UNIVERSITY OF LIEGE
Department ArGEnCO
Service Structural Engineering
USER’S MANUAL FOR SAFIR
2019
A COMPUTER PROGRAM FOR ANALYSIS OF STRUCTURES
SUBJECTED TO FIRE
by
Jean-Marc Franssen [1] & Thomas Gernay [2]
January 2019
[1] jm.franssen@uliege.be – Liege University, Liege, Belgium
[2] tgernay@jhu.edu – Johns Hopkins University, Baltimore, MD, U.S.A.
TABLE OF CONTENTS
1
1.1
1.2
1.7
INTRODUCTION
General
Analysis Procedure.
1.2.1 Thermal analysis.
1.2.2 Analysis of torsional stiffness of BEAM elements.
1.2.3 Structural analysis at elevated temperature.
Capabilities of SAFIR.
1.3.1 Capabilities concerning the temperature analysis.
1.3.2 Capabilities concerning the torsional analysis.
1.3.3 Capabilities concerning the structural analysis.
Common Features in all Analyses.
SAFIR functions and user defined functions.
1.5.1 General principle.
1.5.2 SAFIR defined functions.
1.5.3 User defined functions.
Sign Conventions.
1.6.1 Global and local axes.
1.6.2 Stresses.
Units.
3
3
4
4
7
7
8
8
8
8
10
11
11
11
12
13
13
13
14
2
2.1
2.2
2.3
2.4
INPUT DESCRIPTION.
Input for SAFIR.
General Data for Structural Analysis.
Material Properties.
Convergence Criteria.
15
15
16
19
20
1.3
1.4
1.5
1.6
3
DETAILED INPUT DATA AND FORMAT.
21
3.1
Description and Format of the .IN file for Thermal Analysis.
21
3.2
Description and Format of the eventual file describing the local HASEMI
fire(s) for a thermal analysis.
22
3.3
Description and Format of the .IN file for Structural Analysis.
23
3.4
Description and Format of the .IN file for Torsional Analysis.
59
3.5
Structure of the .TEM files used with the BEAM F. E.
66
3.6
Structure of the .TSH files used with the shell F. E.
68
3.7
Structure of the temperature files used with the truss F. E.
72
1 INTRODUCTION
1.1 General
SAFIR is a special purpose computer program for the analysis of structures
under ambient and elevated temperature conditions. The program, which is based on
the Finite Element Method (FEM), can be used to study the behaviour of one, two and
three-dimensional structures. The program (SAFIR) was developed at the University of
Liège, Belgium, and is today viewed as the second generation of structural fire codes
developed in Liège, the first generation being another computer program called
Computer Engineering of the Fire design of Composite and Steel Structures
(CEFICOSS)1,2.
As a finite element program, SAFIR accommodates various elements for
different idealization, calculation procedures and various material models for
incorporating stress-strain behaviour. The elements include the 2-D SOLID elements,
3-D SOLID elements, BEAM elements, SHELL elements and TRUSS elements. The
stress-strain material laws are generally linear-elliptic for steel and non-linear for
concrete.
The analysis procedure and the program capability are presented in this
Chapter. Details of the data files, material properties and cross sectional shapes are
presented in Chapter 2. The detail input and format used in the program are given in
Chapter 3, while Chapter 4 presents the theory and formulations of the elements
available in the program.
1.2
Analysis Procedure
Using the program, the analysis of a structure exposed to fire may consist of
several steps. The first step involves predicting the temperature distribution inside the
structural members, referred to as ‘thermal analysis’. The torsional analysis may be
necessary for 3-D BEAM elements, a section subject to warping and where the warping
function table and torsional stiffness of the cross section are not available. The last
part of the analysis, termed the ‘structural analysis’, is carried out for the main purpose
of determining the response of the structure due to static and thermal loading. The
various stages of analysis are briefly outlined in the following sections.
1.2.1
Thermal analysis
This analysis is usually performed while the structure is exposed to fire. For a
complex structure, the sub-structuring technique is used, where the total structure is
divided into several substructures and a temperature calculation is performed
successively for each of the substructures.
This kind of situation does arise in a
structure where the members are made of different section types, or made of sections
submitted to different fire exposures. The thermal analysis is made using 2-D SOLID
elements, to be used later on cross sections of BEAM elements or on the thickness of
SHELL elements.
a) Temperatures in beams
The temperature is non-uniform in the sections of the beam, but there is no heat transfer
along the axis of the beams. As an example, a frame structure with reinforced concrete
columns, pre-stressed main beams and structural steel secondary beams, will require
separate temperature analyses for each of these section types. From these analyses,
the temperatures across the cross section are obtained and are stored for subsequent
structural analysis where these sections are present.
b) Temperatures in shells
The temperature is non uniform on the thickness of the shell, but there is no heat
transfer in the plane of the shell. The temperature analysis is performed on a section
having the thickness of the shell and an arbitrary width, 1 cm for example. The node
numbering is from 1 to NNODE / 2 from the bottom to the top of the section and, again,
from NNODE / 2 + 1 to NNODE for the second row of nodes.
Figure 1 : Model for the thermal analysis of a slab
For example, the Figure above has been created with the following lines for a 10 cm
thick slab.
NODES
NODE
GNODE
REPEAT
1 -0.05
11 0.05
11 0.00
0.00
0.00
0.01
1
1
A .TSH file is created in which the temperatures of the first NNODE / 2 nodes are written.
Bellow is given an example of such a file. Note that the temperatures are calculated from
–t/2 to +t/2. For particular cases, like for example a uniform temperature distribution, a
similar file can be created with a text editor. The number of elements and the size (here
the thickness) of these elements is independent from the number and position of the
points of integration that will be used later in the structural analysis. For the structural
analysis, the temperatures at the points of integration are linearly interpolated from the
temperatures of the nodes.
THIS IS A COMMENT LINE
THICKNESS
MATERIAL 1
0.10
REBARS 0
HOT
POSITIONS OF THE NODES.
=======================
NUMBER OF POSITIONS:
11
-0.5000E-01 -0.4000E-01 -0.3000E-01 -0.2000E-01 -0.1000E-01
0.2000E-01
0.3000E-01
TIME=
0.4000E-01
0.5000E-01
60.0000 SECONDS
OR
1 MIN.
0 SEC.
OR
2 MIN.
0 SEC.
===========================================
-0.0500
56.41
-0.0300
20.15
-0.0400
25.16
-0.0200
-0.0100
0.0000
19.96
20.00
20.00
0.0100 20.00
0.0200 20.00
0.0300 20.00
0.0400 20.00
0.0500 20.00
TIME=
120.0000 SECONDS
===========================================
-0.0500
95.76
-0.0300
24.43
-0.0400
-0.0200
-0.0100
41.51
20.54
20.01
0.0000E+00
0.1000E-01
0.0000 20.00
0.0100 20.00
0.0200 20.00
0.0300 20.00
0.0400 20.00
0.0500 20.00
1.2.2
Analysis of torsional stiffness of BEAM elements
This analysis is usually performed when analyzing structures with 3-D BEAM
elements, either because non-uniform torsion and beam cross-section were subject to
warping (warping function is not equal to zero) or because the torsional stiffness is not
available from tables or formulas. The 2-D SOLID elements are used to calculate the
warping function and the torsional stiffness of the cross section. The torsional properties
obtained from this calculation are added to the results obtained from the temperature
analysis of the same cross section for subsequent structural analysis. In cases where
the warping function is not necessary, such as in the case of uniform torsion or a cross
section with a warping function equal to zero, and if the torsional stiffness can be found
in standard tables or by analytical formula, then this analysis need not be performed. In
such situations, the torsional stiffness is simply introduced as a property of the
cross-section for the structural analysis.
1.2.3
Structural analysis at elevated temperature
For each calculation, the loads are applied to the structure, described as
BEAM, TRUSS and SHELL elements. The temperature history of the structure, due to
fire, is read from the files created during the temperature analysis. As the computation
strategy is based on a step-by-step procedure, the following information can be obtained
until failure occurs in the structure:
·
Displacement at each node of the structure.
·
Axial and shear forces and bending moments at integration points in each finite
element.
·
Strains, stresses and tangent modulus in each mesh at integration points of each
finite element.
1.3
Capabilities of SAFIR
SAFIR can be used for performing three different types of calculations,
namely, thermal, torsional and structural analysis.
The capabilities of the program
concerning these three analysis types are outlined in this section.
1.3.1
Capabilities concerning the temperature analysis
●
Plane sections as well as three-dimensional structures can be analyzed.
●
Plane sections are discretized by triangular and/or quadrilateral (rectangular
and non-rectangular) elements, allowing representation of virtually all cross
sectional shapes.
●
Three-dimensional structures are discretized by solid elements (prismatic and
non-prismatic) with 6 or 8 nodes. This allows the representation of virtually all
structure shapes.
●
Variation of material from element to element is possible.
●
The fire temperature, defined as a function of time, can either be the standard
curves predefined in the code (ISO 834, ASTM E119, ULC S-101) or any
other curve can be introduced through data points.
●
Cooling down phases can be considered.
●
Variation of material properties with temperatures, as well as the evaporation
of moisture, can be considered.
●
Can analyze thermal performance of materials such as steel, reinforced
concrete and composite steel-concrete sections. Other materials can also be
analyzed provided their physical properties at elevated temperatures are
known.
1.3.2
Capabilities concerning the torsional analysis
●
Allows virtually all cross section shapes to be represented.
●
Materials are considered to be in the elastic stage, at ambient temperature.
The user may adjust the obtained torsional stiffness in order to take into
account an increase of temperature during the fire. The torsional stiffness
remains constant during the simulation of the structural behaviour.
1.3.3
Capabilities concerning the structural analysis
●
Plane or 3-D structures can be analyzed.
●
The structure is discretized by means of four different element types: Truss
elements, made of one single material with one uniform temperature per
element; beam elements, shell elements; and solid elements.
●
Large displacements are considered in the truss, beam and shell elements.
●
The effects of thermal strains (thermal restraint) can be accounted for.
●
Material properties are non-linearly temperature dependent.
●
Unloading of material is parallel to the elastic-loading branch.
●
Local failure of a structural member that does not endanger the safety of the
whole structure can be handled by means of a dynamic analysis.
●
Nodal coordinates can be introduced in the Cartesian or cylindrical system of
axes.
●
Imposed displacement (prescribed degrees of freedom) can be introduced.
●
Structures with external support inclined at an angle to the global axes can be
analyzed.
●
Residual stresses (initial strains) can be accounted for.
●
Pre-stressed structures can be analyzed.
Automatic adaptation of time step is possible and structural calculation continues until
failure or, alternatively, when the maximum deflection has reached a value defined by
the user.
1.4
Common Features in all Analyses
The common features in all computations are listed as follows:
●
●
●
●
1.5
Optimization of the matrix in order to reduce the computer storage and
calculation time can be performed by the program using internal re-numbering
of the system equations. This re-numbering is transparent to the user.
The same temperature or the same displacement can be imposed at two
different nodes by the use of master-slave relations.
Thermal and mechanical properties of the steel and concrete according to
Eurocodes 2, 3 and 4 are embedded in the code and can be used directly.
Graphic pre-processing and post-processing capabilities are by the
SAFIRwizard and DIAMONDXL codes, respectively. When needed, SAFIR
could be adapted so as to give the results in a format compatible with
commercial graphic software, such as I‑DEAS.
SAFIR functions and user defined functions
1.5.1
General principle
In different locations of the .IN file, some functions of time can be introduced.
They are used either to describe the evolution of the gas temperature in case of a
thermal analysis, or to prescribe the evolution of the solution in different nodes and
degrees of freedom (either be it a displacement, a temperature or a value of the warping
function).
There are two different types of functions:
1. SAFIR defined functions. These functions are embedded in the code. Each
function is represented by a name. The comprehensive list is given here bellow.
2. User defined functions. If the name (maximum 10 characters) is not one of the
SAFIR defined function, SAFIR will assume that it represents the filename.filetype
of a file in which the user has described the evolution of the function with time by a
series of (timei, valuei) pairs (free format). Linear interpolation is made between the
defined points. This file describing the function must be in the same folder as the
input file.
1.5.2
SAFIR defined functions
The comprehensive list of SAFIR defined function is (with t in seconds):
·
F0
f = 0
F1PS
f = t
f = 1
F1
·
·
·
MOINSF1PS
f = - t
F2PM
f = 2 t / 60
f = - t
FMOINS1PS
·
·
·
F20
f = 20
F1000
f = 1000
f = 100
F100
·
·
f = 1000 t
F1000PS
·
·
·
F1PSM1000
f = 0
F1000THPS
f = t / 1000
f = t – 1000
for t > 1000
f = t / 20
for t <= 20
FLOAD2
f = 0.1 t - 0.0025 t²
for t <= 20
FISO
f = 20 + 345 log10 (8 t / 60 + 1)
HYDROCARB
f = 20 + 1080 ( 1 – 0.325 e-0.167 t / 60 – 0.675 e-2.5 t / 60 )
FLOAD
·
·
·
f = 1
f = 1
·
ASTME119
·
for t > 20
for t > 20
f = 345 log10 (8 t / 60 + 1)
FISO0
·
for t <= 1000
f defined by linear interpolation between a set of ( time ;
temperature ) pairs, with time in minutes.
