Introduction to

Eurocode 0 and Eurocode 1

Cem Topkaya, Ph.D

Middle East Technical University

Department of Civil Engineering
Structural Engineering Division

1
© TOPKAYA, 2024

A General Look at the European Norms

Number given to
the specification

Release date

Description

2
© TOPKAYA, 2024
 European Standards Related to Structural Design

• EN 1990 – Eurocode 0 : Basis of Structural Design

• EN 1991 – Eurocode 1 : Actions on Structures
• EN 1992 – Eurocode 2 : Design of Concrete Structures
• EN 1993 – Eurocode 3 : Design of Steel Structures
• EN 1994 – Eurocode 4 : Design of Composite Steel/Concrete Structures
• EN 1995 – Eurocode 5 : Design of Timber Structures
• EN 1996 – Eurocode 6 : Design of Masonry Structures
• EN 1997 – Eurocode 7 : Geotechnical Design
• EN 1998 – Eurocode 8 : Design of Structures for Earthquake Resistance
• EN 1999 – Eurocode 9 : Design of Aluminum Structures

3
© TOPKAYA, 2024

Parts and Sub-parts

Part
Eurocode 3 : Design of Steel Structures

Sub-Part
EN 1993-1 : General Rules and Rules for Buildings
EN 1993-2 : Steel Bridges
EN 1993-3 : Towers, Masts and Chimneys
EN 1993-4 : Silos, Tanks and Pipelines
EN 1993-5 : Piling
EN 1993-6 : Crane Supporting Structures

4
© TOPKAYA, 2024
 EN 1993-1 : General Rules and Rules for Buildings

Sub-Parts
• EN 1993-1-1 : General Rules and Rules for Buildings
• EN 1993-1-2 : Structural Fire Design
• EN 1993-1-3 : Cold-formed thin gauge members and sheeting
• EN 1993-1-4 : Stainless Steel
• EN 1993-1-5 : Plated Structural Elements
• EN 1993-1-6 : Strength and Stability of Shell Structures
• EN 1993-1-7 : Strength and Stability of Planar Plated Structures
Transversely Loaded
• EN 1993-1-8 : Design of Joints
• EN 1993-1-9 : Fatigue Strength of Steel Structures
• EN 1993-1-10 : Selection of Steel for Fracture Toughness and through-thickness prop.
• EN 1993-1-11 : Design of Steel Structures with Tension Components Made of Steel

5
© TOPKAYA, 2024

Eurocode 1990 – EC 0 : Basis of Structural Design

6
© TOPKAYA, 2024
 SECTION 1: GENERAL

SCOPE

• Principles and requirements for the safety, serviceability, and durability

of structures
• Guidelines for structural safety
• Should be used together with EN 1991 – EN 1999 specifications
• Can be used for construction materials that are not covered
by EN 1991 – EN 1999
• It is possible to use EC0 to evaluate existing structures

Principles and Application Rules

In general the statements given in Euronorms are divided into two categories
7
© TOPKAYA, 2024

An example from EC8

“P”rinciples

Application
Rules

8
© TOPKAYA, 2024
 Differences Between Principles and Application Rules
• Principles cannot be changed
• An alternative design rule can be substituted for an Application Rule
based on the state-of-the-art
• If the Application Rules are changed then the resulting
design cannot be claimed to be wholly in accordance with Euronorms

Terms and Definitions

The terminology that is used in entire Euronorms is given

in Part 1 of EC0

Example:
Persistent Design Situation: Design situation that is relevant during
a period of the same order as the design working life of the structure
9
© TOPKAYA, 2024

Symbols

G: permanent action
Q: variable action
A: accidental action
AEk: characteristic value of seismic action

Analysis Methods and Terms Related to Structural Analysis

• First order linear-elastic analysis without redistribution

• First order linear-elastic analysis with redistribution
• Second order linear-elastic analysis
• First order non-linear analysis
• Second order non-linear analysis
• First order elastic-perfectly plastic analysis
• Second order elastic-perfectly plastic analysis
• Elasto-plastic analysis (first or second order)
• Rigid plastic analysis
10
© TOPKAYA, 2024
 SECTION 2: REQUIREMENTS

Basic Requirements :
• A structure shall be designed to have adequate structural resistance,
serviceability, and durability.
• Economic
• Resistance against fire
• Resistance against explosions, impact and consequences
of human errors

Reliability Management:
• Clauses to manage reliability of structures

11
© TOPKAYA, 2024

Design Working Life

Design Indicative Design Examples

Working Working Life
Life (years)
Category
1 10 Temporary Structures

2 10 to 25 Replaceable structural parts

3 15 to 30 Agricultural and similar structures

4 50 Building structures and other

common structures
5 100 Monumental building structures,
bridges, and other civil engineering
structures
© TOPKAYA, 2024 12
 SECTION 3: PRINCIPLES OF LIMIT STATES DESIGN
GENERAL:

• A distinction shall be made between ultimate limit states and

serviceability limit states
Limit state = a limit of structural usefulness, a failure mode
• Design situations shall be classified as follows:
Persistent Design Situation → Normal use
Transient Design Situation → Temporary conditions, e.g. during execution
Accidental Design Situation → Exceptional conditions, e.g. fire,
explosion, impact
Seismic Design Situation → Seismic events

13
© TOPKAYA, 2024

ULTIMATE LIMIT STATES

The limit states that concern the safety of people and/or the
safety of structure. Example = Collapse

Source: Google 14
© TOPKAYA, 2024
 SERVICEABILITY LIMIT STATES
The limit states that concern the functioning of the structure;
comfort of people, appearance of the construction works.
Example = Excessive vibrations, excessive deflection

Excessive
deflection

Source: Google
15
© TOPKAYA, 2024

SECTION 4: BASIC VARIABLES

Classification of Actions (Loads)

• Permanent Actions G
• Variable Actions Q
• Accidental Actions A

Permanent Actions Variable Actions Accidental Actions

G Q A
Self weight, fixed Imposed (Live) loads Explosions
equipment on building floors
Prestressing force Snow load Fire
Earth or water Wind load Impact
pressure
Foundation settlement Temperature effects
© TOPKAYA, 2024 16
 Certain actions such as seismic actions and snow loads may be considered
as either accidental and/or variable actions.

Characteristic Values of Actions

The characteristic value Fk of an action is its main representative value
and shall be specified as a :
• mean value
• upper or lower value
• Nominal value (which does not refer to a known statistical distribution)

17
© TOPKAYA, 2024

Characteristic Value of a Permanent Action

If the variability of G can be considered small, one single Gk that is the
mean value may be used

Gk PG
VG PG Mean

VG Standard
deviation

PG
18
© TOPKAYA, 2024
 If the variability of G cannot be considered as small, two values
shall be used: an upper value Gk,sup and a lower value Gk,inf
(5% and 95% percentile values)

Characteristic lower value

VG
Gk ,inf PG  1.64V G

Characteristic upper value

1.2
PG
1

0.8
Gk ,sup PG  1.64V G
0.6

0.4

0.2

Gk ,inf Gk ,sup 19
© TOPKAYA, 2024

Characteristic Values of Variable Actions

Values that are derived using probabilistic methods. These are
usually based on measurements. Determined based on probability of
exceedance within a reference time period.

Example: Snow load value in which the annual probability

of exceedance is 2 percent (50 year return period)

Qk
Other Representative Values of Variable Actions
EC0 recommends three different variable actions. These values
are generally used in load combinations

\ 0Qk Combination value

\ 1Qk Frequent value
\ 2Qk Quasi-permanent value
20
© TOPKAYA, 2024
Instantaneous
Value, Q Characteristic value
Qk

Combination value
\ 0Qk
Frequent value
\ 1Qk
Quasi
Permanent
value
\ 2Qk
Time
Reference duration

\ 0Qk Combination value: Used for verification of ultimate limit states and
irreversible serviceability limit states
\ 1Qk Frequent value: Used for verification of ultimate limit states involving
accidental actions and verification of reversible
serviceability limit states
\ 2QkQuasi-permanent value: Used for verification of ultimate limit states involving
21
accidental actions, also long-term effects
© TOPKAYA, 2024

Annex A1: Application for Buildings

\0
\1
\2
values

22
© TOPKAYA, 2024
 SECTION 5: STRUCTURAL ANALYSIS AND DESIGN
ASSISTED BY TESTING
Section 5 outlines the basic principles of structural analysis
• Structural analysis models and boundary conditions
• Modeling of static and dynamic actions
• Modeling of fire
• Information on design assisted by testing

Effects of displacements and deformations shall be taken into account

in the context of ultimate limit state verifications if they result in significant
increase of the effects of actions. (Second-order analysis)

Further details are in EN 1991 -- EN1999

23
© TOPKAYA, 2024

SECTION 6: VERIFICATION BY THE PARTIAL

FACTOR METHOD
It is dangerous from a safety point of view to use directly
the characteristic values in design

Ek d Rk
Characteristic action Characteristic resistance
Actions and resistances are variable and there are uncertainties
associated with them
Uncertainties in actions:
• Uncertainties that depend on the type of actions
• Uncertainties in combination of actions
• Uncertainties in modeling and analysis
24
© TOPKAYA, 2024
 Uncertainties in Resistance:
• Uncertainties in member properties (material properties, dimensions,
imperfections, residual stresses)
• Uncertainties in predicting structural behavior (depends on type of
behavior, i.e. Column buckling, lateral torsional buckling, etc.)

A margin of safety is incorporated during the design by making use

of the partial factors method

Ed d Rd
Design action Design resistance

25
© TOPKAYA, 2024

The basic principle is to convert characteristic values to design values

Characteristic
Action
Characteristic
Resistance

Area that
quantifies
probability of
failure
Action or Resistance Value

The case where characteristic action values are close to characteristic

resistance values

HIGHER PROBABILTY OF FAILURE

26
© TOPKAYA, 2024
 In deriving design values, the action values are increased while
the resistance values are decreased by making use of partial factors

Ed J F u Ek
Rk
Rd
JM
JF Partial factor for the actions, also accounting for model
uncertainties and dimensional variations

JM Partial factor for a material property, also accounting for model

uncertainties and dimensional variations

27
© TOPKAYA, 2024

Probability distribution for design values after the application of partial factors

Design
Action
Design
Resistance

Area that
quantifies
probability of
failure
Action or Resistance value

The case where action values and resistance values are separated

LOWER PROBABILITY OF FAILURE

28
© TOPKAYA, 2024
 JM Partial Factor for Resistance

Described in EN 1991 – EN 1999 depending on the type of material used

Example: Steel Structures EN 1993

J M 0 1.0 Resistance of cross-sections

J M 1 1.0 Resistance of members to instability

J M 2 1.25 Resistance of cross-sections in tension fracture

29
© TOPKAYA, 2024

ULTIMATE LIMIT STATES

The following ultimate limit states shall be verified as relevant:

• EQU: loss of static equilibrium

• STR: internal failure or excessive deformation of the structure
• GEO: failure or excessive deformation of the ground
• FAT: fatigue failure

Basic Principles

Ed ,dst d Ed ,stb Equilibrium limit state

Ed d Rd Structural or geotechnical limit states

30
© TOPKAYA, 2024
 Combination of Actions (Load Combinations)

Basic principle is to combine actions that can act together

Each combination of actions should include:

a leading variable action or an accidental action

EC 1990 utilizes the concept of leading variable action and

accompanying variable action

Combination of Actions for Persistent or Transient Design Situations

The basic combination is given in EN 1990 equation (6.10)

¦J
j t1
G, j u Gk , j ""J p u P""J Q,1 u Qk ,1"" ¦ J Q,i u\ 0,i u Qk ,i
i !1

Leading variable Accompanying variable

Permanent Action Prestressing
action action

"" implies “to be combined with” 31

© TOPKAYA, 2024

Recommendations for partial factors are not given in the main body
of EN 1990. These values are specific to the country of use and are
usually given in the National Annex.

On the other hand, some recommendations are given in Annex A1

of EN 1990 in the absence of a National Annex.
Permanent actions are classified as unfavorable and favorable.
Unfavorable permanent actions act in the same direction as the
variable actions
Favorable permanent actions act in the opposite direction to the
variable actions
According to Annex A1 A1.2.1 (1) Note 1: depending on its uses and the form
and the location of a building, the combination of actions may be based on
not more than two variable actions.
32
© TOPKAYA, 2024
 Design Values of Actions (STR/GEO)

Annex A1: Table A1.2(B)

J G ,inf1.00 J G ,sup 1.35

J Q,1 1.50 J Q,i 1.50 33
© TOPKAYA, 2024

Basic combination in the absence of prestressing

¦J
j t1
G, j u Gk , j "" J Q,1 u Qk ,1 "" ¦J
i !1
Q ,i u\ 0,i u Qk ,i

1.35 or 1.0 1.50 1.50

34
© TOPKAYA, 2024
 Derivation of Load Combinations

Example: Roof of a Stadium

D: self weight S: snow load W: wind load
Elevation 100 meters
\0 0.6 Wind Load \ 0 0.5 Snow Load
Cases where self weight in creates unfavorable action:

Snow is the leading variable action and wind is the accompanying action

1.35u D  1.50 u S  1.50 u 0.6 uW 1.35u D  1.50 u S  0.9 uW

Wind is the leading variable action and snow is the accompanying action

1.35u D  1.50 uW  1.50 u 0.5 u S 1.35u D  1.50 uW  0.75u S

35
© TOPKAYA, 2024

Example: Tower Structure D: self weight W: wind load

D
W

Reaction (compressive)
Reaction (Tensile)
Self weight (unfavorable)
Self weight (favorable)
1.00 u D  1.50 uW 1.35u D  1.50 uW

36
© TOPKAYA, 2024
 EN 1990 recommends two other load combinations as an
alternative to Equation 6.10
These combinations are for STR and GEO limit states
National Annex provides recommendations on which equation to use
Most unfavorable case from Equations 6.10a and 610b is used
Equation 6.10

¦J
j t1
G, j u Gk , j ""J p u P""J Q,1 u Qk ,1"" ¦ J Q,i u\ 0,i u Qk ,i
i !1

Equation 6.10a

¦J
j t1
G, j u Gk , j ""J p u P""J Q,1 u\ 0,1 u Qk ,1"" ¦ J Q,i u\ 0,i u Qk ,i
i !1

Equation 6.10b

¦[ J
j t1
j G, j u Gk , j ""J p u P""J Q,1 u Qk ,1"" ¦ J Q,i u\ 0,i u Qk ,i
i !1

[j Reduction factor for unfavorable permanent actions

37
© TOPKAYA, 2024

Load Combination for Accidental Actions

¦G
j t1
k, j "" P "" Ad "" (\ 1,1 or \ 2,1 ) u Qk ,1"" ¦\ 2,i u Qk ,i
i !1

Accidental Action

Load Combination including Seismic Action

¦G
j t1
k, j "" P "" AEd "" ¦\ 2,i u Qk ,i
i t1

Seismic Action

38
© TOPKAYA, 2024
 Example: Office building D: self weight E: earthquake L: live load

\2 0.3

1.00 u D  E  0.3 u L
39
© TOPKAYA, 2024

COMBINATIONS OF ACTIONS FOR SERVICEABILITY LIMIT STATES

EN 1990 defines three combinations to assess serviceability limit states

These combinations differ depending on how frequent the loads
act on a structure

Characteristic Combination

¦G
j t1
k, j "" P"" Qk ,1"" ¦\ 0,i u Qk ,i
i !1

All partial factors are equated to unity (1.0)

