tecluiology Applications.

Proceedings

of I. C. E. , May 1974.
R.

24.

Austerberry,

The Use of Fibres as Reinforcement in Concrete MSc. Thesis, 1972, University of Salford.

25.

McCurrich, L. H. and Adams, M. A. J. Fibres in Cement and Concrete Concrete, April 1973.

26.

S wamy R. N. , Delft Conference on Fibre Reinforced Concrete, December 1973.

Materials

27.

Fibre Reinforced Concrete Conference on Properties and Applications of Fibre Reinforced and Other Fibre Reinforced Building Materials. Delft, September 1973

Concrete

28.

Allen,

H. G.

The Purpose and Methods of Fibre Reinforcement. Materials. for Fibre Reinforced Construction Prospects 1972. B. R. E., Department of the Environment,

29.

Steel, B. R. Glass Fibre Reinforced Cement. Prospects for Fibre Reinforced Construction Materials. 1972. B. R. E. Department of the Environment, ,
Hanson, J. A.

30.

Shear Strength of Lightweight Reinforced Concrete Beams Journal of A. C. I. Proc., September 1958.

31.

Hanson, J. A. Tensile Strength and Diagonal Tension Resistance of Structural Concrete A. C. I. Journal, July 1961.

Lightweight

32.

Welch, G. B. and Patten, B. J. F. Concrete. Structural Lightweight-Aggregate Constructional Review. February 1964.

33.

Evans, R. H. and Hardwick,


Lightweight Concrete R. C. R. June 1960. ,

T. R.
Clay Aggregate.

with Sintered

34.

Richart, F. and Jensen, V. Tests of Plain and Reinforced Concrete Made With Haydite Aggregate. Engineering Experiment Station, University of Illinois, 1931.

35.

Evans, 11.11. and Orangun, C. O. Behaviour in Flexural of Reinforced Lightweight-Aggregate Concrete Beams. Civil Engineering and Public Works Review, May 1964.

(Lytag)

36.

Evans, R. H. and Paterson, W. S. Long-Term Deformation Characteristics Concrete.

of Lytag Lightweight-Aggregate
of Structural

Paper presented at a joint meeting of the Institution Engineers and the Concrete Society.

London, January 1967.

37.

European Committee for Concrete Information Bulletin No. 24, Part 1. Proceedings of the fifth working session.

Vienna, April

1959.

38.

Roberts, N. P. The Use of Deformed Bars as Reinforcement in Lightweight One-day symposium on structural lightweight concrete.
June 1962, Brighton.

Concrete.

39.

Kanoh, Y.
The Long-Term Deflection of Reinforced Artificial Lightweight Aggregate. Concrete Beams made with

I. C. L. C. Vol. 2, London, May 1968.

40.

Swarny, R. N. and Ibrahim, A. B. Deflection Characteristics of Structural Lightweight Reinforced and Prestressed Concrete Beams. Deflection of concrete structures, A. C. I. sp 43-15, April 1974.

41.

Romualdi, J. P. and Batson, G. B. Behaviour of Reinforced Concrete Beams With Closely Spaced Reinforcement. A. C. I. Journal, June 1963.

42.

Biryukovich, Steldotsement,

K. L., Biryukovich, kiev, Budivelnik,

Yu.

L.

and Biryukovich,

D. L.

Glass Fibre-Reinforced Cement, Translated by G. L. Cairns, CERA Translation No. 12, Civil Engineering Research Association, London 1965.

43.

Taylor, R. An Idea for a New Structural Material. Engineering, London, December 1971.

Composite Reinforced

Concrete.

44.

Kaar,

P. H. and Mattock,
Bars

A. H.
Reinforcement, Part 4, Control of

High Strength Cracking.

as Concrete

Journal of the P. C. A. Research and Development January 1963.

Laboratories.

45.

Hognestad, E. High Strength Bars as Concrete Reinforcement. Part 2, Control of Flexural Cracking. Journal of the P. C. A. Research and Development Laboratories. January 1962.

46.

Nervi, P. L. Ferro-Cement: 'Its Characteristics and Potentialities". (In Italian), L'Ingegnere (Rome), No. 1,1951.

47.

Romualdi, J. P., Frasier, J. T. and Irwin, G. R. Crack Extension Force Near a Riveted Stiffener. 1957. Report 4956, U. S. Naval Research Laboratory,

48.

Broms, B. B. and Shah, S. P. Discussion of Reference (41) and (51) Proceedings, A. S. C. E. Vol. 90, E. M. 1. February , , Abeles, P. W. Discussion of Reference (41) and (51) Proceedings A. S. C. E., Vol. 90, E. M. 1.

1964.

49.

February

1964.

50.

Kar,

J. N. and Pal, Division,

A. K.

Strength of Fibre-Reinforced
Structure

Concrete
of A. S. C. E. , May 1972.

Proceedings

51.

Romualdi, G. B. J. P. and Batson, Mechanics of Crack Arrest in Concrete Proceedings, A. S. C. E. June 1963. ,

52.

Untraur, R. E. and Works, R. E. Discussion of Reference (20) A. C. I. Journal, December 1964.

53.

Abolitz, A. L. Discussion of Reference (20) Journal of A. C. I., December 1964.

54.

Agbim, C. C. Discussion of Reference (20) A. C. I. Journal, December 1964.

55.

Shah, S. P. and Rang-an, B. V. Fibre Reinforced Concrete Properties. A. C. I. Journal, February 1971.

56.

Swamy, R. N. and Lankard, D. R. Some Practical Applications of Steel Fibre Reinforced Proceedings of I. C. E., August 1974.

Concrete

57.

Hannant, D. J.
Steel Fibres and Lightweight Beams Concrete. August 1972. Vol. 6.

58.

Samarrai, M. A. and Elvery, R. H. The Influence of Fibres Upon Crack Development in Reinforced Concrete Subject to Uniaxial Tension. Magazine of Concrete Research, Vol. 26: No. 89, December 1974.

59.

Taylor, R. and Burdon, P. Tests On A New Form of Composite Construction. I. C. E., Vol. 53. December 1972.

60.

Durham,

J.

The Tensile Strength of Fibre Reinforced Concrete MSc. Dissertation, University of Salford, October 1971

G1.

O'Leary, D. C., Dave, N. J. and Saunders, J. Steel Fibres in Partially Prestressed Composite Concrete Beams A. C. I. Special Publication No. SP-44-27, Detroit 1974.

62.

Dave, N. J., O'Leary, D. C., Al-Sanjary, K. A. A. and Saunders, Asbestos and Glass Fibre Reinforced Cement Structural Concrete Composites, Parts I, II and III

J.

Fibre Reinforced Concrete - Conference on Properties and Applications of Fibre Reinforced Concrete and Other Fibre Reinforced Building Delft, September 1973. Materials.

63.

Dave, N. J., O'Leary, D. C. and Saunders, J. The Structural Use of Fibrous Cement in Composite Concrete Construction A. C. I. Special Publication No. SP 44-29, Detroit, 1974.

64.

Dave, N. J. Limited Prestressing As A Means of Economy in Structural PhD. Thesis, University of Leeds, July 1967.

Concrete.

65.

Clark, L. A. and Eastwood, W. The Flexural Strength of Concrete Beams Reinforced Strength Steel. The Structural Engineer. July 1970.

With Very High-

GG.

Kollek, J. J. Fibres in Cement-Based Materials. Advances in Concrete. Proc. of a symposium held at the University

of Birmingham,

September 1971.

67.

TAC Construction Materials Turner and Newall Limited. Products Manual Lytag

Ltd.

68.

Lytag Structural LYTAG Ltd. ,

Precast and Refractory

Concrete

69.

Nesbit, J. K. Structural Lightweight-Aggregate London, 1966.

Concrete

70.

British Standard 3797: 1964. Specification for Lightweight Aggregates

For Concrete.

71.

British Standard 882 & 1201: 1965 Specification for Aggregates From Natural Sources For Concrete.

72.

British Standard 812: 1967. Methods for Sampling and Testing of Mineral

Aggregates,

Sands and Filters.

73.

Attisha, H. P. High Tensile-Steel As Normal Reinforcement In Concrete PhD. Thesis, University of Salford, May 1972.

74.

Baker,
A Plastic Concrete Magazine

A. L. L.
Reinforced Theory of Design for Ordinary and Prestressed in Continuous Members. Moment Re-Distribution Including 1949. June Research. Concrete of

75.

Regan, P. E. and Yu, C. W.


Limit State Design 1973. London, of Structural Concrete.

76.

Rowe, R. E.

Ghncepts in The Design -New

Granston, W. B. and Best, B. C. , Concrete.


of Structural

The Structural

Engineer,

No. 12. Vol. 43.

December

1965.

77.

Anchor, R. D. The Application of Limit Concrete, March 1968.

State Design.

78.

Huges, B. P. Limit State Theory for Reinforced Pitman Publishing, London 1971.

Concrete.

79.

Baker, A. L. L. The Ultimate Load Theory Applied To The Design of Reinforced Prestressed Concrete Frames.
Concrete Publications Ltd., 1958.

and

80.

Mayer, li. Bauschaden Als Folge Der Durchbiegung Von Stanhlbeton Bauteilen. Fur Das Bauwesen Der Technischen Report No. 68. Materialprufungsamt Hochschule Munchen, 1966.

81.

Beeby, A. W. and Miles,


Proposals Structural For The Control Concrete.

J. R.
of Deflection In The New Unified Code For

B. R. E.,

May 1969.

82.

Branson Dan, E. Deflections of Reinforced Concrete Flexural A. C. I. Committee 435, June 1966.

Members.

83.

Yu, W. W. and Winter, G. Instantaneous and Long-Time. Deflections Under Working Loads. A. C. I. Journal, July 1960.

of Reinforced

Concrete Beams

84.

Beeby,

A. W.

Short-Term Deformations of Reinforced C and C. A., TRA 408, March 1968.

Concrete Members

85.

Srinivasa, Rao, P. and Subrahmanyam, B. V. Trisegmental Relationships Moment-Curvature Members. A. C. I. Journal, May 1973.

For Reinforced

Concrete

86.

A. C. I. Committee 224 Control of Cracking in Concrete Structures. A. C. I. Journal, No. 12, December 1972.

87.

Reis, E. E, John, J. R, Mozer, J. D., Bianchini, A. C. and Kesler, C. E. Causes and Control of Cracking in Concrete Reinforcedwith High Strength Steel Bars. A Review of Research.
University of Illinois, Bulletin 479,1965.

88.

Base, G. D. Read, G. B. Beeby, A. W. and Taylor, H. P. J. , , An Investigation of the Crack Control Characteristics of Various Types of Bars in Reinforced Concrete Beams. Research Report No. 18, Part 1, C&C. A., London, December 1966.

89.

British Standard 1881. Part Methods of Testing Concrete

4;

1970

90.

Orangun,

C. O.

Influence of Properties of Lightweight-Aggregate (Lytag) Concrete on Behaviour of Reinforced and Prestressed Concrete Members. PhD. Thesis, University of Leeds, 1963.

91.

Hajnal-Konyi, K. Tests on Square Twisted Steel Bars and Their Applications of Concrete. The Structural Engineer, Vol. 21, No. 9,1943.

As Reinforcement

92.

Swamy, R. N. and Anand, K. L. Influence of Steel Stress and Concrete Strength on the Deflection Characteristics of Reinforced and Prestressed Beams. A. C. I. Publication. SP43-18, April 1974.

93.

Sparks, P. R., Menzies, J. B. The Deflection of Reinforced Concrete Beams Under Fluctuating With A Sustained Component.
The Structural Engineer, November 1973.

Load

94.

Abeles, P. W. Static and Fatigue Tests of Partially Prestressed Concrete Constructions. A. C. I. Journal, Vol. 26, No. 4, December 1954.

95.

Bate, S. C. C. A Comparison Between Prestressed Beams Under Repeated Loading. I. C. E. Vol. 24, March 1963.

Concrete and Reinforced

Concrete

96.

Garwood,

T. G.

The Flexural Behaviour of "Class 3" Post-Tensioned Beams Using Various Types of Untensioned Steel. PhD Thesis, University of Salford, August 1972.

Prestressed

Concrete

97.

Snowdon,

L. C.

The Static and Fatigue Performance Deformed Bars B. R. E. CP 7/71, March 1971. ,

of Concrete Beams with High Strength

98.

Stevens, R. F., and Bryden-Smith, D. W. Deformed Bars In Concrete. Deflections of Reinforced Concrete Beams. B. R. E., No. 225/70

99.

Corely, W. G. and Sozen, M. A. Time Dependent Deflections of Reinforced Concrete Beams. A. C. I. Journal, Vol. 63, No. 3, March 1966.

100.

Hainal-Kony, K. Tests on Beams with Sustained Loading. Magazine of Concrete Research, March 1963.

101.

Lutz, L. A., Sharma, N. K. and Gergely, P. Increase in Crack Widths in Reinforced Concrete Beams Under Sustained Loading. A. C. I. Journal, Vol. 64, September 1967.

102.

Soretz, S. Sustained Loading Tests. RILEM Symposium on Bond and Crack Formation Stockholm, 1957.

in Reinforced

Concrete,

TABLE

(1) DETAILS

OF ORDINARY AND COMPOSITE BEAMS (STATIC TESTS)

Steel Reinforcement Nominal yield or 0.2% proof stress N/mm2 275 Applied
working

Beam

mark

Type

Number

and size

Percentage

Efective depth mm

Applied
ultimate

moment KN. m 12.8

moment KN. m 20.5

ST 1-0 ST1-C ST2-0 ST2-C


ST3-0

Mild Steel UNI 410


UNI

2-16mm diameter 2-16mm. diameter


2-16mm.

