CVE 505 – Structural Design (3 Credit Units)

Columns:

The columns in a structure carry the loads from the beams and slabs down to the foundations, and
therefore they are primarily compression members, although they may also have to resist bending
forces due to the continuity of the structure. Slenderness, and the risk of lateral deflections leading to
buckling, is an important consideration in determining failure modes and special requirements will
apply to the design of slender columns. Columns which are required to carry significant moments
about both axes, such as corner columns in buildings, may also need special consideration when they
have to support a large axial load. Design of columns is governed by the ultimate limit state;
deflections and cracking during service conditions are not usually a problem, but nevertheless correct
detailing of the reinforcement and adequate cover are important.

Most columns are termed short columns and fail when the material reaches its ultimate capacity under
the applied loads and moments. Slender columns buckle and the additional moments caused by
deflection must be taken into account in design.

Loading and moments

it is necessary to classify structures into one of the following types:

1. braced – where the lateral loads are resisted by shear walls or other forms of bracing capable of
transmitting all horizontal loading to the foundations, and

2. unbraced – where horizontal loads are resisted by the frame action of rigidly connected columns,
beams and slabs.

With a braced structure the axial forces and moments in the columns are caused by the vertical
permanent and variable actions only, whereas with an unbraced structure the loading arrangements
which include the effects of the lateral loads must also be considered.

Both braced and unbraced structures can be further classified as sway or non-sway. In a sway structure
sidesway is likely to significantly increase the magnitude of the bending moments in the columns
whereas in a non-sway structure this effect is less significant. This increase of moments due to sway,
known as a ‘second order’ effect, is not considered to be significant if there is less than a 10 per cent
increase in the normal (‘first order’) design moments as a result of the sidesway displacements of the
structure. Substantially braced structures can normally be considered to be non-sway. EC2 gives
further guidance concerning the classification of unbraced structures. Only the design of braced non-
sway structures will be considered.

For a braced structure the critical arrangement of the ultimate load is usually that which causes the
largest moment in the column, together with a large axial load. As an example, figure 9.1 shows a
building frame with the critical loading arrangement for the design of its centre column at the first-
floor level and also the left-hand column at all floor levels. When the moments in columns are large
and particularly with unbraced columns, it may also be necessary to check the case of maximum
moment combined with the minimum axial load.

In the case of braced frames, the axial column forces due to the vertical loading may be calculated as
though the beams and slabs are simply supported, provided that the spans on either side of the column
differ by no more than 30 per cent and there is not a cantilever span. In some structures it is unlikely
that all the floors of a building will carry the full imposed load at the same instant, therefore a
reduction is allowed in the total imposed load when designing columns or foundations in buildings
which are greater than two storeys in height. Further guidance on this can be found in BS EN1991-
1-1 (Actions on structures).

Column Classification and Failure Modes

(1) Slenderness ratio of a column

The slenderness ratio 𝜆 of a column bent about an axis is given by

𝑙 𝑙
𝜆 1
𝑖 𝐼 ⁄𝐴

where

𝑙 is the effective height of the column

𝑖 is the radius of gyration about the axis considered

𝐼 is the second moment of area of the section about the axis

𝐴 is the cross-sectional area of the column

(2) Effective height 𝑙 of a column

The effective height of a column, 𝑙 , is the height of a theoretical column of equivalent section but
pinned at both ends. This depends on the degree of fixity at each end of the column, which itself
depends on the relative stiffnesses of the columns and beams connected to either end of the column
under consideration.

EC2 gives two formulae for calculating the effective height:

For braced members:

𝑘 𝑘
𝑙 0.5𝑙 1 1 2
0.45 𝑘 0.45 𝑘
For unbraced members the larger of :

𝑘 𝑘
𝑙 𝑙 1 10 3a
𝑘 𝑘

And

𝑘 𝑘
𝑙 𝑙 1 1 3b
1 𝑘 1 𝑘
where 𝑙 is the clear height of the column between end restraints.

In the above formulae, 𝑘 and 𝑘 are the relative flexibilities of the rotational restraints at ends ‘1’
and ‘2’ of the column respectively. At each end 𝑘 and 𝑘 can be taken as:

𝑐𝑜𝑙𝑢𝑚𝑛 𝑠𝑡𝑖𝑓𝑓𝑛𝑒𝑠𝑠 𝐸𝐼 ⁄𝑙 𝐼 ⁄𝑙
𝑘
∑ 𝑏𝑒𝑎𝑚 𝑠𝑡𝑖𝑓𝑓𝑒𝑠𝑠 ∑ 2 𝐸𝐼⁄𝑙 ∑ 2 𝐼 ⁄𝑙

It is assumed that any column above or below the column under consideration does not contribute
anything to the rotational restraint of the joint and that the stiffness of each connecting beam is taken
as 2𝐸𝐼/𝑙 to allow for cracking effects in the beam.

Fig. 1: Effective length calculation for a column in a symmetrical frame

Table 1: Column effective lengths

1 I⁄𝑙 0
𝑘
4 I⁄ 𝑙 (fixed 0.0625 0.125 0.25 0.50 1.0 1.5 2.0
end)
𝑙 𝑏𝑟𝑎𝑐𝑒𝑑
0.5 0.56 0.61 0.68 0.76 0.84 0.88 0.91
𝑒𝑞𝑢 2 𝑙
 𝑙 𝑢𝑛𝑏𝑟𝑎𝑐𝑒𝑑
𝑒𝑞𝑢 3𝑎 1.0 1.14 1.27 1.50 1.87 2.45 2.92 3.32

𝑒𝑞𝑢 3𝑏 Use greater

1.0 1.12 1.13 1.44 1.78 2.25 2.56 2.78
value 𝑙

Hence, for a typical column in a symmetrical frame with spans of approximately equal length, as
shown in figure 9.2, 𝑘 and 𝑘 can be calculated as:

𝑐𝑜𝑙𝑢𝑚𝑛 𝑠𝑡𝑖𝑓𝑓𝑛𝑒𝑠𝑠 𝐼 ⁄𝑙 1 𝐼 ⁄𝑙 1 𝐼 ⁄𝑙
𝑘 𝑘 𝑘
∑ 𝑏𝑒𝑎𝑚 𝑠𝑡𝑖𝑓𝑓𝑒𝑠𝑠 ∑ 2 𝐼 ⁄𝑙 2 2 𝐼 ⁄𝑙 4 𝐼 ⁄𝑙

