Design Guide 8 - For RHS and CHS Under Fatigue Load

DESIGN GUIDE
FOR CIRCULAR AND RECTANGULAR

HOLLOW SECTION WELDED JOINTS
UNDER FATIGUE LOADING
CONSTRUCTION
WITH HOLLOW STEEL
SECTIONS
Edited by:
Comit International pour le Dveloppement et lEtude de la

Construction Tubulaire
Authors:
X.-L. Zhao, Monash University

S. Herion, University of Karlsruhe
J. A. Packer, University of Toronto
R. S. Puthli, University of Karlsruhe
G. Sedlacek, University of Aachen
J. Wardenier, Delft University of Technology
K. Weynand, University of Aachen
A. M. van Wingerde, Delft University of Technology
N. F. Yeomans, Chairman CIDECT Technical Commission
DESIGN GUIDE
FOR CIRCULAR AND RECTANGULAR
HOLLOW SECTION WELDED JOINTS
UNDER FATIGUE LOADING
X.-L. Zhao, S. Herion, J. A. Packer, R. S. Puhtli, G. Sedlacek,
J. Wardenier, K. Weynand, A. M. van Wingerde, N. F. Yeomans
TV-Verlag
Die Deutsche Bibliothek CIP-Einheitsaufnahme

Design guide for circular and rectangular hollow section
welded joints under fatigue loading / [ed. by: Comit
International pour le Dveloppment et lEtude de la Construction
Tubulaire]. X.-L. Zhao Kln:
TV-Verlag, 2000
(Construction with hollow steel sections ; 8)
ISBN 3-8249-0565-5
ISBN 3-8249-0565-5
by TV-Verlag GmbH,
Unternehmensgruppe TV Rheinland/Berlin-Brandenburg, Kln 2001
Entirely printed by: TV-Verlag GmbH, Kln
Printed in Germany 2001
Preface
Structural hollow sections are widely used in many applications in the field of construction
and mechanical engineering, where fatigue is an essential aspect in design and fabrication.
Basically, the same fatigue design aspects and design principles apply for hollow sections
as they do for open profiles. However, the welded connections between hollow sections
(for example K-joints) need to be considered in a different way, based on the non-uniform
stress distribution around the welded intersection and the secondary bending stresses in
the joint.
The objective of this design guide is to give design recommendations for structural hollow
sections under fatigue loading. Principally the theory applied in this book is based on the
hot spot stress approach. The latest results of research work carried out by CIDECT and
other research organisations, particularly dealing with the stress concentration factors,
have been utilised in this design guide.
This design guide is the eighth in the series Construction with Hollow Steel Sections,
which CIDECT has published:
1. Design guide for circular hollow section (CHS) joints under predominantly static loading
2. Structural stability of hollow sections
3. Design guide for rectangular hollow section (RHS) joints under predominantly static
loading
4. Design guide for structural hollow section columns exposed to fire
5. Design guide for concrete filled hollow section columns under static and seismic loading
6. Design guide for structural hollow sections in mechanical applications
7. Design guide for fabrication, assembly and erection of hollow section structures
8. Design guide for circular and rectangular hollow section welded joints under fatigue
loading
We express our sincere thanks to Dipl.-Ing. D. Dutta, Germany, Dr.-Ing. D. Grotmann of
RWTH Aachen, Germany, Dr.-Ing. S. Herion of University of Karlsruhe, Germany, Prof. Dr.Ing. F. Mang of University of Karlsruhe, Germany, Prof. Dr. J.A. Packer of University of
Toronto, Canada, Dr.Ir. E. Panjeh Shahi of Vekoma, The Netherlands, Prof. Dr.-Ing. R. S.
Puthli of University of Karlsruhe, Germany, Dr.Ir. A. Romeijn of Delft University of
Technology, The Netherlands, Prof. Dr.-Ing. G. Sedlacek of RWTH Aachen, Germany, Dr.Ing. N. Stranghner of RWTH Aachen, Germany, Prof.Dr.Ir. J. Wardenier of Delft University
of Technology, The Netherlands, Dr.-Ing. K. Weynand of RWTH Aachen, Germany, Dr.Ir. A.
M. van Wingerde of Delft University of Technology, The Netherlands, Mr. N.F. Yeomans of
British Steel Tubes and Pipes, United Kingdom, and Dr. X.-L. Zhao of Monash University,
Australia, for their valuable contributions and comments. Further, the support of the
CIDECT member firms is gratefully acknowledged.
Reijo Ilvonen
Chairman of the Technical Commission
CIDECT
CONTENTS
Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.1
1.2
1.3
1.4
1.5
1.6
Applications . . . . . . . . . . . . .
Types of Joints and Loading .
Fatigue Life Estimation . . . . .
Fatigue Resistance . . . . . . . .
Fatigue Damage Accumulation
Partial Safety Factors . . . . . .
Classification Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.1
2.2
2.3
2.4
General . . . . . . . . . . . .
Detail Categories . . . . .
Nominal Stress Ranges
Fatigue Strength Curves
Hot Spot Stress Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
3.1
3.2
3.3
3.4
3.5
3.6
General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Member forces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Nominal Stress Ranges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
SCF Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Hot Spot Stress Ranges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Fatigue Strength Curves with Thickness Correction . . . . . . . . . . . . . . . . . . . . . 29
SCF Calculations for CHS Joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4.1
4.2
4.3
4.4
4.5
Uniplanar T and Y-Joints . . . . .

Uniplanar X-Joints . . . . . . . . . .
Uniplanar K-Joints with Gap . .
Multiplanar XX-Joints . . . . . . .
Multiplanar KK-Joints with Gap
SCF Calculations for RHS Joints
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
5.1
5.2
5.3
5.4
Uniplanar T and X-Joints . . . . .

Uniplanar K-Joints with Gap . .
Uniplanar K-Joints with Overlap
Multiplanar KK-Joints with Gap
.
.
.
.
Structural Detailing for Fatigue and Reinforcement
6.1
6.1.1
6.1.2
6.1.3
6.2
6.2.1
6.2.2
6.2.3
6.2.4
Structural Detailing for Fatigue . . . .

Design parameters . . . . . . . . . . . . .
Structural details . . . . . . . . . . . . . . .
Weld improvement methods . . . . . .
Structural Reinforcement and Repair
General . . . . . . . . . . . . . . . . . . . . . .
Simple repair . . . . . . . . . . . . . . . . . .
Reinforcement of CHS T-joints . . . .
Reinforcement of RHS T-joints . . . .
....
....
....
...
.
.
.
.
.
.
.
.
.
.
.
.
.
..
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
...
...
..
...
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
11
16
19
19
20
21
22
22
22
23
31
34
37
38
39
42
46
47
48
. . . . . . . . . . . . . . . . . . 50
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
50
50
50
50
53
53
54
54
55
6.2.5 Reinforcement of K and N-joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

6.2.6 Effect of reinforcement/repair on fatigue life . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
7
Design Examples for CHS Joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
7.1
7.2
Example 1: Uniplanar CHS K-Joints with Gap . . . . . . . . . . . . . . . . . . . . . . . . . 60

Example 2: Multiplanar CHS KK-Joints with Gap . . . . . . . . . . . . . . . . . . . . . . 64
Design Examples for RHS Joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
8.1
8.2
8.3
Example 1: Uniplanar RHS T-Joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

Example 2: Uniplanar RHS K-Joints with Gap . . . . . . . . . . . . . . . . . . . . . . . . . 69
Example 3: Multiplanar RHS KK-Joints with Gap . . . . . . . . . . . . . . . . . . . . . . . 73
9.
References
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
Appendix A: Fatigue Actions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

Appendix B: Detailed Categories for Classification Method
. . . . . . . . . . . . . . . . . 83
Appendix C: The Determination of SCFs by Testing and Finite Element Analysis

C.1 Hot Spot Stress and SCF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C.2 Experimental Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
C.3 Finite Element Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Appendix
D.1
D.2
D.3
D.4
. . 87
. . 87
. . 88
. . 90
D: SCF Formulae and Graphs for CHS Joints . . . . . . . . . . . . . . . . . . . . 92

Uniplanar CHS T and Y-Joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Uniplanar CHS X-Joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Uniplanar CHS K-Joints with Gap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Multiplanar CHS XX-Joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
Appendix E: SCF Formulae and Graphs for RHS Joints

E.1 Uniplanar RHS T and X-Joints . . . . . . . . . . . . .
E.2 Uniplanar RHS K-Joints with Gap . . . . . . . . . .
E.3 Uniplanar RHS K-Joints with Overlap . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
104
104
106
112
General information about CIDECT

objectives, activities, publications, members, etc. . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Notation
Definition of geometrical parameters
Abbreviations
CHS
RHS
SHS
FE
MCF
MF
SCF
SNCF
Circular Hollow Section

Rectangular Hollow Section including square hollow section
Structural Hollow Section
Finite Element
Multiplaner Correction Factor
Magnification Factor
Stress Concentration Factor
Strain Concentration Factor
Notation
A
D
L
Lr
Mipb
Mopb
N
Nf
R
Pax
Sn
Srhs
Wo,Wi
area
damage accumulation index
chord length between effective support points
distance from weld toe
in-plane bending moment
out-of-plane bending moment
number of cycles
number of cycles to failure
ratio of minimum to maximum stress in a cycle
axial force
nominal stress range
hot spot stress range
elastic section modulus of chord, brace
bo
do
bi
di
e
g
g
ho
hi
m
p
q
ro
ri
to
ti
chord width of RHS

chord diameter of CHS
width of brace i (RHS)
diameter of brace i (CHS)
joint eccentricity
gap length
g/to
chord depth of RHS
brace depth of RHS
slope in S-N curves or ratio of brace axial load
projected connecting length to chord of overlapping brace in overlapped joint
overlap length
external radius of chord member
internal radius of brace member
chord wall thickness
brace wall thickness

Ff
Mf
Ov

relative chord length (2L/do or 2L/bo)

diameter or width ratio (di/do or bi /bo)
half chord diameter or width to thickness ratio (do/(2to) or bo/(2to))
partial safety factor for fatigue loading
partial safety factor for fatigue strength
overlapping percentage of braces (q/p in %)
relative gap (g/do or g/bo)
angle between planes with braces in multiplanar joints
acute angle between brace and chord axes (in Y-, X-, N-, K- and KT-joints)
normal stress due to axial force
ax
ipb
opb

normal stress due to in-plane bending

normal stress due to out-of-plane bending
wall thickness ratio (ti /to)
Subscripts
0
i
ax
ipb
opb
ref
cov
chord
brace number (1, 2, 3, etc.)
axial
in plane
out of plane
reference brace
carry-over brace
Abbreviations of organisations, institutes etc.

API
AWS
CEN
DEn
EN
EC3
ECCS
IIW
10
American Petroleum Institute

American Welding Society
European Standards Commission
Department of Energy
European Standards
Eurocode 3
European Convention for Constructional Steelwork
International Institute of Welding
1 Introduction
1.1 Applications
Structural hollow sections, both circular (CHS) and rectangular (RHS), are widely used in
all kinds of structures under different types of loading, as shown in previous CIDECT
Design Guides (Wardenier et al. [1991], Rondal et al. [1991], Packer et al. [1992], Twilt et
al. [1996], Bergmann et al. [1995], Wardenier et al. [1995], Dutta et al. [1997]). The published CIDECT Design Guides for structural hollow section joints have mainly dealt with
CHS and RHS joints under static loading (Wardenier et al. [1991], Packer et al. [1992]).
Many tubular structures are subjected to fatigue loading. Some typical examples are
shown in Figure 1.1 to Figure 1.11. The aim of this guide is to provide design recommendations
for welded CHS and RHS joints under fatigue loading.
Figure 1.1 Full revolving plough machine
Figure 1.2 Air seeder farming equipment
11
Figure 1.3 Mobile crane
Figure 1.4 Giant wheel in an amusement park
12
Figure 1.5 Offshore platforms
Figure 1.6 Grab bucket
13
Figure 1.7 Ski installation
Figure 1.8 Bridge
14
Figure 1.9 Bridge
Figure 1.10 Bridge
15
Figure 1.11 Communication tower
1.2 Types of Joints and Loading

The types of joints and loading covered in this design guide are summarised in Table 1.1.
A yes in the table means that design rules for these joint types are given in the guide
whereas a no indicates that no design rules are available or necessary. The types of
joints are shown schematically in Figure 1.12 and Figure 1.13, and the types of loading in
Figure 1.14. For K-joints both bracings are identical and at the same angle.
16
Table 1.1 Types of joints and loading covered in this design guide
Joint type:
u = uniplanar
m = multiplanar
Section type:
CHS = circular hollow sections
RHS = rectangular hollow sections
Brace load
Chord load
Axial
force
Axial
force
InOut-ofplane
plane
bending bending
Inplane
bending
CHS
T/Y
X
K(gap)
XX
KK(gap)
u
u
u
m
m
yes
yes
yes
yes
yes
yes
yes
no
yes
no
yes
yes
no
yes
no
no
no
yes
yes
yes
no
no
yes
no
yes
RHS
T/X
K(gap)
K(overlap)
KK(gap)
u
u
u
m
yes
yes
yes
yes
yes
no
no
no
no
no
no
no
yes
yes
yes
yes
yes
yes
yes
yes
(a) CHS T-Joints
(b) CHS Y-Joints
(c) CHS X-joints
(d) CHS K-joints with gap
(e) RHS T-joints
(f) RHS X-joints
(g) RHS K-joints with gap
(h) RHS K-joints with overlap
Figure 1.12 Types of uniplanar joints considered in this guide
17
(a) CHS XX-joints
(b) CHS KK-joints with gap
(c) RHS KK-joints with gap

Figure 1.13 Types of multiplanar joints considered in this guide
Figure 1.14 Types of Loading
18
1.3 Fatigue Life Estimation

The fatigue phenomenon is characterised by a progressive degradation of strength under
time variant stresses that result in the appearance of visible cracks and subsequent crack
growth, which eventually may include fracture of a member or even the collapse of a structure.
The fatigue life of a structure may be divided into two parts: a crack initiation phase, during
which micro-cracks may form, and a crack propagation phase, which may lead to visible
cracks or even critical crack sizes that result in failure.
The fatigue life of welded joints depends upon the joint type, joint loading and structural
detailing of the joint. The loading (axial force or bending moment) in the joint depends on
the type of structure and the fatigue actions, as described in Appendix A. The effect of
structural detailing of joints on fatigue life is discussed in Chapter 6. This guide mainly
concentrates on determining the fatigue resistance of CHS and RHS joints, i.e. the number of cycles to fatigue failure under a given loading condition for different joint types.
1.4 Fatigue Resistance

Several methods have been developed to determine the fatigue resistance of welded
structural hollow section joints. They include:
a)
b)
c)
d)
e)
f)
Classification method
Punching shear method
Failure criterion method
Static strength method
Hot spot stress method (also called geometric stress method)
Fracture mechanics method
Each method is described briefly below.

a) The classification method is based on structural details for different types of joints
which are classified into various detail categories with about the same fatigue life. Each
detail category corresponds to a nominal stress range under which a joint will fail after
2 million cycles. This method will be described in detail in Chapter 2. The classification
method has been adopted by many standards (EC3 [1992], SAA [1990], JSSC [1995],
AISC [1993], CSA [1994]).
b) The punching shear method is nearly similar to the classification method except that
the punching shear stress range is used instead of the nominal stress range. It has
been adopted by the American Petroleum Institute (API [1991]) and the American
Welding Society (AWS [1998]). This method is described in detail by Marshall [1992].
c) The failure criterion method provides diagrams showing the nominal stress ranges or
maximum stresses at 2 million cycles in relation to joint geometry and loading
parameters. The critical member in a joint can be determined using these diagrams.
This method is only applicable to certain types of joints with a limited validity range.
More details can be found in Mang and Bucak (1982).
d) The static strength method relates the fatigue behaviour to the static behaviour of a
joint. Within certain parameter ranges, a reasonable relationship can be obtained. There
are a few theoretical objections against using this approach. For example, fatigue
behaviour is a weakest link driven mechanism (i.e. a strong joint will still have a low
19
fatigue life if there is one point of weakness), whereas the static behaviour is more
dependent on the total strength and allows stress redistribution (van Wingerde et al.
[1997a]). The static strength approach has been described in detail in Kurobane (1989) and
Niemi (1995). It may be applied as a preliminary design tool, before a better selection
of stress concentration factors based on finite element analysis is available.
e) The hot spot stress (also called geometric stress) method relates the fatigue life of a
joint to the so-called hot spot stress at the joint. It takes the uneven stress distribution
around the perimeter of the joint into account directly. This method will be described in
detail in Chapter 3. The hot spot stress method has been recommended by the
International Institute of Welding Subcommission XV-E (IIW [1985]) for design of welded
tubular joints under fatigue loading.
f)
The fracture mechanics method can be used to estimate the fatigue crack propagation
life of a structural component with crack-like defects. It has been mainly applied to
welded simple joints (Fisher et al. [1970], Gurney [1979], Bell et al. [1989], Swanmidas
et al. [1989], Maddox [1991], Sedlacek et al. [1992], Nguyen and Wahab [1995], Mori
et al. [1997], Mashiri et al [1998]). This method demands much higher computing capacity and more sophisticated software in order to predict the fatigue life of welded tubular joints (Sedlacek et al. [1998]).
In this guide only the classification method (see Chapter 2) and the hot spot stress method
(see chapter 3) are discussed in detail.
1.5 Fatigue Damage Accumulation

