MB Structural Design Compendium May16

Structural Design of
Concrete and Masonry:

A compendium of technical papers
Structural Design of Concrete and Masonry
Contents
4
How to calculate anchorage and lap lengths to Eurocode 2
12
Deflection the span-to-effective-depth method and Eurocode 2
17
Fire Design of concrete columns and walls to Eurocode 2
23
Eurocode 6: Design of masonry structures for vertical loads
29
Eurocode 6: Design of masonry structures for lateral loads and other factors
34
Design of post-tensioned slabs
40
Guidance on the design of liquid-retaining structures
45
An introduction to strut-and-tie modelling
About The Concrete Centre

The Concrete Centre provides material, design and construction guidance with the aim
of enabling those involved in the design, use and performance of concrete and masonry
to realise the potential of the material.
Through funding from the cement, aggregates, ready-mixed and precast concrete
sectors, The Concrete Centre is able to invest in the development of services and
resources that support the design and construction of robust, sustainable, cost-effective
structures throughout the built environment.
Resources available for structural engineers are highlighted on the inside back cover
of this document (page 51) and there is a wealth of material available at
www.concretecentre.com.
2 I www.concretecentre.com

This compendium includes articles that were first
published in the renowned journal The Structural
Engineer following its invitation to The Concrete Centre
to write a series of technical papers on structural design
in concrete.
The series includes topics chosen represent topical issues
and respond to frequently asked questions that we
receive from designers, such as guidance on anchorage
and lap lengths, post-tensioning and column fire design.
The compendium also includes papers on deflections,
Eurocode 6, liquid retaining structures and strut-and-tie.
Acknowledgements
The Concrete Centre would like to thank the authors
and peer reviewers for their contribution to these
technical papers including:
John Roberts; RS Narayanan; Robert Vollum,
Imperial College.
www.concretecentre.com I 3
How to calculate anchorage and lap

lengths to Eurocode 2
This article provides guidance on how to calculate

anchorage and lap lengths to Eurocode 2.
Introduction
EC2 provides information about reinforcement detailing in Sections 8
and 9 of Part 1-1 (BS EN 1992-1-1)1. Section 8 provides information on
the general aspects of detailing and this is where the rules for anchorage
and lap lengths are given. Section 9 sets out the rules for detailing
different types of elements, such as beams, slabs and columns.
In EC2, anchorage and lap lengths are proportional to the stress in the
bar at the start of the anchorage or lap. Therefore, if the bar is stressed
to only half its ultimate capacity, the lap or anchorage length will be half
what it would have needed to be if the bar were fully stressed.
The calculation for anchorage and lap lengths is as described in EC2

and is fairly extensive. There are shortcuts to the process, the first
being to use one of the tables produced by others24. These are based
on the bar being fully stressed and the cover being 25mm or normal.
These assumptions are conservative, particularly the assumption that
the bar is fully stressed, as bars are normally anchored or lapped away
from the points of high stress. Engineering judgement should be used
when applying any of the tables to ensure that the assumptions are
reasonable and not overly conservative.
This article discusses how to calculate an anchorage and lap length for
steel ribbed reinforcement subjected to predominantly static loading
using the information in Section 8. Coated steel bars (e.g. coated with
paint, epoxy or zinc) are not considered. The rules are applicable to
normal buildings and bridges.
An anchorage length is the length of bar required to transfer the force in
the bar into the concrete. A lap length is the length required to transfer
the force in one bar to another bar. Anchorage and lap lengths are
both calculated slightly differently depending on whether the bar is in
compression or tension.
For bars in tension, the anchorage length is measured along the
centreline of the bar. Figure 1 shows a tension anchorage for a bar
in a pad base. The anchorage length for bars in tension can include
bends and hooks (Figure 2), but bends and hooks do not contribute to
compression anchorages. For a foundation, such as a pile cap or pad
base, this can affect the depth of concrete that has to be provided.
Most tables that have been produced in the UK for anchorage and lap
lengths have been based on the assumption that the bar is fully stressed
at the start of the anchorage or at the lap length. This is rarely the case,
as good detailing principles put laps at locations of low stress and the
area of steel provided tends to be greater than the area of steel required.
Figure 1 Tension Anchorage
Ultimate bond stress

Both anchorage and lap lengths are determined by the ultimate bond
stress fbd which depends on the concrete strength and whether the
anchorage or lap length is in a good or poor bond condition.
bd = 2.2512ctd (Expression 8.2 from BS EN 1992-1-1)
where:
is the design tensile strength of concrete, ctd = ctctk,0,05/C
ctd
ctk,0,05
is the characteristic tensile strength of concrete,
ctk,0,05 = 0.7 ctm
ctm
is the mean tensile strength of concrete, ctm = 0.3 ck (2/3)
is the characteristic cylinder strength of concrete
ck
C
is the partial safety factor for concrete
(C = 1.5 in UK National Annex5)
ct
is a coefficient taking account of long-term effects on the

tensile strength and of unfavourable effects resulting from

the way the load is applied (ct = 1.0 in UK National Annex)
5
5
5
5
150
150
a
a
90
90
a
a<
< 150
150
a) Bend
Bend or
or L
L bar
bar
a)
b) Hook
Hook
b)
c) Loop
Loop or
or U
U bar
bar
c)
Source: EC2-1-1 Figure 8.1, b, c and d.

Figure 2 Typical bends and hooks bent through 90o or more
Direction of concreting
Confinement of concrete results in the characteristic compression

strength being greater than ck and is known as ck.c. If the concrete
surrounding a steel reinforcing bar is confined, the characteristic
strength of the concrete is increased and so will be the ultimate bond
stress between the bar and the concrete. Increasing the ultimate bond
stress will reduce the anchorage length. Concrete can be confined by
external pressure, internal stresses or reinforcement.
250
c) h > 250 mm
a) 45 < a < 90
Anchorage lengths
300
h
b) h < 250 mm
d) h > 600 mm
Key
Good bond conditions
Poor bond conditions
Figure 3 Good and Poor bond conditions
Table 1 gives the design tensile strengths for structural concretes up to

C50/60.
is the coefficient relating to the bond condition and
1
1 = 1 when the bond condition is good and 1 = 0.7 when

the bond condition is poor
It has been found by experiment that the top section of a concrete pour
provides less bond capacity than the rest of the concrete and therefore
the coefficient reduces in the top of a section. Figure 8.2 in BS EN 19921-1 gives the locations where the bond condition can be considered
poor (Figure 3). Any reinforcement that is vertical or in the bottom of a
section can be considered to be in good bond condition. Any horizontal
reinforcement in a slab 275mm thick or thinner can be considered to
be in good bond condition. Any horizontal reinforcement in the top
of a thicker slab or beam should be considered as being in poor bond
condition.
2
2

Figure 4 gives the basic design procedure for calculating the anchorage
length for a bar. There are various shortcuts, such as making all
coefficients = 1, that can be made to this procedure in order to ease the
design process, although this will result in a more conservative answer.
Both anchorage and lap lengths are determined from the ultimate bond
strength bd. The basic required anchorage length lb,rqd can be calculated
from:
lb,rqd = (/4) (sd/bd)
where sd is the design stress in the bar at the position from where the
anchorage is measured. If the design stress sd is taken as the maximum
allowable design stress:
sd = yd = yk/s = 500/1.15 = 435MPa
This number is used for most of the published anchorage and lap length
tables, but the design stress in the bar is seldom the maximum allowable
design stress, as bars are normally anchored and lapped away from
positions of maximum stress and the As,prov is normally greater than As,req.
The design anchorage length lbd is taken from the basic required
anchorage length lb,rqd multiplied by up to five coefficients, 1 to 5.
lbd = 1 2 3 4 5 lb,rqd lb,min
where the coefficients 1 to 5 are influenced by:
1 shape of the bar
= 1.0 for bar diameters 32mm

= (132)/100 for > 32mm (2 = 0.92 for 40mm bars)
is the diameter of the bar
Figure 4: flow chart for anchorage lengths.
1, 2, 3 and
5=1.0
Start
Determine fctd from Table 1
Is the bar in good

position?
Yes
No
No
Is the bar in
compression?
1= 0.7
4 = 0.7
Yes
4 = 1.0
No
1 = 1.0
Yes
Is bar diameter
32mm
No
Does the bar have

transverse reinforcement
welded to it?
2 = (132)/100
Take lbd = lb,rqd
Yes
Determine the coefficients 1 to 5 (see Table 2)
2 = 1.0
No
Determine ultimate bond stress
fbd = 2.25 1 2 fctd
Yes
Can lb,rqd be used as the design

anchorage length lbd?
Determine As,req and As,prov where the anchorage starts
Determine ultimate design stress in bar

sd = 435 As,req / As,prov
Determine basic anchorage length

lb,rqd = (/4) (sd/fbd)
(This can be conservatively used as the design anchorage length, lbd)
Figure 4 Flow chart for anchorage lengths
2 concrete cover
3 confinement by transverse reinforcement
4 confinement by welded transverse reinforcement
5 confinement by transverse pressure
The minimum anchorage length lb,min is:
max {0.3lb,rqd; 10; 100mm} for a tension anchorage
max {0.6lb,rqd; 10; 100mm} for a compression anchorage
The maximum value of all the five alpha coefficients is 1.0. The minimum
is never less than 0.7. The value to use is given in Table 8.2 of BS EN 19921-1. In this table there are different values for 1 and 2 for straight bars
and bars called other than straight. The other shapes are bars with
a bend of 90 or more in the anchorage length. Any benefit in the
coefficients from the bent bars is often negated by the effects of cover.
Note that the product of 2 3 and 5 has to be 0.7.
To calculate the values of 1 and 2 the value of cd is needed. cd is

obtained from Figure 8.3 in BS EN 1992-1-1 and shown here in Figure 5.
cd is often the nominal cover to the bars. In any published anchorage
tables, a conservative value for the nominal bar cover has to be assumed
and 25mm is used in the Concrete Centre tables. If the cover is larger
than 25mm, the anchorage length may be less than the value quoted in
most published tables. For hooked or bent bars in wide elements, such
as slabs or walls, cd is governed by the spacing between the bars.
In Table 8.2 of BS EN 1992-1-1 anchorage length alpha coefficients
are given for bars in tension and compression. The alpha values for a
compression anchorage are all 1.0, the maximum value, except for 4
which is 0.7, the same as a tension anchorage. Hence, the anchorage
length for a compression anchorage can always conservatively be used
as the anchorage length for a bar in tension.
Yes
1 = 1.0
2 =10.15(cd)/
0.7 2 1.0
Is the bar
straight?
No
END
1= 0.7 if cd > 3
1=1.0 if cd 3
2 =1 0.15 (cd3)/
0.7 2 1.0
Check lbd >

max{0.3lb,rqd;10;100mm}
lbd = 12345lb,rqd
Yes
3 = 1 K
0.7 3 1.0
Does the bar have

another bar between the
surface of the concrete
and itself?
Take 235 = 0.7

Yes
No
No
Is 235 < 0.7
3 = 1.0
Is the bar confined

by transverse
pressure?
Yes
5=1 0.04p
0.7 5 1.0
5=1.0
No
Alpha values for tension anchorage

Alpha values for tension anchorage are provided in Table 8.2 of BS EN
1992-1-1.
1 shape of the bar
Straight bar, 1 = 1.0
There is no benefit for straight bars; 1 is the maximum value of 1.0.
Table 1: Design tensile strength, ctd

C20/25
C25/30
C28/35
C30/37
C32/40
C35/45
C40/50
C50/60
ctm
2.21
2.56
2.77
2.90
3.02
3.21
3.51
4.07
ctk, 0.05
1.55
1.80
1.94
2.03
2.12
2.25
2.46
2.85
ctd
1.03
1.20
1.29
1.35
1.41
1.50
1.64
1.90
Bars other than straight, 1 = 0.7 if cd > 3; otherwise 1 = 1.0

If we assume that the value of cd is 25mm, then the only benefit for bars
other than straight is for bars that are 8mm in diameter or less. For bars
larger than 8mm, 1 = 1.0. However, for hooked or bobbed bars in wide
elements, where cd is based on the spacing of the bars, 1 will be 0.7 if
the spacing of the bars is equal to or greater than 7.
2 concrete cover
Straight bar, 2 = 1 0.15(cd )/ 0.7 1.0
There is no benefit in the value of 2 for straight bars unless (cd ) is
positive, which it will be for small diameter bars. If cd is 25mm, then
there will be some benefit for bars less than 25mm in diameter, i.e. for
20mm diameter bars and smaller, 2 will be less than 1.0. Bars other than
straight, 2 = 1 0.15(cd 3)/ 0.7 1.0
C1
C1
b) Bent or hooked bars

cd = min (a/2, c1)
a) Straight bars
cd = min (a/2, c1, c)
c) Looped bars
cd = c
Source: EC2-1-1, Figure 8.3.
Figure 5 Values of cd (c and c1 are taken to be cnom)
t,
As
Ast
As
K = 0.1
t,
Ast
As
K = 0.05
t,
Ast
K=0

Figure 6 Values of K
Table 2: Anchorage and lap lengths for locations of maximum stress

Reinforcement in tension, bar diameter, (mm)
Bond
Condition
Anchorage
length, lbd
Lap length,
lo
10
12
16
20
25
Reinforcement
32
40
in compression
Straight
bars only
Good
230
320
410
600
780
1010
1300
1760
40
Poor
330
450
580
850
1120
1450
1850
2510
58
Other
bars only
Good
320
410
490
650
810
1010
1300
1760
40
Poor
460
580
700
930
1160
1450
1850
2510
58
50% lapped
in one
location
(a6=1.4)
Good
320
440
570
830
1090
1420
1810
2460
57
Poor
460
630
820
1190
1560
2020
2590
3520
81
Good
340
470
610
890
1170
1520
1940
2640
61
Poor
490
680
870
1270
1670
2170
2770
3770
87
100%
lapped
in one
location
(a6=1.5)
Notes
1) Nominal cover to all sides and distance between bars 2mm (i.e. 2<1). At laps, clear distance between bars 50mm.
2) 1 = 3 = 4 = 5 = 1.0. For the beneficial effects of shape of bar, cover and confinement see Eurocode 2, Table 8.2.
3) Design stress has been taken as 435MPa. Where the design stress in the bar at the position from where the anchorage is measured,
sd, is less than 435MPa the figures in this table can be factored by sd/435. The minimum lap length is given in cl. 8.7.3 of Eurocode 2.
4) The anchorage and lap lengths have been rounded up to the nearest 10mm.
5) Where 33% of bars are lapped in one location, decrease the lap lengths for 50% lapped in one location by a factor of 0.82.
6) The figures in this table have been prepared for concrete class C25/30.
Concrete class
Factor
C20/25
C28/35
C30/37
C32/40
C35/45
C40/50
C45/55
C50/60
1.16
0.93
0.89
0.85
0.80
0.73
0.68
0.63
For example, if anchoring an H25 bar in a beam with H10 links at 300mm
centres:
As = 491mm2 for a 25mm diameter bar
lbd
Ast,min = 0.25 491 = 123mm2

Ast = 4 78.5 = 314mm2, assuming links will provide at least four 10mm
diameter transverse bars in the anchorage length
= (Ast Ast,min)/ As = (314 123)/491 = 0.38
3 = 1 K = 1 0.1 0.38 = 0.96
Figure 7 Anchorage of bottom reinforcement at end supports in beams and slabs where
directly supported by wall or column
Figure 8 Plan view of slab illustrating transverse tension
There is no benefit in the value of 2 for bars other than straight unless
(cd 3) is positive. If we assume that the value of cd is 25mm, then
the only benefit for bars other than straight is for bars that are 8mm in
diameter or less. For bars larger than 8mm 2 = 1.0. Again, for hooked
or bobbed bars in wide elements, where cd is based on the spacing of
the bars, 2 will be less than 1.0 if the spacing of the bars is equal to or
greater than 7.
3 confinement by transverse reinforcement
All bar types, 3 = 1 K 0.7 1.0
where:
K

depends on the position of the confining reinforcement.

The value of K is given in Figure 8.4 of BS EN 1991-1-1 and
shown here in Figure 6. A corner bar in a beam has the
highest value for K of 0.1. Bars which are in the outermost
layer in a slab are not confined and the K value is zero

is the amount of transverse reinforcement providing

confinement to a single anchored bar of area
As = (Ast Ast,min) / As
Ast

is the cross-sectional area of the transverse reinforcement

with diameter t along the design anchorage length
lbd
Ast,min

is the cross-sectional area of the minimum transverse

reinforcement = 0.25 As for beams and zero for slabs
4 confinement by welded transverse reinforcement

4 = 0.7 if the welded transverse reinforcement satisfies the requirements
given in Figure 8.1e of BS EN 1992-1-1. Otherwise 4 = 1.0.
5 confinement by transverse pressure
All bar types, 5 = 1 0.04p 0.7 1.0 where p is the transverse pressure
(MPa) at the ultimate limit state along the design anchorage length, lbd.
One place where the benefit of 5 can be used is when calculating the
design anchorage length lbd of bottom bars at end supports. This benefit
is given in BS EN 1992-1-1 cl. 9.2.1.4(3) and Figure 9.3, and is shown here
in Figure 7. It applies to beams and slabs.
Lap lengths
A lap length is the length two bars need to overlap each other to transfer
a force F from one bar to the other. If the bars are of different diameter,
the lap length is based on the smaller bar. The bars are typically placed
next to each other with no gap between them. There can be a gap, but
if the gap is greater than 50mm or four times the bar diameter, the gap
distance is added to the lap length.
Lapping bars, transferring a force from one bar to another via concrete,
results in transverse tension and this is illustrated in Figure 8 which is a
plan view of a slab. Cl.8.7.4.1 of BS EN 1992-1-1 gives guidance on the
amount and position of the transverse reinforcement that should be
provided. Following these rules can cause practical detailing issues if you
have to lap bars where the stress in the bar is at its maximum. If possible,
lapping bars where they are fully stressed should be avoided and, in
Figure 9: flow chart for lap lengths.
Start
Determine fctd from Table 1
END
No
Is the bar in
good position?
1 = 0.7
Yes
Check l0 >
max{0.36lb,rqd; 15; 200mm}
1 = 1.0
No
Is smaller bar diameter

32mm
2 = (132-)/100
l0 = 12356lb,rqd
Yes
2 = 1.0
Determine ultimate bond stress

fbd = 2.25 1 2 fctd
Take 235 = 0.7
Determine As,req and As,prov where the lap starts
Yes
Determine ultimate design stress in bar

sd = 435 As,req / As,prov
Is 235 < 0.7
No
Determine basic anchorage length

lb,rqd = (/4) (sd/fbd)
5 =1.0
Determine 6
6 = 1.4 for 50% lapped at a section
6 = 1.5 for 100% lapped at a section
Is lb,rqd 6
satisfactory as the
lap length?
No
Is the bar confined

by transverse
pressure?
Yes
5 = 1 0.04p
0.751.0
Yes
Take l0 = lb,rqd6
No
3 =1.0
Determine the coefficients 1, 2, 3 and 5

(see Table 2)
Is the bar in
compression?
No
Yes
Is the bar
straight?
1 =1.0
2 = 1-0.15(cd)/
0.7 2 1.0
No
Figure 9 Flow chart for lap lengths
No
Yes
Does the bar have

another bar between the
surface of the concrete
and itself?
1, 2, 3 and
5=1.0
1 = 0.7 if cd > 3
1 = 1.0 if cd 3
2 = 1 0.15(cd3)/
0.7 2 1.0
3 = 1 K
0.7 3 1.0
Yes
The largest possible savings in lap and anchorage length can be obtained
by considering the stress in the bar where it is lapped or anchored.
typical building structures, there is usually no need to lap bars where
they are fully stressed, e.g lapping bars in the bottom of a beam or slab
near mid-span. Examples where bars are fully stressed and laps are
needed are in raft foundations and in long-span bridges.
The wording of this clause regarding guidance on the provision of
transverse reinforcement is that it should be followed rather than
it must be followed. This may allow the designer some scope to use
engineering judgement when detailing the transverse reinforcement,
e.g increasing the lap length may reduce the amount of transverse
reinforcement.
All the bars in a section can be lapped at one location if the bars are in
one layer. If more than one layer is required, then the laps should be
staggered.
A design procedure to determine a lap length is given in Figure 9 and,
as can be seen in the flow chart, the initial steps are the same as for the
calculation of an anchorage length.
Design lap length, l0 = 1 2 3 5 6 lb,rqd l0,min
(Eq. 8.10 in BS EN 1992-1-1)
Recommendations
The largest possible savings in lap and anchorage length can be
obtained by considering the stress in the bar where it is lapped or
anchored.
For most locations, the old rule of thumb of lap lengths being equal to
40 should be sufficient. For this to be the case, the engineer should
use their judgement and should satisfy themselves that the lap and
anchorage locations are away from locations of high stress for the bars
being lapped or anchored. Where it is not possible to lap or anchor away
from those areas of high stress, the lengths will need to be up to the
values given in Table 2.
This article presents the rules currently set out in EC2. However, there
has been significant recent research which may find its way into the
next revision of the Eurocode. For example, research into the effect of
staggering on the strength of the lap (6) was discussed by John Cairns
in Structural Concrete (the fib journal) in 20146. In the review of the
Eurocodes, the detailing rules have been the subject of 208 comments
(18% of the total for EC2) and it is acknowledged that the rules need to
be simplified in the next revision.
The coefficients 1, 2, and 5 are calculated in the same way as for

anchorage lengths and, again, all the coefficients can be taken as = 1.0
as a simplification.
References:
3 is calculated slightly differently. When calculating 3 for a lap length

Ast,min = As(sd /fyd), with As = area of one lapped bar.
1) British Standards Institution (2004) BS EN 1992-1-1:2004 Design of

concrete structures. General rules and rules for buildings, London, UK: BSI
The design lap length can therefore be determined by multiplying the

design anchorage length by one more alpha coefficient 6, provided 3
has been calculated for a lap rather than an anchorage.
2) Bond A. J., Brooker O., Harris A. J. et al. (2011) How to Design Concrete
Structures using Eurocode 2, Camberley, UK: MPA The Concrete Centre
Minimum anchorage length, l0,min = max {0.3 6 lb,rqd; 15; 200mm}
3) The Institution of Structural Engineers and the Concrete Society

(2006) Standard method of detailing structural concrete: A manual
for best practice. (3rd ed.), London, UK: The Institution of Structural
Engineers
6 coefficient based on the percentage of lapped bars in one

lapped section, 1
6 = (1/25)0.5 1.0 1.5
4) The Institution of Structural Engineers (2006) Manual for the

design of concrete building structures to Eurocode 2, London, UK: The
Institution of Structural Engineers
Design lap length, l0 = 6 lbd l0,min
where:
1

is the percentage of reinforcement lapped within 0.65l0 from

the centre of the lap length considered
In most cases either the laps will all occur at the same location, which is
100% lapped and where 6 = 1.5, or the laps will be staggered, which is
50% lapped and where 6 = 1.4.
5) British Standards Institution (2005) NA to BS EN 1992-1-1:2004 UK

National Annex to Eurocode 2. Design of concrete structures. General
rules and rules for buildings, London, UK: BSI
6) Cairns J. (2014) Staggered lap joints for tension reinforcement,
Structural Concrete, 15 (1), pp 4554
For vertically cast columns, good bond conditions exist at laps.
Deflection the span-to-effective-depth

method and EC2
The span-to-effective-depth (L/d) method is a very

popular way of verifying the limit state of deformation
(i.e. deflection) of concrete slabs and beams.
Introduction
Essentially, the span-to-effective-depth method is a hand method
based on experience, justified by various reports1,2. The L/d method also
serves as a very useful and valuable hand check on computer outputs.
According to Section 7.4.2 of BS EN 1992-1-13 (Eurocode 2) and fib Model
Code 20104, its use will be adequate for avoiding deflection problems
in normal circumstances. The main attraction of the method is that it
avoids the need to undertake laborious calculations.
While according to Eurocode 05, deflection limits should be agreed with
clients, generally the limits implicit in the L/d verification of deflection
of concrete structures are L/250 overall and L/500 post partitions (i.e. for
deflection affecting partitions, brittle finishes, etc.).
The current L/d method
In simple terms, the current BS EN 1992 L/d method means verifying
that:
Allowable L/d = N x K x F1 x F2 x F3 actual L/d (1)
where:
N = basic span-to-effective-depth ratio derived for K = 1.0 from the
formulae:
if 0
F1 = factor to account for flanged sections. When beff/bw = 1.0, factor

F1 = 1.0. When beff/bw > 3.0, factor F1 = 0.80. For values of beff/bw
between 1.0 and 3.0, interpolation may be used
F2 = factor to account for brittle partitions in association with long
spans. Generally F2 = 1.0 but if brittle partitions are liable to be
damaged by excessive deflection, F2 should be determined as
follows:
a) in flat slabs in which the longer span is greater than 8.5m,
F2 = 8.5/leff
b) in beams and other slabs with spans in excess of 7.0m,
F2 = 7.0/leff
F3 = factor to account for service stress in tensile reinforcement =
310/s. It is considered conservative to assume that
310/s = 500As,prov/(ykAs,req) where:
s = tensile stress in reinforcement at mid-span (at support
for cantilevers) under design load at serviceability limit state
(SLS) calculated using the characteristic value of serviceability
load6
F3 is restricted to 1.56
Notes
or if > 0
Factors F1, F2 and F3 have been used here for convenience, they are
not symbols used in BS EN 1992-1-1. According to the notes to Table
NA.5 of the UK National Annex (NA)6 warnings are given that the values
of K may not be appropriate when formwork is struck at an early age.
L/d may not exceed 40K
N = L/d = K[11 + 1.5ck0.5 0/( ) + ck0.5 ( / 0)0.5 /12] (2b)
Basis and current issues
N = L/d = K[11 + 1.5ck0.5 0/ + 3.2ck0.5 (0 / 1)1.5] (2a)
for = 0, N may be determined from Figure 1

where:
L = span
d = effective depth
ck = characteristic compressive cylinder strength of
concrete at 28 days
0 = fck0.5/1000
= As,req/bd
= As2req/bd
K = factor to account for structural system (Table 1)
The L/d method is outlined in Eurocode 2 Commentary7. The method is

based on parametric studies by Corres et al.2, rather than theory. There
have been many comments relating to the soundness of the method,
which is now acknowledged to have some limitations and deficiencies8,9:

The expressions (7.16a) and (7.16b) in BS EN 1992-1-1 (Equations

2a and 2b) assume a certain ratio between total load and dead
load, superimposed dead load (SDL) and imposed load (IL).
It would be desirable to introduce different possibilities for these
ratios in order to widen the application field of these formulae
Notes
1 For two-way spanning slabs, the check should be
carried out on the basis of the shorter span.
2 This graph assumes simply supported span
condition (K = 1.0).
K = 1.5 for interior span condition
K = 1.3 for end span condition
K = 0.4 for cantilevers
3 Compression reinforcement, r, has been taken as 0.
4 Curves based on the following expressions:
l
1.5 fck r 0
+ 3.2
= K 11 +
r
d
fck
( )]
1.5
r0
1
r
where r r 0
and
l
1.5 fck r 0
+
= K 11 +
d
( r r )
fck
12
r
r0
where r > r 0 .
Percentage of tension reinforcement (As,req/bd)
Figure 1 Basic span-to-effective-depth ratios, N, for K = 1, = 0
Figure 2 Typical loading and deflection history for slab in multistorey building
Figure 3 L/d for simply supported slabs
Figure 4 L/d for simply supported slabs supporting imposed load of 2.5kN/m2
The expressions do not account for excess reinforcement in tension

or compression. (UK practice allowed up to 100% additional
reinforcement.) This parameter should be included
The analysis of the section to determine whether the section was

cracked or not looked at the centre span of the beam only, and
conservatively used those properties throughout
The expressions do not account for peak loading during

construction and the cracking induced during that process
(Figure 2). This parameter should also be introduced
The assessment of Ec,eff (effective modulus) was questionable
The relative humidity (RH) was taken as 70%. In the UK, RH is

often taken as being 50% internally and 80% or 85% externally
The effects of ctm,l (mean flexural tensile strength of concrete)

were ignored in the background document, whereas the effects
are very noticeable for sections with <0.6% reinforcement, i.e.
they are very noticeable in slabs. The mean 28-day direct
concrete tensile strength was used in deflection calculations
Results using this method do not give a good match with span-todepth ratios derived by calculating deflections rigorously under
quasi-permanent loading (Figures 3 and 4)
No allowance appears to have been made for the use of loading

expressions (6.10a) and (6.10b) in BS EN 1990
Table 1: K factors to be applied to basic ratios of span to effective depth for

different structural systems
Element
Simply supported beams or slabs
1.0
End span of continuous beams or slabs
1.3
Interior spans of continuous beams or slabs
1.5
Flat slabs (based on longer span)
1.2
Cantilevers
0.4