This combination is generally used for irreversible cases
Example: exceeding a certain crack width in a reinforced concrete structure

40
© TOPKAYA, 2024
 Frequent Combination

¦G
j t1
k, j "" P""\ 1,1 u Qk ,1"" ¦\ 2,i u Qk ,i
i !1

All partial factors are equated to unity (1.0)

This combination is generally used for reversible cases

Quasi-permanent combination

¦G
j t1
k, j "" P"" ¦\ 2,i u Qk ,i
i t1

All partial factors are equated to unity (1.0)

This combination is generally used for long-term effects
Example: creep in a reinforced concrete structure
41
© TOPKAYA, 2024

Annex A: Serviceability Limit States

Annex A provides recommendations on serviceability

Examples:
Vertical and horizontal deflection limits
Limits on vibrations
Additional rules can be found in EN 1991-EN1999 for serviceability

42
© TOPKAYA, 2024
 Eurocode 1991 – EC 1 : Actions on Structures

A general look at EN 1991 specification

• EN 1991-1: General Actions

• EN 1991-2: Traffic Loads on Bridges
• EN 1991-3: Loads of Cranes and Machines
• EN 1991-4: Loads on Silos and Tanks

EN 1991-1: General Actions have 7 sub-parts

43
© TOPKAYA, 2024

EN 1991-1 General Actions: Sub-Parts

• EN 1991-1-1: Densities, self weight, imposed loads for buildings

• EN 1991-1-2: Actions on structures exposed to fire
• EN 1991-1-3: Snow loads
• EN 1991-1-4: Wind loads
• EN 1991-1-5: Thermal actions
• EN 1991-1-6: Actions during execution
• EN 1991-1-7: Accidental actions

44
© TOPKAYA, 2024
 EN 1991-1-1: Densities, self weight,
imposed loads for buildings

45
© TOPKAYA, 2024

GENERAL

• Densities of construction materials and stored materials

• Self weight of construction works
• Imposed loads for buildings (live loads)

CLASSIFICATION OF ACTIONS

• Self-weight: Loads which generally do not vary over time

• Imposed Loads: Loads that vary over time and space

46
© TOPKAYA, 2024
 DENSITIES OF CONSTRUCTION MATERIALS AND
STORED MATERIALS
Densities can be obtained from tables in Annex A

Structural steel
77.0 to 78.5 kN/m3

47
© TOPKAYA, 2024

Stored materials

Angle
of
repose

48
© TOPKAYA, 2024
 IMPOSED LOADS ON BUILDINGS

REPRESENTATION OF ACTIONS

• Normal use by persons

• Furniture of movable objects

• Vehicles
• Anticipating rare events, such as concentrations of persons or of
furniture, or the moving or stacking of objects which may occur during
reorganization or redecoration

Imposed loads are modeled by uniformly distributed loads,

line loads or concentrated loads, or combinations of these loads

49
© TOPKAYA, 2024

CHARACTERISTIC VALUES FOR IMPOSED LOADS

Residential, social, commercial and administration areas
Categories of use are given in Table 6.1

Category B
Office areas

Category D
Shopping areas

50
© TOPKAYA, 2024
 Values of imposed loads are given in Table 6.2 for each category

Distributed load
qk

Concentrated load

Qk

• Distributed and concentrated loads can be considered separately

• National Annex can provide specific load values for the range of values
given in Table 6.2
• In the absence of National Annex the underlined value shall be used 51
© TOPKAYA, 2024

Imposed Load Reduction

• For floors, beams and roofs a reduction based on area is possible
• National Annex provides specific rules
• In the absence of National Annex the following equation can be adopted

5 A0
DA \ 0  d 1.0
7 A
A0= 10m2

A = loaded area

\0 Factor according to EN 1990

52
© TOPKAYA, 2024
 • Loads on columns and walls can be reduced depending on the
number of stories
• National Annex provides specific rules
• In the absence of National Annex the following equation can be adopted

2  2  n u\ 0
Dn
n

n = number of stories (n>2) above the loaded structural elements

\0 In accordance with EN 1990

53
© TOPKAYA, 2024

For other areas different categories and load values are

recommended

Categories of storage and industrial use

Category

Recommended value of imposed load

54
© TOPKAYA, 2024
 Garages and vehicle traffic areas

Category

Recommended value
of the imposed load

55
© TOPKAYA, 2024

Roofs

Category

Recommended value
of the imposed load

56
© TOPKAYA, 2024
 EN 1991-1-3: Snow Loads

EN 1991-1-3 specification can be used to determine snow loads

on structures that are on sites with an altitude below 1500 meters

• Impact snow loads resulting from snow sliding

• Ice loading
• Snow loads on bridges and etc.
are not covered

57
© TOPKAYA, 2024

Need to know the value of snow load on ground to be able to determine

the snow load on the roof.

sk Characteristic value of snow on the ground at the relevant site

Annual probability of exceedance of 0.02

• Determined based on meteorological measurements

• No snow load map developed for Turkey yet
• There exists snow load maps for countries in Europe

58
© TOPKAYA, 2024
 Example: Snow load map for Mediterranean Region

Snow load
on ground

59
© TOPKAYA, 2024

In some cases or some regions exceptional snow fall or

snow drift may occur

According to EN 1991-1-3 specification exceptional cases can be

treated as accidental actions

s Ad Cesl u sk

Cesl = coefficient for exceptional snow loads

(Recommended value = 2.0)

s Ad = design value of exceptional snow load on the ground

60
© TOPKAYA, 2024
 Snow Load on Roofs
The amount of snow on the ground and on the roof are different.
The following factors must be taken into account to relate them

• Shape of the roof

• Its thermal properties
• The roughness of its surface
• The amount of heat generated under the roof
• The proximity of nearby buildings
• The surrounding terrain
• The local meteorological climate, in particular its windiness,
temperature variations, and likelihood of precipitation

61
© TOPKAYA, 2024

Load Arrangements

• Undrifted snow load on roofs: Considered as uniformly distributed.

Affected only by the shape of the roof,
before any redistribution of snow due to
other climatic actions

• Drifted snow load on roofs: Describes the snow load distribution resulting
from snow having been moved from one
location to another location on a roof.
Example: by the action of the wind

62
© TOPKAYA, 2024
 Examples of Undrifted Snow

Photo courtesy of Prof. Gulvanessian

63
© TOPKAYA, 2024

Examples of Drifted Snow

Photo courtesy of Prof. Gulvanessian

64
© TOPKAYA, 2024
 Determining Snow Loads on Roofs

For the persistent and transient design situations:

s Pi u Ce u Ct u sk
For the accidental design situations where exceptional snow load
is the accidental action:

s Pi u Ce u Ct u sAd
Pi Snow load shape coefficient

Ce Exposure coefficient Ct Thermal coefficient

65
© TOPKAYA, 2024

Ce Exposure coefficient

• Depends on the terrain conditions around the structure

• Recommended value equal to 1.0 unless proved otherwise
• Takes into account the effect of surrounding terrain (such as trees)
in blocking the wind and therefore allowing snow accumulation

Windswept Sheltered

Ce 0.8 Ce 1.0 Ce 1.2

Photo courtesy of Prof. Gulvanessian 66
© TOPKAYA, 2024
 Ct Thermal coefficient

• Snow on the roof melts due to heating of the structure underneath

resulting in a reduction in the amount of snow
• Thermal coefficient takes into account the amount of heating
• Recommended value equal to 1.0 unless detailed calculations
prove otherwise
• The use of ISO 4355 is recommended for roof having various
heating properties

67
© TOPKAYA, 2024

Pi Roof shape coefficient

Section 5.3 outlines roof shape coefficients depending on the

type of the roof

68
© TOPKAYA, 2024
 69
© TOPKAYA, 2024

EN 1991-1-4: Wind Actions

EN 1991-1-4 specification can be used to determine wind actions

on structures that are less than 200 meters in height and on bridges
that are less than 200 meters long

EN 1991-1-4 specification can be used to calculate the wind actions

That correspond to an annual probability of exceedance of 0.02
(Return period of 50 years)

The single most important variables used in wind action calculations

are the wind speed and wind pressure
Wind speed is dependent on the regional wind climate and local
factors such as terrain roughness and orography (terrain topography)
70
© TOPKAYA, 2024
 Terrain categories according to EN 1991-1-4

0 Sea, coastal area 1 Lakes or flat area 2 Areas with low

with negligible vegetation
vegetation

3 Area with regular cover 4 Area in which at least 15%

of vegetation or buildings of the surface is covered
with buildings
71
© TOPKAYA, 2024

• Basic wind velocity is needed to determine wind actions

• Characteristic 10-minute mean wind velocity at 10 meter

above ground of a terrain with low vegetation (Terrain Category 2)
• Measurements are used to develop wind maps for a particular region
or country

vb Basic wind velocity

U
qb u vb2 Basic wind velocity pressure
2

U Density of air 1.25kg/m3

72
© TOPKAYA, 2024
 Wind velocity changes with height

Mean wind velocity can be determined as follows:

vm ( z ) cr ( z ) u co ( z ) u vb

vm (z ) Mean velocity at a height of z

cr (z ) Roughness factor

co (z ) Orography factor (Usually taken as 1.0)

73
© TOPKAYA, 2024

cr (z ) Roughness factor

Accounts for the variability of the mean wind velocity at the site
of the structure due to the height above the ground level and
the ground roughness of the terrain

§ z·
cr ( z ) k r u ln¨¨ ¸¸ zmin d z d zmax
© z0 ¹
cr ( z ) cr ( zmin ) z d zmin
0.07 z0 roughness length defined in Table 4.1
§ z · kr terrain factor
kr 0.19¨¨ 0 ¸
¸ zmin the minimum height defined in Table 4.1
© z0, II ¹ zmax=200 meters

74
© TOPKAYA, 2024
 Table 4.1

z0 ve zmin values

75
© TOPKAYA, 2024

Variation of roughness factor cr(z) with the height

above the terrain

A0 A1 A2 A3 A4

250

200

150
z (m)

100

50

0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

cr(z)

76
© TOPKAYA, 2024
 Wind Turbulence
Turbulence should be taken into account in calculations

Turbulence intensity I v (z )
kI
I v ( z) z min d z d z max
§ z ·
co ( z ) u ln¨¨ ¸¸
© z0 ¹
I v ( z) I v ( z min ) z d z min

kI Turbulence factor (Taken as 1.0 unless specified

otherwise in the National Annex)

77
© TOPKAYA, 2024

Variation of Iv(z) factor with the height above the terrain

A0 A1 A2 A3 A4

0.5
0.45
0.4
0.35
0.3
Iv (z)

0.25
0.2
0.15
0.1
0.05
0
0 50 100 150 200 250

z (m)

78
© TOPKAYA, 2024
 Peak Velocity Pressure

Wind velocity and turbulence is taken into account

in calculating peak velocity pressure

q p ( z) >1  7 u I v ( z )@u U u vm2 ce ( z ) u qb

2
q p ( z)
ce ( z ) Exposure factor
qb

79
© TOPKAYA, 2024

Variation of exposure factor with height above the terrain

A0 A1 A2 A3 A4

250

200

150
z (m)

100

50

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5

ce (z)

ce(z)
cr(z)
80
© TOPKAYA, 2024
 WIND ACTIONS
Wind actions depend on the type of structure considered and
the following equation can be used for force calculations

Fw c s u cd ¦c
elements
f u q p ( ze ) u Aref
Reference area

Structural factor Force coefficient Peak velocity pressure

Positive pressure
Negative suction

81
© TOPKAYA, 2024

Structural Factor cs u cd

• Accounts for the effect of non-simultaneous occurrence of peak

pressures over the surface combined with the effect of dynamic
response of the structure or element due to turbulence (See Chapter 6)
• Can be considered equal to 1.0 for most of the simple structures
• Annex B, C, D provides detailed information

An example of a steel building

Values change between 1.10 and 0.85

82
© TOPKAYA, 2024
 Pressure and Force Coefficients cf
• See EN 1991-1-4: Chapter 7 for pressure and force coefficients

• Internal and external pressure coefficients

• Net pressure coefficients
• Friction coefficients
• Force coefficients
• Coefficients for buildings, canopies, circular cylinders etc.

83
© TOPKAYA, 2024

Values

84
© TOPKAYA, 2024
 References
• EN 1990, Basis of Structural Design, July 2001, European Committee
for Standardization, Brussels

• EN 1991-1-1, Actions on Structures – General Actions – Part 1-1:

Densities, self-weight, imposed loads for buildings, July 2001,
European Committee for Standardization, Brussels

• EN 1991-1-3, Actions on Structures – General Actions – Part 1-3:

Snow Loads, July 2003, European Committee for Standardization, Brussels

• EN 1991-1-4, Actions on Structures – General Actions – Part 1-4:

Wind actions, July 2001, European Committee for Standardization, Brussels

• Designers’ Guide to EN 1990 Eurocode: Basis of Structural Design

Gulvanessian H, Calgaro JA, Holicky M, Thomas Telford, 2002

• Designers’ Guide to Eurocode 1: Actions on Buildings

Gulvanessian H, Formichi P, Calgaro JA, Thomas Telford, 2009
85
© TOPKAYA, 2024
 Introduction to
Eurocode 8
Compiled by Topkaya, Akbaş, Zeybek

1
© TOPKAYA, 2024

Calculation of Earthquake Loads According to EC8

Consider a single story structure subjected to ground shaking

u
m
m = mass
c = damping
k = stiffness
k
u = displacement of story

u = velocity of story
u = acceleration of story
ug = acceleration of the ground

acceleration record ug

2
© TOPKAYA, 2024
 The governing differential equation can be written as follows:

mu  cu  ku mug (t )

Solution of the differential equation yields, u(t ) u(t ) u(t )

m u(t ) m F mu(t )

Once u
(t ) is known the forces on the structure can be calculated

The single most important parameter that characterizes

system response is the natural period of vibration.

3
© TOPKAYA, 2024

Consider an undamped system subjected to an initial displacement

u(0)
m

t
k

c=0

Time required for the undamped system to complete one cycle of free vibration

Natural period of vibration

4
© TOPKAYA, 2024
 2S k T 2S
m
T Zn
Zn m k

ωn= natural circular frequency of vibration (radians per second)

T = natural period of vibration (second)

1
The system executes cycles in 1 sec
Tn

1 Zn 1 k
fn
Tn 2S 2S m

fn = cyclic frequency of vibration (cycles per second, Hz, hertz)

For a given earthquake record systems with different periods of vibration
respond differently.
The response can be displayed using a response spectrum.
5
© TOPKAYA, 2024

A typical response spectrum According to EC8

Horizontal Elastic Response Spectrum

u Se
ag

T T
TB TC TD

ª T º ªT º
0 d T d TB Se (t ) ag S «1  K 2.5  1 » TC d T d TD S e (t ) a g SK 2.5« C »
¬ TB ¼ ¬T ¼
ªT T º
TB d T d TC Se (t ) ag SK 2.5 TD d T d 4s S e (t ) a g SK 2.5« C 2D »
¬ T ¼
S = soil factor
η = damping correction
ag = design ground acceleration on type A ground (ag=γI age) 6
© TOPKAYA, 2024
 Seismic hazard and soil conditions are important factors to be considered

agR = reference peak ground acceleration on type A ground

Usually a return period of 475 years is used

For essential facilities the return period can be 2500 years

To take the difference into account the importance factor, γI , is used

In EC8 recommended values of γI can go up to 1.4.

