Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

Unit 8 – Slender reinforced concrete columns.

(Study time allocation - 15 hours)

Introduction.

In this unit, the effect of geometry and external actions on elements is considered.

How do Columns fail.

Consider a column subject to an increasing axial load. Eventually the compressive

strength of the column is exceeded and the column fails.

Now consider a column which is thinner in one direction and again subject to increasing
axial load.

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.1
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

When the load reaches a certain value, the column begins to bend sideways about the
weaker axis. The column is said to have buckled. Increasing the load causes the column
to deflect further until bending failure occurs.

The effect of the sideways deflection is to increase the moment at any section by an
amount equal to the axial load x deflection at the section :-

Madditional = Na

In most practical situations the effect of deflections are small so can be ignored. Where it
is significant, the element is described as being slender.

When we consider a perfect column at first sight it ought not to buckle when subject to
vertical loads. However Euler, a theoretical engineer from Switzerland, discovered that
there is a critical load Ncr, now called the Euler load when a column is in a state of
neutral equilibrium. When the load is less the column is stable, when the load is above
the critical load the column becomes unstable. The formula for the critical load is :-

π 2 EI
N cr =
L2

Thus a column is more likely to buckle if the length is increased or the flexural rigidity
(EI) is reduced. π2 indicates the deflected shape will be sinusoidal.

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.2
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

Slenderness effect on structures.

In structures slenderness effects mainly occur as below:-

In beams which are

narrow compared to the
span or depth

In columns in In walls about

both directions the minor axis

Classification of structures.

The deflection of elements in a structure (particularly columns) depends on the overall

layout and detailing of the structure. To classify the full range of buildings is impractical
so the code has chosen to divide or classify structures into braced or unbraced
structures.

Braced structure. This is a structure which contains bracing elements. These are
usually vertical elements often walls which are so stiff relative to other vertical elements,
that they may be assumed to attract almost all the horizontal forces. With braced
structures, sidesway of the structure is insignificant. A braced structure is defined as one
in which the bracing element(s) attract and transmit to the foundations, at least 90% of all
horizontal forces applied to the structure.

Unbraced structure. This type of structure relies on the frame to transmit horizontal
loads to the foundations. In unbraced structures sidesway is significant, meaning that
lateral displacement of the ends of the columns increases the critical design moments by
more than 10% above that calculated by ignoring the displacements.

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.3
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

Importance of braced and unbraced structures.

The classification of a structure determines the mode of deflection of the structure which
gives very different bending moments in the columns. In an unbraced frame all columns
in a particular storey height are subject to nearly the same deflection.

a a
Bending moment Unbraced Bending moment
due to deflection due to deflection

Effective length of elements (braced and unbraced columns).

The effective length of a column is defined as the equivalent length of a pin-ended strut
subject to the same load and of identical cross section.

Effective length l0 = βl where l is the clear height between end restraints. The effective
length depends on the deflected shape of the element and the end restraints and is the
distance between points of contraflexure.

Braced structure. If the end supports are fixed, two points of contraflexure exist
between the supports and the theoretical effective length is 0.5l0 (β=0.5). If both ends are
pinned β =1.0

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.4
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

This section Pin

N behaves as a N ended
pin ended strut
strut

l Points of l0
contraflexure l0

N N
Braced – fixed ends Braced – pinned ends

Unbraced structure. Only one point of contraflexure exists between the element ends.
If both ends are fully restrained, β = 1, if the top end free, β =2.0. In the example below,
fixity has been judged as being intermediate between fully fixed and pinned.

Virtual point of
contraflexure

N This section
will behave
as a pin
sidesway ended strut

Point of
contraflexure
N

Un braced N

Code requirements – effective length.

Clause 5.8.3.2 (2) – (7) gives guidance on determining the effective length of ‘isolated’
members. These are members that for the purposes of design need to be designed in
isolation.

(2) For a general definition of the effective length, see 5.8.1. Examples of effective length
for isolated members with constant cross section are given in Figure 5.7.

