Steel Buildings Design Notes PDF

Ferenc Papp
Ph.D., Dr habil
Steel Buildings
DESIGN NOTES
Reviewed by
Dr. Bla Verci honorary lecturer
This work is to the scientific program of the Development of quality-oriented and harmonized R+D+I
strategy and functional model at BME.
This work is supported by the New Szchenyi Plan.
The rroject ID is TMOP-4.2.1/B-09/1/KMR-2010-0002.
2010-2011 Budapest
Ferenc Papp
Steel Buildings Design Notes
Practice 1
PRELIMINARY DESIGN
Written in the framework of the project TMOP 421.B JLK 29.
2010-2011 Budapest
Ferenc Papp
1.1 The aim of the design task

The objective of the design task is the steel structure of a simple hall. The main properties of
the structure are the following:
- Symmetric form with duopitch roof;
- Equal interval between main frames;
- Equal interval between purlins;
- Horizontal wall girder system, with wall columns in the end walls;
- Covering system with double trapezoidal plates and heat isolation.
1.2 The initial data for the design
The work starts with the preliminary design of the structure. It is based on the initial data
which are determined and supplied by the architectural engineer which satisfy both the
appropriate building regulations and the requirements of the owner. In the case of the present
design project the initial data concerns to the outer surfaces of the flanges of the steel main
frames (see Figure 1):
Base area to be built: A [m2];
Horizontal distance between the flanges of the main frame: b [m];
Height of the side walls: Hw [m]
Slope of the roof: [deg]
b [m]
[deg]
Hw
[m]
A [m2]
0.0
Fig.1 Initial data for the design
1.3 The theoretical parameters of the main structure

The main theoretical parameters of the steel structure should be determined for the structural
analysis (see Figure 2 which considers both the prismatic and the tapered members). The
theoretical span of the main frame is equal to the horizontal distance between the central
(reference) axes of the columns:
L = b hc
where hc is the initial height of the column section, b is the outer distance of the columns
prescribed by the architectural engineer. The theoretical height of the columns is equal to the
distance between the theoretical column base point and the intersectional point of the column
and the beam central axis. This parameter may be calculated approximately by the following
expression:
Ferenc Papp
(a)
(b)
hc
hb
hb
covering system
Ht
Hw Hc
Hf
Hf
hc
b/2
b/2
L/2
L/2
Fig.2 Theoretical parameters of the main frame

Hc = Hw
hb
2
where Hw is the initial height of the side walls, hb is the initial depth of the beam section. The
theoretical ridge (top) point of the frame may be calculated by the following expression:
H f = Hc +
L
tg ( )
2
1.4 The number of the main frames and their interval

The architectural concept has prescribed the basic area of the building (A), from which the
theoretical length of the steel structure may be calculated,
d0 =
A
b
where the parameters are defined in the Section 1.2. The required number of the main frames
may be determined as following:
nn =
d0
+1
c
In the expression c denotes the interval between the main frames, where the optimal value is
c=57 meters. Different distance may be used in special circumstances only. The applied
number na of the main frames should be an integer, which is determined on the base of the
required number of frames nn. The real theoretical length between the final frames is the
following (see Figure 3):
d = (na 1) c
Ferenc Papp
d
Fig.3 The applied number of main frames and the real theoretical
length of the structure
Since the distance between the main frames is normally uniform, therefore the initial basic
area (A) of the building may be kept only approximately. The real basic area can be calculated
by the main parameters of the structures which were determined previously:
Aa = (L + hc ) (d + bbf + 2 hcsw )
where bbf [m] is the flange width of the beam section, hcsw is the depth of the column section
in the end wall system (see Figure 4). It should be noted that the previous expression is valid
for the structural solution illustrated in the Figure 4.
purlin
beam of the frame
wall beam
wall column
hcsw
bbf
d
Fig.4 Structural system of the end wall
1.5 The initial cross-sectional parameters for the main frames

As a matter of fact, the main frames at the end walls carry lower loads than the intermediate
ones. This is why these two frames might be fabricated from weaker cross-sections. In order
Ferenc Papp
for easy enlarging the building longitudinally in the future, these frames are suggested being
as strong as the intermediate ones. The parameters of the cross-sections of the main frames
are related to the span L and the height Hf. Assuming that the building to be designed is
relatively low and the roof is relatively flat, i.e:
Hc
0 .5
L
15o ,
and the structure is loaded only by dead load and meteorological loads, the initial types and
parameters of the cross-sections may be determined as the function of the theoretical span L.
Table 1 contains the suggested initial values for the cross-sectional parameters which are
based on practical experiences. The used symbols of the geometrical parameters can be found
in the Table 2.
Tab.1 The initial sizes of the cross-sections of the main frames in the function of the span L
span L [m]
type of the section
size* [mm]
1216
1624
hot rolled (IPE/HEA)

welded I
2432
tapered I
300450/200260
flange: 200300 1620
web: 400600 810
flange: 300340 1620
web: 8001200 68
* in the case of hot rolled sections the values mean depth of the section for the lower and the upper limits of the
span L; in the case of welded sections the values mean the width and thickness of the plates for the lower and
the upper limits of the span L
Tab. 2 The signs of the geometrical properties of the cross-sections

structural member
column
beam
property
bcf
tcf
hcw
tcw
bbf
tbf
hbw
tbw
meaning
width of the flange

thickness of the flange
width of the web
thickness of the web
width of the flange
thickness of the flange
width of the web
thickness of the web
1.6 The initial grade of material

The main structural elements are normally made from S235 or S355 steel. Unless there is any
previous reason to use S355 steel grade, the grade of S235 is suggested using. If it is
reasonable, the initial grade of steel may be changed during the analysis and design of the
structure.
Ferenc Papp
1.7 Application
1. PRELIMINARY DESIGN
1.1 Initial parameters
- area to be built [m 2 ]
725
- width of the building [m]
20.0
- heigth of the side walls [m]
Hw
- slope of the roof [deg]
7.5
10
1.2 Initial data for the main structural members

- main frames (welded I section)
column [mm]
flange
b cf
web
240
h cw
depth
468
hc
h cw
beam [mm]
flange
b bf
240
web
h bw
depth
hb
2. t
t cf
16
t cw
8
h c = 500
cf
368
t bf
16
t bw
2. t bf
h bw
h b = 400
- columns in side walls
HEA160
h csw
- purlin
Lindab Z 200
hp
- beams in walls
Lindab C 200
h bsw
150
200
200
1.3 Theoretical properties of the structural model

- span of the frames [m]
hc
L = 19.5
1000
- height of the columns [m]
- heigth of the frame [m]
Hc
Hw
Hf
Hc
hb
2
L.
2
. 1
1000
tan .
H c = 7.3
H f = 9.019
180
1.4 Number of the main frames

- prescribed length of the building [m]
d0
- interval of the frames [m]
- required number of the frames
nn
- applied number of the frames
na
d 0 = 36.25
b
6.0
d0
n n = 7.042
c
7
The building consists of 7 frames!

1.5 Area of the bulding
- length of the building [m]
- actual area of the building [m 2 ]
A tny
na
1 .c
L
d = 36
hc
1000
. d
b bf
1000
2.
h csw
1000
A tny = 730.8
The actual area of the building satisfies the
official plan!
Ferenc Papp
1.8 Arrangement of the purlin system

In our case (the building is under static loading) the purlin system and the wind bracing
system may be independent ones, but they also may form a unified system. In the present
phase of the design it should be determined which concept is followed:
Concept A: Purlin system is independent to the wind bracing system
Concept B: Purlin and wind bracing members form a unified static system
In case of Concept A it is assumed that the purlin system carry the loads and effects which
act directly to the roof, and it does not take part in the bracing of the building. In this case the
wind bracing system is a trussed structure, which consist of two neighboring main frames, the
diagonals and the longitudinal bars are placed under the purlins independently to them. In the
case of Concept B the longitudinal bracing bars are replaced by the purlin system (besides the
primary rule the purlins are part of the wind bracing system). Which concept to be followed in
the design may be supported by the following rules and suggestions:
Application of the Concept A may be suggested in the case of L 18 m , since the
solution is not economical for relatively small spans with relatively low design loads
and effects.
The application of the Concept B may not be suggested for structures loaded by other
considerable loads (e.g. crane load) besides the dead load and the meteorological
loads.
In the framework of this design project the Lindab Z purlin is suggested for the roof system. It
is a practical experience that the optimal distance between two neighboring purlins is
e=1,53,0 meters. The depth of the purlin may change form 200 mm to 300 mm, while the
thickness from 1,5 mm to 2,5 mm. The distance is determined also by the rule that the optimal
value of the angle of the bracing diagonals to the axis of the frame beam is about 45 degrees,
but it is not greater than 60 degrees and not lower than 30 degrees. The suggested numbers for
intermediate units are 4, 6 or 8, since the application of a half-bracing unit can be avoided by
this way (see Figure 5).
Ls1224m
Ls
Ls18-36m
Ls
Ls24-48m
Ls
Ls - distance between the ridge point of the roof and the outer point of the edge purlin in
the plane of the roof system (see Figure 6)
Fig.5 Optimal arrangement of purlin system in term of the distance Ls
Ferenc Papp
The practical purlin arrangement shown in Figure 6 may differ from the theoretical
arrangement shown in Figure 5:
(i)
at the ridge double purlins are used (Figure 6a);
(ii)
at the edge of the roof special edge shape is used (Figure 6b).
The distances denoted in Figure 6 may be calculated by the following expressions:
Ls =
L
h sin
+ b
+ f where
2 cos
2
hc
+ hbsw
2
f =
cos
where hbsw is the depth of the wall beams and g=150200 mm.
Ls
e
g
e
e
f
(a)
(b)
Fig. 6 The scheme of the practical purlin arrangement:

(a) double purlins at the ridge; (b) special shaped edge purlin
1.9 Wind bracing system
The wind bracing system is shown in Figure 7. According to the Concept A (see Section 1.8)
the purlin system is independent to the wind bracing system, therefore the main frames should
be connected to each other by bracing bars (see the dashed lines).
Fig. 7 The wind bracing system which is independent to the purlin system
(dashed lines denote the bracing bars)
9
Ferenc Papp
These bars may be connected to the beams under every second purlin, close to the upper
flange of them. It is noted that according to the Concept B the longitudinal bracing bars could
be replaced by the purlins (see the continuous lines).
1.10 Application
1.6 Arrangement of the purlin system
- distance between the edge purlin and the theoretical point of the frame corner [mm]
hc
f
h bsw
f = 456.942
cos .
180
- distance between the edge beam and the ridge of the frame [mm]
Ls
1000. L
2 . cos .
h b . sin .
180
4
L s = 1.039. 10
180
- interval of purlins [mm]

four spans
six spans
applied spans
e4
e6
e
Ls
3
e 4 = 2.598. 10
4
Ls
3
e 6 = 1.732. 10
6
e4
3
e = 2.598. 10
The e=2598 mm distance is choosen for the arrangement

of the purlin system (except the last distance at the ridge) !
1.11 Covering system

Two different constructions of the covering system are shown in Figure 8. In any case the
external loads and effects are carried by the external trapezoidal sheet.
(a)
(b)
external trapezoid sheet
vapour permeable leaf
heat isolation (150 mm)
vapour proof leaf
internal trapezoid sheet
external trapezoid sheet

vapour permeable leaf
heat isolation (150 mm)
vapour proof leaf
internal trapezoid sheet
spacer members
Fig. 8 Covering system with heat isolation and double trapezoid sheets:
(a) isolation is placed between the purlins
(b) isolation is placed on the purlins
10
Ferenc Papp
1.12 Preliminary drawings

The aim of the preliminary drawing is to establish the initial parameters of the design in
drawings. The preliminary drawings are the basic documents for the structural analysis and
design. Therefore, these drawings should contain all the initial parameters of the building
used in the procedure of the analysis and design. These drawings should not be confused with
the architectural plans and the scenario of the building. In this design project the following
three drawings should be prepared (the format of the drawings is A4 or A3):
top view of the foundation and the roof structure
side views of the building
side views of the main frame.
1.12.1 Foundation and roof view (M 1:200)
The building is symmetric, therefore the one half of the drawing may show the top view of the
foundation, while the other half of it may show the top view of the roof. If the wind bracing
system follows the Concept A (the bracing system is independent to the purlin system), the
top view side of the drawing may be divided into two symmetrical parts: the upper quarter of
the drawing shows the arrangement of the purlin system, while the lower quarter of the
drawing shows the bars of the bracing system. The view of the foundation and the roof
system is projected to the horizontal plane. The drawing gives exact answer to the following
parameters:
top view of the foundation (right side of the drawing):

- theoretical span (L)
- number of the frames (n)
- distance between the frames (c)
- arrangement and initial parameters of the columns in the side walls
- scheme of the foundation
top view of the roof structure (left side of the drawing):
- arrangement and initial parameters of the purlins
- arrangement and parameters of the wind bracing system.
The drawing of the top view of the foundation and the roof structure which satisfies the
Sections 1.7 and 1.10 (Applications) is shown in the Figure 9. It can be seen that the bracing
system follows the Concept A. Furthermore, it can be seen that the column foundations are
tied up by beams, and this system works together with the concrete slab of the industrial floor.
1.12.2 Side views of the building (M 1:200)
The aim of the side view drawings of the building is to give direct information about the
arrangement of the wall beams and about the area and place of the openings as well. The
building is symmetrical, therefore the right hand side of the drawing may show the
arrangement of the openings, while the left hand side may show the arrangement of the wall
beams and the bracing system. The drawing should give exact answer for the following
parameters:
- places and initial section of the wall beams
- arrangement and initial sections of the bracing system
- place and area of the openings.
11
Ferenc Papp
The drawing of the side view does not contain architectural sceneries, it concentrates to the
above parameters. The drawing which satisfies the Sections 1.7 and 1.10 (Applications) is
shown in the Figure 10.
1.12.3 Side view of the frame (cross section of the building) (M 1:100)
The aim of the side view drawing of the frame is to give direct information to take the
structural and load model for analysis and design. The frame is symmetrical, therefore the
right hand side of the drawing may contain the general parameters, while the left hand side of
it may show the theoretical parameters and the arrangement of the structural members:
general parameters (right hand side)
- distance between the outer flanges of the columns (b)
- height of the facade (Hw);
- slope of the roof (
);
- height of the structure (Ht);
- parameters of the column section (bcf;tcf.hcw;tcw);
- parameters of the beam section (bbf;tbf.hbw;tbw);
- parameters of the haunching (bhf;thf.hhw;thw);
- type of the joints;
- type of the column base;
- layers of the covering system;
arrangement of members and theoretical parameters (left hand side)
- theoretical height of the columns (Hc);
- theoretical height of the frame (Hf);
- arrangement and initial section of the purlins;
- arrangement and initial section of the wall beams;
- length of the haunch.
The drawing which satisfies the Sections 1.7 and 1.10 (Applications) is shown in the Figure
11. It can be seen that the column foundation, the beams between the concrete blocks and the
concrete slab of the industrial floor form a unified structural system.
12
Top view of the roof

(projected to horizontal plane)
2559
Top view of the foundation
Lindab purlin (Z section)
4632
2559
10236
2559
Purlin system
Wall columns
(HEA or IPE)
2559
10126 19500
5118
Bracing cross bars (L or rod section)

Bracing bars (CHS)
9750
Bracing system
4632
4632
6000
Fig. 9 Top view drawing of the roof and the foundation
6000
6000
Department of Structural Engineering BUTE

Steel Buildings
Draw No. 1: Preliminary drawing/Top view
M 1:200
Designer
Clever Student (XYZVW)
Supervisor
Clever Teacher assistant professor
Arrangement of wall beams and bracing system
4,600
Lindab wall beam (C200)
3000
7700
7,900
Arrangement of openings
1200
3,600
Bracing bars (CHS)

Bracing diagonals (L or rod section)
2900
0,0
600
6000
6000
6000
length of the window: 11600

18 400
18 000
9,650
7,900
7700
3000
4,600
1200
3,600
2900
0,0
600
4632
Fig.10 Side view drawing of the building
5118
5000 (door)
3600 (window)

Steel Buildings
Draw No. 002: Preliminary drawing/Side view M 1:200
Designer
Supervisor
10390
2447
2598
150
Welded I section:
- flanges: 240-16
- web: 368-6
2598
2597
Covering system:
- external trapezoidal sheet
- vapour permeable leaf
- heat insulation (150 mm)
- vapour proof leaf
- internal trapezoidal sheet
Slope of roof: 100
3500
3000
Moment resistant
end-plated bolted
connections
Purlins (Lindab Z200)

1200
7300
Wall beams (Lindab C200)
9019 9219
Welded I section:
- flanges: 240-16
- web: 468-8
CHS bracing members
2900
19 500/2
600
Fig.11 Side view drawing of the structural frame
Haunch:
- flanges: 240-20
- web: 330-6
7500
20 000/2
Rigid column base

Steel Buildings
Draw No. 003: Preliminary drawing /Side view of the frame M 1:200
Designer
Supervisor
Ferenc Papp
Practice 2
LOADS AND EFFECTS ON THE
BUILDING
2010-2011 Budapest
16
Ferenc Papp
2.1 General
The loads and effects in general are the subject of the course of Basis of the design
(BMEEOHSAT16) in the framework of the BSc education. Here the application of the
general knowledge to the design of simple halls is presented. The loads and effects should be
determined using the following design standards:
EN 1991-1-1:2005 Eurocode 1: Actions on structures Part 1-1: General actions.
Densities, self-weight, imposed loads for buildings (EC1-1-1);
EN 1991-1-2:2005 Eurocode 1: Actions on structures. Part 1-2: General actions.
Actions on structures exposed to fire (EC1-1-2);
EN 1991-1-3:2005 Eurocode 1: Actions on structures. Part 1-3: General actions. Snow
loads (EC1-1-3);
EN 1991-1-4:2005 Eurocode 1: Actions on structures - General actions - Part 1-4:
Wind actions (EC1-1-4);
EN 1998-1:2005 Eurocode 8: Design of structures for earthquake resistance. Part 1:
General rules, seismic actions and rules for buildings (EC8-1).
In the present phase of the design procedure we are dealing with the basic loads and effects
which act on the building. The applied load cases and load combinations are discussed in the
sections which are denoted to the design of the structural members. In general the following
loads and effects should be taken into consideration in the case of a symmetric and duopitch
building:
dead loads;
o weight of the structural members;
o weight of the covering system;
o other dead load type loads;
meteorological loads and effects;
o snow load;
o wind effect;
imposed loads;
seismic effect;
fire effect.
2.2 Dead loads
2.2.1 Weight of the structural members
The self weight of the structural members should be taken on the base of the initial structural
parameters. The evaluation should follow the specifications of EC1-1-1. The density of the
steel material is 78,5 kN/m3. The dead loads which are based on the initial design parameters
should not be changed unless these initial design parameters have changed considerably. The
change is considerable if the effect of the change of any parameter on the design forces
exceeds by 3%. If the effect of the change is at the safe side, the modification of the initial
loads may be neglected. The theoretical self weight of the structural members of the frame is
automatically taken into consideration by the analysis software (Axis, ConSteel, FEMDesign), but the self weight of the purlins and trapezoidal sheets or panels should be given by
the designer (DimRoof). The self weights of the additional elements (stiffeners, bolts, ect.) are
usually taken into consideration by 510% of the theoretical self weight.
2.2.2 Weight of the covering system

17
Ferenc Papp
The weight of the covering system of the roof and the walls should be evaluated according to
the layers specified in the preliminary drawing (see Figure 8 in Practice 1). The densities of
the materials may be found in the appropriate tables of EC1-1-1. The weights structural
sections (purlin, wall beam, etc.) may be found in the product information of the producers.
2.2.3 Other dead load type loads
This type of loads refers to the loads which are acting regularly. Such loads are the weights of
the electrical and mechanical equipments, for example the weights of lighting, climate
technology. Such dead load is the weight of the earth layer of the special greenroof. These
type of loads should be specified by the mechanical engineer and the architectural engineer,
respectively. The applied intensity and the distribution of this type of loads should satisfy the
specifications of EC1-1-1. In present design project in lack of precise information we can
apply approximately 0,25kN/m20,45kN/m2 dead load which is totally distributed on the roof.
2.2.4 Application
2. LOADS AND EFFECTS

2.1 Dead loads
2.1.1 Weights of the structural members and the layers of the covering system
- external trapizoidal sheet : LTP 85 t=0.75mm [kN/m 2 ]
- internal trapizoidal sheet: LPT 20 t=0.4mm
[kN/m 2]
q tr.ext
0.0804
q tr.int
0.0390
- heat insulation (mineral rockwool) [kN/m 2 ]

density [kN/m 3] and thickness [m]
thickness [m]
iso 1.5
t iso 0.150
t iso . iso
q iso.other 0.100
q iso
- further layers for insulation [kN/m 2 ]

- purlin: LINDAB Z 200 (t=2,0) [kN/m]
q purlin
0.0579
- main frame: automatically considered

2.1.2 Installation loads
Installation load projected to the total area of the roof
- lightning [kN/m 2 ]
q light
- building equipments [kN/m 2]
q equip
0.15
q other
0.20
- other loads
[kN/m 2 ]
18
0.10
q iso = 0.225
Ferenc Papp
2.3 Meteorological loads and effects

2.3.1 Snow load
2.3.1.1 Surface snow load
The snow loads on the building are determined by the specifications of EC1-1-4. In Hungary
the additional specifications of the Hungarian National Annex (HNA) should be considered.
The surface snow load may be calculated as follows:
- persistent and transient design situations: s = i Ce Ct sk
- exceptional design situation:
s = i Ce Ct s Ad
where
s
snow load on the horizontal ground [kN/m2];
i
shape coefficient;
Ce
exposure coefficient;
Ct
thermal coefficient;
characteristic value of the ground snow load [kN/m2]-ben;
sk
sAd
exceptional value of the ground snow load [kN/m2]-ben.
The characteristic value of the ground snow load according to the specification HNA 1.5 is
the following:
A
sk = 0 ,25 1 +
100
but sk 1,25
where A is the height of the ground above the sea level in [m]. The exceptional value of the
ground snow load according to the specifications HNA 1.2 and 1.7 is the following:
s Ad = Cesl sk
where Cesl is the exceptional snow load factor which is 2,0. The exposure factor Ce depends on
the topography:
-
windswept:
normal:
sheltered:
Ce = 0,8
Ce = 1,0
Ce = 1,2
Windswept topography: flat unobstructed areas exposed on all sides without, or little shelter
afforded by terrain, higher construction works or trees.
Normal topography: areas where there is no significant removal of snow by wind on construction
work, because of terrain, other construction works or trees.
Sheltered topography: areas in which the construction work being considered is considerably
lower than the surrounding terrain or surrounded by high trees and/or surrounded by higher
construction works.
In the present design project it is assumed that the snow is not prevented from sliding off the
roof, and the shape factor i may be taken from the Table 3.
19
Ferenc Papp
Tab.3 Shape factor for duopitch roof (free slip of the snow)
tet hajlsszge (
)
0 30
30 < < 60
60
0,8
0,8(60-)/30
0,0
The thermal coefficient Ct should be used to account for the reduction of snow loads on roofs
with high thermal transmittance (> 1 W/m2K), in particular for some glass covered roofs,
because of melting caused by heat loss. In the present design Ct=1,0 may be applied.
In regions with possible rainfalls on the snow and consecutive melting and freezing, snow
loads on roofs should be increased, especially in cases where snow and ice can block the
drainage system of the roof. In the present design this effect may be neglected.
2.3.1.2 Application
2.2 Snow load
2.2.1 Snow load for the persistent design situation
- height of the building ground [m]
A see
- charactheristic ground snow load [kN/m 2 ]
sk
sk
1.25
- exposure coefficient (normal)
Ce
1.0
- thermal coefficient
Ct
1.0
- shape coefficient (<30 deg)
1 0.8
s 1. C e .C t . s k
- ground snow load [kN/m 2 ]
300
0.25. 1
A see
100
sk= 1
s=1
2.2.2 Snow load for the exceptional design situation

- exceptional snow load coefficient
C esl
- exceptional snow load [kN/m 2]
s Ad C esl .s k
s r 1 . C e . C t . s Ad
- exceptional ground snow load
[kN/m 2]
2.0
s Ad = 2.5
sr=2
2.3.2 Wind effect

2.3.2.1 Wind pressure on surfaces
The effect is specified by the EC1-1-4. The wind load is the compressive or the sucking load
which is caused by the wind effect. The wind load is perpendicular to the surface. The load
may affect on the external and the internal surfaces as well. Besides the normal wind load the
friction load of the wind effect may be considered. Any wind effect may be considered by a
simplified set of loads which is equivalent to the effect of the turbulent peak velocity. The
wind load belongs to the group of imposed loads. The wind effect depends on the following
parameters of the building:
dimensions;
shape;
dynamic properties.
20
Ferenc Papp
The external and internal wind pressure may be calculated by the following expressions:
we = q p ( ze ) c pe
wi = q p ( zi ) c pi
where
qp( z )
is the peak velocity pressure;
z e , zi
c pe ,c pi
is the external and internal reference heights;

is the external and internal pressure coefficients.
Figure 12 shows the physical direction of the wind loads in the cases of wind sucking (-) and
wind pressure (+). It is noted that the summation of the wind loads should be done by these
physical directions.
(-)
sz
(+)
Fig.12 Physical direction of the wind loads in the cases of wind

sucking (-) and wind pressure (+)
The reference heights may be determined using the following rules (see Figure 13):
if the height of the building (h) is not greater than the width (b) of the windward surface of
the building:
ze = h and zi = ze ;
if the height of the building (h) is greater than b but it is not greater than 2b:
- zone for height of b:
- zone for height of (h-b):
ze = b and zi = ze ;
ze = h and zi = ze .
hb
b < h 2b
ze=h
maximum
height
h
h
ze=h
b
ze=b
Fig.13 Reference heights for plane buildings
21
maximum
height
Ferenc Papp
2.3.2.2 Peak velocity pressure