Time
Temp.
Time
Temp.
Time
Temp.
Time
Temp.
Time
Temp.
0
20
5
538
65
937
130
1017
250
1100
370
1184
10
704
70
946
140
1024
260
1107
380
1191
15
760
75
955
150
1031
270
1114
390
1198
20
795
80
963
160
1038
280
1121
400
1204
25
821
85
971
170
1045
290
1128
410
1211
30
843
90
978
180
1052
300
1135
420
1218
35
862
95
985
190
1059
310
1142
430
1225
40
878
100
991
200
1066
320
1149
440
1232
45
892
105
996
210
1072
330
1156
450
1239
50
905
110
1001
220
1079
340
1163
460
1246
55
916
115
1006
230
1086
350
1170
470
1253
60
927
120
1010
240
1093
360
1177
480
1260
1.5.3
User defined functions
An example of user defined function could be
·
myfire.fct
and the content of the file myfire.fct would be:
0.
20.
600.
200.
2400.
900.
720.
3600.
7200.
10800.
800.
300.
20.
20.
for a natural fire reaching a maximum temperature of 900°C after 40 minutes and
decreasing thereafter.
1.6
Sign Conventions
The following sign conventions are applied.
1.6.1
Global and local axes
Global axes are employed when defining a structure that is to be analyzed
using SAFIR. This is done using the Cartesian system of coordinates. For the 2-D
(plane) problems, the axes are named G1 and G2, while the local axes are named L1
and L2. Applied force and the displacements are positive in the direction of G1 and G2;
the applied moments and rotations are positive in a counter-clockwise direction. For the
3-D problem, the global axes are named G1, G2 and G3 and the local axes are named
L1, L2 and L3.
The movement G1-G2-G3 is dextrorsum; the applied force and
moments, displacements and rotations are all positive in the G1, G2 and G3 directions.
1.6.2
Stresses
The stresses are positive in tension. Axial forces, obtained as a summation of
the stresses, are also positive in tension. Bending moments in the beam elements,
obtained as a summation of yi si, with yi measured on the local axis L1, are positive
when fibres having a positive local coordinate are in tension.
1.7
Units
The international system of units is adopted. Hence, for instance, all length
quantities are in meter. Forces are in Newton. Time is in seconds.
Temperature is in degree Celsius (°C).
2
2.1
INPUT DESCRIPTION
Input for SAFIR
For any analysis using SAFIR, data files acting as input files to the program
are prepared. For each analysis type (thermal, torsional or structural analysis), the user
prepares one data file.
This is an ASCII file, created with a text editor, in a word
processor, or by the pre-processor GID (to be obtained separately), and it must have the
filetype .IN.
This file with a .IN extension contains information such as calculation strategy,
time discretization, loads, node coordinates, types of finite elements used, material
properties, etc. For structural analysis, the .IN file specifies the name of the files created
during thermal and torsional analyses and in which the temperature data is stored.
Figure 2 shows a schematic representation of the different steps and files that
may be involved in the case of a frame structure comprised of two types of different
sections, one for the columns and one for the beam. The structure is modelled by beam
finite elements. The user must create the .IN files. The commands, format and number
of lines required for a section in the input files are briefly given in the following sections,
whereas the detailed structure of these files is given in Chapter 3.
Figure 2 : files and steps
2.2 General Data for Structural Analysis
The general data for the .IN file of a structural analysis is briefly presented in Table 2.
In each input line, a command is given followed by the parameters for the command. Full
details of all the commands are given in Chapter 3.
Table 2: Input data file (.IN) format for structural analysis
Command
Parameter Format
<A80>
Blank line for end of comments
“NNODE”
Notes
Comments, multiple lines possible
“NDIM”
“NDDLMAX”
“STATIC”
OR
“DYNAMIC”
The user has to indicate (following
the word “STATIC” or “DYNAMIC”)
the resolution technic choosen. Either
“PURE_NR” or “APPR_NR”.
“NLOAD”
“OBLIQUE”
“COMEBACK”
OR
“NOCOMEBACK”
“NORENUM”
OR
“RENUMPERM”
OR
“RENUMGEO”
OR
“RENUM”
OR
“READRENUM”
Only COMEBACK needs a
parameter
Only RENUMGEO needs a
parameter, either 0 or a node number
“NMAT”
“ELEMENTS”
“BEAM”
Optional
“NG”
For beam elements
“NFIBER”
For beam elements
“TRUSS”
Optional
“SHELL”
Optional
“NGTHICK”
For shell elements
“NGREBARS”
For shell elements
“SOLID”
“NG”
Optional
For solid elements
"ENDELEM3
end of ELEMENT section
“NODES” or
"NODES_CYL"
“NODE” or "GNODE"
or
“REPEAT”
Choose Cartesian or Cylindrical
coordinates
REPEAT N1 TO N2 STEP N3 TIMES N4
Multiple lines possible.
Last I5 not present if NODE
command is used
“FIXATIONS”
“BLOCK”
Optional, multiple lines possible
“SAME”, "SAMEALL"
OR
“REPEAT”
Optional, multiple lines possible
"END_FIX"
line for end of section
“NODOFBEAM”
Filename.TEM
For beam elements
A20
“TRANSLATE”
"END_TRAN"
Multiple lines possible for beams
end of section
"ELEM" or "GELEM"
Entry to list nodes of all elements and
material type
“NODOFSOLID”
Optional
File name for .OUT file related to
solid elements
Filename.OUT
“ELEM” or "GELEM"
OR
“REPEAT”
“
Multiple lines possible for solid
elements
ENDSYM”
For solid elements
“NODOFSHELL”
Optional
Filename.TSH
<A20>
“ TRANSLATE”
<I5><I5>
“ENDTRANSLA”
Left justified file name of .TEM file for
beams
File name for .TSH file related to
shell elements
As many line as necessary for shell
elements
End of translation for shell elements
“ELEM”
OR
“REPEAT”
<9*I5>
“NODOFTRUSS”
Filename.TRS
Multiple lines possible for shell
elements
Optional
<A20><3*G10.0><I5>
<6*I5>
File name for .TRS file related to
truss elements
Nodes of truss elements
Oblique supports
RELAX_ELEM
<A10>
Optional. Relaxation of end forces
BEAMS
<A5>
Relaxation in beams
ELEM
<A4><I10><6 or 14 reals>
END_BEAMS
<A9>
END_RELAX
<A9>
Mulitiple lines possible, one for each
relaxed beam
“ PRECISION”
“
LOADS”
“ FUNCTION”
“ NODELOAD”
<I10><6*G10.0>
“ DISTRBEAM”
<I10><NDIM*G10.0><I10>
Optional
“DISTRSH”
<I10><NDIM*G10.0><I10>
Optional
“DISTRSOLID”
<I10><NDIM*G10.0><I10>
Optional
“END_LOAD ”
line for end of section
“
MASS”
Optional, multiple lines possible
Optional, only for dynamic analysis
“
M_NODE”
Optional
“
M_BEAM”
Optional
“ M_SHELL”
“END_MASS ”
Optional
line for end of section
Blank line for end of comments
“ MATERIALS”
“
<A10><I5>
Material name, Number of
temperatures
<8*G10.0>
Material properties, multiple
name-properties pairs possible
TIME”
“ TIMESTEP”
<G10.0><G10.0>
Multiple lines possible
“ ENDTIME”
“ NOEPSTH”
OR
“
EPSTH”
“OUTPUT”
Optional
“ TIMEPRINT”
<G10.0><G10.0>
Multiple lines possible
“END_TIMEPR”
“ PRINTDEPL”
Optional
“PRINTTMPRT”
Optional
“ PRINTFHE”
Optional
“PRINTREACT”
Optional
“ PRINTMN”
Optional
“PRNSIGMASL”
Optional
“PRINTVELAC”
Optional
“PRNSIGMASH”
Optional
“PRNNXSHELL”
Optional
“PRNEASHELL”
Optional
“PRNEISHELL”
Optional
“PRNSIGMABM”
<I><I>
Optional
“ PRINTET”
<I><I>
Optional
" PRNEPSMBM
<I><I>
Optional
"PRNEIBEAM"
optional
“PRNSIGMASL”
Optional
Blank line for end of comments
2.3 Material Properties
Material names are provided in the program by command CMAT(NM). the
values of the parameters associated with this material are introduced in the PARACOLD
vector.
There is a maximum of eight values of PARACOLD(I,NM) available in the
program, depending on the material name introduced in the CMAT(NM). Valid material
names are:
·
INSULATION, USER1, USER2, USER3, USER4, USER5, C_GYPSUM,
X_GYPSUM and SFRM_PROBA (these materials have only thermal properties;
they do not carry any load),
·
ELASTIC, BILIN, PARABCONC, RAMBOSGOOD, SILCO_COLD and
CALCO_COLD (these materials have only 1D mechanical properties at room
temperature),
·
STEELEC3, STEELEC3EN, STEELEC3DC, PSTEELA16, STEELEC2,
STEELEC2EN, STEEL_WPB, STEELSL,
STEC3PROBA, USER_STEEL,
CALCONCEC2, SILCONCEC2, LWCONC_EN, SILCONC_EN, CALCONC_EN,
SILCON_ETC, CALCON_ETC, SILHSC1_EN, SILHSC2_EN, SILHSC3_EN,
CALHSC1_EN, CALHSC2_EN, CALHSC3_EN, SILHSC1ETC, SILHSC2ETC,
SILHSC3ETC, CALHSC1ETC, CALHSC2ETC, CALHSC3ETC, SILCONC_PR,
CALCONC_PR, WOODEC5, SLS1.4301, SLS1.4401, SLS1.4404, SLS1.4571,
SLS1.4003, SLS1.4462, SLS1.4311, AL6061T6C, AL5083SUP, AL5083INF,
AL7020SUP and AL7020INF (these materials have thermal properties and 1D
mechanical properties at elevated temperatures),
·
STEELEC32D, SILCOETC2D, CALCOETC2D, LWCONC2D, SILCONC2D,
CALCONC2D,
STEELEC3PS,
ELPLANESTR, PLSTRVML, BLPLSTRVM,
BLPLSTRDP, VMRANK2D (these materials have 2D plane stress mechanical
properties).
·
SILCOETC3D, CALCOETC3D, STEELEC23D, STEELEC33D (these materials
have thermal properties and 3D mechanical properties)
The stress-strain relationships in the load bearing materials are non-linear and
are temperature dependent. In structures exposed to fire, the materials are subjected to
initial strains (ei), thermal effects (eth) and stress related effects (es). The stresses are,
therefore, caused by the difference between the total strain (etotal), obtained from the
nodal displacements, and the initial and thermal strains.
2.4 Convergence Criteria
In order to converge to a solution, a tolerance value has to be specified in
the program. SAFIR uses an iterative procedure to converge on the correct
solution for each increment. The precision given in the data file is a small value
that must be reached at different times in SAFIR calculations in order to have
convergence. A good precision value is dependent on the type of structure that is
being analyzed and information from preliminary runs. However, if the user does
not know which to choose, a value of 0.001 can be used as a starting point (In
case of a dynamic analysis, the default value of 0.0005 is recommended). After
the first run, an examination in the output of the out-of-balance forces and
increments of displacement during subsequent iterations can help the user to
modify the corresponding precision value to obtain an acceptable solution.
3
DETAILED INPUT DATA AND
FORMAT
3.1 Description and Format of the .IN file for Thermal
Analysis
See “Users manual of Safir - Thermal.docx”, Section D.1.2
3.2 Description and Format of the eventual file
describing the local HASEMI fire(s) for a thermal
analysis
See “Users manual of Safir - Thermal.docx”, Section D.1.3
3.3 Description and Format of the .IN file for Structural
Analysis
SERIES 1:
SERIES 2:
Comments.
One line for each comment (can be 0 line).
One blank line to mark end of comments.
SERIES 3:
Number of nodes.
One line.
"NNODE", NNODE
NNODE = Number of nodes of the structure.
SERIES 4:
Number of axes.
One line.
"NDIM", NDIM
NDIM = Number of global axes, 2 for plane structures, 3 for 3-D
structures.
SERIES 5:
Degrees of freedom.
One line, first line in series.
"NDOFMAX", NDOFMAX
NDOFMAX = Maximum number of degrees of freedom per node.
if NDIM = 2
for truss elements, NDOFMAX ≥ 2
for solid elements, NDOFMAX ≥ 2
for beam elements, NDOFMAX ≥ 3
if NDIM = 3
for truss elements, NDOFMAX ≥ 3
for solid elements, NDOFMAX ≥ 3
for shell elements, NDOFMAX ≥ 6
for beam elements, NDOFMAX ≥ 7
SERIES 6 (optional): MATRIX SOLVER
1 card (optional)
SOLVER, CHOLESKY
Ø SOLVER
[A6]
command
Ø CHOLESKY
[A8]
command used to force SAFIR to solve the system of
equations by the method of Cholesky with a storage of the
matrix by a skyline method. This method is not as efficient as
the default method of Pardiso based on a sparse matrix
solver. It is thus not recommended to use this card. The
possibility has been given as a safety measure because
Pardiso has been introduced recently in the code; this card
allows going back to the previous but outdated method in
case any problem would appear with the new method.
If this card is used, the next card on NCORES cannot be used,
because Cholesky systematically uses only one core of the
computer.
