ATENA Engineering 3D Tutorial
ATENA Engineering 3D Tutorial
Na Hrebenkach 55
150 00 Prague
Czech Republic
Phone: +420 220 610 018
E-mail: [email protected]
Web: http://www.cervenka.cz
Written by:
Jan Červenka, Zdenka Procházková,
Tereza Sajdlová
1. INTRODUCTION 1
2. STARTING PROGRAM 3
3. PRE-PROCESSING 5
3.1 Introduction 5
3.1.1 Introduction of The Graphical User Interface 6
3.1.2 Definition of the Geometrical Model 9
3.1.3 Saving of Data 10
3.6 Bar-reinforcement 42
3.6.1 First Bar 42
3.6.2 Second Bar 46
4. FE NON-LINEAR ANALYSIS 67
4.1 Introduction 67
5.1 Introduction 75
6. CONCLUSIONS 89
8. LITERATURE 93
ii
1. INTRODUCTION
This tutorial provides a basic introduction to the usage of the program ATENA 3D, and it
is specifically targeted for ATENA 3D beginners. This tutorial contains a step by step
explanation how to perform a non-linear analysis on an example problem of a reinforced
beam without smeared reinforcement. The geometrical and material properties correspond
to the experimental setup by Leonhard in 1962. More details about the problem or
experiment can be also obtained from the original report [5] or from the program developer
or distributor.
The step by step demonstration is performed on an example of simply supported beam,
which is loaded by two loads as it is shown in Figure 1. The problem is symmetric around
its vertical axis; therefore, only one symmetric half of the beam will be analyzed.
The steps necessary for the data preparation, non-linear analysis and post-processing are
depicted on subsequent figures, which show the computer screen for each step and user
action. There is always also a short description for each figure. In the post-processing stage
only some basic post-processing methods are described. ATENA offers many options for
viewing results from FE non-linear analysis. These options can be easily accessed from the
post-processing window by self-explanatory buttons and toolbars. For more details, it is
recommended to consult the ATENA 3D user's manual or the hotline desk at the program
distributor or developer.
Main Menu
File toolbar:
6
Free Zoom out Zoom Zoom
rotation at mouse window extend
location to view
Translation and rotation
Select
Selection toolbar: by
crossing
Invert selection
Select by allmode
Selection objects of the
skewed selected type
rectangle add, remove, invert
8
3.1.2 Definition of the Geometrical Model
After examination of the user interface layout, it is possible to start with the definition of
the geometrical model of the analyzed structure. It is a good practice to provide a short
description of the problem to be analyzed. In ATENA 3D this can be done by selecting the
General data | Analysis information item in the Input data tree.
In this tutorial problem, the Input for global structural parameters as well as solution
parameters is shown in Figure 4 and Figure 5.
Figure 7: The Materials table, from which new materials can be added or existing materials can be modified or
removed.
Clicking the Add button on the material table window creates a new material. For the
current problem, it is necessary to define three material types: one elastic material for the
steel plates at support and loading points, concrete material for the beam and reinforcement
material. There exist three methods for creating new materials (see Figure 8). A new
material can be defined directly using various ATENA material models, or a previously
saved material definition can be used. The third method is to use a material definition from
the available catalogue of materials. The catalogue contains various material definitions
based on the various national or international standards. For the purpose of this tutorial
let’s use the direct definition.
10
3.2.1 Steel Plates
Figure 8: After selecting the Add button it is possible to specify how the new material will be created.
Figure 10: The dialog for the definition of material properties for the steel plates.
Figure 12: The dialog for the definition of reinforcement material parameters. The bi-linear elastic-perfectly
plastic stress-strain diagram is selected for this problem.
Parameter input:
Type: Bilinear
Elastic modulus: 208 000 MPa
σy : 560 MPa
12
3.2.3 Concrete Beam
Figure 13: Selection of 3D Nonlinear Cementitious 2 material model for the concrete beam.
Parameter input:
Cubic fcu: 33.5 MPa
Figure 14: Default values of material parameters are generated based on the cube strength of concrete. For this
case, the cube strength should be 33.5 MPa.
NOTE: There are predefined parameters in dialog windows for the definitions of
parameters. The table named “Parameter input:” shows the parameters which should be
changed.
Parameter input:
Tensile strength ft: 1.64 MPa
Figure 16: The dialog window for the tensile properties for the concrete material.
Parameter input:
Specific fracture energy Gf: 5.5e-5 MN/m
14
Figure 17: The dialog window for the compressive properties of concrete material.
Figure 18: The dialog window for the miscellaneous properties of concrete material. In this window it is
recommended to verify that the fixed crack model coefficient is set to 1.0.