As shown in Appendix A, the stress range (S = Smax - Smin) is a governing parameter for
fatigue design. Due to the presence of residual stresses, the stress ratio R (= Smin/Smax)
is not taken into account in modern fatigue design. Only if the structure is fully stress relieved, might it be advantageous to take the stress ratios into account.
For constant amplitude loading, it is assumed that there is no fatigue damage when the
stress ranges are below the Constant Amplitude Fatigue Limit, which is typically defined
as the stress range for a specific S-N curve (Figures 2.1 and 3.3) when the number of
cycles is N = 5 106.
For variable amplitude loading, the stress ranges below the Cut-Off Limit at N = 108
(Figures 2.1, 2.2 and 3.3) do not contribute to fatigue damage.
When the stress range for a constant amplitude loaded structure or when the maximum
stress range for a variable amplitude loaded structure is above the Constant Amplitude
Fatigue Limit, the fatigue damage accumulation (D) can be assessed using the PalmgrenMiner linear rule, i.e.
D =
ni /Ni
in which ni is the number of cycles of a particular stress range Si and Ni is the number of
cycles to failure for that particular stress range.
The allowable fatigue damage (D) for structures in a non-aggressive environment is generally taken as 1.0, if the effect of fatigue cracks and the possibility for inspection are taken
into account by partial safety factors.
20
1.6 Partial Safety Factors

In limit states design codes, partial safety factors for fatigue loading (Ff) and for fatigue
strength (Mf) need to be taken into consideration during design. For example Eurocode3:
Section 9.3 (EC3 [1992]) recommends that Ff = 1.0 and Mf, which is dependant upon
both the consequences of failure and the inspection routine, takes the values shown in
Table 1.2. For a fail-safe structural component, the failure of a joint has reduced consequences of failure, such that the local failure of one component does not result in failure of
the structure.
Table 1.2 Partial safety factor for fatigue strength Mf according to Eurocode 3
Inspection and access
Fail-safe component
Non fail-safe component
Periodic inspection and

maintenance.
Accessible joint detail.
1.25
Periodic inspection and

maintenance.
Poor accessibility.
1.15
1.35
21
2 Classification Method
2.1 General
The classification method is based on structural details for different types of joints which
are classified into various detail categories. Each detail category corresponds to a nominal
stress range under which a joint will fail after 2 million cycles. The classification is derived
on the basis of an analysis of relevant test results, taking account of the chord to brace
thickness ratio (to/t1) and using a lower bound. In this method, the effects of other parameters
and the thickness effects are combined to some extent (Noordhoek et al. [1980], Wardenier
[1982]).
This method is simple to use. The design procedures can be summarised as follows:
Determine the detail category from the types of joints and the detail geometry, as
described in Section 2.2
Determine the nominal stress ranges using an elastic analysis as described in Section 2.3
Determine the permissible load cycles at this stress range, using the fatigue strength
curve shown in Section 2.4 relating to the corresponding detail category
The application of this method is limited to the tubular joint types (attachments and lattice
girders) and parameter ranges given in Appendix B. For lattice girders, detail categories
are only available for uniplanar K- and N-joints, but parameters are very limited. A large
variation in fatigue behaviour may occur for joints within the same category, which may
result in a considerable variation in fatigue life (van Wingerde et al. [1997b]).
2.2 Detail Categories

The detail categories for the classification method are listed in Appendix B for both attachments
and lattice girder joints. They are also given in Eurocode 3 (EC3 [1992]).
The construction details with descriptions and the corresponding detail categories are
given in the tables in Appendix B. It should be noted that the arrow in the construction detail
indicates the direction of the applied stress range while the thick curved line perpendicular
to the arrow indicates the fatigue crack. For lattice girder joints, thickness ratio (to/t1) has
a great effect on the detail category.
2.3 Nominal Stress Ranges

For fatigue design, the nominal stress ranges in the members have to be determined. This
can be done easily for attachments. For lattice girders and all triangulated truss systems
(both 2D and 3D) axial forces and bending moments in the members can be determined
using a structural analysis method as described in Section 3.2. The nominal stress ranges
can be determined in accordance with Section 3.3, where MF is the magnification factor
to account for secondary bending moments in lattice girders. Recommended MF factors
are given in Eurocode 3 (EC3 [1992]). They are summarised in Table 2.1 for CHS joints and
Table 2.2 for RHS joints.
22
Table 2.1 Magnification factors to account for secondary bending moments in CHS joints of lattice
girders
Braces
(vertical members)
Braces
(diagonal members)
1.3
1.8
1.4
1.2
1.65
1.25
Type of joint
Gap
joints
Overlap
joints
Chords
K
N
1.5
Table 2.2 Magnification factors to account for secondary bending moments in RHS joints of lattice
girders
Braces
(vertical members)
Braces
(diagonal members)
1.5
2.2
1.6
1.3
2.0
1.4
Type of joint
Gap
joints
Overlap
joints
Chords
K
N
1.5
2.4 Fatigue Strength Curves

The fatigue strength curves are usually called Sn-Nf curves where Sn is the nominal stress
range and Nf is the corresponding number of cycles to failure. They are plotted on a log-log
scale. The Sn-Nf curves in Eurocode 3 are plotted in Figure 2.1 for attachments under normal
stress ranges and in Figure 2.2 for tubular joints in lattice girders. The Sn-Nf curves given
in other national codes are similar to those in Eurocode 3. Figure 2.1 should be used in
conjunction with Table B.1 in Appendix B, while Figure 2.2 should be used in conjunction
with Table B.2 in Appendix B.
In Figure 2.1, attention should be paid to three important values, i.e. Detail Category,
Constant Amplitude Fatigue Limit and Cut-Off Limit, as explained in Section 1.5. A summary of these values is given in Table 2.3. In Figure 2.1, all the Sn-Nf curves have a slope
of m = 3 when Nf is less than 5 106 and a slope of m = 5 when Nf is between 5 106 and 108.
In Figure 2.2, only a single slope of m = 5 is used. The Cut-Off Limits in Figure 2.2 are given
in Table 2.4.
23
24
Figure 2.1 Fatigue strength curves for attachments under normal

stress ranges
Figure 2.2 Fatigue strength curves for tubular joints in lattice girders
according to the classification method
Table 2.3 Constant Amplitude Fatigue Limit and Cut-Off Limit for Attachments
Detail Category
(N/mm2)
Constant Amplitude
Fatigue Limit (N/mm2)
Cut-Off Limit
(N/mm2)
160
117
64
140
104
57
125
93
51
112
83
45
100
74
40
90
66
36
80
59
32
71
52
29
63
46
26
56
41
23
50
37
20
45
33
18
40
29
16
36
26
14
Table 2.4 Cut-Off Limit for Lattice Girder Joints
Detail Category (N/mm2)
Cut-Off Limit (N/mm2)
90
41
71
32
56
26
50
23
45
20
36
16
25
3 Hot Spot Stress Method

3.1 General
In welded tubular joints the stiffness around the intersection is not uniform, resulting in a
geometrical non-uniform stress distribution. An example is shown in Figure 3.1.
Figure 3.1 Geometrical stress distribution in an axially loaded CHS X-joint
The hot spot stress (also called geometric stress) method relates the fatigue life of a joint
to the so-called hot spot stress at the joint rather than the nominal stress. It takes the uneven
stress distribution around the perimeter of the joint into account directly. The hot spot
stress range includes the influences of the geometry and type of load but excludes the
effects related to fabrication such as the configuration of the weld (flat, convex, concave)
and the local condition of the weld toe (radius of weld toe, undercut, etc.). The hot spot
stress is the maximum geometrical stress occurring in the joint where the cracks are usually
initiated. In the case of welded joints, this generally occurs at the weld toe. More information
about hot-spot stresses can be found in Appendix C.
The design procedures can be summarised as follows:
1. Determine axial forces and bending moments in chord and braces using a structural
analysis as described in Section 3.2.
2. Determine nominal stress ranges (Sn or n) as described in detail in Section 3.3
3. Determine the Stress Concentration Factors (SCFs) as described in Section 3.4
4. Determine the hot spot stress ranges (Srhs) as described in Section 3.5
5. Determine the permissible number of load cycles for a given hot spot stress range at a
specific joint location from a fatigue strength curve given in Section 3.6
This method can be applied to the types of joints and loading summarised in Table 1.1.
3.2 Member Forces
For welded hollow section structures, member forces must be obtained by analysis of the
complete structure, in which noding eccentricity of the member centrelines at the joint
(connection) as well as local joint flexibility is taken into account (Romeijn et al [1997] and
Herion and Puthli [1998]). This can be achieved by the methods described in 3.2.1 to 3.2.3.
26
3.2.1 Sophisticated three dimensional finite element modelling where plate, shell and solid
elements are used at the joints (appropriate for experienced analysts), or
3.2.2 Simplified structural analysis using frame analysis for triangulated trusses or lattice
girders. Axial forces and bending moments in the members can be determined using a
structural analysis assuming a continuous chord and pin-ended braces (see Figure 3.2).
This produces axial forces in the braces, and both axial forces and bending moments in
the chord. This modelling assumption is particularly appropriate for moving loads along the
chord members in structures such as cranes and bridges.
3.2.3 Rigid frame analysis for two- or three-dimensional Vierendeel type girders
Figure 3.2 Plane frame joint modelling assumptions
3.3 Nominal Stress Ranges

The determination of nominal stress ranges depends on the method used to determine
member forces.
3.3.1 For analysis undertaken using the approach in Section 3.2.1, the nominal stress
range in any member can be determined by
P
Sr, ax = ax
A
Sr, ipb =
Mipb
Wipb
Sr, opb =
Mopb
Wopb
range in any member can be determined by
P
Sr, ax = MF ax
A
Sr, ipb =
Mipb
Wipb
where MF is the magnification factor given in Tables 2.1 and 2.2.

range in any member can be determined by the same formulae given in Section 3.3.1.
3.4 SCF Calculations

To determine the hot spot stress ranges along a large number of lines perpendicular to the
weld toe by means of strain gauges on test specimens or Finite Element (FE) analysis is
typically not feasible for designers. Therefore stress concentration factors (SCFs) are used
as simple multiplication factors on the nominal stress range. The SCF is the ratio between
the hot spot stress at the joint and the nominal stress in the member due to a basic member load which causes this hot spot stress. It may vary around the perimeter of the joint.
27
Several fixed lines (called locations of interest) are chosen for a joint, along which the SCFs
are determined. The commonly used lines of measurement are given for CHS K-joints in
Romeijn et al. (1992) and Karamanos et al. (1997), for RHS T and X-joints in van Wingerde
(1992) and for RHS K-joints in Mang et al. (1989) and van Wingerde et al. (1997a). They are
shown in detail in Chapters 4 and 5.
Three different levels of SCF calculations are available. They are
Determine SCFs by experimental testing or FE simulation as described in Appendix C

Determine SCFs using detailed parametric formulae
Determine SCFs using simplified parametric formulae or graphs
This design guide will mainly utilise the simplified parametric formulae or graphs for SCF
calculation as shown in Chapters 4 and 5. For the detailed parametric formulae, references
will be made to the original publications.
If the analysis has been undertaken using the approach in Section 3.2.1, the SCFs can be
determined from the analysis or using Chapter 4 (for CHS joints) or Chapter 5 (for RHS
joints). If the analysis has been undertaken using the approach in Section 3.2.2 or Section
3.2.3, the SCFs can be calculated using Chapters 4 or 5.
3.5 Hot Spot Stress Ranges

For analysis undertaken using the approach in Section 3.2.1, the hot spot stress ranges
can be obtained directly from the analysis for each load combination. In all other cases the
following procedures should be used to determine the hot spot stress ranges.
The hot spot stress range at one location under one load case is the product of the nominal stress range and the corresponding stress concentration factor (SCF). Superposition
of the hot spot stress ranges at the same location can be used for combined load cases.
If the position of the maximum hot spot stress in a member, for the relevant loading condition, cannot be determined, then the maximum SCF values must be applied to all points
around the periphery of the member at a joint. Hot spot stress ranges must be calculated
for both the chord member and brace members.
Under general loading conditions, the hot spot stress range at any location, in the chord
member, is given by:
For all joints except CHS XX-joints
S rhs = SCFaxial-force-in-brace Sr, axial-force-in-brace + SCFipb-in-brace Sr, ipb-in-brace
+SCFopb-in-brace Sr, opb-in-brace + SCFaxial-force-in-chord Sr, axial-force-in-chord
+ SCFipb-in-chord Sr, ipb-in-chord
For K-joints, Sr,axial-force-in-chord refers to the additional stress range caused by Pch shown
in Tables D.3, E.2 and E.3.
For CHS XX-joints where no MCFs (multiplanar correction factors) are as yet available
Srhs = SCFaxial-force-in-REF-brace Sr, axial-force-in-REF-brace + SCFipb-in-REF-brace Sr, ipb-in-REF-brace
+SCFopb-in-REF-brace Sr, opb-in-REF-brace + SCFaxial-force-in-chord Sr, axial-force-in-chord
+SCFaxial-force-in-COV-brace Sr, axial-force-in-COV-brace + SCFopb-in-cov-brace Sr, opb-in-cov-brace
28
Under general loading conditions, the hot spot stress range at any location, in the brace
member, is given by:
For all joints except for CHS XX-joints
Srhs = SCFaxial-force-in-brace Sr, axial-force-in-brace + SCFipb-in-brace Sr, ipb-in-brace
+ SCFopb-in-brace Sr,opb-in-brace
For CHS XX-joints where no MCFs (multiplanar correction factors) are as yet availabale
Srhs = SCFaxial-in-force-in-REF-brace Sr, axial-force-in-REF-brace + SCFipb-in-REF-brace Sr, ipb-in-REF-brace
+SCFopb-in-REF-brace Sr, opb-in-REF-brace + SCFaxial-force-in-COV-brace Sr, axial-force-in-COV-brace
+ SCFopb-in-COV-brace Sr, opb-in-COV-brace
For multiplanar joints the load in one brace plane may affect the hot spot stress range in
another brace plane. This is called the carry-over effect and is discussed in Sections 4.4
and 4.5.
3.6 Fatigue Strength Curves with Thickness Correction

Similar to the classification method, fatigue strength curves (Srhs-Nf curves) are used in the
design, where Srhs is the hot spot stress range. A basic Srhs-Nf curve is used for hollow
section joints with a wall thickness of 16 mm (Thorpe and Sharp [1989], DEn [1993],
Dimitrakis et al. [1995], van Wingerde et al. [1996, 1997a, 1997b]). For joints with wall
thicknesses other than 16 mm, thickness correction factors are introduced. The influence
of the thickness effect on fatigue behaviour of hollow section joints has been widely
investigated (Gurney [1979], van Delft [1981], Marshall [1984,1992], van Delft et al. [1985],
Berge and Webster [1987], Haagensen [1989], Thorpe and Sharp [1989], van Wingerde
[1992]). The thickness effect is also recognized in most design recommendations (IIW
[1985], DEn [1990], EC3 [1992], AWS [1998]), usually resulting in higher Srhs-Nf curves for
smaller wall thicknesses. Two different sets of Srhs-Nf curves were originally developed for
RHS joints and CHS joints respectively (van Wingerde et al. [1997b] and Wardenier et al. [1995]).
Most recently, a common set of Srhs-Nf curves and thickness correction formulae have been
established (van Wingerde et al. [1997c, 1998a]). Based on the analysis in the data sets for
square and circular hollow sections, the DEn T line (equivalent to a class 114 in EC3, following the criteria of EC3) is recommended as a reference Srhs-Nf line (for 16 mm wall
thickness) for joints between both circular and square members. The Srhs-Nf curves with
thickness corrections are shown in Figure 3.3. The equations for the Srhs-Nf curves are
presented in Table 3.1.
Attention should be paid to the following notes when using Figure 3.3 and Table 3.1.
(1) Figure 3.3 and Table 3.1 only apply to CHS joints with thickness between 4 mm and
50 mm and RHS joints with thickness between 4 mm and 16 mm.
(2) For welded joints with a thickness below 4 mm possible weld defects may overrule the
geometrical influence, and sometimes may result in considerably lower fatigue strengths
(Wardenier [1982], Puthli et al. [1989], van Wingerde et al. [1996], Mashiri et al. [1998]).
(3) The Constant Amplitude Fatigue Limit and Cut-Off Limit in Figure 3.3 are summarised
in Table 3.2.
29
Table 3.1 Equations for the Srhs-Nf curves for CHS joints (4 mm t 50 mm) and RHS joints
(4 mm t 16 mm)
for 103 < Nf < 5 106

log(Srhs) = 1 (12.476 log(Nf)) + 0.06 log(Nf) log 16

3
t
12.476 3 log(Srhs)
or log(Nf) =
1 0.18 log 16
t

for 5 106 < Nf < 108

(variable amplitude only)

log(Srhs) = 1 (16.327 log(Nf)) + 0.402 log 16

5
t

or log(Nf) = 16.327 5 log(Srhs) + 2.01 log 16

t
Figure 3.3 Fatigue strength curves for CHS joints (4 mm t 50 mm) and RHS joints
(4 mm t 16 mm) according to the hot spot stress method
Table 3.2 The Constant Amplitude Fatigue Limit and Cut-Off Limit in Figure 3.3
Section Type
CHS
&
RHS
CHS
30
Thickness
(mm)
Constant Amplitude
Fatigue Limit (N/mm2)
Cut-Off Limit
(N/mm2)
4
5
8
12
16
25
32
50
147
134
111
95
84
71
64
53
81
74
61
52
46
39
35
29
4 SCF Calculations for CHS Joints

The SCF calculations for CHS joints are described in this chapter. A summary is given in
Table 4.1 where relevant Tables and Figures can be found. A minimum SCF = 2.0 is
recommended as explained in Appendix C.1 unless otherwise specified such as negligible
or no minimum SCF value required.
Table 4.1 Summary of SCF calculations for CHS joints
Type of joints
Tables and Figures for SCF calculations
uniplanar CHS T and Y-joints
Table D.1
Figures 4.2 to 4.4
uniplanar CHS X-joints
Table D.2
Figures 4.6 to 4.8
uniplanar CHS K-joints with gap
Table D.3
multiplanar CHS XX-joints
Table D.4
multiplanar CHS KK-joints with gap
Table D.3 and Table 4.2
4.1 Uniplanar T and Y-Joints