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.5
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

(3) For compression members in regular frames, the slenderness criterion (see 5.8.3.1)
should be checked with an effective length l0 determined in the following way:

Braced members (see Figure 5.7 (f)):

 k1  k2 
l 0 = 0.5l 1 + 1 +  (5.15)
 0.45 + k1  0.45 + k 2 

Unbraced members (see Figure 5.7 (g)):

 kk  k   k 
l 0 = l × max  1 + 10 1 2 ; 1 + 1  × 1 + 2  (5.16)
 k1 + k 2  1 + k1   1 + k 2 

where:

k1, k2 are the relative flexibilities of rotational restraints at ends 1 and 2 respectively:
k = (θ / M)· (EΙ / l).
θ is the rotation of restraining members under bending moment M;
see also Figure 5.7 (f) and (g)
EΙ is the bending stiffness of compression member, see also 5.8.3.2 (4) and (5)
l is the clear height of compression member between end restraints

Note: k = 0 is the theoretical limit for rigid rotational restraint, and k = ∞ represents the limit for no
restraint at all. Since fully rigid restraint is rare in practise, a minimum value of 0,1 is recommended for k1
and k2. [The concrete centre recommend that when θ = 0 then k =0, but a rigid rotational restraint is
unlikely in practice so take k = 0.1.]

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.6
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

(4) If an adjacent compression member (column) in a node is likely to contribute to the

rotation at buckling, then (EΙ/l) in the definition of k should be replaced by [(EΙ / l)a+(EΙ /
l)b], a and b representing the compression member (column) above and below the node.

(5) In the definition of effective lengths, the stiffness of restraining members should
include the effect of cracking, unless they can be shown to be uncracked in ULS.

(6) For other cases than those in (2) and (3), e.g. members with varying normal force
and/or cross section, the criterion in 5.8.3.1 should be checked with an effective length
based on the buckling load (calculated e.g. by a numerical method):

l 0 = π EI / N B (5.17)

where:
EI is a representative bending stiffness
NB is buckling load expressed in terms of this EI
(in Expression (5.14), i should also correspond to this EI)

(7) The restraining effect of transverse walls may be allowed for in the calculation of the
effective length of walls by the factor β given in 12.6.5.1. In Expression (12.9) and Table
12.1, lw is then substituted by l0 determined according to 5.8.3.2.

Note. β can be obtained by interpreting between equations (5.15) and (5.16)

Full detailed analysis is necessary in order to obtain θ, but when this is not known k1 and
k2 can be determined from the ratio of column bending stiffnesses to beam - slab bending
stiffnesses but taking only 50% of beam stiffness to allow for cracking. Only columns
fully framed into the column in the direction of bending should be considered.

SC1
k1 =
S1 S2 S1 + S 2

SC1 l

SC1 + SC2
k2 =
S3 S4 S3 + S 4

SC2 l

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.7
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

Simplified method of determining effective length.

A simplified method of determining the effective length of columns using factors is given
in “How to design concrete structures using Eurocode 2 : Columns – Moss R and
Brooker O, The Concrete Centre 2006. Conservative effective length factors are given in
the table below where l0 = l x factor.

End condition at End condition at bottom

top 1 2 3
1 0.75 0.8 0.9
2 0.8 0.85 0.95
3 0.90 0.95 1.0
Key.
Condition 1. Column connected monolithically to beams on each side that are at least as deep as the
overall depth of the column in the plane considered.
When the column is connected to a foundation, this should be designed to carry moment in order to
satisfy this condition.
Condition 2. Column connected monolithically to beams on each side that are shallower than the
overall depth of the column in the plane considered by generally not less than half the column depth.
Condition 3. Column connected to members that do not provide more than nominal restraint to
rotation.
Table taken from Institution of Structural Engineers, Manual for the design of concrete structures to
Eurocode 2, IStructE 2006.

Slenderness Ratio.

Slenderness ratio λ = effective length / radius of gyration.

Radius of gyration.

I = (I/A)0.5

Example. If the effective length of a column is 2.63m and it is 350mm square, what is
its slenderness ratio?