Figure 19: The three materials, which were defined previously, can be viewed or modified from the Materials
table window at the bottom part of the ATENA 3D window.
16
Buttons to switch between geometry Grid toolbar can be used to
view and macroelement properties modify the work-plane and
dialog. grid settings.
Input data access tree activates the This toolbar is used for
definition of various entities. selection of perspective/
isoparametric projection or
Selection of visible entities. quick view changes.
Figure 21: The parameters for the entity generation can be modified by clicking the highlighted items. The
numerical input is finalized by clicking the checkbox button.
Parameter input:
entity type:
Entity type: column/beam,
column/beam
coordinate system:
Coordinate system: global,
global
reference point: (0,0,0)
Reference
size
point: (V (0,0,0)
=1.275,V =0.190,V =0.320)
x y z
18
Immediately after all parameters for the beam are specified, the preview of the beam
geometry is visible in the main window. In some cases it is necessary to zoom in to see
clearly the beam geometry. This can be accomplished by selecting the Zoom extend
button .
Zoom extend
button to fit the
structure into the
window
Figure 22: Preview of the specified parametric entity before it is created. In some cases the previewed structure
may be very small in the middle of the screen. The display can be fitted to the whole window by selecting the
Zoom extend button .
The preview in this case shows that the beam geometry is correct so it is possible to
finalize the beam definition by clicking the Generate button in the right part of the
generation window. The resulting display is shown in Figure 23.
Figure 23: Program display after the generation of the parameterized beam element.
Figure 24: The material type for the generated beam is specified by clicking the Properties button in top
left corner of the macro-element definition window.
In this case “Concrete” is the correct material for the created beam.
In this window, it is also possible to specify smeared reinforcement. This is one of the two
possible methods for reinforcement modeling that are implemented in ATENA.
Reinforcement can be modeled either by modeling each individual bar or in an average
sense by reinforcing a macro-element in certain directions by specifying an appropriate
reinforcement ratio [%]. This type of reinforcement model is called “smeared
reinforcement” in ATENA, and it can be inserted into each macro/element by selecting the
button in the smeared reinforcement section of this window. The smeared
reinforcement feature is useful especially for modeling reinforcement mats or stirrups. The
analyzed beam, however, is without stirrups so the smeared reinforcement feature will not
be used, and therefore, the smeared reinforcement list should be left empty.
20
After the definition of material model for the created beam it is possible to finalize the
macro-element definition by returning back to the Topology tab and by selecting the
button in the most bottom-right corner of the macro-element definition window.
Please note that this is a different button than the one used for the definition of smeared
reinforcement that was discussed before.
At this point the created beam is added to the model, and it is possible to start defining
other macro-elements. It is possible to look at the so far created model by selecting the
button at the most bottom-right part of the macro-element definition window to
return to the main program view (Figure 25).
Figure 25: The program display after the definition of the first macro-element.
22
3.4.1 Grid Setting
When defining two-dimensional macro-element cross-section it is advantageous to utilize
the working plane grid. The grid settings are controlled by the toolbar shown in Figure 27.
Select the button for the manual specification of grid properties, and set the grid
specifications according to Figure 28. The plate cross-section should be defined in the
plane XZ and since the plate dimensions are 0.03x0.10, the following values for dx = 0.05
and dz = 0.03 should be used (see Figure 28).
Parameter input:
Origin: Step:
X: 0.165 X: 0.050
Y: 0.000 Y: 0.500
Z: 0.320 Z: 0.030
Working plane: XZ
The grid origin should be moved to (0.165, 0, 0.32) in order to place the grid origin into
the center of bottom part of the loading steel plate. At this point it is more convenient to
modify the view of the structure in order to start defining the support plate geometry. First
let’s change the view such that the structure is viewed from the negative Y axis by
selecting the button . This view is perpendicular to the beam geometry as well as the
grid plane. By selecting the zoom extend button , the display of the whole beam
appears. It is also more convenient to switch to parallel view by clicking the button . If
the parallel projection is selected the exact projection of the beam geometry into the X-Z
plane is obtained.
Figure 30: The table of joint’s coordinates which can be used in the case the grid is not displayed correctly.
24
3.4.3 Lines Definition
The next step is to connect these joints via lines. The line input is activated by the item in
the Input data tree Entities | Lines | Add in the Input data tree on left side of the program
window. Each boundary line is defined by first selecting the beginning joint and then the
end joint. It does not matter in which order are the joints selected, however, for subsequent
definition of surfaces, it is important that the created lines form a closed surface. The
process of line definition is shown in Figure 31, and the Figure 32 depicts the program
display after all boundary lines are created. Altogether 5 lines need to be created to form a
closed surface. Naturally, the next step is to create the surface that will represent the cross-
section of the loading plate.