The method for adjustment when providing more reinforcement

than required for flexure (based on steel service stress) is not
conservative
The most substantiated comments came from Vollum10. The significant
reductions in slab thickness initially allowed by BS EN 1992-1-1,
compared to those allowed by BS 8110, were met with some scepticism
in the UK and modifications were made via the UK NA to EC26 (as
outlined earlier). Vollum showed that the EC2 span-to-depth rules do
not account for cracking during construction; variations in effective
depth over thickness (d/h), varying serviceability/ultimate loading ratio
(w/wu) or the effect of restraint at the external supports.
These issues have led some to suggest that the L/d method should be
deleted from standards. However, to do so would be to deny designers
a valuable tool and feel for their designs, although there is clearly room
for improvement.
Rigorous method of assessing deflection
Here, it is worth explaining the rigorous method according to BS EN
1992-1-1, Cl. 7.4.3. A section will crack if it experiences a serviceability
moment exceeding its moment capacity at the time Mcr(t). If a section
is cracked, then its inertia is much less than that of the uncracked section
and so curvature is much greater in cracked sections. Cracked sections
and greater degrees of cracking lead to larger curvatures and deflections
(Figure 5).
Economically designed horizontal elements act somewhere between
wholly uncracked and wholly cracked. Slabs tend to be less highly
stressed and are cracked along only part of their spans. Beams tend
to be more highly stressed and crack along much more of their spans.
Actions are applied at different times and these actions may or may
not cause cracking depending upon the flexural tensile strength of the
concrete at the time. Once cracked, a section is assumed to stay cracked
but some tensile stiffening occurs in the concrete between cracks. So
the mean inertia of the segment is somewhere between those for wholly
uncracked or wholly cracked sections. When considering curvatures,
these different actions incur different creep coefficients, which affect
the applicable effective modulus of the concrete used in assessing
curvatures.
BS EN 1992-1-1 (and MC2010) state that an adequate prediction of
behaviour and the mean curvature in a discreet element (Figure 6) is
given by:
1/rm = (2 + 2cs) + (1 )(1 + 1cs)
where:
rm = mean radius
= 1 (Mcr/M)2
where:

= 1.0 for short-term and = 0.5 for long-term loading. For

construction loads, conservatively10 = 0.70
Mcr = cracking moment

M = SLS moment
1 = M/EceffI1 = curvature of uncracked section
2 = M/EceffI2 = curvature of cracked section
where:
Eceff = Ecm/(1 + )
where:
Ecm = modulus at 28 days

= creep coefficient
I1, I2 = inertias of the uncracked and cracked sections
1cs, 2cs = shrinkage curvature
This rigorous method is described in greater detail elsewhere11,12
and is supported by site-based research10,13,14.
Greater accuracy may be achieved by considering small increments of
span and computing relevant curvatures and thus overall deflections.
The method involves numerical integration, which is tedious by hand
but can, of course, be undertaken by computer, notably by spreadsheet
software.
Default assumptions for rigorous analysis

Deflections depend significantly on cracking, material properties and
loading: all of which makes for difficulties and uncertainties at the
design stage.
However, Vollum10 suggested that in the absence of better information,
the following assumptions should be made in deflection calculations of
slabs in multistorey construction:

The slab is struck at seven days; the superimposed dead load is

applied at 60 days
Creep and shrinkage strains are calculated with a relative humidity

of 50% (internal environment assumed)
Two levels of backprops are used
The floor above is cast after 10 days
When slabs are supported by slabs below during construction, the

peak construction load peak is the peak UDL action for the SLS,
which should be taken as 0.04h kN/m2 where h is the slab thickness
in mm
The permanent load perm should be taken as the quasi permanent

load combination and be applied at one year
Peak deflections are calculated under the frequent load case; the
increment in load freq perm should be treated as an instantaneous
load in the calculation of ELT
where:
freq = the frequent UDL action for the SLS and
perm = the permanent UDL action, including quasi-permanent
variable actions, for the SLS
ELT = the equivalent long-term modulus of the concrete, dependant
on loading and age at time of loading11,12

It is difficult to assess the effective tensile strength of concrete in

slabs due to its inherent variability, and there are uncertainties in
the tensile stress induced by internal and external restraint and
shrinkage. However, back-analysis of deflection data showed that
the effective flexural strength of concrete in reinforced concrete
slabs typically lies somewhere between the indirect and flexural
strengths
Using these default values, rigorous methods of calculating deflection

can be applied in order to judge the span-to-depth method. The
differences between the L/d and the rigorous methods can then be
compared.
Differences in values between methods
The data in Table 2 were derived for simply supported slabs by using:
spreadsheet TCC31R15 to determine outcome L/d ratios using the

rigorous method (Section 7.4.3 of BS EN 1992-1-1) and the default
values described earlier, and
spreadsheet TCC3115 to determine those using the L/d method in

Section 7.4.2 of BS EN 1992-1-1
For each span and imposed load the depth of the slab was iterated
such that all design criteria were met and As,prov = As,req.
www.thestructuralengineer.org
33
EC2 Moment curvature response

Moment
Uncracked
response
E=0.5
Mu
Tension stiffening
Fully cracked
response
Curvature \
\cs1 \cs2
Figure 5 Typical momentcurvature response
Figure 6 Curvature in simply supported slab
Curvature in simply supported slab

Conclusion
Figure 5
Typical momentcurvature response

As may
been seen from Fig. 3, the agreement between the current L/d
method and the current rigorous analysis method is not good at low
spans or low imposed
loads.
As may
be seen
Fig. 4, the
currentL/190
Cl.
construction
deection
would
be L/487.
Forfrom
non-brittle
nishes,
7.4.2(2)
method
appears
to
underestimate
the
L/d
required
by
as
much
compares with limits of L/200 for variable actions for steel beams and
2
as 15%for
fortimber.
an imposed
of 2.5kN/m
at about 8m.Actual deections
L/150
L/487load
would
appear acceptable.
are often moderated by end restraints, stronger concrete, lower loads,
Using this case as a worse case and using the L/d = 26 indicated by
etc.
Cl.The
7.4.2(2)
lead theoretically
to long-term
deflection
of 42mm
datawould
and graphs
show an apparent
anomaly.
The L/d
required
or L/190
in a 2340mm
thick
simply
supported
2 slab (d = 308mm). Postfor
2.5kN/m
is smaller
than
for 5.0kN/m
(Fig. 4). Close examination
construction
would
be L/487.
For non-brittle
finishes,
L/190
revealed
that,deflection
in line with
Vollum,
construction
load was
critical.
If the
compares
L/200 for variable
steel beamswould
and be
slabs
werewith
the limits
sameof
thickness,
crackingactions
duringfor
construction
L/150
for timber.
L/487
would
appear
acceptable.
are
the
same,
but the
effect
of lower
cracked
inertiaActual
of thedeflections
2.5kN/m2 slab
is
greater
than
the
additional
creep
in
the
more
heavily
reinforced
and
often moderated by end restraints, stronger concrete, lower
loads, etc.
loaded slab.
The
datarespect
and graphs
show an apparent
The L/d required
With
to continuity,
rigorous anomaly.
analysis showed
good
2
is smaller
than for in
5.0kN/m
for 2.5kN/m2with
correlation
the K factors
Table 1. (Fig. 4). Close examination
revealed that, in line with Vollum, construction load was critical. If the
Conclusion
slabs were the same thickness, cracking during construction would be
2
Given
thebut
complexity
variability
concrete
a material,
the same,
the effect and
of lower
crackedofinertia
of theas
2.5kN/m
slab
loading
and
the
environment,
it
is
perhaps
unsurprising
that
the
is greater than the additional creep in the more heavily reinforced and
loaded slab.
With
respect
to continuity,
rigorous analysis showed good correlation
Table
2: Basic
L/d ratios
with the K factors in Table 1.
Span (m)
4.0
5.0
6.0
Figure 6
Given the complexity and variability of concrete as a material, loading

current
L/d method itisisinaccurate.
Nonetheless,
as current
discussed,
and
the environment,
perhaps unsurprising
that the
L/d
it
appears
that
the
use
of
L/d
methods
will
be
adequate
method is inaccurate. Nonetheless, as discussed, it appears that the use
for avoiding deection problems in normal circumstances.
of L/d methods will be adequate for avoiding deflection problems in
Compliance with span-to-depth ratios means that deections
normal circumstances. Compliance with span-to-depth ratios means
in members may be considered not to exceed the implicit limits
that
deflections in members may be considered not to exceed the
stated.
implicit
limits more
stated.rigorous methods are necessary in unusual
However,
circumstances
or where
deection
limits other
than those
However,
more rigorous
methods
are necessary
in unusual
implicit in the simplied methods are appropriate.
circumstances or where deflection limits other than those implicit in the
Work continues to provide a more accurate L/d method
simplified methods are appropriate.
particularly at low imposed loads. Part of that process is
to consider
the
of deection
limits
across
all at
Work
continues
toharmonisation
provide a more accurate
L/d method
particularly
materials.
low
imposed loads. Part of that process is to consider the harmonisation
of deflection limits across all materials.
Acknowledgement
Parts of this paper were included in: Goodchild C., Vollum R.

Acknowledgement
and Webster R. (2014) Improving the L/d method, b-Congress,
Parts
of thisIndia
paper were included in: Goodchild C., Vollum R. and Webster
Mumbai,
R. (2014) Improving the L/d method, fib-Congress, Mumbai, India
7.0
8.0
9.0
10.0
11.0
12.0
13.0
14.0
24.3
22.2
20.7
19.6
18.4
17.8
17.2
16.7
L/d ratio using rigorous analysis:

Imposed load = 2.5kN/m2
Table 2: Basic L/d ratios
Imposed load = 5kN/m
Span (m)
L/d ratio using rigorous analysis:
Imposed load = 10kN/m2
37.8
30.7
27.1
37.8
31.2
27.6
25.0
4.0
5.0
6.0
7.0
30.3
24.6
27.5
23.2
25.4
22.5
23.6
21.5
22.9
8.0
22.0
20.4
21.4
9.0
20.5
19.5
20.1
10.0
19.4
18.6
37.8
30.7
27.1
24.3
22.2
20.7
19.6
Imposed load = 5kN/m
37.8
31.2
27.6
25.0
22.9
21.4
30.3
27.5
30.6
25.4
23.6
22.0
20.5
2
2
Imposed
= 5kN/m
Imposed
loadload
= 10kN/m
28.3
24.6
27.3
23.
26.5
22.5
25.6
21.5
23.8
20.4
25.9
25.2
24.5
23.9
24.1
31.9
23.5
30.6
23.0
29.3
22.4
28.2
L/d ratio using Cl

7.4.2(2):
2
Imposed
load
= 7.5kN/m
L/d
ratio using
Cl 7.4.2(2):
Imposed
= 10kN/m
Imposed
loadload
= 2.5kN/m
2
31.9
29.3
28.2
18.9
11.0
18.4
17.8
18.2
12.0
17.7
17.2
17.6
13.0
17.2
16.7
17.2
14.0
16.8
16.3
18.4
17.8
17.2
16.7
20.1
18.9
18.2
17.6
17.2
19.4
22.2
20.8
18.4
17.7
17.2
16.8
22.3
19.5
20.9
18.6
19.6
17.8
18.5
17.2
17.5
16.7
16.4
16.3
22.3
20.9
19.7
18.6
17.6
16.7
15.9
21.0
26.0
19.7
23.9
18.6
22.2
17.6
20.8
16.7
19.5
15.9
18.3
15.1
17.3
26.0
23.9
19.5
18.3
17.3
Imposed
5kN/m2 fyk = 500MPa; As,prov = A
28.3
26.5 long-term
25.6deection
23.8limit L/250,
22.3 post-construction
20.9
19.6
18.5 L/500 17.5
Notes: fload
30MPa;
1.5kN/m2 and
deection limit
ck = =
s,req; SDL =27.3
16.4
25.9
25.2
24.5
23.9
22.3
20.9
19.7
18.6
17.6
16.7
15.9
Imposed load = 10kN/m2
24.1
23.5
23.0
22.4
21.0
19.7
18.6
17.6
16.7
15.9
15.1
Notes: ck = 30MPa; yk = 500MPa; As,prov = As,req; SDL = 1.5kN/m2 and long-term deflection limit L/250, post-construction deflection limit L/500
TSE44_30-34 CDG v1.indd 33
24/07/2015 09:10
Reinforced concrete frame residential building. Courtesy of Coinford Construction.
References:
1) Beeby A. W. (1971) TR456: Modified Proposals for Controlling
Deflections by Means of Ratios of Span to Effective Depth, Wexham
Springs, UK: Cement and Concrete Association
2) Corres Peiretti H., Prez Caldentey A., Lpez Ag J. C. and Edtbauer
J. (2002) EC2 serviceability limit states: deflections. Supporting
document: first draft, 15 June 2002, Madrid, Spain: Grupo de Hormign
Estructural ETSICCP UPM
3) British Standards Institution (2004) BS EN 1992-1-1:2004+A1:2014
Eurocode 2: Design of concrete structures. General rules and rules for
buildings, London, UK: BSI
4) fib (2013) Model Code for Concrete Structures 2010, Berlin, Germany:
Ernst & Sohn
9) Goodchild C. and Webster R. (2012) BSI Committee B525/2 paper:

Interpretation of BS EN 1992-1-1 with respect to span:depth (L/d) ratios
(Unpublished)
10) Vollum R. L. (2009) Comparison of deflection calculations and
span-to-depth ratios in BS 8110 and Eurocode 2, Magazine of Concrete
Research, 61 (6), pp. 465476
11) The Concrete Society (2005) TR58: Deflections in concrete slabs
and beams, Camberley, UK: Concrete Society
12) Webster R. and Brooker O. (2006) How to design concrete
structures using Eurocode 2: No. 8. Deflection calculations, Camberley,
UK: The Concrete Centre
5) British Standards Institution (2002) BS EN 1990:2002+A1:2005

Eurocode 0. Basis of structural design, London, UK: BSI
13) Vollum R. L., Moss R. M. and Hossain T. R. (2002) Slab deflections

in the Cardington in-situ concrete frame building, Magazine of Concrete
Research, 54 (1), pp. 2334
6) British Standards Institution (2004) BS EN 1992-1-1:2004+A1:2014 UK

National Annex to Eurocode 2. Design of concrete structures. General rules
and rules for buildings, London, UK: BSI
14) Vollum R. L. (2003) Investigation into backprop forces and

deflections at St George Wharf, Magazine of Concrete Research, 55 (5),
pp. 449460
7) European Concrete Platform (2008) Eurocode 2 Commentary [Online]

Available at: www.europeanconcrete.eu/publications/
eurocodes/114-commentarytoeurocode2 (Accessed: June 2015)
15) Goodchild C. H. and Webster R. M. (2006) User Guide to RC

Spreadsheets: v3, Camberley, UK: The Concrete Centre
8) Beal A. N. (2009) Eurocode 2: Span/depth ratios for RC slabs and

beams, The Structural Engineer, 87 (20), pp. 3540
Fire design of concrete columns and

walls to Eurocode 2
In fire, concrete does not burn and performs well, both as

an engineered structure and as a material.
Concrete does not normally need any further protection against fire due
to its thermal conductivity properties and the fact that it does not burn.
The design of concrete slabs and beams is not generally affected by fire
design requirements. However, fire design requirements can be
a governing factor in the sizing of columns, particularly in multistorey
buildings. This article therefore concentrates on the guidance given
in Eurocode 2 on the sizing of concrete columns for different fire
resistance periods.
Methods
Guidance on fire design to EC2 is given in part 1-2 (BS EN 1992-1-2)1
and is much more extensive than in the previous codes. For the design
of columns and walls there are basically three design methods available
to the engineer:

tabular data
simplified calculation methods
advanced calculation methods
This article covers some of the tabular methods and simplified

calculation methods for columns and walls. Table 1 shows limitations
on the different tabulated data for columns. Outside these limitations
the simplified calculation methods can be used.
The tabulated data for columns are given in Chapter 5 of part 1-2, split
into Method A and Method B. Both methods are based on tests and
either can be used for the design of columns, but they have
slightly different limitations on their use.
The two simplified methods given in Annex B are the 500C isotherm
method and the zone method. The zone method gives a more accurate
analysis of the effect of the fire on the element than the 500C isotherm
method, but both can provide savings to the sizing of columns
compared to the tabulated data in Methods A and B.
Most of the columns that have been tested have been square columns;
therefore, the tabulated data for columns assume square or circular
columns. Rectangular columns are not covered in Method B, but can be
modelled, to a certain extent, in Method A.
Fin or blade columns are not covered by the tabulated data until they
are greater than a width-to-thickness ratio of 4:1 (h:b). At this point,
EC2 part 1-1 (BS EN 1992-1-1)2 states that they are walls, and the column
should be designed as a wall at both normal temperatures and in a fire.
If a column needs to be designed to fit into a partition, the use of blade
columns with a ratio of 4:1 or greater has been common for many years,
as by definition these are walls.
The tabulated data are given for braced structures only. However, the
background document for the UK, PD 6687-13, states that the tabulated
data can be used to size unbraced columns, at the discretion of the
designer. In critical cases it recommends that Annex B, which details
the simplified methods, be used. It justifies the use of tabulated data for
both braced and unbraced columns on historical grounds.
0.8
0.7
Reduction factor fi
Introduction
fi = 0.9
0.6
0.7
0.5
0.5
0.4
0.3
0.2
0
0.2
G = 1.35
Q = 1.50
0.5
1.0
1.5
Qk,1/Gk
2.0
2.5
3.0
Figure 1 Reduction factor fi when Exp. 6.10 of EC2 has been used
For all the different method types, the axial load on the element
compared to the capacity of the column or wall is key to the design.
A lightly loaded column will be able to resist a fire for a much longer
period than the same column when fully loaded.
T
bfi
b
T - Tension
a) Fire exposure on three sides

with tension zone exposed
Previous codes did not distinguish between braced and
unbraced columns and the tabulated data from BS EN 1992-1-2 give
larger column sizes than those from previous codes. There is also an
argument that, in most unbraced structures, the fire will only affect
a few of the columns at any one time. The columns in the fire can
therefore be said to be braced by the columns unaffected by the fire.
b) Fire expos
the compr
500 C
hfi
Loading
The load under fire conditions can be reduced from the loads taken
for normal temperature design. Generally, the effect of the loads
Ed, = Ed where Ed is the design moment, axial load, shear force, etc.
under normal temperature loads.
bfi
b
The factor can simply be taken as 0.7, or can be calculated:

i) If Expression 6.10 of EC04 has been used in the normal
temperature design (Figure 1)
=
c) Fire exposure on four sides (beam or column)
Figure 2 Reduced concrete section for column exposed on four sides
Gk + 1,1 Qk,1
1.35Gk + 1.5 Qk,1
ii) If Expression 6.10b of EC0 has been used in the normal

temperature design
=
Gk + 1,1 Qk,1
1.25Gk + 1.5 Qk,1
where:
Qk,1 is the main variable action under consideration. Only one variable

action need be considered in the fire design
1,1 is the appropriate factor for the frequent value of the main

variable action
Tabular methods
Columns: Method A
Method A has the more stringent limitations of the two tabulated
data methods:

the effective length of column under fire conditions l0, 3m.
For a braced structure, the effective length can be taken as
0.5l, i.e. l 6m for intermediate floors and 0.5l l0, 0.7l, i.e. l
4.2m for top floors, where l is the actual length of the column
the first order eccentricity M0Ed, / N0Ed, 0.15h or 0.15b, where

M0Ed, is the first order bending moment for the fire condition and
N0Ed, is the axial load under the fire condition
the amount of reinforcement As < 0.04Ac
The fire resistance period is based on the degree of utilisation

= NEd, / NRd and the table gives values for = 0.2, 0.5 and 0.7. Table
5.2a in BS EN 1992-1-2 assumes that cc = 1.0. In the UK a value of cc =
0.85 for bending and compression has been chosen5. However, the
values in the table are conservative for the UK so can be used.
Table 2 gives values for the UK.
Other values can be calculated from BS EN 1992-1-2 Expression
5.7, and this expression and method can be used for rectangular
columns:
R = 120((R + Ra+ Rl + Rb + Rn) / 120)1.8
where R is the fire resistance period in minutes
Table 1: Summary of the tabulated data for columns in BS EN 1992-1-2

Slenderness ratio
Effective length 3m
30
30 80
30 80
30 80
200 b 450
150 b 600
150 b 600
150 b 600
150 b 600
e 0.025b but
e 10mm
e 0.25b but
e 100mm
e 0.5b but
e 200mm
Table C1
Table C2
Table C3
Table C4
Table C5
Table C6
Table C7
Table C8
Table C9
Minimum dimensions
Eccentricity
e 0.15b
= 0.1
Table 5.2a*
= 0.5
= 1.0
Jenny Burridge
e 0.25b
Concrete & Fire
Version 1
Chap 7 Fig
7.7
Table 5.2b
10.09.08
Amendments
23.02.09, 25.06.09
Note
* 4% reinforcement
All columns must be braced
b is the smallest dimension of a rectangular column, or the diameter of the column
Mechanical reinforcement ratio
Slenderness ratio
Lofi
i
Asfyd
Acfcd
Table 2: Minimum column dimensions and axis distances for square and circular columns
Exposed condition
Load level fi
More than one side

exposed
Only one side

exposed
R, = 83 1
Fire resistance period (minutes)

R 30
R 60
R 90
R 120
R 180
R 240
0.2
200/25
200/25
200/30
300/25
250/38
350/33
350/52
450/47
350/68
0.5
200/25
200/32
300/27
300/40
400/35
350/49
450/43
350/68
450/62
450/78
0.7
200/26
300/25
250/40
350/35
350/48
450/43
350/59
450/54
450/75
450/73
0.7
155/25
155/25
155/25
175/35
230/55
295/70
fcd,fi (20)
(1 + )
(0.85 / acc) +
This simplifies to R, =83(1) in the UK as cc = 0.85.

Ra = 1.6(a30), where a is the axis distance
Rl = 9.6(5l0,) where 2.0m l0, 6.0m
Rn = 0 for columns with four longitudinal bars and Rn = 12 for columns

with more than four bars
Rb = 0.09b where b is the modified column width
b = 2Ac/(b + h) for rectangular sections and the diameter for circular
sections
h
is limited to h 1.5b and 200mm b 450mm
However, as proved by tests, blade columns have a longer period
of fire resistance than columns of the same thickness but less width.
It therefore seems reasonable to use Exp. 5.7 for columns where
h > 1.5b, but to limit b in this expression. If h > 1.5b then b should
be limited to 2b 1.5b/(b + 1.5b) = 1.2b. This will give a conservative
answer for the fire resistance.
Example: Blade column design
Assume a 600 200 column, fully loaded in the normal
temperature design condition, designed in the UK (cc = 0.85)
with an effective length in fire of 2m (4m floor-to-floor height).
= 0.7 as the column is fully loaded R = 83(1) = 24.9
Axis distance a = 25mm cover + 10mm link + 8mm (H16 bar) =
43mm
Ra = 1.6(a30) = 20.8
l0, = 2m Rl = 9.6(5l0,) = 28.8
b is kept to 1.2b = 240mm Rb = 0.09b = 21.6
Rn = 12 as there are more than four bars in the column
R = 120 ((R + Ra + Rl + Rb + Rn)/120)1.8
R = 120((24.9 + 20.8 + 28.8 + 21.6 + 12)/120)1.8 = 99min
Columns: Method B
Method B provides a more comprehensive method for the design
of columns in that the restrictions on eccentricity of the first order
moments are less onerous. For most columns Table 5.2b willJenny
be Burridge
adequate, but there are tables in Annex C of EC2 which give Concrete
more & Fire
Version 1
options where the limitations of Table 5.2b are exceeded. Chap 7 Fig 7.8
xbfifcd,fi (20)
Fs = As fscd,fi (
As
z
dfi
As
As1fsd,fi (
m)
Fs = As2fsd,fi (
m)
bfi
Mu = Mu1 + Mu2
As = As1 + As2
Figure 3 Stress distribution at ultimate limit state for rectangular concrete cross-section
with compression reinforcement
The restrictions on the use of Table 5.2b are that:

the slenderness of the column under fire conditions should be
= l0, / i 30, where i is the minimum radius of inertia

the first order eccentricity under fire conditions should satisfy the
limit: e = M0Ed, / N0Ed, emax, where emax = 100mm, e/b 0.25 and
b = minimum column dimension
The load level at normal temperature conditions, n, is used in the
determination of the minimum values (Table 3).
n = N0Ed, / [0.7(Ac cd + As yd)]
Note that in the table the mechanical reinforcement ratio, , is one
of the required parameters:
= ASyd / Ac cd
In BS EN 1992-1-1, a conservative value in the determination of
limiting slenderness for the column takes = 0.1. For a class C30/37
concrete, this represents approximately 0.4% reinforcement,
whereas when = 1.0, the column would require approximately 4%
reinforcement.
Walls
Tabulated data for load-bearing walls are given in Table 5.4 of BS EN
1992-1-2 (Table 4). The degree of utilisation is the same as that for
Method A for columns. Another restriction is that:
clear wall height
wall thickness
Amendments
24.02.09, 26.02.09, 29.06.09, 13.11.09
40
Simplified calculation methods

500C isotherm method
In the isotherm method, concrete at a temperature above 500C is
neglected in the calculation of section resistance, while concrete at
or below 500C is assumed to retain its full, ambient temperature
strength. In BS EN 1992-1-2 the method is illustrated with reference
to rectangular sections. Thus, the calculation process is to first
check that the section meets the minimum cross-sectional width
requirements in Table 5.
Wall
kc (
The rounded corners of the residual section reflect the real profile
of the isotherm and may be approximated to a rectangle (Figure 2);
some interpretation may be required.
m1)
a z1
a z1
a z1
a z1
a z1
W1
If the minimum requirements are met, the area not damaged by

heat, i.e. within the 500C isotherm, is determined to give a reduced
section size (b, d) where the concrete retains its original properties.
All the reinforcement can be taken as acting with the section,
including the reinforcement in the zone outside the 500C isotherm,
but the strength of the bars is reduced. The strength can be taken
from Figure 4.2a of BS EN 1992-1-2.
While the temperature gradient through a section denoted by
isotherms may be determined from testing, BS EN 1992-1-2 provides
temperature profiles for a number of typical member types and
cross-sections in Annex A.
Wall end
M1
W1
W1
W1
Column
kc (
m1)
a z1
a z1
a z1
W1
W1
Figure 4 Reduction of cross-section when using zone method
The section resistance may then be determined using conventional

calculation methods (Figure 3) and compared against the design
load in the fire situation in this figure, where:
Table 2: Minimum column dimensions and axis distances for square and circular columns
Load level at normal
temperature
conditions
(n)
0.15
0.3
0.5
0.7
Reinforcement ratio