I = bd3/12 = 350 x 3503/12 = 1250.5 x 106 mm4

A = 3502 = 122.5 x 103 mm2

Radius of gyration, i = (I/A)0.5 = (1250.5 x 106/ 122.5 x 103)0.5

= 101.04 mm

Effective length = 2630mm

Slenderness ratio = λ = 2630/101.04 = 26.03

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.8
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

Slenderness ratio – Background.

In the example above, a 350mm square column of length 2.63m has a slenderness ratio λ
of 26. To examine the effect of slenderness consider this 350mm square column but of
different lengths. Column is simply supported top and bottom

Column length Slenderness ratio λ

0.1m 0.99 (say 1)
1.0m 9.9 (say 10)
2.5m 24.7 (say 25)
9m 89.1 (say 90)
20m 197.9 (say 200)

Slenderness ratio 1.

Apply an axial load Ned at an eccentricity of eo. Moment on column = Ned eo. Actually
there will be a very small additional moment due to the deflection induced in the column
by Ned but this is almost 0 in this case. The load carrying capacity of the column is
unaffected by the eccentric load and resultant deflection. The axial load and moment are
plotted on the figure below.

Buckling deflection.

For a range of slenderness ratios (column lengths) the buckling deflection is included as a
factor in the load carrying capacity of the columns. The buckling deflection produces an
additional moment in the column, NEde2 which reduces the load carrying capacity of the
column. The greater the slenderness the greater the reduction.

NEd

e2 e0

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.9
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

Slenderness ratio 10.

Again apply an axial load Ned at an eccentricity of eo.

Moment on column = Ned eo + NEde2.
A small barely measurable additional moment will be induced in the column due to the
deflection in the column. The load carrying capacity of the column is reduced marginally
due to this additional deflection but from the practical point of view this is insignificant.
The point is plotted on the figure.

Slenderness ratio 25.

Again apply an axial load Ned at an eccentricity of eo.

Moment on column = Ned eo + NEde2.
A small but beasureable additional moment will be induced in the column due to the
deflection caused by the eccentric Ned. Still from the practical point of view the
reduction in the load carrying capacity of the column is small and can usually be ignored.
See figure.

Slenderness ratio 90.

Again apply an axial load Ned at an eccentricity of eo.

Moment on column = Ned eo+ NEde2. A significant additional moment will be
induced in the column due to the deflection caused by the eccentric Ned because the
column is so long. This moment will reduce the load carrying capacity of the column and
must be considered. See figure.

Slenderness ratio 200.

Again apply an axial load Ned at an eccentricity of eo.

Moment on column = Ned eo+ NEde2. In this case the behaviour is difficult to
predict and slenderness effects are significant.

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.10
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

λ=1
λ=10
λ=25
λ=90
Axial load NEd

λ=200

Moment MEd

Design Consideration.

Clearly design decisions need to be made to ensure that the slenderness or second order
effects as they are sometimes called are included when necessary but not when
unnecessary. According to the figure above, the effects of slenderness ratio λ on axial
load NEd is limited if λ < 25 but becomes important if λ > 90. The challenge for
designers is to decide at what value of λ should slenderness effects be included in the
design process.

When to ignore slenderness effects.

The code states that second order slenderness effects can be ignored if they are less than
10% of the first order effects or as an alternative, for an isolated member if it slenderness
ratio λ is less than λlim . Details on this requirement can be found in Clause 5.8.3.1 which
follows :-

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.11
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

5.8.3.1 Slenderness criterion for isolated members

(1) As an alternative to 5.8.2 (6), second order effects may be ignored if the slenderness λ
(as defined in 5.8.3.2) is below a certain value λlim.