1.044

257

1.044

410

257

19.31

30.9

1.044

550

257

25.61

40.97

ST3-C ST4-0 ST4-C ST5-0

550 KAM 60 KAM 90 Lanes 60 Mild Steel UNI 410 UNI


550

diameter 2-16mm. diameter 2-16mm. diameter 2-16mm. diameter 2-20mm. diameter 2-20mm. diameter 2-19.05mm
diameter

1.044

590

257

27.31

43.7

1.044

875

257

38.43

61.48

ST6-0

1.044

410

257

19.31

30.9

ST7-0 ST7-C ST8-0

1.643

275

255

20.04

32.07

1.643

410

255

29.15

46.64

ST9-0

1.483

550

255.5

34.68

55.48

ST10-0 ST10-C ST11-C

KAM 60 Mild Steel UNI 410

2-12mm. diameter 3-12mm. diameter 3-12mm. diameter

0.582

590

259

15.75

25.2

0.874

275

259

10.75

17.2

ST12-C

0.874

410

259

16.44

26.3

Concrete Section: Nominal

150mm width x 300mm depth. = 50 N/mm2

Concrete Cube Strength

Cover to main reinforcements

= 35mm

TABLE TESTS)

(2) DETAILS

OF ORDINARY AND COMPOSITE BEAMS (FATIGUE LOADING

Steel Reinforcement Nominal yield or Str proof ess N/mm2 275 Effective depthApplied working moment KN. m 12.8 Applied ultimate moment KN. m 20.5

Beam mark

Type

Number and size

Percentage

FA1-0 FA1-C FA2-0 FA2-C FA4-0

Mild Steel UNI 410 KAM


60

2-16mm. diameter 2-16mm. diameter 2-16mm.


diameter

1.044

257 .

1.044

410

257

19.31

30.9

1.044

590

257

25.61

40.97

TABLE TESTS)

(3) DETAILS

OF ORDINARY AND COMPOSITE BEAMS (SUSTAINED LOADING

Steel Reinforcement Nominal yield or 0.2% proof stress N/mm2


275

Beam mark

Type

Number and size

Percentage

Effective depth mm

Working moment KN. m

Ultimate moment KN. m

SU1-0

Mild

2-16mm.

1.044

257

14.95

23.3

SU1-C SU2-0 SU2-C

Steel UNI
410

diameter
12-16mm.

1.044

410

257

21.45

33.7

diameter

*includesself.

weight of beam 150mm width x 300mm depth = 50 N/mm2

Concrete Section: Nominal

Concrete Cube Strength

TABLE

(4) MECHANICAL

PROPERTIES OF f. r. c. UNITS

(TENSILE AND

BENDING TESTS)

Direct Sample No.

tensile

tests

Bending

tests

size of sample (250 x 50 x 6) mm Initial tangent modulus of elasticity KN/mm Maximum tensile stress N/mm2 Maximum measured extensibility 6 x 10-

size of sample (240 x 30 x 6) mm Initial* modulus of elasticity 2 KN/mm Modulus of rupture N/mm2 Maximum measured deflection mm

1 2 3 4 "5 6

11.75 12.00 13.50 10.75 11.25 11.80

18.40 16.66 16.12 15.60 14.93 14.50

1860 1760 1760 1700 1300 1250

14.47 12.24 11.13 17.23 14.65 18.10

37.00 29.40 29.40 35.80 33.70 33.70

7 7 8.1 6.2 8.15 6.15

*Obtained

from

the equation,

deflection

23 216

M EI

L2

TABLE

(5) TENSILE

PROPERTIES OF STEEL BARS USED

Size of bars is 16mm diameter

Type of steel

Nominal yield or 0.2% proof stress N/mm2

Ultimate tensile yield or 0.2% proof 2 stress 2 N/mm stress N/mm 280 465

Observed

Fracture stress N/mm2

Modulus of elasticity KN/mm

Mild

steel
Uni
410

275

360

200

410

425

581

476

210

Uni 550 Kam 60 Kam 90

550

590

712

515

225

590

655

925

911

200

875

870

915

" 890

225

TABLE

(6) PROPERTIES

OF CONCRETE

Beam mark

7 Days cube strength N/mm2 45.6 45.6 43.2 43.3 42.0 46.2 42.7 44.4 43.0 46.6 45.0 51.0 44.6 43.0 42.5 53.0 48.4 46.2 45.3 44.7 46.6 45.4 41.2 44.1 -

Pro rties at Testing Time Age of Cube Cylinder Modulus of beams at strength elasticity strength 2 2 2 testing N/mm N/mm KN/mm time Days 71 34 35 34 75 35 60 41 53 35 60 37 43 68 43 35 45 44 40 41 48 30 37 31 37 42 35 53.1 54.2 52.0 52.0 50.2 52.7 47.0 54.6 50.7 47.0 46.7 54.2 48.6 47.5 51.0 58.5 53.0 48.2 49.1 53 51.2 54.6 50.9 59 58.0 57.0 53.0 43.0 47.5 42.0 40.5 37.8 40.8 39.0 45.0 40.0 42.0 40.0 42.0 32.0 45.0 48.6 44.3 38.6 40.8 53.0 43.6 38.2 46 44.8 42 41.5 18.5 17.4 16.2 17.6 17.2 18.3 16.8 18.2 17.7 15.2 14.0 16.2 18.8 18 19 16.8 17.2 20.0 18.8 17.5 19.2 19 18.4 17.4

Direct tensile strength N/mm2 2.8 1.70 2.66 3.00 1.40 2.80 2.46 1.81 2.80 2.70 2.40 1.13 3.20 1.8 2.72 0.84 2.34 2.50 2.2 2.1 2.20 0.98 2.00 1.14 1.06 0.94 0.96

Modulus of rupture N/mm 2

---Ar-dry um rof density cycles or Kg/m3 days

ST 1-0 ST1-C ST2-0 ST2-C ST3-0 ST3-C ST4-0 ST4-C ST 5-0 ST6-0 ST7-0 ST 7-C ST8-0 ST9-0 ST 10-0 ST10-C ST11-C ST12-C FA1-0 FA1-C FA2-0 FA2-C FA4-0 5U1-0 SUI-C SU2-0 SU2-C

3.00 2.50 2.36 4.20 1.72 3.36 3.24 2.80 3.50 2.55 3.16 1.4 4.00 3.74 4.00 2.0 2.15 3.5 3.56 3.30 2.28 2.66 1.60 2.22 1.90 1.78

1760 1820 1820 1780 1750 1840 1700 1820 1750 1730 1680 1800 1760 1750 1780 1820 1780 1860 1790 1780 1675 1826 1800 1780 1840 1860 1825

3 3 3 3 3 3 3 3 3 2 3
3

cycles it if if it it it
It

it it
It

to

3
3

it
is

3n 3 3 3 2.96 x 106 3.0 x 106 3.0 x 106 3.02 x 106 3.04 x 106 1' 543 543 528 528 Days " " "

~j. tfJ Of

OC M m m

NO O . O

O .d iV

IP7 CO C)

4O

00

N Qf CV d) 00

M NN OO

N N (D

rl

Cl Cl .t

Co Cl V -1

. -1 M OO

. a N O

1! O C7

CD t0

O M

NN OO

O N .1

r1 N

N O

N
m p' O , .
aa VI bN Cl CO N T to O) M Of N W O) N . -1 Co MO t0 W Cl O) CI) CD

NM C) CD OO

Cl CD

M CD

OO

OO

OO

OO

OO

Q) Cp " i G.
U

"1

o) Cl
Cl

N O) "I"!

CO tn OO

. -d .t ii

Cl . .1

N r t

Q) CD M Cl

M 0

A O

CO N O Cl r/ . -i
M .iO

tn CA

CO M . -i

m 10 d m in O O

M N N
VI

IO CO
eI

N ICf CD

IO b yI CD

II)

N sll

Mw

tn

u2 u)

-e Ie to t0

n a

., t0

Co Ig

O t0

N CD

Nr M

ti
e4

N el l

ve 29
0 O

qq

C) .r C
, j CI

m +_' ,.. i
AM

NN
MM

cl

Cl

CD d

ICI M .A tO Co Co
It) O

NM CD Oi
10 M

O m

vm

0 O) MM

N 4 N

Cl Cl
CD

Cl
4 l ee

Co CO
en

CD Cl
CD to

N CO
Il! N

cm CO
Uli 'dI MM

t0 N
CD N

Cl

M N N M

y p

m '" a +c E 'I cO q 4 p Co

Z% . N
c0 fD

O N 0 Cl

O Q) 0 M

23 Of 0 VI

O N M er

Co .M -I . t0

O W Cl

N O Cl M

le t0 CD N

Co

w
Il IO

O N Co Cl

0 N t` 1.1

0 M to Cl

CI p
m "'I ym 0 Cl 0N .1 to Cl M Cl 1 1 N O " 1 w C

4
j 4) F
Ud.

to .,

.,

..

..

..

..

Cl cd

'Cl

O Co + le. O "+ O Cl . -1 CD CD .. 1 t0 MN 92 en NO MM

V'" d0 w

yd ,.\

tQ

. reD CO Cl
Cl N

"

a)N .c CD N .4

I l l ' "J' CD 1.4 M Cl

mz
N
W C;

Umm 91 w 5 -

to .1
1t!
P/

N 1 1
ob

t
.1

"

Co

.q
Um ^' m Cl

N. z

Em

MO
N .I

M .t

0N "O
Cl

Cl

OM ~O
M

O Cl

C]

MM MM

It)

yO
OM 1.4 .1

-e

Cl

C) Cl

Cl

O Cl

. y M

Ig
Cl

CO
.I

to Cl

3 le O
b

O
.1

M 00
. -I

M y I
.1 dl c

le M O O
..

d . yyw N O + Li 'W
1c-I eM O 1I) -1 'd l O

pd

O Co " O

O O

O tO
A U d1

M
eN m

O
.r

O Lo O

O t0 r+

+ i

a
O 1 tj. Dy : /1

p
OU 11 Nr N . N
O 1 C. Dt 4i O [ "! ] 1 OO [y "! ]

,
t .y

p
1 N F W

FQ

= ~

OU 11

OU 11
F /] tA

O U, I1
VI /]

OU 11
"/i

O 1 A I[

f/i UI

-1 "! ]

. I "/

MN
C r+

O p
CD "

N CM

ti t`

to M
" .

rf

r+

. -4

1%4 e-1 "

v-4 CC) "

M '

rI

,r+

In (D
m

U Cd V+
" N F
CO
y. y v

r"

m CO
Cy r-1

co O
t0 rf

CM CYi
O rf

M r-4 O rf

CM CO N rl

CM
r-4 rf rf

M M
M rf

E-1

`. t,

Cd

"
b-

.
0 tm \ C)

' aN

co
O

LMC)

to

LO

co NP

t0 O

co 111

co 1.44

w41 az U2 bo
C)

r.
C)

O] A

z w 0)

to co r-4

. Idl r -t

LO

CO C;

O
O

r -t

co

V
V;
rl

N
ri

O
r-1

w
4)

cD
O
r, 4

O WF i i4 M M m rf

Co
Lri C> r-1

r-4 rl

co rl

rf

La r-I

"-boi

V]

0w + Zi

t0 M co ri

t0

t0 CC> co

lf O l t0 r

LO N

ti N

CO ri

0
r1

to
M N

QU0U

bb
a) 0 10

'

CO

1-4 M

CD

ri CM
t4 N

N r-t

er O 8N

tA
r-

tf
t.`

eN
<f1

eM-1

10-4

r0-t

r0-1

U
1.4

U
N

0
C)

U
V1

U
t
r0-1

U
t
r-1

VN-4

ai

V3

UH]

to

U2

to

TABLE

(9) SUMMARY OF DEFLECTIONS

(STATIC TESTS)

Observed deflection (mm)

Calculated deflection at working moment (mm) Based on l experimenta curvature 4 6.8 5.5 13.2 9.65 20 14.4 18 21 23.6 12 9.6 9.7 14.4 20.4 15.2 10.1 4.2 9.7 Proposed th eor eti ca l method 5 7.93 5.6 12.53 10 17.33 15.2 18.7 18.0 27.42 12.53 10.04 9.2 15.15 19.84 14.4 9.2 4.4 8.8 Col 5 Col. (2) S pan d efl ec ti on (working moment) 7 592 833 369 469 245 300 256 245 165 375 469 500 296 212 317 489 1071 536

At Beam mark 1 ST1-0 . ST 1-C ST2-0 ST2-C ST3-0 ST 3-C ST4-0 ST4-C ST5-0 ST6-0 ST 7-0 ST 7-C ST8-0 ST9-0 ST10-0 ST10-C -ST11-C ST 12-C working moment 2 7.6 5.4 12.2 9.6 18.4 15.0 17.6 18.4 27.2 12 9.6 9.0 15.2 21.2 14.2 9.2 4.2 8.4

Remaining after first cycle 3 2 1.22 2.39 1.82 2.92 2.55 2.88 4.1 3.95 2.39 . 1.83 1.42 2.52 2.85 3.24 1.76 0.8 1.2

6 1.04 1.04 1.03 1.04 0.94 1.01 1.06 0.98 1.00 1.04 1.05 1.02 1.00 0.94 1.01 1.00 1.05 1.05

^O .y x W >+ O i`7 m a

N ~I

NI

N MI

m i9

M V

N O

m rl

00 N

ai rd<
V

6
Z m

,0

vU

Cm
0

0
O

41)1 " Co I
d e

N N

ca .