(3) Limiting slenderness ratio – short or slender columns

EC2 places an upper limit on the slenderness ratio of a single member below which second order
effects may be ignored. This limit is given by:

𝜆 20 𝐴 𝐵 𝐶 ⁄√𝑛 4

where:

𝐴 1⁄ 1 0.2𝜙

𝐵 √1 2𝑤

𝐶 1.7 𝑟

𝜙 effective creep ratio (if not known A can be taken as 0.7)

𝑤 A f ⁄ A f (if not known B can be taken as 1.1)

f the design yield strength of the reinforcement

f the design compressive strength of the concrete

𝐴 the total area of longitudinal reinforcement

𝑛 𝑁 ⁄ 𝐴 𝑓

𝑁 the design ultimate axial load in the column

𝑟 M ⁄M (if 𝑟 not known then C can be taken as 0.7)

M ,M are the first order moments at the end of the column with |M | |M |

Whilst use of the default values for A and B are unlikely to introduce major errors, C may have a
large effect and should be calculated wherever possible.

The following conditions apply to the value of C:

(a) If the end moments, M and M , give rise to tension on the same side of the column 𝑟 should
be taken as positive from which it follows that 𝐶 1.7.

(b) If the converse to (a) is true, i.e., the column is in a state of double curvature, then 𝑟 should be
taken as negative from which it follows that 𝐶 1.7.

(c) For braced members in which the first order moments arise only from transverse loads or
imperfections; C can be taken as 0.7.

(d) For unbraced members; C can be taken as 0.7.

For an unbraced column an approximation to the limiting value of 𝜆 will be given by:

𝜆 20 𝐴 𝐵 𝐶 ⁄√𝑛 20 0.7 1.1 0.7⁄ 𝑁 ⁄A f 10.8⁄ 𝑁 ⁄A f

The limiting value of 𝜆 for a braced column will depend on the relative value of the column’s end
moments that will normally act in the same clockwise or anti-clockwise direction as in case (b) above.
If these moments are of approximately equal value, then 𝑟 1, 𝐶 1.7 1 2.7 and a typical,
approximate limit on 𝜆 will be given by:

𝜆 20 𝐴 𝐵 𝐶 ⁄√𝑛 20 0.7 1.1 2.7⁄ 𝑁 ⁄A f 41.6⁄ 𝑁 ⁄A f

Alternatively for a braced column the minimum limiting value of 𝜆 will be given by taking 𝐶 1.7.
Hence:

𝜆 20 𝐴 𝐵 𝐶 ⁄√𝑛 20 0.7 1.1 1.7⁄ 𝑁 ⁄A f 26.2⁄ 𝑁 ⁄A f

If the actual slenderness ratio is less than the calculated value of 𝜆 then the column can be treated
as short. Otherwise, the column must be treated as slender and second order effects must be accounted
for in the design of the column.

EXAMPLE

Determine if the column in the braced frame shown in Fig. 2 is short or slender. The concrete strength
𝑓 25 N/mm2, and the ultimate axial load 1280 kN.

It can be seen that the column will have the highest slenderness ratio for bending about axes YY
where ℎ 300 mm and also the end restraints are the less stiff 300 500 beams.
 Fig. 2: Column end support details
𝑙 3.0 0.50 5.50 m

𝐼 400 300 /12 900 10 mm4

𝐼 300 500 /12 3125 10 mm4

𝐼 ⁄𝑙 900 10 ⁄2.5 10
𝑘 𝑘 0.115
∑ 2𝐼 ⁄𝑙 2 2 3125 10 ⁄4.0 10

From Table 1 and by interpolation; effective column height 𝑙 0.60 2.5 1.50 m.

Slenderness ratio 𝜆

⁄
Radius of gyration, 𝑖
.

Slenderness ratio 𝜆 𝑙 /𝑖 1.50 10 /86.6 17.32

For a braced column the minimum limiting value of 𝜆 will be given by

𝜆 26.2/ 𝑁 / 𝐴 𝑓

where:

𝑁 / 𝐴 𝑓 1280 10 / 400 300 0.85 25/1.5 0.75

thus

𝜆 26.2/√0.75 30.2 17.32

Hence, compared with the minimum limiting value of 𝜆 the column is short and second order moment
effects would not have to be taken into account.

4) Failure modes

Short columns usually fail by crushing but a slender column is liable to fail by buckling. The end
moments on a slender column cause it to deflect sideways and thus bring into play an additional
moment 𝑁𝑒 as illustrated in Fig. 3. The moment 𝑁𝑒 causes a further lateral deflection and if
the axial load (N) exceeds a critical value this deflection, and the additional moment become self-
propagating until the column buckles. Euler derived the critical load for a pin-ended strut as

𝜋 𝐸𝐼
𝑁
𝑙

Fig. 3: Slender column with lateral de lection

The crushing load 𝑁 of a truly axially loaded column may be taken as

𝑁 0.567𝑓 𝐴 0: 87𝐴 𝑓

where 𝐴 is the area of the concrete and 𝐴 is the area of the longitudinal steel.

Values of 𝑁 𝑁 and 𝑙 𝑖 have been calculated and plotted in Fig. 4 for a typical

column cross-section.

The ratio of 𝑁 /𝑁 in Fig. 4 determines the type of failure of the column. With 𝑙/𝑖 less than, say,
50 the load will probably cause crushing, 𝑁 is much less than 𝑁 , the load that causes buckling
– and therefore a buckling failure will not occur. This is not true with higher value of 𝑙/𝑖 and so a
buckling failure is possible, depending on such factors as the initial curvature of the column and the
actual eccentricity of the load. When 𝑙/𝑖 is greater than 110 then 𝑁 is less than 𝑁 and in this
case a buckling failure will occur for the column considered.
 Fig. 4: Column failure modes
The mode of failure of a column can be one of the following:

1. Material failure with negligible lateral deflection, which usually occurs with short columns
but can also occur when there are large end moments on a column with an intermediate
slenderness ratio.
2. Material failure intensified by the lateral deflection and the additional moment. This type of
failure is typical of intermediate columns.
3. Instability failure which occurs with slender columns and is liable to be preceded by excessive
deflections.

