EC 2 Prestressed Members Considering The Temperature PDF
EC 2 Prestressed Members Considering The Temperature PDF
EC 2 Prestressed Members Considering The Temperature PDF
Introduction
September 2008
Authorship declaration
1. Where I have consulted the published work of others this is always clearly
attributed.
2. Where I have quoted from the work of others the source is always given.
With the exception of such quotations this dissertation is entirely my own
work.
3. I have acknowledged all main sources of help.
4. If my research follows on from previous work or is part of a larger
collaborative research project I have made clear exactly what was done by
others and what I have contributed myself.
1
Acknowledgements
Two years ago I have been introduced to the prestressed concrete design which led to
the choice of this dissertation. Credit for that teaching goes to Ben Zhang. I thank
also Johnson Zhang, my tutor, for his advice. My gratitude goes to my partner, Amy
Walters for help and patient and finally my parents, Gerard and Françoise Rochefort,
which supported my journey to United Kingdom.
2
Abstract
3
Contents
Authorship declaration 1
Acknowledgement 2
Abstract 3
Contents 4
List of tables 8
List of figures 9
Symbols 12
Introduction 15
1- Basic Explanation 17
1.1- Prestressed tendons can be internal or external 19
1.2- Pre-tensioning and Post-tensioning prestress 19
1.3- Bonded systems 21
1.4- Unbonded systems 21
1.5- Structural behaviour 23
1.6- Mechanism of prestressing force system 25
1.7- The loss of prestress 26
2- The Eurocodes 28
Eurocode 0: Basis of structural design 29
2.1- Basic requirements 29
2.2- Design working life 30
2.3- Representative factors and the combinations value 31
2.4- Combinations of actions 32
2.5- Ultimate limit state(ULS) 32
2.6- Serviceability limit state 34
2.7- Annex A1 and Annex A2 34
2.8- Relevant to prestress concrete inside Eurocode 0 35
2.9- Design values of the effects of actions 36
2.10- Combinations of actions 36
Eurocode 1: Actions on structures 37
2.11- Relevant to prestress concrete inside the Eurocode 1 38
4
Eurocode 2: Design of concrete structures 39
2.12- Familiarisation to Eurocode 2 42
3- Design of Prestress Concrete to Eurocode 2 45
3.1- Prestress 45
3.2- Design process and prerequisite relating to prestress 45
3.3- Load arrangements 46
3.4- Combination of actions 47
3.5- Partial factors for prestress 47
3.6- Material properties 48
3.6.1- Concrete 48
3.6.2- Creep of concrete 50
3.6.3- Shrinkage of concrete 51
3.6.4- Fire resistance 53
3.6.5- Prestressing steel, tendons and strands 53
3.6.6- Relaxation of steel 53
3.6.7- Stress-strain curves for prestressing steel 55
3.6.8- Corrosion of steel 56
3.6.9- Ductility characteristics 57
3.6.10- Design assumptions 59
3.6.11- Prestressing tendons in sheaths 61
3.6.12- Prestressing General devices 61
3.7- Mechanical properties 61
3.7.1- Anchored tendons 61
3.8- External non-bonded tendons, the device in Eurocode 2 61
3.9- Analysis of particular prestressed structural member 63
3.9.1- Generalisation to Eurocode 2 63
4- Temperature effect 65
4.1- Introduction 65
4.2- The Eurocodes and temperature effect 65
4.2.1- Scope to Eurocode 1, part 1.5 65
4.3- Why temperature effect is so important in the design 70
4.4- Response of the concrete to the temperature variation 70
5
4.5- Response of the structural member to
the temperature variation 70
4.6- Solar radiation 71
4.7- Thermal gradient 72
4.8- Temperature range 75
4.9- conclusion 76
5- Creep and Shrinkage effect 77
5.1- Introduction 77
5.2- Creep and shrinkage mechanism 77
5.3- General aspect of the creep an shrinkage 78
5.3.1-The microstructure of hardened pf cement paste 78
5.3.2- Used of mechanics of concrete systems and current approaches
to assessing material behaviour to models creep and shrinkage. 79
5.3.3- Time-dependence 80
5.3.4- Migration function 80
5.3.5- Water function 80
5.3.6- Temperature function 81
5.4- Concrete drying at different humidity and temperatures 81
5.5- Results of measurement and modelling of creep 82
5.6- Physical origins of shrinkage 82
5.7- Creep and shrinkage interrelation 83
5.8- Prediction of cracking at early-age concrete due thermal, drying
and creep behavior 84
5.9- Recapitulative shrinkage 85
5.10- Recapitulative creep 86
5.11- Redistribution of the creep force 87
5.12- Effects of autogenous shrinkage on post-tensioning slab 93
5.13- Modern techniques to improve prestressed concrete
member 94
5.14- Conclusion 95
6- Loss of prestres 96
6.1- Degrees of prestressin 96
6.2- The prestress force 96
6.3-Loss of prestress 98
6
6.3.1-Short-term losses 98
6.3.2- Long-term losses . 129
6.3.3-Total prestress loss 133
7- Basis of design 138
7.1- Introduction 138
7.2- Concrete parameters affecting prestressed concrete 139
7.3- Concrete stress limits 139
7.4- Limit state design 140
7.5- Concrete beam under prestress force only 142
7.6- Deflection 146
7.7- Shear forces 147
7.8- Inequalities of design of the members 151
7.9- Design of prestress force 153
7.10- Magnel diagram 155
7.11- Cable Zone 156
8- Examples 158
8.1- Example 1 159
8.2- Example 2 170
8.3- Example 3 178
Conclusion 181
References 182
Appendix 186
7
List of tables :
Table 1.1: Benefits of prestressing p17
Table 1.2: Advantages and disadvantages of pre- and post-tensioning p20
Table 1.3: Advantages and disadvantages of bonded and unbonded systems p22
Table 1.4: Mechanism of the system members under load p25
Table 1.5: Mechanism of the system members cracking p26
Table 2.1: Indicate design working life p30
Table 2.2: Representation values p31
Table 2.3: Value of y factors for buildings p31
Table 2.4: Design values of actions, ultimate limit state p33
Table 2.5: Design values of actions, derived for UK design, ULS p33
Table 2.6: Combinations of actions p34
Table 2.7: Eurocode 1, its parts and dates publication p37
Table 2.8: Selected bulk density of materials p37
Table 2.9: Eurocode 2, its parts and dates publication p39
Table 2.10: Benefits of using Eurocode 2 p40
Table 3.1: Value of partial factor for prestress p48
Table 3.2: Strength and elastic modulus concrete p49
Table 3.3: Performance of lightweight concrete p50
Table 3.4: Creep coefficient φ to Eurocode 2 p51
Table 3.5.: Eurocode 2 shrinkage strains p52
Table 3.6: Classes of relaxation p54
Table 3.7: Classes of relaxation and K1 and K2 p55
Table 3.8: S-N curves parameters values p59
Table 3.9: Structural classification p62
Table 3.10: Values of minimum cover p62
Table 3.11: Exposure classes related to environment p63
Table 6.1: Prestress loss p98
Table 6.2: Relaxation factors p136
Table 6.3: Global prestress losses p137
Table 7.1: Basic concrete shear strength p149
8
List of figures:
9
Figure 5.6: Flexural deformation under creep and support reaction p91
Figure 5.7: Bending moment and shear force after creep p92
Figure 5.8: Diagrams according to three hypotheses of construction p92
Figure 6.1: Tendon and duct relative position p100
Figure 6.2: Curve reaction and friction p101
Figure 6.3: Successive deviations p102
Figure 6.4: True path of the duct p102
Figure 6.5: Cable geometry p104
Figure 6.6: Force in the cable after friction p107
Figure 6.7: Anchorage draw-in effect p108
Figure 6.8: Integration representing the area p110
Figure 6.9: Part interested by the draw-in p111
Figure 6.10: Calculation of partial area p112
Figure 6.11: Calculation of the area ABCB’A” p113
Figure 6.12: Draw-in between B and C p114
Figure 6.13: Cable section p115
Figure 6.14: Linear strength in the cable p116
Figure 6.15: Effect of a small anchorage draw-in p117
Figure 6.16: Effect of an important anchorage draw-in p117
Figure 6.17: Anchorage draw-in reaching the middle of the beam p117
Figure 6.18: Calculation of an important anchorage draw-in effect p118
Figure 6.19: Significant draw-in for a cable tensioned in one extremity p119
Figure 6.20: Comparison of tensioning by one or two extremity p121
Figure 6.21: Prestress by several cables p122
Figure 6.22: Force in the cable after friction and anchorage draw-in p127
Figure 6.23: Force in the middle cable after none simultaneous tensioning p127
Figure 6.24: Shrinkage evolution p129
Figure 6.25: Force in the cable at t=∞ p137
Figure 7.1: Beam under prestress force only p142
Figure 7.2: Stress in the beam p142
Figure 7.3: Tranversal view p143
Figure 7.4: Beam with tendon in the bottom p144
10
Figure 7.6: Application of the stresses p145
Figure 7.7: Cable eccentricity and stresses in concrete p146
Figure 7.8: Transversal flexural deflection p147
Figure 7.9: Moment in the bottom for one cable p148
Figure 7.10: Moment in the bottom for several cable p148
Figure 7.11: Rectangular beam p149
Figure 7.12: flow chart for uncracked and cracked members p151
11
Symbols
12
strength 1/r Curvature at a particular section
fck Characteristic compressive cylinder 1/rb Curvature at midspan of beam or support of
strength of concrete at 28 days cantilever
fcm Mean value of concrete cylinder 1/rm Effective curvature
compressive strength sv Spacing of links
fctk Characteristic axial tensile strength of SLS Serviceability limit state
concrete T Tension
fctm Mean value of axial tensile strength Td Force in longitudinal reinforcement
of concrete Tf Final cable tension
fcu Characteristic compressive cube To Initial cable tension
strength of concrete at 28 days u Perimeter of concrete cross-section,
fp Tensile strength of prestressing steel having area Ac
fpk Characteristic tensile strength of ULS Ultimate limit state
prestressing steel V Shear force
fp0,1 0,1% proof-stress of prestressing steel Veff Effective shear force in a slab
fp0,1k Characteristic 0,1% proof-stress of VRd1 Shear resistance of section without
prestressing steel reinforcement
ft Tensile strength of reinforcement VRd2 Maximum shear resistance of a section
ftk Characteristic tensile strength of Vsd Ultimate shear force
reinforcement w Uniformly distributed load
13
γm Partial factor of safety for materials σcp Axial stress in the concrete
γp Partial factor of safety for prestressing σcpo Initial stress in the concrete adjacent to the
δ Beam deflection tendons due to prestress
δad Anchorage draw-in σt Stress at top of section
δm Deflection of beam due to end-moments σt,beam Stress at top of beam in composite section
δo Deflection at transfer
σt,slab Stress at top of slab in composite section
δsh Deflection due to shrinkage
ΤRd Basic concrete shear strength
εc Concrete strain
φ Creep coefficient
εcu Ultimate concrete strain
εp Total strain in the tendons
14
Introduction
Different approaches have been proposed to design the prestress concrete, from
Freyssinet the creator of the prestress concrete in 1928, to the young generation of
engineering. Nowadays in Europe the prestress concrete has to be designed with the
latest code, the Eurocodes. The way to design and the philosophy of the design has
not changed since the beginning of prestress concrete but the design has become
more scientifically accurate, economically efficient and all effects are more and more
understandable.
The purpose of this dissertation is to introduce the Eurocodes and to explain the
principles of design for prestressed concrete structures and also to introduce the
thermal, creep and shrinkage effect as it is explained in the Eurocodes and described
by researchers in the past. Finally the dissertation gives some guidance of the design
of prestress structure related to Eurocode 2. .
The context of the dissertation can be divided in several parts, the prestressed
concrete introduction, the Eurocodes, the temperature, creep and shrinkage effects,
the loss of prestress and finally some basic design. The issues of the dissertation is to
find out how to understand different phenomenon happening on the prestress force
and to analysis in the best way, Question such as “What is the effect of the
temperature, creep and shrinkage in the prestress concrete?” will be answered. The
study and observation of the topic have been revising and simplifying to ask the good
question and find the good answer. For each of these matters, it is imperative to be
critics on the past research for the most excellent realisation of the dissertation.
There are three main difficulty for this project. The first one is to read the Eurocodes
and also surrounding text on this matter and to be able to identify the useful
information. The difficulty is to be able to understand the peace of principles and
rules. The second difficulty is to have a complete comprehension and knowledge of
the prestress concrete to undertaken the subject. And finally the third difficulty is on
the effect of temperature, creep and shrinkage to find data and make sure the value of
15
this data are correct and to understand it. It is also part of physics matters less as
structural engineering, than the improvement of the knowledge of the thermal actions
and time-dependence effects are important.
The research on prestressed concrete has never stopped since the discovery of this
technique in the 20th century. Because of the difficulty and the wide surrounding
information needs for understanding completely each part of the prestress concrete
the researcher investigate step by step, every little part to improve, accurate and
familiarise the knowledge of this technology. It still remains unclear the effects of
the variation of the temperature, or the extreme temperature on the structures.
Mechanism as joist has been used for building for the deformation but not real
advancement in the design has been done. Research is difficult in this matter because
you can not create a huge prestress beam and apply cold or/and hot in a laboratory.
The problem as the same with the shrinkage and the creep phenomenon.
16
1- Basic Explanation
The prestressing concrete is today one of the most common ballast used for structure
in building and civil engineering. Elegant and impressive structures have been built,
from bridges across rivers to skyscrapers. The history (see annexe one for
prestresssed concrete history) of prestressed concrete is rich with innovation and
engineering brilliance. There are lists of benefits (see Table 1.1) to use prestressed
concrete. The mechanism and the techniques for the prestressed running are
particular so the prestressed concrete becomes unique technology.
1. The loads on the prestress structures may be more heavy as no prestress structure
(i.e. Reinforced concrete members)
2. Tension and cracking under service loads may be eliminated or reduced, depending
on the magnitude of the prestressing force.
3. Downward deflections of beams and slabs under service loads may be avoided or
greatly reduced.
4. Fatigue resistance (i.e. the ability to resist the effect of repeated live loading due to,
for instance, road and rail traffic) is considerably enhanced.
5. Segmental forms of construction, as in the row of books, become a practical reality.
6. Very high strength steel may be used to form the tendons (see figure 2)
7. Beam and slab sections may be smaller than in reinforced concrete, due mainly to
capacity to reduce deflection.
The concrete has strong resistance to compression but not to in tension. If we apply
the axial compressive force (see Figure 1.1), the concrete is compressed everywhere
and cancels the tension caused by the bending on the beam due to the load P. The
concrete is then to be compressed under the neutral axis.
17
Figure 1.1: How the prestressing works
Axial Load P
Compressive Axial
Force Compressive
Force ACF
The main difference between prestressed or reinforced concrete, it is the steel. The
prestressing steel is active and no passive. The steel used in the prestressed concrete
is called tendon (strand, wire), (see Figure 1.6). The tendons apply loads (see Figure
1.3) to the concrete as a result of their prestress force, while in reinforced concrete
the stresses in the reinforcement result from the loads applied to the structure. A
proportion of the external loads is therefore resisted by applying a load in the
opposite sense through the prestressing whilst the balance has to be resisted by
ordinary reinforcement.
Figure 1.2:Relative unit cost (per unit force) Figure 1.3: Prestressed concrete beam
of reinforcing and prestressing steel with steel tendons
18
- Internal, i.e. within the concrete, either bonded to the concrete or
unbonded (single strand in a plastic tube filled with grease).
- External, i.e. outside the concrete but inside the envelope of the
member.
- The post-tensioned members: it is when the ducts are cast into the
concrete through which the tendons are threaded and then stressed
after the concrete has gained sufficient strength.
Pre-tensioning:
- The prestressing steels are tensioning on a casting bed.
- The concrete poured touches the steel wires.
- The concrete element is then put under stress by cutting the
prestressed wires.
- During the curing stage the tendons bond to the surrounding concrete
and compress the concrete.
19
This technique is often used for the building with precast members with the tendons
bonded to the concrete. The precasting factories technique off site have permanent
casting bed for reducing the cost of the pre-tensioning. Nevertheless, the pre-
tensioned members are limited by transportation requirements and not used for
construction like bridges.
Post-tensionning:
- The tendons are putted by jack at the extremity of the element on the
concrete by anchorage after the concrete is poured and hardened.
- Often the tendons are not inside the concrete, which is external
prestressing of course they are only for the extremity.
20
1.3- Bonded systems (see Figure 1.5):
Bonded tendons are installed in galvanised steel or plastic ducts that are cast into the
concrete, which can be either circular or oval-shaped and change in size for the
different number of steel strands within each duct. Metal ducts are made from either
spirally wound or seam folded galvanised metal strip. The use of plastic ducts should
be considered when designing car parks. The oval duct is used in conjunction with an
anchorage, which ensures once the strands have been stressed the void around the
strands is filled by cementations, which fully bonds the strands to the concrete. The
strands are retained in the same plane in order to achieve maximum eccentricity.