0 O

eh ..

a .1

.4 N1

CO

M 1

"d

01

.+

m >

OI 0

m "r

e'J N1

" NI

N 02

to .+

rl

to .r

V N

M N1

N N ""1

Zx..
Uf

43
9

ti . cc, pi ~ dU

a O
0

.1

a .e1
0

"'

N .yI 0

Co N

r1

eO

.t .1

a ..

a .11

-g

xmO O ++ m y
..

rl 0

m OO 0

Co

.dO 0

rl OO

.d

rl O

O O

.iO O

.i O

.1 O

ro
0 O
mN OO OO m OO OO V aN OO OO 00 O O Co O O

a OO
O

Co O O

a O
O

m OO

.4

N O O

d
W

.+m
0

OO

Oi O Z
41 5 d

0 F.

NO
./

..

0 0

m0
./

'

..

.+

N .e

d . +

r1 . 1

m .1

V ..

.O.. -_ __ _mN N1 . ___ 0-1 . 1 .

o m O

.+
O O

Ne .

OO

Q _ .0

0 e F'

a ..

Nw 41!

m 0

dl

f o

" X a.
N.
)O-1 m

z
t-i mN OO Na .1O m .1O 0 OO Q)
!N . .4 . "1

aN OO

N O

. '1 .

. -1

e"! O

E eo OO eM 0

eN O

0 .r

. )' ..

Co O

{. U

N O

.1

m .r O

. -1 CV

N O

N .1 O

NN .1O OO

.. O

le .1

Co .yO OO

0 u
eM 0

Co O

"'

.9
O

'm

eo

aND "T 09 ..

m M

OND

e7

.e

m eM MM

oNO m M

aNO m V

m eM P7

.t ob
M

at'
. -1

"r 0 O)

0 m

ma O
g

.y

eo
.. .1

ei

ed

.i . -1

p a U .7

' [! -

N
01

w
N . -1

Uw

O! .r

a y

a N .+ c" N

a N v3
O
1 Qf

eh
O O . -1 E V If

.r

N .1

N ..

OU
c4 11 .r
[

OU
N
!

i N
1! i

OU
11 M H iA e] Ul

OU
e1 f%]

1 d. V]

o1

10

O
! W

oU
N1 fA N V]

O { CL f/J

.i E l

U 1
. -1 .M

U
.1 (

ai

.+

f/7

1/7

VFJ

fg

% (2

TABLE TESTS)

(11) STEEL STRESSES AND ULTIMATE

MOMENTS

(FATIGUE

LOADING

Steel stress at working moment N/mm2 Type and percentage of steel first cycle 2 N/mm 3x 106 Col. 4 cycle Col . 3

Ultimate moment KN. m

Beam mark

'

(Calculated) l ues actua l wi of strength without partial safety factors 6 31.92 32.44 39.02 39.56 58.81

(Observed)

Col. 7 Col6

1 FA1-0 FA1-C FA2-0 FA2-C FA4-0

2 M. S 1.044 UNI 410 1.044 KAM 60 1.044 .

3 173.4 143.96 240.7 208 332.97

4 172.2 168.36 245.2 254.18 343.9

5 0.99 1.17 . 1.02 1.22 1.03

7 36.125 38.375 44. -7 50.5 65.625

8 1.13 1.18 1.15

1.28
1.12

TABLE

(12) SUMMARY OF DEFLECTIONS Deflection " Beam mark 1 FA1-0 FA1FA2-0 FA2-C FA4-0 . At working moment 2 7.52 5.78 13.48 8.82 18.46 Remaining 3 1.42 1.1 2.87 0.8 3.2 first cycle (mm)

(FATIGUE

LOADING TESTS)

3x 106 cycles At working moment 4 11.66 11.4 17.64 16.47 22.55 Remaining 5 5.3 4.6 5.2 5.6 5.4 Col. 4 Col. 2 6 1.55 1.97 1.31 1.87 1.22

cd zM U
cd a' U

. -- ci U

N T-4

OO

r+ ".

r-1

cV 1.14 i
r "1

r 1

OO

Co

co

q,
V-4 r-1

1 r-1

L C. ri

--

>' t+ w Cd 0 8 1; $4

.-+ O r+

0 U U
C11 "

II "

Cl! e-4

.a

C) tc
O

LO cq I .

di

c
C)

ti rf r-1 N
O

wm r-I N
O

,- O

E! cd 0 C) p

El

m 4
U co O
T"4

.0 4-b

y ., c'

co

N GV M OO

00 00 GV M O

d1 N O

C)
00 0 OO r-i r-1 ri e-1 rl

OO

QO

44

:3

. 44 VO r i e- I

00

N OO

T-1

C17 O

OO

C)

w 0
a) C) 0 0 " C) Q Q' O O .0
rM-I O

C7

cd
.. r

'' O

C7 O

C) U
C) +a V2

C) U M

00

Q)

O . O

OO ." OO

m
O

rl

U2

Q)
N C O.O

to r-I "" OO

rf

M r-1 W

a a H

x F+ Cd

oU '
r Ir

oU
Cc

o W
144

al

WW

WW

en

'
M o CY)

.'
C10

3 cq
S

Cd +

trO-4 1-1

m cli
NN

m5

++

to

an

t`

c> <g ri
r-1

c NN

ti .

CO
CD

tn

. -+ . -+

o0

""

rl eM

N" M

N d

. oO

m t ,o0 cd O

LD
G) dl

Co t
r-1

N d M LOO C+M

er

-dk
N CO Om C ti O ei C) CO . "'1 ri

'M e}1 C. r OO .-i CO

UU

-
""

M
t0 M "" MM

M
l0

Y}+

ra OO r"+

"" NN

P'M

N tl0

UU

"d a
g

-5 Cd O
fa

' O; C. "t3 U] a > t+ <U


CA

q..,

cd Oy,
rr.

.
cd
a ' r

'

ON O

CO

O eN O
ri

er O
r"1

ri

r-I

7-4 Oa (D
U .r0 Fa M <W.

. tA ..
U> a v . -4 j

. M
CD 1-1 MM

CD N

O L

(D

(D
t1 CG

'L7 Cll
an 02 O
*-1

1OO
" i- et1 r-1

4.4 E-+
oU

oU NN
U2 u2

CO2M

MN
LO

LO
N C

N d+
ri N

r4

'C1
p 0
. 4-1

*8
4 rn G"+ + v
no

CD

N
r-1

r f

P4

O+wp

-I

zs

lO

14

14

r.

0
,O O r Fi

s
0'Vr. +

T%
-C4 O O 14 d+ M CV N

wCO O to ^
. .. , CQ 4-b

LS

3- 4 Gy 4 . -.
N

rl

Cd

N 00

co

4d4

eq I: -

O 4-4
V Cd

O b A cd 10. Q

e-1 r-4

r1

N O

(D O

5 '0

9 Cd

O r-+ My

T-4 O

r-1

P4

o
o :R0

4-02UaanV m uznuzN
00

LO
O O O

m
N

., ti 0

U) g w
0

. Qc

; C
ri

c
M Lr-1 O

'cl %-,

p
Cd
0

o (D W

r' XaeV 02U.


uznuI{GL1i

Cl

4-3P-4

.4 .6.4

ca

Cl1 -d N ri

CO

-
x
Cd -0

-1.3 to
'0 .4a e. O
.

92V.I0nV
uznuxrey
.a

o
4 t;

rf
r'
o . o

C) :

Cd
0
U

0.
0Cfl o0 00

't3 >

rrr

Fy

ma i

02UJaeV uznuz}xmyg

o
N

00

* * c3 O o
r-i r-4

U
1-1 R

oU
NN

63 R

900

800

700

600
n Z
N

500

d
4-P

400

U)

300

O
E O

NO

E 0

O Ln Ln
"C

200
.J

er

l00

1. O

2. O

3. O Strain

4.0 o/o

5.0

6.0

7.0

I1iiI1 0

1.0 Kam 60 III 0

2. O

0 Mild

I. 0 steel

2. O

I. O Uni 550

2.0

l11 0

1.0 Uni 410

2.0

FIG. I

TENSILE STRESS-STRAIN CHARACTERISTICS OF REINFORCEMENTS USED

W tW V Z
LL WO

OOU
Q

La IF-r- :"2
.... d WWZ
N

E Y

vv

- d

'c

-C

Z~O

LL

>vZ JZQ

O 0

II

>WO
w

Z Wp

3u F-- O W=tm

cD o z-Q w 030

QT iti J-W .... _ OC J J ..

N 0 10 0

Ln
o

vo
U+1

0 (Y)
34613M UI

0
6uinnS

U-

--'d

-I

3IN
-T

s+

n.
VV NC gn , KY 0

er w f+

9
E C d .n

U) N

O V E E Ln n

tn W
iZ

JIM
E O d .G L. w

E E Ln N.

Z v

0 _1
w L9

31 cm
- -! +

O w L. LL W.

F- a

aw wm
t U c d . w O
N f+ C d

aw

E E

UO

0
J' c)

0 in dII

0 -4: --

Q.
O tn U On

er v U, v 3

.Y V

E od .- vw v o ov E

LJL.

Na
z
wQ ?

31 04
N Y
.+

v
d r.. er v
c. V

:2 Wp

C) cr Q0

Jjc*)

n.
V Y N Nd YC

0 1
1

1 E .o

OC p

LL

ti

Uo rv ,."
EE
_1 3iN L N1N N

ch
I ^"

Ln

lJ.

E E

ccx oO Et 00) V
.G V

.o

O ci `N lmm

r. - c - O O +a , E er evr O v1 uwI
. y y

13 Z

0 o LL

ac
V
3

z 0 cc

3
3

0
-o
Hp Za W>

CL

C a N

v
3
I

L 'v E c. 0

E E

2W W - V,,, cc

'E E O 0
i+

0 in N
E E 0 0 Ln

z
WQ m
Ow
JQ

h IO n. O .6..

O
N

C d
E d
L

0 .
L I"1 VI d d N UI 3

C!

O .4C .
L

Qa
wO

er
4. YI

er

C v

o
It

-Uc CA G G

E E o I N

E E 'o

I
:3

0
LL

I I E OE 0

----

iT

U)

ul N

[I.
-

ww OO

y
I

.
E E 0 Ln
E . -t-

- d
f+

E dz

ca

a
}-

to .
E0z
0

a z

ac
e

LA-

N L

. G. O 4' V V ++ W

i
.

N G. `C

;,

i
`

?W
ww SE
Cp C

o E

Z;

E
e

ad dV dNL
dO

Cx rO u0 vx L: O r-.

H Z, Wu)

4, E u+ Ec

20 w U.u Otn
Z
w cc in U.
OW J0
E

E E 'O

O O Z vc 12
E~
O

ww SE

i i I.
ww O0E

"I.
-Tx

- _1
L/ .

E E tn CV)

a
H2
W0

E E 0 Ln

0 b
.A

i ri
i

. L

0 C
'C L

Ln

6
LL

fcu=50
40

N/mms

Q=0-67f 0
C4 E

cu

E 30 z
20 w
w r

Parabolic

curve
rfCU

N 10

r
(ci)
N

-C

F_-1

77 "371f_..

ej _
1I1I O"002 Strain e

2510

= 0.00282

-- Y"V. u t; 0"003 0"0035

1-0"001

Theoretical

ultimate
0.77

design stress-strain
fcu

curve

E E

I%ft40 z
N N

d3 r
44 L. N

v c u
d >

20

N 10
E
0

a.

U0

0.001 (b)
Experimental

0.002 Strain

-1 cj O"003
e

i1

O"0035

stress-strain

curve

FIG. 6

THEORETICAL AND EXPERIMENTAL STRESS-STRAIN RELATIONSHIP FOR LIGHTWEIGHT CONCRETE IN COMPRESSION

Parabolic

curve

0"67fcu _ va = ym

i
b
N N L i)

_ _ J'
/. 5.5 ^ KN/mm2

4100

f
J

ym u ,

Yc

Strain

0.0035

(a)

Design
concrete

stress-strain

curve

for

normal

weight

(Ym'5)
fy/ Ym Tension

0"8

fr/ym

b
N N N L

200

KN/mm2

0.002

Strain

b)

Design <Ym5)

stress-strain

curves

for

reinforcements

FIG-7

DESIGN STRESS-STRAIN CURVES FOR CONCRETE AND STEEL AS PER B. S. CPIIO

N
++

N d

*p

E E Z

E E

- p G. Ov oZ

v 2f G.

c.
E 0 LL t! '1

q
0 "M
J

n q1t 4
-i

N `_. o
N..
LL

III
WL r

0 '0

0 N

O 0 N

OO ao

'o

0 It

O
N

Pool O .O O
N

n. O ac a J x w J

LL

00 "
I. dW
0, o O1O a Q. .J Z

O Oa co N

xy 4=c O
4+ 4'' `p er 01 rn

FFV UJ

O`
O d'

ti i o It

m
lL

0
W /N LW

Co

14o

sl'

ssaJ3s allsual.

E tn nE

ddd

s. +

CA
b f+

E r1 OEx0 rn WZ
I \ N d(

NE E c!

v
v b
L

'E

v
0
b
N

OE

O "E

L 0

OE 6E

L1 W

Ln D

` ` ` t` \

I1

OE r= tn
er N

Iko E cr)
C

> d

NE

v)
C6

LL 0

C7
N

0 Z

t7 er
\`CC

ED E _E

0 rn er ..

0C

01-

016 6
Yf OC

LL
L

0O U

Oo Ln ,
u. E

0 0

0 a

0000000
co

%O

to

It

cq

t, u OE

L. Ln 0

6uissnd

a6nIu",

t3d

E
O

E O .n

.a
O
C

N 0

F-

z
w LL :2 0 0 J W w W J J W

E
0

CL

aJnl!