Short columns resisting moments and axial forces

The area of longitudinal steel for these columns is determined by:

1. using design charts or constructing M–N interaction diagrams

2. a solution of the basic design equations, or

3. an approximate method.

Design charts are usually used for columns having a rectangular or circular cross- section and a
symmetrical arrangement of reinforcement, but interaction diagrams can be constructed for any
arrangement of cross-section as illustrated in examples 4.10 and 4.11. The basic equations or the
approximate method can be used when an unsymmetrical arrangement of reinforcement is required,
or when the cross-section is non-rectangular.
Whichever design method is used, EC2 requires that geometric imperfections in the structure should
be taken into account. In the case of a single column it is assumed that the geometric imperfections
will result in the column being slightly out of plumb such that the axial force acting at each end of
the column will be eccentric to the other end by an amount taken as 𝑙 /400 , whre 𝑙 is the effective
height of the column. This will result in an additional design moment of 𝑁 𝑙 /400. Furthermore,
a column should not be designed for a moment less than 𝑁 e , where e is the larger of
ℎ/30 or 20 mm where h is the overall size of the column cross-section in the plane of bending.

DESIGN CHARTS AND INTERACTION DIAGRAMS

Bending Plus Axial Load at the Ultimate Limit State

The applied axial force may be tensile or compressive. In the analysis that follows, a compressive
force is considered. For a tensile load the same basic principles of equilibrium, compatibility of
strains, and stress–strain relationships would apply, but it would be necessary to change the sign of
the applied load (N) when we consider the equilibrium of forces on the cross-section. (The area of
concrete in compression has not been reduced to allow for the concrete displaced by the compression
steel. This could be taken into account by reducing the stress 𝑓 in the compression steel by an
amount equal to 0: 567𝑓 .)

Fig. 5 represents the cross-section of a member with typical strain and stress distributions for varying
positions of the neutral axis. The cross-section is subject to a moment M and an axial compressive
force N, and in the figure the direction of the moment is such as to cause compression on the upper
part of the section and tension on the lower part. For cases where there is tension in the section (Fig.
5a) the limiting concrete strain is taken as 0.0035 – the value used in the design and analysis of
sections for bending. However, for cases where there is no tension in the section (Fig. 5b) the limiting
strain is taken as a value of 0.00175 at the level of 1/2 of the depth of the section.

Let

𝐹 be the compressive force developed in the concrete and acting through the centroid of the
stress block

𝐹 be the compressive force in the reinforcement area 𝐴 and acting through its centroid

𝐹 be the tensile or compressive force in the reinforcement area 𝐴 and acting through its
centroid.
 Fig. 5: Bending plus axial load with varying position of the neutral axis
(i) Basic equations and design charts

The applied force (N) must be balanced by the forces developed within the cross-section, therefore
𝑁 𝐹 𝐹 𝐹

In this equation, 𝐹 will be negative whenever the position of the neutral axis is such that the
reinforcement 𝐴 is in tension, as in Fig. 5a. Substituting into this equation the terms for the stresses
and areas

𝑁 0.567𝑓 𝑏𝑠 𝑓 𝐴 𝑓𝐴
5

where 𝑓 is the compressive stress in reinforcement 𝐴 and 𝑓 is the tensile or compressive stress in
reinforcement 𝐴 .

The design moment 𝑀 must be balanced by the moment of resistance of the forces developed within
the cross-section. Hence, taking moments about the mid-depth of the section

ℎ 𝑠 ℎ ℎ
𝑀 𝐹 𝐹 𝑑 𝐹 𝑑
2 2 2 2

ℎ 𝑠 ℎ ℎ
𝑀 0.567𝑓 𝑏𝑠 𝑓 𝐴 𝑑 𝑓𝐴 𝑑
2 2 2 2 6
When the depth of neutral axis is such that 0.8𝑥 ℎ, as in part (b) of Fig. 5, then the whole concrete
section is subject to a uniform compressive stress of 0.567𝑓 . In this case, the concrete provides no
contribution to the moment of resistance and the first term on the right side of the (6 disappears.

For a symmetrical arrangement of reinforcement 𝐴 𝐴 𝐴 /2 𝑎𝑛𝑑 𝑑 ℎ 𝑑 , (5) and (6)

can be rewritten in the following form

𝑁 0.567𝑠 𝑓 𝐴 𝑓 𝐴
7
𝑏ℎ𝑓 ℎ 𝑓 𝑏ℎ 𝑓 𝑏ℎ

𝑀 0.567𝑠 𝑠 𝑓 𝐴 𝑑 𝑓 𝐴 𝑑
0.5 0.5 0.5 8
𝑏𝑑 𝑓 ℎ 2ℎ 𝑓 𝑏ℎ ℎ 𝑓 𝑏ℎ ℎ

In these equations the steel strains, and hence the stresses 𝑓 and 𝑓 , vary with the depth of the neutral
axis 𝑥 . Thus, 𝑁/𝑏ℎ𝑓 and 𝑀/𝑏ℎ 𝑓 can be calculated for specified ratios of 𝐴 /𝑏ℎ and 𝑥/ℎ so
that column design charts for a symmetrical arrangement of reinforcement such as the one shown in
Fig. 6 can be plotted.

Fig. 6: Typical column design chart

Or,

The design of a section subjected to bending plus axial load should also be as follows.
 𝑁 𝐹 𝐹 𝐹 0.567𝑓 𝑏𝑠 𝑓 𝐴 𝑓𝐴

ℎ 𝑠 ℎ ℎ
𝑀 𝐹 𝐹 𝑑 𝐹 𝑑
2 2 2 2

Where,

𝑁 design ultimate axial load

𝑀 design ultimate moment

𝑠 the depth of the stress block 0.8𝑥

𝐴 the area of longitudinal reinforcement in the more highly compressed face

𝐴 the area of reinforcement in the other face

𝑓 the stress in reinforcement 𝐴

𝑓 the stress in reinforcement 𝐴 , negative when tensile.

 Fig. 7: Rectangular column 𝑑 ⁄ℎ 0.20
These equations are not suitable for direct solution and the design of a column with symmetrical
reinforcement in each face is best carried out using design charts as illustrated in Fig. 7: .

Modes of Failure

The relative magnitude of the moment (M) and the axial load (N) governs whether the section will
fail in tension or in compression. With large effective eccentricity 𝑒 𝑀/𝑁 a tensile failure is
likely, but with a small eccentricity a compressive failure is more likely. The magnitude of the
eccentricity affects the position of the neutral axis and hence the strains and stresses in the
reinforcement.

