Prestressed Concrete

BASIC CONCEPTS – PART TWO

STRESS CONTROL BY PRESTRESSING

 Many important features of prestressing can be

illustrated by a simple illustration.
 Consider first the plain, unreinforced concrete beam as
shown in Figure that carries a single concentrated load
at the center of its span. (The self-weight of the member
will be neglected here).
 As the load W is gradually applied, longitudinal flexural
stresses are induced.
 Assuming that the concrete is stressed only within its
elastic range, the flexural stress distribution at mid-span
will be linear. W
h

fr
STRESS CONTROL BY PRESTRESSING

 At a relatively low load, the tensile stress in the

concrete at the bottom of the member will reach
the tensile strength of the material, fr , and a
crack will form.

 Since no restraint is provided against upward

extension of the crack, the member will collapse
without further increase of load.
STRESS CONTROL BY PRESTRESSING

 Now consider an otherwise identical beam, in which a longitudinal

axial force P is introduced prior to the vertical loading.
 The longitudinal prestressing force will produce a uniform axial
compressive stress fc = P/Ac, where Ac is the cross-sectional area
of the concrete.
 The force can be adjusted in magnitude, so that, when the
transverse load Q is applied, the superposition of stresses due to P
and Q will result in zero tensile stress at the bottom of the beam.
 Tensile stress in the concrete may be eliminated in this way or
reduced to a specified amount.

Q
h/2 fc fc 2fc
P P
+ =
ft = fc 0
STRESS CONTROL BY PRESTRESSING

 But it would be more logical to apply the

prestressing force near the bottom of the beam, so
as to compensate more effectively for the load-
induced tension.

A possible design specification might be to

introduce the maximum compression at the
bottom of the member without causing tension at
the top, when only the prestressing force acts.

 It is easily shown that, for a rectangular cross-

section beam, the corresponding point of
application of the force is at the lower third point
of the section depth.
STRESS CONTROL BY PRESTRESSING

 The force P, with the same value as before, but applied with
eccentricity e = h/6 relative to the concrete centroid, will
produce a longitudinal compressive stress distribution varying
from zero at the top surface to a maximum value of
2fc, = (P / Ac) + (P e c2 / Ic.), at the bottom, where fc is the concrete
stress at the section centroid, c2 is the distance from concrete
centroid to the bottom face of the concrete, and Ic is the
moment of inertia of the cross section.
 This is shown in Fig. that stress at the bottom will be exactly
twice the value produced before by axial prestressing.

2Q 0 2fc 2fc
2h/3
P P + =
2fc 2ft = 2fc 0
STRESS CONTROL BY PRESTRESSING

 Post-tensioning method of prestressing uses high

strength steel wires passing through a conduit
embedded in the concrete beam.

 The tendon is anchored to the concrete at one end, and

is stretched at the far end by a hydraulic jack that
reacts against the concrete.

 When the desired tension in the tendon is obtained, it

is anchored against the concrete at the jacking end as
well, and the jack is removed. The result is a self-
contained system by which the force P of previous
Figure may be applied.
STRESS CONTROL BY PRESTRESSING

 If such a system is used, a significant

improvement can be made by using a variable
eccentricity of prestress force with respect to the
centroid of the concrete section along the length
of the member.

 The load 2Q produces a bending moment that

varies linearly along the span from zero at the
supports to maximum at the center.

 Best arrangement of prestressing would produce a

counter-moment, acting in the opposite sense, that
would vary in the same way.
 STRESS CONTROL BY PRESTRESSING

 The prestress moment is directly proportional to the eccentricity of

the tendon, measured from the steel centroid to the concrete centroid.
 The tendon is now given an eccentricity that varies linearly from zero
at the supports to maximum at the center of the span.
 Such an arrangement is shown in Fig.
 The stresses at midspan are the same as before, both when the load
2Q acts and when it does not.
 At the supports, where only the prestress force acts, with zero
eccentricity, a uniform compressive stress fc , is obtained.
0 2fc 2fc
Midspan + =
2Q 2fc 2ft = 2fc 0

P P
fc fc
h/3 Ends + 0 =
STRESS CONTROL BY PRESTRESSING

 For each characteristic load arrangement, there is a

"best" tendon profile in the sense that it produces a
prestress moment diagram that corresponds to that of
the applied load.
 If the prestress counter-moment should be made
exactly equal and opposite to the moment from the
loads all along the span, the result is a beam that is
subject only to uniform axial compressive stress
throughout for that particular loading.
 The beam would not only be free of cracking but also
(neglecting the influence of concrete shrinkage and
creep) would deflect neither up nor down when that
load is in place.
STRESS CONTROL BY PRESTRESSING

Such a situation would be obtained for a load

of ½ x (2Q) = Q, as in Figure. This condition is
referred to as the balanced load stage.