1 card (optional)
NCORES, ncores
Ø NCORES
[A6]
command
Ø ncores [integer]
Number of cores of the CPU of the computer used by matrix
solver. The default value is 1, in which case this card may be
omitted. This card can be used to force SAFIR to use more
than 1 core, if present on the computer. Recent experience
has shown that using more than 1 core hardly reduces the
time of the runs with the present version of Pardiso and this
card can thus be omitted as a common practice. The
possibility of using the card has nevertheless been given in
order to allow users to perform their own test on their
particular system, and in order to offer the possibility of
working with more than 1 core in the future if new releases of
Pardiso show to exploit several cores more efficiently.
This card cannot be used if the previous card of CHOLESKY
has been used, because Cholesky systematically uses only
one core of the computer.
SERIES 7:
either
Loads.
One line, first line of three possible line series.
"STATIC …* "
if the structure or one part of it is submitted to the fire
and a static analysis is required. This is the standard
option.
or
or
"STATICCOLD …* "
if SAFIR is used to determine the ultimate load bearing
capacity of a structure which is not submitted to the
fire, i.e. at room temperature.
"DYNAMIC …* "
if the structure or one part of it is submitted to the fire
and a dynamic analysis is required.
* Static, staticcold and dynamic must be followed by
the type of convergence procedure required during the
structural analysis. The program can use a pure
Newton-Raphson procedure (“PURE_NR”) or a
modified Newton-Raphson procedure (“APPR_NR”).
“PURE_NR” is recommanded for structures made of
beams, and “APPR_NR” is recommanded for
structures made of shells.
Load number of vectors.
One line, second line of two line series.
‘NLOAD’, NLOAD
NLOAD = Number of load vectors. One load vector is made
of different loads of different types that will vary with
time according to the same function.
NEW SERIES Hydrostatic loads
One line
‘HYDROST’, NHYDROST
NHYDROST = number of hydrostatic loads. Hydrostatic loads may
vary as a function of time during the calculation, depending on the
level of the water table. They are not comprised in any load vector.
SERIES 8:
Inclined supports.
One line.
‘OBLIQUE’, NOBLIQUE
NOBLIQUE = Number of inclined supports. Every node where a
boundary condition is expressed in a local system of coordinates,
instead of the global system of coordinates of the structure, is an
oblique support, see Series 20.
A "0" must be typed if there is no oblique support.
Note:
The utilisation of oblique supports is only possible when the CHOLESKY solver is
being used, see Series 5. It cannot be used with the PARDISO solver.
SERIES 9:
settings.
Convergence strategy.
One line, first line of two line series, choice of two possible
‘COMEBACK’, TIMESTEPMIN
TIMESTEPMIN = Minimum value for the time step in case of
comeback only.
or
‘NOCOMEBACK’
Note:
For static analysis,
If NOCOMEBACK is chosen, the simulation is stopped the first time the stiffness
matrix is not positive definite.
If COMEBACK is chosen, each time the stiffness matrix is negative, time is reset
at the last converged point and the simulation restarts from there with a time step
divided by 2. The division of the time step goes on until the time step is smaller
than TIMESTEPMIN.
For dynamic analysis,
If NOCOMEBACK is chosen, the simulation is stopped the first time the left term
of the equation is not positive.
If COMEBACK is chosen, each time this term is negative or the number of
iterations necessary to obtain the convergence is greater then 3, time is reset at
the last converged point and the simulation restarts from there with a smaller time
step. The division of the time step goes on until the time step is smaller than
TIMESTEPMIN. If, for three simultaneous time steps, the convergence is
obtained in less then three iterations, the time step is multiplied by 2 (limited at
the value of the initial time step).
SERIES 10:
Obsolete. No card needed here
SERIES 11: Number of materials.
One line.
‘NMAT’, NMAT
NMAT = Number of different materials.
Note:
If two materials have the same material law but different characteristics, it makes
two different materials. e.g. S235 and S355 steel.
SERIES 12: Number of different elements.
One line, first line of multiple line series.
‘ELEMENTS’
BEAM elements sub-series, present if beams are used in the structure.
One line
‘BEAM’, NBEAM, NGEOBEAM
NBEAM = Number of BEAM elements in the structure.
NGEOBEAM = Number of different groups of geometrical
properties.
Note:
One group of geometrical properties comprises elements that have
the same materials, the same cross section and the same
temperature history. One .TEM file will be necessary to describe
each of the NGEOBEAM groups.
One line.
‘NG’, NG
NG = Number of longitudinal points of integration in
elements. Cannot be less than 2. Greater than 3 is
not recommended.
One line.
‘NFIBER’, NFIBERBEAM
NFIBERBEAM = Number of longitudinal fibres in the beam
elements (the maximum value for all the different
groups of geometrical properties).
TRUSS elements subseries, present if truss elements are used in the structure
One line
‘TRUSS’, NTRUSS, NGEOTRUSS
NTRUSS = Number of TRUSS elements in the structure.
NGEOTRUSS = Number of different groups of geometrical
properties.
Note:
One group of geometrical properties comprised elements that had
the same materials, the same cross sectional area and the same
temperature history.
SHELL elements subseries, present if shell elements are used.
One line
‘SHELL’, NSHELL, NGEOSHELL
NSHELL = Number of SHELL elements in the structure.
NGEOSHELL = Number of different groups of geometrical
properties.
Note:
One group of geometrical properties comprised elements that had
the same materials, the same thickness, the same reinforcing bars
and the same temperature history.
One line
‘NGTHICK’, NGSHELLTHICK
NGSHELLTHICK = Number of points of integration on the
thickness of the elements. Cannot be less than 2 and
cannot be more than 10.
One line
‘NREBARS’, NREBARS
NREBARS = Number of REBAR
layers in the shell elements.
SOLID elements subseries, present if solid elements are used
.
One line
‘SOLID’, NSOLID
NSOLID = number of SOLID elements in the structure
One line.
‘NG’, NG
NG = Number of integration points. Valid entries are 1, 2 and 3.
SPRING elements subseries, present if spring elements are used
One line
‘SPRING’, NSPRING
NSPRING = number of SPRING elements in the structure
Last line of series.
"END_ELEM"
SERIES 13: The nodes.
One line, first line of multiple line series.
<A10>,[<A10>]
‘NODES’
or
‘NODES_CYL’
‘NODES_ CYL’ is used if the cylindrical system of
co-ordinate is chosen instead of the Cartesian
system for the introduction of the co-ordinates of the
nodes. Cylindrical input are transformed for the
internal solution process by
(r,q) => X = r cos(q), Y = r sin(q).
if NDIM = 2
(r,q,Z) => X = r cos(q), Y = r sin(q), Z, if NDIM = 3
Note:
q is in degrees.
The transformation is made after all the nodes have been read and generated.
CYLINDRIC is omitted if the nodes are directly input in the Cartesian system of
co-ordinates.
Nodes.
One line added for each node.
‘NODE’, NNO, RCOORD(1,NNO), …,RCOORD(NDIM,NNO)
NNO = Number of the specific node.
RCOORD(1,NNO) = First global coordinate of the node NNO.
…
…
RCOORD(NDIM,NNO) = Last global coordinate of node
NNO of NDIM global axis.
or
‘GNODE’, NNO, RCOORD(1,NNO), …,RCOORD(NDIM,NNO)
NNO = Number of the specific node.
RCOORD(1,NNO) = First global coordinate of the node NNO.
…
…
RCOORD(NDIM,NNO) = Last global coordinate of node
NNO of NDIM global axis.
This command is used to generate equidistant nodes
between the previously defined node and the current
node NNO.
or
‘REPEAT’, NNO, DELTAC(1), …, DELTAC(NDIM), KGENE
NNO = Number of nodes to be repeated.
DELTAC(1) = Increment on the first coordinate.
…
…
DELTAC(NDIM) = Increment on the coordinate NDIM.
KGENE = Number of times that this command has to be repeated.
SERIES 14: Supports and imposed displacements.
One line, first line of possible multiple line series.
‘FIXATIONS’
Supports and imposed displacements fixed blocks.
One line for each node where solution follows a defined function of
time and the reaction must be calculated.
‘BLOCK’, NNO, CBLOCK(1,NNO), …, CBLOCK(NDOFMAX,NNO)
NNO = Number of the specific node where the solution must
not be calculated.
CBLOCK(1,NNO) = Function describing displacement for
first D.o.F. at this node with respect to time. Type NO
if the displacement is not prescribed for this DoF.
CBLOCK(2,NNO) = Function describing displacement for
second D.o.F. at this node with respect to time Type
NO if the displacement is not prescribed for this DoF.
…
CBLOCK(NDOFMAX,NNO) = Function describing
displacement for last D.o.F. at this node with respect
to time. Type NO if the displacement is not
prescribed for this DoF.
Note:
For each degrees of freedom NDL, from 1 to NDOFMAX, CBLOCK(NDL,NN0) is
either 'NO' if the displacement is not imposed at this D.o.F. or the name of the
function describing the evolution of the displacement at this node with respect to
time.
'F0' is a common function, used to model a fixed support.
Note:
The next lines with the SAME commands can come only after all BLOCK
commands have been entered.
Supports and imposed displacements slave nodes.
One line added for each slave node.
‘SAME’, NNO1, NNO2, CTRAV(1), …, CTRAV(NDOFMAX)
NNO1 = Number of the specific slave node.
NNO2 = Number of the master node.
CTRAV(1) = ‘YES’ If the solution is the same as at node NNO2
and as at node NNO1 for the D.o.F. 1, = 'NO' otherwise
…
CTRAV(NDOFMAX) = ‘YES’ If the solution is the same as at node
NNO2 and as at node NNO1 for the D.o.F. NDOFMAX, = 'NO'
otherwise.
or one line added for repeating series of slave node nodes
‘REPEAT’, NUMBER, INCR, CTRAV(1), …, CTRAV(NDOFMAX)
NUMBER = Number of times that the preceding SAME
command must be repeated.
INCR = Increment on NNO1 and NNO2.
CTRAV(1) = ‘YES’ If the solution is the same as at node
NNO2 as at node NNO1 for the D.o.F. 1, = ‘NO’ If
there is no master-slave relation for this D.o.F.
…
CTRAV(NDOFMAX) = ‘YES’ If the solution is the same as at
NNO2 as at NNO1 for the last D.o.F, = ‘NO’ If there is
no master-slave relation for this D.o.F.
Or one line to create master-slave relationships
between all nodes with same coordinates
‘SAMEALL’, CTRAV(1), …, CTRAV(NDOFMAX)
All the nodes of the structure that have the same coordinates (with a
precision of 0.1 mm) will automatically be attributed a master-slave
relationship.
CTRAV(1) = ‘YES’ If the solution is the same for the D.o.F. 1, =
'NO' otherwise
…
CTRAV(NDOFMAX) = ‘YES’ If the solution is the same for the
D.o.F. NDOFMAX, = 'NO' otherwise.
Note:
A node cannot be defined as slave node in a relationship and master node in
another relationship.
Therefore, if a given node is used in two or more relationships of slave-master, it
must be defined whether as the master in all these relationships, or as the slave
in all these relationships.
Example:
Consider that nodes 1, 2 and 3, which have 7 DoF, coincide and that these three
nodes must have the same displacement in the z direction (in global axes). The
following relationships are valid:
SAME 2 1 NO NO YES NO NO NO NO
SAME 3 1 NO NO YES NO NO NO NO
The following relationships are not valid:
SAME 2 1 NO NO YES NO NO NO NO
SAME 1 3 NO NO YES NO NO NO NO
Last line, indicating that the series is finished
‘END_FIX’
SERIES 15: BEAM elements.
Note:
This series is skipped if no BEAM element is present in the structure.
One line, first line of possible multiple line series.
‘NODOFBEAM’
Beam elements file name sub-series. One sub-series for each type of element.
One line, first line of sub-series.
<A20>
‘CFILENAME’
CFILENAME = Full name of the file where the information on
this section type can be found. Usually the extension
is .TEM. File name is left justified.
Note:
The name of the .TEM files that describe the sections heated by the HASEMI fire
is, for each section type, the name of ONE of the relevant .TEM file. For example,
"b0017_2.tem".
The information about the torsion properties has to be present only in this file, not
in the other .TEM files of the same section type that describe the temperature at
the other points of integration.
As a consequence, all the beam elements of one section type have the same
torsion stiffness.
Beam elements sub-series material translation.
One line added for each different material used in the section.
‘TRANSLATE’, MATL, MATG
MATL = Local number of this material in this section type.
MATG = Global number of this material in the structure.
Note:
MATL starts from 1 for the first material in this section type. The second line is for
the 2nd local material, etc.
Those lines are necessary because of the strategy used for the data files. One
structure can be made of several BEAM section types, each of them being
described in one .TEM file. In each of those .TEM files, the different materials are
given numbers starting from 1. It is necessary to indicate at the level of the
structure, which global material number corresponds to the numbers given in the
.TEM files.
Beam element sub series last line.
One line to mark end of sub-series.
‘END_TRANS’
Beam elements list (in increasing order, from 1 to NBEAM).
'ELEM', NE, NODOFBEAM(1,NE), …, NODOFBEAM(4,NE),
ITYPEBEAM(NE)
NE = Number of this element.
NODOFBEAM(1,NE) = First end node of this element.