Dj

iD

y3dap
I 0 . I

sixo

jojjnaN

a a
w

tl)

Y ob O

Fz W u1 1u.1

W in w htn 0

. e. _

"' N

. C r

v
M a Y

f
U p

U 4J N

v
V C

E 0 E
V N

u 0 ` U

OO

+' H

a_ 0

a _
0 a U
a z Ln }-

E 41

w 0 z

. C
E E
T7 Y

J W J c 0 *Z c v t-b a'
N VI to ob

W J

Cl.

z E D cc w z 0
w F-

v .
v
b Y

w in

cc 0 LL

4_

I\
-6 ' C U X. vv
?CxX

0 6 EL

lp/ x= u

41dap

3ni132;;

3/4ld3p

S1XD

lojinaN

.+

Elastic phase 1'0

Non

elastic phase

er

7
dc ul., er

vC
E
40

4. '

C d

E 0 E v
G. C36

d? u E
-a vxiE

---

Ordinary Composite

beams beams

Eda..
0

i
Values and

emax Concrete flexural compressive strain cc

of ep for ordinary beams composite

i
i

(ci)

between bilinear Simplified relationship flexural compressive and applied moment for in and ordinary concrete strain beams composite
concrete Cracked fa b Of a concrete

C to

Uncracked T lip

E O E v d HO fd

CL C36

UT

2
C7

Stress

at soffit

fob

(b)

Idealised relationship between applied and maximum stress in f. r. c. channel

moment

FIG. II

WITH APPLIED MOMENT FOR RELATIONSHIPS STRESS IN f. r. c. CHANNEL AND FLEXURAL

nE E 2
V 40
M V a. r M V

ao=O"67tcu

tmax=0"0035

Parcbolic

curve

M V

0. E O

60 6c_1 ei <

c2c 2cj

-cu irf ei = 2510

u
Strain CC for

1
ei cmox

(ci)

Stress-strain relationship concrete in compression

lightweight

o"s
0.7

tmax ei

O"6 cl
3c-CC

O"5

3cC - cj
G=

a= 6e-3ec

3C

i1

'A O"4
O

>

O"3

4th-C
`-

A-

12tj-4CC

O"2

6c-4ct

-cl

s= "
iiIIIII1 0.2 0.4 0.6 cc/ci 0.8 1.0

12CC-

4! cc

O"I L 0

1.2

! "4

(b)

prediction

t)f the coefficients

a and

FIG. 12 CALCULATION OF CENTROID AND AREA . FOR COMPRESSIVE STRESS DISTRIBUTION DIAGRAM IN CONCRETE

Top level (interface and

of upstands between concrete

f. r. c. channel)

i
d

Centroid of stress in f. r. c. channel

distribution

FIG. I3

ASSUMED DISTRIBUTION OF TENSILE STRESSES IN f. r. c. CHANNEL FOR COMPOSITE BEAMS

O cm
IO v..

er rn c v

r f bli O

I1 II
i' I
-

ca

i
11 11
r/ --/
71H

o fwW
i r. Y

v v
4. '

OV' i

UU

D
.o d
N

O cr
UU

za

00 LL
>-

LL Cr
t! ) Z

W
W
11

O a E O U O "r-

/
/

I-I

F- _
WU . t0W t/?

-1-00

0 1-

II

E
Q .. O L

0
0 C v ` O ` 0 w
1
LA

-\
n N

N r\

v 3 -v 0 2

1/'
//A -/ I

d' d' O II

J_ WX

cn tn
W

;Q FU')

LL

E +
N

/-w-, /

E f n.. E

l /
/ I

Ili

fl

I IE nCf + _
LL

+
'I; '.

E
a.

'J "1 ; i//


1

i
0

O --t O Ln FO
Z
LL1 U

\ 0

0 U. Z
LL oW H V } cc Z

co d

n a.

+
ii

wJ LL Z Gr LL wOO

E U
C

"
'C

0 0
u

I
Al

I
In
s1xD
1.0

43d3p

cY) 6

LL

3A13D3;13 01 41d3p

lo. i3nau ;o

oi3oa

LL

O
J
W > W Ln

JZ

ZO
Ln W
Z DW

O }_ V

V)

r 0

L0 ,+- 11. :2 LL 00
I--

>-

V Q JV
W

VWY

cc
cc

LL 0

0 cn 2 DH

Ja Dw 00
ON
mX
LL Q

OJ

a U CJC
W 1WZ
LL D LL W

In

41d3p

3nI333;;

3 0-4 41d3p

sixo

jo. i; nau ;o oi; DU

LL

Composite Ordinary

beam beam

behaviour Probable beam for composite

Deflection

()
Ma

load-deflection Typical and composite ordinary-

behaviour beams

for

MQ2Mc
4'

/ _Mp
2/ 0-1 1.40

l*I

401 wo,

c Y
E O

Mc

.010
---

wo

Composite Ordinary

beam beam

behaviour Probable for composite" beam

Deflection

(b)

Idealised moment-deflection beams for composite

relationship

FIG. 16

BEHAVIOUR LOAD-DEFLECTION FOR ORDINARY AND COMPOSITE LIGHTWEIGHT CONCRETE BEAMS

c0

0u -r
t ei IA
c d Y

-x
D

u 0 ` u

0
Q

W
U.

0
'v .Yd +J U
FV W J LL LtJ in Ltl _ FCC 0 LL
1 '0 'Y UN

:;

o
u=V d N

ru

ch

I>.
:2 (j,

Y U

c.
>.

U
_d O .+0,, OUL

W N
>-

J ZQ
QW

CL au `

CL
U
t! V . O '-

N Co zW O
er
C

C 0 *Z u
N

d d 41
U)

G O t V
L

wO a
Ln O v 0
/

v
Y

.a

M:

u uc
.. _

IX uV0

1/2
x U

0
U.

O d er

Neutral level rb
}
I%.

axis

I
''--

-
1

_ _ " _- -----

Steel

level

(a)

Elevation

of a cracked

beam

Average,

tensile

resistance

of concrete

fs 3,

fsavc

(b)

Distribution of steel stress along a cracked beam


Neutral axis level

I 10

QI Average tensile in concrete

Steel stress

level distribution

(c)

Stress distribution at a cracked concrete section and uncracked

FIG. 18

OF STRESS ASSUMED STRESS DISTRIBUTION IN STEEL AND CONCRETE ALONG CRACKED ORDINARY BEAM

f. r. c. unit concrete

f. r. c. unit concrete

qw,
--

_i
---

--

----

(1)
. __,
'4 .4 .4 4

Longitudinal section (general arrangement)


-Q--T

Cross-section

1bl
al--------

lb

E
formed

(2)

Micro-cracks

-11

-- -- -11-II
ll _IL_
cracked

f---4

--1L(3)

Concrete

(partial

failure)

-7 !_-

II II _JL
(4)
f. r. c. unit cracked

--low

(complete

failure)

FIG. 19

CONCRETE PRISM CONFINED BY TO DIRECT f. r. c. UNITS SUBJECTED TENSION STRESS

_ ISO Mm .

200mm

IOmm

lo mm

_: " 't--1
-F--r

1C
tiM i` r I _ E

I "

r0
_1

0
M

E E 0 ... 0

E E In O P C v O 0

_i_:
p
E E

In N

O
M

Elevation

q V

O' C

Cross-section

v. a n

U 0 a M
h

(ci)

Ordinary

beam
06
150 mm lo mm

200

mm

10 mm -

I
E E O O in

E E O .0
M

E E O M a
0 VI C V O 0 0

E E vf

II N1

II

0
p1 C

N-

-8

o, C V 0 d M

0 Z.

4---

25mm

25 mm

Elsvation

Cross-section

04

Ul

(b)
FIG. 20

Composite

beam ON ORDINARY AND COMPOSITE BEAMS

DEMEC LOCATIONS

WW

LSZ=

(P

I
i.

C_ O

C
E

E E O In

0
O

ww 9"6SI=

onX

E E .o 1 cv

C 0 u

to up

WWOOIcGD X

E a to
In

_ Fa
2 NQ XW Qm

I
, Or4x

a
0 0
Co n u
ax 11 Qx x .. <0

J >" Q CC a ir ? FDp W CC Zuo W

Z ZW

C E O E Y

OOO .o

v v Y IO Y L H

0 E
M

" 4

,qt tn 'o f-i ~dv ts


04 x E3Mn.

41
E vWn.

a
ar

WQ cc
QJ

"0
Vf

"E
00
II
MN

-OQ tl
0

C E O E

d 6

Wf---W

cc

2
v M N

a3
N (7 LL.

d d

0"a

ko -g" 0

v >

0 I

tT

.O

1 CO

0600 41dap 3A1123113%41dap

sIxD IcJinaN

Applied
0 0.2

moment/Ultimate
0.4 0.6

moment
0.8 ISO

150 mm

o"
0.2 O"4

I
E E
E .E t 0

Theoretical I ST9-O
jX..
JCS/TN U

40
1. . _ v X--X-X O fO

f 0
x

0 u0

O"6 0"8 ' dx I. O

5-

vExperimental f ST7-0 0 ST9-0 STe-O

_I

.u, d
W

of reinforcement 2-20 dia. mm bars

'

Beam section at mid-point

Values v

at

M/Mu=I

were

calculated

d. O
EE

Applied 0"2

moment/Ultimate 0"6 0"4

moment 0"8

150 mm 1"0
V

,
E

Z 0"2

"

Theoretical

O"4 O"6
O"8 -p

iv

/v

vv

vv

v--

II

m-L nI
n0
u

L;

11
! -t

x
Experimental V STIO-O

i"o

of reinforcement 2-12 mm dia. bars Beam section at mid-point

FIG. 22

RELATION DEPTH WITH APPLIED (ORDINARY BEAMS)

PROPOSED

OF NEUTRAL MOMENT

AXIS

WWLSZ=lP

6.

C O a E

.0

E E O Ln

0
E

onX 9"851= ww
hf
WW 601= 33 X

L_ ..
-0I 0

10 N

C 0 :+ u
N

E
O
m

~1

aN oQ Nw
X

co

L U

d c C 0
U

Qw QN Q

x "

f-- a D w
zO LL

x x
M

G b

4
VU I) M F- FV) N

V V
t7 tl
.J

E 0 E
b"
"

zz W O - O
J Wp cc w

1o

O d
d u :.
O

c E
b

O u V i.. V

E D
M

"Gx
to

E s. vU
x1 W hU) FU)

C b

E O E" b
d

u 0 cw . 0

II

i--

-H1

Wa Na Oa a= O ac a3
CV)

oJ

ci
N

OS
x

Cl.

0
N

lL.

0
9

0 U
t

X 10

14

tt

0.:

%0

0000=
41dap 3nil30jl3/41dap

en

sixfl

In-J1naN

Iww SSZ=P
o- c .ZW uam tz
>C

E E 0 Ln

nX ww 6"Z91 =
E9

W F_

L. .

IwWIVZI =X
N L

w O .,r,

U)

: I--"
c

O .0
b C C O t U U I.4-

zo
V to 4-b O V

co E 6
0 E or 4-+ O O E
4--$

.
d
U.

6I
"I
oU "" c .`
d Xs C b

tL

.0

O
zz

E E

0
N

E
b

O O
J W W

+1 d L
O or

U
L-

0 y.. C L

d
II

",

C. d0 E
0 E '0

F-

e 4
N

d.
0

i
ItI
CV It
'4O

M. 0
O

Wa in a a a= 0 a3
It N LL

.J

01

. .

0
4ld3P

... OOO

L. . C0

ani3: ) aI13/41dap

sixD IDJ; naN

o=

Applied 0 0.2

moment/Ultimate 0.6 0-4

moment 0.8

150 mm

0
0.2
O"4 Theoretical

i"o

kv-v-o

E E o In cm
'C7

10

O"6

V
O"8
I-0

Experimental V STIO-C

of reinforcement-2-12 mm dia. bars f. r. c. channel


vzzzzz

u
W .+ a
o
"X

Values

at

M/Mu

=1 were

calculated

Beam section at mid-point

Applied

p
.J

0-2 I

moment/Ultimate 0.6 0.4


III

moment 0.8

i"o

O"2 O"4

Theoretical
X0

x x

x0
-.

xx 0 x--0 00 --

O"6
O"8

Experimental STII-C X STI2-C

i"o
Beam section at mid-point

FIG. 25

PROPOSED

RELATION

OF NEUTRAL MOMENT

AXIS

DEPTH WITH APPLIED (COMPOSITE BEAMS)

It

W m >a`
o >
Vf N L

Wt
qt

z E ac
z

a E 0 U

O
1-

z
.+ Ln

Z Y
.o G O J

H
J
LU J

d vi c d F-

Li.

IkO N 6 LL

W C11

"N

>a
NO

to

w O

ii

4-b

re
D.
X W

U)

a
U) C d F-

--- LL

rllN

6
LL

d >
N N

d a E O U
L

CM
fY)

O 0

b v L

0 0 It N

u Cd

1 00
EN
xo OL

x1

0 0 It N

a 0 cc 0 1,n\ %0 tn wQ
p>w

0 0 o . og_ =0pOW
,emW N

U) V) W LL a

CO } a

%lO
FZ W

0 0 CV
(V)

0Z

0o00
V2 LL p

0 oOw 0 0 Nr N Co 00 0 00 CV) %0 0 1 N
H N

z
Q >

LL

O 0

't N

O 0 ~IOD

0
0

CC) N 6 LL

0 (I)
9
Co %o e cm

0 0 OD

Oo00 zuawow aInwI31f1/ 3uawow pailddd

ei Or b >
Yl Vl N0
L

K 0 Ed

1In

E a

E o ._
t1

O1

f..

