Design of Compression

Members
 Compression Member
Terminologies
Column, stanchion, or post : a vertical
compression member to support the floor or
beams in a building.
Strut: used in trusses and bracing
Rafter: a top chord member in roof truss.
Boom: the principal compression member in a
crane.

Image source: Geschwinder,

Unified design of steel structures

Image source: N. Subramanian, Design of Steel Structures Typical cross-sections

2
 Single Structural Shapes
Rolled sections (IS808 & IS1173)

Abbreviated
I Section Classification
symbol
Channel
Column section ISSC SC
Wide Parallel Flange Beams/Columns ISWPB WPB
Angles ISA
Angle Medium weight channels ISMC MC
Light weight channels ISLC LC
Tee
Medium weight channels ISMC MC
Medium weight parallel flange channels ISMCP MCP
Rolled normal T bars ISNT NT
Rolled deep legged T bars ISDT DT
Parallel flange section (IS12778) Light weight T bars ISLT LT
Slit medium weight T bars ISMT MT
Slit T bars from H-sections ISHT HT

3
Compound and Built up sections

Created by connecting two or

more shapes together

Rolled shapes strengthened with

plates.

4
 Modes of Failure

❑ Local buckling: slender members starts to deform excessively locally.

❑ Global flexural buckling: excessive deformation of the overall member.

Local plate
buckling

Source: Trahair et al. Source: Subramaniyan

The behaviour and Design Steel Structures:
of Steel structure to EC3 Design and Practice
❑ Torsional and Flexural-torsional buckling: mono- or asymmetric shapes.
https://www.youtube.com/watch?v=cYRicTk-Q08.
❑ Squashing: relatively short column can attain full strength and fails due to yielding.

5
 Plate Buckling

For a slender isotropic plate under uni-axial compression the plate buckling equation is given as

( )
4 4 4 2
∂ w ∂ w ∂ w ∂ w
D 4
+2 2 2
+ 4
+ N 0 2
=0
∂ x1 ∂ x 1 ∂ x2 ∂ x 2 ∂ x1

where w is the lateral deflection of the plate, D the flexural rigidity of the plate, and N0 the
compressive line load applied along the edge of the plate

Image source: Wang et al.

Exact solutions for buckling of
structural members
6
 Plate Buckling Coefficients K

2 2
Kπ E t
f cr = 2 2
12(1−ν ) b

Source: Alexander Chajes. Principles

of Structural Stability Theory.
7
 Critical Stress and Slenderness Limit

For a uni-axially loaded plate with two opposite edges simply supported
2 2
D2 k π Et
f cr =k π 2 = 2 2
b 12(1−ν ) b
Limiting values of b and t (so that section yields and does not buckle locally)

f y f y 12(1−ν 2 ) b 2
= 2 2
≤1
f cr E kπ t

√
2
b kπ E
≤
t 12(1−ν 2 ) f y
5
E=2 x 10 MPa , ν =0.3 , Let us assume k=4 for uniaxially compressed simply supported rectangular plate

√ √
2
b kπ E 250
≤ 2
t 12(1−ν ) 250 f y

b
≤53.8 ε
t

8
 Cross-section Classification
Class 1 cross-section:
Which can form a plastic hinge with the rotation capacity for plastic analysis.

Class 2 cross-section:
Can form plastic hinge with limited rotation capacity.

Class 3 cross-section:
Stress in extreme fibre can reach yield strength but no local buckling occurs.

Class 4 cross-section:
Local buckling occurs before yielding.

Source: EnggStruct,Quan et al. (2020). Vol 224, 111242

9
 Table 2: Limiting Width to Thickness Ratio (Clauses 3.7.2 & 3.7.4)

Full table is not shown

ε=
√ 250
fy

Actual averageaxial stress (negative if tensile)

r 1=
Design compressive stress of web alone

Actual average axial stress (negative if tensile)

r2 =
Design compressive stress of overall section

Please also read the notes in

IS800

Internal element of flange

but≥42ε but≥42ε
but≥42 ε

Web

Internal
Outstand
element
of flange

d /t Original image
from EC3 Part 1-1

10
 Internal and External Elements
Internal element of flange

Web

Outstand of flange

Internal
Web

Outstand of flange Internal element

of flange

Original image
from EC3 Part 1-1
11
 Flexural (Global) Buckling or Euler Buckling
L
Δx
x P=P cr
0,0 x
w ( x)
EI is constant
Bent equilibrium configuration
When P = Pcr the system is in a neutral equilibrium a slight perturbation causes the column to bend.
Assumptions:
●
Euler Bernoulli kinematics is applicable
●
Material is isotropic, homogeneous, and elastic
●
Deformations are small
●
Stresses are within elastic limit
●
The initial configuration is straight
●
Load is acting through the centroid of the cross-section
●
Axial strain is negligible so length of the column is unchanged