Figure 32: Program display after the definition of boundary lines for the loading plate cross-section.
26
3.4.4 Surface Definition
The surface definition starts in a similar manner by selecting the item Entities | Surfaces |
Add in the Input data tree. This activates the interactive surface definition. The surface is
defined by selecting the lines that form its boundary (see Figure 33). In this case, it is not
necessary to select all the boundary lines forming the surface, since the program
immediately recognizes that in this special case only one solution is possible to create a
closed surface, and automatically includes the other lines into the surface definition.
Figure 34: The final display after the first surface definition.
At this point it should be also noted that a surface could be also defined manually by
specifying the ID numbers of each line forming the surface. The manual input is always
activated from the table in the bottom of the macro-element definition window.
28
3.4.5 Extrusion
The macro-element for the loading plate can be now created by extruding this surface in
the Y direction over the beam thickness (i.e. 0.19 m). The extrusion process can be started
from the Input data tree item Generation | Extrusion. When this item is selected a
window appears along the bottom part of the macro-element definition window, in which
the extrusion parameters can be specified (see Figure 35).
Parameter input:
Surface: number1
Direction: global Y axis
Size: 0.190 m
Each extrusion parameter can be modified by clicking the highlighted fields. The
modifications are saved by hitting the Enter key or by clicking the button .
Immediately after a meaningful set of extrusion parameters is specified a preview of the
generated region can be seen in the main window. If needed, the rotation button can be
used to rotate the structure slightly to get a better view of the new entity.
Figure 36: Preview of the new entity for the modeling of the loading plate.
30
3.4.6 Material Definition
The next step is to assign material properties to the newly created entity. This is
accomplished by switching to the Properties tab of the macro-element definition (see
Figure 38). Here the previously created material “Steel plates” is to be selected.
Now, the macro-element for the modeling of the loading plate is fully defined, and it is
possible to include it into the global model by clicking the button in the most
bottom-right corner of the macro-element definition window. At this time, it is possible to
exit the macro-element definition window, and return to the main program window by
selecting the button. This button is also located at the bottom right corner.
Figure 39: Main window after the definition of the loading plate.
Then it remains to create the support plate on the left side of the beam. Since the geometry
of this plate is identical to the loading plate that we have just created, it should be
advantageous to exploit the copy and rotate options of ATENA 3D.
Figure 40: Selection of loading plate macroelement for copy operation to create a similar macroelement for
support plate.
32
The Figure 41 shows the recommended set of parameters. Once a valid set of parameters is
selected a preview of the new macro-element appears in the main graphical window. If the
preview shows that the new macro-element has been created at the correct position (see
Figure 41), it is possible to click the button Copy at the right bottom corner of the program
window. This action will actually create the new macro-element at the specified location.
It should be noted that no new macro-element is created if the Copy button is not selected.
Figure 41: Shifting the copy of the loading plate along the x axis by the distance of 0.810 m.
Parameter input:
shift value: 0.810 m
3) Select
Move item
1) Deselect the
macroelement 2
2) Select the
macroelement 3
Figure 42: The next step is to move the new macro-element 3, such that it occupies the correct location along
the bottom edge of the beam. Move | Mirror operation is used to accomplish that.
34
This action again changes the content of the bottom window, which now contains the
parameters for the move operation. There are several possible move methods: shift,
rotation and mirror. In this case, it is advantageous to use the mirror method. The mirroring
should occur with respect to the XY plane that should be shifted by 0.16 m along the Z
axis from the origin.
The correct definition of the move parameters is shown in Figure 43, which also shows the
display of the main graphical window after the above set of parameters is used. The
graphical window also shows the preview of the new location of the macro-element 3. If
the correct position is verified it is possible to press the Move button to actually perform
the mirror operation.
Figure 43: Mirror operation for moving the support plate macro-element into the correct position.
Parameter input:
Operation: mirror
Shifted to (Z): 0.16 m
Figure 44: The bottom view of the beam with the support plate.
36
At this point, all geometry is defined. The program automatically recognizes all possible
contacts among the existing macro-elements. It is possible to visualize the recognized
contacts by selecting the item Topology | Contacts in the Input data tree. In order to
properly see the generated contacts it is recommended to deactivate the display of surfaces
in the View window at the bottom left corner of the program screen (see Figure 45). By
editing the contacts it is possible to specify special contact conditions such as for instance
nonlinear interface behavior. In this problem, perfect connection is assumed, which is the
default contact setting, so no editing is necessary.
Figure 45: The program automatically recognizes existing contacts among the created macro-elements.
Parameter input:
Global element size: 0.05m
38
The table in this window shows three items, one for each macro-element (Figure 46).