The peak velocity pressure may be calculated by the following expression:
q p ( z ) = ce ( z ) qb
where
ce ( z )
qb
is the exposure factor;

is the basic velocity pressure.
The basic velocity pressure may be calculated as follows:

qb =
1
vb2 ( z )
2
where the density of the air:
= 1,25
kg
m3
and where the basic wind velocity:

vb = cdir cseason vb ,0
According to the Hungarian National Annex (HNA) the initial basic wind velocity and the
direction and season coefficients may be taken as
m
vb ,0 = 23 ,6
; cdir=0,85 ; cseason=1,0
s
The exposure factor is the ratio of the peak velocity pressure to the basic velocity pressure,
and it may be calculated by the following expression:
ce ( z ) = ( 1 + 7 I v ( z )) cr2 ( z ) c02 ( z )
where
cr ( z )
c0 ( z )
is the roughness factor;

is the orography factor;
Iv( z )
is the turbulence intensity.
The roughness factor depends on the reference height:

z
- if z < zmin than cr ( z ) = k r ln min
z0
z
if z zmin than cr ( z ) = kr ln
z0
where the terrain factor:
22
Ferenc Papp
0 ,07
z
kr = 0 ,19 0
z0 ,II
where z0 ,II = 0 ,05[m] , see the second category (II) in the Table 4. In the expression z0 is the
roughness length and zmin is the minimum height. These constants are given in the Table 4.
Tab.4 Roughness lengths and minimum heights
terrain category
0
I
II
III
IV
Sea or coastal area exposed to the open sea 1
Lakes or flat and horizontal area with negligible vegetation and

without obstacles
Area with low vegetation such as grass and isolated obstacles
(trees, buildings) with separations of at least 20 obstacle heights
Area with regular cover of vegetation or buildings or with isolated
obstacles with separations of maximum 20 obstacle heights (such
as villages, suburban terrain, permanent forest)
Area in which at least 15 % of the surface is covered with
buildings and their average height exceeds 15 m
z0 [m]
zmin [m]
0,003
0,01
1
1
0,05
0,3
1,0
10
When the average slope of the upwind terrain is less than 3, the orography factor may be
c0 ( z ) = 1,0 .
The turbulence intensity may be calculated by the following expressions:
kI
- if z < zmin than I v ( z ) =
z
c0 ( z ) ln min
z0
kI
- if z zmin than I v ( z ) =
z
c0 ( z ) ln
z0
where the turbulence factor may be kI=1,0.
The exposure factor can be calculated using the Figure 4.2 of EC1-1-4 (see the graphics
below).
23
Ferenc Papp
2.3.2.3 Application
2.3 Wind loads

2.3.1 Basic velocity pressure
- initial parameters specified by the Hungarian NA
initial basic velocity [m/s]
v b.0 23.6
direction factor
c dir
season factor
c season
air density [kg/m 3]
0.85
1.0
1.25
- basic velocity [m/s]
vb
- basic velocity pressure [kN/m 2 ]
qb
c dir . c season .v b.0

1 . . 2. 1
vb
2
1000
v b = 20.06
q b = 0.252
2.3.2 Peak velocity pressure

- parameters for terrain category (Category III)
z0
- parameter for category II [m]

- terrain factor
kr
z min
5.0
z 0.II
0.05
0.07
z0
0.19.
- roughness coefficient
z> z min
kI
- turbulence intensity
Iv
- exposure factor
ce
- peak velocity pressure [kN/m 2 ]
qp
z = 9.019
Hf
k r . ln
c r = 0.733
z0
- orography coefficient (plane country, slope less than 3 degs)

- turbulence coefficient (no specific rule)
k r = 0.215
z 0.II
- reference height
cr
0.3
c0
1.0
1.0
kI
c 0. ln
I v = 0.294
z
z0
2
2
7 .I v .c r . c 0
c e .q b
c e = 1.643
q p = 0.413
The peak velocity pressure can be determined or checked using the Figure 4.2 of the EN
1991-1-4:
z = 9.019
reference height
terrain category: III
exposure factor by the graphics
c e.graphics
1.63
peak velocity pressure [kN/m 2 ]
q p.graphics
c e.graphics . q b
q p.graphics = 0.41
24
Ferenc Papp
2.3.2.4 External pressure coefficient

The external pressure coefficients depend on the reference height and the size of the loaded
area A, which is the area of the structure that produces the wind action in the section to be
calculated. The external pressure coefficients are given for two loaded areas:
- c pe ,1 is for area of 1.0 m2 as local coefficient;
- c pe ,10 is for area of 10.0 m2 as overall coefficient.
Between the two limit areas (for 1m2<A<10m2) the following interpolation may be used (see
Figure 14):
c pe , A = c pe ,1 ( c pe ,1 c pe ,10 ) lg10 A
cpe,1
cpe,10
0,1
1,0
10,0
A(m2)
Fig.14 Interpolation of the external pressure coefficient
In the present design project the interpolation may be neglected. For the design of the
trapezoidal sheet the cpe.1 may be used, while for the design of the purlins and the main frames
the cpe.10 may be used. The external pressure coefficients are given in tables. The tables for
symmetric buildings with duopitch roofs are contained in the following Annexes:
- Annex 1: Wind effect on vertical walls of the building
- Annex 2: Cross wind effect on the roof (=0)
- Annex 3: Longitudinal wind effect on the roof (=90)
Notes for application of the tables
The tables of the external pressure coefficients have more rows (one row belongs to one slope)
where there are two sub-rows (for example one + and one - values). It is an important rule
that for one roof plane (actually for the half roof) the sub-rows should not be changed. For
example in the Annex 2 for roof slope of 5o there are two sub-rows which define four
combinations:
=5o
1
2
3
4
Zones of the roof

H
cpe,10
cpe,1
cpe,10
cpe,1
cpe,10
cpe,1
cpe,10
cpe,1
cpe,10
cpe,1
-1,7
-1,7
0
0
-2,5
-2,5
0
0
-1,2
-1,2
0
0
-2,0
-2,0
0
0
-0,6
-0,6
0
0
-1,2
-1,2
0
0
-0,6
-0,6
-0,6
-0,6
-0,6
-0,6
-0,6
-0,6
+0,2
-0,6
+0,2
-0,6
+0,2
-0,6
+0,2
-0,6
The automatic use of the tables may lead to a large number of wind load cases. At the design of
simple buildings the designer may select the most dangerous case by a decision based on his
experience and intuition.
25
Ferenc Papp
2.3.2.5 Application
2.3.3 External wind pressure
2.3.3.1 Cross wind (0 degree)
Initial parameters
- size of the building [m]
width perpendicular to the wind direction
b0
b 0 = 36
width parallel to the wind direction
d0
d 0 = 20
height
h0
h 0 = 9.019
- size factor
Hf
h0
- size of the zones [m]
e0
e 0.4
e 0 = 18.038
0
e0
e 0.10
-slope of the roof (approximately) [deg]
0 = 0.451
d0
2. h
e 0.4 = 4.51
4
e0
e 0.10 = 1.804
10
10
Indeces used below

A,B,...
mark of the wall and the roof zone
0; 90
mark of the wind direction in degree
1;10
mark of the loaded area (1m2 or 10 m2)
Wind pressure on the walls

According to Annex 1:
0
- interpolation factor for the case of 0,25<h/d<1,0

- pressure coefficients
c pe.A.0.10
1.2
c pe.B.0.10
0.8
c pe.C.0.10
0.5
c pe.D.0.10
c pe.E.0.10
- wind pressures [kN/m2 ]
0.7 0.1. 0
0.3 0.2. 0
0.25
0.75
0 = 0.268
c pe.D.0.10 = 0.727
c pe.E.0.10 = 0.354
w A.0.10
c pe.A.0.10. q p
w A.0.10 = 0.496
w B.0.10
c pe.B.0.10. q p
w B.0.10 = 0.33
w C.0.10 c pe.C.0.10. q p
w D.0.10 c pe.D.0.10. q p
w E.0.10 c pe.E.0.10. q p
w C.0.10 = 0.207
w D.0.10 = 0.3
w E.0.10 = 0.146
Wind pressure on the roof

Annex 2 contains the pressure coefficients for roof slope of 10 deg which were interpolated
linearly between 5 and 15 degrees given by the EN 1991-1-4.
For roof zones of F-G-H there are two cases: wind sucking and wind pressure.
zones of F-G-H
- wind sucking [kN/m 2]
c pe.F.0.1
2.25
c pe.F.0.10
1.30
w F.0.1
w F.0.10
26
c pe.F.0.1. q p
c pe.F.0.10. q p
w F.0.1 = 0.929
w F.0.10 = 0.537
Ferenc Papp
c pe.G.0.1
1.75
c pe.G.0.10
1.0
c pe.H.0.1
0.75
c pe.H.0.10
w G.0.1
c pe.I.0.10
w G.0.1 = 0.723
w G.0.10 c pe.G.0.10. q p
w H.0.1
0.45
- wind pressure [kN/m 2]

c pe.FGH.0 0.1
c pe.I.0.1
c pe.G.0.1. q p
0.50
w H.0.1 = 0.31
w H.0.10
c pe.H.0.10.q p
w H.0.10 = 0.186
w FGH.0
c pe.FGH.0.q p
w FGH.0 = 0.041
c pe.I.0.1. q p
w I.0.1
0.50
c pe.H.0.1.q p
w G.0.10 = 0.413
w I.0.10
w I.0.1 = 0.207
c pe.I.0.10. q p
w I.0.10 = 0.207
c pe.J.0.1.q p
c pe.J.0.1
0.65
w J.0.1
c pe.J.0.10
0.4
w J.0.10
w J.0.1 = 0.269
c pe.J.0.10.q p
w J.0.10 = 0.165
2.3.3.2 Longitudinal wind direction (90 degrees)

Initial parameters
- size of the building [m]
width perpendicular to the wind direction
b 90
b 90 = 20
width parallel to the wind direction
d 90
d 90 = 36
height
h 90
Hf
h 90 = 9.019
h 90
90
- size factor
- size of the zones [m]
e 90
d 90
2 .h 90
e 90.2
e 90.4
e 90.5
e 90.10
- slope of the roof (approximately) [deg]
e 90
2
e 90
4
e 90
5
e 90
10
90 = 0.251
e 90 = 18.038
e 90.2 = 9.019
e 90.4 = 4.51
e 90.5 = 3.608
e 90.10 = 1.804
10
Wind pressure on the walls

According to Annex 1
90
- size factor
h 90
c pe.A.90.10
1.2
c pe.B.90.10
0.8
c pe.C.90.10
0.5
c pe.D.90.10
c pe.E.90.10
27
90 = 0.251
d 90
0.7
0.3
Ferenc Papp
- wind pressures [kN/m2]
w A.90.10
c pe.A.90.10. q p
w A.90.10 = 0.496
w B.90.10 c pe.B.90.10. q p
w C.90.10 c pe.C.90.10. q p
w D.90.10 c pe.D.90.10. q p
w B.90.10 = 0.33
w E.90.10 c pe.E.90.10. q p
w E.90.10 = 0.124
w C.90.10 = 0.207
w D.90.10 = 0.289
Wind pressure on the roof

According to Annex 3
c pe.F.90.1
c pe.F.90.10
c pe.G.90.1
c pe.G.90.10
c pe.H.90.1
c pe.H.90.10
c pe.I.90.1
c pe.I.90.10
2.1
1.45
2.0
1.30
1.2
0.65
0.55
0.55
w F.90.1
w F.90.10
c pe.F.90.1. q p
c pe.F.90.10. q p
w G.90.1 c pe.G.90.1. q p
w G.90.10 c pe.G.90.10. q p
w H.90.1
w H.90.10
w I.90.1
w I.90.10
c pe.H.90.1. q p
c pe.H.90.10. q p
c pe.I.90.1. q p
c pe.I.90.10. q p
w F.90.1 = 0.868
w F.90.10 = 0.599
w G.90.1 = 0.826
w G.90.10= 0.537
w H.90.1 = 0.496
w H.90.10 = 0.269
w I.90.1 = 0.227
w I.90.10 = 0.227
2.3.2.6 Internal pressure coefficient

Internal and external pressures shall be considered to act at the same time (but external
pressure may act without internal pressure). The internal pressure coefficient (cpi) depends on
the size and distribution of the openings (windows and doors). When in at least two sides of
the buildings (walls or roof) the total area of openings in each side is more than 30 % of the
area of that side, the actions on the structure should not be calculated from the rules given
here.
For a building with a dominant face the internal pressure should be taken as a fraction of the
external pressure at the openings of the dominant face. A face of a building should be
regarded as dominant when the area of openings at that face is at least twice the area of
openings and leakages in the remaining faces of the building considered. When the area of the
openings at the dominant face is twice the area of the openings in the remaining
faces,
c pi = 0 ,75 c pe
When the area of the openings at the dominant face is at least 3 times the area of the openings
in the remaining faces,
c pi = 0 ,90 c pe
where cpe is the value for the external pressure coefficient at the openings in the dominant
face. When these openings are located in zones with different values of external pressures an
area weighted average value of cpe should be used.
In the present design project we may assume that there is no dominant face and the
distribution of the openings is uniform. In this case the internal pressure coefficient may be
calculated as follows:
28
Ferenc Papp
if h / d 0 ,25
- if 0 ,33
than
c pi = 0 ,35
- if > 0 ,9
than
c pi = 0 ,3
- if 0 ,33 < 0 ,9
than
c pi = 0 ,726 1,14
if h / d 1,0
- if 0 ,33
than
c pi = 0 ,35
- if > 0 ,95
than
c pi = 0 ,5
- if 0 ,33 < 0 ,95
than
c pi = 0 ,802 1,37
The opening ratio in the expressions may be calculated with the following term:
A
A
=
where
areas.
neg
neg
is the area of openings where cpe is negative or zero and A is the area of all
2.3.2.7 Application
2.3.4 Internal wind pressure
2.3.4.1 Parameters of the openings
- area of openings in the side walls [m 2 ]
width of the area of windows [m]
L w.s
23.2
height of the area of windows [m]
h w.s
1.2
L w.s .h w.s
As
- area of openings in the end
A s = 27.84
walls [m 2 ]
windows
width of the area of windows [m]
height of the area of windows [m]
industrial door
width of the door [m]
height of the door [m]
L w.e
12.6
h w.e
1.2
A e.w
L w.e .h w.e
b w.d
5.0
h w.d
4.6
A e.d
b w.d . h w.d
Af
A e.w
A e.d
A e.w = 15.12
A e.d = 23
A f = 38.12
2.3.4.2 Cross wind effect (0 degree)

- initial parameters
2. A s A f
area of openings with negative or zero external pressure [m2 ]
area of all openings [m 2 ]
A sum
A neg.0
29
As
2.A f
A sum = 131.92
A neg.0 = 104.08
Ferenc Papp
opening ratio
A neg.0
0 = 0.789
A sum
for h/d=0.25
for h/d=1.00
c pi.0.0.25 0.726 1.14. 0

c pi.0.1 0.802 1.37. 0
c pi.0
- wind pressure [kN/m2 ]
c pi.0.0.25 0. c pi.0.1 c pi.0.0.25

w i.0
c pi.0.q p
c pi.0.0.25 = 0.173
c pi.0.1 = 0.279
c pi.0 = 0.202
w i.0 = 0.083
2.3.4.3 Longitudinal wind effect (90 degrees)

- initial parameters
area of openings with negative and zero external wind pressure coefficient [m 2]
A neg.90 A f 2 . A s
A neg.90 = 93.8
A
opening ratio
neg.90
90
90 = 0.711
A sum
- internal pressure coefficient
for h/d<0.25
- wind pressure [kN/m2 ]
c pi.90
w i.90
0.726 1.14. 90
c pi.90.q p
c pi.90 = 0.085
w i.90 = 0.035
2.4 Imposed loads

The imposed loads are specified by the EC1-1-1. The determination of the imposed loads
should be based on careful examination of the design situation and extended consultations
with the design partners (mechanical designer, electrical designer, etc.). The roof structures
are classified into categories. The standard orders a distributed and a concentrated fictive load
to every category. In the present design situation the walking on the roof is not allowed,
except maintenance and repairing work, therefore the roof belongs to the category H. Table 5
shows the design imposed loads for the category H.
Tab.5 Imposed loads for roof of category H
slope of roof
distributed load
concentrated load
Qk [kN ]
kN
qk 2
m
0,4
1,0
10 o
0
0
20 o
Notes: between the two limits linear interpolation may be used
The imposed load and the snow load shall not be considered to act at the same time. Since the
effect of the snow load is greater, the imposed load may be neglected in the present design. It
is noted that the concentrated imposed load (Qk) may be relevant at the design of the trapezoid
sheet and the purlins, but it is considered by the design software (DimRoof).
30
Ferenc Papp
2.5 Seismic effect

The seismic effect is specified by the EC8-1. Due to the earthquake the displacement and the
acceleration of the ground is changing in time. The seismic design of the buildings is based on
the consideration of the ground acceleration. The acceleration has vertical and horizontal
components, but in Hungary the vertical component may be neglected. The horizontal
component of the ground acceleration depends on the reference peak ground acceleration of
type A ground:
a g = I a gR
where agR is the reference peak ground acceleration of type A ground (see Figure 15), I is the
importance factor given in Table 6. The building in the present design project may belong to
importance category I or II.
Tab.6 Importance categories of buildings
importance category
I.
II.
III.
IV.
importance factor
I
Buildings of minor importance for public safety, e.g. agricultural

buildings, etc.
Ordinary buildings, not belonging in the other categories.
Buildings whose seismic resistance is of importance in view of
the consequences associated with a collapse, e.g. schools,
assembly halls, cultural institutions etc.
Buildings whose integrity during earthquakes is of vital
importance for civil protection, e.g. hospitals, fire stations,
power plants, etc.
0,8
1,0
1,2
1,4
Fig.15 The reference peak ground acceleration of type A ground in Hungary

2.6 Fire effect
The fire effect on the building is specified by the EC1-1-2. In the present design project the
standard (ISO) fire curve should be considered at the design of the main frame. The
required fire resistance is 15 minutes, R15, which means of fire resistance class IV and one
floor building. All the steel structural members (I sections) of the main frame are unprotected,
31
Ferenc Papp
and they are imposed to fire effect at four sides. The main frame is examined for fire effect as
an isolated structure, and the room which is specified by the frame is a unified fire
compartment (see the Figure 16).
Standard fire curve

Required resistance: 15 min
Unprotected
Fig.16 Unified fire compartment of the isolated main frame structure

2.7 Application
2.4 Imposed load
- service class of the roof: H
- slope of the roof: =10o
- imposed load
surface distributed load [kN/m 2 ]
concentrated load [kN]
qk
0.4
Qk
1.0
1.0
2.5 Seismic effect

- importance category of the building: II.
- importance factor
-seismic zone: region of Esztergom, Hungary
- horizontal component of the reference peak ground acceleration [m/s 2 ]
a gR
1.5
[m/s 2]
2.6 Fire effect

- applied temperature-time curve: standard (ISO)
- category of fire resistance: IV (simple building)
- required limit for fire resistance (R15)
t fi
15 [min]
- type of the passive fire protection: "unprotected I section exposed to fire at four sides"
- fire compartment: "internal room determined by the main frame structure"
32
Ferenc Papp
Annex 1
External pressure coefficient for vertical walls
(it is valid for case of h<b)
h/d
1
0,25
cpe,10
-1,2
-1,2
zones
C
cpe,1
-1,4
-1,4
cpe,10
-0,8
-0,8
cpe,1
-1,1
-1,1
cpe,10
cpe,1
-0,5
-0,5
cpe,10
0,8
0,7
cpe,1
1,0
1,0
cpe,10
cpe,1
-0,5
-0,3
e = min( b;2 h )
Top view
Side zones for e<d:
h
w
b
e/5
e
side
Side zones for e>d:
e/5
Note
In the case of rectangular building b is the width of the side which is affected by the wind,
and d is the width of the perpendicular side. The wind may affect to the longitudinal side
(=00) and to the front side (=900), respectively.
0
33
Ferenc Papp
Annex 2
External pressure coefficients of the roof due to cross wind (
=00)
0*
cpe,10
-1,8
zones
H
cpe,1
-2,5
cpe,10
-1,2
cpe,1
-2,0
cpe,10
-0,7
cpe,1
-1,2
cpe,10
+0,2
-0,2
-0,6
cpe,1
+0,2
-0,2
-0,6
-1,7
-2,5
-1,2
-2,0
-0,6
-1,2
+0,0
+0,0
+0,0
+0,0
+0,0
+0,0
-1,3
-2,25
-1,0
-1,75 -0,45 -0,75
-0,5
-0,5
10**
+0,1
+0,1
+0,1
+0,1
+0,1
+0,1
-0,9
-2,0
-0,8
-1,5
-0,3
-0,3
-0,4
-0,4
15
+0,2
+0,2
+0,2
+0,2
+0,2
+0,2
+0,0
+0,0
* given for the case of sharp eaves of flat roof (no parapet or curved eaves)
** given by linear interpolation between slopes of =50 and =150
5
=00
Ridge
e/4
w
G
e/4
e/10
e/10
34
cpe,10
+0,2
-0,2
+0,2
-0,6
-0,4
-0,3
-1,0
+0,0
cpe,1
+0,2
-0,2
+0,2
-0,6
-0,65
-0,3
-1,5
+0,0
Ferenc Papp
Annex 3
External pressure coefficients of the roof due to longitudinal
wind (
=900)
cpe,10
-1,8
0*
cpe,1
-2,5
cpe,10
-1,2
cpe,1
-2,0
cpe,10
-0,7
cpe,1
-1,2
cpe,10
+0,2
cpe,1
+0,2
-0,2
-0,2
5
10**
15
-1,6
-2,2
-1,3
-2,0
-0,7
-1,2
-0,6
-0,6
-1,45
-2,1
-1,3
-2,0
-0,65
-1,2
-0,55 -0,55
-1,3
-2,0
-1,3
-2,0
-0,6
-1,2
-0,5
-0,5
* given for the case of sharp eaves of flat roof (no parapet or curved eaves)
** given by linear interpolation between slopes of =50 and =150
=900
w
h
Ridge
e/4
w
F
G
G
e/4
F
e/10
e/2
35
Ferenc Papp
Practice 4
DESIGN OF THE SECONDARY
ELEMENTS
2010-2011 Budapest
36
Ferenc Papp
4.1 General
In this project the external trapezoidal sheet and the purlins as secondary elements are
designed. The design of the elements of the faade is based on the same methods, therefore it
is neglected. The design methods are specified by the EN 1993-1-3 Eurocode 3: Design of
Steel structures Part 1-3: Cold-formed thin gauge members and sheeting (EC3-1-3). The
theoretical background of this code is the objective of the MSc courses. The most important
expressions used in the design are summarized in the Table 7. In the practice the direct use of
the theory and methods given by the code may be neglected since the producers of the
products (purlins, sheeting) supply design tables and design software. Design of the Lindab
elements is supported by the DimRoof software which is suggested using in this project. The
software may be safely used without any deep knowledge of theoretical background on the
design of cold-formed thin gauge elements.
Tab.7: Most important expressions used in design theory of cold-formed thin gauge elements
(informative)
special structural properties
Large plate slenderness (b/t)
Partially stiffened plates
One symmetric or no symmetric

cross sections
Relatively thin plates
Special connections
special structural behaviors
Plate buckling
Shear lag effect
Flange induced buckling
Distorsional instability
Lateral torsional buckling

Flexural-torsional buckling
Relatively large initial
geometrical imperfections
Special structural failing modes
specialties in design
Class 4 cross-sections
Buckling of the stiffeners
Global instability
Plate thickness as the
parameter of the design
New design methods based
on tests
The loads for design of the purlins and sheeting should be calculated taking the effects of the
slope of roof, the directions of the loads and the constructions into consideration by the
following rules:
dead load and snow load are gravity loads with vertical direction;
wind pressure is perpendicular to the structural plane;
design load consists of transverse loads only (the loads which are parallel with the axis
of elements may be neglected).
These rules lead to the reduction of the loads (see Figure 17). It is noted that in case of
relatively small (5o10o) slope of roof the approximation of coscos21 may be on the safe
side.
37
Ferenc Papp
Basic load
[kN/m2]
Design load
[kN/m2]
ps
ps cos 2
Snow load
p g cos
pg
Dead laod
pw
pw
Slope of roof: [deg]

Wind load
Edge point
Ridge point
Fig. 17 Reduction of loads

4.2 Design of the external trapezoidal sheet
Figure 8 shows the alternative constructions for covering system. Both systems have external
trapezoidal sheet which should be designed.
4.2.1 Structural model
The structural model of the external trapezoidal sheet may be approximated by a multispan
continuous beam (see Figure 18), which is perpendicular to the purlins supporting it rigidly.
The reference axis of the model lies in the plane of the roof. The length c1 depends on the
covering system, the length c2 is the distance between the last purlin and the ridge point
(150200 mm).
c2
c1
Edge of building
Purlins as rigid supports
Ridge point
Fig. 18 Structural model for external trapezoidal sheet

Figure 19 shows the cross-section of the beam (as 1000 mm wide part of sheet). In the case
of Lindab products the cross-section is defined by the nominal depth (eg. LTP45) and the
thickness of the plate (eg. t=0,5 mm).
38
Ferenc Papp
Plate thickness (t)
Depth
1000 mm
Fig. 19 Cross-section for the beam model