12
 Euler Buckling Load
Equation of equilibrium:
2
d w
M (x)= Pcr w (x)=− EI 2
dx
Displacement boundary conditions:
At x=0, w(x) = 0; and at x=L, w(x) = 0.
Force boundary conditions are already satisfied.
d 2 w P cr
2
+ w=0
dx EI
Effective length of column
Assume , w ( x)= A sin ( ) nπ x
L
2 2
n π EI
Non trivial solution , P cr = 2
L
π 2 EI
Lowest critical load is the buckling load Pcr = 2
L
Effective length (Clause 7.2) and slenderness

π 2 EI π 2 E A r 2 P cr π2E π2E
P cr = = ⇒ =f cr = = 2
( )
2 2 2
( KL) (KL) A KL λ
r : Radius of gyration r
K: Effective length factor
λ=KL/r: Effective slenderness
Source: Subramaniyan
fcr : Euler Buckling stress 13
 Effective Length (KL) of Prismatic Column (Clause 7.2)
Actual length, L (Clause 7.2.1): shall be taken as the length from centre-to- centre of its intersections
with the supporting members in the plane of the buckling deformation.

Effective lengths where boundary conditions can be assessed (Table 11, IS800)

Source: Duggal 14
 KL of Trusses (Clause 7.2)

Clause (7.2.4)
❑ For bolted, riveted, and welded compression members in trusses, and braced frame KL shall be
taken as 0.7 to 1 times the distance between center of connections depending on the restraint
provided
❑ For buckling in the plane perpendicular to the plane of truss KL is taken as distance between
center to center of intersections of longitudinal axes of the member at the joint

Distance between center of

intersections at the joints
Distance between center of
connections

Source: https://engineeringbasic.com/different-types-of-roof-roof-
trusses-and-there-components/ 15
 Bounds on Effective Length of Frame Elements
Non-sway frame Sway frame

Upper Upper
bound bound

Read clause 4.1.2 for definition of non-sway and sway frames

Lower Lower
bound bound

Image source: Salmon and Johnson, Steel

structures, design & behaviour
16
 Effective Length of Compression Member in Braced Frame (Annex D)

Kc, Kb: Effective flexural stiffness of the columns and beams

meeting at the joint at the ends of the columns and rigidly connected
at the joints given by K=CI/L.
I: Moment of inertia of the member about an axis perpendicular to
the plan of the frame.
L: Length of the member equal to centre to centre distance of the
intersecting member.
C: Correction factor

Source: Subramaniyan

Iteration to find K
 Assume a value of K
 Calculate Pcr
 Calculate K using the formula
above
 Continue until converges 17
Effective Length Chart of Compression Member in Braced Frame (Annex D)

18
 Effective Length of Compression Member in Un-Braced Frame (Annex D)

Kc, Kb: Effective flexural stiffness of the columns and beams

meeting at the joint at the ends of the columns and rigidly connected
at the joints given by K=CI/L.
I: Moment of inertia of the member about an axis perpendicular to
the plan of the frame.
L: Length of the member equal to centre to centre distance of the
intersecting member.
C: Correction factor

Source: Subramaniyan
Iteration to find K
 Assume a value of K
 Calculate Pcr
 Calculate K using the formula above
 Continue until converges
19
Effective Length Chart of Compression Member in Un-Braced Frame (Annex D)

20
Effective Length of Stepped Column (Annex D)

21
 Effective Length Coefficient (Annex D)

Please see Annex D in code for more details

22
 Buckling Stress vs Slenderness Ratio

π 2E
f cr =
π 2 Et ( KL/r )2
f cr = 2
( KL/r )

Stress strain response of elastic-plastic steel

Source:Subramaniyan

❑ Stocky column grossly yields before buckling

Squash Source: Duggal
❑ Columns with intermediate slenderness region
buckles inelastically
❑ Slender columns buckle elastically
π2E f y
f cr = 2 ⇒ =
λ f cr
π2E
fy √
when f cr =f y ; λ = λ̄ ⇒ λ̄ =
√π2E
fy