The individual steel strands are encapsulated in a PVC sheath and the voids between
the sheath and the strand are filled with a rust-inhibiting grease that protect the
unbonded tendons. The individual tendons are anchored at each end with anchorage
castings. The sheath and grease are applied under factory conditions and the
completed tendon is electronically tested to ensure that the process has been carried
out successfully. The tendons are cast into the concrete section and are jacked to
apply the required prestress force once the concrete has achieved the required
strength.
Table 1.3 lists the main characteristics of bonded and unbonded systems.
21
Table 1.3: Advantages and disadvantages of bonded and unbonded systems
22
Figure 1.6: Nomenclature associated with prestressed concrete members
Regard as a rectangular concrete member (see Figure 1.7) under load, prestress P and
self weight (see Figure 1.8, 1.9 and 1.10). The whole section of the member is under
stress, everywhere the section is P/Ac. Ac is the area cross-section.
e
P P
23
Figure 1.8: Eccentrically prestressed member
P _ Pe
Ac Zt σt
-
Pe σb
Zb
M
P _ Pe σt
Zt Zt
Ac
- + +
+ + + =
+ -
Pe _M σb
Zb Zb
_ Pe Mo
P
Ac Zt Zt σt
- +
+ + + =
+ - +
Pe _ Mo σb
Zb Zb
1.6- Mechanism of prestressing force system (see Tables 1.4 and 1.5):
24
In order to understand how the magnitude and position of the prestressing force
affects the strength and stability of a beam, the mechanism of the system must be
investigated. Table 1.4 illustrates a member under load and Table 1.5 illustrates the
cracking.
Table 1.4: Mechanism of the systems, members under load
horizontal
Beam with central
Deflected component
point load and
prestress prestress Q =
deflected tendons
Pcosθ
Beam with
horizontal
uniformly
Draped component
distributed load
prestress prestress Q =
and draped
Pcosθ
tendons
25
Craking Definition Scheme
Cracking at top of
unit due to The load-carrying capacity of the unit and
prestressing force the location of the prestressing force are
being located limited by the compressive stresses in the
below the kern concrete and also by tension and cracking
considerations. If tension is to be avoided
entirely in a rectangular section, the force
Cracking at
diagram should be kept within the middle-
bottom of unit
third of the section depth. This region,
due to apex of
which varies with the shape of the cross-
force diagram
section, is called the kern of the section.
rising above the
kern
The prestressing force in the member progressively reduces with time. The loss of
prestress has different causes. The loss can be immediate or short term and time
related or long term. They need to be previewed and included in the calculation of
pretressed concrete.
26
Pj is the jacking force
Po is the force immediately after
transfer
Pe is the final or effective force
Pj-Po is the short-term loss
Po-Pe is the long-term loss Figure 1.11: Loss of prestress
2- The Eurocodes
The Eurocode family started in 1975, There are ten Eurocodes (see
Figure 2.1) covering all main structural materials produced by the
27
European Committee for Standardization (CEN) for replacing the national
standards in 28 countries. I will be concentrate on Eurocode 0, Eurocode
1, Eurocode 2 and because these codes will be used for my design.
Figure 2.1: Eurocodes
BS EN 1991 (EUROCODE 1)
Actions on structures
Actions on structures
BS EN 1992 (EUROCODE 2)
Design of concrete structures
BS EN 1993 (EUROCODE 3)
Design of steel structure
BS EN 1994 (EUROCODE 4)
Design of composite steel
and concrete structures
Structural dign and detailing
BS EN 1995 (EUROCODE 5)
Design of timber structures
BS EN 1996 (EUROCODE 6)
Design of masonry structures
BS EN 1999 (EUROCODE 9)
Design of aluminium structures
BS EN 1997 (EUROCODE 7)
Geotechnical design
Geotechnical and seismic
BS EN 1998 (EUROCODE 8) design
Design of structures for earthquake
resistance
28
and durability of the structures and also the best start for the approach to the
Eurocodes to my project. I have scoped more important points for my project.
Every structure has some needs. For my project I will follow the main basic
requirements explained in the Section 2 of the Eurocode 0.
29
- by the choice of suitable materials,
- by appropriate design and detailing,
- by specifying control procedures for design and use
relevant to the partular project.
7) The provisions of Section 2 of Eurocode 2 should be interpreted
on the basis that due skill and care appropriate to the circumstances
is exercised in the design, based on such knowledge and good
practice as is generally available at the time that the design of the
structure is carried out.
The design life for the structure should be specified (Table 2.1).
Table 2.1: Indicative design working Life ( From NA to BS EN 1990:2002 – Table NA.2.1)
Qk (Table 2.2) the characteristic value of single variable action is the principal
representative value. All the other representative values are combination, which are
obtained by applying the factors ψ 0, ψ 1, ψ 2 to the characteristic value. The factor ψ
30
differ with the type of imposed load and kind of construction ( Example for Building
Table 2.3).
Table 2.2: Representatives values
Table 2.3: Value of y factors for buildings (from UK NA for Eurocode 0 - Table NA.A1.1)
Action ψ0 ψ1 ψ2
Imposed loads in buildings (see BS EN 1991-1-1)
Category A: domestic, residential areas 0.7 0.5 0.3
Category B: office areas 0.7 0.5 0.3
Category C: congregation areas 0.7 0.7 0.6
Category D: shopping areas 0.7 0.7 0.6
Category E: storage areas 1.0 0.9 0.8
Category F traffic area, vehicle weight < 30 kn 0.7 0.7 0.6
Category G: traffic area, 30 kN < vehicle weight < 160 kN 0.7 0.5 0.3
Category H: roof (see BS EN 1991-1-1, Clause 3.3.2(1)) 0.7 0 0
Snow loads on buildings (see BS EN 1991-3)
For sites located at altitude H > 1000 m above sea level 0.7 0.5 0.2
For sites located at altitude H < 1000 m above sea level 0.5 0.2 0
Wind loads on buildings (see BS EN 1991-1-4) 0.5 0.2 0
Temperature (non-fire) in buildings (see BS EN 1991-1-5) 0.6 0.5 0
The combinations of actions are specifically used for the definition of the magnitude
of the actions to be applied when a limit state is under the influenced by different
31
actions. The following process can be used to determine the value of actions used for
analysis:
For one variable action (e.g. imposed load) in a combination, the magnitude of the
actions will be the multiplication between the correct partial factors and the value of
the variable action.
For two or more variable actions in a combination, it needs to identify the leading
action (Qk,1) and the other accompanying actions (Qk,i). The accompanying action is
always taken as the combination value.
The Eurocode lists four ultimate limit state to be considered in the design process:
STR is the design process that will be used in my project. This is because I will
calculate the deformation inside the structural prestressing member (e.g. beam).
The design values of actions (for the UK) used for persistent and transient design
situations under the STR limit state, are listed in Table 2.4 and 2.5 as follows:
32
Table 2.4: Design values of actions, ultimate limit state – persistent and transient design situations (Table A1.2
(B) Eurocode)
Combination Prestress Accompanying variable
Permanent actions Leading variable
Expression ( From table actions
action
reference Unfavourable Favourable 2.4(B)) Main (if any) Others
Exp. (6.10) γGj,supGkj,sup γGj,infGkj,inf γPP γQ,1Qk,1 γQ,i ψ0,iQk,i
Exp. (6.10a) γGj,supGkj,sup γGj,infGkj,inf γPP γQ,1ψ0,1Qk,1 γQ,i ψ0,iQk,i
Exp. (6.10b) ξγGj,supGkj,sup γGj,infGkj,inf γPP γQ,1Qk,1 γQ,i ψ0,iQk,i
Note : Design for either Expression (6.10) or the less favourable of Expressions (6.10a) and (6.10b)
Table 2.5: Design values of actions, derived for UK design, ultimate limit state –
persistent and transient design situations
The expressions (6.10b) can be used when the permanent actions are not superior
than 4.5 times the variables actions. (Except for storage loads (Category E, Table 3)
where Expression (6.10a) always applies). For the typical concrete frame building,
the expression (6.10b) will be the most structurally economical combination of
actions.
33
(0.925 * 1.35 Gk)+ 1.5 Qk = 1.2487 Gk + 1.5 Qk
The eurocode list three combinations of actions at the serviceability state following
table 2.6
Table 2.6: Combinations of actions
Cracking-
Frequent Gkj,sup Gkj,inf P ψι,1Θ1,κ ψι,2Θι,κ prestressing
concrete
Quasi-
Gkj,sup Gkj,inf P ψ2,iQk,1 ψι,2Θι,κ Deflection
permanent
Inside the Eurocode 0 there are two Annexes, the Annex A1 and Annex A2, applied
for buildings and bridges, respectively.
Field of application:
- Rules and methods for the combinations of actions,
- Design values of permanent, variable and accidental actions and ψ factors.
Annex A2: Application for bridges (National Annex for EN 1990 Annex A2)
Field of application:
34
- Rules and methods for the combinations of actions for serviceability and
ultimate limit state verifications,
- Design values of permanent, variable and accidental actions and ψ
factors for every kind of bridge,
- Rules and methods for the verifications relating to some material-
independent serviceability limit states.
NOTE The characteristic values of prestress, at a given time t, may be an upper value Pk,sup(t) and a
lower value Pk,inf(t). For ultimate limit states, a mean value Pm(t) can be used. Detailed information is
given in EN 1992 to EN 1996 and EN 1999.
2.9- Design values of the effects of actions (from Section 6, Clause 6.3.2):
- In those cases where more refined methods are detailed in the relevant EN
1991 to EN 1999 (e.g. for prestressed structures) and reference to the next point.
35
- For non-linear analysis (i.e. when the relationship between actions and their
effects is not linear), the following simplified rules may be considered in the case of
a single predominant action :
a) When the action effect increases more than the action, the partial
factor ΥF should be applied to the representative value of the action.
b) When the action effect increases less than the action, the partial
factor ΥF should be applied to the action effect of the representative
value of the action.
NOTE Except for rope, cable and membrane structures, most structures or structural elements are in
category.
For the representative value of the prestressing action (i.e. Pk or Pm), reference
should be made to the relevant design Eurocode for the type of prestress under
consideration.
Eurocode 1 contains ten parts (Table 2.7) giving details of a wide variety of actions.
It is similar to the British Standards for the anticipated actions used in UK.
36
Table 2.7: Eurocode 1, its parts and dates publication
Publication date
Reference Title
Eurocode National Annex
One of the noteworthy modification is the bulk density of reinforced concrete, which
has been increased to 25 kN/m3(Table 2.8). The National Annexes to Eurocode 1
give the imposed loads for UK.
Table 2.8: Selected bulk density of materials ( from Eurocode 1, Part 1-1, Annex A)
Material Bulk density (kN/m3)
Normal weight concrete 24.0
Reinforced or prestressed normal weight concrete 25.0
Wet normal weight reinforced concrete 26.0
37
a) Actions due to prestressing should be taken into account, including the
effects of interactions between the structure and auxiliary
construction works (e.g. falsework) where relevant and possible
specific requirements defined for the individual project.
38
Eurocode 2 contains four parts (Table 2.9). This new code has many benefits (Table
2.10), but there are some differences between Eurocode 2 and BS 8110 with which
we need to be familiar. Eurocode 2 applies to design of buildings and civil
engineering works. It conforms and combines with the other Eurocodes especially
with EN 1990: Basis of structural design and EN 1991: Actions on structures,
required for resistance, serviceability, durability and fire resistance of concrete
structures.
Part 1-1 of Eurocode 2 gives a general basis for the design of structures in plain,
reinforced and prestressed concrete made with normal and light weight aggregates
together with specific rules for buildings.
Part 2 of Eurocode 2 uses the general rules given in Part 1-1, (i.e. all the clauses for
Basis of design and elements of materials and structural analysis etc …) to the design
of concrete bridges. This part gives a basis for the design of bridges and parts of
bridges in plain, reinforced and prestressed concrete made with normal and light
weight aggregates.
- Learning to use the new Eurocodes will require time and effort on behalf
39
of the designer, so what benefits will there be?
1. The new Eurocodes are claimed to be the most technically advanced codes in
the world.
2. Eurocode 2 should result in more economic structures than BS 8110.
3. The eurocodes are logical and organised to avoid repetition.
4. Eurocode 2 is less restrictive than existing codes.
5. Eurocode 2 is more extensive than existing codes.
6. Use of the Eurocodes will provide more opportunity for designers to work
throughout Europe.
7. In Europe all public works must allow the Eurocodes to be used.
40
Flow chart of process to prestressed concrete
to the Eurocode 2
BS EN 1990 (EC0)
Basics of structure
BS EN 1991 (EC1)
BS EN 1997 (EC7) Actions on structures BS EN 206-1
1991-1-1: General actions Densities, self Concrete: Specification,
Geotechnical design weight, imposed loads for performance, production
BS EN 1998 (EC8) buildings and conformity BS 8500
Seismic design 1991-1-2: General actions Actions on Concrete. Complementary
structures exposed to fire British Standard to BS EN
1991-1-3: General actions Snow loads 206-1
1991-1-4: General actions Wind loads
1991-1-5: General actions Thermal actions
1991-1-6: General actions Construction
loads
1991-1-7: General actions Accidental
loads
41
2.12- Familiarisation to Eurocode 2 (from How design concrete structures using
Eurocode 2):
42
Eurocode, but guidance is given in the background paper to
the UK National Annex10.
13) The punching shear checks are carried at 2d from the face of
the column and for a rectangular column, the perimeter is
rounded at the corners.
43
14) Serviceability checks can still be carried out using ‘deemed to
satisfy’ span to effective depth rules similar to BS 8110.
However, if a more detailed check is required, Eurocode 2
guidance varies from the rules in BS 8110 Part 2.
15) The rules for determining the anchorage and lap lengths are
more complex than the simple tables in BS 8110. Eurocode 2
considers the effects of, amongst other things, the position of
bars during concreting, the shape of the bar and cover.
44
3- Design of Prestress Concrete to Eurocode 2
The design process for the elements covers the design life, actions on structure, load
arrangements, combinations of actions, analysis methods, materials properties,
stability and imperfections, verifications. The particularities of prestressed concrete
make it distinction with reinforced concrete. It explains inside clause 5.10 of
Eurocode 2 Part 1-1.
1. The load arrangements: imposed snow and wind loads acting on a structure.
45
3.3- The Load arrangements:
The variable actions are set up giving the most critical forces in structural members.
There are three options decided by the UK for the national determination parameters
(NDPs). (from national annex to Eurocode 2 Part1-1, Table NA.1)
46
Figure 3.3: All spans loaded
a) In a one-way spanning slab the
area of each bay exceeds 30 m2.
b) The ratio of the variable load Qk to
the permanent load Gk does not
exceed 1,25.
c) The variable load Qk does not
exceed 5 kN/m2 excluding
partitions.
The combinations of actions at ULS and SLS are given in EN 1990, Section 6.This
means that the value of actions to be applied at a limit state is under the influence of
different actions. The annex A1 for buildings and the annex A2 for bridges give the
detail of expressions of the combination of actions. Explain part ‘Eurocodes’ of my
report.
The design value of prestress may be based on the mean value of the prestressing
force given in EN 1990 Section 4.
Prestress in most situations is intended to be favourable and for the ultimate limit
state verification the value of γP,fav (Table 3.1) should be used. In the verification of
the limit state for stability with external prestress, where an increase of the value of
prestress can be unfavourable, γP,unfav (Table 3.1) should be used. In the verification
of local effects γP,unfav (Table 3.1) should also be used.
47
Table 3.1: Value of partial factor for prestress (National Annex, Table NA.1)
The strength of concrete when the transfer of prestress is applied must be adequate,
this means a concrete with a high early strength. The strength of hardened concrete,
which increases with age should be taken at 28 days, for example:
48
- With post-tensioning, the age at transfer is less critical and
accelerated curing is normally not necessary.
fck (Mpa) 12 16 20 25 30 35 40 45 50
fck,cube (MPa) 15 20 25 30 37 45 50 55 60
fctm (MPa) 1.6 1.9 2.2 2.6 2.9 3.2 3.5 3.8 4.1 fctm = 0.30fck^(2/3) < C50/60
Lightweight concrete :
This kind of concrete is often used for the prestressed concrete structures because of
the diverse performances (see Table 3.3).
49
Table 3.3 : Performance of lightweight concrete
Characteristics Advantages
with no information from the aggregate supplier Shrinkage and creep are
the shrinkage = 400-600 * 10-6 and the specific usually greater than
-4
creep = 0.7-0.9 * 10 standard concrete
Note: Ecmt is the elasticity modulus of the concrete in the long-term. φ is a creep coefficient (it
can be obtained from table 3.4, when no more specific data are available). Ac and u are respectively
the cross section area and perimeter.