There it is possible to select for which macro-elements the automated mesh generator is to
be started, which generator is to be used (currently only one generator T3D is available),
and what kind of elements are to be generated: linear or quadratic.
Linear elements are low order elements with nodes at each element corner. Quadratic
elements usually have additional nodes on each element edge. Some quadratic elements
may have even nodes in the middle of element sides or inside the element.
In this case, linear elements will be used, and it is recommended to use only brick elements
whenever possible. The close examination of the existing macro-elements clearly shows
that only the first macro-element (i.e. the beam) can be meshed with brick elements. For
the steel plates the tetrahedral elements will have to be used. Even though the linear
tetrahedral elements are generally not recommended for stress analysis, they can be used in
this case for modeling the steel plates, since an accurate modeling of stresses and
deformations in these areas is not as important as the modeling of the concrete beam itself.
The meshing parameters for the beam macro-element are modified by selecting the item
FE Mesh | Macroelements | Add in the Input data tree. This again changes the bottom
table window and opens the dialog that is shown in Figure 47 for defining a prototype of
macro-element mesh properties. Here the “brick” mesh type should be selected. After
clicking OK button, this prototype can be assigned to the macro-element 1 representing the
concrete beam (see Figure 48).
Notes on meshing
The finite element mesh quality has a very important influence on the quality of the
analysis results, the speed, and memory requirements. Refining only the important parts
can save a lot of processor time and disk space.
A bad mesh, like a single layer of volume elements in a region where bending plays a
significant role, can produce very wrong results – see the "Mesh Study" example in the
ATENA Engineering Example Manual. A minimum of 4-6 elements per thickness is
recommended for at least qualitative results in bending. Alternatively, shell elements may
be used (see section Shell Macroelements in the User’s Manual for ATENA 3D).
Other items in the FE Mesh section of the Input data tree allow the user to define similar
mesh parameters for joints, lines, surfaces or contacts. They can be used to specify certain
areas with mesh refinement. However, if areas with mesh refinement are selected, it is
often impossible to mesh the adjacent regions with hexahedral elements (i.e. brick
elements), and tetrahedral finite elements must be used instead.
In the case of contacts, it is possible to enforce compatible meshes on both side of the
contact. In general case, ATENA supports contacts with incompatible meshes, but this
feature should not be used if it is important to properly model stresses and deformation in
the contact area. In the analyzed case, the contact regions between concrete beam and steel
plates should not have a great influence on the beam behavior, so it is not necessary to
enforce the full mesh compatibility on the two contacts. Due to this assumption, it is also
possible to mesh the beam with brick elements and the plates with tetras. This greatly
simplifies the model definition, but it is necessary to understand that this will result in
certain incompatibilities in the displacement field on these contacts. In this case, it is not a
big problem, since in reality the connection among the steel plates and concrete would not
be perfect as well.
40
At this point, it is possible to generate the finite element mesh by selection the button
Generate. This button is visible from the main mesh generation window that is accessible
by selecting the FE Mesh | Generation in the Input data tree.
Mesh incompatibilities
The close examination of the mesh that was created at the contacts of the steel plates with
the concrete beam clearly shows that the meshes in the neighboring regions are not
compatible. This is due to the fact that we have not enforced this compatibility as it was
discussed in the previous paragraphs. The incompatible meshes should be used with great
care since the results close to these regions have lower level of accuracy. The program
internally applies certain special constraint conditions to enforce a proper connection of
these regions, but such a connection is less accurate than in the case of compatible meshes.
Figure 50: The program window at the beginning of the reinforcement bar definition.
42
In this example, there are two reinforcing bars along the bottom side of the beam with
diameter 26 mm. The bar distance from the beam bottom surface is 0.05 m. In this case,
the bar definition will start by defining the first bar, which will be then copied to create the
second one.
The definition of the first bar will start by direct numerical definition of the coordinates for
the bar beginning and end. The numerical definition is activated by selecting the Add
button on the right from the Table of joints along the bottom part of the screen. This opens
a dialog that is shown in Figure 51, into which the coordinates of the two joints should be
specified. The first joint should have the coordinates: (0.0; 0.05; 0.05) and the second one:
(1.225; 0.05; 0.05).
Figure 51: The coordinates of the first and second joint. The Add button should be selected after the definition
of each joint.
Parameter input:
Number: 1 Number: 2
X: 0.000 X: 1.225
Y: 0.050 Y: 0.050
Z: 0.050 Z: 0.050
Figure 52: The program view after the definition of the two reinforcement joints.
Figure 53: The creation of the first reinforcement bar by selection of the first and second bar joint.
44
When the geometry of the first bar is defined,
the Properties tab should be selected to specify
the remaining properties.
Figure 55: The reinforcement bar properties. The cross-sectional area can be comfortably calculated by using the
available area calculator.