4.2.2 Load model
The sizes of the trapezoidal sheets (external and internal) normally uniform on the whole roof.
For the load model the adequate loading area (the load band with 1000 mm width) along the
longitudinal direction of the roof should be found (see Figure 20). The dead load and the
snow load are uniformly distributed, therefore the position of the load band does not matter.
The intensity of the wind load is changing from zone to zone of the roof, therefore the
maximum wind pressure (Wind Load Case 1) and the maximum wind sucking (Wind Load
Case 2) should be found. The wind pressure, the dead load and the snow load may be the
components of Load Combination 1, while the wind sucking and the dead load may be the
components of Load Combination 2. For example, in case of 50 slope of and assuming cross
wind, both of wind pressure and wind sucking occur on zone J but wind sucking occurs only
on zones F and G (see Annex 2) which are greater than that on zone J. Assuming longitudinal
wind there are no zones where wind pressure occurs and the maximum wind sucking may
occur on zones F and G (see Annex 3). Consequently, the adequate place of the load band
may be considered as it is shown in Figure 20. We have to take the conclusion that the
determination of the load model needs an enthusiastic work.
wind load case 1
I
J
wind load case 2
Ridge line
F
Wind direction: 0 degree
Wind direction: 90 degrees

Ridge line
G
I
H
F
Fig. 20 Adequate positions of load bands in case of wind effect

(for 50 slope of roof)
39
Ferenc Papp
In the practice we may use approximations at the load model but it should be noted that the
engineer is responsible for the consequences: approximation at side of safe may lead to extra
costs, while approximation at side of unsafe may be against the law. The following two load
combinations may be adequate for the examination of the external trapezoidal sheet in
persistent and transient design situations:
Load Combination 1 (pressure load)
- ultimate limit state (ULS):
- serviceability limit state (SLS):
Load Combination 2 (sucking load)
- ultimate limit state (ULS):
G ,sup pg + s ps + w.0 w pw.k . p

q g + p s + w .0 p w.k . p
G ,inf p g " +" w pw.k .s
- serviceability limit state (SLS):

p g " +" pw.k .s
The following indices are used in the above expressions:
pg
characteristic value of the uniformly distributed dead load in [kN/m];
ps
p w .k
characteristic value of the uniformly distributed snow load in [kN/m];

characteristic value of the non uniformly (but uniformly within a zone) distributed
wind load in [kN/m] due to cross or longitudinal wind direction, which is the
relevant, and p denotes wind pressure, s denotes wind sucking.
These loads can be calculated directly from the basic loads which were defined for the
building (see Practice 2). The accidental snow load is neglected in this project.
The following partial factors should be used:
G ,sup = 1,35 , G,inf = 1,0 , s = 1,5 , w = 1,5
The following combination factor should be used: w.0=0,6
The ULS load combination is relevant for checking the resistances, while the SLS load
combination is relevant for checking the deflections, where the suggested limit is L/150 (L is
the span of sheet).
4.2.3 Design and Documentation
Design of thin gauged Lindab trapezoidal sheet may be carried out using the DimRoof design
software. The application of the software is described in Annex 4. Before starting design the
following General Settings (design parameters) should be defined:
Function
Trapezoidal sheet belongs to Roof category.
Country
By option of Hungary we specify the Hungarian National Annex to be used.
Standard
The design is governed by the Eurocode standard system.
Profile
The depth of the Lindab trapezoidal sheet (LTP) may be 20 mm150 mm. Here the
adequate depth should be selected.
Extra Sidelap
The type of the sidelap (no; 1 trough; 2 trough or double) is determined by the
architectural engineer. In this project 1 trough is suggested using.
Default statical model
Lap of sheeting should be avoided in longitudinal direction. The length of the sheets is
limited by the transportability. In this project continuous model is suggested.
40
Ferenc Papp
The Geometry (geometrical parameters) is defined by the following parameters:

Spans
Thickness of the plate (t=0.4 mm0.7 mm);
Type and width of the supports
The Loads (parameters of loading) are defined by the following parameters:
Type of load
- U - uniform: the load acts along the whole beam as uniformly distributed load;
- L - linear: the load acts along a part of the beam, and/or its intensity is changing
linearly;
(in the first case only the intensity should be defined, in the second case the
coordinates and the load intensities at the starting and end points of the loaded domain
should be defined)
Load attribute
- ULS: load to examine the resistances of element;
- SLS: load to examine the deflection of element.
It is noted that the software can take one ULS and one SLS load combinations into
consideration at the same time. Both of the combinations may contain more load components
(one load component is defined by one row in the table).

The analysis and check can be run by the Calculate! button. The results of ULS and SLS are
given in [%]. More details of design may be available by the Relative Results and the
Absolute Results buttons. If one of the results exceeds 100%, or it is less than 50-60%, the
profile should be modified. The modification means change in plate thickness or/and in depth
of sheet. After the modification the analysis and design should be repeated. The final result of
the design may be documented by the following parameters:
Depth of the profile (eg. LTP 85)
Plate thickness (eg. t=0,75 mm)
ULS capacity in [%];
SLS capacity in [%] (with the limit for deflections, such as L/150)
4.2.4 Application
3. DESIGN OF COVERING ELEMENTS

In the design project the external trapezoidal sheet and the purlins of the roof system are
designed. Design of the wall elements is neglected.
3.1 Design of external trapezoidal sheet
3.1.1 Geometric model
The external trapezoidal sheet of the roof system is modelled by multispan beam as shown
below.
Parameters of the trapezoidal sheet: LTP85 t=0,75 mm.
250
2597
2597
41
2597
2447
150
Ferenc Papp
3.1.2 Load model
Loads are given in [kN/m] for 1000 mm width of sheet.
3.1.2.1 Characteristic loads
- dead load (self weight of the trapizoidal sheet according to Section 2.1.1)
q dead
q tr.ext. cos .
p snow
s . cos .
q dead = 0.084
180
- snow load (totally distributed load according to Section 2.2.1)
- wind load
p snow = 0.97
180
cross wind (wind effect according to Section 2.3.4.1)
H - zna
F - zna
1,85
4,625
a) wind sucking
p wind.0.F.s
w F.0.1
p wind.0.F.s = 0.929
w H.0.1
p wind.0.H.s = 0.31
p wind.0.H.s
b) wind pressure
p wind.0.FGH.p
p wind.0.FGH.p = 0.041
w FGH.0
longitudinal wind (wind effect according to Section 2.3.4.2)
G - zna
4,625
F - zna
1,850
a) wind sucking
p wind.90.F
w F.90.1
p wind.90.F = 0.868
p wind.90.G
w G.90.1
p wind.90.G = 0.826
b) wind pressure: negligable

3.1.2.2 Design load combinations
Partial factors
G.sup
dead load
1.35
s 1.5
w 1.5
snow load
wind effect
Combination factors
wind effect
w.0
42
0.6
G.inf
1.0
Ferenc Papp
Load Combination 1 : 'Pressure effect' (signed by 'p')
Effect of gravity (dead and snow) loads is increased by wind effect on zones F and G
p p.ULS
p p.SLS
G.sup . q dead s . p snow w.0. w . p wind.0.FGH.p

q dead p snow w.0. p wind.0.FGH.p
p p.ULS = 1.599
p p.SLS = 1.074
pp.ULS ; pp.SLS
Load Combination 2 : 'Sucking effect' (signed by 's')

Maximum wind sucking from longitudinal wind effect
Zone F
p s.ULS.F G.inf. q dead w . p wind.90.F
p s.SLS.F
q dead
Zone G
p s.ULS.G
p s.SLS.G
p s.SLS.F = 0.784
p wind.90.F
G.inf. q dead w .p wind.90.G

q dead p wind.90.G
ps.ULS.F ; ps.SLS.F
p s.ULS.F = 1.218
p s.ULS.G = 1.156
p s.SLS.G = 0.743
ps.ULS.G ; ps.SLS.G
4 625
3.1.2.3 Check of the limit states
External trapezoidal sheet is checked for ultimate and servicebility limit states using the
DimRoof 3.3 design software.
Results of the calculations:
- ULS: 49% (Case 1)
- SLS (with maximum deflection L/150): 36 % (Case 1)
For details see the Annex.
Lindab LTP85 t=0,75mm trapezoidal sheet is adequate!
(LTP45 trapezoidal sheet would be not adequate for SLS.)
4.3 Design of purlins

4.3.1 Geometrical model
The structural model of the purlin may be approximated by a multispan continuous beam (see
Figure 21) which is perpendicular to the main frames. The frames support the beam rigidly.
The top flange of the Z purlin is restrained laterally by the trapezoidal sheet. The bottom
flange may be restrained laterally, if the insulation is placed in the room of purlins (see
Figure 8a). The load acts in the intersection point of the web and top flange. These
sophisticated conditions are modeled by the design software.
43
Ferenc Papp
p [kN/m]
Effect of
external
trapezoidal
sheet.
Main frames as supports
Effect of
internal
trapezoidal
sheet, if
there is.
Fig. 21 Structural model for design of purlin

4.3.2 Load model
The roof structure normally consists of uniformly sized purlins. The dead load and the snow
load are totally and uniformly distributed, therefore that purlin should be examined which has
the greatest width of loading area. The wind load acts on zones with different wind pressures,
therefore the place of the maximum pressure and the maximum sucking should be found. The
maximum wind pressure acts with the dead load and the snow load together (Load
Combination 1), while the maximum wind sucking acts with the dead load (Load
Combination 2). It is noted that all the possible load combinations should be examined
carefully.
Load Combination 1 (pressure)
Ridge line
H
F
purlin
Load Combination 2/a (sucking)

Ridge line
purlin
Wind
direction
(0 deg)
Load Combination 2/b (sucking)
Ridge line
G
Wind
direction
(90 deg)
purlin
Fig. 22 Load bands for check of purlins

(valid for purlin system with uniform interval and 100 slope of roof)
44
Ferenc Papp
For example, in case of 100 slope of roof and assuming cross wind, wind pressure occurs only
on zones F, G and H but on the other zones wind sucking acts only (see Annex 2). Assuming
longitudinal wind sucking occurs on all the zones (see Annex 3). The maximum wind sucking
acts on zones F and G. In case of purlin design the internal wind effect should be considered.
According to the above statements, the following load combinations should be taken into
consideration (Figure 22):
Load Combination 1:
Load Combination 2:
Load combination consists of dead load and snow load, and the
effects of external wind pressure (ext.p) and internal wind
sucking (int.s):
ULS: G ,sup p g + s ps + 0 w ( pw .ext . p + p w.int .s )
SLS: p g + p s + 0 ( p w.ext . p + p w.int .s )
Load combination consists of dead load and the effects of
external wind sucking (ext.s) and internal wind pressure (int.p)
at zones F and G:
ULS: G ,inf p g " +" w ( p w.ext .s + pw .int . p )
SLS: p g " +" ( p w .ext .s + p w.int . p )

The load components used in the above load combinations may be calculated from the basic
loads which were determined in Practice 2:
pg
uniformly distributed dead load in [kN/m];
ps
uniformly distributed snow load in [kN/m];
pw
distributed wind load on the actual zone due to cross or longitudinal wind effect in
[kN/m], which is the relevant.
Furthermore, index ext denotes the external wind effect, index int the internal one, while
index p denotes pressure and index s sucking.
The partial factors to be used: G,sup=1,35; G,sup=1,0, S=1,5 s w=1,5
The combination factor to be used: 0=0,6.
The ULS load combination is relevant for checking the member resistances, while the SLS
load combination is relevant for check deflection of members, where the usually used limit is
L/150 (L is the span of the beam). The accidental snow load is neglected in this project.
4.3.3 Design and Documentation
Design of thin gauged Lindab purlin may be carried out by the DimRoof design software. The
application of the software is described in Annex 5. Before the stating of the design the
following General Settings should be defined:
Function
Design of purlin belongs to the option Z-beam).
Country
Selecting Hungary use of the Hungarian National Annex is specified.
Standard
Design is governed by the Eurocode standard system.
Anti-seg bars
The purlins may be supported laterally by anti-seg bars at midspans in plane of roof. In
this design project, due to the small slope, these bars are not used, therefore option None
should be selected.
Profile
The depth of Lindab Z purlin may change between 100 mm and 350 mm. The initial
depth should be selected here.
45
Ferenc Papp
Default static model

Different static models may be used. Relatively short elements offer easy transportation,
therefore the overlapped solution is suggested for using (10-10% overlapping at every
support, except the second ones at the ends where it is 20%). For this the option of
Overlapped Standard should be used.
Lateral Support
Z purlin may be supported laterally at both flanges (see Figure 8a) or at top flange (see
Figure 8b). The option Both flanges or Top flange may be used.
Sheeting
Lateral restraint effect depends on the size of the trapezoidal sheet is used (Profile,
Thickness).
Screws
The load carrying capacity of purlin may depend on the size and the grade of the bolts
(screws) which connect the trapezoidal sheet to the purlins and the purlins to the main
frames:
- Plate:
grade of screws which connect the trapezoidal sheet to the top
flange of the purlins (4.8 or 5.5);
- Overlap:
grade of screws which connect the webs of purlins to each other close
to the supports (4.8 , 5.5 or 6.3 );
- Support:
grade of bolt which connect the webs of the purlin to the main
frame (5.5 or 6.3 );
- Distance:
distance between bolts which connect the trapezoidal sheet to
the top flange of the purlins (one/bottom or one/2nd bottom).
The Geometry of the purlin should be defined by the following parameters:
Spans
Length
Purlins are supported by the main frames. The length of spans should be defined from the
left end to the right end of the beam.
Th.1
Plate thickness may be as follows: 1,0 1,2 1,5 2,0 2,5
Supports
L1 and L2
According to option Overlapped Standard the software takes 10-10% overlapping at
both sides of the supports.
Type
Different types of support may be used. According to option Overlapped Standard the
software takes hinged (H) supports at the ends of the beam, and overlapped (O )
supports at the intermediate ones.
Width
Width of the supports should be defined by the user. For type H it may be the width of
the flange of the frame, for type O it may be the distance between the bolts which
connect the purlin to the stub (stub is welded to the top flange of the frame).
The loads should be defined by the following parameters. It is noted that the software can
consider one ULS and one SLS load combinations at the same time. Both of the combinations
may contain more load components (one load component is defined by one row in the table).
Type
The software uses more types of loads. In this project we use the following types:
- uniformly distributed load along the whole beam (option U);
- linearly distributed load on a part of the beam (option L).
46
Ferenc Papp
Start p./End p.
In case of L type load they are the coordinates of the start point and the end point of the
linearly distributed load, taking the direction from left to right.
Start int./End int.
In case of U type of load it is the load intensity at the start point. In case of L type
load both of the start and end load intensities should be defined.
ULS/SLS
For every load component (for every row) the type of load combination should be defined
(ULS for load carrying capacities, SLS for serviceability capacity calculations).
The analysis and check can be start by the Calculate! button. The results of ULS and SLS
calculations are given in [%]. More details may be available by the Relative Results and the
Absolute Results buttons. If one of the results exceeds 100%, or it is less than 50-60%, the
profile should be modified. The plate thickness or/and the depth of the purlin may be
modified. After the modification the analysis and design should be repeated. The final result
of the design may be documented by the following parameters:
Depth of profile (eg. Z250)
Plate thickness (eg. t=1,5 mm)
ULS used part of capacity in [%];
SLS used part of capacity in [%] (with the limit for deflections, such as L/150)
4.3.4 Application
3.2 Design of purlins
3.2.1 Geometric model
Purlins are modeled by a many supported beam shown below:
X
6 000
6 000
6 000
6 000
6 000
6 000
Z 250 (t=2.5 mm)

Y
Z
3.2.2 Load model
3.2.2.1 Load combinations in general
Intervals between purlins are uniform, therefore the place of the examined purlin does not
matter. Wind pressure and maximum wind sucking occur at zones F and G. Consequently,
the purlin next to the edge beam should be examined.
Load Combination 1: Combination for pressure loads (signed by 'p')
Load Combination 2: Combination for wind sucking (signed by 's')
Width of load area [m]
cL
e4
c L = 2.598
1000
47
Ferenc Papp
cL
purlin to be designed
3.2.2.2 Characteristic loads

Dead load
Self weight of purlin, external trapezoidal sheet, insulation and internal trapezoidal sheet
according to Section 2.1.1 in [kN/m]:
q dead c L. q tr.ext q tr.int q iso q iso.other . cos .

q purlin .cos .
180
180
q dead = 1.194
Snow load
Totally distrubuted snow load according to Section 2.2.1 in [kN/m]:
p snow
c L. s . cos .
p snow = 2.52
180
Wind load
Case 1: pressure due to cross wind on zones F and G
- external wind effect
p wind.1.ext c L.w FGH.0
- internal wind effect
p wind.1.int
c L. w i.0
p wind.1.ext = 0.107
p wind.1.int = 0.216
Case 2: a) sucking due to cross wind on zones F and G

zone F
p wind.2.a.F c L. w F.0.10
p wind.2.a.F = 1.395
c L. w G.0.10
p wind.2.a.G = 1.073
zone G
p wind.2.a.G
- internal wind effect: opposite to external wind effect, it is negligable

b) sucking due to the longitudinal wind on zones F and G
zone F
p wind.2.b.F c L. w F.90.10
p wind.2.b.F = 1.556
zone G
p wind.2.b.G
c L. w G.90.10
p wind.2.b.G = 1.395
- internal wind effect: opposite to the external wind effect, it is negligable

Partial factors
dead load
G.sup 1.35
G.inf 1.0
snow load
wind load
s 1.5
w 1.5
w.0 0.6
Load combinations for two load cases [kN/m]
Load Combination 1 : Pressure effect
ULS p p.ULS G.sup . q dead s . p snow w.0. w . p wind.1.ext p wind.1.int
SLS p p.SLS q dead p snow w.0. p wind.1.ext p wind.1.int
p p.ULS = 5.683
p p.SLS = 3.908
Combination factor for wind effect
pp.ULS ; pp.SLS
48
Ferenc Papp
Load Combination 2 : Sucking effect - longitudinal wind is dominant

zone F
zone G
ULS
p s.ULS.F
SLS
p s.SLS.F
ULS
p s.ULS.G
SLS
p s.SLS.G
ps.ULS,F ps.SLS,F
G.inf. q dead w . p wind.2.b.F

q dead p wind.2.b.F
p s.ULS.F = 1.118
G.inf. q dead w .p wind.2.b.G

q dead p wind.2.b.G
p s.ULS.G = 0.877
ps.ULS,G ps.SLS.G
p s.SLS.F = 0.34
p s.SLS.G = 0.179
ps.ULS.F ; ps.SLS.F
4625
4625
3.2.2.4 Check of the limit states

Check of the ultimate and servicebility limit states was carried out by DimRoof 3.3 software:
Results for pressure effect:
- ULS: 71 %
- SLS: 35 % (for maximum deflection of L/150)
Results for sucking effect:
- ULS: 14 %
- SLS: 3 % (for maximum deflection of L/150)
Details of calculation can be found in Annex.
LINDAB Z250 (t=2.5mm) purlin is adequate (Z200 would be not adequate)
49
Ferenc Papp
ANNEX 4
Design of trapezoidal sheet using DimRoof software
(Users Manual)
A4.1 Setting
- download the dimroof_xx_telepits.zip file from the departmental portal
(www.hsz.bme.hu/Oktats/Magasptsi aclszerkezetek/Gyakorlat)
- unzip the file and run the Dimroof_33_install.exe
- send the required data to the responsible teacher
- run the program using the code given by the responsible teacher
A4.2 How to use the software
After start and selecting language the design window appears. The window is divided into the
following blocks:
General Settings
Structural Settings
Geometry
Loads
Parameters for Deflection Check
Results
A4.2.1 General Settings
Design of trapezoidal sheet should be carried out with general settings shown in Figure A4.1.
Fig. A4.1 Genaral settings for design of trapezoidal sheet

A4.2.2 Structural Settings
In Profile box the parameters of cross-section should be defined (eg. LTP45), where the
number denotes the depth of the profile. Cross-sectional parameters are available by Profile
tool (see Figure A4.2).
Fig. A4.2 Cross-sectional parameters available by Profile tool

50
Ferenc Papp
Position of sheet should be defined in Flange Up box, where the adequate option is Common
(narrow). This means that the narrower flange is positioned up. The measure of sidelap is
defined in Extra Sidelap box. In our case option 1 trough may be selected. This type of
overlap ensures required waterproof. Static model can be defined in Default Static Model
box. Continuous beam model is suggested for using, therefore option Continuous should be
selected. This means that the sheet is continuous between ridge point and edge beam. These
settings are shown in Figure A4.3.
Fig. A4.3 Structural Settings for design of trapezoidal sheet

A4.2.3 Geometry
Firstly, we should define the parameters for Span/Length. The lengths of spans should be
written into the boxes, where the stating point is the edge beam. The actual static model is
drawn automatically in the graphical window (see Figure A4.4).
Fig. A4.4 Geometrical parameters with the graphics of actual model

The program fulfils the column of Th.1 automatically. The column contains the uniform plate
thickness of the cross-section. The thickness can be changed by clicking on the first row and
choosing the relevant thickness from the list of usable thicknesses. This thickness will appear
in all the boxes. The program fulfils the column of Supports. The Width is equal to the width
of flange of the purlin. The Width in the first row is greater than in the others because the
width of flange of C formed edge beam is greater than the width of flange of Z formed purlin.
The L1 is the length of the cantilever at the edge beam, while L2 is the distance between the
ridge point and the double purlins.
A4.2.4 Loads
At the same time (within a run of the program) only one ULS and one SLS load combination
can be considered. The program should be run more times, if more load combinations would
be considered. Both of ULS and SLS loads may contain more load components. Every row in
the table defines a load component for ULS or for SLS combinations. For example, Figure
51
Ferenc Papp
A4.5 shows the definition of an ULS load combination which contains two load components.
One of them is a totally and uniformly distributed load with 4,2 kN/m intensity, the other is a
partially and uniformly distributed load with 2,5 kN/m intensity (the load starts at the edge
beam and ends at 3000 mm distance).
Fig. A4.5 Definition of an ULS load combination with two components

The possible types of load components are shown in Figure A4.6. The type can be selected by
clicking on the first box of the actual row. Option U denotes totally and uniformly
distributed load, option L denotes partially and linearly distributed load.
Fig. A4.6 Types of load components

A4.2.5 Parameters for Deflection Check
Limit for deflections is not specified by Eurocodes therefore the limit should be defined by
the designer. In this project deflection limit of L/150 is suggested for using (see Figure A4.7).
In the Defl. type box Maximum option is suggested choosing. This means that maximum
deflection along the span will be considered as design deflection.
Fig. A4.7 Limit values for deflection check

A4.2.6 Check
First step of the check of member is the analysis. By Calculation button the program carries
out the analysis and check of model. The used parts of capacities are shown in the Results
boxes (see Fig. A4.8). The member may be considered being designed if the used part both of
the ULS and SLS capacities are less than 100% but one of them is as close to 100% as
possible. The detailed results can be available by the Relative Results and Absolute Results.
Design parameters are the depth and the thickness of the trapezoidal sheet.
52
Ferenc Papp
Fig. A4.8 Tools to get results of check and documents

A4.2.7 Documents
Results of design should be documented. The type and form of the document should be
economic in cost. It is acceptable if the student presents the design to the teacher on his/her
notebook.
53
Ferenc Papp
ANNEX 5
Design of purlin using DimRoof software
(Users Manual)
General description of DimRoof software can be found in Annex 4. In this Annex specific
knowledge of purlin design is described.
A5.1 Settings
Design of purlin can be carried out with general settings shown in Figure A5.1.
Fig. A5.1 General settings for design of purlin

A5.2 Structural Settings
Size of purlin (size of the Z section) should be chosen in the Profile box (see Figure A5.2).
Fig. A5.2 Structural Settings for design of the purlin

In Default Statical Model box option Overlapped Standard is suggested for select on. This
means that every purlin has 10-10% overlap at right and left sides of support. Overlap makes
the beam being continuous and stronger at the supports. In Manufacturer box option
Hungary should be selected. Adequate option for Lateral Support is Top Flange, if
bottom flange is not supported by internal trapezoidal sheet. Option Both Flanges means
that both of the flanges are supported by trapezoidal sheets. In Sheeting frame the size of the
external trapezoidal sheet should be given. Frame of Screws contains details for the
connection between purlin and trapezoidal sheet and between purlin and frame, as well.
54
Ferenc Papp
A5.3 Geometry
First, Spans/Length column should be fulfilled (see Figure A5.4). Lengths should be defined
from the left end of beam to the right end. Actual static model will appear in the graphical
window.
Fig. A5.4 Geometry and static model
Column Th.1 is automatically fulfilled by the software. This parameter denotes the uniform
plate thickness of the purlin. Thickness can be changed by clicking on the first row and
choosing the relevant thickness from the list of usable ones. This thickness will appear in all
the rows. The program fulfils the Supports column. Columns Type, L1 and L2 contain the
parameters of type of support and the overlaps at supports. Software fulfils these columns
automatically. Default overlap of the first and last purlin parts in adjacent intermediate spans
is double of the usual one (20%).
Fig. A5.5 ULS and SLS loads with one component, respectively
55
Ferenc Papp
A5.4 Loads
At the same time (with a run of the program) only one ULS and one SLS load combination
can be considered. The program should be run more times if more load combinations should
be considered. Both of ULS and SLS loads may contain more load components. Every row in
the table defines a load component for ULS or for SLS load combination. For example,
Figure A5.5 shows a totally distributed load for ULS combination with 3,4 kN/m intensity,
and a totally distributed load for SLS one with 2,6 kN/m intensity.
A5.5 Parameters for Deflection Check
Limit for deflections is not specified by Eurocodes, therefore the limit should be defined by
the designer. In this project deflection limit of L/150 is suggested for using (see Figure A5.6).
In Defl. type box option Maximum is suggested choosing, which means that the maximum
deflection along the span is considered as design deflection.
Fig. A5.6 Settings for check of deflections