Let

𝜀 be the compressive strain in reinforcement 𝐴

𝜀 be the tensile or compressive strain in reinforcement 𝐴

𝜀 be the tensile yield strain of steel as shown in the stress–strain curve of Fig. 5.
 Fig. 8: Short-term design stress–strain curve for reinforcement
From the linear strain distribution of Fig. 5(a)

𝑥 𝑑
ε 0.0035
𝑥
9
and
𝑑 x
ε 0.0035
𝑥

For values of 𝑥 greater than ℎ, when the neutral axis extends below the section, as shown in Fig. 5b,
the steel strains are given by the alternative expressions:

ε 0.00175 and ε 0.00175

. .

For pure axial compression, a uniform compressive concrete strain of 0.002 may be taken where
Grade 500 reinforcement is used, as in the UK, and for practical purposes this may be considered to
approximate to the yield strain for such steel. The steel stresses and strains are then related according
to the stress–strain curve of Fig. 8.

Consider the following modes of failure of the section as shown on the interaction diagram of Fig. 5.

(a) Tension failure, 𝜀 𝜀

This type of failure is associated with large eccentricities 𝑒 and small depths of neutral axis 𝑥 .
Failure begins with yielding of the tensile reinforcement, followed by crushing of the concrete as the
tensile strains rapidly increase.

(b) Balanced failure, 𝜀 𝜀 , point b on Fig. 9

When failure occurs with yielding of the tension steel and crushing of the concrete at the same instant
it is described as a ‘balanced’ failure. With 𝜀 𝜀 and from Equ. (9)
 𝑑
𝑥 𝑥 𝜀
1
0.0035

For example, substituting the values of 𝜀 0.00217 for grade 500 steel

𝑥 0.617𝑑

Equations (5) and (6) become

𝑁 𝐹 𝐹 𝐹
10
0.567𝑓 𝑏 0.8𝑥 𝑓 𝐴 0.87𝑓 𝐴
and

ℎ 0.8𝑥 ℎ ℎ
𝑀 𝐹 𝐹 𝑑 𝐹 𝑑
2 2 2 2

where

𝑓 0.87𝑓

Fig. 9: Bending plus axial load chart with modes of failure

At point b on the interaction diagram of Fig. 9, 𝑁 𝑁 ,𝑀 𝑀 𝑎𝑛𝑑 𝑓 0.87𝑓 . When the
design load 𝑁 𝑁 the section will fail in compression, while if 𝑁 𝑁 there will be an initial
tensile failure, with yielding of reinforcement 𝐴 .

(c) Compression failure

In this case 𝑥 𝑥 and 𝑁 𝑁 . The change in slope at point r in Fig. 9 occurs when

𝜀 𝜀

and from Equ. (9)

𝑥 0.0035𝑑 / 0.0035 𝜀
 2.63𝑑 for grade 500 steel

Point r will occur in the tension failure zone of the interaction diagram if 𝑥 𝑥 .

When 𝑥 𝑑

𝑓 0.87𝑓 and tensile

When 𝑥 𝑑

𝑓 0

When 𝑥 𝑑

𝑓 0.87𝑓 and compressive

When x becomes very large and the section approaches a state of uniform axial compression it can
be assumed that

𝜀 0.00217 𝜀 for grade 500 steel

At this stage, both layers of steel will have yielded and there will be zero moment of resistance with
a symmetrical section, so that

𝑁 0.567𝑓 𝑏ℎ 0.87𝑓 𝐴 𝐴

At the stage where the neutral axis coincides with the bottom of the section the strain diagram changes
from that shown in Fig. 5a to the alternative strain diagram shown in Fig. 5b. To calculate 𝑁 and 𝑀
at this stage, corresponding to point s in Fig. 9, Equs (5) and (6) should be used, taking the neutral
axis depth equal to the overall section depth, ℎ.

Such 𝑀– 𝑁 interaction diagrams can be constructed for any shape of cross-section which has an axis
of symmetry by applying the basic equilibrium and strain compatibility equations with the stress–
strain relations, as demonstrated in the following examples. These diagrams can be very useful for
design purposes.

EXAMPLE

M–N interaction diagram for a non-symmetrical section

Construct the interaction diagram for the section shown in Fig. 10 with 𝑓 25 N/mm2 and 𝑓
500 N/mm2. The bending causes maximum compression on the face adjacent to the steel area 𝐴 .
 Fig. 10: Non-symmetrical section M–N interaction example
For a symmetrical cross-section, taking moments about the centre-line of the concrete section will
give 𝑀 0 with 𝑁 𝑁 and both areas of steel at the yield stress. This is no longer true for
unsymmetrical steel areas as 𝐹 𝐹 at yield therefore, theoretically, moments should be calculated
about an axis referred to as the ‘plastic centroid’. The ultimate axial load 𝑁 acting through the plastic
centroid causes a uniform strain across the section with compression yielding of all the reinforcement,
and thus there is zero moment of resistance. With uniform strain the neutral-axis depth, 𝑥,is at infinity.

The location of the plastic centroid is determined by taking moments of all the stress resultants about
an arbitrary axis such as AA in Fig. 10 so that

The fundamental equations for calculating points on the interaction diagram with varying depths of
neutral axis are:

(i) Compatibility of strains (used in Table 2, columns 2 and 3):

𝑥 𝑑
𝜀 0.0035
𝑥 11
𝑑 x
𝜀 0.0035
𝑥
 Table 2: M–N interaction values for example Above
(4) (5) (6) (7)
(1) x (2) (3)
𝑓 𝑓 𝑀 𝑀
(mm) 𝜀 𝜀
𝑁/𝑚𝑚 𝑁/𝑚𝑚 𝑘𝑁 𝑘𝑁𝑚
𝑑 60 0 0.00217 0 0.87𝑓 -189 121
2.63𝑑
0.00217 0.00217 0.87𝑓 0.87𝑓 899 275
158
𝑥
0.617𝑑 0.00217 0.00217 0.87𝑓 0.87𝑓 1229 292
241
𝑑 390 0.00217 0 0.87𝑓 0 2248 192
ℎ 450 0.00217 0.00047 0.87𝑓 93.3 2580 146
∞ 0.00217 0.00217 0.87𝑓 0.87𝑓 3361 0

or when the neutral axis depth extends below the bottom of the section 𝑥 ℎ :