Note: The value of λlim for use in a Country may be found in its National Annex. The recommended value
follows from:

λlim = 20ABC/n0.5 (5.13N)

where:
A = 1 / (1+0,2ϕef) (if ϕef is not known, A = 0,7 may be used)
B = (1+ 2ω)0.5 (if ω is not known, B = 1,1 may be used)
C = 1,7 - rm (if rm is not known, C = 0,7 may be used)
ϕef effective creep ratio; see 5.8.4 [Dealt with in unit 12].
ω = Asfyd / (Acfcd); mechanical reinforcement ratio;
As is the total area of longitudinal reinforcement
n = NEd / (Acfcd); relative normal force
rm = M01/M02; moment ratio
M01, M02 are the first order end moments, │M02│ ≥ │M01│

If the end moments M01 and M02 give tension on the same side, rm should be taken positive (i.e. C ≤ 1,7),
otherwise negative (i.e. C > 1,7).

In the following cases, rm should be taken as 1,0 (i.e. C = 0,7):

- for braced members in which the first order moments arise only from or predominantly due to
imperfections or transverse loading
- for unbraced members in general

(2) In cases with biaxial bending, the slenderness criterion may be checked separately for
each direction. Depending on the outcome of this check, second order effects (a) may be
ignored in both directions, (b) should be taken into account in one direction, or (c) should
be taken into account in both directions. The section on bi-axial bending deals with this.

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.12
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

Design guidance. Determining C and rm – Concrete centre Concise Eurocode.

If rm is not known, C may be taken as 2.7 for columns in double curvature in braced
structures through to 0.7 for constant moment as indicated in the Figure below. For
unbraced structures, C = 0.7.

x x x

x x

C 0.7 1.7 2.7

rm 1.0 0.0 -1.0

The inclusion of slenderness effects.

If the above checks (i.e. λlim ≥ λ) indicate that slenderness effects cannot be ignored, then
slenderness needs to be designed for. In these cases, the effects of cracking, creep, non-
linear material properties and geometric imperfections are also to be considered. This
normally means that in addition to slenderness effects accounted for through the
eccentricity e2, an additional eccentricity ei of the axial load Ned is included which allows
for the other effects. The code defines ei as

ei = 0.5θil0

Where θi is the inclination used to represent imperfections.

Analysis of slender columns.

Three approaches are recommended in the code.

1. Undertake a rigorous non-linear analysis of the structure. This is beyond the

scope of this module but is undertaken elsewhere in this course.
2. Use a simplified method based on the estimation of curvatures
3. Use a simplified second order analysis based on nominal stiffnesses.

In this unit, the simplified method based on estimation of curvatures is used. This
method is recommended by the Concrete Centre.

Nominal curvature method of analysing slender columns.

A rigorous method of analysis will ascertain the load deflection relationship of a structure
which includes say slender columns. A simplified method must aim to find a close
estimate of this but for use in a design office.

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.13
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

The nominal curvature method recommended in the code aims to predict the deflection at
which failure of the concrete commences. i.e. when maximum compressive strain occurs
in the concrete. This will usually represent the actual failure moment and axial load of
the column but may in certain circumstances represent a lower load on the rising branch
of the N – M curve where two solutions are possible from the curve.

The diagram indicates that the strain limits for concrete change from εc3 when the column
is predominantly subject to compressive forces to εcu3 when bending becomes significant.
See unit 6 [EC2 Clause 6.1, Figure 6.1], unit 7 and the following figure.

Concrete in the element is at a strain of

εc3 and εcu3. Neutral axis is outside the
section. Curvature is below maximum
Concrete is at its maximum
strain εcu3 and the steel starts to
carry tensile forces.
Axial load NEd

Both concrete
and steel at
maximum strains

Moment MEd

Assumed curvature of a pin ended strut – Nominal curvature method.

The deflection of a pin ended strut is calculated using the curvature but this varies along
the length of the element. If a central deflection represented by βl02(1/r) is assumed, then
when β = 1/10 is selected close approximation to the actual deflection is achieved and
this value has been selected by EC2.

So e2 = 0.1 l02(1/r)

Derivation of above.

