BS5400 Part 4 Training Course - TE Epsom
BS5400 Part 4 Training Course - TE Epsom
BS5400 Part 4 Training Course - TE Epsom
Revision: 1
Status: Final
369691590.doc Page 1 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
12.1 Introduction
This session deals with the shear resistance of prestressed concrete sections. It
introduces the general shear verification methods adopted by BS5400:Part 4 for
prestressed concrete sections and highlights the differences to the rules for reinforced
concrete sections discussed in a previous session. Although the general rules covering
the shear design of post-tensioned concrete structures are discussed, much of the
session (including the worked example) is focussed on the shear design of pre-
tensioned concrete beams since these are encountered in design more often. A
flowchart covering the shear design of prestressed concrete sections is included as
Figure 12.4 at the end of the session. The code does not explicitly cover the use of
external or un-bonded prestressing tendons and thus the shear design of such sections
should be treated as reinforced concrete columns using BD58.
BS5400:Part 4 clause 6.3.4 deals with the shear resistance of prestressed concrete
beams. As for reinforced concrete sections, shear verifications are only required at
the ultimate limit state but, for prestressed sections, the calculations are generally
carried out to determine resistance forces to resist applied shear forces (as opposed to
stresses). The BS5400:Part 4 rules for shear design in prestressed concrete also refer
to the Class of section considered (introduced in session 9) as discussed in the notes
below.
In determining the applied ultimate shear force, BS5400:Part 4 allows the designer to
take account of any vertical component of prestress for sections un-cracked in flexure
(but not for those cracked in flexure) as discussed further in the sections below. It is
also noted that, although not explicitly covered in the code, the designer may choose
to take account of the vertical components of inclined tension and compression chord
forces. For members with shear reinforcement, these forces should be determined
from the actual chord forces obtained from the truss analogy and not the value of
M z from beam theory since the latter would overestimate the force in the
compression chord and underestimate the force in the tension chord. For sections
without the need for design shear reinforcement, however, it is reasonable to calculate
the vertical components of the forces from the value of M z .
The designer should be careful when considering these inclined chord forces in
conjunction with inclined prestress to ensure it is not double-counted. Often, the
prestressing force forms the tension chord. Care should also be taken with variable
depth sections when a varying neutral axis level has been modelled in the global
analysis. In this case, the output shear components will then be perpendicular to this
inclined axis. The inclined chord forces should relate to a consistent axis.
369691590.doc Page 2 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
breadth of the member or for T, I and L beams as the breadth of the rib as illustrated in
Figure 12.1.
bw bw
For prestressed concrete sections, the web width should be reduced appropriately to
allow for webs containing grouted or non-grouted ducts where the position of such
ducts coincides with the position of maximum principle tensile stress. The note to
clause 6.3.4.2 defines these reductions as:
The full duct diameter for un-grouted ducts;
of the duct diameter for grouted ducts.
BS5400:Part 4 makes no distinction between the use of steel or plastic ducts (unlike
the forthcoming Eurocodes for example) and thus the use of plastic ducts is preferred
for reasons of enhanced durability.
Two types of shear failure can occur in prestressed concrete beams, one due to web
cracking, and the other due to the formation of flexural shear cracks. For sections un-
cracked in flexure, a shear failure is caused by web cracks forming when the tensile
strength of the concrete is exceeded by the applied principal tensile stress. For
sections cracked in flexure, a shear failure is caused either by web cracks or the
development of a flexural crack into a shear crack. Thus it is necessary to check a
section for both types of shear failure and design taking the corresponding minimum
ultimate resistance in accordance with clause 6.3.4.1. For sections cracked in flexure,
it is also not clear whether the section is more critical under the maximum shear force
with coexistent bending moment or under the (absolute) maximum bending moment
with coexistent shear force, thus it is necessary to check both considerations for each
section designed.