𝑥 𝑑
𝜀 0.00175
𝑥 0.5ℎ
𝑥 d
𝜀 0.00175
𝑥 0.5ℎ

(ii) Stress–strain relations for the steel (Table 2, columns 4 and 5):

ε 𝜀 0.00217 𝑓 0.87𝑓
12
ε 𝜀 𝑓 E ε
(iii) Equilibrium (Table 2, columns 6 and 7):

𝑁 𝐹 𝐹 𝐹

0.8𝑥 ℎ 𝑁 0.567𝑓 𝑏 0.8𝑥 𝑓 𝐴 𝑓𝐴

0.8𝑥 ℎ 𝑁 0.567𝑓 𝑏ℎ 𝑓 𝐴 𝑓𝐴

Taking moments about the plastic centroid

0.8𝑥 ℎ 𝑀 𝐹 𝑥̅ 0.8𝑥/2 𝐹 𝑥̅ 𝑑 𝐹 𝑑 𝑥̅

0.8𝑥 ℎ 𝑀 𝐹 𝑥̅ ℎ/2 𝐹 𝑥̅ 𝑑 𝐹 𝑑 𝑥̅

𝐹 is negative when 𝑓 is a tensile stress.

These equations have been applied to provide the values in Table 2 for a range of key values of x.
Then the M–N interaction diagram has been plotted in Fig. 11 from the values in Table 2 as a series
of straight lines. Of course, N and M could have been calculated for intermediate values of x to
provide a more accurate curve.

Fig. 11: M–N interaction diagram for a non-symmetrical section

EXAMPLE

Column design using design charts

Fig. 12 shows a frame of a heavily loaded industrial structure for which the centre columns along
line PQ are to be designed in this example. The frames at 4 m centres, are braced against lateral
forces, and support the following floor loads:

permanent action 𝑔 10 kN/m2 variable action 𝑞 15 kN/m2

Characteristic material strengths are 𝑓 25 N/mm2 for the concrete and 𝑓 500 N/mm2 for the
steel.

The effective height of the ground floor column has been estimated to be 2.34 m and for the first and
second floor columns to be 1.80 m.
 Fig. 12: Columns in an industrial structure
Maximum ultimate load at each floor 4.0 1.35𝑔 1.5𝑞 per metre length of beam

4 1.35 10 1.5 15 144 kN/m

Minimum ultimate load at each floor 4.0 1.35𝑔

4.0 1.35 10 54 kN per metre length of beam

Consider first the design of the centre column at the underside (u.s.) of the first floor. The critical
arrangement of load that will cause the maximum moment in the column is shown in Fig. 13a.

Column loads

Second and third floors 2 144 10⁄2 1440 kN

first floor 144 54 540

Column self-weight, say 2 14 28

𝑁 2008 kN
Similar arrangements of load will give the axial load in the column at the underside (u.s.) and top
side (t.s.) of each floor level and these values of 𝑁 are shown in Table 3.

(a) Critical loading arrangement for centre columns at 1st floor

(b) Substitute frame

(c) Fixed end moments

Fig. 13: Substitute frame for column design example

 Table 3
𝑁 𝑙 𝑁
𝑁 𝑀 𝑁 𝑀 𝐴𝑓 𝐴
Floor 400 𝑏ℎ𝑓
(kN) (kN m) (kN m) 𝑏ℎ 𝑓 𝑏ℎ𝑓 (mm2)
(kNm)
3rd u.s. 540 82.6 3.30 85.90 0.18 0.07 0 240
2nd t.s. 734 68.4 3.30 71.70 0.24 0.06 0 240
+ 540
2nd u.s. 1274 68.4 6.61 75.01 0.06 0.06 0 240
1st t.s. 1468 68.4 6.61 75.01 0.06 0.06 0.10 600
+ 540
1st t.s. 2008 68.4 11.75 80.15 0.07 0.07 0.32 1920

Member stiffnesses are

therefore

∑𝑘 0.71 1.07 2 0.53 10 2.84 10

and

.
distribution factor for the column ∑
0.19
.

Fixed end moments at B are

Thus

column moment 𝑀 0.19 432 72 68.4kNm

Design moment allowing for geometric imperfections

 𝑁 𝑙
𝑀 𝑀
400
.
For underside of first floor, 𝑀 68.4 68.4 11.75 80.15 kNm

.
For topside of first floor, 𝑀 68.4 68.4 6.61 75.01 kNm

The minimum moment in both cases is 𝑁 𝑒 where 𝑒 20 mm

400/30 13: 3 mm which in neither case is critical.

At the 3rd floor

∑𝑘 0.71 1.07 0.53 10 2.31 10

and

.
column moment 𝑀 432 72 82.6 kNm
.

.
𝑀 𝑀 82.6 82.6 3.30 85.90 kNm 𝑁 𝑒

The areas of reinforcement in Table 3 are determined by using the design chart of Fig. 7. Sections
through the column are shown in figure 9.11.

Fig. 14: Column sections in design example

Note: the link spacing is reduced to 0.60 x these values for 400mm above and below each floor level
and at laps below 1st floor level

Cover for the reinforcement is taken as 50 mm and 𝑑 /ℎ 80/400 0.2. The minimum area of
reinforcement allowed in the section is given by:
𝐴 0.002𝑏ℎ 0.002 300 400 240 mm2

and the maximum area is

𝐴 0.08 300 400 9600 mm2

and the reinforcement provided is within these limits.

Although EC2 permits the use of 12 mm main steel, 16 mm bars have been used to ensure adequate
rigidity of the reinforcing cage. A smaller column section could have been used above the first floor
but this would have involved changes in formwork and possibly also increased areas of
reinforcement.
Slabs

Reinforced concrete slabs are used in floors, roofs and walls of buildings and as the deck of bridges.
The floor system of a structure can take many forms such as in situ solid slabs, ribbed slabs or precast
units. Slabs may span in one direction or in two directions and they may be supported on monolithic
concrete beams, steel beams, walls or directly by the structure’s columns.