0 fc fc
Midspan + =
Q 2fc ft = fc

P P
fc fc
b/3 Ends + 0 =
 STRESS CONTROL BY PRESTRESSING

 Although this brief discussion has been presented with

reference to the elimination of flexural tension and
control of cracking and deflection in concrete beams, it
should be recognized that prestressing may be used
effectively for many other reasons, such as to reduce
or eliminate diagonal tensile stresses in beams, hoop
tension in liquid storage vessels or pipes, tensile
stresses due to loading or shrinkage in pavements, or
tension from the eccentric loading of columns.

 The fundamental principles are broadly applicable and

provide design engineers with a powerful means to
improve the performance of structures of many types.
 EQUIVALENT LOADS

 The effect of a change in the vertical alignment of a

prestressing tendon is to produce a transverse vertical
force on the concrete member known as Equivalent
Loads.
 This force, together with the prestressing forces acting
at the ends of the member through the tendon
anchorages, may be looked upon as a system of
external forces in studying the effect of prestressing.

 In Fig. 1.8a, for example, a tendon that applies force P

at the centroid of the concrete section at the ends of a
beam and that has a uniform slope at angle θ between
the ends and midspan introduces the transverse force
2Psinθ at point of change in tendon alignment at
midspan.
 EQUIVALENT LOADS

FIGURE 1.8
(Book: Design of Prestressed Concrete by Arthur H. Nilson)
 EQUIVALENT LOADS

 At the anchorages, the vertical component of the

prestressing force is P sinθ and the horizontal
component P cosθ.

 The horizontal component is very nearly equal to

the force P for usual small slope angles.

 The moment diagram for the beam of Fig. 1.8a is

seen to have the same form as that for any center-
loaded simple span.
 EQUIVALENT LOADS

 The beam of Fig. 1.8b, with a curved tendon, is

subject to a transverse distributed load from the
tendon, as well as the forces P at each end.

 The exact distribution of the load depends on the

alignment of the tendon.

 A tendon with parabolic profile, for example, will

produce a uniformly distributed transverse load. In
this case, the moment diagram will have a parabolic
shape, as for a uniformly loaded simple span beam.
 EQUIVALENT LOADS

 If a straight tendon is used with constant

eccentricity e, as in Fig. 1.8c, there are no
transverse forces on the concrete. But the
member is subject to a moment Pe at each
end, as well as the axial force P, and a
diagram of constant moment results.
 EQUIVALENT LOADS

 The end moment must also be accounted for in

considering the beam of Fig. 1.8d, in which a parabolic
tendon is used that does not pass through the concrete
centroid at the ends of the span.

 In this case, a uniformly distributed transverse load and

end anchorage forces are produced, just as in Fig. 1.8b.
But, in addition, the end moments M = P e cos θ must be
considered.
 EQUIVALENT LOADS

 The concept of equivalent transverse loading is a

useful one, but it must be applied with care.
 In all of the cases considered thus far, the profile of the
concrete centroid was straight.
 The concrete thrust was horizontal, consequently, and
any change in alignment of the tendon produced an
unbalanced force acting on the concrete at that section.
 If the beam axis is curved, as in Figs. 1.8e and 1.8f, and
if the tendon and concrete centroid coincide at all
sections of such beams, the lateral force produced by
the steel at any section is balanced by a resultant force
in the opposite direction produced by the thrust from
the adjacent concrete, and no bending moment results.
EQUIVALENT LOADS
 EQUIVALENT LOADS

On the other hand, if the tendon is

straight but the concrete centroidal
axis has some other alignment, as in
Fig. 1.8g, then the lateral force
produced by the concrete thrust is
not balanced by the lateral forces
from the steel, and bending moment
is produced as shown.
 EQUIVALENT LOADS

 For any arrangement of applied loads, a tendon

profile can be selected such that the equivalent
loads acting on the beam from the tendon are just
equal and opposite to the applied loads.