Usually taken as 1.5 in most other specifications.

agR can be conveniently found from maps

Soil classes A, B, C, D, and E are defined in EC8 Table 3.1

Recommended values for S, TB, TC, and TD are also tabulated.
Two types (Type 1, Type 2) of response spectrum are given
where Type 1 is the most suitable for Turkey.

7
© TOPKAYA, 2024

Assume the following conditions:

agR = 0.4 g , γI = 1.0 , soil type D , T = 0.4 sec , Type 1 spectrum, η = 1

For soil D, S = 1.35 , TB = 0.2 , TC =0.8 , TD =2.0 sec

TB d T d TC Se (t ) ag SK 2.5 0.4 u1.35 u 2.5 1.35g

This physically means that the lateral force is equal to mass times 1.35g

1.35g

8
© TOPKAYA, 2024
 It should be uneconomical to design structures to behave elastically
under such high lateral forces.

Usually a reduction of lateral forces is considered and inelastic behavior

is inherently assumed. According to Newmark the displacements of
the elastic and inelastic systems are equal.

Lateral
Force
Elastic Elastic System
Capacity

Reduced Inelastic System

Capacity

Displacement

9
© TOPKAYA, 2024

In EC8 Behavior Factor (q) is used to reduce forces and arrive at the design spectrum.
For steel structural systems q can be between 4 and 6.

Design Spectrum for Elastic Analysis

ª2 T § 2.5 2 ·º
0 d T d TB S d (T ) ag S «  ¨¨  ¸¸»
¬ 3 TB © q 3 ¹¼

2.5
TB d T d TC S d (T ) ag S
q

2.5 ª TC º
TC d T d TD S d (T ) ag S t Ea g
q «¬ T »¼

2.5 ª TCTD º
TD d T S d (T ) ag S t Ea g
q «¬ T 2 »¼

β = lower bound factor. (Recommended value = 0.2)

10
© TOPKAYA, 2024
 Consider a multi degree of freedom system (MDOF)

m3 m3 >M @ u  >C @ u  >K @ u 0

k3

m2 m2 Solution of a system of equations

k2
Eigenvalues and eigenvectors provide
m1 m1
vibration periods and vibration modes.

k1

11
© TOPKAYA, 2024

T1 ,mode 1 T2 ,mode 2 T3 ,mode 3

T1 > T2 > T3

Each of these modes contribute to the response differently.

The most general method is the modal spectrum analysis
which takes into account the contribution of these modes. 12
© TOPKAYA, 2024
 Lateral Force Method of Analysis:
Applied to buildings whose response is not significantly affected
by contributions from modes of vibration higher than the fundamental mode.

u3
Applied when T1 d 4 u TC
m3 F3
T1 d 2 sec

u2 Base shear force

m2 F2
Fb Sd (T1 ) u m u O

u1 m = total mass of the building

m1 F1 λ = correction factor
λ = 0.85 if T1 ≤ 2TC and more than 2-storey

13
© TOPKAYA, 2024

T1 can be found using methods based on structural dynamics or

3
T1 Ct H 4

Ct = 0.085 for moment resisting steel frames

Ct = 0.075 for eccentrically braced frames
Ct = 0.05 for all others

H = height of building in meters

Distribution of horizontal seismic forces

si mi zi mi
Fi Fb Fi Fb
¦s jmj ¦ z jmj
s = displacements of masses in fundamental mode
z = location of masses (used when the fundamental mode is approximated
by horizontal displacements increasing linearly along the height)
14
© TOPKAYA, 2024
 Calculation of masses

¦G k, j "" ¦\ E ,i u Qk ,i

\ E ,i = combination coefficient

\ E ,i M u\ 2i M vary between 0.5 and 1.0

Recall: for office areas \2 0.3

If M 1.0 then G + 0.3Q should be used for calculating masses.

15
© TOPKAYA, 2024
 CE 482

DESIGN OF STEEL STRUCTURES

LECTURE SLIDES

by Cem Topkaya, Ph.D. and Mehmet Emrah Eryaşar, M.S.

Emphasizing AISC360-22

MIDDLE EAST TECHNICAL UNIVERSITY

DEPARTMENT OF CIVIL ENGINEERING
1
© TOPKAYA, 2024

BEAMS

Definition of a beam (also called “flexural member”)

V
M Beam = member subjected to
bending (flexure)
and shear; no axial force
V M

Beams are typically the horizontal members in a building frame

Beams support gravity loads (dead, live, snow etc)

& transfer these loads to the columns.

Beams also participate the resisting lateral loads

(wind, earthquake) in rigid frames.

2
© TOPKAYA, 2024
 Example of Beams resisting Gravity Loads Pinned Beam-
Column
Connection

Rigid Beam-
Column
Connection
Rigid or Braced
Frame
M
Example of Beams Resisting Lateral Loads in a Rigid Frame

M
3
© TOPKAYA, 2024

GENERAL DESIGN STEPS FOR BEAMS (LRFD)

1. Using LRFD load combinations, determine the bending moment, Mu,
and shear force, Vu, along the beam.

wu

wu = factored load

Mu Mu = Required Flexural Strength

Vu
Vu = Required Shear Strength
2. Compute the nominal flexural strength , Mn, for the beam
(= max moment beam can sustain without a flexural failure)
3. Strength Design Requirement:

Mu ≤ ΦMn
Φ = 0.9 for flexure and shear
Vu ≤ ΦVn

4. Check serviceability, if required

Beam deflection under service load, floor vibrations under service loads. 4
© TOPKAYA, 2024
 REVIEW OF BENDING THEORY – BEAMS
Objective: Given a beam subjected to bending moment M
M M Find: Bending Stress: σ
Bending Strains: ε
Curvature: Ø

BASIS FOR ANALYSIS

1. Stress-Strain Law 2. Plane Sections Remain Plane

Assume no σres
Fy
E
ε
εy

Strain varies linearly over height of X-section 5

© TOPKAYA, 2024 (holds for elastic and inelastic behavior)

3. Equilibrium
For stresses on cross section

P ³ V ˜ dA 0 (Fi = σ dA ΣFi = 0)

M ³ y ˜ V ˜ dA (Mi = ΣỹiFi)

(holds for elastic or inelastic behavior)

ELASTIC BEHAVIOR
σmax
M M
C
y ymax C
Neutral Ø
Axis

T T

ε σ = E.ε

Ø = curvature (1/mm)
ε = y.Ø (plane sections remain plane)
6
© TOPKAYA, 2024
 For elastic behavior
M
• Neutral axis is at centroid I
EI

My Mymax Mc
V V max V max
I I I

Define ymax=c Define

I
S (elastic section modulus mm³)
c

Mc M
V max
I S
Limit of elastic behavior (in absence of σres) σmax = Fy

Define My = moment of first yield (When σmax = Fy)

My=SFy
7
© TOPKAYA, 2024

Condition at M = My
εmax = εy = Fy/E Vmax=Fy

Øy=My/(EI)

Strains Stresses

INELASTIC BEHAVIOR (M > My)

εmax > εy Vmax=Fy

εy
Ø

εy

Strains Stresses

8
© TOPKAYA, 2024
 °E ˜ H
­ for H d H y
ε = y.Ø V ®
°̄ Fy for H t H y

P ³ V ˜ dA 0 => this equation no longer simplifies to

³ y ˜ dA 0
the neutral axis is not guaranteed to be at the
centroidal axis for inelastic behavior.

M ³ y ˜ V ˜ dA Ÿ this equation no longer simplifies to

M = EIØ in the inelastic range

My
V does not apply in the inelastic range.
I

9
© TOPKAYA, 2024

Consider bending stresses as curvature is increased:

Fy Fy Fy Fy

Fy Fy Fy Fy

M = My M = Mp

Increasing Ø and M

In the absence of strain hardening, the maximum moment which can be

sustained occurs when every fiber of the cross-section has yielded.

When every fiber has yielded, the section is said to be “fully plastic”. The moment
required to achieve a fully plastic section is defined as Mp, “plastic moment”.

In AISC360, Mp is taken as an upper bound to Mn. That is Mn ≤ Mp

10
© TOPKAYA, 2024
 Observations on Fully Plastic Section
Plastic Neutral Axis (PNA)
The neutral axis for the cross-section in the fully plastic condition is often
referred to as the “plastic neutral axis”. (PNA)
To compute Mp, it is first necessary to find the PNA location.
The PNA is not necessarily at the centroid.

The PNA can be located from the condition that P = 0

P ³ V ˜ dA ³ VAcomp
comp ˜ dA  ³V
Atension
tension ˜ dA 0

For a fully plastic cross section: σcomp = -Fy σtension = +Fy

If Fy is the same for the entire cross-section (i.e. same grade steel used for
all elements of the cross-section) then the above equation can be simplified
P  Fy ³ dA  F ³ dA
Acomp
y
Atension
0

Acomp = Atension

This result indicates that the PNA is located at the equal area axis for
the cross-section. (if Fy is the same for the entire cross-section)
11
© TOPKAYA, 2024

Thus, the PNA can be located by the requirement that the area above
the PNA must equal the area below it.

A1
PNA
Locate PNA so that A1=A2
A2

Beam Cross Section

Observations:
PNA
1. If bending is about an axis of symmetry,
c.g.
then the PNA is at the centroid.
PNA (equal
2. If bending is about an axis which is area axis)
not a axis of symmetry for the
cross-section, then the PNA is not at the centroid. Centroidal axis =
c.g. Neutral axis for
elastic bending

3. If different grades of steel are used for portions

of the cross-section, then the PNA must be
located from the basic requirement that P V ˜ dA ³ 0 12
© TOPKAYA, 2024
 Computing Mp
Mp can be computed from the basic definition of bending moment.

M ³ y ˜ V ˜ dA
A
As long as P = 0, y can be taken from any convenient location.
Measuring y from the PNA is usually most convenient.

For a fully plastic section, σ = Fy. If Fy is the same for the entire cross-section then,

M ³ y ˜ V ˜ dA ³ F
A A
y ˜ y ˜ dA Fy ˜ ³ y ˜ dA
A

Define Z ³ y ˜ dA
A
plastic section modulus
Mp = [Link]
Then

For most beam cross-sections, it is not necessary to compute Z from the integral
expression. Rather, the cross-section can be divided into several simple
shapes (usually rectangles), and the integral can be replaced by a summation:
Z = ΣAi.ỹi Ai = area of the element “i” ỹi = distance from PNA to centroid of Ai (ỹi is always
taken as a positive number) 13
© TOPKAYA, 2024

MOMENT-CURVATURE RELATIONSHIP
In the absence of strain hardening and residual stresses, the moment curvature
relationship for beam cross-section can be qualitatively illustrated as follows:
M

Define “shape factor” = Mp/My

Mp

My
EI
Ø (1/mm)

The shape factor expresses how much additional moment can be carried
by a cross-section after first yield occurs at an extreme fiber,
up to a fully yielded condition

For a rectangular cross-section => shape factor=1.5

For most rolled I-Shapes, the shape factor is within the range of 1.09 to 1.15

14
© TOPKAYA, 2024
 EFFECT OF RESIDUAL STRESSES
The presence of residual stresses cause first yield to occur at an applied
moment less than My. After first yield, the residual stresses cause
a reduction in stiffness (a reduction in the slope of the M-Ø diagram),
which, in turn, may cause earlier buckling of the beams compression flange.
Thus, the effect of residual stresses in beams is analogous to the effect of
residual stresses in columns.

First Yield
Define Mr = moment at the first yield in the presence of residual stresses.
Fr = maximum compressive residual stress in flange.
For design, Fr can be estimated as : 69 MPa for hot rolled shaped
114 MPa for welded built up shapes

First yield occurs when the applied stress (Mr/S) plus the residual stress (Fr) equals Fy
Mr
?  Fr Fy Ÿ M r S ( Fy  Fr )
S
Note that if no residual stress is present then Mr = My
In AISC 2022 Fr is conveniently taken equal to 0.3 Fy ? Mr 0.7 Fy S x
15
© TOPKAYA, 2024

Moment Curvature
M
Example:
No residual stresses
HEM 450
Mp S235 (Fy = 235MPa)
My Zx = 6340cm3 Sx = 5500cm³
Mr With residual stresses
Take Fr = 69MPa for hot rolled shapes
Ø (1/mm)

Mr = 5500(235 – 69) = 913 kN-m Mr=5500×(0.7×235)= 905 kN-m (AISC)

My = 5500×235 = 1293 kN-m
Mp = 6340×235 = 1490 kN-m

A beam section with residual stresses is still capable of resisting the fully plastic
moment Mp. However, first yield occurs earlier (Mr rather than My),
and the slope of the M-Ø diagram is reduced. This reduced slope (stiffness)
results in an increased likelihood of beam instability prior to developing Mp.

16
© TOPKAYA, 2024
 BEAM BEHAVIOR-FAILURE IN FLEXURE
Consider an experiment on a beam:
P
Beam flexural failure

Pmax
P

Mc = PL/4 Δ
+ + Mc
M Beam flexural failure

Pmax.L/4

= Mmax = Mn

17
© TOPKAYA, 2024 Δ

In the experiment shown, a load P is applied to a simply supported beam.

P is plotted against the beam’s deflection Δ.

The load is increased until it reaches a maximum, and begins to drop,

i.e. until the beam collapses. At a load of Pmax, the beam can resist no
additional moment, i.e. the beam experiences a flexural failure.
The maximum bending moment which the beam can sustain without
a flexural failure is defined as Mn, the “nominal flexural strength”.

Mn is determined by considering the following possible flexural failure modes:

1. Local Buckling
Flange Local Buckling (FLB)
Web Local Buckling (WLB)

2. Lateral Torsional Buckling (LTB)

3. Development of Fully Plastic Cross-Section

18
© TOPKAYA, 2024
 When a beam is subjected to a bending moment, a portion of the cross section
is in compression, and a portion is in tension. The portion of the cross-section
in compression can fail by instability. (FLB,WLB or LTB)
A
Compression
M M
buckling
possible

Tension
A
no buckling

A-A
Flange Local Buckling (FLB)
A

M M

A-A 19
© TOPKAYA, 2024 A

Web Local Buckling (WLB)

A
M M

A
A-A
Lateral Torsional Buckling (LTB)
A Lateral Brace

M Torsional
M Displacement
Lateral
Displacement

Lateral
Brace
View looking Down on Beam 20
© TOPKAYA, 2024
 In LTB, the compression flange buckles laterally (like a column) between
lateral braces. The tension flange does not want to buckle, and therefore
resists the lateral movement of the compression flange. The end result is that
the beam displaces laterally and twists; hence the name is “lateral torsional buckling”.
To prevent LTB, the compression flange of the beam must be provided
with lateral braces.
Lb Lateral Brace

M M

The amount of moment a beam can sustain before LTB occurs between
lateral braces depends on the unbraced length, Lb. As Lb is decreased,
the amount of moment a beam can sustain without LTB increases.
(This is analogous to column bucking)

21
© TOPKAYA, 2024

BEAM LATERAL BRACES

In an actual building, lateral bracing for a beam is typically provided
by cross-members or by a concrete floor slab.
A lateral brace must satisfy one of the following requirements:
- prevent lateral displacement of the compression flange
- prevent twist of the cross section
Examples
Compression
Flange
Compression Concrete floor slab
Flange connected to
compression flange of
the girder. The floor slab
provides continuous
lateral bracing for the
Cross Beam: Acts as a lateral compression flange Lb= 0
brace for girder, because it will Tension Flange LTB not possible
prevent lateral displacement of
Tension
the girder’s compression flange
Flange
Note that if the bottom flange of girder was in
compression, the concrete floor slab would not
be a lateral brace, since it cannot effectively
prevent lateral displacement of the bottom flange.