M5.6 Check
First step of the check of member is the analysis. By Calculation bottom the program carries
out the analysis and check of model. The used parts of capacities are shown in Results boxes
(see Fig. A5.7). The member may be considered being designed if both of the ULS and SLS
capacities are less than 100% but one of them is as close to 100% as possible. The detailed
results can be available by the Relative Results and Absolute Results. Design parameters are
the depth and the thickness of the purlin.
Fig. A5.7 Results and Documents

M5.7 Documents
Results of design should be documented. The type and form of the document should be
economic in cost. It is acceptable if the student presents the design to the teacher on his/her
notebook.
56
Ferenc Papp
Practice 5
ANALYSIS OF THE MAIN FRAME AND
DESIGN OF THE CROSS-SECTIONS
2010-2011 Budapest
57
Ferenc Papp
Practice 5
ANALYSIS OF THE MAIN FRAME AND DESIGN OF THE CROSSSECTIONS
5.1 Design model
5.1.1 General
Design model of steel structures may be created in the following steps:
Structural model
Load model
Static model
The main components of the structural model:
- reference axis of members
- initial cross-sections
- supports
The main components of the load model:
- system of load groups and cases
- loads of load cases
- design load combinations
The static model (mesh) is automatically generated by the design software on the base of the
structural model. The applied software may be based on 2D or 3D. The 2D based software
neglect the out-of-plane deformation as well as the twisting of the members. The 2D model is
used to calculate the in-plane design displacements and forces. The 3D based software may
use the so called 12DOF element which neglects the warping effect, or the so called 14DOF
element which takes the warping effect into consideration. The last type of software may
analyse the spatial stability behaviour of the structure. The check of the stability of members
may be based on in-plane forces and interaction design formula, or it may be based on out-ofplane stability analysis and general design formula (see the details in Practice 7). Table 8
assumes the above discussion. In the present design notes follows the 3D and out-of-plane
model based stability checking which is summarized in the third row of the table.
Tab. 8: Relationship between structural model and mechanical model as well as the applied software
(informative)
structural
type of element
DOF
warping
stability analysis
suggested
model
(mechanical
(degrees of
effect
software
model)
freedom)
in-plane
(2D)
out-of-plane
(3D)
spatial
(3D)
in-plane
no
in-plane buckling
AXIS
out-of-plane
12
no
out-of-plane
14
yes
in-plane buckling
(about minor and
major axis)
complex spatial
AXIS
FEM-Design
ConSteel
ConSteel

The main geometry of the frame and the initial cross-sections of the structural members are
known from the preliminary drawing. The aim is to create the 3D structural model which is
adequate for computer analysis. The structural members of the main frame may be as follows:
Members with hot rolled profiles and short/long haunch;
58
Ferenc Papp
Members with welded profiles and long haunch;

Tapered welded members.
First it is shown how to define the structural model, then how to take the equivalent geometric
imperfections into consideration. Finally, it is discussed how to support the model.
5.1.2.1 Structural model of main frame composed of hot-rolled profiles
The ends of the beams of main frames composed of hot-rolled profiles are normally
constructed with haunch. The types of haunch are distinguished:
Short haunch
Long haunch.
The short haunch is used to increase the distance between the tensioned bolts and the
compressed flange of the end plated moment resistant connection. Applying short haunch the
joint may be stiff and full resistant (see later in Practice 8). The short haunch may be
neglected in the structural model which means that the structural members can be modeled
with uniform cross-sections (see Figure 23a). Long haunch is used to increase the capacity of
the joint (like at short haunch) and partially the beams. The long haunch should be considered
in the structural model (see Figure 23b).
(a)
Uniform members
(b)
Partially non-uniform
members
Fig. 23 Structural model of main frame: (a) with short haunch; (b) with long haunch
5.1.2.2 Structural model of main frame composed of welded profiles
I and H profiles may be fabricated by welding technology. The geometric parameters (width
and thickness of plates) of the welded profiles may be similar than those of the hot-rolled I
and H profiles (equivalent welded sections), but the welded profiles may be constructed with
thinner plates. The fabrication of welded profiles may be economical in great amount (see for
59
Ferenc Papp
example www.dongbusteel.co.kr/eng/product ). The creation of the structural model is based

on same rules than those were described for hot-rolled profiles (see Section 5.1.2.1).
5.1.2.3 Structural model of main frame composed of tapered members
The tapered members has normally relatively thin web with linear change of depth. The web
plate belongs to Class 4 in bending, while the flanges belong to Class 2 or 3. The reference
axis starts at the centroid of the lower end of the members. The model is centric, if the
reference axis follows the centroid of the cross-sections (see Figure 24a). If the reference axis
is out of the centroid, for example it is parallel with the outer vertical flange, the model is
eccentric (see Figure 24b). The centric model allows more accurate analysis but it is
complicated to build it up. It is easy to build up the eccentric model, but eccentric finite
element should be used in the finite element model (see Section 5.2). The outer flanges of the
columns should be vertical at both of the models.
(a)
(b)
Fig. 24 Structural model composed of tapered members:

(a) centric model; (b) eccentric model
5.1.2.4 Equivalent geometric imperfections
Depending on the design method the equivalent geometric imperfections should be
considered in the structural model. In general the equivalent geometric imperfection is given
by the appropriate buckling shape (eigenvector). According to the design standard the
equivalent geometric imperfection may be replaced by two components:
Global imperfection
Local imperfection
The shapes of the components do not depend on the buckling shape.
The global imperfection for frames is shown by Figure 25, where the inclination is
= 0 h m
and
0 = 0 ,005
h =
2
h
de
2
h 1,0
3
60
Ferenc Papp
m = 0 ,5 1 +
m
e
e
e = h
Fig.25 Global equivalent geometric imperfection

and where m denotes the number of columns (for the frame shown in the Figure m=2). The
global equivalent imperfection may be neglected if
H Ed 0 ,15 VEd
where HEd is the sum of the horizontal loads, VEd is the sum of the vertical loads. It is noted
that in case of simple building the wind pressure on the walls satisfies the above condition.
Equivalent horizontal load may be applied to replace the global imperfection. The equivalent
load (VEd) is shown in the Figure 26.
VEd
VEd
Fig.26 Equivalent load to replace the global equivalent imperfections

The local imperfection is defined as an initial out-of-straightness of the structural member.
The shape of the out-of-straightness may be sinus half-wave or parabolic curve, where the
amplitude should be taken from the Figure 27. If the structure is relatively stiff, the probable
amplitude of the out-of-straightness of the member is not greater than L/800 and the reduction
factor method is used to determine the buckling resistance (see Practice 7), then the local
imperfection should not be used (the reduction factor contains the effect of the out-of
straightness with characteristic value of amplitude).
5.1.2.5 Supports
Supports are very important feature of the structural member. The right support model of the
main frame is determined by the bracing system of the whole structure of the building. The
supports can be defined on the base of the accepted preliminary configuration. Concerning the
support model the following topics should be discussed:
61
Ferenc Papp
e0
e0
amplitude for the

out-of-straightness
buckling curve
e0
a0
a
b
c
d
L/350
L/300
L/250
L/200
L/150
Fig.27 Equivalent local imperfection

Support model for column base
Supports on beams
Supports on columns
Knee bars
In the discussion we assume that the structure is placed in the X-Z global system plane,
consequently the support in direction which is perpendicular to the plane of structure is
direction Y.
Support model for column base
The support model for a column base depends on the structural construction of the column
base. The support is normally idealized, which means that the displacements in directions and
rotations about directions are assumed being absolutely free or restrained. In more
sophisticated models the support model may be assumed being semi-rigid (linearly elastic).
Figure 28 illustrates the idealized support model of the simple pinned column base, while
Figure 29 shows the idealized support model of the common rigid column base.
(a) Structural construction
(b) structural model
(c) static model

The end of the column is fixed in
all the directions, and the
rotation about the axis of the
member is restricted.
The support model (code) in

ConSteel software is:
x , y , z , zz
Fig.28 Support model for simple pinned column base
62
Ferenc Papp
(a) Structural construction
(b) structural model
(c) static model
The end of the column is fixed in

all the directions, and is restricted
in all the rotations about the axes.
The support model (code) in
ConSteel software is:
fix
Fig.29 Support model for common rigid column base

Supports on beams
The supports which are perpendicular to the plane of the examined structure (lateral supports)
should be modeled realistically if spatial stability analysis are wanted to calculate on the
model. In contrary the lateral supports should only ensure the running of the second order
analysis. For this it is enough to support the model in the break joints and in quarters of the
members.
Lateral restraint (lateral support) may be ensured by the following two structural members:
Bracing bar connected to the bracing system
Purlin
The places of the bracing bars are defined in the preliminary draw. The supporting effects of
the purlins are illustrated in Figure 30. The purlin is normally sitting on the top flange of the
frame beams but it may be assumed that they support the beam in the centroid (see
Figure 31).
It is a question whether all the purlins may be assumed as lateral support. There are two
answers (models) for the question:
Model 1
Purlin can be assumed as lateral support for the main beam that is connected to the bracing
system. Figure 32 shows this situation where every second purlin is connected to the joint of
the bracing system (in the figure yellow color indicates the purlins that are assumed as lateral
support).
Model 2
All the purlins may be assumed as lateral support, even if some of them are not connected to
any joint of the main beam (see Figure 32).
Model 1 leads to conservative design which is suggested flowing if the structure has large
scale (considerable depth of cross sections), and it has high importance (ex. hangar, power
station). Model 2 leads to economic design which is the number one rule of the mass
production (decreasing costs to minimum). The theoretical background of Model 2 may be the
shear stiffness of the covering system. Figure 33 illustrates this model where the covering
system restraints the purlins that are not connected directly to the join of the bracing system.
63
Ferenc Papp
Fig.30 Modeling the effect of the purlins on the main beam
Fig.31 Place of the support (simplified modeling)
Fig.32 Two models to take the effect of purlins into consideration:

Model 1 (conservative); Model 2 (economical)
64
Ferenc Papp
Purlin which is connected to

the bracing system
Fig.33 Restraint of the purlin which is not connected to the bracing system
(Roofing in bottom view)
Supports on columns
The columns are supported by the wall beams that run parallel to the plane of the main frame.
The wall beams may be connected to the bracing system of the wall structure. The modeling
rules of the lateral supports of columns follows the above rules that were established for the
roofing.
Knee bars
Many times the supporting model of the main frame based on the modeling rules established
above is not sufficient enough, in other words the critical load amplifier is too low (see
Section 5.2.2). The critical load amplifier may be increased by more lateral support, but more
times applying knee bars leads to a right solution. The knee bars are applied at the bracing
members where they connect the compressed flange of the main structural member (beam,
column) to the bracing member (see Figure 34). This construction may restrict the twist of
the cross-section of the structural member. The knee bars may be sufficient if the buckling
mode of the main structure is lateral torsional buckling (LTB). Section A6.9 of Annex 6
shows how to model the knee bars.
65
Ferenc Papp
Fig.34 Knee bars against twist of the main member

5.1.2.6 Simplified model or full 3D structural model?
Figure 35b shows the isolated model of one of the main frames of a simple building. From
point of view of the model it does not matter whether purlins or bracing members supply the
lateral supports. Figure 35a illustrates the full 3D model of the building structure which may
be constructed with relatively few amount of work. It is a practical knowledge that this full
3D model does not supply more information for the main frame than those which is isolated
(simplified model) and supported due to the effect of purlins or bracing members.
The full 3D model may be useful at the design the bracing system. It is noted that the fine
tuning of the full 3D model as well as the creation of the load model may take considerable
amount of work but this model may be a sufficient design tool for engineers who have some
years practice. In this design project we do not suggest using full 3D model.
(a)
(b)
Fig.35 Structural models for simple building: (a) full 3D model; (b) simplified model
66
Ferenc Papp
5.1.2.7 Application
The example below shows the modeling of the main frame in ConSteel design software.
Annex 6 describes the modeling step by step.
4. ANALYSIS AND DESIGN OF MAIN FRAME
In this design project the main frame which is situated close to the side wall is designed. The
result of the design is considered being valid for all the main frames. The analysis and design is
carried out with the ConSteel design software.
4.1 Design model
The initial structural model of the examined main frame was created on the base of the
Preliminary Draw found in Fig. 9-11 of this Design Notes. The model was constructed on the 3D
modeling window of the ConSteel software:
Knee bar
Knee bar
The eccentricities of the lateral supoports were neglected in the structural model. Supports
were placed on to the reference axes and in the break points. The intermediate supports on
columns and beams were placed at the middle points of the members. At the intermediate
supports of the beams knee bars were assumed.
4.1.2 Cross-sectional models and properties of cross-sections
The models of cross sections were developed in the ConSteel software, and the
cross-sectional properties were imported from the software.
4.1.2.1 Cross-section of columns
67
Ferenc Papp
Class of cross-section
flange
web
pure compression
1
4
pure bending
1
1
Cross-sectional properties
- area [mm2]
- moments of inertia
A c.pl
[mm 4 ]
I c.y
- sectional moduli [ mm3 ]
11424
518270000
W c.y.pl
- St. Venant inertia moment
[mm 4]
- warping constant [mm 6 ]
I c.t
I c.w
2296600
A c.eff
I c.z
10609
36884000
W c.z.pl
460800
729500
2157000000000
4.1.2.2 Cross-section of beams
flange
web
pure compression
1
4
pure bending
1
1
- area [mm2]
- moments of inertia
A b.pl
[mm 4 ]
I b.y
- sectional moduli [ mm3 ]

- St. Venant inertia moment
9888
308197000
W b.y.pl
[mm 4]
- warping constant [mm 6 ]
I b.t
I b.w
1677000
A b.eff
I b.z
9346
36870000
W b.z.pl
460800
671166
1358000000000
4.1.2.3 Cross-section of haunched beam

Haunch plates
- flange: 240-20
- web: 310-6
The cross-section of the haunched beam is approximated by an I section where the
intermediate flange is neglected. The height of the cross-section is considered in the
theoretical corner of the frame.
68
Ferenc Papp
flange
web
pure compression
1
4
pure bending
1
1
- depth
- area
h bh
[mm2]
730
A bh.pl
- moments of inertia [mm 4 ]

- sectional moduli
I bh.y
[ mm3 ]
1253000000
W bh.y.pl
- St. Venant inertia moment [mm 4]

- warping constant
12804
I bh.t
[mm 6 ]
I bh.w
3716000
A bh.eff
I bh.z
10501
41472000
W bh.z.pl
518400
1017000
5191000000000
4.1.3 Material properties

- grade of steel: S235
- design strength [N/mm 2 ]
fy
235
- elastic moduli [N/mm 2]
210000
0.3
- shear moduli
[N/mm 2 ]
E
.
2 (1 )
4
G = 8.077. 10
4.1.4 Particial factors

- for cross-sectional resistances
M0
1.0
f y0
- for stability resistances
M1
1.0
f y1
fy
M0
fy
M1
5.1.3 Load model

5.1.3.1 General
The basic loads of the building were determined in Practice 2. Using these loads the design
loads of the examined frame should be determined in this Chapter. The loads which have
effect on the covering system (dead load of covering, snow load and wind load) are
transmitted by the purlins as concentrated loads, see Figure 36a. If there are at least 3-4
purlins (wall beams) on the beam (column), the applied load may be considered as distributed
load, see Figure 36b. In this design project distributed load is suggested applying.
5.1.3.2 Adequate frame to design
The wind load on the roof is changing from zone to zone. Therefore the frames of the building
have different design loads. The greatest wind effect can be found on the frames at the end
walls of the building but the width of the loading area is only the half of the frame distance.
Generally, it is for safe and economical design if the second frame is examined, as it is shown
in Figure 37.
5.1.3.3 Load groups and load cases
The design software supports the generation of the load combinations. In order to use this tool
the load cases should be collected into load groups. Load group contains load cases which
69
Ferenc Papp
(a)
(b)
Fig.36 Loads on the main frame transmitted as:

(a) concentrated loads; (b) distributed load
Frame at wall
Uniform intermediate frames
Adequate frame for design
Width of loading area

Fig.37 Adequate frame and width of loading area for design
(top view of the building)
belong to same load type (ex. snow, wind) and may neglect each others in the load
combinations. Load case contains load items which act together in the same time (ex. cross
wind loads on the frame). In this design project the following load groups and load cases are
suggested applying:
Dead loads
o Weight of structural members (WSM)
o Weight of covering system (WCS)
o Installation loads (IL)
Snow load
o Totally distributed
o Non symmetrically distributed (left sided)
o Non symmetrically distributed (right sided)
70
Ferenc Papp
Wind load
o windload_1
o windload_2
o windload_3
o windload_4
o
Dead loads
The weight of the members of the frame is considered by the software. The user should select
the load case which will contain this load (WSM). The weight of purlins, wall beams and
covering layers may be collected into an individual load case (WCS). The installation loads
(equipments, lightings, and so on) can be collected into another load case (IL). In this project
the following dead load cases are suggested using (see Figure 38):
o WSM:
pg ,s [kN/m] (loads considered by the software)
o WCS:
p g ,c = c qc [kN/m]
o IL:
p g ,i = c qi [kN/m]
where c [m] is the distance between the frames, qc [kN/m2] is the weight of the covering
system, qi [kN/m2] is the installation load.
c q c ,r ; c q i
c qc,w
c qc,w
Fig.38 Model of dead loads

(where r denotes on roof and w denotes on wall)
Snow load
The snow load is gravity load which means that it acts vertically. The basic snow load is
related to 1 m2 horizontal area. The snow load cases are illustrated in Figure 39. In this
design project it is enough to take the totally distributed load (Case 1) into consideration. It is
noted that in cases of unsymmetrical duopitch roofs the unsymmetrical load cases may be
adequate. The frame is loaded by the following totally distributed snow load:
ps = 1 c s
where s [kN/m2] was determined in Practice 2, 1 is a factor given in Section 2.2.1.1. This
snow load case is related to horizontal surface. The reduction of this load may be neglected
for the safe, see Figure 40.
71
Ferenc Papp
Case 1: Totally distributed
Case 2: Unsymmetrical (left)
0.51
Case 2: Unsymmetrical (right)
0.51
Fig.39 Snow load cases
1 c s
Fig.40 Model of totally distributed snow load

Wind load
The wind may affect on the external and the internal surfaces of the building. The loads due to
the wind effect may be calculated as following:
- external wind load: pw ,e = cs cd c we
- internal wind load: pw ,i = c wi
where cscd is the structural factor, which is 1.0 for buildings being not higher than 15 m. In
this design project it is suggested applying the following wind load cases:
Cross wind effect (=0), see Annex 7
Cross wind effect + internal wind effect, see Annex 8a
Longitudinal wind effect (=90), see Annex 8b
Longitudinal wind effect + internal wind effect, see Annex 8c
It can be seen that a wind load case might have more cases. The number of cases (number of
design load combinations) may be reduced neglecting the non adequate load cases.
Unfortunately, there are no general rules how to reduce the number of load cases.
Design load combinations for transient and persistent design situations (basic combinations)
are given by the following expressions (EN 1990 6.4.3.2 6.10):
72
Ferenc Papp
G, j
Gk , j + Q ,1 Qk ,1 + Q ,i 0 ,i Qk ,i
where
G
is the partial factor for dead loads (normally: 1,35)
Q,i
partial factor for load case i (snow and wind load: 1,5)
0,i
combination factor for load case i (snow load: 0,5; wind load: 0,6)
These load combinations can be generated by
Automatically
Engineer
Generation of all the possible load combinations is supported by the software. This automatic
method has some disadvantages:
- second order and stability analysis (it is suggested) should be executed for all of the load
combinations (superposition must not applied);
- results may be reviewed difficultly;
- runtime may be considerable.
Method of generation by Engineer is preferred by senior engineers since using simple
considerations (neglecting the non adequate load cases and combinations) the analysis may be
executed and the results can be reviewed more easily. In this design project it is suggested
using the two methods together: all the possible load combinations can be generated by the
software, and then the inadequate load combinations may be disclosed.
5.1.3.4 Application
The example below shows the creation of the load model using ConSteel design software.
Annex 9 illustrates the program application, step by step.
4.2 Load model
4.2.1 Load groups and load cases
The load cases are set into the following groups:
- dead loads
- snow loads
- wind loads
The load cases in the snow and wind load groups disclose each other in any load
combination.
4.2.1.1 Load cases in dead load group
Self weights of structural members are taken into consideration automatically by the software.
Effect of roof slope is neglected. The loads of wall covering system are neglected.
Dead load of roof covering system
- load on surface [kN/m2]
q g.c
q tr.ext
q tr.int
q iso
q iso.other
q purlin
e4
q g.c = 0.451
- load for beam [kN/m]

p g.c
c . q g.c
p g.c = 2.706
73
Ferenc Papp
Installation loads
- load acting on roof [kN/m2]
q g.i
q light
q equip
q other
q g.i = 0.45
- load on beam [kN/m]

p g.i
c . q g.i
p g.i = 2.7
4.2.1.2 Load cases of snow load group

Unsymmetric snow load cases are neglected.
Accidental snow load (accidental design state) is not examined.
Effect of roof slope is neglected for safe.
- Totally distributed snow load on beam [kN/m]
ps
c .s
ps =6
74
Ferenc Papp
4.2.1.3 Load cases of wind load group
Wind effect is not considerable because of the given geometry of the building.
After discussion of the load system the above five wind load cases are considered:
- external cross wind effect (0 degree): wind sucking on zones F-G-H
- external cross wind effect (0 degree): wind pressure on zones F-G-H
- external and internal cross wind effects (0 degree)
- external longitudinal wind effect (90 degrees)
- external and internal longitudinal wind effects (90 degrees)
Wind loads are calculated with c sds=1, because the height of the building is less than
15 meters.
4.2.1.3.1 Two load cases due to external cross wind effect (0 degree)
Distributed wind pressure on walls [kN/m]
p w.e.D
c . w D.0.10
p w.e.D = 1.801
Distributed wind sucking on walls [kN/m]
p w.e.E
c .w E.0.10
p w.e.E = 0.876
p w.e.F
c .w F.0.10
p w.e.F = 3.222
p w.e.G
c . w G.0.10
p w.e.G = 2.479
p w.e.H
c . w H.0.10
p w.e.H = 1.115
Distributed wind sucking on the roof [kN/m]

Zones F-G-H
Zones I-J
(i) wind sucking
(ii) wind pressure
p w.e.FGH
c . w FGH.0
p w.e.FGH = 0.248
p w.e.I
c . w I.0.10
p w.e.I = 1.239
p w.e.J
c . w J.0.10
p w.e.J = 0.991
Examined load area (second main frame) is covered by zones F and G, but for safe
effect on zone F is considered on the load area.
Width of the zone F [m]
e 0.10.F
e 0.10
e 0.10.F = 1.804
Width of the zone J [m]
e 0.10.J
e 0.10
e 0.10.J = 1.804
Models of load cases
4.2.1.3.2 External and internal cross wind effects (0 degree)

Internal wind sucking effect reinforces the second case of external cross wind effect,
therefore first case is neglected.
Distributed wind pressure on wall [kN/m]
p w.e.D
c . w D.0.10 w i.0
p w.e.D = 2.301
Distributed wind sucking on wall [kN/m]

p w.e.E
c . w E.0.10 w i.0
p w.e.E = 0.377
75
Ferenc Papp
Distributed wind sucking and pressure on the roof [kN/m]
Zones F-G-H
p w.e.FGH
p w.e.FGH = 0.748
Zones I-J
p w.e.I
c . w FGH.0 w i.0
c . w I.0.10 w i.0
p w.e.I = 0.739
p w.e.J
c . w J.0.10 w i.0
p w.e.J = 0.492
4.2.1.3.3 Load case of the longitudinal external wind effect (90 degrees)
On examined load area (second main frame) zone B is dominant for walls, while zone
H for roof.
- Distributed wind sucking load on walls [kN/m]
p w.e.DE c . w B.90.10
p w.e.DE = 1.983
- Distrubuted wind sucking effect on roof [kN/m]

p w.e.H c .w H.90.10
p w.e.H = 1.611
4.2.1.3.4 Load case of longitudinal external and internal wind effects

Internal wind effect is sucking effect which weaknings the external wind sucking, therefore
this load case may not be relevant.
4.2.2 Load combinations
Load combinations shown below were generated by the ConSteel software. These load
combinations belong to persitent design situation. Dead load in combination 12 may be
favorable, therefore the partial factors were modified to 1.00.
76
Ferenc Papp
5.1.4 Static model

Static model may be generated by software. The generation is based on the structural model
and the basic setting. The professional user may change the basic setting and by this way the
main properties of the static model can be determined. One of the main properties of a beamcolumn model is the number of finite elements (FEs). The ConSteel software uses a general
beam-column FE which has uniform cross-section. Following a simple method (so called
segment method), the program distributes the tapered (or haunch) members into a set of
uniform elements. Figure 41 shows the FE model of a haunch member where eccentric
uniform FEs are used. The optimal number of FEs depends on the ratio of the length of the
haunch and the depth of the cross-section. The model with 4 FEs leads usually acceptable
results. The model with 8 FEs may be exact. Tapered members can be modeled on the same
way but at least 8 FEs should be used. The segments may be placed centrically (see Figure
42a) or eccentrically (see Figure 42b). The centric model may be more exact from static
point of view. The eccentric model may be easier to build up and it is closer to the world of
the CAD/CAM systems but it is less exact from static point of view.
Fig.41 Static model for haunch beam using uniform segments (FEs)
5.2 Analysis
5.2.1 General
During the design process analysis should be executed on the following structural models:
Conceptual model
Detailed model
Final model
77
Ferenc Papp
(a) centric model
(b) eccentric model
Fig.42 Static models for tapered frame:

(a) using centric segments (b) using eccentric segments
Analysis on conceptual model is performed in the tender phase when the architectural
design office launches the structural solution. The sharp race in the market requires very fast
procedure for preliminary design, therefore instead of classical hand made calculations the
more accurate computer calculation is used. In this phase it is not needed to deal with all the
structural details, and only first order analysis is executed on simplified structural and load
models. In this design project the preliminary draw may satisfy the requirements of the
conceptual design. The structure was simple, therefore the initial cross-sections were able to
determine without any analysis. It is noted that in case of more sophisticated structures the
preliminary analysis and design could not have been neglected.
Analysis on detailed model should be performed in the realization phase of the structure,
where there are direct economical and criminal consequences of any design mistake. In this
phase all the details should be examined which have any effect on the safe of the structure.
The static analysis is normally executed by computer program which is usually based on finite
element method. Usually second order theory is applied. It may be assumed that the structural
joints are initially rigid or pinned. If any joint of the structure is semi-rigid, the effect of the
initial stiffness of the joint should be built into the model. If any column base is pinned, it
should be modeled by appropriate external point support. Design of joints follows the analysis
and design of the cross-sections, therefore the design procedure is recursive, if there are semirigid joints in the structure. In this design project semi-rigid joints are not suggested
designing.
78
Ferenc Papp
The properties of the whole structure (for example displacements and column base reactions,
should be determined on the final model. The final 3D structural model should contain all the
structural members which have any effect on the complex structural behavior and response.
5.2.2 Main steps of analysis
The structural model is normally analyzed by computer software. Displacements are the
primary result of the analysis. The displacements are shown on the 3D window as the
deformation of the structural model. The displacements should be considered as fictive
structural properties, unless the analysis was executed with the final cross-sections. The
computer analysis may be performed in the following main steps:
Step 1: Check the structural model
The structural model may be checked for a simple and symmetric load case. First order
analysis may be executed, and then the deformation of the structural model should be
examined. By this way the most critical model mistakes (inadequate support system, false
cross-sections, and so on) can be found and the model can be corrected.
Step 2: Check the design model
Assuming that the structural model is adequate, the full load model may be applied. First
order analysis can be executed for all the relevant design load combinations. The model
should have understandable deformation for all the load combinations. If any irregular
behavior can be realized in the model response (ex. there is bending moment at pined joint),
the structural model should be repaired.
Step 3: Stability analysis (optional)
Out-of-plane stability analysis of the structure requires appropriate software and design
model. The software is appropriate if it uses general beam-column finite element which is
able to take the warping effect into consideration (see the ConSteel). The design model is
appropriate if it is 3D model and the support system satisfies the real conditions. False support
system results in false stability response (false critical load amplifier). Using false critical load
amplifier in design may indicate serious criminal consequences. The critical load amplifier is
an important property of the structure, but the required value is not specified in the standard.
Practically, for design of frames the following values may be suggested applying:
If cr 1,0 , the structural is dangerous!
If 1,0 < cr 2 ,0 , the structure is very slender (it is not suggested)!
If 2 ,0 < cr 3 ,0 , the structure is slender but it is possible to design it in a safe way!
If 3,0 < cr , the structure is stiffness and the design can be safe (suggested value)!
The appropriate buckling mode (which belongs to the critical load amplifier) is also an
important property of the structure. The buckling mode may show the weakest member (or
group of members) which determines the actual value of cr. The examination of the buckling
mode may show how to change the support system to get higher cr (to get higher
performance in stability resistance). The support system can be changed by new supported
points, by moving any support to more optimal place and using knee bars. If the change of the
support system is not sufficient, the cross-sections should be changed (ex. using wider
flanges, higher depth). It is highly suggested for junior designers discussing this problem with
a senior engineer.
79
Ferenc Papp
Step 4: Execute full analysis

Calculating design forces the options of first order or second order should be switched on. For
stability analysis it is suggested asking at least 6 eigenvalues (critical load amplifiers). If the
procedure suggested above is followed and any error occurs during the execution for any load
combination, the probable reason is that cr<1 (see Step 3). The structural background of this
phenomenon may be found in the relative weakness of the structure and/or in the relative
excessive loading.
5.2.3 Design document
Relating to documentation of design there are two typical life situations. One of them is that
when more engineers (or more groups of engineers) are working on the project, and they
should use and or should control each other (for example one of them performs the main static
calculation, the other designs the joints). In this situation full design documentation should
be produced. In the other situation (mainly in case of small scale design projects where
structural designer works in single) is that where there is no requirement for direct
cooperation with other engineers. In this situation partial design documentation may be
enough for internal use. In this design project the last situation is followed.
Cooperative design work (concurrent engineering) will enforce an international
standardization for documentation requirements of static analysis and design. Today it is not
more than an effort, therefore there are no general rules for the content and style of the
document. The basic requirement is that the results of the analysis and design should be
traceable and reproducible by independent experts. This basic requirement may be satisfied by
the documentation tools of the design software. The minimum content of the partial design
documentation for a simple structure (like the frame of the present design project) may be the
following:
description of structural model
description of load model
partial results of analysis
The structural and the load model may be described by tables or draws produced by the
software applied, or by both of them. The finite element analysis provides a great amount of
data. It is enough to document some dominant results. For example the check of cross-section
resistances needs the design internal forces (NEd, MEd, VEd) in some points of the structure.
Therefore these forces may be collected into a simple table. Even a simple frame may have a
lot of design load combinations. In this case it is enough to document the forces given by the
most dominant load combination(s). There are no general method how to do this, but there are
some guidelines what load combinations can be dominant:
which gives maximum My.Ed for beams
which gives maximum My.Ed for columns, but in case of thin web (for example tapered
member) which gives maximum VEd and maximum NEd
which gives bending moment diagram closer to uniform (lower My.Ed may be
dominant if the moment distribution is uniform)
According these guidelines the design forces may be documented with content as following:
(A) Design force diagrams drawn by software for 2-3 dominant load combinations. (Figure
43 shows a pattern how to document a dominant bending moment diagram.)
80
Ferenc Papp
Fig.43 Diagram of My.Ed of Load Combination [kNm]

(B) Design force table for check of cross-section resistances. (Figure 44 shows a pattern how
to define the design force table for a simple frame.)
structural
member
point
column
K1
beam
K2
K3
K4
Load
Combination
M y .Ed
N Ed
Vz .Ed
(kNm)
(kN)
(kN)
* color yellow indicates the maximum value
Fig.44 Design force table for check of cross-section resistances
81
Ferenc Papp
The table and picture in Figure 44 show the intention of the designer:
columns will be checked at the column base and at the top of the columns; in both
cases that Load Combination is assumed being dominant which gives maximum
bending moment;
columns will be checked at the column base for shear force where that Load
Combination is assumed being dominant which gives maximum shear force;
beams will be checked at the frame corner and at the maximum moment along the
member; in both cases that Load Combination is assumed being dominant which gives
maximum bending moment;
beams will be checked at the frame corner for shear force where that Load
Combination is assumed being dominant which gives maximum shear force;
5.2.4 Application
Analysis of the main frame was executed by the ConSteel software. The detailed application
guide can be found in Annex 10. The application shown below is based on the method of
partial design documentation.
4.3 Analysis
Second order analysis was executed by ConSteel software. Diagrams of stress resultants of
two dominant load combinations are presented. Results of analysis may be controlled using
the ConSteel model attached in the Annex.
4.3.1 Diagrams of stress resultants for Load Combination 1
- Bending moment diagram - M y.Ed [kNm]
- Normal force diagram - NEd [kN]
82
Ferenc Papp
- Shear force diagram- V z.Ed [kN]
4.3.2 Diagrams for stress resultants for Load Combination 4

- Moment diagram - My.Ed [kNm]
- Normal force digram - NEd [kN]
- Shear force diagram - V z.Ed [kN]
83
Ferenc Papp
5.3 Design of cross-sections

5.3.1 General
The aim of the cross-section design is to find the optimal structural dimensions. Dimensions
of cross-sections may be optimal if they satisfy the design equations of resistances and the
structure has the less cost (in this case the less weight) as possible. Design of cross-section is
an iterative procedure, where the initial cross-sections are checked, the results are discussed,
and some initial cross-sectional parameters may be modified. Any change in cross-sectional
dimensions leads to change of the static model and the design forces as well. Theoretically,
new analysis should be executed if there is any change in the model. Practically the new
analysis may be neglected if the change is relatively inconsiderable.
It is important to consider that the resistances of members are usually determined by reduction
of the cross-sectional resistances. Therefore the cross-sections can not be designed for 100%
usage of their resistances. In case of relatively stiff frames made from hot-rolled or equivalent
welded profiles the usage of cross-sectional resistances can not be greater than 80-90%. In
case of relatively slender frames (welded and tapered frame with large span) this limit may be
70-80%.
The EC3-1-1 provides multilevel design formula to determine the resistances of crosssections. In general the resistance of a cross-section may be determined by one of the
following design formulas:
general elastic formula (based on design stresses)
conservative interaction formula (based on design forces)
plastic interaction formula (based on design forces and plastic properties)
The theoretical background of these design formulas was the subject of the previous studies
(Steel I and Steel II).
5.3.2 Choosing design formula
The appropriate design formula depends on the type of the structure. Some guidelines are
given below:
Hot-rolled or equivalent welded profiles

In the present design project HEA and IPE profiles with middle depth may be applied.
These cross-sections belong to Class 1 or 2 for pure bending and Class 2 or 3-4 for pure
compression of the web. IPE profile may be Class 3-4 if the depth is greater than 400 mm
and it is purely compressed. Frame which is composed of these types of profiles may be
usually designed by plastic interaction formula where the effect of normal and shear
forces reduce the bending moment resistance. Alternatively conservative interaction
formula can be used (see the next paragraph).
Welded cross-section with relatively thin web

Welded cross-sections with relatively thin webs and IPE profiles with depth is grater than
400 mm belong to Class 1 or 2, if the flanges are considered and Class 3 or 4, if the web is
considered for pure compression. If the normal and shear forces are relatively low, the
conservative interaction formula may be used. In this case Aeff and Wpl.y may be used
together in the same formula.
84
Ferenc Papp
Welded cross-sections with high depth and thin web

Frames with relatively large span are usually made from welded cross-sections composed
of relative slender web plate (large depth and thin plate). In this case the web of the crosssection may belong to Class 3 or 4, while the flanges belong to Class 2 or 3. This type of
cross-sections can be designed by the general elastic formula where the effect of shear
stresses is considered. The pure normal stresses can be calculated with elastic crosssectional properties of the nominal and the effective cross-section, respectively. The
effective cross-sectional properties are suggested calculating by software tool which
assumes the interaction of the normal stresses due to pure axial force and due to pure
bending moment (see the ConSteel/Section module).
5.3.3 Application
The application shown below illustrates the hand design of the cross-sections and compares
the results with those are given by the ConSteel software. Annex 11 contains the application
guide on how to use the ConSteel program designing cross-sections.
4.4 Design of cross-sections
4.4.1 Relevant cross sections and design forces
Relevant cross-sections to design
- Section K1: cross-section of column at columnbase
- Section K2: cross-section of column at frame corner
- Section K3: cross-section of beam at frame corner
- Section K4: cross-section of beam at maximum positive bending moment
Design forces of relevant cross-sections, see 4.3.1-2.

member cross-section LC
design forces
moment [kNm]
normal force [kN]
column
K1
4
M K1.y.Ed 354.36 N K1.Ed 179.66
K2
4
M K2.y.Ed 491.76 N K2.Ed 170.99
shear force [kN]

V K1.Ed 116.73
V K2.Ed
114.25
beam
K3
M K3.y.Ed
469.56
N K3.Ed
142.20
V K3.Ed
148.56
K4
M K4.y.Ed
177.91
N K4.Ed
115.75
V K4.Ed
4.18
85
Ferenc Papp
4.4.2 Design of cross-secctions
Checking shear force effect
- maximum design shear force [kN]
- minimum design shear area [mm 2]
V max.Ed
V K3.Ed
A min.V
h bw . t bw
V max.Ed = 148.56
3
A min.V = 2.208. 10
f y0 1
.
A min.V.
1000
3
V min.Rd = 299.576
Maximum design shear force does not excced 50% of shear resistance of web anywhere,
Vmax.Ed<0.5Vmin.Rd ,
- pure shear resistance [kN]
V min.Rd
therefore effect of shear force may be neglected in any case.

Beam and column cross-sections are related to Class 4 for pure compression and Class 1 for
pure bending. Therefore conservative interaction formula is used. Effective cross-sectional area
is used for pure compression and plastic moduli for pure bending.
Dimensional factors (change kN to N and kNm to Nmm)
N 1000
M 1000000
Column Section K1
N .N K1.Ed
K1
A c.eff .f y0
M . M K1.y.Ed
W c.y.pl .f y0
K1 = 0.729
Adequate!
Column Section K2
N .N K2.Ed
K2
A c.eff .f y0
M . M K2.y.Ed
W c.y.pl .f y0
K2 = 0.98
Adequate, but too much!
Beam Section K3
N .N K3.Ed
K3
A bh.eff .f y0
M . M K3.y.Ed
W bh.y.pl .f y0
K3 = 0.595
Adequate!
Beam Section K4
N .N K4.Ed
K4
A b.eff .f y0
M . M K4.y.Ed
W b.y.pl . f y0
K4 = 0.504
Adequate!
Usage of resistance of the cross-section at top of the column excedes 90% as a practical
limit. Instead of reinforcment of column cross-section design bending moment is taken at the
buttom flange of the haunch beam, where thecolumn has realistic cross-section.
- decreasing of bending moment
h b M K2.y.Ed M K1.y.Ed
.
2
1000. H c
- reduced design bending moment
M
M = 61.431
h bh
M K2.y.Ed.red
M K2.y.Ed
M K2.y.Ed.red = 430.329
- Column Section K2
K2
N .N K2.Ed
A c.eff .f y0
M . M K2.y.Ed.red
W c.y.pl . f y0
The initial cross-sections are adequate!
86
K2 = 0.866
Adequate!
Ferenc Papp
Cross-section checking was executed by the ConSteel software too (see the picture
berlow). Results are basically the same than those of the hand calculation.
4.4.3 Applied cross-sections

Column section
flange
web
Beam section
flange
web
Haunch
flange
web
width [mm]
thickness [mm]
width [mm]
thickness [ mm]
240
16
368
6
width [mm]
thickness [mm]
width [mm]
thickness [mm ]
240
16
468
8
width [mm]
thickness [m m]
width [mm]
thickness [mm]
240
16
300
6
87
Ferenc Papp
Annex 6
Structural modeling with ConSteel software
(application guide)
A6.1 Installation
The install file of the ConSteel design software can be downloaded from the www.consteel.hu
portal. The software shouldbe be installed on the Users computer. First the
ConSteel_install_dtum.exe file should be executed when the ConSteelKey0.bin file will be
created by the software. This file and the required data should be sent to the ConSteel
Software Company (info@consteel.hu). In some days the Company will send back the
ConSteelKey.bin software key which can be added to the main folder of the software.
A6.2 Starting
The program can start by the ConSteel.exe. On the ConSteel Startup panel the Create new
model option may be selected, and then the Model name (1) can be defined, see Figure
A6.1.
Fig. A6.1 Create new model

Using OK the 3D model window will appear. The short description of the main modeling
tools can be available by F1.
A6.3 Setting 3D modeling window
First dimensions of the modeling raster may be defined (2). For example, at 19,5 meters span
20000 may be written into the Size box (3), see Figure A6.2.
Fig.A6.2 Setting modeling raster

88
Ferenc Papp
To actualize the setting Enter should be used (or the panel should be closed). By this way the
size of raster will be 20 meters (while the density will remain 1000 mm and the step 250 mm).
A6.4 Setting initial cross-sections
The Section table of the new model is empty, therefore the first step is to define the initial
cross-sections. First the Structural members label (4) should be selected, secondly the
Section administration tool (5). In case of hot-rolled cross-section From Library option
(6), in case of welded cross-sections Macro section option (7) should be used, see Figure
A6.3.
5
6
Fig.A6.3 Selecting cross-section category
After selecting cross-section category the appropriate parameter panel will appear where the
type of cross-section (8) should be defined (actually it is Welded I). Clicking on Next the
parameter table appears where the geometrical parameters of the cross-section should be
defined (9), see Figure A6.4.
10
11
8
Fig.A6.4 Setting type and geometrical parameters for cross-section
Name(10) and Material (11) of the cross-section may be defined on the top part of the
panel. Using Create the cross-section will be placed into the Section table of the actual
model, see Figure A6.5.
89
Ferenc Papp
12
Fig.A6.5 Section table with actually defined cross-sections

All the cross-sectional properties may be available if the cross-section is selected in the
Section table (12) and the Properties is used, see Figure A6.6. The program uses two
different cross-sectional models in the same time. The General Solid Section model (13)
supplies only elastic properties, while the Elastic Plate Segment model (14) supplies plastic
properties as well.
13
14
Fig.A6.6 Cross-sectional properties

A6.5 Create columns
The first step of the structural modeling is the erection of the two columns. Selecting the
Structural member label (15) and the Column tool (16) the column definition panel will
appear, where the height (17) and the cross-section (18) of the column should be determined
(normally the default values of the other parameters are adequate). The columns may be
placed in the global X-Z plane with the base points, see Figure A6.7. The position of a
column can be defined by coordinates also. For this the column should be positioned such a
way that Y=Z=0, and then the X button should be pushed, and the actual X coordinate should
be written into the data box which can be found below the modeling window (19), see Figure
A6.8. The action is closed by Enter, and the column will appear in the graphics. The form
and the view of the visualization can be changed by the tools (20) found at left hand side of
the window, see Figure A6.9.
90
Ferenc Papp
15
16
17
18
Fig.A6.7 Erecting columns
19
M6.8 bra: Column position defined by coordinates
20
Fig.A6.9 Change visualization

A6.6 Create beam
The next step of the modeling is the erection of the beam. Selecting the Structural member
label and the Beam tool (21) the panel of beam definition will appear, where the cross-section
(22) should be selected (normally the default values of the other parameters are adequate).
Using the left hand side mouse button the beginning and the end of the beam should be
defined (in the example these points are the tops of the columns), see Figure A6.10. In order
91
Ferenc Papp
to get the symmetric duopitch roof the beam should be refracted at middle. For this the raster
may be moved into the plane of the frame (23 , 24), see Figure A6.11, and then shifting to the
Geometry label (25) and using the Refract selecting line element tool (26) the middle (ridge)
point may be moved into the right position. The Z coordinate of the ridge point can be defined
also in the control row (27) which is found below the modeling window, see Figure A6.12.
21
22
Fig.A6.10 Erecting beam
24
23
Fig.A6.11 Shift the raster into the plane of the frame

92
Ferenc Papp
25
26
27
Fig. A6.12 Refract the beam
After modeling the columns and the beam it is suggested shifting the raster into the original
position (28) and saving the model (29) with version number _01, see Figure A6.13.
29
28
Fig.A6.13 Shift the raster into original position and save the model
A6.7 Create haunch
Selecting the Structural member label and the Haunch tool (30) the haunch panel will
appear, where the geometrical parameters should be defined. Following the tips of the
program the beginning point (31) and the direction (32) of the haunch should be defined on
93
Ferenc Papp
the actual structural member (for example on the beam). The direction can be defined by any
get point on the member, see Figure A6.14. New haunch can be created by new definition of
beginning point and direction.
30
32
31
Fig.A6.14 Create haunch

M6.8 Create point support
First the support model of the column bases is defined. According to the preliminary draw the
column bases are rigid. Selecting the Structural member label and the Point support tool
(33) the point support panel will appear, where the Fix option should be selected (34). If the
column base is pinned, the option should be x,y,z,zz. Following the tips of the panel the base
points of the model should be selected by the mouse (35), see Figure A6.15.
33
34
35
35
Fig.A6.15 Create fix (pinned) column base supporting model
94
Ferenc Papp
In the next step the intermediate supports in global direction Y are created. These supports are
placed at the purlins and wall beams or at the bracing bars. These places are determined in the
preliminary draw. Point support may be set at any get point which is defined on the structural
member. There are two ways to define get points:
Dividing member into number n uniform parts.
Dividing member into parts with d uniform length.
The required option can be selected in the function row found below the model window, see
Figure A6.16. The option can be changed by a click on the actual option sign. The new set of
these options (d or n) can be fixed by Enter.
Fig.A6.16 Create get points on member

(n: number of parts; d: relative distance in [mm])
Using setting above the get points with red color will appear on the structural member when
the cursor is moved close to one of the ends of it. On the point support panel the y option
(support in global direction Y) should be selected. The support model can be placed at the get
points by a click on it. In the case of the example n=2 get point system was used. The lateral
supports were placed into the breakpoints and into the middspan points of the members, see
Figure A6.17. This support model approximates the real construction which is established in
the preliminary draw.
Fig.A6.17 Structural model with lateral point supports

(at middspan with knee bar effect, see Section A6.9)
A6.9 Create lateral support with knee bar effect
The critical load amplifier (see Section 7.3.2) may be increased by knee bars which can be
used along the structural members. In the example this type of support is used at middspan
points of the beams. Selecting the Structural member label and the Point support tool the
point support panel will appear, where the New support button should be used (36). By this
operation the Definition point support panel will appear, where the New button should be
used. After this the automatically generated Name of support model may be changed (37)
and the y and Rx degrees of freedoms should be switched as Fix and the others as Free
95
Ferenc Papp
(38). This means that the point will be supported in direction y and in rotation about x axis of
the local system of the structural member, see Figure A6.18. Using the Apply button the
support model will be created and it can be placed on the points of the structural members.
Before placing the support the coordinate system should be shifted from global to local
(39), see Figure A6.19. It is noted that the transversal pin of the support symbol should be
perpendicular to the axis of the structural member (40).
36
37
38
Fig.A6.18 Create lateral point support with knee bar effect
40
39
Fig.A6.19 Place lateral support with knee bar effect
96
Ferenc Papp
Annex 7
External pressure coefficient for cross wind effect
(h/d<0,25; =50 ; A>10m2 )
Case 1
cpe,10=-1,7;-1,2
cpe,10=-0,6
cpe,10=+0,2
cpe,10=-0,6
cpe,10=-0,3
cpe,10=+0,8
Case 2
cpe,10=-1,7;-1,2
cpe,10=-0,6
cpe,10=-0,6
cpe,10=-0,6
cpe,10=-0,3
cpe,10=+0,8
Case 3
cpe,10=+0,2
cpe,10=-0,6
cpe,10=-0,3
cpe,10=+0,8
Case 4
cpe,10=-0,6
cpe,10=-0,6
cpe,10=-0,3
cpe,10=+0,8
Megjegyzs: A fenti rtkeket a ( c q p ) vel megszorozva megkapjuk a szlhatsbl szrmaz vonal

menti megoszl terheket.
97
Ferenc Papp
Annex 8
(a) Internal pressure coefficient for cross wind effect
(h/d<0,25 ; =50 ; A>10m2 )
+cpi,0
(b) External pressure coefficient for longitudinal wind effect

(h/d<0,25 ; =50 ; A>10m2 )
cpe,10,H=-0,7
cpe,10,H=-0,7
cpe,10,B=-0,8
cpe,10,B=-0,8
+cpi
(c) Internal pressure coefficient for longitudinal wind effect

(h/d<0,25 ; =50 ; A>10m2 )
+cpi,90
Notes: Coefficients given above times ( c q p ) provide the wind loads distributed on the structural
members of the frame.
98
Ferenc Papp
Annex 9
Load modeling with ConSteel software
(application guide)
A9.1 Load groups and load cases
The structural model defined in Annex 6 is considered. Selecting the Load label (1) and the
Load cases and groups option (2) the Load definition panel appears, see Figure A9.1. (Load
case: group of loads where the loads act together and in same time; Load group: group of load
cases where the load cases may enclose each other in any load combination.)
1
4
Fig.A9.1 Set and rename of a load group

6
Fig.A9.2 Set new load case
99
Ferenc Papp
In case of new design model the left hand side table contains one load group and one load
case generated by the program. First the name of the default load group should be changed to
the actual name (presently Dead loads) (3), and then using Apply the name of the default
load case can be changed to the actual name (presently weights) (4). If the are more load
cases in the actual load group, the New load case (5) and the Name option (6) can be used to
define a new one, see Figure A9.2. New load group can be defined by New (7). Before this
action the load type should be selected (8). For example the name of the new load group is
Snow loads (9) and the type of the load group is Meteorological, see Figure A9.3.
9
7
8
Fig.A9.3 Set new load group

According to the actual load model (defined previously) all of the load groups and load cases
should be defined, see Figure A9.4.
Fig.A9.4 Completed structure for the actual load model
100
Ferenc Papp
Finally, the theoretical weight of the structure is generated by the program, and it should be
added to one of the load cases. To do this the Load case including self weight box found at
bottom of the panel should be opened and the appropriate load case can be selected (10), see
Figure A9.5. Closing the load definition panel by OK the structure of the load model is
determined. In the next step the load cases should be fulfilled with concrete design loads.
10
Fig.A9.5 Select load case which will include the self weight of the structure
A9.2 Set of design loads
The structure of the load model was set in Section A9.1. Here it will be shown how to set the
design loads of the load cases. First the actual load case should be selected (11). Loads set on
the model will be contained by selected actual load case as far as the actual load case is
changed to another one, see Figure A9.6.
11
Fig.A9.6 Select load case to define design loads