NODOFBEAM(3,NE) = Third (i.e. central) node of this element.
NODOFBEAM(2,NE) = Second end node of this element.
NODOFBEAM(4,NE) = 4th node of this element (present
only if NDIM = 3).
ITYPEBEAM(NE) = The section type of this element.
or
'GELEM', NE, NODOFBEAM(1,NE), …, NODOFBEAM(4,NE),
ITYPEBEAM(NE), KGENE
KGENE allows the generation from the previously defined
element up to this one. KGENE gives the increment
on the first 3 nodes.
or
'REPEAT', NE, Nincr123, Nincr4, NincrType, Ntimes
NE = The NE previously defined elements will be repeated
NINCR123: increment on the nodes 1, 2 and 3
NINCR4: increment on the node 4 (present only if
NDIM
= 3).
NINCRTYPE: increment on the type of the element.
NTIMES: how many times these NE elements will be repeated.
Example:
The following sequence
ELEM
GELEM
ELEM
GELEM
ELEM
GELEM
ELEM
GELEM
ELEM
GELEM
8
9
1
15
18
16
17
24
25
32
33
40
1
32
35
49
52
66
69
83
16
19
2
3 108
17 108 1
20 108 1
33
34 108
36
37 108
50
51 108
53
54 108
67
68 108
70
71 108
84
85 108
1
2
12
1
12
1
12
1
12
can be replaced by the following one
ELEM 1
GELEM
REPEAT 8
12
8
15
17
3
16
108
17
1
108
0
0
1
4
2
to generate
ELEM. NODE 1
1
1
2
3
3
5
4
7
5
9
6
11
7
13
8
15
9
18
10
20
11
22
12
24
13
26
14
28
15
30
16
32
17
35
18
37
19
39
20
41
NODE 3
2
4
6
8
10
12
14
16
19
21
23
25
27
29
31
33
36
38
40
42
NODE 2 NODE 4
3
108
1
5
108
7
108
9
108
1
11
108
13
108
15
108
17
108
20
108
22
108
24
108
26
108
28
108
30
108
32
108
34
108
37
108
1
39
108
41
108
43
108
TYPE LENGTH
0.1288E+01
1
0.1288E+01
1
0.1288E+01
0.1288E+01
1
0.1200E+01
1
0.1200E+01
1
0.1200E+01
1
0.1200E+01
1
0.1288E+01
1
0.1288E+01
1
0.1288E+01
1
0.1288E+01
1
0.1200E+01
1
0.1200E+01
1
0.1200E+01
1
0.1200E+01
0.1288E+01
1
0.1288E+01
1
0.1288E+01
1
0.1288E+01
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
43
45
47
49
52
54
56
58
60
62
64
66
69
71
73
75
77
79
81
83
44
46
48
50
53
55
57
59
61
63
65
67
70
72
74
76
78
80
82
84
45
47
49
51
54
56
58
60
62
64
66
68
71
73
75
77
79
81
83
85
108
108
108
108
108
108
108
108
108
108
108
108
108
108
108
108
108
108
108
108
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
0.1200E+01
0.1200E+01
0.1200E+01
0.1200E+01
0.1288E+01
0.1288E+01
0.1288E+01
0.1288E+01
0.1200E+01
0.1200E+01
0.1200E+01
0.1200E+01
0.1288E+01
0.1288E+01
0.1288E+01
0.1288E+01
0.1200E+01
0.1200E+01
0.1200E+01
0.1200E+01
SERIES 16: SOLID elements.
Note:
This series is skipped if no SOLID element is present in the structure.
One line, first line of possible multiple line series.
<A10>
‘NODOFSOLID’
Solid elements file list. One sub-series for each type of element.
One line, first line of sub-series.
<A20>
left justified
‘CFILENAME’
CFILENAME = Full name of the file where the information on
the temperature distribution in the structure can be
found. Usually the extension is .OUT; this file is the
result of the 3D thermal analysis.
Solid elements list.
One line for each solid element.
<A10>,<9*I5>
‘ELEM’, NSL, N1, N2, N3, N4, N5, N6, N7, N8, NMAT, RES1,
RES2, RES3
NSL = Number of the element.
N1 = Node 1
N2 = Node 2
N3 = Node 3
N4 = Node 4
N5 = Node 5
N6 = Node 6
N7 = Node 7
N8 = Node 8
NMAT = Number of the material.
RES1 = Residual stress in the element in the direction of the
global axis 1.
RES2 = Residual stress in the element in the direction of the
global axis 2.
RES3 = Residual stress in the element in the direction of the
global axis 3.
or
<A10>,<I5>,<I5>,<30b>,<I5>
‘
REPEAT’, NSL, N1, ‘
’, NUMBER
NSL = Number of elements to be repeated.
N1 = Node increment.
NUMBER = Number of times that the NSL elements have to be
repeated.
SERIES 17: SHELL elements.
Note:
This series is skipped if no SHELL element is present in the structure.
One line, first line of possible multiple line series.
<A10>
‘NODOFSHELL’
Shell elements file list.
One line, first part of shell element sub-series, one
sub-series for each element type.
<A20> left justified
‘CFILENAME’
CFILENAME = File name where the information concerning
this section type is read.
Note:
The name of the .TSH files that describe the sections heated by the HASEMI fire
is, for each section type, the name of ONE of the relevant .TSH file. For example,
"s0156_3.tsh".
The information about the re-bar layers has to be present only in this file, not in
the other .TSH files of the same section type that describe the temperature at the
other points of integration.
As a consequence, all the shell elements of one section type have the same
re-bars.
Shell element material translation.
One line for each different material in section, second part of
shell element sub-series.
<A10>,<I5>,<I5>
‘ TRANSLATE’, N1, N2
N1 = Local number of this material in this section type.
N2 = Global number of this material in the structure.
Shell element series end of translation.
One line.
<A10>
‘ENDTRANSLA’
Shell element list.
One line for each shell element.
<A10>,<9*I5>
‘ELEM’, NSH, N1, N2, N3, N4, ITYPESHELL(NSH), KGENE
NSH = Number of the element.
N1 = Node 1
N2 = Node 2
N3 = Node 3
N4 = Node 4
ITYPESHELL(NSH) = Type of geometrical section.
KGENE = Automatic generation on the element number.
or
<A10>,<I5>,<I5>,<30b>,<I5>
‘
REPEAT’, NSH, N1, ‘
’, NUMBER
NSH = Number of elements to be repeated.
N1 = Node increment.
NUMBER = Number of times that the NSH elements have to be
repeated.
SERIES 18: TRUSS elements.
Note:
This series is skipped if no TRUSS element is present in the structure.
One line, first line of possible multiple line series.
‘NODOFTRUSS’
Truss elements files.
One line for each different truss section type used.
‘CFILENAME’, GEOTRUSS(1,NGT), GEOTRUSS(2,NGT),
IMATTRUSS(NGT)
CFILENAME = Name of the file where the temperatures
concerning this section types are read. Left justified.
GEOTRUSS(1,NGT) = Cross sectional area of this section type.
GEOTRUSS(2,NGT) = Residual stress of this section type.
IMATTRUSS(NGT) = Global Number of the material in this section
type.
Note:
If cfilename (NGT) is left blank, then:
·
this must be the case for all the section types
·
there is only one element in each NGT
·
the elements must be linked to nodes which belong to solid elements
·
the temperature of each truss element is the average of the
temperature of its 2 nodes, calculated with solid elements.
Truss elements list.
One line for each truss element.
'ELEM', NTR, NODOFTRUSS(1,NTR),
NODOFTRUSS(2,NTR), IGEOTRUSS(NTR),
KGENE
NTR = Number of the element.
NODOFTRUSS(1,NTR) = First node of this element.
NODOFTRUSS(2,NTR) = Second node of this element.
IGEOTRUSS(NTR) = Number of the section type for this element.
KGENE = Allows for automatic generation.
SERIES 19: SPRING elements.
Note:
This series is skipped if no SPRING element is present in the structure.
One line, first line of possible multiple line series.
‘NDFSPRING’
Each element is first defined by the node of the structure on which it is
applied.
The positive direction of the spring element is always defined from the
foundation to the structure, which means that, in the output, a positive
force in the spring, corresponding to tension, will be pulling on the
structure, whereas a negative force will push on the structure.
In the case of 2D models, the direction of the element is defined by the
angle at the foundation between the global axis X and the element, in
counter clockwise direction, as shown on the figure below.
In the case of 3D models, the direction of the element is defined by a 3D
vector that has the direction from the foundation to the node of the
structure. This vector does not have to be normalised.
The behaviour of the spring is shown on the previous Figure.
The force in the spring is comprised between two limits: the upper limit FS
and the lower limit FI.
The slope of the elastic part is the stiffness K.
The state of the spring at time t=0 is defined by the initial displacement Ui
and initial force Fi.
The behaviour in unloading is plastic.
All forces from the F-u diagram will be multiplied by the area of influence
A. As a consequence, the stiffness is also multiplied by A.
Note: in contradiction with the sign convention that is used in the output,
the values of FS, FI and Fi (as well as K) are positive when the spring
pushes on the structure. This is by analogy with active and passive
pressure when the spring is used to model a soil beside a wall or a
foundation underneath a beam.
One line for each SPRING element
For 2D structures
‘ELEM’, NSPR, NNODE, THETA, FS, FI, K, A, Ui, Fi
NSPR
Number of the element.
NNODE
Node where this element is attached.
THETA
Angle between X axis and this element
[degrees].
FS
FI
K
A
Ui
Superior limit of the force.
Inferior limit of the force.
Stiffness of the element.
Area of influence.
Displacement in the configuration of reference
Fi
0).
Force in the configuration of reference (time t =
(time t = 0).
For 3D structures
‘ELEM’, NSPR, NNODE, CX, CY, CZ, FS, FI, K, A, Ui, Fi
NSPR
Number of the element.
NNODE
Node where this element is attached.
CX
Component X of the vector that defines the
direction.
CY
Component Y of the vector that defines the
direction.
CZ
Component Z of the vector that defines the
direction.
FS
Superior limit of the force.
FI
Inferior limit of the force.
K
A
Ui
Stiffness of the element.
Area of influence.
Displacement in the configuration of reference
Fi
0).
Force in the configuration of reference (time t =
(time t = 0).
SERIES 19-2: Oblique supports (only possible with the CHOLESKY solver)
One line for each oblique support.
For 2D structures
’INCLIN’, Ni, Nj
Ni
is the node where a boundary condition is expressed in a
local system of coordinates.
Nj
is another nodes of the structure.
Ni and Nj define the direction in which the node Ni can move. It
cannot move perpendicularly to this direction.
For 3D structures
Boundary condition applied to the displacements
’INCLIN’, Ni, Nj, Nk
Ni
is the node where a boundary condition is expressed in a
local system of coordinates.
Nj, Nk are 2 other nodes of the structure.
Ni, Nj and Nk define the plane in which the node Ni can move. It
cannot move out of this plane.
Boundary condition applied to the rotations
’INCLIN’, -Ni, Nj, Nk
Ni
is the node where a boundary condition is expressed in a
local system of coordinates.
Nj, Nk are 2 other nodes of the structure.
Ni, Nj and Nk define the plane in which the node Ni can have
rotations. It cannot rotate along an axis perpendicular to this
plane.
One line to indicate that this is the end of the Series 20-0
<A10>
’END_INCLIN’
SERIES 19-3 Relaxation
This series is optional.
It is used to relax the internal force at the end of some elements.
The internal force F at the designated degree of freedom is computed according
to the following equation:
F = K (us - ui)
where K
us
ui
is a stiffness;
is the displacement of the node of the structure, in the local
system of coordinates of the beam;
is the displacement of the internal node at the end of the element,
in the local system of coordinates of the beam;
One line
‘RELAX_ELEM’
One line
‘BEAMS’
This line indicates that at least one BEAM finite element has one
or several degrees of freedom that are relaxed.
One line for each BEAM element that has one or several
relaxed degrees of freedom.
‘ELEM’, NBM, K(1), K(2), …. K(Idimk-1)
NBM = Number of the BEAM element
K(1) = stiffness for the relaxation at the first degree of
freedom of the beam (translation along local axis x of the
beam at node 1).
If the value is 0, there is no spring between the beam
element and the node of the structure. The relaxation
is total and the force at the end of the beam element
is equal to 0.
If the value is negative, there is no relaxation. The
displacement at the end of the beam is equal to the
displacement of the structure.
K(2) = stiffness at the first degree of freedom of the beam.
…
K(idimk-1) = stiffness at the last rotational degree of freedom
of node 2
Notes:
1)
Idimk is the number of degrees of freedom of the
beam, 7 for 2D beams and 15 for 3D beams.
2)
The degree of freedom of the central node cannot be
relaxed.
One line
‘END_BEAMS’
One line
‘END_RELAX’
SERIES 20: Precision.
One line.
‘PRECISION’, PRECISION
PRECISION = Small value that must be reached for convergence.
NEW SERIES: Limiting displacement
This series exists only if a static or a dynamic calculation is being
performed
One line
‘MAX_DISPL’, MAX_DISPL
MAX_DISPL = maximum value admitted for the displacements. If
this value is exceeded for an incremental displacement during
iteration, a COMEBACK is performed, if it has been foreseen, or the
calculation is stopped. If this value is exceeded after convergence,
the calculation is stopped.