 The result would be a state of pure compression in

the beam.

 An advantage of the equivalent loading concept is

that it leads the designer to select what is probably
the best tendon profile for any given load
configuration.
 EQUIVALENT LOADS

 All of the systems shown in Fig. 1.8 are self-

equilibrating, and the application of prestressing
forces produce no external reactions.

 This is always true for statistically determinate

beams, but is not generally true for
indeterminate spans.
 OVERLOAD BEHAVIOR AND STRENGTH IN
FLEXURE

 In describing the effect of prestressing, it was implied that

the beam responded in a linear elastic way and that the
principle of superposition was valid.
 This requires that the beam remain uncracked and that both
the concrete and steel be stressed only within their elastic
ranges.
 This may be the case up to approximately the level of service
load, that is, the actual self-weight of the member plus those
superimposed loads that may reasonably be expect act
during the life of the member.
 But should the loads be increased further, tensile stresses
resulting from flexure will eventually exceed the tensile
strength of the concrete, and cracks will form.
 These cracks do not cause failure, because of the presence
of the steel, and the loads generally can be increased well
beyond the cracking load without producing distress.
 OVERLOAD BEHAVIOR AND STRENGTH
IN FLEXURE

 Eventually, with loads increased still further,

either the steel or the concrete or both, will be
stressed into their nonlinear range.

 The condition at incipient (early) failure is

represented by Fig. 1.9, which shows a beam
carrying a factored load equal to some multiple
of the expected service load.

 In designing a member magnitude of the load

factor can be selected to provide the desired
design safety.
 OVERLOAD BEHAVIOR AND STRENGTH IN
FLEXURE

 For the overloaded condition, the beam

undoubtedly would be in a partially cracked
state; a possible pattern of cracking is shown
in Fig. 1.9.
 Only concrete in compression is considered to
be effective, just as in the analysis of ordinary
reinforced concrete.
 The steel in tension works with the concrete
compression to form an internal force couple,
which resists the moment from applied load.
OVERLOAD BEHAVIOR AND STRENGTH IN
FLEXURE

The concrete stress distribution in the

compression zone at failure can be shown
as the magnitude of the compressive
resultant C, the tensile force T in the steel,
and the distance between the two. If the
internal lever arm is z then the ultimate
resisting moment is

Mn = Cz = Tz (1.1)
OVERLOAD BEHAVIOR AND STRENGTH IN
FLEXURE

 At the ultimate load stage, when the beam is at

the point of incipient failure in flexure, it
behaves very much as an ordinary reinforced
concrete beam.
 The main difference is that the steel used has
very high strength and requires a very large
strain to achieve a high stress level.
 If it (steel) were to be used without being
prestressed (and prestrained) in tension,
unacceptably large deformation and cracking
of the beam would result.
 PARTIAL PRESTRESSING

 Early designers of prestressed concrete focused

on the complete elimination of tensile stresses in
members at normal service load. This is defined as
full prestressing.

 As experience has been gained with prestressed

concrete construction, it has become evident that
a solution intermediate between fully prestressed
concrete and ordinary reinforced concrete offers
many advantages.

 Such an intermediate solution, in which concrete

tension, and usually some flexural cracking, is
permitted at full service load, is termed partial
prestressing.
 PARTIAL PRESTRESSING

 Although full prestressing offers the possibility of

complete elimination of cracks at full service load,
it may at the same time produce members with
objectionably large camber, or negative deflection,
at more typical loads less than the full value.

 However, a smaller amount of prestress force in

partially prestressed beams may produce
improved deflection characteristics at load stages
of interest.

 While cracks will be seen usually in partially

prestressed beams, at specified full service load
be applied. These cracks are small and will close
completely when the load is reduced.
 PARTIAL PRESTRESSING

In addition to improved deflection

characteristics, partial prestressing may
result in significant economy by reducing
the amount of prestressed reinforcement
and by permitting the use of beam cross-
section configurations with certain
practical advantages compared with
those required by full prestressing.
 PARTIAL PRESTRESSING

 Even though the amount of prestress force may be reduced

through use of partial prestressing, a beam must still have
an adequate factor of safety against failure.

 This will often require the addition of ordinary reinforcing

bars in the tension zone.

 Alternatives are to provide the total steel area needed for

strength by high strength tendons, but to stress those
tendons to less than their full permitted value, or to leave
some of the strands unstressed.