23
 Buckling of Real Compression Member

Buckling load of real column as function of

slenderness. Source:Subramaniyan
Influencing factors
❑ Geometric imperfection
❑ Load eccentricity
❑ Residual stress
❑ Non-linearity in stress-strain response
24
 Perry Robertson Approach

L
z x P
0,0 w 0 ( x) x

EI is constant w ( x)
z
w0(x): Initial imperfection
w(x) : Transverse deflection with respect to the imperfect geometry

Equation of equilibrium :
2
d w
M ( x)=P [ w( x )+w 0 (x ) ]=−EI 2
dx
Displacement boundary conditions:
At x=0, w(x) = 0; and at x=L, w(x) = 0.
Force boundary conditions are already satisfied.
2
d w
EI 2 + P w=−P w 0
dx

Assume , w 0 (x)=Δ 0 sin ( )

πx
L
25
 Solution

( )
2
d w πx w ( x)=w (x )+ w ( x)
EI 2 + P w=−P Δ 0 sin cf pi
dx L

Assume , w ( x )=C sin (

L )
πx PΔ PΔ 0 0
pi ⇒C=− =− 2
π EI P−P cr
− +P 2
L

w ( x)= A sin
P
EI √
x+B cos

Displacement boundary conditions, w(x = 0,L) =0

P
EI
x−
√
P Δ0
P−P cr
sin
πx
L ( )
Imposing , w( x=0)=0 , B=0 ;and imposing , w( x=L)=0 , A=0.

( )
P Δ0 πx
w ( x)=− sin
P−Pcr L

| P Δ 0 π 2 EI P P cr P Δ0
2
Maximum moment: −EI d w 0
2 =− =− Δ 0= =M F P Δ 0 =M F M max
dx x= L2 P−Pcr L2 P−P cr P
1−
0
M max , Maximum moment applied due to the imperfection P cr

M F , Magnification factor due to the compressive force

P M max P P P cr Δ 0 c f cd f cr Δ 0 c
f max = + c= + =f cd +
A I A P cr −P I f cr −f cd r 2
Δ0 c
Say , 2 =η
r
26
 Perry Robertson Formula
f cd f cr
f max =f y ⇒ f y =f cd + η
f cr −f cd

2
(f y −f cd )(f cr −f cd )=f cd f cr η ⇒ f cd −f cd (f y + f cr +η f cr )+ f y f cr =0

(f y + f cr +η f cr )± √ (f y + f cr +η f cr ) −4 f y f cr
2

f cd =
2

[ f y +f cr (1+η )]− √ [ f y + f cr (1+η )] −4 f y f cr

2

f cd =
2
Define some parameters (just to get the form that we want) :

√ fy f y + f cr (1+η ) 2
1+η + λ̄
= λ̄ ϕ= ⇒ϕ =
f cr 2 2
λ̄ 2 f cr fy
f cd =ϕ − √ ϕ −f y f cr = ϕ − √ ϕ − λ̄ f cr ⇒ f cd =
2 2 2
=
ϕ + √ ϕ − λ¯2
2
ϕ + √ ϕ 2 − λ¯2
Design compressive stress (introduce the material factor of safety)
f y /γ m0
f cd = = χ f y /γ m0
ϕ + √ϕ −λ
2 ¯2

1
where the reduction factor is , χ =
ϕ + √ ϕ 2 − λ¯2 27
 Multiple Column Curve
For better comparison with test results
η =α ( λ̄ −0.2) ⇒ ϕ =0.5 [ 1+α ( λ̄ −0.2)+ λ̄ 2 ]
λ̄ Depends on the shape of the cross−section
α Imperfection factor
❑ Imperfection factor was calibrated by a special committee for developing BS 5950
❑ Thousands of tests were carried out
❑ It was assumed Δ0 = L/1000
❑ Residual stress distribution idealizes experimental observation
❑ Buckling load determined by using tangent modulus theory