The failure criterion for an un-cracked section with no shear reinforcement is that the
principal tensile stress anywhere in the section exceeds the tensile strength of the
concrete, ft, defined in clause 6.3.4.2 (and taken as positive in the following
derivations). ft is defined as 0.24 f cu in the code based on test results and
includes a material partial safety factor of 1.5. This value for ft was converted into
S.I. units from the original test predictions using imperial units and includes the
reduction introduced in the Australian code (of 20%) to allow for shrinkage cracking,
mild fatigue loading and variations in concrete quality.
369691590.doc Page 3 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
Taking principal tensile stress as positive and equal to the tensile strength of the
concrete gives the following equation:
f f bend f cp f bend
2
2
cp
ft (12.1)
2 2
where:
fcp is the compressive stress due to axial loading or prestressing (after losses and
including appropriate partial safety factors) at the level considered (in N/mm 2,
compression taken as positive) multiplied by a partial safety factor of 1.15 if
the prestressing force adversely affects the shear strength, or 0.87 in other
cases
fbend is the stress due to bending at the level considered (in N/mm 2, compression
taken as positive)
is the applied shear stress, where Vco Ay Ib . (This is strictly only valid
where the cross sectional properties are constant along the beam.)
Vco is the shear resistance determined from the shear force required to cause web
cracking
I is the second moment of area of the section
b is web width at the level considered including allowances for ducts as
discussed in section 12.2 above
Ay is the first moment of area of the concrete above the plane considered about
the cross-section centroid level.
In certain sections, such as I-beams, where the section width varies over height, the
maximum principal stress may occur at a level other than at the centroidal axis and it
may be unsafe to only consider this level. In such cases, the minimum value of shear
resistance should strictly be determined by calculating Vco at various levels in the
cross-section using (12.2) and including the relevant non-zero bending term, fbend
although this is not required by the code. This is mitigated, however, by the fact that
for I-beams, Ib Ay 0.8bh instead of 0.67bh for rectangles. For flanged beams in
which the neutral axis is within the flange, however, it is normally considered
adequate to check the principle stress at the web/flange junction.
At the end of clause 6.3.4.2, the code states that when inclined tendons are used, the
vertical component of prestress at a section may be included by adding (if adverse) or
subtracting (if beneficial) the component from the applied shear force. As noted in
section 12.2 above however, care should be taken when including this component in
combination with inclined chord forces or where an inclined neutral axis has been
modelled to ensure no double-counting. The component must be consistent with the
use of the member axis.
369691590.doc Page 4 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
For continuous beams, a further complication arises in how to deal with secondary
creep or temperature effects for example. It could be argued that these effects be
ignored at the ultimate limit state as for bending analysis (assuming adequate rotation
resistance). However, it is recommended here that these effects are taken into account
since shear failures are not ductile and any bending stresses reduce shear resistance
according to (12.2).
A second form of shear failure in prestressed concrete sections can occur by a flexural
crack developing into an inclined crack. The inclined crack eventually triggers a
shear failure.
Tests that considered the loads at which a flexure-shear crack formed in concrete
beams showed that the following empirical equation gives a good lower-bound
approximation to the ultimate shear force, Vcr:
M cr
Vcr 0.037bd f cu (12.4)
M V d 2
where Mcr is the cracking moment (see below) and M and V are the moment and shear
force at the section considered (both taken as positive). d should be taken as the
distance from the extreme fibre to the centroid of the prestressing steel. The above
equation has been converted into S.I. units from the original equation developed in
imperial units and also includes a material partial safety factor of 1.5 applied to fcu. It
is obviously conservative to ignore the d/2 term in (12.4), in which case the following
expression is obtained which appears in BS5400:Part 4 as equation 29:
M
Vcr 0.037bd f cu V cr (12.5)
M
If the modulus of rupture of the concrete (concrete flexural cracking stress) is fr, then
the cracking moment introduced in expression (12.4) above is given by:
M cr f r f pt I y (12.6)
where fpt is the tensile stress due to prestress alone at an extreme fibre distance y from
the centroid. fpt should be determined from the prestressing force after all losses and
multiplied by the appropriate partial safety factor value (0.87 since beneficial).