JO
1K

10 N 1O
NN

z
Q

Ed

01 O
V^IN
F1

IO
-IV

Ex

-10
M

F-cl "' 2
. _f

.n 1=

0 0
F-

xO O c.

w ->

114L W

en

O0t!
N OD

x 0 Ed
O 1 Q I-'I)

`_

00
0
..

1M
O

00

0-

o
W F-

mZ ...
., v a-

W UW

K 0 Ed

. N m

lo I_00 "U 0- n
I, -v

x a Ed

1N

0 1 CD

I
"r

v -I

Io-%
.d.
N

lo

I e_ LL. o Ow
-. 1 Q.

It

INC

w4

0 N
FU)

00 00_ M-a 00
N

_a Q I-

F>
O

0
N

OO LL

0 I'0
U,

O 0

CD

n _OOO

co

'4o

4uawow aiowi31n/4uawow

paliddy

or >
L4 W or CL E 0 C) d d V C'40 o' V0 E

O 0Z

00

in aW
CO

:3 Ec
xO

C) !i

0 0 1N ..

cn w

IM i
cv

w
aO
On

CL
:2
V

m
!ui

0 O
N cr)

OOdZ

Z
O 0 It N

O0OU co LL

w
J

0 0 N
(r)

00 o 0 CD

O H

a.

U I CV IN

0 0
N

a I oC H 3 >

0 Jo

%0
U 1 IU)

0 c LL

0 0 CD

'O

_0
3uawow

O.

N 6

aiDwiiin/juawow

pailddy

VI

qo O
x O Ed

or O

d 0 >

d >
a tA L

x O II x

0 V
d L d N J2 "1 d L O d t I

E oZ U-V
d 41
U
G O
0

(V M O ..

Q \

O
0

x d Eor

or
- i

U
X

1O

o Rt fV
O P1
VN

ww
cn

cO

w
tlf

0'V I N FN

xi

O -N

WLL

E--

. r,

NC!,
0 0 It N

_O 00w
10 F-FWZ W Z

E-1W

0 0 N
(*)

U 1
XU CI NI

H N

0 0 It N 0 0 0 C\j
(M

0 0 CC) 0

LL.

Ow
Z

u 0
F-

0'0

HQ

o
N

0 0 "I Co

0
U 1 (` FN

J. 140
0 0 CD
LL

co 0000 3u3wow

Io

cv pallddd

a; Dwiufl/}uawow

It S F-

0 a J w
N

nE ZSQ

VW

CD

=u.
C`W Cw O t

m -..
(/j
I

`;
u L: w:
!2

?0
N W
W

LL :2

O
F-

--i-b C b

0
N 1-

--

--

tic
Co '" E NW

LL. :s
v C.J

*j dr
1. u CU

n,.
E

O ...;

z
QJ

o
U

ww o,
+'

o
f+

EEO

CC bbb

N N

OOo EE> rn rn CCb


YYc,

v > v
'v u
+

o>a
E O

ac a. a

uubO Oo a. LUUO

c.

j2 40

N Cr)

^ aC

b F1/1

11 t
.

LL

ea
N c+o cu

i
It Cl
N

.J

rn

U)
N

0
J1 fV

.-

Co

It

w"N>l

3uawow

pallddy

0
U n FN

CD

2 I-

U) J W Z 2 '1
Q

UQ

NE
Cl) E

Wm
S:
"W

Gw
U I N

f-"

Z
aQ

N
ci

(n U) W.
U.
v

__ d c11 dn 41.0
Z u
_.

L: .:

cc !- 1-

cNZ
Y,

-i

r-

wv

co

2 O
Z

LL

Lt1

O u "OO
+ cCN NyN

I
...
+ E N

E V E X O

Oo Hw da a
><

FN

a =

OO

. .
vv1
O f+

EE>dI vD v Ecd....
YYLL

E O b

>a
er
O

uu

-EI
F-

CM CV)

Uu0

LvN

L J3

w O .0

6
LL

00"I
co NN It

u 0
FN

N N b L N N

r t C) . tl 3 w r a

0 N
w'N)t

.o luawow

co

It

pailddV

o so 0 It w CO >cc z
cc

0
(f)

0 N

O
J W J

oz cV Y
v d

a a
F-

Oo
-o d
Q.

F
w

o 00 (i
N N
N >d > M Or

,qt a
%p

0 N

w w

cn

G
.J

c
E
'Z u0 ad IW xt

O u
.+

d
E

d 9-

oc

F-

0 i

Od
NmO

Od ,0..

>
43

.v

NL N d "d

M C

o0 00 %0

in

WW/N

o0 00 It

cq ssaj; s laa; s

0 cm

LL
N

Z Y

n
Q

0 i 0
FN

ow om

OdO).
aQ QZ

+ O o Fa
UI N N v

o u
OO0
`0 NQ

0
J
o0

O >

d2

W
I ao FN

d
I V.

>

-t! v u
Ii d

00 %0
po

Fxc w F-

0 i N
FN

le

N w

%00

o NZ~

o000000
0oO000
I%O V (V) cv

1-i W 0 00 LL -0 o
0
Z

0
O I E t3
N0

_
Ft Q >

In i"
V) IatN

1v

w.
a ' (+r)

o00000 0 00000
rl> L, WIN ss3-J3s 12)alS

p QU.

tO

O co

O
'O

w m
OW It F-

0N
O
LV

Oo
O
J

qe
NY

-o w
J

v d O0

'r
..

t7 d

U) U)

Ow

f
J

o
N N

Ow

vb >O

NH

oo "" CV
_E.,
bO ab b -

>

0
O

LL 0

0
iE C Or
f+0 od y--

t
N

a
cc

WI.
9

.C

:1O
, ar.,

1-1.1

!C

`'d

rn

%0

0 0

0 0 10

o0 00 N
zWW/N

It

oo 00 mN
SSaJlS

0 0
laals

+ 3 . cn v-.
d N

(9
LL

0
O

0
"
N

0
U I
N

co
W i

0 N

O CL O \s/

0
O d.
Q O " 0 ___.H
"N

a
J in w

U
I

Oz
NY

a a
D .O

"

0
" 0 -

-v 01 O CL d Q

O" O
a Qb
'Q CJ ++ C d" do CL

ac f-N

w
J w w

po "
"
"

V)
--

"
O

QQ

0
" NO

LL

Q l7 _

0
On

I
I.,.. N

0 0 N

O
E
O 4-P.0 w O

xr W
"I

d or

>
N :M (7 U-

F-

-0 ,1
v . L. 3 tn v-. d
N N

0 N.

0 0 %0

Ov O0 N
zW

It

o0 00 cn
ssa, ils 1331S

0 0

II-

W/N

W CO

1
O N bO E () uE Z.

E d
,p

L U G O U O

c E O E ob c U O U

V
IY

v b ti -

9y

-v

r,

u
.0Cb

-a

ouo
I3

-t,

oa i^ H
lLn
co

Co c+M "
LL"

ti co qt N qt

.o cr)

miC14 m
W'N)I

_i

0 co

d N-

Co

'0

3uawow pailddy

W co } cc ZE cc 0

Z 0
h

cm E EU
c `o O u eQ w NW

F"
W J li

ca

J U J Q U Z Q .0 W > W N m
CD

o14

Co lt

N It

NO M W *NN

<'7

10 N luawow pailddy

to

No LL

i
0 E u
E Cl b . 92 to t.

iI

I0

U c 0 U 0
.J

OQ IRt

W
}

E 0 E CIL

c to
E
0

v
b > L7 bU .GO

i..

E c

0 i 0 1N

NQ C II

cr)
Nv

z
0

(j

Im

cc
Z 0

c o,
N

Y U
L

OU
v

U
V

FJ W

U W

mE E cW O~

uJ bU
} OO co oUJ

.Q

_Q
1

. F-

p3 Im

z
O

Q
in w

>
W N

0 i CD FN

In N
II
3 v

ul !

%0

0
CD

mm
.o9 t*)

WN

dd

. c

CV)

tt

ta

NO

w"NA

zuawow

pailddb

O
pII

OO co FN

d' N

M I- FLn (n

oXO

N Al

Z 0
i-

w J
-O O
1-\

LL

w cl
.J

- -O

io
-6 x

CD -i

0 Z w cc (n F- w

-lx
O

O
x-O

E E

d'
O
O

0`%

w W
!-Q tn Z W W

U
1-

` t

o1r)

Ln '"O Ox

co

V N .b .F. b

o-O -.
-i.

-1
x

, =

a WW m co
xO
O` Co O

)`O

z U) Z E 0 cc
J W cr.
LL

Ox,

0 ac 0

-0-

`x

0
.1

E 0 b
vO 41

d. 0 Lt.

O
X-0 Ox N
N Vf

r
"b _rn

NM
4-1

O000 0000 ln ww/N

MN is s.sa. IaaIg

OI

rtn d 4A

W m W

tn O a O
V. . Z 0
CD

V
EW EJ LL W

v O .c E

c 0
4.4 U
_b `d

LtJ "-

aJ U J

0 L.
%lj/
.p M to +i

CL

:2

V lz Z

ap >
c. or u C .+0

00

W cc w V) Co

O
N CD 14
CV d'
14O M

W'N>1

Irt N

co

140

U-

Zuawow

pailddy

2
o

V
i
N

W Co w FN-0

0
t

Ftn

UI

0 r
.J

to E v
to N 0 CL 0 L

u
Z

O EI IU)

M IL &+

EJ E co

W
LL

I
bU N
.0Q

ul -5-: + ,

OU I

0-2 44 U
~-

C11
Q

pJ %0

U
U I 0
I H ul

.J Q U
10 Z

W > cc w W in 0

c'! 'Cr

d'

NO cr)

Od CY)

CV

CO

4O

LL

w *NA

Zuawow

pailddV

m N

L!

It 04
`
X\1

W Q]

`O\ x %,

W F-

oE
0 V
d0
w

cn

-- IN
'O

CL :2

`
...

Io

c E 0 E rn c
,G L-

c d E O E

rn c_ O 3 v E O d .a ` O c v
L.

OU
II
NN

3
c1

F- FNN

1---1--%0. -i4, --.


x o` lxc N

"_ ,, ,,, c 0Q aZ t_ co

b
0 cc

I)

I d 3 v >

.. `

a
.0
'' "N

OX 11

11

0 0 0

O I\n
X`X0 0X `% Xa

-U
I

It
I _xJ

0 LL
Z

d
N It

0
`-

0 Cl. E 0 u r0 c 0 u
d

0
0

fr)

It N

co

CV

%O

0
O
N

!J

V W
LL

I JIO csJ

W in

4. b

v d
d 0

%AN 0
x
f

z
'O

EW EW C to

or

II
0 0

11
u O OV F- E- _
NN

i
1\0%
X O
`x `
`f

.2m
uZ
v'V
-

co c

cc

ii U
0
I
_3(I O 0

xO

"o a
O
xfO f n

ox

;i ;I

} U
9
It

IIi1' IIIII
CO Al 0

`x. l O

.,.

I
i LL

It

e cY)

o 'r)
w"N>{

04

3uawow

pailddy

o 1 o '4'
0
\I 'O1

It N

W m W I.If)

0 a
. 0 0
x
N

J`0 Ln N %O

0
V in z >-

++

c E O E rn c_ t. O ++ O E G d

C E O E rn C
L

u d r-.

0 .+ C_ O n.

C1

OU iI d- HH

'0

o.

xf _f `o\ 3 -v
I
i~ N.

If

Z 0 CC 0

I v Co
f `X`

E
O
.0

O c v c.

r
O

0 I
1

IC
,p

II U v 0 0

0 N--ra% `X

LL

o
x

Z .. OV

a E
O
U

O w O
c

d'

O/y
"/

cY)

It N

CD

,opw

O
C

O
u w
N

ow t
Nw c E
X,% O

J LL
o z w
H w
Co

or

U
I~-

il II 0 U 0 d

\ x" `x
f

"t N"
c0
-\I

o
u V rd v :

0u II m cr) (Y) N HH V) N

__

-o%1%

0 0.... x 3 '

JIO N1
'O

Z 0 Ln _

I
I

i IC
\

II Ud
0 v

-i

J--, f
xxxo! %x
le

a: Q

2 11

O 1 %o% x`x o
1 xO`O-

aO V
Co !1) C

N0

le

MMN

i_
juawow

__x. M. XoZoxj0
CD N0

LL

w"N)I

pailddy

tn
F O

li
%-x
O` ` O o` xa0 x Nc

w m
qt
E Ed

W }t! )
:2

cr)

N,o "`

_
N

ZU

oO

-0

C E 0 E c7) C
L-

00 E
0)
+-

x
eJ

OO

O(

----

Of OOf

"C 0 C-

NNx

0
.J

rd o,
d

,_

cn cn

11

..

1\1
x%n

-1Z v_lic
%

3C

O
x"bY_

v
E
O
.D

E, t7
to .0

'r
C\l ON
x` f x .

-v

I I
co cm

___XX`X`xJ o O

cr 0 LL

d
N

C_ Tl L

0
Qe--

0 . E ox t 0x c 0 u

0 H
J

V W

LL
`

0
C V

O%o

0 N
\
ON J
D O%

W 0

Z.