IS 800 adopted similar procedure

28
Buckling Class

29
Compressive Strength (Clause 7.1)
P≤P d
P d =f cd A e

30
Effective Section Area Ae (Clause 7.3.2)
P≤P d where , P d =f cd A e

For Class 4 section

Clause 3.7.2

31
 Class 4 Effective Section Area (Clause 3.7.2)
❑ Correct strength of Class 4 section is determined considering post-buckling reserve in thin plates
under compression as per IS 801. (we will not learn this)
❑ IS800 recommends the use of effective area to determine the cross-sectional properties for estimating
compressive strength of Class 4 section
❑ As recommended by IS800, effective area will be calculated by deducting the width of compression
plate element in excess of the semi-compact section limit

d eff

b eff b eff

Original image from EC3 Part 1-5

d eff =15.7 ε
From Table 2, IS 800, for Class 3
2 b eff =42 ε or semi-compact section

32
 Serviceability Criteria (Clause 3.8)
❑ A member under compression loses its stiffness in transverse direction
❑ The code recommends the limiting slenderness value to limit excessive vibration under service load

33
 Summary of Design Steps

From Subramaniyan
 Design of Angle Struts
(member with no or one axis of symmetry)
 Torsional and Flexural Torsional Buckling
❑ In torsional buckling member twists as it buckles
❑ Doubly symmetric sections may undergo torsional buckling. Hot-rolled sections are not
susceptible to torsional buckling due to high torsional stiffness.
❑ Should be checked for for members with symmetric built-up section.
❑ Member with mono asymmetric section twists and bends when it buckles it is called
flexural torsional buckling
The governing differential equation can be found in Galambos,
Structural members and frames. It has the following solution

2
cr
( )
( P cr −P crz )( P cr −P cry )( P cr −P crx )−P ( P cr − Pcry )
z 20
r̄ 0
2
2
−P ( Pcr −P crz )
cr
( )
y 20
r̄ 0
2
=0

y y Pcr: Buckling load, Pcry: Load causing buckling about minor axis z-z, Pcrz:
Load causing buckling anout major axis y-y, Pcrx: Pure torsional buckling
x x load, z0, and y0 are coordinates of the shear center with respect to the
y y centroid, and
2 2 2 I +I
r̄ 0 = z̄ 0 + ȳ 0 + z y
A
z z
Iz: 2nd Moment of inertia about the minor axis, Iy: 2nd Moment of inertia
Source: Segui, W.T.
about the major axis, and A: The cross-sectional area of the member.
Steel Design

36
 Buckling load for single angle strut loaded through centroid (IS 800)

Single angle strut connected by one bolt

Source: Duggal

However, the IS code guideline is not correct

because the centroid of an angle section is not the
shear center.

We should do flexural-torsional buckling check

Effective length to be used to calculate buckling

load is not specified. It is left to the user’s
judgement

37
 Single Angle Strut Loaded Through One Leg
7.5.1.2 Loaded through one leg

The flexural torsional buckling strength of a single angle loaded in compression through on its legs
may be evaluated using equivalent slenderness ratio λe, as given below, in place of λ in 7.1.2.1 and
using c curve ‘c’ (α = 0.49).
λ e = √ k 1 +k 2 λ vv 2 +k 3 λ ϕ 2 Where from this equation came is not sure
where
k 1 , k 2 , k 3 Given in Table 12
l (b 1 +b 2 ) l: center-to-center distance of supporting member
r vv rvv: radius of gyration about the minor axis
λ vv = 2t

√
λϕ=

√
π2E π2E b1,b2: width of the two legs of the angle
ε ε
250 250 t: thickness of the leg
ε: yield stress ratio (250/fy)0.5
The equivalent slenderness takes care of three limit
states of failure local buckling, flexural buckling, and
flexural-torsional buckling

38
 Discontinuous Double Angle Strut (7.5.2)

Clause 7.5.2.1 Double-angle strut connected back to back on opposite side of gusset

Tacking bolts make sure the two

angles behave in union spacing of
tacking bolts must satisfy requirements
in Clauses 7.8 and 10.2.5.5

Source: Duggal
❑K = 0.7-0.85 for buckling in plane of truss
❑K = 1 for out of plane buckling
❑ Failure through flexural buckling is considered
❑ Flexural torsional buckling is not considered assuming high torsional stiffness

39
 Discontinuous Double Angle Strut (7.5.2)

Clause 7.5.2.2 Double-angle strut connected back to back on one side of gusset

Tacking bolts make sure the two

angles behave in union spacing of
tacking bolts must satisfy requirements
in Clauses 7.8 and 10.2.5.5