Definition of the joint
A uniplanar CHS T or Y-joint is shown in Figure 4.1 where the geometric parameters (, ,
, and ) and locations of interest (crown and saddle) are defined.
Typical SCF formulae
The SCF calculations for this type of joint are based on the work of Efthymiou and Durkin
(1985) and Efthymiou (1988). A typical SCF equation for the chord crown location in a Y-joint
under axial load is given below.
SCF = 0.2 (2.65 + 5( 0.65)2) + (0.5 C
3) sin
It can be seen that the SCF is a function of , , , , and C. The factor C corresponds
to the chord-end fixity. In the case of fully fixed chord-ends, C is taken as 0.5. If the chord-ends
are pinned, C is taken as 1.0. A typical value for C is found to be 0.7 (Efthymiou [1988]).
When is less than 12, a short chord correction factor is used to take account of the reduced deformation and stresses in short chords.
Figure 4.1 A uniplanar CHS T or Y-joint
31
Insight into parameters

The SCFs for limited joint configurations are shown graphically in Figures 4.2 to 4.4 for
CHS T-joints under axial load, in-plane bending and out-of-plane bending respectively. The
following conclusions can be made for the chord and brace:
Generally the highest SCF occurs at the saddle position

Highest SCFs at the saddle are obtained for medium ratios
SCF decreases with decreasing value except for the brace crown position under axial
loading
SCF decreases with decreasing 2 value except for the brace crown position under
axial loading
It should be noted that in the case of 0.95, use SCFs for = 0.95.
Figure 4.2 SCFs for CHS T-joints under axial loading ( = 12 and C = 0.7)
32
Figure 4.3 SCFs for CHS T-joints under in-plane bending moment ( = 12)
Figure 4.4 SCFs for CHS T-joints under out-of-plane bending moment ( = 12)
33
Detailed formulae and graphs

The graphs in Figures 4.2 to 4.4 may be used to get a quick estimation of SCFs. The SCF
equations for all the locations of interest (chord saddle, chord crown, brace saddle and
brace crown) in T and Y-joints under axial load, in-plane bending and out-of-plane bending
are given in Appendix D.1 with the range of validity as listed below.
0.2 1.0
15 2 64
0.2 1.0
4 40
30o 90
4.2 Uniplanar X-Joints
A uniplanar CHS X-joint is shown in Figure 4.5 where the geometric parameters (, , ,
and ) and locations of interest (crown and saddle) are defined.
Figure 4.5 A uniplanar CHS X-joint

The SCF calculations for this type of joint are based on the work of Efthymiou and Durkin
(1985) and Efthymiou (1988). A typical SCF equation for the chord crown location in an X-joint
under axial load is given below.
SCF = 0.2 (2.65 + 5( 0.65)2) 3 sin
It can be seen that the SCF is a function of , , and . When is less than 12, a short
chord correction factor is used to take account of the reduced deformation and stresses
in short chords.
34

The SCFs for certain joint configurations are shown graphically in Figures 4.6 to 4.8 for
CHS X-joints under axial load, in-plane bending and out-of-plane loading respectively.
Similar conclusions can be made as for the T and Y-joints shown in Section 4.1. It should
be noted that in the case of 0.95, use SCFs for = 0.95.
The graphs in Figures 4.6 to 4.8 may be used to get a quick estimation of SCFs. The SCF
equations for all the locations of interest (chord saddle, chord crown, brace saddle and
brace crown) in X-joints under axial load, in-plane bending and out-of-plane bending are
given in Appendix D.2 with the range of validity as listed below.
0.2 1.0
15 2 64
0.2 1.0
4 40
30o 90
Figure 4.6 SCFs for CHS X-joints under axial loading ( = 12 and C = 1)
35
Figure 4.7 SCFs for CHS X-joints under in-plane bending moment ( = 12)
Figure 4.8 SCFs for CHS X-joints under out-of-plane bending moment ( = 12)
36
4.3 Uniplanar K-Joints with Gap

A uniplanar CHS K-joint with gap is shown in Figure 4.9 where the geometric parameters
and locations of interest (1 to 4) are defined. In general, for chord locations and for
balanced load, the hot spot is located at either the crown toe (location 1) or the chord
saddle (location 2). For braces, the hot spot locations vary depending on the joint parameters, but in general, the brace crown heel (location 3) and brace saddle (location 4) are
the hot spot locations.
d
= 1
d0
t
= 1
t0
=
d0
2t0
Figure 4.9 A uniplanar CHS K-joint with gap

The SCF calculations for this type of joint are based on the work of Romeijn (1994), Dijkstra
et al. (1996) and Karamanos et al. (1997) and subsequent simplifications by van Wingerde
et al. (1998a). The general format is similar to that given in IIW (1985), i.e. presenting diagrams of SCFs for different values of and and for some constant basic values of and
(i.e. o, o). The values of 12 and 0.5 were chosen as a basis for o and o respectively, and
the corresponding stress concentration factor will be denoted as the reference value SCFo.
A typical SCF formula can be expressed as
1
1
2
2
SCF =

SCF0 = 12
0.5
SCF0
0
o

The exponents 1 and 2 depend on the loading type and the location of interest, which
may vary from 0 to 1.1. The values of 1, 2 and SCFo are given in Table D.3 in Appendix D.
From the general equation, the following conclusions can be made as for CHS T, Y and X-joints
SCF decreases with decreasing value

SCF decreases with decreasing 2 value
From the graphs in Appendix D.3, the following conclusions can be made
For basic balanced axial load, SCF decreases as decreases

For basic balanced axial load, SCF decreases as increases
For chord load, SCF decreases as increases
37

The SCF equations for chord and braces in CHS K-joints with gap subjected to basic
balanced axial load and chord load are given in Appendix D.3 with the range of validity as
listed below.
No eccentricity
Equal braces
0.3 0.6
24 2 60
0.25 1.0
30o 60
4.4 Multiplanar XX-Joints
A multiplanar CHS XX-joint is shown in Figure 4.10 where the geometric parameters and
locations of interest are defined. Four locations in this type of joint are considered as being
critical, i.e.
location 1: chord crown
location 2: chord saddle
location 3: brace crown in reference brace
location 4: brace saddle in reference brace
Figure 4.10 A multiplanar CHS XX-joint

The SCF calculations for this type of joint are based on the work by Romeijn (1994),
Dijkstra et al. (1996) and Karamanos et al. (1997). Four load cases are considered, namely
1.
2.
3.
4.
38
only
only
only
only
axial balanced brace loading

balanced in-plane bending on reference braces
balanced out-of-plane bending on reference braces
axial balanced chord loading
Similar to uniplanar K-joints, this Guide is aimed at proposing simplified equations and/or
graphs for predicting SCFs in multiplanar XX-joints. The general format for SCFs for all
types of loading can be expressed as

1
1
2
2
SCF =

SCF0 = 12
0.5
SCF0
0 o

The values for the exponents (1 and 2) and the SCFo are given in Appendix D.4.
Similar conclusions can be made as for uniplanar CHS X-joints.
For multiplanar joints the load in one brace plane may effect the hot spot stress range in
another brace plane. This is called the multiplanar effect or Carry-over effect.
This effect is considered only at the saddle locations for axial loading of the brace members
or out-of-plane bending. In-plane bending does not introduce any multiplanar phenomena.
The chord axial loading effects are concentrated only on crown locations of the chord.
The detailed SCF formulae and graphs for multiplanar CHS XX-joints are summarised in
Appendix D.4. The ranges of validity in calculating SCFs for multiplanar XX-joints are listed below.
No eccentricity
Equal braces
0.3 0.60
15 2 64
0.25 1.0
= 90
= 90
= 2 arcsin () 16.2
4.5 Multiplanar KK-Joints with Gap
A multiplanar CHS KK-joint is shown in Figure 4.11 where the geometric parameters and locations
of interest are defined. Six locations in this type of joint are considered as being critical, ie.
location
location
location
location
location
location
1:
2:
3:
4:
5:
6:
chord crown toe

chord near-saddle
chord far-saddle
brace crown heel in reference brace
brace near-saddle in reference brace
brace far-saddle in reference brace

The SCF calculations for this type of joint are based on the work by Romeijn et al. (1993),
Romeijn (1994), Dijkstra et al. (1996) and Karamanos et al. (1997) and subsequent simplifications
by van Wingerde et al. (1998b). Two load cases are considered, namely
1. only axial balanced brace loading
2. only chord loading (axial and bending)
39
Figure 4.11 A multiplanar CHS KK-joint
The axial balanced brace loading condition is defined in Figure 4.12. The braces of the joint
can be considered as either braces in a reference plane or braces in a carry-over plane.
The loads in both planes may be different and are related to each other by a factor m.
value of m
referred to as
symmetrical loading
reference-plane loading
anti-symmetrical loading
Figure 4.12 Axial balanced loading condition in multiplanar CHS KK-joints
The SCFs for multiplanar CHS KK-joints can be determined using the SCFs for uniplanar
CHS K-joints (SCFK ) with two correction factors (fgeom and fload) accounting for the effects
of geometry and loading (Dijkstra et al. [1996] and Karamanos et al. [1997]). The general
format is
SCFKK = fgeom fload SCFK
The product of the two factors (fgeom and fload) varies from 1.0 to 1.25 depending on the
geometric parameters and the load conditions. A single factor called the Multiplanar
Correction Factor (MCF) is adopted for simplicity (van Wingerde et al. [1998b]). The general format is
SCFKK = MCF SCFK
The values of MCF for = 180 are 1.0 for all m values. The values of MCF for 90 are
given in Table 4.2. Interpolation is allowed for m between 0 and 1, and for between 90
and 180.
40
Table 4.2 Multiplanar correction factors (MCFs) on SCFs for CHS KK-joints with gap
( 90)
m = +1
chord
m=0
m = 1
m = +1
brace
m=0
m = 1
axial balanced
brace loading
1.0
1.0
1.25
1.0
1.0
1.25
chord
loading
1.0
1.0
1.0
1.0
1.0
1.0
Load
Case

Similar conclusions can be made as for uniplanar CHS K-joints.
For axial balanced brace loading, the multiplanar effects are only considered for the antisymmetrical load case (m = -1). For chord loading, no multiplanar correction is needed.
The detailed SCF formulae and graphs for uniplanar CHS K-joints are summarised in
Appendix D.3. The Multiplanar Correction Factors (MCFs) are listed in Table 4.2. It must be
noted that the actual SCF value (even if it is less than 2.0) for uniplanar CHS K-joints
should be used. The minimum SCF value of 2.0 for multiplanar CHS KK-joints should be
considered after multiplying the MCF factor. The ranges of validity in calculating SCFs for
multiplanar CHS KK-joints are listed below.
No eccentricity
Equal braces
0.3 cos ()
24 2 48
0.25 1.0
30 60
60 180
41
5 SCF Calculations for RHS Joints

The SCF calculations for RHS joints are described in this chapter. A summary is given in
Table 5.1 where relevant tables and Figures can be found. A minimum SCF = 2.0 is recommended as explained in Appendix C.1 unless otherwise specified such as negligible or
no minimum SCF values required. The SCFs given in this section are valid for square hollow
section braces and rectangular hollow section chords with ho/bo between 0.75 and 1.5.
Table 5.1 Summary of SCF calculations for RHS joints
Type of joints
Tables and Figures for SCF calculations
uniplanar RHS T and X-joints
Table E.1, Figures 5.2 to 5.4
uniplanar RHS K-joints with gap
Table E.2, Figures E.1 to E.8
uniplanar RHS K-joints with overlap
Table E.3, Figures E.9 to E.17
multiplanar RHS KK-joints
Table E.2 and Table 5.2
5.1 Uniplanar T and X-Joints

A uniplanar RHS T-joint is shown in Figure 5.1 where the geometric parameters and locations
of interest (lines of measurement A to E) are defined.
Figure 5.1 A uniplanar RHS T or X-joint

The SCF calculations for this type of joint are based on the work by van Wingerde (1992).
The general format of the SCF formulae is
SCF = (a + b + c 2 +d 2) (2)(e+f+g ) h
2
where the constants a, b, c, d, e, f, g and h change for each location of interest (lines A to E)
and applied loading (in-plane bending on brace, axial loading on brace or chord load).
When the load is applied on the chord, the SCF formula is simplified to
SCF = a (2)e h
42

As an indication, the stress concentration factors are given in Figures 5.2 to 5.4 for some
joint configurations. The following observations can be made:
The highest SCFs generally occur in the chord (for = 1) at locations B and C.
The highest SCFs are found for medium ratios.
The lower the 2 ratio, the lower is the SCF.
The lower the ratio, the lower is the SCF in the chord, whereas it has less influence
for the brace.
Figure 5.2 SCFs for T and X-joints of square hollow sections loaded by an axial force on the brace
43
Figure 5.3 SCFs for T and X-joints of square hollow sections loaded by an in-plane bending moment
on the brace

Detailed equations are given in Appendix E.1 for lines of measurement A to E and for
different loading conditions with the range of validity listed below.
0.35 1.0
12.5 2 25.0
0.25 1.0
Attention should be paid to the following notes when using the SCF formulae:
44
The effect of bending in the chord for a T-joint, due to the axial force on the brace,
should be included separately in the analysis.
For fillet welded connections: multiply SCFs for the brace by 1.4.
For non-90 RHS X-joints, SCFs can be determined using SCFs for 90 RHS X-joints
with some correction factors (Packer and Wardenier [1998]), i. e.:
For lines B, C and D on the chord, SCF = 1.2 SCF = 90 sin2
For lines A and E on the brace, SCF = 1.2 SCF = 90 sin
The above two formulae are valid for 40 80.
Figure 5.4 SCFs for T and X-joints of square hollow sections loaded by an axial force or in-plane
bending moment on the chord
45
5.2 Uniplanar K-Joints with Gap

A uniplanar RHS K-joint with gap is shown in Figure 5.5 where the locations of interest
(lines A to E) are defined.
Figure 5.5 A uniplanar RHS K-joint with gap

The SCF formulae for this type of joint were derived by van Wingerde et al. (1996), where
about 36 formulae were given. These formulae give SCF values for the locations of interest
(lines A to E) shown in Figure 5.5. Simplified formulae are given by van Wingerde et al.
(1997a, 1997b), where the number of formulae is reduced to three. The simplified formulae
correspond to the maximum SCF in chord and brace, and are more conservative (Puthli
and Herion [1996], Zhao and Puthli [1998]).
For balanced axial loading, the general format of SCFs can be expressed as (Puthli and
Herion [1996]):
SCF = f(, 2, , , g ) = SCF0 f(2, )
The reference value SCF0 is the SCF for 2 = 24 and = 0.5. For other values of 2 and
a correction factor f(2, ) should be multiplied to SCF0.
For chord loading, only the SCF in the chord needs to be considered. The formula for the
SCF is
SCF = (2.45 + 1.23 ) g 0.27
The following conclusions can be made based on the graphs shown in Appendix E.2.
46
Highest SCFs occur around medium ratios

For chord, SCF decreases as 2 decreases, and SCF decreases as decreases
For brace, SCF decreases as 2 decreases, and SCF decreases as increases

Detailed expressions of SCF = f(, 2, , , g ) are given in Appendix E.2 with the range of
validity listed below.
0.35 1.0
10 2 35
0.25 1.0
30 60
2 g
0.55 e/h0 0.25
Graphs for the reference value SCF0 and the correction factor f(2, ) are also given in
Appendix E.2. The SCF0 of the chord is plotted against , with different lines for = 30,
45 and 60, in different graphs for g = 1, 2, 4 and 8. For the brace, one graph containing
the SCF0 against for = 30, 45 and 60 is sufficient, since the gap has only a very small
influence on the maximum SCF of the brace. Interpolation is allowed between the lines for
other angles and between the graphs for other gaps.
5.3 Uniplanar K-Joints with Overlap

A uniplanar RHS K-joint with overlap is shown in Figure 5.6 where the locations of interest
(lines A to E) are defined.
Figure 5.6 A uniplanar RHS K-joint with overlap

The SCF formulae for this type of joint were derived by van Wingerde et al. (1996), where
about 60 formulae were given. These formulae give SCF values for the locations of interest (lines A to E) shown in Figure 5.6. Simplified formulae are given by van Wingerde et al.
(1997a, 1997b), where the number of formulae is reduced to three. The simplified formulae correspond to the maximum SCF in chord and brace, and are more conservative.
47
For balanced axial loading, the general format of SCFs can be expressed as
SCF = f(, 2, , , Ov) = SCF0 (, , Ov) f(2, )
The reference value SCF0 is the SCF for 2 = 24 and = 0.5. For other values of 2 and
a correction factor f(2, ) should be multiplied to SCFo.
For chord loading, only the SCF in the chord needs to be considered. The formula for the SCF is
SCF = 1.2 + 1.46 0.028 2
The following conclusions can be made based on the graphs shown in Appendix E.3.
Highest SCFs occur around medium ratios for RHS K-joints with 50% overlap
For chord, SCF decreases as 2 decreases, and SCF decreases as decreases
For brace, SCF decreases as 2 decreases, and SCF decreases as increases
SCFs for overlapped K-joints are generally lower than those for gapped K-joints

Detailed expressions of SCF = f(, 2, , , Ov) are given in Appendix E.3 with the range of
validity listed below.
Equal braces
0.35 1.0
10 2 35
0.25 1.0
30 60
50% Ov 100%
0.55 e/ho 0.25
Graphs for the reference value SCF0 and the correction factor f(2, ) are also given in
Appendix E.3. The graphs for SCF0 show the SCF against , with different lines for = 30,
45, 60, in graphs for 50%, 75% and 100% overlap. Interpolation is allowed between the
lines for other angles and between the graphs for other overlaps.
5.4 Multiplanar KK-Joints with Gap
A multiplanar RHS KK-joint is shown in Figure 5.7 where the geometric parameters are
defined.
The SCFs for multiplanar RHS KK-joints can be determined using the SCFs for uniplanar
RHS K-joints (SCFK) with a Multiplanar Correction Factor (MCF) accounting for the effects
of geometry and loading (van Wingerde et al. [1998b]), i. e.:
SCFKK = MCF SCFK
The values of MCF for 90 are given in Table 5.2 where the term m has the same definition as that used for multiplanar CHS KK-joints shown in Figure 4.12. The values of MCF
for = 180 are 1.0 for all m values. Interpolation is allowed for m between 0 and 1, and
for between 90 and 180.
48
Figure 5.7 A multiplanar RHS KK-joint with gap