Again, an empirical equation is used to predict the value of fr, which stems from test
results published in the America. This value was subsequently reduced by the
Australian code to allow for shrinkage cracking, cyclic loading and variations in
concrete quality. Converting into S.I. units and applying a partial safety factor of 1.5
to fcu gives the value used in BS5400:Part 4 of 0.37 f cu . Thus the following
expression for the cracking moment at the section considered is given in the code:
M cr 0.37 f cu f pt I y (12.7)
Clause 6.3.4.3 gives a minimum value of Vcr of 0.1bd f cu . The reason for this
value is not clear but it originates from the American code having been converted into
S.I. units, converted to concrete cube strengths in place of cylinder strengths and
having a partial safety factor of 1.5 applied to fcu.
369691590.doc Page 5 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
The majority of beams for which (12.5) was shown to give a good lower bound
solution had relatively high levels of prestress (where the ratio of effective prestress
(given in clause 6.3.4.3(b) as the effective prestress in the tendons after all losses have
occurred multiplied by the appropriate partial safety factor, m, of 0.87 or 1.15) to
prestress characteristic strength (fpe/fpu) is greater than around 0.5. These beams were
thus representative of Class 1 or Class 2 sections (see session 9 for definitions of
member classes), but not necessarily for Class 3 sections which sometimes have much
lower levels of prestress. The code therefore introduces a modified expression for Vcr
for Class 3 beams which provides a linear transition from the shear clauses for
reinforced concrete sections (where fpe/fpu=0) to expression (12.5) for Class 1 and
Class 2 sections.
Thus the term B in the expression for Vcr above is written conveniently as
B M 0V M and hence Vcr for reinforced and all classes of prestressed concrete
becomes:
Vcr 1 n f pe f pu v c bd M 0V M (12.11)
For reinforced concrete sections, pe f M 0 0 thus cr V v c bd as given by the
reinforced concrete clauses. In order that the equations (12.9) and (12.11) agree for
Class 1, Class 2 and Class 3 sections for a value of fpe/fpu=0.6, it is necessary that:
I V
0.037bd f cu 0.37 f cu 1 0.6n v c bd (12.12)
y M
Equation 30 in the code is a simplification of the above equation where it is assumed
that M V 4h (since a shear failure is unlikely to occur if M V 4h ), d h ,
I y bh 2 6 (the value for a rectangular section), fcu = 50 N/mm2, vc = 0.55 N/mm2
(i.e. assuming 5% steel) and thus n = 0.55, giving:
Vcr 1 0.55 f pe f pu v c bd M 0V M (12.13)
369691590.doc Page 6 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
Tests have verified that this equation gives generally good agreement with the
experimental shear failure loads with suitable factors of safety despite the large
number of simplifications made in its derivation.
The total area of both bonded tensioned and untensioned steel in the tension zone
should be used when assessing vc in equation (12.13) above in accordance with
BS5400:Part 4 clause 6.3.4.3, with d taken as the distance from the extreme
compression fibre to the centroid of the steel area in the tension zone. The
longitudinal steel contributes to the shear resistance by acting as dowel reinforcement
and by controlling crack widths and thus any bonded steel may be considered. Where
a mixture of tensioned steel of area As(t) of characteristic strength fpu(t) and untensioned
steel of area As(u) and characteristic strength fyL(u), fpe/fpu may be taken by analogy to be
the ratio of the effective prestressing force (Pf) divided by the total ultimate force
developed by both tensioned and untensioned steel, i.e.
Pf
f pe f pu (12.14)
As(t) f pu(t) As(u) f yL(u)
Although not explicitly stated in the code, it is widely assumed that the value of Pf
shall be the factored prestressing force after all losses (i.e. including the 0.87 or 1.15
factor for prestressing force as appropriate for overall beneficial or adverse effects
respectively).
As for Class 1 and Class 2 members, the code defines the same minimum value of Vcr
of 0.1bd f cu for Class 3 members.
Contrary to the principles of statics, the code does not permit the vertical component
of the forces of inclined tendons to be added to Vcr. This is based on test results which
indicated that the drape of the tendons in the range 0 to 10 decreased the shear
strength of prestressed beams. However, it is not clear whether the reductions in the
test results were due to the tendon inclination or the reduced effective depths.