50
Table 3.4 : Creep coefficients φ to Eurocode 2
Notional size h0 (=2Ac/u) (mm)
50 150 600 50 150 600
The excess of water in the concrete that has not been used to hydrate the cement
evaporates. This is called the shrinks of the concrete member. The quantity of
shrinkage varies on the environmental conditions surrounding the concrete but the
external load on the member does not affect the shrinkage. Shrinkage of concrete
changes with time. In the absence of more detailed data, the long-term shrinkage
strain for normal concrete mixes to be used for design purposes may be funded from
Table 3.5
51
Table 3.5.1 : Eurocode 2 shrinkage strains
Location of Notional size h0 (mm)
Relative humidity %
the member <150 600
It distinguishes the volume contraction by drying shrinkage εcd from the evaporation
of water contain in the concrete and the endogen shrinkage εca. The value of εcd is
function of the followings parameters:
The drying shrinkage has generally for value the data in table 3.52
40 0.48 ‰ 0.31 ‰
80 0.25 ‰ 0.16 ‰
The endogen shrinkage εca (in ‰) is evaluated in infinite time by this formula:
52
Creep and shrinkage strains depend on the type of concrete and its environment. The
effects of creep and shrinkage on loss of prestress force will be discussed in loss of
prestress chapter.
Prestressed concrete works as fire resistant because it is ruled by the loss of strength
of the steel with increase in the temperature rather than loss at the concrete strength.
The Part 10 of Eurocode 2 discusses of the fire resistance requirements. The nominal
cover depends of the periods of fire resistance and type of structural elements.
- Strength, denoting the value of the 0,1% proof stress (fp0,1k) and the
value of the ratio of tensile strength to proof strength (fpk/fp0,1k) and
elongation at maximum load (εuk)
- Class of relaxation,
- Size,
- Surface characteristics.
3.6.6- Relaxation of steel:
53
There are three classes of relaxation in Eurocode 2 (see Table 3.6). The design
calculations for the losses due to relaxation of the prestressing steel should be based
on the value of ρ1000. The relaxation loss (in %) at 1000 hours after tensioning and at
a mean temperature of 20 °C (see Figure 3.4). The long term values of the relaxation
losses may be estimated for a time t equal to 500 000 hours (i.e. around 57 years).
14
Relaxation % of σpo
12
12 Class 2
Class 3
10
Class 1
8
8 7
6 4.5
4 4.5
4
1.5
2 2.5
1
0
50 60 70 80
σpo/fpk %
54
Actually, the stress loss from relaxation in function of time t is illustrate in hour as:
t 0.75×(1−µ)
∆σpr = σpi × k1 × ρ1000 × e (k2×µ) × ( ) × 10 −3
1000
The parameters k1 and k2 depend of the steel class. It is usually the manufacturer
which gives the relaxation rate corresponding to the steel.
Table 3.7: classes of relaxation and K1 and K2 values.
For important structure it is usually and quite exclusively use class 2 because the
strand have low relaxation.
The figure 3.5 shows the stress-strain curves for prestressing steel. There is a
difference between the mild steel and the high-strength steel. The high-strength steel
does not possess the same well-defined yield point and also changed the proof stress.
The 0,1% proof stress (fp0,1k ) and the specified value of the tensile strength (fpk) are
defined as the characteristic value of the 0,1% proof load and the characteristic
maximum load in axial tension respectively, divided by the nominal cross sectional
area as shown in figure 3.6.
Figure 3.5: Stress-strain curves for prestressing steel
Stress N/m m ²
2000
1500
1000
Bar
500 As-Draw w ire
N-R strand
L-R strand
0
0 0.5 1 1.5
Strain %
55
The heat-treating process applied to the as-draw wire does not only reduce the
relaxation of the steel, but also augment the proof stress and consequently extending
the linear elastic range. The Eurocode 2 for the design purposes states the modulus of
elasticity equal at 200 × 103 N/mm² for all types of steel.
Prestressing tendons in sheaths (e.g. bonded tendons in ducts, unbonded tendons etc.)
needs to be adequately and permanently protected against corrosion.(from clauses
3.3.7)
The prestressing steel has to be protected from the attack by moisture permeating the
surrounding concrete.
56
There is also the stress corrosion affecting wires and strands, this is from a
breakdown of the structure of the steel itself. Small cracks emerge and the steel
becomes fragile.
UK NA to EC2 Part1-1, Table NA.2 shows the cover to protect the corrosion of steel
according to the normal weight quality of concrete, diverses classes of exposure and
times.
3.6.9- Ductility characteristics (from Clause 3.3.4) and fatigue (Clause 3.3.5):
If the steel is subject to a lot of stresses variations (between σmin and σmax), it is
demonstrate after a high number of cycles, a steel rupture without the failure stress
being rise. This phenomenon called fatigue. It is quite unknown in the theory but it is
possible to determine the rupture parameters by experimentation.
The average number of cycles necessary to obtain the rupture will be smaller as the
stress variation is bigger and the average stress is high.
Woehler studied the prestressing steel under fatigue effect. He takes a very high
number of sample and he takes the steel tensioned to 0.75 time to its elastic limit.
57
Figure 3.7: Resistance of the prestressing steel under the fatigue effect
The risk of fatigue is practically inexistent durin δσp < 100 Mpa which is when the
prestressed concrete without crack.
The Eurocode (6.8.4) present the model of the damage S-N curve by fatigue with
exponentially form.
58
Table 3.8 S-N Curves parameters
S-N curve of prestressing steel used
stress exponent
for
Pre-tensioning
N* k1 k2 at N* cycles
- single strands in
106 5 9 185
plastic ducts
- straight tendons or
curved tendons in 106 5 10 150
plastic ducts
- curved tendons in
106 5 7 120
steel ducts
The Eurocode states seven points to follow for the design which are:
2) The design value for the modulus of elasticity, Ep may be assumed equal
to 205 GPa for wires and bars. The actual value can range from 195 to
210 GPa, depending on the manufacturing process. Certificates
accompanying the consignment should give the appropriate value.
3) The design value for the modulus of elasticity, Ep may be assumed equal
to 195 GPa for strand. The actual value can range from 185 GPa to 205
GPa, depending on the manufacturing process. Certificates accompanying
the consignment should give the appropriate value.
4) The mean density of prestressing tendons for the purposes of design may
normally be taken as 7850 kg/m3.
59
5) The values given above may be assumed to be valid within a temperature
range between -40°C and +100°C for the prestressing steel in the finished
structure.
6) The design value for the steel stress, fpd, is taken as fp0,1k/ γS (see Figure
3.7).
The value of εud for the UK may be found inside the National Annex table N.A.1,
Subclause 3.3.6 (7). The recommended value is 0,9 εuk. If more accurate values are not
known the recommended values are εud = 0,02 and fp0,1k/fpk = 0,9.
Figure 3.7: Idealised and design stress-strain diagrams for prestressing steel
(absolute values are shown for tensile stress and strain)
60
3.6.11- Prestressing tendons in sheaths (Clause 3.3.7):
Anchorages and couplers apply to the anchoring devices and coupling devices for the
application in post-tensioned construction, where:
- anchorages are used to transmit the forces in tendons to the concrete
in the anchorage zone
- couplers are used to connect individual lengths of tendon to make
continuous tendons.
61
cover of the anchorage should be in accordance with the appropriate European
Technical Approval. The minimum cover values for prestressing tendons in normal
weight concrete taking account of the exposure classes and the structural classes is
given by cmin,dur.
Note: Structural classification and values of cmin,dur for use in a Country may be found in its National
Annex. The recommended Structural Class (design working life of 50 years) is S4 for the indicative
concrete strengths and the recommended modifications to the structural class is given in table 3.7
(from Table 4.3N to Eurocode 2). The recommended minimum Structural Class is S1. The
recommended values of cmin,dur for prestressing steel are given in table 3.10 (From Table 4.5N to
Eurocode 2)
Strength Class
≥ C30/37 ≥ C30/37 ≥ C35/45 ≥ C40/50 ≥ C40/50 ≥ C40/50 ≥ C45/55
(reduce class by 1)
UK National Annex part 161 give cmin dependent the cement/combination types in
the table NA.2.
62
Table 3.11: Exposure classes related to environmental
conditions in accordance with EN 206-1
Swimming pools
XD2 Wet, rarely dry Concrete components exposed to spray industrial waters
containing chlorides
Parts of bridges exposed to spray containing chlorides
XD3 Cyclic wet and dry Pavements
Car park slabs
4 Corrosion induced by chlorides from sea water
Exposed to airborne salt but not in direct
XS1 Structures near to or on the coast
contact with sea water
63
3.9- Analysis of particular prestressed structural member (Clause 5.10):
64
Method D: Provide satisfactory evidence concerning the
reliability of the tendons.
Method E: Ensure that if failure were to occur due to either an
increase of load or a reduction of prestress under the frequent
combination of actions, cracking would occur before the
ultimate capacity would be exceeded, taking account of
moment redistribution due to cracking effects.
Note: The selection of Methods to be used in a Country may be found in the National Annex. The UK
National Annex accord any of the method A to E can be used.
65
4- Temperature Effect
4.1- INTRODUCTION
The temperature effect is major for prestressed concrete because the stress from
thermal actions can modify the volume and/or help to create crack in the concrete.
The temperature appears has different levels of the design of prestressed concrete.
The temperature affects for example the creep and the shrinkage of the concrete
during the early age (explain in the creep and shrinkage review) of the structures but
also affect the structure has long term. The variation of temperature along a beam
called gradient thermal can produce differential stress, producing internal efforts.
More technically the changes in temperature can set off additional deformation and
stresses and if they are significant, it can affect the ultimate and serviceability limit
states of structures. This part of the project describes principles and rules provided by
the Eurocodes and by pass researchers papers.
4.2.1- Scope and fields of application to the Eurocode 1: Part 1-5: General actions _
Thermal actions
This part of the Eurocode presents the principles and rules for calculating and design
thermal actions on buildings, bridges and structural elements. Thermal actions on a
structure or a structural element are those actions that arise from the changes of
temperature fields within a specified time interval, (En 1991-1-5, §1.5.5). This part
of the Eurocode gives moreover a description of the changes in the temperature of
structural elements. In addition this part gives the characteristic values of thermal
actions, which are applied on the design of structures exposed to daily and seasonal
climatic variations. Structures with thermal actions as their main function are also
treated for example chimneys, tank, etc… Furthermore the characteristic value of
isotherms of national extremes of shade air temperatures is given in forms. The shade
air temperature is the temperature measured by thermometers placed in a white
painted louvered wooden box as a “Stevenson screen”.
66
The design philosophy adopted in Eurocode 1: Part 1-5 states that the temperature
distribution in a cross-section leads to a deformation of the element. If the
deformation is restrained, the stresses in the element may occur. Part 1-5 needs load-
bearing actions on structure to make sure the thermal actions do not produce
overstressing into the structural elements. And also to provide expansion joints or to
take account the resulting effects in the design.
67
Temperature changes in bridges
For temperature changes in bridges, Part 1.5 of the Eurocode 1 splits the
superstructures into three different groups. Group 1 is about steel deck on steel box
and truss or plate girders. Group 2 is about concrete deck on steel box and truss or
plate girders and finally group 3 is about concrete slab or concrete deck on concrete
beams or box girders. Part 1-5 advice to take into account the characteristic values of
the vertical linear temperature component, which is given by the over a period of
time, cooling and heating of the upper deck’s surface of the bridge will result a
maximum temperature variation in the top and bottom surface, this mean the thermal
gradient. These values are established inside a normative annex for the assessment of
non-linear thermal actions in bridges.
68
The section basis of design and actions and environmental influences to the
Eurocode 2, (En 1992-1-1, §2.3.1.2) display the requirements for the thermal effects.
The thermal effects must be taken into account during the examination of the
serviceability limit states and be considered for the ultimate limit states, the failure
when they are important (e.g. fatigue condition). The Eurocode 2 necessitates
accounting the thermal effects as variable action. This means a partial factor will be
applied on it.
The thermal actions on the elastic deformation of the concrete consider a linear
coefficient of thermal expansion equal to 10.10-6 K-1for traditional concrete, (En
1992-1-1, §3.1.3 (5)) and 8.10-6 K-1.for lightweight concrete, (En 1992-1-1, §11.3.2).
For the linear elastic analysis, the thermal deformation effect is influences the
ultimate limit state as a reduced stiffness corresponding to the cracked sections,
neglecting tension stiffening but including the effects of creep. It also influencing the
serviceability limit state as a gradual evolution of cracking, (En 1992-1-1, §5.4 (3)).
For the estimation of the deflection, it is considered that if there are thermal action
the flexural tensile strength of the concrete is equal to fctm,fl, (En 1992-1-2, §7.4.3).
In the structural analysis, it explains that beneficial effects of horizontal restraint
produced by friction because of the weight of any supported element can be applied
where the bearing arrangements prevent the possibility of permanent growing sliding
of the elements, such as initiated by irregular behaviour under alternate actions as
cyclic thermal effects on the contact edges of simply supported elements, (En 1992-
1-2, §10.5.1).
The thermal effect is one of the causes for shrinkage, creep, relaxation and
deformation. These effects are all causing loss of prestress. The shrinkage and creep
have with the same kind of requirement as thermal effects (En 1992-1-1, §2.3.2.2)
and also the deformation of concrete due to the action and environmental influences.
(En 1992-1-1, §2.3.3).
69
4.3- WHY THE TEMPERATURE EFFECT IS SO IMPORTANT IN THE DESIGN?
Changes in temperatures may cause additional deformations and stresses and may, in
some cases, significantly affect ultimate and serviceability limit states of structures.
Fundamental principles and rules described in this part of the project provide basic
tools for specifications of temperature changes and for the evaluation of thermal
actions effects in buildings, bridges and industrial structures.
Above the height of a structural member, the temperature distribution may be non-
linear. Those bring on thermal stresses, which occurs during curing and in service.
The relative intensity of the thermal stresses that contribute to the concrete stress
depends on the structure properties and the temperature distribution. Nowadays the
intensity of stress for a number of structural members can be resolved by computing
for example a girder (P. J. Barr; J. F. Stanton; and M. O. Eberhard 2005)[12]
70
The temperature effect is often more pronounced on prestressed concrete beams as
any other structures. The temperature profile through the depth of the beam
(Emerson 1973[28]) can be divided into three components for the purposes of
calculation. (Hambly 1991[29]) The first causes a longitudinal expansion, which is
normally released by the articulation of the structure, the second causes curvature
which leads to deflection in all beams and reactant moments in continuous beams,
while the third causes a set of self-equilibrating set of stresses across the cross-
section. The reactant moments can be calculated and allowed-for, but it is the self-
equilibrating stresses that cause the main problems for prestressed concrete beams.
These beams normally have high thermal mass, which means that daily temperature
variations do not penetrate to the core of the structure. The result is a very non-
uniform temperature distribution across the depth, which in turn leads to significant
self-equilibrating stresses. If the core of the structure is warm, while the surface is
cool, such as at night, then quite large tensile stresses can be developed on the top
and bottom surfaces. However, they only penetrate a very short distance into the
concrete and the potential crack width is very small. It can be very expensive to
overcome the tensile stress by changing the section or the prestress, and they are
normally taken into account by the provision of a mesh of fine bars close to the
surface. A larger problem can arise if thermal stresses act as a trigger for more
damaging cracking, such as the release of locked-in heat of hydration effects which
can occur when a thick web is associated with thin slabs.
71
.
4.7- THERMALS GRADIENTS
[1]
Bridge (Le Delliou, 2004 ) beams are under the phenomenon of outside
temperature. The top part of the beam is subject to the sun effect (solar radiation),
which heats by sunrays. It results in this part being more hot as the surrounding
temperature and the bottom part is colder to the surrounding temperature.
Furthermore the road on the deck is usually in black asphalt and the dark body
attracts the sunrays, which increases the temperature.
sun area
shadow area
Figure 4.1: heats on the bridge deck
Considering a rectangular beam with constant height h and length l. Also taking the
temperature distribution law over the height of the beam being a linear function of
ordinal y. Thermal gradient is the name given to the difference between the
temperature θ of the top and bottom fibre.
72
Under the effect of the temperature the top fibre get longer of δlt = p × l × θt and the
bottom fibre of δlb = p × l × θb, p is the coefficient of dilatation of concrete, about
10-5 K-1.
If the distribution of the temperature is linear, a fibre located from the ordinal y from
the bottom of the beam has a different strain as the bottom fibre of δl (y) = plθy/h.
The variation of elongation being constant in the length of the beam, the beam has a
deformation in arc with the deflection to the top. If the beam is isostatic, the
deformation can be done and do not produce new parasite stress.
However if the beam is hyperstatic, the extra connections block the deformation and
create a new distribution of the force. Example a beam over three supports. Under
the thermal gradient the beam curve and rise up from the middle support.
To oblige the beam to stay in contact with the middle support, it is necessary to apply
a vertical force F directed from top to bottom. This force is balanced by the two force
directed from bottom to top in the exterior support. If the two segments are the same,
R0 and R2 are equal to F/2.
The beam is subject to a bending moment which in the any section located in the left
segment of the beam is equal to M (x) = F × x/2.
73
Figure 4.4: Hyperstatic moments due to thermal gradient
The design of the forces causing by the thermal gradient can be done by using the
following relation:
l l + δl ρ×l×θ
ω'2 = - ω"1 = - =− =
2R 2 × (R + h) 2h
ρ×θ
M1 = 3 × E × I ×
2h
R1 ρ×θ
R0 = R2 = − = 3EI
2 2hl
74
Comments: This phenomenon has been one of the factors (for prestressed concrete)
that in the past made failed bridges. The most influence by this phenomenon is the
box beam.
The Eurocodes gives two cases of thermal load corresponding to linear gradient of
+15°C and –8°C. The forces are normally pondered and added to the ELS
combination.