Continuous slabs should in principle be designed to withstand the most unfavourable arrangements
of loads, in the same manner as beams. As for beams, bending moment coefficients, as given in Table
4, may be used for one-way spanning slabs. If these coefficients are used the reinforcement must be
of ductility class B or C and the neutral-axis depth, 𝑥, should be no greater than 0.25 of the effective
depth such that the lever arm, 𝑧 𝑑 0.8𝑥/2 , is not less than 0.9𝑑 to allow for moment
redistribution incorporated in the values given (which may be up to 20 per cent). In addition, as for
beams, Table 4 should only be used when there are at least three spans that do not differ in length by
more than 15 per cent, and 𝑄 should be less than or equal to 1.25𝐺 and also less than 5 kN/m2.

The moments in slabs spanning in two directions can also be determined using tabulated coefficients.
Slabs which are not rectangular in plan or which support an irregular loading arrangement may be
analysed by techniques such as the yield line method or the Hilleborg strip method.

Concrete slabs are defined as members where the breadth is not less than 5 times the overall depth
and behave primarily as flexural members with the design similar to that for beams, although in
general it is somewhat simpler because:

1. the breadth of the slab is already fixed and a unit breadth of 1 m is used in the calculations;

2. the shear stresses are usually low in a slab except when there are heavy concentrated loads; and

3. compression reinforcement is seldom required.

Minimum thicknesses and axis distances for fire resistance are given in Table 6 but deflection
requirements will usually dominate.

Table 4: Ultimate bending moment and shear force coef icients in one-way spanning slabs
End support condition
Pinned Continuous
Near At first At middle At
Outer End
middle of End span interior of interior interior
suppor support
end span suppor spans supports
Moment 0 0.086Fl - 0.07 - 0.063Fl -
0.04Fl 0.086Fl 0.063Fl
 Shear 0.4F - - -
0.46F 0.6F 0.5F
𝑙 effective span.
Area of each bay 30 m2. (A bay is a strip of slab across the structure between adjacent rows of
columns.)
𝐹 total ultimate load 1.35𝐺 1.50𝑄 .

Span–effective depth ratios

The appearance and function of a reinforced concrete beam or slab may be impaired if the deflection
under serviceability loading is excessive. Deflections can be better controlled by placing a limit on
the ratio of the span to the effective depth of the beam or slab. EC2 specifies equations to calculate
basic span–effective depth ratios, to control deflections to a maximum of span/250. Some typical
values are given in Table 5 for rectangular sections of class C30/37 concrete and for grade 500 steel.
The ratios can also be used for flanged sections except where the ratio of the width of flange to the
width of web exceeds 3 when the basic values should be multiplied by 0.8. For two-way spanning
slabs, the check for the basic span–effective depth ratio should be based on the shorter span whereas
for flat slabs calculations should be based on the longer span.

The two columns given in Table 5 correspond to levels of concrete stress under serviceability
conditions: highly stressed when the steel ratio 𝜌 exceeds 1.5 per cent and lightly stressed when 𝜌
equals 0.5 per cent. 𝜌 is given by 100𝐴 , ⁄𝑏𝑑 where 𝐴 , is the area of tension reinforcement
required in the section. Interpolation between the values of 𝜌 indicated is permissible. In the case of
slabs, it is reasonable to assume that they are lightly stressed.

Since the value of allowable span–effective depth ratio is affected by both reinforcement ratio and
concrete strength it may be more convenient to use the chart in Fig. 15 which is for a simply supported
span with no compression steel together with a modification factor K (as shown in Table 5) according
to member type. This approach is based on the same basic equations and offers greater flexibility
than reliance placed on tabulated values.

The basic ratios are modified in particular cases as follows:

(a). For spans longer than 7 m (except flat slabs) and where it is necessary to limit deflections to
ensure that finishes, such as partitions, are not damaged, the basic values should be multiplied
by 7/𝑠𝑝𝑎𝑛.
(b). For flat slabs with spans in excess of 8.5 m, similarly multiply the basic ratios by 8.5/𝑠𝑝𝑎𝑛.
 Table 5: Basic span–effective depth ratios 𝑓 500 N/mm2, C30/37 concrete
Basic span–effective depth ratio
Concrete lightly
Structural system Factor for structural Concrete highly
stressed
system K stressed 𝜌 1.5%
𝜌 0.5%
Simply supported beam or
1 one/two-way spanning simply 1.0 14 20
supported slab
End span of continuous beam or
one-way continuous slab or two-
2 1.3 18 26
way slab continuous over one long
side
Interior span of continuous beam or
3 1.5 20 30
one-way or two-way spanning slab
Slab on columns without beams
4 1.2 17 24
(flat slab) based on longer span
5 Cantilever 0.4 6 8

Fig. 15: Graph of basic span–effective depth ratios for different classes of concrete
(c). For characteristic steel strengths other than 500 N/mm2, multiply the basic ratios by 500
𝑓 .
(d). Where more tension reinforcement is provided 𝐴 , than that calculated 𝐴 , at the
ultimate limit state, multiply the basic ratios by
,
∙ 1.5
,
These basic ratios assume a steel working stress of 𝑓 310 N/mm2 where 𝑓 500 N/mm2. The
absolute value of span–effective depth ratio must not be greater than 40𝐾 (K from Table 5).

Table 6: Minimum dimensions and axis distance for RC slabs for ire resistance

Standard Minimum dimensions (mm)

fire Two-way spanning Ribs in two-way spanning ribbed
resistance 𝑙 ⁄𝑙 1.5 1.5 𝑙 ⁄𝑙 2 slab
REI 60 ℎ 80 80 80 𝑏 100 120 200
𝑎 20 10 15 𝑎 25 15 10
REI 90 ℎ 100 100 100 𝑏 120 160 200
𝑎 30 15 20 𝑎 35 25 15
REI 120 ℎ 120 120 120 𝑏 160 190 200
𝑎 40 20 25 𝑎 45 40 30
REI 240 ℎ 175 175 175 𝑏 450 700 -
𝑎 65 40 50 𝑎 70 60 -
Notes:
1. The slab thickness ℎ is the sum of the slab thickness and the thickness of any non-
combustible flooring
2. In two-way slabs the axis refers to the lower layer of reinforcement
3. The term ‘two-way slabs’ relates to slabs supported at all four edges; if this is not the case they
should be treated as one-way spanning
4. For two-way ribbed slabs:
(a). The axis distance measured to the lateral surface of the rib should be at least 𝑎 10
mm
(b). The values apply where there is predominantly uniformly distributed loading
(c). There should be at least one restrained edge
(d). The top reinforcement should be placed in the upper half of the flange

EXAMPLE

Span–effective depth ratio

A rectangular continuous beam of class C25/30 concrete spans 10 m. If the breadth is 300 mm, check
the acceptability of an effective depth of 600 mm when high yield reinforcement, 𝑓 500 N/mm2,
is used. At the ultimate limit state, it is determined that 1250 mm2 of tension steel is needed and 3
No. 25 mm diameter reinforcing bars (𝐴 , 1470 mm2) are actually provided in an interior span.
 𝜌 100𝐴 , ⁄𝑏𝑑

100 1250 ⁄ 300 600 0.7𝑝𝑒𝑟 𝑐𝑒𝑛𝑡.