22
© TOPKAYA, 2024
 Concrete Floor Slab

Compression Flange
In this example, lateral displacement of the compression flange is prevented
by the diagonal members (typically angles)
Cross - Frame

In this example, there is no concrete floor slab. The cross frame prevents twist of each girder and
therefore acts as a lateral brace for each girder. Either flange of the girders can be in compression or tension.
23
© TOPKAYA, 2024

RANGES OF BEAM BEHAVIOR IN FLEXURE

Depending on when instability (FLB,WLB or LTB) occurs in the loading
history of a beam, we say flexure has occurred in one of the three possible
ranges of beam behavior.
• Failure in plastic range
• Failure in the inelastic range
• Failure in the elastic range
To illustrate these ranges of behavior, consider an experiment on a beam:
w

Plot midspan moment M versus

beam deflection Δ

24
© TOPKAYA, 2024
 Failure in the Elastic Range
M

For failure in the elastic range,

Mp
instability occurs before the
FLB,WLB or LTB moment reaches Mr.
Mr
The beam is elastic when instability occurs

The beam fails with essentially no ductility

Δ

Failure in the Inelastic Range

M

For failure in the inelastic range,

instability occurs at a moment
FLB,WLB or LTB
Mp between Mr and Mp.

Mr The beam is inelastic when

instability occurs. The beam fails
with little or no ductility.
Δ 25
© TOPKAYA, 2024

Failure in the Plastic Range

M For failure in the plastic range,
instability occurs after the beam
FLB,WLB or LTB
reaches its plastic moment, Mp,
Mp and maintains Mp through large
inelastic deformations. Beams
Mr
that fail in the plastic range exhibit
large ductility.
Δ
Inelastic
deformation
Summary:
Failure Range Mn Ductility

Failure in the elastic range Mn ≤ Mr Little or none

Failure in the inelastic range Mr < Mn < Mp Little or none

Failure in the plastic range Mn = Mp Large

The specification has different equations that govern beam design for each
of these ranges of behavior. 26
© TOPKAYA, 2024
 DEGISN OF BEAMS FOR FLEXURE
At each point along the length of a beam, the following equation must be satisfied:

M u ≤ Φ Mn Φ = 0.9

Mu: moment under factored loads (from structural analysis)

Mn: nominal flexural strength of beam.

­M n based on flange local buckling

°M based on web local buckling
° n
Mn = lesser of ®
°M n based on lateral torsional buckling
°M p ( Z ˜ Fy d 1.5 ˜ M y )
¯

27
© TOPKAYA, 2024

PLATE BUCKLING
Comparison with Column Buckling
Simply
Fcr
Supported

Free Edge

Simply
t
b Supported
1 S²˜E S²˜E
I ˜ b ˜ t³ A b˜t Fcr
12 ( L / r )² 12 ˜ ( L / t )²
1 b ˜ t³ t²
r² ˜ For other boundary conditions
12 b ˜ t 12
S²˜E
Fcr k˜ where k = 1/K² K = effective length
12 ˜ ( L / t )²
28
© TOPKAYA, 2024
 Plate Behavior
Plate buckling occurs in waves with the length of the wave L
equal to the plate width b.

L=b
Simply
If L = b, then the column buckling equation Supported

S²˜E
Fcr
12 ˜ (b / t )²
Simply
Simply
In addition, the plate bends in two directions, Supported
Supported
which improves the buckling strength
slightly by (1/(1-v²) )

v = 0.3 (Poisson’s Ratio)

Simply
S²˜E Supported
Fcr k˜
12 ˜ (1  v ²) ˜ (b / t )² k is a factor that accounts for the edge boundary
conditions and loading condition.
b is always the shortest plate dimension.
29
© TOPKAYA, 2024

Theoretical Solutions for k

Plates Under Compression Fixed Fixed

k=6.97 k=5.42
Fixed Simply Supported

Simply Supported Fixed Simply Supported

k=4.00 k=1.277 k=0.425

Simply Supported Free Free

Plates Under Bending

Simply Supported Fixed

k=23.9 k=39.6
Simply Supported Fixed

Plates Under Shear

Simply Supported Edges

k = 5.34 + 4/(a/b)²
b

30
© TOPKAYA, 2024 a
 Various Solutions Given in AISC – LRFD
Local Flange Buckling (Sections with slender flanges) (Sec F3.2)

0.9 ˜ E ˜ kc
Mn S x Fcr Fcr kc 4 / (h / t w ) 0.35 < kc < 0.76
(b / t )²

0.9 ˜ E ˜ kc k ˜S ² ˜ E
k = 0.995 kc
(b / t )² 12 ˜ (1  0.3²) ˜ (b / t )²

0.348 < k < 0.756 For slender webs, the k factor can be less than pinned because
the web can buckle, which will bend the flange.
Web Flexural Buckling
What web slenderness ratio will permit Fcr = Fy ?
S²˜E h k ˜S ² ˜ E
Fcr Fy k AISC uses
12 ˜ (1  0.3²) ˜ (h / t w ) 2
tw 12 ˜ (1  0.3²) ˜ Fy
E
h E 5.70 ˜
0.95 ˜ k ˜ Pinned k = 23.9 Fixed k = 39.6 Fy
tw Fy h E h E
4.64 ˜ 5.98 ˜ (close to Fixed Case)
tw Fy tw Fy 31
© TOPKAYA, 2024

Mn Based on Flange&Web Local Buckling

LRFD Specification: Section B4
In general, Mn based on FLB and WLB depends on two variables
• Width-thickness ratio of flange or web
• Fy
The LRFD Specification uses the symbol “λ” as a generic
symbol for width-thickness ratio.
For a wide-flange shape, the values of λ are defined as follows:

bf h
For FLB: O For WLB: O
2t f tw

h Values for bf/2tf and h/tw are

tw tabulated for rolled shapes.

tf

bf

32
© TOPKAYA, 2024
 There are width-thickness ratios (b/t limits) for compression elements (plates)

33
© TOPKAYA, 2024

AISC – LRFD uses the symbol λr to denote the width-thickness limit

above which the compression element can reach yield stress Fy
without local buckling occurring.

AISC –LRFD uses the symbol λp to denote the width-thickness limit

above which the compression element can reach yield stress Fy
and undergo large plastic strain without local buckling occurring
(For example flanges of beams)
To determine Mn based on WLB and FLB, the value of λ is compared
to λp and λr are defined as follows.

bf E E
FLB: O Op 0.38 O r 1 .0
2t f Fy Fy

h E E
O Op 3.76 Or 5.70
WLB: Fy
tw Fy

34
© TOPKAYA, 2024
 bf E h E
For FLB: d 0.38 For WLB: d 3.76
2˜tf Fy tw Fy
λ ≤ λp
Called “COMPACT” Sections

For this case, failure by local buckling occurs in the plastic range. That is,
local buckling does not occur until the beam reaches Mp, and maintains Mp through
large inelastic deformations.
E bf E E h E
λp ≤ λ ≤ λr FLB: 0.38 d d 1.0 WLB: 3.76 d d 5.70
Fy 2 ˜ t f Fy Fy t w Fy

Called “NONCOMPACT” Sections

For this case, failure by local buckling occurs in the inelastic range.
Local buckling occurs at a moment between Mr and Mp

λ > λr bf
! 1 .0
E h
! 5.70
E
FLB: WLB:
2˜t f Fy tw Fy

Called “SLENDER” Sections

For this case, failure by local buckling occurs in the elastic range.
Local Buckling occurs at a moment less than Mr. 35
© TOPKAYA, 2024

Mn based on FLB or WLB

Mp

Mr

0 λp λr λ

Compact Sections Non-Compact Slender Element Sections Failure

Failure in Plastic Sections in the Elastic Range
Range Failure in the
inelastic range

Practical Design Considerations:

Most of the rolled I-Shapes are compact for FLB and WLB at Fy = 235MPa or 355 MPa.
Thus, in most cases, local buckling will not be a design concern, since Mn = Mp for compact sections. 36
© TOPKAYA, 2024
 Mn Based on Lateral Torsional Buckling
LRFD Specification: Chapter F
Restriction: We will restrict our considerations Moment applied about x-axis
of LTB to I-shaped members bent about
their major axis. X X
For other types of cross-sections,
equations for Mn based on LTB are given
in the LRFD Specification. (Chapter F)

Some Special Cases of Interest

I-shape bent about its Circular sections bent Box sections bent about
minor axis about any axis either principal axis
Moment
applied about y
y-axis
y y

x x

Moment applied
For each of these special cases, LTB cannot occur. y about x-axis or y-
Therefore Mn = Mp based on LTB axis 37
© TOPKAYA, 2024

Mn for I-Shaped Sections with Major Axis Bending

A theoretical solution of critical moment for the constant moment case is available.
M M

Lb

S S ˜E
M cr ˜ E ˜ I y ˜G ˜ J  ( )² ˜ Cw ˜ I y
Lb Lb
To adjust for moment gradient, the above equation could be multiplied by a factor, Cb

S S ˜E
M cr Cb˜ ˜ E ˜ I y ˜G ˜ J  ( )² ˜ Cw ˜ I y
Lb Lb

38
© TOPKAYA, 2024
 In general, Mn based on LTB depends on the following variables:
1. Lb = distance between lateral braces (unbraced length)
2. Shape of the moment diagram between lateral braces.
This is represented by the factor Cb.
3. Fy
4. Beam section properties.

LTB depends primarily on following section and material properties:

E = modulus of elasticity of steel = 200 GPa
G = elastic shear modulus of steel = 77.2 GPa
Sx= elastic section modulus (mm3)
Iy = moment of inertia about y-axis (mm4)
J = St. Venant’s torsion constant (mm4)
Cw = Warping torsion constant (mm6)

1
J
3
¦ bt 3 b= width of plate t=thickness of plate

I y ho2 For I sections ho= distance between the flange centroids

Cw
4
39
© TOPKAYA, 2024

For a given beam section and grade of steel, Mn based on LTB will depend
on Lb and the shape of the moment diagram within the unbraced span.

The basic equation for LTB is written for the case of uniform moment between
lateral braces. For any other shape of moment diagram, these basic equations are
modified by the factor “Cb”.

The value of Cb depends on the shape of the moment diagram

between lateral braces. Cb is dimensionless and varies between 1.0 and about 2.4

ALTERNATE FORMS OF Mcr

Recall
S S ˜E
M cr ˜ E ˜ I y ˜G ˜ J  ( )² ˜ Cw ˜ I y
Lb Lb

St. Venants Term Warping Term

40
© TOPKAYA, 2024
 M cr Fcr S x Fcr = critical stress

S 2E J L
Fcr 1  0.078 ( b )2 I yCw
L S x ho rts rts2
( b )2
rts Sx

Warping Term St. Venants Term

If St. Venants Term is neglected

S 2E
Fcr Can be extremely conservative
L
( b )2
rts

Simplified rts bf Radius of gyration of the

rts compression flange plus one-sixth
§ 1 htw ·
12  ¨1  ¸ of the web
¨ 6b t ¸
© f f ¹ 41
© TOPKAYA, 2024

Uniform Moment Between Lateral Braces (Cb = 1)

Given: Beam Section, Grade of Steel, Lb, Uniform moment within Lb
Find: Mn based on LTB for this unbraced span.

Compression Beam
Flange lateral
brace

Lb

Uniform moment between

lateral braces

42
© TOPKAYA, 2024
 Variation of Mn with Lb:

Mn based on LTB

Mn = Mp
Mp Mn per (EQN F2-2)

Mr
Mn per (EQN F2-4)

0 Lp Lr Lb

LTB in the plastic LTB in the LTB in the elastic range

range inelastic
range
43
© TOPKAYA, 2024

Definition of Terms:
E
Mp = Zx Fy Mr = 0.7 Fy Sx Lp 1.76 ˜ ry ˜ (EQN F 2-5)
Fy

2
E J § J · 0.7 Fy 2 (EQN F 2-6)
Lr 1.95rts  ¨¨ ¸¸  6.76( )
0.7 Fy S x ho S h
© x o¹ E
If Lb ≤ Lp:
Mn = Mp (Note: if beam design is based on plastic analysis, need Lb ≤ Lpd)

If Lp < Lb ≤ Lr :
ª Lb  L p º (EQN F 2-2)
Mn « M p  ( M p  0.7 Fy S x )( )» d M p Note that this is the equation of a straight line.
¬« Lr  L p ¼»

If Lb > Lr:
S 2E J L (EQN F 2-3)
Mn Fcr S x Fcr 1  0.078 ( b ) 2 (EQN F 2-4)
Lb 2 S x ho rts
( )
rts 44
© TOPKAYA, 2024
 Non-Uniform Bending Between Lateral Braces
Given: Beam Section, Grade of Steel, Lb, Shape of the moment diagram within Lb
Find: Mn based on LTB for this unbraced span.

Compression Beam
Flange lateral
brace

Solution:
Let MnCb=1 = max moment beam could sustain
without LTB for uniform moment within Lb.

Lb Then: For any moment diagram with Lb:

M Mn = Cb [MnCb=1] ≤ Mp
Any shape moment diagram and
between lateral braces ΦMn = Cb [ΦMnCb=1] ≤ ΦMp

Cb: factor that accounts for the shape of the moment

diagram between lateral braces.
45
© TOPKAYA, 2024

Computing Cb

12.5 ˜ M max
Cb (EQN F1-1)
2.5 ˜ M max  3 ˜ M A  4 ˜ M B  3 ˜ M C

Mmax = absolute value of maximum moment in the unbraced beam segment

MA = absolute value of moment at quarter point of the unbraced beam segment
MB = absolute value of moment at centerline of the unbraced beam segment
MC = absolute value of moment at the three-quarter point of the unbraced beam segment

MC

MA Mmax MB

Lb/4 Lb/4 Lb/4 Lb/4

Unbraced beam
segment = Lb 46
© TOPKAYA, 2024
 Comments on Cb
The value of Cb reflects the degree of moment gradient within the unbraced span.
As the moment gradient increases, Cb increases, and Mn based on LTB increases.

Moment Diagram Cb

1.0
M
Increasing Moment
Gradient
M
1.25 Increasing Cb
M/2

1.67 Increasing Mn
0

M
2.3

-M
47
© TOPKAYA, 2024

The worst case of LTB is uniform moment. For this case, entire unbraced
length of compression flange is subject to the maximum moment and
maximum compressive stress. If any moment gradient is present,
the maximum compressive stress in the flange occurs at only one point
within the unbraced span, and reduces as you move away from this point.
This lower compressive stress in the flange increases resistance to LTB.