Deflection = (5wL4)/(384EI) [Simply supported beam/strut under udl]

Simply supported beam/strut under udl, M = wL2/8
M/I = σ/y = E/R
wL2/8 = EI/R so 1/R = wL2/(8EI)
Def = (5wL4)/(384EI) = [wL2/(8EI)]x[5L2/48]
Def = (1/R)x[5L2/48] approx 0.1 x L2/R

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.14
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

Determining the curvature – Nominal curvature method.

Consider the M – N curve.

1/rud = 0
Axial load

Nud 1/r

Balance point
NEd

Nbal

MEd

At the balance point the strains are as shown below :-

0.0035
Curvature of section at balance point :-
= change in strain over the depth
d 1/rbal = (0.0035 + fyd/Es)/d

εyd = fyd/Es

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.15
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

Presenting this slightly differently.

1.0

d Beam
dept
(0.0035 + fyd/Es)

Curvature of section at Nud = 1/rud = 0

The method then assumes the curvature varies linearly between 1/rud and 1/rbal so for any
load NEd the curvature can be expressed as :-

1/r = KrKφ(1/rbal)

Where:-
N ud − N Ed
Kr = ≤1
N ud − N bal

The code uses a simplified expression for 1/rbal by assuming that for a balanced section
the neutral axis depth is 0.55d thus :-
f yd
1 / rbal = ε yd / 0.45d =
( E s × 0.45d )

The coefficient Kφ allows for the effects of creep.

K φ = 1 + βφ ef ≥ 1

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.16
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

Where :-

β = 0.35 + fck/200 – λ/150

Creep can be ignored (i.e. Kφ =1) if :-

1. φ(∞,t0) ≤ 2
2. λ ≤ 75
3. M0Ed/Ned ≥ h
Where h is the cross sectional depth in the appropriate direction.

Where :-
φ is the effective creep ratio
λ is the slenderness ratio
M0Ed is the equivalent first order moment (May be taken as M0e)
Ned is the design value of applied axial force

Determining the maximum moment MEd.

MEd = MAX(M0e + NEd(e2 +ei) ; M02 +NEdei)

Where M0e is (0.6M02 + 0.4M01) ≥ 0.4M02

M01, M02 are the first order end moments at ULS not including allowances
for imperfections. M02 is numerically larger than M01. Attention should
be paid to the sign of M01 and M02. If they give tension on the same side,
M01 and M02 should have the same sign.

The maximum moment MEd depends on e2 which in turn depends on Kr which can only
be calculated when the area of steel is known.

Therefore an iterative procedure is used.

1. Assume Kr =1
2. Calculate MEd
3. Find area of steel required in column section for axial load NEd and moment MEd
4. Recalculate Kr
5. If this value of Kr differs significantly from the previous value return to 2.

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.17
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

Slender column design – Example.

The column is from a braced frame.

Assume all beam NEd = 3200kN

and column Applied
stiffnesses are equal moments

210kNm

6m

60kNm

400mm square
d2 = 40mm
fck = 40MPa
fyk = 500MPa
θi = 1/200
Ignore creep i.e. φef = 0

Determine the following :-

1. The effective height of the column?

2. The slenderness ratio λ of the column and its slenderness limit λlim ?
3. The equivalent first order moments M0e?
4. The value of the coefficient Kr?
5. Area of reinforcement required As?

1. Effective height of column?

l0 = βl where l is clear height. (See clause 5.8.3.2)

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.18
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

 k1  k2 
So l 0 = 0.5l 1 + 1 +  (5.15)
 0.45 + k1  0.45 + k 2 

 k1  k2 
or β = 0.5 1 + 1 + 
 0.45 + k1  0.45 + k 2 

Beam and column stiffnesses are equal so :-

S1 S2

SC1 l

SC1 + SC 2
k1 = k 2 = =1
S3 S4 S3 + S4

SC2 l

Therefore, β = 0.5 x [1.69 x 1.69]0.5 = 0.845

l0 = 0.845 x 6 = 5.07m

2. The slenderness ratio λ of the column and its slenderness limit λlim ?

Slenderness ratio = effective length/radius of gyration

= 5070/Radius of gyration

Radius of gyration = (I/A)0.5 = [(400 x 4003/12)/(400 x 400)]0.5 = 115.47mm

Slenderness ratio λ = 5070/115.47 = 43.91

Slenderness limit λlim = 20ABC/η0.5

A = 1/(1 + 0.2φef) But ignoring creep, φef =0, therefore A = 1.0

B = (1 + 2ω)0.5 . As ω is unknown so B = 1.1
C = 1.7 - rm rm = M01/M02 = -60/210 = -0.286
C = 1.986