Table 5.2 Multiplanar correction factors on SCFs for RHS KK-joints with gap ( 90)
m = +1
chord
m=0
m = 1
axial balanced
brace loading
1.0
1.0
chord
loading
1.0
1.0
Load
Case
m = +1
brace
m=0
m = 1
1.25
1.0
1.0
1.25
1.0
1.0
1.0
1.0

Similar conclusions can be made as for uniplanar RHS K-joints with gap.
For axial balanced brace loading, the multiplanar effects are only considered for the antisymmetrical load case (m = 1). For chord loading, no multiplanar correction is needed.
The detailed SCF formulae and graphs for uniplanar RHS K-joints with gap are summarised in Appendix E.2. The Multiplanar Correction Factors (MCFs) are listed in Table 5.2. It
must be noted that the actual SCF value (even if it is less than 2.0) for uniplanar RHS Kjoints should be used. The minimum SCF value of 2.0 for multiplanar RHS KK-joints should
be considered after multiplying the MCF factor. The ranges of validity in calculating SCFs
for multiplanar RHS KK-joints are listed below.
Equal braces
0.25 0.60
12.5 2 25
0.5 1.0
30 60
2 g
0.55 e/ho 0.25
60 180
49
6 Structural Detailing for Fatigue and Reinforcement

6.1 Structural Detailing for Fatigue
6.1.1 Design parameters
The principle that good design and fabrication can considerably reduce the total costs of
a structure applies just as well to those structures made from hollow sections as to those
made of other sections. This is particularly true for structures subjected to fatigue loading,
where both a poor design and poor fabrication and welding can result in significantly more
fatigue damage than expected.
Optimal fatigue design is obtained when the SCFs are as low as possible. Considering the
influence of all parameters mentioned in Chapters 4 and 5, the following guidelines can be
made:
Avoid medium ratios; ratios close to 1.0 give the lowest SCFs
Make the wall thickness of the bracing as small as possible with respect to that of the
chord (i.e. low ratio)
Select relatively thick walled chords (i.e. low 2 ratio)
Select overlap K-joints over gap K-joints
Select butt (groove) welds over fillet welds
Fatigue cracks always occur at a discontinuity or stress raiser. In the case of welded
connections this is nearly always associated with a microscopic defect at the weld toe. As
a result, fatigue cracks initiate at these locations irrespective of the steel grade. Higher
grades of steel can be used, but unless some form of post weld improvement is carried
out the fatigue life will not necessarily increase.
6.1.2 Structural details

Welding details, sequences, procedures, etc. will not be provided in this design guide,
since they have already been extensively described in two previous CIDECT Design
Guides Nos. 6 and 7 (Wardenier et al. [1995], Dutta et al. [1997]). However, for convenience, Figures 6.1 and 6.2 show typical weld details at the main locations of a T-joint constructed with circular and rectangular hollow sections respectively.
Weld start/stop positions for non-continuous welds should not be located at points of high
stress concentration, since these can themselves cause stress concentrations. Figure 6.3
shows the recommended locations for these weld start/stop positions (Wardenier et al.
[1995]). For RHS braces weld start/stop positions should be sufficiently away from the corners
and in CHS braces away from the crown and saddle points.
6.1.3 Weld improvement methods

Various methods of improving the fatigue resistance of welded connections are available
(Haagensen [1989, 1997], Bignonnet [1987], Haagensen and Maddox [2000]). They result
in a change in the weld geometry to minimise weld defects and stress concentrations, or
to reduce the tensile residual stresses in the joint. They can be summarised as two groups.
50
One is the geometry improvement method which includes the grinding method (Knight
[1978]), the remelting method (Haagensen [1978]) and the weld profiling method (AWS
[1998], Kobyashi et al. [1977]). The other is the residual stress method which includes the
hammer peening and shot peening methods (Knight [1978]).
Section 6.2 describes some methods for the repair of structures already subjected to fatigue
cracking.
Figure 6.1 Fillet and butt (groove) welds in lattice joints between circular hollow sections
51
Figure 6.2 Fillet and butt (groove) welds in lattice joints between rectangular hollow sections
52
Figure 6.3 Recommended locations for the weld start/stop positions
6.2 Structural Reinforcement and Repair

6.2.1 General
Steel structures are usually designed for a fatigue life or a fatigue load depending on intended
use. Exceeding the calculated lifetime or fatigue loading can result in fatigue damage to the
structure. If a further use of the structure is planned, measures of repair will be inevitable.
On the other hand it is possible that higher loads have been applied to the structure due
to changed demands. In this case methods of structural reinforcement may have to be
carried out at the critical joints. Because the methods of repair generally mean some type
of reinforcement at the same time, the following sections deal with both methods.
Another possibility is to replace complex welded connections with forgings or casting, as
is often done in the nuclear industry. However, the costs involved can be quite high and
the delivery times longer than would be in the case with normal welded connections. If this
option is used the forging or casting manufacturers should be involved in the project as
early as possible.
Fatigue damage of a structure starts with small cracks. The location is usually the toe of a
weld in welded joints. Regular inspections of the structure are required in order to detect
cracks as early as possible. When a crack is detected in a structural member action should
be taken as soon as possible, to prevent it from propagating further. A simple method is to
drill a hole into the base metal at each end of the crack at the crack tip, as shown in Figures
6.4 (a) and (b). The diameter of the hole should be approximately equal to the material
thickness in order to trap any cracks that run at an angle through the material. At the same
time the hole reduces the stress concentration, because the directional change in the flow
of stress around the edge of the hole is not as severe as it would have been at the tip of
the crack. It should be followed by repair of the member.
Different methods for the repair and/or reinforcement of cracks in hollow section joints are
described in the sections 6.2.2 to 6.2.4.
53
Figure 6.4 Grinding or gouging out cracks and rewelding
6.2.2 Simple repair

This method of repair without the use of reinforcement is equally applicable to both circular
and square hollow sections and is illustrated in Figures 6.4 (a) to (d). It is assumed that the
crack has been stopped by drilling as described above. The next stage is to either grind
out the whole of the crack with a grind stone, or gouge out using a gouging arc weld
electrode, and at the same time to prepare the edges for welding. The repair is completed
by welding up the area using normal welding procedures, and then grinding the repair flush,
if required.
6.2.3 Reinforcement of CHS T-joints

In the case of fatigue damage to circular hollow section T-joints, where the crack runs from
the weld area into the chord, there are three methods of reinforcement.
(a) using internal ring stiffeners
One method is to use internal ring stiffeners, provided access is possible from the inner
side, as shown in Figure 6.5 (a). The solution with internal ring stiffeners is frequently found
at locations where high localized loadings may occur (such as at supports). A load
redistribution to adjacent stiffeners takes place, reducing stress concentrations.
(b) using side plates
The other method is to use side plates as shown in Figure 6.5 (b). Side plates can be welded
to the joint before or after fatigue cracking. The ends of the plates should be situated at
locations with a relatively homogeneous stress distribution in order to reduce the negative
influence of notch stresses.
(c) using castings
Another solution is the replacement of parts of the joint by castings which completely cover
the connection area, see Figure 6.5 (c). These are components with a very homogeneous
structure offering a long life expectancy. Attention should be paid to the manufacture.
Welding details have to be taken into consideration carefully to ensure a good connection
between the casting and the original tube. This is time consuming and expensive. For this
reason this solution is used only in special cases.
54
(a) using internal ring stiffeners
(b) using side plates
(c) using casting

Figure 6.5 Reinforcement of CHS T-joints
6.2.4 Reinforcement of RHS T-joints

If at a T-joint a crack has developed in the chord across the main load direction as shown
in Figure 6.4, different possibilities for the reinforcement of the joint are presented in
Figure 6.6 (a) to Figure 6.6 (c). They are
(a) using chord flange plates
The brace member is separated from the chord and a stiffening plate is welded between
the chord and the brace member as shown in Figure 6.6 (a). Because the intermediate
plate is loaded in the direction of its thickness, it should have good through thickness
properties and be free from laminations.
(b) using lateral gusset plates
Lateral gusset plates are attached in order to reduce the stresses at the connection as
shown in Figure 6.6 (b). The length of the recesses should be larger than the outer width
55
of the brace member. The connection of the brace member and the gusset plates allow the
brace loads to be transferred into the chord. This leads to a load redistribution and therefore
to reduction of stress concentrations.
(c) using haunch stiffeners
Lateral haunch-like stiffening of hollow sections, arranged lengthwise to the chord, with the
same dimensions as the brace member are welded to the joint as shown in Figure 6.6 (c).
The recesses are necessary to avoid a superposition of thermal and structural stresses.
Moreover the fillet welds can be welded all around. From a structural mechanics point of
view this is the best solution, because the connecting surface between chord and brace
member increases considerably. However this method can be costly to fabricate.
6.2.5 Reinforcement of K and N-Joints

The fatigue crack in K and N-joints may occur along the weld or in the chord or brace
member. Reinforcement can be carried out on both CHS and RHS joints by welding a flange
plate as shown in Figure 6.7 (a), a transverse gusset plate as shown in Figure 6.7 (b) or the
combination of the above as shown in Figure 6.7 (c). All these repairs are rather difficult to
apply. Another possibility for the reinforcement or repair of K-joints is the use of boxed
components made from RHS as shown in Figure 6.8 (a) for CHS K-joints and Figure 6.8 (b)
for RHS K-joints.
(a) using chord flange plates
(b) using lateral gusset plates
(c) using haunch stiffeners

Figure 6.6 Reinforcement of RHS T-joints
56
(a) flange plate reinforcement
(b) vertical plate reinforcement
(c) combination of flange and vertical plate reinforcement

Figure 6.7 Reinforcement of RHS K and N-joints using welded plates
57
(a) CHS K-joints
(b) RHS K-joints

Figure 6.8 Reinforcement of K-joints using corner components
58
6.2.6 Effect of reinforcement/repair on fatigue life

Indications on improved fatigue strength due to different repair and reinforcement methods
are shown in Table 6.1. Detailed experimental results are reported by Puthli et al. (1992)
and Mang and Bucak (1996). In Table 6.1, the term Nf is the original life without repair or
reinforcement.
Table 6.1 Extended life according to different repair and reinforement measures
Joint
Repair and reinforcement method
Indications on
additional life
CHS
K-Joint
Repair with RHS pieces

(see Figure 6.8)
> 200% Nf
and
Repair with stiffening plates

(see Figure 6.7)
> 100% Nf
RHS
K-Joint
Repair by gouging out cracks and re-welding

(see Figure 6.4)
> 40% Nf
59
7 Design Examples for CHS Joints

7.1 Example 1: Uniplanar CHS K-Joints with Gap
Given:
A uniplanar truss is shown in Figure 7.1. The layout of the truss is the same as that shown
on page 46 in CIDECT Design Guide for CHS Joints under Predominantly Static Loading
(Wardenier et al. [1991]). The joint eccentricity (e) is zero. An applied constant amplitude
loading, ranging from zero to the loads shown, is given in Figure 7.1.
The member sizes are:
Top chord:
CHS 219.1 x 7.1,
Braces:
CHS 88.9 x 4,
Bottom chord: CHS 177.8 x 7.1,
Ao = 4728 mm2,
A1,2 = 1070 mm2,
Ao = 3807 mm2,
Wo = 0.243 x 106 mm3

W1,2 = 0.0217 x 106 mm3
Wo = 0.156 x 106 mm3
The static strength has been found satisfactory.
Figure 7.1 A uniplanar truss subjected to a constant amplitude load range
Problem:
Determine the fatigue life of joint No. 6 indicated in Figure 7.1.
Solution:
Step 1: Parameters
= d1/do = 88.9/177.8 = 0.5
2 = do/to = 177.8/7.1 = 25
= 12.5
= t1/to = 4/7.1 = 0.563
= arc tan (2.4/3.0) = 38.7
The parameters are within the validity range given in Table D.3.
Step 2: Structural analysis
A structural analysis is carried out assuming a continuous chord and pin-ended braces as
described in Section 3.2.2. The axial forces and bending moments found in joint No. 6 are
given in Figure 7.2. They can be treated as a combination of two load conditions shown in
Figure 7.3, i. e.:
60
Load condition 1: basic balanced axial loading

Load condition 2: chord loading (axial and bending)
0.786 kNm
0.786 kNm
Figure 7.2 Axial forces and bending moments in joint No. 6
Figure 7.3 Two load conditions for joint No. 6
Step 3: Nominal stress ranges in critical members

It can be seen from Figure 7.2 that the critical chord loading is in chord 1 due to a larger
tension force in it. Brace 2 with a tensile force range will be checked only. Note: In general,
it is presumed that only braces which have some parts of their load range in tension will
be liable to cause fatigue failure.
For load condition 1 (basic balanced axial loading):
brace,ax = MF 17.2 103 /1070 = 1.3 16 = 21 N/mm2

For load condition 2 (chord loading):
chord,ch = chord,ax + chord,ipb = MF 228.5 103/3807 0.786 106/(0.156 106)

= 1.5 60 5 = 85 N/mm2
(Note that the chord bending moment relieves the tensile stress on the connecting face of
the chord. The values of MF (magnification factor) are given in Table 2.1).
61
Step 4: SCF calculation for load condition 1 (basic balanced axial loading)
From Table D.3
Chord
SCFch,ax =
0.4
1.1
12 0.5
0.4

12.5
12
SCFo,ch,ax =
1.1
0.563
0.5
SCFo,ch,ax
= 1.16 SCFo,ch,ax
for = 0.5 and = 30, SCFo,ch,ax = 2.6
for = 0.5 and = 45, SCFo,ch,ax = 2.9
so that for = 0.5 and = 38.7, SCFo,ch,ax = 2.77
where
and SCFch,ax = 1.16 2.77 = 3.2

Brace
SCFb,ax =
0.5
0.5

12

0.5
SCFo,b,ax =
0.5

12.5
12
0.5
0.563
0.5
SCFo,b,ax
= 1.08 SCFo,b,ax
for = 0.5 and = 30, SCFo,b,ax = 1.3
for = 0.5 and = 45, SCFo,b,ax = 1.8
so that for = 0.5 and = 38.7, SCFo,b,ax = 1.59
where
and SCFb,ax = 1.08 1.59 = 1.7

Check minimum SCF value:
for = 30,
for = 45,
so that for = 38.7,
min SCFb,ax = 2.64

min SCFb,ax = 2.30
min SCFb,ax = 2.44
so use minimum SCF value, SCFb,ax = 2.4

Step 5: SCF calculation for load condition 2 (chord loading)
From Table D.3
Chord
SCFch,ch = 1.2
0.3

0.5
(sin ) 0.9 = 1.2
use minimum SCF value, SCFch,ch = 2.0
62
0.563
0.5
0.3
(sin 38.7) 0.9 = 1.9
Brace
SCFb,ch = 0 (negligible)
Step 6: Hot spot stress ranges
Load condition 1 (basic balanced axial loading):
Srhs,chord = SCFch,ax brace,ax = 3.2 21 = 67 N/mm2
Srhs,brace = SCFb,ax brace,ax = 2.4 21 = 50 N/mm2
Load condition 2 (chord loading):
Srhs,chord = SCFch,ch chord,ch = 2.0 85 = 170 N/mm2
Srhs,brace = SCFb,ch chord,ch = 0 N/mm2
Superposition of load conditions 1 and 2:
Srhs,chord = 67 + 170 = 237 N/mm2
Srhs,brace = 50 + 0 = 50 N/mm2
Step 7: Hot spot stress ranges for design
A partial safety factor on hot spot stress ranges is required for design. For this example the joint
is assumed to be non fail-safe and accessible. From Table 1.2, the partial safety factor is 1.25.
Srhs,chord = 1.25 237 = 296 N/mm2
Srhs,brace = 1.25 50 = 63 N/mm2
Step 8: Fatigue life of joint No. 6
For Fatigue Cracking in the Chord:
t = 7.1 mm and Srhs,chord = 296 N/mm2
Table 3.1 or Figure 3.3 can now be used to determine the fatigue life.
From Table 3.1
log(Nf) = 12.476 3 log(Srhs) = 12.476 3 log(296) = 5.41
1 0.18 log( 16 )
1 0.18 log( 16 )
t
7.1
Nf = 105.41 = 257,000 cycles
For Fatigue Cracking in the Brace:
t = 4 mm and Srhs,brace = 63 N/mm2 . The hot spot stress range of 63 N/mm2 is lower
than the Constant Amplitude Fatigue Limit of 147 N/mm2 listed in Table 3.2. Therefore
no fatigue damage occurs in the brace.
Hence, the fatigue life expectancy of joint No. 6 is 257,000 cycles, with failure in the chord.
63
7.2 Example 2: Multiplanar CHS KK-Joints with Gap

Given:
It is assumed that a multiplanar CHS KK-joint has the same geometry and internal brace
loads in each plane as joint No. 6 in the previous design example for uniplanar CHS Kjoints described in Chapter 7.1. It is assumed that the internal brace loads are totally antisymmetrical, i.e. m = 1. Therefore the loads in the chord member can be considered negligible, see Figure 7.4. The multiplanar parameter (), i.e. the angle between planes with braces, is 90.
It should be noted that the external constant amplitude load ranges are different to those
given in example 7.1. Therefore it is not the purpose of this example to show the multiplanar
effect on fatigue life. The purpose of this example is to demonstrate the design procedures
for a multiplanar joint under a given internal load condition.
Figure 7.4 Assumed load range on the multiplanar joint
Problem:
Determine the fatigue life of the multiplanar CHS KK-joint.
Solution:
Step 1: Multiplanar Correction Factors (MCFs) on SCFs
From Table 4.2
For load condition 1 (basic balanced axial loading), m = 1
chord:
brace:
MCFch,ax = 1.25
MCFb,ax = 1.25
For load condition 2 (chord loading), m = 1

chord:
brace:
MCFch,ch = 1.0
MCFb,ch = 1.0
Step 2: Modified SCFs

SCFs for uniplanar joints used in this example have been taken from the example
in chapter 7.1.
64
For load condition 1 (basic balanced axial loading)