Note: this series can be omitted (which ensures compatibility with
SAFIR versions older than 2016b). In that case, MAX_DISPL is
given the value of 999.
The SERIES 21 is repeated NLOAD times
SERIES 21: Loading.
One line.
‘LOADS’
Loading function.
One line
‘FUNCTION’, CFORCE(NLO)
CFORCE(NLO) = Function showing how the load
vector NLO varies as a function of
time, see § 1.5.2.
Loading on nodes
One line added for each point load directed at a node.
‘NODELOAD’, NNO, LOAD(1), LOAD(2), …, LOAD(NDOF)
NNO = Number of the node where loads are applied.
LOAD(1) = Load at degrees of freedom 1.
LOAD(2) = Load at degrees of freedom 2.
…
LOAD(NNDL) = Load at degrees of freedom NDOF
Note: iloc = minimum(NDOF,6)
Loading on beam elements.
Uniformly distributed loads
The direction of these loads is according to the global system of
coordinates and they are measured per meter in the direction of the
local axis x of the beam finite element. For example, the dead
weight of a prismatic column that extends along the global axis Z
can be described by a DISTRBEAM load in the direction of the
global axis Z.
One line added for each element with a uniformly distributed load
applied.
‘DISTRBEAM’, NBM, TRAV(1), TRAV(2), …, TRAV(NDIM)
NBM = Number of the specific BEAM under a distributed load.
TRAV(1) = Uniformly distributed load in the direction of the global
axis 1.
TRAV(2) = Uniformly distributed load in the direction of the
global axis 2.
…
TRAV(NDIM) = Uniformly distributed load in the direction of
the final global axis.
or
‘GDISTRBEAM’, NBM, TRAV(1), TRAV(2), …, TRAV(NDIM),
KGENE
NBM = Number of the specific BEAM under a distributed load.
TRAV(1) = Uniformly distributed load in the direction of the global
axis 1.
…
TRAV(NDIM) = Uniformly distributed load in the direction of
the final global axis.
KGENE , distributed loads are generated between the
previously defined element and the present element
with a step on the element numbers of KGENE
Trapezoidal distributed loads in local axes
The direction of these loads is according to the local system of
coordinates of the beam and they are measured per meter in the
direction of the local axis x of the beam. Such load can be used, for
example, to impose a wind pressure that varies in a trapezoidal
manner along an inclined roof.
One line added for each element with a distributed load applied.
‘TRAPLOCBM’, NBM, TRAV(1), TRAV(2), …, TRAV(2*NDIM)
NBM = Number of the BEAM under a trapezoidal distributed load.
TRAV(1): distributed load in the direction of the local axis 1 at node 1 of the
element.
TRAV(2): distributed load in the direction of the local axis 2
at node 1.
…
TRAV(NDIM): distributed load in the direction of the final
local axis at node 1.
TRAV(NDIM+1): distributed load in the direction of the local axis 1 at node 2 of
the element.
TRAV(NDIM+2): distributed load in the direction of the local
axis 2 at node 2.
…
TRAV(2*NDIM): distributed load in the direction of the final
local axis at node 2.
or
‘GTRAPLOCBM’, NBM1,NBM2, TRAV(1), TRAV(2), …,
TRAV(2*NDIM), KGENE
NBM1: Number of the first BEAM under a trapezoidal distributed
load.
NBM2: Number of the last BEAM under a trapezoidal distributed
load.
TRAV(1): distributed load in the direction of the local axis 1 at node 1 of the
element NBM1.
TRAV(2): distributed load in the direction of the local axis 2 at node 1 of the
element NBM1..
…
TRAV(NDIM): distributed load in the direction of the final local axis at node 1 of
the element NBM1..
TRAV(NDIM+1): distributed load in the direction of the local axis 1 at node 2 of
the element NBM2.
TRAV(NDIM+2): distributed load in the direction of the local
axis 2 at node 2 of the element NBM2.
…
TRAV(2*NDIM): distributed load in the direction of the final
local axis at node 2 of the element NBM2..
KGENE , distributed loads are generated with a step on the
element numbers of KGENE. The value of the
trapezoidal loads at the intermediate nodes are
calculated by linear interpolation based on the length
of the elements
Trapezoidal distributed loads in global axes
The direction of these loads is according to the global system of
coordinates and they are measured per meter in the direction of the
local axis x of the beam. Such load can be used, for example, to
impose the dead weight of a tapered member.
One line added for each element with a distributed load applied.
‘TRAPGLOBM’, NBM, TRAV(1), TRAV(2), …, TRAV(2*NDIM)
or
‘GTRAPGLOBM’, NBM1,NBM2, TRAV(1), TRAV(2), …,
TRAV(2*NDIM), KGENE
The format of these commands is exactly the same as the format of
TRAPLOCBM and GTRAPLOCBM, except that the distributed loads
are here described in the direction of the global system of
coordinates.
Loading on shell elements.
One line added for each element with a distributed load applied.
‘DISTRSH’, NSH, TRAV(1), TRAV(2), TRAV(3)
NSH
Number of the specific SHELL element under a distributed
load.
TRAV(1) Uniformly distributed load in the direction of the global
axis 1.
TRAV(2) Uniformly distributed load in the direction of the global
axis 2.
TRAV(3) Uniformly distributed load in the direction of the global
axis 3.
or
‘GDISTRSH’, NSH, TRAV(1), TRAV(2), TRAV(3), KGENE
NSH
Number of the specific SHELL element under a distributed
load.
TRAV(1) Uniformly distributed load in the direction of the global
axis 1.
TRAV(2) Uniformly distributed load in the direction of the global
axis 2.
TRAV(3) Uniformly distributed load in the direction of the global
axis 3.
KGENE Increment on the element number
Loading end of series.
One line
‘END_LOAD’
NEW SERIES: hydrostatic loads
This series is repeated NHYDROST time
One line
‘WATERTABLE’, CWATERTABLE
CWATERTABLE = chain of (maximum 10) characters, function
describing the level (in meters) of the water table
One line
‘SPECWEIGHT’, GAMMA
GAMMA = specific weight of the fluid, in m(N/m³).
Note: The first meter in the dimension of this value represents the
horizontal distance between adjacent beam elements. For example,
if 2D frames are representing frames that are, in fact, 3 meters
away in the direction perpendicular to the plane of the frame, the
specific weight of water to be considered here is not 10.000 N/m³,
but 30.000 m(N/m³)
If the specific weight is positive, the load on each element is in the
direction of the local axis y of the element. It is possible to enter a
negative value of the specific weight to change the direction of the
load.
Multiple lines, giving the list of the beam elements loaded by this
hydrostatic load
One line
‘HYDROBM’,NBM
NBM = number of the element loaded by this hydrostatic load.
or
One line
‘GHYDROBM’, NBM, KGENE
The elements from the previously defined element to the element
NBM, with an increment of KGENE, are loaded by this hydrostatic
load.
One line
‘END_HYDRO’
Note: A beam element loaded by a hydrostatic load will be loaded
by a TRAPLOCBM load. The distributed load at each end node of
the element is given by
P = GAMMA x DEPTH
where DEPTH is the difference between the value of
CWATERTABLE and the coordinate NDIM of the node in the
deformed configuration. It is thus assumed that gravity is in the
direction of –Y for 2D structures and –Z for 3D structures.
If the value of DEPTH at a given node is negative, it is replaced by
0.
The pressure will be applied on the deformed shape of the
structure.
SERIES 22: Mass characteristic.
Notes:
1) This series is present ONLY IF DYNAMIC HAS BEEN CHOSEN
2) In SAFIR, masses and forces are totally independent. The masses introduced
produce no force and the forces are not linked to any mass. As a consequence, if
a force of X Newton is produced by gravity, a mass of X/10 kg must normally be
also introduced in the data; if a force is produced by wind, no mass has to be
introduced.
One line, first line of possible multiple line series.
‘MASS’
Concentrated mass on nodes.
One line added for each concentrated mass linked to a node.
‘M_NODE’, NNO, MASS(1), MASS(2), …, MASS(NDOF)
NNO = Number of the node where the mass are applied.
MASS(1) = Mass linked to degree of freedom 1.
MASS(2) = Mass linked to degree of freedom 2.
…
MASS(NDOF) = Mass linked to degree of freedom NDOF.
Notes:
1) A mass linked to a displacement is in kg. A mass linked to a rotation is in
kgm².
2) Usually, a concentrated mass linked to a displacement is active in all
directions. Only in some particular cases can a mass be inactive in a particular
direction (for example, a ball laying on a horizontal surface might be active in the
direction perpendicular to this surface, and not active in the directions parallel to
this surface)
Distributed mass on beam elements.
One line added for each beam element with a distributed mass
applied.
‘M_BEAM’, NBM, TRAV(1), TRAV(2)
NBM = Number of the specific BEAM under a distributed mass.
TRAV(1) = Uniformly distributed mass applied on the beam
element (kg/m).
TRAV(2) = Rotational inertia around the longitudinal axis of the
element
with TRAV(2) only present if a 3D analysis
is made.
or
‘GM_BEAM’, NBM, TRAV(1), TRAV(2), KGENE
NBM = Number of the specific BEAM under a distributed mass.
TRAV(1) = Uniformly distributed mass applied on the beam
element (kg/m).
TRAV(2) = Rotational inertia around the longitudinal axis of
the element
(TRAV(2) only present if a 3D analysis is made)
KGENE = Increment on the element number
(Distributed mass are generated between the
previously defined element and the present element)
Distributed mass on shell elements.
One line added for each shell element with a distributed mass
applied.
‘M_SHELL’, NSH, TRAV(1)
NSH
Number of the specific SHELL element under a distributed
mass.
TRAV(1) Uniformly distributed mass on the shell element (kg/m²).
or
‘GM_SHELL’, NSH, TRAV(1), KGENE
NSH
Number of the specific SHELL element under a distributed
load.
TRAV(1) Uniformly distributed mass on the shell element (kg/m²).
KGENE Increment on the element number
Mass end of series.
One, last line of series.
‘END_MASS’
SERIES 23: Material description.
One line, first line of possible multiple line series.
MATERIALS
Material description sub-series. One sub-series entered for each NMAT material
type
One line, first line of two line material sub-series.
CMAT
CMAT Name of the material
Valid material names are:
·
INSULATION,
USER1,
USER2,
X_GYPSUM, C_GYPSUM, SFRM_PROBA
·
ELASTIC,
SILCONC_EN,
CALCON_ETC,
SILHSC1ETC,
USER3,
USER4,
SILCON_ETC,
SILHSC2ETC,
CALHSC1ETC,
CALHSC2ETC,
CALHSC3ETC,
CALHSC3_EN,
CALCONC_PR,
SILCONC_PR,
SILHSC2_EN,
SILCONCEC2,
SILHSC3_EN,
LWCONC_EN,
SILCO_COLD,
STEELEC3,
SLS1.4404,
CALHSC1_EN,
PARABCONC,
USER5,
CALCONC_EN,
SILHSC3ETC,
SILHSC1_EN,
CALHSC2_EN,
CALCONCEC2,
CALCO_COLD,
STEELEC3EN,
STEELEC3DC,
SLS1.4571,
SLS1.4003,
SLS1.4462,
AL5083SUP,
AL7020INF,
AL7020SUP,
PLSTRVML,
STEELEC32D,
SILCOETC2D,
STEELSL, STEELEC2, STEELEC2EN, SLS1.4301, SLS1.4401,
SLS1.4311,
AL5083INF,
BILIN,
RAMBOSGOOD,
WOODEC5,
AL6061T6C,
USER_STEEL, PSTEELA16, STEEL_WPB, STEC3PROBA
·
ELPLANESTR,
CALCOETC2D,
SILCONC2D,
CALCONC2D,
STEELEC3PS, VMRANK2D, BLPLSTRVM, BLPLSTRDP
·
LWCONC2D,
SILCOETC3D, CALCOETC3D, STEELEC23D, STEELEC33D
Material description sub-series parameters.
One line, second line of two line material sub-series.
INSULATION MATERIAL TYPES
If CMAT = INSULATION, USER1, USER2, USER3, USER4, USER5,
X_GYPSUM, C_GYPSUM, SFRM_PROBA, no parameter is necessary
because this material does not carry any stress. In this case, the second
line is a blank line.
UNIAXIAL MATERIAL TYPES
If CMAT = ELASTIC (this material is valid only at 20°C.)
PARACOLD(1,NM)
Young’s modulus.
PARACOLD(2,NM)
Poisson ratio.
If CMAT = BILIN (this material is valid only at 20°C.)
PARACOLD(1,NM)
Young’s modulus.
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Yield strength.
PARACOLD(4,NM)
Slope of the hardening branch.
If CMAT = RAMBOSGOOD (this material is valid only at 20°C.)
PARACOLD(1,NM)
E, Young’s modulus.
PARACOLD(1,NM)
Poisson ratio
PARACOLD(3,NM)
lp, the limit of proportionality.
PARACOLD(4,NM)
n, exponent of the law.