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.19
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

3200 × 1000
η = N Ed /( Ac f cd ) = = 0.88
400 × 400 × 0.85 × 40 / 1.5)

λlim = 20 x 1 x 1.1 x 1.986 / 0.880.5 = 41.0

This is < 43.9 therefore the column must be designed for slenderness effects.

3. The equivalent first order moments M0e? [Clause 5.8.8.2]

M0e = 0.6M02 +0.4M01 ≤ 0.4M02

= 0.6 x 210 +0.4 x (-60) = 126 - 24 = 102kNm (> 0.4 x 210 = 84)

Eccentricity due to imperfections is :-

ei = θil0/2 (Equation (5.2), clause 5.2 (7))
= (1/200)5070 x 0.5 = 12.675mm

Additional moment = 3200 x 12.65/1000 = 40.56kNm

4. The value of the coefficient Kr?

Using e2 = 0.1l02(1/r) = 0.1l02 Kr Kφ (1/r)

e2 = KrKφ0.1 l02(eyd/0.45d) (with Kφ = 1),

= 0.1 x 50702 x (500/1.15/200000)/(0.45 x 360)Kr

= 34.49Kr

MEd = M0e + NEdei + NEde2 = 102 + 40.56 +3.2e2

N/bhfck = 3200 x 1000 / (400 x 400 x40) = 0.5

Using chart for d2/h = 0.1 gives, (See Figure 15.5(b) unit 7b

And by iteration the following values are found.

Assumed Kr MEd MEd/bh2fck New Kr

1 252.9 0.099 0.47
0.47 187.8 0.076 0.41
0.41 188.1 0.073 0.40

Kr = 0.40

5. What area of reinforcement is required?

e2 = 34.49Kr = 34.49 x 0.4 = 13.8mm

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.20
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

MEd = M0e + NEdei + NEd e2

= 102 + 40.56 +3200 x 13.8/1000 = 186.7kNm

This is less than the larger end moment + imperfection moment

= 210 + 40.56 = 250.56kNm

Therefore MEd = 250.6kNm

N/(bhfck) = 3200 x 1000 / (400 x 400 x 40) = 0.5

M/bh2fck = 250.6 x 106 / (4003 x 40) = 0.098

For chart d2/h = 0.1

Asfyk/(bhfck) = 0.23

As = 0.23 x 400 x 400 x 40 / 500 = 2944mm2

Bending about both axes.

Up to this point the treatment of slenderness effects has been concerned with unixial
bending. In this circumstance, the second order moments are in the same axis as the first
order moments. In reality, there are two other conditions that need to be considered.

1. First order moments about the major axis with second order effects about the
minor axis which is slender. The major axis may not be slender.
2. First order moments about both axes (bi-axial bending) . Either or both axes are
slender.

Step 1. Design uni-axially in each direction separately. Include geometric imperfections

in the least favourable direction.

Step 2. If two orthogonal axes, y and z exist on the cross section and if

0.5 ≤ λy/λz ≥ 2.0 and for rectangular sections,

0.2 ≥ (ey/heq)/(ez/beq) ≥ 5
Where λy, λz – slenderness ratio l0/i with respect to y and z axes.
ly = MEdy/NEd
heq = 3.46iz (= h for rectangular section)
ez = MEdz/NEd
beq = 3.46iy (= b for rectangular section)
Where NEd = design axial action at ULS
MEdy, MEdz = design moment in the respective direction (Moments due to
imperfections need only be included in the direction where they have the most
unfavourqable effects.