In reality, the distribution of temperature is not linear but it looks like the figure 4.6:
for box girder.
Studies show that the difference between the linear distribution law and the true
distribution produce stress irrelevant for the concrete and can be ignore.
To design the temperature effect, engineer needs daily, seasonal temperature and also
the thermal expansion coefficient of the concrete and to understand the solar
radiation effect leading to linear expansion and finally the temperature range. It is
casting temperature articulated as a 2 hours indicate ambient temperature, as the rise
over ambient and drop back to it have been taken in consideration for early thermal
contraction. Casting happen at any time and any temperature over 5°C.
75
4.9- CONCLUSION
The combination of the movement discussed above happens at any time and cause
the linear expansion or contraction. Initially, contraction in early days will lead but
creep of concrete relatively relieves it. In the first summer of the structure cast in
winter, the temperature is rising and expansion happens. It is often inconsequential
because the deformation will create only very small movement. But when the
seasons changes and especially for structure cast in summer, drying shrinkage will
occur and temperature will drop and the global effect will be contraction. This is
significant because tension is bringing on and cracking may take place.
There are different ways to study the effect of temperature. The structural member
can be in an extreme temperature, for example very cold in the ice or very hot during
a fire. The structural member can be under a variation of temperature short or long
term and finally the structural member can inside is thickness be under variation of
temperature. Example a chimney for a steel company where outside the temperature
can be low and inside the chimney the temperature can go to 300°c. The temperature
effects over concrete are omitted in the global analysis inside the Eurocode 2, which
have a preference to using other material techniques for this. The temperature effects
over prestressing steel relaxation uses a temperature of 20°c after 1000 hours of
tensioning.
The temperature has a major effect on concrete structure and specially prestressed
concrete. For the reason that are any deformation in prestressed concrete member has
for result to producing other deformations effect such as shrinkage and creep of
concrete but also crack and failure of structural member. The calculation of the
thermal gradient for the temperature distribution in prestressed concrete member is
significant for any structures but mainly for heavy thermal mass for example bridges
decks, which is subject easily to deformation such as deflections and also stresses
when the movement is restrained. Finally I advise the engineer during design to
check the thermal gradient distribution.
76
5- CREEP AND SHRINKAGE EFFECT
5.1- INTRODUCTION
Creep and shrinkage effects are related and they have been studied very often
together. In this part it explains the phenomenon of creep and shrinkage through the
knowledge understood during the last 30 years from the International Symposium on
Fundamental Research on Creep and Shrinkage of Concrete in Lauzanne (September
1980) to the newest studies. It has been considered since long time but the
comprehension phenomenon and the consequence on concrete structure are relatively
recent. It is with the ability of powerful computers that researchers and engineers
handle more easily the mechanism by creation of mathematical model and using
finite element. This part gives the conclusions of the more important research done
by the best specialist of these both effects. It is still not used in current practice
because the international codes and recommendations are relying on simplified
material laws.
77
further crystal growth is delayed and thus creates an internal pressure. The author
gives an explanation for the carbonisation shrinkage. He explains the phenomenon,
as during hydration of cement, there is a reaction with the CO2 from the surrounding
air and that under normal climatic condition this happens in the surface of the
concrete. When the water is liberated and evaporates this creates a decrease of
volume. The last group is the drying shrinkage known as wetting swelling and can be
defined as changes of volume when the moisture contents changes.
Wittmans[44] classified two creep mechanisms. The first mechanism is the short-time
creep, which may be described by a redistribution of water in the microstructure
because of a stress such as capillarity pressure in the structure. The second
mechanism is the long-time creep, which may be described by the displacement of
gel particle when the porous system adsorbed the capillary condensed water.
Studies show that the Creep and shrinkage deformation is shown schematically as a
function of time. When creep and shrinkage appears simultaneously, the deformation
is higher than the sum of creep and shrinkage when they are measured separately but
the interrelation between the both are not well know and have to be more discuss.
Also a number of apparent mechanisms are always involved and they modify the
behaviour.
78
component. These micro structural components are difficult to characterise but that
make the understanding of creep and shrinkage phenomenon better.
One way to comprehend the creep and the shrinkage is to understand the behaviour
of the concrete systems. Dougill[40] has examined the different procedures from a
structural analysis point of view to describe the behaviour. For the study a
conventional stress and strain and material properties are used in the analysis
connected to the significant variables such as time and temperature. The procedures
to describe the concrete behaviour are models, which are developed a range of
phenomena and circumstances to a single material. There are two mains models,
which are the non-linear behaviour, including breakdown, failure, degradation or the
linear models including elasticity, creep, shrinkage and thermal effect.
For this project I overview the linear model called self-consistent models. This model
relies on assumption concerning the average stress or strain. It is believed the stress
average and properties are similar in a single particle as a composite material. These
procedures allow to obtain expression for the moduli and to study elasticity.
Dougill[40] used the self-consistent approach to determine the thermal expansion, the
shrinkage of concrete and the creep. This approach has given the opportunity to fit
physical processes into a simple structural context and to extent the models to a
range of new phenomena. It is also demonstrating that models can be linked together.
This means that the effect on one factor affect the others factors and by consequences
the models get complex. The models must be flexible to let different modes of
behaviour as shrinkage or creep to be included.
79
5.3.3- Time-dependence:
The migrations need the presence of water molecules, without them there is almost
no creep. The rate of creep is higher when the number of particles migrating is higher
which is influence and bigger when the water content is higher. If there is a variation
in the water content (in any direction), a lot of water molecules move to the
80
micropores, and they are causing to the solid particles larger mobility and greater
creep rate. Although that there is no load, the phenomenon will create some creep.
5.3.6- Temperature function
Usually if there are elevated levels of energy it is because of high temperature. This
affects the particles and facilitates to jump over their activation energy barriers, water
solid particles get more mobility when the temperature is higher. Consequently the
effect is greater migrations and greater creep rate at the macroscopic level but this
causing faster aging which one after another decrease the rate of creep.
81
5.5- RESULTS OF THE MEASUREMENT AND MODELLING OF CREEP AND
SHRINKAGE EFFECTS
Many different factors are needed and not only the applied load to understand the
time- dependent deformation of concrete. The factors are usually the strength, the
maturity of the concrete, the moisture and temperature histories experienced by the
concrete. It is also the materials properties of the concrete, aggregate properties, the
size of the concrete structural member. Actually the experimental techniques that are
utilized studies of the shrinkage and creep of concrete must be selected as much for
the way the environmental parameters may be controlled and changed as for the
methods of load control and strain measurement. The measurement of creep and
shrinkage can be done on hardened cement paste because it is the constituent, which
contributes the most to the time-dependent dimensional changes. But also the
measurement can be on concrete in laboratory or on site to understand the actual
behaviour of concrete in a structure. From my point of view this research has to be
taken with precaution because the unknown factors are high and a basic
understanding of the causes of creep is necessary for used wisely the concrete on any
circumstances.
Researcher classified three physical origins of shrinkage. The mechanical actions due
external forces and displacement as boundary condition like settlement. The thermal
actions which are come from in one hand the natural origin (climate) or industrial
origin (thermal treatment to accelerate concrete hardening) and in other hand the heat
produced in the concrete by hydration of the cement. Finally the Hygral actions
causing by the properties of the concrete. The concrete being a porous material,
capillary pressure compressed the solid skeleton.
82
5.7- CREEP AND SHRINKAGE INTERRELATION
Some literature evaluates creep and shrinkage together. In their research for the
evaluation of a basic creep model with respect to autogenous shrinkage, Yun Lee,
Seong-Tae Yi, Min-Su Kim ,and Jin-Keun Kim[32] found that the
following conclusion.
decrease.
83
Torrenti[33] highlight the fact that the water is influencing every single factors of
deformation and also interrelated together the factors in their basic behaviour.
Research has been done on the prediction of cracking within early age due thermal
drying and creep behaviour. To predict the early age cracking, numerical simulation
based on micro mechanical model have been realised to understand the property
development of young concrete. In this research, factors as environment influences,
creep, hydration and moisture have been take into account. The authors takes for is
approach microscopic stress as the water vapour, capillaries and also heat and
[34]
moisture diffusion. The authors (Y.Yuan; Z.L.Wan 2002 ) describes that the
temperature and the humidity distribution can be translated and changed as extra
loads to implant to the structure and that will affect the creep effect and the elastic
analysis of the concrete behaviour such as stress and strain. For the calculi of the
creep the additional loads gives an independent deformation response to the past load
response. This mean, there are a superposition of principal and the traditional
interpretation of the concrete must been subdivided from time interval to time
multisteps.
Definition :
84
Depending the mechanism of concrete shrinkage, which can be a plastic shrinkage,
an autogenous shrinkage (process known as self desiccation), drying of shrinkage
and carbonation of shrinkage.
- Plastic and drying shrinkage are due to the removal of the water from
concrete under the condition of humidity gradient between the interior of
concrete and the air. This two phenomenon are factors for the cracking of
concrete in early age.
Definition of creep:
85
If you test concrete under loads for a strain test, and after you de-loaded the concrete
the strain remaining is the creep strain. The creep strain can be divided into two parts,
basic or drying creep.
- True or basic creep happens when the concrete is under the condition of no
moisture movement to or from the ambient medium.
The concrete is not an elastic material. Under the stress σc the concrete undergoes an
instantly deformation εi and with time a postpone deformation εv which is usually
about twice εi. This propriety is important consequences for the geometry of any
86
structural members. Further this propriety changes the forces in the structures and
can be a problem for by example the structures construct by phases.(Le Delliou 2004
[1]
)
In most of the beam the tendon is close to the bottom fibre. During no loadings, in
quasi-permanent time, the compression stresses of the concrete are small in the top
fibre and strong in the bottom fibre. With time the bottom fibres cut down a lot more
than the top fibres and the beam has a bending deformation in height direction.
If for example we study a beam of 40-meter length and 2.20m high. The beam is
subject to the creep stress only, on the top fibre σctop = 0 and on the bottom fibre
σcbot=14 MPa.
The concrete moduli is Ec = 40000Mpa.
Under the creep effect, the bottom fibre get shorter by δl = 2σcbot × l / Ec however
the top fibre keep the same length.
87
Figure 5.2: Beam deformation under the creep effect
l = αR l - δl = α(R-h)
R =l × h / δl α = δl / l
88
homothetique of his deformation elastic. Considering the creep as a linear
phenomenon and if we ignore the effect of the prestressed differ force. The mistake
is ignorable in most of the cases. The redistribution of forces is definitely lower as
the thermal gradient.
In the case of a beam not constructed to run in its definitive state. The beam is in the
end of its construction, the concrete store under the effect of the self weight and the
prestress, a system of stresses very different (in value and in sign) as the beam would
be working in its definitive task. Under the creep effect, the beam has a deformation
according its stress repartition memory. The deformation that will occur will be
disturbed by the connection from the definitive state, and this will produce new
forces redistribution.
The structure is formed of beam of constant section with segment and fixed in the
central column. The beam is prestressed by a central tendon which do not change the
slope of the stress diagrams and by consequence, do not produce a vertical
deformation of the beam during that the prestress is taking place and during time.
The prestress changes are neglected and there is not creep during construction.
89
At the end of the construction, the beam looks like two consoles in symmetry from
the column.
If on the extremity of the beam is disposed of a support, which initially do not apply
any forces on the beam. If p is the linear density of the self-weigh, the diagrams of
the moment will be:
With time the concrete creep and the beam deform. The deformation causing by the
creep in the left segment of the beam is equal to the double of the elastic deformation
of a console subject to a uniformly distribute load. The deflection at the extremity in
the infinite time has for value:
2 × p × l4
f =
8 × Ec × I
90
Under the creep effect, the right segment of the beam would take the same
deformation as the left segment, but the support prevents the deformation. The
support reaction which was inexistent at the end of the construction are growing and
its value in infinite time is the force R which must be apply to the extremity of the
console to cancel the flexural deflection calculated before.
By equalling the two values of deflection, it is possible to obtain the support reaction
at infinite time.
R × l3 2 × p × l4
=
3 × Ev × I 8 × Ec × I
3 Ev
R= × p×l×
4 Ec
p×l
R=
4
In the left segment, the deformation due to the creep did not change the forces and
the moment diagrams. In the right segment of the beam, the creep increases the value
of the extremity support reaction that was inexistent at the end of the construction.
This is traducing by the important modification of the bending moment and shear
force.
91
Figure 5.7: Bending moment and shear force after creep
The figures illustrate the shear and bending diagrams for the right segment in three
cases:
- Beam at the end of construction
- Beam after creep at infinite time
- Beam poor on hanger in is definitive state
The creep curve line is located between the two others. In any section, the bending is
equal to:
2
M(t = ∞ ) = Mconsole + (Mhanger − Mconsole)
3
92
In this example, the prestress does not influence because the prestress create only the
central normal force. In the case the tendon present an eccentricity, it must be taken
into account in the isostatic deformation. The hyperstatique bending of prestress
happens in the calculation of the hyperstatic structure pour over hanger link.
Comments:
This example is very simplified and has a value of qualitative because:
- the concrete is more complex that the hypotheses (Concrete is not linear and
not reversible) take into account and the concrete depend to lots of
parameters, particularly the concrete strength (depending of the age) when the
loadings;
- the prestress is not constant in time, the prestress decrease because of the
effect of creep, relaxation and shrinkage;
- a part of the creep occurs before the systems are hyperstatic.
The creep being unknown and under estimate have been one of the principal causes
of failures in brides construct by phases. It is usually better to consider the worst
conditions:
- Take into consideration the complexity of the concrete for the theory
calculation
- Take into account the half of the difference between the forces in the
construction without creep and if the structure was poor on hanger link.
- Residual stress in concrete of 1 Mpa and in passive longitudinal steel of 1.5
MPa
There are two major behaviours that happen to any slab with an axial compression
93
- Firstly, shortening caused by an axial compression happens in the slab
additionally to whichever shrinkage of temperature effects. Pouring strips are
used to reduce this effect because it allows short time movement but with
time creep and shrinkage augment as the surrounding humidity is reducing.
- Second behaviour it is the compression in concrete. So crack are more
unlikely to happens directly from either restraint forces but the axial stiffness
will be higher. Consequently any forces apply to a restrain structure should be
more higher.
But this effects can be positive to prevent or reduce cracking in the slab and also
improve the visual point of view or the waterproofing of the structure.
Studies on the long-term behaviour of prestressed concrete members present that the
effect of creep and shrinkage of concrete and also relaxation of prestressing tendons
produce slow changes in the stresses in concrete and prestressing tendons. The long-
term prestress loss is the main problem to the design of prestress concrete members.
In the literature it has been show that using prestressing (fiber reinforced polymer)
FRP tendons has a relevant influence than that when using prestressing steel. It is due
to the lower elastic moduli of FRP tendons and as for consequence in the long-term
changes to decrease the concrete stresses and deflection. The FRP tendons have been
used for few bridges already. The advantages of FRP tendons to conventional steel
tendons are for example the non-corrosive and non-conductive properties,
lightweights and high tensile strength.
94
5.14- CONCLUSION
Research to found using stronger concrete can improved its reaction to the creep and
the shrinkage. In their research for the study on creep and drying shrinkage of high
performance concrete, Li Jianyong and Yao Yan[36] shows that ultrafine reduced the
effect of drying shrinkage and creep. The mechanism would be that ultrafine fill the
pores which means the water is blended with the ultrafine and do not create
additional intern pressure in the structure.
The research on the creep and shrinkage implicate different problem and discipline
and moreover for their interrelationship. The compilation of data and information
from different directions motivate and contribute to a greater understanding of this
both subject. The analysing and the calculation of creep and shrinkage and other
effects such as steel relaxation or cable/ducts friction to Eurocode 2 for prestressed
concrete design are considered in the next part of the dissertation.
95
Part 6- Prestress loss to Eurocode 2
The first idea of prestressing concrete was that all cracking should be avoided under
service load then the section should be entirely in a permanent state of compression.
It is called the “the full prestressing”. Nevertheless the experimentations were still to
control cracking in mix the tensioned steel with untensioned reinforcement steel in
the section. It is called “the partial prestressing”.
The EC2 makes the difference between uncracked (see Figure 7.12) and cracked (see
Figure 7.12) concrete member in tension. Previously some tension may be authorized
if the quantity is kept below the tensile strength of the concrete. Subsequently the
cracks are authorised but must be limited in the width so it does not have an effect on
96
the durability of the member. The EC2 have many clauses for the behaviour and the
analysis of both types of members, which are similar.
1) P at a given time t and distance x (or arc length) from the active end of the
tendon the mean prestress force Pm,t(x) is equal to the maximum force Pmax
imposed at the active end, minus the immediate losses and the time dependent
losses. Absolute values are considered for all the losses.
2) The value of the initial prestress force Pm0(x) (at time t = t0) applied to the
concrete immediately after tensioning and anchoring (post-tensioning) or
after transfer of prestressing (pre-tensioning) is obtained by subtracting from
the force at tensioning Pmax the immediate losses ∆Pi(x) and should not
exceed the following value:
Pm0(x) = Ap . σpm0(x) (1)
97
4) The mean value of the prestress force Pm,t(x) at the time t > t0 should be
determined with respect to the prestressing method. In addition to the
immediate losses given in « 3 » the time-dependent losses of prestress
∆Pc+s+r(x) as a result of creep and shrinkage of the concrete and the long term
relaxation of the prestressing steel should be considered and Pm,t(x) = Pm0(x) -
∆Pc+s+r(x).