SCFch,ax,mp = MCFch,ax SCFch,ax,up = 1.25 3.2 = 4.0
SCFb,ax,mp = MCFb,ax SCFb,ax,up = 1.25 1.7 = 2.13
but use minimum SCF value, SCFb,ax,mp = 2.4 (see example in 7.1)
where mp stands for multiplanar and up stands for uniplanar.
Note: SCFb,ax,up of 1.7 is the calculated value, not the required minimum value.
chord:
brace:
For load condition 2 (chord loading)

chord:
brace:
SCFch,ch.mp = SCFch,ch,up = 2.0

SCFb,ch,mp = SCFb,ch,up = 0 (negligible)

Srhs,chord,mp = SCFch,ax,mp brace,ax = 4.0 21 = 84 N/mm2
Srhs,brace,mp = SCFb,ax,mp brace,ax = 2.4 21 = 50 N/mm2
Because the chord stress is zero,
Srhs,chord,mp = 0 N/mm2
Srhs,brace,mp = 0 N/mm2
Srhs,chord,mp = 84 + 0 = 84 N/mm2
Srhs,brace,mp = 50 + 0 = 50 N/mm2
Srhs,chord,mp = 1.25 84 = 105 N/mm2
Srhs,brace,mp = 1.25 50 = 63 N/mm2
Step 5: Fatigue live of multiplanar joint
For Chord:
t = 7.1 mm and Srhs,chord,mp = 105 N/mm2. The hot spot stress range of 105 N/mm2
is lower than the Constant Amplitude Fatigue Limit (for t = 7.1 mm) which is between 111 N/mm2 (for t = 8 mm) and 134 N/mm2 (for t = 5 mm) listed in Table 3.2.
Therefore no fatigue damage occurs in the chord.
For Brace:
t = 4 mm and Srhs,brace,mp = 63 N/mm2. The hot spot stress range of 63 N/mm2 is
lower than the Constant Amplitude Fatigue Limit of 147 N/mm2 listed in Table 3.2.
Therefore no fatigue damage occurs in the brace.
Hence, no fatigue damage would be expected in this multiplanar joint.
65
8 Design Examples for RHS Joints

8.1 Example 1: Uniplanar RHS T-joints
Given:
A T-joint in square hollow sections is shown in Figure 8.1, where a fillet weld with weld throat
thickness of 5 mm is assumed.
Figure 8.1 A uniplanar RHS T-joint
The dimensions are:

Chord member:
Brace member:
Chord length:
RHS 200 x 200 x 12.5,

RHS 140 x 140 x 5,
L = 1500 mm
Ao = 8973 mm2,
A1 = 2661 mm2,
Wo = 513444 mm3
W1 = 114711 mm3
The T-joint is assumed to be fail-safe (failure of the joint does not result in failure of the
whole structure). The joint is assumed to be poorly accessible and subjected to constant
amplitude loading.
Problem:
To determine the nominal range of the brace axial force P for the design of the T-joint at
2 106 cycles, for a constant amplitude loading.
Solution:
Step 1: Parameters
= b1/bo = 140/200 = 0.7
2 = bo/to = 200/12.5 = 16
=8
= t1/to = 5/12.5 = 0.4
The parameters are within the validity range given in Table E.1.
66
Step 2: Nominal stress ranges

The brace nominal axial stress (Sn,ax) causes a bending stress in the chord (Sn,ch) of:
Sn,ch = (1/4) P (Lb1)/Wo = (1/4) Sn,ax A1 (Lb1)/Wo
The general expression for the total hot spot stress becomes
Srhs = SCFax Sn,ax + SCFch Sn,ch = Sn,ax (SCFax + SCFch (1/4) A1 (Lb1)/Wo)
By substituting the geometric dimensions
Srhs = Sn,ax (SCFax + SCFch 1.76)
Therefore
Sn,ax = Srhs/(SCFax + SCFch 1.76)
The nominal axial force range can be determined by
Prange = A1 Sn,ax
Step 3: SCFax for axial force on the brace
For braces (lines A and E)
From Figure 5.2, a quick estimation of 7.5 is obtained.
From Table E.1
SCFA&E,ax
= (0.013 + 0.693 0.278 2) (2)(0.790 + 1.898 2.109 )

= 0.362 161.085
= 7.33
2
In this example, a fillet weld is applied. According to Table E.1, the SCF in the brace (lines
A and E) should be multiplied by 1.4 to correct for the weld type. (This correction factor is
not applied to the chord side of the fillet weld.)
SCFA&E,ax
= 1.4 7.33 = 10.3
For chord (lines B, C and D)

From Figure 5.2, SCFB,ax is around 5.0.
From Table E.1
SCFB,ax
= (0.143 0.204 + 0.064 2) (2)(1.377 + 1.715 1.103 ) 0,75

= 0.03156 162.037 0.40.75
= 4.5
2
Similarly
SCFC,ax
SCFD,ax
= 4.27
= 2.07
67
Step 4: SCFch for loads on the chord

From Table E.1
SCFA&E,ch = 0
SCFB,ch = 0
SCFC,ch = 0.725 (2)0.248 0.19 = 0.725 160.1736 0.40.19 = 0.99
SCFC,ch = 2.0 (minimum value)
SCFD,ch = 1.373 (2)0.205 0.24 = 1.373 160.1435 0.40.24 = 1.64
SCFD,ch = 2.0 (minimum value)
Step 5: SCFbrace and SCFchord
As derived before in step 2, the total SCF = SCFax + SCFch 1.76
SCFA&E = SCFA&E,ax = 10.3
SCFB = SCFB,ax = 4.5
SCFC = SCFC,ax + SCFC,ch 1.76 = 4.27 + 2.0 1.76 = 7.79
SCFD = SCFD,ax + SCFD,ch 1.76 = 2.07 + 2.0 1.76 = 5.59
Therefore the maximum SCFs in brace and chord are
SCFbrace = SCFA&E = 10.3
SCFchord = SCFC = 7.8
Step 6: Hot spot stress Srhs at 2 106 cycles
From Table 3.1.
For braces (t = 5 mm):
log(Srhs,brace) = (12.476 log( Nf))/3 + 0.06 log(Nf) log (16/t)
= (12.476 log(2 106))/3 + 0.06 log(2 106) log(16/5)
= 2.058 + 0.191 = 2.249
Srhs,brace = 102.249 = 177 N/mm2
For chord (t = 12.5 mm):
log(Srhs,chord) = (12.476 log (Nf))/3 + 0.06 log(Nf) log (16/t)
= (12.476 log(2 106))/3 + 0.06 log(2 106) log(16/12.5)
= 2.058 + 0.041 = 2.099
Srhs,chord = 102.099 = 126 N/mm2
68
Step 7: Allowable nominal axial force range

A partial factor of safety should be applied to the hot spot stress range as explained in section
1.6. In this example, a factor of safety of 1.15 should be used based on the conditions
given in Table 1.2. The factored hot spot stress ranges become
Srhs, brace = 177/1.15 = 154 N/mm2
Srhs, chord = 126/1.15 = 110 N/mm2
From Step 2 and Step 5
Sn,ax = Srhs/SCF = Srhs,brace/SCFbrace = 154/10.3 = 15.0 N/mm2
or
Sn,ax = Srhs/SCF = Srhs,chord/SCFchord = 110/7.8 = 14.1 N/mm2
The lower one (14.1 N/mm2) should be used to determine the nominal axial force range.
Prange = A1 Sn,ax = 2661 14.1 = 37520 N = 37.5 kN
So the nominal axial force range in the brace should not be greater than 37.5 kN.
8.2 Example 2: Uniplanar RHS K-Joints with Gap

Given:
A uniplanar truss is shown in Figure 8.2. The layout of the truss is the same as that shown
on page 51 in CIDECT Design Guide for RHS Joints under Predominantly Static Loading
(Packer et al. [1992]). The joint eccentricity (e) is zero. An applied constant amplitude loading ranging from zero to the loads shown, is given in Figure 8.2.
The member sizes are:
Top chord:
Brace:
Bottom chord:
RHS 180 x 180 x 8,

RHS 100 x 100 x 4,
RHS 180 x 180 x 8,
Ao = 5410 mm2,
A1,2 = 1480 mm2,
Ao = 5410 mm2,
Wo =0.282 x 106 mm3

W1,2 =0.0446 x 106 mm3
Wo =0.282 x 106 mm3
The static strength has been found satisfactory.
Figure 8.2 A uniplanar truss subjected to a constant amplitude load range
69
Problem:
Determine the fatigue life of joint No. 6 indicated in Figure 8.2.
Solution:
Step 1: Parameters
= b1/bo = 100/180 = 0.556
2 = bo/to = 180/8 = 22.5
= t1/to = 4/8 = 0.5
= arc tan (2.4/3.0) = 38.7
g = bo/tan() b1/sin() = 180/tan(38.7) 100/sin(38.7) 64 mm
g = g/to = 64/8 = 8
The parameters are within the validity range given in Table E.2.
Step 2: Structural analysis
A structural analysis is carried out assuming a continuous chord and pin-ended braces as
described in Section 3.2.2. The axial forces and bending moments found in joint No. 6 are
given in Figure 8.3. They can be treated as a combination of two load conditions shown in
Figure 8.4, i.e.:
Load condition 1: basic balanced axial loading
Load condition 2: chord loading (axial and bending)
Figure 8.3 Axial forces and bending moments in joint No. 6
Figure 8.4 Two load conditions for joint No. 6
Step 3: Nominal stress ranges in critical members

It can be seen from Figure 8.3 that the critical chord loading is in chord 1 due to a larger
tension force in it. Brace 2 with a tensile force range will be checked only. Note: In gen70
eral, it is presumed that only braces which have some load range in tension will be liable
to cause fatigue failure.
For load condition 1 (basic balanced axial loading):
brace,ax = MF 17.2 10 3/1480 = 1.5 12 = 18 N/mm2
For load condition 2 ( chord loading):
chord,ch = chord,ax + chord,ipb = MF 228.5 103/5410 1.106 106/(0.282 106)
= 1.5 42.2 3.9 = 59 N/mm2
Note that the chord bending moment relieves the tensile stress on the
connecting face of the chord. The values of MF (magnification factor) are given in
Table 2.2.
Step 4: SCF calculation for load condition 1 (basic balanced axial loading)
Chord
From Figure E.4 (reference value SCFo)

for = 0.556 and = 30, SCFo = 4.7
for = 0.556 and = 45, SCFo = 9.5
so that for = 0.556 and = 38.7, SCFo = 7.5
From Figure E.5 (correction
for = 0.5 and 2 = 20,
for = 0.5 and 2 = 25,
so that for = 0.5 and 2 = 22.5,
factor)
Correction Factor = 0.7
SCFch,ax = 7.5 0.9 = 6.8

Brace
From Figure E.6 (reference value SCFo)

for = 0.556 and = 30, SCFo = 6.7
for = 0.556 and = 45, SCFo = 10.5
so that for = 0.556 and = 38.7, SCFo = 8.9
From Figure E.7 (correction
for = 0.5 and 2 = 20,
for = 0.5 and 2 = 25,
so that for = 0.5 and 2 = 22.5,
factor)
SCFb,ax = 8.9 0.93 = 8.3

Step 5: SCF calculation for load condition 2 (chord loading)
Chord
From Figure E.8, SCFch,ch = 1.8,

but use minimum SCF value, SCFch,ch = 2.0
Brace
From Table E.2, SCFb,ch = 0 (negligible)
71

Load condition 1 (basic balanced axial load):
Srhs,chord = SCFch,ax brace,ax = 6.8 18 = 122 N/mm2
Srhs,brace = SCFb,ax brace,ax = 8.3 18 = 149 N/mm2
Srhs,chord = SCFch,ch chord,ch = 2.0 59 = 118 N/mm2
Srhs,brace = SCFb,ch chord,ch = 0 N/mm2
Srhs,chord = 122 + 118 = 240 N/mm2
Srhs,brace = 149 + 0 = 149 N/mm2
A partial safety factor on hot spot stress ranges is required for design. For this example the
joint is assumed to be non fail-safe and accessible. From Table 1.2, the partial safety factor
is 1.25.
Srhs,chord = 1.25 240 = 300 N/mm2
Srhs,brace = 1.25 149 = 186 N/mm2
Step 8: Fatigue life of joint No. 6
For Fatigue Cracking in the Chord:
t = 8 mm and Srhs,chord = 300 N/mm2
Table 3.1 or Figure 3.3 can now be used to determine the fatigue life.
From Table 3.1
log(Nf) =
12.476 3 log(Srhs)
12.476 3 log(300)
=
= 5.33
16
1 0.18 log( )
1 0.18 log( 16 )
t
8
Nf = 105.33 = 214,000 cycles
72
For Fatigue Cracking in the Brace:

t = 4 mm and Srhs,brace = 186 N/mm2
From Table 3.1
log(Nf ) =
12.476 - 3 log(Srhs )
12.476 - 3 log(186)
=
= 6.36
16
1 - 0.18 log( )
1 - 0.18 log(16)
t
4
Nf = 106.36 = 2,270,000 cycles

Hence, the fatigue life expectancy of joint No. 6 is 214,000 cycles, with failure in the chord.
8.3 Example 3: Multiplanar RHS KK-Joints with Gap

Given:
It is assumed that a multiplanar RHS KK-joint has the same geometry and internal brace
loads in each plane as joint No. 6 in the previous design example on uniplanar RHS Kjoints described in Chapter 8.2. It is assumed that the internal brace loads are totally antisymmetrical, i.e. m = 1. Therefore the loads in the chord member can be considered
negligible, see Figure 8.5.
It should be noted that the external constant amplitude load ranges are different to those
given in example 8.2. Therefore it is not the purpose of this example to show the multiplanar
effect on fatigue life. The purpose of this example is to demonstrate the design procedures
for a multiplanar joint under a given internal load condition.
The multiplanar parameter (), i. e. the angle between planes with braces, is 90.
Figure 8.5 Assumed load range on the multiplanar joint
Problem:
Determine the fatigue life of the multiplanar RHS KK-joint.
Solution:
Step 1: Multiplanar Correction Factors (MCFs) on SCFs
73
From Table 5.2

For load condition 1 (basic balanced axial loading), m = 1
chord:
brace:
MCFch,ax = 1.25
MCFb,ax = 1.25
For load condition 2 (chord loading), m = 1

chord:
brace:
MCFch,ch = 1.0
MCFb,ch = 1.0
Step 2: Modified SCFs

SCFs for uniplanar joints used in this example have been taken from the example
in chapter 8.2.
chord: SCFch,ax,mp = MCFch,ax SCFch,ax,up = 1.25 6.8 = 8.5
brace: SCFb,ax,mp = MCFb,ax SCFb,ax,up = 1.25 8.3 = 10.4
where mp stands for multiplanar and up stands for uniplanar.
For load condition 2 (chord loading)
chord:
brace:
SCFch,ch.mp = SCFch,ch,up = 2.0

SCFb,ch,mp = SCFb,ch,up = 0 (negligible)

Srhs,chord,mp = SCFch,ax,mp brace,ax = 8.5 18 = 153 N/mm2
Srhs,brace,mp = SCFb,ax,mp brace,ax = 10.4 18 = 187 N/mm2
Because the chord stress is zero,
Srhs,chord,mp = 0 N/mm2
Srhs,brace,mp = 0 N/mm2
Srhs,chord,mp = 153 + 0 = 153 N/mm2
Srhs,brace,mp = 187 + 0 = 187 N/mm2
74
Srhs,chord,mp = 1.25 153 = 191 N/mm2

Srhs,brace,mp = 1.25 187 = 234 N/mm2
Step 5: Fatigue life of multiplanar joint
For Chord:
t = 8 mm and Srhs,chord,mp = 191 N/mm2
From Table 3.1
log(Nf) =
12.476 3 log(Srhs )
12.476 3 log(191)
=
= 5.96
16
1 0.18 log( )
1 0.18 log( 16 )
t
8
Nf = 105.96 = 903,000 cycles

The fatigue life of the chord in this multiplanar RHS KK-joint is about
903,000 cycles.
For Brace:
t = 4 mm and Srhs,brace,mp = 234 N/mm2
From Table 3.1
log(Nf) =
12.476 3 log(Srhs)
12.476 3 log(234)
=
= 6.02
16
1 0.18 log( )
1 0.18 log( 16 )
t
4
Nf = 106.02 = 1,050,000 cycles