Z
E LtJ EW

--

v-

F-

II
0

x.

oW
V

u 0.

o* I
00 II

o;:t

-1
N

Z O c tY

nn I- Ft! )
U)

N O II I) a 1 0 a

-x`0` . x`x0` 0 . c0` XO


X%X

Co

aa v Z: > V

I ox
I(

Il

al 21

ZE

3 b

I
N
it'

o xx

` 0x
0

%O

c*)

'')

d'
N

CO

%O

0 0

C!, U-

w"N>l

Iuawow

patlddd

\I

fV cr)
`O I '0

c
0
N

frS. a -'

0
14 N a V

C L

d N

k-"O

-,;. i.

\ O \" . x _

-\
\ \

No

%o

i- i-
N
\ N
\ y0
N M

!CD

ZN 02 pVm
cV

.0

Co

Itt N
\c

co

oJ}
CV) EQZ EQQ
N0p

F
b

LL CC

WQ

r 2m

It

.%
o\ cp

\O c'.
I"\
cX

.a b to

14

W tic

u
w

0 c +' s, >.

\
\

"

i`. *%
', "i

V
to 10 Ti

O
LL

\o\

\
CD

'%"

\
\_

N%"N ..
": N\
14

Q FVZ
W

o Z0
1A

12

Q O W U'

IN
(N CV)

11 N

.o

Co v b
E
L

0'

N pl

It N _w \
V

.o

CD

310 O
M

>Z cc

I'

0 \ c

03 rIt
o
W

ca O

0 w C
C O "L 4J
.Ju

_I

-cp 'b
to

C' \`, r :
o
"`

le w .- N

o OC .xN

u 0 \t)

\.
0 cu O
MN 11

GD

N M

It N

o ww

co

'o )

Co

uotioaipp

pa3oInoio:

c*)

0
N " J LLI l1J
l! )

tn tn ui
CIC

%I. .,

-+ S.

o
E_ NE

1U, F-

O
b

>_

4-+ N 41

ot

NV

a cc U

v O tiJ
Or

-`

-<

, %Z

OC

O
V)
N

E Q U -v c d %
IV > c.
.V to 41

F-

_EX.,

.i_,,
\ i

\`

0.

LL. W Q CO

0
.
N

0
V V

.n

>

Co
i
LL.

0 0 th

0 0
W/N ZW

0 0 Cn
pals

0 0 (V
UI ssaJ; s

O 0

r)

0
ti O CM

-r

.J

o=

F-

E _U v C O V
'D b:

B::

E Ep

_ E-

b tn

OdQ
er

14

Um

C) L 7i; v tn J3 d
00 I I.

Q >VQ

"+

oz v_

:2

--

03-
YX UQ... O

OuW
E

lL V)

dnch SO

.xzz aw accr
ui

o
NQ

6>

Ot

6
LL

0 0 o

0 0 ul

0 0 11
zWW/N

c7

O 0

0 0 N
ui ssa. ils

0 0

0
0

jaa}s

_ H

, %Z

I
O_
F(1)

I-

W m

o>
N .CY O

bVQ
bZ
N

:20

0 i Fa
U)

v
L N

E U o c O U

_0 0

" Y U

QJ LL.

v ` u

o
O -'

.,

EZZ :i
E_

_b

IV O
4+ > t. N Cl . 0

cv
-1

U O U

11%.

O 1 m Ftn

Qw C

aC

>. v)
0

0 ul
6

00000 OOO
L()

0
N CT7 cv

LL

WW/N

laals

ul ssai4s

O"4

O"3

--o--x

STI-O STI-C Mdw

i
0.2 E
E
-Os V

i
.., 4000 O

o"
.** O

>
V
41

o.00

11X''

11111
36 42

"1

30 24 18 12 KN. Applied m moment

0.4 c U ., 0 U
O. 3 E
X

--0---

I
X

ST2-O ST2-C

0- il 1., o
x o 00,

O"2

o"i

0 000

--1JkT

i 00 00 i o-000, .
, oO-

I- x-mm-da

00 o6

x i

I
18 moment 24 30 KN. m 36 42

12 Applied

FI G. 51

COMPARISON BETWEEN CRACK FOR ORDINARY AND COMPOSITE

WIDTHS BEAMS

O"4

--o-x-

ST3ST3-

O
C
/ i0 , - O-X //

f, / _

0,

0"3
0.2
E E 1i O

o 04, Q" o-+-x_"__

d
> N'

i
18 24 30 36 42 48

d-N

Z-06 V1

--

12

Applied

moment

KN. m

3
.X...

n.

L. U 0.3

--0-111-.

ST4-O
STdCIIIII/

Mamdw_.

1 _._..
X 0

000400o

0-211

0 '-i-

O"I
.. o.

O"0

1-

-o
-

v'

I 1_

12

18 Applied

24 moment

30 KN. m

36

42

48

FIG. 52

COMPARISON BETWEEN CRACK WIDTHS FOR ORDINARY AND COMPOSITE BEAMS

O"4

--o-O"3

ST7-O ST 7- C

O"2
E E

=-oma
-1 --

o--

o"i d
d
41 M 4-81 cy

i401 .0

X-,
24 18 12 Applied moment 30 KN. m 36 42

t-

v
3

O"4 ,c
u v L
u O. 3 E
X O

--o-x

STIO-O
5T1 OG

11 x J, 0

MdW Ma IIo
0.2
itop 1
-1
" -Y1 X X

Xi

/'

o-o
1i i

I
1-

o"i
0

o'

O000-

1----
0
6

24 18 12 Applied moment

30 KN. m

36

42

FIG. 53

COMPARISON FOR ORDINARY

CRACK BETWEEN AND COMPOSITE

WIDTHS BEAMS

O"4 Md w O. 3 ma STII-C

I, X X

0.2
E E

er O"1
e

,
0
6 12

xI X000000.
18 moment 24 30 KN. m 36 2

d d .JO UI
+ C

Applied

t .+

0 ` u
E
X O

0-4

I
ST12-C

O"3

0.2

o"i

12 Applied

16 'moment

24

30 KN. m

36

42

FIG. 54

VARIATION APPLIED

OF MAXIMUM CRACK WIDTHS (COMPOSITE MOMENT BEAMS)

WITH

b0 >U dX

bU >U bX

,VN

"d

u *Z; "X
N

`n

d Wd L
b C

0
W
L_ ~I I tip

oc
bO

>L bU
dU

F^- Ul) W 1-

V_ II-

cI d
d

Nf

I Z W

ZIVN
v-

O -0 gn
bO

s}

14
I

F----I

Cl) I
14

I Fce)
E
Z Y
vO p C7 IA bO ._

C N

It

VO
1 f)

2
:2

F--"1 4 r It

7N 0 O( V b
a. CL

Z E

CV) 'oi

0
N

dv b
d

O
It m
cr)

.o

a .Q

1 d cq r
N N

=v
L

'c

m I

or . 1-", oz Q. cc Ew 0V

CL
1

i
N C4

N
' . Nr JI

v
It L It
N N

in .

tn Y
V

I IV)

t7

0 0
..+ '--1 c .3

tn 6 UL

CL

0 00 11
LL NM QQ LL LL

ta E C7 to .o

UU II -N QQ
U. LL

E O or .0 N
d

F0_

>L

a W
a N X Q . _i Q F W

iii
140

d C
M L

II

H ii
n
N

E O U

0
x
w O *Z.

Z ZQ

x
a c O Z a d L_ O
L

OW
N
.

ao

U)

or

a
vO
c. d

ZW E _
OO Ja WO
DU
QZ

.o E Z 0

0 'x 0
a

i ii_
m le in `o

LL a LL >O
W

z Fu
LL CC W

'4O Ln 6 LL.

JxO ..

d' ...

Ln

----j

O'00= 10 O. O. sixo O. Io. iinaN O" 10 otzoa

OO

4zdap

3n1ZDaJ13/43dap

W >

CM

Iii cr

Ln (f)
a 0 V
Q DW W JW LL in

OOO II N
QQQ U.. LL LL.

44 E c7 to .0 L

vi E O .o

a LL
ln fV

LL

d 41 II 00
a O d E O U

1Ln
--1 " N

d c v
L

--

ii

i
N

0
d

a
N

tn D0
o

0
x c 0

o="O
x

5a
:2

<

Ln Zd d r-

C O :. *z oi 0. vO L .0 E Z

Z OZ C7 >Zd 2; Q QZ

c.
-0

OE
J EC w0 wQw

E :i Z

9
w
LL

CC U

0 F- U U
N

OZ
wZ

Ln . WZ

cr

o0000

o --

0 CNJ
9

000
01 x
a3aJDUOD

UI
UIDJ3S

O 0 0

o-o--JO O 0
CD IDJflXaIi

Pt11

o0 0
wnwixDlN

0 It

6
LL

anlssaJdwoO

E E
C 0 .. + U d v-

0 i N
11..

U I N LL
E O .0 a 4.3 IA 0 a E O U

,oO .oO

xx

cr)

0V
in

_. II

.
Nt N tI)

i v I
j
L C

0
V cn W W
cQm

O d'

I. v
M L. 0

EE

NoO ,0O
X

O
xx ON O0
11 cr) II cY)

.2 41
U~

1w 00 -i a
Z 00

Z:

dQ

-"` \ \

M0

:2
V

II

OU
4

00
o
r-

d'

o
tn

C Z iz -Z QQ 0

.L

L-1
X X ('7 X

CIC wz
x Ln N U OD

1`

tn tn

.`

tn

%z
N I

Cr) C] CC

fW

OU N.. .
N It

0
d.

m d.

LL 0 0 Hj WQ lL.

N, N
JL
C9

T_
W

%0 10 0

LL Wm
Ln

U U
41

CO It t'r)

%o

6 U-

"_
CD

-_

t! ) CD

L)

I
OO00 CT) N>1
ry
(Y)

I N

m.

00
po61 pailddy

00 -pool pallddb

N>I

oN
W W W

IVW
-

U I CV
X`X \X ' "` I O`x\ I

a
O QV Wp mz W

IX

LL

0
is-il;
11W11

>-

a1( puoas

IuI__
rn c :,o O 0 vO v >. v L LL
rn C_ O O vO

I lo

I ait
-

co

11 w 1'1 ON

10

EZ EQC

rn c_ v L7 0 w 0 er U U
K 1

Q _

cx

V N

Z:
b

Z Z_ Q

Oo le

U U T7 C 0 U

oO
QJ

x x x . x` e,
x

0 1 N
tt. e x ` '\\
rX_

OW

Z. Z

apk)

puo: )as ;o jlw! -1

14D

0
0j

III
OOOOOO N>I pool
(r) cv

I 0

a
0

0
O G

pa! Iddy

LL

3uawow Zo uoi3o31;

6uilioM 2a

uoi3: )al; ap. 6uiuiowaa

i X

r)%_,

IiI 1
oIIp I aI
`'IIIiW

I I
I Ip 11 I ;,

H
cn

"

XIIII
I IIpXp-,. II

LL
3 U)

IIilN iIzW
V
Li.

ow

co

IIII
LL IIOO

~
L IK;
LL lo .
.+W

NNxIi00

O
Q

Ln

to

I!

I1.

I
'\I11

.III-\
.

I I;

tu

l
,

!I

oi11
1I .

\12 iIII
"

.1\1

.o.

da

LL
L1..

}-

Occ-

F-Z
w[.
LL

tl
iii
LL
aC

LL
!Clu

1
0
lo

ILL LL 1

LL 0
O

tr

'u

If

`"
x%.. ...

00,0 It

-'Icv
I 11
II
NNNv

`AI.
% \ 1 ` \II 11

xl= O ww
-0 N rr, c+ n

Co

.4

uoiI: ) a1; aa

xr.-s I%11

00

(Y)

i t t
I
O 1
LL

II
II ! o0 x I

\1? \Ji
i

tn QQ

-i-

t! ) W =m

I I x 1 1

W H
0

I. (
o -I
N %0 O X

V Q

O
00
z C]

oz z

OU

I
NN QQ LL LL 1-` O

N C O i+ i-+

oa QZ
o; -J
D

i "" i
er u v L V

.a

OC
FZ W 0 C Z ` Y 0 3

O
c.
.4

W.

0
U
U 1 CV Q U..

LL
LL 0

OU II QQ LL LL

II '<
XI

0 po \1

d C C v
L

LL

U iT

FU W Li. Li. W

i 00
M

14O 6

1\
I
C
N

i
`
-X---o

U.

0
ww

0 Ianal JaD;s

ZD

0
wnwtxDw

y}p)m

)j: )DJ7

i
I
I"

0 CV)

1 1
I

0
It
Q LL.

I I
I. II

I
to

F--
LI)

i i x i i

_I_ 1 I I
I.
00

S F-

a Ui m
W f-LI)

3
Y

.I
II
I

V Q

0
N

a: V
0 Z

a. O
U

I
I

Ov II
N QQ LL Li.

oz }
CZ O-Q Z-

xI

Q to 1O'

'-'

U 1N

no
lt1 /
rf /^ L

III

0".

4
L

HZ E L11 LL LL

-so - i`. II
QQ LL LL

v d Y
U

OZ

O 1VZ
LL

UU
U _-N aQQ C LL

xII
00 II

i; i
I*

OC
L

11
If, 6

WY LL Cf. LL W

; i-

i.

N (7 LL

X: [ O-

cV

OO IaA31 IaaIs
3D 43pIM
)jODJ3

O= ww

a6DJany

N 0 0 6

Ln

oz o0
d
N

vm in .-

N
vO

> o - Qb "NO O O OJ
EZ OQ Ln u

Ec!) Q DO
xI WN J Z) LL V) Z: WQ

"X
1.

vt
G-

_
-0

O0W OdZ 0. -OdW


NO N "t in in r-

IM W> OC

+' c d Z

wo
.G U b
+ NN

0-- -,

bCo

i
N L

d'
q""

pf. -'p Q + b
f+

CD cr)

1N N

O c7 c*)

CO N N

b C v-

Y
L.