R 30
R 60
R 90
R 120
R 180
R 240
0.1%
150/25
150/30
200/25
200/40
250/25
250/50
350/25
400/50
500/25
500/60
550/25
0.5%
150/25
150/25
150/35
200/25
200/45
300/25
300/45
450/25
450/45
500/25
1.0%
150/25
150/25
200/25
200/40
250/25
300/35
400/25
400/45
500/25
0.1%
150/25
200/40
300/25
300/40
400/25
400/50
550/25
500/60
550/25
550/40
600/25
0.5%
150/25
150/35
200/25
200/45
300/25
300/45
550/25
450/50
600/25
550/55
600/25
1.0%
150/25
150/30
200/25
200/40
300/25
250/50
400/25
450/50
550/25
500/40
600/30
0.1%
200/30
250/25
300/40
500/25
500/50
550/25
550/25
550/60
600/30
600/75
0.5%
150/25
250/35
350/25
300/45
550/25
450/50
600/25
500/60
600/50
600/70
1.0%
150/25
250/40
400/25
250/40
550/25
450/45
600/30
500/60
600/45
600/60
0.1%
300/30
350/25
500/25
550/40
600/25
550/60
600/45
>600
>600
0.5%
200/30
250/25
350/40
550/25
500/50
600/40
500/60
600/50
600/75
>600
1.0%
200/30
300/25
300/50
600/30
500/50
600/45
600/60
>600
>600
Jenny Burridge
Concrete & Fire
Version 1
Chap 7 Fig 7.9
05.09.08
Amendments
ge
ire
.12
80
70
1.0
FRR
60
240 min
0.8
50
FRR
30 min
120 min
30
90 min
k c ( m )
az
180 min
40
0.6
0.4
20
10
0
0
90 min
120 min
60 min
30 min
60 min
0.2
180 min
240 min
50
100
150
w (mm)
200
250
300
0
0
50
100
150
200
250
w (mm)
Figure 5 Fire damaged zone az
Figure 6 Reduction of compression strength
Table 4: Tabulated data for load-bearing walls

Exposed condition
One side exposed
Both sides exposed
Load level fi

REI 30
REI 60
REI 90
REI 120
REI 180
0.35
100/10*
110/10*
120/20*
150/25
180/40
REI 240
230/55
0.7
120/10*
130/10*
140/25
160/35
210/50
270/60
0.35
120/10*
120/10*
140/10*
160/25
200/45
250/55
0.7
120/10*
140/10*
170/25
220/35
270/55
350/60
Note
* Normally the cover required by BE EN 1992-1-1 will control
Design notes according to 5.4.2 of BS EN 1992-1-2:
A. The tabular data can be used for plain concrete walls
B. For calcareous aggregates, the minimum wall thickness can be reduced by 10%
C. To prevent excessive thermal deformation and subsequent integrity failure between wall and slab, the ratio of clear wall height to wall thickness should not exceed 40
b =
width of reduced cross-section
d =
effective depth of reduced cross-section
z=
lever arm between tension reinforcement and concrete
z =
lever arm between tension and compression reinforcement
As =
area of tension reinforcement
As1 =

part of tension reinforcement in equilibrium with concrete

compression block
As2 =

part of tension reinforcement in equilibrium with

compression reinforcement
As =
area of compression reinforcement
scd,(m) = design value of compression reinforcement strength in the

fire situation at mean temperature m in that layer
Jenny Burridge
cd,(20) = design value of compression strength concrete

Concretein&the
Firefire

situation at normal temperature = ck/c, Version 1
Chap 7 Fig 7.11
08.09.08
sd,(m) = design value of tension reinforcement strength

in the fire
Amendments

situation at mean temperature m in that 09.02.09,
layer 24.02.09
total force in compression reinforcement in the fire situation,

Fs =

and is equal to part of the total force in the tension
reinforcement
For UK design: = 0.8 for fck 50MPa, or = 0.8 (fck 50)/400 for
50 < fck 90MPa, = 1.0 for fck 50MPa, or = 1.0 (fck 50)/200 for
50 < fck 90MPa, x is as defined for normal temperature design and
c, = 1.0.
Zone method
In the zone method, the cross-section is divided up into several zones
which are ascribed different temperatures. The strength of each zone
is assessed and the strengths are aggregated to give an assessment of
the strength of the whole section. The zone method is more accurate
than the 500C isotherm method, but is more complicated.
The design procedure for the zone method can be summarised as
follows:
Table 5: Minimum cross-sectional width of columns or walls

Fire resistance
R 60
R 90
R 120
R 180
R 240
Minimum
width of crosssection (mm)
90
120
160
200
280
References:
1) British Standards Institution (2010) BS EN 1992-1-2:2004 Eurocode 2.
Design of concrete structures. General rules. Structural fire design,
London, UK: BSI
1. The cross-section is divided into three or more parallel zones of equal

thickness.
2) British Standards Institution (2014) BS EN 1992-1-1:2004 Eurocode 2:

Design of concrete structures. General rules and rules for buildings,
London, UK: BSI
2. The corresponding mean temperature of each of the zones is checked,

using the temperature graphs in BS EN 1992-1-2 AnnexA, and the
corresponding concrete compressive strength fcd() and elastic
modulus (if applicable) of each zone is calculated.
3) British Standards Institution (2010) PD 6687-1:2010 Background

paper to the National Annexes to BS EN 1992-1 and BS EN 1992-3,
London, UK: BSI
3. The fire-damaged zone az (Figures 4 and 5) is calculated; this will be

ignored in the strength and stiff ness calculation. When calculating
the fire-damaged zone, the width w is taken as either the thickness
of a wall or column that is exposed on one side, half the thickness
of a two-sided exposed wall or column, or half the smallest dimension
of a four-sided exposed column.
4. All the reinforcement, including that in the fire-damaged zone, can be
taken into account in the analysis of the section, but with a reduced
strength calculated using Figure 4.2a of BS EN 1992-1-2.
5. The load-bearing capacity and stiffness are determined based on
the reduced cross-section and strength (Figure 6), using normal
temperature design procedures.

Eurocode. Basis of structural design, London, UK: BSI
rules and rules for buildings, London, UK, BSI
Further reading
The Concrete Centre (2011) How to design concrete structures using
Eurocode 2, Camberley, UK: MPA The Concrete Centre
Bailey C. G. and Khoury G. A. (2011) Performance of concrete structures
in fire, Camberley, UK: MPA The Concrete Centre
The impact of a major fire at Tytherington County High School, Cheshire was limited due to the fire resistance of the concrete structure.
Concrete Design Guide
No. Eurocode
2: Eurocode6:6:Design
Design of
ofmasonry
masonry
structures for vertical loads
structures for vertical loads
This series is produced by The Concrete Centre to enable designers to realise the potential
of concrete.
The Concrete Centre, part of the Mineral Products Association (MPA), is a team of qualied
professionals with expertise in concrete construction, engineering and architecture.
www.concretecentre.com
roduction
BS EN 1996-2 and encompasses

usefulof masonry,
associated
test methods.
ight seem strange that a series on
Thethe
design
whether
blockwork or brickwork,
design information previously contained
The standards
supporting EC6 were
crete design should include articles
is coveredinin Eurocode
6.
BS 56281012, which does not conict with the
developed within a common framework but it
out masonry design. However, much
did not prove possible to standardise all the
sonry design in the UK uses concrete
principles contained in EC6.
test methods used by the different materials
EC6 has been developed to enable
cks, made by members of the Concrete
from
whichand
masonry
units
are made.from
Words
the designer to use the following
types
ck Association,
which is affiliated to the
Introduction
units
are made. Words
like brick
block have
disappeared
the
and
block have
disappeared
of masonry unit: clay, calcium silicate,
eral Products Association.
European vocabularylike
andbrick
they are
all referred
to as masonry
units. from
BS(Eurocode
EN 1996 (Eurocode
6) covers
of masonryconcrete,
for buildings
and
the European vocabulary and they are all
aggregate
autoclaved
aerated
BS EN 1996
6) covers
the the design
civil engineering
worksand
andcivil
is organised into
four parts.
In common
with
Newstone
methods were introduced
foras
determining
compressive
referred to
masonry the
units.
concrete
(aircrete),
manufactured
ign of masonry
for buildings
1
the
other
material
Eurocodes,
Part
1-1
covers
the
structural
design
rules
strength
of
masonry
units
and
the
method
of
determining
New methods were introducedthe
for
and natural stone. European Standards for
gineering works and is organised into four
and Part
1-2the
covers
structural
fire design2.these
Thereafter,
there ishave
somebeen published
characteristic
strengththe
of masonry
changed
from testing
determining
compressive
strength
materials
by thecompressive
ts. In common
with
other
material
divergence
fromthe
other
Eurocodes in that Part
covers
aspects
to masonry
much smaller
masonry
wallette
specimens.
units
and the
method
of
BSI 2and
form
part of
ofdesign,
an array ofstorey-height
standards panelsof
ocodes, Part
1-1 covers
structural
3
materials
while Part 3 relating
looks after
German need
determining the characteristic compressive
tothe
masonry-related
products and the
ign rules1 and
Partand
1-2workmanship
covers structural
for simplifiedthere
calculation
methods4. Masonry bridges are not covered
design2. Thereafter,
is some
by EC6.
Each
part has a corresponding
UK national annex58.
ergence from
other
Eurocodes
in that
t 2 coversBS
aspects
of design,
EN 1996-1-1
was firstmaterials
published in 2005 along with BS EN 1996-1-2.
3
workmanship
while
Part
3 looks
after
BS EN 1996-2 and 1996-3
were
published in 2006. The corresponding
Figure 1
German need
for
simpli
ed
calculation
National Annexes bear the same dates. Corrigenda
were
issued
Modi
cations
to K for units
4
thods . Masonry
bridges
covered
with
generalwas
purpose
mortar
to Part 1-1
in 2006are
andnot
2009,
and in 2012 laid
a new
version
published
EC6. Eachincorporating
part has a corresponding
UK the 2012 changes to BS EN 1996Amendment 1. While
58
onal annex
. relatively small, the opportunity was taken to update the
1-1 are
BS EN 1996-1-1
was rstUK
published
in
corresponding
National Annex
based on feedback from use and
05 along with
BS EN 1996-1-2.
recalibration
of some ofBS
the EN
outcomes. The discussion and observations
6-2 and 1996-3
were
in 2006.
that follow
arepublished
therefore related
to the 2012 version of the UK National
e corresponding
National
Annexes
Annex to BS EN 1996-1-1. bear
same dates. Corrigenda were issued
A further
Standards
Institution (BSI) publication, PD 66979,
Part 1-1 in 2006
andBritish
2009,
and in 2012
published
in 2010.
This covers recommendations for the design
ew versionwas
was
published
incorporating
masonry
to BS EN 1996-1-1 and BS EN 1996-2 and
endment 1.ofWhile
thestructures
2012 changes
encompasses
the useful
design information previously contained
BS EN 1996-1-1
are relatively
small,
1012
in
BS
5628
,
which
does
opportunity was taken to updatenot
theconflict with the principles contained
in UK
EC6.National Annex based
responding
feedback from
use
and
recalibration
EC6 has
been
developed
to enable the designer to use the following
ome of the
outcomes.
The
discussion
types of masonry unit:
clay, calcium silicate, aggregate concrete,
observations
that follow
therefore
autoclaved
aeratedare
concrete
(aircrete), manufactured stone and natural
ted to thestone.
2012European
version of
the UKfor these materials have been published
Standards
ional Annex
to BS
by the
BSI EN
and 1996-1-1.
form part of an array of standards relating to masonryA further British
Standards
Institution
(BSI)test methods.
related products and
the associated
blication, PD 66979, was published in 2010.
The standards supporting
EC6design
were developed within a common
s covers recommendations
for the
framework to
butBS
it did
prove possible
masonry structures
ENnot
1996-1-1
and to standardise all the
test methods used by the different materials from which masonry
Concrete v3.indd 38
Figure 1 Modifications to K for units laid with general purpose mortar
22/01/2015 12:26
Ancillary components are dealt with in a coherent way within the

standards and in BS EN 1996-1-1 suitable values of partial factors have
been introduced. The partial factors for use with masonry are given
in National Annex Table NA.1 and shown here in Table 1. Two levels of
attestation of conformity are recognised: Category I and Category II. This
forms part of the declaration made by the manufacturer of the masonry
units. Two classes of execution control are also recognised: 1 and 2.
Vertical load design
Strength
During the drafting of EC6, a way had to be found to deal with the wide
range of masonry units used across Europe. This range not only includes
different materials such as clay, concrete and stone, but also a variety of
configurations based upon the proportion and direction of any holes
or perforations, web thickness etc. This has resulted in four groupings of
masonry units according to the percentage size and orientation of holes
in the units when laid. The UK only has experience of Group 1 and Group
2 masonry units, but no doubt Group 3 and Group 4 units will find their
way to the UK. In the UK National Annex, information is only provided for
Group 1 and Group 2 units because of this lack of a UK national database
for Groups 3 and 4. Properties for Groups 3 and 4 would normally be
established by testing. Two levels of quality assurance for the
manufacture of masonry units are specified:
Table 1: Value of partial factors for materials for ultimate limit states
Material
When in a state of direct of flexural compression

Unreinforced masonry made with:
Units of category I
2.3
2.7
Units of category II
2.6
3.0
Reinforced masonry made with mortar M6 or M12:

Units of category I
2.0
Units of category II
2.3
2.3
2.7
2.0
2.4
2.5
2.5
When in a state of flexural tension

Units of category I and II
but in laterally loaded wall
panels when removal of
the panel would not affect
the overall stability of the
building
When in a state of shear
Unreinforced masonry made with:
Reinforced masonry made with mortar M6 or M12:
Category I masonry units, which have a declared compressive

strength with a probability of failure to reach it not exceeding 5%
Category II masonry units, which are not intended to comply with

the level of confidence of Category I units
The characteristic compressive strength of masonry is presented in

BS EN 1996-1-1 as Equation 3.1. This equation includes the normalised
strength of the masonry unit b and the strength of the mortar m. The
UK National Annex places limits on the use of this equation for general
purpose mortar as follows:

b is not to be taken to be greater than 110N/mm2
m is not to be taken to be greater than b or 12N/mm2
the coefficient of variation of the strength of the masonry units is

not more than 25%
2*
Masonry
In addition the UK National Annex requires that the coefficient of

variation for the compressive strength of masonry units should not
exceed 25%.
Class of execution control M

1*
2.0
Anchorage of reinforcing
steel
1.5
Reinforcing steel and

prestressing steel
2.0
Ancillary components
wall ties
3.0
3.0
Ancillary components
straps
1.5**
1.5**
See NA to BS EN 845-2
See NA to BS EN 845-2
Steel and other components
Lintels in accordance with

EN 845-213
*
Class 1 of execution control should be assumed whenever the work is carried out following the
recommendations for workmanship in BS EN 19962, including appropriate supervision and inspection, and
in addition:
i) the specification, supervision and control ensure that the construction is compatible with the use of the
appropriate partial safety factors given in BS EN 199611
ii) the mortar conforms to BS EN 998-2, if it is factory made mortar. If the mortar is site mixed, preliminary
compressive strength tests, in accordance with BS EN 1015-2 and 1015-11, are carried on the mixture of sand,
lime (if any) and cement that is intended to be used (the proportions given in Table NA.2 may be used initially
for the tests) in order to confirm that the strength requirements of the specification can be met; the proportions
may need to be changed to achieve the required strengths and the new proportions are then to be used for the
work on site. Regular compressive strength testing is carried out on samples from the site mortar to check that
the required strengths are being achieved.
Class 2 of execution control should be assumed whenever the work is carried out following the
recommendations for workmanship in BS EN 19962, including appropriate supervision.
When considering the effects of misuse or accident these values may be halved.
Class 2 of execution control is not considered appropriate for reinforced masonry and should not be used.
However, masonry wall panels reinforced with bed joint reinforcement used:
i) to enhance the lateral strength of the masonry panel or
ii) to limit or control shrinkage or expansion of the masonry can be considered to be unreinforced masonry for
the purpose of class of execution control and the unreinforced masonry direct or flexural compression M values
are appropriate for use.
When considering the effects of misuse or accident these values should be taken as 1.0.
For horizontal restraint straps, unless otherwise specified, the declared ultimate load capacity depends
on there being a design compressive stress in the masonry of at least 0.4N/mm2. When a lower stress due to
design loads may be acting, for example when autoclaved aerated concrete or lightweight aggregate concrete
masonry is used, the manufacturers advice should be sought and a partial safety factor of 3 should be used.
**
Number 2
40
TheStructuralEngineer
February 2015
Technical
Concrete design
Figure 2
Wall stiffened
by piers
Figure 2 Wall stiffened by piers
Table 2: Values of K to be used with Equation 3.1

Masonry unit
General
purpose
mortar
Thin layer
Lightweight mortar of density d
(kg/m3)
mortar (bed joint
0.5mm and
Table 2: Values of K to be used with
Equation 3.1
3mm)
Masonry unit
General
purpose mortar
Clay
Group 1
0.50
Thin layer 600

Lightweight
of density
d
d 800 mortar800<
d 1300
mortar (bed
(kg/m3)
joint 0.5mm
600 d 800
800< d 1300
and 3mm)
0.75
0.30
0.40
Clay
Group Group
2
1
Group 2
Group 3 and 4
0.40
*
Group 3 and 4
0.50
0.70 0.75
0.25
0.30
0.40 0.30
0.40
0.70
0.25
0.30
0.50
Group 2
Group 2
0.50
0.80 0.80
0.40
0.70
0.40
0.70
Aggregate concrete
Group 1
0.75
0.90
0.45
0.45
Group 1 (units
0.50
0.70
0.40
0.40
Aggregate concrete
Group 1
laid flat)
0.75
Group 2
0.50
Group 1 (units
laid at)
Group 3 and 4
0.70
Group 1
0.70
Group Autoclaved
2
0.70
aerated
concrete
Group 3 and 4
0.90
0.75
Manufactured stone
Group aerated
1
0.75
Autoclaved
concrete
Dimensioned natural stone
Group 1
Group 1
0.75
0.45
0.45
0.76
0.40
0.45
0.45
0.76
*
0.45
0.40
0.45
0.45
0.90
0.45
0.45
0.90
0.90
0.45
*
of masonry (other than shell-bedded
the results of
K b m is determined
[Equationfrom
3.1 of
BS EN 151996-1-1]
k =masonry)
The tests are carried out on small wallette
where:
specimens rather than the storey-height
used
in BS 5628. compressive
The designer has
is the
characteristic
strength of the masonry,
k panels
the option of2either having the units intended
in N/mm
to be
used in a project tested or to use the
K values
is a constant
determined from the UK national
The latter values are provided
, database.
are constants
in the UK National Annex in the form of
b theisconstants
the normalised
mean compressive strength of the units,
to be used in the following
in the direction of the applied action effect, in N/mm2
equation:
m is the compressive strength of the mortar, in N/mm2
fk = K fb fm [Equation 3.1 of BS EN
Values of K to be used1996-1-1]
with Equation 3.1 are provided in the UK National
Annex Table NA.4 and are shown in Table 2.
where:
0.45
*
*
Manufactured
stone
Group 3 and
4 units have not traditionally been used in the UK, so no values are available.
and mortar combinations have not traditionally been usedin the UK, so no values are
Group 1These masonry unit0.75

0.90
available.
If Group 1 aggregate concrete units contain formed vertical voids in the normal direction,
Dimensioned
stone
multiply K natural
by (100 - n)/100,
where n is the percentage of voids, maximum 25%.
When aggregate concrete masonry units are to be used laid flat the normalised strength of the unit should be
calculated using the width and height of the unit in
the upright position along
with the compressive strengthof
Group 1
0.45
the unit tested in the upright position.

Note
* Group 3Where
and 4aunits
have
traditionally
used
in the
UK, sobe
nomodified
values are
available.
mortar
jointnot
is parallel
to thebeen
face of
the wall
K should
(Figure
1).
These masonry unit and mortar combinations have not traditionally been used in the UK, so no values are available.
If Group 1 aggregate concrete units contain formed vertical voids in the normal direction, multiply K by (100 - n)/100,
where n is the percentage of voids, maximum 25%.
When aggregate concrete masonry units are to be used laid at the normalised strength of the unit should be calculated
using the width and height of the unit in the upright position along with the compressive strength of the unit tested in the
upright position.
Note
Where a mortar joint is parallel to the face of the wall K should be modied (Figure 1).
TSE38_38-43 Concrete v3.indd 40
dried compressive strength of an equivalent
wide by 100mm
high masonry
The100mm
characteristic
compressive
strength of masonry (other than shellunit. The detail is contained in the test
bedded
masonry)
is
determined
from the results of tests in accordance
methods for masonry units in BS EN 772The
are carried
with114BS
ENadvantage
1052-115. to
. The
thetests
designer
is that out on small wallette
the normalised
independent panels used in BS 5628. The
specimens
rather strength
than theisstorey-height
of the size of the units used in the nal
designer has the option of either having the units intended to be used
construction thereby obviating the need
in afor
project
tested or
the
values
determined
from the UK national
recalculation
if ato
diffuse
erent
size
of unit
is
selected.
database.
The latter values are provided in the UK National Annex in the
The characteristic compressive strength
form of
the constants to be used in the following equation:
tests in accordance with BS EN 1052-1 .
Calcium
silicate
Calcium
silicate
Group Group
1
1
The normalised strength is new to the UK and relates the compressive

strength
unit
determined
by test
fb a standardised shape
fm is of
notthe
to be
taken
to be greater
than to
12N/mm2content. The normalised compressive strength is the
andormoisture
the coefficient of variation of the strength
compressive
strength of the units converted to the airdried compressive
of the masonry units is not more than 25%
strength of an equivalent 100mm wide by 100mm high masonry
Thedetail
normalised
strength in
is new
to the
UK
unit. The
is contained
the test
methods
for masonry units in
and772-1
relates
14 the compressive strength of the
BS EN
. The advantage to the designer is that the normalised
unit determined by test to a standardised
strength
independent
of theThe
sizenormalised
of the units used in the final
shapeis
and
moisture content.
construction
thereby
obviating
the need for recalculation if a different
compressive
strength
is the compressive
the units converted to the airsizestrength
of unit isofselected.
fk is the characteristic compressive

The value
of K is reduced by multiplying by 0.8 when a mortar joint runs
strength of the masonry, in N/mm2
continuously
intermittently through the masonry at right angles to
K is a or
constant
, joints,
are constants
the cross
as shown in Figure 1. Note also that for blocks laid flat,
fb contains
is the normalised mean
the table
a specific value for K to be used in Equation 3.1.
compressive strength of the units,
in the direction of the applied

Values of ,
for use with Equation 3.1 are shown in Table 3 and are
action effect, in N/mm2
takenffromisclause
NA.2.4 ofstrength
the UK National
Annex to Part 1-1.
the compressive
of the
m
mortar, in N/mm2
Values of K to be used with Equation 3.1

are provided in the UK National Annex Table
NA.4 and are shown in Table 2.
The value of K is reduced by multiplying
by 0.8 when a mortar joint runs continuously
or intermittently through the masonry at right
angles to the cross joints, as shown in Figure
1. Note also that for blocks laid at, the table
contains a specic value for K to be used in
Equation 3.1.
Values of , for use with Equation 3.1 are
shown in Table 3 and are taken from clause
22/01/2015 12:26
41
Figure 3
Values of m against slenderness ratio for different eccentricities, based on E of 1000fk
1996-1-1 and depends on the ratio of the pier

spacing to pier width and the ratio of the pier
depth to the wall thickness.
For awall
cavity
wall
the effective
thickness
For a cavity
the
effective
thickness
in the UK is determined using the
in the UK is determined using the following
following
equation:
equation:
tef =
t 1 + t32 3
3
=
1 + 2
where:
is the effective thickness of the
1
outer or unloaded leaf
ist thethe
effective
thickness of the outer or unloaded leaf
effective thickness of the inner
2
the effective
thickness
of the inner or loaded leaf
or loaded leaf
where: t
t1
t2
Note that
the
effective
thickness
of the
leaf should not be
Note
that
the effective
thickness
ofunloaded
the
leaf should
taken to of
bethe loaded leaf and that ties
takenunloaded
to be greater
than not
thebe
thickness
greater
than the at
thickness
. loaded leaf
should
be provided
2.5 per of
m2the
and that ties should be provided at 2.5 per
m.
The slenderness
ratio of the wall is obtained by dividing the effective
The slenderness ratio of the wall is
height by the effective thickness and should not be greater than 27 for
obtained by dividing the effective height by
wallsthe
subjected
mainly vertical
loading.
effectiveto
thickness
and should
not be
2
greater than 27 for walls subjected to mainly
Figure 3 Values of m against slenderness ratio for different eccentricities, based on E of 1000k
Whenvertical
a wallloading.
is subjected to vertical loads which result in an eccentricity
a wall
is subjected
vertical
at rightWhen
angles
to the
line of thetowall,
EC6loads
requires the resistance of the
which
result in an
at right angles
wall to
be checked
at eccentricity
the top, mid-height
and bottom.
to the line of the wall, EC6 requires the
resistance ofof
the
wall
to be
checked
at the
The eccentricity
the
load
at the
top and
bottom of the wall is given by:
designation
mortars
has1-1.
also changed
the need
for
a of the
top, mid-height and bottom.
NA.2.4 The
of the
UK Nationalof
Annex
to Part
h iswith
the clear
storey
height
eccentricity of the load at the top and
The designation
mortars
also rather than mix
wall
MThe
declarationofbased
onhas
strength
proportions. Thus, an
id
ehethe
+ ewall
0.05t
changed with the need for a declaration
n is a reduction
factor, where n = 2, ei = bottom+of
int is given by:
M12 mortar may be expected to have a strength
of 12N/mm2. Equivalent
Nid
based on strength rather than mix
3 or 4, depending upon the
mixesThus,
are shown
National
Table NA.2
andrestraint
are shown
inening
Tableof4.the
proportions.
an M12in
mortar
may Annex
be
edge
or stiff
where: = id +
i
he + init 0.05
expected to have a strength of 12N/mm2.
wall. The reduction factor to be
Eccentricity
id
applied depends upon the restraint M is the design
Equivalent mixes are shown in National
value of the bending moment at the top or the
id
where:
offered by adjoining elements
Annex Table NA.2 and are shown in Table 4.
A further area of change for vertical load relates to the treatment of
bottom
wallvalue
resulting
eccentricity of the floor load at
Mid is of
thethe
design
of thefrom
bending
eccentricity, where a frame analysis approach
is implied
than
the tmin
the support
The value
for the rather
minimum
thickness
Eccentricity
moment at the top or the bottom
A further
change forofvertical
load that any
of the wall
resulting
of eccentricity
a load-bearing
BSarea
5628ofapproach
assuming
atwall
the should
top of be
thetaken as Nid is the design
value
of thefrom
vertical load at the top or the bottom of
relates to the treatment of eccentricity,
eccentricity of the oor load at the
90mm for a single-leaf wall and 75mm for
wall reduces to zero at the bottom of the
wall. The use of a frame analysis
the wall
where a frame analysis approach is implied
support
the leaves of a cavity wall. For a single-leaf
will not
usually
justifiedofgiven typical
UKa construction
practice
eccentricity
theof
top
bottom of the wall resulting from
rather than
the BS
5628be
approach
wall,
double-leaf wall,
a facedand
wall,ita shell-ehe isNthe
is the designat
value
theorvertical
id
assuming
thatacceptable
any eccentricity
at the
the BS
top5628 approach.
bedded wall and a grouted cavity wall the imposed
load
at the top loads
or the bottom of the
is still
to use
horizontal
of the wall reduces to zero at the bottom
effective thickness tef is taken as the actuale is the initial
wall eccentricity resulting from construction imperfections
init
is used to allow
for anyofinaccuracies
An initial
eccentricity
of the wall.
The use
of a frame eanalysis
thickness
the wall t. in the
ehe is the eccentricity at the top or
init
and may
be taken as h /450
will notconstruction
usually be justi
given
typicaland is takenWhen
a wall
stiffened
(Figure
bottom of the wallefresulting from
ofed
the
masonry
as einit
= hefis/450
wherebyhpiers
is the
ef
UK construction practice and it is still
2), the effective thickness is given by the
imposed horizontal loads
effective height of the wall. The initial eccentricity is applied to the full
acceptable to use the BS 5628 approach.
coefficients contained in Table 5.1 of BS EN
einit is the initial eccentricity resulting
height
of the wall
to result
in the greatest value of eccentricity.
An initial
eccentricity
Table 3: Values of K to be used with Equation 3.1
einitsois as
used
to allow
for any inaccuracies in the construction of
Masonry unit
Masonry unit
is obtained
applying
The effective
height
a masonry wall hTable
3: Values by
to be
used inaEquation 3.1
the masonry
and is taken
as eof
ef
init = hef/450
factor
toeff
the
clearheight
height
of the
wall such that:
where h
ective
of the
wall.
ef is the
Type of mortar
The initial eccentricity is applied to the
h
hef =
full height
ofthe
n wall so as to result in the
General purpose mortar
greatest value of eccentricity.