The fatigue life of the brace in this multiplanar RHS KK-joint is about
1,050,000 cycles.
Hence, the fatigue life of the multiplanar joint is 903,000 cycles, with failure in the chord.
75
9 References
AISC: Load and resistance factor design specification for structural steel buildings,
American Institute of Steel Construction, Chicago, USA, 1993.
API: Recommended practice for planning, designing and constructing fixed offshore
platforms, API-PR2A, American Petroleum Institute, Dallas, USA, 1991.
AWS: Structural welding code steel, ANSI/AWS D1.1-98, American Welding Society,
Miami, USA, 1998.
Bell, R., Vosikovsky, O. and Bain, S.A.: The significance of weld toe undercut in the fatigue
of steel plate T-joints. Int. J. Fatigue, 11(1), 1989, pp 311.
Berge, S. and Webster, S.E.: The size effect on the fatigue behaviour of welded joints,
Proceedings, Steel in Marine Structures (SIMS87), 1987, pp 179203.
Bergmann, R., Mutsui, C., Meinsma, C. and Dutta, D.: Design guide for concrete filled
hollow section columns under static and seismic loading. CIDECT-series Construction
with hollow steel sections, Serial no. 5, Verlag TV Rheinland, Cologne, Federal Republic
of Germany, 1995.
Bignonnet, A.: Improving the fatigue strength of welded steel structures, PS4, International
Conference on Steel in Marine Structures, Delft, The Netherlands, June, 1987.
CSA: Limit states design of steel structures, CAN/CSA-S16.1-94, Canadian Standards
Association, Toronto, Canada, 1994.
DEn: Offshore installation: guidance on design and construction, Department of Energy,
London, UK, 1990.
DEn: Background to new fatigue design guidance for steel joints in offshore structures,
Internal Report, Department of Energy, London, UK, 1993.
Dijkstra, O.D., van Foeken, R.J., Romeijn, A., Karamanos, S.A., van Wingerde, A.M., Puthli,
R.S., Herion, S. and Wardenier, J.: Fatigue design guide for circular and rectangular hollow section
multiplanar joints, Draft Final Report, TNO-Report, 91-CON-R1331, Delft, The Netherlands, 1996.
Dimitrakis, S.D., Lawrence, F.V. and Mohr, W.C.: S-N curves for welded tubular joints,
Proceedings, 14th International Offshore Mechanics and Arctic Engineering Symposium
(OMAE95), Volume III, 1995, pp 209222.
Dutta, D., Wardenier, J., Yeomans, N., Sakae, K., Bucak, . and Packer, J.A.: Design guide
for fabrication, assembly and erection of hollow section structures. CIDECT-series
Construction with hollow steel sections, Serial no. 7, Verlag TV Rheinland, Cologne,
Federal Republic of Germany, 1997.
EC3, ENV 1993-1-1, 1992, Eurocode 3: Design of steel structures Part 1.1: General rules
and rules for buildings. European Committee for Standardisation (CEN), London, UK, 1992.
EC1, Eurocode 1: Basis of Design and Actions on Structures, European Committee for
Standardisation (CEN), London, UK, 1994.
76
ECCS-TC6: Recommendations for the fatigue design of structures. European Convention

for Constructional Steelwork, 1985.
Efthymiou, M.: Development of SCF formulae and generalized functions for use in fatigue
analyses, OTJ88, Surrey, UK, 1988.
Efthymiou, M. and Durkin, S.: Stress concentration in T/Y and gap/overlap K-joints, in
Behaviour of Offshore Structures, Elsevier Science Publishers B.V., Amsterdam, The
Netherlands, 1985, pp 429440.
Fisher, J., Frank, K.H., Hirt, M.A. and McNamee, B.M.: Effect of weldment on fatigue
strength of steel beam, NCHRP Report No. 102, Lehigh University, Bethlehem,
Pennsylvania, USA, 1970.
Frater, G.S.: Performance of welded rectangular hollow structural section trusses, PhD
thesis, University of Toronto, Canada, 1991.
Gurney, T.R.: Fatigue of welded structures, Cambridge University Press, 2nd edition,
Cambridge, UK, 1979.
Haagensen, P.J.: Effect of TIG dressing on fatigue performance and hardness of steel
weldment, ASTM STP 648, 1978.
Haagensen, P.J.: Improvement techniques, Proceedings, International Symposium on the
Occasion of the Retirement of Prof. J de Back, Delft, The Netherlands, 1989, pp 7795.
Haagensen, P.J.: IIWs Round Robin and Design Recommendations for Improvement
Methods, IIW International Conference on Performance of Dynamically Loaded Welded
Structures, July, San Francisco, USA, 1997, pp 305316.
Haagensen, P. J. and Maddox, S. J.: IIW Recommendations for Weld toe improvement by
Burr Grinding, TIG dressing and hammer peening for steel and aluminium Structures, IIW.
Doc. XIII-1815-00, June, 2000.
Herion, S.: Multiplanar K-joints made of RHS, PhD thesis, University of Karlsruhe, Germany, 1994.
Herion, S. and Puthli, R.S.: Fatigue design and secondary bending moments in RHS Kjoints with gap, The Eighth International Symposium on Tubular Structures, Singapore,
August, 1998, pp 315322.
IIW: Recommended fatigue design procedure for hollow section joints. Part 1 hot spot
stress method for nodal joints. International Institute of Welding Subcommission XV-E, IIW
Doc. XV-582-85, IIW Assembly, Strasbourg, France, 1985.
JSSC: Fatigue design recommendations for steel structures, Japanese Society of Steel
Construction, English version, December, 1995.
Karamanos, S.A., Romeijn, A. and Wardenier, J.: Stress concentrations and joint flexibility
effects in multi-planar welded tubular connections for fatigue design, Stevin Report 6-98-05,
CIDECT Report 7R-17/98, Delft University of Technology, Delft, The Netherlands, 1997.
Knight, J.W.: Improving the fatigue strength of fillet welded joints by grinding and peening,
Welded Research Int., Vol. 8, No. 6, 1978.
77
Kobyashi, K. et al.: Improvements in the fatigue strength of fillet welded joint by use of the
new welding electrode, IIW Document XIII-828-77, 1977.
Kurobane, Y.: Recent development in the fatigue design rules in Japan. Proceedings,
International Symposium on the Occasion of the Retirement of Prof. J. de Back, Delft, The
Netherlands, 1989, pp 173187.
Maddox, S.J.: Fatigue strength of welded structures, Abington Publishing, Cambridge, UK,
1991.
Mang, F. and Bucak, .: Fatigue behaviour of welded joints in trusses of steel hollow
sections. IABSE Colloquium on Fatigue of Steel and Concrete Structures, Lausanne, 1982.
Mang, F., Herion, S., Bucak, ., Dutta, D.: Fatigue behaviour of K-joints with gap and with
overlap made of rectangular hollow sections. The Third International Symposium on
Tubular Structures, Lappeenranta, Finland 1989, pp 297309.
Mang, F. and Bucak, .: Tauglichkeit und Lebensdauer von bestehenden Stahlbauwerken,
Stahlbau Handbuch, Band 1 Teil B, Stahlbau, Verlagsgesellschaft mbH, Kln, Germany,
1996.
Marshall, P.W.: Connections for welded tubular structures, IIW International Conference on
Welding of Tubular Structures, Boston, USA,1984, pp 154.
Marshall, P.W.: Design of welded tubular connections basis and use of AWS code
provisions. Elsevier Science Publishers, Amsterdam, The Netherlands, 1992.
Mashiri, F.R., Zhao, X.-L. and Grundy, P.: Effect of weld profile on the fatigue life of thin-walled
cruciform joint, The Eighth International Symposium on Tubular Structures, Singapore,
1998, pp 331340.
Mori, T., Zhao, X.-L. and Grundy, P.: Fatigue strength of transverse single-sided fillet
welded joints, Australian Civil/Structural Engineering Transactions, Vol. CE39, No. 2 and
No. 3, 1997, pp 95105.
Nguyen, T.N. and Wahab, M.A.: A theoretical study of the effect of weld geometry parameters
on fatigue crack propagation life, Engineering Fracture Mechanics, 51 (1), 1995, pp 118.
Niemi, E.J.: A novel fatigue analysis approach for tubular welded joints. Proceedings,
Fatigue Design95, Helsinki, Finland, 1995, pp 189201.
Noordhoek, C., Wardenier, J. and Dutta, D.: The fatigue behaviour of welded joints in square
hollow sections, Part 2 Analysis, Stevin Report 6-80-4, TNO-IBBC-Report
BI-80-10/0063.4.3821, Delft, The Netherlands, 1980.
Packer, J.A., Wardenier, J., Kurobane, Y., Dutta, D. and Yeomans, N.: Design guide for
rectangular hollow section (RHS) joints under predominantly static loading. CIDECT-series
Packer, J.A. and Wardenier, J.: Stress concentration factors for non-90 X-Connections
made of square hollow sections, Can. J. Civ. Eng., 25(2), 1998, pp 370375.
78
Puthli, R.S., de Koning, C.H.M., van Wingerde, A.M., Wardenier, J. and Dutta, D.: Fatigue
strength of welded unstiffened RHS-joints in latticed structures and vierendeel girders, Final
Report Part III: Evaluation for Design Rules, TNO-IBBC Report No. BI-89-097/63.5.3820,
Stevin Report No. 25-6-89-36/A1, June, Delft, The Netherlands, 1989.
Puthli, R.S., Wardenier, J., Mang, F. and Dutta, D.: Fatigue behaviour of multiplanar welded
hollow section joints and reinforcement measures for repair, Final report Part V Evaluation
and design recommendation, TNO-Bouw Report No. BI-92-0079/21.4.6394, Stevin Report
No. 6.92.17/A1/12.06, Sept., Delft, The Netherlands, 1992.
Puthli, R. and Herion, S. (1996): Stress concentration and secondary moment distribution
in RHS joints for fatigue design. Background document for fatigue design guide on RHS,
University of Karlsruhe, Germany, 1996.
Romeijn, A., Puthli, R.S., de Koning, C.H.M. and Wardenier, J.: Stress and strain concentration
factors of multiplanar joints made of circular hollow sections, The Second International
Offshore and Polar Engineering Conference, San Francisco, USA, Vol. IV, 1992, pp 384393.
Romeijn, A., Wardenier, J., de Koning, C.H.M., Puthli, R.S. and Dutta, D.: Fatigue behaviour
and influence of repair on multi planar K-joints made of circular hollow sections. The Third
International Offshore and Polar Engineering Conference, Singapore, Vol. IV, 1993, pp 2736.
Romeijn, A.: Stress and strain concentration factors of welded multiplanar tubular joints,
PhD Thesis, Delft, The Netherlands, 1994.
Romeijn, A., Karamanos, S.A. and Wardenier, J.: Effects of joint flexibility on the fatigue
design of welded tubular lattice structures, The Seventh International Offshore and Polar
Engineering Conference, Honolulu, USA, Vol. IV, 1997, pp 9097.
Rondal, J., Wrker, K.G., Dutta, D., Wardenier, J. and Yeomans, N.: Structural stability of
hollow sections. CIDECT-series Construction with hollow steel sections, Serial no. 2,
Verlag TV Rheinland, Cologne, Federal Republic of Germany, 1991.
SAA: Steel structures, Australian Standard AS4100, Standards Association of Australia,
Sydney, Australia, 1990.
Sedlacek, G., Grotmann, D., Gusgen, J., Jo, J.B. and Dutta, D.: The determination of the
fatigue resistance of steel structures on the basis of the combined methods, The Second
International Offshore and Polar Engineering Conference, San Francisco, USA, Vol. IV,
1992, pp 296300.
Sedlacek, G., Grotmann, D., Schfers, M. and Zhao, X.-L.: Fatigue Behaviour of hollow
section joints, CIDECT 7N Project Report, RWTH, Aachen, Germany, 1998.
Swanmidas, A.S.J., Cheema, P.S. and Muggeridge, D.B.: Fatigue strength of fillet-welded
cruciform joints. Can. J. Civ. Eng., 16(2), 1989, pp 162171.
Thorpe, T.W. and Sharp, J.V.: The fatigue performance of tubular joints in air and sea water,
MaTSU Report, Harwell Laboratory, Oxfordshire, UK, 1989.
Twilt, L., Hass, R., Klingsch, W., Edwards, M. and Dutta, D.: Design guide for structural hollow
section columns exposed to fire. CIDECT-series Construction with hollow steel sections,
Serial no. 4, Verlag TV Rheinland, Cologne, Federal Republic of Germany, 1996.
79
Wardenier, J.: Hollow section joints, Delft University Press, Delft, The Netherlands, 1982.
Wardenier, J., Kurobane, Y., Packer, J.A., Dutta, D. and Yeomans, N.: Design guide for circular hollow section (CHS) joints under predominantly static loading. CIDECT-series
Wardenier, J., Dutta, D., Yeomans, N., Packer, J.A. and Bucak, .: Design guide for structural
hollow sections in mechanical applications. CIDECT-series Construction with hollow steel
sections, Serial no. 6, Verlag TV Rheinland, Cologne, Federal Republic of Germany, 1995.
van Delft, D.R.V.: A two dimensional analysis of the stresses at the vicinity of the weld toes
of tubular structures, Stevin Report 6-18-8, Delft University of Technology, Delft, The
Netherlands, 1981.
van Delft, D.R.V., Noordhoek, C. and de Back, J.: Evaluation of the European fatigue test
data on large-sized welded tubular joints for offshore structures, Offshore Technology
Conference, Houston, USA, paper OTC 4999,1985.
van Delft, D.R.V., Noordhoek, C. and Da Re, M.L.: The results of the European fatigue tests
on welded tubular joints compared with SCF formulas and design lines, Steel in Marine
Structures, Elsevier Applied Science Publishers, Ltd., Delft, The Netherlands, 1987,
pp 565577.
van Wingerde, A.M.: The fatigue behaviour of T and X joints made of square hollow sections,
Heron, The Netherlands, Vol. 37, No. 2, 1992, pp 1180.
van Wingerde, A.M., Packer, J.A., Wardenier, J. and Dutta, D.: The fatigue behaviour of
K-joints made of square hollow sections, CIDECT Report 7P-19/96, University of Toronto,
Canada, 1996.
van Wingerde, A.M., Packer, J.A., Wardenier, J. and Dutta, D.: Simplified design graphs for
the fatigue design of multiplanar K-joints with gap. CIDECT Report 7R-01/97, January,
University of Toronto, Canada, 1997a.
van Wingerde, A.M., Packer, J.A. and Wardenier, J.: IIW fatigue rules for tubular joints, IIW
International Conference on Performance of Dynamically Loaded Welded Structures, July,
San Francisco, USA, 1997b, pp 98107.
van Wingerde, A.M., van Delft, D.R.V., Wardenier, J. and Packer, J.A.: Scale Effects on the
Fatigue Behaviour of Tubular Structures, IIW International Conference on Performance of
Dynamically Loaded Welded Structures, July, San Francisco, USA, 1997c, pp 123135.
van Wingerde, A.M., Wardenier, J. and Packer, J.A.: Commentary on the Draft
Specification for Fatigue Design of Hollow Section Joints, The Eighth International
Symposium on Tubular Structures, Singapore, August, 1998a, pp 117127.
van Wingerde, A.M., Wardenier, J. and Packer, J.A.: Simplified design graphs for the fatigue design of multiplanar K-joints with gap, Final Report, Stevin Report 6-98-34, Delft,
The Netherlands, 1998b.
Zhao, X.-L. and Puthli, R.S.: Comparison of SCF formulae and fatigue strength for uniplanar RHS K-joints with gap, IIW Document XV-E-98-235, 1998.
80
Appendix A: Fatigue Actions

The fatigue actions can be included in a design by considering only the service or working
conditions, e.g. by specifying the particular ways that cranes are loaded and operated or
by specifying particular actions of machinery. These indications may be used to determine
the relevant fatigue loading specifically for the structure. The fatigue loading can be taken
from appropriate recommendations or guidelines, e.g. the ECCS TC6-Recommendations
(1985).
The fatigue actions can also be included in a design by referring to a relevant code for the
fatigue loading of railway or road bridges and cranes or that for wind loading, e.g.
Eurocode 1 (1994). Additional data, such as spectra, damage equivalent loads and safety
factors, may be given in the project specifications, and may supplement or replace code
provisions.
A simple example is given in Figure A.1 to demonstrate the general procedures of
determination of fatigue loading and fatigue damage assessment. The steps are shown below.
(a) Determination of one or more service sequences, typical for the total service life of the
structure, e.g. loads PA and PB shown in Figure A.1 (a) are typical load cycles.
(b) Determination of the stress history at the relevant structural detail for these service
sequences, e.g. stress range (S) history at the detail X-X location shown in Figure A.1 (b).
(c) Breaking the stress range history into a stress spectrum using a cycle counting method
if the stress history is of a variable amplitude type. Two commonly used cycle counting
methods are the Rainflow method and the Reservoir method. The Rainflow method is
more convenient for analyzing long stress histories using a computer. The Reservoir
method is easy to use by hand for short stress histories as shown in Figure A.1 (c).
(d) Simplifying the stress spectrum obtained in step (c) into a manageable number of
bands as shown in Figure A.1 (d).
(e) Determination of endurance (Ni) for each stress level (Si) according to an appropriate
fatigue strength curve as shown in Figure A.1 (e).
(f) Application of an appropriate damage accumulation rule such as the Palmgren-Miners
Rule shown in Figure A.1 (f). The rule states that the damage done by all bands together
must not exceed unity. If failure is to be prevented before the end of the specified
design life, compliance with the Palmgren-Miners Rule is necessary.
81

Figure A.1 A simple example on fatigue assessment procedures
82
Appendix B: Detailed Categories for Classification Method

The detail categories for the classification method are listed in Table B.1 for simple
connections and Table B.2 for lattice girder joints. Note that the values for detail categories
shown here are those given in Eurocode 3 [EC3 (1992]). The values may change slightly
in other national codes. The tube thickness in Table B.1 is between 4 mm and 12.5 mm.
The tube thickness in Table B.2 is between 4 mm and 8 mm.
Table B.1 Detail Categories for Hollow Section and Simple Connections
Details loaded by nominal normal stresses
Detail
category
m=3
Constructional detail
Description
160
Rolled and extruded products

Non-welded elements. Sharp edges
and surface flaws to be improved by
grinding.
140
Continuous longitudinal welds

Automatic longitudinal welds
with no stop-start positions,
proven free of detectable
discontinuities.
71
Transverse butt welds

Butt welded end-to-end connection
of circular hollow sections
Requirements:
Height of the weld reinforcement
less than 10 % of weld with smooth
transitions to the plate surface
Welds made in flat position and
proven free of detectable discontinuities
Details with wall thickness greater
than 8 mm may be classified two
Detail categories higher (=> 90).
56
Transverse butt welds

Butt welded end-to-end connection
of rectangular hollow sections
Requirements:
Height of the weld reinforcement
less than 10 % of weld with smooth
transitions to the plate surface
Welds made in flat position and
proven free of detectable
discontinuities
Details with wall thickness greater
than 8 mm may be classified two
Detail categories higher (=> 71).
71
Welded attachments (non-loadcarrying welds)