Equations (12.5) and (12.13) do predict a reduction in shear strength for draped
tendons (at all locations but the lowest point) thus, if the test results were due to the
latter, the code effectively double-counts the reduction by excluding the vertical
prestress component. BS5400:Part 4 is thus illogical in its treatment of inclined
tendons, albeit conservative.
From clause 6.3.4.1, the ultimate shear resistance of the concrete alone, Vc, should be
taken as the lesser of Vco and Vcr for every section considered. If Vc exceeds the
applied shear force, V, then theoretically, no shear reinforcement is required.
However, as for reinforced concrete design, when no shear reinforcement is required
on the basis of the design calculations for prestressed concrete beams, minimum shear
reinforcement should nevertheless be provided in accordance with clause 6.3.4.4 to
satisfy the following:
Asv 0.87 f yv
0.4 N/mm2 (12.15)
sv b
This follows from the same discussion for reinforced concrete sections (see session 2)
where sufficient reinforcement is required to raise the shear resistance above the shear
cracking load.
369691590.doc Page 7 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
If V exceeds Vc then sufficient shear reinforcement must be provided to ensure that the
total shear resistance of the section is greater than the applied shear force. As for
reinforced concrete beams and slabs, the equation below given in the code is based on
a similar development to the shear reinforcement equations in Table 7 of the code (see
discussion of (2.08) in session 2):
Asv V 0.4bd t Vc
(12.16)
sv 0.87 f yv d t
The value of dt should be taken as the effective depth from the extreme compression
fibre to either the level of the longitudinal corner bars in the tension zone around
which the shear links are anchored or the centroid of the tendons, whichever is the
greater. As for reinforced concrete sections, the design shear reinforcement required
by this equation now includes an additional excess nominal shear stress of 0.4 N/mm2
which was introduced to allow for the reduction of aggregate interlock under repeated
loading.
Care needs to be taken in deciding at what level to consider the additional longitudinal
reinforcement discussed above where a mixture of untensioned and tensioned
reinforcement is provided. If the shear design is based on the prestressing tendons
alone, the conventional longitudinal reinforcement can then be taken into account in
providing the additional longitudinal reinforcement required for equilibrium. Since
the centroid of the longitudinal reinforcement is usually at a greater effective depth
369691590.doc Page 8 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
than the centorid of the prestressing tendons, designing the reinforcement for the
calculated additional longitudinal force from the shear design is conservative.
As for reinforced concrete design, the code defines a maximum shear force that
prestressed concrete sections can resist to prevent premature crushing of the web
concrete. Clause 6.3.4.5 requires the ultimate applied shear force, V, to be less than
the maximum shear stress values given in Table 28 multiplied by bd. b is the effective
web width (reduced by the appropriate value for grouted or ungrouted ducts as
discussed in section 12.2 above) and d is the effective depth from the extreme
compression fibre to the centroid of the steel in the tension zone, irrespective of its
characteristic strength. The tabulated maximum shear strengths are base on the same
formula used for reinforced concrete design ( 0.75 f cu ), but the code recognises that
higher concrete strengths are often used in prestressed concrete (and have been
verified by tests) and limits the maximum stress at a greater concrete strength of
60 N/mm2.
It should be noted again that when checking the maximum shear force above, it
appears that any vertical component of prestress should only be included for sections
un-cracked in flexure which would be consistent with the code calculations of Vco and
Vcr. It is unclear if this restriction applies to the check of maximum shear force (it
doesnt in Eurocode 2).
The design of precast elements to resist flexural shear can be carried out as discussed
above for prestressed concrete members. The design of a composite section to resist
vertical shear, however, is more complicated and there is little appropriate test data.
The following discussion assumes that precast concrete units are prestressed and that
the problem is therefore to determine the shear resistance of a prestressed-reinforced
composite concrete section when it is un-cracked in flexure and also when it is
cracked. There are two general cases to consider: firstly a precast prestressed
concrete beam with an in-situ reinforced concrete deck slab (composite beam and
slab) and secondly precast prestressed concrete beams with in-situ concrete placed
over and between the beams (composite slab).