From Table 5, for an interior span 𝐾 1.5

Basic span–effective depth ratio (Fig. 15) = 16.3

Therefore, for an interior span, basic span–effective depth ratio 1.5 16.3 24.4

To avoid damage to finishes for span greater than 7 m:

Modified ratio 24.4 17.1

Modification for steel area provided:

Modified ratio 17.1 20.1

Span-effective depth ratio provided =16.7

which is less than the allowable upper limit, thus deflection requirements are likely to be satisfied.

As a slab is usually a slender member, the restrictions on the span–depth ratio become more important
and this can often control the depth of slab required. In terms of the span–effective depth ratio, the
depth of slab is given by

span
minimum effective depth
basic ratio correction factors

The correction factors account for slab type and support conditions as well as cases of spans greater
than 7 metres and for flat slabs greater than 8.5 metres. The basic ratio may also be corrected to
account for grades of steel other than grade 500 and for when more reinforcement is provided than
that required for design at the ultimate limit state.

The concrete stress may be reduced by providing an area of tension reinforcement greater than that
required to resist the design moment up to a maximum of 1.5 that required.

In the case of two-way spanning slabs, the check on the span–effective depth ratio should be based
on the shorter span length. This does not apply to flat slabs where the longer span should be checked.

Solid slabs spanning in one direction

The slabs are designed as if they consist of a series of beams of 1 m breadth. The main steel is in the
direction of the span and secondary or distribution steel is required in the transverse direction. The
main steel should form the outer layer of reinforcement to give it the maximum lever arm.
The calculations for bending reinforcement follow a similar procedure to that used in beam design.
The lever arm curve of Fig. 16 is used to determine the lever arm (z) and the area of tension
reinforcement is then given by

𝑀
𝐴
0.87𝑓 𝑧

For solid slabs spanning one-way the simplified rules for curtailing bars as shown in Fig. 17 may be
used provided that the loads are uniformly distributed. With a continuous slab it is also necessary that
the spans are approximately equal.

𝑘 0.16
0.05 0.06 0.07 0.08 0.09 0.10 0.11 0.12 0.13 0.14 0.15 0.16
𝑀⁄𝑏𝑑 𝑓 7
0.95 0.93 0.93 0.92 0.91 0.90 0.89 0.88 0.86 0.85 0.84 0.83 0.82
𝑙 𝑧⁄𝑑
4 4 4 4 3 2 1 0 8 6 3 0 0

The percentage values on the K axis mark the limits for singly reinforced sections with moment
redistribution applied

Fig. 16: Lever-arm curve

 Fig. 17: Simpli ied rules for curtailment of bars in slab spanning in one direction
Single-span solid slabs

The basic span–effective depth ratio for this type of slab is 20:1 on the basis that it is ‘lightly stressed’
and that grade 500 steel is used in the design. For a start-point in design a value above this can usually
be estimated (unless the slab is known to be heavily loaded) and subsequently checked once the main
tension reinforcement has been designed.

The effective span of the slab may be taken as the clear distance between the face of the supports
plus a distance at both ends taken as the lesser of (a) the distance from the face of the support to its
centreline and (b) one-half of the overall depth of the slab.

Shear in slabs

Experimental work has indicated that, compared with beams, shallow slabs fail at slightly higher
shear stresses and this is incorporated into the values of the ultimate concrete shear resistance, 𝑉 , ,
as given by Equs (13) and (14). Calculations are usually based on a strip of slab 1 m wide.

The shear capacity of the concrete, 𝑉 , , in such situations is given by an empirical expression:

/
𝑉 , 0.12𝑘 100𝜌 𝑓 𝑏 𝑑 13
with a minimum value of:

/
𝑉 , 0.035𝑘 /
𝑓 𝑏 𝑑 14

where:

𝑉 , the design shear resistance of the section without shear reinforcement

𝑘 1 2.0 with 𝑑 expressed in mm

𝜌 0.02
 𝐴 the area of tensile reinforcement that extends beyond the section being considered by
at least a full anchorage length plus one effective depth (d)

𝑏 the smallest width of the section in the tensile area (mm)

Since shear stresses in slabs subject to uniformly distributed loads are generally small, shear
reinforcement will seldom be required and it would be usual to design the slab such that the design
ultimate shear force, 𝑉 , is less than the shear strength of the unreinforced section, 𝑉 , . In this case
it is not necessary to provide any shear reinforcement. This can conveniently be checked using Table
7 which has been derived from Equs (13) and (14) for class C30/37 concrete on the basis that the
allowable shear stress in the unreinforced slab is given by

,
𝑣 ,

In this case, the applied ultimate shear stress

𝑣 𝑣 ,

Table 7 also clearly illustrates the effect of increasing slab thickness on the depth related factor k,
used in calculating 𝑉 , (see Eqs (13)), as noted above. Where different concrete strengths are used,
the values in Table 7 may be modified by the factors in Table 7 provided 𝜌 0.4%.

As for beams, the section should also be checked to ensure that 𝑉 does not exceed the maximum
permissible shear force, 𝑉 ;, . Practical difficulties concerned with bending and fixing shear
reinforcement make it unlikely that shear reinforcement could be provided in slabs less than 200 mm
thick.