Source: Duggal
❑ Again, the effective length calculation is left to user’s judgement
❑ Failure through flexural-torsional buckling is considered
❑ Strength check as per flexural-torsional buckling

40
 Compression Members With Two Components Back to Back (Clause 7.8)
Clause 7.8.1
Can be composed of two angles, or channels, or tees
❑ Shall be connected together by riveting, bolting, or welding
❑ The un-favorable slenderness of each member between the intermediate connections should not be
greater than min(40,or 0.6*most un-favorable slenderness ratio of the whole member)
❑ Tacking bolt spacing S in line shall not exceed 600 mm (Clause 10.2.5.5)
CL L CL
S

❑ At the ends at-least two bolts or rivets or equivalent welding must be provided
❑ If
leg widths of the connected angles or tees ≥ 125 mm at-least two bolts or rivets on line of each gauge
mark must be provided
❑ If
web width of connected channels ≥ 150 mm at-least two bolts or rivets on line of each gauge mark
must be provided

Gauge line 1 Gauge line 1

≥125 mm ≥150 mm Gauge line 2
Gauge line 2

Angle Channel 41
Design of Laced Columns

42
 Mutli-scale Lattice Structure
❑ Light-weight
❑ Tailoredstrength and stiffness
❑ Can be multi-functional

Multi-scale metal lattice structure Losing Weight to Get to Mars

Micro-lattice metal for future planes

and vehicles by Boeing
Source:https://www.asme.org/topics-resources/content/
losing-weight-get-to-mars

Watch the next video if interested

https://www.boeing.com/features/2015/10/innovation-lightest-metal-10-15.page

Source: Zheng et al. (2016), Nature Materials,

Vol 15,p 1100 ) 43
 Laced Columns (Clause 7.6)

Lacings
Main
members

San Francisco–Oakland Bay Bridge

Source:https://en.wikipedia.org/wiki/Lattice_girder#/media/
File:San_Francisco_Oakland_Bay_Bridge_Retrofit_3.jpg

44
 Behavior of Laced Columns

dM dW
V= +N
dx dx
In absence of bending
moment and under
compression N = -P
dW
V =−P
dx

Origin of shear in bent column

Modes of failure Source: Bažant and ❑ Thelacing members must be designed for the axial force
Cedolin Stability of Structures induced due to transverse shear
❑ To determine such forces 2nd order geometric nonlinear
analysis with geometric imperfection needs to be performed
F
V ❑ IS800 adopts a simplified procedure
V
F=
cos α
V

45
 Laced Columns (Clause 7.6)
❑ Clause 7.6.1.1 (radius of gyration, r): ry>=rz
y
❑ Clause7.6.1.2 : Lacing system should be uniform
throughout the length of the column

❑ Clause 7.6.1.3: Cross-member should be avoided z z

unless all forces resulting from deformation are
accounted for in design of lacing and fasteners
y
❑ Clause7.6.1.4: For single lacing, lacing on one face
should be mirror image of the lacing on the other
face

❑ Clause 7.6.1.5: The effective slenderness ratio of

laced column is increased due to shear deformation
effect

(KL/r)eff =1.05∗( KL/r 0 )

where, KL/r0 is the maximum slenderness ratio

Image source:Duggal

46
 Design of Lacings (Clause 7.6)
❑ Clause 7.6.6.1 (shear force due to axial force)
Design shear force Vt=2.5% of P to be divided equally among all lacing members in parallel plane
❑ Clause 7.6.6.2 (additional shear due to bending)
For member carrying bending moment due to load eccentricity, boundary condition, or external force
additional shear due to bending must be considered
❑ Clause 7.6.6.3 (slenderness ratio to calculate compressive strength)
●
To be designed as compression member
●
The slenderness ratio KL/r ≤ 145
▷
L is the distance between inner end bolts or the distance between inner ends of the weld
▷
If bolted ends, K = 1 for single lacing, K = 0.7 for double lacing
▷
If welded ends, K=0.7

Image source:Duggal 47
 Lacing General Requirements (Clause 7.6)
❑ Clause 7.6.2 : In bolted and riveted construction the minimum
width w of lacing bar shall be 3d where d is the nominal diameter
of bolt/rivet
w
❑ Clause 7.6.3 :
●
Thickness ≥ L/40 for single lacing
●
Thickness ≥ L/60 for double lacing
●
Rolled sections or tubes can be used as lacing
where L is the effective length of the lacing
❑ Clause 7.6.4:
Inclination of lacing member θ shall be 40o≤θ≤70o
❑ Clause 7.6.5:

[
Spacing of lacing bars a 1 ≤r 1∗max 50,0.7∗( )]
kL
ry

where a1, is the unsupported length of individual members, r1 is

the minimum radius of gyration of individual, and ry is the
minimum radius of gyration of the whole cross-section

48
 Design of Lacing Connection (Clause 7.6)
❑ Clause7.6.6.1 (shear force due to axial force): Design shear force Vt=2.5% of P to be divided equally
among all lacing members in parallel plane
❑ Clause 7.6.6.2 (additional shear due to bending): For member carrying bending moment due to load
eccentricity, boundary condition, or external force additional shear due to bending must be considered
❑ Clause 7.6.6.3 (slenderness ratio to calculate compressive strength): The slenderness ratio KL/r ≤ 145
●
L is the distance between inner end bolts or the distance between inner ends of the weld
●
If bolted ends, K = 1 for single lacing, K = 0.7 for double lacing
●
If welded ends, K=0.7
❑ Design procedure: Lacing to be designed as compression or tension member (Clause 5 & 7)
❑ Clause 7.6.7 (connection):
●
Connections are designed to resist member force
●
For welded connection amount of lap at each end = 4*min(thickness of the bar or of the column)

Laced faces:
AA, BB

Force transfer Types of laced column

Image source:Duggal 49
 Battened Columns (Clause 7.7)
❑ Clause 7.7.1.1 (radius of gyration, r): ry>=rz

❑ Clause 7.7.1.2 : If battened column is subjected to eccentric

loading, moment, or lateral force then design must be done based
on exact theory or empirical model verified through test
❑ Clause 7.6.1.3: At-least 3 equally spaced batten at each face
must be provided
❑ Clause 7.6.1.4: The effective slenderness ratio of laced column is
increased due to shear deformation effect

(KL/r)eff =1.10∗(kL /r 0 )

where, KL/r0 is the maximum slenderness ratio Image source:Duggal

50
 Design of Battens (Clause 7.7)
❑ Clause 7.7.1.1 (radius of gyration, r): ry>=rz

❑ Clause 7.7.1.2 : If battened column is subjected to eccentric

Periodic deformation
loading, moment, or lateral force then design must be done based
on exact theory or empirical model verified through test
Clause 7.7.1.3: To be designed for bending moment and shear
force arising due to a transverse force Vt=2.5% of P to be divided
equally among all lacing members in parallel plane

V t/2 N V t/2 N
C /2
`
VtC
VtC VtC 4N
C /2
NS NS
V t/2 N
V t/2 N
S /2
C
S /2
VtC
2N
N: no. of parallel planes
of battens

❑ Clause 7.7.2.1
VtC
Design value of shear force:
NS S /2 S /2
V C
Design value of bending moment: t
2N Source: Bažant and Cedolin `
51
 Design of Battens Contd. (Clause 7.7)
Clause 7.7.2.3 (Size)
●
Effective depth De, is the longitudinal distance between outermost connectors at the ends
●
For end battens, De ≥ distance between centroid of main members
●
For intermediate battens, De ≥ 0.75*distance between centroid of main members
●
D ≥ 2*width of one member
●
Thickness batten and tie plates ≥ 1/50*distance between innermost connecting lines of rivets, welds, and
bolts perpendicular to the main member
Clause 7.7.2.4: Above requirements do not hold if angles, channels, and I-sections are used for battens
Clause 7.7.2.5 (Spacing of battens):

[
C≤r 1∗max 50,0.7∗
kL
ry ( )]
where a1, is the unsupported length of individual members, r1 is the minimum radius of gyration of
individual, and ry is the minimum radius of gyration of the whole cross-section
❑ Clause 7.6.4 (connection):
≥4∗t
●
Length of weld connection at each end ≥ 0.5*depth of batten plate L1

●
1/3rd of the weld must be placed at each end of the plate L2≥LT /3
D
●
Lap at each end ≥ 4* thickness of batten plate
2∗L1 + L2=LT ≥0.5∗D

52
Cross-sectional Instabilities

Cross-sectional instabilities of a channel beam. Image

source: Silvestre and Camotim, J. Struct. Eng., 2006,
132(4): 529-540

53