369691590.doc Page 9 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
Option (1) is simpler (and can sometimes give a better answer!) than option (2). In
applying the rules of clause 6.3.4.4 for shear reinforcement calculations, clause
7.4.2.2(c) requires dt to be derived for the composite section. Method (2) is illustrated
in example 12.1 below. As an alternative to these two options, the designer may wish
to consider the problem from first principles as discussed below.
The shear force (Vc1) due to self weight and the loads acting on the precast section
alone produces a shear stress distribution as shown in Figure 12.2(b). The shear stress
produced by Vc1 in the precast member at the level considered is s. Figure 12.2(c)
shows the shear stress distribution in the composite member with the addition of the
second stage shear force (Vc2) acting on the composite section. The shear stress at the
same level under this loading case is s . Figure 12.2(d) shows the total shear stress
distribution given by s s .
Insitu slab
level checked s s s + s
(typ.composite
Precast centroid)
beam
(a) Section (b) Shear stress due (c) Shear stress due (d) Total shear stress
to loads on precast to additional loads
beam alone on composite beam
Figure 12.2: Shear stress distribution in a composite beam and slab section
If the compressive stress at the level checked due to the build up of stress from loads
acting on the beam alone (including prestress) and on the composite section (with
each load component multiplied by appropriate partial safety factors) is ftot then the
major principal stress at this level is given by:
2
f tot f
tot s s
2
2 2
2 2
369691590.doc Page 10 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
Ac y c
Vc2
However, s where Ac, y c and Ic relate to the composite section. (This
b
Ic
assumes that the cross sectional properties are constant along the beam.) Thus
Ib
Vc2 c f t f t f tot s
2
(12.18)
Ac y c
Vc1 Ab y b
with s where Ab, y b and Ib relate to the precast beam section alone.
b I b
The total shear resistance of the section is thus given by:
Vco Vc1 Vc2
(12.19)
Note that Vc2 could be negative representing cracking of the precast beam section
before being made composite with the deck slab.
As for non-composite members discussed in section 12.3 above, care must also be
taken in dealing with secondary creep or temperature effects for example. This is
further complicated in composite members by the primary effects of differential
temperature, differential shrinkage and differential creep. However, the code requires
only the neutral axis to be checked for a simple, non-composite rectangular beam
without a deck slab and (12.3) ignores the primary effects of differential temperature.
For consistency, one could therefore argue that the designer could neglect all self-
equilibriating stresses in composite construction from differential temperature,
shrinkage and creep, but there is no theoretical justification for this.
369691590.doc Page 11 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
Consider the same M beam and composite deck slab used in worked example 11.1:
1000 mm
160 mm
400 mm
200 mm
60 mm
160 mm
200 mm
330 mm
80 mm
50 mm 160 mm
110 mm
160 mm
60 mm
0 mm
970 mm
Assuming the tendons are deflected over a 3975 mm length, the angle of deflection is
tan 1 370 3975 5.3 .
For section 1, assume an initial prestressing of 3325 kN and a factor for losses of
0.806. For section 2, assume an initial prestressing force of 3331 kN and a factor for
losses of 0.836. Thus the nominal prestressing forces after losses are 2680 kN for
section 1 and 2785 kN for section 2 and the vertical component of prestress at section
369691590.doc Page 12 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
6
2 is 2785 sin 5.3 74 kN. Use a partial safety factor for prestressing of 0.87 or
21
1.15 throughout as appropriate.
Assume the following design shear forces and moments split into dead load and live
load components and assume that all the dead loads are resisted by the beam alone
and that all the live loads are resisted by the composite section:
VDL 73 kN M DL 490 kNm
Section 1: V 357 kN , M 1300 kNm
VLL 284 kN M LL 810 kNm
VDL 132 kN M DL 104 kNm
Section 2: V 574 kN , M 443 kNm .