An interpretation of Cb:

Mmax Mmax/Cb

Mmax/Cb = equivalent uniform moment for LTB

analysis

48
© TOPKAYA, 2024
 Computing Mn for Any Moment Diagram Mn = Cb [MnCb=1] ≤ Mp
Mn ØMn = Cb [ØMnCb=1] ≤Ø Mp

[Link]

Mp

[Link]
Mr

Cb > 1
Cb = 1

Lb
0 Lp Lm Lr

Mn = Mp
Lm = maximum unbraced length for which Mn = Mp based on lateral torsional buckling
Lm depends on Cb: For Cb = 1 Lm = L p
49
For Cb > 1 Lm > L p
© TOPKAYA, 2024

So far we have dealt with flexural members which were compact with respect
to FLB and WLB. Our discussion will continue with members having
slender webs and flanges.
Web Local Buckling
Let’s recall the expressions we derived earlier
Simply Supported
S ²E Fixed
FCR k
12(1  v ²)(b / t )² k = 23.9
k = 39.6

Simply Supported
Fixed
What web slenderness ratio will permit Fcr = Fy?
FCR Fy k
S ²E k ˜S ²E E
(h / t w )² h / tw 0.95 k
12(1  0.3²)(h / t w )² 12(1  0.3²) Fy Fy
E E
Pinned k = 23.9 h / tw 4.64 Fixed k = 39.6 h / tw 5.98
Fy Fy
AISC360 uses
E Beams with compact or
h / t w d 5.70
E Fy Noncompact webs
5.70
Fy
E
(close to the fixed case) h / t w ! 5.70 Beams with slender webs
Fy 50
© TOPKAYA, 2024
 Doubly Symmetric I-Shaped Members with Compact webs and E
h / t w d 3.76
Noncompact or Slender Flanges (Section F3) Fy
Lateral Torsional Buckling
Same as in Section F2.2

Compression Flange Local Buckling

E bf E
For noncompact flanges 0.38 d d 1.0
Fy 2t f Fy
ª O  O pf º
Mn «M p  ( M p  0.7 Fy S x )( )»
«¬ Orf  O pf »¼

For slender flanges

0.9 Ek c S x bf
Mn O kc 4 / (h / t w ) 0.35 < kc < 0.76
O2 2t f

λpf= the limiting slenderness for a compact flange

λrf= the limiting slenderness for a noncompact flange 51

© TOPKAYA, 2024

Other I-Shaped Members with Compact and Noncompact Webs (Section F4)
Compression
hp/2
PNA hc/2
ENA

Doubly Symmetric Shapes

Tension
E h E
3.76 d d 5.70 Singly Symmetric Shapes
Fy t w Fy
hc E
h p Fy E
Op 2
d Or Or 5.70
§
¨ 0.54
Mp · Fy
 0.09 ¸
¨ M ¸
© y ¹
Consider: O  Op
Compression Flange Yield
Lateral Torsional Buckling O p d O d Or
Compression Flange Local Buckling
Tension Flange Yielding
52
© TOPKAYA, 2024
 Compression Flange Yield Mn R pc M yc R pc Fy S xc
Rpc: web plastification factor
Myc, Myt: yield moment referred to compression and tension flanges
Sxc, Sxt: elastic section modulus referred to compression and tension flanges
Lateral Torsional Buckling

Lb d Lp No LTB

ª § Lb  Lp ·º
Lp d Lb d Lr M n Cb « R pc M yc  ( R pc M yc  FL S xc )¨
¨
¸» d R pc M yc
¸
«¬ L  L
© r p ¹»¼

Lb ! Lr Mn Fcr S xc d Rpc M yc M yc Fy S xc

CbS 2 E J Lb 2 S xt
Fcr 1  0.078 ( ) t 0 .7 FL 0.7 Fy
Lb 2 S h rt S xc
( ) xc o
rt S xt S xt
 0 .7 FL Fy t 0.5Fy
I yc S xc S xc
For d 0.23 J=0
Iy 53
© TOPKAYA, 2024

Critical Lengths
2
E E J § J · F
Lp 1.1rt Lr 1.95rt  ¨¨ ¸¸  6.76( L ) 2
Fy FL S xc ho © S xc ho ¹ E

Web Plastification Factor, Rpc

hc Mp Mp Z x Fy d 1.6S xc Fy
d O pw R pc
tw M yc
hc ªMp Mp O  O pw º M p hc
! O pw R pc « (  1)( )» d O
tw «¬ M yc M yc Orw  O pw »¼ M yc tw

Effective radius of gyration for LTB

b fc hc t w
rt aw
1 b fct fc
12(1  aw )
6
54
© TOPKAYA, 2024
 Compression Flange Buckling

Compact Flanges: Local buckling does not apply

ª § O  O pf ·º
Noncompact Flanges M n « R pc M yc  ( R pc M yc  FL S xc )¨¨ ¸»
¸»
¬« © Orf  O pf ¹¼

0.9 Ek c S x b fc
Slender Flanges Mn O
O2 2t fc
Tension Flange Yielding

For S xt t S xc Tension flange yielding does not apply

For S xt  S xc Mn Rpt M yt M yt Fy S xt
hc Mp
d O pw R pt
tw M yt
hc ªMp Mp O  O pw º M p
! O pw R pt « (  1)( )» d
tw ¬« M yt M yt Orw  O pw ¼» M yt

55
© TOPKAYA, 2024

Vertical Flange Buckling

Compression flange can buckle vertically if the restraint from the web is
lost due to instability. Develop a limit to prevent vertical flange buckling

When a girder is bent, the bending force on the flange has a vertical
component that acts on the web.

εf : flange unit strain

dΘ Afσf ε[Link] : deformation
εxdx
accumulated over dx

Afσf
dΘ

dx Afσf
h/2

Afσf dΘ
dΘ dΘ
Afσf

56
© TOPKAYA, 2024
 h 2H f
H f ˜ dx dT ˜ dT ˜ dx
2 h

Vertical force on the web : A f ˜ V f ˜ dT

Dividing by the area t w ˜ dx

A f ˜ V f ˜ dT A f ˜ V f 2H f A f ˜ V f ˜ 2H f
Fc = compressive stress = dx FC
t w ˜ dx t w ˜ dx h tw ˜ h

dΘ S ²E
FCR k
12(1  v ²)(b / t )²

b=h t = tw k=1

dx
S ²E
FCR
12(1  v ²)(h / t w )²
dΘ
57
© TOPKAYA, 2024 Fc

A f ˜ V f ˜ 2H f S ²E h S ²E § Aw ·§ 1 ·
tw ˜ h 12(1  v ²)(h / t w )²
twh = Aw ¨ ¸¨ ¸
tw 24(1  v ²) © A f ¸¹¨© V f ˜ H f
¨ ¸
¹
Suppose flange will reach yield stress Fyf and residual
stresses
Fr Fr

Fyf+Fr
Fr + =

Fr
Fyf Fyf (Final State)
( Fyf  Fr ) Substitute Vf Fyf
Hf
E

h S ²E § Aw · E S² § Aw · E
¨ ¸ ¨ ¸
tw ¨ ¸
24(1  v ²) © A f ¹ Fyf ( Fyf  Fr ) 24(1  v ²) ¨© A f ¸¹ Fyf ( Fyf  Fr )
58
© TOPKAYA, 2024
 Basler suggested that Aw/Af is rarely below 0.5
Fr 0.3Fyf
Therefore, the code equations are as follows:

§ h · 0.40E
For a/h > 1.5 ¨
¨t ¸
¸
© w ¹ max Fy

For a/h ≤ 1.5 (presence of stiffeners allow a higher h/tw)

§h· E
¨¨ ¸¸ 12
© t w ¹ max Fy

See section F13.2 on proportioning limits for I-shaped Members

For a/h > 1.5 (h/tw)max = 340 for S235 and 225 for S355

For a/h < 1.5 (h/tw)max = 350 for S235 and 284 for S355

(h/tw) < 260 for unstiffened girders

59
© TOPKAYA, 2024

Doubly Symmetric I-Shaped Members with Slender Webs

h E
Let’s investigate the behavior of the web for ! 5.70
tw Fy

According to the theoretical solution

S ²E
32.5E
FCR k
If k value of 36 is taken FCR
12(1  0.3²)(h / t w )² (close to fixed case) (h / t w )²
This relationship suggests that the moment capacity would be significantly
reduced for sections with high h/tw ratios.
FCR

Fy

h/t 60
h E
5.70
© TOPKAYA, 2024 tw Fy
 However the observations from the tests are different from our initial
theoretical predictions

MN

Mp

My
10 – 20%
Bend reduction
buckling from My
may occur
No bend
buckling

h/tw
(h/tw)max

61
© TOPKAYA, 2024

Test results revealed that the girders could reach to higher moment capacities
than predicted. This is due to the post buckling behavior of plates.
Unlike columns, plates have significant post buckling capacity.

P P
Post buckling
Post buckling capacity
capacity

Δ Δ
Column Plate

The variation of moment capacity is observed to be linear between

h E
5.70 to (h/tw)max
tw Fy

62
© TOPKAYA, 2024
 Let’s develop a relationship to quantify the capacity in this region
Take (h/tw)max=330

Basler suggested this effective section.

30tw 1/11h
(1-k)h h/tw = 330
NA
10/11h
kh

Location of the neutral axis (static moments about NA)

t w (k ˜ h)² 1 ª 1 º divide by Af.k
Af k ˜ h  A f (1  k )h  h ˜ t w «(1  k )h  h
2 11 ¬ 22 »¼

twh 1 t w h 21 tw h = Aw aw = Aw/Af
k  k² (1  k )  (  k)
2 Af 11 A f 22
63
© TOPKAYA, 2024

4 2 2 21 4 26 22 ª 2 1 º
k  k ²(  )(  ) 0 k   «  »
aw 11 aw 121 aw ² 11aw 121 ¬ aw 11¼

Effective moment of inertia

1 1 21
Ie A f (kh)²  (kh)³t w  A f (1  k )²h²  h ˜ t w ( h  kh)²
3 11 22

ªa a 21 º
Ie A f h² « w k ³  k ²  (1  k )²  w (  k )²»
¬3 11 22 ¼
If Mn/My is plotted with respect to aw
Fy I e (awmax = 3)
Mn Mn
(1  k )h 1 0.082aw
My
Aw aw Mn/My
My Fy ( A f  )h Fy A f h(1  )
6 6
1
aw a 21
k ³  k ²  (1  k )²  w (  k )²
Mn 3 11 22
My aw
(1  k )(1  )
6
64
3 aw
© TOPKAYA, 2024
 @ h E
5.70 Mn = My
tw Fy

h
@ 330 Mn = My[1-0.082aw]
tw
0.082aw
Slope for S235 0.0005aw
330  166

Mn Aw h E
1.0  0.0005 (  5.70 )
My Af tw Fy

ª A h E º
Mn S x Fy «1.0  0.0005 w (  5.70 )»
«¬ Af tw Fy »¼
This equation was further modified to accommodate cases up to aw = 10
ª aw h E º
MN S x Fy «1.0  (  5.70 )»
«¬ 1200  300aw t w Fy »¼

ª aw h E º
Mn R pg Fy S x R pg «1.0  (  5.70 )» d 1.0
«¬ 1200  300aw t w Fy »¼ 65
© TOPKAYA, 2024

Nominal Moment Strength (Mn) for Beams with Slender Webs

The flexural strength is calculated as follows:
Mn is the lower value obtained from the limit states of tension flange yielding,
compression flange yielding, lateral torsional buckling, and compression flange
local buckling.

Tension Flange Yielding

When S xt t S xc Tension flange yield does not apply

When S xt d S xc Mn Fy S xt

Compression Flange Yielding

ª aw h E º
Mn R pg Fy S xc R pg «1.0  (  5.70 )» d 1.0
«¬ 1200  300aw t w Fy »¼
hc t w
aw d 10
b fc t fc

Sxc : section modulus referred to compression flange

Sxt : Section modulus referred to tension flange 66
© TOPKAYA, 2024
 Lateral Torsional Buckling Mn R pg Fcr S xc
When Lb d Lp LTB does not apply

ª Lb  L p º
When Lp d Lb d Lr Fcr Cb « Fy  0.3Fy ( )» d Fy
«¬ Lr  L p »¼

CbS 2 E
When Lb ! Lr Fcr d Fy
Lb 2
( )
rt
E E
Lp 1.1rt Lr Srt
Fy 0.7 Fy

b fc hc t w
rt aw
1 b fct fc
12(1  aw )
6

67
© TOPKAYA, 2024

FCR
ª °º
° Lb  L p ½
­
Cb Fyf «1  0.3® ¾»
«¬ °̄ Lr  L p°
¿»¼
Fyf

C bS 2 E
0.7CbFyf L
( b )2
rt

Lb
E E
Lp 1.1rt Lr Srt
Fy 0.7 Fy

68
© TOPKAYA, 2024
 Compression Flange Local Buckling

Mn R pg Fcr S xc
bf E
When d 0.38 Compression Flange Local Buckling does not apply
2t f Fy

E bf k E ª O  Op º
When 0.38 d d 0.95 c Fcr « Fy  0.3Fy ( )»
Fy 2t f FL ¬« Or  O p ¼»

bf k E
0.9 Ekc
t 0.95 c Fcr
When 2
2t f FL § bf ·
¨ ¸
¨ 2t f ¸
© ¹
4
kc
h
tw

69
© TOPKAYA, 2024

FCR
ª ­° O  O p ½°º
Fyf «1  0.3® ¾»
«¬ °̄ Or  O p°
¿»¼
Fyf
0.9 Ek c
b
0.7Fyf ( f )2
2t f

bf
O
Op 0.38
E
Or 0.95
E 2t f
Fy FL / kc

70
© TOPKAYA, 2024
 Web Behavior under Shear
If the web is too slender, it can buckle under the applied shear stresses.
The theoretical buckling of plates under shear was explained before.

Simply Supported Edges

b 4
k 5.34 
(a / b)²

a
S ²E S ²E short dimension
FCR k W CR kv ˜ k 5.34  4( )²
12(1  v²)(b / t )² 12(1  v ²)(short dimension / t )² long dimension

For stiffened girders two cases are possible

a/h ≤ 1

S ² E (5.34  4.0(a / h)²)

h W CR
12(1  v ²)(a / t )²

a 71
© TOPKAYA, 2024

a/h > 1
S ² E (5.34  4.0(h / a)²)
W CR
h 12(1  v ²)(h / t )²

a
In order to use (h/t) as the stability criterion multiply and divide the first equation by (h/a)².

a/h ≤ 1 a/h > 1

S ² E (5.34 /(a / h)²  4.0) S ² E (5.34  4.0(h / a)²)

W CR W CR
12(1  v²)(h / t )² 12(1  v ²)(h / t )²

Therefore
S ² E ˜ kv 5.34 4.0
W CR kv  4.0 For a / h d 1 kv  5.34 For a / h ! 1
12(1  v ²)(h / t )² (a / h)² (a / h)²

These two equations are replaced with one equation in AISC360

5.0 For a/h > 3

kv 5.0 
(a / h)² girder is assumed unstiffened and kv = 5.34
72
© TOPKAYA, 2024
 For design the buckling equation is put into nondimensional form
W CR S ² E ˜ kv 1.51E ˜ kv
CV 1
Wy 0.6 Fy 12(1  v ²)(h / t w )² Fy (h / t w )²

Depending on the type of buckling two regimes are present

1. Plastic
2. Inelastic

h k E Then Cv1 = 1
For d 1.10 v
tw Fyw
Vn 0.6 Fy AwCv1 0.6 Fy Aw

kv E
h k E 1.10
! 1.10 v Fy
For CV 1
tw Fy (h / t w )

kv E
1.10
Fy
Vn 0.6 Fy Aw ( ) 73
(h / t w )
© TOPKAYA, 2024

Vn
Fy Aw

0.6 For
Unstiffened
(kv = 5.34)

All rolled
sections

0 E h/tw
2.54
Fy

74
© TOPKAYA, 2024
 Nominal Shear Strength – Including Tension Field Action
A typical web behavior under shear was explained before
Post Buckling
Strength

1.0

0.8

W CR
CV 2
Wy

h/tw

Observations from tests revealed that webs have higher shear carrying
capacity than expected due to their post buckling strength.