The paragraphs below describe how to set concrete loads for a load case:
Distributed load on structural member - in global directions
Selecting the Loads label (12) and the Line load option (13) the Load parameter panel
appears, where the starting and end intensities of the line load should be written into the
appropriate boxes (14). The line load can be set on a structural member by a click (15), see
Figure A9.7.
Distributed load on a part of the structural member - in global directions
Selecting the Load label (16) and the Line load option (17) the Load parameter panel
appears, where the Draw option (18) should be selected, see Figure A9.8. The starting and
end load intensities should be written into the appropriate boxes (19) and the starting (20) and
end (21) points of the load should be selected on the loaded member. The points can be
selected by get points setting on the member, see Section A6.8.
101
Ferenc Papp
12
13
14
15
Fig.A9.7 Set line distributed load on structural member in global directions

16
17
18
19
20
21
Fig.A9.8 Set line load distributed on a part of the member
102
Ferenc Papp
Set loads in local system of the member

Basically the loads are assumed acting in the [X,Y,Z] global directions. There are loads (for
example wind load) which are perpendicular to the structural member. These loads may be set
in the local system of the member. The local system can be chosen by the Local coordinate
system option (22), see Figure A9.9. In the local system direction X coincides with the axis
of the member, directions Y and Z are the main axes of the cross-section.
22
Fig.A9.9 Set local system for loads

M9.3 Set load combinations
The structure of the load model and the design loads of the load cases are given. The next step
is to define the design load combinations. Selecting the Loads label (23) and the Load
combination option (24) the Load combinations panel appears where the Automatic
generation of load combinations option (25) should be selected. The relevant design
situation(s) should be chosen on the appeared panel. For example to check the cross-sections
the Load combinations in persistent and transient design situations (6.10) option (26)
should be selected, see Figure A9.10.
23
24
25
26
Fig.A9.10 Automatic generation of the design load combinations
103
Ferenc Papp
The panel should be closed by Apply, and then all the possible load combinations will appear
in the table, see Figure A9.11. Any load combination might be cleared (27), and new
combinations may be defined (28). The factors of the combinations might be changed (29).
28
27
29
Fig.A9.11 Automatically generated design load combinations and tools for modification
104
Ferenc Papp
Annex 10
Analysis with ConSteel software
(application guide)
M10.1 Check structural model

Before analysis for all the design load combinations the design model should be controlled.
Selecting the Analysis label (1) and the Analysis parameters option (2) the Set analysis
parameters panel appears, where only the First order analysis option (3) should be switched
on, see Figure A10.1.
1
2
Fig.A10.1 Set analysis

Shifting to the Load combination label (4) and switching all of the load combinations off (5),
and selecting the Load cases label (6), where one of the load cases which contains
symmetrical effects should be switched on, see Figure A10.2. Using Apply and Calculation
the deformed model will appear on the graphics. The method how to discuss the results of the
4
5
Fig.A10.2 Selecting load case with symmetrical effects to control the structural model
105
Ferenc Papp
analysis is described in the Paragraph A10.2. If any problem is realized in the response of
the structural model, a senior engineer (presently the supervisor) should be involved into
repairing the model.
M10.2 Discussion of model response
When the analysis is performed the deformed geometry appears in the graphics. The scale of
the deformation may be changed by the Scale roller (7), see Figure A10.3.
8
10
11
Fig.A10.3 Scaling deformed geometry

The content of the graphics can be defined by the following selections:
Type of analysis (first order, second order, buckling, ), (8)
Load case or Load combination (9)
Category of response (deformation, design forces, reactions, ), (10)
Type of draw (diagram, color, ), (11)
Using appropriate content of graphics the deformed geometry should be examined. The
guidelines for examination may be the follows:
- if the load is symmetric, the deformed geometry should be symmetric too,
- if the load acts in the plane of the frame, the deformed model should be remained in-plane,
- maximum deflection should be realistic,
- deformed geometry should be realistic from engineering point of view.
As the next step, the in-plane My.Ed bending moment diagram (12) may be examined, see
Figure A10.4. The guidelines for examination may be the follows:
- at pinned column base the bending moment should be zero,
- break/jump in bending moment diagram may occur only at concentrated load/moment,
- maximum bending moment should be realistic,
- shape of moment diagram should be realistic from engineering point of view.
106
Ferenc Papp
12
Fig.A10.4 Select content for graphics

M10.3 Execute analysis
If the model control is negative, the complex analysis can be executed. Switching all the load
cases off and all the load combinations on, see Paragraph A10.1-2, and then the first order
analysis can be executed. If there is no error message, the second order analysis (13) may be
switched on and the analysis may be executed again, see Figure A10.5.
13
Fig.A10.5 Set option of second order analysis

If there is error message, it can be concluded that the structure is too weak and/or the load is
too high for the structure. Assuming that the load model is adequate (there are no duplicated
loads which cover each other on the graphics, and so on), and the support system is also
adequate (all the supports established which are located on the structural model in the
preliminary draw) the structural should be reinforced. If there is no error message, the
analysis can be considered being completed, and the design forces can be documented.
M10.4 Document design forces
Comprehensive description of the document tools and making full static document are not the
objectives of this teaching material. The aim of this paragraph is to shown how to create a
picture with specified content in order to copy it into another (Word or Mathcad) document.
In the first step the dominant load combinations should be selected from the Table of crosssectional resistances. The table can be available selecting the Global check label (14) and the
Global resistance option (15). Cross-section check (16) and Second order analysis (17)
options should be switched on, see Figure A10.6. The cross-sectional check will be executed
by Calculation according to the setting. The results are shown by color graphics and in tables
as well, see Figure A10.7. In the table the utilizations of resistances are shown in the last
107
Ferenc Papp
column (18) and the appropriate load combinations in the fourth column (19), see Figure
A10.7. The most dominant load combinations can be selected from the table. In this design
project it is satisfactory to document the design force diagrams of the most dominant two or
three load combinations. More details how to check of cross-sectional resistances can be
found in Annex 11.
14
15
16
17
Fig.A10.6 Setting for examination of cross-sectional resistances

19
18
Fig.A10.7 Table of the utilization of the cross-sectional resistances

M10.5 Document pictures
In this paragraph it is shown how to document a picture. For example the My,Ed bending
moment diagram which belongs to the Load Combination 4 and which was computed by 2nd
order analysis will be documented. The diagram can be scaled by the Grapics zoom (20). Any
ordinate of the diagram can be visualized by a click on the appropriate point of the diagram
using right hand side button of mouse. Selecting the Marker option (21) the value of the
diagram will be fixed to the diagram (22), see Figure A10.8. The required number of markers
can be fixed to the diagram, see Figure A10.9.
108
Ferenc Papp
20
22
21
Fig.A10.8 Visualization values for the diagram
Fig.A10.9 Bending moment diagram with appropriate number of values

The diagram can be saved by the following way. Selecting the Document label (23) and the
Create snapshot from current state option (24) the Create snapshot panel appears where
the Name of the picture (25) and the Dimension of picture (26) can be set. The size of the text
on the picture may be changed by the Letter size option (27), see Figure A10.10. The diagram
should be completed and placed into the frame draw by dashed lines, and the picture can be
saved by Create (or Modify) into the picture folder. Selecting the Handling of figures option
(28) the content of picture folder appears, where the actual picture may be selected (29) and
using the Save selected picture into file option (30) the picture can be saved as a file into the
selected folder (31) with specified file name (32) and format (33), see Figure A10.11.
109
Ferenc Papp
23
24
25
26
27
Fig.A10.10 Creating picture which can be saved into the picture map
28
30
29
31
32
33
Fig.A10.11 Selecting picture from picture folder and saving it as a file
110
Ferenc Papp
Annex 11
Design of cross-sections by ConSteel software
(application guide)
Paragraph A10.4 presents how to Check cross-sectional resistances by the ConSteel
software. Utilization of resistances is shown by color graphics and in table, see Figure A11.1.
Fig.A11.1 Visualization of cross-sectional resistances by graphics and in table

ConSteel software allows analyzing the results more deeply. Selecting the actual cross-section
by right hand side mouse button on the graphics or in the table (1), and choosing the
Calculate section option (2), see Figure A11.2, the window of Section module appears, see
Figure A11.3.
1
2
Fig.A11.2 Selection of actual cross-section for examination in deep
111
Ferenc Papp
Fig.A11.3 Detailed information about the check of cross-sectional resistance

It is known that Eurocode 3 defines different design equations for determining cross-sectional
resistances. These equations may be the follows:
General elastic resistance theoretically this formula is allowed using for all the crosssections, but design may be uneconomical for Class 1 and 2;
Pure resistances pure resistance formulas should be used in case of interaction of forces
too;
Plastic interaction resistance this formula is valid for Class 1 and 2 cross-sections; the
base of the formula is the bending resistance which should be reduced by the effects of the
normal force and the shear force, if it is necessary (this formula ensures economic design);
Conservative interaction resistance this formula my be valid for all the classes providing
the shear force can be neglected; in the parts of the formula related to pure compression and
pure bending the cross-sectional properties given for the appropriate pure effects may be used
(ex. if the cross-section is related to Class 4 for pure compression and Class 2 for pure
bending, Aeff and Wpl might be used in the appropriate parts of the formula).
ConSteel software selects the adequate formula, but it may be controlled by the engineer.
According to Figure A11.4 the plastic interaction formula is the valid (3) for the actual case.
The table shows all the design parameters of the formula. For example it can be seen that both
of the normal force (4) and shear force (5) effects are allowed neglecting in the given
situation. The decision of the software (what design formula is dominant) may be changed by
the engineer. For example the designer may consider the conservative interaction formula as
the base of the design. Figure 11.5 shows the table of parameters (6) for this case. Utilization
of cross-sectional resistances computed by a selected formula can be visualized for the whole
model too. For this the Box of categories (7) should be opened and the actual category (ex.
Conservative interaction resistance (8)) may be selected, see Figure11.6. This selection will
be valid for all the cross-sections of the model. ConSteel software determines a unified
effective cross-sectional model for Class 4 cross-sections. The considered normal stress
distribution is the sum of the normal stresses due to the axial force and the bending moment.
Therefore the result may differ from the result of the hand calculation where there are
different effective cross-sectional models for pure compression and for pure bending moment.
112
Ferenc Papp
Fig.11.4 Parameters of plastic interaction resistance formula

6
Fig.11.5 Parameters of conservative interaction resistance formula

7
8
Fig.A11.6 Utilization of cross-sectional resistances of the whole model

based on selected design formula
(actually the conservative interaction formula is selected)
113
Ferenc Papp
Practice 7
GLOBAL STABILITY RESISTANCE OF
THE FRAME
2010-2011 Budapest
114
Ferenc Papp
7.1 General
Global stability resistance of the structural members of the frame can be performed by elastic
design method using one of the following methods:
Method of reduction factor
In this method in-plane or spatial static model is applied, where the phenomena of lateral torsional
buckling and flexural torsional buckling is not examined. The design model does not include the
equivalent imperfections but may include real geometric imperfections if their effects are
considerable. The structural members or the parts of them can be examined (simple method).
Alternatively the members or the frame may be examined comprehensively (general method).
Determination of the buckling lengths or the critical forces should be based on the behaviour of
the structural. Annex 12 illustrates the application of the method.
Method of equivalent geometric imperfection
In this method spatial static model should be applied, where the design model includes the effect
of torsion (warping). The design model includes the equivalent geometric imperfections. The
cross-sectional resistances should be computed on design forces given by second order analysis.
The cross-sectional resistances include the effect of the global instability. Annex 13 illustrates the
application of the method.
Method of partial equivalent geometric imperfection
In this method in-plane or spatial static model is applied, where the phenomena of the lateral
torsional buckling and the flexural torsional buckling are not examined. The design model
includes the global equivalent geometric imperfection but neglects the local. The structural
members or the parts of them are examined by the method of reduction factor. As buckling lengths
The buckling lengths of the members may be equal to the structural lengths. Annex 14 illustrates
the application of the method.
The expressions used above are defined as follows:

Elastic design method
Design forces are computed by linear elastic analysis (where elastic material law is applied). The method
allows using plastic cross-sectional properties in the design formulas.
Design model
Design model consists of a structural model and a load model. The structural model is the virtual model of
the real structure. The load model is the virtual model of the specified loads and effects of the structure.
Static model
Static model is generated from the design model following the conditions of the method of analysis. Static
model may determine the type of the model response, such as in-plane, out-of-plane or spatial behaviour.
Equivalent geometric imperfections
Equivalent geometric imperfection is local if the structural member has a specified initial out-ofstraightness, and it is global if the structural member has a specified initial inclination. The effect of the
equivalent imperfections is equal to the effect of the reduction factor used in the method of reduction factor.
Second order analysis
Second order analysis takes the effect of the structural deformation in the equilibrium equations into
consideration. Basically the idea of small displacements is used. The analysis leads to non-linear problem
from mathematical point of view.
Global instability
The stability behaviour of the structure (or structural member) is global if the whole structure or a part of it
(ex. one of the bars of a truss) takes a global buckling mode, such as flexural, flexural-torsional or lateral
torsional buckling, or interaction of them. The global indicates that the buckling mode does not include
local buckling phenomena, for example local plate buckling.
115
Ferenc Papp
In the design practice mostly the method of reduction factor is used. The method of equivalent
geometric imperfection is suggested using for senior engineers who have the experience. The
method of partial equivalent geometric imperfection is normally used in preliminary design.
Within the method of reduction factor the following two methods can be used:
Method of equivalent members (simple method)

The design forces are computed on in-plane or spatial design model, where the instability
responses are neglected. The stability resistance formulas (ex. interaction formulae) are
applied on equivalent structural members isolated from the design model and supported
appropriately. The details of this method are discussed in the Paragraph 7.2.
General method
The calculation of the design forces and the elastic stability analysis are performed on
spatial static model which includes the effect of torsion (warping). Any software can be
applied which uses general beam-column finite element method (14 DOF FE method).
The support system of the design model should be based on the real conditions of the
structure (see Paragraph 5.1.2.5). The general design formulae examines the structure as
a super member. The details of this method are discussed in the Paragraph 7.3.
7.2 Method of equivalent members (simple method)

7.2.1 Interaction design formulae
The stability resistance of uniform members with double symmetric cross sections should be
checked by the interaction design formulae, where a distinction is made for:
members that are not susceptible to torsional deformations, e.g. circular hollow sections or
sections restraint from torsion;
members that are susceptible to torsional deformations, e.g. members with open crosssections and not restraint from torsion.
The interaction formulae is based on the modelling of simply supported single span members
with end fork conditions and with or without continuous lateral restraints, which are subjected
to compression forces and end moment in-plane:
(1)
N Ed
+ k yy
y A fy
M1
(2)
M1
N Ed
+ k zy
z A f y
M1
where
NEd
My,Ed
y , z , LT
M y ,Ed
1
Wy f y
LT
M y ,Ed
1
Wy f y
LT
M1
is the design compression force;

is the maximum design moment about y-y axis (including the moment due to
the shift of the centroidal axis of Class 4 cross-sections);
are the reduction factors due to flexural buckling about y-y and z-z axes, and
due to lateral torsional buckling;
116
Ferenc Papp
kyy , kzy
A , Wy
fy
M1
are the interaction factors;

are the cross-sectional properties due to the class of the cross-section (plastic,
elastic or effective);
is the characteristic design strength;
is the partial safety factor.
The kyy and kzy interaction factors have been derived from two alternative approaches. Method
1 was developed by the so called French-Belgian group. The method provides a continuous
design curve between the cross-sectional and the stability resistances but the expressions are
basically complicated and no understandable. Method 2 was developed by the so called
German-Austrian group. The method provides simple and understandable expressions but
the results are less sophisticated. Method 2 is used for simple hand design, while Method 1
may be used by software tools. The expressions of Method 2 can be found in the Annex 15.
It is easy to realise that no one of the beams or columns of the frame satisfies the conditions
of the interaction formulae. For example, if the column of the frame is taken as an equivalent
member with simple fork supports at the ends, at least two basic conditions is not satisfied:
(i) the column is elastically supported at top;
(ii) the column is supported intermediately by wall beams (or bracing members).
The problem may be solved by the wider meaning of the interaction formulae. The formulae
consists of three pure buckling modes, such as the flexural buckling about y-y axis, the
flexural buckling about z-z axis, and the lateral torsional buckling. It is allowed to take
different equivalent members for these pure buckling modes. The engineer is responsible for
the use of the interaction formulae in wide meaning. Determination of the y , z and LT
reductions factors using the wide meaning of the interaction formulae is illustrated in the
following paragraphs.
7.2.2 In-plane buckling (y)
7.2.2.1 Buckling of columns
The in-plane slenderness of the column may be determined by the stability analysis of the
frame. The slenderness can be calculated using buckling length factor or critical force:
-
Slenderness due to buckling length factor

The buckling length factor of the columns of any simple frame may be found in the
literature, see Figure 45. The duopitch roof structure may be modelled by the simple
portal frame model shown in the figure.
Ig
Io; Ao
csukls
keret:
Pinned
frame:
y = 4 + 1.4 (c + 6) + 0.02 (c + 6 ) 2
befogott
keret:frame:
Fix
(clamped)
y = 1 + 0.35 (c + 6) 0.017 (c + 6) 2
I L
c= o
10
Ig H
4I
= 2 o 0.2
L A
N cr . y
2 EI y
=
( y H )2
y =
A fy
N cr . y
Fig.45 Slenderness due to buckling length factor

117
Ferenc Papp
Slenderness due to critical force

The slenderness of the column may be determined by global stability analysis of the
appropriate model. The numerical procedure is illustrated in the Figure 46.
(a) design model
(b) analysis (design forces)
N Ed
(c) analysis (buckling)
(d) slenderness
N cr = cr N Ed ,max
y =
cr
A fy
N cr
Fig.46 Slenderness due to critical force

7.2.3 Out-of-plane buckling (z)
The out-of-plane stability responses of the columns and the beams are similar. The members
are supported laterally by purlins or wall beams, optionally by bracing members. Out-of plane
flexural buckling mode may develop between two neighbouring lateral supports in form of
half sinus curve. The equivalent members (buckling lengths) are the parts of the examined
member located between two lateral supports. It is more complicated problem if the lateral
supports have considerable eccentricity (for example in case of high web depth with lateral
support at the tensioned flange of the I section). In this situation the out-of-plane flexural
buckling mode and the lateral torsional buckling mode can not be independent phenomena
(see Paragraph 7.2.4). In the lack of more exact stability analysis method (such as general
method), the equivalent member of the LTB problem may be used for the flexural buckling
problem. This means that the appropriate buckling length may be longer than the distance
between two lateral supports. This conservative design method is for the safe. The examples
of Paragraphs 7.2.5.1 and 7.2.5.2 illustrate the design method.
7.2.4 Lateral torsional buckling (LT)
Lateral torsional buckling mode may develop between two neighbouring lateral supports
where the cross-sections are restrained in rotation around the axis of the member (fork
support). Fork support can be taken into consideration in two cases:
- the lateral support is placed at the compressed flange, see Figure 47;
- knee bars support the compressed flange, see Figure 34.
118
Ferenc Papp
(a)
(b)
Fig.47 Equivalent member for the beam at the frame corner: the second lateral support can be
considered as fork support since the support is placed at the compressed flange;
(a) structural model with lateral supports
(b) equivalent member for the beam
7.2.5 How to determine the equivalent members for columns (buckling lengths)
Figure 48 shows the column structure which has pinned supports at the ends. The column is
supported laterally at the middle cross-section and loaded by vertical concentrated force and
concentrated moment at top. Lets determine the equivalent members for the global stability
check of the column with different cross-sections.
7.2.5.1 HEA300 section
The lateral support at the middle cross-section of the column may be considered as fork
support due to the following reasons: (i) the torsional stiffness of the HEA300 profile is
relatively high; (ii) the eccentricity of the lateral support is relatively low.
Structural and load model
Analysis
Equivalent members
Fig.48 Equivalent members of the column made from HEA300

(example for fork lateral support)
119
Checking
Ferenc Papp
However, the column is not sensitive to the flexural torsional and the lateral-torsional
buckling. In this case the global stability of the column may be examined by two equivalent
members: (1) part of the column between the top and the middle supports; (2) part of the
column between the middle of the column and the column base. It is noted, that in this example
the structural models of the two equivalent members coincide and the normal force is uniform,
but the moments are higher on the upper part. Consequently it is enough to examine the upper
equivalent member, see Figure 48..
7.2.5.2 IPE600 section

The lateral support at the middle cross-section of the column can not be considered as fork
support due to the following reasons: (i) the torsional stiffness of the IPE600 profile is
relatively low; (ii) the eccentricity of the lateral support is relatively high. Consequently the
column is sensitive to the flexural torsional and the lateral-torsional buckling modes. In this
case the global stability resistance of the column may be examined on the equivalent member
as the whole member (the equivalent member is the member itself), see Figure 49. It is noted
that a knee bar construction at the middle of the column may lead to previous example, see the
Paragraph 7.5.2.1.
Structural and load model
Analysis
Equivalent member
Checking
Fig.49 Equivalent member of the column made form IPE600

(example for no fork lateral support)
7.2.5.3 Frames
The columns and beams of the frame structure may have more equivalent members. Figure
50 shows a frame model where the lateral supports are fork supports due to knee bars
construction. Lets determine the equivalent members.
The parts of the structural members denoted by yellow colour are the equivalent members for
the global stability check of the frame. The solution may be explained as follows:
- rotation of cross-sections at the lateral supports are restrained due to the knee bars
constructions;
120
Ferenc Papp
Design model
Analysis
My
Equivalent members
Global stability check
G2
G1
O1
Fork lateral supports
Fig.50 Equivalent members for global stability check of the frame supported by fork lateral
supports
In-plane buckling
N y .cr = cr N Ed vagy N y .cr =
y vagy cr
y =
A fy
N y .cr
2 EI y
( y H )2
Out-of-Plane buckling and LTB
N z .cr =
O1
z =
2 EI z
M cr = C1
(Lz .cr )2
A fy
z
N z .cr
LT =
2 EI z
(Lz .cr )2
I
L
GI t
w + z .cr2
EI z
Iz
2
Wy f y
LT
M cr
Fig.51 Reduction factors for the examination of the O1 equivalent member
121
Ferenc Papp
- maximum design bending moment of the columns is found on the O1 equivalent

member;
- maximum design bending moment of the beams is found on the G1 equivalent member;
- design bending moment diagram closed to constant is found on the G2 equivalent member.
It is noted that in case of fix column bases the bottom part of the column (O2) should be
examined too. Figure 51 illustrates the main steps of the examination of the O1 equivalent
member.
7.2.6 Members with changing cross-section

Structural members with changing cross-sections may not be checked correctly by the
interaction formula and the equivalent members. Approximate check may be performed
following the design idea described below. The cross-section of the member is changing due
to the following structural reasons:
Short haunched members
Long haunched members
Tapered members
7.2.6.1 Short haunched members
The effect of the short haunch of beam to the critical forces can be neglected, see Figure 52b.
Assuming that the increasing of the stiffened part of the beam is relatively higher than the
increasing of the design moment, the interaction design formula may be evaluated at the end
of the haunch, see Figure 52a.
(a)
(b)
Adequate cross-section to check the global stability
Fig.52 Global stability check of haunched beam

7.2.6.2 Long haunched and tapered members
The effect of the long haunch should be taken into consideration in the global stability
analysis. As a draft approximation the global stability analysis may be performed with an
equivalent cross-section as following:
if the length of the haunch is equal to about the length of the examined equivalent
member, the depth of the equivalent cross-section is equal to the mean depth, see
Figure 53a;
if the length of the haunch is definitely shorter than the length of the equivalent
member, the depth of the equivalent cross-section may be equal to the cross-section at
the 1/3 haunch depth, see Figure 53.b.
The intermediate flange of the equivalent I cross-section may be neglected. The design
interaction formula should be performed at the real cross-section which has the maximum
utilization for cross-sectional resistance. The tapered members may be examined by the
concept of mean depth of cross-section.
122
Ferenc Papp
(b)
(a)
cross-section at 1/3 depth haunch

mean depth cross-section
Fig.53 Equivalent cross-section for global stability analysis of the haunched equivalent
member: (a) mean depth concept; (b) 1/3 haunch depth concept
7.2.7 Application
The example below shows the global stability examination of the frame using the design interaction
formulae and the equivalent member concept. This examination may be preformed by the ConSteel
software too. The application guide can be found in the Annex 16.
4.5 Global stabilty resistance
The global stability resistance of the frame is examined with the design interaction formula as
well as the general method. The examination based on the interaction design formula is
presented for the columns only. It is noted that the examionation should be extended to the
whole structure in the actual design project.
4.5.1 Examination of resistance of columns with design interaction formula
4.5.1.1 Basic assumptions
In the global stability analysis of columns the following assuptions are followed:
- the reduced slenderness for the in-plane buckling is determined due to the global stability
analysis of the main frame;
- at the inermediate points of the columns offset lateral supports are applied;
- rotation of the column shape at the supports is restrained by the offset
supports, therefore the out-of-plane buckling and the LTB can be examined for
the O1 upper part and the O2 top part.
4.5.1.2 In-plane buckling
buckling length [mm]
I c.y . L
4 .I c.y
L cr.y
N cr.y
2.
0.35.( c
y .1000.H c
2
.E. I c.y
L cr.y
reduced slenderness
= 4.772.10
( 1000. L) A c.pl
critical forcce [N]
c = 4.492
I b.y .H c
A c.eff . f y
N cr.y
123
6 . )
0.017.( c
6 . )
y = 1.493
4
L cr.y = 1.09. 10
6
N cr.y = 9.04. 10
y = 0.525
Ferenc Papp
reduction factor
0.34
0.5. 1
y
y
buckling resistance [N]
N b.Rd.y
y. y
1
y
0.2
y . A c.eff .
y = 0.873
fy
6
N b.Rd.y = 2.176. 10
M1
4.5.1.3 Out-of-plane buckling