Lightweight mortar
= 0.7 and = 0.3
Values to be used
= 0.7 and = 0.3
Thin layer mortar in bed jointsof=thickness

0.7 and 0.53mm
= 0.3 (using
clay units of Group 1, calciumsilicate

where:
The eff
ective height of a masonry wall hef
Lightweight mortar
= 0.7units,
and aggregate
= 0.3
concrete units and autoclaved aerated concrete units)
is obtained by applying a factor to the clear
Thin
layer
mortar
in
bed
joints
of
thickness
0.53mm
(using
=
0.85
and = 0
h
is the
clear
storey height of the wall
height of
the wall
such
that:
Thin layer mortar in bed joints of thickness 0.53mm (using
clay units of Group 1, calcium silicate units, aggregate
3 or 4, depending
upon the aerated
edge concrete
n is a reduction factor, where n = 2,concrete
units and autoclaved
units)
clay units
of Group 2)
hef = nh restraint or stiffening of the wall. The reduction factor to be

Thin layer mortar in bed joints of thickness 0.53mm (using
adjoining
elements
clayoffered
units ofby
Group
2)
where: applied depends upon the restraint
and is given in BS EN 1996-1-1 clause 5.5.1.2 (11).
The value for the minimum thickness tmin of a load-bearing wall
should be taken as 90mm for a single-leaf wall and 75mm for the leaves
of a cavity wall. For a single-leaf wall, a double-leaf wall, a faced wall,
a shellbedded wall and a grouted cavity wall the effective thickness
tef is taken as the actual thickness of the wall t.
When a wall is stiffened by piers (Figure 2), the effective thickness is
given by the coefficients contained in Table 5.1 of BS EN 1996-1-1 and
depends on the ratio of the pier spacing to pier width and the ratio of
the pier depth to the wall thickness.
= 0.85 and = 0
= 0.7 and = 0
= 0.7 and = 0
22/01/2015 12:26
Table 4: Acceptable assumed equivalent mixes for prescribed masonry mortars for Class 2 of execution control
Compressive
strength class*
Prescribed mortars (traditional proportion of materials by volume) (see Notes)

Cement: lime: condition
sand with or without air
entrainment
Cement: sand with or

without air entrainment
Masonry cement: sand
Masonry cement: sand
Mortar designation
Suitable for use

in environmental
condition
Severe(S)
M12
1 : 0 to : 3
1:3
Not suitable
Not suitable
(i)
M6
1 : : 4 to 4
1 : 3 to 4
1 : 2 to 3
1:3
(ii)
Severe(S)
M4
1 : 1 : 5 to 6
1 : 5 to 6
1 : 4 to 5
1 : 3 to 4
(iii)
Moderate(M)
M2
1 : 2 : 8 to 9
1 : 7 to 8
1 : 5 to 6
1 : 4
(iv)
Passive(P)
The number following the M is the compressive strength for the class at 28 days in N/mm2 that may be assumed for the proportions given in columns 2 to 4; site compressive strength testing is not required for these traditional mixes.
Checking of prescribed mortars should only be done by testing the proportions of the constituents.
Cement or combinations of cement (which include CEM I and many CEM IIs) in accordance with NA.2.3.2, except masonry cements
Masonry cement in accordance with NA.2.3.2 (inorganic filler other than lime)
Masonry cement in accordance with NA.2.3.2 (lime)

*
Notes
When the sand portion is given as, for example, 5 to 6, the lower figure should be used with sands containing a higher proportion of fines whilst the higher figure should be used with sands containing a lower proportion of fines.
For Class 2 of execution control site compressive strength testing is not required for these traditional mixes and checking of prescribed mortars should only be done by testing the proportions of the constituents.
Characteristic vertical actions
Masonry unit properties

Type and group
Dimensions
Strength
Determine requirements
for mortar strength and
durability. See tables 5
& 6 of Introduction to
Eurocode 6
Determine normalized
compressive strength, fb.
Determine characteristic compressive strength of masonry,

fk, from Equation (3.1) of Eurocode 6 and Tables 1 & 2
Determine effective height, hef,

of the wall (see page 4) .
Determine effective thickness, tef,

of the wall (see page 4)
Check area 0.04 m2
Check slenderness ratio hef /tef 27
i = 1 2
Determine design value of
vertical actions (per unit
length), Ed, using Expression
(6.10), (6.10a) or (6.10b) of
Eurocode (see Introduction to
Eurocode 6)
Obtain gM from table 1 of

Introduction to Eurocode 6
Check complete
Figure4 Flow chart for design of masonry walls to resist vertical actions
The eccentricity at the mid-height of the wall emk includes the initial
eccentricity einit, the horizontal load eccentricity ehm, and the load
eccentricity em.
The mid-height eccentricity emk is:
emk = em + ek which must be greater than or equal to 0.05t
em =
Mmd
Nmd
where:
i is the reduction factor at the top or bottom of the wall
t
Check Ed NRd
Determine eccentricity (see page 5)
Determine capacity reduction factors, Fm and Fi from (see page 6)
ei
ei is the eccentricity at the top or bottom of the wall
Where cross-sectional area,

A < 0.1 m2, factor fk by (0.7 + 3A)
Calculate design resistance

(per unit length) from least
favourable of:
NRd = Fm t fk / gM
and
NRd = Fi t fk / gM
At the top or bottom of the wall, the reduction factor for slenderness and
eccentricity is given by:
+ ehm + ei
where:
Mmd is the design value of the greatest moment at the middle height of
the wall resulting from the moments at the top and bottom of the
wall, including any load applied eccentrically to the face of the wall
is the thickness of the wall
The capacity reduction factor in the middle of the wall m may be

determined by using either the equation or the graph (Figure 3), which
is given in Annex G of BS EN 1996-1-1. The value of the modulus of
elasticity to be taken in the UK is 1000k.
The design resistance of a single-leaf wall per unit length NRd is given by:
NRd = td
where:
is the capacity reduction factor allowing for the effects of
slenderness and eccentricity of loading (the least favourable value
obtained for the top, bottom and mid-height, i or m)
t
is the thickness of the wall
d is the design compressive strength of the masonry and is taken

as k/M
Nmd is the design value of the vertical load at the middle height of the
wall, including any load applied eccentrically to the face of the wall
ehm is the eccentricity at the middle of the wall resulting from imposed
horizontal loads
ek is the creep eccentricity and, in the UK, can be ignored if the
slenderness ratio is not greater than 27
For sections of small plan area, less than 0.1m2, d should be multiplied
by (0.7 + 3A), where A is the loadbearing horizontal cross-sectional area
of the wall in m2.
A flow chart for vertical load design is provided in Figure 4.
References:
Eurocode 6. Design of masonry structures. General rules for reinforced
and unreinforced masonry structures, London, UK: BSI
Concentrated loads
Concentrated loads are dealt with by a calculation approach using a
dispersion angle of 60. Only Group 1 masonry units may be enhanced.
2) British Standards Institution (2005) BS EN 1996-1-2:2005 Eurocode

6. Design of masonry structures. General rules. Structural fire design,
London, UK: BSI
For a Group 1 unit (not shell bedded) the vertical load resistance is
given by:
3) British Standards Institution (2006) BS EN 1996-2:2006 Eurocode

6. Design of masonry structures. Design considerations, selection of
materials and execution of masonry, London, UK: BSI
NRdc = Abd (Ex. 6.10: EN 1996-1-1)

where:
=
1 + 0.3
a1
hc
)(
1.5 1.1
Ab
Aef
(where Ab/Aef is not to be taken as greater than 0.45) which should not be
less than 1.0 nor taken to be greater than:
1.25 +
a1
2hc
or 1.5, whichever is the lesser
where:
is an enhancement factor for load
a1 is the distance from the end of the wall to the nearer edge of the
loaded area
hc is the height of the wall to the level of the load
Ab is the loaded area
Aef is the effective area of the bearing
lefm is the effective length of the bearing as determined at the midheight of the wall or pier
t is the thickness of the wall, taking into account the depth of recesses
in joints greater than 5mm
For walls built with Groups 2, 3 and 4 masonry units, and when shell
bedding is used, it is necessary to check that, locally under the bearing
of a concentrated load, the design compressive stress does not exceed
the design compressive strength of the masonry fd (i.e. is taken to
be 1.0).
In any case, the eccentricity of the load from the centre line of the wall
should not be greater than t/4.
Further Reading:
1) Roberts J. J. and Brooker O. (2013) How to design masonry structures
using Eurocode 6. 1. Introduction to EC6, London, UK: MPA The Concrete
Centre
using Eurocode 6. 3. Vertical Resistance, London, UK: MPA The Concrete
Centre
using Eurocode 6. 3. Lateral resistance, London, UK: MPA The Concrete
Centre

6. Design of masonry structures. Simplified calculation methods for
unreinforced masonry structures, London, UK: BSI
5) British Standards Institution (2013) NA to BS EN 1996-1 1:2005
+A1:2012 UK National Annex to Eurocode 6. Design of masonry
structures. General rules for reinforced and unreinforced masonry
structures, London, UK: BSI
6) British Standards Institution (2013) NA to BS EN 1996-1-2:2005
UK National Annex to Eurocode 6. Design of masonry structures.
General rules. Structural fire design, London, UK: BSI
7) British Standards Institution (2007) NA to BS EN 1996-2:2006 UK
National Annex to Eurocode 6. Design of masonry structures. Design
considerations, selection of materials and execution of masonry,
London, UK: BSI
8) British Standards Institution (2007) NA+A1:2014 to BS EN 19963:2006 UK National Annex to Eurocode 6. Design of masonry structures.
Simplified calculation methods for unreinforced masonry structures,
London, UK: BSI
9) British Standards Institution (2010) PD 6697:2010 Recommendations
for the design of masonry structures to BS EN 1996-1-1 and BS EN 1996-2,
London, UK: BSI
10) British Standards Institution (2005) BS 5628-1:2005 Code of
practice for the use of masonry. Structural use of unreinforced masonry,
London, UK: BSI
11) British Standards Institution (2005) BS 5628-2:2005 Code of practice
for the use of masonry. Structural use of reinforced and prestressed
masonry, London, UK: BSI
12) British Standards Institution (2005) BS 5628-3:2005 Code of
practice for the use of masonry. Materials and components, design and
workmanship, London, UK: BSI
13) British Standards Institution (2013) BS EN 845-2:2013 Specification
for ancillary components for masonry. Lintels, London, UK: BSI
14) British Standards Institution (2000) BS EN 772-1:2000 Methods
of test for masonry units. Determination of compressive strength,
London, UK: BSI
15) British Standards Institution (1999) BS EN 1052-1:1999 Methods of
test for masonry. Determination of compressive strength, London, UK: BSI
Eurocode 6: Design of masonry structures

for lateral loads and other factors
The previous article in the series explained how to

design masonry for vertical actions1.
Introduction
Throughout this article the Nationally Determined Parameters (NDPs)
from the UK National Annexes (NAs) have been used. These enable
Eurocode 6 (BS EN 1996-1-1)2 to be applied in the UK.
Lateral loads
EC6 offers two approaches to the design of laterally loaded panels. The
first method relies on the flexural strength of the masonry and makes
use of yieldline analysis. The second method is an approach based on
arching and the assumption of a three-pinned arch being formed within
the wall. Figure 1 shows a flow chart for lateral load design.
The flexural strength approach is the most widely used and does
not depend upon rigid supports to resist arch thrust. Values of the
characteristic flexural strength of masonry are provided in Table NA.6 of
EC6 and are shown here in Table 1.
The assessment of the edge conditions is a requirement for the flexural
strength approach. A free edge is easily identified but some judgment
on the part of the engineer is necessary in deciding between simply
supported or fixed conditions. When considering the vertical support
condition, attention also needs to be paid to the potential position of
movement joints, and to the changes the provision of such joints makes
to the panel size and restraint conditions.
Where the walls are not rectangular (e.g. a trapezoidal-shaped wall to
a mono-pitched structure), engineering judgement may be applied to
determine the effective wall height.
Wall panels with openings need to be treated with care and may
typically be subdivided into smaller panels around the opening. It is
beyond the scope of this article to deal with the topic in detail and
reference should be made to suitable handbooks3. Alternatively, a yieldline analysis from first principles may be applied to wall panels4.
If a damp-proof course (DPC) is present in a wall subjected to flexure,
then the degree of fixity may be altered. The bending moment
coefficient at a DPC may be taken as that for an edge over which
full continuity exists, provided that there is sufficient vertical load on
the DPC to ensure that the flexural strength capacity is not exceeded.
Walls may be either horizontally or vertically spanning, or both, and the
ultimate strength of the wall is governed by the capacity of the masonry
to resist flexural tension. This capacity is enhanced by the presence of
vertical load, as the masonry can resist significantly more in compression

than it can in tension. Clearly, the potential flexural strength is greater if
the potential plane of failure is perpendicular rather than parallel to the
bed joint, due to the interlocking of the masonry units.
The designer needs to assess the panel support conditions (or assume a
free edge) and decide whether these provide simple or continuous (fully
restrained) support. Care also needs to be exercised in considering the
effect of DPCs, movement joints, openings in walls etc.3
For panels without openings, the bending moments per unit length MEd are:
MEd1 = a1 WEdl2 when the plane of failure is parallel to the bed joints
MEd2 = a2 WEdl2 when the plane of failure is perpendicular to the bed joints
where:
a1 is the bending moment coefficient parallel to the bed joints
(= a2, Annex G of BS EN 1996-1-1)
a2 is the bending moment coefficient perpendicular to the bed
joints (Annex G of BS EN 1996-1-1)
WEd is the design wind load per unit area Q Wk
l
is the length of panel between supports
is the orthogonal ratio xk1/xk2

The presence of a vertical load increases the flexural strength
of a panel in the direction parallel to the bed joints. The design
moment of resistance within the height of the wall is given by:
MRd =
xk1
M
+ d Z
where:
xk1 is the characteristic flexural strength of masonry bending about an
axis parallel to bed joints (Table 1)
M is the appropriate partial factor for materials
d is the design vertical load per unit area (<0.15NRd) taken where

the bending resistance is being considered
Z is the section modulus of the plan shape of the wall
k is the characteristic compressive strength
The analysis can be based upon a three-pin arch, and the bearing of the
arch thrust at the supports and at the central hinge should be assumed
to be 0.1 times the thickness of the wall.
The design procedure is iterative and may be summarised as:

1. make initial assumption of support condition
Shear
2. make assumptions as to strength and thickness of masonry unit

required; the minimum wall thickness or thickness of one leaf of a
cavity wall is 100mm
The characteristic shear strength of masonry is a function of the

characteristic initial shear strength of the masonry and the design
compressive stress orthogonal to the shear plane being considered.
The values of the initial shear strength of masonry are given in Table
NA.5 of EC6 and are shown here in Table 2.
3. check serviceability slenderness limits. For wall panels supported

top and bottom only, h should be limited to 30t. For other support
conditions, use figures shown in Annex F of BS EN 1996-1-1
The characteristic shear strength is given by the following relationships:
4. determine orthogonal ratio and hence bending moment

coefficient appropriate to panel shape (see Table 2 in Roberts and
Brooker5)
vk = vko + 0.4d but not greater than 0.065b or vlt for fully filled mortar
joints
5. determine design value of applied moment MEd
vk = 0.5vko + 0.4d but not greater than 0.045b or vlt for unfilled
perpend joints
6. check design value of moment of resistance MRd
where:
7. if MEd MRd then the wall is acceptable; if not, return to either step 1
or 2 and modify
vk is the characteristic shear strength of masonry
8. check shear
vko is the characteristic initial shear strength of masonry, under zero

compressive stress
In a cavity wall, the design lateral load per unit area WEd may be
apportioned (either according to capacity or stiffness) between
the two leaves, provided that the wall ties are capable of transmitting
the actions that result from the apportionment.
d is the design compressive stress perpendicular to the shear in the

member at the level under consideration, using the appropriate
load combination based on the average vertical stress over the
compressed part of the wall that is providing shear resistance
EC6 indicates that the flexural strength of masonry should not be used
in design of walls subjected to lateral earth pressure.
b is the normalised compressive strength of the masonry units (as

described in clause 3.1.2.1 of EC6 for the direction of application of
the load on the test specimens being perpendicular to the bed face
Where a masonry wall is built between supports capable of resisting

an arch thrust, it may be assumed that a horizontal or vertical arch
develops within the thickness of the wall in resisting a lateral load.
vlt is the limit value of vk which in the UK is taken as the limit 0.065b
and 0.045b respectively for filled and unfilled joints
Masonry unit properties

Type and group
Dimensions
Strength
h
Determine requirements
for mortar strength and
durability. See tables 5 & 6
of Introduction to Eurocode 6
Characteristic lateral actions
Obtain height, h
Determine design value of

lateral actions, WEd, using
Expressions (6.10), (6.10a) or
(6.10b) of Eurocode. (see
Introduction to Eurocode 6)
Obtain water absorption

for clay masonry unit or
declared compressive
strength from
manufacturer for other
masonry types
Obtain fxk1 and fxk2 from Table 1 and calculate

orthogonal ratio, m where: m = fxk1 / fxk2
Obtain thickness, t
Check slenderness ratio h/t 30 for walls

supported top and bottom only
Obtain a from Table 2 and calculate, MEd1 and MEd2, where:

MEd1 = a1 WEd l 2, parallel to the bed joints or
MEd2 = a2 WEd l 2, perpendicular to the bed joints
Check serviceability slenderness limits using Figures 2, 3 or 4 as appropriate
Characteristic vertical
actions
Obtain, gM from table 1

Introduction to Eurocode 6
Determine design value of vertical load, GEd, using

Expressions (6.10), (6.10a) or (6.10b) of Eurocode.
(see Introduction to Eurocode 6)
Figure 1 Flow chart for design of masonry walls to resist lateral actions
Obtain fxk1 from Table 1 and calculate design moment

of resistance parallel to the bed joints, MRd1, where:
Check MEd1 MRd1

and MEd2 MRd2
MRd1 = gxk1 + s d Z
M
Obtain fxk2 from Table 1 and calculate design moment
of resistance perpendicular to the bed joints, MRd2, where: z
f
MRd2 = gxk2 Z
M
Check shear
Table 2: Characteristic flexural strength of masonry, xk1 and xk2, in N/mm2

Values of xk1
Plane of failure parallel to bed joints
Values of xk2
Plane of failure perpendicular to bed joints
Mortar strength class:

M12
M6 & M4
M2
M12
M6 & M4
M2
0.4
2.0
1.5
1.2
Clay masonry units of Groups 1 and 2 having a water absorption* of:

Less than 7%
0.7
0.5
Between 7% & 12%
0.5
0.4
0.35
1.5
1.1
1.0
Over 12%
0.4
0.3
0.25
1.1
0.9
0.8
Calcium silicate brick-sized masonry units

0.3
0.2
0.9
0.6
0.3
0.2
0.9
0.6
Aggregate concrete brick-sized masonry units
Aggregate concrete masonry units and manufactured stone of Groups 1 and 2 and AAC masonry units used in walls of thickness up to 100mm,** of declared compressive strength (N/mm2):
2.9
0.4
0.4
3.6
0.25
0.2
0.45
0.4
7.3
0.6
0.5
Aggregate concrete masonry units and manufactured stone of Groups 1 and 2 and AAC masonry units used in walls of thickness of 250mm or greater ,** of declared compressive strength (N/mm2):
2.9
0.15
0.25
0.2
3.6
0.1
0.25
0.2
7.3
0.35
0.3
Aggregate concrete masonry units and manufactured stone of Groups 1 and 2 and AAC masonry units used in walls of any thickness of declared compressive strength (N/mm2):
10.4
0.25
0.2
17.5
0.75
0.6
0.9
0.7
Key
* Tests to determine the water absorption of clay masonry units are to be conducted in accordance with BS EN 77273
Units not exceeding 337.5mm 225mm 112.5mm
Autoclaved aerated concrete (aircrete)
The thickness should be taken as the thickness of the wall, for a single-leaf wall, or the thickness of the leaf, for a cavity wall
** Linear interpolation may be used to obtain the values of xk1 and xk2 for:
1) wall thicknesses greater than 100mm and less than 250mm
2) compressive strengths between 2.9N/mm2 and 7.3N/mm2 in a wall of given thickness
When used with flexural strength in the parallel direction, assume the orthogonal ratio = 0.3
Table 2: Values of initial shear strength of masonry, vko

Masonry units
Strength class of
general purpose
mortar
Clay
Calcium silicate
Aggregate concrete, autoclaved

aerated concrete, manufactured
stone and dimensioned natural
stone
Initial shear strength, vko (N/mm2)

Thin layer mortar (bed joint 0.5mm

and 3mm)
Lightweight mortar
0.30
0.15
0.40
0.15
0.30
0.15
M12
0.30
M4 & M6
0.20
M2
0.10
M12
0.20
M4 & M6
0.15
M2
0.10
M12
0.20
M4 & M6
0.15
M2
0.10
Reinforced and prestressed masonry

BS EN 1996-1-1 contains a limited amount of information relating to
the design of reinforced masonry and no detailed information on the
design of prestressed masonry. The clauses relating to reinforced
masonry contain information similar to that which was provided in
BS 5628-26 and Annex J permits enhancement to shear in a similar
way to BS 5628-2.
Fire design
BS EN 1996-1-27 provides information on the passive fire resistance
of masonry walls so that the designer can ensure that the loadbearing
performance is maintained for the necessary period of time and that the
fire is appropriately contained.
For the UK, fire design largely remains in the form of tables addressing
the requirements of the Building Regulations in the UK. The fire
resistance of a loadbearing wall comprises two values depending
on how highly loaded the wall is; it is further enhanced if the wall
is plastered. Most designers will find that the tabulated data cover
most situations, but there is also provision for testing. The tables
cover loadbearing and non-loadbearing walls, single-leaf, cavity and
separating walls.
Materials and execution
BS EN 1996-28 gives basic rules for the selection of materials and
execution to enable it to comply with the other parts of BS EN 1996.
It deals with:
The micro conditions of exposure to which completed masonry is

subjected are divided into five main classes, namely:
MX1 in a dry environment
MX2 exposed to moisture or wetting
MX3 exposed to moisture or wetting plus freeze/thaw cycling
MX4 exposed to saturated salt air or seawater
MX5 in an aggressive chemical environment
BS EN 998-29 defines the durability classes for factory and semi-factory
produced mortar. These may be expressed as follows:
P mortar for use in masonry subjected to passive exposure,

i.e. which is not intended to be exposed to moisture and

freezing conditions
M mortar for use in masonry subjected to moderate exposure,

i.e. which is exposed to moisture and freeze/thaw cycling,

excluding constructions subjected to severe exposure
S mortar for use in masonry subjected to severe exposure,

i.e. which is subjected to saturation with water (driving rain,

ground water) combined with frequent freeze/thaw cycling

due to climatic conditions, and an absence of protective features
These designations may also be used for site-made mortar and are
related in detail in PD 669710 to the durability of masonry in any given
exposure situation.
factors affecting the performance and durability of masonry
Of necessity, given the wide range of products and usages in Europe,

the information contained in BS EN 1996-2 is limited, but for UK use the
detail is provided in PD 6697.
resistance of buildings to moisture penetration
Simplified calculation methods
storage, preparation and use of materials on site
BS EN 1996-311 deals with simplified calculation methods for

unreinforced masonry which are derived from BS EN 1996-1-1 and will
tend to result in less economic design. They should not be confused
with the simple rules found in Approved Document A of the Building
Regulations for England and Wales12 and the Low Rise Housing Code
(BS 8103-2)13, both of which are likely to produce more cost-effective
outcomes.
the selection of masonry materials
the execution of masonry

masonry protection during execution
References:
Eurocode 6. Design of masonry structures. General rules for reinforced
and unreinforced masonry structures, London, UK: BSI
2) British Standards Institution (1998) BS EN 772-7:1998 Methods
of test for masonry units. Determination of water absorption of clay
masonry damp proof course units by boiling in water, London, UK: BSI
3) Roberts J., Tovey A. and Fried A. (2001) Concrete Masonry
Designers Handbook (2nd ed.), London, UK: Spon Press
4) Johansen, K. W. (1972) Yield-line formulae for slabs, Leatherhead,
UK: Eyre and Spottiswoode
using Eurocode 6. 3. Lateral resistance, London, UK: MPA The Concrete
Centre
6) British Standards Institution (2005) BS 5628-1:2005 Code of practice
for the use of masonry. Structural use of unreinforced masonry, London,
UK: BSI

6. Design of masonry structures. General rules. Structural fire design,
London, UK: BSI
6. Design of masonry structures. Design considerations, selection of
materials and execution of masonry, London, UK: BSI
9) British Standards Institution (2003) BS EN 998-2:2003 Specification for
mortar for masonry. Masonry mortar, London, UK: BSI
10) British Standards Institution (2010) PD 6697:2010
Recommendations for the design of masonry structures to
BS EN 1996-1-1 and BS EN 1996-2, London, UK: BSI
6. Design of masonry structures. Simplified calculation methods for
unreinforced masonry structures, London, UK: BSI
12) HM Government (2013) The Building Regulations 2010. Approved
Document A: Structure [Online] Available at: www.planningportal.gov.
uk/uploads/br/BR_PDF_AD_A_2013.pdf (Accessed: February 2015)
13) British Standards Institution (2013) BS 8103-2:2013 Structural
design of low-rise buildings. Code of practice for masonry walls for
housing, London, UK: BSI
Further reading
There are a number of guidance documents available:
Morton J. (2011) Designers guide to Eurocode 6: Design of masonry
structures: EN 1996-1-1, London, UK: Thomas Telford
The Institution of Structural Engineers (2008) Manual for the design
of plain masonry in building structures to Eurocode 6, London, UK:
The Institution of Structural Engineers
International Masonry Society (2009) Eurocode for masonry,
EN1996-1-1 and EN 1996-2: Guidance and worked examples,
Penkhull, UK: IMS
Roberts J. J. and Brooker O. (2013) How to design masonry structures
using Eurocode 6. 1. Introduction to Eurocode 6, London, UK: MPA
The Concrete Centre
Roberts J. J. and Brooker O. (2013) How to design masonry structures
using Eurocode 6. 2. Vertical resistance, London, UK: MPA The Concrete
Centre
NHBC (2011) Technical Guidance Note: The Building Regulations
2004 Edition England and Wales. Requirement A3 Disproportionate
Collapse [Online] Available at: www.nhbc.co.uk/NHBCPublications/
LiteratureLibrary/Technical/filedownload,23676,en.pdf (Accessed:
February 2015)
Brick Development Association, Autoclaved Aerated Concrete