75
© TOPKAYA, 2024

The girder will act like a truss under the applied loads. Buckled
Web will buckle under high shear but tension struts compression strut
continue to carry the excess load.

Tension
struts

Compression elements (stiffeners)

The nominal buckling strength is the sum of the buckling and post-buckling strengths.

Vn VCR  VTF VCR CV 2W y Aw

76
© TOPKAYA, 2024
 Consider a typical panel. Tensile stresses occur in a band having a width s.

s h
σt

From Geometry
s γ h
S = hcos γ - asin γ

a 77
© TOPKAYA, 2024 asin γ
The partial shear force in the stiffener is

'VTF V t s ˜ t w sin J d'VTF

For maximum 'VTF dJ
0
'VTF V t ˜ t w (h cosJ  a sin J ) sin J
d'VTF
V t ˜ t w (h cos 2J  2a sin J cos J ) 0
h dJ
'VTF V t ˜ t w ( sin 2J  a sin ²J )
2 h 1 1
tan 2J sin 2J
Consider the FBD suggested by Basler to find out VTF a a/h 1  (a / h)²

asin γ
'F f (V t ˜ t w ˜ a sin J ) cosJ
σt tw ˜ a
O 'F f Vt ˜ sin 2J
2
6M o 0
h VTF
VTF/2 'F f  a 0
VTF/2 2 2
h/2
VTF a tw ˜ a h
Vt ˜ sin 2J
Fw Fw 2 2 2

'Ff+Ff Ff twh ª 1 º
a VTF Vt ˜ « »
2 ¬« 1  (a / h)² ¼»
ΔFw is omitted because web contributes little to the flexural strength 78
© TOPKAYA, 2024
 The energy of distortion theory could be approximated as follows for the action of σt and τCR together

Vt W CR ª Vt º
1 1  CV 2 Therefore, Vn h ˜ t w «W y CV 2  »
Fy Fy / 3 «¬ 2 1  (a / h)² »¼

ªC 1  CV 2 º
Vn Fyw ˜ h ˜ t w « V 2  » Approximating Fy / 3 as 0.6Fy
«¬ 3 2 1  ( a / h)² »¼

ª 1  CV 2 º
Vn 0.6 Fyw Aw «CV 2  »
«¬ 1.15 1  ( a / h)² »¼

TFA Term

TFA is not permitted if a/h > 3

79
© TOPKAYA, 2024

AISC Approach for TFA

h k E
For d 1.10 v Then Cv2 = 1 Vn 0.6 Fy AwCv 2 0.6 Fy Aw
tw Fyw

h kE
For ! 1.10 v
tw Fyw
When all the following geometrical conditions are satisfied
2 Aw h h
d 2.5 d 6.0 d 6.0
( Afc  Aft ) b fc b ft
ª 1  CV 2 º
Vn 0.6 Fy Aw «CV 2  »
¬« 1.15 1  ( a / h )² ¼»
Otherwise
ª 1  CV 2 º
Vn 0.6 Fy Aw «CV 2  »
«¬ 1.15 a / h  1  ( a / h)² »¼
80
© TOPKAYA, 2024
 Web shear buckling coefficient

h kE
For d 1.10 v Then Cv2 = 1
tw Fy

kv E
kE h kE 1.10
For 1.10 v  d 1.37 v Fy
Fy tw Fy CV 2
(h / t w )

h kv E 1.51˜ kv ˜ E
For ! 1.37 CV 2
tw Fy (h / t w )² Fyw

81
© TOPKAYA, 2024

TFA for END PANELS

Not allowed prior to 2022 – Research demonstrated that this approach

is conservative

82
© TOPKAYA, 2024
 ª 1  CV 2 º
TFA for END PANELS Vn 0.6 Fy Aw «CV 2  E v »
¬« 1.15 1  ( a / h )² ¼»
2.8 M pf  M pm  M pst  M pm
Ev d 1.0
h Fywt w 1  Cv 2

83
© TOPKAYA, 2024

Intermediate Transverse Stiffeners

Requirements to Omit Intermediate Stiffeners
An unstiffened plate girder must have an h/tw value less than 260.
If the required strength Vu is less than the factored nominal capacity
(based on kv = 5.35) no stiffeners are required.

h / t w d 260 Vu d I 0.6 Aw FywCV 1

Requirements to Include Intermediate Stiffeners
§b· E Width thickness ratio of stiffener
¨ ¸ d 0.56 ts ts
© t ¹ st Fyst

I st t I st 2  ( I st1  I st 2 ) U w ls
ls

1
Ist : moment of inertia of the I t s ls ³
cross sectional area of a
3
transverse stiffener taken about
the center of the web. 1
I t s ls ³
12 84
© TOPKAYA, 2024
 1.5
h 4 U st1.3 § Fyw ·
I st1 ¨ ¸
40 ¨© E ¸¹

ª 2.5 º 3
I st 2 « 2
 2 »bp t w ! 0.5bp t w ³
¬ah ¼

bp: is the smaller of the dimensions

a and h

Fyst: specified minimum yield

stress of stiffener material

85
© TOPKAYA, 2024

Efficient Girder Design

Develop practical formulas for Mp and My
Start the design by assuming d = h
ª Aw h º
tf Mp « A f h  2 ˜ 4 ˜ 2» Fy
¬ ¼

ª hº
d h Mp «¬ A f h  Aw ˜ 4 »¼ Fy
tw (Web worths only 1/4)

b
I ˜ M p t M max
M max ª hº
Af b ˜ t f (Area of Flange) I
Mp «¬ A f h  Aw ˜ 4 »¼ Fy

M max ª Aw º
Aw h ˜ tw (Area of web) I
Mp « A f  4 » hFy
¬ ¼

M max A
Af  w Mn Mp
I ˜ h ˜ Fy 4
86
© TOPKAYA, 2024
 What if Mn Mr S X ˜ Fy
t w ˜ h³ Ix
Ix 2 ˜ A f (h / 2)²  SX
12 (h / 2)

t w ˜ h² Aw h
Sx Af h  Af h  (Web worths only 1/6)
6 6
S x Aw
Af 
h 6

Mn Mr I ˜ M r t M max I ˜ S x Fy t M max

I ˜ S x Fy M max
M max Aw h
Sx Af h 
I ˜ Fy 6

M max A
Af  w
I ˜ h ˜ Fy 6
87
© TOPKAYA, 2024

Design of Girders with Compact Webs

h E
h
3.76 t min
For compact web tw Fy 3.76 E / Fy
M max A ATOTAL 2 Af  Aw
Mn = Mp Use Af  w
I ˜ h ˜ Fy 4
2 ˜ M max 2 Aw 2 ˜ M max Aw 2 ˜ M max h ˜ t w
ATOTAL   Aw  
I ˜ h ˜ Fy 4 I ˜ h ˜ Fy 2 I ˜ h ˜ Fy 2
Set
2 ˜ M max h h 2 ˜ M max h²
ATOTAL  ˜  dATOTAL
I ˜ h ˜ Fy 2 3.76 E / Fy I ˜ h ˜ Fy 7.52 E / Fy 0
dh
 2 ˜ M max ˜ h 2 h
 0
I ˜ Fy 3.76 E / Fy
h 2 ˜ M max 7.52 E / Fy M max
3.76 E / Fy I ˜ Fy ˜ h 2 h³
I ˜ Fy
7.52 E M max 7.52 E 1 / 6 ˜ M max
1/ 3

h³ heff ( )1 / 3 ˜
I ˜ ( Fy ) 3 / 2 I Fy

1/ 3
E 1 / 6 ˜ M max
heff 2.029 ˜
Fy 88
© TOPKAYA, 2024
 Estimated Dead Weight of a Member (Compact Section)

1/ 3
2 ˜ M max h² E 1 / 6 ˜ M max
ATOTAL  heff 2.029 ˜
I ˜ h ˜ Fy 7.52 E / Fy Fy

2M max Fy 1 2.029² ˜ E 1 / 3 ˜ M max

2/3
AT ˜  ˜
I ˜ Fy 2.029 ˜ E 1/ 6 ˜ M max1/ 3 7.52 E / Fy Fy

2M max Fy 1 2.029² ˜ E 1 / 3 ˜ M max

2/3
AT ˜  ˜
I ˜ Fy 2.029 ˜ E 1/ 6 ˜ M max1 / 3 7.52 E / Fy Fy
2/3
M max 2 2.029²
AT ˜(  )
I ˜ Fy 2.029 ˜ E ˜ I 7.52 ˜ E 1/ 6
1/ 6

2/3
M max
AT 1.642 ˜
Fy ˜ E 1 / 6 Wt 0.00784kg/mm² .1/m

2/3
M max
Wt kg/m 0.01287 ˜
Fy ˜ E 1/ 6
89
© TOPKAYA, 2024

Efficient Girder Design-Unstiffened Web

Previous equation is only satisfactory for bending capacity and
the web is forced to just satisfy the web compactness requirement.
Develop another formula that includes both bending and shear requirements
First Assume Mn = My (i.e. resulting web will be non-compact)
M max A M max h ˜ tw
Af  w 
I ˜ h ˜ Fy 6 I ˜ h ˜ Fy 6
Second Assume elastic range of behavior in shear kv = 5.34

1.1 k v E Fy h kv E
Vn 0.6 Fy Aw ! 1.10
h tw tw Fyw

Vmax
Vmax  IVn 0.9Vn tw
1.373 E Fy
2M max h ˜ t w 2M max 2 ˜ h ˜ t w
AT   h ˜ tw 
I ˜ h ˜ Fy 3 I ˜ h ˜ Fy 3
dAT  2M max 2t w

3M max
Set 0 I ˜ h² ˜ Fy 3
0 heff
dh 0.9t w Fy 90
© TOPKAYA, 2024
 Web Behavior Under Concentrated Loads
Three limit states that influence the strength or stiffness of webs under
concentrated loads should be checked during design.

I ˜ Rn t Ru Rn : nominal reaction strength

Ru : Factored reaction
Local Web Yielding
The web can yield at the tow level due to high concentrated forces.
The mechanical model assumes a load distribution with a slope of 2.5 to 1.0

R R
1
2.5

k
k
lb+5k
lb+2.5k
k

R 91
© TOPKAYA, 2024

For interior loads Rn (5k  lb ) Fywt w lb : length of bearing

k : distance from outer face
of the flange to the web toe
of the fillet
For end reactions Rn (2.5k  lb ) Fywt w Φ : 1.0

Web Crippling
Web might buckle locally under concentrated loads.
The equations presented in LRFD are from test results and simple mechanical models

92
© TOPKAYA, 2024
For interior loads (loads that act d/2 or apart from the end “d : overall depth of member”)

ª l t º E ˜ Fyw ˜ t f
Rn 0.8t w ² «1  3( b )( w )1.5 » Qf
«¬ d tf »¼ tw

For end reactions (loads that act less than d/2 from member end)

ª l t º E ˜ Fyw ˜ t f
lb / d d 0.2 RN 0.4t w ² «1  3( b )( w )1.5 »
«¬ d tf »¼ tw
Qf

ª 4l t º E ˜ Fyw ˜ t f
lb / d ! 0.2 RN 0.4t w ² «1  ( b  0.2)( w )1.5 »
d tf tw
Qf
¬« ¼»

Φ = 0.75 Qf = 1.0 for I-sections

93
© TOPKAYA, 2024

Sidesway Web Buckling

Under concentrated loads the web can buckle sideways.
This causes a relative movement between the compression flange and
the tension flange.

Two cases are possible according to the restraint on the compression flange

Case 1 Compression flange is not restrained against rotation

ª h / tw º
If « » ! 1.7 Sidesway web buckling does not control
«¬ Lb / b f »¼
94
© TOPKAYA, 2024
 ª h / tw º Cr ˜ t w ³ ˜ t f ª h / tw º
If « » d 1.7 Rn
h²
« 0.4( )³»
Lb / b f »¼
L / b «¬
¬« b f »¼

Cr = 6.6×106 MPa when Mu < My at the location of the force

= 3.3×106 MPa when Mu > My

Case 2 Compression flange is restrained against rotation

ª h / tw º
If « » ! 2.3 Sidesway web buckling does not control
¬ Lb / b f
« »
¼

ª h / tw º Cr ˜ t w ³ ˜ t f ª h / tw º
If « » d 2.3 Rn «1  0.4( )³»
«¬ Lb / b f »¼ h² ¬« Lb / b f »¼

95
© TOPKAYA, 2024

Stiffener Design at Concentrated Loads

When Pu t I ˜ Rn Use provisions of J 10.8 which are based on past experience

Design for the excess load (Pu – ΦRn)

The width of each stiffener plus one half the thickness of the column web
shall not be less than one third of the width of the flange or moment
connection plate delivering the concentrated force
w

w/3

96
© TOPKAYA, 2024
 The thickness of a stiffener shall not be less than one-half the thickness
of the flange or moment connection plate delivering the concentrated load, and
greater than or equal to the width divided by 16

t b
t t
b 16

Transverse full depth bearing stiffeners for compressive forces applied to a beam
or plate girder flange(s) shall be designed as axially compressed members (columns)
in accordance with the requirements of Sections E6.2 and J4.4.

The member properties shall be determined using and effective length of 0.75h
And a cross section composed of two stiffeners and a strip of the web having
a width of 25tw at interior stiffeners and 12tw at the ends of members.
tw

web Buckles like a column

97
© TOPKAYA, 2024 25 tw
 Composite Steel-Concrete Construction
Review of Composite Behavior
Consider a beam with two rectangular sections (b x h cross section)
put on top of each other.

Slip

Non - Composite
b ˜ h³ 1
Equal strain diagrams I 2˜ ˜ b ˜ h3
12 6

Plastic stress distribution; Ultimate (plastic) moment of section

§ h · ªh hº
e Mu 2¨ V y b ¸e V y h ˜ b«  »
© 2 ¹ ¬4 4¼

h2
e Mu V y ˜b˜
2
1
© TOPKAYA, 2024

Consider a beam with nailed sections

b ˜ (2h)³ 4
I ˜ b ˜ h3
12 6

e ªh hº
Mu V y ˜ h ˜ b e V y h ˜ b «  » V y ˜ b ˜ h²
¬2 2¼

2
© TOPKAYA, 2024
 Composite Steel-Concrete Flexural Members (CHAPTER I)
A typical floor construction in buildings and bridges consists of a reinforced
concrete slab supported on steel beams. In addition, filled or encased composite
members can be used.

Encased Composite Members Filled Composite Members

Steel Headed
Stud Anchors

Composite Beams with Steel Headed Stud or Steel Channel Anchors

3
© TOPKAYA, 2024

Composite Beams with Steel Headed Stud or Steel Channel Anchors

Often times a metal deck is used in composite construction.

A typical floor construction with metal deck

Trapezoidal
form of a
typical metal deck

Refer to manufacturers
catalogs for more
details 4
© TOPKAYA, 2024
 Concrete slab

Steel
section

The metal deck can be oriented in two diffrent ways. Ribs parallel or
ribs perpendicular

Ribs Parallel Ribs (Corrugations)

Perpendicular

5
© TOPKAYA, 2024

Only a certain portion of the concrete deck is active in carrying the stresses.
This portion is called the effective width, be

Effective width
for one side Effective width

S2 S1

Effective width = be = Σ (effective width on both sides)

Per AISC-LRFD (Chapter I 3.1a)
­L / 8 ½ L = Length of the beam
° °
Effective width on one side  ®S1 / 2¾ S1 = Spacing between beams
°S ° S2 = Edge distance
¯ 2 ¿ 6
© TOPKAYA, 2024
 Steel Anchors
Several types of connectors are used to transmit shear between steel and concrete.