The tension in the prestressing steel at one point and at one instant given is different
of the stress apply on the tendon by the jack from when it has taking place. Indeed,
several phenomenon come adds it. There are two types of phenomenon which enter
as part of the loss of prestress. The first is the immediate losses (Short-term losses,
see table 6.1), this happens directly after the transfer of the prestress force to the
concrete member. The second is the losses that happen gradually with time (Long-
term losses).
The losses of prestress are calculated with a lot of approximations. It is then not
appropriate to design with great precision and with a lots of significant figures.
In the tendon, the tension force is maximum when the tension in the jack occurs. The
tension in the jack is limited in the Eurocode to minimum 0.8fpk and 0.9 fp0.1k. The
problems of cable strength under regular loads occur normally at this moment. It is
because of that, there is no verification of the cable strength. The tension force in the
98
cables after anchorage is in addition, limited to the smaller of this two values, 0.75
fpk or 0.85 fpk0.1k.
The Eurocode 2 makes the difference between the pre-and post-tensioned immediate
losses. The immediate losses of prestress for pre-tensioning are from clause 5.10.4 to
Eurocode 2.
Elastic shortening :
The prestress force Po transferred to a structural member (i.e beam) for a
pretensioned member. At the level of the prestressing tendons, the strain in the
concrete must equal the change in the strain of the steel.
In the case where the tendons are closely concentred in the tensile zone the elastic
shortening can be given with sufficient accuracy by taking the stress in the concrete
at the level of the tendons σcg as the stress in the concrete at the level of the centroid
of the tendons. In the case where the tendons are widely distributed throughout the
section, then the above estimation is no longer valid, the influences of the tendons or
groups of tendons ought to be determined individually and then superimposed to give
the total effective prestress force.
Just after transfer of the change in strain in the tendons in the post-tensioned member,
this steel strain may be assumed to be equal to the strain in the concrete at the same
level, even if the ducts have not been grouted and there is no bond between the steel
and concrete. Po is assumed to be constant along the member in spite of reality the
force at the transfer is not constant owing to friction.
At mid span:
fpo M 0e
fcg = −
Ac Ic
m+
(2)
Ap (1 + e / r )
2 2
99
where:
- fpo is Po/ Ap ;
- m is the modular ratio Es/Ecm (where Ecm is based on the long-term value of concrete
strength);
- r is (Ic/Ac) 0.5;
- e is the eccentricity of the tendons.
Friction:
The post-tensioned members are the only concerned by the friction losses. The
friction is situated between the prestressing tendons and the inside of the ducts during
tensioning. The type of tendons and duct former modify the magnitude of the friction.
Curvature friction:
During tensioning, the tendons is tackle against the duct, in the curve part of the path,
the tendon temp to put itself in straight line.
There is a displacement of the tendon under the jack action, it produce a friction
between the tendon and the duct, which reduce the tension in the tendon from the
anchorage. A part of the force P applied by the jack is balance by the friction
100
between A and B and between C and D (see figure 6.1) and is not pass on the
extremity of the tendon.
Consider a curving fraction with a radius r and an angle at the centre dα (see figure
6.2). Be µ the tendon friction coefficient on the duct.
P is the tension in the tendon at the extremity of the curve. By projection, on two
perpendicular direction, the total of forces on the tendon, demonstrates that the
tendon applies on the duct a radial stress p = P/r by length unity and an elementary
friction force by length unity t = µp = µ × P/r.
At the extremity of the curving fraction, the duct friction on the tendon will balance a
force dP such as dP = -µP × dα.
P ( α) = Po × e −µα
Where Po represents the initial tension of the curve trace and P(α) the tension at any
point located in the curve at the angular deviation α from the origin. The angle α is
point given in the tendon, and it is the total angular deviation of the tendon tangent. α
is independent from the exact curve between the origin and a given point, if there is
absence of intermediary inflexion point. The calculation of α does not require a
circular curve. In practise, if the trace is defined by an analytic function, the
calculation of α will be done from the derivation of the tendon curve trace equation.
101
Figure 6.2: Curve reaction and friction
If the tendon present several curve, α represent the total of arithmetic deviation.
PB = PA × e − µ(α1+α 2)
In the case where the tendon has successive curve in both direction, α represent the
total of arithmetic deviation in the both direction between the origin of the tendon
and the point where the calculation of the tension. If the tendon bends, α is
102
calculated equal to the sum of αx and αy, this method over estimates the true
deviation but also major the loss.
Parasites deviations
In the area where the duct is straight, there is in theory no contact between the cable
and the duct so there is no friction. In practise, a duct is not rigorously straight. It can
present some imperfection more or less important according to the execution quality.
The trace is close to figure 6.4.
α1 + α2 + ... + αn
k=
l
The tension P in the straight part of the tendon, calculated at a length x from the
origin is written:
− µkx
P = Po × e
103
Even in the curve part, the tendon trace can change from the hypothesis trace,
because of the imperfection of construction. In this area, it exist some additional
parasites deviation and friction.
The angular deviation α is determinate by the trace of the cable defined by the
projector. The parasite deviation represented by k is unpredictable and dependent
essentially of the nature and the size of the reinforcement and the care gives at the
execution, and the intensity of the vibration of concrete.
The tension P in any point of the cable represents the function the tension at the
anchorage by the formula:
− µ(α + kx)
P = Po × e
Nota :X being the abscise of the section, where p is calculated. In theory, it is the curvilinear abscise
of the tendon. In practise, we take in consideration the abscise in the length of the longitudinal axe of
the beam.
Usually, µ × (α+kx) is small; the tension variation in the cable is a function quasi-
linear to the abscise.
Estimate during the design of the project, the values of µ and k can be controlled
during the execution. It is in fact, a global verification at the beam scale. During the
tensioning, the measures are for the cable strain, elongation, in function of the force
applied to the jack and also of the residual force at the passive extremity of the cable
( measure the transmission coefficient) in one side. With enough different cables
traces, the measure of transmission coefficient allow to determinate acceptable
values, in the static way of parameters µ and k.
104
Tension of a tendon in its two extremity
If a tendon is under tension in its both extremity, it exist a point in the central zone,
which is not subject to any displacement. In this area, the tensions are balancing each
other. In the common case the tendon trace is symmetric, the point is located at the
half of the length. In diverse case, it is located at the abscise such as the loss at the
left would be equal to the loss at the right ( the forces at the jack are supposed
similar).
Example:
In this example, I follow the process of Patrick Le Delliou[1], which used the
Eurocodes to find the losses, this example will carry on with the loss by anchorage
draw-in and by the losses by creep, shrinkage and relaxation phenomenon. The
example has different numerical data but takes the same hypothesis.
Consider a tendon trace is defined by two lean straight segment of 1 m just after the
anchorage, and two parabolic segment connected to one horizontal straight segment
near to the bottom fibre of the beam.
We try to determinate the values of the tension all a long of the tendon with the
following hypotheses:
Po = 700 kN α1 = 0.52 rd
105
µ = 0.18 rd-1 α2 = 0.50 rd
k = 0.008 rd/m
PB = Po × e − µk xb = 699.00 kN
PB' = PB = 699.00 kN
Fixed position D
In D, it is:
so,
α1 + k × xd = α2 + k × (1 − xd)
106
(k × l − α1 + α2) (0.008 × 29 − 0.52 + 0.50)
xd = =
2× k 2 × 0.008
xd = 13.25m
PD = 625.40 kN
The diagrams the tension after friction loses looks like figure 6.6. The minimum is
located in fixed point D. The tension in both extremity is by hypo these equal to the
initial tension in the jack.
A B B' A'
635.2
632
625.4
C'
C
m
D
0 1 6 13.25 24 28 29
Anchorage draw-in:
Anchorage draw-in is the mean from the jack to the anchorage. This is during the
process of transferring the tensioning force a little contraction feel by the prestressing
tendons. The kind of anchorage does depend on the exact quantity of contraction. For
the pretensioning the contraction may be waged by initially over-extending the
107
tendons by the calculated quantity of anchorage draw-in. This effect is much greater
on a short prestressed concrete member than on a long one. For the post tensioning
the effect is reduced by the friction of that exists between the tendons and the ducts
because the tendons return due to the draw-in.
Many anchorage systems can use wedges to grip the tendon and transfer the tendon
force to a solid steel anchorage poor in the concrete. The solid anchorage has some
small deformation and most of the contraction in the length of the tendon takes place
as a result of slip between the tendon and the wedges. For this system the typical
value is 5mm.
Actually during the force transfer from the jack to the anchorage, there are small
movements from the tendon to the concrete. These movements can be inexistent or
neglected by using certain anchorage but for the conic anchorage the movement is
important. The displacement g of the tendon to the interior of the duct is disturb by
the friction forces. At a distance X, the friction forces sum balance the tension loss
due to tendon slide and there are not anymore the displacement of the prestressing
steel in the duct. The squeeze of anchorage do not product loss of tension after the
section located at the abscise x.
The friction duct/tendon is symmetric: There is the same loss of friction when the
tendon moves right to left or left to right.
If it is admitted that the variations of prestress are linear in x, the stresses diagram in
the steel, before and after anchorage draw in can be presented by the symmetric line
as the parallel of the axe x. It is possible also to put in ordinal axe not the tension but
the force in the tendon or in the tendons by multiplied the stress by steel section
appropriate.
108
y
Before anchorage
draw-in
After anchorage
draw-in
In the absence of hypothesis on the linear curve formed, it will be easy to manipulate
the hyperbolic function but this will make very harder and heavier the calculation for
a small finale improvement result.
During the anchorage draw in, a segment of the tendon with a length dx get shorter
of δ dx. It is admitted that the steel behaviour is elastic and follows the Hook law.
The prestress loss in the steel, because of the draw in has for value:
δP δ(dx)
= Ep ×
Ap dx
δp
δ(dx) = dx ×
ApEp
109
Ap is the tendon section and δp is the difference between the force before and after
the anchorage draw-in.
Then, it is evidently:
x
g = ∫ δ (dx)
0
x x
1
A p E p ∫0
So: g= × δP × dx → g A p E p = ∫ δP × dx
0
This integration represents the area between the two forces diagrams, before and
after anchorage draw-in.
P
x
g × A p × E p = ∫ δP × dx
0
X x
Figure 6.8: Integration representing the area before and after anchorage draw-in
110
Generally, the tendon trace is composed of straight segments and curve segments.
The linear tension diagrams before the anchorage blockage is constituted of variable
slope. To do the loss calculation by anchorage draw-in. It must know in what area is
the X abscise, which in the figure 6.9 can be between A and B, or B and C, or after C.
A
B
A'
A" B'
The simple solution will consist to compare gApEp with the surfaces ABA’,
ABCB’A”… In this way, it is possible to determine the segment where is located X.
111
Example:
The cable studied previously has a section equal to 462 mm² (cable 12 φ 7)and it is
subject to an anchorage draw-in to 5mm.
A 700 kN
699 kN
A' B
632 kN
625.4 kN
C
D
A" B'
0 1 2 3 4 5 6 7 8 9 10 11 12 13
g = 0.005 m
Ap = 462 mm
Ep = 200000 Mpa (value from the manufacturer)
112
The area ABA’ is smaller to g Ap Ep. So X is located after the point B.
113
For the calculation of X, it is taken: Y = X – 1 (Y is the distance from the point E
from BB’).
It can be written:
PE - PC 5 − Y
=
PB - PC 5
Y
PE = PB − × (PB - PC )
5
PC = 632kN PB = 699kN
Y
PE = 699 − × (699 - 632)
5
PE = 699 − 13.4 × Y
The area (1), (2) and (3) are for value (in kN)
(1) = PA - PB= 1
(2) = (PB – PE) × Y = 13.4 × Y²
(3) = 2 × (PB – PE) = 26.8 × Y
This second-degree polynomial equation has only one positive result, so Y = 4.95m.
It is possible now to calculate PE, PA’, PB’:
114
PB' = PB − 2 × (PB − PE ) = 566.34 kN
Example:
A symmetric cable of 18 m, which looks like figure 6.13. The cable has for section
Ap =462 mm², it is under a prestress force PA to the anchorage of 700 kN and has the
followings parameters:
PA
PA = 700 kN g = 8 mm
µ = 0.18 rd-1 k = 0.008 rd/m
115
The cable is symmetric, the fix point C is at the middle (x = 9 m).
So,
Before the anchorage draw-in and supposing the linear variation between the
anchorage, the diagrams of tensions is as figure 6.14:
A B
700 kN 700 kN
648.8 kN
m
0 9 18
The cable being tensioned in both extremity, the anchorage draw-in cannot be
possible after the point C. Two cases may happened:
The anchorage draw-in in A does not have repercussion after the point C
116
Figure 6.15: Effect of a small anchorage draw-in
117
Figure 6.17: Anchorage draw-in reaching the middle of the beam
The tension diagrams, after anchorage draw-in is like figure 6.16. The unknown is
not the fix point but the additional loss Pc-Pc’.
where;
118
PA’ = PB’= 700 - 2 × (700 – 648.8) – 30.9 = 566.7 kN
The anchorage draw-in has a repercussion until the point D located at the abscise X
such as:
Y
PD = 648.8 − × (648.8 - 640.44)
9
PD = 648.8 − 0.93Y
The area (1), (2) and (3) are for value (in kN)
119
(1) = Xc × (PA – PC) = 9 × (700 –648.8) = 460.80 kN
(2) = 2 × Xc × (PC – PD) = 2 × 9 × (648.8 – 648.8 – 0.93 Y) = 16.74 Y
(3) = (Y – XC) × (PC – PD) = (Y – 9) × (648.8 – 648.8 – 0.93Y) = 0.93Y²- 8.37Y
This second-degree polynomial equation has only one positive result, so x = 13.38m.
It is possible now to calculate PD, PA’, PB’, PC’:
PD = 636.35 kN
PC' = 2 × PD − PC = 623.90 kN
This tension is bigger as the tension obtained the cable in both extremity.
If a prestressed beam is stress by two similar cables, the figure 6.20 gives the values
of the final tensions in the beam. When the two cables have tension by both
extremities and when the two cables are tensioned once by the left extremity and
once by the right extremity.
120
618 kN
636.4 kN
623.9 kN
601.3 kN
567 kN
+ 567 kN 572.7 kN
+
618 kN
624 kN
636 kN
601 kN
567 kN
= 567 kN = 573 kN
1248 kN
1236 kN
1235 kN
1235 kN
1133.4
1133.4
1174 kN 1174 kN
The bigger value on the right part of the figure 6.20 illustrates that it is better to
tensioned by only one extremity because the loss of prestress
Some engineers consider one more short loss of prestress during the tensioning if
there are several tendons, they are not tensioned at the same time. This problem is a
practical matter from three reasons:
121
- practically, the space of the jack will be so large, this will necessitate
distance between anchorage axe very important;
- technically, for the structure construct by phases, this involves
successive tensioning on the members.
The fact to not tension the tendons of the same beam produce loss of prestress.
Indeed, after the first tendon been tensioned and been in the anchorage, the
tensioning of the second tendon provokes strain in concrete. The first tendon
tensioned is subject to the same deformation and this incite tension reduction. If the
beam has n tendons, the first is subject to n – 1 tendon effect following and so on.
Only the last tendon does not have this effect.
The exact calculation of the prestress loss produce from the non-simultaneity of
tensioning is in theory complex. It must calculate the concrete shorting strain in the
tendon 1 at j when the tendon J+1 is tensioning, and after the tendon j and j+1 when
the tendon j+2 is tensioning, etcetera. In some special case, the tensioning of a
tendon can provoke the elongation to an another tendon tensioned then there are not
loss but adds of tension.
Supposing in the first time we are neglected the self-weight. In the tendons gravity
centre, each tensioning occur the stress δσc on the concrete.
122
P Peo²
δσ c = +
Ac I
The relative concrete shortening supposed to follow Hook law has for value:
δl δσ
= εc = c
l Ec
nota: Ec being the instantaneous concrete moduli
The cables already tensioned are subject to the same relative concrete shortening εp =
-εb and consequently there is losses of stresses:
Ep
δσ p = ε p × E p = −δσ c ×
Ec
The first cable tensioned is subjected to a tension variation because n-1 cables
followings. Stress speaking, the total loss in the first cable will be:
Ep
δσ p1 = −(n − 1) × δσ c ×
Ec
The second cable received the effect from the (n-2) followings, equal to:
Ep
δσ p2 = −(n − 2) × δσ c ×
Ec
123
Ep
δσ p1 = −(n − 1) × δσ c ×
Ec
Ep
δσ p2 = −(n − 2) × δσ c ×
Ec
Ep
δσ p3 = −(n − 3) × δσ c ×
Ec
…………………………………….
Ep
δσ pj = −(n − j) × δσ c ×
Ec
…………………………………….