PARACOLD(5,NM)
K, factor of the law.
eps = sigma/E
for sigma <= lp
eps = (sigma/E) + ((sigma-lp)/K)^n
for sigma > lp
If CMAT = CALCON_ETC, SILCON_ETC
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Compressive strength
PARACOLD(4,NM)
Tensile strength (≥0)
The CALCON_ETC,
SILCON_ETC materials take into account
transient creep strain explicitly. The models are described in:
Gernay T., Franssen J.M. (2012). “A formulation of the Eurocode 2
concrete model at elevated temperature that includes an explicit
term for transient creep”. Fire Safety Journal, 51, 1-9.
If CMAT = CALCONC_EN, SILCONC_EN, LWCONC_EN
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Compressive strength
PARACOLD(4,NM)
Tensile strength (≥0)
If CMAT = CALCONCEC2, SILCONCEC2
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Compressive strength
PARACOLD(4,NM)
Tensile strength (≥0)
PARACOLD(5,NM)
< 0 if peak stress strain ec1 = minimum value (stiffer)
= 0 if peak stress strain ec1 = recommended value
> 0 if peak stress strain ec1 = maximum value (more
ductile)
If CMAT =
SILHSC1ETC, SILHSC2ETC, SILHSC3ETC,
CALHSC1ETC, CALHSC2ETC, CALHSC3ETC
If CMAT =
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Compressive strength
PARACOLD(4,NM)
Tensile strength (≥0)
SILHSC1_EN, SILHSC2_EN, SILHSC3_EN,
CALHSC1_EN, CALHSC2_EN, CALHSC3_EN
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Compressive strength
PARACOLD(4,NM)
Tensile strength (≥0)
If CMAT = CALCONC_PR, SILCONC_PR
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Compressive strength
PARACOLD(4,NM)
Tensile strength (≥0)
PARACOLD(5,NM)
Time at which this concrete is cast.
Before and until this time, the material does not carry
any stress or have any stiffness.
If CMAT = CALCO_COLD, SILCO_COLD
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Compressive strength, fcm, in N/m²
Note: the following equations are embedded in SAFIR
If CMAT = PARABCONC
PARACOLD(1,NM)
E, Young’s modulus.
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Compressive strength
PARACOLD(4,NM)
Tensile strength
PARACOLD(5,NM)
Strain at compressive strength
PARACOLD(6,NM)
Ultimate strain
If CMAT = AL6061T6C, AL5083SUP, AL5083INF, AL7020SUP,
AL7020INF (aluminium)
PARACOLD(2,NM)
PARACOLD(3,NM)
PARACOLD(4,NM)
f0.2
fp
erupture
in %
If CMAT = SLS1.4301, SLS1.4401, SLS1.4404, SLS1.4571,
SLS1.4003, SLS1.4462, SLS1.4311 (stainless steel)
PARACOLD(1,NM)
Young’s modulus.
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Yield strength
PARACOLD(4,NM)
Ultimate tensile strength
If CMAT = STEELEC2EN
PARACOLD(1,NM)
Young’s modulus.
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Yield strength
PARACOLD(4,NM)
Maximum temperature for a
reversible behaviour during cooling.
PARACOLD(5,NM)
Rate of decrease of the residual
yield strength when the maximum temperature has
been greater than PARACOLD(4,NM) [N/m²K]
PROCESS
Chain of character that indicates the
fabrication process of the reinforcing bar. It can be:
HOTROLLED
for hot rolled bars (columns 2, 4 and
6 in Table 3.2a of EN 1992-1-2)
COLDWORKED for cold worked bars (columns 3, 5
and 7 in Table 3.2a of EN 1992-1-2)
CLASS
Chain of character that indicates the class
of ductility of the reinforcing bars. It can be:
CLASS_A
for Class A (low) ductility
CLASS_C
for Class C (very high) ductility
CLASS_B
for Class B (high) ductility
If CMAT = STEELEC3, STEELEC3EN, STEELEC3DC, STEELEC2,
PSTEELA16, STEEL_WPB
PARACOLD(1,NM)
Young’s modulus.
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Yield strength
PARACOLD(4,NM)
Maximum temperature for a
reversible behaviour during cooling.
PARACOLD(5,NM)
Rate of decrease of the residual
yield strength when the maximum temperature has
been greater than PARACOLD(4,NM) [N/m²K]
If CMAT = STEELSL
PARACOLD(1,NM)
Young’s modulus.
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Yield strength
PARACOLD(4,NM)
Maximum temperature for a
reversible behaviour during cooling.
PARACOLD(5,NM)
Rate of decrease of the residual
yield strength when the maximum temperature has
been greater than PARACOLD(4,NM) [N/m²K]
PARACOLD(9,NM)
Slenderness of the plate where the
material is present
PARACOLD(10,NM)
Number of supports of the plate
where the material is present
= 3 for oustand plates (flanges)
= 4 for internal plates (webs)
The STEELSL material is described in the reference: Franssen,
J.M., Cowez, B., Gernay, T. (2014), “Effective stress method to be
used in beam finite elements to take local instabilities into account”,
Fire Safety Science 11, in Proc. of the 11th IAFSS Symposium,
Christchurch, New Zealand, Feb 10-14, pp. 544-557.
If CMAT = STEC3PROBA
PARACOLD(1,NM)
Young’s modulus.
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Yield strength
PARACOLD(4,NM)
Maximum temperature for a
reversible behaviour during cooling.
PARACOLD(5,NM)
Rate of decrease of the residual
yield strength when the maximum temperature has
been greater than PARACOLD(4,NM) [N/m²K]
PARACOLD(6,NM)
Value of the standard normal
parameter ε for the probabilistic reduction of yield
strength with temperature
The
STEC3PROBA
material
has
the
same
expression
of
stress-strain relationship as steel of Eurocodes but the reduction of
yield strength with temperature follows the logistic EC3-based
probabilistic model proposed in: Khorasani N.E., Gardoni P.,
Garlock M. (2015). “Probabilistic fire analysis: material models and
evaluation of steel structural members”. JSE, 141(12).
If CMAT = WOODEC5
PARACOLD(1,NM)
Young’s modulus.
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Compressive strength.
PARACOLD(4,NM)
Tensile strength.
If CMAT = USER_STEEL
PARACOLD(1,NM)
Young’s modulus at 20°C.
PARACOLD(2,NM)
Poisson’s ratio at 20°C.
PARACOLD(3,NM)
Yield strength at 20°C.
PARACOLD(4,NM)
critical temperature (in °C) beyond
which the yield strength is not fully recovered during cooling.
PARACOLD(5,NM)
the rate of decrease of the residual
yield strength if the temperature has exceeded the critical
temperature.
This
USER_STEEL
material
has
the
same
expression of
stress-strain relationship as steel of Eurocodes but it will behave at
elevated temperatures according to the decreasing curves specified
in the file "USER_STEEL.TXT" that the user has to create and
locate in the same folder as the input file.
In the file "USER_STEEL.TXT", kE, kfy, kfp, eth, ey, et and eu are
given at different temperatures. Between two temperatures, a linear
interpolation is performed by SAFIR.
kE, kfy and kfp are the reduction factors at elevated temperatures
relative to the values E, fy and fp at 20°C.
eth is the value of the thermal elongation at elevated temperature.
ey is the value of the yield strain at elevated temperature.
et is the value of the limiting strain for yield strength at elevated
temperature.
eu is the value of ultimate strain at elevated temperature.
Structure of the file "USER_STEEL.TXT"
One line.
Number_of_T:, NUMBER_OF_T
NUMBER_OF_T number of elevated temperatures at which the values of the
reduction factors are given.
One line
T
KE
Kfy
Kfp
EPSth
EPSy
EPSt
EPSu
One line for each temperature added to series.
T, kE(T), kfy(T), kfp(T), eth(T), ey(T), et(T), eu(T)
T = Temperature at which the reduction factors are given
kE(T) = reduction factor relative to the value of E (Young’s modulus) at 20°C
kfy(T) = reduction factor relative to the value of fy (effective yield strength) at 20°C
kfp(T) = reduction factor relative to the value of fp (limit of proportionality) at 20°C
eth (T) = thermal elongation at temperature T
ey (T) = yield strain at temperature T
et (T) = limiting strain for yield strength at temperature T
eu (T) = ultimate strain at temperature T
Note:
To have the same thermal elongation as in the material
STEELEC3EN for all temperatures, the first value written in the file
must be equal to -1.
Example: The following file describes a material that has user
defined variations of the E, fy and fp, but the same thermal
elongation as the steel of Eurocode 3, and the same yield strain,
limiting strain and ultimate strain as Eurocode 3.
Number_of_T: 4
T
EPSu
KE
Kfy
0.
1.0
1.00
200.
1.0
0.95
800.
0.1
1200.
0.0
0.20
0.20
0.20
0.20
Kfp
1.00
EPSth
EPSy
EPSt
-1.
0.02
0.15
0.90
-1.
0.02
0.15
0.15
0.10
-1.
0.02
0.15
0.00
0.00
-1.
0.02
0.15
BIAXIAL PLANE STRESS MATERIAL TYPES
If CMAT = ELPLANESTR
PARACOLD(1,NM)
Young’s modulus.
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Coefficient of thermal expansion
Elastic plane stress material law. The material is valid for steel at
elevated temperature and the Young's modulus and thermal strain
vary according to the Eurocode 3 part 1.2.
If CMAT = PLSTRVML
PARACOLD(1,NM)
Young’s modulus.
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Yield strength
PARACOLD(3,NM)
Strain hardening modulus
This model is a simplified model for steel at elevated temperature,
with a bilinear equivalent stress-strain relationship. The model
STEELEC32D is to be preferred, if no problem of convergence is
encountered.
The parameters vary according to the Eurocode 3 part 1.2.
(variation of the strain hardening modulus as for the Young’s
modulus).
If CMAT = STEELEC32D
PARACOLD(1,NM)
Young’s modulus.
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Yield strength
PARACOLD(4,NM)
Maximum temperature for a
reversible behaviour during cooling.
PARACOLD(5,NM)
Rate of decrease of the residual
yield strength when the maximum temperature has
been greater than PARACOLD(4,NM) [N/m²K]
If CMAT = STEELEC3PS
PARACOLD(1,NM)
Young’s modulus.
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Yield strength
PARACOLD(4,NM)
Maximum temperature for a
reversible behaviour during cooling.
PARACOLD(5,NM)
Rate of decrease of the residual
yield strength when the maximum temperature has
been greater than PARACOLD(4,NM) [N/m²K]
If CMAT = CALCOETC2D, SILCOETC2D
PARACOLD(2,NM)
Poisson ratio.
[-]
Recommended value: 0.2
PARACOLD(3,NM)
Compressive strength
[N/m²]
PARACOLD(4,NM)
Tensile strength (≥0)
[N/m²]
PARACOLD(5,NM)
Strain at peak stress
[-]
Recommended value: 0.0025
PARACOLD(18,NM)
Dilatancy parameter
[-]
Recommended value: 0.25 (0.20 –
0.30)
PARACOLD(19,NM)
Compressive ductility parameter
[-]
Recommended value: 0.19 (0.15 –
0.25)
PARACOLD(20,NM)
Compressive damage at peak stress
[-]
Recommended value: 0.30 (0.18 –
0.32)
Condition: < 0.50
PARACOLD(21,NM)
Tensile ductility parameter
[N/m²]
Recommended value: 400 N/m²
N.B. This parameter can be estimated as 100
[N.m/m²] / where is the area of the shell element.
The CALCOETC2D, SILCOETC2D materials are plastic-damage
constitutive models for concrete. The models are described in:
Gernay T., Millard A., Franssen J.M. (2013). “A multiaxial
constitutive model for concrete in the fire situation: Theoretical
formulation”. Int J Solids Structures, 50(22-23), 3659-3673.
If CMAT = CALCONC2D, SILCONC2D
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Compressive strength
PARACOLD(4,NM)
Tensile strength (≥0)
PARACOLD(5,NM)
< 0 if peak stress strain ec1 = minimum value (stiffer)
= 0 if peak stress strain ec1 = recommended value
> 0 if peak stress strain ec1 = maximum value (more
ductile)
If CMAT = LWCONC2D
PARACOLD(2,NM)
[real]
Poisson ratio
PARACOLD(3,NM)
Compressive strength
[real]
PARACOLD(4,NM)
Tensile strength
[real]
If CMAT = VMRANK2D
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Compressive strength
PARACOLD(4,NM)
Tensile strength (≥0)
If CMAT = BLPLSTRVM
PARACOLD(1,NM)
Young’s modulus
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Yield strength
PARACOLD(4,NM)
slope of the hardening branch
If CMAT = BLPLSTRDP
PARACOLD(1,NM)
Young’s modulus
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Yield strength
PARACOLD(4,NM)
slope of the hardening branch
PARACOLD(5,NM)
a
If CMAT(NM) = ‘BLPLSTRVM’
PARACOLD(1,NM) = E, the Young’s modulus.
PARACOLD(2,NM) = The Poisson’s ratio.
PARACOLD(3,NM) = fp, the limit of proportionality.
PARACOLD(4,NM) = ???
Bi-linear plane stress Von Mises material law. The material is valid
at room temperature.
TRIAXIAL MATERIAL TYPES
If CMAT = STEELEC23D, STEELEC33D
PARACOLD(1,NM)
Young’s modulus.
PARACOLD(2,NM)
Poisson ratio.
PARACOLD(3,NM)
Yield strength
PARACOLD(4,NM)
Maximum temperature for a
reversible behaviour during cooling.