VLL 442 kN M LL 547 kNm
(a) Consider section 1 (subjected to a sagging moment) using method (2):
Firstly, check the maximum vertical shear capacity using Table 28 of the code:
Vmax 0.75 f cu bd 0.75 50 160 930 100 10 3 704.3 kN (conservatively
taking d ignoring any beam reinforcement).
V < Vmax therefore section size is adequate.
Method (2) in the code requires the full shear force to be taken on the composite
section and the stresses to be checked at the level of the composite section centroid.
369691590.doc Page 13 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
P y ybeam
f pt fl Pe
Abeam I beam
2680 10 3
0.87 2680 10 3 100 310
0 310
349 10 23.02 10 9
3
0.87 7.679 7.579 13.274 N/mm 2 (compression)
From (12.7) working to the composite section properties:
M cr 0.37 f cu f pt I y 0.37 50 13.274 54.5 10 9
474
10 6 1827.1 kNm
Finally, the section should be checked for longitudinal shear across the beam/ slab
interface (which would be likely to govern in this case) but this check is not
performed in this example.
369691590.doc Page 14 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
(which is greater than the (absolute) applied total moment of 443 kNm). Note that
this appears un-conservative (since the slab will be cracked) but is commonly used
following the examples in the BCA publication Detailed design of composite
concrete bridge superstructures by A. Kumar. It is advisable to use the cracked slab
properties if Mcr differs greatly as discussed in section 12.7 above.
Thus using (12.5) for Class 1 or Class 2 sections:
369691590.doc Page 15 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
M cr
Vcr 0.037bd f cu V
M
701.4
0.037 160 830 50 10 3 574 34.7 908.8 943.5 kN
443
Finally, the section should be checked for longitudinal shear across the beam/ slab
interface but this check is not performed in this example.
Composite slab
For inverted T-beam precast units with transverse reinforcement placed through
standard holes in the bottom of the webs of the units and completely filled with
concrete between and over the units to form a composite slab, BS5400:Part 4 clause
7.4.2.2(b) requires the shear resistance to be based on either:
(1) The same assumptions as option (1) above for composite beam and
slab sections; or
(2) Assuming the ultimate applied shear force, V, is apportioned between
the infill concrete and the precast prestressed units on the basis of the cross
sectional area of each with due allowance being made for the differing
grades of concrete where appropriate. In calculating the resistance in this
manner, the breadth of infill concrete shall be taken as the distance between
adjacent precast unit webs and the depth taken as the mean depth of infill
concrete or the mean effective depth to the longitudinal reinforcement
where this is provided in the infill concrete section.
As for composite beam and slab sections, in applying the rules of clause 6.3.4.4 for
shear reinforcement calculations, clause 7.4.2.2(c) requires dt to be derived for the
composite section. Option (2) is not based on any test results and it seems illogical to
apportion the total ultimate shear force on the basis of cross sectional areas. For
sections with large areas of infill concrete such as edge beams for example, option (2)
369691590.doc Page 16 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
can result in links being required in the infill concrete therefore it is better to use the
method developed from first principles discussed above for composite beam and slab
sections.
It could again be argued that when the section is subjected to hogging moments, the
in-situ concrete cracks and should be ignored.
For composite slabs where the precast beams are cracked in flexure, it is
recommended that either the insitu concrete is ignored and the precast beams designed
to carry the shear forces alone, or that the full composite section is used with the
reinforced concrete clauses, taking care when including the benefit of axial load. As
before, when calculating the effective web widths, it is usual that due allowance is
given to the potentially different classes of concrete by weighting the component
widths according to their design compressive strengths.
369691590.doc Page 17 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
Design adequate
369691590.doc Page 18 of 19
BS5400:Part 4 Course Notes D A Smith
25/10/17
Class Exercises
1. Repeat worked example 12.1 above to find the ultimate shear resistances and
shear reinforcement requirements for the same two loadcases but using option (1)
given in the code i.e. assuming all the load is carried by the precast beam alone.
Is there any benefit gained from using option (2) in this instance?
369691590.doc Page 19 of 19