Parameter input:
Material: (2) Reinforcement
Bar diameter: 0.026 m
Number of bars: 1
Figure 56: Before exiting the bar definition window it is important to add the created bar into the global model.
After the return into the main program window, it is recommended to deactivate the
display of surfaces and the FE mesh in order to see the reinforcement in the interior (see
Figure 57).
The next step is to select the reinforcement bar for copying. This process is described step
by step in Figure 58. If the three steps that are described in this figure are performed, the
appearance of the bottom window changes and it can be used now to define the parameters
necessary for the copy operation. This process as well as the copy parameters to be used
are shown in Figure 59.
46
The new reinforcement
bar is here.
Figure 57: The global view of the model with the new reinforcing bar.
Parameter input:
Operation: shift
Dir.: Y axis
Shift value: 0.09 m
Now all the reinforcement is generated. Later on when the analysis is started, the program
will decompose each reinforcement bar into individual truss finite elements, which will be
embedded into the solid elements. In this way the bar stiffness will be included into the
numerical analysis. This process is however fully automatic and the user does not have to
deal with it. The automatically created individual truss elements will be visible in the post-
processing phase of the analysis.
48
3.7 Supports and Actions
This section describes the definition of supports and loads for this example problem.
The analyzed beam is supported at the bottom steel plate in the vertical direction. Since we
are analyzing only a symmetric half of the beam, it is necessary to enforce the axis of
symmetry along the right side of the beam. This means that the horizontal x-displacements
along this side should be equal to zero.
The beam is loaded at the top steel plate. We are interested in determining the maximal
load-carrying capacity of the beam, which means we want to be able to trace the structural
response also in the post-peak regime. The easiest method to accomplish this is by loading
the beam by prescribed displacements at the top steel plate. It is also possible to apply the
loading by vertical forces, which will be increased in each load step. In order to be able to
go into post-peak, advanced non-linear solution strategies such as Arc-length method
would be necessary. Such techniques are available in ATENA 3D, but they will not be
used in this example, where Newton-Raphson method and displacement load control is
sufficient and will provide more robust results.
A loading history in ATENA 3D is defined in analogy to previous versions ATENA 2D
and SBETA. This means that first load cases are defined, and then they are combined
together to form a loading history for an analyzed structure. In ATENA each loading step
then represents a loading increment, which is added to the previous loading history.
The load-case definition starts by highlighting the Loading | Load cases item in the Input
data tree and clicking the Add button in the Load cases tables (Figure 60).
Figure 61: The first load case will contain the horizontal and vertical supports.
Figure 62: The second load/case will contain the prescribed deformation at the top steel plate.
50
The table along the bottom part of the program window shows the list of created load-cases
(see Figure 63). Each of them can be edited or deleted by selection the appropriate buttons
on the right side of this table. The active load-case is selected using the Load case dialog
above the Input data tree (see Figure 64). At first the load case “Supports” should be
selected. When the load-case is active, it is possible to start defining its boundary
conditions. The definition of the symmetric boundary condition is described in Figure 65,
and the application of the vertical support at the bottom steel plate is shown in Figure 67.
Figure 63: The list of created load-cases in the Load cases table.
Figure 64: An appropriate active load case must be selected prior to the support definition. Supports should be
in the load case 1.
Figure 65: The definition of the horizontal support at the right side of the beam.
52
2) In this dialog the
1) Select attributes for the required
Loading|Lines|Add
boundary condition are to
item in the Input data tree. be specified: support in y
and z directions. Then
click OK (see Figure 68).
Figure 67: The definition of vertical support along the bottom steel plate. If necessary the button can be
used to rotate the structure.
It should be noted that the support steel plate is fixed also in the y direction in order to
prevent any rigid body displacements.
Next, the second load case should be activated in a similar manner to the first one as it was
shown in Figure 64. Now it would be advantageous to apply the prescribed displacement to
54
The button Add allows the manual
specification of the coordinates for the new
joint in the middle-top of the loading steel
plate.
Parameter input:
Coordinate:
X: 0.165 m
Y: 0.095 m
Z: 0.350 m
Recommended Method
ALTERNATIVE METHOD: Sometimes in the case of too coarse grid, the program cannot
automatically recognize that the joint lies on the line. The joint has to be part of the all contour
lines of the macroelement. Therefore it is necessary to add joint in a different way. It is done by
removing the line (surfaces will be automatically removed with the line) lying on the top loading
plate and then add the new joint, lines and surfaces again.
This is accomplished by selection the item Topology | Macroelements | Edit in the Input data
tree. Then the macro-element 2 should be selected and press button Edit. This brings up the
window for macro-element editing that is shown in Figure 69. In this window the line is removed
by selection the item Entities | Line | Remove in the input data tree and then by selected line.