- Examination of the O1 column part (upper part of the column)
equivalent structural length [mm]
L z.1
buckling length factor
z.1 1.0
L cr.z.1 z.1. L z.1
2
. E. I c.z
N cr.z.1
2
L cr.z.1

critical force [N]
z.1
reduced slenderness
reduction factor
z.1
3650
z.1 = 0.659
N cr.z.1
0.49
0.5. 1
z. z.1 0.2
1
2
z.1
N b.Rd.z.1
6
N cr.z.1 = 5.738. 10
A c.eff . f y
z.1
buckling resistance [ N]
3
L cr.z.1 = 3.65. 10
z.1. A c.eff.
z.1
fy
z.1
z.1
z.1 = 0.75
6
N b.Rd.z.1 = 1.869. 10
M1
- Examination of the O2 column part (buttom part of the column)

equivalent structural length [mm]
L z.2
z.2 0.7
L cr.z.2 z.2. L z.2
2
. E. I c.z
N cr.z.2
2
L cr.z.2

critical force [N]
z.2
reduced slenderness
reduction factor
z.2
3650
0.49
0.5. 1
z.2. A c.eff.
124
z.2 = 0.461
N cr.z.2
z. z.2 0.2
1
2
z.2
N b.Rd.z.2
7
N cr.z.2 = 1.171. 10
A c.eff . f y
z.2
reduction factor [N]
3
L cr.z.2 = 2.555. 10
z.2
fy
M1
z.2
z.2
z.2 = 0.864
6
N b.Rd.z.2 = 2.155. 10
Ferenc Papp
4.5.1.4 Lateral torsional buckling (LTB)
Examination of the O1 column part
equivalent structual length [mm]
L LT
LT 1.0
L cr.LT LT.L LT
LTB length factor

LTB length[mm]
ctritical moment [Nmm]
relevant load combination: LC 4
design bending moments [kNm]
M max.1
492
M min.1
72
moment coefficient
---bra---
1 = 0.146
M max.1
1.88 1.4. 1
C1
2
. E. I c.z I c.w
.
C 1.
2
I c.z
L cr.LT
3
L cr.LT = 3.65. 10
M min.1
gradient of moment
M cr
3650
2
0.52. 1
2
L cr.LT . G. I c.t
C 1 = 1.686
9
M cr = 2.537.10
2.
E. I c.z
A cskkent tnyez szmtsnl felttelezzk, hogy az oszlopszelvny a "hengrelt

szelvnyekkel egyenrtk" kategriba sorolhat:
LT.0 0.4
0.75
LT
reduced slenderness for LTB

reduction factor
W c.y.pl . f y
LT
0.76
LT
0.5. 1
LT. LT
1
LT.1
LT.2
LT
kc
f
125
LT
. LT
LT.0
LT.1 = 0.948
. LT
LT.2 = 4.701
LT
LT.1
1
1.33 0.33. 1
1 0.5. 1 k c . 1
LT.mod
M b.Rd.1
LT
1
LT.mod
LTB resistance [Nmm]
LT = 0.461
M cr
k c = 0.78
2. LT
0.8
LT
LT.mod = 1.036
f
1
LT.mod. W c.y.pl .
fy
M1
8
M b.Rd.1 = 5.397.10
Ferenc Papp
Examination of O2 column part
structural length [mm]
L LT
3650
LT 0.7
L cr.LT LT. L LT
LTB length factor

LTB length [mm]
critical bending moment [Nmm]

relevant load combination: LC 4
design bending moment [kNm] M max.2
354
M min.2
72
moment coefficient
M cr
M min.2
moment gradient
C2
2
. E.I c.z I c.w
.
C 1.
2
I c.z
L cr.LT
3
L cr.LT = 2.555. 10
2 = 0.203
M max.2
1.88 1.4. 2
0.52. 2
2
L cr.LT .G.I c.t
E.I c.z
W c.y.pl . f y
LT
reduction factor
M cr
LT
0.76
LT
0.5. 1
LT. LT
1
LT
LT = 0.329
LT.0
. LT
LT = 1.062
2
LT
LTB resistance [Nmm]
C 2 = 2.186
9
M cr = 4.977. 10
2.
reduced LTB slenderness
LT
. LT
M b.Rd.2
LT
LT. W c.y.pl .
4.5.1.5 Interaction of Flexural Bucklings and LTB

moment factor
C my
design normal force [kN]
N Ed
fy
M1
1.0
8
M b.Rd.2 = 5.397. 10
0.9
175
interaction factors (cross-section Class 1 and 2)

N . N Ed
k yy.1 C my. 1 y 0.2 .
y . A c.eff . f y
N . N Ed
k yy.2 C my. 1 0.8.
y . A c.eff . f y
k yy k yy.1
- Examination of the O1 column part
C mLT
k yy.1 = 0.924
k yy.2 = 0.958
0.6 0.4. 1
C mLT = 0.659
interaction factors
k zy.1
k zy.2
k zy
0.1. z.1
N . N Ed
C mLT 0.25 z.1.A c.eff .f y
N . N Ed
0.1
.
C mLT 0.25 z.1.A c.eff .f y
.
k zy.1
126
k zy.1 = 0.985
k zy.2 = 0.977
Ferenc Papp
used capacity
O1.1
O1.2
N .N Ed
N b.Rd.y
N . N Ed
N b.Rd.z.1
k yy .
M .M max.1
M b.Rd.1
k zy .
M . M max.1
M b.Rd.1
- Examination for the O2 column part
C mLT
O1.1 = 0.922
O1.2 = 0.991
0.6 0.4. 2
Megfelel!
Adequate!
C mLT = 0.519
interaction factors
k zy.1
k zy.2
k zy
0.1. z.1
N . N Ed
C mLT 0.25 z.2.A c.eff . f y
N . N Ed
0.1
.
C mLT 0.25 z.2.A c.eff . f y
.
k zy.1 = 0.98
k zy.2 = 0.97
k zy.1
used capacity
O2.1
O2.2
N .N Ed
N b.Rd.y
N . N Ed
N b.Rd.z.2
k yy .
M .M max.2
k zy .
M b.Rd.2
M . M max.2
M b.Rd.2
O2.1 = 0.686
O2.2 = 0.724
Megfelel!
Adequate!
The examination of the columns was preformed by the Member Designer Module of the
ConSteel 6.0 design software too:
127
Ferenc Papp
7.3 General method

7.3.1 Introduction
The general method may be used where the simple method (method of equivalent members)
is not applicable. It allows the verification of the resistance to lateral and lateral torsional
buckling for structural components such as
single members which are built-up or not, uniform or not, with complex support conditions
or not;
plane frames or subframes composed of such members.
The frame structure of this design project satisfies these conditions. The milestone of the
method is the global elastic stability analysis of the structure. If the model response contains
flexural torsional or lateral torsional buckling modes, the method requires finite element
analysis using general beam-column element with warping effect (14 DOF element) or shell
finite element.
7.3.2 Application of the method
Step 1: Load amplifier
Firstly the cross-sectional resistances of the structure should be calculated applying the
conservative interaction design formulae. The load amplifier should be calculated in the
critical cross-section due to the adequate load combination, where the utilization of the
resistance is the greatest:
1
ult ,k =
M y .Ed
N Ed
+
A f y Wy f y
It is noted that the load amplifier as a design parameter which is relevant for the whole
structure.
Step 2: Critical load amplifier
Global elastic stability analysis should be performed on the adequate load combination (see
Step 1). The structural model should have realistic lateral support system. The critical load
amplifier (cr,op) is equal to the lowest linear eigenvalue where the appropriate buckling mode
is out-of-plane mode (such as out-of-plane flexural buckling or lateral torsional buckling).
Step 3: Reduced slenderness
The reduced slenderness is relevant for the whole structure and related to out-of-plane
buckling mode:
op =
ult ,k
cr ,op
Step 4: Reduction factors

On the base of the reduced slenderness (see Step 3) the z out-of-plane buckling reduction
factor and the LT lateral torsional buckling reduction factor should be calculated for the
critical cross-section (see Step 1).
128
Ferenc Papp
Step 5: Checking
The global stability resistance of the structure should be checked at the critical cross-section
for the design forces of adequate load combination (see Step 1). The design formulae is
derived form the conservative design interaction formulae:
M y .Ed
N Ed
+
z A f y / M 1 LT Wy f y / M 1
glob .stab =
The resistance of the structure is adequate if glob .stab 1.0 .

7.3.3 Application
The follow example shows the application of the general method. The examination was
performed by the ConSteel software too. The application guide can be found in the Annex 17.
4.5.2 Global stability resistance by general method
4.5.2.1 Load amplifier
According to the examination of the cross-sectional resistances (see Section 4.4) it can be
seen that the critical cross-section is the K2 section, which is situated at the top of the
column. The adeauate load combination for the examination is the Load Combination 4.
- utilization of cross-sectional resistance in K2 section
N .N K2.Ed
A c.eff . f y
K2
M . M K2.y.Ed.red
W c.y.pl . f y
K2 = 0.866
- load amplifier
ult.k
1
K2
ult.k = 1.155
4.5.2.2 Critical load amplifier

The global stability analysis was performed on the reastically supported design model using
ConSteel software (see Section 4.5.2.1).
- ctirical load amplifier (first eigenvalue)
cr
5.37
- buckling mode (relevant eigenvector)
129
Ferenc Papp
4.5.2.3 Reduced slenderness
The reduced slenderness is relevant for the whole structure:
ult.k
op
op = 0.464
cr.op
4.5.2.4 Reductions factors

Reduction factors for both of the pure lateral buckling and LTB are calculated with the
general slenderness computed in Section 4.5.2.3.
- buckling about weak axis
0.49
0.5. 1
. op
1
2
op
- lateral torsional buckling
LT 0.76
LT
0.2
0.5. 1
LT
LT
= 0.672
z = 0.863
LT. op
1
LT
op
LT.0
. op
. op
LT = 0.605
LT = 0.946
4.5.2.5 Global stability resistance of the frame

The global stability resistance of the frame is calculated in the cross-section determioned in
Section 4.5.2.1 using the reduction factor method and the conservative interaction formulae:
N . N K2.Ed
N
N = 0.079
z. A c.eff . f y
M .M K2.y.Ed.red
M
M = 0.843
LT.W c.y.pl . f y
glob.stab
glob.stab = 0.922
The examination was performed by the ConSteel software too.
130
Megfelel!
Adequate!
Ferenc Papp
The utilization of global buckling resistance at the bottom flange of the beam haunch
may be calculated by linear interpolation. The result is approximately: 0,89.
131
Ferenc Papp
Annex 12
Method of reduction factor
The method of reduction factor is based on the linear elastic analysis of the geometrically
perfect design model. The buckling phenomenon is taken into consideration by the
appropriate buckling reduction factor. The main properties of the method are summarized in
the table:
model & analysis
details
imperfections
type of analysis
design formula
no
1st order
reduction factor
Example: Column buckling resistance

The column has rigid column base. The top of the column is free in direction y and fix
in direction z. The column is loaded by 160kN concentrated load at top. The crosssection of the column is HEA200, the grade of material is S235 and the height is 6
meters.
L0
elastic moduli [N/mm2]
210000
cross-sectional aera [mm2]
5383
moment of inertia [mm4]
Iz
13360000

design force [kN]
z 2.0
N Ed 160
critical force [kN]
N cr.z
cross-sectional resistance [kN]
reduced slenderness
6000
2
. E. I z
. 1
N cr.z = 192.293
2
z. L 0 1000
1
3
N pl.Rk = 1.265. 10
N pl.Rk A . f y .
1000
N pl.Rk
N cr.z
coefficient
z 0.49
0.5. 1
buckling reduction factor
imperfection factor
z. z
1
partial factor
M1
buckling resistance [kN]
N b.Rd.z
utilization of resistance
0.2
z = 0.127
1.0
z. N pl.Rk
N Ed
N b.Rd.z
M1
N b.Rd.z = 160.023
=1
Fig. M12 Column buckling resistance by the method of reduction factor
132
Ferenc Papp
Annex 13
Method of equivalent imperfection
The method of equivalent imperfection is based on the non-linear elastic (second order)
analysis of the geometrically imperfect design model. The buckling phenomenon is taken into
consideration by the appropriate local (initial out-of-curvature) and global (initial inclination)
imperfections. The examination of the cross-section resistance involves the effect of the
global buckling phenomenon. The main properties of the method are summarized in the table:
model & analysis
details
imperfections
type of analysis
design formula
local & global

2nd order
conservative interaction formulae
Example: Column buckling

The structure can be seen in Annex 12.
The design model with initial out-of-curvature (local imperfection) and inclination
(global imperfection) is shown by the Figure M13. The method leads the result which
was given by the method of reduction factor:
Initial equivalent imperfections

- global imperfection
0 =
h =
1
200
2
L
m = 1,0
2
6
= 0 ,816
= 0 h m = 0 ,0041
- local imperfection
' c' curve
e0 =
L
200
Design forces in the appropriate cross-section due to the design force and the
equivalent imperfections, see Figure M13:
N Ed = 160 kN
M y .Ed = 0
M z .Ed = 41,33kNm
133
Ferenc Papp

A .f y 1
.
N pl.Rd
M1 1000
3
N pl.Rd = 1.265. 10
sectional moduli [mm3]

200000
pl.z Column
Fig.WM13
buckling resistance by the method of equivalent imperfections
partial factor
M0
1.0
nyomatki ellenlls [kNm]

W pl.z. f y
1
.
M pl.Rd.z
1000000
M0
design force [kN]
M pl.Rd.z = 47
N Ed 160
design moment [kNm]
M z.Ed
41.33
N Ed
M z.Ed
N pl.Rd
M pl.Rd.z
= 1.006
Fig. M13 Column buckling resistance by the method of equivalent imperfections
134
Ferenc Papp
Annex 14
Method of partial equivalent imperfection
The method of partial equivalent imperfection is based on the non-linear elastic (second
order) analysis of the geometrically partial imperfect design model. The buckling
phenomenon is taken into consideration partially by the appropriate reduction factor and
partially by the equivalent global imperfection. The main properties of the method are
summarized in the table:
model & analysis
imperfections
type of analysis
design formula
details
global
2nd order
buckling interaction formulae (the
structural length is the buckling
length)
Example: Column buckling

The structure can be seen in Annex 12.
The design model with the initial inclination (global imperfection) and the calculation
are shown by the Figure M14. It can be seen that the method overestimates the
buckling resistance of the column with 7%. This method may be used for preliminary
design.
Global equivalent imperfection

1
0 =
200
2
2
=
= 0 ,816
h =
L
6
m = 1,0
= 0 h m = 0 ,0041
Design forces calculated on imperfect model with non-linear elastic analysis
N Ed = 160 kN
M y .Ed = 0
M z .Ed = 19 ,71kNm
135
Ferenc Papp
L0
elastic moduli [N/mm2]
design strength [N/mm2]
fy
235
cross-sectional area [mm2]
5383
moment of inertia [mm4]
Iz
13360000
sectional moduli [mm3]
W pl.z
design force [kN]
N Ed
design moment [kNm]
M z.Ed
elastic ctritical force [kN]
N cr.z
6000
210000
200000
160
19.71
2
. E. I z
2
L0
reduced slenderness
N pl.Rk
A .f y .
N pl.Rk
N cr.z
coefficient
z 0.49
0.5. 1
buckling reduction factor
imperfection factor
1.0
coefficient
C Mz
0.9
interaction factor
k zz
buckling resistance [kN]
bending resistance [kNm]
N cr.z = 769.171
3
N pl.Rk = 1.265. 10
1000
z = 1.282
z. z
1
M0
partial factors
. 1
1000
0.2
z = 0.396
M1 1.0
z. N pl.Rk
N b.Rd.z
N b.Rd.z = 501.344
M1
W pl.z.f y
. 1
M pl.Rd.z
M pl.Rd.z = 47
M0 1000000
C Mz . 1
N Ed
N b.Rd.z
2. z
k zz.
0.6 .
M z.Ed
M pl.Rd.z
N Ed
z. N pl.Rk
k zz = 1.464
= 0.933
Fig. M14 Column buckling resistance by the method of partial equivalent imperfection
136
Ferenc Papp
Annex 15
Interaction factors for Method 2
The following expressions are from EN 1993-1-1:2005 Eurocode 3: Design of steel
structures, Part 1.1: General rules and rules for buildings, Annex B. The expressions and
parameters are valid for compressed and bended members with hot-rolled or welded I sections
susceptible to torsional deformations.
Class 1 & 2 cross-sections
N
N
k yy = min Cmy 1 + y 0.2 M 1 Ed ; Cmy 1 + 0.8 M 1 Ed
y A f y
y A f y
If z 0.4 then
0.1 z M 1 N Ed
0.1 M 1 N Ed
k zy = max 1
;1
(CmLT 0.25) z A f y
(CmLT 0.25) z A f y
If z < 0.4 then
0.1 z M 1 N Ed
k zy = min 0.6 + z ;1
(CmLT 0.25) z A f y
Class 3 & 4 cross-sections
N
N
k yy = min Cmy 1 + 0.6 y M 1 Ed ;Cmy 1 + 0.6 M 1 Ed
y A f y
y A f y
0.05 z M 1 N Ed
0.05 M 1 N Ed
k zy = max 1
;1
(CmLT 0.25) z A f y
(CmLT 0.25) z A f y
The Cmy and CmLT equivalent moment factors depend on the shape of the bending moment
diagram:
Linear moment diagram (-11)
C m = 0 .6 + 0 .4 0 . 4
Non-linear moment diagram (01)

01 s 0-1
M
- distributed design force

C m = 0 .1 0 .8 s 0 . 4
- concentrated design forces
Cm = 0.8 s 0.4
137
Ferenc Papp
-11 and 01
- distributed design force
Cm = 0.95 + 0.05 h
- concentrated design forces
Cm = 0.90 + 0.10 h
M
M/
Important rules for the equivalent moment factors:
For sway frame Cmy = 0.9 should be applied!
For CmLT the moment diagram between two neighbouring lateral and fork supports should
be considered!
For Cmy the moment diagram on the column or on the whole beam should be considered
(between two neighbouring supports in the plane of the structure)!
138
Ferenc Papp
Annex 16
Simple method using ConSteel software
(application guide)
The simple method for global stability resistance of structural members (columns, beams)
may be preformed by the ConSteel program. Before the stability examination the First order
and the Second order analysis as well as the Cross-section resistance options should be
executed on the structure, see Figure M16.1. The options of the stability examination can be
found under Member checks (1) label. Selecting the Start designer module option (2) the
structural model appears on the graphical window where the member to be examined should
be selected (3). The selected member should be added to the Member table (4) where more
members can be collected. Selecting the actual member in the table (5) the Select option (6)
can activate the Design table, see Figure M16.2. The load combination related to the critical
cross-section resistance is selected automatically as actual load combination (7). In the next
step the appropriate examination should be selected. The examination can be one of the Pure
cases (8) or one of the Interaction cases (9). In the case of the actual design project the
Interaction cases, the Method 2 (10) and the Interaction of buckling and lateral torsional
buckling (11) options should be selected.
2
4
5
6
Fig.M16.1 Select member to examine by the interaction design formulae
After setting the initial parameters of the examination, the design procedure can be started by
the Next option (11). First the design parameters of the flexural buckling about the y-y major
axis should be determined, see Figure M16.3. The graphical window shows the member (12)
with the appropriate supports in the plane of the structure (in-plane buckling). The in-plane
buckling length is assumed automatically equal to the structural length (13).
139
Ferenc Papp
7
8
9
10
11
11
Fig. M16.2 Set design properties for global stability examination
The default buckling length can be modified by the engineer (14). Figure M16.4 shows the
table of modification. In the table the buckling length factor (15) or the critical load amplifier
(16) can be defined. (The critical load amplifier may be determined by partial stability
analysis too. The method is over the objectives of this teaching material, therefore the details
are neglected.) Using the Next option on the design table (Figure M16.3), the procedure steps
to the out-of-plane flexural buckling problem, see Figure M16.5. The content of the table is
similar to the table of the in-plane buckling problem. The graphics shows the lateral supports
(17) which divide the member into equivalent members (18). Accepting or modifying the
design parameters, the next stability problem is the lateral torsional buckling, see Figure
M16.6. The content of the table is similar to the table of the in-plane buckling problem, but it
is a bit complicated.
12
14
13
Fig.M16.3 Design parameters for in-plane flexural buckling
140
Ferenc Papp
15
16
Fig.M16.4 Modification of the initial design parameters
17
18
Fig.M16.5 Design parameters for out-of-plane flexural buckling
19
Fig.M16.6 Design parameters for lateral torsional buckling
141
Ferenc Papp
It is noted that the program assumes fork support (restricted rotation) at the lateral supports.
Accepting or modifying the design parameters, the Check option (19) executes the evaluation
of the interaction design formulae. The table of results contains the following information, see
Figure M16.6:
- the adequate case (member, load combination and buckling case) (20);
- the adequate equivalent members (21);
- the summary of the result (22);
- the design parameters for the pure buckling modes, respectively (23).
20
21
22
23
23
Fig.M16.7 Result of the interaction design formulae for global stability resistance
Notes: The actual version of the design module does not allow the examination of haunched beams and tapered
members. The new version which allows the design of these members will be launched at the end of 2011.
142
Ferenc Papp
Annex 17
General method using ConSteel software
(application guide)
The application of the general method requires a design model which has realistic lateral
support system. First the Analysis label (1) should be selected, and then the Analysis
parameters option (2), see Figure M17.1. On the table the Buckling analysis option (3)
should be switched on, and then the analysis can be executed.
1
2
Fig.M17.1 Set design parameters for global stability analysis

After the execution of the stability analysis it should be checked whether the buckling mode
and the critical load amplifier related to the adequate load combination satisfy the conditions
required by the general method (see Paragraph 5.2.2), or not. For visualisation the Buckling
option (4) and then the adequate load combination (5) should be selected. The graphics will
show the buckling mode with the selected graphical style (6), see Figure M17.2.
4
Fig.M17.2 Check the buckling mode and the critical load amplifier
143
Ferenc Papp
The number of the critical load amplifiers was specified at the setting of the analysis. In the
eigenvalue box (7) the lowest value is selected by program. The program uses this lowest
critical load amplifier in the formulae of the general method, unless the engineer changes it.
The default critical load amplifier can be changed selecting a new one from the list (8).
Clicking on the graphics (9) the Select eigenvalue for design option (10) should be selected in
the opened menu, see Figure M17.3.
7
10
Fig.M17.3 Select critical load amplifier for the general method

Accepting the default lowest critical load amplifier or selecting a higher one, the Global
checks label (11) should be selected, and then the Global resistance option (12). In the
appeared Design table (13) only the appropriate load combination (which is related to the
critical cross section resistance) should be switched on (14), see Figure M17.4.
11
12
13
14
Fig.M17.4 Select the adequate load combination for the examination
144
Ferenc Papp
In the Design settings table the Buckling check option (15) should be switched on, and then
the appropriate design options should be selected (16,17,18), see Figure M17.5. Check of the
global stability resistance of the structure by general method can be executed by the
Calculation option (19).
15
16
17
18
19
Fig.M17.5 Set the parameters for the general method
The result of the general method appears in the graphics, see Figure M17.6. The utilization of
the global stability resistance of the cross-sections can be visualized by the cursor (20). The
actual value can be fixed to the cross-section using the right button and the Marker option
(21). The parameters of the general design formula can be available by the Calculate section
option (22). The table contains all the parameters of the general formulae, including the
utilization (23) and the appropriate standard paragraph (24), see Figure M17.7.
20
21
22
Fig.M17.6 Visualise utilization of resistance and design parameters

145
Ferenc Papp
23
Fig.M17.7 Design parameters
146
24
Ferenc Papp
Practice 8
DESIGN OF THE JOINTS
2010-2011 Budapest
147
Ferenc Papp
8.1 General
The frame structure should be divided into structural members in order to reasonable
transmission conditions and erection. Both points of view may be satisfied if the frame is
divided into straight members such as columns and beams. In this case the following joints
should be designed (see the Figure 54):
Column bases
Beam-to-column joints
Beam-to-beam joint
Fig.54 Joints of the frame to be designed

The column bases can be rigid or pinned. Rigid column bases are favorable from steel
structural point of view, but the cost of the basement may be higher. The beam-to-column
joints are located at the maximum of the in-plane bending moment. The beam-to-beam joint is
located at the ridge point. The suggested constructions of the joints are shown in the following
paragraphs. Design of the joints is performed by the ConSteel/Joint software which follows
the specifications of the EC3-1-8 standard. The application of the software is illustrated in the
Annex 18.
8.2 Design of the column bases
Theoretically there are three types of column bases:
Simple (pinned)
Semi-rigid
Rigid
The specific problems of the construction and the design of the column bases are discussed in
the following paragraphs.
148
Ferenc Papp
8.2.1 Simple (pinned) column base

Figure 55 shows the construction of the simple column base. The base plate is connected to
the concrete basement by two anchor bolts. This construction may be used for hot-rolled or
welded I or H sections as well as for tapered columns. The following sizes may lead to the
optimal solution:
12-16 mm thick base plate
4.6, 4.8 or 5.6 grade of anchor bolt
20, 24 or 30 mm diameter of anchor bolt.
This column base has been assumed as pinned connection in the static calculation for many
years. According to EC3-1-8 this column bases construction belongs to the class of semirigid joints. However, the previous assumption (such as the simple column base is pinned)
has not been led to any degradation or structural failure. For favorable practice we can
declare that the simple column base can be considered as pinned joint. The resistance of the
frame with simple column bases is higher than the resistance of the same frame with real
pinned column bases.
In case of relatively high depth sections (more than 400-500 mm) and thick base plate
(thicker than 16 mm) the construction is not suggested applying. The moment resistance of
the column base may be relatively large due to the considerable distance between the
rotational point (compressed flange) and the anchor bolts. This constructional response may
lead to considerable tension force in the anchor bolts. The tension force may destroy the
anchor bolts or/and the base plate.
Fig.55 Simple column base construction with anchor bolts