Products Association and Concrete Block Association (2005)
Masonry design for disproportionate collapse requirements under
Regulation A3 of the Building Regulations (England & Wales)
[Online] Available at: www.planningportal.gov.uk/uploads/br/
disproportionate_collapse.pdf (Accessed: February2015)
For more guidance on lateral load design and the use of yieldline
theory, see Roberts et al.4 and Johansen5.
Further design information and design examples are available
from: www.eurocode6.org
General information on masonry design and construction is
available from:
www.masonry.org.uk
www.brick.org.uk
www.cba-blocks.org.uk
www.aircrete.co.uk
www.mortar.org.uk
www.modernmasonry.co.uk
Design of post-tensioned slabs
This article provides information on how to scheme

a PT slab and how the use of post-tensioning
affects the rest of the structure.
Introduction
Sheath
Grease
PT flat slabs provide the thinnest readily available structural option for
spans of 7m or more, and can economically be used for spans up to 13m
13m. For longer spans, a one-way spanning slab onto band beams is
frequently used.
Strand
Most PT floors are designed in the UK by specialist designers as part

of a performance specification procurement route. However, the
design is not necessarily complicated and the main designer should
have knowledge of the benefits and limitations of PT design so that a
reasonable scheme design might be considered as a structural option
and produced for the tender documentation, and also so that
the designer can factor the design and its benefits into the overall
stability and robustness of the structure.
Unbonded PT tendon
Bonded PT components
Figure 1 Unbonded and bonded tendons and components
Table 1: Comparison of PT systems

Bonded
Bonded and unbonded systems

Unbonded
Localises effect of accidental damage
Reduced covers to strand
Develops higher ultimate strength
Reduced prestressing force
Does not depend on anchorages

after grouting
Tendons can be prefabricated

leading to faster construction
Can be demolished in same way as

reinforced concrete structures
Tendons can be deflected around

obstructions more easily
Greater eccentricity of strand
Grouting not required
Useful when only one strand is
required, e.g. in rib in ribbed slab
This article provides information on how to scheme a PT slab and how

the use of posttensioning affects the rest of the structure. A more
detailed guide to the design of PT floors can be found in the Concrete
Society Technical Report 43 (TR43), Post-tensioned concrete floors:
Design handbook1. This guide can be used for the design of PT floors
to Eurocode 22 as it is quoted as noncontradictory complementary
information (NCCI) in the UK National Annex3.
There are two types of post-tensioning system available to the engineer:

bonded and unbonded (Figure 1). Most of the post-tensioning work in
the UK is bonded, being about 90% of the market.
Bonded systems have prestressing tendons running through a duct
which is then grouted after prestress has been applied. The ducts can
be circular or flat and hold a number of tendons. The benefit of using
a bonded system is that the anchorages are no longer live after the
grouting has set, which means that any damage to a tendon (e.g. if a
tendon is cut through by a post-drilled fixing) is limited to the bond
length of the tendon on either side of the cut. The tendon is also
protected from corrosion by the grout, in the same way as normal
reinforcement is protected in reinforced concrete. Finally, bonded
systems are as easy to demolish as normal reinforced concrete or
possibly slightly easier, due to less reinforcement within the structure.
FIGURE 9: TYPICAL FLOOR LAYOUTS
a) Favourable layout of restraining walls (low restraint)
FIGURE 9: TYPICAL FLOOR LAYOUTS
b) Unfavourable layout of restraining walls (high restraint)
Figure 2 Typical floor layouts
a) Favourable layout of restraining walls (low restraint)
Unbonded systems are more prevalent in North America and in other

parts of Europe. Here the tendons run through a greased sheath and
are always independent of the concrete. This has no effect on the
serviceability design or performance of a structure under normal
working conditions. It does, however, change both the design
theory and structural performance at the ultimate limit state (ULS).
The anchorages for an unbonded system are live throughout the lifetime
of the structure and if the tendon suffers damage, the prestress provided
by that tendon is lost along the whole of its length.
Where the walls are unfavourably arranged, a calculation of the effects

of movement should be carried out and suitable measures taken to
overcome them. This could involve:
Both systems can be used in the same slab if the design dictates it.
Table 1 gives a comparison between the two systems.
FIGURE 10 : TYPICAL INFILL STRIP

Restraint
At the early stages of a project using PT floors, care must be taken to

avoid the problems
of restraint. This is where the free movement in the
1000 mm
length of the slab under the prestress
forces is restrained, e.g. by the
Post-tensioned
RC infill
strip of shearslab
unfavourable
positioning
walls or lift cores (Figure 2).
All concrete elements shrink due to drying and early thermal effects
but, in addition, prestressing causes elastic shortening and ongoing
shrinkage due to creep. Stiff vertical members, such as stability walls,
restrain the floor slab from shrinking, which prevents the prestress from
developing and thus reduces the strength of the floor.
Where 50mm
the restraining walls are inSlab
a favourable
arrangement and the
to remain
floor isseating
in an internal environment, the length of the floor without
fully propped
movement joints can be up to 50m. However, full consideration should
until infill strip
be given to the effects of shrinkage due to drying, early thermal effects,
cured
elastic shortening and creep in the
design. A strain of 650 should
be considered normal.
using infill strips, also known as pour strips, which are usually cast
around 28 days after the remainder of the floor, to allow initial
shrinkage to occur
(Figure 3)
b) Unfavourable
layout of restraining walls (high restraint)
increasing the quantity of conventional reinforcement, to control

the cracking
using temporary release details
FIGURE
11: TEMPORARY
using a proprietary
temporary
release detail RELEASE DETAIL
reducing the stiffness of the restraining elements
Infill later
The effect of the floor shortening on the columns should also be

Post-tensioned slab
considered in their design, as this may increase the design moments.
FIGURE 10 : TYPICAL INFILL STRIP
FIGURE 11: TEMPO
Infill late
1000 mm
RC infill strip
Post-tensioned
slab
2 layers of slip strip

50mm
seating
Slab to remain
fully propped
until infill strip
cured
100mm bearing
Figure 3 Typical infill strip
Number 7
40
July 2015
Technical
Post-tensioned slabs
Design to prevent disproportionate collapse

PT floor systems are usually designed to resist disproportionate collapse
through detailing of the tendons and reinforcement.
In bonded systems, the tendons can be considered to act as horizontal
Figure 4
ties. In unbonded systems,
thechart
tendons
cannot be relied on and the
Design ow
for PT slabs
conventional reinforcement acts as the horizontal ties.
Materials and specification

In bonded systems, the tendons can
PT slabs do not require

particularly
strength ties.
concrete
and
be considered
to acthigh
as horizontal
In
often class C32/40
is used in
a typical
slab design.
unbonded
systems,
theflat
tendons
cannotFor
be speed of
relied onthe
and
the conventional
reinforcement
construction, however,
concrete
should have
high early strength.
as the horizontal
ties. out as early as possible,
This allows initialacts
prestressing
to be carried
usually after 24 hours,
to prevent
cracking.
Final stressing can take
Materials
and speci
cation
place after threePT
days,
once
the
concrete
has
reached
a predetermined
slabs do not require particularly
highstrength
concrete
and often
classlevels
C32/40
strength, allowing
striking
of formwork.
Higher
of cement
is used
in a typical at slab
design. For
replacements, e.g.
ground-granulated
blastfurnace
slag (GGBS) or fly ash,
of construction the concrete should
can be used, butspeed
will increase
the programme length and may change
have high early strength. This allows initial
the parameters used
in
design,
such
as theout
strains
dueasto creep and early
prestressing to be
carried
as early
age shrinkage. possible, usually after 24 hours, to prevent
cracking. Final stressing can take place
Common strandafter
typesthree
useddays,
in the
UK the
are concrete
given in Table
once
has 2. It is
recommended that
onlyaone
of these strand
typesallowing
is used on any
reached
predetermined
strength,
striking of formwork. Higher levels of cement
project.
replacements, e.g. ground-granulated blast-
furnace
slag (GGBS)
orfloors
y ash,
beinused,
A specification for
the execution
of PT
is can
given
the National
4 programme length and
but will
increase the
, section 7.
Structural Concrete
Specification
may change the parameters used in design,

such as the strains due to creep and early
Cover
age shrinkage.
As with other forms
of reinforced
theincover
is are
determined by
Common
strandconcrete,
types used
the UK
consideration of:given in Table 2. It is recommended that only
one of these strand types is used on any
corrosion protection
project.
A specication for the execution of PT
bond
oors is given in the National Structural
Concrete Specication4, section 7.
fire protection
Cover
The cover required for bond considerations for bonded systems is the
As with other forms of reinforced concrete,
diameter of the duct
for circular
ducts; for
ducts it is the
the cover
is determined
by flat
consideration
of: larger of half
the larger dimension or the smaller dimension. For unbonded systems,
corrosion
the cover required
for bondprotection
is the diameter of the sheath.
$
"

$

"$

"
! " $

"$

"$

!

"$

"
"#

Figure 4 Design flow chart for PT slabs
At the ULS the pre-compression in the section is ignored and checks
checked at the ULS in a similar manner to

detailed guidance, based on Eurocode 2, is
are made to ensure that the section
has sufficient moment capacity.
that for reinforced concrete design, although
available in TR43.
Shear
stresses
are
also
checked
at
the
ULSt in
similar
manner
to that
the bene
of athe
prestress
across
the for
shear
At the serviceability condition, the
reinforced
concrete
design, although
the benefit
ofinto
the account.
prestress across
Figure 4 presents aThe
flowcover
chartrequired
for the for
design
plane may
be taken
concrete section
is checked
at all positions
bondof PT slabs
the serviceability limit state (SLS),
to ensure that both
the compressive
considerations
for of
bonded
systems
is the are given
the shear
plane may and
be taken intoAt
account.
Recommendations
for the design
prestressed
concrete
a prestressed slab is generally always in
tensile stresses lie within the acceptable
diameter
of the duct
forflat
circular
for
in Eurocode 2. Design
methods
for PT
slabs ducts;
are relatively
At the serviceability
limit state (SLS),
a prestressed
slab is generally
compression
and therefore
exural cracking
limits given in Eurocode
2.
at ducts it is the larger of half the larger
straightforward, and detailed guidance, based on Eurocode 2, is
is uncommon.
This allows
the accurate
Stresses are always
checked
the concreteand therefore
dimension or the smaller dimension. For
inincompression
flexural cracking
is uncommon.
available in TR43.
prediction
of deections
the properties
section at the initial
the prediction
unbonded systems, the cover required for
This condition
allows thewhen
accurate
of deflections
as the as
properties
of the uncracked concrete section are easily
prestress is applied, and at serviceability
bond is the diameter of the sheath.
of the uncracked concrete section
are easily determined. Deflections
At the serviceability condition, the concrete section is checkedconditions
at all
determined. Deections can therefore be
when calculations are made to
canections
therefore
estimated,
limited to and
specific
values
rather
than rather
positions to ensure
that both
the compressive and tensile stresses
lie
limited
to speci
c values
determine
the de
andbe
crack
widths andestimated,
Design
process
than purely
controlling
combinations.
controlling the span-to-depth
ratio of
the slab, the
as inspan-to-depth
reinforced ratio
Figure limits
4 presents
chart for
within the acceptable
givenainow
Eurocode
2. the design for various loadpurely
of the slab, as in reinforced concrete design.
At the ULS the
pre-compression
of PT slabs. Recommendations for the
concrete
design. in the
Stresses are checked
inof
the
concrete section
at are
the given
initial condition
In carrying out the above checks, extensive
section is ignored and checks are made
design
prestressed
concrete
use
can be made
of computer
software
to ensure that the
section has
in Eurocode
2. and
Design
methods for PT
at
In carrying
outsuffi
thecient
above checks,
extensive
use can
be made of
when the prestress
is applied,
at serviceability
conditions
when
eitheraccurate
to provide
accurate
models
of the
moment capacity.
Shear stresses
are
also to provide
relatively straightforward,
computer
software
either
models
of the
structure,
calculations are slabs
madeare
to determine
the deflectionsand
and crack widths
bond
Design process re protection
for various load combinations.
18/06/2015 11:22
FIGURE 7: PRINCIPLES OF PRESTRESSED DESIGN
a) Consider a beam with a force P applied at each end along the beams centre line.
P
c) The stress distribution from the flexure of the beam is calculated from
M/Z where M is the bending moment and Z the section modulus. By
considering the deflected shape of the beam it can be seen that the
bottom surface will be in tension. The corresponding stress diagram can
be drawn.
+ M/Z
Compression
P/A
This force applies a uniform compressive stress
across the section equal to P/A, where A is the cross
sectional area. The stress distribution is shown right.
=
Tension
- M/Z
0
d) Concrete is strong in compression but not in tension. Only small tensile
stresses can be applied before cracks that limit the effectiveness of the
section will occur. By combining the stress distributions from the applied
precompression and the applied loading it can be seen there is no longer
any tension, assuming the magnitude of P has been chosen correctly.
b) Consider next a vertical load w applied along the beam and the
corresponding bending moment diagram applied to this alone.
w
Applied load
+ M/Z
P/A
Resultant Moment Diagram
P/A+ M/Z
M (max)
- M/Z
0
Figure 5 Principles of prestress design
taking into account the effect of other elements, or to enable different

load combinations to be applied, or to carry out both the structural
analysis and prestress design.
The basic principles of prestressed concrete design can be simply
understood by considering the stress distribution in a concrete section
under the action of externally applied forces or loads.
Figure 5 illustrates the simplicity of the basic theory. In essence, the
design process for serviceability entails checking the stress distribution
under the combined action of both the prestress and applied loads, at all
positions along the beam, in order to ensure that both the compressive
and tensile stress are kept within the limits stated in design standards.
PT beams and slabs are usually designed to maximise the benefit of
the continuity provided by adjacent spans. In this situation secondary
effects should be considered in the design. The secondary effects are not
necessarily adverse and an experienced designer can use them to refine
a design.
P/A - M/Z
0
FIGURE 8: LOAD BALANCING TECHNIQUE
a) Proposed
loading
b) Unstressed
slab
c) Prestressed
slab
d) Final
condition
Figure 6 Load-balancing technique
w2/m
e2
L1
Cantilever
Span 1:
e3
drape
e1
w3/m
drape
w1/m
L2
e4
L3
Span 2:
Span 3:
Figure 7 Idealised tendon profile for two spans with single cantilever
In the majority of prestressed slabs it will be necessary to add

reinforcement, either to control cracking or to supplement the capacity
of the tendons at the ultimate load condition.
The technique known as load balancing offers the designer a powerful
tool. In this, forces exerted by the prestressing tendons in catenary are
modelled as equivalent upward forces on the slab. These forces are
then proportioned to balance the applied downwards forces (Figure 6).
By balancing a chosen percentage of the applied loading, it is possible
to control deflections and also make the most efficient use of the slab
depth.
In order to use the load-balancing technique, the prestressing tendons
must be set to follow profiles that reflect the bending moment envelope
from the applied loadings. Generally parabolic profiles are used.
In PT concrete floors, the load-balancing technique can enable the
optimum depth to be achieved for any given span. The final thickness
of the slab, as with reinforced concrete flat slabs, may also be controlled
by the punching shear around the column.
For a parabolic profile the upward uniformly distributed load w is:
ws2
8
= Pa
where s is the span, a is the drape and P is the prestress force.

This upward load normally balances the self-weight and the
superimposed dead load. Depending on the design, it is also sometimes
used to balance some of the live loads.
This can be extended to several spans and provides a more economical
design as the drape is larger (Figure 7).
The anchorages are normally placed at the centroid of the section in
order to prevent a moment being placed at the end of the beam or slab.
Initial sizing of PT slabs
PT slabs can initially be sized using spanto-depth ratios. TR43 gives spanto-depth ratios for various different slab types, as does the Concrete
Centre book, Economic Concrete Frame Elements to Eurocode 25. Figure 8
gives typical span-to-depth ratios for flat slabs, band beams and ribbed
slabs for different imposed loads. Table 3 gives the range of spans that
are normally used for PT floors.
Prestress losses
From the time that a post-tensioning tendon is stressed, to its final state
many years after stressing, various losses take place which reduce the
tension in the tendon. These losses are grouped into two categories:
short-term and long-term losses.
Short-term losses
Short-term losses include:

friction losses in the tendon
wedge set or draw-in
elastic shortening of the structure
These losses take place during stressing and anchoring of the tendon.
Long-term losses
Long-term losses include:

shrinkage of the concrete
creep of the concrete, including the effect of the prestress
relaxation of the steel tendon
Although these losses occur over a period of 10 or more years, the bulk
occurs in the first two years following stressing. The loss in prestress force
following stressing can be significant (between 10% and 50% of the
initial jacking force at transfer and between 20% and 60% after all losses)
and therefore the losses should, in all instances, be calculated. TR43 gives
advice on prestress losses in Appendix B.
Table 3: Span ranges for PT floors
Floor Type
Span Range
PT flat slab
6-13m
PT band beam
8-18m
PT ribbed slab
7-18m
PT waffle slab
8-18m

43
Table 2: Specication of commonly used strand in the UK

Strand type
Nominal
Nominal
tensile
diameter
(mm)
Table 2: Specification of strength
commonly used strand
in the UK
(MPa)
Strand Type
12.9 Super
15.7 Super
12.9 Super
15.7 Euro
15.7 Super
15.2 Drawn
15.7 Euro
15.2 Drawn
Nominal tensile
1860(MPa)
strength
1770
1860
1860
1770
1820
1860
1820
Nominal diameter
12.9
(mm)
15.7
12.9
15.7
15.7
15.2
15.7
15.2
Crosssectional area
(mm2)
Cross-sectional area
100(mm2)
150
100
150
150
165
150
165
Nominal
mass (kg/m)
Characteristic
value of maximum
force (kN)
Nominal mass (kg/m)
0.781
1.17
1.17
1.29
1.17
1.29
Characteristic
value of 0.1%
proof force
(kN)
Characteristic valus
186
of maximum
force
(kN)
Maximum value of
213
maximum
force (kN)
186
319
213
240
265
302
228
265
0.781
1.17
Maximum value
of maximum
force (kN)
279
300
279
300
302
342
319
342
Chararacteristic
value of160
0.1% proof
force (kN)
228
160
258
240
258
Figure 8
Span-to-depth ratios for PT oors
Figure 8 Span-to-depth ratios PT floors
Conclusion
Conclusion
This
doesnot
notprovide
providea a
full
explanation
This article
article does
full
explanation
of the design of PT
of
the design
of PT oors.that
It isthe
recommended
floors.
It is recommended
designer makes themselves familiar
that the designer makes themselves familiar
with TR43 when starting the design process. The design of PT floors
with TR43 when starting the design process.
allows the designer to play with the different aspects prestress force,
The design of PT oors allows the designer
tendon profile etc. to arrive at the most economic design. There are
to play with the different aspects prestress
manysoftware
packages
help,atboth
force,
tendon pro
le etc.that
tocan
arrive
the 2D design and finiteelement
analysis.design. There are many
most
economic
software packages that can help, both 2D
The Post-Tensioning Association (PTA) in the UK can also help.
design and nite-element analysis.
It produces technical design guidance and is in the process of producing
The Post-Tensioning Association (PTA) in
a model performance specification for PT floors to help main designers
the UK can also help. It produces technical
understand
what and
is required
byprocess
specialistofdesigners and what can be
design
guidance
is in the
expected from
the specialist
designers
and
contractors. The PTA website
producing
a model
performance
speci
cation
canPT
beaccessed
at www.posttensioning.co.uk
for
oors to help
main designers
understand what is required by specialist
designers and what can be expected from
the specialist designers and contractors.
The PTA website can be accessed at
www.posttensioning.co.uk
References:
1) The Concrete Society

(2005)
Technical Report
No. 43: Posttensioned
References
and
further
reading
concrete floors: Design handbook (2nd ed.), Camberley, UK:

1) The
TheConcrete
ConcreteCentre
Society (2005) Technical Report No. 43: Post-
tensioned concrete oors: Design handbook (2nd ed.), Camberley, UK:

The Concrete Centre
2: Design of concrete structures. General rules and rules for buildings,
UK: BSI Institution (2014) BS EN 1992-1-1:2004 Eurocode
2)London,
British Standards
2:3)
Design
concreteInstitution
structures.(2009)
General
and1992-1-1:2004
rules for buildings,
BritishofStandards
NA rules
to BS EN
UK
London,
UK:
BSI
rules and rules for buildings, London, UK: BSI

National
Annex to(2010)
Eurocode
2. Design
of concrete
4) CONSTRUCT
National
Structural
Concrete structures.
SpecificationGeneral
(4th
rules
rules forUK:
buildings,
London,
UK: BSI
ed.),and
Camberley,
The Concrete
Centre
Goodchild C. H.,
Webster
R. M.Structural
and ElliottConcrete
K. S. (2009)
Economic
4)5)CONSTRUCT
(2010)
National
Speci
cation (4th
Concrete
Frame UK:
Elements
to Eurocode
2, Camberley, UK: The Concrete
ed.),
Camberley,
The Concrete
Centre
Centre
5) Goodchild C. H., Webster R. M. and Elliott K. S. (2009) Economic
Concrete Frame Elements to Eurocode 2, Camberley, UK: The Concrete
Centre
18/06/2015 11:22
Guidance on the design of

liquid-retaining structures
Number 1
44
January 2015
Technical
Concrete design
Concrete Design Guide

No. 1: Guidance on the design of
liquid-retaining structures This article highlights some of the key
considerations for the design and construction
This series is produced by The Concrete Centre to enable designers to realise the potential
of liquid-retaining structures.
of concrete.
The Concrete Centre, part of the Mineral Products Association (MPA), is a team of qualied
professionals with expertise in concrete construction, engineering and architecture.
Introduction
This guidance is based on Eurocodes BS EN 1992-1-11 and

BS EN 1992-32 and the corresponding UK National Annexes. This is
not an exhaustive treatment of the subject and the reader is advised
Introduction
3
to refer
to The
Centre
onsalient
concrete
basements
and
This
short
noteConcrete
highlights
somebook
of the
aspects
of the design
4
CIRIAconstruction
publication of
C660
as basements
and tanks
many aspects
and
liquid-retaining
structures
in have
reinforced
concretein
common.
(Figure
1). This guidance is based on Eurocodes BS EN 1992-1-11 and
BS EN 1992-32 and the corresponding UK National Annexes. This is
Watertightness
not an exhaustive treatment of the subject and the reader is advised
3
to
to TheofConcrete
Centre
book onliquid
concrete
basements
and
In refer
the design
containment
structures,
tightness
is a critical
4
CIRIA
publication
as basements
and tanks
have
many
aspects
consideration.
TheC660
structural
engineer should
discuss
and
agree
with in
common.
the client the liquid tightness requirements. Colloquial phrases such

as waterproof construction are best avoided; instead the engineer
Watertightness
should agree the degree of leakage that can be tolerated using the
In the design of containment structures, liquid tightness is a critical
classification ofThe
tightness
classes
shownshould
in Table
1. This,and
in turn,
willwith
lead
consideration.
structural
engineer
discuss
agree
to
limiting
crack
widths
that
should
be
used
in
structural
design.
the client the liquid tightness requirements. Colloquial phrases such
as waterproof construction are best avoided; instead the engineer

Tightness class 1 is the most usual class for utilitarian water-retaining
should agree the degree of leakage that can be tolerated using the
structures and limiting crack widths is normally sufficient to achieve this
classication of tightness classes shown in Table 1. This, in turn, will lead
class.
For tightness
classes
and 3 itbe
willused
not be
sufficient design.
to limit crack
to
limiting
crack widths
that2 should
in structural
widths
alone.class
Liners1 is
and/or
prestressing
willfor
beutilitarian
required to
meet the
Tightness
the most
usual class
water-retaining
requirements.
structures
and limiting crack widths is normally sufficient to achieve this
class. For tightness classes 2 and 3 it will not be sufficient to limit crack
In addition to correct design, liquid tightness also depends on the use
widths alone. Liners and/or prestressing will be required to meet the
of an appropriate concrete mix and good workmanship on site. Good
requirements.
compaction
is essential.
In additionof
toconcrete
correct design,
liquid tightness also depends on the use
of an appropriate concrete mix and good workmanship on site. Good
Durability and selection of materials
compaction of concrete is essential.
Concrete mix
In liquid-retaining structures, mix design should aim at durability and
minimising the risk of cracking. Strength is rarely critical. Watertightness
and durability can be achieved using good-quality concrete alone
Concrete mix
without
any special
additivesmix
or admixtures.
In
liquid-retaining
structures,
design should aim at durability and
minimising
thespecification
risk of cracking.
Strength
is rarely critical.
Watertightness
The following
is likely
to be satisfactory
for most
cases:
and durability can be achieved using good-quality concrete alone
Consistence
class:
S3
without
any special
additives
or admixtures.
followingwater
speci
cationratio:
is likely
to be satisfactory for most cases:
The
Maximum
cement
0.50
Minimum cement content: 300kg/m3 when aggregate size is 20mm;
Consistence
320kg/m3class:
whenS3
aggregate size is 14mm
Maximum water cement ratio: 0.50
3
Maximum cement content: 400kg/m
for CEM I (OPC) concrete and
Minimum cement
content: 300kg/m3 when aggregate size is 20mm;
3
450kg/m
when
ground
granulated
blast-furnace
slag (GGBS) or fly
320kg/m3 when aggregate size is 14mm
Maximum
ash is used
3
cement content: 400kg/m for CEM I (OPC) concrete and
Concrete
strength
C30/37blast-furnace slag (GGBS) or y ash
3
450kg/m
when
groundclass:
granulated
is used
Use of cement replacement (GGBS or fly ash) is recommended as
Concrete strength class: C30/37
the heat of hydration will be less than that for pure Portland cement
(CEM I) and this in turn assists in crack control. Suitable cement or
Use of cement replacement (GGBS or y ash) is recommended as
combination
types arewill
CEM
(which
contains
2135%
flycement
ash) or
the
heat of hydration
be IIB-V
less than
that
for pure
Portland
CEM IIIA
(which
3665%
GGBS).control.
If high proportions
of cement
(CEM
I) and
thiscontains
in turn assists
in crack
Suitable cement
or
replacementstypes
are used,
thereIIB-V
will be
implications
early strength
and
combination
are CEM
(which
containsfor
2135%
y ash) or
CEM
IIIA resistance,
(which contains
GGBS).
If high
of cement
abrasion
which3665%
might have
an effect
onproportions
the programme.
replacements are used, there will be implications for early strength and
Figure 1
Aeration and clarier tanks at New Hythe, Kent
Concrete should be specified in accordance with BS EN 2065,6 and
Concrete
should
bewith
speci
ed inregulations.
accordanceThe
with
BS
EN 2065,6 and
will need to
comply
specific
Civil
Engineering
7,8
BS 8500 Parts 1 and 2 . All materials
in contact with potable water
Specification for the Water Industry9 provides useful information.
will need to comply with specic regulations. The Civil Engineering
Speci
cation forconcrete
the Water
Industry9for
provides
useful
information.
Well-compacted
is essential
durability.
Generally,
the
Well-compacted
concrete
is essential
for durability.
Generally,
the
thickness of members should be
at least 250mm
to permit
good
thickness
of However,
membersthe
should
be atshould
least 250mm
to permitasgood
compaction.
thickness
not be excessive
the
compaction. However, the thickness should not be excessive as the
reinforcement required increases with the thickness of the concrete.
reinforcement required increases with the thickness of the concrete.
The
likely
exposure
classes
for different
elements
are in
noted in
The
likely
exposure
classes
for different
elements
are noted
Table
2. Cover requirements in BS EN 1992-1-1 and BS 8500 will generally
Table 2.
Cover
in to
BSuse
ENnominal
1992-1-1cover
and BS
will generally
of 45mm
from the
apply.
It isrequirements
good practice
cnom8500
apply. It is good practice to use nominal cover Cnom of 45mm from the
face in contact with liquid and 75mm from any face cast against soil.
face in contact with liquid and 75mm from any face cast against soil.
Figure 1 Aeration and clarifier tanks at New Hythe, Kent
HOWES ATKINSON CROWDER
Durability
selection
of materials
BS 8500 Partsand
1 and
27,8. All materials
in contact with potable water
18/12/2014 12:46
Values of actions
Values of actions should be established using the relevant codes. In
practice the designer is likely to know only the maximum depth of
liquid that it is physically possible to store. The operational depth will
be slightly smaller. It is recommended that all calculations are carried out
using the full depth of the tank. Although slightly conservative, this
approach will result in a reliable design.
Partial factors on actions
Partial factors for permanent and variable actions are given in the
UK NA to BS EN 1990. Tables 3 and 4 show the partial factors
recommended for loads and pressures induced by retained liquids,
noting that designs should be based on the assumption that the
maximum liquid level would be the top of the walls.
Figure 2 Tensile strength and stress over time
Key
Line 1a
Tensile strength of the concrete: transitory loading
Line 1b
Tensile strength of the concrete: sustained loading
Line 2a
Early-age contraction stress, allowing for creep
Line 2b
Addition of drying shrinkage stress to line 2a
When in contact with aggressive soil, provisions to resist sulphate

attack are likely to control the mix and a cement type CEM IIIB or CEM
IVB could be used.
Structural analysis
Design should be based on elastic analysis without redistribution.
In rectangular tanks, direct tension in the plane of the walls arises from
the lateral load supported by adjacent contiguous walls and this should
be taken into account in design. Hoop stresses in circular tanks also lead
to in-plane tension.
Structural design ultimate limit state
Guidance in BS EN 1992-1-1 should be used. The following should be
noted in connection with the calculation of shear resistance:
the shear resistance of a section not reinforced for shear VRd,c should

be calculated making allowance for the presence of any tension
Table 1: Liquid tightness classes*
Tightness class
Where chemically aggressive liquids are stored, expert guidance

should be sought for the selection of appropriate concrete mix9.
Normally, protective liners will be required.
Basis of structural design
Design situations
These are dealt with in a general way in BS EN 199010. For tanks
constructed partially or fully below ground:
adverse effects of soil and groundwater pressures on the walls and

base should be considered during construction and in service; this

will normally require consideration of the tank when it is empty
for the design situation when the tank is full, no relief should be

given for the beneficial soil and groundwater pressure effects
Actions on liquid containment structures
Permanent actions
Common permanent actions to be considered are:
self-weight of tank and contents
weight of plant and equipment
Requirements for leakage
Some degrees of leakage acceptable or leakage of liquids irrelevant
Leakage to be limited to small amount. Some surface staining or damp

patches acceptable
Leakage to be minimal. Appearance not to be impaired by staining
No leakage permitted
*Based on BS EN 1992-3
Table 2: Likely exposure classes for different elements*