Neighboring surfaces will be removed automatically with the removed line.
3) Specify condition
2) Select the item attributes: support in z,
Loading|Joints|Add w = -0.0001 and click
OK button.
Figure 70: The definition of the prescribed displacement at the top steel plate in load case 2.
Parameter input:
Support in Zg: fixed WZg: -0.0001 m
56
The new set of solution parameters can be defined by selecting the button Add on the right
side of this table. In this example, a new set of solution parameters called “My N-R
parameters” will be created as depicted in Figure 71, Figure 72, Figure 74 and Figure 75.
Figure 71: The first property sheet for the new set of solution parameters for Leonhardt's beam analysis.
Parameter input:
Title: My N-R parameters
Solution Method: Newton-Raphson
Parameter input:
Break after step
Displacement error
multiple: 10
Residual error
multiple: 10
Abs. residual error
multiple: 10
Energy error
multiple: 1000
Figure 73: The third property sheet Conditional break criteria can be set to stop the computation if an error
exceeds the prescribed tolerance multiplied by the prescribed factor during the iterations or at the end of an
analysis step.
58
After the required solution properties are prescribed, the Add button will include the new
solution properties into the list of all solution properties of this problem as is shown in
Figure 74.
Figure 74: The table with the newly created solution parameters.
Figure 75: Load steps are specified using the button Add from the table of Analysis steps. This table
appears in the table window after highlighting the Run | Analysis steps item in the Input data tree.
60
Parameter input:
Load cases: 1-2
Number of added load steps: 40
Figure 76: The Analysis steps table after the definition of 40 load steps with the above
parameters. It is possible to add more load steps later during the analysis.
For each analysis step, it is necessary to select the load cases, which should be applied,
solution parameters and a multiplier that is used to scale all forces or prescribed
displacements for the given step. Load case numbers should be separated by comas or
dashes. A dash means that all load cases between the given numbers are to be applied in
this step. It is always possible to add, insert or remove steps from this table. However, once
a step is inserted before a step that had been already analyzed, the results for analysis steps
after the inserted step will be lost. If an analysis shows that a required load level or failure
had not been reached, it is possible to add more load steps and continue with the analysis
up to failure.
62
Monitor’s name for easy identification.
It is important to specify in which macroelement the monitor is located. In this case the
first monitor is located in macroelement 1.
Parameter input:
Title: Deflection
Type: Value at node
Value: Displacement
Item: Component 3
Macroelement: 1
Closest to the point: X: 0.0000 m
Y: 0.0950 m
Z: 0.0000 m
Figure 79: The definition of the second monitoring point for modeling reactions, i.e. loads.
It is important to specify in which macroelement the monitor is located. In this case the
first monitor is located in macroelement 2.
Parameter input:
Title: Load
Type: Value at node
Value: Reactions
Item: Component 3
Macroelement: 2
Closest to the point: X: 0.1650 m
Y: 0.0950 m
Z: 0.3500 m
64
These two monitoring points will allow us to monitor the load-displacement curve during
the non-linear finite element analysis. It makes it possible to see the changes of actions and
displacement at each load step and even at each iteration. The program display after the
definition of the monitoring points is shown in Figure 80.
Figure 80: The program display after the definition of monitoring points.
The finite element analysis is started using the button in the top right part. After
clicking this button, the program will start to generate the input files for each step of the
non-linear analysis. This process is indicated by a progress bar showing the status of this
operation. These input files are stored only the program memory and will not appear in the
current working directory.
Figure 81: The interactive window for monitoring the progress of non-linear analysis.
The graph window on left part of the screen shows the development of monitoring point
values. By default this window shows the evolution of convergence criteria of the non-
linear solution algorithm.
In most cases, it is desirable to modify this window such that it can also show the graphical
evolution of a load-displacement diagram. Such a diagram usually plots deflection on the
horizontal axis and loads on the vertical axis.
In the pre-processing stage, two additional monitors had been defined: one for monitoring
deflections and the other one for monitoring reactions. It is useful to modify the graph on
the left side of the run-time window such that it shows the development of these
monitoring points during the analysis.
68
The contents of the graph window can be modified by pressing the button above the
graph window. This action opens a dialog window that is shown in Figure 82. Here it is
possible to select the monitoring data that are to be displayed on the horizontal and vertical
axis. The “deflection” monitor should be selected for the horizontal axis. Then for the
vertical axis, it is necessary to first select the units. The reactions have the units of force,
i.e. MN. When a proper unit is selected the contents of the bottom part of this dialog
changes, and it is possible to select the monitor “Load”.
Figure 82: The dialog for modifying the graph display in the left part of the run-time window.