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.21
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

Note for square columns – (ey/heq)/(ez/beq) = MEdy/MEdz.

Then no further checks are necessary.

Step 3. If the checks in step 2 are not satisfied, then use Clause 5.8.9(4) [See unit 7]. 7b.

Slender beams.

Where a beam is narrow either with respect to its span or its depth, there is a possibility
that it may fail by lateral buckling.

However the likely occurrence of such beams is relatively rare and so the code employs a
simple conservative check thus :-

For safety the code requires both the following conditions be met :-

1. For persistent design conditions : - l ot ≤ 50b /(h / b) 0.333 and h ≤ 2.5b

2. For transient design conditions : - l ot ≤ 70b /(h / b) 0.333 and h ≤ 3.5b

Where :-
lot is the distance between torsional restraints
h is the total depth of the beam in the central part of lof
b is the width of the compression flange

Should these two conditions not be met, the code offers no further guidance.

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.22
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

Self assessment questions.

1. Discuss how column slenderness affects design. In particular consider how lateral
deflection due to moments and lateral deflection due to imperfections affect
design.
2. Derive the curvature if a column.
3. λlim is used to determine if a column is slender or not. Examine the terms in this
expression and justify why A, B C and n are included.
4. Figure Q4 shows a column subject to axial load and moments about x-x. The
creep effects are unknown, additional moments due to curvature have been
evaluated as 50kNm and those due to imperfections are 135.2kNm. Assuming the
mechanical reinforcement ratio is not known at the design stage, determine the
following :

(a) The effective height of the column.

[2 marks]
(b) The slenderness ratio λ of the column
[3 marks]
(c) The slenderness limit λlim of the column.
[4 marks]
(d) The equivalent first order moments M0e
[2 marks]
(e) The design moment at the column centre
[2 marks]
(f) The maximum design moment at the column end
[2 marks]
(g) Area of reinforcement required As.
[10 marks]

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.23
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

1. All beam and column

NEd = 3200kN stiffnesses are equal
2. The column is from a
braced frame.
210kNm 500mm
x

400mm
10m
x

60kNm d2 = 50mm
fyk = 500MPa
Concrete type C40/50
FIGURE Q4

Learning outcomes.

At the end of this unit you should be conversant with :

• The effect of sideways deflections and geometric imperfections on column

design.
• Slenderness ratios for columns.
• The effect of braced and unbraced frames on column design
• The limiting slenderness ratio λlim above which, slender column design needs to
be undertaken.
• The nominal curvature method of analyzing slender columns.
• Slender columns bending about both axes.

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.24
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

Solution to self assessment questions.

Question 4.
Figure 1 shows a column subject to axial load and moments about x-x. The creep effects
are unknown, additional moments due to curvature have been evaluated as 50kNm and
those due to imperfections are 135.2kNm. Assuming the mechanical reinforcement ratio
is not known at the design stage, determine the following :

(h) The effective height of the column.

[2 marks]
(i) The slenderness ratio λ of the column
[3 marks]
(j) The slenderness limit λlim of the column.
[4 marks]
(k) The equivalent first order moments M0e
[2 marks]
(l) The design moment at the column centre
[2 marks]
(m) The maximum design moment at the column end
[2 marks]
(n) Area of reinforcement required As.
[10 marks]

3. All beam and column

NEd = 3200kN stiffnesses are equal
4. The column is from a
braced frame.
210kNm 500mm
x
400mm

10m
x

60kNm d2 = 50mm
fyk = 500MPa
Concrete type C40/50
FIGURE 1

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.25
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

Additional design information. Clauses from BS EN 1992 : 2004

Clause 5.8.3.2. Effective height of columns = l0 = βl where l is clear height.

 k1  k2 
β = 0.5 1 + 1 + 
 0.45 + k1  0.45 + k 2 
k1, k2 are the relative flexibilities of the rotational restraints at ends 1 and 2 respectively.