PARACOLD(5,NM)
Rate of decrease of the
residual yield strength when the maximum temperature
has been greater than PARACOLD(4,NM) [N/m²K]
If CMAT = CALCOETC3D, SILCOETC3D
PARACOLD(2,NM)
Poisson ratio.
[-]
PARACOLD(3,NM)
Compressive strength
[N/m²]
PARACOLD(4,NM)
Tensile strength (≥0)
[N/m²]
PARACOLD(5,NM)
Strain at peak stress
[-]
Recommended value: 0.0025
PARACOLD(18,NM)
Dilatancy parameter
[-]
Recommended value: 0.25 (0.20 –
0.30)
PARACOLD(19,NM)
Compressive ductility parameter
[-]
Recommended value: 0.19 (0.15 –
0.25)
PARACOLD(20,NM)
Compressive damage at peak stress
[-]
Recommended value: 0.30 (0.18 –
0.32)
Condition: < 0.50
PARACOLD(21,NM) Tensile ductility parameter
[N/m²]
Recommended value: 400 N/m²
N.B. This parameter can be estimated as 100
[N.m/m²] / where is the volume of the solid element.
The CALCOETC3D, SILCOETC3D materials are plastic-damage
constitutive models for concrete. The models are described in:
Gernay T., Millard A., Franssen J.M. (2013). “A multiaxial
constitutive model for concrete in the fire situation: Theoretical
formulation”. Int J Solids Structures, 50(22-23), 3659-3673.
SERIES 24: Time discretization.
One line, first line of possible multiple line series.
‘TIME’
Time frames.
Two cases are possible:
1) In a dynamic analysis with comeback, a single time step must
be used because the program adjusts itself the time steps during
calculation.
One single line is required:
TIMESTEP, UPTIME, TIMESTEPMAX
TIMESTEP = Initial time step in seconds.
UPTIME = Time for end of the calculation.
TIMESTEPMAX = Maximum value of the time step.
2) In other cases several lines can be given (maximum of
IDIMTIMESTEP lines, = 20 in SAFIR2007).
One line added for each time frame added.
TIMESTEP, UPTIME
TIMESTEP = Time step in seconds.
UPTIME = Limit of validity of this time step.
Time last line.
One line, end of time discretization series.
‘ENDTIME’
SERIES 25: Thermal elongation.
One line, choice of two options.
‘NOEPSTH’
If thermal elongation is not considered.
or
‘EPSTH’
If thermal elongation is considered.
SERIES 26: Output results.
One line, first line of multiple line series.
‘OUTPUT’
One line.
‘TIMEPRINT’
Timeprint frames.
One line added for each timeprint frame added (maximum of
IDIMTIMEPRINT lines)
TIMEPRINT, UPTIMEPRINT
TIMEPRINT Time step for the output of the results.
UPTIMEPRINT Limit of validity of this timeprint.
Timeprint last line.
One line, end of time discretization series.
‘END_TIMEPR’
Output optional results.
Add one line for each option chosen.
‘PRINTDEPL’
The increments of displacement or of
temperatures are written at every iteration.
[A9]
Tstart
Time from which the increments will be written.
[Real]
‘PRINTTMPRT’
elements are written.
The temperatures in the fibres of the beam
‘PRINTVELAC’
The velocity and acceleration are written at
every time step (In a dynamic analysis).
‘PRINTFHE’
The out of balance forces are written at every
Tstart
Time from which the increments will be written.
iteration.
[A9]
[Real]
‘PRINTREACT’
The reactions are written for at every node
where at least one degree of freedom is restrained (by a BLOCK
or a SAME command). The sum of the reactions of all nodes is
also written for each degree of freedom. It allows verifying the total
applied load (except when master-slave relationships are used for
the supports, in which case the results may be confusing because
some reactions are counted several times).
elements.
‘PRINTMN’
Print the internal forces of the beam
Axial forces are positive in tension.
Bending moments calculated as , hence My is positive if
tension prevails in the regions of the section with positive
values of y.
Shear forces calculated as
where My,n is the bending
moment at the last longitudinal point of Gauss in the beam
element, My,1 is the bending moment at the first longitudinal
point of Gauss in the beam element, and DL is the
distance between these two points of Gauss.
Similar for Mz and Vz.
elements.
‘PRNSIGMASL’, NSOL
Print the stresses in the solid
NSOL Number of the solid element where the mechanical
strains, the stresses, the damage in tension and in
compression and the accumulated plastic strain will be
printed. Valid only for a 3D structural analysis.
Note: If NSOL = 0, then the stresses are printed for all
solid elements. This produces a large amount of output.
'PRNEIBEAM'
Print the stiffness EA, ES and EI in the
beam elements
‘PRNSIGMABM’, NBM, NG Print the stresses in a beam element
(positive in tension).
NBM
Number of the beam element where stresses are
NG
Integration point of the beam element where
printed.
stresses are printed.
element.
‘PRINTET’, NBM, NG
Print the tangent moduli in a beam
NBM
Number of beam element where moduli are printed.
NG
Integration point of the beam element where moduli
are printed.
‘PRNEPSMBM’, NBM, NG
Print the mechanical strains in a
beam element (, positive in tension).
NBM
Number of beam element where mechanical strains
are printed.
NG
Integration point of the beam element where
mechanical strains are printed.
‘PRNSIGMASH’, NSH
Print the stresses in a shell element.
NSH Number of the solid element where the
stresses are
printed.
'PRINTSHELL'
Equivalent to 'PRNSIGMASH' for all the
‘PRNNXSHELL’
Print the membrane forces Nx, Ny and Nxy,
shell elements (large amount of results).
N1, N2 and a in the shell elements
‘PRNMXSHELL’
Print the bending moments Mx, My and
‘PRNEASHELL’
Print the membrane stiffness EAx, EAy at
Mxy, M1, M2 and a in the shell elements
the 4 integration points on the surface of the shell elements (in an
elastic element, this stiffness would be ).
‘PRNEISHELL’
Print the bending stiffness EIx, EIy at the 4
integration points on the surface of the shell elements (in an
elastic element, this stiffness would be ).
‘PRNSTRAIN’, EPS_LIM
Print a message (with the strain and
the stress) when the absolute value of the strain in a bar of a shell
element exceeds a certain limit.
EPS_LIM
Limit of the strain that triggers the
message.
Output results last line.
One blank line as last line of series.
3.4 Description and Format of the .IN file for Torsional
Analysis
SERIES 1:
Comments.
One line for each comment (can be 0 line).
SERIES 2:
One blank line to mark end of comments.
SERIES 3:
Number of nodes.
One line.
"NNODE", NNODE
NNODE = Number of nodes of the section.
SERIES 4:
Number of axes.
One line.
"NDIM", NDIM
NDIM = 2 For torsion.
SERIES 5:
Does not exist anymore.
SERIES 6:
Degrees of freedom.
One line.
"NDOFMAX", NDOFMAX
NDOFMAX = 1 For torsion calculations.
Degrees of freedom at all the nodes.
One line.
"EVERY_NODE", NDOF
NDOF ( must be 1 for torsional calculations).
Degrees of freedom at specific nodes.
One line for each group of nodes with specific degrees of freedom.
"FROM", NNO1, "TO", NNO2, "STEP", NNO3, "NDOF", NDOF
NNO1 = First node of this group of nodes.
NNO2 = Last node of this group of nodes.
NNO3 = Node step.
NDOF = Number of degrees of freedom for this group of nodes, 0
or 1.
Note:
The nodes NNO1, NNO1+NNO3, NNO1+2xNNO3,....NNO2-2xNNO3,
NNO2-NNO3, NNO2 have NDOF degrees of freedom
or
‘REPEAT’, NNO1, ‘TO’, NNO2, ‘STEP’, NNO3, ‘TIME’, NT
NNO1 = First node to be repeated.
NNO2 = Last node to be repeated.
NNO3 = Increment.
NT = Number of times that the nodes have to be repeated.
Note:
This command will create the groups:
NNO1+NNO3 ,
NNO1+NNO3+1 , …
NNO1+2*NNO3 ,
NNO1+2*NNO3+1 , ...
....
....
NNO1+NT*NNO3, NNO1+NT*NNO3+1, …
NNO2+NNO3
NNO2+2*NNO3
NNO2+NT*NNO3
Note:
The active nodes where the warping function has to be calculated must have
NDOF = 1
Two options are possible for each node where the warping function must not be
calculated:
1.
declare that the node bears 1 D.o.F., then fix it in the series 6 on
FIXATIONS in the .STR file, see § 3.4.2
2. declare that it has 0 D.o.F., which saves times in series 6 on FIXATIONS.
End of series.
One line.
"END_NDOF"
SERIES 7:
Torsion.
One line.
‘TORSION’
SERIES 9:
Renumbering strategy.
One line, choice of options.
‘NORENUM’ No renumbering of the equations.
or
‘RENUMPERM’ Renumbering of the equations by logical
permutations.
or
'RENUMGEO’, NNO1 Renumbering of the equations by
geometrical method.
NNO1 = Node number where geometrical renumbering will start.
NNO1 = 0 (must be typed) then
renumbering started successively
from all the nodes.
or
‘RENUM’ = RENUMGEO + RENUMPERM
or
‘READRENUM’
Use previous renumbering from .REN file.
SERIES 11: Number of materials.
One line.
‘NMAT’, NMAT
NMAT = Number of different materials.
Note:
If two materials have the same material law but different characteristics, it makes
two different materials. e.g. C20 and C25 concrete.
SERIES 12: Number of different elements.
One line, first line of five line series.
‘ELEMENTS’
Number of different elements, solid elements.
One line, second line of five line series.
‘SOLID’, NSOLID
NSOLID = Number of SOLID elements in the section.
Number of points for integration.
One line, third line of five line series.
‘NG’, NGSOLID
NGSOLID = Number of points of integration in each direction
in the elements, cannot be less than 1. Greater than
3 is not recommended.
Number of voids.
One line, fourth line of five line series.
‘NVOID’, NVOID
NVOID = 0
Last line of series.
"END_ELEM"
SERIES 13: The nodes.
One line, of multiple line series.
‘NODES’ or 'NODES_CYL’
‘NODES_CYL’ is used if cylindrical coordinates are used.
(r,q,Z) and are transformed to (X,Y,Z) for the internal
solution process by the formula:
X = r cos(q)
Y = r sin(q)
Note:
q is in degrees.
The transformation is made after all the nodes have been read and generated.
CYLINDRIC is omitted if the nodes are directly input in the Cartesian system of
coordinates.
NODES
One line added for each node described.
‘NODE’, NNO, RCOORDG(1,NNO), RCOORDG(2,NNO)
NNO = Number of the specific node.
RCOORDG(1,NNO) = First global coordinate of the node NNO.
RCOORDG(2,NNO) = Second global coordinate of the node NNO.
or
‘GNODE’, NNO, RCOORDG(1,NNO), RCOORDG(2,NNO)
NNO = Number of the specific node.
RCOORDG(1,NNO) = First global coordinate of the node NNO.
RCOORDG(2,NNO) = Second global coordinate of the node NNO.
This command is used for automatic equidistant generation
between the previously defined node and node NNO
or
‘REPEAT’, NNO, DELTAC(1), DELTAC(2), KGENE
NNO = Number of nodes to be repeated.
DELTAC(1) = Increment on the first coordinate.
DELTAC(2) = Increment on the 2nd coordinate.
KGENE = Number of times that this command has to be repeated.
Figure 3 : Coordinate order
Note:
The first coordinate corresponds to the local y axis and the second coordinate
corresponds to the local z axis of the beam element.
SERIES 14: Torsional centre.
One line, first line of two line series.
<A10>,<5b>,<G10.0>,<G10.0>
‘NODELINE’, Yo, Zo
Yo = First global coordinates of the node line which joins the
beam elements.
Zo = Second global coordinate of the node line.
Torsional centre.
One line, second line of two line series.
‘YC_ZC’, Yc, Zc
Yc = First global coordinate of the centre of torsion.
Zc = Second global coordinate of the centre of torsion.
SERIES 15: Supports and imposed displacements.
One line, first line of possible multiple line series.
‘FIXATIONS’
One line added for every node where no solution is to be
calculated.
‘BLOCK’, NNO, ‘F0’
NNO = Node number where no solution is calculated (for
example, lines of symmetry).
One line added for each slave node described.
‘SAME’, NNO1, NNO2, CTRAV(1)
NNO1 = Number of the slave node.
NNO2 = Number of the master node.
CTRAV(1) = ‘YES’
One line for repeating previous slave node.
‘REPEAT’, NUMBER, INC, CTRAV(1)
NUMBER = Number of times that the preceding SAME
command must be repeated.
INCR = Increment on NNO1 and NNO2.
CTRAV(1) = ‘YES’
Optional line to create master-slave relationships
between all nodes with same coordinates
‘SAMEALL', ‘YES'
All the nodes of the structure that have the same coordinates (with a
precision of 0.1 mm) will automatically be attributed a master-slave
relationship.
Last line of series.
‘END_FIX’
SERIES 16: SOLID elements.
One line, first line of possible multiple line series.
<A10>
‘NODOFSOLID’
Solid element list.
One line added for each solid element.
‘ELEM’, NE, NODE(1,NE), …, NODE(4,NE), MAT,
EPSRSOLID, KGENE
NSOL = Number of this element.