LO O 0 O

Lt.

000u

10

o+

v-

Q,

7-

J"
ww OO

0 0

N 0 O O

o 0 r- F'
-m
ON

[Z

o >_
. zr

O O OC V
XN

DI

WD
-it') LL.

za
Ln

i.-

WW

o-o ozW
"N 0.
bW

co
tn

- o0

Ia w :2
`V

u bv-CA

4J
N
iJ

d'
%O

Ln O 0 6
'0 b CT

(3
LL

OOGU

J"
0
1 0

ww 00

"

v tn

UU
II

i~

V)

ry cn

0 CD it

ii
O O It Z
0

0 N
cr) 'ti

J O W Z c z Z _ V J u

oO
dw

0
z z
LL

Oz d.
N

H in w J

.O
H

0W
%O

cn_ LL Z LL W Lit h-

0
ln %O

0 CD
0

li_

,\
O000
It

L
NQ 01 x-

O
13uuo40

o0 00 Co "7"J";

ui

UIDnS

alisua3

0 0 It

O 0

0
0

wnwlxD{N

I I I I I 00 II >> NN

v It to

v -It
A 0

'I
0 co It

I
I

_1 '

OU II cv NO >>m (n u)

e W N W O O d 1J W W F

I
I

i 00 i
0 0 qt.

i
"0
I

I
I

i II
I I I 0 I 1 II

i i

0 N
(Y)

i S
I I I I I I _I

0O
rx o-m -v

Z e-j;,...
UQ ZW

>-

a b

1 1
I

-v
to

Q O

E_

1
-I.

0
I

0 le N

o~ d'
N,

Z N0 =Z `n

I
1

I
I I

I I II
OO

0 '0

I I I I I
_I 0 I It

O
U?
WQ LL n: W

i
I

0 CC) 0l

O co

O LL

II i
Io
0 0 N
1 0-

0 co

JD 0
It

00 co
NN

O cm

%0

00 N0
NN

00 I I I.. I I I I ( I

I I

I
I

ov
NN' >> W U)

i I I

0 CC) ,Rt
Z

iI i i
0 0
Ou 4I

0
H U W J LL W in

I
I

I I I
I I

I
0 I
0

1
I 00

II

D> aa

O
OZw
N M 0 IA

IU

10

ii
i(

IU I1
Iib
I) 1v

ui _N
c
V)

0Qw Mo

-n

po o EwQ.

I'
II I 00 I
j
0"1
I1

1
i

0o1
0

o
N

Z a.. -
F.

tn p Z cn Q LL

""`

UZ
W LL W
O

""O 1`1 -" ,"" J " 1O `O tinO


`_ L:

r*1%O 6 LL

1Nn 1

'

co N
_, ""

C'

`OO

`
rv cm

0 :
N

l -o
Co

0
0

4u1od-piyY

I I I

0 d' N I

't U)

i I
.1 I

I I

i i
I I

0 Co qe
Ln
H O O OY

I I I
0

i I
I

i i I
i
I
0

0 0 It

I
I I

I I i

i i I i i i
I I

U oC U
ZU)

0 N
CM

0 N
>O v

(D Z C] Q

a W
aD W t-

d M
d E_ F-

i iI
I I 00

i 0_ i t i
O

Eoa

o U) -j 0
w :2
ZO Z
lL

0 It N

o~

It NQU

CJ Q
>-

I`.

O _I

1
(

O
f-

0 %0

II
0

i
i
Ouo II
NN >- > NN'

70 CC u-O w
CD

Oo

i
tn O / I

II
tn

O`

0 CD

II0

III
i

DI

I(
I

rn

N-MN

I 0 oO
0 JaAaI InaIs
;D

1I
A0 e

O00000 ww yjpjM
MODJ: )

0 0

wnwixDV1j,

i . . i i i i i i OO
I

,It tn

I I I

It Ln

-I
I I I I

0 co 't
N F-

0 0 It.

-II
I I

O 0 It

Y U

0
zi
0

I
o. N
(r)

I I

_I
I
I I I I
I
0 I I i 0

-I
vN
>-

0 N
M

>l7

V' Z

Q
W Co

v 0
a E_ F-

0 It C\L

i i i
I

'v

0I
E

W t--

0 d'
N

i=

"-o

._..

tn

w z
DZ

i
I

I I

u)
LL ..

O
Ha
UZ

`'

0 /I
O

W0 LL Li LL

W .
01

OV II

II
-O-COD 0

0
N

D> a in I

II

-4-

U I N N

0
I
I

i 0
6
ww

0
O . O

I I I i
0

I 0

0 CO

%O

LL

\
0
1
0-

00 1

ul
I

Ln 0 6
IanaI IaZIs

Lob
.

Ln 0

0 0
I

115 tli. QIMA:

0
)DJ: )

0
a6o. lany

c ei 0b Ed a xb

O +' x

IIbc
c

OEd>_ 'Z: c. b b. + X

-10

vaC
c
c: v:

Wc

Wc

co
10

d 0

L-i
IN V
'i

w E

0 0

a
0
z

O J

CL

U Z_ in Y
cc 0 >OC

zE w

I,

U I N U)

0 N
N

cc H
D it tf) 0 LL LL 0

I d C C 0
v O 0
L U U .. "v O 41 CC

z cc

w v F---v 0 a 0 Ln

i
't-v
'-

oC U w FG w U

.+

IL

Ya v
. V
Q

0 LL

V 0 !'"' Lf)

ww m
WJ F-

cr
:D

cn

LL

O
O

aw

CY-

UQ

lL

FJ

(a)

Ordinary

reinforced

beam

(sTIo-o)

(b)

Composite 2

beam

(STII-C)

PLATE

CONDITIONS JUST BEFORE

OF

BEAMS

FAILURE

(a)

Ordinary

reinforced

beam

(sT3-o)

b PLATE 3

Composite

beam

(STII-C)

CONDITIONS OF AFTER FAILURE

BEAMS

PA

I M

V1

H
U.

1- ID .

v u
a-j

LU

hLLI

-o
00

dU

u.

vu
c-p

w z U-

O
U
W.

0! +

O UU
C

du-C3

ccY dO LU
C "N _lC U VV

cQ
O1 .C -f--) C

LY
LL 0

vi
d 0
d

-e

o z>
d

`n

'1
-0

L1J F-J

0 W Z FLf) :D Lf) cc 0
LL

Z
W

2
Z

C
CC

z
0

LO

a.

APPENDIX

(A)

Derivation

of stress-strain

relationship

of lightweight for 0<

concrete ec ej is

The equation of a parabolic ac=aec2+bee+c Where

curve Fig. 6a

a, b, and c are constants At At ac = 0, ec =0 dar dec and c=0

eC = ei,

Therefore ( dvc dec =02 ec =e Atee= Therefore 0 ae, +b

aQc ac

Ec

C
dz b=Ec Therefore

da

=Ec=b
cc =0

From the above equations

Ee 2ej :
Ec ec2 = EC (ec e2c ) 2e By substituting Ec = in this equation 2o ej The equation
aC

= Ec cc -

ac

= vo

and ec = ej

of ac 2Qo

can be written sec etc ) 2e

as

ej

The value of ei therefore


e As per CP110 0.67 feu co _2 Ec

can be obtained from the equation

or = w
lieu Ec = 5502300

Ym
Dc 2

Substituting

the values of 2x0.67

ao and Ec as given in CP110.. feu _.


De 2300 1 fcu =fs 4100 De 2300 (for Do

ej

feu Ym

Ym x 55004, fu
Ym 1.29 x 103 2 Dc f cu Ym

= 1800 Kg/m3

2510

Ym

3075

and Ym = 1.5).

for

feu ej

= 50 N/mm2 = 0.0023 the partial fcu .


Dc 2300 2300 1.29 D2 x 103 fcu

Without incorporating 2x0.67 e = 55 00 fcu

safety factor see Fig. 6a fcu = 4100( Dc 2300 (for Dc = 1800 Kg/m3) 2

-3
=f

cu 2510

For

feu ej

= 50 N/mm2 = 0.00282

APPENDIX Derivation Assume Dc of P and a (density Factors

(B) Load Conditions

of the Ultimate = 1800 Kg/m3

of concrete)

(a) Incorporating

the materials

partial

safety factors Ym (1.5 for concrete and

1.15 for steel).


r Refering i, to Fig. 6a The strain and stress 'distribution of the ultimate condition

across the depth of the beam would be as follows:


jE

O. 67 f

F
d As
L1

Ix

1 2510

feu Ym

em =0.0035

cu = 0.45 fcu m

Cross Section

et Strain distribution ultimate condition

_Z'

at

Stress distribution ultimate condition

at

From strain distribution Total compressive force

y=X C=0.45

fcu /10.76

fcu. b. X -3x0.45 b. X. fcu" fcu /10.76 /10.76) b. 0.45 fcu. (0.45 X. = cu 3 =a. b. X, fcu axis:

Taking moments of area of the stress block about the neutral a X. fcuY X= (0.45. X. X fcu 234 0.45 y. . fcu).

Substituting by y and take X out aY=0.45


2

0.45
12

fcu
(10- 76)2 .

694 - fcu 3087 a

For the ultimate

moment based on steel; forces

to obtain the neutral axis depth, equate the internal C= Ts f As fyAs

a b. X. fcu =y Ym X

fcu YmQb. 1-Y


X d1 d1 (lever = z arm) = -p fy As a Ym b. fcu = d1 (1 p1yp) aYmfcu

The ultimate Therefore

resistance Mu = Ym

moment based on steel; As d1 (1 -p) aym f cu in the equation of the total compressive force

For fcu = 50 N/mm2


a=0.351 Hence: Y=0.594

3= 1-Y Substituting ultimate

= 1-

0.594 = 0.406 3 and Y and (Ym 1.15) in the equation of the

the values of

resistance fy Mu = 1.15

moment based on steel: As d1 (1-f, p ) fcu

The ultimate (assume

resistance 2 dl
22

moment based on the concrete would be equal to:

X=.

Mu =ab

fcu
dig

(dl
fcu

(1 - 0.406)

0.351 x1

22 = 0.14 feu bd12 (b) Without incorporating The stress-strain the material partial safety factors:
Ga

relationship

of the concrete is shown in Fig.

With the same method of analysis for fcu = 50 N/mm2

a=0.49 Y=0.61 P=0.39 Therefore, the ultimate resistance moment based on steel is:

Mu = fy As d1 (1-

0.793fyp) fcu

the ultimate resistance moment based on the concrete would be equal to: X= dl ) (assume 2 Mu = 0.197 feu b dig

APPENDIX

(C)

Working and Ultimate

Moments of Test Beams

Beam ST 3-0 and ST 3-C Based on nominal strength

bxd= fy fcu As d1

150mm x 300mm = 550 N/mm2 = 50 N/mm2 = 402.286 mm2 = 257mm

p=1.044% L=4.5m Dc Limit = 1800 Kg/m3 Strength safety factors of the materials the ultimate design

State of Ultimate Incorporating

the partial

moment of the section is: Mu = y _! Ym AS dl (1 - fy p a Ymfcu

For

feu = 50 N/mm2 a=0.351

P=0.406
Mu = 550 1.15 = 43.77 x 402.286 x 257 (1 550 x 1.044) 50 x 10-6

KN. m w12 8

The moment due to the dead weight of the simply supported beam is
= 1800 x 0.15 x 0.3 x (4.5)2 8 x 10 1000 2.05 KN. m (say 2KN. m)

The partial ultimate

safety factor ( yZ ) due to dead load = 1.4, therefore, =2x1.4 = 2.8 IM

the m

design moment due to dead load moment

The applied ultimate 2.8 = 40.97 KN. m. The partial

design moment for the beam (live load) i's 43.77 -

safety factor ( Yt ,) for live load is 1.6, therefore, design moment of the beam (Mdiv) is 40.97 = 25.61 KN. m 1.6

the applied working

The working load (W) is


Wx1.5 2 = 25.61

W=

34.15 KN.

Shear resistance
Reference v=V should be made to 3.3.6.1 in CP110, equation 8.

bd1 resistance C beam (appendix is 43.77 KN. ), the m moment of

The ultimate therefore,

loaded length, 1/3 4.5m the beam span at span of simply supported with a V_ 43.77 = 58.36 KN (Ultimate 0.75 = 1.514 N/mm2 load)

150 x 257

The value of vc = 0.6 N/mm2 for (3.12.4 in CP110) v> therefore vc

p= 1% and concrete grade of 40 or more

shear reinforcement

are needed.

Mild steel stirrups

of 6mm

diameters
Asv sv 56.6 Sv

are to be used.
>v_. vb 87 fyv 150 (1.514 - 0.6) 0.87 x 275

(3.3.6.1

CP110)

SV = 98.77mm

The spacing provided was 100mm.