Ep
δσ p ( n -1) = δσ c ×
Ec
δσ pn = 0
If Ap is the cable section, the total loss δ p from prestressed force for all the cables is
written:
n
n(n − 1) Ep
δP = Ap ∑ δσ pj = − × Ap × δσ c ×
j =1 2 Ec
124
Hence n × δσc represents the concrete stress where the average cable pass, if it will
be tensioned the n cables in the same time.
σc n −1 Ep
δσ p = − × ×
2 n Ec
Where σc = n × δσc correspond to the stress in the concrete after the tensioning of the
n cables supposed simultaneous
The precedent calculation is possible in space. But in reality, the self-weight act on
the beam during the tensioning. The exact calculation of this phenomenon is complex
because it depends of the perfect time when the weight is taking into account and by
consequence incites the structure to be more rigid during tensioning:
- If the hanger is noticeably rigid, the beam raises itself during the
tensioning of the first cable. The concrete weight has no importance
on the tension of the cable (the jack apply a tension and not an
elongation) al well for the others cables.
- If the hanger is noticeably floppy, the concrete is not detach from the
hanger during the end of tensioning. When the hanger is going away
this active the concrete weight which induce effects on the totality of
the cables.
σg is the concrete stress when the concrete is close to the tendon due to the
permanent loads.
125
Mg × e o
σg =
I
The deformation in concrete due to the weigh and is by consequence the deformation
of the cables, is:
σg
δ εp = −δ εb = −
Ec
Ep
δσ p = −σg ×
Ec
If only a full calculation using theory and in the hypotheses it is known the exact
value of the tensioning, It is here admitted that the permanent loads effect applied at
the tensioning is between this two extremes cases. By consequences, the tension loss
in a cable is equal to:
σg Ep
δσ p = − ×
2 Ec
σg is the stress in the concrete under the effect of the permanent loads existing at his
instant.
In most of the case, the permanent load is modified after the tensioning. The
variation matching the force applies itself in whole cables.
Ep n −1 σ
δσ p = − × σ c + cp1 + σ cp2
E c 2n 2
126
σcp1 represent the concrete stress beside the average cable under the effect of the
permanent loads added during the tensioning, and σcp2 is equal to σcp1 with added the
permanent force after the tensioning.
Example
Considering the cable already calculated for the friction and the anchorage draw-in.
After these losses, the tension diagrams in the left part of the cable is equal to:
632.67 kN 632 kN
625.4 kN
E C
D
565.34 kN
566.34 kN
A B
Figure 6.22: Force in the cable after friction and anchorage draw-in
127
σcC = σcE = 7 Mpa
σcD = 9 Mpa
Ep = 200000 Mpa
Ec = 35000 Mpa
There are no permanent loads applied after the tensioning. It is deduct the followings
losses:
5 200000
δσp A = δσp B = − × = −14.29 Mpa
2 35000
7 200000
δσp C = δσp E = − × = −20 Mpa
2 35000
δp C = δp E = −9.24 kN
9 200000
δσp D = − × = −25.71 Mpa
2 35000
δp D = −11.88 kN
The tension diagram in the cable, after all instantaneous losses is:
128
623.43 kN 622.76 kN
E C 613.52 kN
558.74 kN
559.74 kN
A B
Figure 6.23: Force in the middle cable after none simultaneous tensioning
Shrinkage
For the period of concrete hardening, the concrete is subject to a volume shrink
called shrinkage ( see part 3). The shrinkage is not finished when the prestressing
tendons are tensioned. ∆ εs the shorten strain (positive) of a beam causing by the
shrink which appears after the tensioning. The prestressing tendons, associate to the
concrete, are subjected to the same strain, it is follow by a tension loss equal to :
The shrinkage evolution with time has been the object of study in laboratory. Its
value, at the instant t, depend of four major parameters:
- surrounding humidity;
- the ratio W/C in concrete;
- the percentage of steel associated to concrete;
- the sample thickness
The following graphic can represent the shrinkage evolution in function of the time.
129
1.2
0.8
e j/e r
0.6
0.4
0.2
0
0 3 10 90 1 5 10
Days Years
J is in days, ho is the central radius of the sample. in millimetres (ho = 2Ac/U where
Ac is the section and U the perimeter). εcd and εca are formulate is the section 3
(shrinkage).
The part from the endogen shrinkage (εca) rise rapidly its limit value. However it is
not the case with the drying shrinkage, which can be very long. It is notice that the
shrinkage between the pour of concrete and the prestressing do not any effect on the
cables. In the formula giving the shrinkage loss, ∆εr(t) is:
(to) being the concrete age at the time of the prestress take place.
The loss of prestress by shrinkage is the same on the all beam, of course this
hypotheses is not really true but the variation can be neglected.
130
Relaxation
A tendon tensioned at the stress σpi close to the elastic limit and with a length kept
constant, result to the stress diminution with time.
Generally, the relaxation is expressed in function of the stress σpi at the end of the
tensioning, the relaxation ratio at 1000 hours.
The relaxation evolution in function of time can be obtain by the formula indicate in
part 3. It is consider that the calculation at 500000h ( about 57 years) represented the
infinite relaxation (t = ∞). Henceforth we obtained the loss (negative) by relaxation
∆σpr(t).
Creep
The concrete is not a perfect elastic ballast. Subject to a constant stress, the concrete
is under deformation. Usually the deformation is twice in long-term to
instantaneously.
Considering a prestressed beam by tendons with a force P(x), (at the end of
tensioning) with an eccentricity eo and a bending moment from permanent load M(x).
The stress in the concrete, close to the tendon is:
131
P(x) P(x) × e o ² M(x) × e o
σc = + +
Ac I I
Ec being the concrete, the concrete instantaneous deformation close to the cable is
equal to.
σc
εi =
Ec
In time, the concrete has a long-term deformation about the twice to the
instantaneous deformation.
εd= K × εi ≈ 2 × εi
The tendon is in theory after an infinite time, the same additional deformation as the
concrete, equal to εi. The cable strain is like a tension loss δσpc = -2 × εi × Ep:
So,
σc
δσ pc = −2 × Ep
Ec
Shrinkage of concrete, creep in the concrete and steel relaxation was discussed in the
section of the materials properties, and they can be established separately and then
summed. The Eurocode 2 states the following expression:
132
where:
- εsh is estimated shrinkage strain from Table 3.5;
- m is the modular ratio Es/Ecm (where Ecm is based on the long-term value of concrete
strength);
- φ is the creep coefficient from Table 3.4;
- ∆σpr is the variation of stress in the tendons due to steel relaxation(from Figure 3.4);
- σ cg is the stress in the concrete adjacent to the tendons, due to self weight and any
other permanent loads;
- σcpo is the initial stress in the concrete adjacent to the tendons due to prestress;
- e is the eccentricity of the tendons.
This equation is iterative because ∆σpr depends on the final value of prestress but the
examination of Figure 3.4 shows that, with the usual initial stress in tendons of 0.7 fpk.
For long-term losses of more than 15%, the relaxation losses are sensibly constant.
Eurocode 2 giving this very complicate formula for the global loss of prestress hides
the true numerical order of each phenomenon.
The total prestress losses is defined by the initial prestress force applied to a member
Po and by the effective prestress force at transfer αPo and at design load βPo. The
value α indicates the short-term losses due to elastic shortening, anchorage draw-in
and friction, whilst the value β indicates the long-term losses due to concrete creep
and shrinkage and steel relaxation.
133
During the initial design stage it is possible to approximate the prestress loss, which
it will refined later in the design process when more details of the prestressing steel
are available.
It is traditional to take 0.9 and 0.75 for the values of α and β for both pretensioned
and post-tensioned members
Example:
In this example I follow the final steps of Patrick le Delliou[1] process to find the
global loss of prestress. Studying the same cable from above, the tension in the cable
at the end of the construction is equal to figure 6.22. The concrete shrinkage after
tensioning has for value:
∆ε r = 2 × 10 −4
The stress in the concrete beside the cable is equal after instantaneously losses to:
It is taking a cable steel of relaxation class 2. The cable characteristic resistance fpk
is equal to 1660 Mpa.
Whatever the section of prestress concrete the loss is wrote like this:
134
Loss of prestress due to the creep:
Ep Ep
δσp r = −2σσ × δP = −2ApσA×
Ec Ec
The stresses in the steel, after the only instantaneous losses are for value:
PA 558;74 × 10 −3 1209.40
σp A = = −6
= 1209.40 Mpa ⇒ µ = = 0.729
Ap 462 × 10 1660
PB 559;74 × 10 −3 1211.80
σp B = = −6
= 1211.80 Mpa ⇒ µ = = 0.730
Ap 462 × 10 1660
PE 623.43 × 10 −3 1349.41
σp E = = −6
= 1349.41 Mpa ⇒ µ = = 0.813
Ap 462 × 10 1660
PC 622.76 × 10 −3 1347.96
σp C = = −6
= 1347.96 Mpa ⇒ µ = = 0.812
Ap 462 × 10 1660
135
PD 613.52 × 10 −3 1327.97
σp D = = −6
= 1327.97 Mpa ⇒ µ = = 0.799
Ap 462 × 10 1660
At the point A, and t infinite (equivalent to 500000 h), the loss by relaxation
theoretically happening alone is equal to:
t 0.75×(1−µ)
∆σpr = σpi × k1 × ρ1000 × e (k2×µ) × ( ) × 10 −3
1000
500000 0.75×(1−0.729)
δPArel = −σpi × A p × 0.66 × 0.025 × e (9.1×0.729) × ( ) × 10 −3
1000
δPArel = −24.79 kN
It will be the same way for the others, which have for results:
δPBrel = −24.95 kN
δPErel = −40.16 kN
δPCrel = −39.94 kN
δPDrel = −37.14 kN
The long-term effect combination imposes more data. This calculation, effected to
the point A with reasonable values of the different parameters, allow to obtain, ∆σp(t
= ∞) = -136 MPa
136
The simple addition of the differed losses gives: - 64.39/0.462 = - 139 Mpa.
Finally, all losses considered; the prestress forces diagrams by cable are equal to
Figure 6.25.
542.36 kN 541.91 kN
E C
527.56 kN
500.95 kN
501.79 kN
B
A
The table 6.3 summarize the values of instantaneous losses and differed losses in
different part of the section.
To give a numerical order, the total losses are in between 157 and 200 kN, and 22%
to 29% of the force applied at the jack. This percentage may be reduced by
improving the execution, especially the friction and the anchorage draw-in.
137
Part 7- Basis of design
7.1- INTRODUCTION
The prestressed concrete is designed under two critical loading conditions which
occurs where concrete stresses must ensure the specified permissible values. The
initial condition is known as the transfer condition. The concrete is at this point still
moderately young and the concrete compressible strength has not reached its full
design value. The transfer condition happens as soon as there is transfer of the
prestress force to the concrete. At transfer condition the stresses are prestress and
stress due to the moment, Mo, induced by the applied loads (which are present at
transfer). However there is often only the self-weight of the member in that case the
transfer moment is often equal to the self-weigh moment of the member.
The next condition is known as the service (SLS, Service Limit State) condition. The
concrete is at this point mature and has full strength and the member is under a full
service loads. At service condition the applied prestress force has been dropped off
from its initial magnitude, P, due to losses in the concrete and the steel. The total loss
of prestress force reaches 20 per cent between the transfer and the service condition.
138
The stresses taking into account the member in the service condition are the prestress
and the stress from the moment, Ms, provoked by all permanent and variable loads.
The prestress design must assure the limits on tensile and compressive stresses at
transfer and service to guarantee the serviceability of the member. Using tensile
stress limit is to make sure any cracking occur in the member. Also the utilization of
the compressible stress limit is to limit microcraking and avoid extreme loss of
prestress due to creep. In the Eurocode 2, the designer has a freedom to select the
concrete stress limit according to his judgment on the appropriate case of study.
Moreover, the large deformations and loss of prestress from the creep of the
concrete may be prevented. Eurocode 2 suggested that compressive stresses could
not exceed 0.45 fck,0 at transfer and 0.45 fck at service. fck,0 and fck are the
characteristic compressive cylinder strengths of the concrete at transfer and service,
in that order.
139
In the occurrence of fully prestressed members, there is no cracking allowed. The
tensile stress should be limited to the tensile strength of the concrete. According to
the designers the tensile stresses in member will be prohibited at all in the service
condition and evenly tolerate to rise during the transfer condition.
- Strength. The structure has to be able to resist with a suitable factor of safety
against collapse. Different collapses may be possible, as fracture of a single
member, instability of the complete structure, or by buckling of section of the
140
structure. If accidental overloading occurs an adequate factor of safety against
collapse is required while the values are usually lower than that given for the
service loads.
- Fire resistance. The structure should stand for sufficient time to let any
occupants to run away.
- Deflection. Under the service load, the deflection of the structure can not be
excessive, if not damages occurs to finishes, partitions or cladding.
- Cracking. The cracking in the concrete can not only be unsightly but can
conduct to extreme ingress of water into concrete, conducting corrosion of
the steel.
- Durability. There is multiply risk of corrosion of the steel when the concrete
is very permeable. The surrounding environment of the concrete may be
aggressive, such as seawater. In the prestressed concrete the durability is
induced by the concrete mix proportions, the cover around the steel and the
grouting of post-tensioning ducts.
141
According to the factors of the structure, surrounding environment and the loads
acting illustrated by the limit states, the structure should be identified and critically
defined.
The two limit state designs have distinctive degrees of importance. This means that
concrete elements which may have cracking will be affected by unsightly appearance
and durability. However this problem is less serious than collapse of the whole
structure due to the rupture of a critical member. To terminate this point, in limit
state design, the factors used for the serviceability (SLS) are considerably inferior
than factors used for safety (ULS), pointing up the better significance of preventing
collapse of the structure.
Considering a beam in the longitudinal view with a constant section Ac, bent
according to the radius R and prestressing by tendons in the neutral fibre with a force
P. Because of the bending, the tendons strain press on the concrete and develop
centripetal reaction, uniformly spread, with linear intensity of p = P/R.
142
P
σc =
Ac
During the bending effect, this force develops a centrifuge load; uniformly distribute
which have for intensity by length unity of:
q = -Ac × σc = −P = −p
R R
143
There are a risk of cracking along of δ , to guarantee the balance in the system, it
must disposed a link of section As by unity of length of beam:
p
As =
2 × σsadm
Generally P/2 is small and As is equal to zero. Usually this effect creates a stress in
the vertical link.
Considering the same beam but now with the tendon move from the centre to the
bottom and we apply a moment M which reduce the stress in the inferior fibre.
Moving the tendon and adds the moment M do not change the total force on the
section which still equal to P.
144
Figure 7.5: Cable eccentricity and stresses
But the total stress in the concrete and the prestress force are not coaxial. In figure
7.6, d is the distance between the tendons and the total stress. In this case the force is
equal to:
Anyway the equilibrium still remaining because the force in the length have inverse
rapport:
L1 = α × (R + d) and L2 = α × R
145
δ
Figure 7.7: Cable eccentricity and stresses in the concrete
In the transversal view, the force due to concrete is completely located upper the δ
line passing by the tendon. In this case the link must take the total force and they will
equal to:
p
As =
R × σsadm
In the case of beam-box where the bottom part has a curve with the average radius R.
In this case shear force superimpose with the transversal bending over the bottom of
the beam
Considering only one cable, the force by length unity in the bottom of the beam is:
Figure 7.9: Moment in the bottom of the beam for one cable
146
Considering several cables, the force by length unity in the flange of the beam is:
Figure 7.10: Moment in the bottom of the beam for several cables
In the two cases link must be put in the web to balance the reaction from the bottom
flange.
Comment: The phenomenon can be analogue but in the opposite direction if the
bottom flange will not have tendons and will be with only the upward concrete force.
7.6- DEFLECTIONS:
The deflection can be totally eliminated in the prestressed concrete members. This is
realised by the use of a suitable arrangement of prestressing. The CAMBER of
prestressed concrete members is the deflection when no applied load.
Concrete structure cannot predicted the deflections with a high accuracy because
there are to many non-linear factors concerned, for example the concrete has not a
linear stress-strain curve and the load-deflection characteristics of concrete beams are
non-linear in general. The technique to know the exact deflection is to carry out tests
on a model of the structure, using the same materials. The Eurocode 2 recommends
for a structure where the sag of a member would be evident, the deflection under
quasi-permanent load be limited to L/250, where L is the span of a bean or cantilever.
A basic method for finding the maximum deflection of concrete for cracked and
uncracked members is using the shape of the bending moment diagram. The
maximum deflection is given by the following:
147
ymax = KL²×1/rb (4)
Where:
- L is the effective span
- 1/rb is the curvature at midspan, or at the support for a cantilever (1/rb = 1/r1 for
uncracked members and 1/r1 = Mqp/EI)
- Coefficient K can be obtained for simple loading with Table 3.12
Note:
- The deflection for complex loading must use other methods that
follow Clause 7.4.3 checking the deflection by calculation of Part 1 to
Eurocode 2.
0.104
αL
3-4a²
L 48(1-α)
0.25
0.125
148
member is dependent on whether or not the section in the region of greatest shear
force has cracked. The both situations are of the different actual mode of failure.