Steel Headed Steel Channel

Stud Anchor Anchors

Spiral

The most common form of shear

connectors is the welded steel headed
stud anchor usually known as a shear stud.

7
© TOPKAYA, 2024

The behavior of steel headed stud anchors are studied by performing push-out test
P
I-Shape

Concrete Top View

Blocks

Load

Qn

Slip 8
© TOPKAYA, 2024
 The nominal strength of a steel headed stud anchor is determined from the test.
An empirical equation given in the AISC Specification is:

Qn 0.5 ˜ Asa f c' ˜ Ec d Rg R p Asa Fu

Asa : cross-sectional area of the steel headed stud anchor

Fu : minimum specified tensile strength of a steel headed stud anchor
Ec : modulus of elasticity of concrete (See section I2.1b)

f cc : specified compressive strength of concrete

Per Sec I2.1b Ec 0.043˜ w1.5 ˜ f cc w (unit weight kg/m³) f cc (MPa)

For normal weight concrete w # 2300 2500 kg/m³ Ec 5375˜ f cc

Strength of Steel Channel Anchors

Qn 0.3 t f  0.5tw la f c' Ec

la : length of channel anchor

tf : thickness of flange of channel anchor
tw : thickness of channel anchor web 9
© TOPKAYA, 2024

Complicated rules for Rp and Rg which reflect reductions due to stud position

10
© TOPKAYA, 2024
 11
© TOPKAYA, 2024

For construction details see Section I3.2c

12
© TOPKAYA, 2024
 13
© TOPKAYA, 2024

Nominal Strength of Composite Sections

Two methods are available to determine nominal strength of composite sections

• Plastic Stress Distribution Method

• Elastic Stress Distribution Method

Methods of Construction
Two methods are possible, namely shored and unshored

Shored Construction Unshored Construction

- beams are placed - beams are placed
- shoring is placed underneath the beams - concrete is cast on top of beams
- concrete is cast - beams carry their dead load and
- shores are removed the load of wet concrete
- composite action carries the dead and live loads - concrete cures
- additional loads are carried by the
composite section

14
© TOPKAYA, 2024
 Composite Beams With Steel Headed Stud or Channel Anchors

Positive Flexural Strength

Depending on web slenderness the moment capacity calculations change

h E
For d 3.76 (compact web)
tw Fy

Mn is calculated based on plastic stress distribution for the limit state of yielding
(plastic moment)

h E
For ! 3.76
tw Fy

Mn is determined from the superposition of elastic stresses,

considering the effects of shoring, for the limit state of yielding (yield moment)

Ib 0.9
15
© TOPKAYA, 2024

Calculation of Flexural Strength Based on Plastic Stress Distribution

Plastic analysis with Whitney stress block is used to calculate Mn.

Depending on the dimensions plastic neutral axis (PNA) can be
in the slab or in the steel section. Concrete under tension is assumed to be cracked.

Depending on the available strength of anchors the behavior is divided into two:

• Fully composite flexural members

• Partially composite flexural members

The strength of a composite member is determined by the strength of steel anchors

in the case of partially composite beams

6Qn
Composite Ratio
min( As Fy ,0.85 f c' Ac )

As : area of steel section

Ac : area of effective concrete slab
16
© TOPKAYA, 2024
 Concrete slab Applied load Steel anchors

Steel
beam

Moment
diagram
Section of Section of
zero maximum
moment moment

ΣQn
Internal forces on steel anchors

ΣQn

17
© TOPKAYA, 2024

Fully Composite Beams

Case 1 0.85 f c' Ac ! As Fy

be

C = As Fy
Neutral db a 0.85 f c'
Axis
d1
Cracked
region

d2

Fy T = As Fy

Stress Internal Forces

As Fy Distribution
a
0.85 f c'be
18
© TOPKAYA, 2024
 Case 2 0.85 f c' Ac  As Fy

be

db a 0.85 f c' C
d1
d3
Fy Cs
Neutral
Axis
Fy d2

Stress Internal Forces

Distribution
a db C 0.85 f c' A c
19
© TOPKAYA, 2024

Partially Composite Beams

be

a 0.85 f c' C
db
d1
Cracked
region d3
Fy Cs
Neutral
Axis
Fy d2

Stress Internal Forces

6Qn
a C 6Qn Distribution
0.85 f c'be
Placement and Spacing
Studs shall be distributed uniformly between the point of maximum and
point of zero moment
20
© TOPKAYA, 2024
 Negative Flexural Strength
be

db T
d1
d3
Fy Ts
Neutral
axis
d2
Fy
C

Stress Internal Forces

Distribution

The effect of reinforcing bars can safely be neglected

21
© TOPKAYA, 2024

Effective Rigidity of Composite Sections

Fully Composite Flexural Members

I 0.75I tr

Partially Composite Flexural Members

ª 6Qn º
I 0.75« I s  '
I tr  I s »
¬« min( A F
s y ,0.85 f c cA ) »¼

I = moment of inertia of composite section

Is = moment of inertia of steel section
Itr = moment of inertia of transformed section (obtained using area transformation
according to Ec/Es ratio)

See commentary for details

22
© TOPKAYA, 2024
 Encased and Compact Filled Composite Members
Use plastic stress distribution to obtain flexural strength

Fy 0.85f’c
Neutral
axis

Fy

Stress Stress
Distribution Distribution
(Steel section) (Concrete Section)

For Non-compact and Slender filled composite members see AISC360-22

23
© TOPKAYA, 2024
 COMPRESSION MEMBERS
Definition of a “compression member”
P P

Compression member = member loaded purely in axial compression;

member nominally has no bending (except due to minor eccentricities in the load
or minor imperfections in the column)

Examples of Compression Members:

The diagonal braces and the columns of this frame

would typically be idealized as carrying axial load only.
(tension or compression). For loading cases that
place these members in compression, they would
be designed as compression members.

1
© TOPKAYA, 2024

Trusses:

Top chord members =

compression members

Rigid Frame
Members subject to
axial compression + bending
treat as a “beam column”
(LRFD Chap. H)

Note: the term “column” is often used

to denote a compression member.
2
© TOPKAYA, 2024
 Limit States

The primary strength limit state for compression members is: buckling (instability)
Definition of buckling: loss of compressive load carrying capacity
resulting from a change in the geometry of the member.
P

Pmax = buckling strength = Pn

Δ
Typical response of
a steel column
Δ

Pn = nominal compressive strength of member (= buckling strength)

Design Equation Pu d IPn 3

© TOPKAYA, 2024

Study of columns:
-Strength of isolated columns
-Strength of columns as a part of a frame

Strength of Isolated Columns

Theoretical Elastic Solution – Developed by Euler in 1759
L, EI
P
Δ

Buckled
shape

Assumptions:
1. Column is pin-ended
2. Column is initially perfectly straight
3. Load is initially applied to the centroidal axis
4. Material is linear, elastic; with no residual stresses. Material never yields
5. Buckled column bends about a principle axis. (no twisting)

Find: Maximum load P that this column can sustain.

4
© TOPKAYA, 2024
 Solution: The solution of this problem, first developed by Euler, is achieved
by writing the differential equation of equilibrium for the column. The differential
equation is then solved with appropriate boundary conditions. Details of the
solution are provided in strength of materials textbooks.

Result:
(1) Buckled Shape = half sine wave

(2) Buckling Load = S ² EI

PE PE = Euler (Elastic) Buckling Load
L²
P

Behavior

S²˜E˜I
L²

5
© TOPKAYA, 2024

Observations:
PE α I (buckling controlled by Imin)
PE α 1/L² (as a column gets longer, it gets weaker)
PE independent of Fy (by assumption)
y x and y = principle axes
Imax = Ix (x = “strong axis”)
Imin = Iy (y = “weak axis”)
x x Buckling will be about the weak (y-axis),
unless the weak axis is restrained.

y
PE

S ² ˜ E ˜ I x (Strong axis buckling)

L²
S ² ˜ E ˜ I y (Weak axis buckling)
L²
L
6
© TOPKAYA, 2024
 Rewrite Euler Buckling load in terms of stress

S ² EI
PE Divide both sides by A
L²

S ² EI I
PE Substitute r2 r: radius of gyration
A A ˜ L² A
PE S ²E
FE
A ( L / r )²
FE = Euler (elastic) buckling stress L/r = slenderness ratio

Plot FE vs L/r:
FE
Note that buckling is controlled by axis
producing largest value of L/r
rmin corresponds to Imin S²˜E
(L/r)max correspond to rmin Fy ( L / r )²

7
L/r
© TOPKAYA, 2024

Columns with End Conditions other than Pinned-Pinned

Example: Consider a column with both ends fixed against rotation
P To determine the elastic buckling load, establish
the differential equation of equilibrium, and solve
with the appropriate boundary conditions

Result: 4S ² EI
PE
L²
Euler buckling load is increased by a factor of 4
due to fixed ends.

L 4S ² EI S ² EI
Rewrite PE
L² (0.5L)²
Thus, the fixed-fixed column buckles at the same load
as a pinned-pinned column with length of 0.5L
In general, we can solve for the elastic buckling load for any end restraint conditions
by solving the differential equation of equilibrium with the appropriate
boundary conditions.
The result of these computations are usually expressed using the effective length
concept. 8
© TOPKAYA, 2024
 Effective Length
In general for a column with any end restraint conditions, the elastic
buckling load (or stress) can be written as follows:

S ² EI FE
S ²E
PE
( KL )² ( KL / r )²

Where : KL = effective length K = effective length factor

The effective length (KL) of an isolated end restrained

column = the length of an equivalent pin-ended column with
the same elastic buckling load.
The effective length of an isolated column is also equal to the distance
between points of inflection in the buckled shape.

Thus, if you can visualize the buckled shape, then you can estimate the effective length
4 ˜S ² ˜ E ˜ I
PE
L² S²˜E˜I
PE
(0.5L)²

L = 0.5L
9
© TOPKAYA, 2024

K Factors For Various Columns

K =0.7 K =0.5 K =1 K =2 K =2
K =1
FE

Plot FE vs KL/r S²˜E

Fy ( KL / r )²
KL/r = effective slenderness ratio

10
© TOPKAYA, 2024 KL/r
 Column Strength – Effect of Initial Crookedness
L
Δo Δ

Buckled shape
Initial (crooked) shape of
column prior to applying
load P

Δo = initial midspan deflection of crooked column prior to applying P

Typical values of Δo for steel columns:

* ASTM limit for maximum permissible out of straightness of steel members
Δo ≤ L/1000
* Average measured out of straightness for steel columns:
Δo ≈ L/1500
No real column is perfectly straight. However, the amount of initial crookedness
is generally quite small.
11
© TOPKAYA, 2024

Load Deflection S²˜E˜I Δo = 0

Behavior: L²
Elastic
Real Theory
column
Δ
Δo
Effects of Δo:
The column bends right from the start of loading when Δo ≠ 0
Buckling is not an instantaneous event.
There is additional stress on the column from the start of loading, when Δo ≠ 0
P P

Δo ≠ 0
Δo = 0
Δ M=P. Δ
P
V
A P M ˜c
V 
A I

12
© TOPKAYA, 2024
 Based on elastic theory (material never yields), P is asymptotic to PE.
There is no loss of strength due to Δo

For real columns (with a material that can yield) some loss of strength occurs
due to Δo. The additional bending stresses due to Δo cause earlier yielding, and
result in a loss of strength.
If Δo is small, the strength loss due to Δo is small.
If Δo is large, the strength loss due to Δo is large.

Thus, it is important to limit the initial crookedness of compression members

to small values.

Effects of Load Eccentricity

Effects of e are the same as the effects of Δo
P P
e

S²˜E˜I e=0
L²
Δ Elastic
Real Theory
e≠0 column
13
Δ
© TOPKAYA, 2024

RESIDUAL STRESSES
Definition: Self equilibrating stresses that result from the manufacturing
process for a steel member

Self equilibrating ³V
A
res dA 0

Sources of residual stresses:

-Uneven cooling of hot rolled shapes
-Uneven cooling of welded shapes
-Cold straightening of members
Typical residual stresses for hot rolled I shape
σres
+

+
σres - - 14
© TOPKAYA, 2024
 σres
Typical
Welded I-Shape +

+ (σ -res)max = 130 -200 MPa

- -

Stub Column Test

L0 Δ σres
Stub column
(too short to buckle)
15
© TOPKAYA, 2024

σ
Coupon: P

No σres

[Link] 3

Fy
2
A.(Fy-σres) With σres
E 1 First
Yield
ε Δ

1 2 3

Fully Yielded
Yielded 16
© TOPKAYA, 2024 flange tips
 P '
Define V avg H avg
A L0

σavg

No σres

Fy
E

(Fy-(σres)max) With σres

Et

ε
εy

17
© TOPKAYA, 2024

EFFECT OF RESIDUAL STRESSES ON COLUMN STRENGTH

P σavg
Straight column with
residual stresses Stub column average stress-strain
curve

Fy
L
Et
Δ
(Fy-σres)

Buckling load for column with nonlinear σ-ε curve:

Use tangent modulus theory
S² E T I S² E T
PCR PT FCR FT
(KL )² ( KL / r )²

PT : tangent modulus buckling load 18

© TOPKAYA, 2024
 ELASTIC vs INELASTIC BUCKLING

1) Elastic Buckling

If FE d Fy  (V res ) max

E=ET S² E
FE
(KL / r )²
No yielding prior to buckling

2) Inelastic Buckling
 S² E T
If FE ! Fy  (V res ) max FT
( KL / r )²
Yielding occurs prior to buckling

19
© TOPKAYA, 2024

Column Strength Curve

FCR
S² ˜ E T
FCR FT
(KL / r )²

S² ˜ E
Fy FCR FE
(KL / r )²

(Fy-(σres)max)

KL/r
Inelastic buckling Elastic buckling
20
© TOPKAYA, 2024
 STRENGTH OF STEEL COLUMNS
The previous discussion and theoretical background indicates that the buckling
strength of a real steel column depends on a number of factors

• Slenderness (L/r)
• End restraint (K factor)
• Initial crookedness or load eccentricity Must consider variability
• Yielding and residual stresses in these factors

Any method to determine real column strength must consider these factors.

There are generally two approaches for determining real steel column strength:
1) Experiments on Columns
Load column until buckling occurs; measure the maximum load
carried by the column
2) Numerical Analysis
Recently developed numerical analysis techniques
(Finite element method etc.) permit accurate analytical determination of
column buckling strength
These techniques require accurate information on initial crookedness and
residual stresses.
21
© TOPKAYA, 2024

TYPICAL RESULT OF COLUMN EXPERIMENT P Real steel column

with residual
P stresses and initial
crookedness low
(KL/r) (inelastic
buckling)

S² ˜ E ˜ I L
PE
L²
Δ
Strength loss
due to σres
S² ˜ E T ˜ I
PT Strength loss due
L² PCR to Δ0 and
additional yielding

Experiment

Observations: Δ
­V res and yielding
PCR  PE due to ®
¯' 0
Buckling is not ductile. 22
© TOPKAYA, 2024
 RESULTS OF MANY COLUMN EXPERIMENTS:
FCR
S² ˜ E
FE
(KL / r )²

Fy Experimental data
falls within “band”

KL/r

Observations:
Large scatter in results (due to scatter in σres and Δ0)
Large values of KL/r => Data falls close to Euler
Intermediate KL/r => Significant effect of σres
Low KL/r => Can develop Fy
23
For design : fit an equation to the data
© TOPKAYA, 2024

LRFD COLUMN DESIGN STRENGTH

LRFD developed an equation for computing strength of columns by
fitting an equation to available column strength data.