Bond stress Reference should be made to paragraph (3.11.6 in CP110)

Local bond (3.11. G. 1 in CP110)


aDS

V Fusdl 58.3GKN =2x = 257


= 58360 100.57 x 257 = 2.258 N/mm2

V-= us dl
fbs

16 x 22 7

= 100.57mm

The ultimate

local bond stress allowed in CP110 for grade 50 lightweight concrete 2. (3.11.6.1 for deformed bars is 0.8 x 3.4 = 2.72 N/mm and 3.12.11 in CP110)

Anchorage bond (3.11.6.2 in CP110) = f___ Ym x 201.143 = 96.2 As

The force in the bar

550 = 1.15
Anchorage length = 1.5m

effective perimeter
Anchorage

of bars

= 16 x 22
7 96.2 x 103

= 50.28mm = 1.276 N/mm2 concrete for

bond stress

1500 x 50.28 The ultimate

anchorage bond stress for grade 50 N/mm2 lightweight

deformed bar in tension is 2.6 N/mm2 1.275 < 2.6 O. K. (3-12.6.2 and 3.12.11 in CP110)

APPENDIX Derivation

(D)

of Equations for the Neutral Axis Depth composite section (Xuc)

D1_ Neutral axis depth for untracked Uncracked stage for 0< M/Mu <

_b
Co

Take moments of area about the top edge of the section considering transformed section: an untracked

bd(d/2) + (m - 1)Asdi + (ml - 1)Ach d2 From this equation and by substituting and 9=
Xuc

[bd

+ (m - 1)As + (ml - 1)Achl pl = Ach/bd2,

Xuc

p= As/bdl,

A= d/d1

d2/d1
0.5 _ X2+ (m - 1)p + (ml - 1)pl 2 n

n=

dl D2_Neutral

X+

(in - 1)p + (ml - 1)p1 Tl


b =_

axis depth for cracked composite section (Xcc)


K

;N

Cracked stage for

Cl < M/Mu

< C2

Take moments of area about the neutral axis level considering transformed section. a cracked d dl d2

b Xcc (Xcc/2) =mA.

(d1 - Xcc) + ml Ach (d2 - Xcc) p= As/bd1, pl = Ach/bd2,11


1 2) 1) _

For this equation and by substituting


n= Xcc d1 = imp + ml Pi 11 )1+2

= d2/dl

imp + ml P1T1 (mp + ml pil )

D3 Neutral axis depth at the transition Transition stage For Co < MIM: 10

stage 5 Cl (Composite beams)

Reference can be made to Fig.

The equation of the assumed parabolic X= Where: aR2+bR+C R=

curve is

(1)

M/Mu and a, b and c are constants

The boundry conditions are:


X= X= dx/dR =0 XuC XCC at at at R= R= R= Co Cl C1 (a) (b) (c)

Substituting boundry condition (a) in equation 1 Xuc = aCo2 + bCo +C


(2)

Substituting boundry condition (b) Xcc C= Differentiate dx/dR


0= b Substitute

=a

C12 +b Xce -aC12

C1 +C = bC1

(3) (4)

equation (1) and substitute boundry condition (c) (5)


(6) (7) (7) and equation (4) in equation (2)

= 2aR +b
2aC1 =-2aC1 equation +b

Xuc =a
xuc a= Co2

Co2 Xuc Xcc

2aC1Co+ X,

-a

C12 + 2aC12

(8)

_ 2C1Co xcc c + C12

(9)

b=-2

Co
equation (1):

2C1C0 + C12

(10)

From

X=

aR2

+ bR

+ Xcc

bC1 aC12 -

(11) Xcc

X=

Xue - X" 22 C 2C1 Co + C1 01

R2 - 2C1R + C12

J+

APPENDIX

(E)

Derivation

of the coefficients

P and a to determine the centroid and area diagram in concrete. Reference can be

of the compressive made to Fig. 12 (a) Determination

stress distribution

of

(i) When ec S ej
The equation of the parabola is : ao ac =2 e (ec ec ) 2e See appendix (A)

The centrold of the area from the y-axis depth in flexural cc e o CC eo o 2 Qo ej (ec e. 2ej 2) de _ ac dec _f 0 member, ec

which can also represent

the neutral axis

at any value of ec can be obtained from the following: ec dec . (e is the distance for the centroid of the area from the y-axis).

I
ec 8e,

vo ej

(ec -

cc

2)

cc dec .

2e

Co

2334 ec - ec 2 6e

ec _ 3

From this equation: eo = 8ec ej - 3ec 12 ej - 9. ee


By substituting

co =Y

ec

8ej 12 ej -

3ee 4ec to the value ej

The value of is equal to1 Therefore

to calculate the distance of the centroid relative -Y =1-

8e - 3ec 12e - 4ec

4e - ec _ 12e - 4ec

(ii)

When ej <

ec c emax line at ec which also represent zone of a flexural /2+3


a0

Taking moments of area about the vertical at the extreme element of the compression

strain

member.
ej

ha0

(ec - ej) +3

ao ej

(3ec =

vo (ec - ej)2

ej + ec - ej)

From this equation the value of the coefficient 6ec2 - 4ec e" + e2 12ec - 4ec ej (b) Determination (i) When of a e, < ej

at any value of ec is equal to

The area at any value of cc is assumed to be equal to Considering


a ac

a ae e.

the equation of the parabola given in appendix (A)


ee =2 ej ao fee 2) eC 2ej

dee

Substituting
2 Qo

the value of

Qo and integrate vo ei

ej

iec _ 2ej

e _2

e2 I L2

^ e3 I Gej j

From this equation: 3e4Cl 6ej -

eo
3eo

(ii) When ej . ec
The area at any value Therefore

emax
to be equal to a

of ec is assumed

ec CFO

ao ec From this equation: 3ec - e 3ec

Qo ej +

ao (ec - ej)

APPENDIX

(F)

THEORIES F1_ General

FOR THE

CALCULATION

OF DEFLECTION

In order to calculate the deflection of the curvature

of a flexural

member,

the distribution The differential

known. be beam the the, should span of along is given in the following (Al) deflection expression: -

equation of flexural d2a rb where a: s: rb ds2 Vertical

Distance measured along the beam : Radius of curvature relationship :is linear, the equation

For beams in which the moment curvature can be integrated a=k$ 12 to obtain the following (A2)

expression

To calculate the curvature

of a beam, it is important

to classify

whether the

section is in a cracked or uncracked condition. the be conventional using calculated member can 5.3.1. in discussed as concrete section to consider the contribution curvature of the member.

The curvature

of an untracked

method based on a homogeneous

For a cracked section, it is important the

of the concrete in the tensile zone in calculating This, however,

has been thought of in the past and ways.

the contribution

different in in tension the considered was concrete of the various discussed. are proposed methods for a cracked reinforced

In the following, F2_ Assessment

of the curvature

concrete member out assuming

In the past, the analysis of a cracked member was carried a completly

cracked section where the concrete in the tensile zone being ignored. was obtained from the following (A3) equation: -

The curvature
M ECIC

or e. dl - XC (A4) calculated without considering the

es = Strain in the steel reinforcement contribution

of the concrete in tension.

It was realised,

however,

that this approach considerably

over estimated of the concrete in

the amount of deflection. tension

This was mainly due to the contribution

was not considered. of the cracked concrete in the tensile zone of flexural in the subsequent methods proposed by various members

The contribution

was taken into consideration investigators in different

approaches. of the beam under

YU and Winter

(84) proposed a method where the curvature

(Iave) based be load of area second moment average on an calculated can working of the beam. The contribution of the concrete between the cracks in the tensile the average second moment of area as follows: (A5)

zone is considered in calculating Iave =1 IMdw In which

"M

= 0.1

(fc) 3

bd (d - Xc)

Consequently the curvature Mdw Ec Iave fc _ Cylinder The derivation triangular

from beam is the calculated of (A6)

strength of concrete. M theory followed the elastic approach. an of equation of compressive

stress distribution

of the concrete in the tensile zone was assumed by at the soffit of the beam and zero was obtained empirically

the stress being equal to the modulus of rupture at the level of the neutral axis.

The factor 0.1, however,

based on normal concrete beams data. In the CEB (79) recommendations, curvature by considering a bi-linear a method was suggested to calculate the for the load-deflection response,

relationship

I. e. taking into account the initial than the cracking the following
_ Mer

uncracked behaviour.

For a moment (M) greater can be obtained from

moment of concrete (Mcr),

the curvature

expression: +M" Mcr

(A7)

ErIo For q<0.25 only

0.75

ESAS d1 (1 - 2q) (1 - 2/31

In which AS bdl f f of the equaticn followed an elastic theory approach. expression of an average effective

The derivation

Branson (83) presented an empirical

second moment of area (Leff) over the entire length of a simply supported uniformly loaded beam. Ieff The expression is given in the following form: 3 3 Icr Mcr Io Mcr +1 m

(A8)

kM

The equation would only apply when M is greater than or equal to Mcr+ otherwise Ieff on an untracked = Io" It was'suggested that the second moment of area (Io) based transformed section might be more accurately used instead of the The method

gorss section especially

when high percentages of steel were employed.

does allow for the contribution

of the concrete along the span length of the beam. taking into consideration in the Tests (Ec)

Beeby (85) derived an expression for the curvature a bi-linear relationship for the load deflection

response and the variation and cracked conditions.

modulus of elasticity

of concrete at the untracked

on normal weight concrete beams showed that a modified modulus of elasticity equal to 0.57 Ec should be used. for the curvature
b'Ic Ec Io

This however was simplified

and the expression

form the in was given - M0.85 Me Ec Icr at mid span of a cracked flexural

(A9)

In CP110 it is suggested that the curvature member can be estimated following of elasticity distribution for the materials

the elastic theory approach, using the modulus A triangular stress

as recommended by the code.

for the concrete in tension is assumed, having a value of 1 N/mm2 and zero at the level of the neutral axis.

at the level of the tension reinforcement

This would allow the average stresses in concrete and steel to be calculated. The curvature consequently can be calculated from the following :-

fca XEC

sa =f (d1 - X) ES

(A10)

Where:

fca and fsa are the average stresses in concrete and steel as calculated. a trial and error approach to determin the neutral axis

The method requires

depth under the working moment while the tension forces balance the compressive forces in the section.

APPENDIX Sustained-Load System

(G)

(i) Calculation

line between beams the distance the the the centre of and of to produce similar moment at mid span of the beams.

supporting trestle

L= .
[ I

4500

GL
F: Force due to weight of the spring assembly = 1.6 KN. w: Weight of the beam per unit length. vvL=1800x0.15x0.3x 10. x4.5x103=3.6KN.
wL+F 22 F

A
Arrangement

+ L/3

WL+F

Upper beam The bending moment 11 at mid span.


M=FL+wL2(wL+F) L

Upper beam r =+F


11

WZ

+F

2826
FL 3 + wL2 24

wL

+F

wL

+F

Lower beam

Lower beam The bending moment at mid span should be equal to that obtained for the upper beam. trestle is X. Assume the distance between the centre line of the beams and the The value of X can be obtained from the following (WL + F) L+ (wL + F) X_ FL 3' equation:

wL2 826 From this equation:


X= L

+ wL2 24

4F+

C2F+wL
wL

Considering
4.5 4

F=1.6
2x1.6 1.6

KN and wL = 3.6 KN.


+ 3.6 )=1.47m + 3.6

The bending moment due to dead weights at mid span of the beams is: 4.5 3 + 3.6 244.5

M =1.6 The ultimate

= 3.077KN.

m 4.3 KN. m =
W

design moment the to dead load is 3.077 x 1.4 calculation: w


F L/3

Lid) Deflection (Macauley's Upper beam

Method)
at

L/3

L/3

.I

,. .1 -. -

T-

__

cv

-,

The differential EI d2a dx2

equation to calculate the deflection is:


M

EI

ww X =2x-2 dx` d--2 rigidity

2a d

(X-

. S)

EI : flexural X: a: w:

distance along the beam deflection total applied load the above equation: da dx 2rW (X -L )2 J C1 +C

By integrating

EI

4L4311

At

g_L, 2 da _ dx

aa = o, d.<
WX2 43

therefore

WL2
18

EI

-W

(X -L

)2

- WL2 18

Integrating
TT

r. t a=

again gives C VvrX3 JfiX3 -W 12 12

(X -L)33]

WL2 18

X+

C2

At

X3,

a=0,

therefore

C2 =

5WL3 324

Hence

EI

a=

WX3 12

-W 12 a1 at X=0

(X is: Since L2 EI X=L 2 L2 is:

L3 3

XWL2 18

5WL3 324

The deflection

EI a1 = 4WL3 324 therefore al =5M 54 a2 at M EI

M=

WL 6

The deflection a2 =1. 72

The total deflection relative

(a1 + a2) is equal to 7.7 times the central deflection This is shown in the following: 54 23 216 x 72 = 7.7

a2 measured

to the supports. 1+5 72

a2+a1 a2 Lower beam Similar

1 72

approach can be employed for the lower beam.

The boundry conditions are At X= L/2 da dx a

=o
=0 to calculate the deflections are the same

At X=

L/3

Hence, the values of the coefficients as those for the upper beams.

APPENDIX Partially Cracked Composite Section

(H)

HI-Neutral

Axis Depth

Xp

b In the calculation partially of the neutral axis depth for a the elastic theory XP

cracked concrete section,

approach is used. An assumption is made by considering an

\.
-t

d dl d2

is to equivalent which area, concrete uncracked the confined concrete by the presence of the f. r. c. channel, below the neutral discussed in 5.3.4. Taking moments of area about the neutral axis level: bXp 4P" (dl - Xp) + ml Ach (d2 - )>p) + bd d/2 axis level. This is
Cross Section

x-

=mA.

For this equation and by substituting p= As , _, Dal pl = Ach pat rl = d2/dl, 1 = d/dl 2 2 (m p+ mi Pi T1 (m p+ mi Pi Tl )2 ,2 Ti

d1

(m. P+

ml Pl 1 )1

I12_ Second Moment

of Area

(IP)

,
bid3 3

The second moment of area for a partially be obtained from the following: Ip = bp) +m AS (dl - Xn)2

cracked composite section can

+ ml

Ach

(d2 - Xp)

2+