 Partial prestressing is looked upon with increasing favor in

the United States, as it offers the combined advantages of
reinforced and prestressed concrete.
 LOADS

 Loads that act on structures can be divided into three broad

categories: dead loads, live loads, and environmental loads.

 Dead loads are fixed in location and constant in magnitude

throughout the life of the structure. Usually the self-weight
of a structure is the most important part of the dead load.
This can be calculated closely, based on the dimensions of
the structure and the unit weight of the material.

 Concrete density varies from about 90 to 120 pcf (14 to 19

kN/m3) for lightweight concrete and is about 145 pcf (23
kN/m3) for normal concrete.

 In calculating the dead load of structural concrete, usually a

5 pcf (1 kN/m3) increment is included with the weight of the
concrete to account for the presence of the reinforcement.
 LOADS

 Live loads consist chiefly of occupancy loads in

buildings and traffic loads on bridges.

 They may be either fully or partially in place or not

present at all and may change in location.

 The minimum live loads for which the floors and

roof of a building should be designed are usually
specified in the building code that governs at the
site of construction.

 Representative values of minimum live loads to be

used in a wide variety of buildings are found in
Minimum Design Loads for Buildings and Other
Structures.
 LOADS

 The referenced table gives uniformly distributed live loads for various
types of occupancies; these include impact provisions where
necessary.
 These loads are expected maxima and considerably exceed average
values.
 In addition to these uniformly distributed loads, it is recommended
that, as an alternative to the uniform load, floors be designed to
support safely certain concentrated loads if these loads produce a
greater stress.
 Certain reductions are permitted in live load for members supporting
large areas on the premise that it is not likely that the entire area
would be fully loaded at one time.
 LOADS

 Service live loads for highway bridges are specified by

the American Association of State Highway and
Transportation Officials (AASHTO) in its Standard
Specifications for Highway Bridges.

 For railway bridges, the American Railway Engineering

Association (AREA) has published the Manual of
Railway Engineering.

 Environmental loads consist mainly of snow loads,

wind pressure and suction, earthquake loads (i.e.,
inertia forces caused by earthquake motions), soil
pressures on subsurface portions of structures, loads
from possible ponding of rainwater on flat surfaces,
and forces caused by temperature differentials.
 Like live loads, environmental loads at any given time
are uncertain both in magnitude and distribution.
 LOADS

 Much progress has been made in recent years in developing

rational methods for predicting horizontal forces on
structures due to wind and seismic action.

 The sum of the calculated dead load and the specified or

calculated live and environmental loads is called the service load,
because this is the best estimate of the maximum load that can be
expected to act during the service life of the structure.

 The factored load, or failure load, that a structure must be capable

of resisting to ensure an adequate margin of safety against
collapse, is a multiple of the service load.
 SERVICEABILITY, STRENGTH, AND
STRUCTURAL SAFETY

 To serve its purpose, a structure must be serviceable under

ordinary conditions of , use and must be safe against collapse.
 Serviceability requires that deflections be, suitably small, that
crack widths be kept to within acceptable limits, that vibrations be
minimized, and so on.
 In addition, most specifications impose limitations on the stresses
in the concrete and steel.
 Safety requires that the strength of the structure be adequate to
resist overloading should the loads actually expected to act be
increased by a certain amount.
 In designing prestressed concrete structures, the engineer must
consider a number of load stages and must impose certain
limiting conditions defining serviceability and safety.
 This process is sometimes known as limit states design, with
specific consideration of cracking limit state, deflection limit state,
strength limit state, and the like.
 SERVICEABILITY, STRENGTH, AND
STRUCTURAL SAFETY

 In the United States, the design of a prestressed concrete member

usually starts with the consideration of limit stresses in the
concrete and steel when the member is in the unloaded stage
(member self-weight plus prestress) and the full service load
stage.
 Tentative member dimensions, and prestressing steel area and
force are selected based on stress limits imposed by
specifications such as the Building Code Requirements for
Reinforced Concrete of the American Concrete Institute.
 For the trial member, crack widths, deflections, and other
serviceability limit conditions are checked and the design
modified as necessary.
 The ultimate strength of the member is then calculated and
compared with strength required to resist hypothetical overloads.
 Section dimensions, prestressing, and other reinforcement may
be further modified to produce the required margin of safety.
 SERVICEABILITY, STRENGTH, AND
STRUCTURAL SAFETY

 The strength of a structure depends on the strength of

the materials from which it is made.
 Minimum material strengths are specified in certain,
standaradized ways. The properties of concrete and its
components, the methods of mixing, placing, and
curing to obtain the required quality, and the methods
for testing are specified in documents.
 Standards of the American Society for Testing
Materials (ASTM) pertain to reinforcing and
prestressing steels and concrete.
 SERVICEABILITY, STRENGTH, AND
STRUCTURAL SAFETY

 Strength also depends on the care with which the structure

is built, that is, the accuracy with which the drawings and
specifications of the engineer are followed.