Nominal Compression Strength: Pn Ag Fn

Ag: gross area Fcr: buckling stress

Design Compression Strength: IPn (0.9) Ag Fn

Design Criteria Pu d IPn

Computing Fcr
When When
Fy KL E Fy
KL E ! 4.71 (or ! 2.25)
d 4.71 (or d 2.25) r Fy Fe
r Fy Fe
inelastic buckling elastic buckling
ª Fy º
« Fe » Fn 0.877 Fe
Fn «0.658 » Fy
« » 24
© TOPKAYA, ¬2024 ¼
 Fcr/Fy

Fn Fe
Fy Fy

1.0

EQN E 3-2
0.39 EQN E 3-3

Fy
0.5 1.0 1.5 2.0 2.5 Fe

25
© TOPKAYA, 2024

Consider Fcr versus KL/r

Dividing line between elastic and inelastic buckling

KL E KL E
1 .5 ˜ S ˜ For Fy = 235 MPa : 1.5 ˜ S ˜ 137
r Fy r Fy

KL E
For Fy = 355 MPa : 1.5 ˜ S ˜ 112
r Fy

KL E
Elastic Buckling ! 1 .5 ˜ S ˜ Fcr 0.877 ˜ Fe
r Fy

Thus, for elastic buckling, LRFD adopted 0.877 times the Euler Solution
(0.877 accounts for effects of initial crookedness)
Note that Fcr is independent of Fy
KL E
Inelastic Buckling d 1 .5 ˜ S ˜
r Fy Note that Fcr depend on Fy
ª ­ KL F ½º KL
« ® ( )²˜ y ¾ 0 Fn Fy
¯ r S ²E ¿ » At
r
Fn «0.658 » ˜ Fy
« » KL E Fn 0.39 ˜ Fy
¬ ¼ At 1 .5 ˜ S ˜
r Fy
26
© TOPKAYA, 2024
 LOCAL BUCKLING IN COLUMNS
Thus far, we have considered column failure to occur as buckling
of the entire member. However, it is also possible to get failure by
local buckling of the plate elements that make up the cross-section.

PCR local PCR local

PCR member web buckling
buckling flange
buckling

Member Buckling Local Flange Local Web Buckling

Buckling
27
© TOPKAYA, 2024

The column strength equations given in Chapter E of the LRFD Spec are based
on an assumed failure mode of member buckling. In order to use Chapter E
equations, it is first necessary to determine that local buckling does not control.

Local Buckling Strength

Pcr (Local Buckling)

Let Fn ( Local Buckling)
A
b
In general, Fn ( Local Buckling) f( , Fy )
t
" b"
= controlling width-thickness ratio of plate element (Sec B4)
t
For an I-Shape
tf/2 b bf
Flange:
tf t 2t f

tw h
b h
Web:
t tw 28
© TOPKAYA, 2024
 General Relationship between local buckling strength and b/t:
FCR local buckling

Fy

Most rolled I shapes

b/t

Note: As b/t increases, Fcr local buckling decreases

High b/t’s
Low b/t’s
Low FCR local buckling
High FCR local buckling 29
© TOPKAYA, 2024

DESIGN CONSIDERATIONS

I) b d Or
t
If b dO then FCR ( Local Buckling) t Fy and member buckling will control.
t r

λr => See Table B4.1

E
Flange Or 0.56 =16.3(S235) =13.4(S355)
Fy

E
Web Or 1.49 =43.4(S235) =35.6(S355)
Fy
Thus, local buckling of a rolled I-Shape will not control its strength if:

bf E h E
d 0.56 d 1.49
2t f Fy tw Fy

30
© TOPKAYA, 2024
 II ) b ! Or
t
If b !O then Fn( Local Buckling)  Fy
t r

Local Buckling or member buckling may control.

Compute Column Strength using Section E7.

Observations:

Rolled I-Shapes – local buckling rarely controls

Local Buckling may control for:

High Fy
Welded I-Shapes
Other shapes (angles, WT’s, etc.)

31
© TOPKAYA, 2024

COLUMNS IN FRAMES

Consider a member that is a part of a frame or truss. If the member is subject

only to axial compression (no bending moment), then it can be designed as
a compression member (or “column”), using equations given in Chapter E.

Examples

Ponly Ponly Ponly Ponly

BRACED FRAME UNBRACED (RIGID) FRAME

32
© TOPKAYA, 2024
 Ponly

TRUSS

Each of these members shown above as “P only” can be designed as

compression members. The key question is:

What is the effective length, KL?

33
© TOPKAYA, 2024

Each of the members shown as “P&M”

Additional
carry significant axial compression and bending,
Examples
P&M and therefore cannot be designed as
compression members. These members are
called “beam-columns” and are designed per
Chapter H of LRFD.

However, as a part of beam-column design,

it is still necessary to determine an effective
BRACED FRAME length, KL, for these members.

P&M
P&M

UNBRACED FRAMES 34
© TOPKAYA, 2024
 EFFECTIVE LENGTH OF COLUMNS IN FRAMES
Before considering columns as part of frames, consider effective length
factors of isolated columns under idealized boundary conditions

K=1 K = 0.7 K = 0.5 K=1 K=2 K=2 K=∞

Smallest Largest
possible K possible K

Columns ends cannot translate

when buckling occurs; Column buckles Column ends are permitted to translate when
in a “non-sway” mode Buckling occurs; column buckles in a “sway” mode
0.5 ≤ K ≤ 1.0 1.0 ≤ K ≤ ∞
35
© TOPKAYA, 2024

Note: As K ØPn

For purposes of establishing effective length factors, we divide structural

frames into two broad categories:

Frames in which sidesway of columns is prevented

“Braced Frames” 0.5 ≤ K ≤ 1.0

Frames in which sidesway of columns is not prevented

“Unbraced Frames” 1.0 ≤ K ≤ ∞

BRACED FRAMES

In braced frames, sidesway of columns

during buckling is prevented by
bracing or by walls.

Sidesway of
columns is
prevented

36
© TOPKAYA, 2024
 Reinforced
Concrete or
Masonry
Wall

Each column will buckle in a non-sway mode Each column will buckle in a
non-sway mode

Each member will buckle in a

non-sway mode
37
© TOPKAYA, 2024

In braced systems, beam to column connections are generally designed as

“simple” or “pinned” connections.

Pinned connections nominally permit free rotation of the member end.

Pinned beam end connections nominally transfer only shear (no moment)
from the beam to the column.

Example of “Pinned” Connections:

38
© TOPKAYA, 2024
 Consider possible K values

0.5 ≤ K ≤ 0.7 0.7 ≤ K ≤ 1.0

K = 0.7 K = 1.0

“fixed” “pinned” “fixed” “pinned”

column column column column
base base base base

“Pinned” beam to column connections “Rigid” beam to column connections; beam & column
(beam provides no rotational restraint ends must rotate together to maintain original 90° angle.
at column end) The beam provides some rotational restraint at the
column end. Degree of rotational restraint provided
by the beam will be intermediate between a fully pinned
and a fully fixed column end.

In general, for columns in braced frames:

0.5 ≤ K ≤ 1.0
39
© TOPKAYA, 2024

K-Factors Design of Columns in Braced Frames

[Link] Frames with Simple Beam Column Connections (“pinned connections”)

USE K=1
[Link] Frames with Rigid Beam Column Connections Two design options
A. USE K=1

Using K=1 provides an upper bound of K values for braced frames.

Therefore, using K=1 is conservative.
B. Account for rotational restraint at column ends:
USE K<1

To understand the approach for computing a K factor for this case,

consider two limiting cases:
1. Frame with Igirder>>Icolumn [Link] with Igirder<<Icolumn

Tops of
Tops of columns are columns are
nearly “fixed” against nearly “pinned”
40
rotation
© TOPKAYA, 2024
 In general, the degree of rotational restraint that a girder provides to the end
of a column, and the K factor, depends on the relative stiffness of the girder
and column. That is
§ Column Stiffness ·
K f¨ ¸
© Beam Stiffness ¹
To numerically determine the K factor, use the “Alignment Chart” for Braced Frames.

Alignment Chart for Braced Frames

This chart provides a very simple method for computing K factors for columns
in braced frames, when rigid beam-column connections are used.

Procedure:
Compute “G” at the top of a column (GA) and at the bottom of the column (GB)

§ I ·
6¨ ¸ I: moment of inertia of member
L L: length of member
© ¹ Columns
G
§ I ·
6¨ ¸
© L ¹ Girders 41
© TOPKAYA, 2024

Enter alignment chart; draw straight line between GA and GB Read K

Computing “G”

I C3 I
 C2
Top: L C3 L C 2
C3 GA
I G3 I
G3 G4  G4
L G3 L G 4

C2
I C2 I
G1 G2  C1
Bottom: L C 2 L C1
GB
I G1 I
C1
 G2
L G1 L G 2

Enter chart with GA and GB => Read K

42
© TOPKAYA, 2024
 Notes on “G” G
G is inversely proportional to the degree of
rotational restraint provided at the column end Rotational Restraint

I
In 6 , only include members that frame rigidly into the joint
L
§ I ·
Do not include right girder in 6¨ ¸
© L ¹ girder

Use in plane stiffness for I Column Bases:

Nominally “Pinned”
Columns: Use Ix Gtheo = ∞
Beams: Use Ix Recommended G = 10 (AISC)

Nominally “Fixed”
Gtheo = 0
Columns: Use Iy Recommended G = 1 (AISC)
Beams: Use Ix 43
© TOPKAYA, 2024

UNBRACED FRAMES

Sidesway not prevented by For stability need rigid

bracing or walls beam-column connections

KL
L

Sway buckling K≥1 To stabilize, provide

rigid beam-column connections

44
© TOPKAYA, 2024
 RANGE of K VALUES
1. Frame with Igirder>>Icolumn [Link] with Icolumn>>Igirder

K=2

K=1 K=2

Lowest K for
unbraced
frames

Column tops nearly fixed against rotation

K=∞

Column tops nearly pinned

45
© TOPKAYA, 2024

K values for Unbraced Frames

1d K  f
no simple upper bound for design

§ Column Stiffness ·
K f¨ ¸
© Girder Stiffness ¹

To compute K, use Alignment Chart for Unbraced Frames

Procedure
6( I ) columns
G L
Compute I
at top & bottom of column
6( )
L girders

Read K from Chart

46
© TOPKAYA, 2024
 BASIS FOR ALIGNMENT CHART – UNBRACED FRAMES

The alignment chart is derived from

conducting a sidesway stability analysis
on a highly idealized frame.
Assumptions:

Rectangular framing
A
Elastic Behavior

All beam-to column joints are rigid

(column bases may be pinned)

B Beams deform in reverse curvature

with equal end rotations

All columns buckle simultaneously in

a sway mode; no column provides a bracing
effect for any other column

47
© TOPKAYA, 2024

In the derivation of the alignment chart, a typical column is modeled as follows:

The column is modeled with rotational springs at each end. The stiffness of
each spring is a function of the rotational stiffness of girders rigidly framing
into the column end. No translational restraint is assumed to exist at the column end.

An elastic buckling analysis is conducted on column AB, and the elastic

buckling load, PE, is computed. P

The length of an equivalent pin ended column with the same elastic
buckling load is then derived. A

i.e. L
EI
S² EI ŸSolve for K
PE
(KL)² This value of K is given by the alignment chart.
B

48
© TOPKAYA, 2024
 Alignment Chart for Unbraced Frames
BASIS: highly idealized assumptions not satisfied by any real building frames
Errors in the “K” factors predicted by Alignment Chart

Why use alignment charts?

1. Don’t need precise K for design
2. Alignment chart errors are often conservative
3. Alignment Chart is only a simple tool available for design

Alignment Chart errors are sometimes

a) unconservative => handle by modifying Alignment Chart
b) too conservative

MODIFICATIONS TO ALIGNMENT CHART

I) Girder Stiffness Modification

A.C. assumptions : all beam column connections are rigid

Girders deform in reverse curvature (Unbraced Frames)
Girders deform in single curvature (Braced Frames)

49
© TOPKAYA, 2024

M
M 6EI
Θ T L
M

M 3EI
Θ T L
M
M 2EI
Θ T L
M M
M 4EI
Θ T L
M
6( I ) columns
Modify “G” G L m: stiffness modification
6( m ˜ I ) girders
L
Most common girder stiffness modification; far end of girder pinned
3EI / L 1
m (For braced frames safe to use m=1)
6EI / L 2 50
© TOPKAYA, 2024
 II) Modification Factor for Leaning Columns
“Leaning Column” => pinned-pinned column in an unbraced frame.

PB PA

A => leaning column

A
B G BOT f½
¾K f IPn 0
G TOP f¿

Theoretical Approach to Handle Leaning Columns

Use ΣP Concept

In an unbraced frame the following requirements must be satisfied.

51
© TOPKAYA, 2024

1. Each column must support the load that is on it in the no sway (K=1) mode
Pi

No sway buckling

Pi d IPni>K 1.0 @

Check both the x-x and y-y axes. This mode of buckling is exactly the
same as if the structure had diagonal bracing to prevent sway.

52
© TOPKAYA, 2024
 2. Check sway buckling of an entire story in all the directions the frame can sway

Pi

These columns do not bend in this particular

case (pinned connections at top) because their
stiffness is zero in the sway mode (K=∞ , PCR=0)

¦ Pi d ¦ IPni >K t1.0@ ØPni is the individual column capacity based on

the alignment chart value of K for sway
permitted
53
© TOPKAYA, 2024

Composite Compression Members

Encased Composite Members Filled Composite Members

Local buckling must be

considered for filled
composite members

54
© TOPKAYA, 2024
 Compressive Strength

Encased Composite Members

Pno ª Pno
º
For d 2.25 Pn Pno «0.658 Pe »
Pe «¬ »¼

For
Pno
! 2.25 Pn 0.877Pe Ic 0.75
Pe

P no Fy As  Fysr Asr  0.85 f c' Ac

S 2 ( EI eff )
Pe
( KL ) 2

EI eff Es I s  0.5Es I sr  C1Ec I c

§ A  Asr ·
C1 0.25  3¨ s ¸ d 0.3
¨ Ag ¸
© ¹
55
© TOPKAYA, 2024

Filled Composite Members

Use the strength expressions for encased composite members with

minor modifications
For compact sections
§ E ·
Pno Pp Pp Fy As  C2 f c' ¨¨ Ac  Asr s ¸¸ C2 0.85 rectangular sections
© Ec ¹
C2 0.95 round sections

For non-compact sections

Pp  Py § E ·
Pno Pp  O  Op 2
Py Fy As  0.7 f c' ¨¨ Ac  Asr s ¸¸
Or  O p 2
© Ec ¹

For slender sections

0.72Fy
9 Es Fcr round
§ E · Fcr rectangular 0.2
Pno Fcr As  0.7 f c' ¨¨ Ac  Asr s ¸¸ §b·
2
§ § D · Fy · sections
© Ec ¹ ¨ ¸ sections ¨¨ ¨ ¸ ¸¸
©t¹ © © t ¹ Es ¹

§ A  Asr ·
EI eff Es I s  Es I sr  C3 Ec I c C3 0.45  3¨ s ¸ d 0.9
¨ Ag ¸
© ¹ 56
© TOPKAYA, 2024