Element
Likely exposure class
Wall in contact with liquid
Cyclic wet and dry XC3 and XC4
Underside of roofs to
reservoirs
Moderately humid environment XC3 and XC4
Surfaces in contact with

soil (walls and slabs)
DC class depending on the aggressiveness of the soil
Unprotected surfaces
of walls and roofs
Will depend on the circumstances XC, XD, XS and XF may all be relevant
* Exposure classes are in accordance with BS EN 1992-1-1 and BS 8500
Variable actions
Common variable actions to be considered are:
loads due to liquid pressures
wind loads on structures at or above ground
snow loads on covered structures
uplift forces on underground tanks due to ground water
lateral loads due to earth and water pressures on underground tanks
Table 3: Partial factors at ULS for tanks below ground

Design
situation
Verification
Permanent
actions
Loads
from
contained
liquid
Earth and
groundwater
pressures
Equilibrium
0.90
1.50
Strength
1.35
Strength
1.35
1.35
Tank empty
Tank full
*The more critical values obtained using combinations given in BS EN 1997-111 should be used here
caused by loading (as opposed to that caused by restraint of intrinsic

deformation such as shrinkage and temperature effects). In such cases
cp in expressions (6.2.a) and (6.2.b) will be negative and therefore
reduce the shear resistance
a value of cot = 1.0 should be used for the calculation of shear

reinforcement as recommended by the Eurocode
Structural design serviceability limit state
Crack widths and watertightness
Table 5 gives the crack width limits and recommendations for the
watertightness classes chosen from Table 1. The approach to crack
control and the performance implications of the chosen method should
be agreed with the client.
Estimation of crack widths
Crack widths are normally calculated for:
cracking caused by restraint to movement (also referred to as

imposed deformations)
cracking caused by loading
Examples of imposed deformations include early thermal effects,
autogenous shrinkage and drying shrinkage.
CIRIA publication C660 contains extremely useful information for the
estimation of crack widths. The formulae for crack width calculation are
not included here and the reader should consult the reference for fuller
details.
Cracking caused by restraint
While cracking is accepted in concrete structures, it is expected to
be controlled such that cracks will occur at intervals and their width
will be small. This requires the presence of a minimum amount of
reinforcement in the structural element. Just prior to the occurrence
of the first crack, the concrete and the reinforcement will be in tension.
At the crack the tension can be carried only by the reinforcement. The
minimum reinforcement is calculated such that the reinforcement is
able to transfer the tension without yielding. It follows therefore that
the stronger the concrete, the greater the amount of reinforcement
that will be required to achieve controlled cracking. The tensile strength
of concrete at the time when first cracking might be expected is the
appropriate strength that should be used. Often, early age cracking
at three days is considered critical. Figure 2 shows the relationship
between the increase in stress and strength of the concrete over time.
Minimum reinforcement does not guarantee any specific crack

width. It is a necessary amount presumed by models for crack width
calculations; but not necessarily a sufficient amount to limit the crack
widths. Additional reinforcement may well be necessary.
For cracking to occur there has to be restraint to movement. The
degree of restraint is another source of uncertainty. Restraint may
be internal or external to the element. Internal restraint is caused by
differential expansion and will generally be significant in thick sections.
External restraint is either end restraint or edge restraint. The
mechanism of crack formation in these two cases is different and this is
reflected in the formulae used for the calculation of restrained strain. In
the case of members restrained at ends, cracking occurs progressively.
Each crack occurs to its full potential width before successive cracks
occur. In this case crack-inducing strain is specifically related to the
strength of the concrete and the steel ratio.
In the case of members restrained at one edge, the reinforcement
and adjacent concrete act as crack distributors and the crack width
is a function of the restrained strain rather than the tensile capacity of
concrete. See Figure 3 for a typical edge restraint crack pattern.
The amount of reinforcement required to limit the crack width to a
given value will be considerably higher in the case of members with end
restraint compared to those with edge restraint. This is the reason why
specifications prohibit casting in alternative bays.
Cracking caused by loading
Procedures set out in BS EN 1992-1-1 may be used to control cracking
caused by loading.
Minimising the risk of cracking
BS EN 1992-3 suggests a number of strategies to minimise the risk of
cracking. CIRIA publication C660 also provides tips for control of early
thermal cracking. The reader should refer to these.
Deflection control
In general, deflections are unlikely to be critical. The procedures in BS
EN 1992-1-1, including the span-to-depth formulae, may be used. Where
finishes are applied to the structure, manufacturers should be consulted
on any limitations on the strains.
Construction
Joints
There are essentially two types of joints: those required for minimising
Table 4: Partial factors at ULS for elevated towers
Design
situation
Verification
Permanent
actions
Loads
from
contained
liquid
Wind
Snow
Tank empty
Equilibrium
0.90
1.50
Strength walls
and base
1.35
1.35
1.50
Strength
supporting
structures and
foundations
1.35
1.35
1.50
Tank full
*For the value of 0 see the UK NA to BS EN 1990
0 1.50*
It is good practice to protect all joints by incorporating water bars. If

the joints are sealed (recommended when another line of defence is
deemed necessary), the selection of sealants should be undertaken by
a specialist and should take into account:

the chemical compatibility with other materials or soil with which

they are likely to be in contact
expected movement in the joint
ease of repair and replacement
Table 5: Requirements for water-tightness classes*

Tightness class
0
Figure 3 Edge restraint of wall cast on base
Suggested measures to meet the requirements

Structure may be designed using the provisions of clause 7.3.1
of BS EN 1992-1-1
Width of any cracks that can be expected to pass through the full thickness of
the section should be limited to w1*
Where the full thickness of the section is not cracked, provisions of clause 7.3.1
of BS EN 1992-1-1 may be used subject to conditions
Cracks that may be expected to pass through the section should be avoided,
unless special measures are incorporated (e.g. water bars or liners). There is an
implication in the code that it may be adequate to provide water bars to
safeguard against leakage through cracks and verify the above-mentioned
conditions for cracks that do not penetrate the whole depth of the section
Special measures will be required (e.g. liners or prestress)
*The crack width w1 is related to hd/h values where hd is the hydraulic head and h is the overall thickness of the wall as
shown in Table 6
Where the whole section is not cracked, it must be checked that the depth of the compression zone should be at least
xmin. In the UK the value of xmin is given as the lesser of 50mm or 0.2h
The crack width limit w1 noted above* is only acceptable if the range of strain at a section under service conditions is
150 x 10-6. If this is not satisfied self-healing is unlikely to occur at the crack and this may lead to leakage
The size of individual pours will be governed by the site constraints,

such as ease of access for concreting and the geometry of the element.
The National Structural Concrete Specification12 recommends default
pour sizes, but these can be modified if agreed between the designer
and the contractor.
At construction joints, structural continuity is required and no
relative movement between the sections should occur at the joint. The
reinforcement should pass through. The surface of the first pour should
be roughened to increase the bond strength and to provide aggregate
interlock. Powerful hammers should not be used as they may dislodge
the aggregate particles.
Figure 4 Typical detail of preformed PVC strip backstop water bar
the risk of cracking (movement joints) and those required for

convenience of construction (construction joints). Expansion joints
should be provided when reversible movements are expected and
contraction joints are suitable when only contraction has to be
accommodated.
Joints are potentially vulnerable locations for water penetration.
In water-retaining structures, the number of joints should be kept
to a minimum. Joints also require maintenance for continued good
performance.
Water bars
Water bars are preformed strips of durable impermeable material that
are wholly or partially embedded in the concrete during construction.
They are located across joints to provide a permanent liquid-tight seal
during the whole range of movements. Water bars may be metal strips
or proprietary products made of rubber or flexible plastics such as PVC
(Figure 4).
At construction joints (both horizontal and vertical), a rigid water bar
formed from a strip of black steel (unpainted and non-galvanised) has
proved effective. The water bar is placed centrally across the joint. In
horizontal joints, the water bar is gently pushed in when the concrete is
still green. Separate lengths of the metal water bars need not be welded
together. At butt joints between two water bars, a gap should be left
equal to aggregate size + 5mm. A cover strip overlapping the two water
bars should be placed, again leaving a gap of aggregate size + 5mm.
may also be encountered and may be cut, or the relocation of the

fixing will increase the area of damaged concrete. This is not a
satisfactory procedure in water-retaining structures.
Inspection and testing
BS EN 1992-3 does not provide any guidance on inspection and testing
nor does it state any acceptance criterion for test results. In the UK, the
procedure given in BS 800713 has been used successfully in the past
and it can continue to be used.
References:
1) British Standards Institution (2004) BS EN 1992-1-1:2004
Eurocode 2. Design of concrete structures General rules and rules
for buildings, London, UK: BSI
2. Design of concrete structures. Liquid retaining and containing
structures, London, UK: BSI
3) Narayanan R. S. and Goodchild C. H. (2012) Concrete Basements:
Guidance on the design and construction of in situ concrete
basement structures, London, UK: MPA The Concrete Centre
Figure 5 Types of formwork ties for use in watertight construction
Table 6: Limiting values of w1

hd/h
w1 (mm)
10
15
20
25
30
35
0.200
0.175
0.150
0.125
0.100
0.075
0.050
Similar detail should be used at corners. Because of the stiffness of

the metal water bars, there is no risk of the water bar collapsing during
concreting.
PVC rear guard water bars are especially suitable for protecting
joints in base slabs cast on ground.
Miscellaneous
Kickers
In liquid-tight construction, it is recommended that a kicker with a
height of 150250mm is constructed monolithically with the base
slab. While kickerless construction may be preferred by some
contractors in the interest of speed of construction, the associated
additional risk of water penetration is not warranted in liquid-tight
construction.
Formwork ties
Formwork ties should be selected carefully so that there is no risk
of moisture penetration. There are two types available in the market:
lost tie system and hollow tube system (Figure 5).
Service penetrations through the structure
In water-retaining structures, it is common for service pipes to
penetrate the concrete. They are also common locations where
leakage occurs. The locations should be pre-planned and the pipes
should incorporate puddle flanges (effectively incorporating a
prewelded water bar on the outer face). Casting a sleeve or boxing and
locating the pipes subsequently will result in more interfaces and will
increase the risk of leakage; this should ideally be avoided.
Post-drilling concrete and any internal membranes for fixings
could also damage the concrete in the local area. Reinforcement
4) CIRIA (2007) Publication C660: Early-age thermal crack control

in concrete [Online] Available at: www.ciria.org/CMDownload.
aspx?ContentKey=dc8dcc64-7feb-4342-a06a-63f68aa598b4&Con
tentItemKey=15398ecc-7629-4f3a-a528-5e11eee0ca7c (Accessed:
December 2014)
5) British Cement Association (2001) Specifying concrete to BS EN
206 1/BS 8500: Concrete resistant to chemical attack, Camberley,
UK: BCA
6) British Standards Institution (2013) BS EN 206:2013 Concrete.
Specification, performance, production and conformity, London, UK:
BSI
7) British Standards Institution (2012) BS 8500-1:2015
Concrete. Complementary British Standard to BS EN 206-1. Method
of specifying and guidance for the specifier, London, UK: BSI
8) British Standards Institution (2015) BS 8500-2:2015
Concrete. Complementary British Standard to BS EN 206-1.
Specification for constituent materials and concrete, London, UK: BSI
9) UK Water Industry Research Ltd (2004) Civil Engineering
Specifi cation for the Water Industry (6th ed.), Swindon, UK: WRC plc
Eurocode. Basis of structural design, London, UK: BSI
11) British Standards Institution (2013) BS EN 1997-1:2004+A1:2013
Eurocode 7. Geotechnical design. General rules, London, UK: BSI
12) CONSTRUCT (2010) National Structural Concrete Specification
for Building Construction (4th ed.) [Online] www.construct.org.uk/
media/National_Structural_Concrete_Specification_for_Building_
Construction.pdf (Accessed: August 2015)
13) British Standards Institution (1987) BS 8007:1987 Code of
practice for design of concrete structures for retaining aqueous
liquids, London, UK: BSI
An introduction to strut-and-tie modelling
Robert Vollum, of Imperial College, London, discusses

strut-and-tie modelling, a simple method of modelling
complex stress patterns in reinforced concrete as
triangulated models.
Introduction
It is based on the same truss analogy as the design for shear in Eurocode
21 and can be applied to many elements, but is particularly useful where
normal beam theory does not apply, i.e. where plane sections do not
remain plane, e.g. in deep beams, corbels and pile caps. EC2 provides
information about the use of strut-and-tie modelling and this article is
an introduction for engineers who want to take advantage of this useful
analysis method.
Regions where normal beam theory cannot be used arise at geometrical
discontinuities, supports and concentrated loads and are known as D or
disturbed regions. D regions are dimensioned using Bernoullis principle
(Figure 1). Regions where normal beam theory can be used are called B
or beam or Bernoulli regions. In strut-and-tie modelling, the structure
can be split into D and B regions and a strut-and-tie model (STM)
prepared for the D regions. The method uses the truss analogy where
the actual flow of stresses is represented by a series of struts and ties
(Figure 2). The method is an application of the lower bound theory of
plasticity. Consequently, there is no unique STM for any given problem.
The main requirements are to ensure that equilibrium is satisfied and
to proportion the cross-sectional areas of struts and ties such that their
resistances are everywhere greater than or equal to the internal forces.
Adequate ductility is typically ensured through the provision of codespecified minimum reinforcement areas.
The scope of this article is restricted to strut-and-tie modelling of planar
structures. The reader is referred to specialised texts27 for more in-depth
treatment of the method.
Development of STM
The first step is to determine the basic STM geometry within the
element. The serviceability limit state (SLS) of cracking is usually satisfied
if the ultimate limit state (ULS) STM geometry is based on the elastic
stress field, with members positioned at the centroids of the
compressive and tensile elastic stress distributions (Fig. 2a).
An elastically based STM can lead to overly conservative designs, since
such models do not recognise the redistribution in stress that occurs
when the concrete cracks, putting more stress into the reinforcement.
Therefore, some deviation from the elastic solution is permissible, as in
the design of continuous beams. Consideration of Fig. 2a and Fig. 2b
Figure 1 Dimensioning of D regions
shows that the STM geometry changes as the load is increased to failure,
due to the internal redistribution of forces that occurs on cracking and
subsequent yielding of reinforcement, which in the ultimate case lies
within the plastic limit. The STM in Fig. 2b neglects the contribution of
the minimal web reinforcement required by EC21 to control cracking,
which further increases load resistance.
Load path method
The load path method2 is a useful tool for developing the STM geometry
within elements. The starting point is to determine the stress distribution
at the boundaries of the D region after which the STM can be developed
using the following rules of Schlaich and Schfer2:
1. Load paths do not cross each other
2. Load paths take the shortest streamlined route between the centres
of gravity of stress diagrams
3. Curvatures concentrate near stress concentrations, reactions and
concentrated loads
4. In the absence of elastic analysis, position struts from experience
and the application of simple rules
37
Figure 2
Development of STM for uniformly loaded deep beam
37
a) based on elastic stress eld b) at ULS
Figure
shows howof
rule
can bemay
usedbetoallowed
distinguish
between
4. The5 orientation
the2 model
to depart
fromgood
the
37
elastic
eld
at thebeam.
ULS ifThe
the STM
SLS in
is checked
and
badstress
STMs for
a deep
Fig. 5a is bad because
tensile
is provided
Figure
2 insufficient
4. The
orientation
of the model
may reinforcement
be allowed to depart
from the to control cracking at the
Development of STM
for uniformly
loadedhow
deep beam
Figure
5atshows
shows
rule 2
be used
to distinguish
between
elastic
eld
atthe
ULS
if the that
SLS isthere
checked
a) based stress
on elastic
stress
eld5b
b)
ULS
SLS.
Fig.
is can
an area
of concrete
in the top
section
good and bad STMs for a deep beam. The STM in Fig. 5a is bad
of
the
STM
which
is
in
tension
and
therefore
this
section
should
be
5 shows
how
rule
2 can
be be
used
to distinguish
4. Figure
The orientation
of the
model
may
allowed
to
departbetween
from the is provided to control
because
insuffi
cient
tensile
reinforcement
reinforced
for
the
tensile
stress.
It
can
be
shown
from
finite-element
good and
bad
STMs
forULS
a
deep
beam.
The
STM
in
Fig.
5a
is
bad
elastic
stress
cracking
eld
at the
if
the
SLS
is
checked
at the SLS. Fig. 5b shows that there is an area of concrete
because insuffi
cient tensile
reinforcement
is provided
to control
models
thatsection
this tensile
zone
is which
created.
Normally
stress
in the top
of the
STM
is in
tension the
andtensile
therefore
thisis
cracking
the SLS.
Fig.
5b2shows
there
is an area of
concrete
Figure at
5 shows
how
rule
can bethat
used
to distinguish
between
section
should
be
reinforced
for
the
tensile
can beam.
be shown
taken
the
nominal
provided
in the Itdeep
in the top
of by
the
STM
which
is inreinforcement
tension
therefore
this stress.
good
andsection
bad
STMs
for
a deep
beam.
The
STMand
in Fig.
5a
is bad
Figure 2
Development of STM for uniformly loaded deep beam
a) based on elastic stress eld b) at ULS
nite-element
models
that
this
section should
be reinforced
for the tensile
It can
betensile
shown zone is created. Normally
because
insuffifrom
cient
tensile
reinforcement
isstress.
provided
to control
Compressive
stress
from nite-element
models
thisthat
tensile
zone
isthe
created.
Normally
cracking
at the
SLS.tensile
Fig. 5bthat
shows
there
isby
an
area
of concrete
the
stress
is fields
taken
nominal
reinforcement provided in
thethe
tensile
stress
taken
by
the
nominal
reinforcement
provided
in the compressive stress in
in
top section
ofdeep
the STM
which
is
in tension
and way
therefore
this
theis
beam.
Compressive
stress
fields
are the
in which
the deepshould
beam.be reinforced for the tensile stress. It can be shown
section
the
idealised
struts
spreads
through
the
concrete.
They are categorised
from nite-element models that this tensile zone is created. Normally
stress
elds (Figure
Figure 2 Development of STM for uniformly loaded deep beam
Compressive
stress
elds
the
tensile stress
is takenby
the nominal
reinforcement
provided6).
in Prismatic stress fields
asCompressive
prismatic,
fanor
bottle-shaped
a) based on elastic stress field b) at ULS
Compressive
stress
elds are
the way in which the compressive
Compressive
stress
elds are the
way inwhich
the compressive
the
deep beam.
typicallystruts
arise in B regions,
whereas
fan- andare
bottleshaped stress fields
stress in the idealised
through
the spreads
concrete. They
stress in thespreads
idealised
struts
through the concrete. They are
categorised as
prismatic,
fanor bottle-shaped
(Figure
6). Prismatic
arise
in D
regions
due to thefandispersion
of the stress
paths6).radiating
out
Compressive
stress
elds
categorised
as
prismatic,
or bottle-shaped
(Figure
Prismatic
stress
elds typically
in Bthe
regions,
fanand bottleCompressive
stressconcentrated
arise
elds are
way
inwhereas
which
the
compressive
from
loads
or reactions.
stress elds typically
arise
in B regions, whereas fan- and bottleshaped
stress
elds
arise
in
D
regions
due
to
the
dispersion
of
the
stress in the idealised struts spreads through
concrete. They are
shaped
bottle-shaped
elds ariseloads
in(Figure
D or
regions
due to the dispersion of the
stress paths as
radiating
outstress
fromor
concentrated
reactions.
categorised
prismatic,
fan6). Prismatic
Bottle
stress
fields
are awhereas
feature
of
elastic
stress loads
field analysis.
The
stress
paths
out
from
concentrated
or reactions.
Bottle
stress
elds arise
are
a in
feature
of elastic
stress
eld
stress
elds
typically
Bradiating
regions,
fanandanalysis.
bottlecurvature
of
the
principal
compressive
stress
trajectories
of
the
bottle
The curvature
of
the
principal
compressive
stress
trajectories
of
the
shaped
stress
elds
arise
in
D
regions
due
to
the
dispersion
of
the
Bottle stress elds are a feature of elastic stress eld analysis.
bottle paths
eld develops
signi
cant concentrated
transverse
tension.
The
compressive
stress
radiating
out from
loads compressive
or reactions.
Thedevelops
curvature
of the principal
trajectoriesresistance
of the
field
significant
transverse
tension. stress
The compressive
resistance
of bottle
elds is limited
by stress
splitting
along
the axis
Bottle stress
eldsstress
are a feature
of elastic
eld
analysis.
Figure 3
bottle
stress
eld develops
cant
tension.
compressive
ofof
bottle
fields issigni
limited
bytransverse
splitting
the The
axis of
the strut,
of the
strut, unless
transverse
reinforcement
is provided
to maintain
Application of load path method (adapted from EC21)
The
curvature
the
principal
compressive
stress
trajectories
of along
the
resistance
of bottle
stress
elds
limited by splitting along the axis
equilibrium
after
cracking.
bottle
eld develops
signi
cant
transverse
tension.
The
compressive
unless
transverse
reinforcement
isisprovided
to maintain equilibrium
Figure 3
of the
strut,
unless
transverse
reinforcement
Application of load path method (adapted from
EC21)
The fan-shaped
stress
eld
is an
idealisation
in which
the the axis is provided to maintain
resistance
of bottle
stress
elds
is limited
by splitting
along
1
Figure 3 Application
Figure 3 of load path method (adapted from EC2 )
after
cracking.
compressive
stress
trajectories
arecracking.
assumed is
toprovided
be straight.
of
the strut, unless
transverse
reinforcement
to maintain
Application of load path method (adapted from EC21)
equilibrium
after
Consequently,
nocracking.
tension
develops at stress
right angles
toisthe
The procedure is illustrated in Figure 3 for the anchorage zone
equilibrium
after
The
fan-shaped
iseld
anstress
idealisation in which the
The
fan-shaped
stress
field
an
idealisation
The procedure
is
illustrated
in
Figure
3
for
the
anchorage
zone
of
a
posttrajectories
of the fan.
Fan-shaped
elds in
are
assumed
of a post-tensioned beam. A similar loading arrangement arises
The fan-shaped
stress
eld is an stress
idealisation
which
the to in which the compressive
compressive
stress loads
trajectories
are
assumed
to be straight.
develop
at supports
and concentrated
intoplastic
stress
eld Consequently, no tension
whenbeam.
a wall isAloaded
with
a concentrated
load at its
corner.
Note a wall compressive
stress trajectories
areare
assumed
beto
straight.
stress
trajectories
assumed
be
straight.
tensioned
similar
loading
arrangement
arises
when
is
Consequently,
noattension
develops
at right angles to the stress
The
procedure
isF2illustrated
in Figure
for geometry
the anchorage
zone
analysis. Failureno
istension
assumed
to occur
at
supports
and
concentrated
that
theprocedure
forces
F1 isand
are equal
and
opposite.
The
Consequently,
develops
right
angles
to the
stress
The
illustrated
in Figure
3
for the3anchorage
zoneof
develops
atgreatest.
right
angles
to
the stressstress
trajectories
of the fan. Fan-shaped
loadedof
with
a is
concentrated
load
its
corner.
thatde
the
forces
Ftrajectories
andwhereofstresses
trajectories
of the
fan. Fan-shaped
of
post-tensioned
beam.
Adimensions
similar
arises
1
loads
the
fully dened
once
theat
z1 Note
and zarrangement
ned.
the
fan.are
Fan-shaped
stress
elds are assumed
to elds are assumed to
aaSTM
post-tensioned
beam.
A similar
loading loading
arrangement
2 arearises
Prismatic
or
parallel-sided
stress
elds
are
adevelop
specialstress
case
of
the in and
The
dimension
z2 loaded
can with
beThe
found
by
assuming
that
angle
ofits
the
stress
fields
are
assumed
to
at
supports
concentrated
F2 are equal
opposite.
geometry
of
the
STM
is
fully
defined
once
develop
at supports
and concentrated
loads
in
plastic
loads
eld
when
aand
is loaded
a concentrated
load
atthe
itsload
corner.
Note
develop
at
supports
and
concentrated
plastic
stress eld loads
when
awall
wall
is
with
a
concentrated
at
corner.
Note
o
fanand bottle-shaped
stress
elds.
7 shows
aconcentrated
deep at
beam
diagonal
strut
Fis1 and
45
Alternatively,
the opposite.
strut
canopposite.
be
analysis.
is assumed
tooccur
at
supports
andoccur
that
F
and
Thepositioned
geometry
ofgeometry
2 are
Failure
is Figure
assumed
to
supports
concentrated
thatthe
theforces
forces
Fz12.and
Fequal
and
Thebe
of Failure
inanalysis.
plastic
stress
field
analysis.
Failure
is assumed
to and
occur
at supports and
the dimensions
z1 and
are
defined.
The
dimension
z2 can
found
2 are equal
loaded
with
a
concentrated
load.
In
reality,
the
compressive
stress
at
the
centroid
of
the
corresponding
elastic
compressive
stress
loads where stresses are greatest.
the STM is fully dened once the dimensions z1 and z2 are dened.
loads as
where
stresses
are
greatest.
the STM
is
fully
de
ned
once
theofdimensions
z
and
z
are de
ned.
1
2
concentrated
loads
where
stresses
are
greatest.
by assuming
that
the
angle
of
the
diagonal
strut
is
45.
Alternatively,
eld
is
bottle-shaped
indicated
with
the
dotted
line.
However,
in
distribution
as
done
for
the
deep
beam
Fig.
1a.
Prismatic or parallel-sided stress elds are a special case of the
The dimension z2 can be found by assuming that the angle of the
Prismatic
or
parallel-sided
stress
elds
are a special case of the
The
dimension
z2 ocan
be found
bystrut
assuming
thatBoth
the angle offantheand bottle-shaped
practice,
struts are
often
idealised
asFigure
prismatic
(Fig.
7).
EC2
requires
Figure
4 shows
STMs
for
a can
deepbebeam.
stress
elds.
7 shows
a deep
beam
diagonal
strut
is two
45
. alternative
Alternatively,
the
positioned
the strut
can
be
positioned
at
o the centroid of the corresponding elastic
fan-ofand
bottle-shaped
stress
fields
elds. Figure
7 shows
a deep
diagonal
strut
the
iscorresponding
45 .but
Alternatively,
the
be positioned
a minimum
horizontal
vertical
web
reinforcement
are
lower
bound
the STM
in Fig.
4bstrut
is notcan
optimal,
or
parallel-sided
stress
are
atospecial
case
of thebeam
fan- and
loaded
withamount
aPrismatic
concentrated
load.and
In reality,
the
compressive
stress
at
the
centroid
ofsolutions
elastic
compressive
stress
compressive
stress
distribution
as beam
doneoffor
the
deep
beam
of Fig.
1a.
be
provided
in deep
beams
toacontrol
The
code
requires
an compressive stress
or
since
load
path
is incompatible
with
elastic
stress
eld
is bottle-shaped
aswith
indicated
withcracking.
the dotted
line.
However,
distribution
as the
done
thecorresponding
deep
Fig.
1a.the
loaded
concentrated
load.
Inshows
reality,in
the
atbad,
the centroid
offor
the
elastic
compressive
stress
bottle-shaped
stress
fields.
Figure
7
a
deep
beam
loaded
with
isotropic struts
mesh with
a minimum
area
eld.
The4STM
in Fig.
is more appropriate,
or good,
because
0.001A
or 0.001t
mm2/
practice,
idealised
asofprismatic
(Fig.2 7).
EC2
Figure
shows
two4a
alternative
STMs for
a deep
beam.
Both the
c mm
are
eldoften
is bottle-shaped
indicated
withrequires
the
dotted line. However, in
distribution
done
for the deep
of Fig.
1a.
concentrated
In as
reality,
the compressive
stress field is bottleFigure are
4 shows
twoasalternative
STMs
forinbeam
aFig.
deep
beam.
Both are lower
strut
orientation
follows
the
elastic
compressive
trajectories.
mm
(where amount
Aa
cross-sectional
of the
and t is to
its
a minimum
horizontalload.
and area
vertical
webmember
reinforcement
lower
bound
solutions
but
the STM
4bstress
is
not optimal,
c is theof
practice,
struts
are
often
idealised
asAnnex
prismatic
(Fig. 7). EC2 requires
Figure
4 shows and
twoSchfer
alternative
STMs to
for
a deep
beam. Both
2
As
notedsince
by Schlaich
be
provided
into
each
face,with
but the
National
to
it isishelpful
realise
that
be
providedtoin
deep
beams
control
cracking.
The
code line.
requires
an
bad,
thethe
loadSTM
path in
is incompatible
with
the
elastic
stress
shaped
as indicated
theUK
dotted
However,
in practice, struts are
boundor
solutions
but
Fig., 4b
not
optimal,
or
bad, sincethickness)
the
9
minimum
amount
horizontal
and vertical
web reinforcement to
bound
solutions
the
STM
in
Fig. because
4b is not
2
2
EC2
loads
try to
use
with
thebut
least
forces
and
deformations.
increases
this
area
to 0.002A
isotropic
mesh a
with
a
minimum
areac. ofof
are
eld. lower
The
STM
inthe
Fig.path
4a is
more
appropriate,
or
good,
theoptimal,
0.001A
c mm or 0.001t mm /
idealisedinas
prismatic
7). EC2 cracking.
requires aThe
minimum
amountan
of
load path
isorientation
incompatible
with
the
elastic
stressstress
field.
The
in Fig.
be provided
deep
beams(Fig.
to control
code requires
or bad,
since
the load
path
is
incompatible
with
theSTM
elastic
stress
Consequently,
a follows
crude
rule
thumb
for distinguishing
between
strut
theof
elastic
compressive
trajectories.
mm (where Aoften
c is the cross-sectional area of the member and t is its
Figure 4
2
2
good
and
bad
models
is Schfer
to choose
model
with
shortest
ties, because
As
noted
by
Schlaich
and
thickness)
be
provided
face,
but
the
UK National
Annex
, itthe
isappropriate,
helpful
to realise
horizontal
and
vertical
web
reinforcement
totobec provided
in deep
beams
4a is more
appropriate,
good
the strut
orientation
follows
isotropic
mesh
with
a
minimum
area of
eld.
The
STM
in or
Fig.
4a
is, because
more
or that
good,
the andtobad
0.001A
mm2 or 0.001t
mm
/
Good
model
based in
oneach
minimising
length
of
ties
9
since ties
are
much
morewith
deformable
than
struts.
loads
try
to
use
the path
the least
forces
deformations.
this (where
area to 0.002A
c.
strutcompressive
orientation
follows
the
elastic
compressive
stress trajectories.
Ac is the
cross-sectional
of the member
andatminimum
is its
tomm
control
cracking.
The
code requires area
an isotropic
mesh with
the elastic
stress trajectories.
Asand
noted
by Schlaich
andEC2 increases
Consequently, a crude rule of thumb for distinguishing
between
2
2
2 noted by Schlaich and Schfer , it is helpful to realise that Figure 4
As
thickness)
to
be
in each
face,(where
but theAUK
Annex to
or 0.001t
mm2/mm
is National
the cross-sectional
area
of 0.001Ac mmprovided
, it is
helpful
to realise
that loads
try toand
useshortest
the path
Schfergood
Choice
ofbad
limiting
angle
ties
and
models
is to between
choose
thestruts
model
with
ties,with the Good and bad
c
model9based on minimising
length of ties
EC2 increases this area to 0.002Ac.
loads
trybetween
tomuch
use the
with
the
least
forces
andtodeformations.
The
angle
struts
andConsequently,
ties should
be
large
enough
since
ties
morepath
deformable
than
struts.
area of the member and t is its thickness)
to be provided in each face,
least forces
andare
deformations.
a crude
ruleavoid
of thumb for
strain
incompatibilities
due to
tiesof
extending
struts shorteningbetween
Consequently,
a crude
rule
thumb and
for distinguishing
9
8
but
the
UK
National
Annex
to
EC2
increases
this area to 0.002Ac.
distinguishing
between
good
and
bad
models
is
to
choose
the
model
Figure
4
in
almostand
thelimiting
same direction.
Model
Code
1990
gives
thewith
following
Choice
of
angle between
struts
ties
good
bad
models
is
to choose
theand
model
shortest ties,
Good and bad model based on minimising length of ties
useful
rules
ofsince
thumb
for developing
STMs
which
broadly
follow
the
The
angle
between
struts
and much
ties
should
be
large
enough
to
avoid
with shortest
ties,
ties
are
more
deformable
than
struts.
since
ties
are
much
more
deformable
than
struts.
elastic
stress eld:
strain incompatibilities
due to ties extending and struts shortening
8
almost
the same
direction.
Modelstruts
Code 1990
following
ChoiceinChoice
of
limiting
angle
between
andgives
ties theand
limiting
angle
ties
1. Orientation
thefor
elastic
stressbetween
STMs
eld is which
morestruts
important
for struts
useful
rulesof
ofwith
thumb
developing
broadly follow
the
than
ties,
which
can usually
be arranged
to be
the large
edges enough
of the
Thebetween
angle
between
struts
tiesparallel
should
to avoid
elastic
stress
eld:
The angle
struts
and
tiesand
should
be large
enough
to avoid
member
following practical considerations
due to ties extending and struts shortening
due
to ties
extending
and
struts
shortening
in
2. Orientation
highly stressed
regions,
main
struts
ties
should
at
1.
with
the
elastic
stress
eld and
is more
important
struts
8 for
inInalmost
the
same
direction.
Model
Code
1990meet
gives
the following
angles
of which
about
60 and
not
less
than
45
(this8isgives
a guideline
and
not
can
usually
be
arranged
parallel
to
the the
edges
of the
following
useful
almostthan
theties,
same
direction.
Model
Code
1990
useful
rules
of
thumb
for
developing
STMs
which
broadly
follow
the
mandatory)
member
following practical considerations
rules of2.
thumb
for
developing
STMsstruts
follow
elastic
stress
eld:
3.
If the
arrangement
of the model
iswhich
made
inbroadly
accordance
withthe
the
In
highly
stressed
regions,
main
and
ties
should
meet
at
elastic
eld,
ULS
not require
checking
of
about
60the
and
notdoes
less than
45 (this
is a guideline and not
elastic angles
stressstress
field:
mandatory)
1. Orientation with the elastic stress eld is more important for struts
3. If the arrangement
of the model
isfield
madeisinmore
accordance
with the
1. Orientation
with
the can
elastic
stressbe
important
foredges
strutsof the
than ties,
which
usually
arranged
parallel
to the
elastic stress eld, the ULS does not require checking
S
S
following
practical
considerations
thanmember
ties, which
can usually
be arranged
parallel to the edges of the
2. In
highly
regions,
main struts and ties should meet at
CDG
v1.indd stressed
37 practical
TSE40_36-41
member
following
considerations
angles of about 60 and not less than 45 (this is a guideline and not
mandatory)
2. In highly
stressed regions, main struts, ties should meet at angles of
3. 60
If the
arrangement
of45
the(this
model
made in accordance
with the
about
and
not less than
is aisguideline
and not mandatory)
elastic stress eld, the ULS does not require checking
3. If the arrangement of the model is made in accordance with the