Circular or rectangular section,
fillet welded to another section.
Section width parallel to stress
direction 100 mm.
83
Table B.1 Detail Categories for Hollow Section and Simple Connections (continued)
Detail
category
m=3
84
Description
50
Welded connections (load-carrying

welds)
Circular hollow sections, end-to-end
butt welded with an intermediate plate.
Requirements:
Welds proven free of detectable
discontinuities
Details with wall thicknesses greater than 8 mm may be classified
one Detail category higher (=> 56).
45

welds)
Rectangular hollow sections, end-toend butt welded with an intermediate
plate.
Requirements:
Welds proven free of detectable
discontinuities
Details with wall thicknesses greater than 8 mm may be classified
one Detail category higher (=> 50).
40

welds)
Circular hollow sections, end-to-end
fillet welded with an intermediate plate.
Requirements:
Wall thickness less than 8 mm.
36

welds)
Rectangular hollow sections, end-to-end
fillet welded with an intermediate plate.
Requirements:
Wall thickness less than 8 mm.
80
l 50mm
71
50 < l < 80 mm
63
80 < l < 100 mm
56
l > 100 mm
80
t 12 mm
71
t > 12 mm
80
t 12 mm
71
t > 12 mm
Longitudinal attachments (non-loadcarrying welds)

The Detail category varies according
to the length of the attachment l.
Transverse attachments
The end of weld more than 10 mm
from the edge of the plate.
Transverse attachments
Diaphragms of rectangular girders
welded to the flange or web.
Table B.1 Detail Categories for Hollow Section and Simple Connections (continued)
Detail
category
m=3
Description
80
Transverse attachment
The effect of welded shear
connectors on base material.
71
Cruciform joints (load-carrying

welds)
Full penetration weld. Inspected free
of detectable discontinuities.
Requirements:
The maximum misalignment of the
load-carrying plates should be
less than 15% of the thickness of
the intermediate plate.
36
50
Cruciform joints (load-carrying

welds)
Fillet welded connection. Two fatigue
assessments are required. Firstly,
root cracking is evaluated by
determining the stress range in the
weld throat area. Category 36.
Secondly, toe cracking is evaluated
by determining the stress range in
the load-carrying plates. Category 71.
Requirements:
The maximum misalignment of
the load-carrying plates should
be less than 15% of the thickness
of the intermediate plate.
t 20 mm
Cover plates (load-carrying welds)
End zones of single or multiple
welded cover plates, with or without
frontal weld. When the reinforcing
plate is wider then the flange, a frontal weld, carefully ground to remove
undercut, is necessary.
85
Table B.2 Detail Categories for Lattice Girder Joints

Detail
category
m=5
90
t0/ti = 2.0
45
t0/ti = 1.0
71
t0/ti 2.0
36
t0/ti = 1.0
71
t0/ti 1.4
56
t0/ti = 1.0
71
t0/ti 1.4
50
t0/ti = 1.0
Description
Joints with gap

Circular hollow sections, K and Njoints
Joints with gap

Rectangular hollow sections, K and
N-joints
Requirements
0.5 (b0 bi) g 1.1 (b0 bi)
g 2 t0
Joints with overlap

K-joints
Requirements
overlap between 30 and 100 %
Joints with overlap

N-joints
Requirements
overlap between 30 and 100 %
General Requirements
4 to 8 mm*
bo 200 mm
do 300 mm
4 t1 8 mm*
0.4 b1/bo 1.0
0.25 d1/do 1.0
35 50
0.5 ho e 0.25ho
0.5 do e 0.25do
(bo/to) (to/t1) 25*

(do/to) (to/t1) 25*
Out-of-plane eccentricity: 0.02 b0 or 0.02 d0

Fillet welds are permitted in braces with wall thicknesses 8 mm.
For Intermediate t0/ti values, use linear interpolation between nearest Detail Categories
Note that the braces and chords require separate fatigue assessments
* Based on actual test results. Different from Eurocode 3 where to 12.5 mm, t1 12.5 mm, bo/to 25
and do/to 25.
86
Appendix C: The Determination of SCFs by Testing

and Finite Element Analysis
C.1 Hot Spot Stress and SCF
The hot spot stress (also called geometric stress) method relates the fatigue life of a joint
to the so-called hot spot stress at the joint rather than the nominal stress. It takes the uneven stress distribution around the perimeter of the joint into account directly. The hot spot
stress range includes the geometrical influences but excludes the effects related to fabrication such as the configuration of the weld (flat, convex, concave) and the local
condition of the weld toe (radius of weld toe, undercut, etc.). The hot spot stress is the
maximum geometrical stress occurring in the joint where the cracks are usually initiated.
In the case of welded joints, this generally occurs at the toe of the weld.
The stress concentration factor (SCF) is the ratio between the hot spot stress at the joint
and the nominal stress in the member due to a basic member load which causes this hot
spot stress. The hot spot stress has to be determined at the weld toe position from the
stress field outside the region influenced by the local weld toe geometry. The location from
which the stresses have to be extrapolated, the so-called extrapolation region, depends
on the dimensions of the joint and on the position around the intersection. The boundaries
of the extrapolation region for CHS and RHS joints are defined in Table C.1 and Figure C.1
(Romeijn [1994]).
Table C.1 Boundaries of extrapolation region for CHS and RHS joints
distances from
weld toe
chord
saddle
Lr,min *)
brace
crown
0.4 t0
saddle
crown
0.4 t1
CHS
Lr,max **)
Lr,min *)
0.09 r0
0.4 r0 t0 r1 t1
0.65 r1 t1
0.4 t0
0.4 t1
Lr,min + t0
Lr,max + t
1
CHS
Lr,max **)
*) Minimum value for Lr,min is 4 mm.

**) Minimum value for Lr,max is Lr, min + 0.6 t1
The hot spot stress or the SCF can be determined using strain gauges or by finite element
analysis. This appendix will provide advice and suggestions on using these two methods
to determine the hot spot stress and SCF.
A minimum SCF = 2.0 is recommended unless it is negligible. The bases for minimum SCF
values are:
87
The SCFs are determined along limited fixed lines or locations of interest. The hot spot
stresses found may underestimate the true hot spot stress if the direction of the principal stresses deviates from these lines, especially if the stress concentration is less
pronounced.
Difficulties in FE modelling, such as the case where = 1.0 and the case where weld
shapes have a strong influence on SCFs.
Crack initiation from the root of the weld for low SCF values
Figure C.1 Definition of extrapolation region
C.2 Experimental Approach

Advice and suggestions are given for the following aspects:
a) Stress and strain components
Different views exist with regard to which stress (strain) component should be used to
determine the SCF: the principal stress (strain) or a stress perpendicular to the weld toe.
The principal stress is used in IIW, DEn and EC3, whereas the stress perpendicular to the
weld toe is used in AWS and API. The differences between the two stresses become less
significant near the weld toe (Marshall [1992], Romeijn et al. [1992]). Strains perpendicular
to the weld toe can be measured by simple strain gauges instead of strain gauge rosettes
which are required to determine the principal strains. The use of stresses (strains) perpendicular to the weld toe is recommended.
b) Location of strain gauges
Normal strains have to be measured in the extrapolation region perpendicular to the weld
toe for the chord and parallel to the brace axis for the brace locations. The extrapolation
region is defined in Figure C.1 and Table C.1. Lines of measurement or locations of interest are suggested in Figures 4.1, 4.5, 4.9, 4.10, 4.11, 5.1, 5.5 and 5.6, along which strip
strain gauges can be applied to measure the strain distribution.
88
c) Effect of bending moments

In addition to the applied in-plane bending (IPB) or out-of-plane bending (OPB) moments,
secondary bending moments exist in joints, especially those in lattice girders. The secondary bending moment is caused by the inevitable loading eccentricity and the joint flexibility
(Wardenier [1982], Romeijn et al. [1997], Herion and Puthli [1998]). It is necessary to measure the strains caused by bending moments and those caused by axial forces in a test.
This can be done by applying strain gauges in two or three cross-sections along each
brace member. The strains caused by bending and axial force can be separated by a simple superposition technique. The distance between the measuring sections in the brace
and the weld toe should be kept to at least 3d1 or 3b1 where d1 or b1 is the diameter or
width of the brace member.
d) Extrapolation method
As mentioned before, the hot spot stress range includes the geometrical influences but
excludes the effects related to fabrication such as the configuration of the weld (flat, convex,
concave) and the local condition of the weld toe (radius of weld toe, undercut, etc.). On the
one hand, the extrapolation of strain distribution should be carried out in the extrapolation
region defined in Table C.1 and Figure C.1. On the other hand, the extrapolation points
must be positioned in such a way that the strain gradients due to the global geometry
effects are represented. Two extrapolation methods are commonly used, namely linear and
quadratic, for the determination of the hot spot stress. They are shown in Figure C.2. For
joints in CHS, the linear extrapolation method can be used since the gradient is nearly linear
(Romeijn [1994]). For joints in RHS, the quadratic extrapolation method is required because
of the strong non-linear strain distribution observed (van Wingerde [1992]).
Figure C.2 Methods of extrapolation
e) Relationship between SNCF and SCF

Because only strains can be directly measured using strain gauges, the strain concentration
factor (SNCF) is determined first. Then the SNCF is converted to the stress concentration
factor (SCF).
89
The formulae for SNCF can be expressed as (Herion [1994])

SNCFRHS =
max/( ax + IPB + OPB)
for joints in RHS
SNCFCHS =
max/( ax + ( 2IPB + 2OPB))
for joints in CHS
where max is the extrapolated maximum strain, ax, IPB, and OPB are nominal strain
components caused by axial force, in-plane bending and out-of-plane bending respectively.
The SNCF can be converted to SCF using (Frater [1991], van Delft et al. [1987])
SCFRHS = 1.1 SNCFRHS
for joints in RHS
SCFCHS = 1.2 SNCFCHS
for joints in CHS
f) Others
The experiments have to be carried out in a well-equipped laboratory by specialists. One
should ensure that the boundary conditions and the loading positions are correct.
C.3 Finite Element Analysis

Advice and suggestions are given for the following aspects:
a) Preconsideration about the FE-Model
The FE analysis should be carried out with a validated FE package and by FE analysts
experienced in the use of FE programs for determining SCFs. Some of the FE packages
used in the past to determine SCFs include ABAQUS, ANSYS, DIANA and MARC. It is also
important to ascertain the scope of the problem with regard to the required computer
capacities. Geometrical symmetry and the symmetry of applied forces should be utilised
to simplify the problem. Attention should be paid to the boundary conditions in the plane
of symmetry.
b) Elements
The refinement of the FE mesh of a hollow section joint depends on the type of elements
and on the stress/strain gradient over the element. The refinement of the mesh should be
such that any further refinement does not result in a substantial change of the stress
distribution (outside the notch effect area). Element dimensions of 0.5t1 to 0.5to may be
used (Herion [1994]). With the use of solid elements, there should be no problem with this
ratio of side length to element thickness. In the case of thin shell elements with this dimension,
the reliability of the calculation needs to be checked. The element size in regions of less
interest can be increased to save computing time.
c) Modelling of details
Modelling the weld shape greatly improves the accuracy of SCFs. Therefore, using solid
elements to model the weld area and the extrapolation region is recommended. Because
of the high accuracy requirements, the use of 20-node solid elements with an integration
scheme of 2 x 2 x 2 is recommended (Romeijn [1994]).
90
For rectangular hollow sections, a stress redistribution around the corners can be observed
especially for butt (groove) welded joints. Modelling of the corners with several elements
is recommended. The minimum number of elements required depends on the thickness of
the tube (t), i.e. 2 for t 8 mm, 3 for 8 < t < 16 mm and 4 for t 16 mm (Herion [1994]).
The radius of corners should also be taken into account. Inside and outside corner radii
are different for hot-formed and cold-formed sections. Different values also exist between
manufacturers and countries.
d) Judgement of results
The comparison of predicted strain concentration factors (SNCF) with experimental values
should be carried out. Before comparing with experimental results, it is recommended to
check the following aspects first.
Simple geometry and loading. This can be done by using the check routines of the
pre-processor used to generate the model.
Boundary conditions. This can be done by checking the global static equilibrium condition,
i.e. comparing the calculated node forces at fixed nodes with applied loads.
Mesh division. This can be done by comparing the stresses in neighbouring elements. If
there is a big difference, it is often caused by the big differences in the element dimensions.
e) Others
Stresses and strains have to be determined in the extrapolation region perpendicular to the
weld toe or in the direction of the principal strain.
91
Appendix D: SCF Formulae and Graphs for CHS Joints

A minimum SCF of 2.0 is recommended for all joint types, all locations and all load
conditions unless otherwise specified.
D.1 Uniplanar CHS T and Y-Joints
Table D.1 SCFs for uniplanar CHS T and Y-joints
Locations
Range of validity
0.2 1.0
15 2 64
0.2 1.0
4 40
30 90
Chord-end fixity parameters C
C1 = 2 (C 0.5)
C2 = C/2
C3 = C/5
For fixed chord ends, C = 0.5
For pinned chord ends, C = 1.0
Otherwise
C = 0.7
Load conditions
92
Load condition 1
axial load with chord ends fixed
Load condition 2
axial load with general chord fixity
Load condition 3
in-plane bending
Load condition 4
out-of-plane bending
Table D.1 SCFs for uniplanar CHS T and Y-joints (continued)
Load condition 1
axial load with chord ends fixed

(uniplanar CHS T and Y-joints)
chord (saddle and crown)
brace (saddle and crown)
SCFch_saddle,ax = T1 F1
SCFch_crown,ax = T2
SCFb_saddle,ax = T3 F1
SCFb_crown,ax = T4
T1 = 1.1[1.113 (0.52)2] sin1.6

T2 = 0.2 [2.65+5 (0.65)2] + (0.25 3) sin
T3 = 1.3+ 0.52 0.1[0.1871.25 1.1(0.96)] sin(2.70.01)
T4 = 3+1.2 [0.12 exp (4)+0.011 2 0.045]+ (0.11.2)
if 12:
F1 = 1.0
if < 12:
F1 = 1(0.83 0.56 20.02) 0.23 exp[0.21 1.16 2.5]
where exp[x] = ex
Load condition 2
axial load with general chord fixity conditions

SCFch_saddle,ax = T5 F2
SCFch_crown,ax = T6
SCFb_saddle,ax = T3 F2
SCFb_crown,ax = T7
T5 = 1.1[1.113 (0.52)2] sin1.6 + C1 (0.86) 2(12)0.5 sin22

T6 = 0.2 [2.65+5 (0.65)2]+ (C2 3) sin
T7 = 3+1.2 [0.12 exp (4 )+0.011 20.045]+ (C3 1.2)
if 12:
F2 = 1.0
if < 12:
F2 = 1(1.43 0.97 20.03) 0.04 exp[0.71 1.38 2.5]
where exp[x] = ex
93
Table D.1 SCFs for uniplanar CHS T and Y-joints (continued)
Load condition 3
in-plane bending
SCFch_saddle,ipb = 0 (negligible)
SCFch_crown,ipb = T8
SCFb_saddle,ipb = 0 (negligible)
SCFb_crown,ipb = T9
T8 = 1.45 0.85 (10.68) sin0.7

T9 = 1 +0.65 0.4 (1.090.77) sin(0.061.16)
Load contition 4
SCFch_saddle,opb = T10 F3
SCFch_crown,opb = 0 (negligible)
SCFb_saddle,opb = T11 F3
SCFb_crown,opb = 0 (negligible)
T10 = (1.71.053) sin1.6

T11 = 0.95 0.46 (1.71.053) (0.990.47+0.084) sin1.6
if 12:
F3 = 1.0
if < 12:
F3 =10.55 1.8 0.16 exp[0.49 0.89 1.8]
where exp[x] = ex
94
D.2 Uniplanar CHS X-Joints

Table D.2 SCFs for uniplanar CHS X-joints
Locations
Range of validity
0.2 1.0
15 2 64
0.2 1.0
4 40
30 90
Load conditions
Load condition 1
balanced axial load with chord ends pinned
Load condition 2
in-plane bending
Load condition 3
95
Table D.2 SCFs for uniplanar CHS X-joints (continued)
Load condition 1
balanced axial load with chord ends pinned

(uniplanar CHS X-joints)
SCFch_saddle,ax = X1 F2
SCFch_crown,ax = X2
SCFb_saddle,ax = X3 F2
SCFb_crown,ax = X4
X1 = 3.87 [1.101.8] (sin )
1.7
X2 = 0.2 [2.65+5 (0.65)2] 3 sin

X3 = 1 + 1.9 0.5 0.9 (1.091.7) sin2.5
X4 = 3 + 1.2 [0.12 exp (4 )+0.011 20.045]
if 12:
F2 = 1.0
if < 12:
F2 = 1(1.43 0.97 20.03) 0.04 exp[0.71 1.38 2.5]
where exp[x] = ex
Load condition 2
in-plane bending
SCFs are the same as those for uniplanar T-joints subjected to in-plane bending,
as given in Table D.1.
Load condition 3
SCFch_saddle,opb = X5 F3
SCFch_crown,opb = 0 (negligible)
SCFb_saddle,opb = X6 F3
SCFb_crown,opb = 0 (negligible)
X5 = (1.561.34 4 ) (sin )
1.6
X6 = 0.95 0.46 (1.561.34 4 ) (0.990.47 +0.08 4) (sin )

if 12:
F3 = 1.0
if < 12:
F3 =10.55 1.8 0.16 exp[0.49 0.89 1.8]
where exp[x] = ex
96
1.6
D.3 Uniplanar CHS K-Joints with Gap