8.2.2 Semi-rigid column base
Figure 56 shows the typical construction of the semi-rigid column base. This column base
may be applied for hot-rolled or welded columns. Basically, this construction leads to semirigid class. The rigid class may not be reached due to the following reasons:
the cross-sectional area of the tensioned anchor bolts is relatively small
the 8.8 or 10.9 grade of anchor bolts may lead large elongation
too large base plate would be required.
The moment resistance of this column base depends on the diameter and the grade of the
anchor bolts, and the thickness and the extension of the base plate. Relatively large extension
of base plate may disturb the architectural construction of the wall system. However, any
construction with reasonable design parameters may lead to semi-rigid behavior, and the
initial elastic stiffness of the joint should be taken in the analysis of the frame into
consideration. Because of these problems this column base construction is not suggested
using within the present design project.
149
Ferenc Papp
Fig.56 Semi-rigid column base construction

8.2.3 Rigid column base
Figure 57 shows a special construction for rigid column bases. In the lack of design
specifications the design of this construction requires special consideration. The algorithm of
the calculation of the moment resistance is shown in the Annex 19. The ConSteel/Joint
software follows this procedure. The stiffness classification of this construction is not
performed, but it can be assumed that this column bases is rigid. Following design guidelines
may lead to adequate and economical construction:
First the design forces and moments should be defining (the quadratic equation for the
length of the compressed concrete area can not be solved for excessive or extremely
low design moment).
The optimal height of the stiffener plates is about 200-300 mm, depending on the
depth of the column section.
The thickness of the stiffeners should be equal to, or grater than, the thickness of the
flanges of the column section.
The optimal size of the base plate has at least 100-100 mm extensions. The width of
the base plate should be the minimum which is possible.
The optimal thickness of the base plate is 20-30 mm, depending on the size of the
column section.
The grade of the anchor bolts should be 4.6, 4.8 or 5.6. The optimal diameter of the
bolts depends on the thickness of the base plate. Extremely the diameter can be 48 or
56 mm.
The type of the welds connecting the base plate to the stiffener is basically double
fillet welds. In case of sections with narrow flanges (ex. IPE 300-400) a single fillet
weld is used outside of the stiffener. For the safe the program takes the welds which
located at the extended part of the base plate into consideration.
Fig.57 Rigid column base construction
150
Ferenc Papp
8.2.4 Application
The details of the design of the rigid column bases of the actual example are shown in this
following paragraph. The design was performed by the ConSteel/Joint software. The
application guide of the software can be found in the Annex 18.
4.6 Design of the joints
The frame consists of straight fabrication units (columns and beams) which are connected by
moment resistance joints. Following joints are designed:
- column bases
- beam-to-column joints
- beam-to-beam joint
Design is performed by the ConSteel/Joint software.
4.6.1 Design of the column bases
Fix column bases were assumed in the analysis. Therefore rigid column bases sholud be
designed.
4.6.1.1 Initial parameters
Design forces
- relevant load combination: LC 5
- design forces [kN;m]
N Ed
179.7
M y.Ed
354.4
V z.Ed
116.7
Geometrical parameters [mm]

- base plate: 460-780 (t=20)
- stiffener plates: 240-780 (t=20)
- anchor bolts: grade 5.6 M30
Welds [mm]
- fillit welds which connect the column section to the base beam
double welds at the flanges: 5
double welds at the web : 4
- fillit welds which connect the stiffener plates to base plate
flange-to-stiffener weld: 5
stiffener-to-base plate weld: 6
4.6.1.2 Computer based design
Results of the computer based design are shown in the picture below. The full design
documentation can be found in the Annex.
151
Ferenc Papp
8.3 Design of the beam-to-column joints

8.3.1 Types of joint construction
From constructional point of view there are two solutions for the beam-to-column joint:
end-plated
end-plated with haunch.
The aim of the design is a rigid and full strength joint excepting the situation where the
construction does not allow full strength behavior. For example the beam-to-column joint
with tapered structural members is normally partial strength joint.
Figure 58 shows the joint construction with end-plate. The end-plate has 10-15 mm
extension to ensure room for the double fillet welds. The extension may be neglected using
deep penetration fillet or but welds, but the solution may be expensive.
Fig.58 Construction for beam-to-column joint with end-plate

Figure 59 shows the beam-to-column solution with haunch. From the joint design point of
view it does not matter that the haunch is short or long. The haunch is made from a piece of
beam section which is cut across as it is illustrated in the figure.
cross cut
piece of beam
section
Fig.59 Construction for beam-to-column joint with haunch
152
Ferenc Papp
8.3.2 Stiffeners
Stiffeners are the important parts of the joint. The types of the stiffener can be the following:
Web stiffener
Flange stiffener
Shear stiffener.
The web stiffener is welded to the web of the column in the line of the beam flanges, see
Figure 60. Stiffeners mean higher cost but basically they are needed to get rigid and full
strength joint.
(a)
(b)
Fig.60 Web stiffeners in the beam-to-column joints

(a) end-plated joint; (b) haunched joint
The flange of the hot-rolled column may be reinforced by flange stiffener (backing plate),
see Figure 61. The backing plate can be considered in the calculation if the plate is
connected to the web of the column with appropriate size of weld.
Fig.61 Reinforced column flange using backing plate

The resistance of the sheared column panel may be reinforced by shear stiffener, see Figure
62. The shear stiffeners may lead to much labor work.
153
Ferenc Papp
(a)
(b)
Fig.62 Column web panel with shear stiffener

(a) end-plated joint; (b) haunched joint
8.3.3 Design parameters
The construction of the joint may lead to rigid and full strength joint if the next guidelines are
followed:
The Lb beam length may be taken as the distance between the two columns.
The thickness of the end-plate should be equal to (or greater than) the thickness of the
column flange. Thicker end-plate may be inefficient.
The grade of bolts should be 10.9 rather than 8.8. The diameter of the bolts should
exceed the thickness of the end-plate.
The thickness of the web stiffeners should be equal to (or greater than) the thickness
of the beam flange. Thicker stiffener may be inefficient.
Shear stiffener is normally not applied. If it is applied, its thickness should be equal to
the thickness of the web stiffeners.
In many cases the resistance of the joint is determined by the resistance of the column
flange for bending. The thickness of the flange of the welded sections may be
increased using thicker plate at the joint, see the Paragraph 8.3.4. The flange of hotrolled sections may be reinforced by backing plate.
The arrangement of the bolts may be based on two methods. The conservative method
uses uniformly distributed bolts, while the progressive method applies the required
number of tensioned bolts and one row of bolts at the compressed flange, see the
Figure 60.
In case of haunched beam the size of the room between the beam flange and the
haunch flange should be checked: the bolt should be placed and turned freely.
8.3.4 Special construction for welded sections with high depth
Extended end-plate or haunch are normally not used in the joints of tapered members, see the
Figure 63a. Tapered members can be constructed with relatively thin web (6-8 mm) and
flanges (12-16 mm). These thicknesses often lead to semi-rigid and partial strength joint. It is
not suggested applying web and shear stiffeners to reinforce the thin web and flange plates.
Instead of stiffeners special end construction is used where the web panel and the flange plate
are replaced by thicker plates, see the Figure 63b. The replacing plates are 1,52,0 times
thicker than the original ones.
154
Ferenc Papp
(a)
(b)
Replacing
web plate
Replacing
flange plate
Fig.63 Beam-to-column joint of welded members with slender plates

(a) original construction;
(b) construction with replacing plates.
8.3.5 Application
The details of the design of the rigid beam-to-column joint are shown in this paragraph. The
design is performed by the ConSteel/Joint software. The application guide of the software
can be found in the Annex 18.
4.6.2 Design of the beam-to-column joint
In the analysis rigid beam-to-column connections were assumed. Therefore moment
resistant connections sholud be desined. The web panels should be reinforced by shear
stiffener.
4.6.2.1 Initial design parameters
Design forces
- relevant load combination: LC 4.
in the beam
N Ed
142.2
M y.Ed.1
368.0
M y.Ed.2
469.6
V z.Ed
148.6
reduced moment is used at the connection

(interpolation between two points)
L k 3500
M y.Ed.red
in the column
h c.
M y.Ed.2
N Ed
171.0

- end-plate: 240-750 (t=20)
- flange of the haunch: 240-3500 (t=20)
- web of the haunch : 300-3500 (t=6)
- web stiffeners: 112-468 (t=16)
- grade of the bolts: M27 10.9
- shear stiffener: 460-700 (t=8)
- backing plate: 112-700 (t=8)
155
M y.Ed.2
M y.Ed.1
Lk
4
M y.Ed
491.8
M y.Ed.red = 411.543
V z.Ed
114.3
Ferenc Papp
Welds [mm]
- welds
double fillet welds for the upper flange: 6
double fillet welds for the web: 3
double fillet welds for the bottom flange: 6
- welds for the haunch
double fillet welds for the flange: 7
double fillet welds for the web: 3
4.6.2.2 Checking
The results of the computer based design are shown in the picture (see below).
The full computer documentation of the design can be found in the Annex.
8.4 Design of the beam-to-beam joint

8.4.1 Joint constructions
Design of the beam-to-beam joint is based on the design rules and constructional details
which were used for to the design of the beam-to-column joints (see the Paragraph 8.3).
Figure 64 shows the possible constructions.
(a)
(b)
(c)
Fig.64 Constructions for the beam-to-beam joint

156
Ferenc Papp
Construction (a) leads to the lowest cost. If the utilization of the cross-sectional resistance
of the beam is relatively low as well as the depth of the section is high (ex. IPE section), the
construction may have adequate moment resistance. The moment resistance can be increased
by two rows of tension bolts. Construction (b) with extended end-plate and tensioned bolts
may give higher resistance. The size of the extension of the end-plate should be the minimum
which is allowed by the bolts. Conservatively, extended web plate may be used too.
Construction (c) gives the strongest joint, but the placing of the bolts in the room between
the flanges of the beam and the haunch may lead to problems. Following the next guidelines
the construction may lead to optimal joint:
the Lb beam length may be taken as the distance between the two columns;
the thickness of the end-plates are 16-25 mm, respectively, depending on the size of
the beam section;
it is suggested using the same grade and size which were used at the beam-to-column
joints;
construction (c) might be avoided;
uniformly distributed and high density bolt arrangement might be avoided.
8.4.2 Application
Details of the design of the rigid beam-to-beam joint are shown in this paragraph. The design
is performed by the ConSteel/Joint software. Guidelines for the software application can be
found in the Annex 18.
4.6.3 Design of the beam-to-beam joint

In the analysis rigid beam-to-beam connection was assumed. Therefore moment resistant
end-plated connection should be designed.
4.6.3.1Initial parameters
Design forces
- relevant load combination: LC 4
N Ed
109.2
M y.Ed
172.6
V z.Ed
19.3

- end-plate: 240-426 (t=20)
- grade of bolts: 10.9 M27
Welds [mm]
- double fillet welds at upper flange: 6
- double fillet welds at web: 3
- double fillet welds at bottom flange: 6
4.6.3.2 Checking
Results of the computer based design procedure are shown by the picture (see below).
The full computer documentation can be found in Annex.
157
Ferenc Papp
158
Ferenc Papp
Annex 18
Design of the end-plated moment resistant joints using
the ConSteel software
(application guide)
M.18.1 Execute the csJoint program (starting)
The csJoint program can be executed in two modes:
within the ConSteel program
directly (csJoint.exe)
There two modes to execute the program within the ConSteel program:
model independent
based on structural model
Both of the two modes are shown below, but details are given for the structural model based
design method.
M.18.1.1 Model independent mode
The csJoint program can be executed within the ConSteel program. Actual structural model
is not needed for this application. First a new structural model folder should be created, and
then the Structural members label (1), and then the Edit joints option (2) should be selected,
see Figure M18.1. If there are joint models in the folder, the actual model can be selected
from the list (3). If there is no model in the folder or a new model is wanted creating, the
Create option (4) should be selected.
1
Fig.M18.1 Execute csJoint program within the ConSteel and

create new joint model
In the appeared table the Name (5) of the joint model should be given firstly, and then the
sections of the joint can be selected by the Load sections option (6). By the Next option (7)
the design of the joint can be started, see Figure M18.2.
M.19.1.2 Structural model based mode
The structural model based mode requires a structural model which has the joints to be
designed. First the Structural members label (8) and then the Create joint by model option
(9) should be selected, see Figure M18.3. The joint can be created by clicking on the
159
Ferenc Papp
Fig.M18.2 Give the name of the joint model and select sections
8
9
Fig.M18.3 Create joint model by model based mode

actual node of the structural model. The program realizes the type of the joint and the sections
which compose the joint. Figure M18.4 shows the procedure where the column base is
created by the model based mode (10-12).
11
13
12
10
Fig.M18.4 Create column base by model based mode
If the joint consists of more than two members, the joint model can be simplified by
neglecting members (13).
160
Ferenc Papp
M.18.1.3 Model independent mode

The ConSteel/Joint software can be executed directly by the csJoint. This mode is useful
when there is no ConSteel structural model or the model was built up in other software.
Figure M18.5 shows the starting panel where the actual existing joint folder (14) can be
opened, or a new joint folder can be created (15). In case of new folder the name of it should
be given, and then the design procedure can be started by the Create option. From this
point the program follows the steps described in the Paragraph M18.1.1.
14
15
Fig.M18.5 Starting panel of the csJoint software

M.18.2 Design of simple column base
First the type of the joint (16) and then the type of the column base (17) should be selected,
see Figure M18.6. Using the Create option (18) the design panel appears with the initial
construction of the simple column base, see Figure M18.7.
17
16
18
Fig.M18.6 Select the type of the joint and the type of the column base
First the End-Plate option (19) of the menu may be selected, and then the design parameters
of the base plate and the initial bolt arrangement can be defined, see Figure M18.8. The
relevant values of the design parameters may be the follows:
Number of bolt rows (20): 1
Position of end-plate (21): -15
Height (22): actual value
Thickness (23): 12 or 16
Bolt distances (24): actual value
161
Ferenc Papp
19
Fig.M18.7 The design panel with the initial column base construction
22
23
21
20
24
Fig.M18.8 Initial and final settings of the design parameters

In the next step the Joint loading option (25) may be selected and then the Input of joint
loading (26) and the User defined joint loading (27) options, see Figure M18.9.
162
Ferenc Papp
26
27
25
28
29
30
31
Fig.M18.9 Design joint loading and results of design

The design force table can be opened by the New option (28). According to the rules of
directions (see the figure below the force table) the normal force (29) and the shear force (30)
can be defined. The results of the checking (31) can be found in the table at the right hand
side part of the design panel.
M.18.3 Design of rigid column base
First the type of the joint and then the type of the column base (32) should be selected, see the
Figure M18.10. The design panel and the initial construction of the rigid column base appear
by the Create option (33), see the Figure M18.11. In the next step the Joint loading option
(25) can be selected, and then the Input of joint loading (26) and the User defined joint
loading (27) options can be switched on, see the Figure M18.9. The design force table can be
opened by the New option (28). According to the rules of directions the design forces can be
given, see the Figure M18.12. The results of the actual checking (34) can be found in the
table at the right hand side part of the design panel, where the actual error message (red
color message) means that the size of the base beam is too low. Selecting the Base plate
option (35) the height of the plate (36) can be increased as far as the error message
disappears, see the Figure M18.13.
32
33
Fig.M18.10 Select the type of column base
163
Ferenc Papp
Fig.M18.11 Initial construction for the rigid column base
34
Fig.M18.12 Set design loading and discuss the result of the checking
36
35
Fig.M18.13 Set the adequate height of the base beam

164
Ferenc Papp
In the next step the size of the stiffeners (37-39) and the welds (40-41) can be set, see the
Figure M18.14.
37
40
39
41
38
Fig.M18.14 Set the adequate size of the stiffeners and the welds
The optimal construction may be reached by the change of the design parameters, see
Figure M18.15. The construction of the column base may be optimal if the following
guidelines are met:
the maximum utilization is close to 100%;
the other utilizations are high as possible;
the height of the stiffeners is low as possible;
the thickness of the base plate is low as possible;
the width and the height of the base plate is low as possible;
the diameter of the anchor bolts is low as possible.
Fig.M18.15 The optimal construction
165
Ferenc Papp
M.18.4 Design of beam-to-column joint with haunched beam

First the type of the joint (42) and then the type of the connection (43) should be selected, see
Figure M18.16. The design panel and the initial construction of the beam-to-column joint
appears by the OK option (44), see the Figure M18.17.
42
43
44
Fig.M18.16 Type of joint and connection
Fig.M18.17 Initial construction for the beam-to-column joint

The required joint construction can be reached by the following steps:
Height of the column

The Column option of the menu should be selected and then the Lc Average height
of column (45) should be given. The Lsr Position of reference plane (46) may be
taken as 10-15 mm, see the Figure M18.18.
166
Ferenc Papp
45
46
Fig.M18.18 Height of the column
Slope of the beam

The slope of the beam can be set in the Position of beam sub-panel (47), see the
Figure M18.19.
47
Fig.M18.19 Slope of the beam
Size of the haunch

Selecting the Beam option (48) and the Haunch on bottom flange label (49) and then
selecting the Haunch with flange option (50) the size of the haunch can be set. Initial
size can be generated by the Default option (51). The optional bottom web stiffener
can be positioned to the flange of the haunch (it is suggested) if the Assign the stiffener
to haunch flange option (52) is switched on, see the Figure M18.20.
49
48
50
52
51
Fig.M18.20 Set the haunch for the beam
167
Ferenc Papp
Size and position of the end-plate

Selecting the End-plate option (53) the position (54) and the height (55) of the endplate can be set, see Figure M.18.21.
55
54
53
56
Fig.M18.21 Size of the end-plate
Grade and size of the bolts

The grade and the size of the bolts can be defined using the Modify option (56). In
the appeared table first of all the size of the bolt (57) and the grade of the material (58)
can be set, see the Figure M18.22.
57
58
Fig.M18.22 Size and grade of the bolts
Arrangement of the bolts

Selecting the End-plate option (53) the arrangement of the bolts can be set in the
Vertical positions of bolts sub-table (59), Figure M.18.23. First the number of the
bolts (60) should be set and then the positions of the rows (61). The setting is
controlled by graphics (62).
Web stiffeners
Web stiffeners are applied optionally. Selecting the Stiffeners option (63) the upper
(64) and the lower (65) web stiffeners can be switched on, respectively, see Figure
M.18.24. The initial thicknesses of the stiffeners are the minimum (they are equal to
the thickness of the flanges).
168
Ferenc Papp
59
60
62
61
Fig.M18.23 Set arrangement of the bolts
63
65
64
Fig.M18.24 Web stiffeners
Design joint loading

Selecting the Joint Loading option (66) the Input of joint loading (67) and the User
defined joint loading (68) options can be switched on. In the next step the design
forces should be wrote into the force table (69), see Figure M18.25.
67
68
66
69
Fig.M18.25 Design joint loading
169
Ferenc Papp
The construction of the joint and the result of the checking are shown on the design panel, see
Figure M18.26. The no adequate utilizations of the components are written by red color (70).
70
Fig.M18.26 Evaluation of the design

The next step is to refine the design parameters. Keeping the following guidelines the
optimal construction can be reached:
The thickness of the end-plate is optimal if the moment resistance is not increased
applying thicker plate.
The size of the bolts is optimal if the moment resistance is not increased applying
greater diameter.
The number of the bolt rows is optimal if the moment resistance is not increased or
slightly increased applying more rows.
It is checked that the Lb beam length is equal to the distance between the columns.
If the design is not adequate, the sizes of the joint components determined by the program
should be changed. Figure M18.27 shows a program message as an example.
Fig.M18.27 The weakest components determined by the program
170
Ferenc Papp
Higher shear resistance for the web component of hot rolled section may be reached by shear
stiffeners (71) such as diagonal plates (72) or additional web plate (73), see Figure M18.28.
Thicker column web plate rather than web stiffener may be used in case of welded section.
71
72
73
Fig.M18.28 Web stiffeners

Sometimes the weakest component of the joint is the column flange. In case of hot rolled
sections the flange may be reinforced by backing plate (74-76), see Figure M18.29. In case of
welded sections thicker flange plate rather than backing plate may be used.
74
75
76
Fig.M18.29 Backing plate to reinforce the column flange

of the hot rolled section
M.18.5 Design of beam-to-column joint with tapered members
The creation of the beam-to-column joint can be seen in the Figure M18.16. The beam and
the column are welded sections with 240-16 flanges and 868-8 web, respectively. The initial
construction which is generated by the program is shown in Figure M18.30. Because of the
relatively high web haunch or extended end-plate are not used. If the following guidelines are
met in the design, an optimal construction can be reached:
171
Ferenc Papp
Fig.M18.30 Initial construction for the beam-to-column

joint of tapered members
Selecting the Column menu option the Lsr (length of the column up to the reference
plane) may be defined as 15-20 mm, and the Lc (actual length of the column) should
be defined, see the Figure M18.31.
Fig.M18.31 The actual length of the column
Selecting the Beam menu option the Lb (length of the beam) may be defined as the
distance between the two columns, and the (slope of beam) can be given, see the
Figure M18.32.
Fig.M18.32 The length and the slope of the beam
172
Ferenc Papp
Selecting the Stiffeners menu option the Use upper stiffener and the Use lower
stiffener options should be switched on, see the Figure M18.33.
Fig.M18.33 The web stiffeners
Selecting the Joint loading menu option the design forces should be defined according
to the rules of directions which are shown by the figure, see the Figure M18.34.
Fig.M18.34 Design joint loading
Selecting the End-plate menu option the Position of end-plate should be defined as
15-20 mm, according to the actual value of Lsr parameter. Later the Height end-plate
parameter should be defined which is adequate if there is room for the outer fillet weld
at the bottom flange of the beam, see the Figure M18.35. Selecting the Modify
option the grade of the bolts can be defined as 10.9 (optionally 8.8) and the diameter
of the bolts can be defined as 24 or 27 mm. The number of bolt rows can be selected
as 3 or 4. The adequate bolt arrangement is defined in the Pitch of holes table.
After the first setting of the parameters the result of the checking should be examined. Figure
M18.36 shows the actual design situation of the example. The problems emerged by the
program should be solved by the next steps:
The column web thickness does not satisfy the requirements. This problem may be
solved if 1,5-2,0 times thicker web plate is used, see the Figure 63. In the example 14
mm thick column web is used. New reinforced column section can be defined
according to the Figure M18.37.
173
Ferenc Papp
Fig.M18.35 Size of the end-plate and the bolt arrangement
Fig.M18.36 Result of the checking (not adequate!)
174
Ferenc Papp
Fig.M18.37 New cross-section
The utilization of the moment resistance of the joint is more than 100%. The
resistance can be increased using 1,5-2,0 times thicker column flange at the
connection, see the Figure 63. In the example 20 mm thick column flange is used.
The shear stiffness of the web is too low, therefore the joint is semi-rigid. The shear
stiffness can be increased if shear stiffener is applied, see the Figure M18.38.
Fig.M18.38 Shear stiffeners

After refining the joint parameters the checking of the joint may show an adequate beamto-column construction, see Figure M18.39.
Fig.M18.39 The adequate beam-to-column joint

(reinforced column end and shear stiffener are used in the example)
175

Steel Buildings Design Notes PDF

Uploaded by

Document Information

Original Description:

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

Steel Buildings Design Notes PDF

Uploaded by

Copyright:

Available Formats

Ferenc Papp

Written in the framework of the project TMOP 421.B JLK 29.

1.1 The aim of the design task

1.3 The theoretical parameters of the main structure

Fig.2 Theoretical parameters of the main frame

1.4 The number of the main frames and their interval

beam of the frame

Fig.4 Structural system of the end wall

1.5 The initial cross-sectional parameters for the main frames

type of the section

hot rolled (IPE/HEA)

Tab. 2 The signs of the geometrical properties of the cross-sections

width of the flange

1.6 The initial grade of material

- width of the building [m]

- heigth of the side walls [m]

- slope of the roof [deg]

1.2 Initial data for the main structural members

- columns in side walls

1.3 Theoretical properties of the structural model

1.4 Number of the main frames

- interval of the frames [m]

- required number of the frames

- applied number of the frames

The building consists of 7 frames!

- actual area of the building [m 2 ]

1.8 Arrangement of the purlin system

Fig.5 Optimal arrangement of purlin system in term of the distance Ls

Fig. 6 The scheme of the practical purlin arrangement:

- interval of purlins [mm]

The e=2598 mm distance is choosen for the arrangement

1.11 Covering system

external trapezoid sheet

1.12 Preliminary drawings

top view of the foundation (right side of the drawing):

Top view of the roof

Top view of the foundation

Lindab purlin (Z section)

Bracing cross bars (L or rod section)

Fig. 9 Top view drawing of the roof and the foundation

Department of Structural Engineering BUTE

Arrangement of wall beams and bracing system

Lindab wall beam (C200)

Bracing bars (CHS)

length of the window: 11600

Fig.10 Side view drawing of the building

Department of Structural Engineering BUTE

Slope of roof: 100

Purlins (Lindab Z200)

Wall beams (Lindab C200)

CHS bracing members

Fig.11 Side view drawing of the structural frame

Rigid column base

Department of Structural Engineering BUTE

Written in the framework of the project TMOP 421.B JLK 29.

2.2.2 Weight of the covering system

2. LOADS AND EFFECTS

- heat insulation (mineral rockwool) [kN/m 2 ]

- further layers for insulation [kN/m 2 ]

- main frame: automatically considered

- building equipments [kN/m 2]

2.3 Meteorological loads and effects