Table 7: Shear resistance of slabs without shear reinforcement 𝑉 , N/mm2 Class C30/37
concrete

𝐴 Effective depth, 𝑑 (mm)

𝜌
𝑏 𝑑 200 225 250 300 350 400 500 600 750
0.25% 0.54 0.52 0.5 0.47 0.45 0.43 0.40 0.38 0.36
0.50% 0.59 0.57 0.56 0.54 0.52 0.51 0.48 0.47 0.45
0.75% 0.68 0.66 0.64 0.62 0.59 0.58 0.55 0.53 0.51
1.00% 0.75 0.72 0.71 0.68 0.65 0.64 0.61 0.59 0.57
1.25% 0.80 0.78 0.76 0.73 0.71 0.69 0.66 0.63 0.61
1.50% 0.85 0.83 0.81 0.78 0.75 0.73 0.70 0.67 0.65
2.00% 0.94 0.91 0.89 0.85 0.82 0.80 0.77 0.74 0.71
k 2.000 1.943 1.894 1.816 1.756 1.707 1.632 1.577 1.516
 Table 8: Concrete strength modi ication factor
𝑓 N/mm 25 30 35 40 45 50
Modification factor 0.94 1.00 1.05 1.10 114 1.19

EXAMPLE

Design of a simply supported slab

The slab shown in Fig. 18 is to be designed to carry a variable load of 3.0 kN/m2 plus floor finishes
and ceiling loads of 1.0 kN/m2. The characteristic material strengths are 𝑓 25 N/mm2 and 𝑓
500N/mm2. Basic span–effective depth ratio 19 for a lightly stressed slab from Fig. 15 for class
C25/30 concrete and 𝜌 0.5%.

Fig. 18: Simply supported slab example

For simplicity, take the effective span to be 4.5 m between centrelines of supports.

(1) First design solution

Estimate of slab depth

Try a basic span-depth ratio of 27 (approx. 40% above value from Fig. 15):

minimum effective depth

4500 167
27 c. f c. f

As high yield steel is being used and the span is less than 7 m the correction factors can be taken as
unity. Try an effective depth of 170 mm. For a class XC-1 exposure the cover 25 mm. Allowing,
say, 5 mm as half the bar diameter of the reinforcing bar:
overall depth of slab 170 25 5 200 mm Slab loading

self-weight of slab 200 25 10 5: 0 kN/m2

total permanent load 1.0 5.0 6.0 kN/m2

For a 1 m width of slab:

ultimate load 1.35𝑔 1.5𝑞 5

1.35 6.0 1.5 3.0 4.5 56.7𝑘𝑁

𝑀 56.7 4.5/8 31.9 kNm

Bending reinforcement

𝑀 3.9 10
0.044
𝑏𝑑 𝑓 1000 170 25

From the lever-arm curve of Fig. 16, 𝑙 0.95. Therefore, adopt upper limit of 0.95 and lever-arm
𝑧 𝑙 0.95 170 161 mm:

.
𝐴 455 mm2/m
. .

Provide H10 bars at 150 mm centres, 𝐴 523 mm2/m.

Check span–effective depth ratio

,
𝜌 0.268% 0.13% minimum requirement)

From Fig. 15, this corresponds to a basic span–effective depth ratio of 32. The actual ratio
4500/170 26.5; hence the chosen effective depth is acceptable.

Shear

At the face of the support

. . . .
Shear 𝑉 26.5 kN
.

𝜌 0.31

𝑉 , 𝑣 , 𝑏𝑑 where 𝑣 , from Table 7 0.55 (note: no concrete strength adjustment since 𝜌

0.4% . Thus:

𝑉 , 0.55 1000 170 93.5 kN

as 𝑉 is less than 𝑉 , then no shear reinforcement is required.

Continuous solid slab spanning in one direction

For a continuous slab, bottom reinforcement is required within the span and top reinforcement over
the supports. The effective span is the distance between the centreline of the supports and the basic
span–effective depth ratio of an interior span is 30.0 for ‘lightly stressed’ where grade 500 steel and
class C30/37 concrete are used. The corresponding limit for an end span is 26.0. The bending moment
and shear force coefficients given in Table 4 may be used.

EXAMPLE

Design of a continuous solid slab

The four-span slab shown in figure 8.4 supports a variable load of 3.0 kN/m2 plus floor finishes and
a ceiling load of 1.0 kN/m2. The characteristic material strengths are 𝑓 25 N/mm2 and 𝑓 500
N/mm2.

Fig. 19: Continuous slab

Estimate of slab depth

As the end span is more critical than the interior spans, try a basic span–effective depth ratio 30 per
cent above the end-span limit of 26.0 (i.e., 33.0):

minimum effective depth

.. ..

. .
As high yield steel is being used and the span is less than 7 m the correction factor can be taken as
unity. Try an effective depth of 140 mm. For a class XC-1 exposure the cover 25 mm. Allowing,
say, 5 mm as half the bar diameter of the reinforcing bar:

overall depth of slab 140 25 5 170 mm

Slab loading

self-weight of slab 170 25 10 4.25 kN/m2

total permanent load 1.0 4.25 5.25 kN/m2

For a 1 m width of slab

ultimate load, 𝐹 1.35𝑔 1.5𝑞 4.5

1.35 5.25 1.5 3.0 4.5 52.14 kN

Using the coefficients of Table 4, assuming the end support is pinned, the moment at the middle of
the end span is given by

𝑀 0.086𝐹𝑙 0.086 52.14 4.5 ¼ 20.18 kN m

Bending reinforcement

.
0.0412

From the lever-arm curve, 𝑙 0.96. Therefore, lever-arm 𝑧 𝑙 𝑑 0.95 140 133 mm:

.
𝐴 349 mm2/m
. .

Provide H10 bars at 200 mm centres, 𝐴 393 mm2/m.

Check span–effective depth ratio

100𝐴 , 100 349

0.249
𝑏𝑑 1000 140

From Fig. 15 this corresponds to a basic span–effective depth ratio in excess of 32 1.3 (for an end
span) 41. The actual ratio 4500/140 32.1; hence the chosen effective depth is acceptable.

Similar calculations for the supports and the interior span give the steel areas shown in figure 8.5.
 Fig. 20: Reinforcement in a continuous slab
At the end supports there is a monolithic connection between the slab and the beam, therefore top
steel should be provided to resist any negative moment. The moment to be designed for is a minimum
of 25 per cent of the span moment, that is 5.1 kN m. In fact, to provide a minimum of 0.13 per cent
of steel, H10 bars at 400 mm centres have been specified.

Transverse reinforcement 0.0013𝑏𝑑

0.0013 1000 140 182 mm2/m

Provide H10 at 400 mm centres top and bottom, wherever there is main reinforcement (196 mm2/m).