Parameter input:
X axis
Axis label: Displacement
Value: Deflection
Switch axis orientation
Y axis
Axis label: Load
Value: [MN]
Switch axis orientation
83 and
Figure 86. When the OK or Apply button is selected the content of the graph window
changes as is shown in Figure 83.
It is useful to save these graphs settings, by clicking the button above the graph
window and name it “LD”. This enables the saving of the current graph settings under a
user defined name. The saved graph settings are accessible from the list box above the
graph window, and they become available every time the same input file is opened.
70
In this toolbar, it is possible to
modify the run-time display.
Figure 84: The run-time window after the completion of all 40 steps and selecting a crack
filter of 0.1 mm.
Normally the program displays all cracks, even very small cracks that are normally no
visible. A somewhat cleaner display of the main crack can be obtained by introducing a
proper crack filter. A crack filter can be introduced by selecting the button in the
toolbar to the left of the main run-time window (see Figure 85). Often an appropriate
minimal crack width to be displayed is 0.0001 m, i.e. 0.1 mm.
72
Figure 86: The graph display when monitoring after steps is selected.
After the analysis is completed it is possible to return to the pre-processor for making any
necessary changes, such as for instance adding more load steps, or proceed to the post-
processing stage of the analysis by clicking the post-processing button in the top-right
corner.
Other important button is the move button . If this button is selected and the cursor is
inside the main graphical window and the left mouse button is pressed, the structure
follows the mouse movements.
The button activates the zoom. The zoom area is selected by pressing the left mouse
button once; selecting the desired area and the zooming operation is performed by again
pressing the left mouse button. Any zoom operation can be cancelled by selecting the
zoom extent button , which changes the view such that the whole structure fits into the
main window.
The first step in post-processing is to select the analysis step (i.e. load step), from which
the results are requested. The program loads the data for the requested load step into the
computer memory, and fills in appropriately the lists of available output quantities. The
type of analysis and used material models determine the available output data. The process
of selecting a display showing the deformed shape with contour areas of maximal principal
strain is depicted in Figure 87.
In the case of reinforced concrete structures, it is often important to display result
quantities along the reinforcement bars. Reinforcement data can be visualized by
deselecting 3D results as it is described in Figure 89 and Figure 90. There it is possible to
see the two available methods for visualizing reinforcement data: either by using different
colors or 2D diagrams.
Figure 87: The post-processing window containing contour areas of maximal principal strain, cracks and
deformed shape for the last load step 40.
76
The best crack visualization is obtained by turning of the
deformed shape.
Figure 88: Filtered crack display along with the contour areas of maximal principal strains.
Figure 89: The display of reinforcement bar stresses activated by clicking an appropriate labels and check boxes
in the toolbar along the left side of the program window.
Another important feature is the possibility to cut the analyzed structure by an arbitrary
plane and display results on this plane. The option is activated by selecting the button
below in the Section toolbar. This opens a window showing a list of currently defined cuts
(see Figure 91). There are no cuts now, but a new one can be created by selecting the Add
button. This opens another window that is shown in Figure 92, for cut definition.
78
Cut definition is started by clicking this button in
the Section toolbar.
Figure 92: Cut definition window. A cut parallel to YZ plane is selected with origin at 0.5 m and named “Cut 1”.
In this window (see Figure 92), it is possible to see the cut plane as it is intersecting the
structure. After the cut is created and saved using the Add button, it is possible to select it
in the list box in the Section toolbar. This hides the whole structure, and shows the
selected output quantity only on the predefined cut plane as it is shown in Figure 93.
Figure 93: The display of maximal principal strains on the predefined cut.
There are many possible displays of results in ATENA, the user is encouraged to explore
the available menus in ATENA post-processor or to consult the ATENA User’s Manual
for more details. The subsequent figures summarize some of the possible methods for
displaying the results from ATENA analysis.
80
Figure 94: The display of un-deformed mesh outline and element cracks in the interior of the structure. The
crack filter of 0.3 mm is used.
Figure 95: Un-deformed mesh outline with displacement vectors at finite element nodes.
82
The Figure 96 and Figure 97 show that there exist a large stress concentration at the
loading and support steel plates. It is understandable, but this affects the color scale that is
automatically selected such that it covers the whole stress range in the current figure. Very
often this is not desirable, since it would be more interesting to learn about the stress
distribution in the beam. With this scale setting, it is not possible since almost the whole
beam is covered by a single color. There are two ways how to resolve this problem and
obtain a better color distribution. One method is to deactivate the automatic color scale and
define a new color scale (see Figure 98) that can be for instance saved for future use.
Figure 98: The boxes at the bottom of the Result toolbar can be used to create a user-defined color scale.