 Member sizes may differ from specified dimensions,

reinforcement may be out of position, prestressing force
may be improperly applied, or poor placement of the
concrete may result in voids.

 An important part of the Engineer's job is to provide proper

supervision of construction, and slighting of this
responsibility has had disastrous consequences in more
than one instance.
 SERVICEABILITY, STRENGTH, AND
STRUCTURAL SAFETY

 If strength could be predicted accurately and if loads were

known with equal certainty, then safety could be assured by
providing strength just barely in excess of the requirements
of the loads.

 But there are many sources of uncertainty in the estimation

of loads, as well as in analysis and design of the structure and
in construction.

 These uncertainties require a larger margin of safety. The

selection of an appropriate safety margin is not a simple
matter, but much progress has been made in recent years
toward rational safety provisions in design codes.
 SERVICEABILITY, STRENGTH, AND
STRUCTURAL SAFETY

 The load factors specified in the ACI Code are summarized

in Table 1.2.

 The required strength U is calculated by applying individual

load factors to the respective service loads: dead load (D),
live load (L), wind load (W), earthquake load (£), earth
pressure (H), fluid pressure (F), impact allowance (I), and
environmental effects (T) that may include settlement, creep,
shrinkage, and temperature change.

 Lower factors are used for loads known with greater

certainty, for example, dead loads, compared with loads of
greater variability, for example, live loads.

 Furthermore, for load combinations such as dead plus live

loads plus wind forces, a reduction coefficient is introduced
that reflects the improbability that an excessively large live
load coincides with an unusually high windstorm.
 SERVICEABILITY, STRENGTH, AND
STRUCTURAL SAFETY

 The factors also reflect, in a general way, uncertainties

with which internal load effects are calculated from
external loads in systems as complex as are highly
indeterminate, inelastic, reinforced or prestressed
concrete structures.
 Finally, the load factors distinguish between two
situations: (1) where the effects of all simultaneous
forces are additive, or (2) where the various load
effects tend to counteract each other, such as when
horizontal forces are present in addition to gravity
forces.
 In all cases in Table 1.2, the controlling equation is the
one that gives the largest factored load effect (U).
 SERVICEABILITY, STRENGTH, AND
STRUCTURAL SAFETY

 The required strength, should the member be overloaded,

must not exceed a conservative estimate of the actual
strength.
 To obtain that estimate, the nominal strength (Sn) is
calculated according to the best available information
concerning materials and member behavior.
 That nominal strength is reduced by applying strength
reduction factor (φ) to obtain what is called the design
strength of th member.
 The design strength must be at least equal to the required
strength calculated from the factored loads, that is

φSn ≥ U (1.2)
 SERVICEABILITY, STRENGTH, AND
STRUCTURAL SAFETY

 Equation (1.2) is stated in general terms. It can be

interpreted as applying either to loads acting on a member
or to the related internal effects such as; moment, shear,
and thrust.

 In specific terms for a member subjected, say to moment,

shear, and thrust:

φMn ≥ Mu (1.3a)
φVn ≥ Vu (1.3b)
φPn ≥ Pu (1.3c)

where terms with the subscript n are the nominal strengths

in flexure, shear, and thrust, respectively, and terms with the
subscript u are the factored load moment, shear, and thrust.
 SERVICEABILITY, STRENGTH, AND
STRUCTURAL SAFETY

 The strength reduction factors (φ) found in the ACI

Code are summarized Table 1.3.
 These are given different values depending on the
state of knowledge of particular kinds of member
behavior, that is, the confidence with which various
strengths can be calculated.
 The φ value for bending is higher than that for shear or
bearing.
 (φ) values reflect the probable importance, for survival
of the structure, of the particular member, as well as
the probable quality control achievable. For both these
reasons, a lower φ value is used for columns than for
beams.