The Eurocode 2 considers only the cracked shear resistance of prestressed concrete
members. This means there is the action of any longitudinal reinforcement crossing
the shear crack, aggregate interlock across the two faces of the crack and the
contribution of any vertical or inclined shear reinforcement crossing the crack.
It is habitually for beams with distributed loading to establish the shear resistance at
the distance from the face of a support equal to d, the effective depth of the
prestressing tendons.
Where:
- τRd is the basic shear strength taken from table 3.13,
- k is equal to (1.6 – d), where d is not greater than 0.6 m,
- bw is the minimum width of the section,
- ϑ 1 is the tension reinforcement ratio, with a maximum value of 0.02,
defined as (Ap + As) / bwd, where As is the area of untensioned
reinforcement and Ap the area of any non-inclined prestressing
tendons at the section considered,
- σ cp is the axial stress arising from the prestressing force and any
applied axial load.
149
For a section of a prestressed concrete member with shear reinforcement there are
two methods, but only the requirements of what is termed the ‘standard’ method:
This reinforcement may be a mix between the links and inclined bars, but at least
50% ought to be in the form of links.
Where:
- v = 0.7 - fck /200 (v > 0.5)
- If σcp > 0.27 fck, the value of VRd2 should be further reduced to 1.67 (1 – 1.5σcp / fck)
VRd2
If the applied shear force surpasses VRd2 then the section size has to be increased.
If Vsd is less than VRd1 then a minimum amount of shear reinforcement must be
provided and given by the following:
150
- ϑw is the minimum shear reinforcement ratio, found from Table 3.14
The maximum size of links should be 12mm for mild steel and the maximum spacing
should not exceed the smaller of:
The beginning of this design process is to take a simply supported beam (see figure
3.8) carrying a uniform load. P0 and e are respectively the prestress force and the
eccentricity. The midspan stresses at the top and bottom fibres of the beam are at
transfer, under the quasi-permanent and rare load combinations and showed by the
equation given below. In the Eurocode 2, there is one moment at transfer but two
moments at service, the rare load moment and the quasi-permanent load moment.
Thus the inequalities work by five.
e
P
P
151
Figure 7.11: Rectangular beam
Transfer:
α P 0 α P0e M0
σt = − + (9a)
Ac Zt Zt
α P 0 α P 0e M0
σb = + − (9b)
Ac Zb Zb
β P0 β P 0e Mqp
σtqp = − + (9c)
Ac Zt Zt
β P0 β P0e Mra
σtra = − + (9d)
Ac Zt Zt
β P0 β P 0e Mra
σb = + − (9e)
Ac Zb Zb
Where:
- Zt, Zb are the elastic section moduli for the top and the bottom
fibres respectively
- Ac is the concrete cross-section
- α and β are the short- and long-term prestress loss factors
respectively
- These equations assume that the transfer, quasi-permanent and rare
load bending moments, Mo, Mqp and Mra respectively, are sagging
moments.
152
The maximum allowable compressive stresses in the concrete are f’max, (fmax)qp and
(fmax)ra at transfer, quasi-permanent and rare load respectively. The minimum stresses
at transfer and rare load are f’min and fmin respectively. If fmin is negative it ought
indicate an permissible tensile stress. Then the equations can be written as
inequalities:
αP0 αP0e M0
− + ≥ f ' min (10a)
Ac Zt Zt
α P 0 α P 0e M0
+ − ≤ f ' max (10b)
Ac Zb Zb
β P0 β P 0e Mqp
− + ≤(f )
max qp (10c)
Ac Zt Zt
β P0 β P 0e Mra
− + ≤(f )
max ra (10d)
Ac Zt Zt
β P0 β P 0e Mra
+ − ≥ f min (10e)
Ac Zb Zb
In the reality the inequalities (a) and (e) and the inequalities derived from them in the
following sections, are the most critical cases, which is for minimum stress at
transfer and under the quasi-permanent and rare load combinations. Yet for another
member (i.e Double-T section with narrow ribs), the inequalities (b) can occasionally
be critical. The criterion for determining the prestress force and eccentricity for some
sections may be described in the following sections.
By combining Inequalities 9.2(a) and (c), and 9.2 (a) and (d), the expressions for Zt
might be derived:
αMqp − βMo
Zt ≥
α ( f max) qp − βf ' min ( 11a)
153
αMra − βMo
Zt ≥
α ( f max)ra − βf ' min (11b)
And by combining Inequalities 9.2(b) and (e), the expression for Zb might be derived:
αMra − β Mo
Zb ≥
β f ' max − αf min (11c)
These two inequalities for the minimum values of Zt, Zb depend only on the
difference between the maximum and minimum bending moments and allowable
stresses, and not on their absolute values.
At this time the design process is to find the prestress force, based on a maximum
eccentricity determined from the section properties.
( Zt f ' min − M 0)
P0 ≥ (12a)
α ( Z t / Ac − e )
( Zb f ' max + M 0)
P0 ≤ (12b)
α ( Z b / Ac + e )
( Zt ( fmax)qp − Mqp)
P0 ≤ (12c)
β ( Zt / Ac − e)
( Zt ( fmax)ra − Mra )
P0 ≤ (12d)
β ( Zt / Ac − e)
154
( Zt fmin − Mra )
P0 ≥ (12e)
β ( Z b / Ac − e )
These five inequalities have three upper and two lower bounds to the value of the
prestress force. In general the minimum value is required since the cost of the
prestressing steel is a significant proportion of the total cost of prestressed concrete
structures.
The magnel diagram is designed with 1/Po and not with Po. The inequalities for
prestress force are just rearranging. Nevertheless the value e may not be such a range,
since the innermost of the bounds could overlap. In this case another value of e must
be chosen and the limits for Po found again, the process being repeated until a
satisfactory combination of Po and e is found. The inequalities can be given in the
following form:
1 α ( Zt / Ac − e)
≤ (13a)
P0 ( Zt f ' min − M 0)
1 α ( Zb / Ac + e)
≥ (13b)
P0 ( Zb f ' max + M 0)
1 β ( Z t / Ac − e )
≥ (13c)
P0 ( Zt ( fmax)qp − Mqp)
1 β ( Z t / Ac − e )
≥ (13d)
P0 ( Zt ( fmax)ra − Mra )
1 β ( Zb / Ac + e)
≤ (13e)
P0 ( Zb fmin + Mra )
Note:
155
- This inequalities are only valid if the denominators are positive, if
it’s the contrary (a), (c), (d) has been multiplied by a negative
number and the sense of the inequality have to be reversed.
The magnel diagram has been introduced by a Belgian engineer, Magnel. The
relations between 1/Po and e are linear, and if plotted graphically, they offer very
useful meanings to find the appropriate values of Po and e.
Following the choice of the value of the prestress force, the limit of the eccentricity e
now may be found. At this step the term ‘cable’ is used to indicate the resultant of all
the individual tendons. At the same time as the cable lies in the zone the stresses at
the different loading stages will not exceed the allowed values, even though some
tendons physically lie outside the cable zone.
Zt 1
e≤ + (M 0 − Zt f ' min) (14a)
Ac α P 0
1 Zb
e≤ ( M 0 + Zb f ' max ) − (14b)
αP 0 Ac
Zt 1
e≥ + (Mqp − Zt ( f ) )
max qp (14c)
Ac β P 0
Zt 1
e≥ + (Mra − Zt ( f ) )
max ra (14d)
Ac βP 0
1 Zb
e≥ (Mra + Zb fmin ) −
βP 0 Ac
(14e)
156
157
Figure 7.12: Flow chart for uncracked and cracked members
Uncracked Cracked
Imposed load Imposed load
Bending moment and shear force diagrams Bending moment and shear force diagrams
Choose concrete grade and allowable stresses - Choose concrete grade and allowable stresses
Choose min. cover and determine max. e Check Ultimate flexural strength
Determine cable zone and choose cable profile Determine prestress losses
This part presents the designs of three prestressed concrete member to the Eurocodes.
The first example is a post-tensioned prestressed concrete I-beam. This member was
designed to British code and showed by Doctor Ben Zhang during his courses. I
followed the same process but I designed the members following the Eurocode 2.
The second example is a pretensioned uncracked prestressed concrete rectangular
beam. I designed this member following the design process of uncracked members to
the Eurocode 2. These two first example come from my honours project, which was
on prestressed concrete. Most of my calculations are of the losses of prestressed, the
shear force and the deflection. The inequalities and equations used in this section are
from part 4. The example 3 is an example of shear force design, for uncracked area
158
Project: Example 1
School of the Built Environment
Merchiston Campus Job Title: Design post-tensioned beam
Napier University Design of prestressed concrete I-beam
10 Colinton Road Calculated by: Anthony Rochefort
Edinburgh EH10 5DT
159
Design of post-tensioned prestressed concrete I-beam simply supported
over a span of 10.0 meter
Using the design data provided, I will determine the tendons required and
their eccentricity at mid-span(transfer at 7 days) and check the ultimate
moment capacity of the section.
Sectional properties of a prestressed concrete I-beam
b = 500
D F = 150
y T = 320
D = 750
b r = 100
y B = 430
D F = 150
1. b B = 300
Loading:
- dead load (excluding self-weight) = 10kN/m
- superimposed dead load = 6kN/m
- imposed load = 20kN/m
Material Properties:
- Concrete
fck = 50N/mm²
f’ck = 36N/mm²
- Steel
fpk = 1700N/mm²
Short term prestress loss at transfer = 12%
Long term prestress loss = 22%
Project: Example 1
School of the Built Environment
Merchiston Campus Job Title: Design post-tensioned beam
Napier University Design of prestressed concrete I-beam
10 Colinton Road Calculated by: Anthony Rochefort
Edinburgh EH10 5DT
160
No. Calculations Refs/Remarks
2. SOLUTION
Design parameters for the prestressed concrete beams
2.1
For transfer at 7 days, the compressive strength of concrete for C50 is
f’ck = 36 N/mm²
From part allowable stresses, the maximum allowable compressive
concrete stress at transfer, f’max, is
f’max = 0.6 fck = 0.6 × 36 = 21.6 N/mm²
From part allowable stresses, the maximum allowable tensile concrete
stress at transfer, f’min, is
f’min = -3.2 N/mm²
For serviceability state, the compressive strength of concrete, fck, is
fck = 50 N/mm²
From part allowable stresses, the maximum allowable concrete stress under
quasi-permanent load, (fmax)qp, is
(fmax)qp = 0.45 fck = 0.45 × 50 = 22.5 N/mm²
From part allowable stresses, the maximum allowable concrete stress under
rare load, (fmax)ra, is
(fmax)ra = 0.6 fck = 0.6 × 50 = 30 N/mm²
From part allowable stresses, the maximum tensile stress allowed in
The concreta at service is,
fmin = - 4.1 N/mm²
We take : α = 1 – 0.12 = 0.88 β = 1 – 0.22 = 0.78
γ = 24 kN/m3 L = 10 m
Design loads for the prestressed concrete beams
2.2
UDL at transfer due to self-weight of the beam per metre width, wo
wo = γ Ac + wd= 24 × 165×103×10-6 + 10 = 13.96 kN/m
UDL at service due to self-weight and service load per metre width for
quasi-permanent load (wqp), frequent load (wfr) and rare load (wra).
wqp = wo + 0.3 (wimp + wimp,d ) = 13.96 + 0.3 (20+6)= 21.76 kN/m
wfr = wqp + 0.3 (wimp + wimp,d ) = 21.76 + 0.3 (20+6)= 29.56 kN/m
wra = wfr + 0.4 (wimp + wimp,d ) = 29.56 + 0.4 (20+6)= 39.96 kN/m
Bending moment at transfer due to self-weight of the beam, M0
M0 = wo L² / 8 = 13.96 ×10² / 8 = 174.5 kNm
Bending moment at service due to self-weight and service load, for quasi-
permanent load (Mqp), frequent load (Mfr) and rare load (Mra).
Mqp = wd L² / 8 = 21.76 × 10² / 8 = 272.00 kNm
Mfr = wd L² / 8 = 29.56 × 10² / 8 = 369.50 kNm
Mra = wd L² / 8 = 39.96 × 10² / 8 = 499.50 kNm
161
Edinburgh EH10 5DT
No. Calculations Refs/Remarks
162
10 Colinton Road Calculated by: Anthony Rochefort
Edinburgh EH10 5DT
In this case, the distance from the neutral axis to the soffit is 430mm. If
We logically take the distance from the centre of prestressing steel to the soffit as
80 mm, the maximum possible value of eccentricity for the permissible zone is
e = 430 – 80 = 350mm
The corresponding prestressing force will be obtained by
108 / Po = 0.3063 e – 74.351 = 0.3063 × 350 – 74.351 = 32.854 1/N
Po ≤ 108 / 32.854 = 3043.76 KN
163
Napier University Design of prestressed concrete I-beam
10 Colinton Road Calculated by: Anthony Rochefort
Edinburgh EH10 5DT
140
120
100
10^8/pi (1/N)
80
60 Ine (5a)
Ine (5b)
40 Ine (5c)
Ine (5d)
20 Ine (5e)
e taken
0
-50 50 150 250 350 450 550 650
e (m m )
Magnel diagram
2.5 Selection of prestressing steel
I use the BS because there are not more explanation for this case inside
Eurocode 2
BS8110:1997’At transfer the initial prestress should not exceed 70%’
Preq = Po /0.7 = 1058/0.7 = 1511.42kN
Here we assume 4No.7-wire super strands, so the required characteristic load per
strand will be:
Preq / 4 = 1511.42/4 = 377.85kN
Try 4No.7-wire drawn strands of 18 nominal diameter with
Pu = 380kN/strand
With a nominal strength fpk = 1700N/mm² and a cross-sectional area of
Ap = 223mm²
164
Merchiston Campus Job Title: Design post-tensioned beam
Napier University Design of prestressed concrete I-beam
10 Colinton Road
Edinburgh EH10 5DT Calculated by: Anthony Rochefort
= 4.84 + 10.63 – 18.99 = - 3.52 N/mm² > fmin = -4.1 N/mm² OK!
165
Project: Example 1
School of the Built Environment
Merchiston Campus Job Title: Design post-tensioned beam
Napier University Design of prestressed concrete I-beam
0 Colinton Road Calculated by: Anthony Rochefort
Edinburgh EH10 5DT
166
2.8 Bending moment resistance
The initial stress in tendons is fpo = 0.7 fpu = 0.7 × 1700 = 1190 N/mm²
The elastic modulus of steel is known as Ep = 200 kN/mm²
The stress-strain curve for the particular grade of steel used is shown in
Figure with two typical points as follows
- transition point from linearly elastic to non-linear hardening
σp,1 = 0.8 fpu / γm =,0.8 × 1700 / 1.05 = 1295 N/mm²
εp,1 = σp,1 / Ep = 1295/(200×103) = 0.00648
- transition point from non-linear hardening to perfectly plastic
σp,2 = fpu / γm = 1700/ 1.05 = 1619 N/mm²
εp,2 = σp,2 / Ep + 0.005 = 1619/(200×103) + 0.005 = 0.01310
The stress and strain distribution are shown in Figure 4.3. The strain in
prestressing steel at the ultimate limit stat due to prestress only, εpe, is
given by
εpe = 0.8 fpo / Ep =0.8 × 1190/(200×103) = 0.00476
1800
1619
1600
1400
1295
1200
Stress-Strain
Stress Sigma p (N/mm²)
1000
800
600
400
200
0 0.0131
0.00648
0 0.005 0.01 0.015 0.02 0.025 0.03
Strain ep (m m )
Stress-strain diagram
or εp = 0.0035 × (670-x)
x
Project: Example 1
School of the Built Environment
Merchiston Campus Job Title: Design post-tensioned beam
Napier University Design of prestressed concrete I-beam
10 Colinton Road Calculated by: Anthony Rochefort
Edinburgh EH10 5DT
167
No. Calculations Refs/Remarks
D F = 150
x y T = 320
D = 750
b r = 100
350
y B = 430
D F = 150 80
b B = 300
d=670
z
0.00476 εp fpb T
168
Edinburgh EH10 5DT
No. Calculations Refs/Remarks
The stress in the steel can be found from the stress-strain curve and the
tensile and compressive forces in the steel and concrete, T and C, can
then be determined according to the principles for the design of
prestressed concrete sections. Here, we assume 0.9 x > Df, so we
can get
T = fpb Aps = εpb Ep Aps = ( εpe + εp) EP Aps
And C = [b Df + br (0.9 x - Df )] (0.45 fck)
We need to try different values of y or x to make sure T = C first and then
calculate the ultimate bending moment resistance
The following table 4.1 shows these forces for different values of y and x.
when x =209 mm
εp = (670 – 209) × 0.0035/209 = 0.00772
εpb = 0.00476 + 0.00772 = 0.01248
fpb = 1295 +(1619-1295)×(0.01248-0.00648)/(0.01310-0.00648)
jjjjjjjjjjjjjjjjjjjjjjjjj = 1589.0 N/mm2
T = fpb × Ap = 1589 * 0.001115 = 1771.7 kN
C = [b Df + br (0.9 x - Df )] (0.45 fck)
= [500 × 150 + 100 x(0.9 x 209-150)] x (0.45 x 50) x10-3
= 1773.2kN
The neutral axis depth x may thus be taken with sufficient accuracy to be
209mm. In this case, the prestressing steel has not yielded.