elastic stress field, the ULS does not require checking
4. The orientation of the model may be allowed to depart from the

ULS if the SLS is checked
TSE40_36-41
CDG v1.indd
37the
elastic stress
field at
19/03/2015 14:19
19/03/2015 14:19
Figure 4 Good and bad model based on minimising length of ties
19/03/2015 14:19
Number 4
38
April 2015
Technical
Concrete design
Figure 6
Figure 5 Use of Model Code 19908 rules to distinguish between good and poor STM
Figure 6 Compressive
stress fields
a) prismatic
b) fan-shaped c) bottle-shaped
Compressive
stress
elds
Figure 5
Use of Model Code 19908 rules to distinguish between good and poor STM
Table 1: Eurocode 2 node strengths

Code
EC2
Type of node
Strength
Table 1: Eurocode 2 node strengths
1.0cd Strength
Code C-C-C
Type of node
EC2
C-C-T
C-C-C
0.85cd1.0f
C-T-T
C-C-T
0.75cd0.85f
C-T-T
Note: = 1ck/250
Note: = 1fck/250
cd
cd
0.75fcd
a) prismatic
b) fan-shaped
c) bottle-shaped
boundaries are less than or equal to the design concrete strengths given
in Table 1. It is necessary to ensure that the design stresses are less than
the design node
strengths
in both
nodes
at node-strut
interfaces
where
dimensions
need
to be and
chosen
to ensure that
stresses on
node
boundaries
are lessThe
thandesign
or equal
to the design
concrete
strengths on
strut resistance
is checked.
strength
of the
strut depends
given in Table 1. It is necessary to ensure that the design stresses
the reinforcement
provided within the depth of the element and is often
are less than the design strengths in both nodes and at node-strut
less than the
node strength.
interfaces
where strut resistance is checked. The design strength of
the strut depends on the reinforcement provided within the depth of
Dimensioning
of nodes
the element
and is often less than the node strength.
Compressive strength of struts
The dimensions
of concentrated
nodes are frequently determined by the
Dimensioning
of nodes
The and
dimensions
of concentrated
nodes
areposition
frequentlyofdetermined
by
widths of load
support
plates as well
as the
ties (Fig. 8).
EC2 gives design concreteCompressive
strengths for struts
which
are
based
on
the widths of load and support plates as well as the position of ties
strength of struts
simple criteria that account
forgives
the eff
ectsconcrete
of cracking
and confi
nement.
(Fig. 8).
EC2
design
strengths
for struts
which are based
Sizing of C-C-C
nodes
on simple criteria that account for the effects of cracking and
In general:
connement. In general:
Sizing
of C-C-C
nodes
The width of the loading
plate
of the deep beam in Figure 9 needs to
The width of the loading plate of the deep beam in Figure 9 needs to
transverse compression is favourable. Especially confinement

which
be
chosen
to
limit
the
bearing
stress to
a maximum
of 1.0
be chosen to limit the bearing
stress
to a maximum
of 1.0f
transverse compression is favourable. Especially connement from
cd cd which
from transverse reinforcement
or surrounding
concreteconcrete
or surrounding
is the
design for
strength
C-C-C The
nodes.
The depth
x ofC-C-C
the C-C-C
is the design
strength
C-C-Cfornodes.
depth
x of the
node
node in Fig. 9 depends upon the assumed exural compressive
transverse tension is detrimental
transverse tension is detrimental

in Fig. 9 depends
upon the assumed flexural compressive stress which
stress which cannot exceed 1.0fcd. The choice of exural
cannot exceed
1.0cd.stress
The choice
flexural compressive
stress
in
compressive
in Fig. 9of
determines
the widths of the
incoming
cracks are
detrimental
equilibrium requires
Skew cracks are particularly Skew
detrimental
asparticularly
equilibrium
requiresasshear
diagonalthe
struts.
Clause
(8) of EC2 states
that struts.
C-C-C nodes
shear forces to be transferred across cracks by aggregate interlock.
Fig. 9 determines
widths
of 6.5.4
the incoming
diagonal
Clause
forces to be transferred across
cracks by aggregate interlock. Struts
can normally be sized on the basis that stresses are equal on all
Struts are checked at their ends where the cross-sectional area is
6.5.4 (8) of EC2
states that C-C-C nodes can normally be sized on the
are checked at their ends where
cross-sectional
area
is reduces
definedthe
bydesign concrete
node boundaries. In this event, the stress distribution is hydrostatic
dened the
by the
node dimensions.
EC2
basis
that
stresses
equal
node boundaries.
In are
thisnormal
event,tothe
within
theare
node
and on
the all
incoming
strut centrelines
strength
in
prismatic
struts
with
transverse
tension
to
0.6f
where
cd
the node dimensions. EC2 reduces the design concrete strength in
the nodeisboundaries.
In practice,
the width
thethe
loaded
area is strut
= 1fck/250. The strength of 0.6fcd also applies to the ends
of distribution
stress
hydrostatic
within the
nodeofand
incoming
where
=
1
/250.
prismatic struts with transverse
tension
to
0.6
cd diagonal struts of
cd Fig. 7 if insufficient
normally governed by other considerations than the design concrete
bottle-shaped struts like the
centrelines
are
normal
to
the
node
boundaries.
In
practice,
the
width
The strength of 0.6cd also
appliesreinforcement
to the ends isofprovided
bottle-shaped
struts
strength. Consequently, the bearing stress can be signicantly less
transverse
to maintain
equilibrium after
of
the
loaded
area
is
normally
governed
by
other
considerations
than
than
cracking
of
concrete.
f
.
It
is
suggested
that
in
the
absence
of
shear
reinforcement,
cd
like the diagonal struts of Fig. 7 if insufficient transverse reinforcement is
the
exuralstrength.
compressive
stress is takenthe
as f
the widths
cd, to avoid
the
design
concrete
Consequently,
bearing
stress
can
be
provided to maintain equilibrium
after cracking
of concrete.
of the incoming struts being overestimated, but this is not an EC2
Classication
of nodes
signifi
less than cd. It is suggested that in the absence of shear
requirement.
Nodes are dened as regions were struts change direction or
strutscantly
Classification of nodes and ties intersect (Fig.7). Smeared nodes arise in regions where
reinforcement, the flexural compressive stress is taken as cd, to avoid
the orientation of struts is diverted by a band of evenly distributed
of C-C-T nodes
the
widths Sizing
of the incoming
struts being overestimated, but this is not an
Nodes are defined as regions
were struts
change
direction
reinforcement
bars
and is not
checkedor
in struts
design. Concentrated
The design bearing stress at C-C-T nodes is limited to a maximum
EC2
requirement.
nodes
are
typically
highly
stressed
and
need
to
be
designed
to
resist
of
0.85f
.
The
node
width normal to the inclined strut centreline in
and ties intersect (Fig. 7). Smeared nodes arise in regions where
cd
the incoming forces without concrete failing in compression. EC2
Fig. 8b is given by:
the orientation of struts is diverted by a band of evenly distributed

classies nodes as C-C-C (three compressive struts), C-C-TSizing
(two of C-C-T nodes
reinforcement bars and is compressive
not checkedstruts
in design.
nodes
and oneConcentrated
tie), and C-T-T (one
compressive strut
(1)
Thecap
design bearing stress at C-C-T nodes is limited to a maximum of
andand
twoneed
ties) (Figure
8). The node
column meets a pile
are typically highly stressed
to be designed
towhere
resistathe
would typically be a C-C-C node, and the node where a two-pile
pile . The
where
Lb width
is the length
of the
plate, u
is the
height of the
node
normal
to support
the inclined
strut
centreline
in
0.85
incoming forces without concrete
failing in compression. EC2 classifies
cd
cap meets the pile would typically be a C-C-T node.
back face of the node and is the angle of the strut centreline to the
Fig.
8b
is
given
by:
nodes as C-C-C (three compressive
struts),
C-C-T
(two
compressive
struts
Forces are transferred from ties into nodes through a combination
horizontal. EC2 does not require stresses to be checked at the back
of compressive
bearing and bond
stresses
within
the(Figure
node. The
and one tie), and C-T-T (one
strut
and two
ties)
8).dimensions of the face of C-C-T nodes.
nodes in Fig. 7 are governed by the dimensions of the loading and
It is benecial to provide the tensile reinforcement in several layers
(1)
w = Lbsin + ucos
The node where a columnsupport
meets plates
a pile as
cap
would
typically
be
a
C-C-C
well as the cover to the exural reinforcement. The
in highly stressed concentrated nodes since this increases u and
node, and the node where a two-pile pile cap meets the pile would
typically be a C-C-T node.
where Lb is the length of the support plate, u is the height of the
back face of the node and is the angle of the strut centreline to the
Forces are transferred from ties into nodes through a combination of
horizontal. EC2 does not require stresses to be checked at the back19/03/2015
bearing and bond stresses within the node. The dimensions of the nodes
face of C-C-T nodes.
in Fig. 7 are governed by the dimensions of the loading and support
plates as well as the cover to the flexural reinforcement. The node
It is beneficial to provide the tensile reinforcement in several layers
dimensions need to be chosen to ensure that stresses on node
in highly stressed concentrated nodes since this increases u and
14:19
39
39
hence the strut width w. The reinforcement should be sufficiently
anchored to develop its design tensile force at the node.
anchored
to develop its design tensile
force at the
Additional
recommendations
with respect
to node.
STM
8. If necessary, increase the design shear resistance by increasing

the node dimensions or providing shear reinforcement if C > CRd
8. If necessary,
increase
the design
shear resistance by
increasing
9. 7.
Calculate
the
exural
the
and wb with Eq.required
1 and Eq.to2resist
and hence
Calculate
thearea
strutofwidths
wtreinforcement
the node
dimensions
or/fproviding
shear reinforcement if C > CRd
design
force
T
.
(
A
=
T
)
1
s
1 with
yd Eq. 7
anchored to develop its design tensile force at the node.
9. the
strut
capacity
C
Calculate the area Rd
of exural reinforcement required to resist the
PD6687-1 (2010)10, the background document to the UK National
design
force Tincrease
Additional
recommendations
with respect
to
STM
= Tthe
)
8.
If necessary,
design
resistance
by3000mm,
increasing
the
9
1. (A s in
1/f9
yd
Annex
to EC2
Consider
the beam
Fig.
with Lshear
, gives 10the designwith
compressive
strength
= 5000mm,
h=
t=
Additional
recommendations
respect toconcrete
STM
PD6687-1 (2010) , the background document to the UK National
node
dimensions
or
providing
if C >and
CRd
as fcd = 0.85fck/c for
200mm, Lt = 300mm, Lb = 150mm,shear
STM. It also states that for shear, the shear
d = reinforcement
70mm, P = 500kN
9
10 , gives the design compressive concrete strength
Annex(2010)
to EC2
the beam in Fig. 9 with L = 5000mm, h = 3000mm, t =
, the background
document
toBS
theEN
UK1992-1-1:2004,
National Annex f =Consider
PD6687-1
resistance
of sections
should be veri
ed using
30MPa. the area of flexural reinforcement required to resist the
ck9. Calculate
as9, fgives
200mm, Lt =
STM. It alsoconcrete
states that
for shear,
the shear
300mm, Lb = 150mm, d = 70mm, P = 500kN and
cd = 0.85f
ck/ c for
to
EC2
the
design
compressive
strength
as
6.2.2 at all sections where av > 1.5d, where av is the distance of the
The design concrete strengths
force T1. (As = T1/yd) are as follows:
resistance of sections should be veried using BS EN 1992-1-1:2004,
fckdesign
= 30MPa.
cd = 0.85
/ac for
STM. It also states
that
for shear,
resistance
section
from
concentrated
load (or
support)
andthe
d isshear
the eff
ective
ck all
6.2.2 at
sections where av > 1.5d, where av is the distance of the
The design concrete strengths are as follows:
depth
of the
section.verified using BS EN 1992-1-1:2004, 6.2.2 at all
fConsider
/250)(0.85f
= 15.0MPa
of sections
should
theckbeam
in Fig. 9 with
L = 5000mm,
h = 3000mm, t = 200mm,
ck/1.5)
cd = (1f
section
from abe
concentrated load (or support) and d is the effective
(bearing
strength
top node)
sections
where
av section.
> 1.5d, where av is the distance of the section from
= cd
300mm,
Lb ck
=at
150mm,
d = ck
70mm,
= 500kN
and ck = 30MPa.
Lf
depth
of the
=
(1f
/250)(0.85f
/1.5)P=
15.0MPa
t
Example
1. Design
deep beam
centraldepth
pointofload
(bearing strength at top node)
a concentrated
load (orof
support)
and d iswith
the effective
the
Thef
design
concrete strengths are as follows:
Fig. 9Example
shows an 1.
STM
for a of
deep
beam
of thickness
t loaded
withload
a
0.85
cd = 0.85(1fck/250)(0.85fck/1.5) = 12.7MPa
Design
deep
beam
with central
point
section.
central
load an
P. The
strut
width
the bottom
nodewith
is a
(bearing
strength
at bottom
node)
= (1
/250)(0.85
/1.5)
= 15.0MPa
Fig.point
9 shows
STMinclined
for a deep
beam
of at
thickness
t loaded
0.85
=
0.85(1f
cd f cd ck
ckck/250)(0.85fck/1.5) = 12.7MPa
given
by Equation
1.of deep
(bearing
node)
Example
1. Design
with
central
central
point load
P. Thebeam
inclined
strut
widthpoint
at theload
bottom node is
(bearing strength
strength at
attop
bottom
node)
At given
the top
the inclined
strut width is given by:
0.6fcd = 0.6(1fck/250)(0.85fck/1.5) = 9.0MPa
bynode,
Equation
1.
0.85(1
/250)(0.85
/1.5) /1.5)
= 12.7MPa
0.85
Fig. 9 shows
STM
for athe
deep
beamstrut
of thickness
loaded
(strut
without
reinforcement)
ck calculated
cktransverse
At thean
top
node,
inclined
width is tgiven
by:with a
0.6strength
fcdcd==
0.6(1f
= 9.0MPa
ck/250)(0.85f
ck
(2)
(bearing
strength
at bottom
node)transverse reinforcement)
central point load P. The inclined strut width at the bottom
node is
(strut strength
without
calculated
The minimum required widths of the bearing plates are:
(2)
given by Equation 1.
= 0.6(1required
/250)(0.85
/1.5) = 9.0MPa
0.6
ck
ck of the bearing plates are:
where:
Thecd minimum
widths
(strut strength without calculated transverse reinforcement)
At thewhere:
top node, the inclined strut width is given by:
(3)
(2)
The minimum required widths of the bearing plates are:
wt = 0.5Ltsin + xcos
(3)
in
which:
500 x 103
where:
= 167mm < 300m provided
Lt
in which:

200 x 15
(4)
T1
x = (3)
(4)
Trial and
250error
x 103gives: x = 70mm, = 50, C = 326kN, wb = 205mm,
tco

= 98mm
150mprovided
and error gives:
x = <70mm,
= 50, C and
= 326kN,
wb = 205mm,
b Trial
where c0 fcd is the exural compressive stress.
wt L=
160mm,
T1 = 209kN.
200 x 12.7
where c0 fcd is the exural compressive stress.
wt = 160mm,
and T1 =is209kN.
in which:
Therefore,
the required area of exural reinforcement
required
area
exural
is
2gives:
and
errorthe
xinclined
= 70mm,
of=is
50,
C =reinforcement
326kN, windicating
= 205mm,
T1Trial
cot = (0.5L0.25Lt)/(h0.5xd)
(5)
/fydTherefore,
=
480mm
. The
strut
overstressed,
b
2
T1 = 0.5Pcot (4)
9
T1shear
cot = (0.5L0.25Lt)/(h0.5xd)
(5)
/f160mm,
480mm
strut
is
overstressed,
indicating
yd =reinforcement
that
is
required.
UKand
National
Annex to
w
=
CRd =. The
160 xinclined
200
x 1000
=The
287kN
T1 = 209kN.
t
that
shear reinforcement
is required.
The UK
National
Annex to of
EC2
requires
nominal horizontal
and vertical
web
reinforcement
The compressive
force
in the
strut is given
by:
flexural
compressive
stress.
where
c0 cd is the
2
EC2
requires
nominal
horizontal
and
vertical
web
reinforcement
of
The compressive force in the strut is given by:
= 480mm
.
Therefore,
the
required
area
of
flexural
reinforcement
is
T1/of
yd this
area 0.002Ac to be provided in each face. The effectiveness
area
0.002
A
to
be
provided
in
each
face.
The
eff
ectiveness
of
this
c strut is overstressed, indicating that shear reinforcement is
The
inclined
)/(h0.5xd)