Table D.3 SCFs for uniplanar CHS K-joints with gap
Locations
Geometrical conditions:
no eccentricity
equal braces
Range of validity
0.30 0.60
24 2 60.0
0.25 1.00
30 60
Load conditions
Load condition 1
basic balanced axial loading
Load condition 2
chord loading (axial and bending)
97
Table D.3 SCFs for uniplanar CHS K-joints with gap (continued)
Load condition 1

(uniplanar CHS K-joints with gap)
chord
SCFch,ax =
12 0.5
0.4
1.1
SCFo,ch,ax = correction factor SCFo,ch,ax
SCFo,ch,ax of chord, balanced axial loading Correction factor for other values of 2 and
brace
SCFb,ax =

12
0.5

0.5
0.5

SCFo,b,ax = correction factor SCFo,b,ax
Minimum values of SCFb,ax are 2.64, 2.30 and 2.12 for = 30, 45, and 60 respectively.
SCFo,b,ax of braces, balanced axial loading
98
Correction factor for other values of 2 and
Table D.3 SCFs for uniplanar CHS K-joints with gap (continued)
Load condition 2

(uniplanar CHS K-joints with gap)
chord
0.3
SCFch,ch = 1.2

0.5

(sin )0.9
SCFch,ch is also given in the graph below where a minimum SCF of 2.0 is adopted.
brace
99
D.4 Multiplanar CHS XX-Joints

Table D.4 SCFs for multiplanar CHS XX-joints
Locations
no eccentricity
equal braces
Range of validity
0.3 0.6
15 2 64
0.25 1.0
= 90
= 90
= 2 arcsin() 16.2
Load conditions
Load condition 1
Load condition 2
balanced in-plane bending on braces
Load condition 3
balanced out-of-plane bending on braces
Load condition 4
100
Table D.4 SCFs for multiplanar CHS XX-joints (continued)
Load condition 1

(multiplanar CHS XX-joints)
Axial Load in Reference Braces (Pref)

axial load in reference braces Pref
chord (locations 1 and 2)
SCF1,ref,ax = 5
SCF2,ref,ax =
0.9
12
1.1
(1 )
1.15
12 0.5
SCFo,2,ref
brace (locations 3 and 4)

SCF3,ref,ax = 2.0
SCF4,ref,ax =
0.5
0.75
12 0.5
SCFo,4,ref
Axial Load in Carry-Over Braces (Pcov)

axial load in carry-over braces Pcov

SCF1,cov,ax = 0 (negligible)
SCF2,cov,ax =
1.1
1.15
12 0.5
SCFo,2,cov
No minimum value for SCF2,cov,ax

is required.
SCF3,cov,ax = 0 (negligible)
12 0.5
0.5
SCF4,cov,ax =
0.75
SCFo,4,cov
No minimum value for SCF4,cov,ax is required.
101
Load condition 2
balanced in-plane bending on braces

In-Plane Bending in Reference Braces (Mref)

0.6
0.8
SCF1,ref,ipb =
SCFo,1,ref
12
0.5

where SCFo,1,ref = 1.45 0.85 (1 0.68) sin0.7

SCF2,ref,ipb = 0 (negligible)
SCF3,ref,ipb = 2.0
SCF4,ref,ipb = 0 (negligible)
In-Plane Bending in Carry-Over Braces (Mcov)
SCF1,cov,ipb = SCF2,cov,ipb = 0 (negligible)
SCF3,cov,ipb = SCF4,cov,ipb = 0 (negligible)
Load condition 3

Out-of-Plane Bending in Reference

Braces (Mref)
SCF1,ref,opb = 0 (negligible)
SCF2,ref,opb =
1.25
12
1.05
0.5
SCFo,2,ref

SCF3,ref,opb = 2.0

12

0.65
SCF4,ref,opb =
102

0.5

0.65
SCFo,4,ref
out-of-plane bending in reference braces
Load condition 3

Out-of-Plane Bending in Carry-Over

Braces (Mcov)
out-of-plane bending in carry-over braces

SCF1,cov,opb = 0 (negligible)
SCF2,cov,opb =
1.25
1.05
12 0.5
SCFo,2,cov
No minimum value for SCF2,cov,opb

is required.
SCF3,cov,opb = 0 (negligible)
SCF4,cov,opb =

12
0.65

0.5
0.65

SCFo,4,cov
No minimum value for SCF4,cov,opb

is required.
Load condition 4


SCF1,ch = 2.0
SCF2,ch = 0 (negligible)
103
Appendix E: SCF Formulae and Graphs for RHS Joints

A minimum SCF of 2.0 is recommended for all joint types, all locations and all load conditions unless otherwise specified.
E.1 Uniplanar RHS T and X-Joints
Table E.1 SCFs for uniplanar RHS T and X-joints
Locations
Range of validity
0.35 1.0
12.5 2 25.0
0.25 1.0
Fabrication
For joints with fillet welds:
Multiply brace SCFs by 1.40.
Load conditions
1. axial force on the brace
2. in-plane bending on the brace
3. chord loading (axial and bending)
Load condition 1
axial force on the brace

(uniplanar RHS T and X-joints)
chord (lines B, C and D)

2
SCFB,ax = (0.143 0.204 + 0.064 2) (2)(1.377 + 1.715 1.103 ) 0.75

2
SCFc,ax = (0.077 0.129 + 0.061 2 0.0003 2) (2)(1.565 + 1.874 1.028 ) 0.75
2
SCFD,ax = (0.208 0.387 + 0.209 2) (2)(0.925 + 2.389 1.881 ) 0.75

for X-joints with = 1.0:
SCFC,ax is multiplied by a factor of 0.65
SCFD,ax is multiplied by a factor of 0.50
brace (lines A and E)
2
SCFA,ax = SCFE,ax = (0.013 + 0.693 0.278 2) (2)(0.790 + 1.898 2.109 )

Multiply brace SCFs (SCFA,ax, SCFE,ax) by 1.40 for brace side of weld.
104
Table E.1 SCFs for uniplanar RHS T and X-joints (continued)
Load condition 2
in-plane bending on the brace


2
SCFB,ipb = (0.011 + 0.085 0.073 2) (2)(1.722 + 1.151 0.697 ) 0.75

2
SCFC,ipb = (0.952 3.062 + 2.382 2 + 0.0228 2) (2)(0.690 + 5.817 4.685 ) 0.75
2
SCFD,ipb = (0.054 + 0.332 0.258 2) (2)(2.084 1.062 + 0.527 ) 0.75

2
SCFA,ipb = SCFE,ipb = (0.390 1.054 + 1.115 2) (2)(0.154 + 4.555 3.809 )

Multiply brace SCFs (SCFA,ipb and SCFE,ipb) by 1.40 for brace side of weld.
Load condition 3


SCFB,ch = 0 (negligible)
SCFC,ch = 0.725 (2)0.248 0.19
SCFD,ch = 1.373 (2)0.205 0.24
SCFA,ch = SCFE,ch = 0 (negligible)
105
E.2 Uniplanar RHS K-Joints with Gap

Table E.2 SCFs for uniplanar RHS K-joints with gap
Locations
equal braces
Range of validity
0.35 1.0
10 2 35
0.25 1.0
30 60
2 g
0.55 e/h0 0.25
Only the maximum SCFs for the braces

(among lines A and E) and chord (among
lines B, C and D) are given.
Load conditions
Load condition 1
Load condition 1
Load condition 2
(uniplanar RHS K-joints with gap)
chord (maximum SCF)

SCFch,ax = (0.48 0.5 2 0.012/ + 0.012/g ) (2)1.72 0.78 (g )0.2 (sin ())2.09
brace (maximum SCF)
SCFb,ax = (0.008 + 0.45 0.34 2) (2)1.36 0.66 (sin ())1.29
Load condition 2
chord (maximum SCF)
SCFch,ch = (2.45 + 1.23 ) (g )0.27
brace (maximum SCF)
106

Table E.2 SCFs for uniplanar RHS K-joints with gap (continued)
General format
(for graphical presentation)
SCF = SCFo Correction factor
where SCFo is the SCF for 2 = 24 and = 0.5. The correction factor depends on 2 and .
Load condition 1

brace (SCFo)
chord (SCFo)
use
use
use
use
Figure
Figure
Figure
Figure
E.1
E.2
E.3
E.4
for
for
for
for
g
g
g
g
=
=
=
=
1.0
2.0
4.0
8.0
use Figure E.6 for all g values
chord (correction factor)
brace (correction factor)
use Figure E.5
use Figure E.7
Load condition 2

chord (SCFch,ch)
brace
use Figure E.8
107
Figure E.1 The reference value SCFo for the chord of RHS K-joints with gap g = 1.0 (balanced axial loading)
108
109
Figure E.5 The correction factor on SCFo for the chord of RHS K-joints with gap (balanced axial loading)
Figure E.6 The reference value SCFo for the braces of RHS K-joints with gap all g (balanced axial loading)
110
Figure E.7 The correction factor for SCF for the braces of RHS K-joints with gap (balanced axial loading)
Figure E.8 The SCFch,ch for the chord of RHS K-joints with gap (chord loading)
111
E.3 Uniplanar RHS K-Joints with Overlap

Table E.3 SCFs for uniplanar RHS K-joints with overlap
Locations
equal braces
Range of validity
0.35 1.0
10 2 35
0.25 1.0
30 60
50 % Ov 100 %
0.55 e/hO 0.25
Only the maximum SCFs for the braces
(among lines A and E) and chord (among
lines B, C and D) are given.
Load conditions
Load condition 1
Load condition 2
Table E.3 SCFs for uniplanar RHS K-joints with overlap (continued using equations)
Load condition 1

(uniplanar RHS K-joints with overlap)
chord (maximum SCF)

SCFch,ax = (0.5 + 2.38 - 2.87 2 + 2.18 Ov + 0.39 Ov - 1.43sin()) (2)0.29 0.7
2
Ov0.73 - 5.53 sin () (sin())-0.4 0.08 OV
brace (maximum SCF)
SCFb,ax = (0.15 + 1.1 - 0.48 2- 0.14/Ov) (2)0.55 -0.3 Ov-2.57 + 1.62 (sin())0.31
2
Load condition 2

chord (maximum SCF)

SCFch,ch = (1.2 + 1.46 0.028 2)
brace (maximum SCF)
112
Table E.3 SCFs for uniplanar RHS K-joints with overlap (continued using graphs)
General format
(for graphical presentation)
SCF = SCFo Correction factor
where SCFo is the SCF for 2 = 24 and = 0.5. The correction factor depends on 2 and .
Load condition 1

chord (SCFo)
brace (SCFo)
use Figure E.9 for 50% overlap


chord (correction factor)
brace (correction factor)
use Figure E.12
use Figure E.16
Load condition

chord (SCFch,ch)
brace
use Figure E.17
113
Figure E.9 The reference value SCFo for the chord of RHS K-joints with 50% overlap (balanced axial loading)
114
Figure E.12 The correction factor on SCFo for the chord of RHS K-joints with overlap (balanced axial loading)
115
Figure E.13 The reference value SCFo for the braces of RHS K-joints with 50% overlap (balanced axial loading)
116
Figure E.16 The correction factor on SCFo for the braces of RHS K-joints with overlap (balanced axial loading)
117
Figure E.17 The SCFch,ch for the chord of RHS K-joints with overlap (chord loading)
118
CIDECT Comit International

pour le Dveloppement et lEtude
de la Construction Tubulaire
International Committee
for the Development and Study
of Tubular Structures
CIDECT, founded in 1962 as an international association, joins together the research
resources of the principal hollow steel section manufacturers to create a major force in the
research and application of hollow steel sections worldwide.
The objectives of CIDECT are:
to increase the knowledge of hollow steel sections and their potential application by
initiating and participating in appropriate research and studies
to establish and maintain contacts and exchanges between the producers of the hollow steel sections and the ever increasing number of architects and engineers using
hollow steel sections throughout the world
to promote hollow steel section usage wherever this makes for good engineering practice and suitable architecture, in general by disseminating information, organizing congresses, etc.
to co-operate with organizations concerned with practical design recommendations,

regulations or standards at national and international level.
Technical activities
The technical activities of CIDECT have centred on the following research aspects of hollow steel section design:
Buckling behaviour of empty and concrete-filled columns

Effective buckling lengths of members in trusses
Fire resistance of concrete-filled columns
Static strength of welded and bolted joints
Fatigue resistance of joints
Aerodynamic properties
Bending strength of hollow steel section beams
Corrosion resistance
Workshop fabrication, including section bending
The results of CIDECT research form the basis of many national and international design
requirements for hollow steel sections.
119
CIDECT Publications
The current situation relating to CIDECT publications reflects the ever increasing emphasis
on the dissemination of research results.
The list of CIDECT Design Guides, in the series Construction with Hollow Steel Sections,
already published, or in preparation, is given below. These design guides are available in
English, French, German and Spanish.
1. Design guide for circular hollow section (CHS) joints under predominantly static loading
(1991)
2. Structural stability of hollow sections (1992, reprinted 1996)
3. Design guide for rectangular hollow section (RHS) joints under predominantly static loading
(1992)
4. Design guide for structural hollow section columns in fire (1995, reprinted 1996)
5. Design guide for concrete filled hollow section columns under static and seismic loading
(1995)
6. Design guide for structural hollow sections in mechanical applications (1995)
7. Design guide for fabrication, assembly and erection of hollow section structures (1998)
8. Design guide for circular and rectangular hollow section welded joints under fatigue
loading (2000)
9. Design guide for structural hollow section column connections (in preparation)
In addition, taking into account the ever increasing place of steel hollow sections in
internationally acclaimed high tech structures a new book Tubular Structures in
Architecture has been published with the sponsorship of the European Community. This
is also available in English, French, German and Spanish.
Copies of the design guides, the architectural book and research papers may be obtained
from members or from:
The Steel Construction Institute
Silwood Park
Ascot
Berkshire SL5 7QN
England
Tel.:
Fax:
E-mail:
URL:
120
+44 (0) 13 44 62 33 45
+44 (0) 13 44 62 29 44
f.awan@steel-sci.com
http//www.steel-sci.org
CIDECT Organisation (2000)
President: B. Becher (Germany)

Vice-President: C. L. Bijl (The Netherlands)
A General Assembly of all members meeting once a year and appointing an Executive
Committee responsible for administration and execution of established policy
A Technical Commission and Working Groups meeting at least once a year and directly responsible for the research and technical promotion work
Present members of CIDECT are (2000)
Aceralia Transformados, Spain

BHP Steel, Australia
British Steel Tubes and Pipes, United Kingdom
Hoogovens Buizen, The Netherlands
IPSCO Inc., Canada
Mannstaedt Werke GmbH & Co., Germany
Rautaruukki Oy, Finland
Tata Iron and Steel Co., India
Tubeurop, France
A.G. Tubos Europa, Spain
Vallourec & Mannesmann Tubes, Germany
Voest Alpine Krems, Austria
Care has been taken to ensure that all data and information herein is factual and that
numerical values are accurate. To the best of our knowledge, all information in this book is
accurate at the time of publication.
CIDECT, its members and the authors assume no responsibility for errors or misinterpretation of the information contained in this book or in its use.
Acknowledgements for photographs:
The authors express their appreciation to the following firms for making available the photographs used in this Design Guide:
British Steel Tubes and Pipes
IPSCO Inc. Canada
Tubeurop France
Vallourec & Mannesmann Tubes Germany
Voest Alpine Krems
121

Design Guide 8 - For RHS and CHS Under Fatigue Load

Uploaded by

Copyright:

Available Formats

Design Guide 8 - For RHS and CHS Under Fatigue Load

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Design Guide 8 - For RHS and CHS Under Fatigue Load

Uploaded by

Copyright:

Available Formats

What is the purpose of this document?

What is the purpose of this document?

What topics does this document cover in its research?

What topics does this document cover in its research?

DESIGN GUIDE

FOR CIRCULAR AND RECTANGULAR

Comit International pour le Dveloppement et lEtude de la

X.-L. Zhao, Monash University

Die Deutsche Bibliothek CIP-Einheitsaufnahme

Hot Spot Stress Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

SCF Calculations for CHS Joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

Uniplanar T and Y-Joints . . . . .

SCF Calculations for RHS Joints

Uniplanar T and X-Joints . . . . .

Structural Detailing for Fatigue and Reinforcement

Structural Detailing for Fatigue . . . .

6.2.5 Reinforcement of K and N-joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

Design Examples for CHS Joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60

Example 1: Uniplanar CHS K-Joints with Gap . . . . . . . . . . . . . . . . . . . . . . . . . 60

Design Examples for RHS Joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

Example 1: Uniplanar RHS T-Joints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

Appendix A: Fatigue Actions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

Appendix C: The Determination of SCFs by Testing and Finite Element Analysis

D: SCF Formulae and Graphs for CHS Joints . . . . . . . . . . . . . . . . . . . . 92

Appendix E: SCF Formulae and Graphs for RHS Joints

General information about CIDECT

Circular Hollow Section

chord width of RHS

relative chord length (2L/do or 2L/bo)

normal stress due to in-plane bending

Abbreviations of organisations, institutes etc.

American Petroleum Institute

Figure 1.1 Full revolving plough machine

Figure 1.2 Air seeder farming equipment

Figure 1.3 Mobile crane

Figure 1.4 Giant wheel in an amusement park

Figure 1.5 Offshore platforms

Figure 1.6 Grab bucket

Figure 1.7 Ski installation

Figure 1.8 Bridge

Figure 1.9 Bridge

Figure 1.10 Bridge

Figure 1.11 Communication tower

1.2 Types of Joints and Loading

(a) CHS T-Joints

(b) CHS Y-Joints

(c) CHS X-joints

(d) CHS K-joints with gap

(e) RHS T-joints

(f) RHS X-joints

(g) RHS K-joints with gap

(h) RHS K-joints with overlap

Figure 1.12 Types of uniplanar joints considered in this guide

(a) CHS XX-joints

(b) CHS KK-joints with gap

(c) RHS KK-joints with gap

Figure 1.14 Types of Loading

1.3 Fatigue Life Estimation

1.4 Fatigue Resistance

For basic balanced axial load, SCF decreases as decreases