Ultimate moment of resistance, Mult may be calculated as
169
10 Colinton Road Calculated by: Anthony Rochefort
Edinburgh EH10 5DT
2.10 Summary
Here. The bending moment resistance, Mult, is larger than the maximum
bending moment due to applied loads, Mmax, by 56%, so the present
design is adequate to resist the ultimate bending moment.
170
Napier University : Design of rectangular uncracked beam
10 Colinton Road Calculated by: Anthony Rochefort
Edinburgh EH10 5DT
Loading:
- Concrete
fck = 40N/mm²
f’ck = 30N/mm²
Ec = 35kN/mm²
- Steel
fpk = 1820N/mm²
links fyk =250N/mm²
Reinforcement fyk = 460N/mm²
171
Merchiston Campus Job Title: Design pretensioned beam
Napier University Design of rectangular uncracked beam
10 Colinton Road
Edinburgh EH10 5DT Calculated by: Anthony Rochefort
Ben Zhang
No. Calculations Refs/Remarks
2. SOLUTION
2.1 Design parameters for the prestressed concrete beams
For transfer at 7 days, the compressive strength of concrete for C40 is
f’ck = 30 N/mm²
From part allowable stresses, the maximum allowable compressive
concrete stress at transfer, f’max, is
f’max = 0.6 fck = 0.6 × 30 = 18 N/mm²
From part allowable stresses, the maximum allowable tensile concrete
stress at transfer, f’min, is
f’min = -2.9 N/mm²
From part allowable stresses, the maximum allowable concrete stress under
rare load, (fmax)ra, is
(fmax)ra = 0.6 fck = 0.6 × 40 = 24 N/mm²
From part allowable stresses, the maximum tensile stress allowed in
The concreta at service is,
fmin = - 3.5 N/mm²
172
Project: Example 2
School of the Built Environment
Merchiston Campus Job Title: Design pretensioned beam
Napier University Design of rectangular uncracked beam
10 Colinton Road Calculated by: Anthony Rochefort
Edinburgh EH10 5DT
173
2.3 Elastic sectional moduli
From Inequalities (11a) and (11b) and (11c), we can obtain the elastic section
moduli
Required about the top and bottom fibres, Zt and Zb, as
174
No. Calculations Refs/Remarks
175
Edinburgh EH10 5DT
No. Calculations Refs/Remarks
140
120
100
10^8/pi (1/N)
80 Magnel diagram
60 Ine (5a)
Ine (5b)
40 Ine (5c)
Ine (5d)
20
Ine (5e)
0 e taken
-50 50 150 250 350 450 550 650
e (m m )
176
10 Colinton Road Calculated by: Anthony Rochefort
Edinburgh EH10 5DT
At support:
fcg = fpo - = 1274 = 4.50mm²
[m + Ac ] [6.25+ 319200 ]
Ap 1155
∆po1 = m× fcgaverage × Ap = 6.25 × (13.27+4.5)/2 × 115510-3 = 64.14N/mm²
Low-relaxation steel:
∆σpr= 38.22 N/mm²
At midspan:
σcg = P0 × (1/AC+e^2/Ic) – Mqp×e/Ic
= 1471.47 103 × (1/319200 + 3502/1.8769 10^10) – 445.5 106 ×350
/1.8769 1010
= 14.21 – 8.30 =5.91N/mm²
σcpo = 14.21N/mm²
Long-term losses = Shrinkage Strain+∆σpr + m × Creep coeff × (σcg+ σcpo)
(1+m × (P0 / Ac) ×(1+Ac×e²/Ic)× (1+0.8 × Creep coeff)
= 6 10-4 × 200 103 +38.22 + 6.25 × 1.5 × (5.91 +14.21)
(1+6.25 × (1471.47/319200)×(1 + 319200×350²/1.8769 10^10)×(1+0.8 ×1.5)
= 346.845 = 290.25N/mm²
1.195
At support:
σcg = σcpo = 14.21 (7-3)/7 = 8.12 N/mm²
177
Napier University Design of rectangular uncracked beam
10 Colinton Road Calculated by: Anthony Rochefort
Edinburgh EH10 5DT
178
Merchiston Campus Job Title: Design pretensioned beam
Napier University Design of rectangular uncracked beam
10 Colinton Road
Edinburgh EH10 5DT Calculated by: Anthony Rochefort
500.00
300.00
Cable zone
100.00
0 1 2 3 4 5 6 7 8 9
-100.00
Série1
-300.00 Série2
Série3
Série4
-500.00 Série5
179
Project: Example 2
School of the Built Environment
Merchiston Campus Job Title: Design pretensioned beam
Napier University Design of rectangular uncracked beam
10 Colinton Road Calculated by: Anthony Rochefort
Edinburgh EH10 5DT
180
= 0.80 × 1471×103 - 0.80 × 1471×103 × 350 + 510.3×106
319200 44.69×106 44.69×106
=3.69 – 9.22 + 11.42 = 5.90 N/mm² < (fmax)ra = 24 N/mm² OK!
From Inequality (1e), the stress at the bottom fb is calculated as
f’b, = βPo + βPoe - Mra
Ac Zb Zb
= 0.80 × 1471×103 + 0.80 × 1471×103 × 350 – 510.3×106
319200 44.69×106 44.69×106
= 3.69 + 9.22 – 11.42 = 1.48 N/mm² > fmin = -3.5 N/mm² OK!
2.8 Shear force:
The effective depth of tendons is equal to 770mm
wult = 1.35×7.6 +1.5×5= 17.76kN/m (Mult = 719.28kNm)
Vsd = wult × ((L/2)- 770×10-3) = 17.96 × (9 –0.77) = 209.92kN
Msd = (wult×770×18×0.77)/103 = 17.96 × 770 ×18 ×0.77/103 = 191.67kNm
Ro =Ap/bwd = 319200/770×380 = 1.09%
k = 1.6 – 0.77 =0.83 v = 0.7 –40/200 = 0.5
σcp = P0 × β1 /Ac = 1471.47 103×0.71 /319200 = 3.15N/mm²
Basic concrete shear strength τRd = 0.41N/mm² (from table13,part 3)
From equation (5)
VRd1 = [ τRdk (1.2 + 40 ϑ1) + 0.15 σcp] bwd
Project: Example 2
School of the Built Environment
Merchiston Campus Job Title: Design pretensioned beam
Napier University Design of rectangular uncracked beam
10 Colinton Road Calculated by: Anthony Rochefort
Edinburgh EH10 5DT
181
No. Calculations Refs/Remarks
2.9 Deflection:
The basic method for finding the deflection of concrete members is using
the shape of the bending moment:
Example 3
500
18 m
Here I going to check when there are no bending crack but when the tensile
stress in the section is high; because it is often the case where the tensile is
high as the bending. This is known has stage two of the shear force check
182
- Check shear resistance of the uncracked web 0,5 m from the support.
Shear resistance of the web based on the principal tensile stress criterion
Vrd ,c = 9085 × 10 6 × 100 / 16.38 × 10 6 (2.732 + 0.611 + 5.93 × 2.73) 0.5 / 1000
where
S = B( D / 2) × ( D / 4) + ( B − bW ) × ( D − tf − 25) 2 / 2
lx= 650 mm
183
To finish to check the member, it must check the shear stress as a cracked (as a
separately reinforced beam) by determine the section the closest to the support,
where the cracking happens, in the bottom of the beam. It is possible at the end to
calculate the compression reinforcement to resist shear force at the cracked section.
It is known as stage one of the shear force check and the calculation above as stage
two of the shear force check.
184
General conclusion
To conclude, the Eurocodes give a good design approach for the prestress concrete
and a good approach of the thermal action but the Eurocodes does not bring the two
together. It is not easily comprehensible. One of the projects for an engineer is to link
the Eurocodes to each other. I hope my dissertation gives a good start for the design
of prestress concrete and thermal action. The creep and shrinkage are analysised in
the Eurocodes, but the both phenomenon are not completely understood.
The key of my dissertation is to open the view of the engineer and explain to them, it
is not because there is a code it is applicable and efficient for all kind of structures
and researchers are trying to improve the knowledge on this point. It is also to make
realise to the engineer that the influence of the temperature is most important.
Finally the prestress concrete is influenced by the temperature and deep research
must be done to understand better and better the relationship between all the effects.
For the future, researchers have to improve the knowledge of the temperature effect
to accurate the design and the Eurocodes.
185
References
1. Hurst M.K. (1998), Prestressed concrete design 2nd ed. E and FN Spon
2. Le Delliou (2003), Béton Précontraint aux Eurocodes. ENTPE PUL
3. Nawy E.G.(1999), Prestressed concrete A fundamental approach 3rded.
Prentice Hakk
4. O’Brien E.J., Dixon A.S. (1995), Reinforced and Prestressed concrete
design. Lonman Scientific and Technical
5. Zhang B.S., Bicanic N., Pearce C.J., Phillips, D.V. (2002),
Relationship between brittleness and moisture loss of concrete
exposed to high temperature. Cement and concrete., pp363-371
6. Zhang B.S., Bicanic N. (2001), Fracture energy of performance
concrete at temperature up to 450°C. Proc. 4th Inter.Conf. on Fract.
Mech. Of Conc. And Conc. Struc. Cachan, France, A.A. Balkema
Publishers, pp 461-468
7. British Standard Institution (2004) BS EN 1992-1-1 General rules and
rules for buildings, BSI.
8. British Standard Institution (2005) BS EN 1992-1-2 General rules-
Structural fire design, BSI.
9. British Standard Institution (2004) BS EN 1991-1-1 Densities, self-
weight and imposed loads, BSI.
10. British Standard Institution (2003) BS EN 1991-1-5 Thermal actions
design, BSI.
11. British Standard Institution (2006) BS EN 1990 Basis of structure
design, BSI.
12. Barr P.J., Stanton J.F., and Eberhard M.O., (2005) Effects of
Temperature Variations on Precast, Prestressed Concrete Bridge
Girders, JOURNAL OF BRIDGE ENGINEERING, Avril, pp.186-194
13. Lier H.P., Wittmann F.H. (1995) Coupled heat and mass transfer in
concrete elements at elevated temperatures [Internet] Elsevier, institute
for Building Materials. Swiss Federal Institute of Technology, CH-
8093 Zirich, Switzerland. Available from:
<http___www.emeraldinsight.com_Insight_ViewContentServlet_Filen
186
ame=_published_emeraldfulltextarticle_pdf_1100200403> [Accessed
6 March, 2008]
14. Powers T.C., (1968) The thermodynamics of volume change and creep,
Materials an Structures: Research and Testing, Vol.1 No, 6, pp. 487-
507
15. Feldman R.F., Sereda P.J., (1968) A model for hydrated Portland
cement paste as deduced from sorption-length change and mechanical
properties. Materials and Structures: Research and Testing, Vol.1,
No.6, pp. 509-520
16. Rusch H., (1959) Physikalische Fragen der Betonprufung. (Physical
problems in testing concrete) Zement-Klk-Gips, Vol.12, No,1, pp. 1-9
17. Jungwirth D., (1970) Determination of time varying creep and
shrinkage deformation. An evaluation of field test and laboratory
results. Sympostum of the Design of concrete structures for creep,
shrinkage and temperature changes.
18. Acker P. and Hassan M., (1978) Prestress loss of thermal action during
production of pretensioned concrete elements, Rep. LPC No, 78,
Laboratoire Central des Ponts et Chaussees, Paris
19. AASHTO, (1989) AASHTO guide specification thermal effect in
concrete brides superstructures, Washington DC
20. Concrete Q & A., (2007) “Estimating Evaporation Rates to Prevent
Plastic Shrinkage Cracking”, Concrete International
21. Menzies JB. and Gulvanessian H., (1998) Eurocode 1 The code for
structural loading, Part 2. DETR, pp.1-8
22. Brown R.D. and al, (1973) The creep of structural concrete, Concrete
Society Technical Paper No.101, pp.1-54
23. Nethercot D.A. and al, (2004) National strategy for implementation of
the structural Eurocodes : Design Guidance, The Instituion of
Structural Engineers, UK.
24. Concrete Society Technical report No,43, (2005) Post tensioned
concrete floors Design handbook, Second Edition, Concrete Society,
Camberley
187
25. Xu, Q and Burgoyne C., (2005) Effect of temperature and construction
sequence on creep of concrete bridges. Proc.Inst. Civ. Engrs, Bridge
Engineering
26. Burgoyne C., (2005) Analysis of continous Prestressed concrete beam.
Proc.Inst. Civ. Engrs, Bridge Engineering
27. Freyssinet E., (1956). Birth of prestressing. Library translatiion 59,
Cement and Concrete Association
28. Magnel G., (1951) Prestressed Concrete Ctatically Indeterminate
Structures, Cement and Concrete Association, pp.77-86
29. Hambly E.C., (1991) Brigde deck behaiour (2 ed.). London
30. Emerson M., (1973) The calculation of the distribution of temperature
in bridges, Technical report, Transport and Road Research Laboratory,
LR561
31. Concrete society (2008), Movement, restraint and cracking in concrete
structures, Technical report No.67, pp. 6-27
32. Yun L., Seong-Tae Y., Min-Su K. Jin-Keun K., (2006) Evaluation of a
basic creep model with respect to autogenous shrinkage, Cement and
concrete research 36, pp.1268-1278
33. Benboudjema F., Meftah F. and Torrenti J.M., (2005) Interaction
between drying, shrinkage, creep and cracking phenomena in concrete,
Engineering Structures 27,pp.239-250
34. Yuan Y. and Wan Z.L., (2002) Prediction of cracking within early-age
concrete due thermal, drying and creep behaviour, Cement and
concrete research 32,pp.1053-1059
35. Pane I. And Hansen W., (2008) Investigation on key properties
controlling early-age stress development of blended cement concrete,
Cement and concrete research, doi:10.1016/j.cemconres.2008.05.002
36. Yan Y. and Jianyon L.I., (2001) A study on creep and drying
shrinkage of high performance concrete, Cement and concrete research
31, pp.1203-1206
37. Tia M., Liu Y. and Brown D., (2005) Modulus of elasticity, creep and
shrinkage of concrete, Final report, U.F. Project No.49104504973-12
188
38. Acker P. and Ulm F., (2000) Creep and shrinkage of concrete:
physical origins and practical measurements, Nuclear Engineering and
Design 203, pp.143-158
39. Young J.F., (1980) The microstructure of hardened Portland Cement
Paste, Creep and Shrinkage in concrete structures, pp. 3-22
40. Dougill J.W., (1980) Mechanics of concrete Systems; Current
Approaches to Assessing Material Behaviour and Some Possible
Extensions, Creep and Shrinkage in concrete structures, pp. 23-50
41. Cinlar E., (1980) Probabilistic Approach to Deformations of Concrete,
Creep and Shrinkage in concrete structures, pp. 51-86
42. Pihlajavaara S.E., (1980) Estimation of Drying of Concrete at
Different Relative Humidities and Temperatures, Creep and Shrinkage
in concrete structures, pp. 87-107
43. Pomeroy C.D., (1980) Experimental Techniques and Results, Creep
and Shrinkage in concrete structures, pp. 111-128
44. Wittmann F.H., (1980) Creep and Shrinkage Mechanisms, Creep and
Shrinkage in concrete structures, pp. 129-162
45. Bazant Z.P., (1980) Mathematical Models for Creep and Shrinkage of
Concrete, Creep and Shrinkage in concrete structures, pp. 163-257
46. Dilger W.H., (1980) Methods of structural Creep Analysis, Creep and
Shrinkage in concrete structures, pp. 305-337
47. Samer A.Y. and Vistasp M.K., (2007)An approach to determine long-
term behavior of concrete members prestressed with FRP tendons,
Construction and buildings Materials 21,pp.1052-1060
189
Appendix
HISTORY OF PRESTRESSED CONCRETE:
The ideas of prestressed concrete is not new. The engineer wanted to find a solution
for the low resistance in tension for the concrete long time ago.
In 1886, P. Jackson (USA) deposited the brevet for compressed the road, and in 1888,
a German, W. Dohring deposited a brevet for prestressing the slab.
In 1907, Koenen and Lundt tested the cracking limitation under tension.
All experience was unsuccessful because the concrete was not of good quality, with a
bad resistance.
In 1919, W.Wettstein used for the first time high strength steel.
In 1927, P. Forber invented a system for the sliding of the prestressing steel during
the tension.
In October 1928, Eugène Freyssinet deposited the First Brevet for the prestressing.
Inside the technology, the steel was able to resist the tension.
In 1929, Freyssinet requested his brevet for the firm Forclum who build electrics
columns.
In 1933, Freyssinet reinforceed with success using prestressing the maritime station
of the Havre.
In 1934, F. Dischinger and U. Finsterwalder gave the basic principle for external
unbonded prestressing.
In 1937-1938, the first systems of external prestressing was used for the bridges.
At that moment the external prestressing was used for contoured the brevets of
Freyssinet.
In 1940, E. Freyssinet found the process of wire and anchorage by concrete cone.
After the Second World War the prestressing systems rapidly extended with the
works of Abeles in Great Britain, Magnel in Belgium, Leaonhardt in Germany, Lin
in United States and Freyssinet and Guyon in France.
190