(5)
cot
C ==(0.5L0.25L
0.5P/sin
(6)
t
C = 0.5P/sin
(6)
The
in the strut
is given by:within the depth of the
In compressive
the absenceforce
of minimal
reinforcement
In the absence of minimal reinforcement within the depth of the

element,
the compressive resistance of the strut is given by:
C = 0.5P/sin
element, (6)
the compressive resistance of the strut is given by:
required. The UK National Annex to EC2 requires nominal horizontal and

vertical
reinforcement of area 0.002Ac to be provided in each face.
Figure web
7
SS
Figure
7
STM
for simple
deep beam
STM for simple deep beam
CRdabsence
= 0.6f
In the
of minimal
reinforcement
within the (7)
depth
of the
cdmin(w
t,w b)t
C
(7)
Rd = 0.6f cdmin(w t,w b)t
element, the compressive resistance of the strut is given by:
EC2 does
not de
how
the strut
resistance
CRdCshould
be
EC2 does
notne
de
ne how
the strut
resistance
Rd should be
min(wt,wbstrength
)t
(7) C-C-T node-to-strut
CRd = 0.6cdThe
calculated.
is 0.6f
cd at the
calculated. strut
The strut strength
is 0.6f
at
the
C-C-T
node-to-strut
cd
interface,
but EC2
doesdoes
not de
the
strength
at C-C-C
nodes.
interface,
but EC2
not ne
de
ne strut
the strut
strength
C-C-C
nodes.
beat
calculated.
EC2 does
not define
how the strut
resistance
CRd should
It is suggested
that, that,
in the
of code-prescribed
minimal
web
It is suggested
in absence
the absence
of code-prescribed
minimal
web
The
strut
strength
is
0.6
at
the
C-C-T
node-to-strut
interface,
but
cd
reinforcement,
the strut
strength
is also
taken
as 0.6f
C-C-CEC2
cd at
reinforcement,
the strut
strength
is also
taken
as 0.6f
cd at C-C-C
does nodes.
not define the strut strength at C-C-C nodes. It is suggested that,
nodes.
in the absence of code-prescribed minimal web reinforcement, the strut
Typical
design
solution
procedure
strength
is also
taken
as 0.6
at C-C-C nodes.
Typical
design
solution
procedure
cd
Equations
27 can
using
the following
step-by-step
Equations
27 be
cansolved
be solved
using
the following
step-by-step
Typical
design solution procedure
procedure:
procedure:
Equations 27 can be solved using the following step-by-step

1. Find
allowable
lengths
of the
top top
andand
bottom
nodes.
1. the
Findminimum
the minimum
allowable
lengths
of the
bottom
nodes.
procedure:
(The
maximum
allowable
bearing
stress
at the
bottom
nodes
(The
maximum
allowable
bearing
stress
at the
top top
andand
bottom
nodes
1. Find
the
allowable
lengths of the top and bottom nodes.
are
fminimum
andf
0.85
fcd respectively)
are
f
cd 0.85
cd and
cd respectively)
Choose
(The
maximum
bearing stress at the top and bottom nodes
2. Choose
2.
x (e.g.x (e.g.
uallowable
) u)
3.
Calculate
with
3.
Calculate
cot0.85
cot
with
Eq.
5Eq. 5
and
respectively)
are
cd
cd
4. Calculate
T1 with
4. Calculate
T1 with
Eq. 4Eq. 4
2. Choose
x (e.g.
u)
5. Calculate
co with
3. If minimal
reinforcement
is provided,
5. Calculate
co with
Eq. 3.Eq.
If minimal
webweb
reinforcement
is provided,
the chosen
ofacceptable
x5is acceptable
if P/(L
t) [bearing
stress
at top
3. Calculate
cotvalue
with
Eq.
t
the
chosen
value
of x is
if P/(L
t)
[bearing
stress
at top
t
node]
adjust
by trial
error
cof
= fcd
co . f
cd. Otherwise,
node]
co
T
f
adjust
x byxtrial
andand
error
untiluntil
co =
cd Otherwise,
cd
4. Calculate
with
Eq.
4
1
6. Calculate
the strut
in terms
of applied
the applied
P with
6. Calculate
the strut
forceforce
C in C
terms
of the
loadload
P with
Eq.Eq.
6 6
5. Calculate
Calculate
with
3. If
minimal
web
isEq.
provided,
7. Calculate
theEq.
strut
widths
wt and
wreinforcement
Eq.
1 and
2 and
hence
b with
co strut widths w and
7.
the
w
with
Eq.
1
and
Eq.
2
and
hence
t
b
the
strut
capacity
with Eq. 7 if P/(L t) [bearing stress at top
the
chosen
valueCof
xCisRdacceptable
the
strut
capacity
t
Rd with Eq. 7
node] co cd. Otherwise, adjust x by trial and error until co = cd
Figure 7 STM for simple deep beam
Figure 8
Figure
8 cation of node types
Classi
Classication of node types
a) C-T-T
a)b)
C-T-T
C-C-T
b)c)
C-C-T
C-C-C
c) C-C-C
Figure 8 Classification of node types a) C-T-T b) C-C-T c) C-C-C
6. Calculate the strut force C in terms of the applied load P with Eq. 6
19/03/2015 14:19
19/03/2015 14:19
Number 4
40
April 2015
Figure 9
STM of deep beam
Figure 9 STM of deep beam
reinforcement can be assessed using the EC2 design equation for
Technical
Concrete design
Figure 10
Bottle stress elds
a) partial discontinuity
b) full discontinuity
Figure 10 Bottle stress fields a) partial discontinuity b) full discontinuity
reinforcement with minimum area of 0.002Ac = 400mm2/m in each
reinforcement, the normal resistance nRd per unit

The effectiveness of
can be
assessed
using
the EC2
design
fullreinforcement
discontinuities, stress
eld
models or
the EC2
design
equations For orthogonal
face in each direction.
for shear enhancement
withinmodels
2d of supports.
reinforcement,
the normal
equation for full discontinuities,
stress field
or the EC2 design
lengthFor
to orthogonal
the strut centreline
is given
by: resistance nRd per unit
length to the strut centreline is given by:
equations for shear enhancement within 2d of supports.
2
2
Transverse reinforcement in bottle stress elds
(8)
nRd = (As1sin /s1 + As2cos /s2)fyd
Equations 6.58 and 6.59 of EC2 are used to calculate the area of
nRd = (As1sin2/s1 + As2cos2/s2)fyd
(8)
Transverse reinforcement
in bottle stress fields
transverse reinforcement required to equilibrate the transverse
where s1/s2 = spacing of horizontal bars of cross-sectional area As1 and
in bottle-shaped struts with partial and full discontinuity.
where s1/s2 = spacing of horizontal bars of cross-sectional area As1
vertical
bars of cross-sectional area As2 and is the angle of the strut
Equations 6.58 andtension
6.59 of
EC2 are used to calculate the area of
and vertical bars of cross-sectional area
As2 and is the angle of the
centreline
to the horizontal.
required to equilibrate the transverse tension
Partial discontinuity
strut centreline
to the horizontal.
For cases
partial and
discontinuity
the design tensile force for the
For isotropic reinforcement As1 = As2 and nRd = Asfyd/s in all
in bottle-shaped struts
withofpartial
full discontinuity.
For isotropic
transverse reinforcement (Figure 10) is given by:
directions.reinforcement As1 = As2 and nRd = Asyd/s in all directions.
Hence, the required minimum reinforcement grid of 0.002Ac (i.e.
Partial discontinuity
Hence,
the 2required
minimum
reinforcement
gridcan
of resist
0.002A
400mm
T = F(ba)/4b
(Exp 6.58)
/m vertically
and horizontally)
in each face
a c (i.e.
For cases of partial discontinuity the design tensile force for the
2
transverse
force: and horizontally) in each face can resist a
400mm
/m vertically
10)inisthe
given
where F (Figure
= the force
strut,by:
b = the available strut width and a =
transverse force:
the node width.
T = F(ba)/4b
(Exp 6.58)
Alternatively, the strut capacity corresponding to minimum web
435
3.78
reinforcement
given
Full discontinuity
T = (800
x 1000 ) xis0.6
x by:
=394kN >> 79kN
2
where F = the force
incases
the strut,
= the available
strut(Fig.
width
and
For
of fullbdiscontinuity
(b > H/2)
10b):
(9)
a = the node width.
T = F[1.00.7a/H]/4
(Exp 6.59R)
Alternatively, the strut capacity corresponding to minimum web
in which a =iswgiven
, H is the
t
Full discontinuity
reinforcement
by:strut length and nRd is given by Equation 8.
The maximum possible shear force corresponding to crushing of
in which H is the strut length and a is the width of the loaded area.
For cases of full discontinuity
(b
>
H/2)
(Fig.
10b):
the concrete at the bottom node of the deep beam of Example 1 is
(Expression 6.59R replaces the term a/h in Expression 6.59 of EC2
1.2n H
a/H as directed by BSI paper
381kN.
resistance
can be increased
T = F[1.00.7a/H]/4with

(Exp B525/2
6.59R) 11 0034.)
CRD =Vminimum(0.85w
,wThe
tvshear
,
) (9)
max = 0.85f cdtLb = tv
b
cd
t
cd
Rd
1.0 H ] and loading plates as

EC2 requires the area of transverse reinforcement resisting T to
further by increasing the length of the[bearing
be provided
the compression
trajectories
in which H is the strut
length over
and the
a islength
the width
of the loaded
area. are curved, well as d.
but does not dene this distance. Schlaich and Schafer2 show the in which a = wt, H is the strut length and nRd is given by Equation 8.
(Expression 6.59R replaces
the term a/h in Expression 6.59 of EC2
transverse reinforcement being distributed over the central 0.8H of
Use of EC2 beam equations for design of shear
with a/H as directed
BSI paper
B525/2
11 0034.)
maximum possible shear force corresponding to crushing of the
theby
partial
discontinuity
shown
in Fig. 10b. However, this paper takesThe reinforcement
the length of the transverse tension zone as 0.6H as suggested by concrete
Alternatively,
shear reinforcement
can
be designed
a truss1 is
at the bottom
node of the
deep
beam ofusing
Example
EC2 requires the area
of transverse
Hendy
and Smith11. reinforcement resisting T to
model like that of Schlaich and Schfer12 (Figure 11) or with the EC2
= 0.85
tL
=
381kN.
The
shear
resistance
can
be
increased
Vmax design
cd b
equations
for shear enhancement in beams. In the case
be provided over the length the compression trajectories are curved,
further
byEC2
increasing
the lengthequations,
of the bearing
and loading
of the
shear enhancement
no calculated
shear plates as
Example
2.
Inclined
strut
with
full
discontinuity
2
but does not define this distance. Schlaich and Schafer show the
is required, provided the design shear stress
The inclined strut in Fig. 9 is a full discontinuity of length:
wellreinforcement
as d.
transverse reinforcement being distributed over the central 0.8H of
vEd vRdc where vRdc is given by Equation 6.2a in EC2 and = av/2d
av beam
is the clear
shear span.
mainof
advantage
of using a
the partial discontinuity shown in Fig. 10b. However, this paper takes
Usewhere
of EC2
equations
for The
design
shearreinforcement
truss model over the bottle stress eld is to steepen the angle of the
the length of the transverse tension zone as 0.6H as suggested by
inclined strut
at the
support. For example,
effect ofusing
introducing
Hence, the design tensile force to be resisted by transverse
Alternatively,
shear
reinforcement
can bethe
designed
a truss model
Hendy and Smith11reinforcement
.
the vertical tie in the simplied truss
model of Fig. 11 is to increase
over each half of the strut length is:
12
(Figure
11)
or
with
the
EC2
design
like that
of
Schlaich
and
Schfer
the resultant angle of the inclined strut from:
0.7a
Example 2. InclinedTstrut
with full discontinuity
= F[1.00.7a/H]/4
= 326[1.00.7160/3776]/4 = 79kNequations for shear enhancement in beams. In the case of the EC2 shear
enhancement equations, no calculated shear reinforcement is required,
The inclined strut in The
Fig. tensile
9 is a full
discontinuity
force
T is provided of
by length:
a mesh of horizontal and vertical
provided the design shear stress vEd vRdc where vRdc is given by
(3000 70 70/2)
Equation 6.2a in EC2 and = av/2d where av is the clear shear span. The
H =
= 3776mm
main advantage of using a truss model over the bottle stress field is to
sin50
steepen the angle of the inclined strut at the support. For example,
the
TSE40_36-41
CDG
v1.indd
40
19/03/2015 14:19
Hence, the design tensile force to be resisted by transverse
effect of introducing the vertical tie in the simplified truss model of Fig.
reinforcement over each half of the strut length is:
11 is to increase the resultant angle of the inclined strut from:
T = F[1.00.7a/H]/4 = 326[1.00.7160/3776]/4 = 79kN
z
tan -1 ( )
a
The tensile force T is provided by a mesh of horizontal and vertical
reinforcement with minimum area of 0.002Ac = 400mm2/m in each
face in each direction.
The strut strength can be increased if necessary by increasing node

dimensions, providing shear reinforcement or a combination of the
two. The minimum reinforcement required by the UK National Annex
within the depth of deep beams is often sufficient to increase the
strut strength at C-C-T and C-C-C nodes to the node strengths of
0.85cd and cd respectively.
to:
tan1
F1
Tmax F1cot
where cot = 0.5a/z.

Model Code 19908 estimates the tie force F1 as:
F1
= (2
1) /3 for 0.5z a 2z (10)
References:
1) British Standards Institution (2004) EN-1992-1-1:2004. Eurocode 2.
where z is the flexural lever arm.
41
Design of concrete structures. Part 1. General rules and rules for buildings,
Clause 6.2.3 (8) of EC2 reduces the design shear force due to loads
London, UK: BSI
applied to the top surface of beams within 0.5d av 2d of supports by
2) Schlaich J. and Schfer K. (1991) Design and detailing of structural
node, wt = strut width at top node, t = member thickness, fsb = strut
=to:
av /2d where av is the clear shear span. Shear reinforcement
needs
to
concrete using strut-and-tie models, The Structural Engineer, 69 (6), pp.
strength at bottom node and fst = strut strength at top node). In the
shear reinforcement,
take fst = fsb = 0.6(1fck/250)fcd.
be provided within the central of av to resist the reducedabsence
designofshear
113125
The strut strength can be increased if necessary by increasing node
force.
Consideration of Clause 6.2.3 (8) shows that the reduced
design
dimensions, providing shear
reinforcement
or a Federation
combination offor
theStructural Concrete (2011) fib
where
3) International
8
The(8)
minimum
Model
Code
estimates
tie tie
force
F1 as:F1 of Fig. 11. Clausetwo.
shear
force
is1990
equivalent
tothe
the
force
6.2.3
is reinforcement required by the UK National Annex
Bulletin
No.suffi
61:cient
Design
examples
within the depth of deep beams
is often
to increase
the for strut-and-tie models, Lausanne,
typically less onerous than Equation 10,(10)
with both being less
onerous
strut strength at C-C-T andSwitzerland:
C-C-C nodes tofib
the node strengths of
than Expression 6.59 of EC2 for a full discontinuity. It is also0.85
possible
tocd respectively.
fcd and f
where z is the exural lever arm.
design
the shear reinforcement using stress field models7,13.
4) Thurlimann B., Muttoni A. and Schwartz J. (1989) Design and
Clause 6.2.3 (8) of EC2 reduces the design shear force due
References anddetailing
furtherofreading
to loads applied to the top surface of beams within 0.5d av 2d
reinforced concrete structures using stress fields, Zurich,
General
application
of supports
by = av/2d where av is the clear shear span. Shear
Switzerland:
Swiss Federal Institute of Technology
reinforcement needs to be provided within the central of av to
1) British Standards Institution (2004) EN-1992-1-1:2004. Eurocode
2. Design
resist
the reduced
shear method
force. Consideration
Clause be applied
The
STM
is a verydesign
versatile
that canofreadily
to aof concrete structures. Part 1. General rules and rules for
buildings, London, UK: BSI 5) Goodchild C., Morrison J. and Vollum R. L. (2015) Strut-and-tie
6.2.3 (8) shows that the reduced design shear force is equivalent to
wide
range of structures that would otherwise only be designable
with
the tie force F1 of Fig. 11. Clause 6.2.3 (8) is typically less onerous
Models, London, UK: MPA The Concrete Centre
2) Schlaich J. of
and Schfer K. (1991) Design and detailing of structural
than Equation 10, analysis
with both being
less onerous
than Expression
6.59
finite-element
or empirical
design
methods.
The advantage
concrete using strut-and-tie models, The Structural Engineer, 69 (6), pp.
of EC2 for a full discontinuity. It is also possible to design the shear
an
STM over finite-element analysis is that an STM requires113125
load paths
6) Schlaich J., Schfer K. and Jennewein M. (1987) Towards a
reinforcement using stress eld models7,13.
to be clearly identified. Hence, structures can readily be checked for
consistent design of structural concrete, PCI Journal, 32 (3), pp. 74150
3) International Federation for Structural Concrete (2011) fib Bulletin No.
General application
equilibrium.
An STM also has the advantage of clearly identifying
the
61: Design examples for strut-and-tie
models,J.Lausanne,
Switzerland:
b
The STM is a very versatile method that can readily be applied to a
7) Sagaseta
and Vollum
R. L. (2010)
Shear design of short-span
wide range ofrequirements
structures that would
otherwise only beunlike
designable
anchorage
of reinforcement,
finite-element analysis.
, Magazine
Concrete
Research, 62 (4), pp. 267282
4) Thurlimann B., Muttoni A. beams
and Schwartz
J. (1989) of
Design
and detailing
with nite-element analysis or empirical design methods. The
of reinforced concrete structures using stress elds, Zurich, Switzerland:
advantage of an STM over nite-element analysis is that an STM
Summary
Swiss Federal Institute of Technology
requires load paths to be clearly identied. Hence, structures can
8) CEB-FIP (1990) Model Code for Concrete Structures, Lausanne,
readily be checked for equilibrium. An STM also has the advantage
Itofisclearly
oftenidentifying
convenient
to base the STM geometry at the ULS5)on
the elastic
Switzerland: CEB-FIP
Goodchild C., Morrison J. and Vollum R. L. (2015) Strut-and-tie
the anchorage requirements of reinforcement,
Models, at
London,
unlike field,
nite-element
analysis.
stress
since this
typically ensures satisfactory performance
the UK: MPA The Concrete Centre
SLS.
However, some deviation from the elastic solution is permissible.
6) Schlaich J., Schfer K. and
JenneweinAnnex
M. (1987)
a 2. Design of concrete structures. Part 1.
Summary
National
toTowards
Eurocode
consistentcan
design of structural concrete, PCI Journal, 32 (3), pp. 74150
is often
convenient
base the STMis
geometry
at the ULS
on STM geometry
IfItthe
elastic
stresstodistribution
not available,
the
General
rules
and
rules
for
buildings, London, UK: BSI
the elastic stress eld, since this typically ensures satisfactory
2
.
be
developed using the load path method of Schlaich and7)Schfer
Sagaseta J. and Vollum R. L. (2010) Shear design of short-span
performance at the SLS. However, some deviation from the elastic
beams, for
Magazine of Concrete
(4), pp. 267282
solutionCode
is permissible.
If the elasticastress
distribution isrule
not which is useful
Model
19908 suggests
2:1 dispersion
10)Research,
British 62
Standards
Institution (2010) PD 6687-1:2010 Background
available, the STM geometry can be developed using the load path
establishing
the
basic
STM
geometry.
The
axial
resistance
of
struts
in
paper
to
the
National
8) CEB-FIP (1990) Model Code for Concrete Structures,Annexes
Lausanne, to BS EN 1992-1 and BS EN 1992-3,
method of Schlaich and Schfer2. Model Code 19908 suggests a
Switzerland:
CEB-FIP
dispersion
rule which
is useful
establishingof
the
STM w t (in
a2:1
planar
member
equals
thefor
minimum
wbasic
tf
and
which
London,
UK:
BSI
b sb
t st
axial resistance of struts in a planar member equals
wgeometry.
= strutThe
width
at bottom node, w = strut width at top node,
b minimum of wbtfsb and wttfst (in which wb t= strut width at bottom
the
11) Hendy C. R. and Smith D. A. (2007) Designers Guide to EN 1992
t = member thickness, sb = strut strength at bottom node National
and Annex to Eurocode 2. Design of concrete structures. Part 1.
Eurocode
2: Design
General rules and rules for buildings,
London,
UK: BSIof concrete structures. Part 2: concrete bridges,
st = strut strength at top node). In the absence of shear reinforcement,
London,
UK:
Thomas
Telford
10) British Standards Institution (2010) PD 6687-1:2010 Background
take
st =11ed
sbtruss
= 0.6(1
/250)
.
SFigure
ck
cd and
Simpli
model according
to Schlaich
F
Schfer13
paper to the National Annexes to BS EN 1992-1 and BS EN 1992-3,

12) Schlaich J. and Schfer K. (2001) Konstruieren im Stahlbetonbau
London, UK: BSI
(in German), BetonKalender (Vol 2), Berlin, Germany: Ernst & Sohn, pp.
11) Hendy C. R. and Smith D. A. (2007) Designers Guide to EN 1992

311492
Eurocode 2: Design of concrete
structures. Part 2: concrete bridges,
London, UK: Thomas Telford
13) Vollum R. L. and Fang L. (2014) Shear enhancement in RC beams
12) Schlaich J. and Schfer K. (2001) Konstruieren im Stahlbetonbau

with multiple point loads, Engineering Structures, 80, pp. 389405
(in German), BetonKalender (Vol 2), Berlin, Germany: Ernst & Sohn, pp.
311492
Further reading
13) Vollum R. L. and Fang L. The

(2014)Concrete
Shear enhancement
in RCrecently
beams
Centre has
published a guide to Strut-and-tie
with multiple point loads, Engineering5Structures, 80, pp. 389405
Models , which gives more information on how to construct and use
Figure 11 Simplified truss model according to Schlaich and Schfer13
Further reading
STMs. This is available via the Concrete Centre website:
The Concrete Centre has recently published a guide to Strut-and-tie
Models5, which gives more information
on how to construct and use
STMs. This is available via the Concrete Centre website:
19/03/2015 14:19
Publications from The Concrete Centre

Free to Download
Concrete Floor Solutions for
Passive and Active Cooling
Publication date: 2012
Ref: TCC/05/26
Concrete Framed Buildings
Ref: TCC/03/24
Crosswall Construction
Ref: TCC/03/26
Hybrid Concrete Construction
Ref: TCC/03/53
Post-tensioned Concrete Floors
Ref: TCC/03/33
Precast Concrete in Buildings
Ref: TCC/03/31
Car Parks
Ref: TCC/03/34
Concrete Bridges
Ref: TCC/02/08
Concrete Footbridges
Ref: TCC/02/11
Hospital Construction
Ref: TCC/03/44
Office Cost Study
Ref: TCC/03/57
Schools
Ref: TCC/03/52
Purchase online e-books now available

Concrete and Fire Safety
Ref: TCC/03/43
Material Efficiency
Ref: TCC/05/21
Thermal Mass Explained
Ref: TCC/05/11
Visual Concrete
Ref: TCC/05/29
Whole-Life Carbon and Buildings
Ref: TCC/05/30
Specifying Sustainable Concrete
Ref: TCC/05/24
A Design Guide for Footfall

Induced Vibration of Structures
Ref: CCIP-016
Properties of Concrete for use in

Eurocode 2
Ref: CCIP-029
Concise Eurocode 2
Ref: CCIP-005
Residential Cellular Concrete

Buildings
Ref: CCIP-032
Concrete Basements
Ref: CCIP-044
Concrete Buildings Scheme
Design Manual
Ref: CCIP-051
Strut-and-Tie Models
Ref: CCIP-057
Tall Buildings
Ref: CCIP-058
Cost Model Studies Commercial/Schools/Hospitals

Publication date: 2007/8
Ref: CCIP- 10/11/12
Utilisation of Thermal Mass in

Non-Residential Buildings
Ref: CCIP-020
Design and Construction using

Insulating Concrete Formwork
CCIP-019
Worked Examples to Eurocode 2:

Volume 1
Ref: CCIP-041
Price: 50
Design of Hybrid Concrete

Buildings
Ref: CCIP-030
Designing for Safer Concrete
Structures
Ref: CCIP-043
RC Spreadsheets: version 4B.2

Ref: CCIP-008.
Single user and Site licences are
available
Economic Concrete Frame

Elements
Ref: CCIP-025
How to Design Concrete
Structures using Eurocode 2
Ref: CCIP-006
Performance of Concrete
Structures in Fire
Ref: CCIP-031
Visit www.concretecentre.com/publications to download or purchase these titles.
The Concrete Centres vision is to make

concrete the material of choice. We provide
material, design and construction guidance.
Our aim is to enable all those involved in the
design, use and performance of concrete to
realise the potential of the material.
The Concrete Centre is part of the
Mineral Products Association, the trade
association for the aggregates, asphalt,
cement, concrete, dimension stone, lime,
mortar and silica sand industries.
www.mineralproducts.org
The Concrete Centre provides design

guidance, seminars, courses, online resources
and industry research to the design
community.
For more information and downloads, visit:
www.concretecentre.com/publications
www.concretecentre.com/events
www.concretecentre.com/cq
Subscribe to our email updates:
www.concretecentre.com/register
Follow us on Twitter:
@concretecentre
@thisisconcrete
Published by MPA The Concrete Centre
Gillingham House, 38-44 Gillingham Street, London SW1V 1HU
Ref: TCC/03/59
ISBN: 978-1-908257-18-5
MPA The Concrete Centre, 2016
All advice or information from MPA The Concrete Centre is intended only for use in the UK by those who will evaluate the significance and limitations
of its contents and take responsibility for its use and application. No liability (including that for negligence) for any loss resulting from such advice or
information is accepted by Mineral Products Association or its subcontractors, suppliers or advisors. Readers should note that the publications from
MPA The Concrete Centre are subject to revision from time to time and should therefore ensure that they are in possession of the latest version.

MB Structural Design Compendium May16

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

MB Structural Design Compendium May16

Загружено:

Авторское право:

Доступные форматы

Structural Design of

Concrete and Masonry:

Structural Design of Concrete and Masonry

How to calculate anchorage and lap lengths to Eurocode 2

Deflection the span-to-effective-depth method and Eurocode 2

Fire Design of concrete columns and walls to Eurocode 2

Eurocode 6: Design of masonry structures for vertical loads

Design of post-tensioned slabs

Guidance on the design of liquid-retaining structures

An introduction to strut-and-tie modelling

About The Concrete Centre

Structural Design of Concrete and Masonry

Structural Design of Concrete and Masonry

Structural Design of Concrete and Masonry

How to calculate anchorage and lap

This article provides guidance on how to calculate

The calculation for anchorage and lap lengths is as described in EC2

Figure 1 Tension Anchorage

Ultimate bond stress

Structural Design of Concrete and Masonry

Source: EC2-1-1 Figure 8.1, b, c and d.

Confinement of concrete results in the characteristic compression

Poor bond conditions

Figure 3 Good and Poor bond conditions

Table 1 gives the design tensile strengths for structural concretes up to

= 1.0 for bar diameters 32mm

Structural Design of Concrete and Masonry

Figure 4: flow chart for anchorage lengths.

Determine fctd from Table 1

Is the bar in good

Does the bar have

Determine the coefficients 1 to 5 (see Table 2)

Can lb,rqd be used as the design

Determine As,req and As,prov where the anchorage starts

Determine ultimate design stress in bar

Determine basic anchorage length

Figure 4 Flow chart for anchorage lengths

To calculate the values of 1 and 2 the value of cd is needed. cd is

Structural Design of Concrete and Masonry

Check lbd >

Does the bar have

Take 235 = 0.7

Is the bar confined

Alpha values for tension anchorage

Table 1: Design tensile strength, ctd

Bars other than straight, 1 = 0.7 if cd > 3; otherwise 1 = 1.0

Structural Design of Concrete and Masonry

b) Bent or hooked bars

Figure 5 Values of cd (c and c1 are taken to be cnom)

Source: EC2-1-1, Figure 8.4.

Table 2: Anchorage and lap lengths for locations of maximum stress

Structural Design of Concrete and Masonry

Ast,min = 0.25 491 = 123mm2

3 = 1 K = 1 0.1 0.38 = 0.96

Figure 8 Plan view of slab illustrating transverse tension

depends on the position of the confining reinforcement.

is the cross-sectional area of the transverse reinforcement

is the cross-sectional area of the minimum transverse

4 confinement by welded transverse reinforcement

Structural Design of Concrete and Masonry

Figure 9: flow chart for lap lengths.

Determine fctd from Table 1

Is smaller bar diameter

Determine ultimate bond stress