NODE(1,NE) = First node of this element.
NODE(2,NE) = Second node of this element.
…
NODE(4,NE) = Last node of this element.
MAT = Material of this element.
EPSRSOLID = Residual stress in this element.
KGENE = Allows the generation from the previously defined
element up to this one. KGENE gives the increment
on the nodes number.
or
‘
REPEAT’, NER, INC, NODE(2,NE), …, NODE(4,NE), MAT,
EPSRSOLID, KGENE
NER = Number of elements to repeat.
INC = Increment on the node number.
NODE(2,NE) = Any value ( can be 0 ).
NODE(4,NE) = Any value ( can be 0 ).
MAT = Any value ( can be 0 ).
EPSRSOLID = Residual stress in this element.
KGENE = Number of times that this command has to be
repeated. The element numbers increase by 1.
For triangular elements, NODE(4,NE) = 0.
Note:
The following group of lines on symmetry is necessary if symmetry is accounted
for. If not, only the ENDSYM line is present.
Solid elements symmetry.
One line.
<A10>
‘ SYMMETRY’
Solid element axis of symmetry.
One line for each axis, a maximum of six axes can be specified.
<A10>,<2I5>
‘ REALSYM’, N1, N2
N1 = First node on axis.
N2 = Second node on axis.
Note:
This line means that the line passing by the nodes N1 and N2 is an axis of
symmetry. When creating the .TOR file, the fibres located on the other side of the
line are created. This option is used when there is a thermal axis of symmetry,
which will not be a structural axis of symmetry in the structural calculation.
Solid elements axis of symmetry, symmetric about y axis.
One optional line.
<A10>
‘
YSYM’
Note:
This line is used for plane beam elements, which have this symmetry. When
creating the .TOR file, the area of the fibres is simply multiplied by 2.
Solid elements last line.
One line to mark end of series and symmetry.
<A10>
‘
ENDSYM’
SERIES 11: Precision.
One line.
<A10>,<G10.0>
‘PRECISION’, PRECISION
PRECISION = Small tolerance value reached to have
convergence. A ‘good’ value depends on the type of
structure that is analyzed. 10-3 may be used for the
first simulation to look at the incremental
displacements and if out of balance forces needs a
different value.
SERIES 17: Material description.
One line, first line of possible multiple line series.
<A10>
‘ MATERIALS’
Material description line pair added for each different material used.
One line, first line of two line pair.
<A10>
CMAT
CMAT = Name of the material.
Valid material names are:
'
ELASTIC', ' BILIN',
'
STEELEC3', 'STEELEC3EN', '
STEELEC2', 'STEELEC2EN', '
PSTEELA16',
'CALCONC_EN', 'SILCONC_EN', 'CALCONC_PR', 'SILCONC_PR'
Material description properties.
One line, second line of two line pair.
<G10.0>,<G10.0>,<G10.0>
The value of the following three parameters depends on the material
name introduced in CMAT
If CMAT(NM) = ELASTIC, BILIN, or for STEEL type materials.
PARACOLD(1,NM) = Young’s modulus.
PARACOLD(2,NM) = Poisson’s ratio.
For the CONCRETE type materials
PARACOLD(2,NM) = Poisson’s ratio.
PARACOLD(3,NM) = Compressive strength fc.
PARACOLD(4,NM) = Tension strength, not used here
The Young’s modulus for concrete materials is calculated
according to the formula:
SERIES 18: Output results.
One line, first line of multiple line series.
<A10>
‘OUTPUT’
One line,
‘TIMEPRINT’
Timeprint frames.
One line, second line of multiple line series.
TIMEPRINT, UPTIMEPRINT
<A.10>,<G.10.0>
TIMEPRINT = Any value.
UPTIMEPRINT = Any value.
Timeprint last line.
One line, end of time discretization series.
‘END_TIMEPR’
Output results last line.
One blank line to mark end of series.
<A80>
3.5
E.
Structure of the .TEM files used with the BEAM F.
As many lines as needed
<A80>
Comment lines
1 blank line
<A80>
1 line
<A10,I5>
1 line
<A10>
"NFIBERBEAM"
NFIBERBEAM(NGB) : # of fibres in this section
"
1 line
<A10,2G10.0>
FIBERS"
"NODELINE"
Y0
Series for the position
of the node of the beam element in the (y,z)
system of co-ordinates.
y co-ordinate of the node in the (y,z)
Z0
z co-ordinate of the node in the (y,z) system
system
1 line
"YC_ZC"
Yc
system
Series for the position
of the centre of rotation of the beam element
in the (y,z) system of co-ordinates.
y co-ordinate of the centre of rotation in the (y,z)
Zc
z co-ordinate of the centre of rotation in the (y,z)
system
NFIBERBEAM lines
<3E13.6,I5,E13.6>
RCOORDYZINBEAM(1,NFB,NGB) : y co-ordinate of this fibre
RCOORDYZINBEAM(2,NFB,NGB) : z co-ordinate of this fibre
FIBERSECTION(NFB,NGB)
: cross sectional area
of this fibre
MATBEAM(NFB)
: local number of the material
present in this fibre
EPSRBEAM(NFB,NGB)
: residual stress in this fibre
1 group of lines
if this group is absent, then the warping function is equal to 0 on the cross section
1 line
<A10>
"
w"
NFIBERBEAM(NGB) lines
<F12.6>
WARPING(NFB,NGB) : value of the warping function in this fibre
1 line
<A8,E12.6>
"
GJ="
GJ(NGB) : torsional stiffness of the cross section
End of the group
1 line
<A10>
Either
"
or
COLD" : This section is not
heated. The temperature in
all fibres remains at 20°C.
===> The .TEM file can be
ended here. No need to
write the next groups (
TIME, NFIBERBEAM )
necessary
"
HOT" : This section is heated. The next groups are
1 group of lines
Repeat this group of lines for each time step
1 blank line
<A80>
1 line
<7x,F8.1>
"
TIME"
1 blank line
<A80>
NFIBERBEAM(NGB) lines
<5x,F6.1>
TEMPBEAM(2,NFB,NBG) : temperature in the fibre NGB
3.6
E.
Structure of the .TSH files used with the shell F.
SERIES 1 : Comments.
Any number of lines
comment cards
<A80>
1 blank line indicating that the comments are
finished
SERIES 2 : Thickness of the shells.
1 line
- "THICKNESS"
<A9>
- THICKSHELL
<Real>
Thickness of this section type
SERIES 3 : Material of the shells.
1 line
- "MATERIAL"
<A8>
- MAT
<Integer>
Local material number of this section
type. This is the material of the plain
section, to which layers of re-bars can be
added.
SERIES 4 : Layers of re-bars.
1 line
- "REBARS"
<A6>
- NBARS
<Integer>
# of re-bar layers in this
section type.
NBARS groups of cards.
1 card.
- "MATERIAL"
<A8>
- MAT
<Integer>
Local # of the material of this layer
1 card.
- "SECTION"
<A7>
- A
<Integer>
Cross sectional
area of this layer (in m²/m)
1 card (optional)
- "SIGMA_RES"
- sigma
initial
stress
of the
bars in
this
layer
(N/m²)
Notes:
1)Residual stresses can be introduced in the bars if
the bars are made of steel. If not, the stress can be
entered as O or this card can be omitted.
2)Positive values are given for tension in the bars.
3)This model is valid for pretensioning. The stress
in the bars will be reduced during the first time
step because of the elastic shortenings developing in
concrete.
4)For introducing residual stresses in the material
of the plain section, see Section 3.7 ‘Instructions
for introducing residual stresses in steel shell
elements’.
1 card.
- "LEVEL"
<A5>
- z
<Integer>
Position of this layer with
respect to the thickness
For each bar layer, there are two methods to give the
orientation of the bars in the plane of the element.
Method 1: with respect to the local system of coordinates
of each element.
1 card.
- "ANGLE"
<A5>
- angle
<Real>
Angle in degrees between the local x
axis and the layer of
rebars, see Figure below.
Figure 4 : Orientation of rebars in shell F.E. in local axes
Method 2: with respect to
coordinates of the structure.
the
global
system
of
1 card.
- "NORMAL"
<A6>
- N1
<Real>
- N2
<Real>
- N3
<Real>
< N1 ; N2 ; N3 > is a vector in the global
system of coordinates of the structure. The norm
of the vector does not have to be 1.
This vector is used to define the position of
the bar layers in the shell elements with
respect to the global system of coordinates
according to the following technique, see figure
below.
The bars have the orientation of the line which is the
intersection
between the shell element and a plane that is
perpendicular to the normal.
If the norm of the vector is 0, then the orientation of this bar
layer is perpendicular, in each element, to the previous bar layer
(not possible for bar layer 1).
SERIES 5 : Cold or Hot section
1 line
Either
"
COLD"
<A10>
This section is not heated. The temperature in the shell remains at 20°C. => The
.TSH file can be ended here. No need to write the next series of cards
or
"
HOT"
<A10>
This section is heated by a time-temperature curve (the same curve for all the
elements of this type). The next groups are necessary
or
"TuserShell"
<A10>
This section is heated according to a function that has been programmed by the
user in the DLL called SAFIRDLL.DLL.
No need to write the next groups
or
"
HASEMI"
<A10>
This section is heated by a Hasemi fire. The next groups are necessary.
SERIES 6 : position of the nodes
1 line
<A24>
- " POSITIONS OF THE NODES."
1 line
<A25>
- " ======================="
1 line
<A21,I4>
- " NUMBER OF POSITIONS:"
- Number_of_positions
Gives the number of nodes which give the temperature of
the slab across its thickness. The positions of these
nodes only depends on the discretisation which was chosen
when the temperature distribution was calculated. It is
independent of the location of the integration points
across the thickness which will be used in the structural
analysis.
1 line
<Number_of_positions Real>
- position of the first node (the one with the
smallest z coordinate)
- position of the second node
- etc
- position of the lest node (the one with the
highest z coordinate)
Series 7 : temperatures
Repeat this group of lines for each time step
1 blank line
1 line
<A6, F12.4>
- " TIME="
- TIME
are given
value of the time when the temperatures
1 line
<A6>
- " ====="
Number_of_position lines
<2 Real>
- position of the node (same as in series 6)
- temperature at this node
The example underneath gives the beginning of a TSH file that is 60 mm thick
and has two layers of rebars of 322 mm²/m located at mid depth of the slab and
perpendicular to the global Y axis (first layer) and X axis (second layer).
3.7 Instructions for introducing residual stresses in
steel shell elements
This section explains how to consider residual stresses in the first material
of the shell element. This procedure has been foreseen to introduce residual
stresses in steel plates, not in the rebars of reinforced concrete slabs. For the
latter, see SERIES 4 in Section 3.6 ‘Structure of the TSH files’.
In order to introduce residual stresses, it is necessary to create a file with
the name “resi_str_shell.txt”, which must be placed in the same folder as the
structural input file.
In this file, the residual stresses should be placed in Pascal, for each point
of integration of the elements in which there are residual stresses.
Figure 5: Position of the integration point in the plane
There are 4 points of integration in the plane of the element located as
indicated by Figure 1. The integration in the plane is by the method of Gauss.
The number of integration points on the thickness NGTHICK is chosen by
the user, from 2 to 9. The integration on the thickness is also by the method of
Gauss.
Format of the file “resi_str_shell.txt”:
One line for shell element.
'ELEM', NE,
NE = Number of this element.
One line for each gauss point of the shell element. There are (4 x
NGTHICK) Gauss points.
'PG', NPG, Sx, Sy, Sxy
NE = Number of this gauss point.
Sx = Residual strength in x axis.
Sy = Residual strength in y axis.
Sxy = Residual tangential strength.
or for automatic generation of the gauss points
‘Gpg’, NLPG, KGENE
NLPG = Number of the last gauss point with automatic generation.
KGENE = Step for the automatic generation of the gauss points.
or for automatic generation of the elements
‘GELEM’, NLE, KGENE
NLE = Number of the last element with automatic generation.
KGENE = Step for the automatic generation of the elements.
Example:
ELEM 1
PG 1
100000
PG 2
100000
PG 3
100000
PG 4
100000
PG 5
100000
PG 6
100000
PG 7
100000
PG 8
100000
ELEM 2
PG 1
-100000
PG 2
-100000
PG 3
-100000
PG 4
-100000
PG 5
-100000
PG 6
-100000
PG 7
-100000
PG
8
-100000
ELEM 3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
PG
PG
PG
PG
PG
PG
PG
PG
ELEM
PG
PG
PG
PG
PG
PG
PG
PG
1
2
3
4
5
6
7
8
4
1
2
3
4
5
6
7
8
-100000
-100000
-100000
-100000
-100000
-100000
-100000
-100000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
100000
100000
100000
100000
100000
100000
100000
100000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Or with automatic generation:
ELEM
1
PG 1
Gpg 8
GELEM 4
ELEM 2
PG 1
Gpg 8
GELEM 3
100000
1
3
-100000
1
1
0
0
3.8 Structure of the temperature files used with the
truss F. E.
As many lines as necessary, each line being, in a free format, a pair of values in
the form:
TIME
TEMPERATURE
Example:
600.
1200.
1500.
1800.
3600.
0.
300.
800.
1000.
900.
20.
20.
20.
600.