5.8.3.1 Slenderness criterion for isolated members

…..Second order effects may be ignored if the slenderness λ is below the value λlim.
λlim = 20ABC/η0.5
where:
A = 1 / (1+0,2ϕef) (if ϕef is not known, A = 0,7 may be used)
B = (1+ 2ω)0.5 (if ω is not known, B = 1,1 may be used)
C = 1,7 - rm (if rm is not known, C = 0,7 may be used)
ϕef effective creep ratio.
ω = Asfyd / (Acfcd); mechanical reinforcement ratio;
As is the total area of longitudinal reinforcement
Ac is the cross sectional area of the concrete column
fyd is the design strength of the reinforcement
η = NEd / (Acfcd); relative normal force
NEd is the design axial load
fcd is the design strength of the concrete
rm = M01/M02; moment ratio
M01, M02 are the first order end moments, │M02│ ≥ │M01│

If the end moments M01 and M02 give tension on the same side, rm should be taken
positive (i.e. C ≤ 1,7), otherwise negative (i.e. C > 1,7).

5.8.8.2 Bending moments

…(2) Differing first order end moments M01 and M02 may be replaced by an equivalent
first order end moment M0e near mid height of the column.
M0e = 0,6 M02 + 0,4 M01 ≥ 0,4 M02
M01 and M02 should have the same sign if they give tension on the same side, otherwise
opposite signs. Furthermore, │M02│≥│M01│.

5.2 Geometric imperfections

…(7) For isolated members the effect of imperfections may be taken into account in as
follows. Use an eccentricity, ei, given by ei = θi l0/2
where l0 is the effective length, θi is the inclination of the column due to imperfections.

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.26
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.27
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

Solution.

1. Effective height of column = βl

 k1  k2 
β = 0.5 1 + 1 + 
 0.45 + k1  0.45 + k 2 

Beam and column stiffnesses are equal so :-

k1 = k2 = 1.0
Therefore β = 0.5 x [1.69 x 1.69]0.5 = 0.845
Effective length l0 = 0.845 x 10 = 8.45m

2. Slenderness ratio of column = Effective length / radius of gyration

Effective length = 8.45m

Radius of gyration = (I/A)0.5
I = (400 x 5003)/12 = 4167 x 106mm4
A = 0.2 x 106 = mm2
Radius of gyration = (4167/0.2)0.5 = 144.3mm
Slenderness ratio = 8450/144.3 = 58.6

3. Slenderness limit of column.

λlim = 20ABC/η0.5

Value of A = 0.7 {Ignoring creep effects}

Value of B = 1.1 {Reinforcement ratio not known}
Value of C = 1.7 – rm
rm = M01/M02 = -60/210 = -0.286
C = 1.7 – (-0.286) = 1.986
Value of η = NEd/(Acfcd)
η = (3200 x 103)/((400 x 500) x 0.85 x 40 /1.5) = 0.706

λlim = 20ABC/η0.5 = 20 x 0.7 x 1.1 x 1.986 / (0.706)0.5 = 36.4

Hence the column is slender.

4. Equivalent first order moment M0e

M0e = 0.6M02 +0.4M01 ≤ 0.4M02

= 0.6 x 210 + 0.4 x (-60) = 126 - 24 = 102kNm (> 0.4 x 210 = 84)

Design moments.
Moment at column centre = M0e + 50 + 135.2 = 287.2kNm

Maximum moment at column top = 210 + 135.2 = 345.2kNm

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.28
Civil Engineering, Faculty of Engineering & Physical Science, University of Surrey

So design moment = 345.2kNm

5. Area of main steel.

N/(bhfck) = (3200 x 103)/ (400 x 500 x 40) = 0.4

M/(bh2fck) =( 345.2 x 106)/ (400 x 5002 x 40) = 0.0863

Asfyk/(bhfck) = 0.12 = (As x 500)/(400 x 500 x 40)

As = 1920mm2

Use 4 No. T25 bars as the main reinforcement.

PG Cert/PG Diploma/MSc in Civil, Bridge & Structural Engineering Spring 2010 Page SE1M80.8.29