Another method for changing the color scale is to activate the display of only certain parts
of the structure. Then the automatic color scale is adjusted based on the maximal and
minimal values of the active part of the analyzed structure. In this case, for instance, it
would be desirable to display only the concrete beam without the steel plates. This feature
is called “activity” in ATENA 3D and an activity can be selected in the Activity toolbar in
the toolbar window on the left. The activity list is currently empty since no activities have
been defined so far. New activities can be selected only in the pre-processor.
It is possible to return to the pre-processing part of the program by selecting the button
in the top right corner of the post-processing window. This operation does
not delete any of the calculated results. The results can be however deleted automatically
by the program, if certain editing operations are performed in the post-processor. However,
the user is always notified and warned if certain operation can result in the loss of
calculated results. At any time it is possible to return to the post-processor and continue
with the post-processing of the analysis results.
Once the pre-processor is selected, and ATENA changes to the pre-processing mode, a
new activity can be defined by selecting the Activity item in the Input data tree on the left
(see Figure 100).
Figure 101: Display of minimal principal stresses on the activity “Concrete beam”.
84
After the activity definition, it is possible to return to the post-processor by selecting the
button . The Figure 101 shows the program display if the previously defined
activity “Concrete beam” is selected as well as the display of minimal principal stresses. It
can be clearly seen that the new display is much more representative and gives a better
understanding about stress distribution in the structure.
The active post-processing window can be printed from the menu item File | Print
graphic… or copied to the clipboard from Edit | Copy. The copied picture can be for
instance pasted to a Microsoft Word document.
It is possible to modify some parameters controlling the display on the screen or on paper
with the help of the dialog Options | Display and Options | Settings.
Figure 103: The window for editing graph parameters with the description of some of its important features.
The selected diagram can be printed in the same manner as it was described on page 85.
The numerical values of the monitored quantities can be obtained from the text output that
is described in Section 5.4.
86
5.4 Text Output
This section describes another form of output from the program ATENA 3D. The text
output can be used to obtain numerical data at finite element nodes, elements, integration
points or monitoring points.
The text output is selected from the menu item Files | Print text…. This selection opens
the window that is shown in Figure 104. The window is composed of two main sub-
windows.
The left-hand window contains a tree structure of the available data types and load steps.
The requested data should be checked in this tree, and then by selecting those data an
alpha-numerical output will be automatically created in the right-hand window.
The contents of this window can be printed, saved to a file or copied to another program
using the system clipboard.
Figure 104: The program window for the definition of alpha-numerical output.
Figure 105: The step information window contains the input and output files from the finite element analysis.
It is possible to view the contents of the various data streams for each analysis step, which
can be selected from the pull-down list at the top of the window. The content of each data
stream can be examined by selecting an appropriate tab at the top part of the window.
The input stream contains the commands that were passed from the pre-processor to the
analysis module. In the first step, it contains the definition of the numerical model. In the
subsequent load steps it contains the definition of supports, loads and solution parameters.
The format of this file is described in the ATENA Input File Format manual [4]. The
advanced users can modify the contents of this file by copying and pasting it into an
external editor.
Then such an input file (it is recommended to use the extension *.inp for these files) can be
loaded into the ATENA 3D pre-processor using the command File | Open other |
Analysis control file. Only users experienced with the program ATENA and the format
of this file should modify the input file, otherwise they can damage their data, which may
then become unusable.
The output stream contains the output from the analysis module. Normally this stream is
empty since it is used later when text output is requested.
The message stream contains the information about the analysis progress as they appeared
also in the interactive window during the non-linear analysis.
The error stream contains error and warning messages from the analysis modules. This
stream should be examined for errors that might have occurred during the numerical
calculations.
88
6. CONCLUSIONS
This tutorial provided a step by step introduction to the usage of ATENA 3D on an
example of a reinforced concrete beam without shear reinforcement. Although this
example is relatively simple from geometrical and topological point of view, it is not a
simple problem from the numerical point of view. Due to the missing shear reinforcement
the beam fails by a diagonal shear crack, which is very difficult to capture using smeared
crack approach.
This example demonstrates the powerful simulation capabilities of ATENA for modeling
the brittle failure of concrete structures. Even with a rather coarse mesh, which was used in
this demonstration example, the diagonal shear crack was successfully captured. Further
improvement of the results can be achieved by decreasing the finite element size to for
instance 8 elements over the beam height.
The objective of this tutorial is to provide the user with basic understanding of the program
behavior and usage. For more information the user should consult the user’s manual [2] or
contact the program distributor or developer. Our team is ready to answer your questions
and help you to resolve your problems.
The theoretical derivations and formulations that are used in the program are described in
the theory manual [1].
The experienced users can also find useful information in the manual for the analysis
module only [4].